Taxus and Taxol - A Compilation of Research Findings

Special Circular 150-99

Taxus Populations and Clippings Yields at Commercial Nurseries

Robert C. Hansen,
Department of Food, Agricultural, and Biological Engineering,
Ohio Agricultural Research and Development Center,
The Ohio State University,
Wooster, Ohio.

Kenneth D. Cochran,
Secrest Arboretum,
Ohio Agricultural Research and Development Center,
The Ohio State University,
and The Ohio State University Agricultural Technical Institute,
Wooster, Ohio.

Harold M. Keener,
Department of Food, Agricultural, and Biological Engineering,
Ohio Agricultural Research and Development Center,
The Ohio State University,
Wooster, Ohio.

Edward M. Croom Jr.,
Research Institute of Pharmaceutical Sciences,
School of Pharmacy,
The University of Mississippi,
University, Miss.

Originally published in the Transactions of the ASAE. Nov.-Dec. 1993, Vol. 36 No. 6, 1873-1877. American Society of Agricultural Engineers. St. Joseph, Mich. Reprinted with permission.

Summary

Commercial nurseries were surveyed during the summer and fall of 1991 as a basis for estimating populations of Taxus cultivars growing in the United States. Clippings of selected cultivars were sampled from nursery fields in Ohio and Michigan in order to estimate expected clippings yields as a function of cultivar and cultivar age.

More than 30 million Taxus plants were reported to be grown by the 19 major nurseries that responded to the survey. Approximately 88% of all Taxus plants reported in the survey were grown in the three-state area of Ohio, Michigan, and Pennsylvania. Taxus x media 'Densiformis,' 'Hicksii,' and 'Brownii' were found to be grown by nearly all nurseries in the survey. More than half grew Taxus x media 'Wardii' and Taxus cuspidata 'Capitata,' while other well-known cultivars seemed to have been specialties of one or two nurseries.

Annual clippings yields on a dry-weight basis (db) ranged from approximately 20 grams per plant to 140 grams per plant. Expected yields were found to be very dependent upon plant age and cultivar.

Taxus x media 'Hicksii' appeared to be the most ideal ornamental yew that could provide a renewable source of Taxol because of immediate availability and potential for mechanical harvesting of upright clippings. An estimated 3,000 to 4,000 ovarian cancer patients could have been treated annually with the Taxol available for extraction from Taxus x media 'Hicksii' clippings.

Introduction

A popular ornamental hybrid yew, Taxus x media 'Hicksii,' was evaluated extensively as a source of Taxol and taxanes (Croom et al., 1991). New-growth shoots, which are typically pruned on an annual basis as a part of standard cultural practices in nurseries, and roots and needles from cull plants were the preferred sources of Taxol and other taxanes from ornamental yews.

Taxus x media 'Hicksii' had been targeted for priority evaluation for the following reasons:

• Taxol and taxane yields were comparable to or greater than those from other Taxus cultivars (Witherup et al., 1990; ElSohly et al., 1992; Mattina and Paiva, 1992).
• Upright shoots produced clippings that facilitated mechanical harvesting.
• Taxus x media 'Hicksii' was among the more abundantly grown ornamental yews in the United States (see survey results later in this report).
• Taxus x media 'Hicksii' was grown as a nursery field crop and therefore offered an immediate renewable resource for annually pruned cuttings which were otherwise discarded.
• Clippings from ornamental yews offered the possibility of being less costly to harvest and process than bark from wild plants such as the Pacific yew tree (Taxus brevifolia).

While Taxus x media 'Hicksii' appeared to be a promising cultivar, many other Taxus cultivars were being collected and evaluated (Croom et al., 1992a; ElSohly et al., 1992; Mattina and Paiva, 1992). More than 100 cultivars are included in the Chadwick Living Herbarium of Taxus on the Wooster campus of The Ohio State University's Ohio Agricultural Research and Development Center.

An example of the concentrations of Taxol and 10-deacetylbaccatin III are shown in Table 1 for needles from six Taxus cultivars. While the highest concentration of Taxol came from Taxus x media 'Hicksii,' the highest concentration of the precursor came from Taxus baccata 'Repandens' Parsons and Sons.

In more recent research, Wheeler et al. (1992) reported that epigenetic factors (timing of harvest, plant tissue type, etc.) and environmental factors (soil type, climate, fertility, etc.) may be confounding efforts to establish expected Taxol and taxane yields for selected Taxus cultivars. They observed that higher Taxol yields were obtained from Taxus brevifolia that grew in cool, moist, shady environments than from trees that grew in warm, dry, sunny climates.

Protocols pertaining to harvesting, handling, drying, processing, and laboratory analyses needed to be defined. Each part of the system could have a significant effect on expected Taxol and/or taxane yields (Hansen et al., 1992; Croom et al., 1992b; Richheimer et al., 1992; Hansen et al., 1993; Ketchum and Gibson, 1993).

Along with identification of ideal cultivars and protocols for Taxol and/or taxane production on a renewable basis, an estimate of populations or acreage of each Taxus cultivar that was being grown in nurseries in the United States and other countries needed to be established. Other than for purposes of propagation, Taxus clippings had not been routinely harvested or stored; therefore, no estimates of potential yields for clippings and/or needles were available.

Objectives

The objectives of this research were:

• To survey nursery growers and account for the diversity and abundance of Taxus cultivars currently being propagated and cultivated for the ornamental market.
• To measure clippings yields as a function of Taxus cultivar and cultivar age.

Procedures

Survey forms, designed to determine types, ages, and quantities of ornamental Taxus grown by nurseries in the United States, were distributed to 68 nurseries in May 1991. The nurseries selected for the survey included all nurseries in the United States that were thought to be growing significant collections of one or more Taxus cultivars. The survey forms requested an accounting of propagation, field-grown, and container-grown Taxus plants. Results were returned from 18 nurseries during summer and autumn of 1991. One additional nursery returned a survey March 15, 1992. The data collected for field-grown plants for 17 of the 19 nurseries that responded to the survey were selected for the summaries and analyses included in this report. The two that were excluded reported that they did not grow Taxus.

During April and May 1992, the quantity of clippings to be expected during annual pruning of different ornamental Taxus plants at various ages was measured (see Table 2). Fifty plants were randomly selected from each field (selection was based on random number tables) independent of field size or number of plants in the field. In fields with plant populations greater than 3,000, four to six rows were randomly selected followed by random selection of 50 plants to be pruned from the identified rows. The degree of pruning and desired plant shape was determined according to the preferences of the grower. Soil conditions permitted retrieval of nearly all plant material pruned from the plants. Pruning was consistently done by one person skilled with the use of hand-pruning shears. All clippings, needles, and bits and pieces of plant material were retrieved from the soil surface below each plant and immediately placed in sealed plastic storage bags. The 50 bags were coded according to nursery, type of cultivar, and age (Table 2).

Upon returning to the laboratory, fresh weights of each of the 50 individual samples were recorded. Five of the 50 samples were randomly selected to be oven dried as a basis for calculating moisture content. Samples were dried at 60°C for 48 hours in standard, brown-paper grocery bags. Based on the average moisture content of the five dried samples, the average yields per plant in grams (dry basis) were calculated for each field along with the standard deviation.

Discussion of Results

Results for the 18 most numerous Taxus x media and the four most numerous Taxus cuspidata were tabulated in Table 3 based on responses from the 17 nurseries. Taxus baccata 'Repandens' is also shown. Rooted cuttings were counted as plants; cutting wood was not. The No. 1 cultivar in terms of total plants growing of all ages at the time of the survey was Taxus x media 'Densiformis' with a total population of 8.36 million plants. Taxus x media 'Hicksii' was second with 5.30 million plants while Taxus cuspidata 'Capitata' was third with 3.77 million. The fourth most numerous plant was Taxus x media 'Brownii' with 3.62 million plants followed by Taxus x media 'Dark Green Spreader' with 2.97 million.

Plant Population by Nursery and Cultivar

Table 3 is a summary of all cultivars that were grown in significant quantities. For the purpose of limiting the complexity of this analysis, a cultivar was designated significant if 100,000 plants (or more) of the cultivar were grown in at least one of the nurseries that responded to the survey. This number included rooted cuttings, all plants ranging from one to 10 years old, and plants greater than 10 years old. Cutting wood was excluded.

Some cultivars may have been left out because more than 40 nurseries did not respond to the census report. There may also be cultivars that were grown in small numbers (less than 100,000 plants) in a number of nurseries where combined production of that cultivar would exceed 100,000 plants. Although the designation of 100,000 plants was arbitrary, it was a conservative estimate of the minimum population of plants in any one nursery that would be required to produce enough clippings to economically justify collecting, handling, drying, and storing clippings for the Taxol/taxane market.

Review of Table 3 revealed that the three most numerous medias (hybrids) ' Taxus x media 'Densiformis,' 'Hicksii,' and 'Brownii' - were grown by at least 15 of the 17 nurseries that responded to the survey. Eleven out of the 17 nurseries evaluated grew Taxus x media 'Wardii,' and 10 nurseries grew Taxus cuspidata 'Capitata.' Otherwise, most cultivars that were grown in significant quantities seemed to be specialties of one or two nurseries.

Plant Population by Age

A summary of plant populations of the more commonly grown cultivars as a function of age is provided in Table 4. A wide diversity in age of plantings was evident from the survey results. Grower estimates were subject to error; however, the survey results provided a good idea of what existing plant populations might have been.

Clippings Yields

A tabulation of measured and estimated clippings yields for three- to seven-year-old plants from five nurseries is shown in Table 5 for Taxus x media 'Hicksii' and 'Densiformis' as a function of age. Table 2 provides the resource data for Table 5. Data available from Table 2 for Taxus x media 'Runyan' were not included in Table 5. The yields per plant listed in Table 2 were samples of growth that occurred during 1991. The 1991 growing season was limited by low rainfall in Ohio and surrounding states. Average rainfall in Ohio is 100 cm; in 1991 the state average was 61 cm. These yields were therefore, at best, rough estimates of expected clippings yields. The large standard deviations were evidence of the wide variation in yields among plants. The moisture contents listed in Table 2 were used to determine clippings yields on a dry-weight basis.

Referring to the codes in column 3 of Table 2, BH3 and RH5 were the only results for expected clippings yields for three-year-old and five-year-old Taxus x media 'Hicksii' plants. The results given for BH3 and RH5 in column 8 - 18.8 and 26.8 - were rounded off to 19 and 27 g per plant (db) and recorded in Table 5. The result for seven-year-old Taxus x media 'Hicksii' plants, 58 g per plant, is the average of SH7 and RH7. Expected yields for four- and six-year-old plants were estimated by interpolation.

For three-, five- and six-year-old Taxus x media 'Densiformis' plants, results from only one field each (SD3, BD5, and LD6) were available for estimating expected yields in Table 2. Again, the raw data 23.6, 135.7, and 142.4 were rounded off and entered in Table 5. The result for four-year-old Taxus x media 'Densiformis' (BD4) was 190 g per plant. This yield was very high relative to all other results; therefore, it was excluded from the analysis. The field could have been unusually fertile or the age was improperly recorded. The yield for four-year-old plants was therefore interpolated from the results for three- and five-year-old plants. The result for seven-year-old Taxus x media 'Densiformis' plants, 60 g per plant, was the average of three fields - SD7, LD7, and ZD7. Recognizing that this data was a rough estimate or "first look" at expected yields, the results showed clippings from Taxus x media 'Densiformis' generally were more than double the expected yield of Taxus x media 'Hicksii.'

No information of this type is known to exist in nursery crop literature. Plants are pruned every year to develop compactness, i.e., a high density of limbs and branches. Clippings have generally been left on the ground. If these same clippings could be harvested for Taxol or taxane production, then what was heretofore waste has the potential to become a second crop.

Table 6 is a tabulation of expected clippings yields for three- to seven-year-old Taxus x media 'Densiformis' plants as a function of age and nursery. Plants that are three- to seven-years-old would have the greatest potential as a renewable source of Taxus clippings since they are in the productive cycle and have to be clipped each year. The results were calculated by multiplying expected yields (Table 5) times plant populations. These results indicated that an estimated 307,208 kg (db) of clippings could have been obtained from the 15 nurseries reporting that they grew Taxus x media 'Densiformis.'

Table 7 reveals expected yields for Taxus x media 'Hicksii' that were developed by following the same procedure described earlier for Table 6. Comparison of the total yields for all nurseries for ages three to seven showed expected yields for Taxus x media 'Hicksii' plants were only 20 percent of expected yields for Taxus x media 'Densiformis.' The lower total plant population (Table 3), combined with much lower yields per plant (Table 5), led to the lower total expected clippings yield.

Traditionally, most field crops are evaluated in terms of yields per acre. When plant populations per acre are known, the yield data in Table 5 can then be used to estimate expected yield per acre. The section entitled Land Area Required (later in this report) can be used (or customized according to plant populations grown by individual nurseries) to facilitate the estimation of clippings yields per acre.

Based on plant population and taxane yield potential from needles only (see Table 1), Taxus x media 'Densiformis' might appear to have promise as a viable source of 10-deacetylbaccatin III. However, existing isolation technologies cannot produce taxanes cost effectively from such low concentrations (Croom et al., 1991). Either more efficient chemical processing technologies must be developed or additional compounds in high concentrations that can be efficiently converted to Taxol must be found in Taxus x media 'Densiformis' for this material to become a viable commercial source.

While Taxus baccata 'Repandens' is apparently the best source of 10-deacetylbaccatin III, the results (based on the nurseries that responded to the survey) showed that only an estimated 221,000 plants were being grown (see Tables 1 and 3). While other ornamental Taxus plants may eventually be identified as promising sources of Taxol or Taxol precursors, Taxus x media 'Hicksii' prevails at present as the most ideal ornamental yew that could provide a renewable source of the drug because of immediate availability and potential for mechanically harvesting upright clippings.

In addition, based on a new process developed by Croom et al. (1991) (patent pending), Taxol and taxane yields for Taxus x media 'Hicksii' may be more than double those shown in Table 1. Even if isolated Taxol yields of 0.010 to 0.014% are obtainable from Taxus x media 'Hicksii,' Croom et al. (1991) estimated 14 to 20 kg (db) of Taxus x media 'Hicksii' clippings would supply enough Taxol (2 g) to provide one full course of treatments for one ovarian cancer patient. Using only the Taxus x media 'Hicksii' clippings identified in Table 7, some 3,000 to 4,000 patients could be treated with the Taxol isolated from these clippings. The isolation and conversion of other taxanes to Taxol from Taxus x media 'Hicksii' and other cultivars could increase Taxol supplies several fold beyond these estimates.

Land Area Required

Some estimate of land area required to grow various Taxus cultivars should be noted. As an example, when plants were transplanted to fields at about three years of age, Zelenka Nursery, Inc., Grand Haven, Michigan, typically planted two rows to a bed with plants spaced 18 inches apart in the row. The rows were spaced 44 inches apart in beds which were laid out 84 inches apart. These spacings led to a theoretical population of 8,297 plants per acre.

The trend is to plant three rows in each bed with all other specifications the same as outlined earlier. This practice leads to a theoretical population of 12,446 plants per acre. After adjusting these populations for turn rows, irrigation ditches, and so forth, reasonable estimates for two-row and three-row plantings in beds may be 8,000 and 12,000 plants per acre, respectively.

Sources of Error

• An attempt was made to survey all nurseries with known significant populations of Taxus cultivars. Some nurseries may have been overlooked; some may not have received the survey forms; some may have lost the forms; and some may have elected not to fill out the forms and return them. Also, no attempt was made to contact nurseries outside the United States. In any event, the 17 nurseries referred to previously may still represent 75 to 90% of all commercially grown Taxus in the United States and therefore the information provided in this report would be relevant and useful.
• There may be Taxus cultivars that are grown in significant quantities somewhere (100,000 plants or more) that are not among the plants grown by the reporting nurseries.
• Because of historical propagation practices, growers may not know which cultivar they are growing. Plant populations within a given field may be generally one cultivar, but extraneous, unexpected cultivars may be dispersed throughout a given field. Even experts have difficulty identifying and distinguishing some cultivars.
• Growers vary in their assessment of plant age. Some determine age based on the time when cutting wood is "stuck" and rooting is initiated. Others may refer to a three-year-old plant based on the time it was transplanted in the field. The survey forms were carefully laid out to minimize these sources of error, but general agreement on plant age does not seem to exist.
• Tracking plant inventory is difficult for many nurserymen. Inventory tracking requires unusual dedication because of plant mortality in the field and the standard practice of selective harvest. Some plants may be alive, but are sufficiently defective that they may not be a viable source of clippings or may not be marketable as an ornamental plant. Accurate inventory is complicated further because of problems associated with distinguishing cultivars.

Recommendations for Future Research

• Survey Taxus plant populations periodically (at least once every five years).
• Focus surveys on cultivars shown to be promising sources of Taxol and taxanes.
• Continue to sample clippings yields annually for plants older than three years.
• Study the potential for harvesting other plant organs and whole plants for Taxol along with clippings.

Questions to be answered include:

• What would be the optimum plant age to harvest?
• What time of year should harvesting be done?
• Should plants be dried, after which needles only are stored and marketed?
• How can agronomic practices be adjusted (row spacing, fertilization, irrigation, choice of climate, etc.) to maximize Taxol/taxane yields per acre?
• How does the dedicated "whole-plant-for-Taxol" system compare economically to harvesting clippings and cull plants as by-products of traditional production and marketing of ornamental yews?

Literature Cited

Broder, S. 1992. Opening remarks. Second National Cancer Institute Workshop on Taxol and Taxus. Sept. 23-24, 1992. Alexandria, Va.

Cragg, G. M., S. A. Scherpartz, M. Suffness, and M. R. Grever. 1993. The Taxol supply crisis. New NCI policies for handling the large-scale production of novel natural product anticancer and anti-HIV agents. J. of Natural Products. 56(10): 1657-1668.

Croom, E. M. Jr., H. N. ElSohly, T. R. Sharpe, and J. D. McChesney. 1991. Research Institute of Pharmaceutical Sciences. School of Pharmacy. The University of Mississippi. Private Communication.

Croom, E. M. Jr., H. N. ElSohly, M. A. ElSohly, J. D. McChesney, and T. R. Sharpe. 1992a. Development of reliable and economic sources of Taxol - clippings of ornamental yews. Second National Cancer Institute Workshop on Taxol and Taxus. Sept. 23-24, 1992. Alexandria, Va.

Croom, E. M. Jr., E. S. El-Kashoury, and H. N. ElSohly. 1992b. Effect of drying conditions on the Taxol content of the needles of ornamental Taxus. Second National Cancer Institute Workshop on Taxol and Taxus. Sept. 23-24, 1992. Alexandria, Va.

ElSohly, H. N., E. A. El-Kashoury, E. M. Croom Jr., M. A. ElSohly, and J. D. McChesney. 1992. Taxol content of the needles of various cultivars of ornamental Taxus. Second National Cancer Institute Workshop on Taxol and Taxus. Sept. 23-24, 1992. Alexandria, Va.

Hansen, R. C., H. M. Keener, and H. N. ElSohly. 1993. Thin-layer drying of cultivated Taxus clippings. Transactions of the ASAE. 36(6):1873-1877.

Hansen, R. C., R. B. Shugert Jr., E. M. Croom Jr., H. N. ElSohly, and H. M. Keener. 1992. Evaluation of bin drying of Taxus biomass. Paper No. 926534. American Society of Agricultural Engineers.

Ketchum, R. E. B. and D. M. Gibson. 1993. Rapid isocratic reversed phase HPLC of taxanes on new columns developed specifically for Taxol analysis. J. of Liquid Chromatography. 16(12):2519-2530.

Mattina, M. J. I. and A. A. Paiva. 1992. Taxol concentration in Taxus cultivars. J. Environ. Hort. 10(4): 187-191.

Richheimer, S. L., D. M. Tinnermeier, and D. W. Timmons. 1992. High-performance liquid chromatographic assay of Taxol. Analytical Chemistry. 64(20): 2323-2326.

Wheeler, N. C., K. Jech, S. Masters, S. W. Brobst, A. B. Alvarado, A. J. Hoover, and K. M. Snader. 1992. Effects of genetic, epigenetic, and environmental factors on Taxol content in Taxus brevifolia and related species. Journal of Natural Products. 55(4): 432-440.

Witherup, K. M., S. A. Look, M. W. Stasko, T. J. Ghiozi, and G. M. Muschik. 1990. Taxus spp. needles contain amounts of Taxol comparable to the bark of Taxus brevifolia: analysis and isolation. Journal of Natural Products. 53(5): 1249-1255.