Paclitaxel

Production

Bark processing

From 1967 to 1993, almost all paclitaxel produced was derived from bark of the Pacific yew, Taxus brevifolia, the harvesting of which kills the tree in the process. The processes used were descendants of the original isolation method of Monroe Wall and Mansukh Wani; by 1987, the U.S. National Cancer Institute had contracted Hauser Chemical Research of Boulder, Colorado, to handle bark on the scale needed for phase II and III trials. While both the size of the wild population of the Pacific yew and the magnitude of the eventual demand for paclitaxel were uncertain, it was clear that an alternative, sustainable source of the natural product would be needed. Initial attempts to broaden its sourcing used needles from the tree, or material from other related Taxus species, including cultivated ones, but these attempts were challenged by the relatively low and often highly variable yields obtained. Early in the 1990s, coincident with increased sensitivity to the ecology of the forests of the Pacific Northwest, paclitaxel was extracted on a clinically useful scale from these sources.

Semisynthesis

Concurrently, synthetic chemists in the U.S. and France had been interested in paclitaxel, beginning in the late 1970s. As noted, by 1992 extensive efforts were underway to accomplish the total synthesis of paclitaxel, efforts motivated by the desire to generate new chemical understanding rather than to achieve practical commercial production. In contrast, the French group of Pierre Potier at the Centre national de la recherche scientifique addressed the matter of overall process yield, showing that it was feasible to isolate relatively large quantities of the compound 10-deacetylbaccatin from the European yew, Taxus baccata, which grew on the CNRS campus and whose needles were available in large quantity. By virtue of its structure, 10-deacetylbaccatin was seen as a viable starting material for a short semisynthesis to produce paclitaxel. By 1988, Poitier and collaborators had published a semisynthetic route from needles of the European yew to paclitaxel.
The view of the NCI, however, was that even this route was not practical. The group of Robert A. Holton had also pursued a practical semisynthetic production route; by late 1989, Holton's group had developed a semisynthetic route to paclitaxel with twice the yield of the Potier process. The main innovation was "Ojima−Holton coupling", a ring-opening method independently discovered by Holton and Ojima. Florida State University, where Holton worked, signed a deal with Bristol-Myers Squibb to license their semisynthesis and future patents. In 1992, Holton patented an improved process with an 80% yield, and BMS took the process in-house and started to manufacture paclitaxel in Ireland from 10-deacetylbaccatin isolated from the needles of the European yew. In early 1993, BMS announced that it would cease reliance on Pacific yew bark by the end of 1995, effectively terminating ecological controversy over its use. This announcement also made good their commitment to develop an alternative supply route, made to the NCI in their cooperative research and development agreement application of 1989.
As of 2013, BMS was using the semisynthetic method utilizing needles from the European yew to produce paclitaxel. Another company which worked with BMS until 2012, Phyton Biotech, Inc., uses plant cell fermentation technology. By cultivating a specific Taxus cell line in fermentation tanks, they no longer need ongoing sourcing of material from actual yew tree plantations. Paclitaxel is then captured directly from the suspension broth by a resin allowing concentration to highly enriched powder containing about 40% paclitaxel. The compound is then purified by one chromatographic step followed by crystallization. Compared to the semisynthesis method, PCF eliminates the need for many hazardous chemicals and saves a considerable amount of energy.
In 1993, paclitaxel was discovered as a natural product in Taxomyces andreanae, a newly described endophytic fungus living in the yew tree. It has since been reported in a number of other endophytic fungi, including Nodulisporium sylviforme, Alternaria taxi, Cladosporium cladosporioides MD2, Metarhizium anisopliae, Aspergillus candidus MD3, Mucor rouxianus, Chaetomella raphigera, Phyllosticta tabernaemontanae, Phomopsis, Pestalotiopsis pauciseta, Phyllosticta citricarpa, Podocarpus sp., Fusarium solani, Pestalotiopsis terminaliae, Pestalotiopsis breviseta, Botryodiplodia theobromae, Gliocladium sp., Alternaria alternata var. monosporus, Cladosporium cladosporioides, Nigrospora sp. and Pestalotiopsis versicolor. However, there has been contradictory evidence for its production by endophytes, with other studies finding independent production is unlikely.