The Early Years-Across the Bench from Bruce (1963-1966)

Home , Leonidas Zervas, Max Bergmann, Stanford Moore, Vincent du Vigneaud

The Early Years—Across the Bench From Bruce (1963–1966) The Early Years—Across the Bench From Bruce (1963–1966)

Garland R. Marshall1,2 1Department of Biochemistry and Molecular Biophysics, Center for Computational Biology, Washington University, St. Louis, MO 63110 2Department of Biomedical Engineering, Center for Computational Biology, Washington University, St. Louis, MO 63110

Received 14 July 2007; revised 20 September 2007; accepted 5 October 2007 Published online 16 October 2007 in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/bip.20870

a Nobel Laureate, Chairman of the Department of Biology at ABSTRACT: Caltech and a member of the National Academy of Science, and was still willing to recommend me for graduate studies This personal reflection on the author’s experience as at Rockefeller. Bruce Merrifield’s first graduate student has been I was convinced at the time that I was chosen to study adapted from a talk given at the Merrifield Memorial neurophysiology, having failed miserably to isolate the acetyl- Symposium at the Rockefeller University on November choline receptor from denervated rabbit muscle as an undergraduate at Caltech. The outstanding neurophysiologists at 13, 2006. # 2007 Wiley Periodicals, Inc. Biopolymers Rockefeller including H. Keffer Hartline, Nobel Laureate, (Pept Sci) 90: 190–199, 2008. were more interested, however, in the wiring diagrams of the Keywords: solid phase synthesis; Merrifield; DNA synthe- eye of the horseshoe crab2 than in how a small molecule sis; combinatorial chemistry could trigger the action potential. Thus, my first laboratory experience at Rockefeller was with Prof. Henry Kunkel, a This article was originally published online as an accepted prominent immunologist.3 Prof. Kunkel suggested that I try preprint. The ‘‘Published Online’’ date corresponds to the preprint to develop a radioimmunoassay for angiotensin II, an eight- version. You can request a copy of the preprint by emailing the residue vasoactive peptide hormone, based on the seminal Biopolymers editorial office at [email protected] work of Berson and Yalow who had devised a radioimmunoassay for insulin4 (for which Yalow shared a Nobel Prize). Karl Landsteiner, the prominent immunochemist who had explored antibodies against a variety of small-molecule hapt- INTRODUCTION ens,5 was still a legend at Rockefeller. The bottleneck for the he Rockefeller Institute for Medical Research1 was a experiments planned was a source of angiotensin II and a formidable institution in 1962, the year I was strategy for selectively attaching it to a protein carrier before accepted for admission with an undergraduate degree injecting rabbits. Henry suggested in early 1963 that I go see T in biology from the California Institute for Technol- Bruce Merrifield (Figure 1), a young faculty member in the ogy. In those days, admission effectively required a Woolley group, whose reputation indicated some experience letter of recommendation from a Nobel Laureate, from a in peptide/protein chemistry, a Rockefeller dominated area chairman of a department at a ‘‘recognized’’ university, or of research (Figure 2). Little did I know this suggestion from a member of the National Academy of Science. Rockef- would change the direction of my thesis research as well as eller President Detlev Bronk controlled the admission pro- dominate the rest of my career. cess, and candidates were personally interviewed before acceptance. I was fortunate in that Prof. George W. Beadle was THE WOOLLEY GROUP First, I need to explain about the hierarchical system in place

Correspondence to: Garland R. Marshall, Department of Biochemistry and Molecu- at Rockefeller at the time. Each research group was organized lar Biophysics, Center for Computational Biology, Washington University, 700 S. under the leadership of a Professor in a pyramidal/Prussian Euclid Ave., St. Louis, MO 63110, USA. e-mail: [email protected] or [email protected] manner. Prof. D. Wayne Woolley was a founder of the ﬁeld of 6,7 VC 2007 Wiley Periodicals, Inc. antimetabolites and had two junior faculty working in his

190 PeptideScience Volume 90 / Number 3 The Early Years—Across the Bench From Bruce 191

FIGURE 1 Bruce Merrifield with new manual solid phase shaker as seen from across the lab bench in 1963. This picture appears on page 115 of Merrifield’s autobiography and on the cover of this special issue of Biopolymers: Peptide Science in honor of Professor Bruce Merrifield. group as well as postdocs and technicians. He was a powerful as there were few formal courses at Rockefeller in those days intellect as well as blind at 26 from diabetic retinopathy (only 12 graduate students were admitted in 1962). The title shortly after receiving his Ph.D. in biochemistry from the of the Institute was changed during my tenure to ‘‘The Rock- University of Wisconsin and moving to Rockefeller. Prof. efeller University.’’ It was a joke among the students that Woolley not only worked daily in the lab doing organic President Detlev Bronk sent an announcement to all the out- chemistry with assistance from his technician, but also had a standing universities in the world welcoming them to the ‘‘photographic’’ memory of the chemical literature. His wife same status as Rockefeller. and others spent hours each day reading the latest journals to Prof. Woolley to keep him up to date. I remember with amazement, when during a discussion of the mechanism of SOLID PHASE PEPTIDE SYNTHESIS cleavage of a benzyl ester by hydrogen bromide, Prof. Wool- Bruce had gotten his Ph.D. developing bacterial bioassays for ley cited the volume, page number, and location on the page amino acids8 in the lab of Prof. Max Dunn at UCLA and was of an experimental graph in an old JACS issue that was an extending this work on other growth factors,9 including pep- essential part of the evidence. I was terrified that he and tides under Dr. Woolley’s direction. But the synthetic tedium others in the lab would realize that my synthetic chemical and repetitiveness of solution peptide synthesis was problem- background was inadequate at best. Bruce had been recruited atic. With his characteristic insight, Bruce had decided that from UCLA, and John M. Stewart from the University of Illi- there had to be a better way to synthesize peptides and, ulti- nois and both had progressed to faculty appointments in mately, proteins. Thus, the famous entry ‘‘A New Approach Woolley’s group. John’s Ph.D. was in synthetic organic chem- to the Continuous, Stepwise Synthesis of Peptides’’ in his istry under the legendary Roger Adams, and I and a few other notebook of May 26, 1959 (Figure 3). Questions regarding students got John to tutor us in advanced organic chemistry the four years of research effort that Bruce had expended

Biopolymers (Peptide Science) 192 Marshall

technician, had already started to explore benzyl ester linkage to the polymer and t-butyloxycarbonyl (BOC) amino protection (Figure 7) recently disclosed by Carpino.11 Angela related later that Bruce had jokingly suggested the use of a hammer to remove the peptide from the nitrobenzyl polymer due to possible overnitration of the polystyrene and potential production of analogs of TNT. Even though my interest in synthetic chemistry was nascent, I immediately realized the power of Bruce’s invention and never completed the radioimmunoassay project in the Kunkel lab. Having a graduate student in his lab allowed Bruce the luxury of having a new shaker designed and built (shown proudly with Bruce in Figure 1). His old reciprocal-arm shaker, noisy and slinging droplets of oil, was relocated across the bench from Bruce, and I began my life as a peptide chemist apprenticed to a future Nobel Laureate. As Bruce had no teaching responsibilities per se and Rockefeller scientists did not apply for grants at that time, I had the daily privilege of working across the lab bench from Bruce and Angela, while trying hard not to expose my ig- norance of science in general and chemistry in particular. Being treated as an intellectual equal from day one was a great incen- tive for me to learn as quickly as possible. Once Bruce’s first FIGURE 2 The Rockefeller heritage in peptide/protein chemistry note on SPPS appeared in 1963, a steady stream of prominent was very strong. Bergmann, who was trained in Emil Fisher’s lab scientists visited the lab. Almost each one mentioned at some where the first peptide bond was synthesized, and Zervas were refu- point in the visit how he had thought of using a filterable poly- gees from Germany. Bergmann trained most of the next generation meric support as a protecting group (the essence of solid phase of protein chemists [Stanford Moore, William Stein, Joseph Fruton synthesis), but, of course, none had spent the years exploring (Yale), Klaus Hoffman (Pittsburg), etc.] Lyman Craig brought ex- pertise in chemistry of unusual natural products including peptides alternative approaches until a practical solution was found. I as well as separation science to the Institute. (Copied from a picture particularly remember the visit of Sir Robert Robinson, the in the hallway of the Merrifield lab). British Nobel Laureate in Chemistry, who pulled me aside and told me how lucky I was to be Bruce’s student. He said that he exploring numerous alternative supports, linkages, amino was going to nominate Bruce for the Nobel Prize in Chemistry. protection, etc. was never discussed to any extent during my If I remember correctly that was in 1964; I listened every Octo- days across the bench. One could not buy functionalized ber for 20 years to learn that R. Bruce Merrifield had finally polymeric supports per se, for example, but had to modify been recognized by the Nobel Committee. Obviously, the delay the polystyrene-divinylbenzene beads (Figure 4) to allow for was due in some part to the reluctance of the entrenched solu- covalent attachment of the first amino acid. Only when Bruce tion chemistry community to rebuke solid phase chemistry as published his autobiography10 was this frustrating period nontraditional, i.e., no isolation and characterization of chemi- illuminated in any detail. cal intermediates. Bruce once showed me the scathing review When I went to Bruce to discuss the synthesis of angioten- his first article in JACS had received in which the reviewer char- sin II in early 1963, he described the current state (Figure 5) acterized SPPS as a ‘‘travesty,... not chemistry at all, a concept of solid phase synthesis using a nitrobenzyl ester linkage to which should be suppressed by the community.’’ I always the polymer and carbobenzoxy (Cbz, Z) amino protection wanted a copy of that review as an inspiration to persevere in about to be published. It was clear that the original proc- spite of conservative objections and prejudice. edure in which the tetrapeptide LeuÀÀAlaÀÀGlyÀÀVal was prepared has deficiencies including racemization of the C- terminal amino acid during saponification from the support ROCKEFELLER, A BASTION OF PEPTIDE (Figure 6) based on its chromatographic profile. Because of CHEMISTRY the harsh deprotection and cleavage from the polymeric sup- The Rockefeller heritage in peptide/protein chemistry was port involved, Bruce and Angela Coregliano Murphy, his very strong, particularly with regard to synthetic peptide

Biopolymers (Peptide Science) The Early Years—Across the Bench From Bruce 193

FIGURE 3 Bruce Merrifield reviewing his notebook entry of 5/26/59 ‘‘A New Approach to the Continuous, Stepwise Synthesis of Peptides’’ (Courtesy of Prof. John M. Stewart). chemistry (Figure 2) and bovine pancreatic ribonuclease A peptide hormone, angiotensin II, by myself, a naive graduate (EC 3.1.27.5, RNase). Prof. Max Bergmann, a student of Emil student, with a totally novel approach. Although I had the Fisher (credited with the synthesis of the first peptide bond) illusion that I was working independently, Bruce obviously and refugee from Nazi Germany, had arrived at Rockefeller was a dominant influence by his logical approach to all ex- in 1933 along with his colleague Leonidas Zervas. They had perimental questions, discussed regularly across the bench- recently invented the carbobenzoxy (Cbz, Z) group,12 an eas- top. Bruce was a superb experimentalist who designed ily removable amino protecting group for peptide chemistry experiments that unambiguously focused on the question to in 1932, and the rest was history. Prof. Vincent du Vigneaud, be resolved. next door at Cornell Medical School, and his large group of Several research groups at Rockefeller had a strong practi- synthetic chemists used the Z group repetitively in his syn- cal influence on our work. Professors Stanford Moore and thesis of oxytocin13 and vasopressin, the nonapeptide neuro- William Stein had trained with Bergmann and focused their hypohyseal hormones, for which he received the Nobel prize. research on the enzyme, RNase A. Ribonuclease A was dis- Here I was, just 10 years later, trying to synthesize an octa- covered and isolated at Rockefeller by Dubos and Thompson

FIGURE 4 Modiﬁcation of polystyrene-divinylbenze polymer by Friedel-Craft alkylation with chloromethylmethyl ether (known now to be a carcinogen) to provide a site for benzyl ester forma- tion with N-protected C-terminal amino acid. Copied from original hand-colored slide used in author’s thesis defense in 1966.

Biopolymers (Peptide Science) 194 Marshall

Rockefeller, where graduate students could mingle with these legends of protein chemistry and enzymology. I can remember several lunches where I and other graduate students got to discuss the raging dispute in the 1930s over enzymes being proteins or nucleic acids with Prof. Kunitz. Now it is clear that both can do catalysis. Moore and Stein developed the methodology to determine the amino acid sequence for RNase A, an accomplishment for which they shared the No- bel Prize.16 In the course of their research, they developed automated amino acid analysis,17 later commercialized by Beckmann instruments. Bruce, of course, had followed the work on RNase A with great interest and later turned to its synthesis with his postdoc Gutte and Merrifield18 as a proof of principle for SPPS after I had left the lab. RNase A is, perhaps, the most thoroughly investigated enzyme of the 20th century,19 and offers many opportunities for validation of methodology, etc., because of the plethora of experimental data. The Merrifield lab contributed significantly to RNase A studies, for example, developing a three-component system, by combining equimolar amounts of RNase A fragments 1– 20, 21–118, and 111–124, that retained 30% of the enzymatic activity of intact RNase A.10 One area of current interest in FIGURE 5 The first paper (Merrifield, JACS 85:2149, 1963) used carbobenzoxy (Cbz, Z) amino acids with HBr/HOAc deprotection after each elongation step. This required a more stable link to the polymer that the benzyl ester so the polystyrene was nitrated. Cleav- age from the nitrobenzyl polymer was achieved by saponification. in 193814 and crystallized by Kunitz in 1940.15 Professors Dubos, Kunitz, Moore, and Stein were still daily participants in lunches at long tables held in the Faculty Lunchroom at

FIGURE 6 Ion exchange purification of cleaved tetrapeptide FIGURE 7 Merrifield SPPS using the Boc group for amino pro- H2NÀÀLeuÀÀAlaÀÀGlyÀÀValÀÀCOOH from first JACS paper using tection and a benzyl ester linkage to the polymeric support. This saponification to remove peptide from nitrobenzyl-linked polymeric was the basic scheme used for synthesis of many biologically active support. Besides many examples of deletion sequences where one or peptides–bradykinin, angiotensin II, oxytocin, insulin, etc. Signifi- more amino acid is omitted form the desired peptide sequence, rac- cant improvements were the introduction of TFA for deprotection emization of the C-terminal valine residue was also observed, prob- of Boc groups and HF for cleavage of polymeric benzyl ester linkage ably due to saponification. to support. (Copied from slide used at author’s thesis defense).

Biopolymers (Peptide Science) The Early Years—Across the Bench From Bruce 195 my own research is the role of protein dynamics on enzyme Gisin24,25 from Prof. Brenner’s lab in Basel soon joined the kinetics and catalysis using RNase A as a model system.20 Wooley/Merrifield group as postdocs as well as another grad- The Woolley lab did not have an amino acid analyzer, and uate student, Arnold Marglin, M.D. Arnie successfully syn- we had to do analysis of our synthetic peptides the hard way. thesized insulin by SPPS for his Ph.D. thesis,26 and we shared Acid hydrolysis followed by an ion-exchange column separa- a lab in the mid-sixties; he is the only physician I have known tion lasting 24 h, addition of ninhydrin, heating and cooling, who would carefully remove JAMA from his desk with a pair followed by reading each individual tube in a Klett colorime- of forceps and deposit it in the round file (trashcan). ter. The values were plotted and graphed, the peaks then cut Despite the immediate success of this methodology in out and weighed. It was accurate but inefficient, and a bottle- generating many synthetic biologically active peptides, a neck for my thesis research. I appealed to the graduate office number of troublesome, sometimes unpredictable, side reac- and was able to purchase a Beckman amino acid analyzer for tions plagued early SPPS. The original chromatogram of Merrifield’s research. Quite a coup for a graduate student, LeuÀÀAlaÀÀGlyÀÀVal prepared by SPPS published in JACS but the limited number of graduate students and our special illustrated a significant problem (Figure 6). If a protected relationship with President Bronk was something I could amino acid was not completely deprotected and neutralized capitalize to the lab’s advantage. before coupling the next residue, or if the coupling reaction Another strong influence at Rockefeller was Prof. Lyman was not forced to completion by excess reagents and/or mul- Craig, the natural product chemist and inventor of counter- tiple couplings, then ‘‘deletion peptides’’ missing one or more current distribution (CCD), basically an interconnected se- residues would be generated. SPPS depended on excess ries of separation funnels each of which mimicked a theoreti- reagents, multiple couplings, etc. to drive each reaction to cal plate in chromatography.21 This apparatus was routinely completion, or as near to 100% as possible. Otherwise, the used to purify peptides in the Woolley lab, following separa- multiple, stepwise elongation procedures would produce a tion by colorimetric reactions. A CCD machine was my first horrendous mixture. The manifestation of the heterogeneity major purchase when I started my own lab in St. Louis in of the solid phase support as well as the dynamic change in 1966. Thank goodness for labor-saving HPLCs that became physical properties of the polymer as the peptide chain was available a decade or so later. elongated was a sudden loss of reactivity of the growing chain that occurred occasionally and lead to truncated peptides. Often this lack of reactivity was reversed with reappear- THE MERRIFIELD METHOD ance of near quantitative chain elongation after one or more The use of polystyrene crosslinked with 1 or 2% divinylben- amino acids were added. The growing chains that ‘‘disap- zene (DVB) and then chloromethylated (Figure 4) with peared’’ and then reappeared would be missing the interven- chloromethyl methyl ether (now a known carcinogen) ing amino acid sequence—thus the designation ‘‘deletion became the routine polymeric support in the Merrifield lab. sequence.’’ Some European peptide chemists consistently Attachment of a Boc-amino acid by a benzyl ester linkage, referred to such peptides as ‘‘failure sequences,’’ a term that followed by Boc removal and chain elongation was the stand- never failed to cause Bruce some angst. ard methodology (Figure 7) for the synthesis of many biolog- Bruce spent extensive effort within his research group ically active peptides and their analogs. Making side-chain exploring this issue. One instructive example was his syn- protected Boc amino acids with the azide generated from t- thesis of angiotensinylbradykinin in which bradykinin was butylcarbazate as described by Carpino,11 however, had its prepared by SPPS, and then the octapeptide sequence of drawbacks. A very potent vasodilator was generated as a side angiotensin continued to give a 17-residue peptide. Since product that left me unconscious in the cold room one day. I both bradykinin and angiotensin analogs were routinely became highly sensitized to the side product and could detect prepared in the lab without apparent ‘‘deletion sequences,’’ whether the reaction was being performed on the fourth the apparent loss of reactivity for coupling residue 12, floor upon entering the building. Fortunately, SPPS became BocÀÀHis(Bzl)ÀÀOH, was anomalous (Figure 8). By simply sufficiently popular that both the polymeric support and reducing the cross-linking to 1% DVB and changing the sol- Boc-protected amino acids became commercially available. vent for Boc deprotection, this reproducible change in acces- Prof. John M. Stewart dramatically increased the productiv- sibility for extension of the growing peptide chain was elimi- ity of his search for bradykinin antagonists using SPPS,22 and nated. The SPPS protocol from the early 1960’s has a shrink/ access to the CCD apparatus for purification of cleaved prod- swell cycle included to increase washing efficiency by using a ucts became a bottleneck for all. Drs. Maurice Manning23 solvent such as ethanol that shrank the polymeric polystyrene from Prof. du Vigneaud’s group at Cornell and Balz support.

Biopolymers (Peptide Science) 196 Marshall

protein (ACP), isolated and characterized in the lab of our collaborator Prof. P. Roy Vagelos in St. Louis. The growing peptide chain of C-terminal 67–74 segment of ACP would reproducibly change to limited accessibility followed by a full return of reactive sites for coupling. Monitoring the synthesis of ACP 67–74 by 36Cl showed that chloride ion was seques- tered in growing polymer matrix, and could then be liberated after a shrink/swell wash.29 This implied that some portion of polymer’s amine hydrochloride groups became inaccessi- ble to base during chain elongation. Many improvements in protocols for SPPS have largely eliminated this problem, and quantitative monitoring of FIGURE 8 Synthesis of angiotensinylbradykinin showing dra- deprotection and coupling efficiencies are more routinely matic effects of the reagent used for Boc removal (HCl-acetic acid vs used to detect such problems should they occur. One must HCl-dioxane) and cross-linking (1 vs 2% divinylbenzene) on cou- acknowledge the many improvements and variations in SPPS pling efficiency as the growing peptide chain was extended. that were the results of the efforts of talented students and postdocs in Bruce’s lab and elsewhere in the peptide chemis- It became increasingly evident that the polymeric support try community. was not inert in SPPS because of the lack of homogeneity of the reactive sites within the polymer. While the DVB cross- linking of the styrene polymer used was 1 or 2% on average, AUTOMATION OF SPPS there was still a distribution of regions within each polymeric On the basis of the successful automation of the amino acid bead that varied in cross-linking. This impacted the sites for analyzer and the CCD apparatus at Rockefeller and the inabil- chloromethylation and introduction of the C-terminal resi- ity of a graduate student to work reliably 24/7, it was inevitable dues. The physical properties of the polymer also changed as that SPPS should be automated to increase both its productiv- the peptide grew. Each amino acid residue added increased the ity and reliability by eliminating human error during the re- ratio of hydrophilic amide bonds to hydrophobic styrene units. petitive process. Bruce and John made an excellent team for Merrifield attributed this change in accessibility to intermolec- the overall design of an automated instrument30 that they per- ular aggregation of the growing peptide chains (see Chapter on sonally constructed using a drum programmer along with Nils ‘‘Properties of the Polymeric Support’’ in Merrifield’s autobio- Jernberg’s (Rockefeller instrument shop) specially designed graphy10), which is still a viable hypothesis. 12-port valve, two of which were used for amino acid and rea- My Ph.D. thesis, ‘‘Synthesis of an octapeptide, angiotensin gent selections (Figure 9). Since I lived in an apartment effec- II’’ seems, in retrospect, embarrassingly trivial, but Prof. du tively on campus, it became my routine task to check on the Vigneaud had received a Nobel Prize for the synthesis with his synthesizer before retiring every evening. Upon arriving in St. large group of colleagues of only two nonapeptides13 a decade Louis, two 12-port Jernberg valves were purchased and the earlier. I presented this research27 at the FASEB meeting at At- second automated SPPS apparatus constructed. There was lantic City in the spring of 1965. Shortly thereafter, Bruce some modification of the circuitry to include subroutines for received a phone call from Prof. John Josse, newly appointed operations like washing that led to an interest in digital elec- chairman of the Department of Physiology and Biophysics at tronics (but that is clearly another story). Washington University School of Medicine in St. Louis. According to Bruce, the conversation was rather brief–Josse, ‘‘I want one of those;’’ Bruce, ‘‘that’s the only one there is.’’ Instead GENERALIZED SCHEME FOR of completing my thesis and taking the postdoctoral position HETEROPOLYMER AND ORGANIC that Stanford Moore had helped arrange with Prof. Per SYNTHESIS Edman28 in Australia to develop solid phase peptide Bruce Merrifield clearly anticipated application of his meth- sequencing, I interviewed for a faculty position in St. Louis odology to the synthesis of other biopolymers such as nucleic that October, and wrote my first NIH grant before complet- acids (DNA and RNA) and oligosaccharides. The first dinu- ing my Ph.D. thesis (but that is another story). cleotide, dTT, was prepared by SPS in the Merrifield lab in A similar problem in chain extension existed when we 1965 (Marshall and Merrifield, unpublished). The methodol- attempted to synthesize a 74-residue enzyme, acyl carrier ogy for coupling nucleotides was primitive; at that time quanti-

Biopolymers (Peptide Science) The Early Years—Across the Bench From Bruce 197

throughput screening forced the synthetic organic chemistry community to reevaluate solid-phase organic chemistry. The rest is history! Much of the synthetic repertoire of reactions found in a modern organic text has been adapted to SPS. Solid-phase organic chemistry has become the accepted tech- nology for lead discovery and optimization in leading aca- demic centers specializing in organic synthesis. One major event the Marshalls shared with the Merriﬁelds (Figure 11) was an invitation from Dan Veber and Michael Moore, a for- mer graduate student of mine, to join them at SmithKline- Beckman to inaugurate an automated SPOC machine they had custom designed and installed for combinatorial chemistry on polymeric supports.

FIGURE 9 The Merrifield/Stewart/Jernberg graduate student replacement. At least, I was trusted to check how my replacement was doing each night before bed. Note drum programmer, two 12- port rotary valves and shaker with polymeric support. tative reactions of nucleic acid couplings were elusive even with large excesses of reagents. It was Letsinger et al.31,32 and Car- uthers,33 however, who developed the methodology that revo- lutionized the synthesis of oligonucleotide sequences by SPS and opened the door for modern molecular biology. It was also clear to Bruce that the filterable, polymeric protecting group was generally applicable to synthetic organic chemistry (Figure 10). He stated in a review in 1969, ‘‘A gold mine awaits discovery by organic chemists,’’ but it was not his objective to generalize the concept beyond biopolymers. One of the pioneers in this effort was Prof. C. C. Lezn- off of York University in Canada who used insoluble polymeric supports to overcome a number of synthetic organic problems. For example, monoreactions of symmetrical bifunctional compounds was demonstrated using a functionalized diol to react with a symmetrical dialdehyde. Leznoff also used his polymeric supports to synthesize insect sex attractants and carotenoids.34 It is my impression that Prof. Leznoff’s successful demonstration of what became solid phase organic chemistry (SPOC) was not generally appreci- FIGURE 10 Generalized scheme for heteropolymer synthesis ated by the synthetic community and was minimized in his using filterable polymeric protecting groups. Merrifield clearly group because of funding difficulties. anticipated expansion of approach to the synthesis of other bioloy- mers such as nucleic acids (DNA and RNA) and oligosaccharides. Only the development of combinatorial peptide chemistry The first dinucleotide, dTT, was prepared by SPS in the Merrifield 35 36 37 by Geysen, Houghten et al., Lam and coworkers, lab in 1965. Copied from hand-colored original used by author at Furka,38 and others in response to the needs of high- thesis defense.

Biopolymers (Peptide Science) 198 Marshall

FIGURE 11 The Marshalls and Merrifields at SKB for the inauguration of the SPOC automated synthesizer used to generate combinatorial libraries. From left to right, Suzanne Marshall, Garland Marshall, Libby Merrifield, and Bruce Merrifield, all with safety glasses. Finally, global acceptance by the synthetic organic community!

It has been somewhat ironic for me to hear leading syn- rather than any overt criticism. His priorities never wav- thetic chemists discuss their ability to generate combinatorial ered—family first. Perhaps, most impressive to many was his libraries of compounds by this approach considering the dif- humble demeanor when faced with praise. The gas station ficulty in acceptance that faced Merrifield and solid phase on Highway 66 was always an option, although none of us synthesis. I personally collected a fair bit of psychological ever believed Bruce could give up his passion for science. The scar tissue, as did Bruce, in spreading the gospel of this para- inseparables, Bruce and Libby, have served as mentors, col- digm shift in chemistry.39 It still amazes me that something leagues, role models and, most importantly, friends to so obvious to a naı¨ve graduate student trained in biology was Suzanne and myself for over 40 years. It’s a hard act to follow, so threatening and misconstrued by the synthetic commu- but Bruce trained many who have tried to duplicate the nity. Thus, the genesis of one of my favorite aphorisms, ‘‘the example he set. As one can imagine, my receipt of the Merri- only thing wrong with science is that it’s done by people.’’ field Award from the American Peptide Society in 200140 is the highpoint of my scientific career.

HOMAGE TO BRUCE AND LIBBY I thank the editors (Svetlana Mojsov and George Barany) of this It’s impossible to separate my memories of Bruce from those special issue who requested that I commit my Symposium presenta- of Libby and his commitment to their family. Bruce serves as tion to press. Second, my gratitude to all my colleagues who spent a perfect example of the encouraging fact that good guys can time in the Merrifield lab and paid their pound of salt to see Bruce vindicated as a visionary. Thirdly, my sincere appreciation to all my finish first. Bruce taught the most effective way, by example. students and postdocs who have made my own scientific career so There was never a question regarding integrity, scientific or enjoyable. (But then, they never followed my advice to work for otherwise. He found ways to constructively offer a suggestion someone who would obviously win the Nobel Prize; it was a great

Biopolymers (Peptide Science) The Early Years—Across the Bench From Bruce 199 way to jumpstart a career.) Finally, my thanks to the Rockefeller 20. Marshall, G. R.; Feng, J. A.; Kuster, D. J. Biopolymers (Peptide Institute for Medical Research and Prof. D. Wayne Woolley for Sci), in press. giving Bruce the time and freedom to implement his radical con- 21. Kresge, N.; Simoni, R. D.; Hill, R. L. J Biol Chem 2005, 280, e4. cept. In addition, support by taxpayers around the world has 22. Stewart, J. M.; Vavrek, R. J. Adv Exp Med Biol A 1989, 247, enabled our scientific community collectively to realize the potential 81–86. of that dream. I personally think it was a wise investment. 23. Manning, M.; Wuu, T. C.; Baxter, J. W.; Sawyer, W. H. Experien- tia 1968, 24, 659–660. 24. Gisin, B. F.; Merrifield, R. B.; Tosteson, D. C. J Am Chem Soc REFERENCES 1969, 91, 2691–2695. 1. Corner, G. W. A History of the Rockefeller Institute for Medical 25. Gisin, B. F.; Merrifield, R. B. J Am Chem Soc 1972, 94, 6165– Research; The Rockefeller Institute Press: New York, NY, 1964. 6170. 2. Ratliff, F. Biogr Mem Natl Acad Sci 1990, 59, 197–213. 26. Marglin, A.; Merrifield, R. B. J Am Chem Soc 1966, 88, 5051–5052. 3. Henry, K. F. Lupus 2003, 12, 153–250. 27. Marshall, G. R.; Merrifield, R. B. Biochemistry 1965, 4, 2394– 4. Yalow, R. S.; Berson, S. A. Diabetes 1961, 10, 339–344. 2401. 5. Wiener, A. S.; Karl, L. Acta Genet Med Gemellol (Roma) 1968, 28. Edman, P.; Begg, G. Eur J Biochem 1967, 1, 80–91. 17, 641–646. 29. Hancock, W. S.; Prescott, D. J.; Vagelos, P. R.; Marshall, G. R. J 6. Jukes, T. H. J Nutr 1974, 104, 507–511. Org Chem 1973, 38, 774–78l. 7. Roe, D. A. J Nutr 1987, 117, 1324. 30. Merrifield, R. B.; Stewart, J. M.; Jernberg, N. Anal Chem 1966, 8. Dunn, M. S.; McClure, L. E.; Merrifield, R. B. J Biol Chem 1949, 38, 1905–1914. 179, 11–18. 31. Letsinger, R. L.; Caruthers, M. H.; Jerina, D. M. Biochemistry 9. Merrifield, R. B.; Woolley, D. W. J Biol Chem 1952, 197, 521– 1967, 6, 1379–1388. 537. 32. Letsinger, R. L.; Mahadevan, V. J Am Chem Soc 1966, 88, 5319– 10. Merrifield, B. Life During a Golden Age of Peptide Chemistry; 5324. American Chemical Society: Washington, DC, 1993. 33. Caruthers, M. H. Science 1985, 230, 281–285. 11. Carpino, L. A. J Am Chem Soc 1957, 79, 4427–4431. 34. Leznoff, C. C. Accts Chem Res 1978, 11, 327–333. 12. Bergmann, M.; Zervas, L. Berichte der deutschen chemischen 35. Geysen, H. M. Immunol Today 1985, 6, 364–369. Gesellschaft 1932, 65, 1192–1201. 36. Houghten, R. A.; Pinilla, C.; Blondelle, S. E.; Appel, J. R.; 13. du Vigneaud, V.; Ressler, C.; Swan, J. M.; Roberts, C. W.; Dooley, C. T.; Cuervo, J. H. Nature 1991, 354, 84–86. Katsoyannis, P. G.; Gordon, S. J Am Chem Soc 1953, 75, 4879– 37. Lebl, M.; Krchnak, V.; Sepetov, N. F.; Seligmann, B.; Strop, P.; 4880. Feler, S.; Lam, K. S. Biopolymers 1995, 37, 177–198. 14. Dubos, R. J.; Thompson, R. H. S. J Biol Chem 1938, 124, 501–510. 38. Furka, A. In Combinatorial Peptide and Nonpeptide Libraries: 15. Kunitz, M. J Gen Physiol 1940, 24, 15–32. A Handbook; Jung, G., Ed.; VCH: Wenheim, 1996. 16. Moore, S.; Stein, W. H. Science 1973, 180, 458–464. 39. Marshall, G. R. J Pept Sci 2003, 9, 534–544. 17. Hirs, C. H.; Stein, W. H.; Moore, S. J Biol Chem 1954, 211, 40. Marshall, G. R. In Peptides: The Wave of the Future; 941–950. Houghten,R.A.;Lebl,M.,Eds.;AmericanPeptideSociety: 18. Gutte, B.; Merrifield, R. B. J Biol Chem 1971, 246, 1922–1941. San Diego, 2001 (Proceeding of the 17th American Peptide 19. Raines, R. T. Chem Rev 1998, 98, 1045–1066. Symposium).

Biopolymers (Peptide Science)