PROFILE PROFILE Profile of Martin Karplus, Michael Levitt, and Arieh Warshel, 2013 Nobel Laureates in Chemistry

Alan R. Fersht1 Medical Research Council Laboratory of Molecular Biology, Cambridge CB2 0QH, United Kingdom

You can read the Nobel Prize Web site for affectionately called “LMB,” thehomeof the best description of the science behind the Sydney Brenner, Francis Crick, John Ken- 2013 Nobel Prize in Chemistry. However, let drew, Aaron Klug, Cesar Milstein, Max me tell you what the beginnings of the sci- Perutz, and Fred Sanger—all now legends ence and its immediate impact were really in molecular biology. LMB was small and like–a personal account from a close by- overcrowded. Theoreticians were crammed stander and indirect collaborator in a neigh- side by side in small offices; experimentalists boring field who grew up scientifically with did not have offices, but sat at the end of the two of the laureates, and knows well the bench for paper work. Nobel laureates occu- third. The new laureates’ papers in the mid- pied tiny cubbyholes, and Sydney Brenner 1970s changed the way we think about pro- and Francis Crick shared a small office. We Michael Levitt. Credit: Linda A. Cicero/Stan- teins and set up a new area of science, which metallofthetimeincorridors,andinthe ford News Service. immediately and radically influenced and in- canteen for coffee, lunch, and tea. There spired me, along with many others. was continual cross-fertilization of ideas and The 1960s and 1970s were a time of great discussions among the different strands of time was rather esoteric and mainly appealed excitement in the world of proteins. Their scientists. Above all, the legendary figures to a small number of X-ray crystallographers. amino acid sequences were by then routinely, were central to this passionate intellectual Warshel then came to work with Levitt but laboriously, being determined. The first activity and drew us, the much younger sci- at LMB. They collaborated first on enzyme tranche of 3D structures from X-ray crystal- entists, in and treated us as equals. They mechanisms, which was my passion. Sepa- lography had been published. And the impli- gave us our independence early on and en- rately and jointly they came up with ideas cationsofthestructureofDNAandits thusiastically encouraged us by being living that were revolutionary from their calcula- transcription and translation, and the folding examples of what scientists should be. I was tions on the mechanism of catalysis by hen of its protein products were being explored. recruited at the age of 26, after my one post- egg-white lysozyme—a landmark study from The center for these activities was the Medi- doctoral year with W. P. Jencks (another David Phillips’ laboratory (4) not from LMB. cal Research Laboratory of Molecular Biology, great scientist and man) to work on the mech- Phillips was another great scientist who anism of enzymes, as there was no mechanistic worked hard to further the careers of young chemist in LMB. I was given a laboratory and scientists. The Phillips paper was remarkable a technician and told to do anything I wanted. because the mechanism of lysozyme was un- Independence at that age is now so rare. known until the crystal structure had been de- Into this milieu came a wunderkind, termined, and the crystallographers proposed Michael Levitt, to do a PhD. By great fortune, a mechanism involving strained interactions John Kendrew, a member of the Scien- in one of the binding sites and electrostatic fi ti c Academic Advisory Committee of the stabilization of an oxocarbenium ion. Weizmann Institute suggested that Levitt go I had come back in 1969 from my post- fi – there rst to work with Shneior Lifson (1914 doctoral year with Bill Jencks, having been 2001), as related by Levitt (1). Kendrew exposed to the “strain theory” of enzyme ca- fi realized that the consistent force eld that talysis, dating back to J. B. S. Haldane (5). Lifson was working on from data on small Many enzymologists believed that enzymes crystalline molecules (2) could be applied to were so sufficiently rigid that they could dis- macromolecules and analyzed their energet- tort a substrate toward the structure of its ics. Levitt teamed up with Arieh Warshel in transition state, which was the driving force Lifson’s laboratory and produced the first real program that could compute noncovalent in- teraction energies within proteins and nucleic Author contributions: A.R.F. wrote the paper. Martin Karplus. Image Copyright Emmanuel acids, which is the basis of current energy The author declares no conflict of interest. Nguyen Ngoc, BnF. refinement methods (3). The work at the 1E-mail: [email protected].

www.pnas.org/cgi/doi/10.1073/pnas.1320569110 PNAS Early Edition | 1of2 Downloaded by guest on October 6, 2021 for enzymatic catalysis. However, Levitt’scal- ended with a section: “It is most important to culations had shown that this was unlikely. have available detailed tests of the dynamic Hidden in a book (6) but fortunately told simulation.” This was where our paths crossed directly to me, he had written: “Small distor- again and I could become part of a confeder- tions of a substrate conformation that cause ation between experimentalists and theoreti- large increases in strain energy cannot be cians. Warshel et al. used an experiment of caused by binding to the enzyme.” Warshel mine on the energetics of the salt bridge in and Levitt (7) produced a paper, “Theoretical chymotrypsin to benchmark electrostatic studies of enzymic reactions: Dielectric, elec- simulations (12). The advent of protein engi- trostatic, and steric stabilization of the car- neering allowed Martin Karplus to calibrate bonium ion in the reaction of lysozyme,” his program CHARMM by simulating the which calculated the interaction energies in energetics of the mutational energies we had the enzyme–substrate complex (7). Arieh measured on the protein barnase. Valerie stressed the importance of electrostatic effects Daggett and Levitt performed the first atom- in enzymatic catalysis, a theme he has subse- istic simulation of protein unfolding in 1992, quently consistently promoted, and drummed using Levitt’s program ENCAD (13). This the general importance into me. coincided with my introduction of ϕ-value Inspired by their work and the early X-ray analysis for the experimental near-atomistic crystallography, I wrote a book that at- Arieh Warshel. analysis of protein folding/unfolding transi- tempted to synthesize radical, new ideas into tion states. So began a long collaboration be- a modern look at enzyme mechanisms (8). tween Daggett and me, combining simulation The Warshel and Levitt key papers and anal- protein structure (10). Levitt has subsequently with experiment, fathered by Levitt. yses were frequently cited throughout the first concentrated on protein folding, using both On a personal note of pride, but with and subsequent editions. I coined the term simplified and atomistic models. a continuing message, I am Master of a Cam- “stress” rather than “strain” because of War- I still remember the frisson of excitement bridge College, Gonville and Caius, whose shel and Levitt’s calculations. Radically new when the first paper on the atomistic molec- members are numbered in the hundreds, as ideas often take a long while to be accepted ular dynamics simulation of a protein, again well as my being at LMB. We elect a few re- (9). Warshel and Levitt’s lysozyme paper took BPTI, was published by Martin Karplus and search fellows each year toward the end of considerable time to catch on—the number colleagues in Nature in 1977 (11). It was the their PhD so they can do independent post- of citations each year averaged only ∼17 for next landmark in computer simulation of doctoral work in the sciences and the human- the first 10 years of publication but has proteins. Karplus was already a young full ities. Four have gone on to win Nobel Prizes, climbed to a remarkable and consistent 120 professor at Harvard, having earned his including Roger Tsien, and now Michael per annum for the past 10 years; over 2,000 PhD at the age of 23, and with a famous Levitt. And we did have Francis Crick as a citations in all, so far. Their influence made equation in NMR spectroscopy named after mature graduate student. The message is, we me write in my book that enzyme mecha- him. Karplus has bestrode the world of the- must allow scientists to be independent when nisms should be solved by a confederation oretical chemistry like a colossus. The list of they are young and not wait until they are in of chemists and theoreticians, based on hisformerstudentsandpostdoctoralscholars their 30s and 40s for their first independent structure: “The chemist seeks to identify looks like a who’s who of theoreticians. And grants. To me, all three of this year’swinners the chemical nature of the intermediates, he is a gourmet chef and professional-stan- deserved their Nobel Prizes just for their by what chemical paths they form and decay, dard photographer with books and exhibi- earlier work, and have taken it to even and the types of catalysis that are involved. tions to his name. greater heights since. The prize is a triumph These results can then be combined with I listed the Levitt and Lifson (3) 1969 and for youth. Martin Karplus earned his PhD at those from X-ray diffraction (and NMR) the McCammon et al. (11) MD simulation as an age when many are just starting it. Michael studies and calculations by theoretical chem- two of the 12 key events in technological Levitt and Arieh Warshel showed what ists to give a complete description of the innovations in studying the folding of pro- young scientists can achieve in a laboratory mechanism.” That is generally accepted teins(9).TheKarplusandcolleaguespaper that fosters talent and independence. now, but it was not universal in 1977, and different subjects defended their own patches. As an experimentalist, I wanted to acquire 1 Levitt M (2001) The birth of computational structural biology. Nat carbonium ion in the reaction of lysozyme. J Mol Biol 103(2): experimental data at high structural resolu- Struct Biol 8(5):392–393. 227–249. tion so the theoreticians could take over. I 2 Lifson S, Warshel A (1968) A consistent force field for calculation 8 Fersht A (1977) Structure and Mechanism in Protein Science knew I had gone as far as I could with experi- on conformations, vibrational spectra and enthalpies of cycloalkanes (W. H. Freeman, New York). and n-alkane molecules. J Chem Phys 49(11):5116–5129. 9 Fersht AR (2008) From the first protein structures to our current ments on catalysis and it required the likes of 3 Levitt M, Lifson S (1969) Refinement of protein conformations knowledge of protein folding: Delights and scepticisms. Nat Rev Mol Warshel who would take it to the next level, using a macromolecular energy minimization procedure. J Mol Biol Cell Biol 9(8):650–654. which he has done in so many areas since the 46(2):269–279. 10 Levitt M, Warshel A (1975) Computer simulation of protein fi 4 Blake CC, et al. (1965) Structure of hen egg-white lysozyme. A folding. Nature 253(5494):694–698. 1970s and has dominated the eld. three-dimensional Fourier synthesis at 2 Angstrom resolution. Nature 11 McCammon JA, Gelin BR, Karplus M (1977) Dynamics of folded In the mid-1970s, promoted by Crick, 206(4986):757–761. proteins. Nature 267(5612):585–590. “ 5 Haldane JBS (1930) Enzymes (Longmans, Green and Co., London). 12 Warshel A, Russell ST, Churg AK (1984) Macroscopic models LMB became interested in the protein-fold- 6 Levitt M (1974) On the nature of the binding of hexa-N-acetyl for studies of electrostatic interactions in proteins: Limita- ing problem.” Levitt and Warshel published glucosamine substrate to lysozyme. Peptides, Polypeptides and tions and applicability. Proc Natl Acad Sci USA 81(15): a molecular dynamics simulation of the fold- Proteins, eds Blout ER, Bovey FA, Goodman M, Lotan N (Wiley, 4785–4789. New York), pp 99–113. 13 Daggett V, Levitt M (1992) A model of the molten globule state ing of the pancreatic trysin inhibitor (BPTI), 7 Warshel A, Levitt M (1976) Theoretical studies of enzymic from molecular dynamics simulations. Proc Natl Acad Sci USA 89(11): based on a simplified representation of reactions: Dielectric, electrostatic and steric stabilization of the 5142–5146.

2of2 | www.pnas.org/cgi/doi/10.1073/pnas.1320569110 Fersht Downloaded by guest on October 6, 2021