BOOK REVIEWED - Chemical Biology: From Small Molecules to Systems Biology and Drug Design

• Edited by Stuart Schreiber, Tarun Kapoor & Günther Wess
• Wiley, : 2007 1280 pp., 3 vol., pp. $625.00

• ISBN 9783527311507

Perhaps the central challenge of any book on chemical biology is how to define the field. Over the past decade, the field of chemical biology has accrued large numbers of adherents, spawned new journals and spurred the rebranding of academic departments. Nonetheless, chemical biology remains poorly defined, perhaps reflecting its inclusive nature (some say that all of biology is fundamentally chemical). In commissioning the chapters for a new chemical biology handbook, Stuart Schreiber, Tarun Kapoor and Günther Wess have deftly outlined a broad yet well-delineated vision of the field and its future. The three-volume set Chemical Biology: From Small Molecules to Systems Biology and Drug Design ties together contemporary ideas in the practice of chemical biology, both in academics and, notably, in the pharmaceutical industry. Two important themes emerge from this handbook: (i) the ongoing integration of chemical and systems biology and (ii) the essential role of chemical biology in drug discovery and development. Both themes are accompanied by significant challenges associated with managing and interpreting large quantities of information.

The editors' organization of the three-volume set corresponds loosely to the vantage points from which the chapters view chemical biology. The chapters in the first volume are united in their focus on the interactions of molecules and cells—a theme that matches well with popular definitions of what constitutes chemical biology. This volume contains historical accounts of the use of natural products and their analogs, strategies for small-molecule control of protein function and properties, and narratives describing the subdiscipline of chemical genetics. Volume 2 takes a closer look at the chemistry side of the chemical biology equation, including the synthesis of bioactive molecules and methods for predicting which molecules will display desirable biological properties. In volume 3, the reader sees chemical biology from the perspective of the cell or organism: here the chapters survey target protein families, cellular and organismal delivery of molecules, and the use of systems biology to model cellular responses to small molecule–induced perturbations.

This handbook is by no means a compendium of all of chemical biology, nor should it be. Rather, it consists of a well-chosen assortment of examples in which the use of small molecules or other chemical tools has led to key biological insights. These volumes will be an essential reference work for libraries, but the hefty price tag may impede their entry into individual chemical biology labs. As a whole, the handbook is more of a source book and less well suited to classroom teaching. Each chapter is contributed by a different author (or set of authors), so the style, perspective and level of detail vary. Nonetheless, a number of the chapters conform to a standard outline that includes historical developments, general comments on strengths and limitations of particular approaches, and pivotal examples of successes in the topic under consideration. These structured chapters will be useful in graduate courses because they offer introductions to specific topics through both historical accounts and jumping-off places for critical analysis. Clearly not all topics can be covered in the space of three volumes, and those who study DNA and RNA may feel somewhat shortchanged by this handbook—the focus here is on small molecules and carbohydrates that interact with protein targets. Nucleic acids play a decidedly minor role in these volumes.

One of the strengths of the handbook is the attention given to practical details for implementing and interpreting chemical biology experiments. Many of the contributors do an excellent job of pointing out the unavoidable caveats associated with particular techniques, and their cautionary tales offer a glimpse of future challenges in the field. For example, in a chapter entitled "Engineering Control over Protein Function Using Chemistry," Simon and Shokat describe how the inhibitor resistance and substrate specificity of HIV protease can be inextricably linked. This observation provides the warning that researchers attempting to redesign enzyme active sites should always be cognizant of the possibility that changes designed to selectively affect one activity may have unintended consequences. Similarly, a chapter on "Tags and Probes for Chemical Biology" will be essential reading for those who are interested in sequence-selective chemical tags. The chapter contains a history of the design and use of the powerful FlAsH tag. Also included is invaluable information concerning related compounds that were not successful tags and about situations where FlAsH labeling is less efficient (the oxidizing compartments of the secretory pathway). This comprehensive story will be invaluable to anyone using and developing new chemical tags for specific applications.

Though the chapters have been contributed by different authors, discussions of key molecules recur throughout the books. Rapamycin makes an appearance in no fewer than six chapters, reflecting its immunosuppressive effects, its role in immunophilin discovery and its use as a chemical inducer of dimerization. Similarly, colchicine and capsaicin were essential tools in the discovery of components of the cytoskeleton and heat-sensing receptors, respectively, and these archetypal examples of chemical biology are discussed from multiple perspectives. For example, in the chapter by Lampson and Kapoor ("Using Natural Products to Unravel Biological Mechanisms"), colchicine provides a historical lesson on the indispensable role that small molecules have played in the discovery of the cellular components and mechanisms of mitosis. The authors describe experiments conducted in the 1950s, long before the term chemical biology was in vogue, in which the use of colchicine provided the temporal control that facilitated the discovery of microtubule polymerization and depolymerization. They go on to chronicle the parts that small molecules continue to play in advancing the understanding of mitotic processes, especially in studying the properties of essential proteins, such as tubulin, for which mutagenesis experiments are infeasible.

A key theme that emerges from this handbook is the close relationship between chemical biology practices in academia and drug development methods in the pharmaceutical industry. Technologies such as high-throughput screening, chemical informatics and mathematical modeling of cellular pathways are foundational both to academic chemical biology and to the pharmaceutical industry. The overlap between these endeavors is a common topic of conversation within the chemical biology community: should drug discovery be an academic pursuit? Does it offer appropriate training for students? Can it be effectively practiced using academic resources? Opinions may vary, but this handbook reflects the current reality that there is no clear demarcation between academic and industrial chemical biology: a spectrum of technology development and drug discovery is taking place in both locations. Indeed, authors with industrial affiliations contributed many of the volume 3 chapters that describe target families, small-molecule delivery and systems biology. Graduate students who plan to enter the pharmaceutical industry may find that these chapters provide a useful window into the industrial practice of chemical biology.

Looking toward chemical biology's future, a significant fraction of the handbook is devoted to the computational methods that are sure to shape the future practice of the field. Chief among these are chemical informatics and systems biology. Like all modern biologists, chemical biologists are faced with an explosion of data emanating from high-throughput experimental techniques. Future directions in chemical biology will rely heavily on the development of ways to manage the production and analysis of this torrent of information. Chemical informatics methods seek to parse large quantities of data describing drug-induced phenotypes or activities and to discover the molecular features that confer these properties. A major challenge associated with this endeavor is how to describe molecules computationally: what coordinate system should be used to map chemical space? This pervasive cartographic question appears in chapters on "Diversity-oriented Synthesis," "Forward Chemical Genetics" and "The Target Family Approach," but is discussed most extensively in the "Chemical Informatics" chapter authored by Paul Clemons. In addition to a clear historical account, this chapter provides an overview of the difficulties associated with producing accurate and reliable computational descriptions of molecules. Although a universal standard has not yet emerged, these articles describe the variables that are candidate coordinates for representing relationships among three-dimensional molecules and essential drug-like properties.

Systems-wide models of cellular phenotypes are equally important to modern chemical biology. Whether a phenotype is induced by disease states or by small-molecule perturbations, computational descriptions potentially provide routes to identify the best points for pharmacological intervention. Rather than discovering molecules that inhibit or modulate the activity of a single enzyme, the goal is to discover the relationship between a small molecule and its specific cellular response, which may be mediated through the actions of tens of different proteins. For example, the development of mathematical models of signal transduction pathways is discussed in two chapters on "Computational Methods and Modeling." One important debate is over the utility of phenomenological models not defined in terms of molecular rate constants. Through the use of simplifying assumptions, these models offer portability and minimize adjustable parameters, but their use may obscure mechanistic variations that occur in complex biological systems. In addition, because biological systems are inadequately described by deterministic methods, stochastic techniques are becoming increasingly important. Again, the ideal strategy remains to be defined. Nonetheless, computational methods represent a path away from 'one molecule–one target' approaches and toward a 'one molecule–one phenotype' paradigm.

In 1994, Tim Mitchison outlined a vision for chemists and biologists to collaborate in a new type of genetics, in which libraries of small molecules are used to induce phenotypes of interest and then to discover the relevant biological players (Chem. Biol. 1, 3–6, 1994). The proposed pharmacological genetics has become an important approach both for the discovery of biological mechanisms and for drug development. The next step for chemical biology will be to take full advantage of current methods in systems-wide modeling of cellular behavior. This nascent chemical biology–systems biology alliance offers the promise of predictable methods to discover selective small-molecule perturbagens, which will be essential tools for both academic and industrial chemical biology.