lasker~koshland special achievement in medical science award ESSAY

On the road from classical to modern molecular biology

Tom Maniatis

I graduated from the University of Colorado in a Boulder in 1965, and, while browsing through the university bookstore, I noticed a new book called Molecular Biology of the Gene by James Watson. When I read it, I was captivated by the elegance and power of an exciting new field that combined structure, genetics and biochemistry. Delving more deeply, I was fas- cinated by the classic paper of Francois Jacob and Jacque Monod on the operon theory and the subsequent identification of the Lac and l phage repressors by Walter Gilbert and Mark Ptashne, respectively. These breakthroughs in our understanding of gene regulation in b prokaryotes, which were based on a combi- nation of ingenuity, genetics and biochemis- try, defined classical molecular biology and inspired me to seek a PhD in the new field and, ultimately, to study the mechanisms of gene regulation in eukaryotes. On the basis of my interest in molecular biology, Joseph Daniel, my undergraduate research advisor at Colorado, suggested that I contact Leonard Lerman, who had recently moved from the University of Colorado Medical School to the new Department of Figure 1 Photos of me and my colleagues taken circa 1976–1978. (a) Tom Maniatis, Richard Axel and Molecular Biology at Vanderbilt University. Hal Weintraub. (b) Argiris Efstratiadis and Fotis Kafatos. Courtesy of the Cold Spring Harbor Archives. I took Joe’s advice and ultimately joined Leonard’s lab, where I studied the structure of My postdoctoral project was the characteriza- and mentor. Fred’s quiet brilliance and modesty highly compacted DNA by small-angle X-ray tion of the l repressor/operator protein–DNA affected everyone fortunate to work with him. scattering. Leonard was an excellent men- complex. I found that the repressor binds coop- It was a particularly exciting time in Fred’s lab, tor who encouraged critical thinking, and I eratively to multiple adjacent DNA sequences in as he and Allen Coulson were developing the enjoyed learning a new field. Toward the end the l operators and that mutations in these sites Sanger sequencing method (and back home, at of my thesis research, Mark Ptashne visited weakened DNA binding in vitro and increased Harvard, Allan Maxam and Walter were devel- Vanderbilt, and during his dinner with the expression in vivo. This early use of restriction oping an alternative approach to sequencing). graduate students I had the opportunity to enzymes and DNA binding assays presaged As these methods were still in development, discuss my interests in his work. That conver- my later studies of eukaryotic promoters and Mark and I worked with Bart Barrell and John sation ultimately led to an offer to join Mark’s enhancers. To determine the DNA sequence Donelson using an older, more cumbersome lab at Harvard for postdoctoral studies. of the l operators, Mark trapped Fred Sanger procedure to determine the DNA sequence of a on a ski lift at Keystone and coerced him into 27–base-pair λ operator, which required nearly Tom Maniatis is in the Department of Biochemistry hosting us in his lab at the UK Medical Research a full year to complete. Over the next several & Biophysics, Columbia University, New York, Council laboratory in Cambridge. We were ini- years, Mark and his lab went on to show how New York, USA. tially the unwelcome American duo, but Fred the organization of the operators has a funda- e-mail: [email protected] quickly warmed to us and was a gracious host mental role in the l ‘genetic switch’. I benefited

nature medicine volume 18 | number 10 | october 2012 xxix essay

Globin mRNA Plasmid PMB9 agated in bacteria, and Dale Kaiser and Paul 5ʹ 3ʹ AAAA Berg’s labs had shown that double-stranded dT 15 Reverse DNA molecules could be joined by dA:dT transcriptase joining. David Hogness exploited this method to clone and characterize random fragments of AAAA dT Drosophila genomic DNA, and his laboratory cDNA 15 EcoRI went on to transform Drosophila developmen- Alkali 5ʹ 3ʹ tal biology through the elegant application of genetics and genomic cloning methods. dT Our strategy for cDNA cloning was to 15 λ exonuclease perfect a method for generating full-length DNA polymerase I double-stranded cDNA, which could then be inserted into a plasmid by dA:dT tailing. cDNA Terminal However, our goal of cloning cDNAs was sDNA transferase derailed by the recombinant DNA controversy in Cambridge, Massachussetts, which led to a SI nuclease dA 100 moratorium on recombinant DNA research in dA 100 July 1976. The moratorium was the result of a complex mix of genuine concerns for safety, Anneal Double-stranded post-Vietnam social activism, historic town- globin DNA and-gown antagonisms, city and scientific poli- tics and intellectual dishonesty. The latter led to λ Exonuclease AA an infamous recital of a ‘slow-death scenario’ A T T A TT A T T A T at a city council meeting by an opponent of the A A T T T A T A research in the guise of providing ‘scientific’ advice to the mayor. The end result was that we Terminal transferase were unable to insert our in vitro–synthesized DNA into a bacterial plasmid. dT 100 Fortunately, however, we were rescued dT 100 by Jim Watson, who offered me a place to work at the CSHL, where recombinant DNA Figure 2 Schematic illustrating the synthesis and cloning of double-stranded cDNA from purified rabbit b-globin mRNA using the sequential activities of reverse transcriptase and DNA polymerase I. Reverse research was allowed. CSHL was a beehive transcriptase generates a short hairpin that functions as a primer for second-strand synthesis. S1 nuclease of activity when we arrived. Jim Watson had cleaves the hairpin, the ends are prepared for the terminal transferase reaction by l exonuclease treatment, recruited an exceptional group of young and oligo-dT is added to the ends. The plasmid PM9 vector is cleaved by EcoRI and subjected to oligo-dA investigators, including Joe Sambrook, Phil addition by terminal transferase. The cDNA is annealed to the plasmid DNA through dA:dT base pairing, Sharp, Rich Roberts, Mike Botchan and and the hybrid DNA molecule is introduced into E. coli. Redrawn from ref. 2. many more. Restriction enzymes were being used to create genetic and physical maps of enormously from my time in Mark’s lab. Mark ball shoes, and only rarely could we accom- viral and eukaryotic genomic DNA in Joe was an excellent and intellectually demanding modate Richard’s expensive tastes like dining Sambrook’s lab, and these tools contrib- mentor, and he deeply influenced my think- at a Michelin-rated restaurant. On one occa- uted to the discovery of RNA splicing by ing and writing. He is the best I have known sion, Hal appeared very much out of place in a Rich Roberts and his colleagues, who were at designing clever, incisive experiments that spectacular restaurant when an elegant French also producing dozens of new restriction address deep, mechanistic questions. gentlemen sitting at the next table turned to enzymes (RNA splicing was independently While still a postdoc at Harvard, I attended Hal and said “nice shoes.” The gentleman was discovered in Phil Sharp’s lab at MIT). CSHL the 1976 Cold Spring Harbor Symposium on François Mitterrand, the President of France. also provided an endless source of new infor- chromatin. This meeting would profoundly Hal was the most imaginative and creative mation and interesting people and science influence the future directions of my science biologist I have known. through its meetings program. It was in this and lead to life-long friendships. During the Upon returning to Harvard from the UK, atmosphere that we succeeded in generating annual lobster bake on the Cold Spring Harbor I met Argiris (Arg) Efstratiadis, who was a and thoroughly characterizing the first full- Laboratory (CSHL) beach, I met Richard Axel graduate student (and MD) in Fotis Kafatos’s length cDNA clones of b-globin mRNA1,2. and Hal Weintraub (Fig. 1a). I felt an imme- lab at Harvard (Fig. 1b). Arg and I connected This required the purification of a number diate intellectual and personal connection immediately, and I came to enjoy his humor, of enzymes that were not commercially avail- with them, and we remained close scientific intellectual intensity and madness. An initial able, the observation that reverse transcrip- colleagues and friends until Hal’s untimely brief conversation led to a highly productive tase leaves a short hairpin DNA at the 3ʹ end death at the age of 50 in 1995. We spoke collaboration with Arg and Fotis, the ultimate of the transcript and the establishment of every week on the phone, often late at night goal of which was to clone full-length comple- conditions for cleaving the hairpin by the discussing the latest data, paper or idea. We mentary DNAs (cDNAs). At that time, Herb single-strand–specific S1 nuclease. (Fig. 2). often attended meetings together, including Boyer and Stanley Cohen had shown that a The production and faithful propagation of some in fancy locales overseas. Hal was always eukaryotic ribosomal gene provided by Don full double-stranded DNA clones provided dressed in jeans, t-shirt and high-top basket- Brown could be joined to a plasmid and prop- powerful tools for isolating genes from

xxx volume 18 | number 10 | october 2012 nature medicine essay genomic DNA and for the production of transient transfection method in collabora- DNA technology to a generation of emerging mammalian proteins (see ref. 1 for a more tion with Yasha Gluzman at the CSHL6 and molecular biologists. detailed discussion of cDNA cloning in the to stably introduce cloned genes into cultured In the 1980s, our interest in globin genes mid-1970s). cells with Richard Axel and Michael Wigler7. continued after my return to Harvard, where When the Cambridge moratorium on These recombinant DNA tools not only made I was surrounded by extraordinary colleagues recombinant DNA was further extended in it possible to study eukaryotic gene regulation and benefited enormously from the outstand- October 1976, it became clear that I could not at a level comparable to that of the earlier pro- ing students and postdocs attracted to the establish a competitive research program in karyotic studies but also played an important Department of Biochemistry and Molecular recombinant DNA at Harvard, and adequate part in the emergence of the biotechnology Biology. I taught at Harvard College for nearly space was not available at CSHL. I was con- industry. 30 years and was constantly stimulated and tacted by the Chairman of Biology at Caltech, It was at this time that Ed Fritsch, a postdoc- challenged by brilliant students. During this Robert Sinsheimer (a nationally recognized toral fellow in my lab (who had a key role in time, our interest in globin genes led to the and highly respected opponent of recombi- genomic cloning), and I organized a successful development of an in vitro splicing system9, nant DNA research at the time), who offered CSHL summer course in molecular cloning. A the discovery of the lariat splicing interme- me a faculty position. This prompted an remarkable byproduct of this summer course diate10 (also reported by Phil Sharp). We 8 angry phone call to Sinsheimer from George was the Molecular Cloning Manual , which we also discovered the serine/arginine splicing Wald, one of the leaders of the anti–recombi- wrote with Joe Sambrook. At the time, none of factor SC35 (SRSF2)11 and found that exon nant DNA group in Cambridge. Sinsheimer us (perhaps with the exception of Jim Watson, sequences are required for accurate RNA splic- responded by saying that he did not see a con- who orchestrated the effort) could foresee the ing12. Finally, we identified and characterized flict between his opposition to recombinant impact the cloning manual would have on the splicing enhancers and their role in alternative DNA as “a private citizen” and his respon- training and dissemination of recombinant splicing13. sibility as department chairman to recruit RI RI RI outstanding faculty. Sinsheimer later became > 100 kb High molecular weight supportive of recombinant DNA research Charon 4A eukaryotic DNA and ultimately became an early proponent of DNA Fragment the Human Genome Project, which relied on Anneal cohesive ends recombinant DNA technology. At Caltech I was greeted by a highly collegial faculty, and thus began an intense and exciting 20 kb DNA period of activity that coincided with impor- EcoRI methylase tant advances in understanding gene regula- to block EcoRI sites tion during development in Eric Davidson’s CH CH EcoRI and Norman Davidson’s labs and immu- 3 3 CH CH noglobulin diversity in Lee Hood’s lab, the 3 3 Ends birth of behavioral neurogenetics in Seymour EcoRI methylase Benzer’s lab, and the discovery of homeotic to block EcoRI sites + CH CH genes in Ed Lewis’ lab. Norman was an excep- 3 3 Internal CH CH tional mentor who played an important part 3 3 fragments in the success of my lab with sage advice and EcoRI digestion critical input. Every scientific discussion with Size fractionate CH CH Norman started with a yellow pad on which 3 3 EcoRI cohesive ends 31 kb he would carefully write down the key facts. CH CH 3 3 I also enjoyed my brief interactions with Max Delbruck, but I was shocked by his aggressive questions in seminars. Cohesive-end ligation My lab focused on the development of CH CH 3 CH 3 CH genomic DNA libraries. The promise of using 3 3 CH CH CH CH specific cDNA hybridization probes and iso- 3 3 3 3 lating individual genes by screening genomic In vitro packaging libraries was realized with the isolation and determination of the genomic organization of the rabbit and human globin gene clusters3,4. However, the generation of a fully represen- Amplification tative genomic library with the necessary and screening physical and biological containment condi- tions required considerable effort and the use Figure 3 Schematic illustrating our strategy for the construction of a genomic DNA library. High molecular weight genomic DNA was fragmented, size fractionated to enrich for 20-kb fragments and of highly efficient means of introducing the 3 modified by EcoRI methylase to block the EcoRI sites from cleavage. We then added EcoRI linkers by recombinant phage into bacteria (Fig. 3). blunt-end ligation and generated EcoRI cohesive ends by EcoRI digestion of the linkers. Bacteriophage This advance made it possible to identify RNA l ‘arms’ were generated by EcoRI cleavage, size fractionated and then ligated to genomic DNA to form splicing mutations that cause the inherited long concatemers, a substrate for efficient in vitro packaging of the recombinant DNA into phage heads. blood disease thalassemia5, to establish the The library was then amplified in bacteria. Redrawn from ref. 3.

nature medicine volume 18 | number 10 | october 2012 xxxi Essay

The transfection methods also made it Millennium Pharmaceuticals, Julian Adams, unusually diverse, but the common thread is a possible to investigate mechanisms of globin the chief scientific officer of ProScript, moved deep interest in the fundamental mechanisms and interferon gene regulation, leading to the to Millennium, where he championed the clini- of gene expression and how these mechanisms identification and characterization of tran- cal development of Velcade and, along with relate to the underlying biology. scriptional promoters and enhancers and Ken Anderson of the Dana Farber Cancer the discovery of the combinatorial nature of Institute, succeeded in gaining FDA approval COMPETING FINANCIAL INTERESTS The author declares competing financial interests: cis-regulatory elements and the assembly of of Velcade (bortezomid) for the treatment of details are available at http://www.nature.com/ multicomponent splicing enhancers (enhan- multiple myeloma in 2003. Thus, in less than doifinder/10.1038/nm.2931. 14 ceosomes ). Interest in the mechanism by ten years, a basic research discovery (the role 1. Maniatis, T. in Life Illuminated: Selected Papers from which the enhanceosome is assembled led to of the proteasome in signal transduction) was Cold Spring Harbor, Volume 2 (1972–1994). (eds. Witkowski, J., Gann, A. and Sambrook, J.) 227 (Cold studies of the signaling pathway required for translated into an effective treatment of a fatal Spring Harbor Press, 2008). the activation of interferon gene expression in blood disease18. 2. Maniatis, T., Sim, G.K., Efstratiadis, A. & Kafatos, F.C. response to viral infection. This, in turn, led In 1990, my sister was diagnosed with Amplification and characterization of a b-globin gene synthesized in vitro. Cell 8, 163–182 (1976). to the finding that the processing of the p105 amyotrophic lateral sclerosis (ALS) and died 3. Maniatis, T. et al. The isolation of structural genes from protein precursor of the p50 subunit of nuclear three years later. My exposure to this devastat- libraries of eukaryotic DNA. Cell 15, 687–701 (1978). 15 4. Fritsch, E.F., Lawn, R.M. & Maniatis, T. Molecular clon- factor-kB is ATP dependent , to the first evi- ing neurological disorder had a transforming ing and characterization of the human b-like globin dence for a role of the ubiquitin-proteasome impact in my life. Initially, I chaired an ALS gene cluster. Cell 19, 959–972 (1980). pathway in signal transduction16 and to the Association committee to search for new direc- 5. Treisman, R., Orkin, S.H. & Maniatis, T. Specific transcription and RNA splicing defects in five cloned discovery of the ubiquitin-dependent IkB tions in ALS research, and later my laboratory b-thalassemia genes. Nature 302, 591–596 (1983). kinase with James Chen17. became directly involved in the search for an 6. Mellon, P., Parker, V., Gluzman, Y. & Maniatis, T. Identification of DNA sequences required for tran- In 1980, Mark Ptashne and I cofounded understanding of ALS disease mechanisms. I scription of the human a1-globin gene in a new SV40 Genetics Institute, which quickly established am hopeful that my experience in innate immu- host-vector system. Cell 27, 279–288 (1981). and improved gene cloning and expression nity, RNA biology and gene regulation and my 7. Wold, B. et al. Introduction and expression of a rabbit b-globin gene in mouse fibroblasts. Proc. Natl. Acad. methods directed toward the treatment of experiences in drug development will lead to Sci. USA 76, 5684–5688 (1979). human diseases. We were fortunate to have a treatment or cure for ALS. Three years ago, 8. Maniatis, T., Fritsch, E.F. & Sambrook, J. Molecular Cloning: A Laboratory Manual. (Cold Spring Harbor assembled an incredible group of young sci- I moved to Columbia University to join an Press, 1982). entists working at the forefront of technologies incredible neuroscience program, where I inter- 9. Krainer, A.R., Maniatis, T., Ruskin, B. & Green, M.R. to produce protein drugs using recombinant act closely with Richard Axel, Charles Zuker Normal and mutant human b-globin pre-mRNAs are faithfully and efficiently spliced in vitro. Cell 36, 993– DNA methods. Among the drugs developed and Tom Jessell. In addition to studying ALS, 1005 (1984). and approved by the US Food and Drug my laboratory studies the role of cadherin-like 10. Ruskin, B., Krainer, A.R., Maniatis, T. & Green, M.R. Excision of an intact intron as a novel lariat structure Administration (FDA) at the Genetics Institute genes in the generation and function of single- during pre-mRNA splicing in vitro. Cell 38, 317–331 were the blood clotting factors VIII and IX, cell diversity in the nervous system. My move (1984). erythropoietin and bone morphogenic pro- to Columbia has transformed my personal and 11. Fu, X.-D. & Maniatis, T. Isolation of a complementary DNA that encodes the mammalian splicing factor teins. As chairman of the scientific board of scientific life, and I very much look forward to SC35. Science 256, 535–538 (1992). the Genetics Institute for over 15 years, I was many more years of exciting science in the com- 12. Reed, R. & Maniatis, T. A role for exon sequences and splice-site proximity in splice-site selection. Cell 46, exposed to a broad spectrum of human biology pany of my wife, Rachel, and my close friends 681–690 (1986). and medicine and experienced first hand the and colleagues. 13. Tian, M. & Maniatis, T. A splicing enhancer complex translation of basic research into treatments of I have not mentioned the extraordinary controls alternative splicing of doublesex pre-mRNA. Cell 74, 105–114 (1993). human diseases. graduate students and postdoctoral fellows 14. Thanos, D. & Maniatis, T. Virus induction of human On the basis of our finding that the ubiq- I have been fortunate to work with. Some of IFN-g gene expression requires the assembly of an enhanceosome. Cell 83, 1091–1100 (1995). uitin-proteasome pathway is required for the their names are found in the reference list, but 15. Fan, C.-M. & Maniatis, T. Generation of p50 subunit 16 activation of nuclear factor-κB in 1994, Fred all current and past members of my lab are of NF-kB by processing of p105 through an ATP- Goldberg and I, along with Michael Rosenblatt listed on my website (http://www.maniatislab. dependent pathway. Nature 354, 395–398 (1991). 16. Palombella, V.J., Rando, O.J., Goldberg, A.L. & and Kenneth Rock, founded the biotechnol- columbia.edu/index.php.html). I am deeply Maniatis, T. The ubiquitin/proteasome pathway is ogy company ProScript. The objective was indebted to them for enriching my intellectual required for processing the NFkB1 precursor protein to develop proteasome inhibitors to treat life, for all of their work and accomplishments and the activation of NFkB. Cell 78, 773–785 (1994). 17. Chen, Z.J., Parent, L. & Maniatis, T. Site-specific phos- inflammatory diseases and cancer. Research that made my laboratory a special place and phorylation of IkBa by a novel ubiquitination-dependent at ProScript led to the development of a small- for going on to make enormous contributions protein kinase activity. Cell 84, 853–862 (1996). 18. Sánchez-Serrano I. Success in translational research: molecule inhibitor of the proteasome called to research and teaching. On the surface, the lessons from the development of bortezomib. Nat. Rev. Velcade. When ProScript was acquired by work in my lab over the past 36 years seems Drug Discov. 5, 107–114 (2007).

xxxii volume 18 | number 10 | october 2012 nature medicine