"This trend may weaken the university training of molecular biologists, and impair the development of protein engineering, which might remain limited to those projects targeted by industry," contended Daniel J. Goldstein of the University of Buenos Aires at the annual meeting of the American Association for the Advancement of Science here in February.
While not all observers see the situation in such stark terms, there is agreement that the trend is significant. "I think it's a problem now," said George A. Jeffrey of the University of Pittsburgh. "Industry seems to be most interested in taking people with postdoctoral experience; they've robbed the postdoc level, if you will."
Keith Watenpaugh, who is starting a crystallography group at the Upjohn Company in Kalamazoo, Mich., said the hiring is triggered by a realization "that structural information will be very important in the rational design of drugs." X-ray crystallographers who specialize in proteins and small biological molecules are valuable to industry both to identify structures and for their ability to envision molecular structures in three-dimensional terms in, for example, the setting up of graphic displays.
A typical industry team consists of three or four Ph.D. crystallographers and as many as eight technicians with bachelor's or master's degrees. Such a group can take from one to two years to puzzle out an entirely new structure, said Watenpaugh, who combines his Upjohn post with an associate professorship at the University of Washington. More routine jobs, such as exploring the ways in which drugs bind to proteins, normally require a few weeks.
The move to sign up such teams has created a bandwagon effect. "In the pharmaceutical industry, if one company decides to invest in protein crystallography, they all do," said Jeffrey. "They can't afford not to, in case someone makes a big hit." William Duax, of the Medical Foundation of Buffalo, estimated that 70 drug companies worldwide are developing identification tests for proteins and small organic molecules—tests that require Ph.D. crystallographers.
Paying Top Dollar
Exacerbating the problem for university departments is the relative paucity of protein X-ray crystallographers. With few job opportunities outside academia until the past five years, fewer scientists entered the field. Duax estimates that the United States has at most 1,200 crystallographers who specialize in biological molecules.
The net effect of the increased demand by the private sector will be fewer scientists to train tomorrow's crystallographers. Duax said universities have been unable to find replacements for retiring faculty in half a dozen cases.
Goldstein expressed concern that corporations that have hired large numbers of crystallographers will set the agenda in the field. "Proprietary rights over protein structures," he warned, "could hinder scientific knowledge and limit market competition."
Others are more optimistic about the long-term situation. "I think it will stabilize itself," said Jeffrey, noting that robotics should enable crystallographers to improve their output significantly.
What the trend away from academia does show, argued Jeffrey, is the need to give more science students a better grounding in crystallography. "I think that departments of chemistry, biochemistry, and biological sciences ought to teach the subject at a more sophisticated level than they do," he said.