Discovering promising anticancer agents requires experience and the science of pharmacognosy.
BY DARRELL E. WARD
In August 2000, Drs. Soedarsono Riswan and Leonardus B.S. Kardono, both of the Indonesian Institute of Science, led a small crew into a remote rainforest of Indonesian Borneo on a plant-collecting expedition for a U.S. National Cooperative Drug Discovery Group (NCDDG) project. The team made their way through thick undergrowth, ignoring heat, humidity, insects and rain, as they gathered twigs, leaves, roots, fruits and flowers from selected plants.
They took careful notes, recorded each plant’s global positioning coordinates, photographed the leaves and flowers and preserved reference specimens in herbarium paper. They packed the samples in porous sacks, dried the material in the sun, then shipped it to the NCDDG, based then at the University of Illinois, Chicago.
The project was headed by A. Douglas Kinghorn, PhD, DSc, who in 2004 joined Ohio State as the Jack Beal Professor and Chair of Natural Products Chemistry and Pharmacognosy in the College of Pharmacy, and as a researcher with The Ohio State University Comprehensive Cancer Center – Arthur G. James Cancer Hospital and Richard J. Solove Research Institute (OSUCCC – James).
Kinghorn and his laboratory are internationally recognized in the field of natural product drug discovery. He has isolated more than 250 natural product compounds that show potential anticancer or chemopreventive activity, with more than 50 having a novel structure. They include aromatase inhibitors and pervilleines A-F, potent inhibitors of the drug-resistance protein, P-glycoprotein.
His laboratory evaluates some 400 plant samples per year for promising “lead compounds,” molecules with biological activity and interesting structures. They may investigate a dozen or more plants at a time.
For 15 years he has been editor-in-chief of the Journal of Natural Products, the leading journal in his field. He is series-editor-in-chief of the book series Progress in the Chemistry of Organic Natural Products, and he chairs the Dietary Supplements – Botanicals Expert Committee of the U.S. Pharmacopeia.
Today, Kinghorn directs a $7 million, five-year, National Cancer Institute (NCI) program project grant, titled Discovery of Anticancer Agents of Diverse Natural Origin, which funds the collection and analysis of tropical rainforest plants, and of cyanobacteria and fungi. His collaborators include researchers at the University of Illinois, Chicago; University of North Carolina-Greensboro; and Bristol-Myers Squibb (see sidebar).
That same team was involved in the NCDDG grant and the Borneo collection, which, as it turned out, would yield an extraordinarily interesting compound from a tree in the mahogany family, Aglaia foveolata. Kinghorn named the new substance silvestrol.
Natural products–substances made by living organisms that often have pharmacological activity–have historically been an important source of anticancer drugs. Taxol is one example. According to a recent study, of 155 antineoplastic agents marketed in Western countries and Japan since the 1940s, 47 percent were either unmodified natural products or semi-synthetic derivatives of natural products.
These plant-derived anticancer agents fall into four structural classes: the vinca alkaloids (vinblastine, vincristine, vinorelbine), the epipodophylotoxins (etopside, etoposide phosphate, teniposide), the taxanes (paclitaxel and docetaxel), and the camptothecin derivatives (irinotecan and topotecan). Silvestrol may add yet another.
Hooked on Pharmacognosy
Kinghorn came to his career reluctantly. As a teenager in the United Kingdom, he planned to enter medical school. Then his father, a pharmacist, became seriously ill. He called Kinghorn, the oldest of five children, to his bedside. Concerned that he might die, the father wanted his son to attend pharmacy school so he could look after the family.
Kinghorn enrolled in the pharmacy program at the University of Bradford. There, pharmacognosy–the study of drugs or substances of natural origin, and the search for new drugs from natural sources–captured his imagination. He went above and beyond, completing a literature review of hallucinogens, a microscopic study of parsley, and a gas chromatography analysis of a fixed oil. He graduated with special honors.
“I was hooked,” Kinghorn says. He focused on drug discovery and searching for biologically active compounds in graduate school. “I’ve been faithful to that throughout my career,” he says.
His father? He survived the illness. “He was a fellow of the Royal Pharmaceutical Society of Great Britain, and I became one too, just before he died,” says Kinghorn. “We were the only father and son with the same name to ever be fellows together, and that meant a tremendous amount to him.”
As a postgraduate, Kinghorn studied plant-derived anticancer agents with Norman R. Farnsworth at the University of Illinois, Chicago. In subsequent independent work on identifying sweet plant substances, a review of ancient Mexican botanical literature turned up a 16th century monograph that led Kinghorn to the Aztec sweet herb, Lippia dulcis. From that, he isolated hernandulcin, an oil 1,000 times sweeter than table sugar.
“I got very lucky with that,” he says. “It really was a major discovery among sweeteners.” Science published the findings in 1985.
After discovering several novel sweet substances, Kinghorn–who remains a recognized world authority on phytochemical sweeteners–changed his focus to natural product anticancer agents.
“It’s all drug discovery,” he says. “The trick is to recognize really great leads and dump the not-so-promising stuff. It’s a skill that comes with experience.”
In 1992, Kinghorn became principal investigator on an NCDDG award, and it was that grant which funded the 2000 plant-collecting trip to Indonesian Borneo.
Cutting Through Thickets
Kinghorn and his colleagues chose the Borneo site because his former graduate student, Leonardus Kardono, could help arrange and coordinate the formal plant collection agreement with Indonesia.
To plan collections, the researchers check NAPRALERT, a natural products research database, to identify plants in the area that are underinvestigated. They also give the local collectors a botanical manual indicating which genera to sample. Flowers and fruits are of special interest. “Someone may have to climb 250 feet up to reach them, so it’s not always possible,” Kinghorn says. “A lot depends on how well you can motivate the people out in the field.”
A given plant may produce some 5,000 to 10,000 individual compounds, which can be separated into “primary” and “secondary” metabolites.
Primary metabolites include nucleic acids, amino acids, fatty acids, sugars and other materials required for growth.
Plants produce secondary metabolites, or simply “natural products,” for physiological and ecological reasons. They include the chlorophylls and carotenoids, as well as specialized compounds such as silvestrol that are made by only a restricted group of plants. Secondary metabolites that influence human cells in vitro are potential drugs.
Kinghorn’s search for such bioactive compounds begins with powdered plant samples, which are usually prepared by colleagues at the University of Illinois, Chicago. A pound or two of material will usually yield 100 mg of purified compound, enough for in vitro and initial in vivo testing.
For silvestrol, Kinghorn’s lab made aqueous and chloroform extracts from powdered twigs and fruit and tested them against several cancer cell lines and rapid in vitro assays that provide mechanistic data.
“Crude extracts can contain factors that inhibit or accentuate activity,” Kinghorn explains, “but our screening assays quickly provide large amounts of biological information. If there is something there, we will see it in 90 percent of cases.”
Only the chloroform extract showed cytotoxic activity. Using column chromatography, they divided that portion into eight fractions, then evaporated each of those down and tested the residue against a panel of cancer cells. Fraction five showed the strongest activity, so it was divided into six sub-fractions. Of those, sub-fraction two showed the greatest activity. Continued fractionation yielded pure silvestrol.
In all, Kinghorn’s lab purified two new cytotoxic compounds: silvestrol, a white powder, and episilvestrol, a yellowish gum. Tested against a panel of cancer cell lines, silvestrol had three times the cytotoxicity of episilvestrol, which itself had cytotoxicity comparable to paclitaxel (Taxol).
In Vivo Testing
Silvestrol was both potent and promising. “But we don’t get enthusiastic about a compound until we test it in an animal system,” Kinghorn says.
First is the hollow-fiber assay. Developed at the NCI in the mid-1990s by Melinda G. Hollingshead, DVM, PhD, and Michael Grever, MD, now the Charles Austin Doan Chair of Medicine and co-director of the OSUCCC Experimental Therapeutics program, it uses human tumor cells growing in hollow plastic fibers that are implanted into test animals (see sidebar).
“We rely heavily on this assay,” Kinghorn says. “It requires only up to about 25 mg of the compound versus perhaps 100 mg for an average xenograft study.”
Tested at four doses, silvestrol inhibited the growth of a lung cancer cell line by 15 to 82 percent. “The outcome of the hollow-fiber assay was beautiful. Silvestrol came up trumps,” Kinghorn says.
Finally, the compound was tested in a mouse lymphocytic leukemia model, where it increased survival by 150 and 129 percent over controls, depending on the route of administration.
By then, Kinghorn and his group had worked out the compound’s structure and stereochemistry using nuclear magnetic resonance spectroscopy and X-ray crystallography.
They published the characteristics of the agent in the Journal of Organic Chemistry in 2004. “Our findings suggested that silvestrol should be investigated further as a potential new cancer chemotherapeutic agent,” he says.
Interested Clinician Wanted
Kinghorn had taken silvestrol as far as he could. “We can get a new agent to a certain level,” Kinghorn says, “then a physician-researcher must take it on and foster it.”
He found that physician in Ohio State’s Michael Grever, who also heads the OSUCCC – James Phase I Clinical Trials Program.
“Dr. Grever is one reason I came to Ohio State, and I’m so glad I did,” he says. “Physicians at Ohio State are very open to collaboration. This is part of the whole overall philosophy from top to bottom here. People are actively trying to find collaborations. It’s wonderful.”
Grever was intrigued by silvestrol’s structure and activity. Furthermore, studies by his research associate, David Lucas, PhD, examining silvestrol’s activity in special chronic lymphocytic leukemia (CLL) cell lines and an acute lymphoblastic leukemia mouse model suggested that the agent was more active against B cells than T cells. Those 2009 findings were published in the journal Blood (see also "The Right Collaboration").
To ensure that Grever and Lucas had an adequate supply of silvestrol for their work, Kinghorn organized a recollection of the plant’s bark in 2005.
Purifying larger amounts of a compound from a recollection is a tedious, months-long effort that falls to post-doctoral students, Kinghorn says. “It’s not very interesting and there are few prospects for getting a publication, but it requires care and detailed documentation of the spectroscopy to ensure purity.” Kinghorn becomes part psychologist and part cheerleader to move the work along.
Silvestrol’s development received an important boost when, in 2007, the NCI advanced it to Drug Development Group IIA status. The NCI would examine the agent in xenografts, perform range-finding toxicology studies and develop a clinical formulation. It was an important step toward a phase I clinical trial.
More compound was needed. Dr. Soedarsono Riswan returned to the original collection site and shipped back about 50 kg of dried stem bark, bringing many members of Kinghorn’s NCI drug development grant team into play. The dried material was sent to Dr. Doel Soejarto, a taxonomist at the University of Illinois, Chicago, and a methanol extract was prepared at the University of Illinois Pharmacognosy Field Station, in Downers Grove, Illinois.
The laboratory of Dr. Jimmy Orjala, one of the grant’s project leaders at the University of Illinois, Chicago, concentrated the extract to a residue and shipped most of it to Dr. David Newman, chief of the NCI’s Natural Products Branch in Frederick, Maryland. Newman worked with Thomas McCloud, head of the isolation group at the NCI’s SAIC-Frederick research center, and they produced two grams of 97 percent pure silvestrol for further biological testing.
“Our work is highly multidisciplinary,” Kinghorn says.
The American Cancer Society estimates that more than 1.4 million Americans will be diagnosed with cancer and that 562,300 people will die of cancer in 2009. Most will succumb to recurrent or advanced disease that lacks effective therapy. “We need new agents that have novel mechanisms of action and don’t overlap with drugs currently in use,” Kinghorn says.
Parsing plants for anticancer agents has its challenges. Active compounds tend to occur in low concentration, and they are often chemically unstable or have solubility problems. Political unrest in the nation of origin can delay recollection, and even with recollection, the compound may be missing due to biological variation.
On the other hand, “Because natural products are made by living things, they offer unique molecular scaffolds that are unlikely to arise in synthetic drug-discovery laboratories,” Kinghorn says.
“Natural products research offers a very real opportunity to discover chemical entities that may cure some of our most threatening diseases.”
The Hollow-Fiber Assay
In the mid-1990s, Michael R. Grever, MD, now co-director of the OSUCCC-James Experimental Therapeutics program, was associate director of Developmental Therapeutics at the National Cancer Institute (NCI). He needed a more rapid in vivo method to evaluate new agents produced by the NCI Cancer Drug Screen, one that required a small amount of each novel agent.
Work by Melissa Hollingshead, DVM, PhD, presented an interesting opportunity. To study protein production by the human immunodeficiency virus, she was placing viral-infected cells into hollow plastic fibers, then implanting the fibers in experimental animals.
After recruiting Hollingshead to the NCI, Grever and his colleagues worked with her to turn the idea into today’s screening assay for possible anticancer agents. Tumor cells are grown in fine polyvinylidene fluoride fibers that are implanted in immune-deficient mice. The mice are treated with an experimental agent, which enters the fibers through small pores in the plastic. The fibers are then removed and the effects on cell-growth determined by optical density. Implanting fibers in both the peritoneal cavity and subcutaneously provides a “two-compartment” test that estimates an agent’s ability to withstand systemic circulation. Results are usually available much sooner than with typical in vivo models, which facilitates the advancement of exciting new agents into more detailed animal studies.
The Hunt is On
Grant to fund the discovery of novel bioactive compounds
When A. Douglas Kinghorn, the Jack L. Beal Professor and Chair in Natural Products Chemistry and Pharmacognosy in the College of Pharmacy, arrived at Ohio State in 2004, he directed an NCI-funded, multi-institutional National Cooperative Drug Discovery Group (NCDDG) project aimed at discovering new anticancer agents from tropical plants. Kinghorn headed that program for 15 years and oversaw the collection of more than 2,600 plant acquisitions.
The NCDDG grant was succeeded in 2007 by a $7 million, five-year, NCI program project grant awarded to Kinghorn titled Discovery of Anticancer Agents of Diverse Natural Origin. The new grant funds the collecting and sampling of tropical rainforest plants, blue-green algae, and filamentous fungi for novel bioactive compounds.
The research involves the collaboration of two academic centers, a nonprofit organization, a pharmaceutical company and a biotechnology company, listed here with their areas of responsibility:
The Ohio State University
Project administration, biostatistics
University of Illinois, Chicago
Collecting tropical plants
Research Triangle Institute
Brystol-Myers Squibb, Inc.
Maintains a library of 55,000 filamentous fungi from targeted ecosystems around the world