Peering into Carnegie

A culture of tough but supportive scrutiny has propelled imaginative research at the nearly 95-year-old embryology department in Baltimore. Will a modern architectural makeover change the science? By Brendan Maher

Delivering the results of some recent experiments before a half-filled auditorium, a graduate student starts off with some nervous chatter: "Can everyone hear me OK?" The entire staff of Carnegie Institution of Washington Department of Embryology is invited to these weekly progress reports. Like many of the other seminars planned this week, the turnout is pretty good - of the 115 at the department, 45-50 researchers, from techs to principal investigators, pick at boxed lunches as the student prepares. And prepare he must.

The researcher had been using morpholinos to disrupt the activity of an acyl CoA long-chain protein, ACSL4, in zebrafish and he barely makes it through his third slide when the interruptions start. Marnie Halpern, who has trained dozens of researchers on the zebrafish system including the student's mentor, Steven Farber, wants clarification on the embryonic stage the student was looking at. Another scientist questions if he knew what the off-target effects of morpholinos might be in this scenario. Could he be sure that there wasn't incomplete knockdown? As he struggles to field the questions and get back on track with his presentation, a preliminary result that seems to contradict his hypothetical model throws the room into a tizzy. "It's confusing," chastens Halpern.

The firing squad treatment of presenters is a celebrated ritual at Carnegie, and it's just the beginning of an institutional program that includes feedback forms and reviews of videotaped seminars. Everyone at the department, through these and other interactions, has influence on the work of everyone else.

Intense skepticism is expected at Carnegie meetings, and while this student's experience isn't necessarily typical, the level of scrutiny and input from the entire community is something that sets the department apart from other institutions. A cornerstone of the environment, say Carnegie employees, is a spirit of both support and critical thinking, fostered by close interaction.

New facilities around the world tout their open lab architectures and designs intended to encourage creative thinking and fruitful collaborations. Whether something so tangible as architecture can actually promote such activity is debatable. Researchers here pride themselves on the interactive atmosphere, and the embryology department in Baltimore has been considered a leader in innovative approaches to developmental biology since long before moving into its $24 million facilities in December 2005.

The Carnegie endowment, valued at $635 million last year, supports research at the embryology department on the Johns Hopkins University campus, but also at five other departments: the Terrestrial Magnetism department and the Geophysical Laboratory both in Washington, DC; the Observatories in Pasadena, Calif., and Las Campanas, Chile; and the Plant Biology and the Global Ecology departments both located on the campus of Stanford University.

The embryology department's success may be due to the collaborative spirit, but also to a track record of putting few restrictions on the investment in people with open minds who are daring in their thinking. Gary Borisy, director of the Marine Biological Laboratory in Woods Hole, Mass., offers his impression: "I believe it's a place which really fosters open-ended, important long-term research ... It seeks to identify important problems and allows members to pursue them freely, with very little administrative burden."

Two feathers adorn the Carnegie cap this past year: 2006 Lasker Award winner, Joseph Gall, since 1983 has had a joint appointment at Carnegie and Johns Hopkins. And 2006 Nobel laureate, Andrew Fire, discovered the mechanisms behind RNAi while a staff associate here, pushing past some criticisms at his own departmental progress reports that he was spending too much time chasing down an artifact.

"The real advantage is the style of the place ... the traditions," says director Allan Spradling. While those traditions include interrupting speakers, custom goes deeper for the nearly 95-year-old department. Johns Hopkins embryologist, Franklin Paine Mall, created the institute in 1914 in part to foster the growth and study of a vast collection of human embryo specimens (see "Foundations,"), but also for "the formation and solution of problems," free from the restriction of immediate application.

Spradling, one of three Howard Hughes Medical Institute (HHMI) investigators here, is well known for his work identifying a stem cell niche in Drosophila. With a mop of grey hair and broad, thick glasses, he seems quiet and reserved but comes to life when asking about unexplored angles of someone's project. Fresh ideas wind him up, even if his views on science tend to drift toward the traditional. He bristles at talk of systems biology, for example, and says he hates the term postgenomics. "I don't believe we're in some new era," he says shortly before leading a tour of the state-of-the-art facilities. While pushing for more quantitative approaches from those at his department, he clings to the idea of biology as an observational science, not quite ready for theory. That physicists and engineers are going to march in and solve the outstanding biological questions with computational modeling goes "against a lot of big grains," he says. It's just not that simple: "If you hate complexity, you're going to hate biology."

Though dubious of popular "big-science" approaches, his certainly is not a conservative approach. Taking risks are de rigueur in what Spradling likes to call "Carnegie-style science," and shortly after taking the torch from Donald Brown in 1994, Spradling, already an accomplished scientist, needed to take an administrative risk.

He tapped Alejandro S￡nchez Alvarado, at the time a postdoc in Brown's lab, to become a staff associate and pursue a project that was unfundable at the time. S￡nchez Alvarado had been working on regeneration and was frustrated by the dearth of good model systems. The staff associate position, a five-year post with guaranteed support from the Institute in the neighborhood of $1 million, gave the then 30-year-old free reign to explore open avenues.

"I was given a lab and resources to hire a technician and do experiments," says S￡nchez Alvarado, now an HHMI investigator at the University of Utah. In hindsight he says he recognizes how unique the opportunity was. "I essentially had no hypothesis. I essentially had no research program to speak of. I essentially just had an idea." That idea was to comb texts looking for an invertebrate model of regeneration that was cheap, easy to work with, and met several developmental criteria. Taking a year to hit the books, S￡nchez Alvarado found an extensive pocket of literature on planarians. In what he calls an historical accident, research on the flatworm largely ceased after Thomas Hunt Morgan abandoned the organism in 1910.

S￡nchez Alvarado says that colleagues including Brown and Spradling were supportive. "We were never told not to do these experiments," he says. "That's not to say we were never criticized, but it was a good kind of criticism."

With help from then postdoc Philip Newmark, they overturned several myths about planarians. For example, neoblasts, the extraordinary pluripotent stem cells responsible for planarian regeneration, were difficult to stain and observe. Many in the field believed that the cells would not take up bromodeoxyuridine (BrdU), a marker for active mitotic cells. The inability likely caused some to abandon the model. S￡nchez Alvarado and Newmark, now a principal investigator at the University of Illinois, Urbana-Champaign, showed that by feeding them food laced with BrdU, the cells would take up the stain and neoblast migration could be traced. ¹ S￡nchez Alvarado says the two have since essentially tripled the flatworm research community with their trainees.

Spradling calls the venture a "perfect example" of the kinds of projects that would have little chance of success elsewhere. A current seemingly unfundable project continues in the basement of the Carnegie. Staff associate Alex Schreiber has been looking not at flat worms but at flatfishes, those odd bottom dwellers such as flounder and sole that start out just like any other fish with eyes on each side of their heads, but through the effects of a single hormone, undergo a craniofacial remodeling program that leaves them the most asymmetric vertebrates known on the planet. (see "Research goes flat,", and look at videos of metamorphosing flatfish)

Inside, Mari Moren is examining some young specimens in a 10-gallon tank. The visiting scholar studies halibut in her native Norway, where the aquaculture industry has been troubled by farm-raised fish whose eyes don't migrate like their counterparts in the wild. Aside from the aesthetic problems, infections common in the "down eyes" make the fish less likely to eat and put on weight. Moren suspects dietetic differences are keeping Norwegian halibut from developing normally.

Although outlandish organisms are part of the Carnegie culture, standard models also play a major role, often in applications that originally seemed unwarranted. Doug Koshland, now an HHMI investigator and staff member at the department, submitted an R01 to the National Institutes of Health in 1988 to develop the budding yeast Saccharomyces cerevisiae into a model for studying chromosomal structure during division. Yeast are smaller than most eukaryotes, however, and their chromosomes appeared to be different. The NIH rejected it, he says, calling the proposal interesting but not backed by enough results. The funding cushion supplied by Carnegie money meant that he had the time and breathing room to produce the results that eventually got the project funded. He's now a leader in the field of higher-order chromosome structure.

That's not to say that the Carnegie is a place where one can get comfortable. Without tenure, staff members such as Koshland are reviewed every five years. Spradling estimates that in the past 25 years, 90% of these reviews resulted in the staff member staying on. Associate positions end after about five years, although a few notable exceptions have been asked to stay. (Fire made the transition from associate to member in 1989). Moreover, says Koshland, while there isn't an emphasis on productivity or visibility, there is an absolute requirement to constantly push the boundaries of their fields. Anyone can take one moment and milk it he says, but to remain continually creative, "I consider that a high-pressure environment."

The labs are smaller here than in many medical schools. Staff member labs have on average seven postdocs and grad students along with a few additional technicians, visitors, rotating students, or undergrads. That small size can mean a nimble group: Gall says his group had no problems when his work on Cajal bodies brought him back to studying Drosophila from the amphibians where he had worked for decades. "We're versatile" he says.

Tiny labs can put lab heads in a precarious position. Koshland says that when three out of five people leave (as happened to him in 2001), it can have a tremendous impact on the direction of the science. He took the cue as impetus to take his science in new directions. This uncertain position fostered his current collaboration with Yixian Zheng, another staff member and HHMI investigator. Together, the two have been looking at genome evolution in hybrids between two yeast species: S. cerevisiae and Saccharomyces bayanus. The chromosomal rearrangements observed in the cross may provide clues to the steps of successful speciation.

Zheng says she's had similar experiences with an unexpected lab exodus, but that's not among the most stressful experiences in her opinion. Even with her second five-year review coming soon, she says she gets more anxious about her progress reports to the entire department. Zheng says the environment is not terribly different from what she experienced at the University of California, San Francisco, where she was a postdoc with Bruce Alberts. The only difference, she says is that the size of UCSF meant that many people were studying the same thing. At Carnegie, where competition in the same field is rare, communications are more open. "The openness is really the key."

Openness seems to be part of the culture. An artichoke-shaped chandelier accents a large atrium with a two-story window overlooking a steep wooded slope down to a shallow creek. The atrium was designed, according to architect notes, as a general meeting area for scientists moving between the three wings that jut out from it. It's a common architectural notion, but with some lab members spread across wings in so-called postdoc clusters, the traffic through the embryology department's atrium is steady, and people constantly stop each other in the halls. Staff associates, postdocs, even a rotation student comment that even legends like Gall have an open door policy. Add that to the dozens of departmental, inter-, and intralaboratory meetings that go on each week, and schedules fill up.

One morning in early December, a department administrator scoured the building for Zheng. She had missed an important meeting, she later confessed, because she was in her car in the parking lot, working on a paper. Halpern could empathize, because she's done the same.

Halpern joined the embryology department 12 years ago. Friendly if not a bit frantic at times, she's a pioneer in zebrafish genetics. She says that what sets people apart at Carnegie is that they're "still motivated by the results rather than the prizes, the papers." It's a liberating experience, she says. "I feel like a postdoc in my mind." The freedom she gives to her postdocs, however, hasn't always been the best for her own research interests.

Farber, now a staff member with his own lab in an adjacent wing, became a postdoc in Halpern's lab in 1995. He developed a system whereby fluorescently quenched phospholipids given to early embryos light up in the presence of lipase activity.

In a bleary-eyed 2:00 a.m. experiment, Farber fed the lipids to a plate of five-day embryos - far older than the specimens they had been using - and found bright fluorescence in the gall bladder. "I thought, hey, this might be a great indicator for cholesterol synthesis," Farber says. Halpern, whose research deals primarily with neural development, was in the lab at the time. Hearing his excitement about the gall bladder, Halpern worried that Farber might be moving on to "yet another research directive." She says, "I was frustrated. I wanted him to fall on a project, but being at the Carnegie, I just supported him to do it."

Halpern and Farber pursued it together, publishing the findings with University of Pennsylvania researcher Michael Pack in Science in 2001, ² "That would have been impossible elsewhere," says Farber.

Farber had moved on to Thomas Jefferson University in Philadelphia. But, when he heard there was a likely opening at the embryology department in 2003, he jumped at the chance to return, ultimately taking the place of Fire who moved to Stanford that November. Farber jokes that his office was "Andy's design." (see "Making Outreach Work)"

Desk-design is hardly the only Fire vestige, though. When Fire was leaving in 2003, postdoc Judith Yanowitz was faced with the prospect of entering the job market or applying for a staff associate position. She wasn't sure she wanted to spend another five years at Carnegie, so she consulted her thesis advisor at Princeton, Paul Schedl, who she says encouraged her to go for it: "He said that if he could he would take it. Five years of pure research without all these other idiotic obligations like teaching and committees sounded like the perfect job." Yanowitz had to go through the normal application process including interviewing with faculty and giving a talk in front of that famous Carnegie firing line.

Now three years as a staff associate, she studies aging and reproduction in Caenorhabditis elegans, and is penning a K-award, a career-development grant from the NIH, to develop the system as a model for age-related nondisjunction. Although technically considered junior faculty, Yanowitz says she feels more like a postdoc, which isn't such a bad thing. "People always say, 'At Carnegie, it's postdoc heaven,'" she says. Although she doesn't detect much difference in the interactions here and at Princeton, she says that at Carnegie the faculty are fewer, and you can't avoid running into very smart people who are very interested in what you're doing.

The rigor of that constant feedback may seem intimidating at times. The student who gave his progress report to a deluge of sharp, pointed questions asked that his name be withheld because he's now exploring other career options. His struggles with the project, made apparent in the talk, culminated in his decision to leave graduate school.

That student is in the minority, though: Maybe five percent of grad students who start at Carnegie don't finish, according to Spradling. And the rigor of the feedback is part of a process that nurtures good science. Support permeates the Carnegie culture, and the student says he left on amicable terms. "I think people do want everyone here to succeed. I don't think anything was ever done in the spirit of, 'You don't belong here or you shouldn't be doing this.'"

Remaining on the cutting edge with a slew of scientists whose research interests don't always appear relevant to human health can be a challenge. Spradling quips that the department can coast a bit since Fire won the Nobel; RNAi gives "some bang for your buck for a little while." But he realizes he has his work cut out for him in proving to the world the value of descriptive biology, especially projects that go against the grain. Spradling says his department is following history: Small science and creative, agile scientists following their results rather than the funding will continue to contribute the biggest findings. It can be stressful, however, following something not in favor at the moment. The best science, he says, "never looks like a guarantee. Sometimes it doesn't even look good."