yew.gif (4624 bytes)


yew.gif (4624 bytes)

4. Synthesis

When the US Food and Drug Administration (FDA) approved the use of taxol for treating advanced ovarian cancer in 1992 supplies of taxol were scarce.  Obtaining sufficient quantities of the drug became the prime concern and sparked a great deal of research into the complete synthesis of taxol. Meanwhile a potential solution was found when 10-deacetylbaccatin III (3), a precursor of taxol lacking the C-13 side chain. and the C10 acetoxy group , was discovered in the twigs and needles of the European Yew tree (Taxus baccata).

baccatin.gif (3471 bytes)
3: 10-deacetylbaccatin III

Harvesting this renewable source followed by semi-synthesis9 allowed the production of taxol on a relatively large scale and provided sufficient quantities of the drug to complete clinical trials and begin treating the first patients.   This method is, however, very laborious and low yielding. Hence the price of taxol remained prohibitively high for its wide scale use to be implemented in the fight against cancer. The pressure on groups researching the synthesis of taxol therefore increased concluding with the complete synthesis by three independent groups in the autumn of 1995.

The Complete Synthesis of Taxol
Although the main interests in taxol lay with it's unique method of action and potential as an anti cancer agent, synthetic chemists were most impressed with it's structure10.  The molecule is distinguished by a characteristic ester side chain, a 6-8-6 tricyclic carbon framework and a dense pattern of oxygenated functionality. Taxol contains a total of 11 stereoisomers marked * on the diagram below (4).

taxol_stereo.gif (2262 bytes)
4: Taxol, 11 Stereo-centres.

This means that there are potentially 211 (2048) different stereoisomers.   As with most biologically active chemicals only a few of these have the desired activity.  This presented synthetic chemists with a very large problem that required the development of new strategies and tactics for chemical synthesis.  To date three research groups have disclosed distinct total syntheses.

i) The Nicolaou Approach11,12,13
This route was the first complete synthesis of taxol to be disclosed, being published initially in the journal Nature11.   The work was conducted by the Nicolaou group in the Nicolaou laboratories at the Scripps Research Institute and resulted in K.C. Nicolaou Ph.D. being elected to membership in the National Academy of Sciences on May 2, 1996,   in recognition for his outstanding scientific achievements14.

The route devised by Nicolaou involved the construction of the A and C rings separately and then coupling the two molecules together using a Shapiro and a McMurry coupling to form the B ring.  Further reactions were then carried out to produce the final product taxol.  The initial retrosynthetic analysis carried out by Nicolaou suggested the following disconnections (5).

taxol_retro1.gif (3417 bytes)
5: Nicolaou's Proposed Retrosynthetic Disconnections.

.The A and C rings were constructed and then fused via the reaction sequence (6).

nicolaou1.gif (3988 bytes)
6: The Nicolaou Route to Formation of the B Ring.

Having created the fused A, B and C ring system Nicolaou and colleagues went on to complete the total synthesis of taxol.  For an in depth discussion of this synthesis see Classics in total synthesis by K.C. Nicolaou10.

ii) The Holton Approach15,16
Holton took a different approach to that use by Nicolaou choosing (-)-borneol as his starting material.  This he converted to an unsaturated ketone (7) over a total of 13 synthetic steps.

holton1.gif (1354 bytes)
7: The Holton Route, Stage 1

This ketone was then converted into -patchouline oxide which Holton then epoxidised and treated with a lewis acid to induce a rearrangement to yield a tertiary alcohol (8)

holton2.gif (1412 bytes)
8: The Holton Route, Stage 2

This alcohol was then epoxidised once more before undergoing a fragmentation reaction to produce the A and B rings of taxol.  The C ring was then introduced using the Robinson-Stork method of annulation (9).

holton3.gif (1916 bytes)
9: The Holton Route, Stage 3

From this building block Holton carried out a variety of other reactions resulting in the full stereoselective synthesis of taxol.  For a more detailed description of this synthesis refer to J.Am.Chem.Soc (1994), 116, 1597 and 159915, 16.

iii) The Danishefsky Approach17
This method of synthesis is the most recent to be published and requires less steps than the Holton or Nicolaou routes.  It is, however, still too complicated to be utilised for large scale production.  The method employed by Danishefsky involved starting with the Wieland-Miescher ketone.  This was then converted to a complex enol triflate containing an olefin on the C-ring which allows for the development of taxol via an intramolecular Heck reaction (10).

danishefsky.gif (2926 bytes)
10: The Danishefsky Route.

For a more detailed description of this synthesis refer to J.Am.Chem.Soc (1996), 118, pp284317.

The Future
Unfortunately the three complete syntheses of taxol that have been devised to date are far too complicated and involved to be utilised for large scale production.   Consequently the only current source of taxol available for treatment of cancer patients is through the semi synthesis method mentioned earlier.

Research into the synthesis of taxol is still ongoing with a number of groups around the world carrying out work in order to develop newer and shorter routes to this natural product, but also with a view to creating a range of structures based on taxol but which may be more biologically active and / or easier to synthesise. Such groups include the Magnus researchers at Austin, Texas, and the Wender group at Stanford who have published their results recently18,19,20,21. Current research by Wender and coworkers involves the use of pinene as a precursor to taxol (11) and research into the development of superior agents.

taxol_retro2.gif (4353 bytes)
11: The Pinene Path to Taxanes21