Relationships between Inhibitory Activity against a Cancer Cell Line Panel, Profiles of Plants Collected, and Compound Classes Isolated in an Anticancer Drug Discovery Project (original) (raw)

2006, Chemistry & Biodiversity

In an attempt to determine the relationships between the plant profiles (country of collection, taxonomy, plant part) and the compound classes isolated with cytotoxic activity against a panel of human tumor cell lines, the data compiled from a 15-year anticancer drug-discovery project were subjected to an analysis of variance (ANOVA). The results indicate significant trends in cytotoxic activity relative to collection location, taxonomy, plant part, and compound classes isolated. Plant collections were made in tropical forests in six countries, with collections from Ecuador resulting in higher activity than those from Indonesia and Peru. Interestingly, collections from Florida were not statistically different than those from the countries with higher biodiversity. One hundred and forty-five families were represented in the collections, with the Clusiaceae, Elaeocarpaceae, Meliaceae, and Rubiaceae having low ED 50 (half maximal effective dose) values. Especially active genera included Aglaia, Casearia, Exostema, Mallotus, and Trichosanthes. Roots and below-ground plant materials were significantly more active than aboveground materials. Cucurbitacins, flavaglines, anthraquinones, fatty acids, tropane alkaloids, lignans, and sesquiterpenoids were significantly more active than xanthones and oligorhamnosides. The results from this study should serve as a guide for future plant collection endeavors for anticancer drug discovery. isolation [2] [4], and the elimination or scaling down of natural-product research programs at pharmaceutical companies and governmental agencies [1] [3]. A majority of new medicines derived from plant secondary metabolites have been applied toward the treatment and/or prevention of cancer [2] [5]. Since 1990, there has been a 22% increase in cancer incidence and mortality, with over 10 million new cases and over 6 million deaths worldwide in 2000 (all tissues excluding non-melanoma skin) [8]. Progress has been made in cancer chemotherapy, a considerable portion of which can be attributed to plant-derived drugs [5]. Anticancer agents from plants currently in clinical use comprise four main compound classes (including their synthetic and/or semi-synthetic derivatives): Vinca (or Catharanthus) alkaloids from Catharanthus roseus (L.) G. Don (Apocynaceae; formerly Vinca rosea L.) [9] [10], epipodophyllotoxins from Podophyllum peltatum L. (Berberidaceae) [11], taxanes from Taxus brevifolia Nutt. (Taxaceae) [12-14], and camptothecins from Camptotheca acuminata Decne. (Nyssaceae) [12] [15]. Taxane and camptothecin analogues accounted for estimated sales of over 2.75 billion dollars in 2002, or approximately one third of the global market [13]. Two other compound classes from plants are showing promising results in advanced clinical trials, namely, the combretastatins and their derivatives from Combretum caffrum Kuntze (Combretaceae) [16] [17] and the homoharringtonine alkaloids from Cephalotaxus harringtonia C. Koch (Cephalotaxaceae) [18]. Despite considerable advances in cancer treatment, new drugs are still needed, since some cancers have become resistant to currently available treatments and certain types of cancer lack appropriate drug treatments. As part of a multidisciplinary, multi-institutional research project entitled Novel Strategies for the Discovery of Plant-Derived Anticancer Agents, funded through the National Cooperative Drug Discovery Group (NCDDG) from the U.S. National Cancer Institute (NCI), we have been investigating the anticancer activity of higher plant species since 1990 (see the most recent review [19]). The project has involved plant collection, extract preparation, screening of extracts in cytotoxicity and mechanism-based in vitro bioassays, dereplication of active plant species, activityguided fractionation, compound isolation and structure elucidation, in vivo testing in hollow-fiber and xenograft mouse models, lead optimization, and compound development. Although extensive data have been collected throughout the course of this project, this paper represents the first in-depth statistical analysis concerning the relationship between the plant samples collected, the compound classes isolated, and the resultant in vitro biological test data obtained. This study examines the statistical relationships between the plant profiles, including country of collection, taxonomy, and plant part, of the samples collected through our NCDDG project over a ten-year period (1995-2005) and their cytotoxicity expressed as ED 50 (half-maximal effective dose) values in a panel of human cancer cell lines. Statistical analyses of such a data set may provide insight into taxonomic and ecological patterns of the plants collected related to biological activity, and as such, may help guide future plant-collection efforts in a drug-discovery endeavor. This study also examines the statistical relationship between compound classes isolated from plants during the entire project period (1990-2005) and their cytotoxicity in a panel of human cancer cell lines. Those compound classes found to be most active can be considered during dereplication processes in future drug-discovery efforts from plants. Informa