Reflections: Learning How to Be a Scientist Kensal E. van Holde J. Biol. Chem. 2008, 283:4461-4463. doi: 10.1074/jbc.X700002200 originally published online December 19, 2007 Find articles, minireviews, Reflections and Classics on similar topics on the JBC Affinity Sites. Alerts: • When this article is cited • When a correction for this article is posted Click here to choose from all of JBC's e-mail alerts This article cites 12 references, 5 of which can be accessed free at http://www.jbc.org/content/283/8/4461.full.html#ref-list-1 Downloaded from http://www.jbc.org/ by guest on September 22, 2014 Access the most updated version of this article at doi: 10.1074/jbc.X700002200 REFLECTIONS This paper is available online at www.jbc.org THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 283, NO. 8, pp. 4461–4463, February 22, 2008 © 2008 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A. Learning How to Be a Scientist Published, JBC Papers in Press, December 19, 2007, DOI 10.1074/jbc.X700002200 Kensal E. van Holde From the Department of Biochemistry and Biophysics, Oregon State University, Corvallis, Oregon 97331 The unexamined life is not worth living—Socrates I Downloaded from http://www.jbc.org/ by guest on September 22, 2014 ndividual scientists approach and develop their careers in quite different ways. Some, for example, fix upon a major, unsolved problem at an early age and for decades throw all their resources into its resolution. Others may quickly become adept in some particular technique and center their interest on its development and applications. Such work serves a very important function in the overall scientific enterprise; indeed, one can argue that most major advances are stimulated by new techniques. However, I have come to believe that true science lies in seeking answers to questions presented by nature. Upon reflection, I have also come to realize that in fact 3 decades of my own career were dominated by a concentration on methods, not questions. My research changed its nature very abruptly in the early 1970s. I think the story has some interest not so much as a personal narrative, but rather as an example of how one negative experimental result can wholly redirect a research career. The Ultracentrifuge Analytical ultracentrifugation was my consuming interest for nearly 30 years. I became fascinated with the ultracentrifuge during undergraduate and graduate research in the laboratory of J. W. Williams at the University of Wisconsin. Williams had worked with T. Svedberg and was the possessor of two of the very few Svedberg ultracentrifuges still operational in the 1940s. At the time, ultracentrifugation was one of a small number of methods that could provide quantitative data about macromolecules, and I found this combination appealing. I learned all I could about the technique and its applications from the exceptional group of students and scientists in the Williams laboratory at that time. When I left the Williams laboratory, I should normally have had to abandon this interest; there were virtually no places I could have gone to that had such equipment. But a miracle occurred in the form of the Spinco Model E, the first commercially available ultracentrifuge. Not only was it easier to use than the Svedberg monsters, but it could be purchased and installed in any laboratory. In fact, when I returned to Wisconsin for a postdoctoral degree, it developed that my friend and colleague Robert Baldwin had such an instrument available. We proceeded to collaborate on an idea to make sedimentation equilibrium experiments (which formerly took 1 week) doable, with precision, overnight. The resulting paper (1) received a lot of attention, and the work probably got my academic appointment at the University of Illinois in 1957. There, a Model E awaited me. For the next 15 years, my research was centered on the ultracentrifuge and other physical techniques such as light scattering and circular dichroism. It was, I still think, good and useful work, but certainly not at the core of biochemistry or molecular biology. As one prominent (and sarcastic) molecular biologist is said to have put it, “He does elegant experiments on uninteresting problems.” Alas, there was a good deal of truth in the statement. However, in the early 1970s, something happened that wholly changed the orientation and even the character of my research. Chromatin: Mysteries and Excitement Ironically, it was my old addiction to physical techniques that led to a new orientation. I had recently become interested in the possibilities of electric dichroism, a method of seeming potential, FEBRUARY 22, 2008 • VOLUME 283 • NUMBER 8 JOURNAL OF BIOLOGICAL CHEMISTRY 4461 REFLECTIONS: Learning How to Be a Scientist 4462 JOURNAL OF BIOLOGICAL CHEMISTRY fiber.” The two major models, one by Bram and Ris (3) and the other by Pardon and Wilkins (4), differed markedly in fiber dimensions and helical pitch. It occurred to us that our new electric dichroism instrument might shed light on this argument. The two models predicted quite different values for the average orientation angle of the DNA bases with respect to the supercoil axis. The proximity of the Isenberg laboratory made the project feasible for us. So in 1972, we obtained some calf thymus chromatin, and a new postdoctoral student, Randolf Rill, began measurements. To our great surprise, the results could not be fitted to either model; in fact, even the sign of the dichroism was wrong! The data could be reconciled with uniform supercoils only if just part of the structure was in this form, with the rest being more extended. For the first time, I had encountered a seemingly solid result that did not seem to make sense. This negative result put us face to face with a mystery that demanded resolutions. At this point, I recalled a recent paper by Clark and Felsenfeld (5) indicating that part of the DNA in chromatin was much more accessible to nuclease digestion than the remainder. Perhaps this could explain our perplexing results. So Rill began carrying out limited digestions of chromatin using micrococcal nuclease. From these digests, he obtained, by a combination of salt precipitation and resolubilization at low salt, particles (we called them PS particles) that had bizarre and interesting properties. They had a higher protein/DNA ratio compared with whole chromatin, an unusual CD spectrum, and electric dichroism markedly lower compared with either whole chromatin or DNA. Most important, sedimentation in the ultracentrifuge suggested that these particles had a compact structure (6). At this juncture, the very nature of my research program suddenly changed. True, we still relied heavily on physical techniques, but were now using them as tools to attack an intriguing, probably important problem. We alone (at least to our knowledge) had our hands on particles that must bear some relation to chromatin structure. What were these particles, and what did they mean in terms of the overall structure and function? I was fortunate at this point to gain a very talented graduate student, Chintamin Sahasrabuddhe. Sahasrabuddhe further improved the PS particle preparation to where it represented ⬃50% of the DNA in chromatin; it was now clearly not just a minor fraction. He then set about on a thorough physical characterization. Upon purification, the particles were found to sediment as a homogeneous 12 S component. They had a molecular mass (by sedimentation equilibrium) of ⬃180,000 Da, of which ⬃80,000 Da VOLUME 283 • NUMBER 8 • FEBRUARY 22, 2008 Downloaded from http://www.jbc.org/ by guest on September 22, 2014 but little employed. The idea is simple: if you subject a solution of polar macromolecules to an intense electric field, they will partially orient. Then, light polarized parallel or perpendicular to the field may show a difference in absorption (dichroism) depending on the orientation of the absorbing groups with respect to the molecular axis. Thus, you may be able thereby to learn something about internal macromolecular structure in solution. I had an idea for a new and sensitive instrument and a very capable graduate student, Fritz Allen, who was able to design and build it. Again, note that the emphasis was on the method rather than any specific problem. Having the new instrument, we began casting about for things to study with it. Tobacco mosaic virus was a first choice and behaved as it ought to, in accord with its known structure. DNA was tried next and showed pretty much the expected behavior: the bases appeared to be aligned very nearly (but not quite) perpendicular to the helix axis (2). So far, we had shown that the instrument worked beautifully, but had learned very little of scientific import. Because we were really focusing on the technique, this seemed a satisfactory state of affairs at the time. At this point (1970 –1971), I had already been a few years in the Department of Biochemistry and Biophysics at Oregon State University. A colleague in the department, Irvin Isenberg, was working on chromatin. Now what was really known about chromatin in the early 1970s could be described in a few paragraphs: it was known to be a complex of eukaryotic DNA with proteins, mainly five small basic proteins (histones) and smaller amounts of a host of other, largely unidentified proteins (cleverly called “nonhistone proteins”). A few of the histones had been sequenced, and something was known of their covalent modifications. There was the peculiar observation that four of the histones appeared to always be present in equimolar quantities. In 1972, Isenberg’s group was beginning to obtain evidence for specific, noncovalent interactions between pairs of histones, but none of this fitted together into a coherent picture. A good overview of the paucity of real information on chromatin during this period can be obtained by perusing the papers presented on this topic in the 1973 issues of Cold Spring Harbor Symposia on Quantitative Biology. As far as the structure of the chromatin fiber was concerned, there were only a few relatively uninformative electron micrographs, some x-ray fiber scattering data, and lots of speculation. The prevalent concept of the fiber was of a superhelical DNA, coated with histones to form a uniform fiber. There was considerable controversy concerning the details of this hypothetical “nucleohistone REFLECTIONS: Learning How to Be a Scientist England is green Oregon’s blue We see 142 bp Now Crick does too! The years that followed saw a meteoric increase in our understanding of chromatin structure. We contributed to this, both in our laboratory and via collaborations. With John Pardon and his associates, we were able to show, by neutron scattering experiments, that the DNA was wrapped on the outside of the nucleosome (8). A postdoctoral student, Dennis Lohr, showed that the chromatin of a very simple eukaryote, bakers’ yeast, had nucleosomal structure very much like that of Metazoa (9). A graduate student, Kelley Tatchell, demonstrated that nucleosomes reconstituted from DNA and histones had properties FEBRUARY 22, 2008 • VOLUME 283 • NUMBER 8 identical to those of “native” particles (10), an advance that paved the way for many future studies, including high resolution x-ray diffraction. Even today, chromatin continues to surprise and fascinate. J. Widom’s laboratory has shown that the nucleosome is a dynamicentity, capable of partial DNA uncoiling to allow access by proteins (11). Such results are providing new perspectives on chromatin function (see Ref. 12). The chromatin field has exploded, from its modest beginnings in 1970, to constitute a major part of modern molecular biology. In illustration, for 1970 –1971, the keyword “chromatin” in titles or abstracts picks up only six papers in JBC. For 2006 –2007, the count is 246. The field is now in comprehensibly vast and complex. In the 1980s, I spent 8 years attempting to write a comprehensive monographon chromatin (13). I believe the same project would be impossible today. I consider myself very fortunate to have stumbled into this fascinating area when there was so little known and so much yet to learn. It was an experience that taught me, finally, what science is all about. Finally and above all, I consider myself very fortunate to have had the pleasure of working with many wonderful people over the past 60 years. REFERENCES 1. van Holde, K. E., and Baldwin, R. L. (1958) Rapid attainment of sedimentation equilibrium. J. Phys. Chem. 62, 734 –745 2. Ding, D.-W., Rill, R., and van Holde, K. E. (1972) The dichroism of DNA in electric fields. Biopolymers 11, 2109 –2124 3. Bram, S., and Ris, H. (1971) On the structure of nucleohistone. J. Mol. Biol. 55, 325–336 4. Pardon, J., and Wilkins, M. H. F. (1972) A Supercoil model for nucleohistones. J. Mol. Biol. 68, 115–124 5. Clark, R. I., and Felsenfeld, G. (1971) Structure of chromatin. Nat. New Biol. 229, 101–106 6. Rill, R., and van Holde, K. E. (1973) Properties of nuclease-resistant fragments of calf thymus chromatin. J. Biol. Chem. 248, 1080 –1083 7. Sahasrabuddhe, C. G., and van Holde, K. E. (1974) The effect of trypsin on nuclease-resistant chromatin fragments. J. Biol. Chem. 249, 152–155 8. Pardon, J. F., Worcester, D. I., Wooley, J. C., Tatchell, K., van Holde, K. E., and Richards, B. M. (1975) Low-angle neutron scattering from chromatin subunit particles. Nucleic Acids Res. 2, 2163–2175 9. Lohr, D., and van Holde, K. E. (1975) Yeast chromatin subunit structure. Science 188, 165–166 10. Tatchell, K., and van Holde, K. E. (1977) Reconstitution of chromatin core particles. Biochemistry 16, 5295–5303 11. Polach, K. J., and Widom, J. (1995) Mechanism of protein access to specific DNA sequences in chromatin: a dynamic equilibrium model for gene regulation. J. Mol. Biol. 254, 130 –149 12. van Holde, K., and Zlatanova, J. (2006) Scanning chromatin: a new paradigm? J. Biol. Chem. 281, 12197–12200 13. van Holde, K. E. (1989) Chromatin, Springer-Verlag New York Inc., New York Address correspondence to: mrathja@asbmb.org. JOURNAL OF BIOLOGICAL CHEMISTRY 4463 Downloaded from http://www.jbc.org/ by guest on September 22, 2014 (⬃120 bp by measurements of the time) was DNA and the rest protein. The DNA was in one piece, despite the nuclease digestion. Most amazing, combining the sedimentation coefficient and molecular mass data showed that the particles were nearly spherical, with a diameter of ⬃8 nm! This required that the DNA, which would have had a length of ⬃40 nm when extended, must have been tightly folded or coiled in these particles. Finally, mild trypsin digestion removed only a small fraction of the mass, but decreased the sedimentation coefficient drastically. This meant that the folding of the DNA was maintained by the histones. All of this was reported in an unassuming little JBC paper in January 1974 (7). Unfortunately, the title did not call attention to the importance of the results. What we had, of course, were what later came to be called nucleosomes or, more specifically, nucleosomal core particles. They led us into an exhilarating chase to determine their fine structure and functional significance in chromatin. Never before or since has the atmosphere of our laboratory been so charged with excitement and camaraderie. We knew we were in intense competition with other laboratories, some with much greater resources than our own, but our mood was of exhilaration. A simple example follows: at one point, we were in an argument with laboratories in Cambridge as to how many base pairs of DNA were an integral part of the core particle. I learned that a consensus had been reached when, one morning, a postdoctoral student dropped a note with a bit of doggerel on my desk. She had written the following verse.