A Changing Science

Combinatorial chemistry is alive and well, but it doesn't hold the same position that it did some 10 years ago. Conceived as heavy artillery in the fight to develop new drugs, combichem is today merely a foot soldier in that ongoing battle.

Still, it's a reliable soldier that is being deployed more and more effectively by the pharmaceutical industry's scientists. As chemists attending the International Symposium on Advances in Synthetic, Combinatorial & Medicinal Chemistry learned, combinatorial chemistry is an indispensable tool in drug discovery.

The symposium, held earlier this month in Moscow under the chairmanship of K. C. Nicolaou of Scripps Research Institute, was extensive, with almost 50 speakers and more than 1,000 attendees. While the conference covered a lot of scientific ground, one theme that ran through it was the evolution of combinatorial chemistry and the related high-throughput screening of combichem libraries to find drug candidates.

In a talk about high-throughput screening, Alexander Alanine, medicinal chemistry section head in Roche's discovery chemistry unit, admitted that the drug industry was initially smitten with combichem. "Companies embarked upon it with great enthusiasm," he told the meeting, "but it really didn't work out as intended."

One problem, he said, was that its practitioners didn't understand chemical diversity well enough. In the early days of combichem, companies would create a 1,000-compound library from one basic molecular template or scaffold. Such a library was large, he said, but it wasn't particularly diverse and thus not very effective in broad screening for compounds effective against a drug target.

AS AN ALTERNATIVE, researchers can try rational drug design techniques, but for targets such as G-protein-coupled receptors (GPCRs)--an important, but poorly understood, class of proteins--screening of large compound libraries is the only effective discovery approach. "The challenge," Alanine said, "is to reduce a massive haystack to a smaller one."

At Roche's two-year-old chemistry building in Basel, Switzerland, computational chemists are taking inspiration from marketed GPCR drugs by "virtually" shredding them into molecular fragments and recombining them into new compounds. They triage the resulting compounds with Lipinski's rule, molecular weight cutoff formulas, and other tools. Using the compounds that remain, Alanine said, Roche chemists synthesize much smaller collections of analogs for screening runs that boast hit rates of 6%, 10%, and even 14%--much higher than traditional high-throughput screening would yield.

Magid Abou-Gharbia, head of chemical and screening sciences at Wyeth Research in Princeton, N.J., presented a talk on his group's blending of structure-based drug design and combinatorial techniques to discover TMI005, now in clinical trials as a rheumatoid arthritis treatment.

TMI005, an orally active small molecule that works by inhibiting the TNF-alpha converting enzyme (TACE), was conceived as a complement to Wyeth's TACE-inhibiting protein drug Enbrel. Taking note of a unique "bend" in the enzyme's binding pocket, Wyeth scientists designed molecular scaffolds that had the potential to inhibit the enzyme. Armed with this knowledge, a team of chemists started making related compounds via combinatorial and medicinal chemistry techniques. TMI005 was one of about 3,000 compounds the group made in an iterative process of synthesis and screening.

According to Abou-Gharbia, Wyeth also employed combinatorial chemistry in its receptor/ligand-based design approach to identifying 5-HT6 antagonists for cognition-related ailments such as Alzheimer's disease. Both programs used combichem in a more focused fashion than in the early days. "The industry went from making thousands of compounds to making smaller arrays. That's the paradigm shift," he said.

Combinatorial techniques are also evolving at companies that supply compound libraries to other firms. Thomas R. Webb, vice president of R&D at ChemBridge, a chemistry services company that helped organize the Moscow symposium, told C&EN that combinatorial chemistry, like all new technologies, had to experience growing pains. "It's naive to pursue a technology thinking that a single technology will change things," he said.

Webb delivered a talk on the use of "molecular stencils" to design compound library templates that have nature as a model. Building on work by chemists S. L. Garland and Philip M. Dean, ChemBridge computational and synthetic chemists searched the firm's database for druglike compounds with geometry similar to that of the beta turn, a common feature of peptides used by the body as receptors. Using what they turned up, the chemists built a family of some 200 relevant templates.

"Synthetic chemists are skeptical, and if they don't understand an idea, they won't embrace it."

THROUGH THIS EFFORT, ChemBridge is attempting to bridge a chasm between computational and synthetic chemists that Webb says has impeded compound library design. "Synthetic chemists are skeptical, and if they don't understand an idea, they won't embrace it," he said. "The molecular stencil is a way to meet in the middle."

Drug companies amassing compounds for screening can buy them from a firm such as ChemBridge or they can synthesize them in-house. At the Moscow symposium, Mike T. Cox, director of global compound management at AstraZeneca, described how his firm is marrying the approaches to build up its collection.

AstraZeneca was formed in 1999 through the merger of Sweden's Astra and the U.K.'s Zeneca. One of the new firm's many postmerger tasks was to bring together all of its compounds into a central location--no simple task, Cox said. "Chemists are hoarders," he explained. "They like to hang onto their compounds just in case they will come in handy."

This proclivity meant that some 300,000 compounds were scattered about the AstraZeneca organization and unavailable to all its chemists. After the compounds were centralized at the firm's Alderley Park research center, Cox's group set about cataloging them. They found only a 10% overlap, but also turned up other deficiencies.

About 24% of the compounds were what Cox called "ugly"--they were toxiphores, had bland functionality, or were not druglike. At the same time, only 66% of the compounds passed quality tests. In the end, just half the company's collection was adequate, a testimony, Cox said, to the numbers-driven compound collecting that characterized the early days of combichem and high-throughput screening.

Early last year, AstraZeneca launched a compound collection enhancement program that combines the purchase of high-diversity compounds with the in-house synthesis of more targeted ones. Cox described AstraZeneca's effort with words that could be applied to the combinatorial chemistry field as a whole. "We are trying to emphasize quality over numbers," he said. "Time will tell if we will be successful."

BUSINESS

Moscow Is Chemistry Service Hub

The chemists who traveled to Moscow for the International Symposium on Advances in Synthetic, Combinatorial & Medicinal Chemistry earlier this month have returned home. Remaining behind in the city, however, is a sizable community of chemists making a living by providing services to Western drug companies.

They work for ChemBridge, Chemical Diversity, Asinex, InterBioScreen, and several other chemistry services firms that together employ more than 1,000 chemists and support staff at laboratories throughout Moscow. Were it not for the emergence of such companies during the past 10 years, these chemists would likely be living abroad or working in other professions.

Asinex is typical of the Moscow chemistry providers. It was started in 1994 by three partners who had identified an emerging need among Western companies for compound libraries used in high-throughput drug and agchem screening.

One of those partners was Dmitry Genis, who in 1993 had earned his Ph.D. in chemistry from Moscow State University. Genis, now Asinex' chief executive officer, explains that the firm began as a middleman, connecting companies with university chemists who, strapped for funds in the post-perestroika era, were eager to make the compounds.

Over time, he says, Asinex shifted to buying the compounds itself, stocking them, and assembling libraries. In 1998, Asinex set up a laboratory and began synthesizing many of the compounds on its own. Today, the firm has 300 employees, including more than 120 chemists who are making compounds and, increasingly, conducting medicinal chemistry and drug lead optimization for customers.

While Asinex and its competitors were expanding, the scientific community in Moscow was collapsing along with the government funding that had supported it until Mikhail Gorbachev declared perestroika in 1985. Of the 220 chemistry Ph.D.s who graduated with Genis, only 12 are working as chemists today, he says.

Although perestroika was a disaster for Russian science, it was a boon for the new chemistry providers, which were able to hire the best Russian chemists. As recently as four years ago, Genis says, Asinex could offer salaries two to three times the average for Moscow professionals. Today, with the economy reviving, he says he is still paying roughly 30% more than the going rate.

Of course, for the chemists working at such firms, it's not just about the money. Symposium attendees who toured ChemBridge's facility saw four floors of new laboratories that approach Western academic and corporate standards. ChemBridge, probably the largest of the Moscow chemistry firms, shares a building with the Moscow Institute of Fine Chemical Technology, and the contrast between the two organizations' labs was startling.

Vladimir Trukhan, a ChemBridge laboratory head, says he probably would be working abroad if he wasn't at ChemBridge. Trukhan earned a Ph.D. from Moscow's Institute of Chemical Physics in 2000 and then spent two years as a postdoc at Lund University, in Sweden. He figured his next move would be into industry in Western Europe, a possibility that didn't please his wife.

But Trukhan learned about ChemBridge from friends and, on a vacation back home, decided to visit. He came away realizing he had found a job in the place he least expected. "I liked working in Sweden," he says, "but home is home."

Alexander Kasatkin, a Moscow University Ph.D. who returned home in 2001 after six years abroad, tells a similar tale. Kasatkin was all set to rejoin the Russian Academy of Sciences, where he had worked from 1984 to 1994. Instead, he heard about ChemBridge and is now a lab head there, working in facilities that he says are significantly better than those at the academy.

Indeed, the story is not as happy at Moscow's educational institutions, which continue to struggle to make ends meet. Konstantin V. Kudryavtsev, an R&D group head at Moscow State, says the six chemists in his lab bring in needed funds by generating catalyst possibilities for DuPont and libraries of screening compounds for FMC's ag division.

Kudryavtsev would like to generate more such work for his lab and laments the fact that most of the work is going to private companies like ChemBridge and Asinex. "Before perestroika," he says, "only academics could work for the West."