The research interests of the Organic
Synthesis Group encompass two major overlapping themes. The first
of these is focussed on methods and strategies for the synthesis of
complex natural products which have interesting biological
properties, while the second is concerned with the molecular basis
of plant growth regulation, using organic synthesis as an enabling
technology.
Despite the enormous progress made by organic chemists over the past three decades in developing new methods and strategies for the construction of complex molecules, the efficiency, reliability and predictability of most procedures still leaves much to be desired. It is important, therefore, that we continue to attempt difficult syntheses and thereby test synthetic procedures to the limit. In this way the practitioner is forced to seek a better understanding of the chemical processes involved, to improve existing methodology, and to invent new ways of building organic molecules. Within this context, the group has successfully completed syntheses of numerous complex natural products and developed a number of useful synthetic procedures.
The second major activity in the group is concerned with natural plant bioregulators with special reference to the gibberellins ("GAs"). GAs affect numerous aspects of plant growth and development, e.g. germination and flowering, although their most characteristic property is the induction of stem growth. There are several commercially valuable applications: most seedless table grapes are now grown with the application of GAs. The rind of citrus fruit typically softens at maturity, and is subject to injury by pests which adversely affect the appearance of otherwise marketable fruit. GAs maintain the rind in better condition. Control of russet, a "scabby" skin disorder in apples, especially in "golden delicious", may be achieved by the application of GAs. A variety of ornamental plants can be induced to flower either earlier than usual, or in off-seasons.
Studies on GAs, pursued in collaboration with several groups within the plant development program in the Cooperative Research Centre for Plant Science, have led to the discovery of semi-synthetic derivatives that selectively promote flowering, but not growth. Other analogues interfere with the plant's natural production of phytohormones, thereby inhibiting growth. Because they can be expected to be environmentally benign, these materials have considerable commercial potential and are presently being screened in field tests in N. America, Israel and Europe, as well as in Australia.
Considerable effort has been invested in transforming kaurene derivatives into gibberellins either by incubation with the fungus, Gibberella fujikuroi, or by chemical means. Most kaurenoids, however, are not as easily obtained as the more common gibberellins. We have therefore been exploring the prospect of utilising these compounds as a source of semi-synthetic kaurenoids. Of special interest is the Rabdosia family of diterpenes, many of which have antibacterial and antineoplastic properties, e.g. oridonin (1), enmein (2) and shikodonin (3). We have recently transformed the gibberellin, GA15 (4), into the hemiacetal (6) via hydroxy aldehyde (5) as with a view to preparing the seco-derivatives (10). This sequence is expected to give access to the kaurenoids macrocalyxoformin (7) and longirabdolactone (8), while serving as a model for the preparation, inter alia, of the more complex derivatives 1 - 3.
The probable biosynthetic pathway leading to the highly bioactive gibberellins, GA32 (4) and GA87 (5), isolated from various members of the Prunus genus, are two of the most potent gibberellin bioregulators known. The most likely biosynthetic precursors appeared to be GA5 (1) or GA95 (3), but although both were converted by immature seeds of P. persica (peach) into GA3 (2), neither were incorporated into 4 or 5. The recent identification of two new gibberellins from P. persica , GA120 (8) and GA121 (9), however, strongly suggests that the actual biosynthetic pathway is (7)-->(8)-->(9)-->(6)--> (5)-->(4). The formation of GA120 by the fungus, Gibberella fujikuroi, was suspected by Hanson some 30 years ago, but not proven. As well as prunus, GA120 has also very recently been found in Ornithgalum thyrsoides, a bulbous geophyte, in which it co-occurs with GA7 (13-deoxy-GA3). Thus, the discovery of GA120 may well be indicative of new biosynthetic pathways that proceed via 1,2-dehydrogenation of GA9 (7).
18-Hydroxy gibberellin A4 (4, R = H) is representative of a family of 18-hydroxylated gibberellins found in trace amounts in barley (Hordeum vulgare). In order to confirm tentative structural assignments and to obtain sufficient material to conduct a number of biological investigations, we have developed a general protocol for the introduction of a hydroxy group at C-18 in gibberellins as illustrated by the preparation of ester 4 (R = Me) from gibberellin A4 (1), which may be obtained in good supply from fermentation of the fungus, Gibberella fujikuroi. The key process in the conversion is based on a tandem sequence initiated by the conjugate addition of an alkoxide to the a-methylene lactone function in 2 followed by an intramolecular aldol reaction to form 3.
The unusual diterpenoid tropone, harringtonolide (4), first isolated in North America from seeds of Cephalotaxus harringtonia (Taxaceae), and independently from the bark of the related Chinese species Cephalotaxus hainanensis, has been shown to have promising anti-neoplastic and anti-viral properties. In order to explore its chemistry and therapeutic potential, we have examined a number of strategies for the preparation of 4, culminating in the successful completion of the sequence outlined in Scheme 1.
Scheme 1
A shortcoming of this approach, however, was the need to carry out extensive manipulations in the presence of the highly reactive cycloheptatriene moiety so as to effect the conversion of 3 into 4. In seeking a more efficient synthesis of 4 and its analogues, we have now designed the alternative route outlined in Scheme 2 that should require fewer steps and that postpones the assembly of the cycloheptatriene array to a much later stage. In order to evaluate this new approach we have recently prepared the model diazoketone 10 and studied its transformation into 11 (Scheme 3) . The results have been sufficiently encouraging to embark upon the synthesis itself, and the first stage, namely the preparation of the des-8-methyl analogue of 6 (R = CO2Me) has been successfully completed.
Scheme 2
Scheme 3.
Further studies in total synthesis are being directed towards the more complex C20 diterpene alkaloids, e.g. (12), and the Galbulimima group of alkaloids. Among the latter compounds, himbacine (13), is a potent muscarinic anatagonist and a lead compound in the search for drugs to treat Alzheimer's disease. (13) has been recently synthesised, but our own efforts have been focussed on the more complex members of the family, e.g. himgaline (14) and himaline (15). The advanced intermediate (16) has been prepared and current efforts are directed at cutting open the cyclohexenone ring and inserting the nitrogen atom. Methodology for the conversion of (15) into (14) has already been established.
Last revised 28 September 2000