Main

Some experiments are straightforward, others much less so. Kevan Shokat has good examples for both. He says: “It is easy to relate the dose of a drug to a phenotype..., but it is always difficult to relate inhibitor occupancy to phenotype.” Putting a drug, for example, a kinase inhibitor, on cells and measuring the effect gives researchers important insights into the potency of the drug, but it is impossible to infer how much of the inhibitor physically binds to the ATP pocket of the kinase and what effect partial inhibition has on downstream signaling. Shokat and his team at the University of California in San Francisco realized that to explain the signaling capacity of a kinase at a defined inhibitor occupancy, new tools were needed.

He outlines their strategy: “We wanted a chemical genetics system in which we could clamp the kinase activity at 10% or 20% inhibition, and see how the pathway responds to this inhibitor occupancy and how it is coupled to the downstream elements.”

Shokat's team built upon their previous innovation in chemical genetics where they had introduced a space-creating gatekeeper mutation into the ATP-binding pocket of a kinase, thereby making it receptive for bulky ATP analogs that cannot bind to the wild-type kinase. Now they added a cysteine residue at the entrance of the ATP-binding pocket: the gatekeeper mutation allows a bulky inhibitor to enter, and the cysteine facilitates the formation of an irreversible link between kinase and inhibitor (Fig. 1).

Figure 1: Chemical genetics strategy to irreversibly inhibit a kinase.
figure 1

A point mutation in the ATP-binding pocket creates the space for the inhibitor to enter; a cysteine introduced by a second mutation facilitates irreversible binding of the inhibitor.

Full size image

To correlate kinase activity with downstream signaling, they first incubated cells overexpressing the mutated kinase with a submaximal dose of inhi-bitor, then stimulated the kinase and chased with a saturating dose of the inhibitor linked to a fluorophore. The amount of fluorescent kinase shows how much of the enzyme was available for activation after addition of the first submaximal dose of inhibitor. The measurement of downstream substrates of this kinase indicates the activity that a given inhibitor occupancy allows.

A few wrinkles in the technique remain to be ironed out: the cysteine residue is essential for irreversible inhibitor binding, but it is not always obvious from sequence alignments where the cysteine should be placed. Shokat is also concerned that studies in populations of cells may make a clean separation of pathway activities difficult, and he is therefore pursuing a single-cell model. “That way,” he explains, “we can do single-cell clamping of a kinase and measure the exact downstream phosphoactivation of that single cell.”

Additionally, his team is working on a system in which they can independently inhibit two kinases to answer questions about how kinases in different pathways interact.

Looking beyond the interest of his own laboratory, Shokat sees a good chance that this chemical genetics approach will be applicable for other classes of enzymes. A space-creating gatekeeper mutation and a cysteine in the ATP-binding pocket might make other classes of enzymes susceptible to a high-affinity inhibitor—an attractive strategy, especially for enzymes, such as ATPases, for which no high-affinity inhibitor exists now.

If Shokat's approach finds followers, kinases will not remain the only enzymes that are being clamped.