r. Charles Sherr, a Howard Hughes Medical Institute (HHMI) investigator based at the St. Jude Children’s Research Hospital in Memphis, TN, discusses the origins and consequences of his cell-cycle research, which has made him one of the top-cited scientists of the 1990’s.  Dr. Sherr has contributed to 11 papers, which have been cited a total of 8545 times. Among these papers is Dr. Sherr's review, “Mammalian G-1 cyclins” (Cell, 73[6]: 1059-1065, 18 June, 1993), which has been cited 1,190 times.

Discovery depends on a combination of factors, including brains, intuition, timing, luck, dedication, good colleagues, and the ability to view data in an unbiased way. For example, Hitoshi Matsushime, Martine Roussel, and I discovered the D-type cyclins in a screen for so-called "delayed early" growth factor-responsive genes. Using what are now outmoded (but still powerful) techniques, about 25 cDNAs were obtained, and the decision to study a novel "cyclin-like gene" was not made immediately. A new interleukin-1 homologue (later found to be the IL-1 receptor antagonist) was also identified in our screen. At the time, our lab was poised to study hematopoietic cytokines, but had never been involved in cell cycle research, so the IL-1 homologue at first seemed to be the most interesting clone. I then read an article by Marc Kirschner's group in Nature and realized what mitotic cyclins were all about. The guess, then, was that we had found a mammalian G1 cyclin, and when I asked more knowledgeable colleagues, I realized that no such mammalian genes had so far been identified. Returning from an HHMI meeting, I serendipitously found David Beach sitting next to me on the bus and spontaneously asked whether he'd like to test our mammalian cyclin in G1 cyclin-deficient yeast cells. I hadn't realized that David was using these strains in a genetic screen to identify new cyclin cDNAs, and that he had found a human gene that rescued yeast G1 cyclin-deficiency. We exchanged partial sequences, and his human and our mouse cDNAs were virtually identical -- cyclin D1.

In early 1995, Dawn Quelle was trying to clone the cDNA for mouse p16INK4a, but she kept isolating alternative transcripts that had unusual 5' ends. Their sequence suggested that the novel 5' sequences opened up a second reading frame in a shared exon, and that the predicted mRNA did not encode p16 at all. Several labs saw these "p16" transcripts and wrote them off in print. Others assumed either that the alternative transcripts encoded no protein, encoded shortened versions of p16 initiated from internal AUG codons, or simply existed to somehow regulate p16 mRNA splicing. We tested the obvious—namely, that there was an alternative reading frame protein (ARF)—by making antibodies to peptides predicted from the putative ARF primary amino acid sequence and testing to see whether the protein was produced in mammalian cells. We found ARF and showed that it could induce cell cycle arrest. More than half the lab works on ARF today. It is a bona fide tumor suppressor that regulates p53! No one would have predicted this result.

I work at St. Jude Children's Research Hospital (SJ) in Memphis. When Martine Roussel and I moved here from the NIH in 1983 to start a new Department of Tumor Cell Biology, SJ provided the resources to get us started. Memphis wasn't on the map as a basic research center, although SJ had a fine reputation clinically, so we knew that we were taking a risk to come here. However, the problem that we were studying at the time—what is the identity of the FMS oncogene?—was moving along nicely, and by 1985, we demonstrated that FMS encoded the receptor for colony-stimulating factor-1. This was our first "hot" result and allowed us to get an Outstanding Investigator Grant from the NCI with "real money" and 7 years of guaranteed support. We could try riskier projects afterwards, and by 1987, HHMI expressed an interest in funding the lab. When we found the D-type cyclins in early 1990, I called HHMI to ask whether we might move the lab in a completely different direction. Fortunately, Max Cowan's answer was "We hired you—not the project. You should do what you want." With their support, we switched our efforts into the cell cycle field. No one asked us to document why we had changed our Specific Aims. Being at a "new" place like SJ carries unusual pressures but also rare opportunities. During our tenure here, we have helped to define the culture of the organization and have recruited many friends who are excellent scientists. SJ supports part of their work, too.

What are the implications of your work for the future of your field in terms of clinical/therapeutic applications/products?

Many people are trying to develop drugs that interfere with cell cycle progression: chemical CDK inhibitors, some now in clinical trials in cancer, are a good example. Even proteins like ARF, if studied in greater detail, might eventually give way to designer drugs that activate the p53 checkpoint in those cancers (~50%) that retain p53 function. It may also prove that the evolutionary differences in cell cycle regulators between yeasts and man will allow the development of inhibitors that target only the simpler eukaryotes—new antifungal compounds might emerge on this basis.

As a postdoctoral fellow at the NIH, I made a terrible mistake that led to an "exciting" paper going into press in the PNAS. Fortunately, my mentor and I reviewed all the data retrospectively (they seemed too good to be true), and we realized that there could be (and was!) a serious problem with the data. I was able to withdraw the paper, but I had to telephone about 40 senior scientists in the retrovirology field who had received preprints and tell them that I had screwed up. ("Is this a good way to start a career?" one wonders). Most of the people I spoke with expressed their disappointment with me. On the other hand, the late Henry Kaplan told me that he respected me for doing the right thing. There are lessons here.

On the scientific side, there is nothing like real discovery: for me (for my lab!), it was finding FMS/CSF-1 receptor, D-type cyclins, CDK4, CDK inhibitors, and ARF. I'm no administrator, but it also gives me immense pleasure to know that my wife and I played a key role in putting St. Jude Children's Research Hospital on the map as a basic research center. Bringing good colleagues here was half the fun.

Science is so absorbing, I can no longer imagine myself wanting to do anything else. But as an M.D., Ph.D., it would be great to see our discoveries "play out" in a way that would ultimately help patients.

Would you like to leave any other comments about your work or share a personal side of yourself to be included in the piece?

Well, maybe, I would have liked to have been a rock star.

Dr. Charles Sherr
Howard Hughes Medical Institute
St. Jude Children's Research Hospital
Memphis, TN, USA