a general prescription for treating disease. Drugs are the primary tools for this purpose, and the synthetic organic chemistry required to fashion them today is much the same as it was a century ago. Finally, the economic hurdles associated with drug discovery are daunting. This Pioneer proposal addresses a technology that can close the gap between basic research discoveries, and the application of such insights to medicine. The approach, called """"""""chemical evolution"""""""" (see below), provides the means to breed drugs out of enormous synthetic small-molecule populations. It has the potential to transform drug discovery from a process requiring hundreds of chemist-years and the infrastructure of a large pharmaceutical company, to something a graduate student with knowledge of basic molecular biology can accomplish in a month. Chemical evolution is closely related to nucleic-acid and protein evolution techniques with proven track records in academia and industry. Moreover, our recent pilot studies have definitively established the feasibility of evolving small molecules[1-3]. These studies were the subject of two Science and Technology review articles in Chemical and Engineering News over the last year, and they were named a """"""""Chemistry Highlight"""""""" for 2004 (a short annual compilation by the American Chemical Society of key advances in chemistry)[4-6]. Despite its enormous potential and the excitement it engenders, three different federal agencies have declined to fund further development of the technology on the grounds that it is too ambitious and too risky. Shaping Small Molecules A single paradigm has dominated drug development in the modern era. Discovery begins with lead compounds, which are either known medicinal agents or molecules identified by high-throughput screening of compound collections. Variants of the lead compounds are synthesized by an army of medicinal chemists, and retested for biological activity. Efficacy, specificity and pharmacological profiles are incrementally optimized until derivatives can be found that perform adequately in animal models. The optimized molecules are then tested in human beings through three stages of clinical trials. Recent economic studies put the total cost of this process in excess of 800 million U.S. dollars for the development of a single successful drug, with the price tag for the pre-clinical phase falling in the range of 300-600 million dollars[7]. Paradoxically, the traditional discovery process is becoming increasingly inefficient. Gross expenditures on drug R&D in the U.S. increased from $15 billion in 1996 to $25 billion in 2002, while the number of 2005 Pioneer Application --- 1 new molecular entities approved yearly by the FDA dropped from 56 to 17 over the same period[8]. A radically different paradigm for molecular discovery, based on mimicking evolution in a test tube, is beginning to challenge the traditional approach. The techniques, collectively termed in vitro evolution1, utilize diverse libraries of gene products, each of which is physically associated with a corresponding DNA blueprint. The library mixtures are subjected to affinity purification over an immobilized macromolecule, to physically isolate gene products with a desired binding property (many alternative functional selections have also been applied). The DNA blueprints in the isolated pool are then amplified and translated to produce a second-generation library. Over multiple iterations of this selection, amplification and translation process, high-affinity and high-specificity binders to the macromolecule target emerge from the population. Isolation of nucleic acid and peptide ligands from libraries of up to one quadrillion different molecules has become a routine practice that is widely used in basic science labs[9]. The approach is simple, effective and inexpensive. Aside from cost and speed, in vitro evolution methods have a huge numerical advantage over traditional screening strategies. Large numbers translate into quality advantages in the products that are isolated[10]. Estimates based on the human antibody, T-cell receptor and olfactory receptor repertoires suggest that libraries with complexities exceeding 1010 should generally provide ligands with nanomolar affinities[11]. A typical small-molecule collection at a pharmaceutical company comprises one million different compounds, and is analyzed at a rate of 300,000 compounds/day. By comparison, in vitro evolution libraries generally comprise 109-1014 distinct molecules. To put the difference in perspective, a bead-based2 library of 1014 compounds would flood an Olympic pool, would weigh hundreds of thousands of kilograms, and would require several hundred millennia to screen at a state-of-the-art facility. The impact of in vitro evolution approaches on drug discovery has been quite limited because the techniques are restricted to biopolymer libraries. Nucleic acid aptamers and peptide ligands generally lack the membrane permeability and metabolic stability required of medicines, although special cases are beginning to appear in the clinics. However, two new technologies, DNA display (developed in my lab) and DNA- templated synthesis (DTS), have recently changed the fundamental nature of the game[1- 3, 12]. DNA display and DTS facilitate direct chemical translation of DNA genes into covalently attached synthetic small-molecule gene products. Thus, in vitro evolution of combinatorial chemistry libraries is now possible. We proved the point by chemically translating a DNA-encoded library of one million synthetic pentapeptides, and affinity maturing it over two generations against the [Leu]enkephalin-specific 3E7 antibody to produce a nanomolar ligand (Figure 1). This accomplishment currently represents one of only two chemically translated libraries (ours being greater than four orders of magnitude larger than that of Liu et. al.[12]), and it is the only chemically translated library to have been propagated over multiple generations through DNA blueprints. We have developed general, highly efficient methods for synthesis on DNA supports, and have carried out in 1 Examples of in vitro evolution methods include aptamer selection, SELEX, phage display, polyribosome display, RNA display and peptides on plasmids. 2 Synthetic libraries used in high-throughput screening are often prepared on inert polystyrene beads, with one compound per bead. 2005 Pioneer Application --- 2 vitro selections against multiple proteins. The work creates enormous, unexplored opportunities in chemical evolution. To transform these pilot studies into a useful tool, the chemical evolution approach must be adapted to operate on complex libraries of """"""""drugable"""""""" small molecules. Were this achieved, basic-science labs could rapidly and cheaply identify cell permeable ligands with nanomolar to picomolar KD's for the biological entities they study, with tremendous benefits for medical research. Fig. 1. Left: The DNA display microcolumn splitting scheme used to encode synthesis. The DNA support library (top) is split by hybridization to anticodon microcolumns (colored), and then transferred to DEAE microcolumns (black and white). The DNA supports in each sub-pool are coupled to a distinct chemical building block through the reactive group at the 5'-end of the support. The DNA is then pooled and re-split based on the subsequent coding region. Right: Affinity maturation of a chemically synthesized library. Approximately 70 DNA genes from each round of translation, selection and amplification were sequenced, and the results are summarized as a histogram plot. The x-axis indicates the number of amino acid residue matches to [Leu]enkephalin encoded by a library sequence. The y-axis indicates the round of library evolution (0: starting material, 1: after round one selection, 2: after round two selection). The z-axis indicates the number of sequences encoding a particular number of matches (x-axis) in a particular round (y-axis). The critical attributes of chemical evolution are that it is inherently """"""""low tech"""""""", inexpensive and easy. Thus, it eliminates economic and technological barriers that impede molecular discovery. It can empower a wide range of scientists with minimal resources, and who lack a lifetime of prior experience in synthetic organic chemistry, to explore """"""""chemical space"""""""" for compounds with rare and desired properties. The reactions used to build small-molecule populations, and to produce individual compounds on a therapeutic scale, are necessarily straightforward (no Mount Everest-type complex natural product synthesis is involved). The cost of producing large quantities of the resulting small-molecule drugs is manageable, as opposed to the case for protein and nucleic-acid therapies. Forward Applications of Chemical Evolution Since completing our pilot studies last summer, we have worked on developing a massively parallel format to evolve drugable libraries of ~1011 compounds (the compounds conform to the """"""""Lipinski rules of five"""""""", an empirical set of physical chemical criteria characterizing successful prescription drugs with good oral bioavailability). We have organized a series of collaborations to test the technology in animal models. Selective kinase inhibitors - We are evolving inhibitors active against the frequently occurring Gleevec-resistant mutant Bcr-Abl(T315I). The mutant kinase arises in acute lymphoblastic leukemia patients, and in chromic myelogenous 2005 Pioneer Application --- 3 leukemia patients experiencing blast crisis. In collaboration with David Lockhart (Ambit Biosciences) we will profile the kinase inhibitors against the entire human kinome, and determine whether we can evolve inhibitors with singular specificity. We have established a second collaboration with Charles Sawyers (UCLA Medical School) to test the molecules in a SCID mouse model of drug resistant CML. Dengue Virus protease inhibitors - In collaboration with Karla Kirkegaard (Stanford Medical School), we are evolving inhibitors of the Dengue Virus NS2/3 protease. We will test the resulting molecules in an interferon ?/? receptor- interferon ? receptor double knockout mouse model of Dengue infection. Interleukin-5 Receptor ? antagonists and asthma - In collaboration with Greg Barsh (Stanford Medical School), we are evolving antagonists of the IL-5R? receptor, which is exclusively expressed in basophils and eosinophils. We will test whether coordinate pharmacological inhibition of the IL-5R? receptor and the eotaxin CCR3 receptor can reverse induced airway eosinophilia and airway hyper-responsiveness in a BALC/c mouse model of asthma. Pathway-specific PET contrast reagents for early detection of cancer - In collaboration with Pat Brown (Stanford Medical School), we are evolving tyrosine-containing substrates that are selective for a single tyrosine kinase. We will test whether cells with dysregulated tyrosine kinase activity selectively accumulate flourine-18 derivatives of these substrates, and whether we can exploit this phenomenon to non-invasively image aberrant growth signals by PET scanning. Countless applications of chemical evolution exist. Some ideas include creating cell-type specific targeting reagents, evolving small molecule affinity tags for """"""""omics"""""""" arrays, producing """"""""smart sensor materials"""""""" that possess molecular recognition properties, and generating small-molecule catalysts for asymmetric and regioselective organic synthesis. Chemical evolution might accelerate the acquisition of knowledge in vertebrate molecular physiology by readily providing small molecule tools for experimentation in animal models (zebrafish, mouse, primates etc.). In addition, chemical evolution could potentially solve the problem of engineering small molecules to disrupt protein-protein interactions, and make small molecules ligands as accessible as antibodies, only cheaper. I consider each of our ongoing projects extremely worthwhile, but efforts to secure funding for them have failed. Reviewers have commended the ideas (""""""""clever"""""""" and """"""""state of the art"""""""") while ultimately dinging the proposals as too risky. A Pioneer award could fund all of the projects at a high level. Protein Footprinting and Computational Design My lab at Stanford includes two research programs unrelated to chemical evolution. One program aims to develop alternatives to x-ray crystallography and NMR spectroscopy for measuring protein structure """"""""in the wild"""""""". A key goal is to obtain structural data on proteins in biologically relevant environments, such as in a membrane 2005 Pioneer Application --- 4 bilayer, on a cytoskeletal filament, or as part of a large complexes. We have developed a protein footrprinting approach called MPAX that combines quantitative aspects of hydrogen deuterium exchange with the versatility of chemical modification techniques[13]. MPAX utilizes a novel translational misincorporation strategy to introduce cysteine structural probes throughout a protein, and allows us to monitor in parallel the accessibility of the probes to solvent, and the identity of the residues neighboring the probes. Early work with MPAX has provided the first detailed unfolding pathway for a large protein, the (?/?)8-barrel triosephosphate isomerase[14]. In ongoing work, we are adapting MPAX for de novo determination of protein three-dimensional structures. We are also developing kinetic footprinting reagents that will allow measurement of protein conformational changes on a sub-millisecond timescale. MPAX is a new and general structural tool that has been adopted by a number of labs in the United States, Europe and Australia. The ultimate goal of the third research program is to computationally engineer enzymes. As a graduate student, I designed the first protein sequence to adopt a backbone topology not observed in nature[15]. My lab at Stanford has gone on to develop more sophisticated protein modeling algorithms, and we have applied them to the problem of automating the design of specificity in molecular recognition[16, 17]. A major aspect of our work is to base all designs on physically rigorous molecular mechanics potentials with a continuum treatment of solvation. This intellectual perspective runs counter to the prevailing trend in the computational design field, which relies almost exclusively on knowledge-based statistical potentials. Total Professional Effort: If I receive a Pioneer award, I will commit at least 75% of my professional effort to the science it supports. All of the projects I have tackled since arriving at Stanford are high-risk/high-payoff;so finding funding has been a constant battle. 1. Halpin, D.R. and P.B. Harbury, DNA display I. Sequence-encoded routing of DNA populations. PLoS Biol, 2004. 2(7): p. E173. 2. Halpin, D.R. and P.B. Harbury, DNA Display II. Genetic Manipulation of Combinatorial Chemistry Libraries for Small- Molecule Evolution. PLoS Biol, 2004. 2(7): p. E174. 3. Halpin, D.R., et al., DNA Display III. Solid-Phase Organic Synthesis on Unprotected DNA. PLoS Biol, 2004. 2(7): p. E175. 4. Borman, S., A Genetic Code for Organic Chemistry. Chemical &Engineering News, 2004. 82(28): p. 23-24. 5. Borman, S., Chemistry Highlights 2004. Chemical &Engineering News, 2004. 82(51): p. 53-61. 6. Henry, C.M., DNA-Programmed Organic Synthesis. Chemical &Engineering News, 2005. 83(5): p. 35-36. 7. Rawlins, M.D., Cutting the cost of drug development? Nat Rev Drug Discov, 2004. 3(4): p. 360-4. 8. Hall, S., Revitalizing Drug Discovery, in Technology Review. 2003. p. 38-45. 9. Roberts, R.W. and W.J. Ja, In vitro selection of nucleic acids and proteins: what are we learning? CURRENT OPINION IN STRUCTURAL BIOLOGY, 1999. 9(4): p. 521-529. 10. Carothers, J.M., et al., Informational complexity and functional activity of RNA structures. J Am Chem Soc, 2004. 126(16): p. 5130-7. 11. Lancet, D., E. Sadovsky, and E. Seidemann, Probability model for molecular recognition in biological receptor repertoires: significance to the olfactory system. Proc Natl Acad Sci U S A, 1993. 90(8): p. 3715-9. 12. Gartner, Z.J., et al., DNA-templated organic synthesis and selection of a library of macrocycles. Science, 2004. 305(5690): p. 1601-5. 13. Silverman, J.A. and P.B. Harbury, Rapid mapping of protein structure, interactions, and ligand binding by misincorporation proton-alkyl exchange. J Biol Chem, 2002. 277(34): p. 30968-75. 14. Silverman, J.A. and P.B. Harbury, The equilibrium unfolding pathway of a (beta/alpha)8 barrel. J Mol Biol, 2002. 324(5): p. 1031-40. 15. Harbury, P.B., et al., High-resolution protein design with backbone freedom. Science, 1998. 282(5393): p. 1462-7. 16. Havranek, J.J. and P.B. Harbury, Tanford-Kirkwood electrostatics for protein modeling. Proc Natl Acad Sci U S A, 1999. 96(20): p. 11145-50. 17. Havranek, J.J. and P.B. Harbury, Automated design of specificity in molecular recognition. Nat Struct Biol, 2003. 10(1): p. 45-52. 2005 Pioneer Application --- 5 Principal Investigator/Program Director (Last, First, Middle): Harbury, Pehr A. B. BIOGRAPHICAL SKETCH Provide the following information for the key personnel in the order listed on Form Page 2. Follow this format for each person. DO NOT EXCEED FOUR PAGES. NAME POSITION TITLE Pehr A.B. Harbury Associate Professor, Biochemistry EDUCATION/TRAINING (Begin with baccalaureate or other initial professional education, such as nursing, and include postdoctoral training.) INSTITUTION AND LOCATION Harvard University, Cambridge, MA Harvard University, Cambridge, MA DEGREE (if applicable) YEAR(s) B.A. Ph.D. 1987 1994 FIELD OF STUDY Biochemistry Biological Chemistry RESEARCH AND PROFESSIONAL EXPERIENCE: Concluding with present position, list, in chronological order, previous employment, experience, and honors. Include present membership on any Federal Government public advisory committee. List, in chronological order, the titles, all authors, and complete references to all publications during the past three years and to representative earlier publications pertinent to this application. If the list of publications in the last three years exceeds two pages, select the most pertinent publications. DO NOT EXCEED TWO PAGES. Research Background: 7/04 ?Present Associate Professor Department of Biochemistry Stanford University 8/97 - present Assistant Professor Department of Biochemistry Stanford University 9/95 - 8/97 Damon Runyon-Walter Winchell Postdoctoral Fellow in the laboratory of Professor Peter Schultz University of California at Berkeley 3/90 - 8/95 Graduate Student and Postdoctoral Associate in the laboratory of Professor Peter Kim Whitehead Institute at MIT 2/89 - 2/90 Rotation Student in the laboratories of Professor James Wang and Professor Stephen C. Harrison Harvard University 9/87 - 9/88 Technician in the laboratory of Professor Kevin Struhl Harvard Medical School 1986 ?1987 Undergraduate Research Assistant in the laboratory of Professor Mark Ptashne Harvard University Awards: National Merit Scholar, 1984-1987 Magna cum laude with highest honors in biochemistry, Harvard University (1987) Damon Runyon-Walter Winchell Cancer Research Fellow (1995-1997) Searle Scholar of The Chicago Community Trust (1999-2002) Terman Fellow of the Lucille Packard Charitable Trust (1998-2001) Technology Review Magazine's 100 Young Innovators of 1999 Agency: Massachusetts Institute of Technology Burroughs Wellcome Young Investigator in the Pharmacological Sciences (2000-2003) Schering-Plough Awardee of the ASBMB (2004) PHS 398/2590 (Rev. 05/01) Page Biographical Sketch Format Page

National Institute of Health (NIH)
National Institute of Biomedical Imaging and Bioengineering (NIBIB)
Biotechnology Resource Grants (P41)
Project #
Application #
Study Section
Special Emphasis Panel (ZRG1-SBIB-C (40))
Program Officer
Peng, Grace
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of Southern California
Biomedical Engineering
Schools of Engineering
Los Angeles
United States
Zip Code
Chalacheva, Patjanaporn; Khaleel, Maha; Sunwoo, John et al. (2017) Biophysical markers of the peripheral vasoconstriction response to pain in sickle cell disease. PLoS One 12:e0178353
Zhang, Yaping; Hu, Kaiqiang; Beumer, Jan H et al. (2017) RAD-ADAPT: Software for modelling clonogenic assay data in radiation biology. DNA Repair (Amst) 52:24-30
Song, Gina; Pacini, Giovanni; Ahrén, Bo et al. (2017) Glucagon increases insulin levels by stimulating insulin secretion without effect on insulin clearance in mice. Peptides 88:74-79
Henley, Brandon Christian; Shin, Dae C; Zhang, Rong et al. (2017) Compartmental and Data-Based Modeling of Cerebral Hemodynamics: Nonlinear Analysis. IEEE Trans Biomed Eng 64:1078-1088
Lytton, William W; Arle, Jeff; Bobashev, Georgiy et al. (2017) Multiscale modeling in the clinic: diseases of the brain and nervous system. Brain Inform 4:219-230
Dolton, Michael J; D'Argenio, David Z (2017) Population-based meta-analysis of roxithromycin pharmacokinetics: dosing implications of saturable absorption and protein binding. J Antimicrob Chemother 72:1129-1136
Cole, Marc; Eikenberry, Steffen; Kato, Takahide et al. (2017) Nonparametric Model of Smooth Muscle Force Production During Electrical Stimulation. J Comput Biol 24:229-237
Bensman, Timothy J; Wang, Joshua; Jayne, Jordanna et al. (2017) Pharmacokinetic-Pharmacodynamic Target Attainment Analyses To Determine Optimal Dosing of Ceftazidime-Avibactam for the Treatment of Acute Pulmonary Exacerbations in Patients with Cystic Fibrosis. Antimicrob Agents Chemother 61:
Geng, Kunling; Marmarelis, Vasilis Z (2017) Methodology of Recurrent Laguerre-Volterra Network for Modeling Nonlinear Dynamic Systems. IEEE Trans Neural Netw Learn Syst 28:2196-2208
Marmarelis, Vasilis Z; Shin, Dae C; Tarumi, Takashi et al. (2017) Comparison of Model-Based Indices of Cerebral Autoregulation and Vasomotor Reactivity Using Transcranial Doppler versus Near-Infrared Spectroscopy in Patients with Amnestic Mild Cognitive Impairment. J Alzheimers Dis 56:89-105

Showing the most recent 10 out of 143 publications