Chemical genetic approaches for the elucidation of signaling pathways

Abstract

New chemical methods that use small molecules to perturb cellular function in ways analogous to genetics have recently been developed. These approaches include both synthetic methods for discovering small molecules capable of acting like genetic mutations, and techniques that combine the advantages of genetics and chemistry to optimize the potency and specificity of small-molecule inhibitors. Both approaches have been used to study protein function in vivo and have provided insights into complex signaling cascades.

Introduction

Renewed appreciation for the power of small organic molecules to address questions in cell biology has fueled an explosion of interest in chemical biology. The idea that small (MW<700 Da) drug-like molecules can perturb the function of specific proteins is a central tenet of pharmacology. This is further bolstered by the fact that many cellular functions are carried out by small molecules (e.g. ATP, neurotransmitters, steroid hormones, prostaglandins and phosphoinositides). Using small molecules to perturb protein function is particularly useful because the effects of drugs are:

• Rapid, potentially diffusion-limited.

• Often reversible because of metabolism/clearing.

• Tunable, enabling graded phenotypes by varying concentration.

• Conditional, because they can be introduced at any point in development.

Despite these advantages, the use of small molecules to probe cellular signaling has lagged behind that of genetic or biochemical methods. Two types of genetic experiments have provided a wealth of information about cellular processes. When identification of new components of a pathway are desired, forward genetics is used. This involves generating large numbers of mutants, screening to identify those with either gain- or loss-of-function phenotypes in a process of interest, and then identifying the mutations in the specific genes that underlie the phenotype. In order to study the function of one component, reverse genetics is used. Mutations are targeted to a particular protein and the role of the protein is inferred from the phenotype of the resulting mutant.

The advantages of genetics are that it is both highly specific (a single nucleotide change in 3 billion base pairs [bps]) and highly portable (any organism can, in principle, be genetically modified). In contrast, it is often difficult to identify a small molecule that binds specifically to a single enzyme active site out of up to 30,000 proteins, some of which have highly homologous active sites. Additionally, unlike genetic systems, the synthesis of each small molecule presents unique challenges and is not sufficiently general to systematically inactivate every gene product in an organism.

Nonetheless, despite the important advantages of high specificity and portability there can be problems with traditional genetics. For example, when gene knock-outs are lethal, further study of the mutant organism is complicated. Also, most genetic mutations are not conditional; they cannot be turned on or off at will. Although conditional mutations can be introduced through the use of an inducible promoter, generation of the mutated protein occurs over hours to days. Conditional mutations can also be found that are rapidly manifested by an external stress, such as heat (TS alleles), but this stress can often have unwanted side-effects such as induction of heat stress proteins. Thirdly, knock-out phenotypes of non-essential genes can often be masked by functional compensation by related genes during development by the organism. Chemical genetic strategies using small molecules that act as mutations would complement traditional genetic studies by providing a general means to rapidly and conditionally inactivate proteins.

The key question becomes, how can we identify specific chemical inhibitors of every gene product in the yeast, worm, fly, mouse and human proteomes? The ideal drug is one that shows perfect target-specific behavior and can be given to an animal to inactivate its target almost instantaneously. In this review, we highlight two complementary approaches to generate and identify such molecules, and describe ways in which these molecules have been used to study biological processes.