Drug hunt

Researchers with a history of problem-solving are highly prized for early-stage drug discovery.ASTRAZENECA

If industries were gardens, the biopharmaceutical sector would be uneven and rocky, with only the most persistent plants thriving in an often unpredictable environment. Last year was particularly rough: market withdrawals of blockbuster drugs, the looming end of patent protection on key products, and constant reminders of nearly empty pipelines. Some companies have announced significant layoffs over the next few years; others look set to outsource some services to lower-cost locations. But in terms of employment, which stages of the drug pipeline stay steady? For the time being, chemical and biological scientists eager to get their hands wet in drug discovery can rest easy.

Recruitment has been slow at biotechnology and pharmaceutical firms for the past couple of years. But companies continue to be on the lookout for talent that can bring new drugs into the pipeline.

"We always recruit year by year, even through the thin years," says Martin Braddock, associate director of discovery bioscience at AstraZeneca, UK. According to the US Bureau of Labor Statistics, employment for chemists and biologists in the US pharmaceutical and medicine manufacturing sector is expected to grow by about 25% until 2014. Jobs still exist at the early stages of the drug pipeline, but many have been redistributed and may be more competitive than ever.

In the earliest stages of pharmaceutical research, scientists are part of medical reconnaissance missions to identify biological targets that could be attacked by a new drug. Biologists play a much larger role than chemists during this stage. They track down targets using literature searches, database mining and state-of-the-art techniques, including antisense and small RNAs to reduce the expression of targeted genes.

The biologists who bring these targets to light are often plucked directly from academia. "We don't necessarily recruit for expert skills. What we really look for are broad-based scientists who have shown a history of problem-solving in a biological system," says Gary O'Neill, senior director of biology at Merck Frosst, in Montreal, Canada.

With the abundance of data now available from genome mapping, work at this stage is heavily supported by those with experience in bioinformatics to sift through the information. Structural chemists may also get involved in providing information about the three-dimensional shape of the target, says Gerd Schnorrenberg, head of chemical research at Boehringer Ingelheim in Biberach, Germany. If X-ray crystallography or NMR spectroscopy aren't possible, computational chemists are called on to produce a three-dimensional model of the target.

During the lead identification stage, drug candidates are found in a number of ways. These days, drug leads rarely come from natural products. Instead they're found within proprietary databases containing hundreds of thousands, or even millions, of compounds, and the scientific and patent literature. These high-throughput screens give a discovery team a hit list of compounds that show interactions with the target. The compounds are then clustered using bioinformatics tools. "These scientists need a broad knowledge of molecular biology, cell biology and automation," says Schnorrenberg.

David Zembower sees assessing drug metabolism as a pipeline bottleneck.L. LAUTH

The essential skills required for lead generation are found in medicinal chemistry. "Making molecules is what it's all about," says Paul Leeson, director of medicinal chemistry at AstraZeneca. The task at this stage is to design molecules with appropriate activity against the target, and selectivity against the counter-targets, to improve the potency of the hits generated during screening. New graduates or postdocs keen to pursue organic synthesis can find an entry point into the lead identification stage, says Leeson. The medicinal chemistry can be taught on the job or through short courses (see Nature 424, 594–596; 200310.1038/nj6948-594a).

Structural chemists and those with computational skills will also find jobs in lead identification. Their ability to determine molecular structures using X-ray crystallography or NMR, or to pick out salient information from structural databases, can help medicinal chemists design new drugs. Although there are fewer positions available, chemists with these skills may be recruited directly out of graduate school.

Smaller biotech firms, which frequently don't have access to vast chemical libraries for high-throughput screening, can jump-start their programmes by focusing on families or subfamilies that are likely to produce good leads, says David Campbell, vice-president of drug discovery at Phenomix in San Diego, California. "You need a good medicinal-chemistry background, and knowledge of what sorts of molecules have worked in the past against the targets and which ones have a higher potential of having drug-like properties," he says.

Biologists who hope to break into drug discovery by generating leads for new drugs need to like working with cell-based systems and to be able to improve an already well-oiled machine.

Moving a lead to the optimization stage requires greater investment. This step refines compounds to see whether they are worth developing further. At this point, the number of leads has been boiled down to a few: a primary lead plus a couple of back-ups. Pharmacologists and scientists with training in pharmacokinetics and metabolism work with medicinal chemists during this stage, testing and tweaking compounds until the optimal molecule is made.

Companies are putting increased emphasis on a lead compound's absorption, distribution, metabolism, excretion and toxicology profile during the lead optimization stage. These data help project leaders identify — and if necessary kill — lacklustre drug candidates earlier in the pipeline.

At deCODE Chemistry in Woodbridge, Illinois, the drug-metabolism team has doubled in size in the past year. "We see that as being a real bottleneck," says David Zembower, vice-president of chemistry. "It's an area where you separate the interesting chemicals from the potential drugs."

Pushing drug candidates through the discovery pipeline requires a range of skills.ASTRAZENECA/AMRI

Chemists would be well-advised to look for jobs at small and medium-sized service-oriented companies (see 'Think small'). Limited capital is forcing small biotech firms to outsource aspects of their discovery operations, meaning that service jobs are available. "The wisest use of their cash is not building lab facilities and hiring more people, it's outsourcing their science to companies such as us to create intellectual property and further value," says Zembower.

And when it comes to intellectual property, patent lawyers tend to be involved at the earliest stages of drug discovery. "It has become more important than it used to be," says Schnorrenberg. "In the past it was mainly dedicated to chemistry patents; now it is also going into targets. The strategy is more aggressive and it's a higher workload."

Patenting a novel target occurs early on, but patenting compounds is done as late as possible so that the patent will last longer. It's risky, as it could leave the way open for a competitor. Lawyers involved in these non-research aspects of drug discovery will almost always have experience in biology or chemistry.

The relationship between start-ups and larger firms is also changing. Rather than scooping up the best leads through mergers and acquisitions, large firms are contracting out an increasing proportion of drug discovery to smaller outfits.

Both deCODE Chemistry and Albany Molecular Research (AMRI) of Albany, New York, say they are planning for some growth in 2006. AMRI plans to hire PhD-level medicinal chemists and bachelor's or master's-level physical chemists at its US facilities this year, says Bruce Sargent, vice-president of discovery research and development. AMRI also recently set up subsidiaries in India and Singapore. Although still in their infancy, these are growing fast to provide resources in discovery and synthesis to the US branch. As such strategic outsourcing gains steam — fuelling job growth in India, China and the former Soviet Union (see Nature 433, 902–903; 200510.1038/nj7028-902a) — North American and European firms may have fewer jobs to offer.

Interaction among scientists is ubiquitous in drug discovery, meaning that strong communication and interpersonal skills are essential. Increasingly, scientists need to have an understanding of a discipline outside their own. Project leaders need to keep the final goal in mind: what does the customer, the clinician, the principal investigator or the patient want in a drug?

Flexibility is another important trait, especially in the biotech industry, says Campbell, where a company can change direction and focus several times. Project leaders need to be able to carry a team through the highs and lows. "Keeping momentum is key for a project leader," says Braddock.

Think Small

Between 1990 and 2004, the American Chemical Society tracked chemistry graduates to see whether they wound up in small or large industry. Over the years, the changing results have proved significant. In 1990, more than 70% of graduates were working for large employers, but by 2004 the distribution was nearly 50:50.

Working for a large pharmaceutical firm has its advantages, as the vast in-house experience provides plenty of learning opportunities. But you may see only a small part of the programme. Smaller companies provide the opportunity to focus on a wide variety of skills, techniques and protocols.

R. STARKMAN

"If you are young, bright, hardworking and ambitious, you can progress rapidly in biotech," says David Campbell (right), now vice-president of drug discovery at Phenomix, who previously worked at Bayer. "If you can step into that vacuum and add value and drive programmes, you can be promoted very rapidly." H.H.