News and Publications

omewhere deep in primordial time, a tiny creature performed a chemical reaction that changed the world: it somehow read a piece of nucleic acid like RNA or DNA as if it were a computer tape and created a corresponding protein.

Every second, every cell in our bodies reenacts this leap into life, creating robust workhorse proteins out of delicate strands of RNA and DNA. It became known as the "central dogma" of molecular biology: DNA serves as a template for what came to be called "messenger RNA," and this RNA is in turn translated into protein. But ever since this mechanism was proposed in the 1950s, there have been questions. How do cells alive nowadays translate message into protein? How could it first have occurred? And how could knowledge about translation itself be translated into practical drugs and other therapies?

That Schimmel would become a master of such theoretical questions was unlikely, given his original career path. After attending college at Ohio Wesleyan University, Schimmel moved to Boston to attend medical school. But after less than two years at Tufts, Schimmel moved across the river to Cambridge to attend graduate school at MIT. "I wanted to take graduate level courses in physical chemistry, especially quantum mechanics and statistical mechanics. My father was horrified. His son a med school dropout! But I did it anyway." His relationship with MIT would last for most of his professional life so far -- he was a faculty member at MIT for thirty years, most recently the John D. and Catherine T. MacArthur Professor of Biochemistry and Biophysics -- before he came to TSRI.

Given Schimmel's interest in the physical end of chemistry, it is unusual that he wound up in a biology department. But in that sense, medical school left its mark. "I wanted to keep my roots in biology. So that's what I did my graduate degree in, even though my advisor was in chemistry." Schimmel went on to work with Chemistry Nobel Laureate Paul Flory at Stanford for a year before returning to MIT.

STRUCTURE OF BIOLOGICAL MOLECULES

But it is Schimmel's interest in the physical structures of biological molecules that has given his research career its distinct character. For many biologists, the nucleic acid molecules that make up the genetic code are no more than symbols to be analyzed for their content, like words on the page of a book. But for Schimmel, biological molecules -- including both nucleic acids and the enzymes that help make proteins -- are space-filling physical structures. Schimmel has also distinguished himself by working on both nucleic acids and the proteins produced from their templates. "Most people who work on macromolecules usually choose one or the other," observes Charles Cantor, who co-authored a 1980 three-volume textbook with Schimmel in biophysical chemistry that has become known as the authoritative book in its field. Scientists who study nucleic acids and those who study proteins "belong to very different cultures," says Cantor. "It requires pretty good breadth to cover them both."

PROTEIN SYNTHESIS

When Schimmel went to work on the translation problem in the middle 1960s, barely anything was known about it. Put simply, protein synthesis looked like this: One group of enzymes "reads" the genetic "words" coded in the DNA of a gene and creates RNA copies of the gene, the so-called "messenger RNA" or "mRNA." Then the message is sent to the cell's protein factory, the ribosome, where another set of enzymes takes the message and translates it into protein, which is made of subunits called amino acids. These ribosomal enzymes read each "letter" of the copied mRNA "word" and seek out the amino acid that corresponds to it in the surrounding intracellular soup.

But just as important is how the tRNAs know which amino acid they are supposed to recognize. Protein synthesis would collapse into a shambles if the tRNAs brought the wrong amino acids into the ribosome for assembly. Early in his career, Schimmel set out to determine what it was about the structures of the tRNAs (and the enzymes that manipulate them) that allow them to be so specific. It was not enough, he says, to figure out how the system works now. "You have to try and figure out how it became that way. That can give you enormous insights into how a cell really works the way it does and how you might intervene."

Schimmel applied all the techniques available to him through the early 1970s -- this was in the days before cloning and genetic engineering -- and got about as far as anyone did. While taking a sabbatical in Santa Barbara, he taught himself molecular biology in its earliest days in 1974 and 1975. His efforts paid off with a series of discoveries. First, in 1981, he cloned and sequenced one of the enzymes that matches up one tRNA with its specific amino acid, alanine. His paper made the cover of Science.

A SECOND GENETIC CODE

But he also pursued biochemical and genetic studies of tRNAs, trying to figure out what conferred upon them their exquisite specificity for particular amino acids. Finally, in 1988 there was a major advance: together with postdoctoral fellow Ya-Ming Hou, Schimmel found the feature that distinguished the tRNA for alanine. It was a single pair of RNA subunits (called bases), a "G" and a "U", that are found near the place where alanine is attached. The discovery was hailed by some as a "second genetic code," since presumably it is a different feature on each tRNA that allows it to select the correct amino acid out of the pool of twenty available. But Schimmel prefers to call it an "operational RNA code," since the specificity is created not necessarily by the letters ("GU") -- but rather by the overall shape of the tRNA in the vicinity of the crucial pair. "It is the texture of the tRNA that contributes to its matching up with a specific amino acid," says Schimmel.

What was exciting about the operational code is that it was a direct link between protein and RNA, which many scientists believed evolved first, before DNA. Furthermore, Schimmel's later work showed that the entire tRNA molecule was not required to specify a particular protein. This combination, which links proteins to short pieces of RNA, come as close to an understanding of the origin of protein-based life as there is.

"It used to be pejorative even to talk about working on evolution," says Schimmel, "because you are just speculating. But what we have been able to do is design a lot of experiments," which show how things might have actually happened.

Schimmel's contributions have multiplied with regard to tRNA and the enzymes that handle it, in particular an enzyme called "tRNA synthetase." He was among the first to establish the modular design of tRNA synthetase by making and studying truncated forms of it. One of the significant results to emerge from his experiments is that one of the enzyme's modules is responsible for selecting the proper tRNA to match with an amino acid, but there is a second, entirely separate module which double-checks the selection and edits it out if it is incorrect.

BIOTECHNOLOGY REVOLUTION

Schimmel's insight that he would have to go to the gene level to understand biology served him in his non-academic pursuits as well. He realized early on what many luminaries of the biological sciences have now come to appreciate: that the scale of research required to make a major impact on therapeutic problems was beyond what the National Institutes of Health were able to fund. "It wasn't so much the commercial side" that attracted Schimmel, he says. "It was more the idea that we were going to transfer technology out of the laboratory. The opportunities were mushrooming."

Schimmel co-founded his first company, Repligen, in 1981 and followed a few years later with Alkermes, a drug delivery company. Neither was closely related in its goals to Schimmel's lab work. Instead, they drew upon his broad scientific vision and from his realization -- rare among scientists -- that it was wiser to let business goals, rather than scientific curiosity alone, dictate what the scientists in biotech had to do. "[Repligen] was a business from day one, unlike some other biotech companies which offered an excuse to do more science, except with investors' money."

Only with his third company, Cubist Pharmaceuticals, which he co-founded along with then-MIT chemist and current Skaggs Institute director Julius Rebek, Ph.D., did Schimmel get the chance to develop drugs based on what he had learned about tRNA and the enzymes that join tRNA to specific amino acids, enzymes called "tRNA synthetases." These enzymes are different in different species, so they provide an attractive target for drugs that inhibit tRNA synthetase activity in infectious organisms like bacteria but are harmless in their human hosts.

Schimmel's work on tRNA synthetases has led him in a new direction that may be the starting point for a fifth company (he helped one of his former MIT postdoctoral fellows found the fourth one, Amira, in Worcester, Massachusetts, in 1989). "We were able to manipulate synthetase shape and turn on editing activity" in the editing module of the enzyme by providing a prototypical drug-like molecule. "I think we can correct some defects in proteins and RNAs" using a similar method. Traditionally, drugs -- such as those designed by Cubist -- are inhibitors, meant to block some harmful activity in the cell. But drugs that correct protein and RNA defects would result in a gain, not a loss, of function of the targets. "I have a hunch we can exploit this in a big way," says Schimmel.

The move to TSRI fills Schimmel with a sense of excitement and adventure; it is the intellectual atmosphere that Schimmel finds most attractive. TSRI is a hotbed of research on the origin of life and RNA. And Schimmel is now reunited with his MIT colleague Rebek. For Schimmel, it is like starting a second career at the age of 57.