Plant Samples for Pharmaceutical Screening

©The World Botanical Associates Home Page
Last Updated: October 2006

The World Botanical Associates provides samples of plants for research and development of new pharmaceutical drugs. The species we collect are usually not offered in alternative medicines, or those alleged to contain beneficial properties that are marketed as nutritive foods or food additives.  The small (general) samples of indigenous species that we provide represent our selection obtained from random, systematic, or more selective approaches, discussed below. 

The following tables indicates costs for general samples obtained randomly or systematically according to numbers of species sampled (50, 100, 250, 500, 1000) and weight class (1–25 g, 25–50 g, 50–100 g, 100–250 g, 250–500 g, 0.5–1 kg, 1–2 kg), and also are further differentiated by samples limited to just the aerial parts (twigs and leaves combined, or stems–leaves-flowers and fruits, usually combined), or when a sample of root is included (with rt).
 

1. Random (biodiversity) Samples are collected as species are encountered in the field with the limitation to avoid duplication of samples from the same species (Perdue and Hartwell 1969), except for species occurring in different phytogeographic regions, or when different varieties or subspecies can be recognized.  Excluded are species not native to the United States (e.g., weeds of European origin), and species that are considered rare, threatened, sensitive or endangered.  Samples are air-dried unless otherwise specified.  All samples obtained from one species belong solely to a single contract; i.e., samples from a particular plant species are not shared by more than one contractor at the time of collection. Vouchers for random samples may be limited to photographs (soft vouchers).  For an additional $10 per species, a hard vouchered specimen will be obtained. Each species sampled is assigned a collection number and a record is made of the location, date collected, ecological data, habit of the plant, color of flowers, and other features that may be pertinent to species identification.  A separate (WBA) number is assigned to each sample.  A further discussion on the limitations to a random screening methodology can be found in Spjut (1985).

Samples may be shipped in lots of 50 or more.  A shipping list will be provided that indicates species name, family name according to the APG (Angiosperm Phylogeny Group system) II, State of collection, date of collection, plant parts, weight of sample to the nearest gram or nearest 10 grams for samples over 100 g.  Plant parts will be abbreviated from base to top of plant as follows: rt (root, bark not separated), rb (rootbark), wr (wood of root without bark), wst (woody stem, bark not practical to separate, usually applied to semi-woody plants), ws-sb (woody-stem with stem-bark attached), ws (woody-stem with bark removed), sb (stem-bark), st (stem, applied to herbaceous plants), tw-lf (twig-leaf combined), tw (twig), lf (leaf), if (inflorescence), fl (flower), fr (fruit), sd (seed).  Shipping is an additional cost that generally runs $1.00 to $1.50 per pound. 

Aerial plant parts. Costs are based on samples that exclude roots.  For shrubs and trees, two samples may be collected, a woody-stem with bark (ws-sb) and twig-leaf.  Herbaceous species and semi-woody species will usually consist of one sample of the aerial parts (stems-leaves and flowers or fruits combined).  The expected minimum number of samples can be calculated by multiplying the number of species by 1.5.  For example, an agreement for 100 species will likely come from trees, shrubs and herbs that would yield on average 150 samples.  For an additional $0.10 per gram, bark can be separated from wood, and leaves can be separated from twigs, when feasible, and when available a separate sample of flowers/and or fruits or seeds may be obtained. 

Unless otherwise specified, collecting will initiate in southern California.  The collecting season generally runs from the beginning of February through October.  Collections from 50-100 species may be completed in one week to one month depending on the size of the sample.  Agreements requiring 500 to 1,000 species can be completed in one season if collecting commences early in the season and is limited in dry weight to no more than 100 g.

Root Samples are obtained from most woody plants and perennial herbs.  Agreements that include root can expect the total number of samples to be twice the number of species; for example, a contract for 50 species that includes root samples should yield 100 samples.  Because all material is exclusive to a single contractor, the sample weights are not uniform.  Also, root will tend to weigh less than aerial parts.

It should be noted that costs are species based, not sample based. In collaborative research programs, which usually operate under a reimbursable agreement, random samples are often justified by the lower cost per sample as a result of dividing plants into as many samples as practical    For example, a flora in the Beltsville region of Maryland of nearly 800 species (Terrell et al. 2000) includes 12 species of oak (Quercus)—mostly trees—that may yield up to 9 samples for each species.  Such samples may include (1) wood of root (wr), (2) bark of root (rb), (3) wood of stem (ws), (4) bark of stem (sb), (5) branch (ws-sb), (6) twig (tw), (7) leaf (lf), (8) inflorescence (catkins), and (9) fruit (fr: acorns).  One may ask whether it is necessary to collect all twelve species of oak, which in this particular case would yield a total of 108 samples, as opposed to a more selective approach that might limit sampling to fewer species of oak and plant parts that are most likely to show activity.  Random samples in a sample based approach are least expensive—because the only restriction is not to duplicate samples of species.

Advocates of random screening often cite the large number of species in the world flora, ~250,000 species of vascular plants, as available for collection, but in reality not all are equally available because many species are rare; thus, screening all of the species in the world flora would appear to be an impossible task.  Those who advocate selection of plants based strictly on use in medicinal folklore often look upon the random method as a blind approach, whereas others refer to it as biodiversity prospecting.  The chances in discovering new drugs from random screening are often compared to that of winning a lottery.

2. Systematic sampling is a methodical approach that obtains the broadest chemical diversity for the least cost—based on phytogeographic, taxonomic, ecological, and pharmacological data—with emphasis on plant parts most likely to show activity (Spjut 1995; Spjut et al. 1992).  In contrast to random samples, systematic collections are more limited in the number of samples obtained from a geographical area, the number of species sampled within a genus, and in the selection of plant parts. 

The basic idea is to systematically eliminate taxa in the search of plants most likely to yield new medicinal drugs.  The cooperative effort between the NCI and ARS during the late 1970's had evolved to the point where much of the world vegetation had been precluded from screening (Spjut 1985).  Consequently, procurement of samples became more expensive.  The logistics involved in predicting which species are collectable and what it will cost for their procurement present a difficult challenge that few botanists seem to recognize. 

From a systematic approach the world flora does not appear all that large.  In contrast to the random approach that considers 250,000 species, there are only approximately 15,000 genera of vascular plants classified in 400–500 families.   The world phytochoria—the division of the world into its natural floristic geographic units—is recognized to have only 58 floristic regions based on distribution patterns of genera (Spjut 1985).  As in population statistics, one does not need to collect every species in each of the regions, but obtain only a “sample” of them.  Spjut (1985) further recommended that the number of species sampled in a genus be limited to seven (7) species.  This was based on the data compiled from the NCI screening of ~20,000 species of plants during 1960–1974.  Nonetheless, a long term program requires a more thorough inventory (library) of extracts in order to more quickly pursue leads on new compounds.  The NCI, which recently suspended its plant acquisition program, has been developing as an extensive library of natural product extracts, which are preserved in cold storage.   These extracts are available to qualified entities.   

Targeted (Selective) Samples (Medicinal plants, recollections, taxonomic samples), Recollections, Taxonomic Surveys, Biomass Surveys.  These are undertaken on a cost reimbursement basis, or may be pursued based on cost estimate contract.  Rates for a reimbursable agreement are as follows:

      Short-term contracts, less than 60 days.

      Salary—$48 per hour—during April through September—based on 10 hour work days, and 7 days per week for up to 6 week periods.  March and October are 9 hour days, and field work undertaken during November through January is based on an 8 hour day. Long-term contracts—those more than 60 days—are charged $40 per hour. 

      Vehicle—$50 per day plus gasoline. We usually drive on average 150 miles per day.  Standard government rates have been $0.48 per mile.  We employ a 4-wheel drive (XUV Envoy) vehicle for field work.  Rather than keep track of mileage, we charge a flat rate, which usually works out to be less than that based on the standard government rate.  Cost for gasoline is figured at $3.25 per gallon, or approximately $35 per day for a 4-wheel drive vehicle.

      Lodging—based on actual cost—averages $100 per day (includes 9-10% taxes and energy sur charges).  Because camping is often restricted to specific areas, it is generally more cost effective to stay at a motel.  Collection schedules will vary according to weather conditions, road conditions, and seasonal activities or holidays.  It is not always possible to shop for the best motel rates.  Nationwide motel franchises generally fluctuate less than small hotel operators who will often triple their prices when room availability suddenly become limited.

      Per diem—$50 per day.  Government rates are often based on travel to a specific area.  These vary widely.  While our field rate may appear higher than government rates for rural areas, we prefer not to obtain our meals at fast-food places.

      Summary for average daily field costs

          Short term—salary $480 + vehicle $85 + lodging $100 + per diem $50 = $715

          Long term—salary $400 + vehicle $85 + lodging $100 + per diem $50 = $635

All costs assume field work is being conducted for a single entity.  Lower costs will be passed on when travel costs can be shared by more than one entity.

3. Plants Used in Folk Medicine. A certain mystique about tribal medicine allures many to a search for magical secret remedies in plants.  This “Indiana Jones” approach is the basis upon which government authorities claim intellectual property rights to (patent) their indigenous (uses of) plants (Witmeyer 1997; Lesser 1997).  However, many species alleged to have therapeutic value are in reality nothing more than common weeds; thus, there appears to be no hidden secrets in these plants, especially since such plants will likely be collected in a random screening program.  

Nevertheless, retrospective studies have indicated that plants used in folk medicine are more likely to show biological activity—generally 1.6–2.0 times greater than that of random collections (Spjut and Perdue1976; Spjut 1985; Spjut 2005), especially if plants are reported to be poisonous (Spjut and Perdue 1976; Spjut 2005).  However, activity in these studies was also due to cytotoxicity (KB in vitro), which in some cases the active agents may cause more harm than good (Cassady and Suffness 1980; Spjut 2005).   Yet, it is interesting to note that Taxus is 9th among 39 genera and species of plants most frequently reported to cause human poisoning in the United States (Marderosian and Liberti 1988, Table 1), and that the discovery of taxol was based on KB data (Cragg et al. 1993).    And while it is easier to draw correlations between plants used in folk medicine and pharmacological activity (Farnsworth et al. 1985; Spjut and Perdue 1976; Spjut 2005), than with classification systems (Latin nomenclature, phytogeographic distribution patterns, bioassay screening, and FDA regulations; Spjut 1985; Spjut et al. 1992), the folklore activity relationship that is evident may be due to compounds that provide nothing more than a general therapeutic effect instead of an effective treatment for a particular disease such as cancer (Spjut 2005). 

A major disadvantage to the strict acquisition of plants used in folk medicine is cost.  Procurement of samples based strictly on alleged (folklore) use in medicine, or other basis such as cytotoxic activity, or any combination of these criteria (e.g., Gustafson et al. 1992), is expensive.  In seeking out such plants, a considerable effort has to be made to define what species and plant parts should be collected and to determine just where the species can be collected.  In North America, for example, one might evaluate 2,147 species reportedly used in American Indian medicine (Moerman 1986) with also consideration to 3,000+ species listed by Hartwell for folk use against cancer.   Whether there are hundreds or thousands of species candidates, one is not likely to obtain more than one or two samples each day in the field, but with random collecting, one can easily obtain up to 50 one kg samples each day (Spjut 2005).

Permits are usually required for collecting samples.  These can be difficult to obtain when it involves protecting the interests of the parties who have invested a significant amount of resources into research and development of a new pharmaceutical drug, while also assuring that the land owners will receive a fair share in the royalties.  Our estimated costs for samples do not include collecting and export permits for plants outside the United States.  The cost for this involves correspondence, travel to the host country to meet with officials and/or sponsors, and legal fees.  For collections of selected taxonomic groups (e.g., family, genus, species) that occur in many countries, blanket agreements may be offered to authorities in each country.  Under such an agreement, all participants would agree to receive a portion of the royalty payment if a new drug were to emerge, regardless the particular species or country that becomes the preferred source for the active compound(s) of interest.  Royalties generally range from 0.5-2% of the gross sales.

Safety factors. Buyer also assumes responsibility for following safety protocol in handling and grinding samples. We cannot always be certain which samples may cause dermatitis or other allergic reactions. As a general rule, extra precaution should be taken when grinding and extracting plants that contain resins or latex (e.g., Anacardiaceae, Apocynaceae, Asclepiadaceae, Burseraceae, Clusiaceae, Euphorbiaceae, Moraceae, Sapotaceae).  We note on our shipping lists problems that we experience in the collection and drying of samples; examples are spines (e.g., Holacantha), stinging hairs (e.g. Eremocarpus), skin reactions (e.g., Teloschistes chrysopthalmus), or breathing difficulties (e.g., Thymelaeaceae).

Other Considerations

Taxonomy

Distribution of Activity According to Plant Classification. The genus is generally the lowest taxonomic level with the greatest chemical diversity; i.e. species within a genus are likely to yield similar active compounds, whereas those in different genera are likely to be different (Spjut 1985). For example, Senecio and Baileya are genera of the Asteraceae that have yielded antitumor active (P388 Leukemia, KB) alkaloids and sesquiterpene lactones, respectively (Cassady and Suffness 1980; Hartwell 1976). However, other active agents may be discovered by other assays, or may be sensitive to more ubiquitous tannins and sterols, which may have to be extracted out before testing (Hartwell 1976). Additionally, certain classes of active compounds can occur more abundantly in some families (e.g., sesquiterpene lactones, Asteraceae, Hartwell, 1976; phorbol esters, Euphorbiaceae, Gustafson et al. 1992), or predominantly characterize a family (e.g., quassinoids, Simaroubaceae; Barclay and Perdue 1976).

Geography

Phytogeographical v. Political Boundaries. Ideally one might obtain representative samples from each of the following continental floristic areas—Western North America, Western Australia, New Caledonia, the Amazon, the Guyana-Venezuela Highlands, South Africa, southwestern China, Asian dipterocarp rain forests, southern hemisphere temperate forests (Chile or Tasmania or New Zealand), and Madagascar. This floristic division relates to the early diversification of flowering plants that occurred on ancient continental land masses that were in closer proximity to one another (Gondwanaland and Laurasia) at a time when dinosaurs were dying out.   Spjut (1985) further divided the world flora into 58 floristic regions based on distribution patterns of genera and vegetation types.  These are regions of biochemical diversity as the result of isolation by continental drift and ecological diversification. Field work is planned where the largest number of samples with the greatest diversity can be obtained with the least cost, i.e. one collects from as many of the different regions and vegetation types in the shortest distance of travel.

Unfortunately, not all geographical areas are not open to foreign plant collectors for political reasons. For instance, in 1992, the WBA established a partnership in Western Australia (WA) as an Australian enterprise with a strong legal foundation, but this partnership had to be dissolved after an agreement could not be reached with the WA government concerning one of our discoveries—conocurvone from Conospermum unilaterale (Proteaceae).   We have encountered a similar situation in Ecuador where we were unable to ship samples of Castela that were collected with the government's permission.  In both instances, local authorities felt they were not going to get their fair share of the royalties if a potential new drug was to emerge from screening our samples.  One official in Ecuador felt they should get 50% of the gross sales from a drug, whereas a royalty of 1-2% was offered. But we could not cite actual cases where royalty money was being paid out to foreign countries for discoveries from plant products so there seemed to be a certain amount of mistrust and confusion towards reaching a mutual agreement.  It might be noted, however, that pharmaceutical companies have provided advanced payments of up to $1,000,000 for licenses to investigate natural products in some areas.  One WA official mentioned this; however, the WBA was conducting work under a reimbursable agreement with a government agency, the NCI, and that agreement provided only for reimbursement of the costs incurred.  Moreover, the WBA (based in Maryland) did not have any royalty agreements with any institution, and was not in any position to gain any monetary value for development of a new drug from smokebush or Castela.

Prior to the establishment of taxol as a commercial cancer drug, the status of which might be correlated with a Federal government award of a taxol CRADA—23 Jan 1991 (Goodman & Walsh 2001), royalty issues were never raised in obtaining plant collecting permits.  During the 1970's and 1980's, government officials freely offered assistance to procuring plant samples.  But today as the world forests continue to disappear from the activities of agriculture, roads, and urban sprawl, university chemists cannot afford to pay out huge sums of money to obtain a license to conduct basic chemical research on foreign plants.  Thus, the investigation of natural products has become more expensive due to political ramifications in obtaining permits.  It is interesting to note that taxol was not a patented compound; however, the U.S. government did express a desire for remuneration in the commercialization of taxol as a result of the taxol CRADA (Goodman & Walsh 2001).

In further regard to the WBA procurement of Australian smokebush, the WA authorities requested that the funds be given directly to them instead of allowing WBA do perform the work after which the WBA had already made a substantial investment in defining the problems and costs (see Lesser 1997).  Notwithstanding, the NIH, USDA, OICD, and other U.S. government agencies have programs that provide opportunities to foreign scientists to visit and conduct research in the U.S. (Cragg et al. 1993), and financial aid for economic development is available.  Unfortunately, the cost to develop a new drug is enormous—generally estimated at $500 million (CBS Evening News Report, June 23, 1999). This has led some to suggest that experimental drugs be marketed as an alternative medicine, or as food nutritives, or food additives (e.g., Gingo biloba). 

Even in the United States, obtaining permits for the collection of plants can become complicated when the custodians of our public lands see their duties as primarily one of protection for aesthetic reasons.  Some national forests in California would not allow collection in wilderness areas, yet proposals to harvest timber in national parks have been submitted.  California chaparral is subject to natural fire, especially in a climax community.  A good example is much of the Los Padres National Forest having been consumed by a summer fire of 2006.

Plant Parts

Pharmacological Activity in Plant Parts as Related to Vegetation Types. Systematic screening also requires consideration to parts of the plant most likely to yield activity (Spjut 1995). The following exemplifies these differences based on samples that were screened during the 1970's by the NCI against P-388 Leukemia and 9KB Cell Culture.  The plants were collected in a broad spectrum of vegetation types, and data were part of another study on whether activity was more .

Percent Active (P388 Leukemia, 9KB) by Plant Parts and Vegetation Types

                             rt/wr   rb   sb    ws    tw    lf   tw-lf
                             _________________________________________

Woody Vegetation (#spp.)

Tropical Rain Forests

  Amazon-Peru 932 spp.        2.8    -    5.0   2.3   2.5   2.0   -  
  Africa-Ghana 107 spp.             3.2   5.7                    1.9

Semi-dry Tropical forests

  Montane, Tanzania 169 spp.  3.7    -    8.7   5.2   0.5   1.2   - 
  Lowland, Kenya 97 spp.      5.1    -    9.7   6.0   2.1   3.0   -

Tropical Grassland/Woodland

  Montane, Tanzania 71 spp.   6.8    -    2.6   3.9   2.8   0     - 

Mediterranean/Desert Scrub

  Turkey, 597 spp.            5.1    -     -     -    -     -    3.7
  Sonoran Desert                                                 2.4
___________________________________________________________________

Herbaceous plants                        aerial parts

  Kenya Montane Rain Forest              1.3
  Tanzania Semi-dry Montane Forest       3.1
  Kenya Lowland Semi-dry Forest          4.5
  Tanzania Grassland/Woodland            6.1
  Sonoran Desert                         7.4

These preliminary screening data indicate that activity was found most often in stem-bark (sb) followed by root (rt).  The relatively higher incidence of activity for root (rt) in plants of drier vegetation types—such as seen in the lowland semi-dry forest tropical in Kenya, the East African savanna, and Mediterranean scrub in Turkey—is related to the physiognomy of the vegetation—in having fewer trees and shrubs that will yield 300–500 g of bark.  Perdue (1976) reported that most antitumor active agents in African plants would have been discovered if sampling had been limited to stem-bark and root.  He further suggested that sampling be limited to fewer plant parts, indicating that twig be substituted for root (rt) if there are no other limitations to procurement.   Examples of significant discoveries by the NCI include taxol that was isolated from stem-bark of Taxus brevifolia (Taxaceae) in the Pacific NW of North America, and conocurvone that was found only in the root of Conospermum (Proteaceae) spp. in Western Australia.

The slightly higher frequency of active species in herbaceous plants of drier regions may relate to evolution of chemical defenses in response to competition for water.  Plants  may increase their chances of survival if chemical byproducts help protect them from being consumed by animals, and/or to inhibit other plants from growing nearby.  For example, the root of Prosopidastrum mexicanum (Mimosaceae), a wide spreading, mesquite-like shrub, 1-5 meters or more in diameter, occurring along the Pacific coastal desert of Baja California, release a strong odor when removed slashed.  The odor is so strong or rank that can be detected at distances of nearly 1,000 meters. The aromatic (terpenoid) compounds in aerial parts of many desert Asteraceae and Mediterranean Myrtaceae undoubtedly help deter grazers or browsers.  In other species, the odors become more noticeable as the plant sample dries, an example is the aerial parts of Penstemon deustus (Scrophulariaceae, western North America) that smells like a wet dog.  An interesting discussion on the various desert life forms and their survival strategies can be found in Frank and Carol Crosswhite, Desert Plants (1984). 

Significance of Active Species from plant samples collected in North America. Spjut (USDA Memorandum 1974) found that nearly two-thirds of all NCI active plants in the United States were collected west of the Continental Divide; however, active agents included ubiquitous tannins and sterols that have had little therapeutic potential.  Other cytotoxic compounds such as sesquiterpene lactones and cucurbitacins account for a notable disproportion of active species in the western region, especially Asteraceae since it is the most abundant dicot family in the Southwest. 

Antitumor active agents relating to plants active in the above table have been reviewed by Cassady and Suffness (1980), Hartwell (1976), Douros and Suffness (1978) and Suffness and Douros (1979). Holacanthone and colubrinol are examples found only in plants of southwestern North America (Holacantha emoryi [Simaroubaceae], Colubrina californica, C. texensis [Rhamnaceae] Wall et al. 1976) that are also related to maytansine and bruceantin, respectively, that were first isolated from East African plants (Brucea antidysenterica, Simaroubaceae; Maytenus spp., Celastraceae). Maytansine, an ansamacrolid usually found in micro-organisms (Actinomycetes, Suffness and Douros 1979), failed phase II clinical trials due to its low therapeutic index, essentially too toxic for humans; however, synthetic derivatives of this and also quassinoid compounds are still being investigated for treating cancer (Spjut 2005). Another novel compound, bouvardin, discovered by Jack Cole (Univ. AZ, Tucson) from Bouvardia ternifolia (Rubiaceae) of the Chihuahua Desert region of Mexico, represents an antitumor active agent reported only from W North America; unfortunately, it failed clinical trials.

The most significant discovery, thus far, is taxol—from Taxus brevifolia (Wani et al. 1971; Cragg et al. 1993; Goodman & Walsh 2001), a species of the Pacific Northwest Coniferous forest region.  Taxol and related taxoids, which have been the subject of much study for treating cancer, are found in other species of the genus outside W North America  (Kingston 1996; Kingston et al. 1990). 

Another active compound—from which derivates are used in cancer chemotherapy is camptothecin—first discovered in Camptotheca acuminata (Wall et al. 1976; Wall and Wani 1995)—a species of the Nyssaceae that occurs in temperate China, but was collected from a plant cultivated in southern California. 

The deciduous forest regions in the eastern United States have yielded several interesting antitumor active compounds such as thalicarpine from Thalictrum dasycarpum (Ranunculaceae), and podophyllotoxin from Podophyllum peltatum (Berberidaceae) (Perdue and Hartwell 1969). Thalicarpine is an example of a compound (alkaloid) that was active in animal test systems that apparently failed to reproduce in Homo sapiens, showing toxicity at doses below those required for activity (Sieber et al. 1976), whereas derivatives of podophyllotoxin, first isolated from Podophyllum peltatum, are currently used to treat the treatment of some cancers (Spjut 2005).

In conclusion, plants of more arid regions may show a higher incidence of activity, but the most significant active compounds come from a broad spectrum of vegetation types.  However, an intensive review by Spjut as reported in his travel reports indicates that roots of desert plants have been infrequently collected for the NCI program.  Indeed, the uses of plants by American Indians indicate root, bark, flower and fruit samples are the most often the parts used.  Future investigators might expect to find novel compounds in desert plants that are thought to have already been investigated by focusing on root and bark samples, and such studies might reinforce this idea by making comparisons to the aerial parts.

Lower Plants as Sources of Antitumor Agents

The first significant antitumor activity in mosses was discovered in Claopodium crispifolium (Thuidiaceae), an endemic to western North America; (Spjut et al. 1986, 1992). However, recollections were inconsistent in activity (Spjut et al. 1988), which may have been due to erratic presence of associated micro-organism(s)—a blue-green alga—Nostoc sp. that was identified in portions of the samples (Spjut et al. 1988). It is interesting that the collectors subsequently obtained samples of Nostoc that showed significant activity against leukemia (P-388; Spjut et al. 1988); that NCI aggressively procured cyanophyte samples for screening; that one of the cyanophyte antitumor agents—cryptophycin from Nostoc sp.—advanced to clinical trials (Smith et al. 1994), and that another—cyanovirin-N—from Nostoc ellipsosorum—inactivates HIV viral strains (Boyd et al. 1997). However, maytansinoid compounds were also isolated from the moss samples (Suwanborirux et al. 1990), suggesting that the active organism may be an actinomycete, which their presence in minute amounts can be highly active; for example, 9 tons of Maytenus buchananii stems yielded ~6 grams of maytansine, which was sufficient for an experimental treatment of several hundred patients (Suffness and Douros1979).

Many lichens have also been screened for antitumor and anti-HIV activity with activity primarily due to screen's sensitivity to ubiquitous lichen substances in aqueous extracts (McCloud pers. comm.), although there has been some interest in activity due to depsidones (e.g., norstictic acid; Akee at al. 1998). 

Home page

RECOLLECTIONS—for Isolation of Novel Compounds

For chemical analysis guided by bioassay screening, it is important to be aware of the many variables. This includes not only the taxonomy but also variation due to genetics, seasonal factors in development stages of the plant, and geography. For example, cardenolides in a desert milkweed (Asclepias albicans) may show cytotoxicity (KB activity) if samples are gathered just before flowering (Spjut 1985).  Recollections in Tanzania of Apodytes dimidiata (Icacinaceae) and Gnidia glauca (Thymelaeaceae) that were obtained during a flush period of growth near the end of the dry season (“pre-rain flush”) showed activity (P-388 Leukemia) in plant parts other than those for which activity was originally discovered; the original samples were taken at the same location but at an earlier time in the dry season—when plant growth was dormant (Spjut 1976 USDA Memorandum).  In Psorospermum febrifugum (Clusiaceae) activity (P-388 Leukemia) reconfirmed only in recollections of root from Tanzania even though collected at different times of the dry season, but failed to reconfirm in samples from other locations such as from Ghana (Spjut 1976 USDA Memorandum).

Active compounds undergoing clinical studies have required massive quantities of samples for isolation and purification; for example, clinical trials of taxol required as much as 30 tons of stem-bark.  At one time this was a critical factor to a decision network study in drug development. Institutions do not want to invest resources to develop a drug if barriers exist as to its supply—such as limited availability of the plant source, or complicated issues in developing agreements as in the case with smoke bush in Australia. Alternative methods such as producing compounds by tissue culture can have problems in procedure as plants often produce secondary metabolites in response to stress, which may be difficult to reproduce in the laboratory, whereas complete synthesis in the lab can be expensive.

Another alternative is cultivation of the plant, but this can take many years, while the need for the compound is often immediate. During the 1970's, Robert Perdue, as Chief of the Medicinal Plant Resources Laboratory in the Agricultural Research Service (ARS), made a special effort to bring into cultivation many plants that contained promising anticancer compounds. These included Tripterygium wilfordii (Celastraceae) from Taiwan, Cephalotaxus spp. (Cephalotaxaceae, from Japan, Korea, and Taiwan), Camptotheca acuminata (Nyssaceae) from cultivated sources in the United States, Brucea antidysenterica (Simaroubaceae) from Kenya, Maytenus buchananii (Celastraceae) from Kenya, Putterlickia verrucosa (Celastraceae) from South Africa, Baccharis megapotamica (Asteraceae) from Brazil, and others.  Cuttings were planted in Kenya, Tanzania, Puerto Rico, Chico (California) and in greenhouses (Glendale, MD). One objective of the ARS was to develop new crops for improving economic development not only in the United States but also in foreign countries. Ironically, the western yew (Taxus brevifolia), the one that did succeed, was not among those under cultivation, however, the western yew was thought to be generally available, and it was also recognized that related species have had a long history in cultivation.

Quassinoids, which are being studied by Professor Paul Grieco at Montana State University (e.g., Grieco et al. 1994, 1995; http://www.chemistry.montana.edu/grieco.html), can also be synthesized; however, Dr. Grieco has found it more economical to use plant material as a starting source. Unfortunately, the best source thus far does not appear suitable for agricultural development, however, similar compounds have been easily produced in tissue culture by scientists in many countries.

Spjut has been involved not only in obtaining recollections for isolation of active agents, but also in the procurement of plant samples required in mass quantities needed for clinical studies.  Samples of this magnitude require considerable study of the taxonomy ecology and geography of the species, and usually related species have to be considered.  Examples include 15 tons of Maytenus buchananii stems from Kenya, 5 tons of Bouvardia ternifolia (whole plant) from Mexico, 4 tons of Thalictrum dasycarpum seed from Wisconsin, Minnesota, and southern Ontario, 1 ton of Colubrina texensis and C. californica stems, and 1 ton of Gnidia subcordata leaves from Kenya. Before undertaking such collections, extensive surveys may be necessary; for example, aerial reconnaissance was conducted by helicopter and by light aircraft over Western Australia for Conospermum spp, and by light aircraft over southern Ontario (Canada) for Thalictrum dasycarpum. This can run as much as $25,000 per mission. Plant material may also require close inspection to insure quality of sample material as was necessary during the collection of many tons of stembark of Taxus brevifolia being provided by many collectors to the NCI contractors. These projects are usually accompanied by detailed reports. In requesting a detailed field study of a species, we need to know time or budgetary restrictions.

TAXONOMY

What you should know about plant species

Species are subject to interpretation. For example, Spjut (1996), recognized 69 lichen species of Niebla and Vermilacinia in California and Baja California (Mexico) where only 17 were previously known (Bowler et al.1994). Taxus has for a long time thought  to include only four species in North America, and four in Eurasia (Bailey 1923), but the genus is much more diverse in Asia (Spjut submitted), consequently, additional species are being recognized (Spjut 1999, unpublished).   It is interesting that Spjut et al. (1993) found the taxoid content in a shrub the Pacific yew to be similar in samples of a shrub from widely scattered sites that was also most similar in morphology and ecology.  This shrubby Taxus is proposed as  new

At the other extreme, Pilger (1903) recognized only 1 species of Taxus. In reviewing species names it may help to understand how they are defined, and which classification best fits one's needs. As a general rule, species that have not been reviewed in more than 50 years may need taxonomic revision. Molecular (DNA) data, which are often utilized by molecular taxonomists to help clarify evolutionary relationships that have been recognized by morphology, are not utilized by us at this time.

What you should know about plant names

The scientific name of a plant includes the genus name followed by the specific epithet; the two together comprise the species name, which may include subspecies, varietal and form names.  In scientific publications, it is often appropriate to make reference to the person (authority) who named the plant, e.g.,  Taxus baccata Linnaeus (Species Plantarum, 1753); the authority is often abbreviated without reference to the publication, e.g. Taxus baccata L. It is customary to mention the authority only the first time when using a species name, and in subsequent citations to abbreviate the genus name, e.g., T. baccata.  

For a scientific name to be accepted, it must conform to the International Code of Botanical Nomenclature (ICBN, published in many languages, Greuter et al. 2000). These rules, which are periodically amended, operate much like legal statutes. Articles of code are numbered, often accompanied by case examples. It is not uncommon to find names that are not in accordance with the code. One reason is that plants often have been misclassified. For example, Rehder (1936) discovered a (type) specimen collected by Maire from Yunnan that Lemeé and Léveillé (Léveillé 1914) used to describe what they thought was a new species of hemlock (Tsuga); it was validly named Tsuga mairei. But Rehder—upon reviewing Maire specimens (at Edinburgh)—recognized it was not a hemlock, but a yew (Taxus), one that he had considered to be the same as that which he had earlier recognized as T. chinensis (Rehder 1919), which Pilger (1903) had first described as a variety (of T. baccata) based on other specimens collected in Sichuan. Although Lemeé and Léveillé made an error in their generic identification, the rules require that the epithet—mairei—be adopted because priority is a primary principle of the code; after all, they were the first to recognize it as a new species even if they were wrong about which genus it belonged to. Although Pilger first described it as a variety, the Code requires that priority be applied only to names within the same taxonomic rank; the correct name, Taxus mairei, was eventually adopted by S. Y. Hu ex Liu (Illustr. Nat. Ind. Lign. Pl. Taiwan 16. 1960), whereas Rehder (1936) and others (Cheng and Fu 1978) had continued to use T. chinensis.

These kind of upsets often lead to confusion, but it would be even more confusing if these rules were not followed.  On the other hand, both Taxus chinensis and T. mairei can be recognized if they can be distinguished from one another, which Spjut (submitted.) does. The taxonomic situation of Taxus has been complicated by the lack of biologically meaningful data to distinguish the species—other than to arbitrarily classify them by geographical data; however, Spjut (1992, 1993, 1998) has discovered leaf anatomical characters to have taxonomic significance.

This is why vouchers are needed, and why it is important to cite the author(s) for the names and the taxonomic reference(s) that were employed for identification. Many journals that publish chemical constituents of plants mention only the Latin name, while others do cite vouchers and acknowledge the person who identified it. But sometimes this is not enough. It may be necessary to describe the plant in some detail, mention where the plant was collected, and the references and taxonomic characters used to identify the plant. After all, this is part of the methodology of the investigation, and a big chunk of the pie is often consumed on getting plant samples for the chemical studies.

The legitimacy of scientific names for use in publications—relating to chemical constituents isolated from plants—may be easily determined if someone has recently published a review, but if the name has not been reviewed in decades, it may require much library work to locate the publication where the species was first described, make a determination on whether a voucher (type) is available, and  how it is being applied taxonomically. For many names published before 1900, the book or journal is often rare, requiring special arrangements with a librarian specialist. Similarly, type specimens are also rare, requiring special arrangements for study. Moreover, many types remain undetermined as the concept of a fixing a name to a type specimen was not routinely practiced until after the 1950’s. Natural product chemists cannot be expected to undertake a taxonomic revision to publish their results, but perhaps a status report on the taxonomy might provide a better perspective of what was investigated.

Some institutions expend a great deal of money checking names of plants without much consideration to the types or voucher specimens. One could easily spend thousands of dollars verifying the validity of a name, but if the taxonomic characteristics of the voucher have been misinterpreted, then the nomenclatural work is meaningless or inapplicable. However, it can be helpful to know synonyms and related names when conducting literature searches on what kinds of chemical constituents may have been previously isolated.

The phytogeographical relationships of Baja California has been the subject of many symposia papers (e.g., Spjut 1997).
Return to Outline

LITERATURE CITED

1. Akee R. K., R. J. Fisher, and T. G. McCloud. P7 Nucleocapsis protein binding natural products: A new category of lead compounds with anti-HIV activity. Amer. Soc. Pharm. 39th Annual Meeting, Orlando, FL.
   
   
2. Anonymous. 1994. Gymnosperms of Sichuan. Department of Biology, Sichuan United University.
   
   
3. Barclay, A. S. and R. E. Perdue, Jr. 1976. Distribution of anticancer activity in higher plants. Cancer Treatment Reports 60: 1081-1113.
   
   
4. Bailey, L. H. 1923. The cultivated evergreens. MacMillan Co., New York.
   
   
5. Bowler, P.A., R. E. Riefner, Jr., P. W. Rundel, J. Marsh and T.H. Nash, III. 1994. New species of Niebla (Ramalinaceae) from western North America. Phytologia 77: 23-37.
   
   
6. Boyd, M. R., K. R. Gustafson, J. B. McMahon, R. H. Schoemaker, B. R. O'Keefe, T. Mori, R. J. Gulakowski, L. Wu, M. I. Rivera, C. M. Laurencot, M. J. Currens, J. H. Cardellina II, R. W. Buckheit, Jr., P. L. Nara, L. K. Pannell, R. C. Sowder II, and finally L. E. Henderson. 1997. Discovery of cyanovirin-N, a novel human immunodeficiency virus-inactivating protein that binds viral surface envelope glycoprotein gp120: Potential applications to microbicide development. Antimicrobial Agents and Chemotherapy 41: 1521-1530.
   
   
7. Boyd, M. R. and K. D. Paull. 1995. Some practical considerations and applications of the National Cancer Institute in vitro anticancer drug discovery screen. Drug Development Research 34:91-109.
   
8. Brummitt, R. K. 1999. A critique of some recent developments in plant nomenclature.  Systematic Botany 24: 108-115.
   
   
9. Carter, S. K. and R. B. Livingston 1976. Plant products in cancer chemotherapy. Cancer Treatment Reports 60: 1141-1156.
   
   
10. Cassady, J. M. and M. Suffness. 1980. Terpenoid antitumor agents. Pp. 201-269 in J. Cassady and J. Douros, eds., Anticancer agents based on natural product models. Academic Press, New York.
    
    
11. Cheng, W-c. and L.-k. Fu. 1978. Taxaceae. In Flora Reipublicae Popularis Sinicae. Tomus 7, Gymnospermae. Agendae Academiae Sinicae. [English Translation].
    
    
12. Chongming, W., A. A. Leslie Gunatilaka, F. L. McCabe, R. K. Johnson, R. W. Spjut, and D. G. I. Kingston. 1997. Bioactive and other sesquiterpenes from Chiloscyphus rivularis. J. Nat. Prod. 60 (12): 1281-1286.
    
    
13. Cragg, G.M., M.R. Boyd, J.H. Cardellina II, M.R. Grever, S. Schepartz, K.M. Snader, and M. Suffness. 1993. The search for new pharmaceutical crops: Drug discovery and development at the National Cancer Institute. p. 161-167. In J. Janick and J.E. Simon (eds.), New crops. Wiley, New York.
    
    
14. Crosswhite, F. S. and C. D. Crosswhite. 1984. A classification of life forms of the Sonoran Desert, with emphasis on the seed plants and their survival strategies. Desert Plants 5(4): 131-161.
    
    
15. Czeczuga, B., B. D. Ryan, R. W. Spjut, J.-A. Flock, W. A. Weber, C. W. Beasley, R. E. Showman, R. D. Worthington and V. L. Boucher. 1997. Carotenoids in lichens from the United States of America and Mexico. Feddes Repertorium, 108 (5-6): 401-417.
    
    
16. Decosterd, L. A., I. C. Parsons, K. R. Gustafson, J. H. Cardellina II, J. B. McMahon, G. M. Cragg, Y. Murata, L. K. Pannell, J. R. Steiner, J. Clardy and M. Boyd. 1993. Structure, absolute stereochemistry, and synthesis of conocurvone, a potent, novel HIV-inhibitory naphthoquinone trimer from a Conospermum sp. J. Amer. Chem. Soc. 115: 6673-6679.
    
    
17. Farnsworth, N. R., O. Akerele, A. Bingel, D. Soejarto & Z. Guo. 1985. Medicinal plants in therapy. Bull. World Health Org. 63: 965-981.
    
    
18. Greuter, W., J. McNeill, F. R. Barrie, H. M. Burdet, V. Demoulin, T. S. Filgueiras, D. Nicolson, P. C. Silva, J. E. Skog, P. Trehane, N. J. Turland and D. L. Hawksworth. 2000.  International Code of Botanical Nomenclature (St. Louis Code) adopted by the Sixteenth International Botanical Congress. St. Louis, MO, July-August 1999, Koeltz Scientific Books, Königstein, Germany.
    
    
19. Grieco, P., J.M. VanderRoest and M. Piñeiro-Nuñez. 1995. Polyandrol, a C₁₉ quassinoid from Castela polyandra, Phytochemistry 38: 1463.
    
    
20. __________, E.D. Moher, M. Seya, J.C. Huffman, and H.J. Grieco. 1994. A quassinoid (peninsularinone) and a steroid from Castela peninsularis. Phytochemistry 37: 1451.
    
    
21. Gustafson, K. R., J. H. Cardellina II., J. B. McMahon, R. J. Gulakowski, J. Ishitoya, Z. Szallasi, N. E. Lewin, P. M. Blumberg, O. S. Weislow, J. A. Beutler, R. W. Buckheit, Jr., G. M. Cragg, P. A. Cox, J. P. Bader, and M. R. Boyd. 1992. A nonpromoting phorbol from the Samoan medicinal plant Homalanthus nutans inhibits cell killing by HIV-1. J. Amer. Chem. Soc. 35: 1978-1986.
    
    
22. Hartwell, J. L. 1976. Types of anticancer agents isolated from plants. Cancer Treatment Reports 60(8): 1031-1067.
    
    
23. Hartwell, J. L. 1967-1971. Plants used against cancer. A survey. Lloydia 30-34, p.p.
    
    
24. Kingston, D. G. I. 1996. Improving on nature. Taxane J. 2: 20-27.
    
    
25. Kingston, D. G. I, G. Samaranayake and C. A. Ivey 1990. The chemistry of taxol, a clinically useful anticancer agent. J. Nat. Prod. 55: 1-12.
    
    
26. Lesser, W. H. 1997. International treaties and other legal and economic issues relating to the ownership and use of genetic resources. Pp. 31-50 in Global Genetic Resources: Access, Ownership, and Intellectual Property Rights, ASC, Washington, DC.
    
    
27. Léveillé, H. 1914. Monde des Plantes sér. 2, 16: 88, 20.
    
    
28. Marderosian, A. D. and L. Liberti. 1988. Natural product medicine. George F. Sticley Co., Philadelphia.
    
    
29. Marin, R. M. 1999. A collection of medicinal plant samples from Africa and elsewhere. World Botanical Associates, Web Page Press.
    
    
30. McCloud, T. G., H. G Sheffield, Jr., J. Nemec, G. Muschik, M. Suffness, G. Cragg, J. Thompson, and K. Snader. 1989. Extraction of bioactive molecules from marine organisms. Amer. Soc. Pharm. 30th Annual Meeting, San Juan, Puerto Rico.
    
    
31. Moerman, D. E. 1986. Medicinal Plants of Native America. 2 Vols., Univ. Michigan Museum Anthropology Tech. Rep. 19, Ann Arbor. 1998. American Indian Botany. Timber Press, Portland.
    
    
32. Perdue, R. E., Jr. 1976. Procurement of plant samples for antitumor screening. Cancer Treatment Reports 60(8): 987-1005.
    
    
33. Perdue, R. E., Jr. and J. L. Hartwell. 1969. The search for plant sources of anticancer drugs. Morris Arboretum Bulletin 20(3): 35-53.
    
    
34. Pilger, R. 1903. Taxaceae-Taxoideae—Taxeae. In Engler, A. (Ed.), Das Pflanzenreich IV: 5.
    
    
35. Rehder, A. 1919. New species, varieties and combinations from the herbarium and the collections of the Arnold Arboretum. J. Arn. Arb. 1: 44-60.
    
    
36. Rehder, A. 1936. Notes on the ligneous plants described by H. Levéillé from eastern Asia. Taxaceae. J. Arn. Arb. 17: 54.
    
    
37. Rowinsky, E. K., L. A. Cazenave, and R. C. Donehower. 1990. Taxol: a novel investigational antimicrotubule agent. J. Natl. Cancer Inst. 82: 1247-1259.
    
    
38. Sieber, S. M., J. A. R. Mead, and R. H. Adamson. 1976. Pharmacology of antitumor agents from higher plants. Cancer Treatment Reports 60: 1127-1139.
    
    
39. Smith, C. D. E., X. Zhang, S. L. Mooberry, G. M. L. Patterson & R. E. Moore. 1994. Cryptophycin: A new antimicrotubule agent active against drug-resistant cells. Cancer Research 54: 3779-3784.
     
40. Spjut, R. W.  2005.  Relationships Between Plant Folklore and Antitumor Activity: An Historical Review.  Sida 21(4): 2205–2241.  For unpublished reports cited in this paper, see also studies on Medicinal and Poisonous Plants.
    
    
41. Spjut, R. Ms under review. Morphological evolution in the Taxus leaf and its significance to recognizing ecological species within the genus. Systematic Botany. Presented at the AIBS Annual Meeting, American Systematic Plant Taxonomists, Baltimore Convention Center, MD, Aug. 5, 1998, Abstract, Botanical Society of America. Also, Species of Taxus (URL: http://www.botany.org/).
    
    
42. Spjut, R. W. Under Review. Nomenclatural and Taxonomic Notes on Three Confusing Species of Taxus (Taxaceae) in Asia. (submitted Mar 1999; after six months the editors of Taxon sent an EMAIL stating they did not get favorable reviews, and disposed of the paper without providing any indication what was not favorable.  The paper is being resubmitted to another journal ).
    
    
43. Spjut, R. W. 1999. Key to the species of Taxus in North America. WBA Web Page Press.
    
    
44. Spjut, R. W. and R. Whitcomb. 1999. Checklist and Key to species of Carex on the Beltsville Agricultural Research Center, Maryland. WBA Web Page Press.
    
    
45. Spjut, R. W. 1997. The California Floristic Element on Isla Cedros. Paper presented at the Baja California Botanical Symposium, Aug 14-16, Museum of Natural History, San Diego, Abstract.
    
    
46. Spjut, R. W. 1996. Niebla and Vermilacinia (Ramalinaceae) from Baja California and California. Sida Miscellany 14: 225 pp., 73 spp. of which 53 are new, 69 color photos, 52 illus., 129 black/white photos. Introduction, phytogeography, two taxonomic keys emphasizing morphological and chemical characters, and detailed descriptions of genera and species.
    
    
47. Spjut, R. W. 1995. Vermilacinia (Ramalinaceae, Lecanorales), a new genus of lichens. Pp. 337-352 in Flechten Follmann; Contr. Lichen. in honor of Gerhard Follmann, F. J. A. Daniels, M. Schulz and J. Peine, eds., Koeltz Scientific Books, Königstein.
    
    
48. Spjut, R. W. 1995. Occurrence of Mobergia calculiformis in peninsular Baja California. Pp. 475-482 in Flechten Follmann; Contr. Lichen. in honor of Gerhard Follmann, F. J. A. Daniels, M. Schulz and J. Peine, eds., Koeltz Scientific Books, Königstein.
    
    
49. Spjut, R. W. 1995. A systematic approach to collecting plant chemical diversity. Abstract. Paper presented at 36th Annual Meeting, Society for Economic Botany, Cornell University, Ithaca, Jun 21-25.
    
    
50. Spjut, R. W. 1994. A Systematic Treatment of Fruit Types. Mem. New York Bot. Gard., Vol 70, 181 pp, 53 plates, 153 illus.
    
    
51. [Spjut in] Hils, M. 1993. Taxaceae Gray. Yew family. Flora of North America 2: 423-427.
    
    
52. Spjut, R. W., J-Y. Zhou & C-J. Chang. Comparison of taxane content between two different native yews from the Pacific northwest. Poster/Abstract, Thirty-fourth Annual Meeting of American Society of Pharmacognosy, San Diego, CA, Jul 1993.
    
    
53. Spjut, R. W. 1993. Abstract. Reliable Morphological Characters for Distinguishing Species of Taxus. Paper presented at the International Yew Resources Conference, Berkeley, CA, Mar 12-13, 1993.
    
    
54. Spjut, R. W. 1992. Abstract. A taxonomic key to the species of Taxus. Workshop on Taxus, Rockville, MD.
    
    
55. Spjut, R. W., D. G. I. Kingston and J. M. Cassady. 1992. Systematic Screening of Bryophytes for Antitumor Agents. Tropical Bryology 6: 193-202.
    
    
56. Spjut, R. W. 1990. Book Review. Institute of Chinese Materia Medica, China Academy of Traditional Chinese Medicine. 1989. Medicinal Plants in China. World Health Organization, Singapore. 327 pp. softbound. Journal of Natural Products 53(6): 1632.
    
    
57. Spjut, R. W. and J. W. Thieret. 1989. Confusion between multiple and aggregate fruits. Botanical Review 55: 53-72.
    
    
58. Spjut, R.W., J. M. Cassady, T. McCloud, D.H. Norris, M. Suffness, G.M. Cragg, and C.F. Edson. 1988. Variation in cytotoxicity and antitumor activity among samples of a moss, Claopodium crispifolium (Hook.) Ren. & Card. (Thuidiaceae). Economic Botany 42(1): 62-72.
    
    
59. Spjut, R. W., M. Suffness, G. M. Cragg, and D. H. Norris. 1986. Mosses, liverworts and hornworts screened for antitumor agents. Economic Botany 40: 310-338.
    
    
60. Spjut, R. W. 1985. Limitations of a random screen: Search for new anticaner drugs in higher plants. Economic Botany: 39(3): 266-288.
    
    
61. Spjut, R. W. and R. E. Perdue, Jr. 1976. Plant folklore: A tool for predicting sources of antitumor activity? Cancer Treatment Reports 60: 979-985.
    
    
62. Spjut, R. W. 1971. Mosses of the Marble Mountain Wilderness Area, Siskiyou County, California. MA Thesis, Humboldt State Univ. 177 species and varieties identified, 37 new reports for the state of California, moss flora largely circumboreal, disjunct occurrences for many species spport the Klamath Mountains refugium.
    
    
63. Suffness, M. and J. Douros. 1982. Current status of the NCI plant and animal product program. J. Nat. Prod. 45: 1-14.
    
    
64. Suffness, M. and J. Douros. 1979. Drugs of plant origin. Pp. 73-126 in V. T. DeVita and H. Busch, ed., Methods in Cancer Research, Vol. 16. Academic Press, New York.
    
    
65. Suwanborirux, K., C.-J. Chang, R. W. Spjut and J. M. Cassady. 1990. Ansamitocin P-3, a maytansinoid, from Claopodium crispifolium and Anomodon attenuatus or associated actinomycetes. Experimentia 46: 117-120.
    
    
66. Terrell, E. E., J. L. Reveal, R. W. Spjut, R. F. Whitcomb, J. H. Kirkbride, Jr., M. T. Cimino, and M. T. Strong. 2000. Annotated list of the flora of the Beltsville Agricultural Research Center, Beltsville, Maryland. USDA ARS-155, Natl. Tech. Info. Serv., Springfield, VA.
    
    
67. Wall, M. E., M. C. Wani. 1995. Camptothecin and taxol discovery to the clinic. Cancer Research 55: 753-760.
    
    
68. Wall, M. E., M. C. Wani and H. Taylor. 1976. Isolation and chemical characterization of antitumor agents from plants. Cancer Treatment Reports 60: 1011-1030.
    
    
69. Wani, M.C., H. L. Taylor, M. E. Wall, M. E. P. Coggon, & A. T. McPhail. 1971. Plant antitumor agents. VI. The isolation and structure of taxol, a novel antileukemic and antitumor agent from Taxus brevifolia. J. Amer. Chem. Soc. 93: 2325-2327.
    
    
70. Witmeyer, D. M. 1997. The north-south politics of genetic resources: Issues and implications. Pp. 13-30 in Global Genetic Resources: Access, Ownership, and Intellectual Property Rights, ASC, Washington, DC.