In some ways, the best targets for therapeutic intervention are the worst ones, too – because they are all things to all people.

“Many of the key regulators of cell function are signaling hubs,” Alexander Hoffmann told BioWorld Today. And because they are signaling hubs, targeting them pharmacologically is a risky business.

NFkB is a prime example. A transcription factor that regulates inflammation when it is activated in response to cytokines and chemokines, NFkB’s levels are elevated in many cancers, and it also plays a role in autoimmune disease.

But targeting NFkB – directly or via its upstream regulators IKK and NEMO – has been challenging to date. “There are good inhibitors,” Hoffmann explained. “But they don’t do well in clinical trials” due to a plethora of side effects.

The reason is that NFkB is one of the strongest transcriptional activators the genome has, possibly the strongest one, and so its inhibition can do more harm than good.

Starkly, NFkB knockout animals die before birth. And while the partial inhibition of a protein long after birth is a different scenario from its total loss before conception, in clinical trials, a big problem has been that the factor is also involved in the response to pathogens, and so globally inhibiting its function leaves patients too vulnerable to infections.

Now, Hoffmann and his team may have found a way to specifically target NF-kappaB’s effects in cancer – and, more generally, a new way to target specific aspects of the activity of signaling hubs.

They published their results in the Oct. 10, 2013, issue of Cell.

Beyond the specific results it described, the paper represents a different way of conceiving of hub signaling, and how to target it, than what is now most typical.

Receptors, kinases and so forth “shouldn’t be seen as on-off switches,” Hoffmann said.

He likened such hubs, instead, to a telephone wire, which transmits a multitude of signals. “You can say different words, even though you only have one wire.”

Specifically, different inputs will activate hubs in different ways, which leads to different activity patterns. For example, a hub might be activated very rapidly, with a strong early peak. Or it might take longer to become active.

“The dynamic pattern of activity represents, essentially, a language,” Hoffmann said. And so if a kinase or other hub protein has a signature activity pattern under certain circumstances, the question is “Can we reduce or eliminate just this feature, rather than inhibiting the whole kinase?”

By targeting such features – “not the kinase itself, but its temporal regulation . . . you may in fact recover [hub proteins] as useful drug targets.”

That approach also implies that looking for the strongest inhibitors of a hub is not the strategy that is most likely to bring success. Instead, it is more likely to cause trouble for two separate reasons – because it wipes out desirable and undesirable activities of the hub alike, and because such strong inhibitors would be more likely than weaker ones to inhibit additional targets as well.

In their paper, Hoffmann and his team began by using mathematical modeling or virtual screening to establish general patterns in which hubs may be activated. In those experiments, they found that the activation of such hubs could be divided, broadly speaking, into two types.

“You can classify mechanisms that impinge on a kinase as controlling either steady-state or dynamics,” Hoffmann said. And so, inhibitors could also be classified as affecting one or the other.

The team next applied those rules to a real-world case, using cell culture experiments to look for inhibitors that would be able to affect specific features of NFkB activation. They were able to identify such specific inhibitors, which means that in principle it could be possible to block NFkB therapeutically without running headlong into side effects.

“Cancer is a state that is quasi-equilibrium – things are moving slowly,” Hoffmann explained, at least compared to the response to a pathogen, where NFkB activator tumor necrosis factor is generated in bursts that lead to NFkB upregulation within minutes.

“If you can find drugs that inhibit the equilibrium,” he said, “then those drugs might not activate the immune response.”