AN INVESTIGATION OF SOME HORMONAL BASES FOR ABSCISSION IN COWPEA (VIGNA UNGUICULATA L. WALP.)’ BY AKINBO AKINWÜMI ADESOMOJü B.Sc. (Hons.) Cheuu (Ibadan) Ä thesis in the Department of Chemistry Submitted to the Faculty of Science in partial fulfilment of the requirements for the degree of DOCTOR OF PHILOSOPHY of the UNIVERSITY OF IBADAN MAY 1977 UNIVERSITY OF IBADAN LIBRARY 2 ABSTRAC! The Investigations carried out on the abscission problem in cowpea are reviewed. The Isolation 5 characterization 9 physiological roles5 chemistry5 biosynthesis and metabolism of the various groups of plant hormones are also reviewed. Using biological assays and combined gas-liquid chromatography- rnass spectrometry (GC-MS)* sorne of the hormones in the extensively purified acidic ethyl acetate extracts ohtained froni 2-day old and 6-day old cowpea fruits were examined. Biological. assays indieated the presence of only Inhibitors in the 2-day old fruits but inhibitors as well as gibberellins and auxins were indieated to be present in the 6-day old fruits, GC-MS analysis of the extract frorn 2-day old fruits afforded the identification of the known inhibitors, abscisic acid and phaseic acid. 61-hydroxymethyl abscisic acid was also identified in the extract and this is the first reported evidence that 6!-hydroxymethyl abscisic acid occurs naturally. Several plant hormones'were identified (GC-MS analysis) in the extract from 6-day old fruits. These.were abscisic acid* phaseic acid5 dihydrophaseic acid* 1iso1dihydrophaseic acid5 61~hydroxyinethy abscisic acid; gibberellins  * A^5 Aß* lisol A.^* and A^. UNIVERSITY OF IBADAN LIBRARY 3 Gibberellins A_ and A^q were also believed to be probably present, Two components5 believed to be two new gibberellins were also identified in-the extraet and were tentatively'called gibberellins X and Y. Tentative structures were assigned to these two new gibberellins, Purified acidic ethyl aeetate extraet obtained from fruits that were over si_x days old was also analysed on the GC-MS. The result was essentially similar to that obtained for the extraet from the 6-day old fruits, The erude acidic ethyl aeetate extraets from 6-day old seeds and the fruit walls of the 6-day old fruits were also examined on the GC-MS. Several gibberellins were tentatively idernzified in the extraet from the seeds but only one gibberellin could be identified in the extraet from the fruit walls, The methyl esters of 16a-hydroxy5 17-hydroxy5 and 16a517~dihydroxy derivatives of gibberellin Aqq and the 16-epimers of the last two compounds were synthesized from gibberellin A . This was done in order to correlate the structures that were tentatively assigned to the two newT gibberellins with. the. natural compounds, The disparity in the hormonal contents of the 6-day old and 2-day old fruits is discussed in relation to the ahscission problem. a_n cowpea. UNIVERSITY OF IBADAN LIBRARY UNIVERSITY OF IBADAN LIBRARY 5 ~ I thank all the members of staff of the Chemistry Department for their help and friendliness. I am grateful to rny fellow researeh students5 particularly my F103 colleagues for their co-operation and friendliness5 and to Mr. Banji Akanni for neatly typing this thesis. I also wish to thank Dr. Ojehomon for his advice and the Director5 National Cereals Institute«, Ibadan for generously supplying the cowpea. seeds . I am grateful to the Rockefeiler Foundation and the Inter- University Council for financial aid which enabled me to carry out this work. Finali}?' I thank all those people5 too numerous to mention who have contributed in one way or the other to the completion of this proje-ct. UNIVERSITY OF IBADAN LIBRARY 6 We certify that this work was carried out by Mr. Akinbo Akinwumi Adesornoju in the Department of Chemistry, University of Ibadan, Nigeria. Supervisors:- . J_ .—I —. V* —•O>- k—ogun, B— .—S c. ~ —S —p e_ c_ ia_ l_- — —( —L —on— d— o~n)~, Ph.D. (Ibadan), D.I.C. (London), Senior Lecturer in the Department of Chemistry, University of Ibadan, Nigeria. D.E.U. Ekong, B.Sc. (London), Dip.Chem.; Dr, rer. nat. (Heidel), Professor in the Department of Chemistry, University of Ibadan, Nigeria. N.O. Adedif>e, B.Sc. (Agric.), Ph.D. (Brit. Colum.), M.I. Biol., Professor in the Department of Agricultural Biology, University of Ibadan, Nigeria. MAY 1977 UNIVERSITY OF IBADAN LIBRARY 7 CONTENTS Pages ABSTRACT ... ... 2 ACKNOWLEDGEMENTS 4 INTRODUCTION ... 12 Gibberellins 16 Auxins 78 Cytokinins ... 91 Abscisic acid 104 Ethylene ... 121 RESULTS AND DISCUSSION 124 CONCLUSIONS ... 211 EXPERIMENTAL 213 APPENDIX 245 REFERENCES ... 252 UNIVERSITY OF IBADAN LIBRARY 8 LIST 0F FIGURES Fig. Page? i. The Gibberellins.......... . ... • . 16a 2. Nomenclature of GAs. ... ... • 18 3. Some Auxins. ... ... ... • 82i 4. Some metabolites of IM. ... ... • 90 5. Some cytokinins ....... ... • 94 & 1 6. Gas chromatogram of the MeTMSi derivative of AE extract from 6-day old fruits • 126a 7. Halo bioassay on column fractions of AE extract from 6~day old fruits 126a 8. Gas chromatogram of the MeTMSi derivative of column fraction *4 of the AE extract from 6-day old fruits. 130a 9. Mass spectrum of MePA ... ... 130a 10. Mass spectrum of MeTMSi ether of DPA * 130a 11. Mass spectrum of the MeTMSi ether of TisoDPAT 132a 12. Mass spectrum of MeTMSi ether of 61-hydroxymethylABA ... ... * 13.4a 13. Mass spectrum of MeABA ... ... 0 134a 14. Mass spectrum of the MeTMSi ether of 1isoGA o! 140a 15. Mass spectrum of MeA oTMSi ... ... 142a• UNIVERSITY OF IBADAN LIBRARY 9 LIST OF FIGURES FFg« ' Pages 16. Hass spectrum of MeTMSi ether of !gibberellin Xf 143a 17. Gas chromatogram of the MeTMSi derivative of colurnn fraction 5 of the AE extract from 6-day old fruits........... ... ... 146a 18. Mass spectrum of MeA 6TMSi ... ... ... 146a 19. Mass spectrum of MeTMSi derivative of lgibberellin Y! ... ... ... ... 149a 20. Gas chromatogram of TLC Rf 0.5 - 0.6 of fractions 5 to 10 of AE extract from 6-day old fruits (MeTMSi derivative) ... ... ... 159a 21. Wheat coleoptile bioassay on TLC Rf zones of the AE fraction from 6-day old fruits ... 164a 22. Wheat coleoptile bioassay on paper Rf zones of the AE fraction from 6-day old fruits. 164a 23. Wheat coleoptile bioassay on TLC Rf zones of the AE fraction from 2-day old fruits. 164a 24. Mass spectrum of Mel6a;>17-dihydrox37' îTMSi 188a 25. Mass spectrum of Mel7-hydroxyA^fMSi ... 198a 26. Mass spectrum of epimer of Me 17-hydroxyA^TMSi 204a UNIVERSITY OF IBADAN LIBRARY 10 LIST 0F TABLES Tables Pages la. Chemical shift (ö5 ppm) of some protons in. some C^^-gibberellin methyl esters. 31 & 32 lb. Chemical phift (o5 pprn) of some protons in some C2o*~gifrbereilin Methyl esters. 2. Results obtained in the halo bioassay carried out on the column fractions of the acidic extract obtained from six-day old fruits. 129 3. Plant hormones identifiea in 6~day old fruits of Vigna unguiculata (L. Walp) by GC-MS. 162 4. Results obtained in the wheat coleoptile Segment bioassay on TLC Rf zones of the acidic ethyl acetate extract from 6-day old fruits, 16*4 5. Results obtained in the wheat coleoptile Segment bioassay on Rf zones of the paper chromatography of the acidic ethyl acetate extract from 6-day old fruits, 167 6. Plant, hormones that were identified in 2-dav old fruits of Vigna unguiculata (L.Walp) by GC-MS. 170 UNIVERSITY OF IBADAN LIBRARY 11 LIST OF TABLES Tables Pages 7. Results obtained in the wheat coleoptile segment bioassay on TLC Rf zones .of acidi,c ethyl acetate extract from 2 day old fruits. 172 UNIVERSITY OF IBADAN LIBRARY 12 I N T R O D U C . T . I O N Cowpea belcngs to the Family Leguininosae of flowering plants. The cowpea is a very important food crop in tropical Africa. S e v e ra l euXVWcsrc, Qf cowpea are cultivated in Nigeria for their dry, edible seeds. These provide a very good souree of the much needed protein. They normaliy grow as spreading, sub-erect or erect annuals with moderately long pods. Ojehomon studied the flowering and fruiting patterns in some cowpea cultivars. The cowpea inflorescence consists of a central peduncle and the flower buds are arranged in four to eight alternate racemes along the peduncle, Ojehomon 2 reported that, in the cowpea, about 100 to 150 flower buds are produced per plant. Most of these flower buds drop off before they develop into mature fruits. Only about six to sixteen per cent of the flower buds formed ultimately developed into mature fruits. ' The cowpea therefore exhibits an excessive abscission of buds and immature fruits and this seriously limits the grain yield of the cowpea. It is believed.that if abscission can be successfully reduced, the grain yield of cowpea could be increased considerably. In the cowpea inflorescence, the flowers of the first (lowest) raceme open first followed by those of the second raceme after two to four days^. Flower opening thus begins at the bottora of the peduncle and proceeds sequentially upwards. In most cowpea cultivars UNIVERSITY OF IBADAN LIBRARY - 13 like "New Eran and "Mala", only the two lateral flower buds of the first raceme develop into mature fruits. The flowers of the other raceme higher up on the peduncle are shed at different stages of development. Ojehomon 1 * 2 5 i| reported that the flower buds of each raceme ij V ‘ • however developed into mature fruits when the older fruits below 1 were removed. This implied that the abscission of the buds and immature fruits of the upper racemes was due to the presence of mature fruits on the lowest raceme. Ojehomon 2 postulated some hypotheses to explain the implication of the lowest, older fruits in the abscission of the upper, younger ones. Two of these hypotheses were: (a) that the older fruits were monopolising the available mobile nutrients and the younger fruits therefore starved and dropped; (b) that the abscission of the flower buds and immature fruits might be stimulated by a substance or substances produced in the older fruits. Ojehomon 5 tested the first hypothesi.s by .studyi.ng the. dist.ribution of 14C-assimilates in the cowpea inflorescence using two cultivars of cowpea ("New Era" and "Adzuki"). He administered radioactive carbon- dioxide to the leaves and then used autcradiography to test for the presence of radioactivity in the inflorescence. He found that radioactive assimilates were translocated to all fruits, flowers and flower buds in the inflorescence. This was taken to impiy that the UNIVERSITY OF IBADAN LIBRARY monopolization of nutrients by the older.fruits was not an important factor in the abscission of the upper, younger fruits and flower buds. The occurrence of a monopoly of nutrients by the older fruits •;V- “ ‘ g was however demonstrated from the results that Adedipe and Ormrod obtained from I a study of the distribution pattern of 32P in two cultivars of cowpea. These two cultivars were !,Early Ramshorn” which exhibits a relatively high degree of abscission of flowers and fruits; and "Adzuki" which exhibits a relatively low degree of abscission. They showed that there was a preferential accumulation of P by the fruits of the lowest raceme in both cultivars. They also showed, quantitatively, that the raceme 1 fruits of "Early Ramshorn", the cultivar that showed a higher abscission degree, were a more potent sink for 32 .r_ than those of "Adzuki". Further studies on the distribution of radioactive nutrients in the cowpea by Adedipe et al 7 also confirmed the above Observation. They studied the distribution of 14C in two cultivars of cowpea, nMalaf! and "Adzuki". * t!Malan exhibits a higher degree of the abscission problem than !,Adzukin. They obtained quantitative results that showed that the raceme 1 fruits were more effective in mobr.l.is.i ng 14C assimilates away from the younger, upper fruits in nMalalt than in !tAdzuki!!. Their results thus showed that the competition for available mobile nutrients is an important factor in the abscission of flowers and immature fruits, UNIVERSITY OF IBADAN LIBRARY 15 Extensive investigations 8 by several workers have shown that, generally, in the plant kingdom, abscission is mainly under the control of five plant hormones. These are the anxins, gibberellins, cytokinins, abscisic acid and ethylene. The first three are growth Promoters while the last two are growth Inhibitors. One important role of the growth promoters in relation to abscission is that of the mobilization of nutrients to developing fruits. Organs which have low concentration of these growth promoters are therefore deprived of nutrients by organs which have higher concentration of the growth promoters. Seith and Wareing reported that in the French bean (Phaseolus vulgaris), auxin caused mobilization of 32.P . They also found that a combination of auxin with gibberellin or auxin with cytokinin was even more effective in the mobilization of 32.P. Adedipe et 7al also reported that exogenously applied benzyl adenine, a synthetic eytokinin, was effective in redirecting li+C to the advantage of treated fruits in the upper raceme of a variety of cowpea that is known to ex/hibit a high degree of abscission. The seed yield of cowpea has been reported to be substantially increased through judicious foliar spray by Aba El-Soad and co-workers 10a. They sprayed cowpea three times with 25ppm of GA.g (gibberellic acid) at 10-day intervals. There was an increase in the number of flowers per plant and a reduction in the percentage of UNIVERSITY OF IBADAN LIBRARY 16 flower drop. A lot of Chemical studies have been carried out leading to the Identification and isolation of the various plant hormones. Sensitive Chemical methods of analysis have been employed on account of the very low endogenous level of the hormones in plants. These include the various Chromatographie and spectroscopic techniques in conjunction with biological Essays. These studies have led to the elucidation of the Chemical nature of the plant hormones. The Gibberellins The gibberellins (GAs) are a group of naturally occurring tetracyclic diterpenoid acids which have a hormonal function in higher plants1^3. They were initially discovered10c as secondary metabolites of the fungus Fusarium moniliforme (Gibberella -fujikuroi). This fungus is the causal agent of the bakanae or "foolish seedling" disease of rice. Many higher plants have been shown to contain gibberellins and it is now believed that the gibberellins are present in most, if not all9 plants. At present fifty-one*^0^ gibberellins are known (Fig.l) of which about forty occur in higher plants. Several glucosyl ethers and esters of gibberellins have also been isolated, and characterized. 1. Physiological roles The major physiological effects 11 ? 12 include; UNIVERSITY OF IBADAN LIBRARY 16a COOH * COOH COOAHA23 m COOAH >'j COOH^NvOH LCHi CH3 COOH A33 A iyt? Fig- 1. THE G IBBER ELL IN S UNIVERSITY OF IBADAN LIBRARY 17 (a) Stimulation of growth. extension in many intact plants, (b) reversa! of genetic dwarfism, (c) induction of Stern growth in rosette plants, (d) Stimulation of flowering, (e) breaking of dormany, and / (f) invölvement in the synthesis of many enzymes, for example a-amvlase in cereal aleurone. 2. Nomenclature In order to prevent confusion and for mere convenience, the gibberellins have been allocated 13 A-numbers. The known gibberellins have thus been allocated the trivial names GA^ to GA51 (Fig,1), Apart from the trivial nomenclature used for the gibberellins, there is also a systematic nomenclature 1*4- based on the I.U.P.A.C. System. This systematic nomenclature is based on the trivial name GIBBERELLANE which is used for the ring system[1] shown in figure 2. The numbering which is as shown is consistent with the numbering of other tetracyclic diterpenes. The gibberellins all have the enantiomeric stereochemistry of the gibberellane skeleton. In the systematic nomenclature, gibberellin A1 [Fig. 1] is ent - 3a, 10, 13-trihydroxy--20-norgibberell~16-ene*-7, 19-*dioic acid 19, 10~lactone. The conformational terms a, and g have their normal stereochemical connotations of below and above the plane of the UNIVERSITY OF IBADAN LIBRARY 18 molecule respectively. The connotations are however reversed if the name of the molecule is preceeded by ent. Both the systematic nomenclature and the trivial System are used in this thesis. C17 FIg . 2 Nomenclature of GAs. UNIVERSITY OF IBADAN LIBRARY 19 3* Chemical Structure The gibberellins can be sub-divided into two groups. These are the C^-gibberellins which have the full complement of diterpenoid carbon atoms and the C 19-gibberellins which have lost carbon - 20. The C^Q-gibberellins are thus derivatives of ent- gibberellane [11] and the C 19-gibberellins are derivatives of ent- 20-norgibberellane [111] . The C^g-gibberellins are characterized by the 19 -> 10-lactone ring, with the exception of GA which has a 19 + 2-lactone. The carbon atoms C-l, C-2, C-3, C-16 and C-17 may be involved in double bond formation while hydroxylation(s) may occur at C-l, C-2, C-3, C-ll, C-12, C-13, C-15 and C-16. Carbon-7 is always present as a carboxyl group. The C .-gibberellins have the C-20 as -CH , -CH OH, -CH0, or /\) O Z. -C00H groups. Hydroxylations have been found at carbon atoms 2,3 and 13 and carbon atoms 7 and 19 are always present as carboxyl groups. Unlike in the case of C^-gibberellins, double bond occurs only between carbon atoms 16 and 17 in the C^^~gibberellins. *4. Structure Determination The methods employed for the determination of the structures of the gibberellins have become progressively more sophisticated with the development of the various physical methods for the determination of structures of organic moiecules. The original UNIVERSITY OF IBADAN LIBRARY 20 studies on the structural elucidation of the gibberellins centered on GA^. Gibberellin A0 5 also called gibberellic acid is obtained on a large scale from the commercial fermentation of the fungus, Fusarium moniliforme. The structure of GA [IV] was determined o mainly by Chemical methods with the aid of ultraviolet and infrared spectroscopy. Cross 15 showed that GA o was a tetracyclic dihydroxylactonic carboxylic acid. He found that GA o forraed a monomethyl ester on treatment with excess diazomethane. It also formed a raonoacetyl derivative on treatment with acetic anhydride in pyridine. The infra red spectrum of the methyl ester of this monoacetyl derivative of GA^ still showed a hydroxyl band at 3510cm \ This showed that there was a second hydroxyl gro'up present and this was considered to be tertiary because of the difficulty of acetylation. The presence of the y-lactone ring was indicated by the presence of a strong band near 1780 cm ^ in the I.R. spectrum of GA^. Hydrognation showed the presence of two carbon, carbon double bonds. Further Information on the structure of GA^ was obtained from the elucidation of the structures of some degradation products of GA . These were allogibberic acid[-V], gibberic acid [VI] , and o gibberene[VII]. Allogibberic acid was obtained 15 when GA^ was treated with dilute - HCl at 55-65° for about 2xp hours. Gibberic acid was obtained as the main product when either GA^ or allogibberic UNIVERSITY OF IBADAN LIBRARY 21 acid was refluxed with dilute hydrochloric acid for about 1 hour. Gibberic acid and allogibberic acid both gave gibberene on dehydrogenation with selenium. SCHEfvfE 1 UNIVERSITY OF IBADAN LIBRARY 22 Mulholland and Ward 16 showed that gibberene was 1,7-dimethyl- fluorene[VII] by oxidative degradation to the known fluorene-1* 7-dicarboxylic acid. This structure for gibberene was confirmed by synthesis^ # % \ . . . . Gibberic acid was deducea to be a keto~acid because it formed an ester and an oxime. The I.R. spectrum showed a band at 1741 cm 1 indicating that the ketone group was present in a five-membered ring 1‘ 5 . The presence of the hexahydrofluorene nucleus was established 17 by selenium dehydrogenation to l97-dimethylfluorene, (gibberene)[VII]. The Position of the carboxyl group was determined 17 by the degradation of gibberic acid to methyl 1,7-dimethyl fluorene-9-carboxylate[VIII]. This was identicai to a specimen prepared by carboxylation of the 9-lithium derivative of l 97-dimethylfluorene with solid carbon dioxide9 followed by methylation with diazomethane. The structure of gibberic acid was finally established 17 to be as shown m structure[VI]. The structure of allogibberic acid [V] was determined as follows. Hydrogenation showed the presence of an ethylenic double bond which was shown to be exocyclic because ozonolysis of allogibberic acid gave formaldehyde and a nor-ketone[IX] 18 . The methyl ester of allogibberic acid was isomerised with acid to methyl gibberate. This showed that the carboxyl group was at the same Position as in gibberic acid. The I.R, spectrum of methyl allogihberate showed an absorption at 3460 cm ^ indicating the UNIVERSITY OF IBADAN LIBRARY 23 presence of a hydroxyl group. This hydroxyl group was deduced to be tertiary because of the difficulty of acylation. The structure of allogibberic acid was established as [V] and the absolute configuration followed from measurements of optical rotatory dispersion on the keto-ester[X] 19 ’ 20 obtained from the nor-ketone[IX] by oxidation with sodium bismuthate followed by methylation with diazomethane. SCHEME 2 H CIXJ UNIVERSITY OF IBADAN LIBRARY 24 The elucidation of the structure of allogibberic acid paved way for more progress in the determination of the structure of gibberellic acid (GA o). As stated earlier, Cross 15 had shown that GA^ was a tetracyclic dihydrox^, lactonic carboxylic acid- with two carbon, carbon double bonds. The methyl esters of both gibberellic acid and allogibberic acid gave methyl gibberate on refluxing with dilute HCl. This indicated that the position of the carboxyl group in both GA o and allogibberic acid were the same. The ozonolysis of methyl ester of GA o gave formaldehyde and uWvvr\^Vsi\^c\ keto-acid[XI]2.A The methyl ester of this keto-acid gave the keto~ester[X] on treatment with acid. This same keto-ester had also been obtained by oxidation, followed by methylation of the nor-ketone[IX] obtained from the ozonolysis of allogibberic acid. This showed 21 that the B/C/D rings of both GA o and allogi .bbe.ric .acid UÄce_ SCHEME 3 Me of CIV3 (1) ch2n 2 CX3 M2GA3 (2) Hof H+ c x n UNIVERSITY OF IBADAN LIBRARY 25 similar, and the formation of allogibberic acid from GA therefore involved only the aromatisation of ring A. The y-lactone and the secondary hydroxyl group must therefore be in ring A of GA O. The hydroxyl group was shown to be allylic by oxidation with manganese dioxide to give an a,ß-unsaturated ketone. The ring A of GA o must I also contain the rnethyl group which appeared in position 1 of the fiuorene degradation products. There was a high yield of acidic hydrogenolysis products on catalytic hydrogenation of the ring A double bond of the rnethyl ester of GA o. This was taken to imply an allylic lactone System. The position of the y-lactone was deduced from this. The position of the allylic hydroxyl group was established by stepwise degradation. The correct structure of GA^ was finally deduced by Cross et_ al 22 and the deduction was supported by nuclear magnetic resonance studies 23. There were some controversies about the orientation of the lactone ring 24 and the 9 - H was considered to have an a-configuration. These controversies were finally resolved through x-ray and circular dichroism studies 25-27 . The full structure and- stereochemistry of GA^ was thus established as shown in [IV]. The structure of many of the other C^-gibberellins were established by relating them chemically to GA^ either directly or indirectly. For example GA^ [XII] was shown to be dihydroGA^ by the UNIVERSITY OF IBADAN LIBRARY 26 fact that controlled hydrogenation of MeGA with palladium-carbon catalyst (stopping after the absorption of 0.94 mole of H^) gave MeGA 28 SCHEME A GA 5 CXIIIl methyl ester UNIVERSITY OF IBADAN LIBRARY 27 Dehydration of MeGA^, by refluxing the 3-toluene~p~sulphonyl derivative of MeGA.1 in collidine gave GA_ [XIII] as the methyl. 0 ester 29 . Thus GA 5 was established as dehydrated GA 1. GA^ [XIV] was related to GA^ as follows: Ozonolysis of MeGA^ gave the nor-ketone [XV] Acetylation of this rior-ketone followed by reductive de-acetoxylation (by boiling for 36 hours with zinc and acetic anhydride) of the resulting diacetate gave the mono acetate of MeA^ nor-ketone [XVI] 30 SCHEME 5 GA\ CXII □ methyl ester R 1 R 2 R 3 C X V D - OAc - c h 3 = 0 CX IVü - O H - H = CH2 DCv'IJj -O H „ H ^ c h 3 " " O H UNIVERSITY OF IBADAN LIBRARY 28 GA^ [XVII] was established as hydrated GA^ [XIV] This was because treatment of GA^ with dilute hydrochloric acid for two days at room temperature gave GA^. Since the time after the determination of the structure of GA o* there have been a lot of progress in the development of physical methods for th!e-determination of the structures of organic compounds. Nuclear magnetic resonance spectroscopy and mass spectrometry in particular have been extensively used in the determination of the structure of the numerous other gibberellins. (a) NMR Spectroscopy Hanson 32 reported the nuclear magnetic resonance of some gibberellin derivatives in deuterochloroform (CLC%) and deuteropyridine (CDJ oN), Since then there have been reports of the NMR of the various other gibberellins that have been isolated5 and the data of the Chemical shift of some gibberellins have been published^ [Table 1] . UNIVERSITY OF IBADAN LIBRARY 29 The 5-H resonance and the 6-H resonance are very important characteristic features of the proton nuclear magnetic resonance spectra of gibberellins. The 5-H is a ß-axial substituent on ring A and therefore 1,3-diaxial transannular effects operate from the 3 position. The position of the 5-H resonance is generally very susceptible to Substitution in ring A. Thus in gibberellin A y methyl ester (XVIII], = H) the 5-H resonance appears at 62.47 in CDC1 ̂. In gibberellin A » methyl ester ([XVIII], R -L = R O = H, R^ = ßOH), there is a clownfield shift of the 5-H resonance to 63.15. This deshielding is consistent with transannular 1,3-diaxial interac- tion. In the 3-epimer of GA^ methyl ester (3-aOH), the position of the 5-H resonance is very near that observed for GA y methyl ester since the 1,3-diaxial interaction is absent in the structure of the 3-epimer of gibberellin. A^ methyl ester. The deshielding due to 1,3-diaxial interaction is amplified in deuteropyridine. This effect is therefore very useful in determining the stereochemistry of the substituent on C-3. Unlike the 5-H, the 6-H is an a-substituent on ring B. The Chemical shift for the 6-H is usually fairly constant within the ränge 62.83 for most C^g-gibberellin methyl esters in deuterochloroform. The C-17 methylene protons usually give two broad peaks at around 65.0 in deuterochloroform. The presence of a 13-hydroxyl group causes a significant downfield shift in the position of one of these UNIVERSITY OF IBADAN LIBRARY 30 peaks. This dovmfield shift is further increased when the NMR spectrum is determined in deuteropyridine. A comparison of the Chemical shifts of the 017 protons in gibberellin methyl ester (XVIII, R -L = R = R O = H) and gibberellin A methyl ester£ ± (XVIII, R = H, R = R = OH) in both deutrochloroform and -L £ c5 deuteropyridine in Table la will illustrate this point. This phenomenon is therefore useful in determining whether or not the 13-hydroxyl is present. The Chemical shift of the C-18 methyl protons is sensitive to the presence of other substituents in ring A, UNIVERSITY OF IBADAN LIBRARY 31 Table la Chemical Shift (S,pprn) of some protons in some C ^ —^ gi, bberellin rnethyl esters, after Takaha' shi, N33 CLrCH3 C16 = CH2 H - 1 H - 2 H - 3 H - 5 H - 6 A1 (i) 1.13 5.23, 4.92 3.75 3.1V 2.65 (ii) 1.44 5.59, 5.04 4.08 3.72 2.97 A2 (ii) 1.45 1.45 (CH > 4.08 3.79 3.02 A3 (1) 1.23 5.25, 4.94 6.30 5.87 4.08 3.17 2.75 (ii) 1.54 5.45, 5.02 6.41 6.11 4,49 3.69 3.05 \ Ci) 1.13 4.95, 4.83 3.80 3.15 2.65 (ii) 1 .4 3 4.96, 4.86 4.06 3.68 2.93 A b (i) 1 .22 5.21, 4.92 5.80 5.65 2.60 2.77 (ii) 1 . 3 3 5.60, 5.06 5.69 5.69 2.88 2.99 A? (i) 1.23 4.95, 4.83 6.32 5.87 4.12 3.25 2.73 (ii) 1.54 4.97, 4.85 6.37 6.04 4.42 3.62 2.97 Ag (ii) 1.71 5.61, 5.03 4.38 4.22 4.01 3.17 A9 CD 1.07 4.91, 4.79 2.47 2.68 (ii) 1.12. 4.97, 4.87 2.61 2.86 Aio (l) 1.1 1.38 (CH O) 2.47 2.77 A16 (i) 1.12 4.95 4.20 3.95 3.20 2.72 A20 1.07 5.21, 4.90 2.50 2.67 (ü) 1.15 5.58, 5.05 2.70 2.91 UNIVERSITY OF IBADAN LIBRARY 32 Table la (contd.) [^- c h3 Ir 16 = UCHl 2 H - 1 H - 2 H - 3 H - 5 H - 6 A21(l) 5.20, 4.90 3.13 2.76 ' (ü) 5.58, 5.06 3.42 3.05 A22 1.71 5.00, 4.87 4.40 4.23 3.98 3.16 A35 (l) 1.15 5.04, 4.91 3.86 3.26 2.74 \ o ^ 1.10 4.96, 4.84 4.28 2.74 2.58 (ii) 1.27 5.00, 4.90 4.49 3.00 2.82 (i) NMR determined in CDC1 ö (ii) NMR determined in C_b D_oN UNIVERSITY OF IBADAN LIBRARY 33 - Table lb Chemical Shifts (6,ppm) of protons in some C^q . gibberellin methyl esters, after Takahashi N33 C,4 -CH 3 C^substituents C16 1= CH2 H -2 H - 3 H - 5 H - 6 A13 1.21 4.80, 4.73 3.89 2.51 3.79(CH OC0) A15 1.15 4.03, 4.42 4.90, 4.75 2.21 2.79 A17 (i) 1.09 5.06, 4.83 3.72 (ü) 1.23 5.42, 4.95 4.05 A19 (l) 1.15 9.73(CH0) 5.02, 4.80 2.43 3.87 (ü) 1.28 5.57, 5.10 2.46 4.16 A23 1.20 9.72(CHO) 5.17, 4.93 4.11 2.77 3.88 (ii) 1.60 5.53, 5.03 3.37 A24 (l) 1.11 9.62(CH0) 4.84, 4.76 2.19 3.81 A25 (l) 1.10 4.80, 4.73 2.07 3.78(CH.OCO) A27 (i) 1.24 4.46,4.42 4.95, 4.83 3.86 3.72 2.76 2.76 (ii) 1.69 4.58, 4.19 4.93, 4.79 4.35 4.18 3.27 3.04 A28 (i) 1.22 5.14, 4.91 3.99 2.58 3.78 (ii) 1.67 ' 5.53, 5.05 4.40 3.22 4.28 A36 (l) 1.22 9.68 (CHO) 4.92, 4.84 4.11 2.75 3.91 (ii) 1.59 9.96 4.94, 4.86 4.48 3.34 4.29 (C.H 0C0) Ag7 (1) 1.19 4.42, 4.07 4.90, 4.78 3.74 2.74 2.74 (i) .CDQ (ii) C5D5N UNIVERSITY OF IBADAN LIBRARY 34 The resonance usually appearing as singlet Is deshielded by the presence of either hydroxyl group or double bond in ring A. (Tables la & 1b). As an Illustration, in CDCl, o, the Chemical shift of the 018 methyl protons of GA y methyl ester (XVIII, R x =R x =R o=H) occurs at 61.07 whereas it occurs at 61.13 in GA^ methyl ester (XVIII, R x =H, R =ßOH, R ~aOH). The downfield shift is even more /. ö prominent in the NMR spectrum of gibberellin A methyl ester o (61.23) which apart from having the 3-hydroxyl also has a 1-2 double bond. The deshielding of the 018 methyl protons caused by the presence of a ring A double bond can be distinguished from that caused by a hydroxyl Substituent in ring A by determining the NMR in CDClg and deuteropyridine. Whereas deuteropyridine increases the deshielding due to hydroxyl substituent it does not affect the deshielding due to the double bond. (b) Mass Spectrometry Mass spectrometry is another technique that has become extremely useful in the determination of the structures of the gibberellins. There have been reports 36*~3° ̂of the high and low resolution mass spectra of gibberellin methyl esters and the low resolution mass spectra of the trimethylsilyl ethers of the methyl esters of many of them have been published39 UNIVERSITY OF IBADAN LIBRARY 35 (i) Methyl Esters Most C iy -gibberellin methyl esters show the molecular ion with moderate intensity. They also show prominent peaks at M^32> M-44, M-f+6, M-60, M-62, M-104, M-106, and M-12238. The M-32 and the M-60 peaks are due to the loss of CH O h and CH COOH respectively from the o o methoxycarbonyl group attached to C-6. The M+- 44 and the M*-v46 peaks are associated with the loss of CO^ and HCOOH respectively from the lactone in ring A. The M-62 peak is a characteristic feature of the mass spectra of GAs that have hydroxyl group in ring A. The peak which is normally prominent is due to the concerted loss of CO^ and H^O from the lactone and the hydroxyl group respectively. The M-62 ion is absent in the mass spectra of gibberellins that lack hydroxyl group in ring A. The prominent peaks at M-104 and M-106 are due to the loss of 'HCOOCH O + CO and HCOOH + HCOOCH o respectively.Z. They are present in the mass spectra of the methyl esters of most C iy -gibberellins. The M-122 ion is due to the loss of HCOoOCH + H^O + CO^* The M-IO*^ M̂ -106 and M-122 ions thus indicate that the loss of CO^ or HCOOH from the lactone in ring A and the elimination of the methoxycarbonyl group attached to C-6 can occur concurrentty. The mass spectra of the methyl esters of the C^Q-gibberellins also show the molecular ion with moderate intensity and prominent peaks at M̂ -32 and M*605 like those of the C^G*-gibberellins. The M■-V H4, M--v46 and the M-62 ions which are associated with the elimination UNIVERSITY OF IBADAN LIBRARY 36 of the lactone in ring A are not normally present. Anöther characteristic feature of the mass spectra of the methyl esters of the C^Q-gibberellins is the presence of a prominent peak at M-119/120 which is due to the loss of two moles of HCOOCH o from two methoxycarbonyl groups. This can be used to distinguish the C^Q-gibberellins from the C^-GAs since the '-•^"gibberellins are monocarboxylic and therefore cannot give rise to the M-119/120 ion. The only exception is GA^ which is the only C^-gibberellin known that'is dicarboxylic. The mass spectra of the methyl esters of the C^^gib• berellins also contain peaks *av t M-91/92 \-7hich are believed to arise from the loss of 59/60 (HCOOCH o) from one methoxycarbonyl group and the loss of 31/32 (CH oOH) from a second methoxycarbonyl group. M*+-■ 18 ions are pres.ent in the mass spectra of the methyl esters of the hydroxylated gibberellins. They are also present in the mass spectra of methyl esters of GA^ and GA^ which have epoxides in ring A. The mass ‘spectra of the methyl esters of gibberellins that ha/e C-13 hydroxyls are characterised by intense peak ät m/e 136. This is believed 38 to be fragment[XIX] formed from rings C and D. UNIVERSITY OF IBADAN LIBRARY 37 HX 1X3 Gibberellins that have the 016 hydroxyl, for example GA^ and GA have intense base peaks at m/e 43 [XX] in the mass spectra of their methyl esters. This ion is believed 39 to arise from ring D by the fragmentation pattern [XXI] -> [XX] . CXXZ3 LXXIIJ CXX3 UNIVERSITY OF IBADAN LIBRARY 4 -0 38 (ii) Trimethylsllyl ethers The mass spectra of the trimethylsllyl ethers of the gibberellin methyl esters are very diagnostic of the structures of the hydroxylated gibberellins. The low resolution mass spectra of the trimethylsllyl ethers of some gibberellins have been published by Binks eit al 39 J The mass spectra contain M-v’*- 31/32 and Mv-59/60 ions associated with the methoxycarbonyl group just as in the mass spectra of the methyl esters. Peaks at M-15 (-CI-oL )> M^-89/90 [(CH o ) oSiO-] , m/e 73 and m/e 75 [(CH o ) oSi-] are associated with the trimethylsllyl ether group. Hydroxylation at C-3 is characterized by the presence of prominent ion at m/e 129 which is accompanied in some cases by an M■-Y 129 peak. The ion at m/e 129 is believed 39 to have the structure [XXIII]> arising from ring A by the process shown in Scheme 7 below. SCHEME 7 R t CXXIVD + (CH3)3 S i -0 m/e ,129 CXX I I Iu UNIVERSITY OF IBADAN LIBRARY 39 The mass spectra of gibberellins having the 2,3-diol are characterized by the presence of an ion at m/e 217 corresponding to [XXVI]5 arising from ring A. They also show an ion [XXVII] at m/e 147 which is believed to come from the rearrangement of the vicinal3 253~trimethylsilyl ether groups. OSi (CH3)3 m/e 217 CXXVI□ m/e 1H7 [XXVII] The mass spectra of gibberellins that have hydroxyl attached to C-13 have intense molecular ions which are usually the base peaks. The molecular ion is of much lower intensity in the mass spectra of gibberellins that lack the 13-hydroxyl group. The mass spectra of the 13-hydroxylated gibberellins are also characterized by prominent peaks at m/e 207/208 [XXVIII] formed from rings C and D • UNIVERSITY OF IBADAN LIBRARY HO c x x v n n 16-hydroxylated gibberellins are characterized by intense peak which is the base peak at m/e 130. This ion [XXIX] is derived from cleavage of ring D probably by the process [XXX] -*■ [XXIX] SCHEME__8_ CH, .+ -> CH 0 Si(CH3>3 m/e 130 fO-Si(CH3)3 [XXIX] c x x x 2 UNIVERSITY OF IBADAN LIBRARY 41 The trirnethylsilyl ethers of the C^Q-gibberellins shov7 promi• nent M*~V" 90, M-91, or M"-V”92 ions just like the methyl esters. These are due to the loss of 59/60 (HrlCOOCH ) from one methoxycarbony 1 ö group and loss of 31/32 (-CH oOH) from a second methoxycarhonyl group. The mass spectra of the C^Q-gibberellins that have the C-20 as aldehyde group1 (GA y , GA z. o , GAo u ) have intense base peaks at M *v-*28 corresponding to the loss of CO from the aldehyde function. ¥ 5. Synthesis of the gibberellins There have been reports 40 5 41 in literature of the stereocontrolled total synthesis of gibberellin A and the formal total synthesis of gibberellin A^. Many partial syntheses of gibberellins have also vb een publish, ed,42,43 (a) Total Synthesis Nagata et al 40 described the stereocontrolled total synthesis of gibberellin A 1 b [XXXI] (a C2 0-GA) in the racemic form. They started with the intermediate enone [XXXII]. This already had the correct stereochemistry for the A, B and the lactone rings. The N-mesyl piperidine ring was expected to be stable to further elaboration VO 'OGL and^eadily converted to the lactone. UNIVERSITY OF IBADAN LIBRARY 42 Ms c x x x d c x x x i n The first stage in the synthesis was the contraction of ring B. This was carried out as follows: The enone starting material [XXXII] was refluxed with isopropenyl acetate in the presence of p-toluenesulphonic acid, yielding the dienol acetate [XXXIII] as the major product. This was reduced with sodium borohydride in alkaline medium giving the hydroxy olefin [XXXIV] as the main product. Oxidation of [XXXIV] with osmium tetroxide gave the triol [XXXV] which gave the ketoaldehyde [XXXVI] on oxidation with periodic acid. The ketoaldehyde was cyclised in contact with neutral alumina to give the tetracyclic aldehyde [XXXVIIa]. UNIVERSITY OF IBADAN LIBRARY 1+3 SCHEME 9 EXXXVJ UNIVERSITY OF IBADAN LIBRARY HU Configurations at C-6, C-8 and C~9 of [XXXVIIa] viere assigned on the following reasons. It was believed that the stereochemistry at 9 was retained throughout the transformation from [XXXIV] since no basic reagents which could have caused epimerization were used. * For the determination of the configurations of the 8-hydroxyl and the 6-forroyl group, some derivatives, [XXXVIIb] , and [XXXVIIc] were prepared and their infra-red spectra measured. The IR- results indicated that in [XXXVIIb], the 8-hydroxyl was hydrogen bonded to the 13-3 acetoxyl and the acetal oxygen in [XXXVIIc]. It also showed that the 13-3 hydroxyl was hydrogen bonded to the 8-hydroxyl in [XXXVIIc]. This showed that all the substituents at C-6, C-8 and C-13 are oriented cis and therefore Decisive evidence for the 3 configuration of the formyl group attached to C-6 was provided by the fact that mild oxidation with chromic anhydride-pyridine complex in methylene Chloride (Collins reagent) gave the 3 lactone [XXXVIII]. The next stage in the synthesis was the construction of the D ring. Compound [XXXVIIb] on Wittig vinylation followed by alkaline hydrolysis gave [XXXIX]. Jones oxidation of [XXXIX] gave [XL]. This was converted to the dienone [XLI] by dehydration with thionyl Chloride in methylene Chloride - pyridine at -73°. The dienone was then hydrocyanated by treatment viith an excess of diethyl aluminium cyanide UH in methylene Chloride at room temperature gi.vmg. th.e cxs- UNIVERSITY OF IBADAN LIBRARY 45 cyano ketone [XLII]. The ß configuration was assigned to the C~6 vinyl group because the IR spectrum of [XL].indicated that hydrogen bond occurred between the 0 6 vinyl and the 8-hydroxyl showing that they are cis. No epimerization was considered to have occurred at 0 6 during the transformation of [XXXIX] to [XLII]. SCHEME 10 UNIVERSITY OF IBADAN LIBRARY - 46 - Scherns 10 contd. V( Et02) PC0 ) ■ Ĥ~ .C..H■ N■“ HCrHf jP ^ V/'IfNah Ms- (11) TsC! (II!) Ac-p CXLVJ (IV) 03 R-j r 2 r 3 (a) ch2 CHO H (b) ch2 CHO Ts (c) 0h2- CH(0Ac)2 Ts (d) 0 CH(0Ac)2 Ts c l d ;: CXLiXS UNIVERSITY OF IBADAN LIBRARY 47 Sehe me 10 contd- K2c o3 , ------- The next stage in the construction of ring D was the Konversion of the 13-keto group in [XLII] into a suitable a-oriented leaving group. Reduction of [XLII] with aluminium isopropoxide gave the alcohol [XLIII] as a main product. Its 13-epimer was a minor product. The angular cyano group in [XLIII] was converted to a formyl group by reduction with diisobutyl lithium aluminium hydride followed by hydrolysis with aqueous acetic acid and sodium acetate in THF giving the hydroxyaldehyde [XLIVa]. The 13~hydroxyl group in [XLIVa] was protected with dihydropyran giving the tetrahydropyranyloxyaldehyde [XLIVb]*Formyl olefination .of [XLIVb] by treatment with sodium diethj^l ß^(cyclohexylamino) vinylphosphonate in dry THF followed by hydrolysis with aqueous oxalic acid and 10% perchloric acid gave the formyl olefin [XLVa]. The hydroxyformyl olefin was then converted to the tosylate [XLVb] by treatment with UNIVERSITY OF IBADAN LIBRARY 48 p-toluenesulphonyl Chloride, thus providing a suitable leaving group at the 13 a-position. Attempts to selectively cleave the 6-vinyl group of [XLVb] to the formyl group by ozonolysis proved unsuccessful because of the apparent higher reactivity of the formyl olefin side-chain at C-8. The formyl group was deactivated by acetylation (The acetyl groups provided steric hinderance). The diacetate [XLVc] was prepared by reacting [XLVb] with acetic anhydride in methylene Chloride in the presence of zinc Chloride. Ozonolysis of [XLVc] gave the formyl derivative [XLVd]. Compound [XLVd] was cyclised to give the correct BCD skeleton as follows: On treatment of [XLVd] with potassium7hydroxide in dry methanol and THF at -8° for 5 minutes, [XLVI] was obtained. What happened was the Tiydrolysis of the acetate groups, conversion of the C-6 formyl group to a hemi-acetal and the Michael addition of the hemi-acetal to the C-8 formyl olefin double bond. [XLVI] was treated with pyrrolidine in dry methanol and N-methylpyrrolidone at room temperature overnight and then at 70° for 1.5 hours giving [XLVII]. through the Intermediate enamine [XLVIII]. The cyclization was interpreted as an intramolecular SN reaction of the intermediate enamine [XLVIII]. [XLVII] was hydrolysed by heating with 50% acetic acid at 100° for one hour giving the hexacyclic formylhemiacetal 40 [XLIX] . Treatment of [XLIX] with Collins reagent effected the UNIVERSITY OF IBADAN LIBRARY 49 selective oxidation of the hydroxyl group giving a mixture of the formyl lactone (LJ epimeric at C-16* The lactone ring was opened by treatment with aqueous potassiurn carbonate giving the carboxylic acid [LI] with the loss of the asymmetrie centres at C-15 and C-16. Wolff-Kishner reduetion of [LI] was accompanied by endo/exo isomerisation of the double bond giving the exo-methylene- carboxylic acid [LII]• The correct B/C/D skeleton was thus achieved. The final stage of the synthesis involved the removal of the N-mesyl protecting group to give the piperidinc ring which was ultimately converted to the 6-lactone. This was effected according to the following scheine« Reductive elimination of the mesyl group SCHEME 11 co2r2 CLIIIJ a C LiVO b R| «2 (a) H H (b) CF3 CO H (c) CF3 CO CH3 W) H CH3 UNIVERSITY OF IBADAN LIBRARY 50 Scherns 11 contd. C L V b D CLVlcQ CLVIbD CLVIaU was achieved by treating [LII] with lithium in liquid ammonia in the presence of tert-butyl alcohol as a proton source, giving compound [Lllla], isolated as the hydrochloriae. (Simultaneous reduction of the carboxyl group occurred to a negligible extent). In other to selectively methylate the carboxyl group, the secondary amino group was first protected as the trifluoroacetylamide [LHIb] . Kethylation UNIVERSITY OF IBADAN LIBRARY 51 of [Llllb] with diazomethane-gave the methyl ester [LIIIc]. The trifluoroacetyl protecting group was seleetively removed by refluxing the methanol solution of [Lilie] with 3N K CO giving [LUId] . Z. o The piperidine ring was converted to the 6-lactone by the following method: [LUId] on dehydrogenation with lead tetraacetate gave a mixture of the isomeric azomethines [LIVa] and [LIVb]. The mixture on treatment with nitrous acid gave a mixture of isomeric hemiacetals [LVa]-and [LVb]. The mixture was oxidized with Collins reagent giving a mixture of the two isomeric lactones [LVIa] and [LVIb]. They were separated by preparative thin-layer chromatography and the desired lactone, [LVIa] which was dl-gibberellin  ,_ methyl ester was obtained pure. The methyl ester x̂ as demethylated by refluxing it in collidine with lithium iodide and triphenylphosphine for one hour. This gave the acid5 dl-GA [XXXI] mp. 235-237°. The acid was proved to be a racemic form of GA_1_5 by comparing its spectral data with those of an authentic sample of gibberellin  j_. Another total synthesis of gibberellin that has been pubiished was the formal total synthesis of gibberellin A [XIV] (a C^Q - gibberellin) by Mori and co-workers 41 . This synthesis can be divided into five stages. The first stage involved the synthesis of epigibberic acid [LVlIIa] from o-xylene [LVII] through twenty-one steps 46 . UNIVERSITY OF IBADAN LIBRARY 52 The second stage involved the activation of ring A of epigibberic acid so that the required functional groups could be introduced. This was accomplished by the following scheme. Nitration of racemic epigibberic acid methyl ester gave the nitro ester [LIX]. This nitro ester gave the amino ester [LX] on hydrogenation over 10% palladium - charcoal catalyst. Diazotization of the amino ester SCHEME 12 2J_si2£s_^ 2 3 C L v i m c L v i m (a) R = H (b) R = CH3 followed by hydrolysis gave the hydroxy ester (LXI). Hydrogenation of this ester over Raney nickel resulted in the reduction of the C-13 carbony1 group giving the dihydrox^ ester (LXII]. Further hydrogenation of this dihydroxy ester over rhodium-platinum oxides gave a cornplex mixture of esters in which the A ri_ng had been cornpletely hydrogenated. The mixture was oxidised with Jones1 reagent5 and the racemic ester [LXIII] was isolated as one of the products. UNIVERSITY OF IBADAN LIBRARY 53 SCHEME 13 L L X I Ü L LXMI3 UNIVERSITY OF IBADAN LIBRARY 54 The third stage involved the conVersion of the dioxo ester [LXIII] to the dienone [LXIV]. In order to introduce a double bond into the 1,2 position, an a-formyl group was used as an activating group for the introduction of a bromine atom. Treatment of [LXIII] in dry THF with sodium methoxide in dry benzene and methyl formate gave on work-up the formyl ketone [LXV] which was brominated to give the bromide [LXVI]. Decarbonylisation of the bromide [LXVI] with sodium hydroxide gave the bromoketone [LXVII]. This was dehydrobrominated to give an a-ß unsaturated ketone [LXVIII]. Boiling dilute hydrochloric acid isomerised [LXVIII] to the ß, y-unsaturated ketone, [LXIX]* Migration of the double bond was effected by treating [LXIX] with 10% pailadium-chareoal to give [LX]. In order to introduce the 1,2-double bond, the diketone, [LXX] was successively brominated and dehydrobrominated according to the sequence described above for the dioxo ester [LXIII], ultimately yielding the dienone [LXIV]. UNIVERSITY OF IBADAN LIBRARY 55 Scheine 14 conid* H CLXIX3 »> LXIV The fourth stage involved the partial synthesis of the methyl ester* [LXXI] from the dienone [LXIV] * This essentially led to the formation of the ring A lactone and the introduction of the 3-hydroxyl group. The dienone [LXIV] was ketalized with excess ethylene glycol and p-toluene sulphonic acid in boiling dichlorcethane giving the inonoketal [LXXII] . In order to introduce the lactone bridge in UNIVERSITY OF IBADAN LIBRARY 56 ring A an a-oriented carboxyl group had to be attached to C-4, of [LXXII]. This earboxylation was carried out with ethereal triphenylmethyl sodium and carbon dioxide. The resulting mixture was esterified with diazomethane after acidification5 giving a mixture of esters from which the diester monoketal [LXXIII] was isolated. The reduetion of the diester monoketal with sodium borohydride gave the hydroxy ester [LXXIV]. Hydrogenation of the hydroxy ester [LXXIV] over palladium-charcoal gave [LXXV]. [LXXV] was lactonised and deketalised by boiling with dilute sulphuric acid and the product treated with diazomethane giving the hydroxyester [LXXVI]. The 3-hydroxy1 group of [LXXVI] was epimerised by treatment with dilute aqueous sodium hydroxide giving the methyl ester [LXXI] on work-up. SCHEME 15 (LXIV) UNIVERSITY OF IBADAN LIBRARY Sehe me 15 contd- NaOH (i )H 2 SQ4 •c l x x i : (i i)CH2 N 2 r l x x v j The fifth and final stage of the synthesis involved the rearrangement of the C/D rings* Cross et_ al_* 47 had converted the methyl ester5 [LXXI] tc gibberellin methyl ester [LXXVII) by the following process. They reduced fLXXI] with sodium borohydride giving the diol [LXXVIII] . This was then rearranged. to gibberellin UNIVERSITY OF IBADAN LIBRARY 58 methyl ester [LXXVIIJ by treatment with phosphorus pentachloride. Mon et ai- 41 then deraethylated the gibberellin methyl ester by boiling with dilute aqueous sodium hydroxide to give gibberellin [XIV] and its 3-hydroxy epimer. •SC H EM E. 16 (LXXI) N q B H A (b) Partial synthesis Some partial synthesis of gibberellins have been described in literature* for example the partial synthesis of gibberellin A ^ [LXXIX] frcin GA 119Io [LXXX] . GA Io was oxidised with Jones reagent to give the 3-keto derivative [LXXXI] which was then converted to the 20,3-lactone [LXXXII] by reduction with sodium borohydride followed by heating at 135° e Reduction of the lactone with lithium borohydride gave the 3~epimer [LXXXIII] of GA^. The synthesis of UNIVERSITY OF IBADAN LIBRARY 59 the lactone [LXXXII] enabled the selective reduction of the 10-carbonyl group of GA [LXXX]. This carbony1 group is the most hindered out of the three carbony1 groups in GA^ and therefore the least susceptible to reduction. The 3-epimer [LXXXIII] of GA ̂was then converted into GA o / [LXXIX] by Oxidation to the 3^-ketone [LXXXIV] with Jones1 reagent and subsequent reduction with aluminium isopropoxide in isopropanol. UNIVERSITY OF IBADAN LIBRARY 60 A general method for the partial synthesis of’2-hydroxy gibberellins from the more abundant 3-hydroxy gibberellins has been described il3. The structure of GA,h ̂-6 [LXXXVa] was established by this method by partial synthesis from gibberellin A _Lo [LXXX]. GA io v/as treated with sodium metaperiodate and osmiurn tetroxide 5 followed by methylation with diazomethane giving the non-ketone methyl ester [LXXXVI]. On dehydration with phosphorus oxychloride in pyridine 5 [LXXXVI] gave the olefin [LXXXVII] . The olefin v/as converted to the bromohydrin [LXXXVIII] on treatment with acetyl hypobromite. The hydroxyl group in [LXXXVIII] was protected as the trimethylsiiyl ether and this derivative was then debrominated with tri-n-butyltin hydride giving the 2~alcohol as the trimethyl silyl ether [LXXXIX]. [LXXXIX] was subjected to the Wittig reaction5 thus Converting it to [XC]. This on oxidation with Jones reagent gave the ketone [XCI]. Reduction of this ketone with aluminium isopropoxide preparea from-aluminium and propan~*2~ol in the presence of mercury[II] Chloride and carbon tetrachloride gave GA TU methyl ester [LXXXVb] and 10% of the 2a~isomer. UNIVERSITY OF IBADAN LIBRARY 61 [LXXXVa] 0 C LXXXIX3 C XC3 UNIVERSITY OF IBADAN LIBRARY -t 62 Scheme 18 contd Several other approaches to the synthesis of the gibberellins have been described. One of these, described by Dasgupta et al 148 involved intramolecular carbene insertion reaction. The diazoketone [XCII] was cyclised to the cyclopropane derivative [X.CIII] . This on acid-catalysed cleavage of the cyclopropane ring gave [XCIV] which has the basic skeleton of the gibberellins. SCHEME 19 UNIVERSITY OF IBADAN LIBRARY 63 Another approach was the synthesis of the hydrofluorenone derivative [XCV] by Hon and Nakanishi . This * was synthesized in high yield in several Steps from ethyl pyruvate and malononitrile. This compound [XCV] was believed to be a possible Intermediate in the total synthesis of the G ^-gibberellins. AcO One interesting approach described by Corey et al had as its key Step the construction of the D ring of the gibberellin skeleton by intramolecular reductive addition of a 6- or e-halo ketone in the presence of an appropriate organometallie reagent. Thus the reaction of [XCVT] with 6 equivalents. of di-n~butyl~ copperlithium in ether (0.15M) at --50° for 2.5hr gave [XCVII] in 73% yield. [XCVI] had been prepared from the tricyclic ketone[XCVIII] UNIVERSITY OF IBADAN LIBRARY 64 S C H E M E 20 H H Another interesting approach was the partial synthesis of gibberellin A,1,5. norketone [XCIXb] from 7-hydroxykaurenolide(C) by Cross and Gatfield 51 . 7~hydroxykaurenolide(C) was tra* nsformed 52 353 to the aldehydo-acid. (Cla) which was then converted to the amide (Clb). The amide was photolysed in benzene in the presence of lead tetra-acetate and iodine 5 giving the lactone [XCIXa] which on Jones oxidation gave gibberellin A norketone [XCIXb]. UNIVERSITY OF IBADAN LIBRARY 65 S C H E M E 21 R- (C) The gibberellins occur at very low endogenous levels in plants. The methods used for the detection5 identification5 and Isolation of the gibberellins are therefore very sensitive. Extensive purification of extracts are normally required. The extraction proeedure normally used for the determination of the gibberellin content of plant materials is summarised in the scheine belov/: UNIVERSITY OF IBADAN LIBRARY - 66 Plant materials are homogenised with 80% aqueous methanol in a blenden. The homogenised material is filtered and the residue further extracted twice with 80% aqueous methanol. ! * The combined filterates are concentrated under reduced pressure to remove all the methanol. The pH of the aqueous layer is adjusted to 8.0 and the aqueous layer is then extracted three times with petroleum ether (60°~80°). The petroleum ether extract is discarded. The aqueous layer is then extracted thrice with ethyl acetate giving the neutral ethyl acetate (NE) fraction. 1 The pH of the aqueous layer is then adjusted to 3.0 and the aqueous layer is extracted three times with ethyl acetate.. This is the acidic ethyl acetate (AE) fraction. 1 UNIVERSITY OF IBADAN LIBRARY 67 The aqueous layer is then extracted with n-butanol giving the acidic n-butanol CAB) fraction. 4* The aqueous layer is discarded... The neutral ethyl acetate (NE) fraction contains the gibberellin glucosyl esters. Most free gibberellins are extracted in the acidic ethyl acetate (AE) fraction. The acidic butanol (AB) fraction contains the gibberellin glucosides and some polar gibberellins 5 such as gibberellins A and A . ÖZ Zo Some of the methods used for purifying and separating the gibberellins further are counter-current distribution 5U 5 55 , column chromatography 56 5 paper chromatography 57 5 gel-filtration 58 and thin-layer chromat o g r a p h y ^ . The gibberellins can be conveniently viewed under U.V. light on thin-layer chromatography plates 59 after sprayin'g with ethanol/concentrated sulphuric acid (95:5) and heating at 120° for IC minutes. Gas chromatography has been extensively used 60 in the Separation and identification of gibberellins since it was first applied by Ikekawa _et al in 196 The most powerful tool in the identification of gibberellins is the combined gas chromatography - mass spectrometry (GC-MS)^5̂ . 1t is a very sensitive and conclusive method of identification of the UNIVERSITY OF IBADAN LIBRARY 68 various known gibberellins without the need of reference compounds This%ecause reference mass spectra of the methyl esters and the trimethylsilyl ethers of the methyl esters. Cwhere appropriate) of the known gibberellins are available 39 5 64. On account of its sensitivity it requires little or no purification where the concentration of the endogenous gibberellins in the extract is relatively high (for example greater than one per cent). Extensive purification is however still necessary with very low concentrations. It has also proved very useful in the detection and identification of new gibberellins 65 5 66 . This is possible because the mass spectrum of any new gibberellin will feature the characteristic gibberellin-like fragmentation patterns which were discussed earlier. It will however be distinguishable from the known gibberellins. It may be possible to suggest a structure for It which can then be confirmed by partial synthesis and nuclear magnetic resonance. An example of the application of these methods is provided by the detection and characterization of GÄ^Q 66' This gibberellin was detected as a new gibberellin ln the endosperm of Echinocystis macrocarpa by GC-MS. The mass spectrum of Its MeTMSi derivative CM t at 580 a.m.u.) exhibited the fragmentation patterns cha* race encountered in gibberellin Chemistry. Ring A of the gibberellins is labile to strong base. If there is no double bond in the l52-position5 epimerization of 3ß-diydroxyl group occurs in dilute alkali to an equilibrium mixture with the 3a-hydroxyl 72 . In the presence of a 1,2-double bond however5 the 19 -> 10-lactone ring undergoes an allylic rearrangement to form a 19 2-lactone without epimerisation of the 3ß-hydroxyl in dilute alkali72 33-hydroxyl groups are smoothly oxidized to 3-ketones only if the exocyclic methylene group had been reduced or removed 21 . The 3-ketones so obtained are reduced by alkali-metal hydrides to the 3a-epimer 72 . However if the reduction of the 3-ketone is effected with aluminium isopropoxide the 3ß-hydroxyl is obtained as the major product14 3 3-hydroxyl groups can readily be dehydrated by reacting the gibberellin (usually as the methyl ester) with phosphorus oxychlorid.e5 to give the 2*3-dehydro compound. The deh^^dration can also be effected via the tosylate. The tosylate is prepared by reacting the gibberellin with p~toluene sulphonyl Chloride in dry pyridine at room temperature. The resulting tosylate is converted to the 2,3-dehydro derivate by refluxing in collidine*^. UNIVERSITY OF IBADAN LIBRARY 72 The 16-exocyclic double bond is readily cleaved by using sodium metaperiodate and catalytic amount of osmium tetroxide in pyridine giving the 16 nor~ketone53 In the absence of a 13-hydroxyl group, the exocyclic methylene group is hydrated by treatment with mineral acid 31. If a 13-hydröxyl group is present a Wagner-Meerwein rearrangement occurs, and the resulting nor-ketone [CIV] has an opposite configuration (a) in ring D18*47. This'nor-ketone [CIV] gives the 16-hydroxyl compound [CV] on reduction, If this alcohol is treated with phosphorus pentachloride, a reVersal of the Wagner-Meerwein rearrangement occurs 1+7, probably via the intermediate ions [CVI] and [CVII] to give the 13-deoxygibberellin system [CVII1] . The 13-hydroxyl S C H E M E 23 UNIVERSITY OF IBADAN LIBRARY 73 Scheme 23 con id . gibberellin System [CIX] can thus be converted to the 13-deoxygibberellin System [CVIII] 47 8. Biosynthesis and metabolism The biosynthesis and metabolism of the gibberellins have been studied in some detail both in the fungus Fusarium moniliforme and in higher plants 7 L‘j . 73. It has been found that the early stages in the biosynthesis of the gibberellins follow the usual pathway for diterpenoids. This is from acetyl co-enzyme A [CX] via mevalonate pyrophosphate [CXI] to geranylgeranyl pyrophosphate [CX1I]. The geranylgeranyl pyrophosphate is then cyclised to .ent-kaur-16-ene [CXIII]* The ent-kaur~lfi-ene is oxidised through ent"kaur~16-en-19"01 [CXIV] and ent~kaur~16~en~19~al [CXV] to ent-kaur-16-en-19-oic acid [CXVI]. The acid is converted via ent^7a~hydroxykaur-16-en-19-oic UNIVERSITY OF IBADAN LIBRARY 74 acid [CXVII] to ent-gibberellan-7-al-19-oic acid [CXVIII] (GA12 aldehyde). This is then converted to ^Q^ikberellins, for example GA^ [CXIX] and C^g-gibberellins, for example GAg [CXX]. 5CHEME 24 OH CH3 CO— SCoA ---- acetyl-Coensyme A c h 2 rcxx CH20PP C02H mevalonate geranytgeranyl pyrophosphate pyrophosphate CCXI 1 LCXttl ent-kaur-16-ene errt-kaur-16-en~19-0l (CXV) R = CHO CCXII13 r c x i v j (CXVl) R = CO OH UNIVERSITY OF IBADAN LIBRARY 75 S e h e me 24 contd • rC X V I1 3 19-oic acid C C X11 i □ C1 9 -gibberellin (GAg) tCX X 3 UNIVERSITY OF IBADAN LIBRARY 76 Most of the Information about the biosynthetic pathway of the gibberellins was obtained through feeding experiments with labelled precursors to Gibberella fujikuroi and cell-free Systems from some higher plants 7 Lj. *- 7 8. Thus various investigations, particularly by West and his group 78 5 based on the feeding of labelled mevalonate to Gibberella fujikuroi provided the 5.nformation on the biosynthetic pathway of fungal gibberellins (Scherne 2*4). Using cell-free Systems of Echinocystis macrocarpa, West and his group 75 also showed that mevalonate was converted to ent-7a- hydroxy kaurenoic acid [CXVII] through ent-kaurene [CXIII] 5 ent- kaurenol [CXIV] 5 ent-kaurenal [CXV] <, and ent-kaurenoic acid [CXVI] . Graebe at al 76 * working with cell-free System prepared from immature seed of Cucurbita pepo reported the conversion of mevalonic acid into gibberellin aldehyde [CXVIII]. They fed labelled mevalonate to the System and one of the labelled products was GA^ aldehyde together with ent-kaurenoic acid [CXVI] and ent-7q- hydroxykaurenoic acid [CXVII]. Further feeding experiments by Graebe et al 77 5 78 using cell-free System from immature seeds of Cucurbia maxima provided evidence on the conversion of inevalonic acid to gibberellins in higher plants, They fed [ 14 C] mevalonic acid to the cell-free System and o1b4tained [ C] GA12-aldehyde * They refed the labelled GA^-aldehyde to the System and obtained the labelled C '-gibberellins [ 14 C] GA.l o_ and [ 14C] GA, ZU ;4 o_ UNIVERSITY OF IBADAN LIBRARY 77 and a C -L -gibberellin [ 14 C] GA 77H . Thus it was shown that the pathway from mevalonic acid to the gibberellins was basically the same in the higher plants and in the fungus. The pathway for the interconversion of the various GAs may however vary in different plants or even in different organs of the ; . ; - same plant. Normally any plant will produee only a few of the large' number of gibberellins. . organ Hydroxylation and glucosidation of the GAs in any plant/occur as the plant organ matures, although not all hydroxylations result in reduction of biological activity. 2ß~hydroxylation is however a deactivation process and it occurs in all higher plants that have been studied. For example Frydman and MacMillan 79 reported that when fed to immature pea seeds* labelled GA was rnetabolised to labelled 2ß~hydroxy GAg (2-epi-GA^). GA^g was a-̂so m^tabolised to GA^g (2ß~hydroxyGAg^). It is believed that glucosides are biologically inactive storage forms5 from which the free GAs are later regenerated during germination. Three sites of bios3mthesis of gibberellins have been identified in higher plants. These are immature seeds, root and shoot apices. 9. Uses • Some of the gibberellins have much use in agriculture 80 and8 1 industry 81 . In agriculture they are used to increase the.yield and UNIVERSITY OF IBADAN LIBRARY 78 quality of some fruits and in the preservation of others. Some examples are given below. They are used for the following: (a) regulating the time of harvest and increasing the yield and quality of grape fruits. The timing of the spraying of the crop with gibberellin solution is very important in this application; (b) increasing fruit set in citrus, delaying the ripening of the fruits on the tree for extended period of many months without deterioration and preservation of the fruits duIring storage, (c) preservation of banana fruits; and (d) increasing yield in pear by spraying the trees at bloom. Gibberellin A0 (gibberellic acid) which is the most commercially available gibberellin is usually used for the various applications. GA^/GA^ mixture have also been used especially for reducing the usual June fruit drop. In industry, the gibberellins particularly GA have been used expensively in the malting stage of beer production81 The Auxins The auxins are another group of plant hormones which are very effective among other things in promoting growth extension in plants. UNIVERSITY OF IBADAN LIBRARY 79 They differ frora the gibberellins however in that whereas they are very effective Cwhen applied exogenously) in promoting growth extension of plant sections, they hav'e little effect on intact plants81a Gibberellins on the other hand promote large growth responses on intact plants but have little effect on plant sections. The discovery of auxins stemmed from the Observation of Charles Darwin in the. last Century on the bending of coleoptiles to unilateral light. This gave rise to a series of investigations which culminated in the isolation 82 5 83 of three compounds early in the Century. They were reported to have growth promoting activity. They were calied "auxin At!, !!auxin B,! and "heteroauxin”. "Auxin A" and "heteroauxin" were isolated by Kögl and co-workers 82* 5 83 from human urine. ”Auxin A and "auxin ß" have since been found to possess no growth promoting activity. Vliegenlbert and Vliegenlbert 84 using X~ray crystallography and mass spectrometry determined the structures of authentic sarnples of"auxin A" and 'auxin B. They found that n aux. m A was cholic acid and i\ aux3.. n B ” was thiosemicarbazide. The so calied heteroauxin is indole-3~acetic acid (IAA) [CXXI]^ and. it is now accepted as the most important natural auxin. The occurrence of IAA has now been reported in a wide ränge of plants and plant tissues^^. UNIVERSITY OF IBADAN LIBRARY 80 1• Physiological roles The physiological role of auxins in plants is multifarious 85-88 •The ma-jor physiological effects„include: (a) promotion of growth extension in plants, (b) promotion of the initiation of roots on cuttings, (c) inducernent of flowering'and fruit set in some plants, for examp-le pineapple and tornato, (d) Inhibition of the abscission of plant organs, such as leaves, flowers and fruits, and (e) inducement of the syntheses or activities of some enzymes, for example they enhance the formation of Cellulose synthetase in coleoptiles. 2. Chemical structure *89 The most important natural auxins are indole compounds. Apart from IAA several other physiologically active indole compounds have been isolated from plants. One of these is indole~3~acetonitrile, (1AN), [CXXII], first isolated by Jones et al 89 from cabbage. It has since been detected mainly by Chromatographie evidence in many plants, for example tomato and grape. It is now widely believed that 1AN is not active per se but owes its activity in certain biological assays to the fact that the tissues concerned can convert it to IAA enzymatlcally. UNIVERSITY OF IBADAN LIBRARY 81 Some of the other naturally occurring indole compounds which may be auxinsprecursors of auxins or metabolites of auxins are ethyl and rnethyl esters of indole-3-carboxylic5 indole-3-acetic? indole-3-propionic and indole-3-butyric acids. The rnethyl ester of 4-chloroindole-3-acetic acid [CXXIII] has also been isolated i frorn methanol extracts of immature pea seeds and shown to be active in auxm tests90 Some non-indole auxins have been isolated from plants. Examples of these are phenylacetamide and p-hydroxybenzoic acid. Many synthetic compounds were tested for physiological activities associated with the natural auxins. This led to the discovery of the synthetic auxins of which a large number are now known. Some of these are naphthalene-l-acetic acid [CXXIVa]* naphthalene-2- acetic acid [CXXIVb] 5 phenoxyacetic acid [CXXVa] and 2,4-dichloro- phenoxyacetic acid (2^-D) [CXXVb] 93 5 92 . UNIVERSITY OF IBADAN LIBRARY 82 rcxxu r1 = ch2cooh, r2 « h CCXXIVal Rj =* ch2 cooh rexxiii r1 = ch2 cn, r2 ~ h R2 - H rcxxiiu Ri = C00CH3, 2= CCXXIVb] Ri = Hr cl R2 = CH2COOH Rt CCXXVaJ R-j = OCH2COOH, R2 = H R3 ccxxvba R] = OCH2COOH, r2 = Cl r 3 Fig- 3 Some Auxins 3. Identification and characterization The auxins just like the GAs occur in very small coneentrations in plants. Their identification therefore involves the use of sensitive methods. UNIVERSITY OF IBADAN LIBRARY 83 The extraction procedure outlined for the gibberellins can also be used for the auxins. The auxin esters and any neutral auxin will.be in the neutral ethyl acetate fraction. The acidic ethyl acetate fraction will contain the free acidic auxins. Initially the various identifications of auxins in plants were based on Chromatographie evidence coupled with biological assays. Paper chromatography and thin-layer chromatography were extensively used 93 5 9i|. and tables of Rf values for several auxins in a number of solvents have been publi. shed9 593 Most of the several bioassays 95 that have been developed for detecting auxins are based on the ability of auxins to promote growth extension in plant sections. The first auxin bioassay* the * Avena coleoptile curvature test* was developed by Went in 1928 Oat grains are grown in a suitable medium in the dark at about 25°C When the coleoptiles are about 25mm long? the tip (about 2mm) of each coleoptile is cut off5 thereby removing the main source of endogenous auxin of the coleoptile. One agar block5 containing the auxin or plant extract to be tested is then placed on one side of the cut tip of each coleoptile. The seedling is then left in the dark for about two hours. The response is determined by measuring the curvature of each coleoptile. UNIVERSITY OF IBADAN LIBRARY 84 Another bioassay also involving the use of oat or wheat coleoptiles has been developed.. In this assay the lengths of cut Segments of coleptiles irnmersed in Solutions of the auxin or plant extract to be tested are measured. Many Chemical tests which give colour reactions with indole auxms have been developed 95. These colour reactions have proved very useful in identifying auxins on paper and thin~layer chromatograms. One example of these Chemical tests is the Ehrlich reaction test. In this test a rnixture of p-dimethylaminobenzaldehyde and hydrochloric acid (HCl) give a ränge of purple and blue colours with indole compounds. The. test has been used for the Identification of auxins on paper and thin-layer chromatograms. For paper chromatograms3 the identification can be carried out by dipping the paper in a 4:1 solution rnixture of acetone and 10 per cent dimethylaminobenzaldehyde in concentrated HCl. The test can detect as low as 1 microgram of indole auxin on a chromatogram. For thin- layer chromatograms5 the plate can be sprayed with an ethanolic solution of the p-dimethylaminobenzaldehyde and then kept over concentrated HCl. A modification of the Erlich test empioys p-dimethylaraino- cinnamaldehyde instead of p-dimethylaminobenzaldehyde * When this is applied as a spray in a one per cent solution in a 50:50 rnixture UNIVERSITY OF IBADAN LIBRARY 85 of ethanol and 6NHC1, it gives a blue co.lour with indoles. Most of the Chemical colour Tests however are positive with indole Compounds in general and are not specific for auxins. Some of themare not even compietely specific for indole compounds. Fluorometry, gas-liquid chromatography and mass spectrometry have also been used extensively for identifying auxins96-98 The mass spectra of the auxin indoles normally show the molecular ion. Auxins of the type indole-3-CH R [CXXVIa] give a characteristic peak at m/e 130 which is usually the base peak. The structure of the ion at m/e 130 is believed 97 to be either [CXXVIb] or the quinolinium ion [CXXVII]. There is also an ion at m/e 103 [CXXVIII] due to the loss of hydrogen cyanide from the quinolinium ion. This may then lose acetvlene moieties to give an ion at m/e 77 [CXXIX] and at m/e 51 [CXXX]^°. 5CHEME 25 CCXXVIcU CCXXVibj LCXXV! 13 UNIVERSITY OF IBADAN LIBRARY 86 Schema 25 contd- —HCN -CHSCH f ^ } C4H 3' m / c 77 m/e 51 c c x x v n n r c x x i x ] c c x x x n 4. Biosynthesis and metabolism The investigations carried out so far on the biosynthesis of auxins have centered mainly on IAA. The amino-acid tryptophan [CXXXI] is believed to be the pr-eCursor of IAA. Many workers have reported that labelled tryptophan (TPP) was converted to labelled IAA in many plants. For example Gibson et al 99 reported that 14-C-tryptophan was converted to 1 LfC-1AA in tornato and barley shoots. More conclusive evidence showing tryptophan as the main precursor of IAA in plants was provided by Erdmann and Schiewer100 using double labelling technique. They fed 3 H-serine and 1*4C-indole to sterile pea seedlings and non-sterile oat coleoptiles. They determined the 3 H/ 14C ratios of the tryptophan and the IAA that UNIVERSITY OF IBADAN LIBRARY 87 were proauced frora the labelled serine and indole. They then supplied 3 H5 IM-C-tryptophan to the seedlings and coleoptiles instead of the labelled serine and indole. They determined the 3 H/ 14-C ratio of the IAA obtained. They found that the relative labelling ratio 3 H/ i -̂’1 C-tryptophan/3 H/ m C-IAA was the same in each case. This showed that the conversion of indole to IAA passed through tryptophan without any significant bypass. It is generally believed that the main pathway from tryptophan to IAA in most plants is through indole-3-pyruvic acid [CXXXII] and indole-3-acetaldehyde [CXXXIII]. The first Intermediate5 indole-3-pyruvic acid (IPyA) has not been isolated from plants but its presence has been eemonstrated in many plants. Gibson et al 99 demonstrated the in vivo conversion of labelled tryptophan to labelled IPyA in barley and tomato shoots. They fed labelled tryptophan to the shoots and then added unlabelled carrier IPyA to the resulting raetabolites that were extracted from the shoots. The 2,4-dinitrophenylhydraz.one of the IPyA was prepared to stabilise it and it was found to be radioactive after Separation on thin-layer chromatogram. The in vivo conversion of IPyA to indole-3~acetaldehyde (lAAld) has not been demonstrated but many plants produce IAA when fed with lAAld. UNIVERSITY OF IBADAN LIBRARY SCHEME 26 „c h 2 ~ c h - cooh n h2 H (CXXXi) 0 // CH2 “ C ~ COOH c h 2 ~ c h 2 ~ n h 2 Ä H COo (CXXXii) (CXXXIV) ^ 0 CH2. ” C - h ,CH2“ C "'OH X N " H (CXXI) UNIVERSITY OF IBADAN LIBRARY Another biosynthetic pathway from tryptopban to 1AÄ which is belleved to occur in some plants is the tryptamine [CXXXIV] pathway. In this pathway the tryptophan is converted to lAAld through tryptamine (TNH^) [CXXXIV] instead of through lPyA as in the first pathway. TNH^ has been found to be endogenous in several plants for exairiple barley and tomato shoots 101 . It has however not been detected in many others like pea and bean 103 Gibson et _al 99 reported that tomato and barley shoots converted labelled tryptophan to labelled TNH^ and also converted labelled TNH^ to labelled IAA. They demonstrated the participation of lAAld in this pathway by adding carrier unlabelled lAAld to the metabolites obtained after feeding 1 14.C-tryptophan to the shoots. They px^epared the 2*M~dinitrophenylhydrazone derivative of the lAAld in each case and found that both hydrazones were radioactive. The sites of biosynthesis of auxins are believed to be apical meristems and other actively ineristematic Organs5 for example immature leaves and flowers. IAA is metabolised to several products 85 5 102 . Some of these are 3~hydroxy-3'~methyl~2~oxindole [CXXXV] 5 3~hydroxym6thyl~-2-oxindole [CXXVI] 5 and 3~methylene--2-oxindole [CXXXVII]. UNIVERSITY OF IBADAN LIBRARY Q k . . “N I ‘R H r c x x x v j R1 := = 0 , r 2 < c ‘ 3 OH c c x x x v j : R ^ ~ 0 , R2 = C H 2 OH CCXXXVÜ 1 - » 0 ) R 2 =* c h 2 Fig. 4 Some metaboSites of IA A 5.. Uses Some auxins are used commercially in agriculture 10' 3 * They are used to induce flowering in some fruits for exampie pinapple. They are also used to induce fruit-set in some others such as tomato where the application of auxin can replace the need for the pollination of the flower* They are used to control fruit drop, for exampie to prevent preharvest drop in apples. Seme of the auxins are also used as herbicides e.g. 2»lHD, [CXXVb] * UNIVERSITY OF IBADAN LIBRARY 91 Cytoklnins The cytoklnins are a group of plant hormones wh ich.promote cell division 10i|. The various investigations leading to the isolation of the cytoklnins followed from the observation that certain plant 'extracts and fluids promote the growth of Tissue cultures1 mainly by cell division. Further investigations led to the detection of a cell division factor in deoxyribonucleic acid (DNA). A new cell division factor called kinetin was isolated from DNA by Miller et al^^. The extraction procedure used involved stirring the DNA in water and autoclaving before extraction with organic solvent. Kinetin was identified as 6-furfurylaminopurine [CXXXVIII]105 and it is now generally believed to be an artefact^ probably produced by Chemical degradation of the DNA molecule during the autoclaving. Many physiologically active compounds, structurally related to kinetin have since been isolated from plants and others have been synthesized. They are all referred to as cytoklnins. 1. Physiological roles The cytoklnins perform various physiological functions in plants 106. These include; UNIVERSITY OF IBADAN LIBRARY 92 Ca) proraotion of cell division in tissue cultures and also in intact plants 5 (b) delay of senescence in plant organs probably by inhibiting protein breakdown or proraoting protein synthesis, (c) mobilization of assimilates, (d) enhancement of the resistance of plants to adverse conditions, (e) promotion of bud development in a variety of plants and tissues, and (f) regulation of the activities of some enzymes. 2. Chemical structure1 7*0 The presence of an adenine molecule with the purine ring intact 0 and with a N -substituent of moderate size appears to be one of the principal structural prerequisites for a high level of cytokinin activity. There are exceptions however, for example diphenylurea and its derivatives are active. The naturally occurring cytokinins are mainly adenine derivatives, The first naturally occurring cytokinin to be isolated was xeatin [CXXXIX]. It was isolated from methanol extract of immature maize kerneis 107 . The extract was purified with ion^-exchange column chromatography and paper chromatography. UNIVERSITY OF IBADAN LIBRARY 93 The structure of zeatin was determined by Letham et al by a combination of Chemical and spectrometric methods. When oxidised with nitric acid or manganese dioxide it yielded adenine. On chromatograms5 zeatin gave indications that it was a purine, The pKa values for zeatin were 4.4 and 9.8 (in water) indicating that positions 1,3,7 and 9 of the adenine molecule were unsubstituted. Substitution was therefore believed to be at the 6-amino-group. Comparison of the ultraviolet and mass spectra of zeatin . with those of 6- (substituted amino) purines indicated that the amino-group was monosubstituted. The n.m.r. spectrum of the picrate determined in deuteropyridine/deuterium oxide had two one-proton singlets at 67.88 and 67.96 assigned to the protons at positions 2 and 8 of adenine. The basic structure of the zeatin molecule was thus believed to be a purine molecule substituted at N 6-position. The structure of the substituent group was deduced from the mass and n.m.r. spectra of zeatin The mass spectrum showed a molecular ion at m/e 219, and a base peak at m/e 202 due to the loss of a hydroxyl group. There was a prominent peak at m/e 188 due to the loss of CH^OH from the molecular ion. Prominent ions at m/e 148 m/e 136 and m/e 135 which are characteristic 109 of adenine derivatives were also obtained. UNIVERSITY OF IBADAN LIBRARY 94 The n.Tiur. spectrum showed eight px^otons in addition to the two already detected in the purine molecule. There v/as a peak at 61.53 (3H, doublet) assigned to a methyl group adjacent to a double bond. A 4H, multiple! at around 63.84 was assigned to two methylene groups bcth adjacent to either oxygen or nitrogen. A peak at 65.42 (IH, multiplet) was assigned to an olefinic proton. T h e .structure of zeatin was thus deduced to be 6-(4-hydroxy-3- methyl-trans-2-butenylamino) purine [CXXXIX]. This structure was confirmed by synthesis^^. The condensation of the amino-alcohol [CXL] with 6-methylmercaptopurine [CXLI] in a sealed tube at 134° gave a product which on purification was identical with zeatin .. 4 J H N - C H j - r 2 16 Fig-5.Some Cytokinins UNIVERSITY OF IBADAN LIBRARY 95 c h 2o h ( CXL II) R] R- 3 = H , R2 = -C H 2 --C H x CHo CH 3 (C XL III) Ri = R3 = H f R2 = -C H = C CH3 /CHoOH (CXLIV) Rt = H . R2 = ~ C H = C < z , R 3 = CH; C H 3 CH<: (CXLV) R-j = CH = C ^ w 0 *» R3 = CH3S • CH20H CH3 OH \, ( CXLVI) R1 = * R2 • R3 = H = " A OH ÖH (CXLVI I) R-j = r 2 = - O Fig. 5 corvtd. Some Cytokinins SCHEME 27 / c h 2oh SCH3 h n - ch2c h = c XhhOH x CH3 nh2ch 2c h =c C ch (CXL) ( C X L I ) (CXXXIX) UNIVERSITY OF IBADAN LIBRARY 96 Another cytokinin, named Öihydrozeatin [CXLII] was isolated from immature seeds of Lupinus luteus by Koshimizu et a l ~ ^ \ The methanol extract of the seeds was purified by column and paper ehromatography and an active compound was obtained. This compound had the same R/f with zeatin on paper ehromatograms in three different kinds of solvent Systems. The infrared spectrum of its picrate was however different from zeatin picrate. The n.m.r. spectrum of the picrate in deuteropyridine contained the following peaks: A 3H-doublet at 61.15 assigned to a methyl \ group adjacent to a secondary carbon atom (H-C-CH^. A 3H multiplet at 61.60-2.40. A 2H-doublet, at 63.78 (H-^-CH^OH), a 2H-multiplet at 64.13 (>N-CH^-), and a 2H-singlet at 68.98 (protons at positions 3 and 5 of picric acid). Also observed were two one-proton singlets at 68.47 and 8.78 assigned to the protons at position 2 and 8 of adenine. The u.v. absorption spectra NaOH 275 mn an(j 282 nm (shoulder), A max 269 nm, 9 and vmax 273 nm indicated* ' a N 6(alkyl-substituted) adenine. The structure was deduced to be 5-(4-hydroxy-3-inethyl butylamino) purine [CXLII]. This structure was confirmed by synthesis. Catalytic hydrogenation of zeatin gave dihydrozeatin as one of the products and this was identical with the natural material. UNIVERSITY OF IBADAN LIBRARY 97 A cytokinin was isolated from liquid cultures of a bacterium, Corynebacterium fascians 3 Ca plant pathogen which induces multiple bud formation in certain trees). The structure was elucidated by Helgeson and Leonard 11' 3 to be 6-(3-methyl-2-butenylamino) purine (isopentenyi adenine, IPA) [CXLIII]. The u.v. spectrum indicated a N 6(alkyl-substituted) adenine and the n.m.r. spectrum showed the presence of two methyl group instead of the one methyl group in zeatin. The mass spectrum showed the general pattern for purine derivatives, similar to that of zeatin. This cytokinin has since been identified in some higher plants for example in immature peas. The riboside of IPi^was isolated 114 as one of the constituent nucleosides of soluble RNA of yeast. Soluble ribonucleic acid extracted from yeast was hydrolysed enzymatically to its constituent nucleosides. The mixture was separated by column chromatography. IPA-riboside was obtained and crystallized. The n.m.r. spectrum (in CD o COCD o -D Z0) exhibited the basic pattern associated with adenosine and also contain^d the following peaks. A split peak at 61.75 (6H-two vinyl methyl groups)5 a one- proton multiplet at 65.4 (vinyl proton) and a three-proton multiplet at 64.2 assigned to the methylene group attached to N 0 and to the C-4 proton of ribose. UNIVERSITY OF IBADAN LIBRARY 98 The mass spectrum had a raolecular ion at m/e 335 and streng peaks at 292 and 203 Cfree base).. There were also prominent peaks at m/e 88, 160, 1*4-0, 136 and 135 (adenine). ' Some other cytokinins which have been identified 115 as constituent nucleosides of transfer RNA after hydrolysis are ribosides of CH O S-äeatin [CX, LIV], and CH S-IPA [CXLV]. Transfer RNAö extracted from wheat was hydrolysed enzymatically and the resulting ribonucleosides were extracted with ethylacetate and separated on Sephadex 1H-20 columns and on paper. Their u.v. spectra were indicative of N 6— (alkyl substituted) adenine. The mass spectra of CH -S-xeatin and CH^»-IPA ribosides had a fragmentation pattem" similar to those of :zeatin and IPA ribosides respectively except that the molecular weights were *4-6 atomic mass units higher in each case. Their structures were deduced to be 6~(14~hydroxy~3~methyl“2-butenylamino)'-2-methylthio~9-ß-D~ribofuranosyl purine, (CH S-Zeatin riboside) [CXLIV] and 6-(3“-methyl~2-butenylamino)- o 2-methylthio-9--ß--D-ribofuranosylpurine, (CH oS-IPA riboside) [CXLV] . Another * cytokinin, 6-(0-hydroxybenzylamino) purine was isolated116 from leaves of Populus robusta as its ribofuränoside [CXLVI]. The u.v. spectrum of. t_h e compound, /Av (EtOH neutral) X (EtOH pH0)max 266nm,|\ max ^ 2 260 nm,/\ max 265 nm, indicated that it was a N ~■ substituted adenosine. UNIVERSITY OF IBADAN LIBRARY 99 The mass spectrum of the compound which had principal peaks at m/e 373, 284, 270, 241, 224, 178, 164, 148, 135 (base peak), 121, 120, 119, 108, 106, 78, 66 as well as a molecular ion at m/e 661 suggested that the compound was a N 0 (hydroxybenzyl) adenosine. This was confirmed by high resolution mass spectrometry. The position of the hydroxyl group on the benzene ring was determined by comparing the mass spectra of the three synthetic isomers with that of the natural product. It was then confirmed by do-chromatography on a GLC System which clearly separated the three- isomers, The syntheses of the three isomers were effected by condensing 6-chloropurine-9-ß-D-r4i bo"side with the appropriate hydroxybenzylamine in refluxing methanol. Many biologically active cytokinins have been synthesized. The most active of these are adenine derivatives for example 6-benzylaminopurine [CXLVII]. The two methods that are usually used for the syntheses of cytokinins are: (a) displacement of alkylthio group 117 or chlorine 19118 at the 6-position of the purine ring and (b) direct alkylation 119 The displacement reaction has been used to synthesize N^-substituted adenines and adenosines. For example, kinetin [CXXXVTII] was synthesized by reacting 6-methylmercaptopurine [CXLI] with UNIVERSITY OF IBADAN LIBRARY 100 furfurylamine at 115-120° for 9 hrs6-(o-hydroxybenzylamino) purine riboside [CXLVI] was synthesized by reacting 6-chloropurine riboside with o~hydroxybenzylamine in refluxing n-butanol. Thus the reaction basically involves the reaction of a 6-thioalkyl purine or a 6-chloropurine with an appropriate amine. The second method of direct alkylation is usually used for synthesizing cytokinin ribosides and ribotides. The reaction is based on the fact that a Substituent at the 9-position of the purine ring directs alkylation to the 1-position. Therefore 1-substituted adenines are readily prepared by reacting adenine (substituted in the 9-position) with an appropriate alkyl halide. The 1-substituted adenines are then readily rearranged to N^-substituted adenines by heating in alkali119. 3. Isolation and Identification The extraction procedure usually used for the cytokinins involves the grinding of the plant material in aqueous methanol. The methanol is then removed under reduced pressure.. The pH of the resulting aqueous phase is adjusted to about 3,0 and the aqueous phase is extracted with ethyl acetate to remove acidic impu^ities. The pH of the aqueous phase is adjusted to 7.0 and it is extracted with water-saturated n-butanol. The cytokinins will be in this n-butanol fraction. The n-butarol is then removed in vacuo. UNIVERSITY OF IBADAN LIBRARY 101 Further purification of the extract is usually carried out by ion-exchange column chromatography and gel-filtration chromatography. Paper chromatography is also usually used for further Separation. Gas chromatography is used for the Separation of trimethylsilyl derivatives of the cytokinins 120 . It is also used for Identification when authentic samples are available. The shape and general appearance of the ultraviolet spectra of the cytokinins (adenine derivatives) under different pH conditions are used to determine the position of the substituent on adenine 121. For example 9-substituted adenines usually have u.v. spectra which are insensitive to pH. The adenines usually absorb u.v. light str• o4n gily in the region 250nm to 280nm. The pka values of the adenine derivatives are also used to determine the position of Substitution on the adenine molecule 1 2 2 . The presence of pKa in the region below 6 indicates that positions 1, 3 and 7 are not substituted. If the pKa lies in the region 8 - 11, it shows that the N-H group is still present in the five- membered ring of adenine. If however there is no pKa value in the region 8 - 11 it means that there is no ionizable hydrogen in the five-membered ring indicating that position 9 of adenine is substituted. UNIVERSITY OF IBADAN LIBRARY 102 Mass spectrometry has been employed extensively for the elucidation of the structures of the various cytokinins 109 The 6-alkylaminopurinyl structures of raost of the cytokinins show some characteristic ions in their mass spectra. Some of these are ions at m/e 148 [CXLVII], 136 [CXLVIII], 135 (adenine ion) [CXLIX]j 119 (purine ion) [CL]5 and m/e 108 (adenine - HCN) [CLI] which may be derived as shown in Scheme 28. Combined gas-chromatography-mass spectrometry has also been used for the Identification of trimethylsilyl derivatives of cytokinins SCHEME 28 m/e 148X (CXLVII) m/e 119 (CL) UNIVERSITY OF IBADAN LIBRARY 103 Scherne 28 contd- m/e 135 (CXLIX) Some bioassays have also been developed for the detection of cytokinins 124 1259 . Some of these are based on the ability of the cytokinins to stimulate cell division in tissue cultures (e.g. tobacco callus assay). Some others are based on the ability of cytokinins to delay senescence and loss of chloroplasts in detached leaves, and to induce the formation of buds in some plants. The bioassays based on cell-division are usually the most reliable because of their sensitivity and specificity. UNIVERSITY OF IBADAN LIBRARY 104 4. Biosynthesis and metabolism The biosynthetic pathway leading to the cytokinins is still not well understood but the isoprenoid side chain is believed to be frorn mevalonate* .jf V Seeds, ijnmature fruits, roots, and root exudate have been identified as rieh sources of cytokinins in higher plants. The cytokinins are rapidly metabolised in plant tissues. Some of the metabolites of zeatin that have been isolated are 7-glucosylzeatin, 9-glucosylzeatin* zeatin riboside and zeatin riboside 5f-monophosphate125 The synthetic cytokinins are similarly metabolised to the glycosides and the glycotides and are in part degraded to adenine. Abscisic acid Abscisic acid (ABA) [CLII 3, a monocarboxylic, monocyclic • sesquiterpenoid was first isolated as an abscission accelerating substance in cotton plants 127 . It was also isolated from sycamore leaves and yellow lupin where it caused the abscission of immature fruitlets. Since then ABA has been isolated from several plants and detected in very many others 128 . UNIVERSITY OF IBADAN LIBRARY 105 1. Physiological roles There have been raany reports 128 5 129 of the physiological roles of ABA in plants. Some of these functions are: (a) abscission acceleration, (b) induction of dormancy, I r (c) regulation of stomatal opening, (d) involvement in some aspects of ripening process, and (e) inhibition of the synthesis of some enzymes, for example a-amylase in cereal aleurone. 2. . Chemical structure *lo Ohkuma et al 130 proposed a structure for ABA based on spectral data. The molecular formula was deduced to be Cl,oC H_2 0JD ,4 from Chemical analysis and the mass spectrum of ABA which showed a molecular ion at 264 a.m.u. The infrared spectrum of ABA showed hydroxyl group (3405 cm carboxylic acid group (2300 cm and a coniugated keto.group (1650 cm ^). The presence of three bands at 1674 cm \ 1623 cm \ and 1600 cm ^ was believed to indicate the presence in ABA of a group similar to sorbic acid (2,4-Hexanedienoic acid). A strong peak at 978 cm ^ indicated the presence of a trans substituted double bond. UNIVERSITY OF IBADAN LIBRARY 106 The n.m.r. spectrum showed the presence of four methyl groups. Two of these were on saturated carbon (61.10 and 1.17) and two were vinyl methyl groups (61.99 and 2.10). Two strong peaks at 62.41 and 62.47 were believad to represent a methylene group adjacent to a carbonyl group. There were four vinyl protons at 65.79, 5.985 6.17 and 7.81. The mass spectrum had prominent M-56 (due to the loss of isobutylene from the ring) and M^lll due to the loss of the side chain. On the basis of the spectal data they deduced the structure of ABA to be 3-methyl-5-(lf-hydroxy-41-oxo-2f,6l,6T-trimethyl-2f- cyclohexen-l-yl)-eis, träns-2»4-pentadienoic acid [CLIL ]. This structure was confirmed by synthesis131 The synthesis consisted of the photo-sensitized epidioxidation of S-methyl-S-^’ ,6T ,6 '-trimethylcyclohexa-l’ 31 -dienyl)-cis ?tran s-2, 4~pentadienoic acid [CLIII] to give an epidioxide [CLIV]. This was rearranged by heating at 100°C in aqueous sodium hydroxide giving racemic ABA [CLV] after acidification. The stereochemistry of the tertiary hydroxyl group of natural ■© ABA is oi and the absolute configuration of natural ABA is (S).128 It is strongly dextro-rotatory ([ctJD + 430°). UNIVERSITY OF IBADAN LIBRARY 107 j l SCHEME 29 132 An improved synthesis of ABA was reported by Roberts et al 132 starting with a-ionone [CLVI]. a-Ionone in t-butyl alcohol was oxidised with' t-butyl chromate to give l-hydroxy-U-keto-a-ionone [CLVII] and *+-keto-a-ionone [CLVIII] in a 4:1 ratio approximately. The side chain was then extended by Wittig reaction using triphenyl carbethoxymethylene phosphorane. Thus refluxing the l~hydroxy-4- keto-a-ionone in toluene with the phosphorane [(C.H ) P=CHC00C K ] gave a 1:1 mixture of racemic ABA ethyl ester [CLIX] and the UNIVERSITY OF IBADAN LIBRARY 108 2-trans isomer [CLX]* The mixture on hydrolysis with base gave racemic ABA [CLVJ and the 2-trans isomer [CLXI] in a 1:1 ratio and these were separated on t sLc. SCHEME 30 i (CLVI) (CLV11) (CLVI 11) CLVI 1 (CLIX) R = - C H 2CH3 (CLX) R = -CH 2 CH3 (CLV) R - H (CL X!) R = H 3. Isolation and Identification The extraction procedure used for the gibherellins and auxins can also be used for abscisic acid. Free ABA will be extracted in the acidic ethyl acetate (AE) fraction. Any ABA ester present will be in the neutral ethyl acetate (NE) fraction UNIVERSITY OF IBADAN LIBRARY 109 and ABA glucoside in the acidic n-butanol (AB) fraction. Further purification of the extract ean be by the various Chromatographie methods. Several rnethods have been developed for identifying ABA in purified extracts. One of these is the spectropolarimetric method 133 9 13*4 which utilises the unique optical rotatory dispersion (ORD) of ABA. This has an intense positive'cotton effect' with extrema ,at 289nm and 2*46nm. The ORD spectrum is affected by pH and so it is usually measured in 0.005N or the spectrum of the methyl ester which is unaffected by pH is measured. The 1cotton effect1 of the ORD is so large that if the extract is well purified, interference from other substances may be negligible. This ORD method can therefore be used to determine ABA present in an extract both qualitatively and quantitatively, provided there is no other optically active substance in the extract. The ultraviolet spectrum of ABA can also be used to identify it in highly purified extracts. The u.v. spectrum 13*4 is also affected by pH. In acidic conditions the absorption maximum occurs at 262nm (c = 21,*400) with a shoulder at 2*40nm. In basic conditions the maximum occurs at 2*45nm and it is about 20% more intense. Milborrow 13. *4 developed a method for the determination of concentrations of ABA in plant tissues. In this method called fracemate dilutionT (RD) method, a known amount of synthetic, racemic UNIVERSITY OF IBADAN LIBRARY 110 ABA is added to the crude plant extract.. The extract is purified and the total amount of ABA remaining after the purification is determined fr-om the u.v. absorption. The C$)-Ct)-ABA remaining after the purification is determined by the ORD method. From these determinations5 the amount of natural* CCS)~C+)) ABA originally present in the plant tissue can be calculated. The main dis- advantage of the use of u.v. and ORD for identifying and estimating ABA is the very high degree of purification that is usually necessary. Thin-layef chromatography has been used for identifying ABA. It can be separated from its 2-trans isomer by this method. GLC is also very useful for identifying ABA in plant extracts* and rigorous purification may not be necessary 135 . Methyl and trimethylsilyl esters of ABA can readily be separated on GLC and identified by comparison with authentic samples. The use of GC-MS makes the Identification of ABA in extracts possible without the availability of authentic samples. The mass spectrum of methyl ester of ABA shows a molecular ion at m/e 278 and prominent peaks at m/e 260 CM* -HzO )* 246 (M+-CH oOH), 222 (M+-isobutylene), 190, (base peak), (M+-CHgOH and isobutylene), 162, 134 and 125. UNIVERSITY OF IBADAN LIBRARY 111 Two Chemical colour tests have been developed for ABA. The first test.> reported by Antoszewski and Rudnicki 136 is based on the fact that ABA gives a yellow-green fluorescent colour after heating with sulphuric acid on silica gel. This is used to detect ABA on silica gel thin-layer chromatograms. The other colour test was first reported by Mallaby and Ryback 137 and is based on the fact that ABA gives a lactone [CLXII] as the major product when heated with acid. This lactone gives an intense characteristic transistory violet-red colour with alkali. This is used as a sensitive qualitative test for ABA. Several bioassay Systems 129 have been used to detect ABA. They include; (a) acceCl er>'a tiön of abscission in exc.ised abscission zones (explants), (b) inhibition of coleoptile curvature or straight growth, (c) inhibition of seed germination, (d) the inhibition of growth of rice seedlings and Ce) stomatal clcsure assay. . Chemistry Much work has not been done on the chemistry of ABA. Mallaby and Ryback 137 reported that when ABA was heated with a mixture of formic acid and concentrated hydrochloric acidd a lactone [CLXII] was obtained as the major product. The lactone gave an intense violet-red colour with alkali. The lactone was believed to be hydrolysed to the keto acid salt [CLXIII] by alkali. UNIVERSITY OF IBADAN LIBRARY .12 SC H EM E 31 ABA (CLIO ABA methyl ester is reduced in aqueous methanolic sodium borohydride to give approximately equal mixtures of the l1 ,4t~cis,~diol [CLXIV] and the 1 ' ,41 -trans-diol [CLXV] TO Q. These * diols are readily reoxidized to ABA by manganese dioxide in dry MeOH. SCHEME 32 Me ABA NaBH4 Me of ( C U iT HO'' (CLXIV) (CLXV) UNIVERSITY OF IBADAN LIBRARY 113 ABA exchanges six of its carbon-borne hydrogen atoms with water at pH 11 and above, These are the C-2*-methyl hydrogens, the C-3' hydrogen and the C-5* methylene hydrogens. This has been used as a basis for preparing deuterated ABA139 5. Biosynthesis The terpenoid nature of ABA was demonstrated by the fact that many plants converted labelled mevalonate into ABA 138 5 140 . This was first demonstrated by Noddle and Robinson 140 who reported that [ mevalonic acid was incorporated into ABA by avocado and tomato fruits. They supplied labelled mevalonic acid to ripening fruits and extracted the ABA two days later. The ABA contained the label from the mevalonic acid. The pathway from mevalonate to ABA is still not very certain. Two pathways have been postulated. In the first pathway known as the "carjotenoid pathway", it is postulated that ABA is derived from a carotenoid, especially violaxanthin [CLXVI] 1^~ 1 . In the second pathway called the "direct synthesis pathway", it is postulated that a C precursor is cyclized and then converted into ABA142 JLO In order to test the ’carotenoid pathway*, Taylor and Smith exposed some carotenoids on damp filter paper to bright light. They found that violaxanthin gave rise to a streng inhibitor which was later characterized 143 to be approximately equal amounts or UNIVERSITY OF IBADAN LIBRARY 114 2-cis-xanthoxin [CLXVII] and 2-trans-xanthoxin [CLXVIII]. The 2-cis isomer was found to be the active component. Xanthoxin has since been detected in extracts of many plants, for example dwarf bean and wheat seedlings1^. # , SCHEME 33 (CLXVI) 1 f ( + ) - 2 -c is Xanthoxin (+•)-2- trans-Xanthoxin (CLXVI!) (CLXVIII) UNIVERSITY OF IBADAN LIBRARY 115 Taylor and Bürden have shown that labelled 2-ci s-xanthoxin is converted to ABA by shoots of tomato and bean m s . They fed 2-[ m C] - cis-xanthoxin to cut shoots of tomato and dwarf bean and found that it was converted within 8 hours to ABA in yields of 10.8% and 7.0% respectively. This result and the fact that xanthoxin had been 1 shown to be enI dogenous in many plants 144 were taken to suggest that carotenoids and xanthoxin may be precursors in the biosynthesis of ABA. The results did not however establish the importance of the fcarotenoid pathway* in normal biosynthesis of ABA. On the other hand, strong evidence has been produced in favour of the 1 direct synthesis* pathway 128 . One of the experiments suggesting that ABA is derived from a C precursor and not from a carotenoid, was carried out by Robinson'1'146. He supplied [^C]- labelled phytoene (an uncyclized carotenoid) to avocado fruit together with [3H] mevalonate. He then extracted and isolated the ABA and carotenoids in the fruit after some hours. The [ 3 H] label was incorporated into the carotenoids and the ABA.. The ["^C] from the phytoene was incorporated into carotenoids but it was not incorporated into ABA. This experiment thus shc./ed that ABA was formed from mevalonate in avocado fruit via a route not involving a carotenoid. This strong evidence in support of the direct synthesis pathway did not however'exclude the possibility that under certain circumstances the carotenoid pathway might also operate. UNIVERSITY OF IBADAN LIBRARY 116 6. Metaboiism Several metabolites of ABA have now been isolated, It was reported 147 that whe'n labelled ABA was supplied to tomato shoots, it was rapidly metabolised into two compounds. The first compound was identified as the glucose ester of ABA. The second compound called ’metabolite Cr [CLXIX] was more polar than ABA. The compound showed a strong positive ORD curve similar to that of ABA but displaced slightly towards longer wavelengths ((t)-extremum = 298nm, (-)-extremum = 254nm). The ultraviolet and infrared spectra of !metabolite Cf were almost identical with those of ABA. The infrared spectrum of 1 fmetabolite C l contained absorption bands at 1585, 1615 and 1666cm 1 indicating that the double bonds in the ring and side chain of ABA were still present in !metabolite CT. The absorption band at 3400cm ^ was more intense in ’metabolite CT than in ABA. 1Metabolite CT was acetylated with acetic anhydride in pyridine, but the tertiary hydroxyl group of ABA could not have been acetylated in that way. 1Metabolite C! was therefore deduced to be a hydroxylated derivative of ABA. Metabolite C1 i%ras reported 147 to be unstable* rearranging easily to give ancther compounds phaseic acid (PA) [CLXX]. Phaseic acid had a plain negative ORD curve between 250 and 350nm. UNIVERSITY OF IBADAN LIBRARY - 117 In the infrared spectrum of phaseic acid* the OH absorption was less intensa than that of Metabolite C M The band ai 1666cm ^ in the infrared spectrum of Metabolite C l was absent in the spectrum of phaseic acid but there was a new band at 1715cm \ These changes were believed to be due to the Saturation of the double bond a* ß to the M-T-keto group of Metabolite C M The n.m.r. spectrum of methyl phaseate showed that the 3-methylpentadienoic acid side-chain of ABA was still present in phaseic acid. The n.m.r. spectrum also contained a new methylene signal and showed the loss of the peak assigned to one of the 6!-gern- dimethyl groups present in ABA. This thus indicated that Metabolite C ! was formed by the hydroxylation of one of the C-6! dimethyl groups of ABA. The structure of Metabolite C f was thus deduced to be 61-hydroxymethyl abscisic acid [CLXIX]. Subsequent attempts to isolate Metabolite CT have been unsuccessful. Phaseic acid was believed to be formed from Metabolite C f by the nucleophilic attack of the hydroxymethyl on C-2M thus giving the structure of phaseic acid as [CLXX]. Phaseic acid was first isolated 1U8 from bean seeds (Phaseolus multiflorus). UNIVERSITY OF IBADAN LIBRARY 118 (CL XX) phaseic acid Tinelli et al 149 reported that two compounds designated M-l and M-2 were isolated as metabolites of ABA in excised axes °f Phaseolus vulgaris. They fed 2-^L|’C~ABA to the axes and isolated the two radiactive metabolites after 48 hours. M-2 was raethylated with diazomethane and the methyl ester on oxidation with CrO o in acetic acid gave a compound which was identical with M-l methyl ester. The mass spectrum of the compound (M 294) was identical with that of methyl phaseate. M-2 was acetylated with acetic anhydride in pyridine at room temperature but M-l did not react with acetic anhydride in this way. M-2 did not react with 2,4-dinitrophenyihydrasirie but M-l did. M-l methyl ester gave two main products when reducea with sodium borohydride and one of these co-chroraatographed with M-2 methyl ester. UNIVERSITY OF IBADAN LIBRARY - 113 - From the above data M-l was deduced to be phaseic acid and M-2;4r-dihydrophaseic acid CDPA) [CLXXI] . The mass spectrurn of the methyl ester of 4!-dihydrophaseic acid showed a molecular ion at m/e 296 and a base peak at m/e 43. Naturally occurring 41-dihydrophaseic acid (DPA) was first isolated from methanol extract of mature bean seeds by Walton et al^^. The epimer of DPA, referred to as epi-41-dihydrophaseic acid, (epi-DPA), [CLXXII] was identified as a minor metabolite of ABA in f^aseolus vulgaris 151 . Racemic ABA (labelled) was fed to bean shoots. After extraction and purification of the acidic fraction by thin-layer chromatography on silica gel, (+)-ABA was found to have been metabolised to phaseic acid, DPA and epi-DPA. The PA was methylated and identified by GC-MS. The methyl esters of DPA and epi-DPA were separated on TLC by multiple development in ethyl acetate-hexane (2:1). DPA was more polar than epi-DPA. DPA methyl ester (Me-DPA) and epi-DPA methyl ester (epi-Me-DPA) gave two separate peaks on GLC with retention times identical to those of the two products obtained by reduction of Me-PA with sodium borohydride 152 The mass spectra of the four compounds were also very similar, showing molecular ions at m/e 296, UNIVERSITY OF IBADAN LIBRARY 120 Epi-DPA was detected as one of the metabolites of endogenous ABA in dry bean seeds 2’ 51. 1t was conclusively identified by GC-MS of its methyl ester. A probable inetabolic pathway for ABA rnay be as follows: . ABA 61 -hydrox^BA PA DPA and epi~DPA. (CL XXI) (CL XXI I) The riebest sourees of ABA and its metabolites in plants are fruit tissues. UNIVERSITY OF IBADAN LIBRARY 121 Ethylene The recognition of ethylene as a plant hormone can be traeed back to the report in 1901 by Neljubow1 53 that the abnormal growth ■ |V;- • (stunting, stem thickening and prostrate nature) of some seedlings /. in laboratory air was caused by traces of coal gas in the air. He identified ethylene and acetylene present in the coal gas as the active substances. Gane 15T in 193M- demonstrated that ethylene was produced by plants. He passed the gaseous mixture from apple into.tubes containing bromine at -65°C. The oil obtained was purified by fractional distillation, The presence of ethylene in the oil was demonstrated by the formation of N,NT-diphenyl ethylene diamine when the oil was heated with aniline. Since then it has been reported that ethylene is produced by many plants particularly during the ripening of fruits 155 1. Physiological roles Many physiological functions have been attributed to ethylene in plants“*^“ 50 5 ~j °c y . Some of these are the following: Ca) breaking of dormancy5 Cb) control of flower induction* Cc) induction of adventitious roots5 Cd) promotion of the ripening process. UNIVERSITY OF IBADAN LIBRARY 122 (e) closure of plumular hook> Cf) induction of epinasty., Cg) Inhibition of leaf expansion, (h) inhibition of extension growth, and (i) acceleration of abscission. Ethylene also has some effects on the physiology of IAA156-158 Generally ethylene reduces the level of IAA in plants. One of the means by which ethylene affects IAA level in plants is believed to be through the inhibition of the biosynthesis of IAA. This was particularly indicated by the work of Valdovinos eit al 158 . They fed [^C] -labelled tryptophan to cell-free preparations from stems of Coleus and Pisum. They found that enzymes from tissues treated with ethylene were much less active in producing labelled IAA and labelled CO^ from the labelled tryptophan. Ethylene may also catalyse the destruction of IAA and inhibit IAA transport m some pl, ant^s 156,157 2. Biosynthesis It is now believed that the amino-acid, methionine [CLXXIII] is the most likely precursor of ethylene in plant• s 159 . The possibility of methionine acting.as a precursor of ethylene was first reported by Lieberman and Mapson 160 . They reported that ethylene was formed from copper catalysed breakdown of methionine. They also reported UNIVERSITY OF IBADAN LIBRARY 123 that apple-tissue slices fed with methionine produced more ethylene than Controls C5 H„ - S- 4C H - C 3 H 2 - C 2 H - C1OOH o Z Methionine [CLXXIII] More significant evidence on the possibility of methionine acting as a precursor of ethylene was obtained through tracer studies. Lieberman et al^^added [*̂ C] -methionine labelled in carbon atoms 1, 2, 3, H or 5 to apple slices. They found that only methionine labelled in position 3 and *4 produced significant amounts of labelled ethylene. The work thus demonstrated that ethylene is derived from C-3 and C-4 of methionine. The work in this thesis covers essentially the examination of some of the hormones in the extensively purified extracts obtained from two groups of cowpea fruits (lNew Era1 cultivar). These were 6-day old fruits from the lowest raceme and 2-day old fruits from the upper racemes. UNIVERSITY OF IBADAN LIBRARY 124 RESULTS AND DISCUSSION As stated in the introduction5 the investigations of Ojehornon 192 9 4 seeraed to implicate the presence of fruits which were older than five days, at the lowest raceme of the cowpea inflorescence, with the abscission of the fruitlets, and flowers of the upper racemes. The work of various other workers 8 had also revealed that the plant hormones are the principal endogenous factors that control abscission in plants. It was therefore decided that a comparative analysis of some of the hormones in the maturing fruits of the lowest raceme and the abscising fruitlets of the upper: racemes should be carried out, It was thought that a comparison of the hormonal contents of these two groups of fruits might lead to the understanding of some aspects of the endogenous factors Controlling abscission in cowpea. This might then be of importance in the control of abscission in cowpea which is an important local source of protein. The work in this thesis covers essentially the analysis of some of the hormones in immature cowpea fruits obtained from different racemes of the cowpea inflorescence. The VNew EraT cultivar of cowpea was used. This cultivar is known to exhibit a high degree of abscission of flower buds, flowers and immature fruits^" \ UNIVERSITY OF IBADAN LIBRARY 125 Extracts were obtained from two groups of fruits.. These weresix-day old fruits from the lowest (first) raceme of the inflorescence and two-day old fruits from the upper racemes. The analysis of the extract obtained fron'fruits that were older than six days (frorn the lowest raceme) was also carried out for comparison. The extraction procedure u s e d ^ ^ for» each group of fruits was carried out as described in the experimental section (page 2.\5 ). The analysis was concentrated on the acidic ethyl acetate (AE) fraction. This was because most of the free acidic plant hormones should be in the AE fraction. Analysis of the acidic ethyl acetate fraction of the extract obtained from six-day old fruits A portion of the acidic ethyl acetate (AE) fraction obtained from six-day old fruits was purified 162 with PVP. In order to analyse the PVP purified acidic extract on gas­ liquid Chromatograph (GLC), suitable volatile derivatives had to be prepared. Aliquot of the purified extract was therefore methylatea with ethereal diazomethane. The methyl ester thus obtained was silylated in pyridine with a mixture of hexamethyldisilazane and trimethylsilyl Chloride. A portion of the methyl ester trimethylsilyl ether (KeTMSi) derivative of the extract was then analysed on gas-liquid UNIVERSITY OF IBADAN LIBRARY 126 Chromatograph directly linked to a mass spectrometer (GC-MS). The gas-liquid chromatogram of the MeTMSi deriyative of . the extract (Fig.6) had many peaks. No Identification could easily be made directly frorn the mass spectra of most of the peaks. Most of the peaks appeared to be mixtures. One of the peaks (Fig. 65 p38) however appeared to be a mixture of the methyl ester trimethylsilyl ether of gibberellin (MeA^TMSi), [CLXXIV] and possibly MeA^TMSi [CLXXV] together with some impurities. The mass spectrum obtained from a scan of the first part of the peak had a molecular ion at m/e 432. There were prominent M+-15(M+~CH o ) and M+-59(M+-CH oC00) ions. There were also prominent ions at m/e 303, 235, 208 and 207. The ions at m/e 303 and 235 were characteristic 39 prominent ions in the mass spectrum of MeA bTMSi. The ions at m/e 208 and 207 were indicative 39 of hydroxylation at C-13. Although there were some other prominent ions (believed to be due to impurities) in the mass spectrum5 the general fragmentation pattern was consistent with that of‘Standard MeA.bTMSi 39 The mass spectrum obtained from a scan of the middle of the peak (Fig. 6, P38), indicated that apart from MeA^TMSi, MeA^TMSi might also be present. Apart from the ions mentioned above, there was an ion at m/e 492 which could be the molecular ion for UNIVERSITY OF IBADAN LIBRARY 126a 18H E E c 1AH o ru o -C ioH 0 k. ■*-* 6H 1 a i5 24- TTTTTT7 TTrnTTIT ip/ n.|i7|/trn fix>! Control Column fractions Fig 7 Halo bioassay on column fractions of AE extract from 6-day old fruits UNIVERSITY OF IBADAN LIBRARY 127 MeA TMSi. There were. also M+-32 and M*-119 Ions., The M -1197120 Ion had been reported 39 to be characteristic of ^20~^S an<̂ WaS kelieved 39 to ke due to the l°ss two moles of acetic acid. This indicated the presence of at least t two methoxycarbonyl groups in the molecule.. The Ions at m/e 207/208 might also be due to MeA^TMSi since GA.^ has a hydroxyl group attached to 013. The peak (Fig. 6, P38) was therefore believed to consist of MeA^TMSi and possibly MeA^TMSi together winh some impurities. MeAg TMSi MeA17TMSi (CLXXiV) ( CLXXV) UNIVERSITY OF IBADAN LIBRARY 128 The mass spectra of some of the other GLC peaks, (Fig. 6, P35, P39 and Pi|-J ) indicated that sorae other GAs like-, GA^q , GA^, and GA O might be present. Each of these mass spectra however had very many prominent ions which could not be accounted for. It was therefore difficult to make reasonable deductions from the mass spectra. 1. Column chromatography on AE fraction obtained from six-day old fruits. On account of the large amount of impurities in the extract> it was decided to purify the extract further by charcoal-celite column chromatography. Twenty 150 cm 3 fractions were collected by gradient elution with increasing concentrations of acetone in water. (a) TLC analysis of column fractions Explox^atory analytical thin-layer chromatography (TLC) was • carried out on all the twenty column fractions. The results of the analytical TLC indicated^•] 36 that ABA and its metabolites might be present in fractions *4 and 5. The results also indicated 59 the probable presence of GAs in fractions H to 9. (b) Barley half-seed Chalo) bioassay of column fractions The column fractions were tested for gibherellin activity by using the barley half-seed (halo) bioassay^"^^ . UNIVERSITY OF IBADAN LIBRARY 129 The results obtained from the halo bioassay are shown in Table 2 and the histogram is shown in figure 7. Gibberellin activity was observed in fractions 5 to 9. TABLE 2 Results obtained in the halo bioassay carried out on the column fractions of the acidic extract obtained from slx-day old fruits ----- 1 Column Diarneter (in mm) of the halo for each half-seed Frac­ Total Mean tions 1 2 3 4 5 6 7 8 9 10 1 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 3 0 0 0 0 0 0 0 0 0 0 0 0 4 0 0 0 0 0 .0 0 0 0 0 0 0 5 16 15 14 ' 12 16 14 12 13 14 14 140 14±1 6 16 15 14 16 17 14 13 13 12 13 143 14±2 7 15 9 14 15 7 13 11 1i10 13 14 121 12131 8 16 17 16 18 16 17 18 15 18 15 166 1711 9 15 15 12 11 13 12 11 9 11 11 120 1212 10 to 20 0 0 0 0 0 0 0 0 0 0 0 0 -Control 0 0 0 0 0 0 0 0 0 0 0 0 Standard 19 20 21 20 19 21 21 21 20 20 202 2011 2ttg/c z ui £ 50- U>i <_i 75 207 238H3 1*7 129 ! 217 103 44 \ I rUh T K l U V 1 _r U + "T_r- "4| i l i— i— r t 50 100 150 200 250 300 350 400 MASS/CHARGE RATIO Fig 15 M ass spcctrum af MtAgTMSi UNIVERSITY OF IBADAN LIBRARY 1*43 McAgTMS* (CLXCI) Cvii) Methyl ester trimethylsilyl ether of gibberellin X The mass spectrum(Fig. 16) of the component (P*49. Fig. 8) had a molecular ion at m/e 68*45 and a very prominent M*~103 ion (base peak). The presence of this prominent M*~103 ion and an m/e 103 ion was indicative 39 of the presence of a -CH / 0Si(CH o ) o group in the molecule. There were prominent ions at m/e 129 3 1*47 and 217 indicating the presence of hydroxyl groups at the 2jS-positions* UNIVERSITY OF IBADAN LIBRARY 143a Fig 16 M ass spectrum of McTMSi ether ot MASS/CHARGE RATIO ’gibberellin X* 100- 5 0 - Mf-15 684VT) — p * i — i— i— r i — I----1----- fHW I * 1— I----1------1-----1 I 700 4Ö0 450 500 551 0 600 650 MASS/CHARGE RATIO Fig 16 (contd) RELAT IVE INTEN S IT V ( ”/.) RELAT IVE I N T E N S I T Y ( % ) UNIVERSITY OF IBADAN LIBRARY The absence of prominent ions at inJe 201 J 20% coupled with the fact that the molecular ion at m/e 681! was not very prominent indicated the absence of a hydroxyl group in the 13~position. The absence of M -119/120 ions Cdue to the loss of 2 rnoles of CH^COOH) and M+~91/92 ions (due to the loss of 59/60 a.m.u. from one methoxycarbonyl group and 31/32 a.m.u. from another methoxycarbonyl group) indicated that the compound was monocarboxylic The mass spectruia of the compound showed the general pattern associated with the mass spectra 39 of MeTMSi derivatives of C^g — gibberellins. The mass spectrurn was however different from the mass spectrurn of the MeTMSi derivative of any of the known C^g _ gibberellins. It was therefore believed to be a new gibberellin and was tentativeiy referred to as gibberellin X. The molecular weight of the MeTMSi derivative of this gibberellin indicated the presence of four hydroxyl groups in the molecule. Two of these should be in the 2,3-positions as discussed above. One other hydroxyl group should be a CH^OH group as discussed above. It could be either the C~VS methyl group or the C-17 methylene group that had been converted to the hydroxymethyl group. The likely sites left for the fourth hydroxyl group were positlons 115 125 15 and 16. 15-hydroxylation was usually characterized 16 6 by a prominent ion at m/e 156. 11- and 12- UNIVERSITY OF IBADAN LIBRARY 145 hydroxylated GAs generally showed^^ prominent M*-90 ions (due to the loss of TMSiOH from the parent ion). The absence of prominent ions at m/e 156 and M+~90 in the mass spectrum of the compound therefore suggested the absence of hydroxyl group at Position 11, 12 or 15. The 016 was therefore left as the most probable site for the fourth hydroxyl group. Normally 16-hydroxylation was indicated by the presence of a prominent ion [XXIX] at mJe 130 which was usually the base peak. This ion was derived from cleavage of ring D (Scheine ^ page ). This ion was absent in the mass spectrum of this compound. This was thought to be due to the fact that the 017 was also hydroxylated. Thus the methyl group in the m/e 130 ion would be replaced by the CH OSi(CH ) group as-shown in [CLXCII]. Therefore no m/e 130 ion should be observed jC H 3 .|C H02S iO(CH_3) 3 CH2 = C - 0+ - Si(CH3)3 CH2 = C - 0+-Si(CH3>3 m/e 130 [XXIX] [CLXCII] The most likely structure of the compound was therefore considered to be the methyl ester trimethylsilyl ether of 2, 3, 16, 17-tetrahydroxy gibberellin [CLXCIII]. The stereochemistry of the 2,3-hydroxyl groups was not certain. Although the Stereo­ UNIVERSITY OF IBADAN LIBRARY 146 chemistry of the 16-hydroxyl group was also not certain, the a configuration was thought to be the more likely by comparison with the structures of the known gibberellins.. An attempt was made to confirm the structure postulated by partial synthesis from gibberellin A^. The results obtained will be discussed later in this chapter. (b) Column fraction 5 The gas-liquid chromatograru of the MeTMSi derivative of this fraction (Fig. 17) had many peaks. This indicated that there UNIVERSITY OF IBADAN LIBRARY 146a Retention time (min) Bg-17. (k» dwömffltogrom ot the MiTMSi derivative ot cotumn traction 5 ot the AE extract trom 6- day old fruits (2 % SE. 130x 04cm coktma Nifrogtn c a r r it r gas 60 m l/m in. temparature programme 180-250° c£ UVmm ) Fig. 16 M o » epectrwn ot MeA^TMSi MASS/CHARGE RATIO i • U RELATIVE IN T E N S IT Y {%)NIVERSITY OF IBADAN LIBRARY -• 147 were many compounds present.. Most of these compounds could not be identified and were thought to be mostly phenolic impurities. Some compounds which were identified earlier in fraction 4 were also present in this fraction. These were (i) methyl ester of phaseic acid [CLXXVI] (P23, Fig., 17), (ii) methyl ester trimethylsilyl ether of 6’hydroxymethyl abscisic acid [CLXXXIII] (P283 Fig. 17) and (iii) methyl ester trimethylsilyl ethep of gibberellin A o [CLXCI] (P45, Fig. 17). Some other compounds which were identified in this fraction are listed below. (i) Methyl ester of abscisic acid [CLXCIV] The mass spectrum (Fig. 13) of the component (P21, Fig. 17) had a weak molecular ion at m/e 278. There were important ions at M+-18, M+-32, M*~56, m/e 190 (base peak), m/e 162, m/e 147, m/e 134, m/e 125 and m/e 112. The M^-18 and the M+-32 ions were due to the loss of H^O and CH oOH respectively from the molecular ion. The other ions (M+-56, m/e 190, m/e 147, m/e 134 and m/e 125) were derived as explained earlier (Scheine 36, pagel"2>?> ). UNIVERSITY OF IBADAN LIBRARY 14-8 Me ABA ( C L X C IV ) The mass spectrum was identiea.1 to that of authentic MeABA and was identical to the published 167 mass spectrum of MeABA. It was therefore conclusively identified as MeABA. (ii) Methyl ester trimethylsilyl ether of gibberellin A -(---M---e---A- .T-M---S---i---)- - ---[--C---L---X--X--I~-V--]--- The mass spectrum (Fig., 18) of the component (PM-O, Fig.. 17) had a strong molecular ion (base peak) at m/e H32.» This coupled with the presence of prominent ions at m/e 207/208 indicated the presence of a hydroxyl group in the 13-position. There were the ions at m/e 735 m/e 75 and M+~15 associated with the TMSi ether group^ UNIVERSITY OF IBADAN LIBRARY 149 There were other prominent ions at M*-59 CM+-CH oC00), m/e 303 and mJe. 235, The mass Spectrum was identical with that of Standard MeA^TMSi and with the published 39 mass spectrum of MeA^TMSi. The compound was therefore conclusively identified as methyl ester trimethylsilyl ether of gibberellin A^ [CLXXIV]. MeA6 TMSj (iii) Methyl ester trimethylsilyl ether of gibberellin Y The mass spectrum (Fig* 19) of ehe component (P^Ö, Fig. 17) had a very strong molecular ion Chase peak) at m/e 596, UNIVERSITY OF IBADAN LIBRARY 149a Fig. 19 (contd) UNIVERSITY OF IBADAN LIBRARY 150 There were prominent ions at m/e 129 Cindicating 3-hydroxy- lation), m/e IM-7 and m/e 217 Cindicating 2,3-diol System). There was a prominent ion at mJe 103. This was taken to indicate the presence of CH^OSiCCH^)^ (103 a.m..u. ) group-in the molecule. There were no m/e 207/208 ions. 13-hydroxylation was therefore considered to be absent. The mass spectrum of the compound had the general characteristic features of the mass spectra 39 of MeTMSi derivatives of C^g- gibberellins. It was however different from the mass spectra of the MeTMSi derivatives of the known gibberellins. It was therefore believed to be a new gibberellin and was tentatively referred to as gibberellin Y. The most likely structure of the compound was considered to be methyl'ester trimethylsilyl ether of 2*3517~trihydroxy gibberellin A y [CLXCV]. The stereochemistry of the 2- and 3~ hydroxyl groups were not certain. 23-, 33-hydroxylation was however thought to be the most likely by analogy with the known gibberellins. UNIVERSITY OF IBADAN LIBRARY 151 (CLXCV) An attempt was made to confirm this postulated structure by partial synthesis from gibberellin A . The results obtained will be discussed later in this chapter. Methyl ester trimethylsilylether of gibberellin A^ (MeA^TMSi) [CLXCVI] and methyl ester trimethylsilyl ether of gibberellin A^g (MeA^gTMSi) [CLXCVII] were also thought to be probably present in fraction 5. The mass spectrum of the peak CPH29 Fig.. 17) that was thought to contain MeA TMSi had a prominent- molecular ion at m/e 506. All the significant ions in the mass spectrum 39 of UNIVERSITY OF IBADAN LIBRARY 152 MeA^TMSi were present. There viere however many other ions present which could not ba due to MeA TMSi. The presence of MeA^TMSi was therefore not certain at this stage5 although it was thought to be likely. j (CLXCVI) . (CLXCVII) The mass spectrum of the peak (PM-3, Fig. 17) which was thopghf to be probably inainly MeA^0TMSi also had a molecular ion at m/e 506. All the significant ions in the mass spectrum 168 of MeA^gTMSi were present. Definite identification as MeA^gTMSi was however impossible because of the presence in the mass spectrum UNIVERSITY OF IBADAN LIBRARY 153 of many prominent ions which. could not be due to MeA 29TMSi. Cc) Column fraction 6 None of the components in the gas-liquid chromatogram of the MeTMSi derivative of fraction 6 could be identified from the mass spectra obtained. Most of them were thought to be phenolic impurities. (d) Column fraction 7 All the peaks in the gas-liquid chromatogram of this fraction except one could not be identified from their mass spectra. One of the peaks however appeared to contain a mixture of MeTMSi derivatives of two gibberellins together with some impurities. In the mass spectrum of the peak, there were some prominent ions at m/e 416 (base peak), m/e 417, m/e 418, and m/e 419. Their relative intensities were 100, 42, 58 and 25 respectively. The pattern of the relative intensities of these ions seemed to indicate that the mass spectrum might be that of a mixture.. The ion at m/e 416 might then be the molecular ion of one of the components of the mixture while the ion at m/e. 418 might be the molecular ion of another component. The component with M .j- 416 was considered to be probably MeA^TMSi [CLXCVIII]. This was because oftyhperesence of other ions UNIVERSITY OF IBADAN LIBRARY 154 at m Je401 (M*-CH ), 372 (>I+-C0 ) 343, 299 275, 208, 207 and 180 o Z. which are important Ions in the mass spectrum of MeA 5TMSi. MeA20 TMSi (CLXCVII I) (CL XC1X) The other component with molecular5/ oant m/e 418 was considered to be probat ly MeA TMSI [CLXCIXJ * This was because sorne other important ions in the mass spectrum 39 of MeA^TMSi were also present. These were the ions at m/e 403 tM+-CH O ) 5 m/e 375, m/e 359 (M+-CH oC00), UNIVERSITY OF IBADAN LIBRARY 155 m/e 314 5 m/e 301 , mJe 208 and m/e 207. Concluslve Identification of these two gibberellins was however not possible at this stage account of the presence of a large number of other prominent ions in the mass spectrum which could not have been derived from either MeA 5TMSi or MeA' TMSi. (e) Column fraction 8 The mass spectra obtained from a scan of most of the peaks in the gas-liquid ehromatogram of the MeTMSi derivative of this fraction could not be interpreted. One of the peaks however was believed to contain the MeTMSi derivative of a gibberellin together with some impurities. The mass spectrum that was obtained from a scan of the peak had a molecular ion at m/e 418. There were M ~j- -15, M -j~-18, . M 4* -32, M 4* -60 and M 4*-90 ions. There were also very prominent M+-i129, M+-134, M*-193 and M+~194 ions. All these ions were important prominent ions in the mass spectrum 39 of Standard MeA^TMSi [CC]. The M 4 -1-344, -M -193 and M 4--194 ions were particularly characteristic of the mass spectra 39 of MeA^TMSi and MeA^TMSi« The M + -134 ion had been shown 39 to be ^^H^gO Si anĉ *t̂ie ^4 - -193/194 ions to be ^3^24/25^^ '*'n case MeA^TMSi. These fragment UNIVERSITY OF IBADAN LIBRARY 156 ions should contain ring A of MeA^TMSi. Also present in the mass spectrnm of the. peak were the ions at m/e 129 Cindicative of 3~hydroxylation) and m/e 73/75 (from the TMSi group). Although there were many other ions which could not have been due to MeA^TMSi in the mass spectrum, the evidence outlined above was thought to be strong enough to indicate the presence of MeA^TMSi [CC] in the peak. ( C C ) UNIVERSITY OF IBADAN LIBRARY 157 Cf) Column fractions 9 and 10 None of the many peaks in the. gas-liquid chromatograms of the MeTMSi derivatives of these two fractions could be identified from their m*ass spectra. The results obtained above from the GC-MS analysis of the selected derivatized coluinn fractions (of the acidic ethyl acetate extract obtained from six-day old fruits) showed that a lot of impurities still interfered seriously with the analysis. Only fraction *4 out of the fractions analysed was reasonably free from impurities. For example, although the thalot bioassay ■( results (Fig. 7 and Table 2) indicated the presence of gibberellin activity in fractions 6 and 9, no gibberellin was detected in these two fractions by GOMS analysis. This was presumably due to the fact that the impurities present in these fractions were masking the presence of the gibberellins. Further purification of the column fractions was therefore considered to be necessary. UNIVERSITY OF IBADAN LIBRARY 158 (3) TLC of selected column fractlons and GC-MS of some Rf zones of the TLC An attempt was made to find out whether further purification of some of the column fractlons by TLC would help to remove some of the impurities in the fractlons. Aliquots of the PVP purified extract obtained from column fractlons 5 to 10 viere mixed together and subjected to TLC. The thin-layer chrornatogram was divided into ten equal zones after development in ethyl acetate/chloroform/acetic acid (15; 5; 1* v/v/v). ’Halo1 bioassay indicated. gibberellin activity in the extract obtained from Rf 0.4 - 0.5 and 0.5 - 0.6, of the thin-layer 4 'Y ' chrornatogram. Extracts from selected Rf zones (0.3 - 0.4, 0.4 - 0.5, and 0.5 - 0.6) were analysed on the G6-MS as the MeTMSi derivatives. The result of the lhalol bioassay was taken into consideration in selecting these Rf zones. Also most of the gibberellins and some of the other plant hormones such as ABA and PA would be expected to occur within the selected Rf zones in the developing mixture that was used. The results obtained in the GC-MS analysis of the derivatized selected Rf zones are given below. (a) Rf 0.3 to 0.4 No gibberellin could be identified in the MeTMSi derivative UNIVERSITY OF IBADAN LIBRARY 159 of the extract obtained from Rf 0.3 to 0..4* The fragmentatlon pattem of the mass spectra of the. various peaks in the gas-liquid chromatogram indicated that a lot of impurities were still present. (b) ^ Q.M- to 0,5 No particular gibberellin could be conOclusively identified in the MeTMSi derivative of the extract from RF 0.4 to 0.5. Some of the mass spectra obtained showed the fragmentation pattern associated with the MeTMSi derivatives of some gibberellins5 for example MeA^TMSi. Each of these mass spectra however also contained many ions which could not be due to the methyl esters or the MeTMSi of gibberellins. Conclusive identification of any gibberellin in this Rf zone was therefore not possible. (c) Rf 0.5 - 0.6 The GC-MS analysis of the MeTMSi derivative of the extract from Rf 0.5 to 0.6 afforded the identification of some plant hormones in this Rf zone., One of the peaks in the gas-liquid chromatogram (Fig. 20) of the MeTMSi derivative of the extract from this Rf zone appeared to consist of a mixture of the MeTMSi derivatives of two gibberellins and some impurities. The mass spectrum that was obtained from a scan of the first part of the peak (PIS, Fig. 20) UNIVERSITY OF IBADAN LIBRARY 159a Retention time (min) Fig. 20 Gas chromatogram ot TLC Rf. 05-0-6 of fractions 5 to 10 of AE extract from 6-day old fruits (MeTMSi derivative) ( 2 % SE 33, 130x0 Ucm carrier 60ml/min, temperature Programme 180-250° ai P/m in) UNIVERSITY OF IBADAN LIBRARY 160 showed a molecular ion at m/e 418. There were sorne other prominent ions at m/e 403 üZ-CHg), 375, 359 CM+-CH3C00), 301, 235 and m/e 208/207. This fraginentation pattern was similar 39- to that of MeA Z.„TMSi. It was thought that the evidence was strong enough r to indicate the presence of MeA^TMSi despite the presence of other prominent ions in the mass spectrum. The mass spectrum obtained from a scan of the centre of the peak indicated the presence of MeA TMSi and possibly MeA TMSi together with some impurities. The mass spectrum of the last part of the peak appeared to consist largely of impurities and possibly MeA^TMSi. There was a molecular ion at m/e 416. Other ions that were indicative of the likely presence of MeA D TMSi were at m/e 401 (M+-CH O)5 372, 357 (M+-CH oC00), 299, 208 and 207. The presence of many other prominent ions which could not be due to MeA^TMSi in the mass spectrum, made conclusive Identification of MeA^TMSi impossible. 1t. was however thought that the presence of MeArTMSi was very likely. Also identified in the MeTMSi derivative of the extract from Rf 0.5 - 0.6 were MeA,4 TMSi [CC], MeA bTMSi [CLXXIVJ and MePA [CLXXVI] . . On the whole it appeared that the further purification of the column fractions on TLC did not significantly remove the impurities that v.Tere interfering with the GC-MS analysis. UNIVERSITY OF IBADAN LIBRARY 161 Several plant hormones were thus identified (by means of GC-MS analysis) in the acidic ethyl acetate extract from 6-day old fruits. These were GA^, GA^, GA^5 ‘iso’GA^, GA^ anĉ ^wo new gibberellins tentatively called gibberellins X and Y. Others were ABA, PA, DPA,^isoDPA and 61-hydroxymethylABA. Also probabl present were GA^, G _, GA.^ and GA^g* They were identified in some of the fractions that were collected during the charcoal- celite column chromatography of the acidic ethyl acetate extract that was obtained from 6-day old fruits as shown in table 3 below. UNIVERSITY OF IBADAN LIBRARY 162 TABLE 3 Plant horraones identifled in 6-day old fruits of Vigna unguiculata CL.Walp) by GC-MS Column Eluant Compounds fractions [%(CH3)2CO in H20] 25 - 38 PA, DPA , 'isoDPA, 61-hydroxymethylABA, GA 8 j visoGA 8 and gibberellin X 5 38 - 4M- PA, 61-hydroxymethylABZ ABA, GAg , GAg, gibberellin Y, GA ? and GA?g? 7 49 - 53 GA5? and GA20 8 53 - 58 GA4 UNIVERSITY OF IBADAN LIBRARY 163 (4) Wheat coleoptile segment bioassay on the acidic ethyl acetate extract that was obtained frora 6-d'ay old fruits Ca) Bioassay on TLC Rf zones A portion of the PVP purified acidic ethyl acetate extract from 6~day old fruits was subjected to thin-layer chromatography using EtOAc/CHCl o/HOAc (15: 5: 1) as the developing solvent mixture. The thin-layer chromatograrn was divided into ten equal zones. Aliquots of the extract from each Rf zone were tested for biological activity in the wheat coleoptile segment bioassay. The wheat coleoptile segment bioassay is basically a test for auxins and growth inhibitors. Gibberellins do not generally respond to the test. This is because gibberellins do not generally have any reasonable promotive effect on the growth extension of plant sections although they are very effective in promoting growth extension of intact plants. The main aim of the test was to find out if there was any active auxin in the extract. The results that were obtained in the bioassay are shown in Table 4 below and the histogram is shown in Figure 21. UNIVERSITY OF IBADAN LIBRARY 164 TABLE 4 Results obtalned in the wheat coleoptile Segment bioassay on TbC Rf Zones of the acidic ethyl acetate extract from 6-day old fruits TLC Rf Length of wheat coleoptile se;gments (in oun) Zones 1 2 3 4 5 6 7 8 9 10 TOTAL MEAN 0.0-0.1 131 125 129 121 134 128 120 130 132 130 1280 12815 0.1-0.2 120 102 102 121 109 105 109 115 115 106 1104 11017 0.2-0.3 140 132 116 110 115 130 120 118 115 126 1222 12219 0.3-0.4 130 130 131 120 116 120 140 120 124 126 1257 12617 0.4-0.5 118 130 130 ■ 122 132 113 113 126 127 116 1227 12318 0.5-0.6 94 89 93 83 90 90 86 92 93 89 899 9013 0.6-0.7 90 100 94 98 94 95 92 98 94 96 951 9513 0.7-0.8 113 108 103 109 104 110 105 102 108 110 1072 10714 0.8-0.9 113 ■104 110 115 120 114 108 110 118 110 1122 11215 0.9-1.0 110 110 110 115 105 108 110 114 106 112 1100 11013 CONTROL . 104 110 119 106 116 110 105 115 115 110 1110 11115 ) UNIVERSITY OF IBADAN LIBRARY 164a TIC Rf. Zontt Fig-21 • Wheat ccleoptile Woassay on Tl£ Rf zont t cf tht AE traction from 6-day old fruRt. FigW-2h2e at R i zones of paper Chromatogramccleoptile bioassay on the paper chromatography Rf zo n tt of the AE traction trom 6 days old fruits- UNIVERSITY OF IBADAN LIBRARY 165 Reasonable growth promoting activi.ty was observed at Rf 0,0-0.1 and Rf 0,3-0,*4, The aetivity at Rf 0.0-0.1 was more prominent, Growth inhibiting aetivity was observed at Rf 0.5-0.6 and Rf 0.6-0.7. The presence of growth inhibiting aetivity at Rf 0.5-0.6 and Rf 0.6-0.7 was not unexpected. ABA and its metabolites occurred at these Rf zones in the solvent System that was used. Since these compounds had been identified in the extract by GC-MS, growth inhibitory aetivity would be expected in these Rf zones. The compound or compounds that exhibited the growth promoting aetivity at Rf 0.0-0.1 and 0.3-0.4 were not known. The common acidic natural auxins for example indole-3-acetic acid(IAA), indole- 3-propionic acid (IPA), and indole-3-butyric acid(IBA) occurred at higher Rf values (0.5 and above) in the solvent mixture that was used for the TLC. The growth promoting aetivity that was observed was therefore not considered to be due to any of these. The MeTMSi derivative of the extract from Rf 0.0-0.1 was analysed on the GC-MS in an attempt to identify the compound or compounds that caused the promotive effect that was observed at Rf 0,0-0.1. Unfortunately none of the compounds that were present could be identified from their mass spectra except the methyl UNIVERSITY OF IBADAN LIBRARY 166 esters of palmitic and stearic acids. The Identity of the growth promoting substance in this Rf zone (0.0-0.1) therefore remained unknown. (b) Bioassay on Rf zones of paper chromatogram The PVP purified acidic ethyl acetate extract was subjected to paper chromatography using Isopropanol/ammonia/water (10:1:1) as the developing solvent mixture. The results that were obtained in the wheat coleoptile Segment bioassay on each of the ten equal Rf zones of the paper chromatogram are shown in Table 5 and the histogram is shown in Figure 22. UNIVERSITY OF IBADAN LIBRARY 167 TABLE 5 Results obtalned in the wheat coleoptile Segment bioassay on Rf zones of the paper chromatograra of the acidic ethyl acetate extract from 6-day old fruits Paper Length o f whe at co. leoptile Chromatogram segm ents (mm) Rf zones 1 2 3 4 5 TOTAL MEAN 0.0 - 0.1 72 72 56 62 70 332 66 ± 7 0.1 - 0.2 75 68 55 68 70 336 67 ± 7 0.2 - 0.3 66 57 72 72 65 332 66 ± 6 0.3 - 0.4 81 74 72 73 62 362 . 72 ± 7 0.4 - 0.5 65 53 50 59 64 291 58 ± 7 0.5 - 0.6 62 58 66 55 57 298 60 ± 4 0.6 - 0.7 45 50 50 51 49 245 49 ± 2 0.7 - 0.8 50 50 49 49 51 249 50 ± 1 0.8 - 0.9 90 88 79 85 95 437 87 ± 6 0.9 - 1.0 84 72 64 60 76 356 71 + 9 CONTROL 63 65 62 63 63 316 63 + 1 UNIVERSITY OF IBADAN LIBRARY 168 Strong growth. inhibiting activity was observed at Rf 0..6 - 0.7 while strong prornoting activity was observed at Rf 0.8 - 0.9. It was thought to be unlikely that any of the common free acidic natural auxins, for example IAA, I.PA, and IBA could have been responsible for the growth prornoting activity that was observed at Rf 0.8 - 0.9. This was because all of thera should occur 169 at lower Rf values of the paper chromatogram in the solvent System that was used for the development. The approximate Rf values for several auxins on paper chromatograms have been publish^ed1 69 . None of the free acidic natural auxins occurred at about Rf 0.8 - 0.9 when isopropanol/ ammonia/water (10:1:1) was used for developing the paper chromatogram The approximate Rf valuCeYs1; ;169 of some of them were IAA, 0.25 - 0.35; IPA, 0.35 - 0.45, and IBA, 0.45 - 0.55. The identity of the substance that was responsible for the growth prornoting activity that was observed at Rf 0.8 - 0.9 of the paper chromatogram remained unknown. Analysis of the acidic ethyl acetate (AE) fraction of the extract obtained from two~day old fruits. The acidic ethyl acetate fraction that was obtained from 2-d.ay old fruits was an<4L3~sed on the GC^MS as the MeTMSi derivative after purification with PVP. No plant hormone was detected in UNIVERSITY OF IBADAN LIBRARY 169 the extract by GC-MS, at this stsge. 1. Colurnn chromatography of the acldic ethyl acetate fraction that was obtalned from 2-day old fruits« Purification of the acidic ethyl acetate fraction that was out obtained from 2-day old fruits was carried /Non a charcoal-celiteJ colurnn eluting with a gradient of increasing concentrations of acetone in water. Seventeen 150cm 3 fractions were collected. (a) TLC analysis of colurnn fractions Analytical TLC indicated that ABA and its metabolites were probably present in colurnn fractions 4 to 6. The TLC analysis however did not indicate the presence of gibberellins in any of the seventeen colurnn fractions. (b) Barley half-seed (halo) bioassay of colurnn fractions. Gibberellin activity was not detected in any of the colurnn fractions. 2. GC-MS analysis of selected colurnn fractions Selected colurnn fractions Ü4 to 10) were analysed on the GC-MS as the MeTMSi derivatives after purification with PVP. Fractions 4 to 6 were selected on the basis of the results of the analytical TLC which indicated that ABA and its metabolites were probably present in these fractions. Fractions 7 to 10 were selected just by analogy with the selection made in the case of the extract UNIVERSITY OF IBADAN LIBRARY 170 from 6~day old fruits. The results that were obtained. in the GC-MS analysis of the selected column fractions of the aeidic ethyl acetate extract that was obtained from 2-day old fruits are shown in Table 6 below. MePA [CLXXVI] and the methyl ester trimethylsilyl ether of 6!-hydroxymethylABA [CLXXXIII] were identified in fraction 4. MeABA was identified in fractions 5 and 6. No gibberellin was identified by GC-MS in any of the fractions 4 to 10. , TABLE 6 < !-------- Plant hormones that were identified in 2-day old fruits of unguiculata (L.Walp) by GC-MS. Column Eluants Compounds fractions [% CCH3)2CO in H203 27 - 35 PA and '61 -hydroxymethylAI 5 35 - HO ABA 6 HO - H5 ABA UNIVERSITY OF IBADAN LIBRARY 171 3. Wheat coleoptile Segment bioassay on the acidic ethyl acetate extract obtalned front Z-day old fruits A portion of the PVP purified acidic ethyl acetate extract that was obtained from 2-day old frnits was subjected to thin-layer chromatography using EtOAc/CHCl^HOAc (15: 5:1) for the development of the chromatogram. The results that were obtained in the wheat coleoptile Segment bioassay that was carried out on each of the ten equal Rf zones of the thin-layer chromatogram are shown in Table 7 below and the histogram is shown in Figure 23. Strong inhibitory activity was observed at Rf 0.5-0.6 and Rf 0.6-0.7. No growth promoting activity was observed in any of the Rf zones. UNIVERSITY OF IBADAN LIBRARY 172 TABLE 7 Results obtained in the wheat coleoptile Segment bioassay on TLC Rf zones of acidic ethyl acetate extract from 2-day old fruits. TLC Rf Length of wheat coleoptile Segments (in mm) Zones i 2 3 4 t- . r5jin ■ 6 7 8 9 10 TOTAL MEAN 0.0-0.1 110 112 112 103 114 108 107 110 113 112 1101 110+3 0.1-0.2 118 115 105 120 110 113 113 109 115 114 1132 113±4 114 110 117 115 114 114 110 120 112 114 1140 114+3 0.3-0.4 125 111 105 101 122 118 120 106 109 112 1129 11318 0.4-0.5 120 120 115 108 110 116 115 110 119 114 1147 11514 0.5-0.6 78 80 90 77 80 85 82 76 80 82 810 8114 0.5-0.7 95 108 98 98 98 97 100 103 97 98 992 9914 0.7-0.8 118 112 113 104 98 110 110 106 115 104 1090 10917 115 110 112 102 100 108 105 105 114 109 1080 10815 0.9-1.0 112 115 112 102 102 106 112 107 110 110 1088 10914 CONTROL 104 110 119 106 116 110 105 115 115 110 1110 11115 0 O 00 U1 roO oi . N 00IVERSITY OF IBADAN LIBRARY 173 Cornparlson of the extracts that were obtained from 2-day old and 6-day old fruits The results that were obtained in the. analysis of some of the horrnones in the 6-day old fruits and the 2-day old fruits clearly showed a disparity in the hormonal contents of these two groups of fruits. While several gibberellins (Table 3* page\b7L) were identified in the 6-day old fruits, no gibberellin (Table 6, pageVl^) could be identified in the 2-day old fruits. Growth promoting activity was observed in the auxin bioassay (wheat coleoptile segment bioassay) carried out on the extract i 2A Q<\ck ' X X , ̂ 0,0^ \ from 6-day old fruits (Tables 4 and 5, pages \WA- and \to^^j>)-No growth promoting activity was however observed when the same test was carried out on the extract from 2-day old fruits (Table. 7* page and Figure 23, page ItAVcO. Thus while auxins could be detected in the 6-day old fruits«, they could not be detected in the 2~day old. fruits. It therefore seemed that while gibberellins and auxins (growth promoting horrnones) were present in the 6-day old fruits* they were absent in the 2-day old fruits. If these growth promoters were present in the 2-day old fruits at all* their concentrations must have been so low that they could not be detected. UNIVERSITY OF IBADAN LIBRARY As mentioned earlier, under* the. Introduction C.page 15 )3 the plant hormones constitute the major interaäl factor governing abscission in plants. The growth promoters (auxins, gibberellins and cytokinins) promote the growth of plant organs and therefore normally inhibit the abscission of these organs. The growth inhibitors on the other hand, inhibit the development of plant organs and so normally promote the abscission of these organs. Abscisic acid and its metabolites therefore accelerate the abscission of plant organs, such as leaves, fiower buds, flowers and fruits. Another internal factor which regulates the abscission of plant organs is competition for available nutrients. This factor is also related to the hormonal content of the plant organs. This is because one major role of the growth promoters in relation to abscission is in the mobilization of nutrients. Thus plant organs which have high concentrations of the growth promoters deprive organs which have low concentrations of these growth promoting hormones of nutrients. The organs that are deprived of nutrients will then starve and abscise. In the light of wkat has been written above about the role of plant hormones in relation to abscission, the results that were obtained in the analysis of the hormonal content of the two UNIVERSITY OF IBADAN LIBRARY 175 groups of cowpea frults could be used to explain the absciscion problem In cowpea. The results that were obtained in the analysis of some of the growth promotlng hormones (GAs and auxins) in the .two groups of cowpea fruits clearly indicated the presence of these growth promoters in the 6-day old fruits. On the other hand5 these growth promoters were not detected in the 2-day old frults. It could therefore be expected that the growth promoters in the older (6-day old) fruits of the lowest raceme would lead to the mobilization of available nutrients towards these fruits. The younger fruits of the upper racemes which were deficient in the growth promoters ( and therefore deficient in nutrient "mobilizers") would then starve and abscise. The Interpretation of the results given above would then explain the observation of Ojehomon 1 2 4 that the abscission of the immature fruits and flowers of the upper racemes of the cowpea inflorescence were associated with the presence of maturing fruits at the lowest raceme. It would also explain his observati. on2 that the lateral flowers of each raceme were capable of growing to mature fruits if the older fruits below were removed. The Interpretation would also agree with. the results of the Investigations by Adedipe and Ormrod 6 and Adedipe et al 7 In which they showed that competition for nutrients was an important factor UNIVERSITY OF IBADAN LIBRARY 176 in the abscission problem in cowpea. Apart from the mobilization of nutrient, the growth promoting hormones can also have direct effect on abscission.. The growth of plants and plant organs is partly controlled by an interaction between the growth promoting hormones and the. growth inhibiting hormones. If the balance of growth regulators is in favour of the growth promoters5 the growth of the organ will progress. If however the growth Inhibitors are favoured relative to promoters, the growth of the plant organ is arrested. The 6-day old fruits had both growth promoting hormones (GAs and auxins) and growth inhibitors (ABA and its metabolites). The 2-day old fruits however had growth inhibitors but no detectable amount of the growth promoters. Thus the older fruits would have an adequate concentration of the growth promoters to counteract the effect of the growth inhibitors. The growth of these fruits would thus progress to maturity. In contrast, the 2-day old fruits (from the upper racemes) had no detectable amount of growth Promoters (GAs and auxins) to counteract the inhibitory effect of ABA and some of its metabolites that were identified in these fruits. The growth of these fruits would therefore be arrested and they would finally abscise. UNIVERSITY OF IBADAN LIBRARY 177 The inducing influence which the maturing fruits at the first raceme had on the abseission of the younger upper fruits could then he due to the abscising influence of hormones which were translocated in the acropetal Capical) direction from the maturing basal fruits./ Both growth Promoters and inhibitors accelerate 8 abseission when applied proximally to the abseission zone. Thus plant hormones (both growth promoters and inhibitors) entering the peducle from the maturing fruits of the lowest raceme and translocated in the acropetal Capical) direction would reach the younger, upper fruits from the proximal ena. They would therefore accelerate the abseission of these younger, upper fruits. This would thus agree with Ojehomoir s second hypothesis that the abseission of the immature fruits were stirnulated by substances produced in the lower, older fruits. Van Steveninck 17Q 5 17/ had postulated a si. mi.lar hypothesis to explain the abscising influence of maturing fruits, at the base of yellow lupin (hupinus luteus) inflorescence on the abseission of younger, upper fruits. (Yellow lupin exhibits an abseission problem quite similar to that of cowpea). Wheat coleoptile straight growth best on the extract which he obtained from 7-day old fruits indicated the presence of growth UNIVERSITY OF IBADAN LIBRARY 178 Promoter1/s. and inhibitor/s in the extract, Th-e main component of the inhibiting fraction was later identified by Cornforth. et al‘U 2.B. to be abscislc acid., Analysis of the acidic ethyi acetate (AE) fraction of the extract obtained from fruits ' ov/er ̂ . 6 days o\d The acidic ethyi acetate extract that was obtained from o\d fruits that were ovc-r six days^was also analysed. This was in order to compare the hormonal content with the hormonal contents of the 6~day old and the 2~day old fruits. The PVP purified extract was analysed on the GC-MS as the MeTMSi derivative. The gas chromatogram had very many peaks indicating the presence of many compounds most of which co.uld not be identified from their mass spectra. The mass spectrum of one of the peaks however indicated that it was a mixture of MeA o TMSi [CLXXIV] and MeA 1 /TMSi [CLXXV]. All the prominent ions in the mass spectra of MeA TM'Si and MeA^^TMSi as discussed earlier in this chapter Cpage ) were present in the mass spectrum of the peak. The extract was further purified on a charcoal-celite column eluted with a gradient of increasing concentrations of acetone in water. UNIVERSITY OF IBADAN LIBRARY 179 GC-MS analysis of the MeTMSi deri.yati.ves of some of the PVP purified colurnn fractions afforded the. identification of MePA [CLXXVI] , MeDPATMSi [CLXXXII] , iso-MeDPATMSi, MeABA [CLXCIV] , Me6'-hydroxymethylABATMSi [CLXXXIII], MeA 6 TMSi [CLXXIV], MeA.oTMSi [CLXCI] and possibly MeA TMSi [CLXCIX]. The hormonal content of the fruits that were ovar 6 days o\d was thus basically similar to the hormonal content of the 6-day old fruits, More gibberellins were however identified in the extract from 6-day old fruits than from the extract from fruits o\ck, This was not surprising since the gibberellin content of plant organs generally decrease 3 72b as the organs get older. Analysis of the extracts obtained separately from the seeds and from the ,v>3q\\ of 6-day old fruits One problem that was constantly encountered during the GC-MS analysis of the extracts that were obtained from the 6-day and 2-day old fruits was the presence of a lot of impurities in these extracts, These impurities interfered seriously with the analysis. The various mathods that were used to purify the extracts did not succeed in removing a substantial amount of the impurities. An attempt was thex-efore made to find out wheLher the impurities were from the seeds or from the ^ . wall. Most of the UNIVERSITY OF IBADAN LIBRARY 180 plant hormones would normally be expected to be in the seeds. The seeds of some 6-day old fruits were separated from the ^v\\\V . walls and separate extracts were then obtained from the seeds and the r̂v\\V walls. Without further purification, the acidic ethyl acetate fraction of each extract was analysed on the GC-MS as the MeTMSi derivative. The GC-MS analysis of the derivatized crude ethyl acetatd fraction obtained from the seeds afforded the identification of MeA 1 TMSi [CLXCVI], MeA,4 TMSi [CC]5 MeA 5 TMSi [CLXCVIII] , MeA 6TMSi [CLXXIV]s MeA o TMSI [CLXCI] 9 MeA 1 / TMSi [CLXXV], MeA 20TMSi [CLXCIX]9 MeTMSi derivative of gibberellin Y [CLXCV] and possibly MeA.' TMSi [CCI] in the extract. < A:lthough the mass spectra obtained for many of these compounds were contaminated with impurities, they were considered-good enough for the identifications to be made. Only MeA^TMSi could be identified in the derivatized (MeTMSi) crude ethyl acetate fraction that was obtained from the walls. All the other compounds could not be identified. Most of them were thought to be phenolic impurities. The result that was obtained for the GC-MS analysis of the crude acidic ethyl acetate fraction obtained from 6-day old seeds was very interesting in that many gibberellins were identified. This result could be compared with the result that was obtained for the UNIVERSITY OF IBADAN LIBRARY 181 GC-MS analysis of the acidic ethyl acetate fraction obtained from 6~day old fruits, after a one step purification with PVP (page\d(o). Only MeA TMSi and MeA TMSi could be easily identified from the extract from 6-day old fruits after the one step purification with PVP, The other honnones were identified only after further purification of the extract by column chromatography. This thus showed that the extract from the seeds contained less amount of impurities than the extract from the fruits. Therefore the major part of the impurities that were encountered in the analysis of the extracts obtained from fruits were from the fruit walls and not from the seeds. This conclusion was further confirmed by the result that was obtained above for the GC-MS analysis of the extract from the fruit walls. It therefore appeared advisable that any further investigations of the hormonal content of cowpea fruits should be concentrated on the seeds. The acidic ethyl acetate extract obtained from some 6-day old fruits which were extracted in a soxhlet extractor with 80% aqueous methanol was used for some preliminary biological assays. UNIVERSITY OF IBADAN LIBRARY 182 MeAigTMSi ( C d ) Partial synthesis of the structures that were tentatively assigned to gibberellins X and Y As stated earlier in this chapter, two new compounds were detected in the extract that was obtained from 6~day old fruits. They were believed to be new gibberellins from the evidence obtained from the mass. spectra of their MeTMSi derivatives* They were tentatively referred to as gibberellins X and Y. On the basis of the mass spectral evidence (page \ ) 5 the structure of gibberellin X was thought to be 16a5 17-dihydroxy derivative of gibberellin A O H [CCIIa], The stereochemistry of the 2,3-hydroxyl groups was not certain, but 2ß5 36 was thought to be UNIVERSITY OF IBADAN LIBRARY 183 the mast probable by analogy with th.e structures of the known CiQ-gibberellin3. The stereochemlstry of the 16-hydroxyl group was also not certain. The a~configuration was however thought to be more likely by comparison with the structures of the known gibberellins , ( F i g 1). Gibberellin X Gibberellin Y (CC IIa) R = OH (CCIIb) R = H The structure of gibberellin Y was also thought to be probably 17-hydroxy derivative of gibberellin A o H [CCIIb]* Although the configurations at C-2, C-3 and C~16 were not certain5 the most likely configuration at these positions v.Tere thought to be as shown in CCIIb. In order to determine whether the tentatively assigned structures [CCIIa] and [CCIIb] for the two new gibberellins viere correct, it was decided to carry out their partial synthesis. UNIVERSITY OF IBADAN LIBRARY 184 Gibberellin a relatively abundant gibberellin was used as the starting material for the partial synthesis. HO ch 2 co2h G A 4 ( X I V ) 1. Partial synthesis of the structure that was tentatively assigned to gibberellin X and the 16-epimer In order to find out whether the structure [CCIIa] that was tentatively assigned to gibberellin X was correct, it was decided to synthesize the methyl ester of 16a? 17-dihydroxyMeGA^ [CCIII] from gibberellin A^ [XIV] as shown in seherne 37 below. UNIVERSITY OF IBADAN LIBRARY 185 S C H E M E 37 a 4' a 7 ( c c m ) Lx W]/[x iV; ä 3 C 0 2 C H 3 ( C C i V ) A mixture of GA^ and GA^ (70% GA^) was methylated with ethereal diazomethane, The crude methyl ester without further purification TOcO., The retention. time. of the MeTMSi derivative of gibberellin X was however different from the retention time of the TMSi ether of CCIII. It thus followed that the two compounds were not identica.1. The fact that their mass spectra were similar however showed that they were closely related structurally. It was thought that the natural compound might be the epimer of the synthetic compound. The epimerization was thought to be more likely at C-16. It was therefore decided to prepare the 16-epimer of CCIII in order to check whether the TMSi ether of this would have the same retention time with the MeTMSi ether of gibberellin X. The preparation of the 16~epimer of CCIII would involve the hydroxylation of the exocyclic double bond of MeGA^^ [CCV] from the more hindered face. An important method for the preparation of cis diois involving the introduction of the cis hydroxyl groups on the more hindered side of the molecule is the Woodward 177 reaetion. This is the reaction of olefins with iodine and silver acetate in wet acetic acid. Although there had been no report of the use of this reaction in gibberellin Chemistry, it had been used“1 78 in the total synthesis of some steroids. The Woodward reaction would thus give a cis UNIVERSITY OF IBADAN LIBRARY 190 glycol with opposite stereochemistry to that derived from osmium tetroxide.- The projected plan for the synthesis of the 16-epimer ofCCIII is shown in scheme 38 below. The first part of the scheme consisted of the synthesis of Me A^ [CCV] from GAfi [Xiyj following the method used by Beeley and MacMillan43 "CH2 M2A34 (CCV) o SCHEME 38 L ^ j UNIVERSITY OF IBADAN LIBRARY 191 Scheme 38 contd. MeOH'/AcOH -------------------*“ (CCV) = MeA34 A. mixture of GA *+ and GA / (70%GA H) was treated il3 in pyridine with sodium metaperiodate and a small amount of osmium tetroxide* GA norketone ‘ [CCVI] was obtained in a good yield* The meiting point (120-136°) was in close agreernent with the reported"^^ m.,p. (120-140°) for GA^ norketone. The mass spectrum of the MeTMSi derivative of the tGA norketone (M 420) gave the fragrnentation pattern expected for the MeTMSi derivative of GA^ norketone. UNIVERSITY OF IBADAN LIBRARY 192 This oxidative cleavage of the 16517-exocyclic double hond was effected in Order to prevent its being attacked by OsGij later in the scheine during the hydroxylation of the 2 «,3-positions. Combinations of periodate with osmium tetroxide or permanganate are usually used for this type of cleavage. This is because the osmium tetroxide or the permanganate converts the olefin to 152 glycol which is then cleaved by the periodate. The norketone was methylated with ethereal diazomethane and the resulting crude methyl ester was dehydrated 43 using phosphorus oxychloride in pyridine. The resulting olefin5 methyl 2>3- dehydroGA 9 norketone [CCVII ] was obtained in about 60% yield. In the n.m.r. spectrum of the olefin [CCVII] in deuterochloroform (CDC1 ), the olefinic protons were observed as a 2H-multiplet at O 65.6 to 65.9. The 5-H and the 6-H protons were observed as two doublets ce^tred at 62.5 (JlOHz) and 62.68 (JlOHz) respectively. The C-18 inethyl protons and the three methoxy protons appeared as two 3H-singlets at 61.23 and 63.73 respectively. Both this n.m.r. spectrum of the olefin and the melting point (158 - 160°) were in close agreement wTith those reported 7 3in literature. Treatment 43 of the olefi#n [CCVII] with osmium tetroxide in pyridine gave methyl GA norketone [CCyill] «, UNIVERSITY OF IBADAN LIBRARY 193 The fragmentation pattern of the mass spectrum of the TMSi derivative of the raethyl GA^ norketone that was obtained was in agreement with what was expected.. It had a molecular ion (base peak) at m/e 508. There were prominent ions at m/e 147 and m/e 217 Cindicative of 2*3 diol System). There were also other important ions at m/e 73, 75, 129, 477 (M+-31) and 493 (M+-15). As stated earlier, the ß5 B configuration of the 233^hydroxyl groups of the norketone [CCVIII] was assumed on the basis of steric considerations. Since hydroxylation of olefins with osmium tetroxide occur from the less hindered side, the attacll of the olefin [CCVII] by the bulky osmium tetroxide would be expected to be from the less hindered ß face of ring A. The 253-vicinal diol System of methyl GA vH norketone [CCVIII] were then protected 43 as the TMSi ethers. This was achieved by silylating the diol with a mixture of hexamethyldisilazane* trimethyisilyl Chloride and pyridine (2: 1: 2). The diol System was protected as the TMSi ethers because it was reported 180 that attempted Wittig 1‘ 81 reaction on the unprotected diol norketone [CCVIII] did not give satisfactory results. This was probably due to the ättack of the ylid on the unprotected hydroxyl groups of the diol norketone* ySV UNIVERSITY OF IBADAN LIBRARY 194 The protected diol norketone was then. treated 43 with methylenetriphenylphosphorane giving the MeTMSi derivative of GA34 [CCIX]. The conversion of aldehydes and ketones to olefins by means of the Wittig reaction 181 i well known in Chemistry. The hydro; lysis of the MeTMSi derivative of GA OH with a 4:1 mixture of methanol/acetic acid gave MeA ^ [CCV] in low yield. In the n.in.r. spectrum of MeA 0*4 in CDC1 O, the exocyclic methylene protons were observed at <54.96. The 5-H and the 6-H protons appeared as two doublets at 63.19 and 62.70 respectively. The downfield shift of the 5-H signal was due to the deshielding 32 by a 3ß-hydroxyl group. This confirmed the 2ß, 3ß-configuration of the 253-diol System. The mass spectrum of the TMSi derivative of the MeGA oH was consistent with what was expected. It showed a molecular ion at m/e 506 (base peak). There were prominent ions at m/e 147 and 2-17 indicative of the 2,3-diol systein. The next stage in the synthesis was the hydroxylation of the exocyclic double bond from the more hindered side of the molecule. As stated earlier 5 the Woodward reaction IV V should lead to the introduction of cis hydroxyl groups from the more hindered side of ring I) of the gibberellin molecule* The mechanism of the Woodward reaction is believed to be as shown in scheme 39 below. UNIVERSITY OF IBADAN LIBRARY 195 An olefin [CCXI] reacts with 1* forming an Intermediate cyclic iodonium ion [CCXII ]* It is believed that the I* is derived frorn acetylhypoiodide? CH COOI, formed from silver acetate and ö iodine. The iodonium ion is openedby an acetate group to give a trans-acetoxy iodide [CCXIII]. The acetate group of the acetoxy ! iodide displaces the adjacent iodide ion giving a cyclic acetoxonium ion [CCXTV]. Water (from wet acetic acid) then adds to the acetoxonium ion, giving a cis hydroxy acetate [CCXV]. Hydrolysis of this gives the cis diol [CCXVI]. SCHEME 39 ü / \ (CCX I) (CCXII) (CCXII I) (COMV) \/ h 2o h o - c " 0H HO — V — -------------------- j AcO — C HO — C / \ / \ (CCXV) (CCXVI) UNIVERSITY OF IBADAN LIBRARY 196 The Woodward hydroxylation was carried out on MeA.0ft* The ' final product was silylated and analysed on the GC-MS, The analysis showed that there were many compounds in the final product of the Woodward reaction. Three of the Compounds however had the expected molecular ion at m/e 684, It was therefore likely that one of these was the TMSi ether of CCX. One of these three had the same retention time with the MeTMSi ether of gibberellin X. Each of the three components however had some prominent ions in its mass spectrum which were not present in the mass spectrum of the MeTMSi of gibberellin X. 2. Partial synthesis of the .tentatively assigned structure of gibberellin Y and the 16-epimer Since gibberellin Y was thought to be probably 17-hydroxyGA^ [CCIIb], it was decided to synthesize the methyl ester of 17-hydroxyGA OH [CCXVII] in order to check whether the tentatively assigned structure was correct. The 17-hydroxyGA^ methyl ester [CCXyil] was synthesized from MeA as shown in scheme 40 below. UNIVERSITY OF IBADAN LIBRARY 197 SCHEME 40 CH2OH Hydroboration 5‘ is a versatile method for the anti- Markoynikov hydration of double bonds. It involves a cis anti- Markcvnikov addition from the less hindered side of the double bond. It was therefore thought that hydroboration should provide a nice method for preparing the methyl ester of the structure that was tentatively assigned to gibberellin Y [CCXVII] from MeÄ^. MeA Ot in dry tetrahydrofuran (THF) was treated with a solution of diborane (B^2 H 6) in THF, under nitrogen gas.. The resulting organoborane was oxidised with hydrogen. peroxide in the presence of aqueous NaOH. The gummy product was analysed on the GC-MS as the UNIVERSITY OF IBADAN LIBRARY 198 TMSI ether.. The GC-MS analysis of the product revealed that it was a single compound. The general fragmentation pattern of the mass spectrum (Fig. 25) of the TMSi of this compound was consistent with what would be expected from the TMSi of [CCXVII]. It had a molecular ion (base peak) at mJe 596. There was a prominent ion at m/e 103 indicating the presence of the -CH TMSi group. There was an ion at m/e 129 indicating the presence of 3-hydroxyl group. Prominent ions at m/e 147 and 217 indicated the presence of the 253-diol System. There was a significant ion at m/e 506 (M+-90) which was probably due to the loss of a TMSiOH group from the molecular ion. The first half of the mass spectrum was.very similar to the mass spectrum of MeA TMSi except that the ion at m/e 103 was much more prominent. Although the mass spectrum (Fig. 25) of the TMSi derivative of the hydroboration product was similar to the mass spectrum (Fig. 19yv/vof the MeTMSi derivative of gibberellin Y (the natural compound)5 there were some significant differences. Moreover their retention times on GLC were not the sarne, although they were close. It therefore followed that the structure [CCIIh] that was tentatively assigned to gibberellin Y was not correct. It was however thought that the natural compound might be the epimer of UNIVERSITY OF IBADAN LIBRARY 198a 100-t 50- 73 75 U7 217 I03 L 241 245 Ä, 28 8 ^g9 -1L29T “ —T i_L ii__Jy I 11 11 i r, I A ,”L— l 50 100 150 200 250 300 350 400 Fig. 25- Mass spcctrum of Me 17-hydroxy A3 4TMS1 MASS'CHARGE RATIO Fig 25 (contd) RELATIVE INTENSITY (•/„) RELATIVE INTENSITY (*/.) UNIVERSITY OF IB DAN LIBRARY 199 the synthetic compound.. This was because of the similarity of the mass spectra of their MeTMSi derivatives. Also although. their retention times were not the same5 they were quite close. The epimerisation was thought to be more like3.y at C-16.. The stereochemistry of the hydroboration product [CCXyil] at C-16 had been assumed to be as shown on the basis that the reagent wou-ld attacH the double bond from the less hindered a face of ring D. The mechanism of the hydroboration. reaction could be envisaged to have taken place as shown in scherne *41 below leading to the formation of CCXVII. SCHEME 41 H H S f 6- CO 1 H (CCV) c- ------BC (CCXVIli) b UNIVERSITY OF IBADAN LIBRARY 200 Scheine 41 contd H? 0? ( C C X V I I ) Nc OH I H \ c c x i x ] The attack on the exocyclic double bond of MeA O H CCCV] would be initiated by the Polarisation of the boron-hydrogen bond. The attack would proceed on the less hindered a~face leading to the formation of the alkyl borane [CCXIX]5 through a four- centred transition state [CCXVIII]. Oxidation of the alkyl borane [CCXIX] with hydrogen peroxide in the presence of alkali would give the alcohol [CCXVII] . This rnechanism for the hydroboration reaction clearly indicated that the stereochemistry at C-16 of the hydroboration product should be as shown in [CCXVII]. That is the Orientation of the lV-CH^OH group should be 3. It was therefore decided to prepare the 16~epimer of CCXVII. The orientation of the IT-CH^OH group would be a in this compound. The 16-epimer of CCXVII was prepared as shown in shceme. .42. UNIVERSITY OF IBADAN LIBRARY 201 Scheine 42 involved the protection of the. 2^3 diol System of CCXVII as the acetonide [CCXX] followed by the oxidation of the 17-hydroxyl to give the. aldehyde [CCXXI] . Hopefully the aldehyde could then be epimerised with base.. Reduction of the epimerised alxlehyde [CCXXII] and the removal of the protecting group of the 253-diol would give the 16-epimer [CCXXIII] of CCXVII. These reactions were done on small scale and were therefore followed with GC-MS analysis. SCHEME 42 (CCXVii) H Coliins reqgerrt l (CCXXO H UNIVERSITY OF IBADAN LIBRARY 202 Sehe ms 42 contd. The acetonide [CCXX] was prepared by dissolving the triol [CCXVII] in dry acetone containing about 0.001% conc.H^SO^ at room temperature for about 2*4 hours. The GC-MS analysis of the proauct indicated the presence of the acetonide as the main product. The mass spectrum had a molecular ion at m/e M-20. There were some other prominent ions at m/e 405 (M t -CH o), m/e 403 (M t-0H), m/e 402 (M+-H 0), m/e 388 s (M+-CH 0H), m/e 387 (M+-H 0,CH ), m/e 344Z ö Z O (M+-CH C0CH , HO), m/e 312 (M+-CH„C0CH„, CH 0H, H.O), m/e 284 o o Z o o o Z (M+-CH3C0CH3, CHgCOOH, H O), m/e 241 (M^-CHgCOC^, CH C0Ö, C02,H20) m/e 239 (M+-CH„C0CE , CH„C00, HC00H, HO), m/e 223 (base peak) ö ö ö Z (M+-CH C0CH , CH o C00H, HC00H, CH o , HO ) and m/e 43 (CH C0+). o ö Z o UNIVERSITY OF IBADAN LIBRARY 203 The acetoni.de linkage is widely used as a protecting group for 152-eis diols"^\ It is stable to alkaline conditions*^^ but easily cleaved with acid186 As the acetonide linkage (protecting group) is labile to acid5 the oxidation of the 17-hydroxyl group had to be effected with nonacidic reagent. A suitable nonacidic reagent which has been used extensively to oxidise primary alcohols to aldehydes is dipyridine-chromium[VI] oxide complex in dichloromethane (Collins reagent) 187 . The acetonide linkage has been reported to be stable to Collins reagent188 The alcohol [CCXI] was oxidised with the complex prepared in situ following the method of Ratcliffe 189 . Analysis of the product on GC-MS indicated the.presence of the aldehyde [CCXXI] as the major product. The mass spectrum of the aldehyde had a molecular ion at m/e 418 There were prominent ions at m/e 403 (M+-CH^)5 300 CH^CO, CH O C00H), 257 (M+~C0 , CH C0ÖH, CH C0CH ), 239 (.base peak) (M+-H 0,£ . 0 0 0 C0o, CH COOH, CH COCH o ) and m Je 43 CCH C0+).Z ö o o The next stage was the epimerization of the aldehyde (CHO) group. It was decided to do this in a basic medium since the acetonide protecting group would be stable to the basic condition. Epimerization of suitable substituents on the gibberellin skeleton UNIVERSITY OF IBADAN LIBRARY 204 have bec.n reported to occur in basic medium* The aldehyde in THF was treated with 5N aqueous NaOH. The product which was presumably [CCXXII] was reduced with sodium borohydride giving an alcohol as the major product, The mass spectrum of this alcohol did not show any significant differen.ee from the mass spectrum of [CCXX] . It was therefore not clear at this stage whether or not epimerization occurred. Removal of the acetonide protecting group in methanol containing a catalytic amount of p-toluene sulphonic acid gave an alcohol which was presumably [CCXXIII]. The mass spectrum (Fig. 26) of the TMSi ether of [CCXXIII] was similar to the mass spectrum of the TMSi ether of CCXVII. The relative intensities of some of the ions were however different. In these respects the mass spectrum of the TMSi ether of CCXXIII resembled the mass spectrum of the MeTMSi of the natural compound (gibberellin Y) more closely than the mass spectrum of the TMSi ether of CCXVII. The differences seemed to suggest that the epimerization of the ll-CH^OH group was achieved in CCXXIII. That is the structure of the triol [CCXXIII] was as shown. In order to be certain that CCXXIII was actually prepared, it was decided to attempt its synthesis via another scheme (scheme 43). The method was similar in principie to the inethod used by Croft et al 190 to prepare ent-Ta, 17^dihydroxykauran-19'-oic acid [CCXXIV] . UNIVERSITY OF IBADAN LIBRARY 204a 100t 8 ĉ : CHjOSi(CHj>3 50- CH3>'Z3S>'& 9cosct*' 581 0-f- r JiL 506 549iL -4- - M - 400 450 500 550 600 MASS CHARGE RATIO Fig- 26 (conto.) RELATIVE INTENSITV (%) UNIVERSITY OF IBADAN LIBRARY 205 SCHEME 43 OsO/4 i OH (CCXXV!) ( i ) Nq BH4 (ii) OH “ (ccxxni) "CHQ (CCXXVI!) UNIVERSITY OF IBADAN LIBRARY 206 The 253-diol System of MeA O H [CCy] was protected as the diacetate by dissolving the MeA O *4 in acetic anhydride containing a catalytic amount of p-toluene sulphonic acid at room temperature. Acetylation is a widely used method for protecting hydroxyl groups agamst reactions proceedmg under acidic conditions 191 5 l°~7 ~. The product of the acetylation [CCXXV] was treated without further analysis with osmium tetroxide followed by the hydrolysis of the resultant osmate ester to give the 16,17-diol [CCXXVI]. Osmylation reactions are known to proceed from the less hindered side of a molecule. It followed therefore that an attack from the less hindered a side of the ring D of [CCXXV] should result in the 16a5 17-dihydroxy compound [CCXXVI]. GC-MS analysis of the product (as the TMSi ether) of the osmylation reaction indicated that CCXXVI was the main product. The mass spectrum of the TMSi derivative had a molecular ion at m/e 624. Some other prominent ions were at m/e 609 (M -CH^), m/e 521 (M -CH^ OTMSi), m/e 403 (M+-CH OTMSi, 2CH C00), m/e 73 and 75 (TMSi). z. was o The next stage which/the dehydration of the 16., 17-diol System was effected by refluxing the diol in a mixture of 1-2% conc. in THF for about 30 mins., Analysis of the product on the GC-MS indicated the presence of the aldehyde [CCXXVIX] . UNIVERSITY OF IBADAN LIBRARY 207 The mass spectrum of the aldehyde had significant ions at m/e 460 tM+~2), 431 o 0), 420 tM+-CH CO), 402 ÜM+~CK C00H),2 ö 298 (M+-2CHoC00H5C0_ ), 238 (,M+-3CHo„ C00K,C0o) and m/e 43 (base peak)2 2 2 'tCH C0). The mechanism of the formation of the aldehyde from the diol might be' as shown in scheme 44 below. The acid induced elimination of the 16-hydroxyl group with the concerted rearrangement shown would lead to the formation of the aldehyde [CCXXVII]. (CCXXVt) (CCXXVII) UNIVERSITY OF IBADAN LIBRARY 208 According to the report by Croft et al 19 o the dehydration was accompanied by the epimerization of the aldehyde group in the acid medium. The epimerization was also assumed to have oceurred in this case since the same conditions were employed. Reduction of the aldehyde with sodium borohydride and the deacetylation of the 253~hydroxyl groups by treatment with 5% potassium methoxide gave an alcohol. The TMSi derivative of the alcohol was analysed on the GC-MS. The mass spectrum was identical to the mass spectrum of the TMSi of the alcohol [CCXXIII] obtained in scheme 42. This thus seemed to indicate that epimerisation of the 17-hydroxymethyl group was achieved through both schemes 42 and 43. Thus it followed that the structure [CCIIb] pcstulated for gibberellin Y was wrong. Gibberellin Y could also not have been just the 16~epimer of [CCIIb]. However the mass spectra of the TMSi ethers of the two synthetic compounds CCXVII and CCXXIII were quite similar to the mass spectrum of the MeTMSi of the natural compound. This suggested that their structures were closely related. On the basis of their mass spectra3 the MeTMSi of gibberellin Y appeared to be more structurally related to the TMSi of CCXXIII than the TMSI of CCXVII, It was therelor'e thought that in the natural compound (gibberellin Y) the 17-methylene hydroxyl group might have UNIVERSITY OF IBADAN LIBRARY 209 the a configuration as in CCXXII. The difference raight therefore be in the orientation of either or both of the 2,3-hydroxyl groups. 2. Partial Synthesis of 16-hydroxy MeA., It was thought ,that it was very unlikely that gibberellin Y was 16-hydroxy GA^. This was because the mass spectrum (Fig. 19, of the MeTMSi of gibberellin Y did not have a prominent ion at m/e 130. An intense ion at m/e 130 [XXIX] is characteristic of the mass spectra of the MeTMSi of 16-hydroxylated gibberellins. It was however decided to synthesize 16-hydroxy MeGA O H [CCXX] in order to compare the mass spectrum of its TMSi ether with that of the M3TMSi ether of gibberellin Y. . . . -/ 1 k, ' The exocyclic 16917~double bond of gibberellins which lack 13-hydroxylation is readily hydrated with dilute aqueous acid to form the 16~a hydroxyl derivative 31■*. This method has been used to synthesize GA^ [XVII] from GA^ [XIV]31. Treatment31 of MeA [CCV] with aqueous 3N HCl at room OHr temperature gave the 16-a hydroxy MeA oH [CCXXVIII] as the major product. UNIVERSITY OF IBADAN LIBRARY 210 SCHEME 45 (CCV) M <2 A 34 (CCXXVIII) The stereochemistry of the 16-hydroxyl group was assumed to be a because the reagent was expected to attack the double hond from the less hindered a face of ring D. Similar hydration of some other gibberellins that have been reported 31 5 193 gave the 16a~hydroxyl derivatives of these gibberellins. The mass spectrum of the TMSi ether of the 16a-hydroxyMeA had a molecular ion at m/e 596. There was the prominent ion (base peak) at m/e 130 indicating the presence of the 13~hydroxyl group. There were prominent ions at m/e 147 and 217 indicative of the 253~diol System. The mass spectrum was quite distinct from the mass spectrum of the.MeTMSi of gibberellin Y. Therefore the natural compound could not have been 16a~hydroxyGA . UNIVERSITY OF IBADAN LIBRARY 211 CONCLUSIONS The results of the coraparative analysis of scme of the hormones in the acidic ethyl acetate extracts obtained from 6~day old fruits (from the lowest raceme) and 2-day old fruits (from the upper racemes) showed 'that there was a disparity in the hormonal contents of these two groups of fruits. The 6-day old fruits had both growth promoting hormones (gibberellins and auxins) and growth inhibiting hormones (abscisic acid and its metabolites). Only the growth inhibiting hormones could however be^ identified in the 2-day old fruits. It can therefore be inferred that the abscission of the younger, upper fruits of the eowpea inflorescence ' ;-,f ' , is due to the lack of adequate concentration of growth promoting hormones in these fruits. Since they have no detectable amount of growth promoting hormones to counteract the effect of the growth inhibitors; their growth will be arrested and they will abscise. Also the older fruits at the lowest raceme will monopolise the available nutrients since they have an adequate concentration of growth promoting hormones which will induce the mobilization 8 of the available nutrients towards these fruits. The inducing influence which the maturing fruits at the first (lowest) raceme have on the abscission of the younger, upper fruits can therefore be partly due to the monopolization of the UNIVERSITY OF IBADAN LIBRARY 212 available nutrients by the oldest (lowest) fruits. This agrees with the observations of Adedipe and Ormrod 6 and Adedipe et al 7 that competition for nutrients is an important faetor in the abscission problem in cowpea. The inducing influence which the oldest (lowest), fruits have on the abscission of the younger, upper fruits can also be partly due to the abscising influence of hormones from the oldest, basal fruits which are translocated along the peduncle in the acropetal (apical) direction. This agrees with Ojehomon 2 !s hypothesis that the abscission of the flower buds and irnmature fruits may be stimulated by a substance or substances produced in the older fruits. The results cf the GOMS analysis of the crude acidic ethyl acetate fractions obtained from 6-day old seeds and 6-day old fruit walls clearly showed that most of the impurities that interfered with the analysis were from the fruit walls and not from the seeds. It therefore appears advisable that any further investigations of the hormonal content of cowpea fruits ought to be concentrated on the seeds despite the difficulties encountered in separating the tiny seeds from the fruit walls. UNIVERSITY OF IBADAN LIBRARY 213 EXPERIMENTAL General Melting points -were determined on a Koefler hot^stage apparatus and were not corrected. Solvents were redistilled prior to use. Petroleum ether refers to light petroleum b.p. 60-80°* The following Chromatographie materials were used for column chromatography: activated charcoal (BDH)5 Celite 545 9 (Johns- Manderville) and silica gel MFC (Hopkins and Williams). For preparative thin layer chromatography, plates were prepared with Kiesel gel G, HF or PF (Merck) and were pre-eluted with ethyl acetate. For analytical thin layer chromatography5 plates (0.25mm) were prepared with kiesel gel G (Merck). After development they were viewed in ultra-violet' light after spraying with five per cent sulphuric acid in ethanol and heating in the oven. Nuclear magnetic resonancelj^*^^) spectra were obtained with a Varian HA 100 spectrometer for deuterochloroform or deuteropyriaine Solutions with tetramethylsilane as internal Standard. absorptions are quoted in 6 units. Methylation was carried out by addition of excess ethereal diazornethane to Solutions of the compounds or extracts to be methylated in a minimum amount of dry methanoi* Trimethylsilyl (TMSi) derivatives of the methyl esters were prepared by adding a UNIVERSITY OF IBADAN LIBRARY 214 silylating reagent 19U- consisting of pyridine* hexamethyldisilizane and trimethylchlorosilane (.2:2:1* vjvjv) to Solutions of the previously dried methylated extracts or compounds in dry pyridine. In each case the resulting mixture was usually left Standing in a sealed tube at room temperature for about thirty minutes before analysis on the combined gas Chromatograph '— mass spectrometer (GCMS). In some cases the mixture in the sealed tube was warmed slightly with a hair- dryer for a few seconds. Gas chromatography and combined gas chromatography - mass spectrometry The methyl esters (Me) and methyl ester trimethyl silyl ethers (MeTMSi) were chromatographed on a Pye 104 gas Chromatograph using 2^o SE 33 or Gas chrom Q (80 - 100 mesh) packed in a glass column (130cm x 0.4cm i.d.). Column temperatures were usually programmed from 180° to 250° at if° per minute with a nitrogen flow rate of 60ml per minute. Gas chromatography - mass spectrometry (GCMS) analysis of the methyl esters and the MeTMSi derivatives was done on an A.E.I.- G.EC. MS 30 mass spectrometer coupled to a Pye 104 gas Chromatograph throu'gh a silicone membrane. The glass column (171cm x 0.2cm i.d.) for the GCMS was packed with 2% SE 33 Cor sometimes 2% QFI) on Gas Chrom Q 080-100 mesh) with a helium flow rate of 25ml per minute.. UNIVERSITY OF IBADAN LIBRARY 215 The mass spectra were obtained at 24- ev and were processed on-line by a Computer. Materials and extraction procedures The New Era cultivar of cowpea was used. The seeds that were j sown and harvested for extraction were obtained from the National Cereals Research Institute5 Moor Plantation5 Ibadan. The seeds were grown in the field at spacings 91 x 30cm (3 x 1 ft,) and this enabled the plants to grow as individuals. Flowers were labelled on the day they opened in order to determine the age of each fruit* The fruits were frozen in liquid nitrogen and then (unless otherwise stated) freeze dried. (a) Extraction of 2-day old fruits The fruits (40g. dry weight or 200g. fresh weight) were ground in a Waring blender in 80 per cent aqueous methanol (15cm 3|JLg.. dry weight of plant material). The mixture was left overnight in the cold room (ca.0°C) and was then filtered. The residue was mixed with fresh 80% aqueous methanol and left in the cold room for twenty-four hours before filtration. The extraction of the residue with 80% aqueous methanol was repeated twice (24 hours each time).. The combined filtrates (2.2 lites) were concentrated (ca..300cm 3) at below 40°C under reducea pressure in order to remove the methanol. UNIVERSITY OF IBADAN LIBRARY 216 The pR (_5.2) of the coneen.trated aqueons extract was ad just ed to between 8*0 and 8..1 with. 2M sodium hydroxide solution. The aqueous extract was then extracted with petroleum ether (3 x 100cm 3 ). The petroleum ether layer was discarded. The aqueous layer was extracted with ethyl acetate (3 x 100cm 3 ). The ethyl acetate layer was backwashed with water (100cm 3) and the water was added to the aqueous layer. The ethyl acetate layer was evaporated at below 40°C under reduced pressure giving the neutral ethyl acetate (NE) fraction as a gum (213mg). The pH of the aqueous layer was adjusted to 3.0 with 2M aqueous hydrochloric acid. The aqueous layer was then extracted with ethyl acetate (4 x 150 cm 3). The ethyl acetate layer was backwashed with water (150cm 3) and the water was added to the aqueous layer. The ethyl acetate layer was evaporated at below 40°C under reduced pressure to give the acidic ethyl acetate (AE) fraction as a gum (430mg). The aqueous layer was extracted with water-saturated (fwet?) n-butanol (3 x 150cm 3 ). The n~butanol layer was backwashed with water C 150 cm 3) and the wrater was added to the aqueous layer. The n-butanol layer was evaporated at below 40°C under reduced pressure to give the acidic n-butanol CAB) fraction as a brown gum (.83 8mg). The aqueous layer was discarded. UNIVERSITY OF IBADAN LIBRARY 217 The backwash of the yarious extracts with water was to remove residual acid or base. Cb) Extraction of 6-day old frults The fruits CllOg, dry weight or 660g fresh weight) were ground in a blender in 80 per cent aqueous methanol (15cm 3Jg dry weight of plant material). The mixture was left overnight in the cold room5 filtered and the residue extracted thrice more with 80 per cent aqueous methanol. The combined filtrates (ca. 6.5 litres) were concentrated (ca.600cm^) at below 40°C under reduced pressure. The pH (4.8) of the concentrated aqueous extract was adjusted to between 8.0 and 8.1 with 2M sodium hydroxide and the extract was washed with petroleum ether (3 x 200cm 3). The petroleum ether extract was discarded- The aqueous layer was then extracted with ethyl acetate (3 x 200cm 3 ). After backwashing with water (. 200cm 3)5 evaporation of the ethyl acetate layer under reduced pressure at below A0°C gave the neutral ethyl acetate (NE) fraction as a gum (526mg). The pH of the aqueous layer was adjusted to 3..0 and the. o aqueous layer was then extracted with ethyl acetate t4 x 250cm^). The ethyl acetate layer was evaporated in the usual way (after backwashing with water) to give the acidic ethyl acetate (AE) fraction as a gum (,700mg. ) UNIVERSITY OF IBADAN LIBRARY 218 The aqueous layer was extracted with 'wet’ n-butanol (3 x 250cm 3 ). After hackwashing with water (200cm 3), the n-butanol layer was evaporated to give the acidic n-butanol (AB) fraction as brown gum (1.4g). (c) Extractiob of 6-day old seeds Seeds were removed from 6-day old fruits. The seeds (21g. fresh weight) were ground in a blender in 80 per cent aqueous methanol. The extraction procedure was then carried out as described above for 6-day old fruits, giving an acidic ethyl acetate fraction (18mg) as a brown gum. (d) Extraction of 6-day old fruit walls The 6-day old fruit walls (34g fresh weight) were extracted by the same procedure described above for the extraction of 6-day old fruits giving an acidic ethyl acetate fraction (80mg) as a brown gum. (3) Another extraction43 >of 6-day old fruits (used for paper chromatography) The 6-day old fruits (5 kg. fresh weight) were ground in a mill and extracted thrice with 70 per cent aqueous methylated spirit. - 3 The combined extract (ca. 5.5 litres) was concentrated (ca*700cm ) under reduced pressure at below 4u°C. UNIVERSITY OF IBADAN LIBRARY 219 The pH (_5,0) of the concentrated aqueous extract was adjusted to between 7.3 and 7.74- with IM sodium carbonate solution.. The aqueous extract was then extracted with ethyl acetate (3 x 250cm 3). Evaporation of the ethyl acetate layer in vacuo gave the neutral ethyl acetate fraction, The pH of the aqueous layer was adjusted to 3.0 with 3M hydrochloric acid, and the aqueous layer was extracted with ethyl acetate (M x 250cm3"). The ethyl acetate layer was backwashed with water until the washings were neutral. The ethyl acetate layer was then concentrated (ca.300cm 3) and extracted with phosphate buffer 195 pH 6.3 (4 x , 1 0v0fc ■m 3). Evaporation of the ethyl acetate \ layer gave the weak acidic ethyl acetate fraction. The pH of the phosphate buffer layer was adjusted to 3.0 with 3M hydrochloric acid the phosphate buffer layer was then extracted with ethyl acetate (4 x 150cm 3). After backwashing with water, the ethyl acetate layer was evaporated jLn vacuo giving the strong acidic ethyl acetate-fraction as a gum (lg). Another modification of the procedure used for the extraction of 6~day old fruits was carried out as described below. Freeze-dried 6-day old fruits (,126g dry weight) were ground and tts<=Lr\ extracted in a soxhlet extractor with 80% aqueous methanol. The flask. conta.ining the aqueous methanol was immersed in a water' bath UNIVERSITY OF IBADAN LIBRARY 220 ma.intained at 35°C. In order to bring the aqueous methanol to reflux5 the condenser on top of the soxhlet extractor was connected to a vacuum pump«, The condenser was cooled with cold methylated spirit (ca. - 15°C) circulated from a cooler. The extraction was carried on unfil no more Chlorophyll was extracted. The aqueous methanol extract (1.5 litres) was concentrated (200'crn̂ ) at below 40°C under reduced pressure* in order to remove the methanol. The pH of the concentrated aqueous extract was adjusted to 7.5 with IM sodium carbonate solution. The aqueous extract was then extracted with petroleum ether using a liquid-liquid extractor. The petroleum ether layer was discarded. The extraction procedure was then continued as described above for the extraction of 6-day old fruits (used for paper chromatography) giving a strong acidic ethyl acetate fraction as a gurn (400mg). This fraction was used for some preliminary biological assays. (f) Extraction of fruits that were over 6 days old The fruits (23kg fresh weight) were fro/zen in liquid nitrogen, ground in a mill and extracted in the cold with 85 per cent aqueous methanol (3 x 10 litres). The combined extracts were concentrated (ca. 3 litres) under reduced pressure.. UNIVERSITY OF IBADAN LIBRARY 221 The pH. C5..6) of the aqueous extract was adjusted to 7.5 with 114 aqueous sodium carbonate solution 'and the aqueous extract was extracted with hexane. The pH of the aqueous extract was then adjus ted further to between 8.0 to 8.1 ana the aqueous layer was extracted thrice with ethyl acetate.. Evaporatiop of this ethyl acetate layer in vacuo gave the neutral ethyl acetate fraction. The pH of the aqueous layer was adjusted to 3.0 with 5M. hydrochloric acid. The aqueous layer was then extracted with ethyl acetate (.6 x 500cm 3 ). The ethyl acetate layer wTas backwashed with water (3 x 3Q0cm 3), and dried overnight over anhydrous sodium sulphate. The dried ethyl acetate layer was then evaporated in vacuo at below 140°C to give the acidic ethyl acetate (AE) fraction as a gummy solid (8g). . The aqueous layer wras discarded. Preparation of phosphate buffers. The phosphate buffers were prepared as described below. (a) Phosphate buffer pH 6.3 This was prepared igc~j by mi. xi. ng M/15 Na^HPO^ solution and M/15 KH^PO^ solution in a ratio of 1:4 (v/v) respectively. The■ M/15 Na^HPO^ solution was prepared by dissolvrng Q.M-ßUg of Na^HPO^ in 1000cm 3 of aistilled water.. UNIVERSITY OF IBADAN LIBRARY - - 222 The M-7ib KH^PO^ solution was prepared by dissolving 9.075g °f KH PO in. lOOOcm^ of distilled water. Cb) Phosphate buffer pH 8.0 This was prepared 19b as follows; 50cm° of 0.2M KH^PO^ solution was mixed with 46.8ml 0.2M NaOH solution and the mixture was diluted to 200cm 3 with distilled water.. Purification of extracts with PVP Aliquots of the plant extracts to be purified were taken into phosphate buffer (0.2M, pH 8.0) using about 1cm 3 of buffer per extract from 2g dry weight Cor 10g fresh weight) of plant material. Pre-washed polyvinylpyrrolidone (PVP or Polyclar AT) was added at concentrations ranging from 50mg to 100mg per cm 3 of phosphate buffer. Each mixture was then thöroughly shaken for thirty minutes. The PVP was filtered off under vacuum filtration and the residue washed thoroughly with aliquots of the phosphate buffer (6 x 2cm 3). Fresh PVP was added to each filtrate. The shaking and filtration procedures were repeated two more tirnes. The combined phosphate buffer filtrates were washed with Petroleum ether (3 x ™o ■volume of buffer). The petroleum ether layer was discarded.. 1 ‘The buffer layer was v/ashed with ethyl acetate (3 x 3 volume of buffer). UNIVERSITY OF IBADAN LIBRARY 223 The pH of the buff er layer was adjusted to pH 3*0 with 2M hydroehloric acid. The buffer layer -was extracted with ethyl aeetate (4 x ~p volume of buffer). The ethyl acetate layer was backwashed with water and evaporated in vavuo to give the purified extract. Column chromatography L 3 Ca) Column chromatography of the acidic ethyl acetate fraction obtained from 2~day old fruits Half (215mg) of the acidic ethyl acetate fraction that was obtained from 2-day old fruits was put on a column (15 x 2.5cm) of activated charcoal (16g) and celite (32g). < -v The column was eluted with a gradient of increasing concentration of acetone in water, obtained by connecting an aspirator of water (1.5 litres) to that of acetone (2 litres) 150cm 3 fractions were collected. The solvent composition of the eluant for each fraction was determined by refractometry. Selected column fractions (4 to 10) were purified with PVP, and then analysed on the GLC and the GC-MS as the methyl esters (Me) and the methyl ester trimethyl silyl ethers (MeTMSi). Cb) Column chromatography of the acidic ethyl acetate fraction obtained from 6-day old fruits Half C350mg) of the acidic ethyl acetate fraction that was UNIVERSITY OF IBADAN LIBRARY 224 obtained frora 6-day old frults was placed on a column C26 x 1.5cm) of activated charcoal C8g) and celite Cl6g). The column was subjected to a gradient elution of increasing concentration of acetone in water* Twenty 150cm 3 fractions viere collect ed.. Selected column fractions (.4 to 10) were purified with PVP, derivatized (Me, MeTMSi) and analysed on the. GLC and the GCMS. (c) Column chromatography of the acidic ethyl acetate fraction obtained from fruits that were over 6 days old The acidic ethyl acetate fraction (7*5g) was chromatographed on a column (54 x 4.5cm) of acti-vated charcoal (70g) and celite (140g). The column was eluted with an increasing gradient of acetone in water, obtained by connecting an aspirator of water (4 litres) to another aspirator containing acetone (5 litres). Forty fractions were collected. The first two were 500cm3 fractions.while the others were 200cm 3 fraction. Some fractions, which were selected on the basis of the results that were obtained for the analytical TLC and the Thalor bioassay carried out on the column fractions, were derivatized (Me, MeTMSi) aller purification with PYP. The deriyatized fractions were then analysed on the GLC and the. GC-MS. UNIVERSITY OF IBADAN LIBRARY 225 Thin-layer chromatography CTLC) of the acidlc ethyl acetate (AE) fractions. for wheat coleophtile segment bioassay Aliquots of the PVP purified AE fractions from 2-day and 6-day old fruits (equivalent to extract from 20g.. fresh weight of fruits) were each applied on a TLC plate (20 x 20cm) which was prepared with kiesel gel G and had been pre-developed in ethyl acetate and dried. Each plate was then developed in ethyl acetate: Chloroform: acetic acid (15; 5: 1, v/v/v) to a distance of 15cm. The Zone of development was divided into ten equal Rf zones. Silica gel from each Rf zone was eluted with f-wetT ethyl acetate. The silica gel from a zone below the base-line was also eluted with f-wetf ethyl acetate and the extract from this zone was used as the control. Half of the extract from each zone was tested for biological activity in the wheat coleoptile segment bioassay described below. Paper chromatography of the AE fraction (from 6-day old fruits) for the wheat coleoptile segment bioassay After purification with PVP, aliquot of the acidic ethyl acetate fraction (equivalent to extract from 5g.. fresh weight of fruits) obtained from 6-day old fruits was strip-loaded on a 10mm wide Strip of Whatraan No.. 1 filter paper.. The paper had been pre-run first in ethanol and then in a solvent mixture consisting of Isopropanol: ammonia (25%); water (10: 1:1, v/v/v), After the UNIVERSITY OF IBADAN LIBRARY 226 extract had been strip-loaded on the paper s the paper was equilibrated overnight with the solvent mixture (Isopropanol: ammonia; water) before it was developed in the same solvent mixture. The development was descending to a distance of about 18cm.. After drying5 the zone of development was cut into ten equal Segments and two Segments were also cut in the area below the base-line. The last two segrnents were used as Controls. The various Rf zones and the Controls were then tested for biological activity in the wheat coleoptile segment bioassay as described below. Bioassays (a) Wheat coleoptile segment bioassay Wheat grains (cultivar Kolibri) viere kept in running water for about 24 hours and then planted in Vermiculite in a dark chamber at room ternperature. After about three days5 coleoptiles which viere between 18 mm to 23 mm long were selected. A 10mm segment was cut at 2mm from the tip of each coleoptile. The selections and euttings were carried out in the dark room with a dirn yellow light. The 10mm coleoptile segrnents were left floating in distilled water in the dark for about two hours. The soaking in water was to reduce the concentration of the endogenous hormones of the coleoptile segrnents. UNIVERSITY OF IBADAN LIBRARY 227 For each assay, five coleoptile Segments were incubated with each paper chroraatogram F.f zone or with the aliquot of the extract from each TLC Rf zone in a specimen bottle containing 1.5cm 3 of a Puffer solution. The buff er solution that was used as the medium was prepared as follows: A stock solution containing 0.1M K HPO and 0.05M z. » citric acid monohydrate was prepared by dissolving 4.485g of K^HPO^ and 2.547g of citric acid monohydrate in 250cm 3 of distilled water. 2g of sucrose was added to 10cm 3 of the stock solution and the mixture was diluted to 100cm 3 with disti. lled water. 1.5cm 3 of this solution was used as the medium in each specimen bottle.. A tiny hole was punched in the centre of the plastic lid of each specimen bottle to ensure adequate supply of air to the coleoptile segments. The specimen bottles with their contents were rotated horizontally about their main axis in the dark for about 24 hours. The rotation was to prevent coleoptile segments curvatures caused by gravitational Stimulation. The lengths of the coleoptile segments were measured after 22 hours. Cb) Halo hioassay^ Barley CHordeum vulgarum 3 cultivar* White naked atlas) seeds were cut transversely into two. The embryo sections were discarded. UNIVERSITY OF IBADAN LIBRARY 228 The emhryoless Cendosperra) half-seads were soaked for twenty minutes In I% (v/v) sodiura hypochlorite in water. After rinsing thoroughly in sterile water, the haif^seeds were left to irabibe water in a beaker for about 20 hours. Meanwhile a mixture consisting of 3% agar and 0.25% soluble starch in water (w/v) was prepared. The mixture was autoclaved for about twenty minutes. Aliquots (1 cm 3 aqueous ethanol Solutions- equivalent to extract from 5g fresh weight of fruits) of the column fractions to be tested were put in 9cm petri dishes. 25cm 3 of the autoclaved agar mixture were added to each petri dish and the mixture was shaken thoroughly. The mixture was then left to set. The barley half-seeds were then placed on the.agar mixture (with the cut surface on the mixture) in the petri dishes. The petri dishes and their contents were left at about 20°C for two days. After the two-day incubation period, a potassium iodide-iodine solution (prepared by dissolving 2.0g of potassium iodide and 200mg of iodine in 100cm 3 of water) was added to the mixture in each petri dish. The presence of active gibberellins in the materials tested was indicated by the presence of a clear circle Chalo) around each half-seed. The diameter of each clear circle was measured as a quantitative estimation of the active gibberellins present. UNIVERSITY OF IBADAN LIBRARY 229 For each. test Controls were prepared consisting of 1 cm3 aqueous ethanol and 25cm 3 agar-starch mixture. Standard gihberellin consisting of 2f4g/cm 3 was also tested. Synthesis ent-10-Hydroxy-20-norgibberel 1-2 ;>16-diene~-7 ,19-dioic Acid 7-Methyl ester 19910-Lactone (2,3rdehydroMeGA 9, CCIV) A GA^/GA^ mixture (7:3; 300mg) was dissolved in a little methanol and methylated with ethereal diazomethane. The methylated mixture without further purification was dissolved in pyridine ( 7cm 3) and treated with phosphorus oxychloride (0.7cm 3). The reaction mixture was left at room temperature for 12 hours and then refluxed for 3 hours. On cooling, the reaction mixture was added to water slowly. The mixture was acidified (pH 3.5) with concentrated hydrochloric acid and it was then extracted with ethyl acetate. The ethyl acetate extract was backwa'shed. v;ith water and evaporated under reduced pressure giving a gummy product (2*+5mg). The gummy product vzas purified on preparative TLC (silica gel G, 0.9mm). After development (twice) in ethyl acetate- Petroleum ether (20; 80? v/v)5 silica gel from Rf zone 0.5 to 0.55 was eluted with ethyl acetate giving the olefin5 CCIV. as colourless UNIVERSITY OF IBADAN LIBRARY 230 crystals (_85rag), m.p. 125-130° (recrystallised from ethyl acetate- petroleum ether). <$ (CDCI3 t UMS) 1.23 (3H, 18-H O), 2.65 ClHj d, 5-H), 2.83 (1H, d, 6-H), 3.73 (3H, S, CO CH o), 4.88 andz 4.99 Ceach broad, 17-H } and 5.75 ( 2H, ra, 2-H and 3-H). m/e 328 (M+ , < 1%), 284 C40), 225 (14), 224 (100), 223 (21), 216 (16), 209 (28), 181 (32), 157 (26), 156 (42), 155 (28), 151 (30), 143 (23), 131 (21), 119 (25), 105 (62), 93 (42), 92 (31), 91 (39), 79 (28), and 44 (24). ent - 2a, 3a, 10, 16ß, 17-Pentahydroxy-20-norgibberellane-7,19-dioic Acid 7-Methyl ester 19,10-Lactone (16a, 17-dihydroxyMeGA^ , CCIII) 2,3-dehydrogibberellin A methyl ester, CCIV (50ing) was dissolved in pyridine (2cm 3 ) and the stirred solntion was treated with osmium tetroxide (100mg). The s.tirring was continued at room temperature for 2 days (when analytical TLC showed no starting material). The reaction inixture was stirred and treated slowly with a solution of sodium metabisulphite (400mg) in water (6 cm 3) and pyridine (4.5 cm 3 ). The mixture was stirred for a further thirty' minutes. The mixture was neutralised with concentrated hydrochloric acid and extracted with ethyl acetate. The ethyl acetate extract was backwashed with water and then evaporated in vacuo to give a solid (40mg). UNIVERSITY OF IBADAN LIBRARY 231 The solid was purified on preparatiye TLC (.silica gel G, 0.4mm), deyeloped twi.ce in ethyl acetate-chloroform-acetic acid (.15: 5: 1, v/v/v). Silica gel from Rf Zone 0.15 to 0.25 was eluted with ethyl acetate-methanol (.9:1, v/v) giving the tetrol, CCIII as a solid (28mg). It was recrystallized from ethyl acetate, m.p. 130-140°. 6 (C 5 D 5 N + TMS) 1.49 (3H, —S, 18-H„o ), 3.05 (1H, —d, 6-H), 3.60 (3H, —S5 C02CH3), 3.78 (1H, d, 5-H), 3.97 (2H, d, 17-H ),. 4.13 (1H, d, 3-H), 4.32 (1H, m, 2-H). The TMSi derivative was prepared and the mass spectrum (Fig. 24) was run. m/e 684 (M+, 7%), 583 (23), 582 (48), 581 (100), 218 (4), 217 (9), 147 (14), 143 (4), 129 (9), 103 (3), 75 (22) and 73 (30). ent-3a > 10-Dihydroxy~16--0xo-17,20-bisnorgibbereliane~7,19-dioic Acid 19,10 Lactone (GA^ norketone, CCVI) A stirred ice-cold solution of GA /GA mixture (7:3? 3g) and osmium tetroxide (20mg) in tetrahydrofuran (THF* 45cm^3) and water 45cm was prepared. Powdered sodium metaperiodate (3*7bg) was added slowly,. (20 minutes) in small portions5 to the stirred solution The mixture was allowed to warm to roora temperature and the stirring was continued for 24 hours. The mixture was concentrated in vacuo to remove the THF. It was then diluted with more water and acidified (pH 3.0) with dilute hyarochloric acid. It was extracted with ethyl acetate and the ethyl acetate layer was backwashed with UNIVERSITY OF IBADAN LIBRARY 232 water to remove the residual hydrochloric acid., Evaporation of the ethyl acetate layer gaye the crude. gibberellin A norketone CCCVI.5 2.0g). This was recrystallised frorn ethyl acetate-petroleum ether m.p. 120-136° (Lit.^^ m.p. 120-140°). ent-10-Hydroxy--16-0xo-17 ?20-bisnorgibberell-2-ene-7 ,19-dioic Acid 7-Methyl ester 19— , 10-—L actone (2 ,3-d.ehydro—G A^9 ” norketo—ne methyl ester CCVII) The gibberellin A^ norketone, CCVI (1.8g) was dissolved in methanol (5 cm 3) and methylated with excess ethereal diazomethane. Evaporation of the mixture gave the methyl ester of gibberellin norketone. The gibberellin A^ norketone methyl ester, without further purification was dissolved in pyridine (45 cm 3) and treated with phosphorus oxychloride (4.5cm 3). The mixture x̂ as left Standing at roorn temperature overnight and was then refluxed for about 3 hours. The reaction mixture, on cooling, was added slowly to water with stirring and then acidified with concentrated hydrochloric acid to pH 4.0. The mixture was extracted with ethyl acetate and the ethyl acetate layer was backwashed with water. Evaporation of the ethyl acetate la3̂ er in vacuo gave a solid Cl. 7g). The solid was purified by column chromatography on silica gel. Elution with 30% ethyl acetate in petroleum ether gave the 2,3-dehydroGA.g norketone methyl ester CCCVII). This was recrystallised UNIVERSITY OF IBADAN LIBRARY 233 in 50%. ethyl acetate: petroleum ether (v/v) to give coiourless crystals (700mg) m.p. 158-160° (Literature^0 m.p. 158-159°, 160-161°). 6 (CDC1 + TMS) 1.23 (3H, S, 18-H ), 2.5 (1H, d, 5-H), o • o 2.68 (1H, d, 6-H), 3.-73 (3H, d, CO^ CH O), 5.6 to 5.9 (2H, m, 2-H andZ. 3-H). This was consistent with the rm-r- spectrum reported in literature 176 for the compound. m/e 330 (M+ , < 1%), 286 (34), 227 (44), 226 (100), 183 (35), 105 (32), and 95 (33). ent-2a, 3a,10-Trihydroxy-l6-0xo-17,2Q-bisnorgibberellane-7,19-dioic Acid 7-Methyl ester 19,10-Lactone (MeA oH norketone, CCVIII) 3* The 2,3-dehydroGAg norketone methyl ester, CCVII (500mg) was dissolved in pyridine (20cm 3). The stirred solution was treated with osmium tetroxide (500mg). The stirring was continued at room temperature for two days. The reaction mixture was stirred and treated slowly with a solution of sodium metabisulphite (Hg), water (60cm 3 ) and pyridine (H5cm 3 ). The mixture was stirred for a further thirty minutes after the addition. The mixture was neutralised with eoncentrated hydrochloric acid and extracted with ethyl acetate. The ethyl acetate layer was backwashed with water and then .̂viâ o^ra cVeji dto give a white solid (H82mg). UNIVERSITY OF IBADAN LIBRARY 234 The solid was recrystallisea from methanol to give the MeA - Ö * T norketone, CCVIII (,350mg). M.p. 113-117° (Litt3 m.p. ca. 115-120°). It was then converted to the TMSi ether and its mass spectrum run. m/e 503 (40%), 508 (M*, 100), 218 (25), 217 (45), 147 (35), 75 (23) and 73 (49). ent-2a , 3a >10-Trihydroxy-20-norgibberell-16-ene--7,19-dioic Acid 7-Methyl ester 19,10-Lactone (MeA,O H"> CCV) The trimethylsilyl ether of MeGA H norketone5 CCIX was preparedö by adding a silylating reagent (500yl) to a solution of the methyl ester of GA norketone (100mg) in dry pyridine (lOOyl) in a tightly covered screw-cap pressure vial. The silylating reagent was made up of hexamethyldisilazane, trimethylchlorosilane and pyridine (2: 1: 25 v/v/v). The pressure vial and its contents were warmed slightly with a hair dryer for a few seconds and then left Standing at room temperature for about 1 hour. The mixture was evaporated and extracted with dry ethyl acetate. The ethyl acetate extract was centrifuged and the supernantant was evaporated to give the * trimethylsilyl ether of the methyl ester of gibberellin A norketones CCIX as a solid. This was dried thoroughly under vacuum. m/e 509 C40%), 508 (M+ , 100), 243 (.11), 225 Cll), 218 (.26), 217 (45), 147 (.36), 75 (23) and 73 (49). UNIVERSITY OF IBADAN LIBRARY 235 Triphenylphosphine (11g) was dlssolved in dry benzene C9 cm 3) and the resulting solution was cooled in an ice-salt mixture.. To the cooled solution in a screw-cap bottle was added pre~cooled methyl bromide (3.23cm 3 = 5.6g). The bottle was covered tightly and left to stand at room temperature for 24 hours. The resulting white solid was collected by suction and washed with hot benzene (ca.200cm 3) to give methyltriphenyl phosphonium bromide (14.9g). This was dried at 100° for 24 hours. M.p.. 233 - 234°. (Lit.^ ̂ m.p. 232 - 233°). Methyltriphenyl phosphonium bromide (3g) was suspended in 30cm 3 of dry peroxide-free tetrahydrofuran. Sodium hydride (400mg) was added to the mixture with stirring under a flow of nitrogen. The mixture was heated until a yellowish green colour was obtained. Stirring was continued for 24 hours under a flow of nitrogen at room temperature. After 24 hours, the stirring was stopped and the sodium bromide was allowed to settle. Aliquot (4cm 3) of the supernantant (methylenetriphenyl phosphorane in THF) was added to a solution of the trimethylsilyl ether of the methyl ester of GA^ norketone (prepared from 100mg of MeGA^ norketone as described above) in a small arnount of dry tetrahydrofuran. The addition was carried out with stirring under nitrogen and the stirring was continued for 24 hours after the addition. UNIVERSITY OF IBADAN LIBRARY 236 The produet was evaporated under nitrogen and the resulting crude trirnethylsilyl ether of the methyl ester of GA O H was hydrolysed with acetic acid-methanol (4;l), (.5cm 3) at room temperature for about 12 hours (TLC nonftoring). The resulting mixture was eyaporated in vacuo after the addition of water and toluene (to remove the acetic acid azeotropically). The residual gum was purified on preparative TLC on kiesel gel with ethyl acetate Petroleum ether (15: 5, v/v) giving gibberellin A O M-methyl ester, CCV, as a gum (25mg). 6(CDC1 O + TMS) 1.18 (3H, S, 18-H o), 3.02 (1H, d, 6-H), 3.25 (1H, d, 5-H), 3.70 (3H, S, CO CH o) 5 H.86 andz. 4.96 (17-H2). The TMSi derivative was prepared and the mass spectrum run. m/e 507 (43%), 506 (M+ , 100), 241 (11), 233 (10), 229 (14), 223 (17) 218 (15), 217 (30), 210 (10), 147 (31), 75 (35) and 73 (58). Woodward reaction 177 on gibberellin A^ ̂methyl ester Silver acetate (H.67mg5 0,28mM) was added to a stirrea solution of gibberellin A methyl ester5 CCV (5mg, O.lUmM) in acetic acid (lern 3 ). Finely powdered iodine (3.55mg? 0.14mM) vias added in small portions to the vigorously stirred mixtured for about thirty minutes Aqueous glacial acetic acid (lOOyl) was then added. (The aqueous glacial acetic acid was prepared by diluting 20yl of water to SOOpl with glacial acetic acid.) The reaction mixture was heated.at UNIVERSITY OF IBADAN LIBRARY 237 90-95 C with. stirring for 3 hours,. The mixture was allowed to cool, diluted with more water, and extracted with ethyl acetate. The ethyl acetate layer was backwashed with a little water and evaporated to give a gum (5.4mg). The gummy product (4mg) obtained above was hydrolysed by treating a solution of the gum in a little methanol with 5% potassium hydroxide in methanol overnight. The mixture was neutralised with dilute hydrochloric acid, more water was added, and the mixture was concentrated. The concentrated mixture was extracted with ethyl acetate and the ethyl acetate layer was evaporated (after backwashing with water) to give a gum (2.5mg) GC-MS analysis of the TMSi derivative of the gum revealed that it was a mixture of several compounds. The mass spectra of three of the GLC peaks had the expected molecular ion at m/e 684. m/e 684 (M+, 9%), 583 (21), 582 (45), 581 (100), 491 (7), 490 (13), 217 (8), 147 (18), 143 (6), 129 (5), 75 (10), and 73 (25). m/e 684 (M+ , 7%), 596 (11), 595 (16), 594 (36), 583 (20), 582 (40), 581 (100), 540 (13), 491 (13), 490 (36), 218 (12), 217 (28), 156 (27), 147 (39), 143 (13), 129 (9), 103 (4), 75 (27) and 73 (72). m/e 684 (M+, 7%), 634 (13), 633 (.27), 632 (65), 594 (33), 582 G34), 581 (66), 540 (28), 522 (19), 491 (19), 490 (27), 456(13), 287 (13), 218 (25), 217 (57), 156 (33), 147 (61), 143(21), 129(19), 103(9), 75(36), and 73(100). UNIVERSITY OF IBADAN LIBRARY 238 ent~2a , 3a , 10,17-Tetrahydroxy-20-norgibberellane-7,19-dioic Acid 7-Methyl ester 19, IQ~Lactone (17~hydroxyGA o H methyl ester, C’CXVII Methyl ester of gibberellin A , CCV (5mg) was dissolved in a öhr \%1- little dry THE and was then treated with 0.5M' diborane in THF (ca. 5cm 3;a stream of nitrogen gas was passed through before the addition). The reaction mixture was left Standing for 2 hours when analytical TLC showed no Starting material. Aqueous 5% sodium hydroxide (Ga. 5cm 3 ) and hydrogen peroxide (30% w/v, 2.5cm 3) were added and the resulting mixture was left Standing for Ihour at room temperature. More water (10cm 3 ) was added and the mixture was extrac- ted with ethyl acetate, The ethyl acetate layer was backwashed with water and evaporated in vacuo to give a gum (5mg). GLC analysis indicated only one product. The TMSi derivative was prepared and the mass spectrum (Fig. 25) run. m/e 598 (21%)5 597 (40), 595 (M+ , 100), 581 (1), 462 (5), 301 (6), 289 (10), 288(9), 284 (9)', 261 (8), 245 (3), 241 (4), 233 (6), 229 (7), 223(7), 219 (5), 218 (10), 217 (17), 211 (6), 201 (5), 181 (6), 147 (19), 143 (5), 129 (6), 103 (12), 75 (20) and 73 (38). Epimerisation of the hydroboration product 1. . Ist Method (a) Acetonisation The 17-hydroxyGA methyl ester, CCXVII (2mg) (obtained from the UNIVERSITY OF IBADAN LIBRARY 239 hydroboration of MeGA ) was acetonated by dissolving in dry acetone Cl cm 3) containing about 0.001% concentrated sulphuric acid. The mixture was left for about 24 hours (reaction followed by TLC). A small amount of sodium bicarbonate -was added to aestroy the sulphuric acid and the mixture was evaporated. A little water was added and the mixture was extracted with ethyl acetate. The ethyl acetate layer was backwashed with a little water and then evaporated to give a gum. Analysis of the gum on the GC-MS showed that it was mainly the 2,3-acetonide of 17-hydroxy gibberellin A Ot methyl ester, CCXX (about 95% by GLC estimation'!) m/e 420 (M+5 6%), 406 (19), 405 (66)*'403 (28), 402 (26), 388 (31), 387 (72), 344(16), 312 (35), 283 (28), 256 (28), 241 (35), 239 (56), 228 (28), 233 (100), 180 (25), 179 (31), 171 (28), 155 (22), 95 (16), 71 (19), 57 (26), 55 (26) and 43.(38). (b) Oxidation of the acetonide The acetonide, CCXX obtained above, without further purif ication. was oxidised with chromium trioxide-pyridine complex prepared“^ in situ as described below. A mixture of dry dichloromethane (250yl) and dry pyridine (lOyl - 0.12mM) was cooled with stirri.ng to 10°C. Chromium trioxide C6mg - 0.06mM) was added to the mixture once. The stirring was continued at 10°C for five minutes and the mixture was then allowed UNIVERSITY OF IBADAN LIBRARY 240 to warm slowly (with s.tirring) to 20°C over about 60 minutes. The crude acetonlde. (4mg ~ 0.01mM)in dry dichloromethane C50pl) was added to the complex with stirring.. The reaction was allowed to go on for about 15 iriinutes. The organic layer was decanted, 5% aqueous sodium bicarbonate was added to it and it was extracted with ethyl acetate. The ethĵ l acetate layer was backwashed with water and evaporated to give a gummy product (3.5mg). Analysis on the GC-MS revealed that the aldehyde, CCXXI was the major product (Ca.80% by GLC estimation). m/e 418 (M+, 9%), 404 (21), 403 (92), 244 (11), 240 (20), 239 (100), 183 (10), and 43 (15). (c) Epimerisation of the aldehyde [CCXXI] and the rcduction of the resulting epimer The crude aldehyde, CCXXI (2mg) was dissolved. in tetr^hydrofuran (50pl). 5M aqueous sodium hydroxide (1 cm 3 ) was added and the mixture was stirred at room temperature overnight. The mixture was acidified and extracted with etlyl acetate. The ethyl acetate layer was evaporated, after backwashing with water to give a gum (ca.2mg), GLC analysis showed that there was one major product (probably CCXXII). The gummy product (ca. 2mg) was dissolved in ethanol (1.5 ein 3) and sodium borohydride (4mg) was added. The mixture was stirred overnight at room temperature. The mixture was neutralised with UNIVERSITY OF IBADAN LIBRARY 241 dilute byarochloric acid and extracted with ethyl acetate. The ethyl acetate extract was evaporated (after backwashing with water) to give a gurnrny product (ca. 2mg). GLC analysis indicated only one major (ca. 70%c)o^mwpiotnhe nat longer GLC retention time than the aldehyde? CCXXI. Cd) Deacetonation The gummy product (ca, 2mg) obtained above was dissolved in methanol’d cm 3) and a catalytic amount of p-toluenesulphonic acid was added. The mixture was left overnight at room temperature. Ethereal diazomethane was added to destroy the p~toluenesuiphonic acid and the mixture was evaporated to give a gum (ca. 2mg). Analysis of the gum on the GC-MS as the trimethylsilyl ether indicated that the alcohol* CCXXIII was the major product. rn/e (of TMSi derivative of CCXXIII5 Fig. 26) 598 (23%). 597 (49), 596 (M+ , 100), 581 (9), 462 (6), 301 (6), 289 (9), 288 (16), 284 (8), 261 (6), 245 (13), 241 (4), 233(8), 231 (9), 229 (9), 223 (9), 219 (6), 218 (12), 217 (23), 211 (6), 201 (5), 181(7) 147 (22), 143 (8), 129 (6), 103 (16), 75 (23) and 73 (51), 2. 2nd method (a) Acetylation of gibberellin methyl ester. CCV Gibberellin A Ot methyl ester.CCV (2mg) was dissolved in acetic anhydride (1 cm 3) and a catalytic amount of p-toluenesulphon UNIVERSITY OF IBADAN LIBRARY 242 acid was added. The mixture was left overnight at rooro temperature after which analytical TLC showed no starting material, Ethereal diazomethane was added to destroy the excess reagent arid the mixture was evaporated to give a gum (ca. 2mg). (b) Hydroxylation of the 2,3-diacetate of MeGA„O,t, , “CCXXV The crude 2,3-diacetate of gibberellin A methyl ester, CCXXV 0*4 obtained above was dissolved in tetrahydrofuran (1 cm 3) and osmium Tetroxide (ca, 2mg) was added, The mixture was stirred overnight. ns The resulting osmate ester was cleaved in the usual way (sodium metabisulphite 400mg, water 500yl, and pyridine 600yl). On work-up a gummy product (2.8mg) was obtained. Analysis of the product on GC-MS as the trimethylsilyl ether showed that it was mainly (over 95%) the 2 ,3-diacetate of 16 ,17-dihydroxy gibberellin A ^ methyl ester [CCXXVI] and a little rnonoacetate. m/e (of the TMSi derivative of CCXXVI) 624 (M+ , 1%), 609 (1), 523 (10), 522 (37), 521 (100), 217 (4), 147 (9), 143 (3), 129 (3), 75 (7), 73 (18) and 43 (6). (c) Dehydration of the 2,3~diacetate of 16,17-dihydroxy MeGA^ The crude 2,3-diacetate of 16,17~dihydroxy gibberellin A ^ methyl ester [CCXXVI] was dissolved in tetrahydrofuran (1 cm 3) containing concentrated sulphuric acid (1-2%). The mixture was refluxed gently for 20 mimrtes, cooled, neutralised wlth 5% aqueous UNIVERSITY OF IBADAN LIBRARY 243 sodium blcarbonate solutlon and extracted with ethyl acetate.. The ethyl acetate layer was evaporated after backwashing with a little water. The gummy product (ca,lmg) was’analys'ed on the GC-MS. The analysis indieated that the aldehyde [CCXXVII] was the major (ca. 60%) product. m/e (of CCXXVII) 462 (M+ , < 1%), 431 (4), 420 (8), 402 (5), 360 (5), 346 (8), 304 (14), 303 (12), 298 (34), 286 (14), 240 (19), 239 (55), 238 (94), 183 (19), 182 (40), 181 (22), 179 (14), 155 (14), 115 (16), 73 (18), 71 (35), 55 (16) and 43 (100). (d) Reduction and deacetylation The gummy product (ca. lmg) obtained above was taken into ethanbl (1cm 3) and sodium borohydride (2mg) was added. The mixture was stirred overnighi. Acetone (2cm 3) was added to destroy the excess sodium borohydride. Water (1cm 3 ) was added and the mixture was concentrated (ca. 0.5cm 3). The concentrated mixture was extracted with ethyl acetate. The ethyl acetate layer was backwashed with water and then evaporated to give a gum (ca.lmg). The gum (ca.lmg) was deacetylated by treating its solution in a little methanol (ca.250yl) with 5% potassium methoxide (da. 750yl) for about 3 hours. The mixture was neutralized with dilute hydrochloric acid. Water (1.5cm 3) was added and the mixture was concentrated (ca. 1 cm 3). The concentrated mixture was extracted with ethyl acetate. The ethyl acetate layer was evaporated (after UNIVERSITY OF IBADAN LIBRARY ~ 24*4 ~ backwashingwithwater) to giye a gummy product (< 1mg). The product was analysed on the GC-MS as the trimetylsilyl derivative, The mass spectrurn of the major product was identical with the mass spectrum (Fig. 26) of the trimethylsilyl derivative of the alcohol5 CCXXIII. ent-2a? 3a? 10, 16ß-Tetrahydroxy-2Q~norgibbereIlane~7 ,19-dloic Acid ~ 7-Methyl ester 19, 10-Lactone (16a-hydroxyMeGAo0)hi , CCXXVIII) 3M dilute hydrochloric acid (600yl) was added to a solution of gibberellin A0î methyl ester, CCV (1 mg) in methahol (150yl). \cl?> The mixture was left at room temperature for two days (rnonitored with TLC). Water (1 cm 3) was added to the mixture and the mixture was extracted with ethyl acetate. The ethyl acetate layer was backwashed with water, and evaporated giving a gummy product (1mg), The product was analysed on the GC-MS as the trimethylsilyl ether derivative. The GC-MS analysis indicated that the 16a~hydroxy * gibberellin A 0 *4 methyl ester [CCXXVIII] was the major (over 90%) product. m/e (of TMSi derivative of CCXXVIII) 597 (44%.), 596 (M+ ,92), 581 (11), 506 (15), 289 (13), 288 (15), 223 (10), 218 (13), 217 (18), 157 (15), 156 (13), 147 (27), 143 (23), 131 (28), 130 (100), 129 (8), 117 (13), 115 (15), 75 (61) and 73 (90). UNIVERSITY OF IBADAN LIBRARY - 2*5 “ A P P E N D I X HALO BIOASSAY ON THE ACIDIC ETHYL ACETATE (AE) EXTRACTS OBTAINED FROH 6-DAY OLD AND 2-DAY OLD ADZUKI FRUITS Introduction The results that were obtained in the analysis of some of the hormones in the acidic ethyl acetate (AE) extracts obtained from 2-day old and 6-day old New Era fruits showed that the 2-day old fruits were deficient in growth promoting hormones. This indicated that the abscission of the younger fruits might be due to the lack of adequate concentration of growth promoting hormones in these fruits. The New Era cultivar of cowpea was known 2 5 6 to exhibit a high degree of abscission of flower buds, flowers and immature fruits. It was decided to carry out a preliminary study of the hormonal content of another cultivar of cowpea that exhibited a relativelv low degree of the abscission probiem.. The Adzuki cultivar of cowpea haa been reported 2 5 6 to exhibit a lower degree of the abscission probiem than New Era. The Adzuki cultivar was therefore used. UNIVERSITY OF IBADAN LIBRARY 2 IVfe - Extracts were obtained from two groupsof fruits. These were 6-day old fruits from the lowest raceme and 2-day öld fruits from the upper racemes. Results and Discussion (a) Halo bioassay on TLC Rf zones of the AE fraction from 6-day old fruits. The results of the halo bioassay carried out on the extracts of the ten equal TLC Rf zones of the PVP purified AE fraction from 6-day old Adzuki fruits are shown in Table 8 below. Gibberellin activity was obtained at Rf zones 0*3 ~ 0.4, 0.4 - 0.5 and 0.5 - 0.6. UNIVERSITY OF IBADAN LIBRARY - 2 in - TABLE 8 Results obtained in the halo bioassay on TLC Rf zones of the AE extract from 6-day old Adzuki fruits. TLC Diameter of halo (in mm) for each half seed Rf Zones i 2 3 4 5 6 7 8 9 10 TOTAL MEAN 0.0 - 0.1 0 0 0 0 0 0 0 0 0 0 0 0 0.1 - 0.2 0 0 0 0 0 0 0 0 0 0 0 0 0.2 - 0.3 0 0 0 0 0 0 0 0 0 0 0 0 0.3 - 0.4 8 8 10 10 8 9 8 8 7 9 85 9 + i 0.4 - 0.5 11 10 • 11 12 10 10 12 9 10 10 105 H + i 0.5 - 0.6 8 9 10 10 10 12 12 13 12 10 106 H + i 0.6 - 0.7 0 0 0 0 0 0 0 0 0 0 0 0 0.7 - 0.8 0 0 0 0 0 0 0 0 0 0 0 0 0.8 - 0.9 0 0 0 0 0 0 . 0 0 0 0 0 0 0.9 - 1.0 0 0 0 0 0 0 0 0 0 0 0 0 CONTROL 0 0 0 0 0 0 0 0 0 0 0 0 Standard 1 1 GA3 3 10 10 12 11 10 9 10 10 9 10 101 10 + 1 0.5 Pg/cm UNIVERSITY OF IBADAN LIBRARY (b) Halo bioassay on TLC Rf zones of the AE fraction from 2-day old fruits* The results of the halo bioassay carried out on the extracts of ten equal TLC Rf zones of the PVP purified AE extract from 2-day old Adzuki fruits are shöwn in Table 9 below.. UNIVERSITY OF IBADAN LIBRARY TABLE 9 Results obtained in the halo bioassay on TLC Rf zones of AE extract from 2-day old Adzuki fruits. TLC Diameter of the halo for each half-seed Rf (in mm) Zones i 2 3 4 5 6 7 8 9 10 TOTAL MEAN 0.0 - 0.1 0 0 0 0 0 0 0 0 0 0 0 0 0.1 - 0.2 0 0 0 0 0 0 0 0 0 0 0 0 0.2 - 0.3 0 0 0 0 0 0 0 0 0 0 0 0 0.3 - 0.4 0 0 0 0 0 ö 0 0 0 0 0 ^ 0 0.4 - 0.5 4 5 6 6 5 5 5 6 5 5 52 5 + 1 0.5 - 0.6 4 5 5 . 6 6 4'" 5 5 5 5 50 5 + 1 0.6 - 0.7 0 0 0 0 0 0 0 0 0 0 0 0 0.7 - 0.8 C 0 0 0 0 0 0 0 0 0 0 0 0.8 - 0.9 0 0 0 0 0 0 0 0 0 0 0 0 0.9 - 1.0 0 0 0 0 0 0 ■ 0 0 0 0 0 0 CONTROL 0 0 •0 0 0 0 0 0 0 0 0 0 Standard GA, 10 10 12 11 10 Q 10 10 2 10 101 0.5ygö /cm3 UNIVERSITY OF IBADAN LIBRARY * 1 tl or~| 250 Comparison of the results obtained for Adzuki fruits with those obtained for New Era fruits The results obtained for Adzuki fruits differed from the results obtained for New Era fruits in that no gibberellin could be detected in 2-day old New Era fruits. • Thus the 2-day old fruits of Adzuki (a cultivar that exhibits a relatively low degree of the abscission problem) had a detectable amount of gibberellins, whereas the 2-day old fruits of New Era (a cultivar that exhibits a relatively high degree of the abscission problem) had no detectable amount of gibberellins. This preliminary results therefore, support the proposition that the abscission of the immature fruits might be^due to lack of adequate concentration of growth promqting hormones. Experimental The 6-day old Adzuki fruits (50g fresh weight) were extracted according to the procedure used for 6-day old New Era fruits (page 217) giving the acidic ethyl acetate fraction (43 mg). The 2-day old Adzuki fruits (50g fresh weight) were extracted according to the procedure used for 2-day old New Era fruits (page 215) giving the acidic ethyl acetate fraction (41mg). UNIVERSITY OF IBADAN LIBRARY 25\ The TLC of the PVP purified extracts was carried out as described for the extracts from New Era fruits Cpage 225) and the halo bioassay was carried out as described in page 227. UNIVERSITY OF IBADAN LIBRARY REFERENCES Ojehomon, 0.0. 1968., The development of the inflorescence and extrafloral nectaries of Vigna unguiculata (LO Walp. J.W, Afr. Sei., Ass.; 13 * 93 - 110. Ojehomon, 0 .0. 1968. Flowering, fruit production and abscission in Cowpea, Vigna unguiculata (L.) Walp. J.W. Afri. Sei. Ass.$ 13, 227 - 234. Ojehomon, 0.0. 1969. A comparison of the vegetative growth, development and seed yield of three varieties of Cowpea, Vigna unguiculata (L.) Walp. J. Agric. Sei.; Camb.; 74, 363 - 374. Ojehomon, 0.0. 1969. Effect of continucus removal of open flowers on the seed yield of two varieties of Cox^pea, Vigna unguiculata (L.) Walp. J. Agric. Sei.; Camb., 74, 375 - 381. Ojehomon, 0.0. 1972. Fruit abscission in Cowpea, Vigna unguiculata (L.) Walp. I. Distribution of 14C-assimilates in ‘the inflorescence, and comparative growth of ovaries from persisting and abscising open flowers. J . Exp . Bot; 23, (76) 751 - 761. Adedipe, N.O. and Ormrod D.P. 1975. Absorption of foliar-applied 32P by successive leaves, and distribution patterns in relation to early fruiting and abscission in the Cowpea UNIVERSITY OF IBADAN LIBRARY - 2sa> - ’Cyi-gna unguiculata CL-). Ann. Bot-, 39, 639 - 646... Adedipe, N.O., Fletcher, R.A,, and Ormrod, D.P. 1976. Distribution of assimilates in the Cowpea (Vigna unguiculata (L.) in relation to fruit abscission and treatment with Benzyladenine. Ann. Bot., 40, 731 - 737 8 . Addicott, F.T. 1970. Plant hormones in the controi of abscission. Biol. Rev., 45, 485 - 524., 9. Seith, A.K., and Wareimg, P.F. 1967, Hormone-directed transport of metabolites and its role in plant senescence. J . Exp. Bot., 18, 65 - 77. 10a. Aba El-Soad, I.A. and Mansour, M.A. 1975. Physiological - 'i \ - response of Cowpea plants to gibberellic acid II. GAoo as foliar spray. Ist Arab Weed Control Conference 1975, 132 - 144. 10b. MaciMillan, J. 1971. Diterpenes - The Gibberellins • Aspects of Terpenoid Chemistry and Biochemistry. (Goodwin, T.W., ed.) Acadeirdc Press Inc. (London) Ltd. 10c. Yabuta, T. and \c\'3>c\- c.cx\ v\c^c\v\s> v \ c j e _ W * 4 - < ^ W \ *VV-v«=L YV \<£ApVe_ Nj J V nnC J K ^ p r O sW-VNckev* v-\ 'S» ̂ ^ A V v Ce--e s- 3>3>^ ) UNIVERSITY OF IBADAN LIBRARY 25M- lOd. MacMillan? J. Personal Communications 11. Paleg? L.G. 1965, Physiological effects. of gibberellins. Ann, Re.v, Plant Physiol. , 16, 291 - 322. 12. Jones, R.L. ”1973,. Gibberellins. Their Physiological Role. \ Arm. Rev. Plant Physiol., 24, 571 - 598. 13. MacMillan, J.; and Takahashi, N. 1958. Proposed procedure for the allocation of trivial names to gibberellins. Nature, 217, 170 - 171. 14. Rowe, J.W. 1968. The common and Systematic nomenclature of cyclic Diterpenes. (Proposal I.U.P.A.C. Cotn̂ t- Org. Nomen. 3rd. rev.) 15. Cross, B.E. 1954. Gibberellic acid. Part I. J. Chem. Soc. ; 4670 - 4676. 16. Mulholland, T.P.C., and Ward, G. 1954. Gibberellic acid. Part II. J. Chem. Soc.; 4676 - 4681. 17. Cross, B.E.; Grove, J.F.; MacMillan, J. and Mulholland, T.P.C. 1958. Gibberellic acid. Part VII. The structure of gibberic acid. J, Chem. Soc. ; 2520 - 2536. 18. Mulholland, T.P.C. 1958. Gibberellic acid. Part IX. The structure of allogibberic acid. J, Chem. Soc. ; 2693 - 2701 19. Stork, G.; and Newman, H. 1959. The stereochemistry of allogibberic acid and of gibberic acid. J. Amer. Chem. Soc. 81, 3168. UNIV RSITY OF IBADAN LIBRARY 2 S 5 - 20. Grove, J.F.; and Mulholland, T.P.C. 1960, Gibberellic acid, Part XII., Stereochemistry of allogibberic acid, J, Chem, Soc.; 3007 - 3022, 21. Cross, B.E. 1960. Gibberellic acid. Part XIII. The structure of ring A. J■ Chem. Soc. ; 3022 - 3038. 22. Cross, B.E.; Grove, J.F.; MacMillan, J.; Moffatt, J.S.; Mulholland, T.P.C.; Seaton, J.C. and Sheppard, N. 1959. A revised structure for gibberellic acid. Proc. Chem. Soc. 302 - 303. 23. Sheppard, N. 1960. Gibberellic acid. Part XV. Nuclear magnetic resonance spectra and structures of gibberellic < • h acid derivatives. J. Chem. Soc.; 3040 - 3045. 24. Stork, G.; and Newman, H. 1959. The stereochemistry of gibberellic acid. J. Amer. Chem. Soc.; 81, 5518 - 5519. 25. Hartsuck, J.A.; and Lipscomb, W.N. 1963. Molecular and crystal structure of Di-p-bromobenzoate of the methyl ester of gibberellic acid. J. Amer. Chem. Soc.; 85, 3414 - 3418. 26. Scott, A.I.; McCapra, F.; Corner, F.; Sutherland, S.A.; and Young, D.W. 1964. Stereochemistry of the Diterpenoids - IV. Structure and stereochemistry of some polycj^clic diterpenoids. Tetrahedron, 20, 1339 - 1358. UNIVERSITY OF IBADAN LIBRARY 254 27. McCapra, F. ; McPhail, A.T,j Scott, A. I.. ; Sjjn, G . A . j and ...... - ' V Young, D.W. 1966. Fungal Metabolites., Part VII. btereochemistry of gibberellic acid., X~ray analysis of methyl bromogibberellate. J. Chem. Soc.; (C.), 577 - 1585. 28. Grove , J.F.; JEFFS, P.W.; and Mulholland, T.P.C. 1958. Gibberellic acid. Part V. The relation between gibberellin and gibberellic acid. J. Chem, Soc.♦ 1236 - 1240. 29. MacMillan, J. ; Seaton, J.C.; Sutter, P.J. 1960. Plant hormones I. Isolation of gibberellin A^ and gibberellin Â_ from Phaseolus rqultiflorus. Tetrahedron, 11, 60 - 66. 30. Kit^mura, H.; Takahashi, N. ; Seta, Y. ; Kawanada, A. ; Sumiki, Y. 1960. Biochemical studies on bakanae fungus. LIV. Chemical structure of gibberellins. 20. Bull. Agric. Chem, Soc. Japan, 23, 344 ~ 346 31. Gro\ve J.-F. 1961. Gibberellin A . J. Chem. Soc., 3545 - 3547. 32. Hanson, J.R. 1965. Gibberellic acid. Part XXXI. The nuclear magnetic resonance spectra of some gibberellin derivatives. J. Chem. Soc.; 5036 - 5040. 33. Takahashi, N. 1974. Recent progress in the chemistry of gibberellins, tn Plant Grovrbh Substances 1974, 228 - 240 (Hirokawa Publishing Co. Inc. Tokyo). UNIVERSITY OF IBADAN LIBRARY 2 5 7 34.. Yamaguchi, I.; Takahashi., N. ; Fujlta, K.. 1975. Application of 13C nuclear magnetic resonance. to the study of 'gibberellins. J, Chem., Soc. ; Perkin I, 992 - 996. 35. Evans 5 R. ; Hanson, J.R.; Siverns, M.. 1975. The "^C nuclear magnetic resonance spectra of some gibbere3_lins. J. Chem. Soc.; Perkin I, 1514 - 1517. 36. Wulfson, N.S.; Zaretskii, V.I.; Papernaja, I.B.; Serebryakov, E.P.; and Kucherov, V.F. 1965. Mass spectrometry of gibberellins. Tetrahedron Leiters, 4209.- 4216. 37. Zaretskii, V.I.; Wulfson, N.S.; Papernaja, I.B.; Gurvich, I.A.; Kucherov, V.F.; Milstein, I.M.; Serebryakov, E.P.; and :v,jr Simolin, AlVV 1968. Mass spectrometryof gibberellins II. The location of the double bond in the gibbane systern. Tetrahedron, 24, 2327 - 2337. 38. Tökahashi, N.; Murofushi, N.; Tamura, S.; Wasada, N.; Hoshino, H Tsuchiya, T.; Sasaki, S. ; Aoyama, T.; and Watanabe, E. 1969 Mass spectrometric studies on gibberellins . Org. Mass Spec. ̂ 2, 711 - 722. 39. Binks, R.; MacMillan, J.; and Pryce, R.J. 1969. Plant hormones VIII. Combined gas chromatography - mass spectrometry of the methvl esters of gibberellins A 1 to A and their trimethylsilyl ethers* Phytochemistry, 8, 271 - 284. 40. Njagata, W. ; Wakabayashi, T.; Hayase, Y. ; Narisada, M.; and Kamata, S. 1971. The stereocontrollea total synthesis UNIVERSITY OF IBADAN LI RARY 25S of dl-gibberellin A . J . Amer. Chem. Soc, ; 93. 57140 - 5758. 41. Mori, K. ; Shiozaki* M«; Haya, N.; Matsui, M.; and Sumiki5 Y. 1969. Synthesis of substances related to gibberellins XXI. Total synthesis of (±) gibberellins A . A 2, a , A 9 , and A1 u . Tetrahearon, 25, 1293 - 1321. 42. Bowen, D.H.; Harrison, D.M.; and MacMillan, J. 1972. Partial synthesis of gibberellin A O / by selective reduction of the hindered 10-Carboxy-group in gibberellin A . I o Chem. Comm. ; 808 - 809. 43. BeeleyL L.J.; and MacMillan5 J. 1976. Partial syntheses of 2-hydroxygibberellins; characterisation of two new gibberellins5 A and A . J. Chem. Soc.; Perkin I5 1022 - 1028. 44. Nagata, W.; and Yoshioka5 M. 1966. Alkylaluminum cyanides as potent reagents for hydrocyanation(I). Tetrahedron Leiters ̂ 1913 - 1918. 45. Nagata5 W.; and Hayase, Y. 1969. Formylolefination of carbony1 compounds. J. Chem. Soc,; (C) 460 -* 466. 46. Mori« K.; Matsui5 M.; and Sumiki5 Y. 1963. Biochemical Studies on bakanae fungus (Gibberella fujikuroi). LXVI. Synthesis of substances related to gibberellins. 10. Total synthesis UNIVERSITY OF IBADAN LIBRARY 25^ of racemic epigibberic acid. Agric. Biol. Chem.; 27 (7)5 537 - 42. 47. Cross5 B.E.; Hanson? J.R.; and Speake5 R.N. 1965. Gibberellic acid. Part XXX. The preparation of 7-deoxygibberellins from gibberellic acid. J. Chem. Soc.; 3555 - 3563. 48. Dasgupta9 S.K.; Dasgupta, R.; Ghosh, S.R.; and Ghatak, U.R. 1969. Bridged bic5?pclo (39 2, 1) octane ring System by intramolecular carbene insertion reaction: a new synthetic approach to tetracyclic diterpenoids. Chem. Comm.; 1253. 49. Hori, T. and Nakanishi9 K. 1969. Synthesis of hydrofluorenone derivative5 a promising intermediate for synthesis of Ci^-gibberellins. Chem. Cornrn. ; 528 - 529. 50. Corey, E. J. ; Narisada, M.; Hiraoka, T.; Ellison., R.A.; 1970. Studies on the total synthesis of gibberellic acids. A simpleroute to the tetracarbocyclic network. J. Amer. Chem. Soc.; 92«, 396 ~ 397. 51. Cross9 B.E.; and Gatfield5 I.L. 1970. Partial synthesis of gibberellin A^r norketone from 7-hydroxykaurenolide. Chem. Comm.; 33. UNIVERSITY OF IBADAN LIBRARY 2&0 52. Galt, and Hanson, J.R. 1965. New metabolites of Gibberella fujikuroi. Part VII. The preparation of some ring -B nor-derivatives. J. Chem. Soc. ; 1565 - 1570. 53. Crossj B.E.; Norton, K. ; and Stewart, J.C. 1968. The biosynthesis of the gibberellins. Part III. JY Chem. Soc.; (C), 1054 - 1063. 54. Crozier, A.; Aoki, H.; Pharis, R.P.; 1969. Efficiency of countercurrent distribution, Sephadex G-10, and silicic acid partition chromatography in the purification and Separation of gibberellin-like substances from plant tissue. J . Exp. Bot.; 20, 786 - 795. 55. Yokota, T.; Murofushi, N.; Takahashi, N.; and Tamura, S. 1971. Gibberellins in immature seeds of Pharbitis nil Part II. Isolation and structures of novel gibberellins, gibberellins A and A^. Agric. Biol. Chem.; 35, 573 - 582. 56. MacMillan, J.; Seaton, J.C.; and Suter, J.J. 1962. Plant hormones II. Isolation and structures of gibberellin A^ and gibberellin A . Tetrahedron, 18, 349 - 355. 57. Skene, K.G.M.; 1970. The gibberellins of developing bean seeds. J, Exp. Bot.; 21 (66), 236 - 46. UNIVERSITY OF IBADAN LIBRARY - 2« - 58. MacMillan, J.; and Wels, C.M. 1973. Partition chromatography of gibberellins and related diterpenes on columns of Sephadex LH-20. 31 Chromatography , 87, 271 - 276. 59. MacMillan, J.; and Sutter, P.J.; 1963. Thin-layer chromatography of t. hi e gibberellins. —Na tu~re, 197, 790. 60. Carveil, B.D.; MacMillan, J.; Pryce, R.J.; and Sheppard, A.C. 1967. Plant hormones - V. Thin-layer and gas-liquid chromatography of the gibberellins, direct identification of the gibberellins in a crude plant extract by gas-liquid *> chromatography. Phytochernistry, 6, 867 - 874. 61. Ikekawa, N.; Sumiki, Y. ; and Takahashi, N. 1963. Gas Chromatographie Separation of gibberellins. Chem. and Ind. , (Lond.), 1728 - 1729. 62. MacMillan, J.; Pryce, R.J.; Eglinton, G.; and McCormick, A. 1967. Identification of gibberellins in crude plant extracts . by combined gas chromatography - mass spectrometry. Tetrahedron Retters, 2241 - 2243. 63. Durley, R.C.; MacMillan, JY, and Pryce, R.J. 1971. Plant hormones - XI. Investigatxons of gibberellins and other growth substances in the seeds of Phaseolus multiflorus and of Phaseolus vulgaris by gas chromatography and by gas chromatography - mass spectrometry. Phytoehernistry 10, 1891 - 1908. UNIVERSITY OF IBADAN LIBRARY 2£>7- 64-. Gaskin, P.; and MacMillan, J. Unpublished data. 65. MacMillan, J. ; and Wels, C.M. 1974. Detailed analysis of metabolites from mevalonic lactone in Glbberella fuj ikuroi. Phytochemistry, 13, 1413 - 1417. 6 6 . Beeley, L.J. ; Gaskin, P. •a,nd MacMillan, J. 1975. Gibberellin A ̂and other terpenes in endosperm of Echinocystis macrocarpa. Phytochemistry, 14, 779 - 783. 67. Phinney, B.O.; and West, C.A. 1960. Gibberellins as native plant growth regulators. Ann. Rev. Plant. Physiol.; 11, 411 - 436. 68. Bailiss, K.W. and Hill, T.A. 1971. Biological assays for gibberellins. Bot. Rev., 37 (4), 437 - 479. 69. Frankland, B. and Wareing, P.F. I960. Effect of gibberellic acid on hypocotyl growth of lettuce seedlings. Nature, 185, 255 - 256. 70. Nicholls, P.B.N.; Paleg, L.G. 1963. A Barley endosperm bioassay for gibberellins. Nature, 199, 823 - 824. 71. Coombe, B.G.; Cohen, D.; Paleg, L,- 1967. Barley endosperm bioassay for gibberellins II. Application of the method. Plant Physiol.; 42, 113 - 119. 72. Cross, B.E.; Grove, J.F.; and Morrison, A. 1961. Gibberellic acid. Part XVIII. Some rearrangements of ring A. J. Chem. Soc.; 2498 - 2515. UNIVERSITY OF IBADAN LIBRARY 73. Cross, B.E.; Galt, R.H.B.; and Hanson, J.R. 1962. ' New metabolites of Gibberella fujikuroi - I. Gibberellin and gibberellin A^. Tetrahedron, 18, 451 - 459. 74. Cross, B.E. 1968. Biosynthesis of the gibberellins in Progress in Phytochemistry, Vol. I. (Reinhold, L.; and Liwschitz, Y.; eds.), 195 - 222. Wiley and Sons, New York. .75. West, C.A. 1973. Biosynthesis of gibberellins in Biosynthesis and its control in plants. (Milborrow, B.U.; Ed.), 143 - 169, Academic Press, London. 76. Graebe, J.E.; Bowen, D.H.; MacMillan, J. 1972. The conversion of mevalonic acid into gibberellin A^-aldehyde in a cell-free System from Cucurbita pepo. Planta, 102, 216 - 271. 77. Graebe, J.E.; Hedden, P.; Gaskin, P.; and MacMillan, J. 1974. The biosynthesis of a C^g-gibberellin from mevalonic acid in a cell free System from a higher plant. Planta, 120, 307 - 309. 78. Graebe, J.E.; Hedden, P.; Gaskin, P. s arid MacMillan, J. 1974. Biosynthesis of gibberellins k ^ »  ,_, ^24’ ^35 an<̂ ‘̂37 by a cell-free System from Cucurbita maxima. Phytochemistry 13, 1433 - 1440. 79. Frydman, V.M.; MacMillan, J. 1975. The metabolism of gibberellin Ag, Aoq, and Aog in immature seeds of Pisum sativum CV. UNIVERSITY OF IBADAN LIBRARY Progress No. 9. Planta, 125, 181 - 195. 80. Turner, J.N. 1972. Practical uses of gibberellins in agriculture and horticulture. Outlook Agric.; 7(1), 14 - 20. 81. Stuart, N^W. and Cothey, H.M. 1961. Applied aspects of the gibberellins. Ann. Rev. Plant Physiol.; 12, 369 - 394. 81a. Grove, J.F. 1961. The Gibberellins. Quart. Rev., 15, 56 - 70. ' " ; ' 82. Kogl, F. and Haagen Smit, A.J. 1931. Uber die Chemie des Wuchsstoffs. Proc. K. ned. Akad. Wet.; 34, 1411 - 1416. ( C , CLU ’ au 55^ 83. Kogl, F. ; Haagen Smit, A.J.; and Erxleben, H. 1934. Uber ein neues Auxin ('Heteroauxin1) aus Harn. Hoppe-Seyl. Z. 228, 90 - 103. Ce-*-, X°\ ) 84. Vliegenthart, J.A. and Vliegenthart, J.F.G. 1966. Reinvestigation of authentic samples of auxins A and B and related products by mass spectrometry. Reel. Trav. Chim. Pays-Bas. Belg.; 85, 1266 - 1272. > 85. Schneider*5 E.A. and Wightman, F. 1974. Metabolism of auxin in higher plants. Ann. Rev. Plant Physiol.; 25, 487 ̂513* 86. Bonner, J. and Bandurski, R.S. 1952. Studies of the physiology, pharmacology and biochemistry of the auxins. Ann. Rev. Plant Physiol.; 3, 59 - 86. UNIVERSITY OF IBADAN LIBRARY 2^5 87. Galstons A.W. and Purves, W.K. 1960. The mechanism of auxin action. Änn. Rev. Plant Physiol.; 11, 230 - 276. 88. . Fawcett, C.H. 1961. Indole auxins. Änn. Rev. Plant Physiol.; 12, 34-5 - 368. 89. Jones, E.R.H.; Henbest, H.B. ; Smith, G.F.; and Bentley, J.A. 1952. 3-Indolylacetonitrile: a naturally occurring plant growth hormone. Nature, 169, 485. 90. Marumo, S. ; Abe, H. ; Hattori, H.; Munakata, K. ; 1968. Isolation of a novel auxin, methyl 4-ehloroindole-acetate from immature seeds of Pisum sativum. Agric. Biol. Chem. ; 32, 117 - 118. 91. Zimmerman, P.W.; Hitchcock, A.E.; and Wilcoxon, F. 1936. Several esters as plant hormones. Contrib. Boyce Thompson Inst.; 8, 105 - 112. 92. Zimmerman, P.W.; and Hitchcock, A.E.; 1942. Substituted phenoxy and benzoic acid growth substances and the relation of structure to physiological activity. Contrib. Boyce Thompson Irist. ; 12, 321 - 343. ^ ) 93. Bentley, J.A. 1961. Extraction and purification of auxins. Encyclopaedia of Plant Physiology. Springer, Berlin, Gottingen, Heidelberg, 14, 501 - 520. ^V A t d 94. Kaldewey, H. and Stahl, E. 1964. Die quantative auswertung UNIVERSITY OF IBADAN LIBRARY 2 dunnschichtchromamatographisch getrennter auxine im Avena-tageslichttest nach Soding (Avena-Silicagel- Krummungstest), Planta, 62, 22 - 38.. ^ V C\ , 95. Audus, L.J. 1972. The natural auxins in his Plant growth Substances. Vol. I. Chemistry and Physiology. Leonard Hill!5 London. 96. Powell, L.E. 1964. Preparation of indole extra cts from plants for gas chromatography and spectrophotofluorometry. Plant. Physiol.; 39, 836 - 842. 97. Jamieson, W.D. and Hutzinger, 0. 1970. Identification of simple naturally occurring indoles by mass spectrometry. Phvtochemistpy , 9, 2029 - 2036. 98. Abe, H. and Marumo, S. 1974. Mass spectra of indoleacetic acid and its chloroderivatives. Agric. Biol. Chem.; 38(8) 1537 - 1538. 99. Gibson, R.A.; Schneider, E.A.; and Wightman, F. 1972. Biosynthesis and metabolism of indol~3yl-labeiled precursors of IAA In tomato and barley shoots. J . Exp. bot.; 23, 381 - 399. 100. Erdmann , N. ; and Schiewer * U. 1971. Tryptopharijdependent Indoleacetic-acid biosynthesis from Indole, demonstrated by double-labeHing experiments.Planta, 97, 135 - 141. 101. Schneider, E.; Gibson, R.A.; Wightman. F. 1972.. Biosynthesis UNIVERSITY OF IBADAN LIBRARY 261 and metabolism of indol-3yl~acetic acid. I* The native indoles of barley and tomato shoots. J. Exp. Bot. ; 23, 152 - 170. 102. Hinman, R.L.; and Lang, A. 1965. Peroxidase catalysed oxidation of indole-3-acetic acid. ‘Biochemistry, 4, 144. 103. Leopold, A.C. 1958. Auxin uses in the control of flowering and fruiting. Ann. Rev. Plant Physiol.; 9, 281 - 310. 104. Skoog, F. and Armstrong, D.J. 1970. Cytokinins. Ann * Rev. Plant Physiol.; 21, 359 - 384. 105. Miller, C.O.; Skoog, F.; Okumura, F.S.; Von Saltza, M.Hi; and Strong, F.M.; 1956. Isolation, structure, and synthesis of kinetin, a substance promoting cell division. J. Amer, Chem. Soc.; 78, 1375 - 1380. 106. Kende, H. 1971. The Cytokinins. Int. Rev. Cytol.; 31, 301 - 338. 107. Miller, C.0. 1961. A kinetinlike compound in maize. Proc. Natt Acad. Sei. U.S.; 47, 170 - 174. 108. Letham, D.S.; Sh^non, J.S. and McDonald, I.R. 1964. The structure of zeatin, a factor inducing cell division. Proc. Chem. Soc.; 230 ~ 231. 109. Shannon, J.S. and Letham, D.S. 1966. Regulators of cell division in plant tissues IV. The mass spectra of cytokinins and other 6-aminopurines. N.Z. J. Sei.; 9C4), 833 - 842. UNIVERSITY OF IBADAN LIBRARY 26% 110. Shaw* G. ; and Wilson, D.V. 196*4. A synthesis of zeatin. Proc. Chem. Soc.; 231. 111. Koshimizu, K. ; Kusaki, T.; Mitsui, T.; Matsubara, S. 1967. Isolation of a cytokinin, (~)-dihydro-zeatin from immature seeds of Lupinus luteus.T.etrahedron Leiters, 1317 - 1320. 112. Klambt, D.; Thies, G.; and Skoog, F. 1966. Isolation of cytokinins from Corynebacterium fascians. Proc. Natt. Aead. Sei., U.S.; 56(1), 52 - 59. 113. Helgeson, ä P.; and Leonard, N.J. 1966. Cytokinins. Identification of compounds isolated from Corynebacterium fascians. Proc. Natl. Acad. Sei. U.S.; 56(1), 60 - 63. 114. Hall, R.H.; Robins, M.J.; Stasiuk, L.; Theuford, R.; 1966. Isolation of N 6~(y,y-Dimethylallyl) adenosine from soluble ribonucleic acid. J. Amer. Chem. Soc.; 88, 261*4 - 2615. 115. Burrows, W.J.; Armstrong, D.J.; Kaminek, M . ; Skoog, F.; Hecht, S.M.; Dammann, L .G. ; Leonard, N.J.; and Qccolowitz, J 1970. Isolation and identification of *4 cytokinins from wheat germ ribonucleic acid.. Biochemistry 9(9), 1867 n 1872 116. Horgan, R.; Hewett, E.W.; Horgan, T.M.; Purse, J.; and Wareing, P.F. 1975. A new cytokinin from Populus robusta . Phytochemistry, 1*4 (*4), 1005 - 1008. UNIVERSITY OF IBADAN LIBRARY - 26^ - 117. Eiion, G.B.; Burgi, E. and Hitchings, G.H. 1952. Studies on Condensed pyrimidine Systems IX. The synthesis of some 6-substituted purines. J. Amer. Chem. Soc. ; 74, 411— 414 118. Bullock, M.W.; Hand, J.J.; and Stokstad, E.L.R. 1956. Synthese of 6-substituted purines. J. Amer. Chem. Soc.; 78, 3693 - 3696. 119. Leonard, N.J.; and Fujii, T. 1964. The synthesis-of compounds possessing kinetin activity. Proc» Natt» Acad. Sei. U.S. 51 (1), 73 - 75. 120. Most, B.H.; Williams, J.C. and Parker, K.J. 1968. Gas Chromatography of cytokinins. J. Chromatography, 38, 136 - 138. / K • 121. Townsend, L.B.; Robins, R.K.; Leoppky, R.N. and Nelson, J.L. 1964. Purine nucleosides VIII. Reinvestigation of the ’ Position of glycosidation in certain svnthetic "7T!- substituted 6-dimethylamino“purine nucleosides related to puromycin. J. Amer. Chem. Soc.; 86, 5320 - 5325. 122. Leonard, N.J. and Deyrup, J-A. 1962. The Chemistry of triacanthine. J. Amer. Chem. Soc.; 84, 2148 - 2160. 123. Horgan, R.; Hewett, E.W.; Purse, J.G.; Horgan J.M.; and Wareing, P.F. 1973. Identification of a cytokinin in Sycarnore sap by gas chromatography mass-spectrometry. UNIVERSITY OF IBADAN LIBRARY 210 Plant Science Leiters, 1, 321 - 324. 124. Helgeson, JiP. 1968. The Cytokinins. Science, 161, 974 - 981. 125. Audus, L.J. 1972. The biological assay of the cytokinins tn his Plant growth Substances Vol. I. Chemistry and Physiology. Leonard Hill, London. 126. Letham, D.S.; Wilson, M.M.; Parker, C.W.; Jenkins, I.D.; Macleod, J.K.; and Summons, R.E. 1975. Regulators of cell division in plant tissues XXIII. The identity of an unusual metabolite of 6'-benzyl aminopurine. Biochimica et Biophysica Acta, 399, 61 - 70. 127. Ohkuma, K. ; Lyon, J.L.; Addicott, F.T. and Smith, O.E. 1963. Abscisin II, an abscission accelerating substance from young cotton fruit. Science N.Y.; 142, 1592 - 1593. 128. Milborrow, B.V. 1974. The chemistry and physiology of abscisic acid. Ann. Rev. Plant Physiol.; 25, 259 - 307. 129. Addicott, F.T. and Lyon, J.L. 1969. Physiology of abscisic acid and related substances. Ann. Rev. Plant Physiol.; 20, 139 - 164. 130. Ohkuma, K.; Addicott, F.T.; Smith, O.E.; Thiessen, W.E. 1965. The structure of abscissin ±1. Tetrahedron Leiters, 2529 - 2535. UNIVERSITY OF IBADAN LIBRARY 211 131. Cornforth, J.W.; Milborrow, B.V.; Ryback, G. 1965. Synthesis of (±) - Abscissin II. Nature , 206, 715. 132. Roberts, D.L.; Heckman, R.A.; Hege, B.P.; Bellin, S.A.; 1968. Synthesis of (RS) - abscisic acid. J . Org. Chem.; 23 3566 - 3569. 133. Cornforth, J.W.; Milborrow, B.V.; Ryback, G. 1966. Identification and estimation of (+) abscisin II Odormin1) in plant extracts by spectropolarimetry. Nature, 20, 627 - 628. 134. Milborrow, B.V. 1967. The Identification of (t)-abscisin II [(+)-dormin] in plants and measurement of its concentrations ■*'. y ■ Planta, 76, 93 - 113. 135. Davis, L.A.; Heinz, D.E. and Addicott, F.T. 1968. Gas-liquid chrornatography. of trimethylsilyl derivatives of abscisic acid and other plant hormones. Plant Physiol.; 43, 1389 - 1394. 136. Antoszewski, R.; and Rudnicki, R. 1969. Spectrofluorometric method for quantitative determination of abscisic acid on thin-layer chromatograms. Anal. Biochem.; 32, 233-237. 137. Mallaby, R.; and Ryback, G. 1972. Chemistry of a colour test for abscisic acid. J. Chem. Soc.; Perkin II, 911 - 921. 138. Milborrow, B.V. and Noddle, R.C. 1970. Conversion of 5~(1,2~ epoxy-2,6,6-trimethylcyclohexyl)-3-methylpenta-cis-2- UNIVERSITY OF IBADAN LIBRARY 272. trans-4~dienoic acid into abscisic acid in plants. Blochern.. J.; 119, 727 - 734. 139. Milborrow, B.V. 1971. Abscisic acid in Aspects of terpenoid chemistry and blochemistry (Goodwin, T.W. ed.). Academic press Lond. and New York. 140. Noddle, R.C.; and Robinson* D.R. 1959. Biosynthesis of abscisic acid; incorporation of radioactivity from (2- 14 C) 'mevalonic acid by intact fruit. Biochem, J.; 112, 547 - 548. 141. Taylor, H.F.; and Smith, T.A. 1967.. Production of plant growth Inhibitors from xanthophylls: a possible source of dormin. Nature, 215, 1513 - 1514. 142. Milborrow, B.V. 1969. The occurrence and function of abscisic acid in plants. Sei. Prog. Oxf.; 57, 533-558. 143. Taylor, H.F. and Bürden, R.S. 1970. Identification of plant r\ growth Inhibitors produced by photolysis of violaxajhin. Phytochemistry, 9, 2217 - 2223. 144. Firn, R.D.; Bürden, R.S.; Taylor, H.F. 1972. The detection and estimation of the growth inhibitor xanthoxin in plants. Planta, 102, 115 - 126. 145. Taylor, H.F., and Bürden, R.S. 1973. Preparation and roetabolism of 2-[14C]-cis,trans-xanthoxin. J. Exp. Bot. ; 24(82), 873 - 880. UNIVERSITY OF IBADAN LIBRARY 273 146. Robinson, D.R. Unpublished results, cited by Milborrow, B.V. 1971 - Ref. 139. 147. Milborrow, B.V. 1969. Identification of ’metabolite Cf from abscisic acid and a new structure for phaseic acid. Chenu Comm.; 966 ~ 967. 148. MacMillan, J.; and Pryce, R.J. 1968. Phaseic acid, a putative relative of abscisic acid from seed of Phaseolus multiflorus Chem. Cornrn, ; 124 - 126. 149. Tinelli, E.T.; Sondheimer, E.; Walton, D.C.; Gaskin, P. and MacMillan, J. 1973. Metabolites of 2- 14C-abscisic acid. Tetrahedron Leiters, 139 ~ 140. 150. Walton, D.C.; Dorn, B.; and Fey, J. 1973. The Isolation of an abscisic acid metabolite, 4*-dihydrophaseic acid from non-imbibed Phaseolus vulgaris seed. Planta, 112(1), 87 - 90. 151. Zeevart, J.A.D.; and Milborrow, B.V. 1976. Metabolism of abscisic acid and the occurrence of epi-dihydrophaseic acid in Phaseolus vulgaris. Phytochemistry, 15, 493 - 500. 152. Milborrow, B.V. 1975. The absolute configuration of phaseic • acid and dihydrophaseic acids. Phytochemistry, 14, 1045 - 1053. UNIVERSITY OF IBADAN LIBRARY 153. Neljubow, D. 1901. Uber die horizontale nutation der stengel von Visum sativum und einiger anderer pflazen. Beih. bot. Zbl.; 105 128 -138 cited by Audus, L.J. 1972, in bis plant grovrth substances Vol.I. Chemistry and Physiology. Leonard Hill, London. 154. Gane, R. 1934. Production of ethylene by some ripening fruits. Nature, 134, 1008. 155. Burg, S.P. 1962. The physiology of ethylene formation. Ann Rev. Plant Physiol.; 13, 265 - 302. 156. Burg, S.P. 1968. Ethylene, plant senescence and abscission. Plant Physiol.; 43, 1503 - 1511. 157. Pratt, H.K. ; and Goeschl, T.D. 1969. Physiologe’cal roles of ethylene in plants. Ann. Rev. Plant Physiol.; 20, 541 - 584. 158. Valdovinos, J.G« ; Ernest, L.C.; and Henry, E.W. 1967. Effect of ethylene and gibberellic acid on auxin synthesis in plant tisŝ es. Plant Physiol.; 42, 1803 - 1806. 159. Abeies, F.B. 1972. Biosynthesis and mechanism of action of ethylene. Ann. Rev. Plant Physiol.; 23, 259 - 292. 160. Lieberman, M. and Mapson, L.W. 1964. Genesis and biogenesis of ethylene. Nature, 204, 343 - 345. UNIVERSITY OF IBADAN LIBRARY 2 1 5 161a.. Lieberman5 M.; Kunishi, A.T.j Mapson, L.W. and Wardale, D.A. 1965. Ethylene produetion from methionine. Biochem. J.; 97, 449 - 459. 161b. Martin, G.C.; and Dennis, F.G. Personal communication. 162. Gleen, J.L.; Kuo, C.C.; Durley, R.C« and Pharis, R.P. 1972. Use of insoluble polyvinylpyrrolidone for purification of plant extracts and chromatography of plant hormones. Phytochemistry, 11, 345 ~ 351. 163. Phinney, B.O. Unpublished work. 164. MacMillan, J. 1972. Current aspects cf gibberellins and plant growth regulations in Hormonal Regulation in Plant Growth and Development. Proc. Adv. Study Inst. Izmir 1971, pp. 175 - 187. Kaldewey, H. and Vardar, Y. (eds.), Verlag Chemie, Weinheim.-1972. 165. MacMillan, J. and Pryce, R.J. 1969. Plant hormones X. The Constitution of phaseic acid, a relative of abscisic acid from Phaseolus multiflorus. An interpretation of the mass spectrum of phaseic acid and a possible structure. Tetrahedron, 25, 5903 - 5914. 166. Bearder, J. ; Gaskin, P..; and MacMillan, J. Unpublished results, cited by Beelev, L.J.; in his Ph.D. thesis 1975. University of Bristol. UNIVERSITY OF IBADAN LIBRARY - 276 - 167. Most9 *B.H.; Gaskin P. and MacMillan, J. 1970. The occurrence of abscisic acid in Inhibitors and C from imraature fruits of Ceratonia siliqua L. (Carob) and in commercial carob syrup. Planta3 92 9 Ul - 49. 168. Frydman, V.M. ̂ and MacMillan, J. 1973. Identifications of gibberellins A^q and A^g in seed of Pisum sativum cv, Progress No. 9 by combined gas~chrornatography-mass spectrometry. Planta , 1155 11 - 15. 169. Sen5 S.P.; and Leopold, A.C. 1954. Paper chromatography of plant grovrth regulators and allied compounds. Physlologla plantarum , 7, 98 - 108. 170, Van Steveninck, R.F.M, 1959. Abscission accelerators in Lupins. (Lupinus luteus L.). Nature , 183, 1246 - 1248. 171. Van Steveninck, R.F. 1959. Factors affecting the abscission of reproductive organs in yellow lupins (Lupinus luteus L. III. Endogenous grovrth substances in virus infected and healthy plants and their effect on abscission. J . Exp. Bot.; 105 367 - 376. 172a. Cornforth, J.W. ; M\lborrow, B.V.; Ryback, G.; Rothvje.il, K. ; and Wain, R.L. 1966. Identification of the yellow lupin grovrth Inhibitor as (+) Abscisin II((+) - Dox^min). Nature , 211s 742 - 743. UNIVERSITY OF IBADAN LIBRARY 277 172b. Moore5 T.C. and Ecklund, P.R., 1975. Role of gibberellins in the development of fruits and seeds m Gibberellins and Plant Growth (Krishnamoorthy> H.N* ed.) Wiley Eastern Ltd. New Delhi. 173. Kirk, D.N. and Hartshorn, M.P, 1968. Olefin-forming elimination reactions in their ARea.ction Mechanisms. Elsevier, Amsterdam. 174. Criegee, R. ; Marchand, B .\ and Wannowius, H. 1942. Zur kenntnis der organischen osmium-Verbindungen II. Mitteilung. Annalen, 550, 99 - 133. 175. Baran, J.S. 1960. Method for the cleavage of osmate esters. J. Org. Chem,; 25, 257. 176. House, H.O. 1972. Oxidation of carbon-carbon double bonds in his Modern Synthetic Reactions ̂ 2nd Edition, W.A. Benjamin Inc. 177. Woodward, R.B. and Bruteher* F.V. Jr. 1958. cis-Hydroxylation of a synthetic steroid intermediate with iodine, silver acetate and wet acetic acid. J. Amer. Chem. Soc.; 80, 209 - 211. 178. Gunstone, F.D. 1960 in Advances in .Organic Chemistry, Vol. I Raphael, R.A.; Taylor, E.C.; and Wynberg, H. (eds.) Interscience, New York. UNIVERSITY OF IBADAN LIBRARY 278 179. Aldridge, D.C.; Hanson, J.R.; and Mulholland, T.P.C. 1965. Gibberellic acid. Part XXVIII. Some derivatives of gibberellins and A^, J. Chem. Soe. , 3539 - 3549. 180. Beeley, L.J. 1975. Ph.D. Thesis, University of Bristol. 181. Maercker, A. 1965. The Wittig reaction Organic Reactions, 14, 270 - 490. 182. Brovm, H.C. and Zweifel, G. 1960. Hydroboration VII. Directive effects in the hydroboration of olefins. J. Amer. Chem. Soc.; 82, 4708 - 4712. 183. Brown, H.C. and Zweifel, G. 1961. Hydroboration IX. The hydroboration of cyclic and bicyclic olefins - stereochemistx'y of the hydroboration reaction. J. Arner. Chem, Soc.; 83, 2544 - 2551. 184. Djerassi, C. 1963. The protection of hydroxyl groups in his Steroid Reactions: An outline for organic chemists. Holden-Day Inc. 185. Tanabe, M. ; and Bigley, B. 1961. 17a,21-isopropylidenedioxy steroids. J. Amer. Chem. Soc.; 83, 756 - 757. 186. Robinson, C.H.; Finchkenor, L.E.; Tiberi, R. ; and Oliveto, E.P. 1961. 17,21-acetonide derivatives of 9,11-disubstituted cortical hormones. J. Org. Chem,; 26, 2863 '--2867. UNIVERSITY OF IBADAN LIBRARY 21H 187. Collins, J.C.; and Hess, W.W. 1972. Aldehydes from primary alcohols by oxidation with chromic trioxide: 1-heptonal. Organic Syntheses, 52; 5 - 10. 188. Zderic, J.A.; Carpio, H.; and Ejerassi, C. 1960. Steroids CXXX. Synthesis of 12ß-hydroxy-prednisolone-llß, 12ß-acetonide 21-acetate. J. Amer, Chem. Soc.; 82, 446 - 451. 189. Ratcliffe, R.W. 1976. Oxidation with the chromiurn trioxide-pyridine complex prepared in situ: 1-decanal. Organic Synthesis, 55, 84 - 86. 190. Croft, K.D.; Ghisalberti, E.L.; Jefferies, P.R.; Knox, J.R.; Mahoney, T.-J.; and Sheppard, P.N. 1974. Chemical and microbiological syntheses of intermediates in gibberellin biosynthesis. Tetrahedron, 30, 3663 - 3667. 191. McOmie, J.F.W. 1963. Protective groups. Adv. Org. Chem.; 3, 191. 192. Satchell, D.P.N. 1963. An outline of acylation. Quart. Rev.; 17, 160 - 203. 193. Hanson, J.R. 1966. New metabolites of Gibberellin fujikuroi - X. Gibberellin Tetrahedron, 22, 701 - 703. UNIVERSITY OF IBADAN LIBRARY 2S0 194* Sweeley, C.C.; Bentiey* R.; Makita, M.; and Wells, W.W. 1963« Gas-liquid chromatography of trimethylsilyl derivatives of sugars and related substances. J. Ämer> Chem. Soc,; 85 ? 2497 - 2507. 195* Vogel 5, A.I. 1360. Buffer Solutions ln his Text-book of Quantitative Inorganic Analysis. Theory and Practice. 2nd Edition, Longmans. 196., Clark5 W.M. and Lubs, H.A. 1916. Hydrogen electrode Potentials of phthalate, phosphate, and borate buffer mixtures. J. Biol, Chem,; 25, 479 - 509. 197. Willig s G. and Schoellkopf, ü. 19. Methylenecyc■fl" o|hexane. Org■ Syntheses, 40, 66 - 68. UNIVERSITY OF IBADAN LIBRARY