The Right Combination

"You might have to speak louder," advised Peter Wipf, chemistry professor at the University of Pittsburgh. "There's a concert going on outside my office. Someone named Bon Jovi."<br > In fact, rock star Jon Bon Jovi was at Pittsburgh this April headlining a rally for presidential hopeful Sen. John F. Kerry (D-Mass.) as part of the candidate's campus tour. The presence of this '80s icon at a major 2004 event was indicative of a trend celebrities have relied on for decades: Evolving your image can expand your popularity well past the 15-minute mark.

In the pharmaceutical industry, combinatorial chemistry was the major pop star of the 1990s. This process of rapidly making millions of compounds through parallel synthesis and screening them against target proteins for drug candidates seemed to be the wave of the future.

During the initial fervor, drugmakers encouraged chemists to explore high-throughput synthesis in hopes of revolutionizing productivity. Ten years later, with the promise of brimming drug pipelines unfulfilled, combinatorial chemistry's 15 minutes of fame appeared to be at an end.

"The initial idea was, 'We'll make millions of compounds, and one will be your drug.' That's just unrealistic," says Samuel Gerritz, group leader for lead synthesis at Bristol-Myers Squibb, in Wallingford, Conn. "In hindsight, I'm surprised how many of us bought into it."

In the early days, combinatorial chemistry created lots of mixtures, and the quality of the libraries, even for discrete compounds, was insufficient, says Rongshi Li, head of high-throughput chemistry at ChemBridge Research Laboratories (CRL) in San Diego. Impurities in these early libraries led to many false-positive hits during screening, which generated negative attitudes toward their use in drug discovery.

Nevertheless, the techniques of combinatorial chemistry have evolved to be incorporated into the standard job of drugmaking, and they continue to change as synthetic organic and other chemists in the field make improvements and adaptations. Like Bon Jovi, the updated version of combinatorial chemistry still has its fans and is still headlining key research in the pharmaceutical industry.

"Right now is a relatively good time for synthetic organic chemists to be looking for a job," says Anthony W. Czarnik, visiting chemistry professor at the University of Nevada, Reno, and editor of the Journal of Combinatorial Chemistry. As chemists move up the industry ladder, "there's a constant need for new blood doing the actual chemistry in pharmaceuticals," he says.

STATISTICALLY, at least, the pharmaceutical industry has been holding its own in terms of hiring. A decrease of 67,000 jobs for the chemical industry as a whole between 2000 and 2003 was tempered by an increase of 27,000 new hires in pharmaceuticals between 2001 and 2003 (C&EN, Jan. 19, page 11).

What's more, Czarnik says, almost every pharmaceutical company uses combinatorial chemistry or parallel synthesis in some fashion. "These tools are really being integrated into the everyday work of medicinal chemists," he says.

Gerritz believes combinatorial chemistry is a powerful tool in drug discovery that shouldn't be abandoned because it didn't live up to the early hype. "Today employers expect that you are familiar with the concepts of combinatorial chemistry," he says. "It's becoming part of the natural skill set for synthetic chemists in industry."

Currently, the majority of openings in pharmaceuticals are for people with good basic skills and innovative ideas, says Wipf, who is also director of Pittsburgh's Combinatorial Chemistry Center. The recent string of mergers at big pharmaceutical companies generated hiring freezes and employee redistribution that impacted the job market for new graduates, he says. Companies are still hiring, but they are able to be more particular, conducting more thorough background checks and emphasizing softer skills.

According to Li, "A good synthetic chemist can always find a job regardless of the market." Good candidates, he says, display creativity, a sound knowledge of modern synthesis methods, problem-solving skills, and the ability to work in multidisciplinary teams.

"Combinatorial chemistry is a skill in addition to the basic synthetic chemistry skills," Li says. "Chemists can use it to synthesize desired compound libraries, move from hit to lead, and conduct hit follow-up in the most efficient manner."

Hans Maag, vice president of chemistry at Roche, Palo Alto, Calif., believes that the heyday for combinatorial chemistry in big pharmaceutical companies has passed, but that parallel synthesis is and will continue to be a useful tool for medicinal chemistry.

"If a candidate has experience in parallel or solid-phase synthesis, that adds to their skill set, but we don't go out looking for someone who has done massive numbers of reactions," Maag says.

Instead, he primarily looks for excellence in synthetic organic chemistry, plus qualities such as independent thinking and an engagement with science.

According to Czarnik, the actual work of synthesizing large libraries of compounds is most often outsourced to smaller start-up firms. These companies make and characterize libraries of diverse compounds that are then transferred to larger companies for screening.

CRL is one of these small contracting firms. Its parent company, ChemBridge Corp., employs about 350 people worldwide and contracts chemistry services with large drug companies such as Pfizer and Merck. Using the existing ChemBridge portfolio, CRL expands beyond pharmaceuticals to contract with biotech companies and to develop its own unique libraries for drug discovery.

At Roche, Maag says, compound libraries are outsourced in this fashion, but internal researchers also apply parallel synthesis to lead optimization. The work done by employees involves much smaller libraries of compounds generated in traditional labs. "The size of the sets we create has gone down to 20 to 50 compounds," he says. "Even the libraries we outsource have become smaller."

Li also thinks smaller, more diverse libraries are a better path to increasing hits. In the early days of combinatorial chemistry, he says, companies were "playing the numbers game," trying to see who could create the largest library, usually from a single core compound. But having thousands of compounds based on the same scaffold doesn't actually improve the hit rate, he says. "Well-designed small-molecule libraries using drug-relevant building blocks and biologically privileged scaffolds can provide better coverage of biological targets and druglike chemical spaces, enhancing the chances of lead discovery," Li says.

"The term 'high-throughput synthesis' is undergoing constant redefinition," Wipf says. When it was first introduced, high-throughput meant hundreds of thousands of similar compounds, he says. Since then, it has been scaled down to mean hundreds to thousands of distinct compounds. Wipf also notes that diversity-oriented synthesis--using combinatorial methods to create many different compound skeletons that have high potential for appendages--could once again transform what high-throughput means for drugmakers.

THE CONSTANT EVOLUTION of what can still count as a fledgling field means that working in combinatorial chemistry can be as much about personality as about technical skill. According to Gerritz, being adaptable and open to new challenges are key attributes for the combinatorial chemist. While earning a Ph.D. from Massachusetts Institute of Technology, Gerritz worked under the late Satoru Masamune, a man he describes as "fearless" about diving into different areas of chemistry.

"I was working on a traditional total synthesis project, but [Masamune] taught me to not be afraid to leave my comfort zone," he says. After graduating in 1993 during the height of the combinatorial craze, Gerritz took a job with Glaxo, where he plunged into the very nontraditional field of high-throughput synthesis. "In 1995, we had to invent or adapt everything," he says, referring to the dearth of specialized lab equipment in the early days of the field.

Reaction protocols for one-at-a-time compound synthesis also had to be reinvented to apply to parallel synthesis churning out thousands of compounds.

"This field attracts the type of person who is never fully satisfied with the way things are," Gerritz says. "You always look for more efficient ways of handling so many compounds--you want to improve the process."

Now, as a group leader at Bristol-Myers Squibb, Gerritz is applying his hands-on experience to planning synthesis efforts and coordinating library creation with the needs of ongoing discovery projects. Although his synthesis projects these days have been scaled back to a mere 50 to 100 compounds produced in a week, Gerritz is appreciative of the speed and efficiency that combinatorial methods have contributed to drug discovery.

"I would be surprised if the majority of compounds entering clinical trials in 2004 have not been impacted in one way or another by combinatorial chemistry," he says.

Li also got his start in combinatorial chemistry in 1993, when he was a postdoctoral fellow at the University of California, San Francisco. His work building combinatorial libraries in a structure-based protease-inhibitor-design program led to a job in the chemistry division of start-up firm Irori, which later became ChemRx, a part of Discovery Partners International. During this time, Li directed sorting technology, developed combinatorial protocols, and synthesized vast discrete-compound libraries for pharmaceutical and agricultural companies.

Since joining CRL in 2001, Li says he has been actively involved in recruiting chemists for the firm, although recruitment has lessened over time. In general, Li says experience in high-throughput methods is a plus for candidates, and whether that experience comes from academe or industry is immaterial. "No matter where a candidate has had training, they will still need some kind of additional training" when starting work at a new company, he says.

Wipf agrees. "From the academic point of view, we try to make sure our graduates have the skill sets to succeed in industry, but we don't want to cater exclusively to industry," he says. Instead, he explains, universities must give students a broad foundation that will allow them to adapt to new technologies over a 30- or 40-year career. "The main job of academia is to turn students into problem solvers," he says.

Within the pharmaceutical industry, synthetic organic chemists can be hired for many different positions with vastly different requirements, so a traditional background remains paramount. Still, Wipf says, the academic environment must provide students with an infrastructure that matches the modern tools of industry. "The equipment they learn with must be representative of the 21st century," he says. "Our facilities and tools must be updated continuously." For example, today one of the basics all his students learn is the ability to access and critically evaluate the vast array of data made available by databases and the Internet.

For today's chemists interested in working at small combinatorial contract firms like CRL, a background in physical organic chemistry might be a better fit than traditional synthesis, Czarnik says. "It's a little like doing production: You optimize rates, optimize yield, test purity--skills that are useful for making libraries but less essential for drug discovery."

According to Czarnik, working for small combinatorial start-ups offers the benefit of owning stock early in your career, but it can challenge chemists not used to rigorous deadline pressure. "At large firms, chemists are expected to work long hours per week and do their best, but in the end whatever you produce is what you produce," he says.

At smaller firms, contract deadlines must meet clients' expectations, although the general rule is that deadlines can be broken for the sake of quality. "High-throughput synthesis shouldn't become a worker-bee environment," Wipf says. Instead, it is a tool that allows students to do auxiliary work more quickly so that they can focus on the design of creative chemistry. "Ultimately, the most important thing is the compound we synthesize, not the devices we use," he says.

No matter what type of employer seems most appealing, Czarnik says, "it's much more important to get that first job than to wait for your first job at a specific company." Simply doing the job and talking with coworkers helps recent graduates learn what it is like to work for different employers. If the first job doesn't satisfy, good training and industry experience make mobility much easier.