CA2108265C - Methods and compositions for isolating taxanes - Google Patents

Abstract

Methods of obtaining renewable sources of taxanes including taxol are provided. Compositions comprising taxanes which are useful as source materials for the further purification of taxanes are also disclosed. Specifically, a method of drying plant matter to preserve their taxane content and facilitate their extraction is disclosed. In addition, methods of extracting and purifying taxol and other taxanes from ornamental cultivars using a series of organic and aqueous solvents and normal phase chromatography columns are also disclosed.

Description

~~.~8?~

2 ~CT/US92/030$8 , TECHNICAL FIELD OF THE INVENTION
This invention relates to methods and compositions for obtaining crude taxane mixtures from renewable sources of plant matter and a method of purifying such taxane mixtures and compositions to obtain specific taxanes. More particularly, this invention relates to a method of treating plant matter in a manner which preserves taxane content. This invention also relates to a method of extracting taxanes from plant matter with solvents to produce a crude mixture of taxanes wherein the crude taxane mixture is present in an aqueous solvent.
Additionally, this invention relates to a method of purifying specific taxanes from a crude taxane mixture i using normal phase chromatography.
BACICGROUND OF THE INVENTION
Taxanes are alkaloids possessing a taxane nucleus. The taxane nucleus comprises the three ring structure shown below which is also identified as ~,8,12,15,15-pentamethyl-tricyclo (9.3.1.03~$]
pentadecane.

'~'O 92/1&t92 PCT/LS92/03088 V
At least 56 different taxanes have been identified in the literature. For example, in 1980, R.W. Miller authored an article which surveyed known taxus alkaloids and other taxane derivatives and reported for these compounds formulas, HNMR, MS, and x-ray data. The article by R.W. Miller, appearing at J.
Nat. Prod. 43(4): 425-437 (1980)a M.G. Begley, E.A. Freeknall, and G.
Pattenden, Acta Crystalloqr, 40; 1745 (1984); D.P.
Della Casa de Marcano and T.G. Halsall, Chem. Comm.
1382 (1970); D.G.I. Kingston, D.A. Hawkins, and L.
Ovington, J. Nat. Prod. 45; 466 (1982); F. Gueritte-Voeglein, D. Guenard, and P. Potter. J. Nat. Prod.
50:(1):9-11 (1987): B. Lythgoe in "The Alkaloids" Ed.
by R.H.F. Manske, Vol. 10, Academic Press, New~York 1968, pp. 597-626; D.P. Della Casa de Marcano and T.G.

~_t08~65 W~ 92/18492 ' PC.'TI~J~92103J8~ ~r~.~, Halsall, Chem. Commun. 1381 (1970) V. Senilh, S.
Hlechert, M. Colin, D. Guenard, F. Picot, P. Potier, arid P. Varenne, J. Nat. Prod. 47(1) pp. 131-137 (1984);
J.L. McLaughlin, R.W. Miller, R.G. Powell and C.R.
Smith, Jr., ~. Nat. Prod. 44:312-19 (1981); D.P. Della Case de Marcano and T.G. Halsall, Chem. Comm. 365 (1975); C.H. Oliver ~Iuang, David G.I. Kingston, Neal F.
Magri, G. Samaranayake and F.E. Boetner, J. Nat. Prod.
49(4): 665-669 (1986).
The taxane series of molecules possess potent antitumor activity. Generally, the taxanes which have been studied for their antitumor activity have found use in the treatment of ovarian cancer and leukemia (W. P. MacGuire et al., Annals of Internal Medicine, vol. 111, pg. 273 (1989)). Taxanes are believed to exert their antitumor activity by inducing tubulin polymerization and forming extremely stable and nonfunctional microtubules which has an antiproliferative effect on taxane sensitive cells.
(Eric K. Rowinsky et al., Journal of the National Cancer Institute, Vol. 82, No. 15, pp. 1247-1259 (1990)). Among the taxane molecules which have been studied most with respect to their antitumor activity are taxol, cephalomannine, desacetylcephalomannine, baccatin III, 10-desacetyl baccatin III and 10-desacetyltaxol: The structures of these taxanes are Shown belowt J
"""
R~ - AC 8aoatln III
~ 4~0~1 b~oo~h pl 01~
O
Rj-~-NH O
2 0 ~~~ _ O
OH
2 5 ~ -,~, p~: CaHb Taoool*
I~ - ofl, I~ - ca H5 ~ a.o...o~n T.~I
R, - ~ ~ ~ ~ s ~ - aa~s W.yvMro~r.,r,.
Rt - aH, A2 ~ ai aai - ass ~ 'o°, trade-mark A

The taxane compound known as taxol, was first reported to be isolated from the stem bark of the western yew Taxus*brevifolia, a slow growing conifer.
Its structure was elucidated by M.C. Wani et al., Journal of the Americas Chemical Society, Vol. 93, pp.
2325-2327, (1971).
Taxanes are commonly isolated from the bark of T. brevifolia collected in the wild. Because the concentration of specific taxanes in T. brevifolia is extremely low (for example, taxol is present in a concentration of between about 0.004% to about 0.02%
based upon the dry weight of bark), large quantities of trees must be harvested and processed to produce even modest amounts of taxanes needed for research purposes.
Furthermore, wild trees grow under very different conditions resulting in highly variable levels of taxanes produced in the bark. In addition, wild populations of trees are an unreliable source for taxanes because they are plagued with many uncertainties and risks such as forest fires, annual climatic variations, natural variations in taxo~
content in the different chemotypes of wild populations. Increased criticism from environmentalists concerning harvesting of wild plants also threatens the availability of T. brevifolia as an adequate source of taxanes. T. brevifolia, therefore, * trade-mark represents a nonrenewable and inconsistent source of taxanes.
Even more critical is the fact that extensive harvesting of wild trees risks the destruction of the genaplasm essential for the future cultivation of T.
brevifolia. Such harvesting could result in the loss of wild genes coding for proteins providing for such characteristics as disease and pest resistance, cold hardiness, high growth rates and tolerance to full sunlight and the extremes of drought and flooding and high taxol*/taxane content. The preservation of these wild genes will be critical to long-term development of cost-effective taxol* and other taxane production whether produced from cultivated plants, tissue culture or genetically modified microorganisms. Because of the critical role wild germplasm will serve in future production strategies, the preservation of wild populations should be considered an essential component of the development strategy for taxol* and other taxane production.
Since the harvesting of wild populations of - ~ brevifolia yields such a limited supply of taxol*, clinical experiments of taxol have been restricted to only a few specific chemotherapeutic applications.
Lack of a stable and reliable source of taxol* at a predictable cost will also significantly impede clinical utilization of the agent. Development of a * trade-mark sustainable, economic and reliable source for taxanes is imperative.
The potential of taxol* as a cancer chemotherapeutic agent and the structural complexity of the taxol molecule has prompted a large effort directed toward its de novo synthesis. However, the molecular complexity of taxol* suggests that a total synthesis of taxol from readily available raw material is not likely to be economically feasible.
A synthesis of the taxane ring skeleton is reported by R.A. Holton et al., at Journal of the ~°erican Chemical Society, ~ø 5731 (1984), and 'b d, Vol. 110, pp. b558-6560 (1988). However, these syntheses are deficient in that the final product lacks sufficient pharmacological activity to serve as an effective antitumor agent.
Semi-synthesis of taxol using 10-desacetyl baccatin III, a more abundant precursor isolated from the leaves of T. baccata (lg/Kg fresh leaves), has been reported by French workers. Additionally, V. Senilh et al., C.R. Seances Acad. Sci. Ser. 2, Vol. 299, pp.
1039-1043 (1984); F. Gueritte-Voegelin, Tetrahedron, Vol. 42, pp. 4451-4460 (1986): Colin et al. European patent application 0 253 278 and Colin et al. European patent application 0 253 739 refer to the semisynthesis of taxol*from 10-desacetyl baccatin III. These methods use taxane derivatives as the starting materials, and, * trade-mark _ g _ therefore, suffer from the disadvantage that the approach requires isolation and purification of taxanes from a plant source followed by conversion of the purified taxane to taxo~': this multistage preparation of taxol is more expensive than isolation of taxol*
directly from the plant material.
There have been efforts to develop new techniques for isolating taxanes such as taxol~ from plant matter. However, none of the methods reported to date provide for the adequate extraction of taxanes from a renewable source. For example, M.C. Wani et al., Journal of the American Chemical Society, Vol. 93, pp. 2325-2327 (1971), refers to the purification of taxanes l-from stem bark of a u* brevifolia using normal phase column chromatography protocols. Normal phase column chromatography entails the use of a polar column packing to effect molecular separation.
National Cancer Institute Natural Products Branch paper NSC #125973 dated July 15, 1983 also refers to the purification of taxol from bark, wood without bark, branches, twigs, needles, seeds/fruits and roots also from Maxus brevifolia.
A disadvantage of the method for isolating taxol described in the NSC #125973 paper and other publications is the reliance on the use.of methylene chloride in the extraction process. Methylene chloride °nd other chlorinated hydrocarbons, such as chloroform, * trade-mark - 1~ -are recognized to be toxic and potentially carcinogenic. It is, therefore, desirable to avoid utilizing these solvents in the extraction and purification of taxar_es from plant matter, both from the standpoint of exposure of these chemicals to personnel who carry out these procedures and from the standpoint of the potential exposure of patients to these chemicals via trace amounts not removed from the purified taxanes in the final dosage form of the taxane medications.
V. Senilh et al, Journal of Natural Products, Vol. 46, No. 1, pp. 131-137 (1984) refers to the purification of taxol*and cephalomannine from the trunk of a s'~baccata L. The purification method which is reported by Senilh et al. is deficient in that further purification of specific taxanes using IiPLC or crystallization is required even after multiple elutions of taxanes on a variety of normal phase chromatography packings has been performed.
There have been attempts to isolate taxanes from plant matter using reverse phase column chromatography protocols. See, e.g., Keith M. Witherup et al., Journal of Natural Products, Vol. 53, No. 5, pp. 1249-1253 (1990): Keith M. Witherup et al., Journal o_f Liquid Chromatography, Vol. 12, No. 11, pp. 2117-2132 (1989). These attempts are undesirable in that the taxanes are isolated in relatively low yield, * trade-mark ~~! (~~;~~;i ,.;,.."W~ 92/18492 PCT/US92/03088 approximately 50~ based. upon the theoretical yield of taxanes.
Reverse phase column chromatography entails the use of a nonpalar column packing to effect molecular separation. Another disadvantage of reverse phase chromatography is the difficulty in separating taxanes from the aqueous elution medium. The evaporation of an aqueous medium is both expensive and time consuming. Furthermore, the use of an aqueous medium hydrolyzes labile bonds, thereby lowering the yield of taxanes. Also, the aqueous medium used in the reverse phase chromatography step partially epimerizes the c-7 stereocenter. The epimers so produced are undesirable in that they have diminished pharmacological properties. Furthermore, these undesirable epimers are separated from pharmacologically useful taxanes only with additional expense arid difficulty.
Separation is further complicated because desired taxane products coelute during the reverse phase chromatography step. Additional chromatography steps must therefore be performed to completely separate the taxanes, resulting in increased expense and effort.
Another shortcoming of known methods for isolating taxanes from plant matter is that the methods require that the plant matter be in substantially desiccated form. Present methods of drying plant matter promote the degradation of taxanes contained within the plant matter. Drying plant matter, therefore, contributes to decreased taxane yields.
SUMMARY OF THE INVENTION
In accordance with the present invention, there are provided novel renewable sources of taxanes from cultivated varieties of Taxus plants in contrast to wild plants which are a limited resource.
Preferably, the plants to be processed for taxane extraction are provided as intact clippings wherein a substantial amount of foliage is attached to the stems.
The plant matter is subsequently treated with one or more solvents to obtain a taxane-rich composition.
This invention also provides a taxane-rich composition which is prepared from one or more cultivars of ornamental Taxus*plants. The taxane-rich composition is prepared by treating the plant matter from one or more cultivars, provided as intact clippings, with one or more solvents to obtain the taxane-rich product. Prior to solvent treatment the plant matter optionally may be subjected to a drying step or may be ground.
In addition, this invention provides a process for preparing a plant material which can be used as a source for subsequent extraction of taxanes * trade-mark ...,,Vb'~ 92/18492 ? ~. ~ ~ ~ ~ ~ pCT/U~92/030&8 r,,:.:.:

a which process preserves taxane content and increases the accessibility of taxanes to extraction. This process comprises providing intact clippings wherein a substantial amount of foliage is attached to the stems of taxane-containing plant matter, drying the intact clippings to reduce the volatile content of the clippings to produce substantially dry clippings, and recovering the substantially dry clippings.
Preferably, the process of preparing the plant material is conducted at a temperature of less than about 70°C
and is not conducted under unobstructed direct sunlight.
The plant materials, which can be used as a source for taxanes, prepared according to the method of this invention which method comprises providing intact clippings, drying the intact clippings to reduce the volatile components to produce substantially dry clippings and recovering the substantially dry clippings are also a part of this invention.
This invention also relates to a process for obtaining a arcade taxane mixture from taxane-containing plants. The process comprises an initial solvent treatment comprising the steps of:
A. extracting taxanes from plant matter by contacting the plant matter with an organic solvent and obtaining a substantially solvent-free first residue 3~
rich in taxanes; .

21~8'~~
Yd0 92/18492 ?CTlU~92/03088 ,..~:.~
;~; : .w,~

B. partitioning the first residue between an organic solvent capable of dissolving the taxanes and an aqueous solvent which does not dissolve significant quantities of taxanes to produce an organic phase and an aqueous phase;
C. separating the organic phase from the aqueous phase and recovering the organic phase substantially free of the aqueous phase; and D. forming a second residue from the second organic phase; and subjecting said second residue to a finishing treatment.
Taxanes present in the second residue are further purified through a finishing treatment selected from either treatment E or F. Finishing treatment E
a 20 comprises the steps of:
dissolving the second residue in a solvent which allows for the reversible attachment of the taxanes to a solid support, introducing a solid support to the solvent comprising the dissolved second residue and allowing the taxanes of the second residue to attach to said 30 solid support, separating said solid support from the solvent;
35 ' sequentially contacting the solid support with attached taxanes with a series of solvents which solvents have differing taxane eluting properties, and eluting a taxane-rich fraction from the solid support as the crude taxane mixture.
Finishing treatment F comprises partitioning the second residue between an aqueous mixture and a nonpolar nonmiscible organic solvent, wherein the aqueous mixture comprises a single phase of water and a polar organic solvent miscible with water which polar organic solvent is present in the aqueous mixture in an amount sufficient to allow the taxanes present in the second residue to enter the aqueous mixture, and recovering the crude taxane mixture from the aqueous mixture.
In addition to the process of preparing a crude taxane mixture, this invention also includes the crude taxane mixture compositions prepared according to the method disclosed herein.
This invention also provides a process for separating two or more taxanes including taxo~ and cephalomannine comprising the steps of:
providing a mixture comprising two or more taxanes in a solvent suitable for loading onto a normal phase chromatography column, * trade-mark loading the taxane comprising mixture onto a normal phase chromatography column packed with a solid support in a solvent suitable as a mobile phase, and separating the taxanes by eluting the normal phase chromatography column with a mobile phase having a sufficient polarity to separately elute taxol and cephalomannine.
Additionally, the present invention provides a process for separating taxol from cephalomannine using normal phase chromatography wherein a mixture of ethyl acetate and methylene chloride is used as the mobile phase.
A process for obtaining taxanes from taxane-containing plant matter and separating taxanes is also a part of this invention. This process comprises:
providing intact clippings wherein a substantial amount of foliage is attached to the stems of taxane-containing plant matter, drying the intact clippings to reduce its volatile content to produce substantially dry clippings and recovering said substantially dry clippings, extracting taxanes from said substantially dry clippings according to the following steps comprising:
A. contacting the plant matter with an organic solvent and obtaining a substantially solvent-free first residue rich in taxanes:
* trade-mark .,WO 92/18492 ? - ~ ~ ~ PCl'/U~92/03~88 B. partitioning the first residue between an organic solvent capable of dissolving the taxanes and an aqueous solvent which does not dissolve significant quantities of taxanes to produce an organic phase and an aqueous phase;
C. separating the organic phase from the aqueous phase and recovering the organic phase substantially free of the aqueous phase:
D. forming a second residue from the second organic phase;
subjecting said organic phase to a finishing treatment selected from either E. dissolving the second residue in a solvent which allows fox the reversible attachment of the taxanes unto a solid support, introducing a solid support to the solvent comprising the dissolved second residue and allowing the taxanes of the second residue to attach to said solid support, separating said solid support from the solvent, sequentially contacting the solid support with attached taxanes with a series of solvents which solvents have differing taxane eluting properties, and eluting a taxane-rich fraction from the solid support as the crude taxane mixture, or F. partitioning the second residue between an aqueous mixture and a nonpolar nonmiscible organic solvent, wherein the aqueous mixture comprises a single phase of water and a polar organic solvent miscible with water which polar organic solvent is present in said mixture in an amount sufficient to allow the taxanes present in the second residue to enter the aqueous mixture, and recovering the crude taxane mixture from the second aqueous mixture, providing the crude taxane mixture obtained from either treatments E or F, said crude taxane mixture comprising two or more taxanes, in a solvent suitable for loading onto a normal phase chromatography column:
loading the taxane comprising mixture onto a normal phase chromatography column packed with a solid support in a solvent suitable as a mobile phase:
separating the taxanes~by eluting the normal phase chromatography column with a mobile phase having a sufficient polarity to separately elute taxol and cephalomannine.
Accordingly, it is an object of this invention to provide a consistent and renewable source of taxanes and method for efficiently recovering taxanes.
* trade-mark WO 92/18492 ~ ~ ~ ~ ? ~ J Pf'TlU592103088 ~C;r~a;.s.
r;., _ 19 It is a further object of this invention to provide a method for the isolation of taxanes from plant matter that reduces the need for multiple elutions on a variety of normal phase chromatography packings.
It is an additional object of this invention to provide a method for the isolation of taxanes which method avoids or minimizes the use of certain chlorinated hydrocarbons such as methylene chloride and chloroform.
It is also an object of this invention to provide a method for the isolation of taxanes from plant matter that produces taxanes in high yields.
It is another object of this invention to provide a method for the drying of taxane-containing plant matter that minimizes loss of taxanes. Another object of this invention is to provide substantially dry plant materials which can be used as a source for taxanes.
It is a further object of this invention to provide a method for the isolation of taxanes from plant matter that permits the use of simple agitation to remove the organic components from the plant matter.
It is also an object of this invention to provide a method for the isolation of taxanes from plant matter that eliminates the need for filtration steps.

~~.~8~~
iW0 92/18492 FCT/LJS92/03088 ,:.. _;:

a It is an additional object of this invention to provide a method for the isolation of taxanes from plant matter that does not reguire the use of aqueous elution media as found in reverse phase chromatography.
It is a further object of this invention to provide a method for the isolation of taxanes from plant matter that prevents the epimerization of C-7 stereocenter.
It is an additional object of this invention to provide a method for the isolation of taxanes from plant matter in which all taxanes possess substantially distinct retention times in normal phase chromatography.
It is another object of this invention to provide a method for obtaining a crude taxane mixture.
It is a further object of this invention to provide a composition comprising a crude taxane mixture.
DETAILED DESCRIPTTON OF THE INVENTION
Any plant matter which contains taxanes is useful in the methods of the present invention.' Furthermore, the plant matter used may be of pure genetic origin, mixed genetic or hybrid origin, or unknown genetic origin. Plant matter harvested from plants growing in the wild may be used as a source for taxanes. Examples include T. brevifolia, T. baccata, T. cuspidata, and T. wallachiana. Preferably, plant matter from cultivars of plants belonging to the genus Taxus* is used as a source for taxanes and in particular certain varieties or species of cultivated ornamental S
Taxus~ As used herein, "cultivars" means an assemblage of cultivated plants which is clearly distinguished by any characters (morphological, physiological, cytological, chemical, or others), and which when reproduced (sexually or asexually), retains its distinguishing characters. Hortus Third: A Concise Dictionary of Plants Cultivated in the United States, MacMillan Publishing Co., Inc., 1976.
Surprisingly, it has been discovered that several varieties of ornamental a s cultivars have been identified with taxol* content in the fresh leaves comparable to or higher than that of the dried stem bark of ~ brevifolia. Among the identified varieties are: Ti X media "Henryi', ,~ X a 'a 'Runyan', T.
c~3_spidata, ~, X media 'Halloran' , ~ X Media .Hatfield', ~ X ed a 'Hicksii', ~ X a a 'Nigra', ~ X Media 'Tauntonii', ~ X media 'Dark Green Spreader' and cultivated ~ cuspidata 'Hrevifolia', cusDidata 'Spreader'. ether varieties with taxol*
content in the dried leaves comparable to that of the dried stem bark of ~ brevifolia include ~ X me_ dia 'Wardii', ~ X media 'Hrownii', and T. X m_gdia 'Densiformis'.
* trade-mark " Preferred cultivars may be selected from -the group consisting of ~ X media 'Densiformis', ~ X
media 'Hicksii', ~ X is 'Dark Green Spreader', T. X
medi 'Runyan', ~ X med'a Brownii', T. X me_ did 'Wardii', ~ X med'a 'Halloran', T. X me ' did Hatfield , T. X em did 'Nigra', ~ X med'a 'Tauntonii', T.

cuspidata 'Brevifolia', and ~ cuspidata. Most preferably, however, are cultivars whose leaves have a higher taxane content than the dried stem bark of brevifolia are used as a source for taxanes. Most preferred cultivars may be selected from the group 15 consisting of T. cuspidata: ~ X me 'a 'Halloran': T,= X
media 'Hatfield'; ~ X media 'Nigra'; T. X med 'Tauntonii': ~ X edi 'Dark Green Spreader'; and ~ X
e~ did 'Hicksii'. The cultivars stated above are 20 available from commercial nurseries.
The plant matter used in the methods of this invention may be any portion of the cultivar that contains taxanes and may comprise the leaves, stems, 25 branches, bark, roots or mixtures thereof. In the preferred embodiment of this invention, leaves are dried on the small stems as intact clippings prior to stripping in order to preserve their taxol content. As used herein, "intact clippings" is meant to include any clipping in which a substantial amount of the original foliage or leaves remains attached to the stems. A
clipping is also referred to by those in the art as a * trade mark "trimming". In addition, the taxol content of the small stems carrying the leaves is found to be approximately 60-70% of that of the leaves.
The plant matter to be extracted according to the methods of this invention is used in either fresh or dried form. As used herein, "fresh" is meant to include plant matter that retains substantially all of its volatile content. Preferably, the plant matter is dried at a temperature of. less than about 70~C under either fully lighted, shaded or dark conditions such that the drying is not conducted~under unobstructed direct sunlight. As used herein, "fully lighted" is meant to include any condition in wh~ch ambient visible light, such as sunlight or that found in a greenhouse or artificially lighted building is not substantially blocked, reflected, or otherwise prevented from reaching the plant matter. More preferably, the plant matter is dried at a temperature of between about 65'C
and about 20'C under shaded or dark conditions. Dry air or low humidity are preferred conditions for drying. Preferably the relative humidity is not above about 80%: more preferably not above about 50% relative humidity. As used herein, "shaded conditions" is meant to include an environment where between about 40% and about 70% of external unobstructed visible sunlight is blocked, reflected, or otherwise prevented from reaching the plant matter that is being dried.
* trade-mark '' Alternatively, exposure of plant matter to light which light is absorbed by the plant and converted to heat, causing the temperature of the plant to rise above about 70°C should be avoided. Preferably, at least a 25% reduction in the weight of the plant matter is obtained during the drying step. In a more preferred embodiment, the plant matter is dried until between about a 40% and about a 70% weight loss is obtained.
while not wishing to be bound by theory, it is believed that drying the plant matter under the conditions of this invention substantially preserves 15~ the taxane content of the plant matter by limiting their decomposition. In addition, it is also be~~eved, while not wishing to be bound by any theory, that the method of drying the plant matter according to the present invention increases the accessibility of taxol~
and other taxanes to extraction.
Those of skill in the art will recognize that there is a relationship between the temperature of ZS drying and the duration of drying. For example, drying at a relatively higher temperature will require a shorter period of time than drying at a relatively lower temperature. Any~method of drying the plant matter to a weight loss as described above, while leaving the taxane content of the plant matter substantially unaffected, is withi.~.~. the scope of this * trade mark . W~ 9Z/18492 ~ -~ ~ ~ ~ ~ '~ PCfI(JS92/03088 .:,~:

invention. Such methods may include freeze-drying and S
microwave drying.
While the pressure at which the plant matter is dried is not deemed to be critical, it is understood that the use of decreased atmospheric pressure can promote drying at lower temperatures or shorter periods of time. The practitioner may therefore make a selection among various temperatures and atmospheric pressures of drying without departing from the scope of this invention.
Plant matter, prepared according to the process of this invention, comprising substantially dry intact clippings provide a plant material also a part of this invention which can be used as a source for taxanes. Following the drying step, the clippings are recovered for subsequent use as a source of taxanes.
While the plant matter to be extracted may be used, in substantially intact form, it is preferably manipulated to increase the surface area of the plant matter and thereby increase the rate of taxane extraction. More preferably, the plant matter is cut, crushed or manipulated to form a powder. Most preferably, the plant matter is ground either by hand or by mechanical means. Suitable methods of grinding include the use of a blender, ball mill grinder, Wiley Mill or Fritz Mill. Other grinding apparati and ',~r~.~~~.~5n~~~
WO 92/18492 PCT/US92/03088 ,,f~..
t, ::::

methods may be used without departing from the scope and spirit of this invention.
Preferably, the plant matter is ground to a particle size of between about 40-80 mesh. More S
preferably, the plant matter is ground to a particle size of about 60 mesh.
To extract the taxanes, the fresh or dried plant matter to be extracted, in any physical form described above, is subjected to one or more initial solvent treatments comprising contacting the plant matter with an organic solvent to form an extract from which a substantially solvent-free first residue rich in taxanes may be obtained. Multiple extractions of.
the plant matter with the organic solvent may enhance taxane extraction. Preferably, the weight: volume ratio of plant matter to organic solvent ranges from about 1:8 to 1:12. A more preferred weight:volume ratio is about 1:10. A weight:volume.ratio of plant matter to total volume of organic solvent used to extract the plant matter ranges from about 1:10 to.about 1:150. A
more preferred weight:volume ratio of plant matter to total volume of organic solvent is about 1:40.
Preferably, contact of the plant matter with the organic solvent is promoted by percolation or soxhlet extraction. More preferably, contact is promoted by shaking. Most preferably, contact is accomplished by agitating the plant. matter as it soaks r WG~,92/18492 PCT/US92/03088 in the organic solvent. In this step, substantially all of the extractable organic matter, including taxanes, is removed from the plant matter and carried into the organic solvent.
The organic solvent which is contacted with the plant matter is preferably one commonly used with extracting organic chemicals from plant matter. 1?referred organic solvents include chloroform, carbon tetrachloride, ethanol, acetone, ethyl acetate, methylene chloride, methanol, methyl ethyl ketone, methyl isobutyl ketone, methyl t-butyl ether, or mixtures thereof. Preferably the organic solvent is ethanol. ~~he~ore, it has been found that between about three to ~'~out six vo7.umes of between about 40 ml and about 200 ml of t:he organic solvent, applied sequentially over a period of about 8 hours to about 48 hours, are sufficient t:o remove essentially 100% of all taxanes from about 10 grams of plant matter. Also, it has been found that between about three to about six valumes of between 2 liters and about 5 liters of the organic solvent, applied . sequentially over a period of about 8 hours to about 48 hours are sufficient to remove essentially 100% of all taxanes from about 100og of plant matter..
In a separate preferred embodiment, the plant matter is-optionally first defatted with hexane, hexanes, pentane, petroleum ether, isooctane, or mixtures thereof, followed by contacting the plant ~~.5~~5 WO 92/y8492 PC1'/US92/03088 _~;, _ 28 _ " matter with the organic solvent. For example, it has been found that the use of two volumes of appraximately 100 ml hexane is sufficient to defat about l0 grams graund, dried plant matter over a period of between about eight hours and about two days.
In a more preferred embodiment the plant material is first defatted with hexane then extracted by shaking with the organic solvent, ethanol, acetone ar ethyl acetate, for a period of 24 hours, changing the solvent four times during this period.
Those of skill in the art will recognize that the multiple uses of relatively smaller volumes of solvent is more effective in extracting materials than the single use of a larger solvent volume. Thus, the practitioner may depart from the solvent volumes used in each of the steps involving extracting, partitioning, or otherwise removing organic materials described herein, as well as the number of iterations of extraction, without departing from the scope of the invention.
Hydrophilic moieties extracted from the pJ.ant matter by the organic solvent are then separated from the taxanes and removed by partitioning the extracted plant matter obtained from the organic solvent between a suitable pair of organic and aqueous solvents such that the taxanes remain in an organic solvent and the hydrophilic moieties are removed in an aqueous solvent.

~~~;<-::WO 92>18492 c~ ~ ~ . ~ PC~C/US921030R~

Organic and aqueous solvents which are immiscible with one another and may be separated from each other following partitioning to the extent that the purity of one phase is not substantially contaminated by the other, are suitable, provided that the taxanes have sufficient solubility in the organic solvent to remain in the organic solvent. Furthermore, the suitable aqueous solvents should not dissolve significant quantities of taxanes. Accordingly, in this and any subsequent step involving the partitioning of the taxanes between two different phases or their further purification, any solvent from a previous step remaining with the taxanes being purified, which solvent interferes with the action of solvents in a subsequent step is removed. For example, a solvent 2p from one step which interferes with the partitioning of an organic and an aqueous solvent in a subsequent step by increasing the miscibility of the organic and aqueous solvents, is removed prior to contacting the taxanes with the organic and aqueous solvents to be partitioned.
Preferably, following extraction of the plant matter with the organic solvent, the extract is 3p evaporated to dryness forming a first residue, with or without heating to promote the rate of solvent evaporation. If heating is used, the temperatures which promote epimerization of the taxanes should b,e 2~.~a?~~
WO 92/18492 9~CT/IJS92/Oa088 a avoided. Preferably, temperatures should not exceed about 35°C to about 40°C.
Once the solvent has been evaporated, the material remaining as the first residue is dissolved in an organic solvent such as methylene chloride or, more preferably, ethyl acetate to form a second extract. A
weight: volume ratio of first residue to organic solvent of about 1 to about 10 to 20 is preferred. For example, one gram of residue would be dissolved in about 10 to 20 ml of organic solvent. The second extract is washed by partitioning with an aqueous solvent component which does not dissolve significant - quantities of taxanes to remove any hydrophilic moieties that may remain in the second extract. Water, similar polar solvents or mixtures thereof are suitable as the aqueous solvent. Preferably the aqueous solvent is water. A volume of aqueous solvent of about one half of that of the organic solvent is generally sufficient to wash the second extract. The aqueous solvent is then separated from the second extract. The aqueous solvent may be discarded.
The second extract, contains the taxanes.
The second extract is optionally treated with a drying agent to remove water found in the extract, followed by removal of the drying agent and solvent evaporation to form a second residua. Preferred drying agents are anhydrous magnesium sulphate, 4A molecular sieves, .a "
WO 92/18492 PC?/US92/03088 calcium chloride or mixtures thereof. A more preferred drying agent is anhydrous sodium sulphate. The practitioner may find it advantageous to add.a small volume of an organic solvent to the drying agent before combining the taxane containing second extract and the drying agent. Adding an organic solvent to the drying agent tends to prevent taxanes from adhering to the drying agent, and thereby improves yields. The drying 1'0 agent may be removed by gravity filtration, vacuum filtration, or by decanting the solvent from the drying agent. Another method to remove residual water from the second extract is to cool the second extract to a temperature at which the water in tha second extract is in the frozen state without freezing the organic solvent. The liquid nonaqueous material could then be ~0 decanted separating it from the aqueous material.
Further purification of the taxanes is accomplished according to the invention by subjecting the second residue to a finishing treatment.
According to the method of one finishing treatment, the second residue is dissolved in a solvent, e.g., ethanol, which allows for the reversible attachment of the taxanes onto a solid support. A solid support is then introduced into the solvent comprising the dissolved second residue and components from the second residue including the taxanes are allowed to attach to the solid support.
Following separation of the solid support from the solvent, preferably by evaporating the solvent so as to leave a residue on the celite* which residue forms a coating comprising the taxanes, the solid support is sequentially eluted with a series of solvents Which results in the elution of a taxane rich fraction. An example of a suitable solid support is celite.
Preferably the celite is coated with components of the second residue which have been dissolved in an organic solvent comprising a mixture of methanol and ethyl acetate by evaporating the solvent from the mixture of solvent, second residue and celite. Preferably the -methanol -and---ethyl--acetate-wa~rewpresent in a volume- , volume ratio of 1:3.
The coated celite is preferably eluted first with --hexane ---which- does not----sigra'if-icantly elute the taxanes but elutes other less polar molecules. A crude taxane mixture of the present invention is then eluted from the celite preferably with ethyl acetate, methylene chloride or another similarly polar solvent, which is more polar than the first solvent. -Another finishing step to further purify the taxanes present in the second residue comprises partitioning the second residue between an aqueous mixture and a nonpolar nonmiscible organic solvent.
.
preferably, the second residue is first dissolved in the aqueous mixture which is then combined with the nonpolar nonmiscible organic solvent. Preferred nonpolar nonmiscible organic solvents are such solvents * trade~rcark ,:a::,~'~ 92/18492 ~ '~ ~ .~ ~) f FC?/1J~92f03088 N . ~. .. r,1 ~~~ ~' - ~3 -as hexane, pentane, petroleum, ether, heptane, iso-octane, and mixtures thereof. preferably, the nonpolar nonmiscible organic solvent is hexane. A preferred ratio of nonpolar nonmiscible organic solvent to aqueous mixture is 2 to 1. The aqueous mixture is a single phase aqueous mixture comprising water and a polar organic solvent miscible with water. The polar organic solvent is present in the aqueous mixture in an amount sufficient to allow the taxanes present in the second residue to enter into the aqueous mixture.
Preferably, the aqueous mixture comprises more than abcsut 50% polar organic solvent and the balance water.
More preferably, the aqueous mixture comprises about 9 parts polar organic solvent and about 1 part water. In a preferred embodiment, the polar organic solvent is acetonitrile. In a more preferred embodiment, the polar organic solvent is methanol..
The aqueous mixture comprising the crude taxane mixture of the present invention is separated from the nonpolar nonmiscible organic solvent. The aqueous mixture may be evaporated to remove the polar organic component and the crude taxane mixture may be combined with an organic solvent to further purify the taxanes. Because the aqueous mixture is predominantly acetonitrile or methanol in composition, it is preferably evaporated directly to form a third residue comprising the crude taxane mixture. Heat, vacuum and combinations thereof may be used to promote evaporation, provided that the temperature of the aqueous mixture does not exceed about 40'C. The direct evaporation of the aqueous mixture avoids the possibility that any taxanes might be lost in unnecessary extraction steps.
Any of the foregoing initial solvent treatment steps or finishing treatment steps of the process to obtain the crude taxane mixture may be repeated one or more times. For example, any of the foregoing steps may be repeated at least one, two, three or four times.
Specific taxanes may be purified from the crude taxane mixtures of this invention by any commonly used purification protocols such as crystallization or trituration. Preferred taxanes which may be present in the crude taxane mixture and may be further purified according to the methods of this invention include taxol, cephalomannine, desacetylcephalomannine, baccatin III, 10-desacetyl baccatin III and 10-desacetyltaxol. Crystallization may be performed by allowing crystal formation from a substantially saturated solution of the crude taxane mixture.
Alternatively, crystals may be grown by adding a seed crystal of the desired taxane product to a substantially saturated solution of crude taxanes.
Ideally, a single solvent is used as the *trade-mark WO 92/18~b92 ~ ~ J P~I'/US92/03088 ~'i2,r ~1 4 ~ i ~.e) i..:-v.r; ':

crystallization solvent, although the use of solvent mixtures is contemplated by this invention. The temperature of the crystallization solvent or solvents is not critical so long as' substantially pure taxanes result.
According to the present invention, specific taxanes may also be purified from the crude taxane mixture using a series of normal phase chromatography columns. The specific taxanes are eluted by passing solvent mixtures having progressively increased elutrophic power through the columns. Generally, in normal phase column chromatography, the mobile phase comprises a two component system. The first component is a nonpolar organic solvent which controls the rate of compound elution. The second component is a polar organic solvent which elutes the compounds to be separated. Increasing amounts of the second component in the mobile phase generally decrease the contact time of the eluting compounds with the column packing.
Higher concentrations of the second component therefore have the effect of decreasing the elution time.
The separation of taxanes may be improved by using a mobile phase that follows a gradient of increasing elutropic power, although isocratic elution is also contemplated by this invention. Preferred normal phase column chromatography packing materials for one or more of the chromatography columns may be ~~a~.~y,~~'J:~
i3~0 92118492 PCT/ iJS92J03088 ,,....

selected from the group consisting of silica gel, florisil, alumina, and celite. A specifically preferred normal phase column chromatography packing material is silica gel 60, 230-400 mesh, manufactured by E. Merck, and available from Brinkman Instrument Co., Westbury, N.Y.
To prepare the crude taxane mixture, third residue, for running on a chromatography column, the third residue may be triturated (digested) repeatedly (1-5 times) with an organic solvent which dissolves the taxanes. Such an organic solvent may be selected from the group consisting of ether, methylene chloride, methanol, chloroform, ethyl acetate and acetone. The preferred organic solvent is methylene chloride. The third residue dissolved in the organic solvent is combined with a solid support such as diatomaceous earth or celite and evaporated to dryness, thereby adhering the taxanes to the solid support.
The taxane-containing solid support is then loaded directly onto the top of a packed normal phase chromatography column.
' Once the solid support has been placed directly on top of the packed normal phase chromatography column, a mobile phase capable of separating the taxanes from one another and from other plant components is passed through the chromatography column. The mobile phase comprises (1) a nonpolar ~e~o 9ai~~n~2 ~crius9ze~o~og8 ,.. ~.;
"... . ;
° 37 -" component selected from the group consisting of hexane, petroleum ether, iso-octane, and solvents having similar polarities, and (2) a polar component selected from acetone, ethyl acetate, ether, methyl t-butyl ether, chloroform, and solvents having similar polarities. Preferably, the mobile phase comprises hexane and acetone.
In a preferred embodiment, the mobile phase comprises initially between about 60% to 85a hexane and between about 15% to 40p acetone and follows a gradient of increasing elutropic power. The mobile phase can have any final composition so long as effective separation of taxanes is accomplished. Most preferably, however, the mobile phase has an initial cancentration of about 75% hexane and about 25%
acetone, and a final concentration of about 0% hexane and about 100% acetone. In the case of a hexane/acetone mixture, it is the hexane which controls the rate of elution and the acetone which elutes the taxanes. The desired taxanes are eluted from fractions comprising about 60% hexane and about 40o acetone, and about 0% hexane and about 100% acetone. Alternatively, any other series of solvent mixtures having polarities similar to those of the hexane/acetone mixtures described above are suitable to elute the taxane molecules from the first normal phase chromatography Column.

Preferably, the normal phase chromatography column is run under pressure. For a 5 X 16 cm column containing 160g silica gel 60 the flow rate is preferably about 100 ml/5 min.
The presence of taxanes in the fractions collected from the columns may be detected by thin *.
layer chromatography ("TLC") using silica gel G W254 (Machery Nagel, Duren) using 5% methanolJchloroform as a developing system and p=anisaldehyde/
sulfuric acid as a visualizing reagent. Taxo~'and cephalomannine appear as a bluish-grey spot with an Rf value of 0.62 in ,the fractions eluted with hexane/acetone ratios of 60:40 and 55:45. Baccatin III
appears as a blue spot with an Rf value of 0.55 in the fractions eluted with hexane/acetone 50:50 and 45:55.
The taxane 10-desacetyl baccatin III appears as a purplish spot with an Rf value of 0.28 in the fraction eluted with hexane/acetone 0:100.
The hexane/acetone solvent from the taxane rich fractions is evaporated to dryness to form a taxane-rich product., Further purification of the taxanes taxol and cephalomannine may be achieved by loading the taxane-rich product comprising taxol and cephalomannine onto another normal phase chromatography column comprising silica gel 60*or similar material packed with about 1%
methanol and 99% methylene chloride. The elutropic * trade-marks power of the mobile phase is increased by increasing the concentration of methanol in increments of about 0.5% methanol, until the mobile phase has a composition of about 2.5% methanol, 97.5% methylene chloride. The mobile phase passing through a 2 cm X 40 cm column preferably has a flow rate of about 8 ml/min. The taxanes, taxol* and cephalomannine, are collected from fractions eluted with about 2.5% methanol and 97.5%
methylene chloride. Alternatively, any other series of solvent mixtures having polarities similar to that of the methanol/methylene chloride solvent mixtures is 15 suitable to elute the taxanes from the second normal *
phase chromatography column. The taxol-cephalomannine rich fractions from the methanol/methylene chloride column are evaporated to dryness to form another taxane-rich product.
Surprisingly, taxol may be separated from cephalomannine according to the method of the present invention by providing a mixture comprising two or more taxanes in a solvent suitable for loading onto a normal phase chromatography column. The taxane comprising mixture is then loaded onto a normal phase chromatography column packed with a solid support in a solvent suitable as a mobile phase. The taxanes, including taxol*and cephalomannine are then separately eluted with a mobile~phase having a sufficient polarity to separately elute taxol and cephalomannine.
*trade-mark preferably, a mixture of ethyl acetate and metHylene chloride is used as the mobile phase. According to a preferred embodiment of the present invention, the second taxane rich product is dissolved in a mixture of about 20% ethyl acetate and 80% methylene chloride and loaded on a normal phase chromatography column packed With silica gel 60, or similar solid support as described above, in the presence of a mobile phase comprising about 20% ethyl acetate and 80% methylene chloride. The elutropic power of the mobile phase is increased by increasing the concentration of ethyl acetate in increments of about 5% ethyl acetate,. until . the mobile phase has a composition of about 50% ethyl acetate, 50% methylene chloride. A flow rate of about 8 ml/min is~preferred for a 1 cm X 32 cm column.
Substantially, pure taxol is collected from fractions eluted with about 45% ethyl acetate, 55% methylene chloride. Substantially, pure cephalomannine is collected from fractions eluted with a mobile phase comprising about 50% ethyl acetate and 50% methylene chloride. Alternatively, any other series of solvent mixtures having polarities similar to those of the ethyl acetate/methylene chloride mixtures described above are suitable to elute the taxane molecules, taxol and cephalomannine, from the normal phase chromatography column.
* trade-rrerk Depending on the degree of separation of taxanes present in the crude taxane mixture achieved as a result of the initial normal phase chromatography step, additional chromatography of the taxanes using ethyl acetate and methylene chloride as the mobile phase, surprisingly, may separate and purify the t taxanes, including the separation of taxol from cephalomannine, obviating the need for intermediate chromatography using methanol-methylene chloride.
Any of the foregoing chromatography steps may be repeated one or more times, for example one, two, three or four times, to further purify or separate the taxanes at a particular~step.
Although column chromatography is a preferred method, any other chromatographic configuration, such as radial normal phase chromatography, as performed on a chromatotron, are useful in the methods of the present invention. Also, other suitable detection means may be used to identify taxanes without departing from the scope of the invention.
EXAMPLES:
Example 1: Quantification of taxol content of different varieties of cultivated ornamental Taxus.
The fresh leaves from clippings of various cultivated varieties of ornamental Taxus were analyzed *trade-mark for their taxol content following the procedure outlined below:
Extraction a. Leaves which had been stripped from clippings were well mixed and weighed.
IO
b. Ten gram samples of each variety of cultivated ornamental Taxus were blended in a blender with 1Q0 ml 95% ethanol, and the mixture transferred quantitatively into a 250 ml Erlenmeyer flask. The leaves and ethanol were allowed to extract by percolation for 24 hours, and then filtered with 15 rinsing (1 X.25 ml). The percolation process was repeated three additional times.
c. The filtrates were combined (500m1) and evaporated to dryness under vacuum at a temperature not 20 exceeding 40'C. The weight of the ethanol extract, or first residue, was recorded.
P~_eparation of the ethanol extract for HPLC analysis 25 a, Approximately 100 mg of the extracts were weighed into screwcap vials. The caps were lined with a piece of aluminum foil to avoid contact of the extract with the cover-lining during partitioning.
b. The extracts were partitioned between water (1 ml) and methylene chloride (2 ml x 5) or until the methylene chloride became colorless. The combined methylene chloride washes were pipetted into 25 ml *trade-mark ?:~ ~~?~5 ,:~?oW~ 92/18497. PCT/US92/!~3088 Erlenmeyer flasks and evaporated to dryness to produce a second residue. The weight of the second residue was recorded.
c. The second residue was dissolved in ethyl acetate (5 ml) and methanol (2 ml) with sonication. Celite (200 mg to 650 mg) was introduced and allowed to be coated with the extracted material.
The solvent was evaporated to dryness in vacuo to produce a celite material coated with the second residue.
d. The residue-coated celite was ~antitatively transferred into a petri-dish. The dish was left under a hood until the last traces of solvent had been removed (10 minutes). The final product was triturated until a free flowing, uniform powder was obtained. The resulting powder was packed into a Pasteur pipet that served as a column.
e. The column was eluted with hexane (7 ml, or until' the eluent was colorless), followed by methylene chloride (5-6 ml).
f. The methylene chloride eluent was evaporated in vaeuo, and the weight of the remaining residue was recorded.
g. The remaining residue was dissolved with sonication in 1 ml methanol (HPLC grade). The solution was withdrawn in a 1 ml syringe and filtered through a ~'~ 92/18492 PCT/IJS92/03088 Millex filtering unit (0.45~C) into a Wheaton vial, and capped.
HPLC conditions Column: ~. Bondapak C-18 (10~C, Waters Associates) Mobile Phase: methanol/water (65:35) Flow rate: 1.2 ml/minute Detection: U.V. at 227 nm AUFS: 0.1 Amount in~Lected: 10,1.
Results of these analyses are shown in Table 1.

*
Table 1 Taxol Content in fresh leaves of certain Taxus Cultivars Cultivar % Ethanol %Taxol~
Extractives Average + S.D.' ~. X media 'Henryi' 17.63 0.00272 ~ X media 'Densiformis' 18,93 0.00342 ~ X media 'Hicksii' 15.43 0.01042 ~ X media 'Dark Green Spreader' 15.32 O, ~ X m i 'Runyan' 17.04 0.00572 ~ X m~ia 'Brownii' 19.55 0.00272 T X m is ' Wardii' 14. 89 0.00522 T cu~idata 'Brevifolia' 14.100.36 0.00970.0037 C.V.=3896 ~ X m i 'Brownii' 16.07~0.93 0.00465~0.00096 C.V. =20.696 1 S ~ ~s 14.7310.46 0.0141 ~0.00226 C.V.=1696 ~ X media 'Densiformis' 13.3 f 1.36 0.00594~0.00039 C.V. =6.596 ~,, X media 'Halloran' 14.79~0.18 0.01183~0.00057 C.V.=4.896 ~ X media 'Hatfield' 12.83~0.29 0.0121 ~0.00042 C.V.=3.5%
~ X media 'Hicksii' 15.67~0.29 0.0152~0.0023 C.V.=15%
2 5 ~ X ~ ~Nigra'4 15.1310.23 0.0291 ~0.0057 C.V.=19.6%
T baccata 'Repandens' 12.83 f 0.57 0.00178~0.00074 C.V.=41.5%
~ n=3, except for ~ X media 'Hicksii', 'Fairview', C.V. - coefficient of variation.
'Wardii' and 'Spreader' where n=4.
Single values based on single determinations.
Synonyms = 'Tautoni', ~ cusgidata 'Tauntonii' Some authorities treat as ~ baccata 'Nigra' Vegetatively propagated unnamed cultivar of cuspidata.
*trade-mark T. X m is 'Tauntonii'3 16.47~0.92 0.0198~0.00245 C.V.=129 T X m is 'Hicksii' 19.71 ~ 1.01 0.0093+0.0013 C.V.=14%
S
T. X m is 'Fairview' 19.25~0.38 0.0013~0.0004 C.V.~30.7%
T. X m is 'VVardii' 18.97~0.45 0.0084~0.0036 C.V.=43%
1 Q T a i ata 'Spreader' 19.52 ~0.49 0.0032 ~0.00036 C.V.=llY6 Example 2: Effect on Taxol recovery of stripDincr leaves from stems before drvin~ leaves.
The fresh leaves (100g) from clippings which 15 were representative samples from large collections (3-6 lbs.) of certain cultivars were analyzed for their taxol content using the method described under Example 1. Table 2 shows the percent taxol extracted from fresh leaves compared to that extracted from dry 20 leaves. The calculated taxol~content is the % taxol*
expected in the dry plant material based on the % taxol content determined for the fresh material and the %
moisture of the plant material. Percent Taxol~
(calculated) - % Taxol fresh /[1-(% moisture/100)].
25 The % moisture is determined from the difference in weight of the plant material before and after drying.
A ratio of found dry % taxol content to calculated dry % taxol content was then calculated (F/C). According to this calculation, a number less 30 than 1.00 indicates a loss of taxol in the drying procedure (e.g., 0.90 indicates a 10% loss), and a number greater than 1.00 indicates that the drying method yields a quantity of taxol greater than is found by extracting fresh clippings (e.g., 1.20 indicates a 35 20% gain).
*trade-mark " 2(a) Fresh clippings of the same cultivars were separated into leaves and small stems. A portion (log) of the leaves of each cultivar were extracted and analyzed as described under Example 1. Results are shown in Table 2(a). Another portion of the leaves of each cultivar was allowed to dry at room temperature until na further weight loss was observed. A sample of the dried leaves was ground in a Wiley Mill; and a log portion was extracted and analyzed as described under Example 1. The results are shown in Table 2(a).
2(b) Alternatively, clippings of ~ X media 'Dark Green Spreader' were dried intact (i.e., without separating the leaves from their stems) at room temperature until no further weight loss was observed.
The dried leaves were then stripped from their stems, ground in a Wiley Mill* and a lOg portiow was extracted and analyzed as described under Example 1. The results are shown in Table 2(b).
The data of Table 2(b) were obtained by analysis of 10 replicates of a large sample of plant material. The 131% retention in taxol content obtained according to the method in 2(b) compared with a mean .
33% retention in taxol content obtained according to the method in 2(a) is reproducible. Surprisingly, drying according to the method of retaining the clippings intact enhances the recovery of taxol . ..
compared to ethanol extractions of leaves dried after they are stripped from their stems. An almost 4-fold increase.in taxol retention occurs by drying the .
clippings intact as compared to drying after stripping the leaves.
* trade-mark f~fN 1 tD P m e'1 \ f'1a f'7N N f'fN Y1 .

V. O O O O O O O r-1 !~

_ D

O

d O

O

*

~~ O rf 0~r1 P N 1 O r O U ~ aoo ~D n i ~D

x .a o o N N ,.~1 0 ..nl ~

x~- 0 0 0 0 0 1 0 0 0 ao d o N

A O

~

1 ~ o -~ w .. ~ ~, a .c~ a N o ' v7 O N f'fCv 1~ f'1iC1N f'1N
~ C

.. 0l x O O O O O O O O f"1 ,~ O

o O o o O o O o O

O x~- 0 0 0 0 0 o O o O

U

N N

d x > _ ~o E O y C U

t -~r L it O vt1n O ~rO t~ t~

~ it m x e n 1. o o~o n ~c Iw .~ 47 .-~N rl N ~1N .1 n-i d U

O
1"~

O

N

O N N .~n a t aDaD n a~

> -~ W D vDv0 vo tn~t1111v0v0 'O

4l U1 -~ N

L O

O O

tl!

'O dl * O

~0 Ip "~ CO IHV ~D P 1 P N m o O O 1 x O O . O O O
~ ~ o 0 -tt0 0 0 0 0 E ~

r 0 0 0 0 0 1 0 0 0 a. d c w rA

-~~ ar c o .N

c ~

c r ~ ~c o m n o w o eo .. y 1J
V Li7 1~ ~ If11t1 P ~ 01 X

rr ta7 .~ .i.i .i .a.r.~ .i x t0 N !r 1r E-~ d d p fl E ..a ' H

W0 -.. Wo ~ w ~ ~ ~ ~

N -..i .. p~ t O -.iC~
~ 1 ~
r ~ ~ ~ N ~ C H

4J ~. l. N ~C = 'p.!L b r i -a ~ t G V fr t ~ O L it cC

3 S ~ -~ a~ a~..,~ ~ A s~ A to N

~ ' = o x v a f..m 3 c v ~ ~

Example 3: Determination of taxol content of fresh stems and stripped dried stems.
* , The percent taxol content of fresh stems (100g) from clippings of certain * ultivars was analyzed and compared to the percent taxol content of stems (lOg) which were dried and ground as described under Example 2(a). The fresh and ground stems were extracted and analyzed as described under * xample 1.
Results are shown in Table 3. The % taxol calculated was determined as described under Example 2. Also, the ratio of found dry % taxol content to calculated dry %
taxol~content (F/C) was determined as described under Example 2.

*trade-mark N i 1 I
vt'1 ar N v e~ 1 ~ 11 U v O r O ~D 1 1 O .-~ O O O i 1 O
'D

d ~o O 1H If1riW D 1 47N

x wo v ~nn m t ~Io A o o .~N o 1 0 .., .-, E. o 0 0 0 0 1 0 0 A

a V . 1 1 x-- 0 0 0 0 0 1 0 0 a~

al a a w o . o ov r o00 ~ o l in to o c x N a o .-m n co~ .r ~

a o A o o .~0 0 0 ~ o c O sr ~ E 0 0 0 0 o o ~ o ~

~

b > D x O O O O O O 1 O

t N

a ~

N V

p1c = o N iwo N o~.-an y e 1S ~x r~ a .~ ...,..~ ~, .,.~.., w~

c x oc ~w y .., c b o V w * -i ~
U ~

uw D O~co O ~t.aao ~ u~

2 0 H s a ae o ' o ..~

o 0 o N o e~ 1 0, a .e x n N ~oo ~ ~ o v a A o 0 0 ., 0 1 0 0 o H o 0 0 0 0 1 0 0 w 1 25 or x o 0 0 0 0 1 0 0 as d ~
>

~ N n It10t G

J ~ N v0 ~
l .

rl rl~ ~ O N O

wy x xw w cd -E v ~ H
'C

fr O -..~Ir c -.i r0 ...1w .~p~ C 0 ~ ~ I
it >

-- ~ N ~ ~C >. C 3 '0'd a W

cd U O

CJ C!1 iC 7 A . yOH
- i O r D ft E m 3 v . _ . _ _ _ _ _ _ " Exam_ple 4: ra) Determination of taxol content of leaves of non-stripped barn-dried clippings.
Fresh intact clippings (stems + leaves). of T.X media 'Densiformis' (Table 4(A)) and T_.X media.
'Dark Green Spreader' (Table 4(B)) were placed on aluminum wire cloth (18 x 16 inch, 0.010 gauge window screen) attached to a wooden frame. The frames were placed in a drying barn on supports 30" above the floor. The plant material was allowed to dry in the' dark without additional heat. A fan provided air movement: The drying apparatus used is a "Roanoke"*
style model 7,5-I-G Bulk Curing Barri (Gregory Manufacturing Company, Inc., Lewistown, Woodville, NC
27849). The temperature ranged between 30'C at 8 a.m.
to 40'C at 3:30 p.m. Drying lasted for 2 days. The moisture of the leaves prior to drying was determined to be 64.5% (~, X ed a 'Densiformis') and 66% (~ X
_ media 'Dark Green Spreader') based on the difference in weight of the leaves before and after drying.
Immediately before analysis, the dried leaves (10 gms) were recovered and stripped from their stems, ground using a Wiley mill; and extracted by percolation with ethanol, acetone or ethyl acetate. Analysis was carried out as described under Example 1.
The taxol content of the dried leaves was compared with fresh leaves of the same variety which were processed immediately after stripping by blending with the extraction solvent. Results are shown in Table 4. Calculated dry % taxol content was determined according to the method described under Example 2. A
ratio of found dry % taxol content to calculated dry taxol~ content (F/C) was then calculated according to the method described under Example *.
jbl Determination of taxol content of leaves of non-stripped a~reenhouse-dried clippincts.
* trade-marks Fresh intact clippings (stems + leaves) of T. X media 'Densiformis' and T_. X media 'Dark Green Spreader' were placed on aluminum wire cloth (18 x 16 inch, 0.010 gauge window screen) attached to a wooden frame. The frames-were placed on supports 30" above the floor. The plant material was allowed to dry in a polyethylene covered greenhouse (length 60', width 20' and height 9') approximately 55% shaded (using a shade fabric) and ventilated with two Arvin air coolers, fan only, without adding water to the cooling pads. The-temperature ranged between 15.5'C at 8 a.m. to 44'C at 3:30 p.m. Drying lasted for 6 days. The moisture of the leaves prior to drying was determined to be 62%
X media 'Densiformis') and 66% (T. X media 'Dark Green Spreader') based on the difference in weight of the leaves before a-nd after drying. The dried leaves were recovered and processed as in Example 4(a), and analyzed as described under Example 1. Results are compared with fresh leaves of the same cultivar and are shown in Table 4.
(c) Determination of taxol* content of leaves of non-stripped cli~ping~s dried outdoors under shaded conditions.
Fresh intact clippings (stems & leaves) of T.
X media 'Densiformis' and T. X media 'Dark Green Spreader' were placed on aluminum wire cloth (18 x 16 -inch, 0.010 gauge window screen) attached to a wooden frame. The frames were placed under a shade structure out of doors on supports 30" above the ground. The plant material was allowed to dry under a structure , constructed by placing'a shade fabric over a 2" x 6"
frame, approximately 7 feet above the ground. The.
shade fabric reduced ambient sunlight by about 80%. No fan was used. The temperature ranged between an average 22'C at 8:00 a.m. to an average 36'C at 3:00 * grade-mark p.m. Drying lasted for 10 days. The moisture of the leaves prior to drying was determined to be 63.3% (T. X
media 'Densiformis') and 64% (T. X media 'Dark Green Spreader') based on the difference in weight of the leaves before and after drying. The dried leaves were recovered and processed as in Example 4(a), and analyzed as described under Example 1. Results are compared with fresh leaves of the same cultivar and are shown in Table 4.
(d) Determination of taxol' content of leaves of nonstripped clippings dried indoors at room temperature.
Fresh intact clippings (stems & leaves) of T.
X media 'Densiformis' and T. X media 'Dark green Spreader' were placed on aluminum wire cloth (18 x 16 inch 0.010 gauge window screen) attached to a wooden frame and dried according to the procedure described under Example 2(b). The moisture of the leaves prior to drying was determined to be 63.3% (T. X media 'Densiformis') and 65% (~ X ed'a 'Dark Green Spreader') based on the difference in weight of the leaves before and after drying. The leaves were recovered and processed according to the procedure described under Example 4(a), and analyzed according to the procedure described under Example 1. Results are compared with fresh leaves of the same cultivar and are shown in Table 4.

*trade-mark N -t W

E

N

1 t 1 m rf o P 0~ O ~D i i o~ 1 1 U 1 1 1 CD v O O n O co 1 1 h 1 1 \ 1 1 1 . . . . . . . l 1 .

W ~ ls.1 1 1 O ~ ~ ~ ~ ~ O 1 1 O 1 I
N

H
G:
Z

5c ~
o N o ~, W ~ >. ..v e~ oD ~ O vD ~D w U

~

~C O D O o O O O O O O
er 11 If ono oar oao on o~ or. 0 0 o 'v o o~ o o o~ o~n or. orr . .

W 411 1 1 Il1 . n . . . 1 1 . 1 I
nl . ,.i a-1 i~1 1 1 O O O O O O O 1 1 O 1 1 r1 ~ ~ ~ e1 a!1 ~ .~~

l01 1 1 +I +t +1 +I +1 tl tl 1 1 +1 1 1 o ~ ~ 1 I I o c ,r In N a o I 1 o 1 1 n a a a a a n a a ~ ~ 1 1 I 0 Il1 m 11'11C1 l"w D I 1 ~0 1 I

V7 ' V .~ o o -~ o a O

Hz E o> o> o> o> o> o> o> p>

lO ~

W U O o o O O O O d V V U V V U V V

~> x ~

N ~ ICf W

W W r Il1v n i N fV v u1 1t1 ~

..a o o e~ v ., er o o .+ rr m ro x rr rr ~ w WN cry or o on o~c o o~ ol~ o~ orr ol oln a 0 o o o o o o o o o o .c .v a n .

+I n ~D n Ilf o ef ef . . . 1 1 . 1 1 rr . O ..a O O O O 11'1O O O O O I 1 O 1 1 15 N N ri r1 ~ ~ ff tf! 1 .-1 +~ +i +I +I O +I +1 +1 tl +I t 1 +I I 1 ~~

~ III on on o~n +In lna u~a Na ern o0 1 1 o,u I I

ty ~p p 115 fV n a to a1 vD t~ P a I I N I 1 E., L C O O O ~ O O ~ O O ~ 1 WW

d = O> O> O> O> O> O> O> O> O> O> O>

' ~ ou ou oti oc ou ou ov oc~ oti oti oti i:.~ i , .. .

, Hr x Hw 20~
Z

*a a oW
R ~

x ,~.~ d ~ .+ al R .~ al .-~ ar z E C O C O C O C O G O C

w 41C O ~ C O ..a C O ~ t O .-rC O -~

> 1p y >. ~O i> >, IC i~ ?, i0 id >,~0 J~ >.

W a t C1 .C t 41 ~ ~ 41 .C t al .CZ C t U

O i~ V il ~ TJ i~ it U i~ ~ O a~~ U iJ

V~W rC W W iC W W it W W 4 iviW 4 W

a ~.

25~

w x '~ ~ o z m o t ~ w . d a~

~ ' A

u -~ a o d~ 1 >.~ o~

C a E
d 'C r 'p r ~ o c ~ : A ~' ,.

o ~ m ~o t~n~ ~c ~

a .

- .

O I 1 1 P O N ~D r n n 1 I .-~ I 1 , V 1 I 1 In r .-1 ~ r1 co e'~ 1 I Ov 1 1 O \ I 1 1 . . . . . . . 1 1 1 1 ,S 1 1 I .-~ .~ rl .1 .-1O .~ 1 1 O 1 1 H

W

C Ir1 r In 0J >~ ~f' CO N fa ~ r1 L

Qt G .1 O O .-1 O O .d O

H O O O O O O O O
~ ~ I' 1r II Ir In C~ 'fl O O O o O O O O
r ~c aD r e~ O~ C, d 1 1 I . . I 1 1 1 . rr .~C~ I 1 1 O O O O O O O 1 1 O 1 I
0~ n r 0~ Ir'1ac aD m la 10 I t 1 +1 +I +1 +I +1 +1 +1 i 1 +1 1 1 l0 ~ I 1 I ~D n r ao In ~o v 1 1 .-i t 1 p p p p p p p p O O 1 i I .m 1 W c1 n r a I 1 1~1 1 I

U N .~ ,.~ N ,.~.,.~N N
. . . . .

~~ O O O O O O O O
9 ,7 ;7 ~ ~ ~ ~ ~

IO ~ i . J O O O O O O O o V V V V V V V V

~

ci X

In o, N O In N c. ao .-i r O

N .-1 ~i !'1 r .'t f'1 O rl N N
~ II

O O O O O O O O O O O
In I1 ~ u1 x rI rr rr O O O O O O O O O O O
rr w ~ aD a~ ~ ~D u1 r . . . . . . O . . . 1 1 . I 1 ~p rl Q . . ~ . . .

N c N G7 n .-1 .-t Iflr CD CO

+I +I tl +I +1 +1 +1 +1 +I +1 I 1 +1 1 I

O ~0 O O N C~ m N aD ~D 1 I O 1 1 II 11 II II a a n II p II 11 ~ O ~ ~ ~ ~ ~ r ~O a N 1 1 n 1 1 +1 C O O O n .a ,.~ N ..a.~ t-f N
. . . .

> > > > 9 ~ '~ > > >

_ . Q' c oci ou ou oc~ ociov oci ociov oci cv r L

G

_ u *

2 .~ , w ~o 0 0 ~o x y ~ y ~ it ,o ev nr ar a O

a ~ <

o a ., a~ ~ ..~ d .~ al .~ d ~~ c o c o c o c ~ o c o ,~ .. ..
ar c o .~ c o . c o > R ~r >. ~C ii >. 10 J~ >, 10 y >,A ~r>.

L G7 L L 01 L L d7 L L d L t GIL

O y U ~ ~ U ~ ~ U s~ y U s~a~ U

V7 W iC W W ~C W W it W W 4 W W 4 W

I

d H

y m A
m 1 ~ H
d d ~ ph A
A

d CL H

~ d L O G

N O L 1 E-~
N

3 ~ ~ d w c 0 ~ m ~

.. I , l ~ H L O
~ H N 1r t9 OG
O D

Ca. ~D

~~~~~~7~
WO 92/18492 PCT/iJS92/03088 Example 5: Solvent of extraction.
Fresh clippings of T. X media 'Hicksii' were stripped into leaves and, small stems which were then left to dry at room temperature under conditions as described under Example 2(a). For analysis, a portion of dried leaves (10g) was ground and then extracted by percolation (4 X 100 ml) with one of several organic solvents utilizing the procedure of Example 1. The moisture content of the fresh leaves was 630. Analysis was carried out as described under Example 1. Results are shown in Table 5.

J

...

n x a r O t~1 .w C
r W

D r1 W a N n w ~ x p ..i ..r .~ .i .-i O O O O O O

O O O O O O

>

.-~ H

ma 0a mm o a H a ' w o 0 0 0 0 0 U O~ m to N pt a N o a, r ~

2 a ~ N .-, ..~ ., X
~

> C

O

G O

dx w a ...

CI E~ ~ N

w v v C .~ d w O O 'C

C -~ C ..i y A

.~ ~ t o va o w or r w y ~ U

w x ar d W w 'a 'D 41 ..r -.~ C

d s r z m U U .I~ i0 E d .O d d ~
~ 0 H ~i.~ 1 C1 0 rr O O C

d >, ?. .r a~

> ~ = ~ d t t .

a , . Cr .. v i ~ ~ a v ~

~~.~R',~5 WO 92/18f92 ' PCT/US92/03058 ,r,,~:>~, Example 6> Method of extraction.
Three methods of extraction were compared using two replicates in.each instance. Leaves of T. X
media °Hicksii" were stripped from fresh clippings and allowed to dry at room temperature under conditions as described under Example 2(a). The dried leaves (10 gms) were then ground and extracted using acetone by percolation, under soxhlet conditions or by shaking (soaking with agitation). Extraction by percolation was performed according to the procedure described under Example 1. Extraction under soxhlet conditions was carried out using one portion (100 ml) of solvent for 12 hours, followed by exchanges with fresh solvent at four hour intervals; a total of 400 ml of solvent was used for each 10g sample of plant material over the 24-hour period of extraction: Extraction by shaking (soaking with agitation) was carried out using the same schedule of solvent exchange as described under soxhlet conditions. The extracts were then processed and analyzed as described under Example 1. Extracts obtained from the percolation method of extraction were combined and analyzed; extracts obtained from the other two methods of extraction were analyzed at each step of the extraction process. Results are shown in Table 6.

c G

a~
o .., ., w ~ n ~ o U

.
r O

.r, o n N
w w N ...1 O

O >

U

~, +1 +, +.
, v~ '~' ' o m .~ c9 0 < u, o m o vo 0 O ~ O N O

> +1 O ~ O ' O

O O O

O O O

1a 1 0 '~ ~ ~' p 1D f !'1 f'1 \D
~

V
rl n iCf 1Cf 10 O N
'i a ..i ~ x .., ..~ ., ..m -, N

O O O O O O

v E" o o 0 0 0 0 f"' ~' ..i ..i .Ix w.

b 1 1 ~ ~' m tl.,~ 1 1 O O N .a y d ~ ~ ~ O O O O

~ 1 1 G GO O O O O

Q1 e~ 1 1 oD e'f r 01 O ~"~ ef c~I

x ~ G, ~ ~ O O O O

y A * 3 ~ 1 1 0 0 0 0 ~r E"' 1 1 c ~ v~ o 0 0 >a x s~
w N

?C E 1 1 2 0 ~ Q' a ~ ~ o o 0 0 w tn 1 1 O O O O

o~
a 'o ., 1 1 .1 0 In ~c oa N n ~o N

s ev Q, .., ..~ ..

v ~ ~ 0 0 0 0 *' xo cn 1 1 0 0 0 0 o er r"

H ~ ~ w -1 N ~ N ~ N

w Lr Ip 0 'w C

o*..~

vl o o sK

H

~, w V .
n ao O o O In 11.1 ~ ~ N f"1 1f 41 01 1D

3 0 o ~ 3 .~ ., ~ .; .~ ~, ~ w ~w ao .~
o a o >m c w a o . o ,~a w 3 5 m L v c se b o .r . !.. . . ~d e x to it it tl d x a O t a ~ w o. In v:

E

~~.~?~,~5 ~'O 92/18492 ~ Pt'T/LJS92/03088 ,~

Example 7: Partitionina of the first residue between water and methylene chloride.
Fresh clippings of ~'. X media °Nigra~ were dried in a bulk tobacco curing barn according to the method described under Example 4(a). A portion of the clippings were ground using a Wiley Mill, then extracted by percolation using a solvent selected from among ethanol, acetone and ethyl acetate and using the procedure described under Example 1. Another portion of clippings was separated into leaves and stems, and the two portions were ground separately using a Wiley Mill. Each ground portion was then extracted by percolation using a solvent selected from among ethanol, acetone and ethyl acetate and using the procedure described under Example 1.
(a) Partitioning° of the first residue of the ethanol extract between water and methylene chloride A known weight of the residue of the ethanol extract was partitioned between water (1 ml) and methylene chloride (2 ml x 5). The methylene chloride phase was then prepared and analyzed by HPLC according to the method described under Example 1. The results are shown in Table 7a.
(b) Partitioning' of the first residue of the acetone extract between water and methylene chloride A known weight of the residue of the acetone extract was partitioned between water (1 ml) and methylene chloride (2 ml x 5). The methylene chloride phase was then prepared and analyzed by HPLC according to the method described under Example 1. The results are shown in Table 7b.
(c) Partitioning of the first residue of the ethyl acetate extract between water and meth~lene chloride.

~.,,.WO 92/18492 ~ ,~_ ~ ~ ~ ~ ',~~ PC I'/US921J3~88 V ' A known weight of the first residue of the ethyl acetate extract was partitioned between water (1 ml) and methylene chloride (2 ml x 5). The methylene chloride phase was then prepared and analyzed by HPZC
according to the method described under Example 1. The results are shown in Table 7c.

_ 62 J

p U

d it c sJ

ar tn a ,r ..
., z cr v~
1~

_ 3 = ~ n O

. .~ 0~ o ~o eoenO 1 ~ . H m l b O P1 ~ N t9~ .1 C N N N N O O

O O O O O O ~ O O O O O O
~

d Ew o 0 0 0 0 o Ew o o 0 0 0 0 c O -..;
X

y N v0O 010~ ,~ O 4141 P1t'1If1 ~

U V t Ip N m n N o V e-ve~.r letN vo .-. t ~

N N . . . . . ~, ~T N
LT

m a 3 ' o,o. v.n ~ v ~ ~'~' c' o, w ' 3 ~

- . . ~
l .i., .., ~" ~ m o~ Irl.r N InO G c o~ n ea n n In ~ ~ w o n N n ~ x ~ eoIn Into 0 o c x In .O p, 'O tr t .

o C C ~ ~ m ~ m o~N N ~ d a~ n a Inn n E A
B

w .", O s 3 ., .r...1.r.-I.-IU x ~ n n n .~ .r '- Z
~

E ~ V y m y CO i~.~i O ao o~.r ofowo Ix ~ n lc1~ o n o~
= d ..., ~ y . a i r, o N o~ t, o ao< voao r' fl V N N
~ --. ,,, x ." . . . . . ~, N
to ~e ~
p~ v~

S t ao N ~ n w .-IC Z N n IW n .i a E t E

V p, n n e-Ie~N n p U O vD1e1lefa a ... G.
v U
w Q d L ta7 a O 0~ ~DN ~O ~C c 1e1O O saa ~D

!'1r1m O~~ n O N N 1D 10a a d 1 S ~ . . . . 1 y ~ ~ 0 0 ~

y o A o 0 0 0 0 o d 0 0 0 0 3 ~
~

., y c 0 0 0 0 0 o N ~ o 0 0 0 0 0 2 ~ c ~ m s ..l.~..l.~..~~., c ..~..l.~ .,., ..
.c - c -->
c N

E
c y y o .p Inae ..le,n O co o~O N In In 'p C ,~ N n la W D f y If1u1.1 1e1u1 In y .1 ~

O C7 n n e~ ~ r ~ d ~ ~lO O n n .. d ~

d Lt~7 N N N N r1ri ~C ~ ~ ~ ~ D O
? ~

w N

w O

'"' ~

o d A

c 1~r .~ rl fV.1 fV.aH1 ~ rl N ~ N rl N
E as ~

t O

pv C

..i 1~ ID ..l o y O r n. m o or o. m x n ~ ~

wy '~ ~ v H t y ~

ww , .,. - , a v~

mo r H
d r d A
E

w v~

~3 ~

m a~ ~ ~ O
o N NN N ~1v-1 O OO O O O

E"''~ O OO O O O

1~

N

p V t n emn ao o~N
~

NN
y ~ ~o nN o ~on x A
E

V U 3 ri .~N N rlr-1 '~

fp is y x w c a~ nv m ~om _ d N nn a N 0t Z S ~ x ~
~

my n aev N N
E

y .~ r erv a n n ' d U

...1,~ ..rOrr 1~1v01f1 d r, v0~ .~ O~n t0 C' a~ ~
' c~ u ~ ~~ ~. ~
o o 2~ n ' N rO N ~ N

.i ~. O Nn n .mn W 4 O ~ 0 O
E

. ~ .. ~ .
r .~ a -i nao c, .-~n 4.-~ o nv m n n O of ao n~o ~o v ,r E.
...r wa 0 00 0 0 0 .a C

~ rt Nrl N ~1N
as N

30 ~ =

m > a .,. ip ...r d N

y U t f l~R

'y1'~ 92/18492 PCT/US92/03088 .fr~a, V
Example 8: Partitioning of the first residue between water and ethy_1 acetate followed by second solvent partitioning between hexane and an aqueous mixture.
Fresh intact~clippings of T. X media 'Nigra' were dried in a bulk'~'tobacco curing barn according to the method described under Example 4(a). A portion of the clippings was ground using a Wiley Mill, then extracted by percolation using a solvent selected from among ethanol, acetone and ethyl acetate and using the procedure described under Example 1. Another portion of clippings was separated into leaves and stems, and the two portions were ground separately using a Wiley Mill. Each ground portion was then extracted by.
percolation using a solvent selected from among ethanol, acetone and ethyl acetate and using the ' procedure described under Example 1.
(a) Partitioning of the first residue of the ethanol extract between water and et~l acetate followed by second solvent partitioning between hexane and an aqueous mixture (methanol: water).
A known weight of the residue of the ethanol extract was partitioned between water (1 ml) and ethyl acetate (2 ml x 6). The ethyl acetate phase was evaporated and the second residue was then partitioned between two parts hexane and one part methanol: water (9:1). The methanol:water phase was evaporated to dryness in vacuo. The residue was dissolved with sonication in 1 ml methanol (HPLC grade) and the ' solution was withdrawn in a 1 ml syringe and filtered through a Millex filtering unit (o.45~c) into a Wheaton vial, and capped. HPLC analysis was then conducted as described. under Example 1. Results are shown in Table sa.
(b) Partitioning of the first residue of the acetone extract between water and ethyl acetate .,~9~~ 92/18492 ~ ~ ~ ~ ~ ~ ~ PC'~'/US92/030~8 followed by second solvent partitioning between hexane and an agueous mixture (methanol: water).
A known weight of the residue of the acetone extract was partitioned between water (1 ml) and ethyl acetate (2 ml x 6). The ethyl acetate phase was evaporated and the second residue was then partitioned between two parts hexane and one part methanol: water (9:1). The methanol:water phase was evaporated to dryness in vacuo. The residue was dissolved with sonication in 1 ml methanol (HPLC grade) and. the solution was withdrawn in a 1 ml syringe and filtered through a Millex filtering unit (0.45~C) into a Wheaton vial, and capped. HPLC analysis was then conducted as described under Example 1. Results are shown in Table sb.
(c) Partitioning of the first residue of the ethyl acetate extract between water and ethyl acetate followed by second solvent partitioning_"between hexane and an aqueous mixture (methanol: water).
A known weight of the residue of the ethyl acetate extract was partitioned between water (1 ml) and ethyl acetate (2 ml x 6). The ethyl acetate phase was evaporated and the second residue was then partitioned between two parts hexane and one part methanol:water (9:1). The methanol:water phase was evaporated to dryness in vacuo. The residue was dissolved with sonication in 1 ml methanol (HPLC grade) and the solution was withdrawn in a 1 ml syringe and filtered through a Millex filter unit (0.45~a) into a Wheaton vial, and capped. HPLC analysis was then conducted as described under Example 1. Results are shown in Table 8c.
Comparison of Tables 7 and 8 indicates that purification by partitioning produces comparable yields of taxol*when compared to the more expensive (in materials and labor) method of purifying using celite.
~ ..

20.

3 5 * trad~rk c 0 2s ~d a >'a U 'O

-.1 I

w C

~ a w wo a o c " 'P

o~ v~

~ +~
~ r I 0~o n r .~o -~ I a o~r 3 = n O e~n eI n .-1.a O N N N N O O

O O O O O O ~ O O O O O O
~

O

Ew o 0 0 0 0 o Ew o 0 0 o c o ~~ ' 00 ~ ~
c E

o .,., .., \ f'!O 1~1e~ N Il1 \ f .r ~p L C Q! Q1 -1 ~ O CD 01 CD~ O -i~D O N ofO~
y O : :
a w .aosr ..;r e m aou; o wi cv~o t t E
y! ~ P7N N f'1N N Z f'1t'1rf~f !'1f'1 On CL

' ' '~

Z ,. O v~ O v, S .~

..aE ..~ E

N d v v V d V ~

p > N a C ..~O n .r eft~a C O~In eDv u1~
fW d ~ ~ ~ N ~ eD r n r ~ 'Wt'1r O~
~

a Qt ~ x .1 C X C1 N O O .-iCC!~X 01 ~-1N N W 0 ~l1 t t p w = .a.i.r .i W Z e~n N N .~rr G. Cr E -a~ .a o y a O
, A E N E

.

W V V
2 ~ v N % ... cW c,vo~-,~-,o ... v c ..~r,.m eo.~
W

O ~ a O~ O O .~'C O 1~1~D n O C~O
A ~

a a. I E N o a n ccr'I E c,o vc.-yrin t t m v .r~ v r,n w vo,..vpw nn c. cs.

1r ~ N
w O

w ~ r1 ..1 d 1r ~

'o d ~
a 7 it ?~ raO O !~ m O C aDu'1a O~ N ..r N O O O N r y ~-1ICIIHN vDn 2 d U =
S -~

!", O O O O O O O 41 O O O O O O
d a Z

t it j o ~ O ~ O O 4 ~ o ~ o O o - ~ ~

f r . . .. ,.. . ., ....
. -1 i .I .I.~ -1r r i a~
a w c o nr w vd i y y,, .i ~

~

w r.1N r1 N ~I!V~C ~ N ~1N ~iN
i w C
~

30 W O ~

C~ a !T C C' t > a ~ > a n .

-. a -. d ~. a ~ ~, b = ~ a ~ ~ ~ a -E

Fr .~. ~l '~ !, J .) VIVO 92/18492 PCT/US92/O~n88 - s8 -..a r.b voo,N

ro 0 0 0 0 0 0 ~

Hro 0 4 0 0 0 0 1~

o U
M ~

ro W 01b' 6f1rd01a"1 .C

~ cha a v v sr x a.

W

b a C CON ~Dt0ST10 x a;~ o~r O
~

Cl I'~N r-1riIDh .C

r1 ~ P1~V'eff7N N
~

dJ

W LT

s vv ~ ~r r,n m r, O N o r r ~rr ro H .-Wc o ~ocon r W COc0 01r I~h ~ , ~ P O t0N srsf U o a~ w e own ~S f O O
O

ft p O d o E o o 0 o O o C

W rle-~r!.-W e-1 ~ -~

sQ,' r4N P9N 'iN
~

I

s s S.~ N -X11 i p 9 O. ~

U ~

V N

WO 92/18492 "s~'' - ~ '~-~' ~ J ~'CT/US92/03~88 n 69 _ -Examt~le 9: Elution of the eelitecolumn with ethyl acetate instead of meth~lene chloride.
Fresh clippings of T. X media 'Hicksii' were separated into leaves and stems and allowed to dry at room temperature until no further weight loss was observed. The dried leaves were then ground using a Wiley Mill and extracted by percolation with acetone.
Analysis was carried out in duplicate using 10g leaves.
The acetone extract was treated and analyzed as described under Example 7(b), except that ethyl acetate was substituted for methyl~ne chloride. Results are shown in Table 9.

J

n p ) N >
0 ~ .

5 ~' o v r a ~ o y ~ ce . .r i o o u , ~u ~

. ~"~ 0 0 d ~
'~

10 ' ~ v 0 0 d t~

~ ro ar y V w ~' ~E =
~"

O~N
t ii O

1. ~.~ a,ry d o 3 ~

Cl 3 4) ~ N
iJ U

i I
d ~ 0 I5 3~0 ~a c ~ .

.

N Cl -~r O C t .~. ~ N

C ?, i0 Vl p~ ~ 01 ~.1 ?~ ~ L ~ ~ tot0 ..1 N N

R t c it -.r l.i p, W

20 x c Gl .i W

Nt * N
O

O > ~0 CI b " o' m..~ c t ~r A , E
' m-.~ ~ y dt U .~ 3 Cr CJ
G. "' tC C! C R

U O

C M

R c iJ

V

C ~.1 tr'"r ~'~ ~
~
H

c ii t ..
, ~
y~ p, '~

~w a ~
r =W m ~ .
~ .

y ..~ w C

-~ it O

~ D

~ a s~

30 td :

W ~ n N N

iJ O QJ ~ ~ rt r1 ci w t ' y ~. w w c c w ~p O .,~ .sd ca N

N

3 5 ~, ,, ~ N

c ?~_~~>~~
,~,;,,, WO 92118492 PCl'I U~92/030$v _ m _ V
Example 10~ (a) Defatting of the plant material prior to extraction by percolation using acetone.
A sample of leaves of T. X media ~Hicksii' was dried and ground using the method described under Example 9. Dried, ground leaves (10g) were defatted with hexane by percolation for 2 days, changing the solvent every 24 hours (2 x 100 ml). The combined hexane extracts were evaporated to dryness (residue average 200 mg which represents 21% of the total hexane and acetone extractives). The marc (remaining residue) was transferred quantitatively into a 250 ml Erlenmeyer flask and extracted with acetone by percolation, changing the solvent every 24 hours for 4 days (4 x 100 ml). The combined acetone extracts were evaporated to dryness. A known weight of the acetone residue was ' partitioned between water (1 ml) and methylene chloride (2 ml x 5). The methylene chloride phase was evaporated to dryness in vacuo and the residue reconstituted in methanol (1 ml). The methanolic solution was filtered through a Millex filtering unit, diluted 1:1 with methanol and used for HPLC analysis according to the method described under Example 1.
Results are shown in Table 10.
(b) Defattina of the plant material~rior to extraction by soakinqlwith agitation (shaking~,~ using acetone.
Dried, ground leaves (log) as in (a) were defatted with hexane by soaking with agitation for 8 hours, changing the solvent every 4 hours (loo ml x 2).
The average hexane extractives was 194 mg or 12% of the total hexane and acetone extractives. The marc was transferred quantitatively into a 250 ml Erlenmeyer flask and extracted with acetone (4 x 100 ml) by soaking with agitation over a 24 hour-period, changing WC~ 92/1492 PC'~'/L1S92/03088 t - 72 -the solvent Zit~l2, 16, 20, and 24 hours. The extracts, collected at each point, were combined, evaporated to dryness in vacuo and a known portion of the resulting residue was partitioned between water (1 ml) and methylene chloride (2 ml x 5). The methylene chloride phase was evaporated to dryness and the residue reconstituted in methanol (1 ml). The methanolic solution was filtered through a Millex filtering unit, diluted 1:1 with methanol and used for HPLC analysis.
Results are shown in Table 10.
(c) Defatting of the plant material prior to extraction by soxhlet usincr acetone.
Dried, ground leaves (lOg) as in (a) were defatted with hexane by soxhlet extraction for eight hours, changing the solvent every four hours (100 ml x 2). The average hexane extractives was 297 mg or 260 of the total hexane and acetone extractives. The mare, in the soxhlet, was further extracted with acetone (4 x 100 ml) according to the method described under Example 6. Extraction, partitioning and analysis are as described under Example 10(b) and the results are shown in Table 10.

n o v m vo ~ off o o vi r.

0 0 ow o.r x +i . .
~ ,a t0 O O O U1 .~1 r1 E~ C) +1 +1 tl cr .n nr ~r 4 ~ ~o . H .. .
.

~ o> o> o>

.. . ..

~t o o o a U V

m w f,, * w oy rH Leo e~r.~

p 'p .. W o~ .. ev pv n x .. .r .. .r .a e~

"'' E' ~ O O C O
O O

O~

U

rr *r l N

i C ~
:
c ~

a d ~ y C m e~ o Ew v t~
U ~

_ ~, O ~0 N N C7 If1 S 1~ C' ~ .-t ~ ~ J3 t d' ' ~

1 V f R 41 a 10 f : i c .
c o.co ~~ U

~

pi ,, . vD v N
x w O
r p .
0 ~ N

. v l1 H m . m ~

V N ~ d '~ N N ~ O
E 0~ ~ r 1.1 ~

y~ N i9 ~ O O ~
'O d O 3 ~ ~ O .H
C ~ Q'' ..r ~ aJ 'C

w O C~

2 0 '~ w y w .p o .~

w .-~ U e-s ae t~
n er v w L ~ L ~
~

G9 'C 0v ; m CO
0~

'~
'~

x p~ O O O
O

~. ~ v O

2 5 '" ' ar c A c 'y o o ~ c x a a y ~

i ~ w H .i ~ H
.~ .i =

..

m 3 ~ ?"

rl ~1 ~ 1V
~s N N

C

r C

U t 35 ~s~.~'ar c rr ~ O ~ t b C
?

d t O

a sn N

2~i~°26~
1W0 92/18492 P~CT/US92/030$$ t., ,.
Example 11: partitionina of the acetone extract of the defatted plant material between water and ethyl acetate.
A sample of leaves of T. X media 'Hicksii' was dried and ground using the method described under Example 9. Dried, ground leaves (lOg) were defatted with hexane and extracted by percolation using acetone using the method described under Example 10(a). A
known weight of the acetone residue was partitioned between water (~. ml) and ethyl acetate (2 ml x 5). The ethyl acetate phase was evaporated to dryness in vacuo and the residue reconstituted in methanol (1 ml). The methanolic solution was filtered, diluted and analyzed as described under Example l0(a). Results are shown in Table 11.

,~=,,,WO 92/1$492 i~ '~ ~ S~ ~ ~ ~ PCT/US92/030$$

a '' ~ O~

O
~' O

f0 C~

~ H ro ..
, ~, ~ to d O ~ O

~ "~
i . O

1 d d~ W N

~o~

a .u ro .G a~ .., ~ . ea ~r.-a eo , ~ 3 ~

3 B o~ 0 w a~ w H ~ ~
o o y ~i ~ N

5 a ro ~o W

~,~~

~ a ~ ~ v Sri N ~ .o oc~

-.~ ro .u c t~ ar ~a ro s~

xvx ~
~ o a~ ~ ro w ~
.. ao ~

ro H o ~ r , cs. ~
m ~ ~21~ u~o~
~ 6 ~p v o0 ~w ,o ~a o v r, c ~, w~N~

O
~ ~ U ~.

a ~ ~ ~ O~
C b.WT O~

U w . ~ V
i x N o O

.~ ~ a~ x ~, ~
w .~ .N .~ U

1~ ?~ d~ -..1 w U

CI 'O

'ty A1 C LT

f0ro C
U

WO -.
i C

I

~,~E ~3~~ ~;

a ~a ~ ~ d x w ~

s ~

m~,wH

N

' r-1 .r..l r1 N

~ '-1 '~' N

QI n"
~

'3 5 ~ ~c 8 92 PCT/US92/03D88 ,..r;>~., t: .
-~ 76 _ U
Example 12: Determination of taxane levels in T. X
media 'Niqra!usinqasolid phase finishincx treatment.
Fresh leaves of T. X media 'Nigra' (Rhode Island Nurseries, Newport, Rhode Island) were stripped from their stems, and the following procedures were carried out using three replicates. The stripped fresh leaves (1o gms) were placed in a blaring blender along with 100 ml of 95~ ethanol to macerate the leaves. The leaves were ground for 2 minutes.
The ground leaves arid the ethanol were quantitatively transferred to a 250 ml Erlenmeyer flask and the leaves were allowed to soak for 24 hours. The ethanol was filtered with rinsing (25 ml ethanol) and the ground leaves were returned to the Erlenmeyer flask, to which another 100 ml volume of 95% ethanol was added. This procedure was repeated three more times until a 500 ml volume of ethanol extract was obtained.
The ethanol extract was evaporated to dryness in vacuo at a temperature not exceeding 40°C to leave a residue ("first residue"). The weight of this residue was 1.47 g ~ 0.059 (C.V'. = 4%).
About l00 mg of the first residue was transferred into a 4 ml screwcap vial. The vial cap was lined with aluminum foil to avoid contact of the vial cap with the residue during subsequent partitioning steps.
The first residue was partitioned between. about 1 ml of water and about five 2 ml volumes of methylene chloride. The methylene chloride was then transferred to a 25 ml Erlenmeyer flask and the solvent was evaporated in vacuo to form a second residue.
The residue from the methylene chloride layer was then dissolved in about 5 ml of ethyl acetate and about a ml of methanol. To assist in the dissolution .~.,;..~0 92/18492 ~ '~ ~ ~ ~ JPC.'T/US92/030~3~
,.~._,~t...,=.~'7~

a ' of the residue the flask was sonicated. When the residue was completely dissolved, about 650 mg of celite was added. The residue was adhered to the celite by allowing the solvent to evaporate under reduced pressure to produce a coated celite material.
The coated celite was transferred to a petri dish, then triturated until a free-flowing powder was obtained. A Pasteur pipette was then packed with the triturated powder.
Hexane was washed through the column until the eluent appeared colorless (7 ml). Methylene chloride (5-6 ml) was then used to elute the taxanes. The methylene chloride was then evaporated in vacuo to produce the crude taxane mixture. The weight of this 1$ mixture was 39.1 mg ~ 7.43 (C.V. = 19%).
The crude taxanes were dissolved in 1 ml HPLC
grade methanol. Sonication was used to assist in the dissolution of the crude taxanes. The solution was filtered through a Millex filtering unit (0.45,) and collected in a flask.
The identity and amount of specific taxanes in 'the solution was determined using HPLC analysis from a 10 g sample. A ~C Hondapak C-18 (10~C) column, obtained from Waters Associates was used fox the HPLC analysis.
A mixture of methanol (650) and water (350) at a flow rate of 1.2 ml per min was used as the mobile phase. Taxanes were detected with a U.V. detector set at a wavelength of 227 nm.
The retention times and percent compositian of the solution is shown below in Table 12.

J
Table 12. Rt Values and Percent Composition of Specific Taxanes Isolated from T. X media 'Nigra' Rt Value Taxane (min.) _$
Composition Taxol~ 11.5 0.291 cephalomannine 10.5 baccatin III 4.7 -10-desacetyltaxol - -10-desacetylcephalomannine - -10-desacetyl baccatin III 4.2 -example 13. Percent Taxol obtained from Barn-Dried leaves extracted accordincLto the method of Example 12.
Fresh intact clippings of T. X e~n,d,~a 'Nigra~
15 leaves, still attached to their stems were placed on an aluminum wire cloth (18 in by 16 in, 0.10 gauge window screen) attached to a wooden frame inside a "Roanoak"
style drying barn, model 7.5-1-G Bulk Curing Barn obtained from Gregory Manufacturing Company, Inc., of 20 Woodville, North Carolina. The frames were placed on supports about~30 in. above floor level. The plant matter was dried in a ventilated barn in the dark.
During the drying step, the temperature varied betweew about 30'C at 8 a.m. to about 40'C at 25 3:30 p.m. Drying was continued for two days. The clippings were analyzed for taxane content following the procedure described in Example 12. The following taxanes were detected: taxol; cephalomannine: baccatin III. 10-desacetyl taxol: 10-desacetylcephalomannine;
30 and 10-desacetyl baccatin III. The percent taxol present in the composition was 0.034.
*trade-mark * trade-maxk .
Example 14. Isolation of taxol from the leaves of T. X
media 'Dark Green S,~reader' using a solid phase finish~n4 treatment.
Fresh clippings of T. X med'a 'Dark Green Spreader' were separated into leaves and small stems and then allowed to dry at room temperature according to the method described under Example 2(a). The dried leaves (500g, showing a taxol content of 0.0074% as determined by the method described under Example .1) were ground in a Wiley Mill to a mesh size of about 60 and extracted by percolation with 95% ethanol, changing the solvent every 24 hours (1.5L,x 5). The combined ethanol extracts were evaporated on a Buchi rotary evaporator under reduced pressure at a temperature not exceeding about 40'C.
The ethanol residue (112 g, residue A) was partitioned between water (about 500 ml) and the following volumes of methylene chloride: one volume of about 1 liter, and three volumes of about 500 ml each.
The methylene chloride layers were separated, combined and dried over anhydrous sodium sulphate. The drying agent was removed from the methylene chloride using vacuum filtration. The methylene chloride solvent was evaporated using a Buchi rotary evaporator under reduced pressure to produce a second residue (27.88, residue H).
The second residue (27.8g) was dissolved in an ethyl acetate/methanol mixture (3:1, 600 ml) and the resulting solution divided into two 300 ml portions.
Each portion was uniformly coated onto about 2008 celite 545 (Fisher) as described under Example 1.
The celite coated with the second residue, Residue B was packed into two columns (5 x 32 cm each) and each washed successively, under pressure, with hexane (2.2L) followed by methylene chloride (1.2L).
*trade mark v Evaporation of the hexane and methylene chloride washes iI vacuo yielded combined residues weighing 19.258 (Residue C) and 6.358 (Residue D), respecti~rely. .
Residue D comprising the crude taxane mixture (6.358) was dissolved in 250 ml ethyl acetate and adsorbed onto celite (508) according to the method described under Example 1. The resulting material was applied onto a silica gel 60 flash column (1608, 5 x 16 cm, 230-400 mesh, E. Merck, 1 column volume=500 ml) packed in hexane/acetone (?5:25). The polarity of the mobile phase was gradually increased to hexane/acetone (55:45) by increasing the percentage of acetone in 5%
increments. Fractions we're collected, under pressure, at a flow rate of 100m1/5-minutes. The column was eluted with one column volume of hexane/acetone (?5:25) collected as one 500 ml-fraction, followed~by two ' column.volumes of hexane/acetone (?0:30) collected as 4 fractions 250 ml-each , two column volumes of hexane/acetone (65:35) as 4 fractions 250m1-each, one column volume of hexane/acetone (60:40) as 2 fractions 250m1-each, one column volume of hexane/acetone (55:45)' as 10 fractions 50m1-each. The collected fractions were examined by thin-layer chromatography on precoated silica gel G W2~~(Machinery Nagel; Duren). Fractions eluted with hexane/acetone (60:40, 500 ml) and hexane/acetone (55:45, 150 ml) were combined and evaporated to dryness 'fir vacuo (Fraction E, 0.4278).
In this fraction, 8 components were identified by thin layer chromatography using 5% methanol in chloroform as a developing system and p-anisa~ldehyde/sulfuric acid as a- visualizing agent. The taxol-cephalomannine mixture appeared as a bluish-grey spot, with an Rf value of 0.62.
Fraction E was further fractionated on a silica gel 60 flash column. The material (42? mg) was * trade-mzrks dissolved in 1% methanol/methylene chloride (2 ml) and applied onto a silica~gel 60 column (55g, 2 x 40cm, 230-400 mesh, E. Merck, 1 column volume=120 ml), packed in 1% methanol/methylene chloride. The column was eluted, at low pressure to provide a flow rate of 8 ml/min., successively with 1% methanol/methylene chloride (2 column volumes as 20 fractions 12 ml-each), 1.5% methanol/methylene chloride (1 column volume as l0 fractions 12 ml-each), 2% methanol/methylene chloride (1 column volume as 10 fractions 12 ml-each) and 2.5%
methanol/methylene chloride (3 column volumes as 74 fractions 5 ml-each). Fractions were monitored by thin layer chromatography using the same developing and visualizing reagent systems described above. The taxol*cephalomannine mixture was identified in fractions 78-86 eluted with 2.5% methanol/methylene chloride. These were combined and evaporated tc dryness in-vacuo to give a residue (40.35 mg, Fraction F). This fraction when examined by HPLC indicated the presence of taxol together with cephalomannine.
Taxol* was separated from cephalomannine by dissolving Fraction F (40.35 mg) in 1 ml of a solvent mixture of 20% ethyl acetate in methylene chloride and applying the solubilized Fraction F onto a silica gel 60 flash column (lOg, 1x32cm, 230-400 mesh E. Merck, 1 column volume = 21 ml) packed in the 20% ethyl acetate/methylene chloride solvent mixture. The column was eluted, at low pressure to provide a flow rate of 8 ml/min., successively with 20% ethyl acetate/methylene chloride (2 column volumes as 4 fractions 11 ml-each), 25% ethyl acetate/methylene chloride tl column volume as 2 fractions 10 ml-each), 30% ethyl acetate/methylene chloride (1 column volume as 5 fractions 4 ml-each), 35% ethyl acetate/methylene chloride (2 column volumes as 9 fractions 4.5 mI each), 40% ethyl *trade-mark V
acetate/methylene chloride (1 column volume as 7 fractions 3 ml-each), and 45% ethyl acetate/methylene chloride (2 column volumes, as 20 fractions 2 ml-each).
The fractions were monitored by thin-layer chromatography on silica gel G Wz~' using ethyl acetate/methylene chloride (1:1) as developing solvent and p-anisaldehyde/sulfuric acid as visualizing reagent. Taxoh'(26mg) was obtained in a pure form, as determined by HPLC analysis, in fraction's 35-50 eluted with 45% ethyl acetate in methylene chloride.
Example 15: Isolation of taxol~from the leaves of T. X
media 'Niara~ using a partitioning finishing treatment ste .
Intact clippings of ~ X media 'Nigra~ were.
barn-dried as described in Example 4(a). The leaves obtained from the barn-dried intact clippings (500g, showing a taxol content of 0.043% dry weight as determined by the method of Example 1) were ground in a Wiley Mill to a mesh size of about 60 and treated by percolation with 95% ethanol (8L). The combined ethanol extracts were evaporated to dryness on a Buchi rotary evaporator under reduced pressure at a temperature not exceeding 40'C.
The ethanol residue (134.4g, first residue, residue A) was partitioned between water (about 600 ml) and the following volumes of ethyl acetate: two volumes of about 1.2 L, and three volumes of about 600 ml each. The ethyl acetate layers-were separated, combined and dried over anhydrous sodium sulfate. The drying agent was removed from the ethyl acetate using vacuum filtration. The ethyl acetate solvent was evaporated using a Buchi rotary evaporator under reduced pressure to produce a second residue (44g, 3 5 Residue 8) .
'* trade-marks Residue B was partitioned between a biphasic mixture of about 1 part water, 9 parts methanol (500 ml), and two volumes of hexane (1000 ml).
Subsequently, each phase was washed with the opposite solvent (100 ml x 5) and the washings combined with their respective phases. The combined hexane layers were evaporate to dryness to yield a reside (17.7g, residue C) and the combined water/methanol layers were evaporated, 'fir vacuo, to produce a residue of a crude i0 taxane mixture (3l.lg, residue D).
Residue D (3l.lg) was washed with sonication using methylene chloride (300 ml x 4). The methylene chloride washings were combined, and evaporated 'fir.
vacuo to yield a methylene chloride soluble portion of Residue D (1?.3g, Residue E, showing a taxol content of 167mg as determined by the method of Example (1)). The weight of the methylene chloride insoluble portion of Residue D was 13.8g and showed a taxol content of 20mg as determined by the method of Example 1.
Residue E (17.3g) was dissolved in 250 ml ethyl acetate and adsorbed onto celite (30g). The resulting material was applied onto a silica gel 60 flash column (315g, 5 x 32 cm, 230-400 mesh, E. Merck, 1 column volume=600 m1) packed in hexane/acetone (70:30). The polarity of mobile phase was gradually increased to hexane/acetone (45:55) by increasing the , acetone content in 5% increments and then the column washed with 100% acetone. Fractions were collected, under pressure, at a flow rate of 100 ml/5-minutes.
The column was eluted with two column volumes of hexane/acetone (70:30) collected as eight fractions:
300 ml, 100 ml, 125 ml, 200 ml, 160 ml, 190 ml and 210 ml, followed by one column volume of hexane/acetone (65:35) collected as 2 fractions: 175 ml and 300 ml, one column volume of hexane/acetone (60:40) as one ~ trade-mark fraction 500 ml-fraction, one column volume of hexane/acetone (50:50) as ? fractions 100 ml-each, one column volume of hexane/acetone (45:55) as 6 fractions:
100 ml-each, and 3 column volumes of 100% acetone as 4 fractions: 450 ml-each. The collected fractions were examined by thin-layer chromatography on precoated silica gel G W25' (Machery Nagel, Duren) using 5%
methanol in chloroform as a developing system and p-anisaldehyde/sulfuric acid as a visualizing reagent.
Fractions eluted with hexane/acetone (55:45, 100 ml) and hexane/acetone (50:50, 450-ml) were combined and evaporated to dryness ~ vacuo (Fraction F, 1.61g).
In this fraction, 8 components were identified using 5% methanol in chloroform as a developing system and p-anisaldehyde/sulfuric acid as a visualizing reagent. Taxol*cephalomannine mixture' appeared as a bluish-gray spot, with a Rf value of 0.62.
While we have hereinbefore described a number of embodiments of this invention, it is apparent that the basic constructions can be altered to provide other embodiments which utilize the methods of this invention. Therefore, it will be appreciated that the scope of this invention is defined by the claims appended hereto rather than by the specific embodiments which have been presented hereinbefore by way of example.

* trade-mark

Claims (78)

We claim:

1. A process for obtaining taxanes from a Taxus plant which comprises:
(a) separating intact clippings from one or more Taxus plant wherein said intact clippings are leaves attached to stems;
(b) drying the intact clippings from step (a) to form dried plant matter;
(c) contacting the dried plant matter from step (b) with an organic solvent to extract taxanes from the dried plant matter and obtain a taxane-containing extract;
(d) evaporating the taxane-containing extract formed in, step (c) to form a residue and partitioning the residue between an aqueous solution and an organic solvent to form a two phase solution comprising a taxane-containing organic phase and an aqueous phase;
(e) removing the aqueous phase of step (d) from the taxane-containing organic phase of step (d);
(f) evaporating liquid from the taxane-containing organic phase of step (e) to form a crude taxane mixture;
(g) separating the taxanes from the crude taxane mixture of step (f); and (h) recovering the taxanes.

2. A process for obtaining a crude taxane mixture, which process comprises:
(a) separating intact clippings from one or more Taxus plant wherein said intact clippings are leaves attached to stems;
(b) drying the intact clippings from step (a) to form dried plant matter;
(c) contacting the dried plant matter from step (b) with an organic solvent to extract taxanes from the dried plant matter and obtain a taxane-containing extract;
(d) evaporating the taxane-containing extract formed in step (c) to form a residue and partitioning the residue between an aqueous solution and an organic solvent to form a two phase solution comprising a taxane-containing organic phase and an aqueous phase;
(e) removing the aqueous phase of step (d) from the taxane-containing organic phase of step (d); and (f) evaporating liquid from the taxane-containing organic phase of step (e) to form a crude taxane mixture.

3. The process according to claim 1 or 2 wherein the drying in step (b) is conducted at a temperature of less than 70°C.

4. The process according to claim 3 wherein the drying in step (b) is conducted at a temperature of between 20° C and 65°C.

5. The process according to claim 1 or 2 further comprising a step of grinding the dried plant matter of step (b).

6. The process according to claim 5, wherein the dried plant matter is ground to a particle size of between 40 to 80 mesh.

7. The process according to claim 6, wherein the dried plant matter is ground to a particle size of 60 mesh.

8. The process according to claim 1 or 2 wherein the drying in step (b) is conducted under shaded or dark conditions such that the drying step is not conducted under unobstructed direct sunlight.

9. The process according to claim 8 wherein between 40%
and 80% of visible sunlight is prevented from reaching the intact clippings during the drying step.

10. The process according to claim 9 wherein between 40%
and 70% of visible sunlight is prevented from reaching the intact clippings during the drying step.

11. The process according to claim 1 or 2 wherein the intact clippings are dried at a temperature between 20°C
and 65°C and for time in step (b) sufficient to achieve at least a 25% reduction in weight of the intact clippings.

12..The process according to claim 11 wherein the intact clippings are dried at a temperature between. 20°C and 65°C
and for time in step (b) sufficient to achieve between a 40% and a 70% weight loss of the intact clippings.

13. The process according to claim 1 or 2 wherein the Taxus plant is selected from one or more cultivars of ornamental Taxus plants.

14. The process according to claim 13 wherein the Taxus plant is one or more Taxus species selected from the group consisting of T. X media 'Henryi', 'T. X media 'Densiformis', T. X media 'Hicksii', T. X media 'Dark Green Spreader', T. X media 'Runyan' , T. X media 'Brownii', T.
X media 'Wardii', T. X media 'Halloran'. T. X media 'Hatfield', T. X media 'Nigra', T. X media 'Tauntonii', T.
X media 'Fairview', T. cuspidata 'Brevifolia', T.
cuspidata, and T. cuspidata 'Spreader'.

15. The process according to claim 14 wherein the Taxus plant is one or more Taxus species selected from the group consisting of T. cuspidata, T. X media 'Halloran', T. X
media 'Hatfield', T. X media 'Nigra', T. X media 'Tauntonii', T. X media 'Dark Green Spreader', and T. X
media 'Hicksii'.

16. The process according to claim 1 or 2 wherein the Taxus plant is a plant growing in the wild.

17. The process according to claim 1 or 2 wherein, in step (c) the dried plant matter is present in a weight:volume ratio to the organic solvent of from 1:8 to 1:12.

18. The process according to claim 1 or 2 wherein the plant matter in step (c) is contacted with an organic solvent selected from the group consisting of ethanol, acetone. ethyl acetate, methylane chloride, chloroform, carbon tetrachloride, methyl ethyl ketone, methyl isobutyl ketone, methyl t-butyl ether, methanol and mixtures thereof.

19. The process according to claim 18 wherein the organic solvent is selected from the group consisting of ethanol.
acetone, ethyl acetate, methylene chloride, methyl ethyl ketone, methyl isobutyl ketone, methyl t-butyl ether.
methanol and mixtures thereof.

20. The process according to claim 19 wherein the organic solvent is selected from the group consisting of acetone, ethanol and ethyl acetate.

21. The process according to claim 20 wherein the organic solvent is ethanol.

22. The process according to claim 1 or 2 wherein contact of the plant matter with the solvent in step (c) is promoted by a further step selected from the group consisting of soaking, percolating, soxhlet extracting, agitating, and a combination thereof.

23. The process according to claim 22 wherein contact of the plant matter with the organic solvent in step (c) is promoted by agitating.

24. The process according to claim 22 wherein contact of the plant matter with the organic solvent in step (c) is promoted by soaking and agitating the plant matter in the organic solvent.

25. The process according to claim 1 or 2 wherein prior to contacting the plant matter with the organic solvent in step (c), the plant matter is defatted with a defatting agent selected from the group consisting of hexane, pentane, petroleum ether, isooctane, and mixtures thereof.

26. The process according to claim 1 or 2 wherein the organic solvent of step (d) is selected from the group consisting of ethyl acetate, ether, methyl-t-butyl ether, methylene chloride, chloroform and mixtures thereof.

27. The process according to claim 26 wherein the organic solvent of step (d) is selected from the group consisting of methylene chloride and ethyl acetate.

28. The process according to claim 27 wherein the organic solvent of step (d) is ethyl, acetate.

29. The process according to claim 1 or 2, wherein step (e) further comprises:
(1) contacting the taxane-containing organic phase of step (e) with a drying agent; and (2) removing the drying agent from the taxane-containing organic phase.

30. The process according to claim 29 wherein the drying agent is selected from the group consisting of anhydrous magnesium sulphate, anhydrous sodium sulphate, 4A molecular sieves, calcium chloride and mixtures thereof.

31. The process according to claim 30, wherein the drying agent is anhydrous sodium sulphate.

32. The process according to claim 1 or 2, wherein step ie) further comprises cooling the solution of step id) to a temperature at or below that at which water freezes and separating the liquid organic phase from the frozen aqueous phase.

33. The process according to claim 1 or 2, further comprising:
(1) dissolving the crude taxane mixture of step (f) in an organic solvent to foam a taxane-containing solution;
i2) contacting the taxane-containing solution of step (11 with a solid support to form a mixture of the taxane-containing solution of step (1) adsoxbed onto the solid support;
t31 evaporating the organic solvent from the mixture of step (2) to form a taxane residue coated solid support;
(4) sequentially eluting the taxane residue coated solid support first with a non-polar organic solvent followed by a polar organic solvent so that the polar organic solvent contains the taxanes; and (5) evaporating the polar organic solvent of step (4) to form a second crude taxane mixture.

34. The process according to claim 33 wherein the organic solvent of step (1) is selected from the group consisting of methanol, ethyl acetate and a mixture of methanol and ethyl acetate.

35. The process according to claim 34 wherein the organic solvent is a mixture of ethyl acetate and methanol.

36. The process according to claim 34 wherein the organic solvent is ethyl acetate.

37. The process according to claim 33 wherein the solid support of step (2) is celite*.

38. The process according to claim 33 wherein the solid support of step (2) is selected from the group consisting of silica gel, florisil, alumina and diatomaceous earth.

39. The process according to claim 33 wherein the organic solvent in step (3) is evaporated from the mixture until dryness.

40. The process according to claim 33 wherein the nonpolar organic solvent of step (4) is selected from the group consisting of hexane, petroleum ether and isooctane.

41. The process according to claim 40 wherein the nonpolar organic solvent is hexane.

42. The process according to claim 33 wherein the polar organic solvent of step (4) is selected from the group consisting of acetone, ethyl acetate, ether, methyl-t-butyl ether, methylene chloride, chloroform and mixtures thereof.

43. The process according to claim 33 wherein the polar organic solvent is selected from the group consisting of ethyl acetate and methylene chloride.

44. The process according to claim 41 or 43 wherein the polar organic solvent of step (4) is ethyl acetate.

45. The process according to claim 33, wherein step (5) further comprises:
(a) dissolving the second taxane-containing residue of step (5) in an aqueous mixture comprising water and a polar organic solvent miscible in water to form an aqueous polar organic taxane-containing mixture;
(b) combining the aqueous polar organic taxane-containing mixture of step (a) with a non-polar organic solvent to form a two phase solution comprising a taxane-containing aqueous polar organic phase and a non-polar organic phase;
(c) separating the taxane-containing aqueous polar organic phase from the non-polar organic phase of step (b); and (d) evaporating the taxane-containing aqueous polar organic phase from step (c) to form a third-taxane-containing residue.

46. The process according to claim 45 wherein the aqueous polar organic mixture of step (1) consists of more than 50%
polar organic solvent and the balance is water.

47. The process according to claim 45 wherein the aqueous polar organic mixture of step (1) consists of about 9 parts polar organic solvent and about 1 part water.

48. The process according to claim 45 wherein the polar organic solvent of step (1) is selected from the group consisting of acetonitrile, methanol, ethanol and isopropanol.

49. The process according to claim 48 wherein the polar organic solvent is selected from the group consisting of acetonitrile and methanol.

50. The process according to any one of claims 45-49 wherein the polar organic solvent of step (1) is methanol.

51. The process according to claim 45 wherein the non-polar organic solvent of step (2) is selected from the group consisting of hexane, pentane, petroleum ether, heptane, iso-octane and mixtures thereof.

52. The process according to claim 51 wherein the non-polar organic solvent is hexane.

53. The process according to claim 45 wherein the nonpolar organic solvent of step (2) is hexane and the aqueous polar organic mixture consists of methanol and water.

54. The process according to claim 45 wherein the second taxane-containing residue is partitioned between a greater volume of nonpolar organic solvent from step (2) than aqueous polar organic mixture from step (1).

55. The process according to claim 54 wherein the second taxane-containing residue is partitioned between about 1 part aqueous polar organic mixture from step (1) and about 2 parts nonpolar organic solvent from step (2).

56. The process according to claim 7.. wherein the separating in step (g) is performed by crystallization.

57. The process according to claim 1, wherein the separating in step (g) is performed by normal phase chromatography.

58. The process according to claim 57, wherein the normal phase chromatography comprises:
(1) dissolving the crude taxane mixture from step (f) in an organic solvent to form a taxane-containing solution;
(2) contacting the taxane-containing solution from step (1) with a solid support to form a mixture;
(3) evaporating the organic solvent from the mixture of step (2) to form a taxane residue coated solid support;
(4) loading the taxane residue coated solid support veto a normal phase chromatography column comprising a solid support;
(5) repeatedly passing over the normal phase chromatography column loaded with the taxane residue-coated solid support a mobile phase comprising a mixture of a nor-polar solvent and a polar solvent; wherein upon each pass, the polarity of the mobile phase is increased by increasing the ratio of polar solvent to non-polar solvent:
(6) eluting the taxanes from the normal phase chromatography column as separate fractions; and (7) recovering the taxanes from the separate fractions of step (6).

59. The process according to claim 58, wherein the organic solvent of step (1) is selected from the group consisting of ether, methylene chloride, methanol, chloroform, ethyl acetate and acetone.

60. The process according to claim 58, wherein the solid support of steps (2) and (4) are selected from the group consisting of silica gel, florisil, alumina and diatomaceous earth.

61. The Process according to claim 58, wherein the organic solvent of step (3) is evaporated from the taxane-coated residue until dryness.

62. The process according to claim 58, wherein the non-polar solvent of step (5) is selected from the group consisting of hexane, petroleum ether and isooctane.

63. The process according to claim 58, wherein the polar solvent of step (5) is selected from the group consisting of acetone, ethyl acetate, ether, methyl t-butyl ether, methylene chloride, methanol and chloroform.

64. The process according to claim 58, wherein the normal phase chromatography is performed under pressure greater than atmospheric pressure.

65. The process according to claim 58, wherein the normal phase chromatography column comprises silica gel.

66. The process according to claim 58 further comprising:
(i) identifying the fractions of step (6) which consist of taxol* and the fractions of step (6) which consist of cephalomannine;
(ii) dissolving the fractions consisting of taxol* and the fractions consisting of cephalomannine in a mixture consisting of ethyl acetate and methylene chloride to form a taxane-containing solution;
(iii)loading the taxane-containing solution from step (i) onto a normal phase chromatography column;
(iv) repeatedly passing over the nnxmal phase chromatography column loaded with the taxane-containing solution, a mobile phase consisting of ethyl acetate and methylene chloride, wherein upon each pass, the ratio of ethyl acetate to methylene chloride is increased;
(v) separately eluting taxol* and cephalomannine from the normal phase chromatography column as separate fractions; and (vi) recovering taxol* and cephalomannine from the fractions of step (v).
* trademark

67. The process according to claim 66 wherein the mixture of step (ii) consists of about 20~ ethyl acetate in methylene chloride.

68. The process according to claim 66 wherein the ratio of ethyl acetate to methylene chloride in step (xii) is increased starting from 20:80 to 50:50.

69. The process according to claim 66 wherein the mixture of step (ii) consists of 45% ethyl acetate and 55%
methylene chloride and the taxane recovered in step (vi) is taxol*.

70. The process according to claim 66 wherein the mixture of step (ii) consists of 50% ethyl acetate and 50%
methylene chloride and the taxane recovered in step (vi) is cephalomannine.

71. The process according to any one of claims 1-70, wherein any of the steps are repeated.

72. The process according to claim 71, wherein any of the steps are repeated 1, 2, 3 or 4 times.

73. The process according to any one of claims 1 to 72, further comprising after step (b) separating dried leaves from dried stems so that the dried plant matter comprises only dried leaves.

74. the process according to any one of claims 1, 2 or 33-70 wherein the crude taxane mixture of step (f) consists of taxanes selected from the group consisting of taxol*.
cephalomannine, desacetyl-cephalomannine, baccatin III, 10-desacetyl baccatin III and 10-desaceryltaxol.
* trademark

75. The process of claim 74 wherein the crude taxane mixture consists of taxol* and cephalomannine.

76. The process according eo any one of claims 1, 2 or 33-70 wherein the taxane recovered is taxol*.

77. The process according to anyone of claims 1, 2 or 33-70 wherein the taxane recovered is cephalomannine.

78. A crude taxane mixture obtained according to the process of claim 2.