In the past 100 years, organic chemistry largely replaced bioprospecting, and the advents of combinatorial chemistry and in vitro assays have increased throughput to the point that the pharmaceutical industry now pumps out dozens of new products each year. But most of these “new” drugs are based on small chemical changes to existing drugs; and of the thousands, perhaps millions, of chemical “shapes” available to pharmaceutical researchers, only a few hundred are being explored. What compounds are being missed?

“In combinatorial chemistry, people felt that we could generate a lot of new structures and we could screen much more rapidly,” says Jay Short, president and CEO of San Diego’s Diversa Corp. “The problem is that whereas natural products have a function in the environment, combinatorial chemistry has no functional filter. These molecules are randomly generated and synthesized.”

“You had to search through many more molecules to find those ones with a function,” Short continues. “That, I think, is why combinatorial chemistry has not delivered what it was originally sold as delivering, even though it clearly has delivered in terms of the area of optimizing molecules.”

Computational biology has picked up some of the pharmacological slack, as exemplified by the SHAPES library developed at Vertex Pharmaceuticals; but even these attempts have rested on what was already in the database (1). And, as described in “2001: A dock odyssey” (Modern Drug Discovery, September 2001, pp 26–32), in silico drug design is still in its infancy and many think that our technological grasp outreaches our intellectual base. What else is available to us? Where should we go for new leads, or how should we reexamine the old ones?

Diversity is strength
When researchers first looked in earnest for biologically active substances in marine products, they were surprised by the chemical diversity not only among the products but also between them and compounds isolated from terrestrial sources. This chemical novelty was likely due to the nature of the environment in which the molecules evolved—saltwater, which provided a rich source of halogens. But beyond the chemical diversity, the sea also provided amazing biological diversity.

“If you look at the fundamental phyla of life, there are 34; and 17 occur on land, whereas 32 occur in the sea [with some overlap],” says William Fenical, researcher at the Scripps Institute of Oceanography (La Jolla, CA) and co-founder of Nereus Pharmaceuticals (San Diego). “From the fundamental point of view of biodiversity, the ocean is far more diverse and really would have been the best place to start to develop a natural pharmacy.”

“People developed an interest in convincing the pharmaceutical industry that they ought to begin to look at submarine sources because of this very high chemical diversity,” Fenical continues. “It was a total failure, because there were already things happening with respect to synthetic chemistry and combinatorial chemistry. The people that run these companies are so far extended from marine systems that they felt uncomfortable doing it. No one ever chose to invest the resources—time and financial resources—that they had done in the last 30 years on terrestrial systems.”

The era of computer-assisted design, combinatorial chemistry, and parallel chemistry began to evolve such that natural product work, especially over the past 10 years, became less of a focal point of big pharma. A couple of companies “dabbled”, as Fenical describes it, and found things, but they never made a commitment, leaving the field open to smaller companies (see box, “Small fish with big plans”). In all fairness, he admits, natural product research is difficult for big pharma because it is by nature slower than combinatorial chemistry and there was a move to higher throughput.

“When you take a combinatorial pathway, you start with a molecule that has a couple of requirements,” says Fenical. The compound must have groups that can be functionalized—carbon-carbon bonds are rarely formed—and you need to start with a scaffold with inherent biological activity. For example, paclitaxel, an anticancer drug marketed as Taxol that was derived from the Pacific yew tree, was a revelation, according to Fenical, in that it did some things that never could have been predicted. “It has a mechanism of action that no one could have designed an assay to discover.”“Breeding” an antibody

Octopus’s garden
Not quite as dramatic an event as the discovery of Taxol was the work of Werner Bergmann. In the early 1950s, Bergmann and his colleagues isolated thymine and uridine pentofuranosides from air-dried sponges, calling the two compounds spongothymidine and spongouridine, respectively (2, 3). Before the discovery of these biologically active nonribose or deoxyribose nucleosides, chemists in search of improved antibiotics had concentrated their efforts on modifying the base moieties. Now, there was an interest in making changes to the sugars.

“With the recognition that the sugar moiety could be modified while still maintaining biological activity,” wrote Stringner Yang and his colleagues at the National Cancer Institute (NCI) in a recent review (4), “a significant number of nucleosides were made with regular bases but modified sugars, or both acyclic and cyclic derivatives, including AZT and ultimately acyclovir.” The same work, according to Alejandro Mayer, a researcher at Midwestern University (Downers Grove, IL), led to the synthesis of arabinosyl cytosine, an anticancer agent that is produced by Pharmacia under the brand name Cytosar-U (5).

But this work was the exception to the rule, as the conventional pharmaceutical industry still concentrated its efforts on synthetic and combinatorial chemistries. Marine biology remained a largely untapped source of compounds until the NCI started a discovery initiative in the 1970s, which led to the discovery of a dozen or so products (e.g., bryostatin 1 and dolastatin 10) that just now are reaching or have recently started clinical trials (Table 1).

Most marine bioprospecting has focused on an eclectic menagerie of invertebrate organisms—sponges, tunicates, mollusks, and sea hares, to name a few (6). And a quick survey of the literature suggests that most of the active products being tested are toxins, typically used as part of a defense mechanism.

“On a coral reef, what you have is chemical warfare,” said the NCI’s David Newman in a recent interview. “Every invertebrate needs to have a toehold on the reef so it can feed, and the way to defend that toehold is to develop a more powerful poison sac than the competition.”

The watery milieu of these organisms also helps account for the potency of their toxins. Surrounded by so much water, the toxins are rapidly diluted and therefore must act quickly. An example of just such a toxin is ziconotide, a compound being developed by Elan Pharmaceuticals that is derived from a cone snail. But whereas the snail uses the chemical to poison its prey, Elan hopes to use it to treat chronic pain.

It would be wrong, however, to assume that all of the biologically active compounds being tested are toxins. “Because there is almost no money available for antibiotic or anti-inflammatory drug discovery,” says Fenical, “the people who are out there exploring have grants from the NCI and by definition are using cytotoxicity end points. It is grossly inaccurate to say that the ocean, therefore, is composed of those molecules. We see a broad cross section of diverse structure types that appear to have biological activities across the board.”

Marine microbes
Although much of the early research on marine natural products focused on organisms such as sponges, tunicates, and mollusks, the sources that have garnered attention recently are marine bacteria and fungi. There is even suspicion that some of the biologically active compounds isolated from the larger organisms may in fact have been products of symbiotic microorganisms.

According to Fenical, as you move away from the continents, the sea chemistry changes, and this is reflected in the microbiology. As you go north, the waters become colder; and the microbes have adapted to deal with the temperature. As you move down from the surface, oxygen levels decrease, light diminishes, and pressure increases. Each of these changes is matched by the organisms, resulting in an incredible diversity—one that almost approaches an alien environment.

“What was really surprising,” says Short, “was the relatively low homology or sequence identity across species and an exceptionally low relatedness to other genes.”

The scientific literature of the past 100 years has documented and described perhaps 10,000 microbial species, according to Short, whereas the genomics approach taken by Diversa suggests that this number will shortly increase 200-fold, bringing with it new biologically active molecules.

“Many of these proteins have entirely different catalytic mechanisms than anything that you’ve seen in eubacteria,” says Short. “That’s really what has allowed our enzyme area to take off and why we have such a rich pipeline.”

And as William Fenical says, “The ocean is the last frontier of natural products research, whether big pharma gets into it or not.”

References

1. Bemis, G. W.; Murcko, M. A. J. Med. Chem. 1996, 39, 2887–2893.
2. Bergmann, W.; Feeney, R. J. J. Am. Chem. Soc. 1950, 72, 2809–2810.
3. Bergmann, W.; Feeney, R. J. J. Org. Chem. 1951, 16, 981–987.
4. Yang, S. S.; et al. J. Nat. Prod. 2001, 64, 265–277.
5. Mayer, A. M. S. Pharmacologist 1999, 41, 159–164.
6. Schwartsmann, G.; et al. Lancet Oncol. 2001, 2, 221–225.

Randall C. Willis is an assistant editor of Modern Drug Discovery. Send your comments or questions about this article to mdd@acs.org or the Editorial Office by fax at 202-776-8166 or by post at 1155 16th Street, NW; Washington, DC 20036.