Research Highlights

How A New Explanation for How cAMP Binding Causes Protein Kinase Activation


In spite of several decades of research, scientists have been unable to describe in rigorous quantitative terms the exact sequence of events that leads to activation of Protein Kinase A (PKA) by cyclic adenosine monophosphate (cAMP). But a research team at UCSD will soon publish results that explain this mechanism for a PKA isoform essential for proper functioning of the heart.

In yet another example of the power of interdisciplinary work, this project leveraged a broad range of expertise. Faculty from the departments of Bioengineering, Chemistry and Biochemistry, and Pharmacology in the School of Medicine all played important roles in nucleating this team.

NBCR also provided critical support. NBCR’s recently renewed five-year grant from the NIH included funding for PKA activation as a driving biomedical project and it provided computational support. And, says Britton Boras, the paper’s first author, three members of the bioengineering doctoral thesis committee overseeing his work – Rommie Amaro, Andrew D. McCulloch, and Susan S. Taylor – are affiliated with NBCR.

Boras works in McCulloch lab, which is arguably known best for cell-to-organ cardiac modeling. “I was trying to develop a whole-cell model of the cardiomyocyte, the basic cell of the heart muscle,” he says. “But I kept getting flawed results. I had to go to smaller and smaller scales to isolate the problem. It turned out that we weren’t accurately defining the protein kinase activation mechanism.”

And so this project was born, and Boras sought the critical guidance of biochemist Taylor, a well known expert on the structure and function of PKA signaling proteins. Together, McCulloch and Taylor co-mentor Boras, who is also enrolled in the Interdisciplinary Ph.D. Specialization in Multi-Scale Biology, more commonly known as the UC San Diego Interfaces Graduate Training Program.

Protein kinases, one of the largest gene families in eukaryotic cells, regulate most biological processes. When activated, they modify other proteins by chemically adding phosphate groups to them, which causes functional changes in those proteins.

PKA is present in all mammalian cells and is a particularly important regulator of cardiac development and function. It performs several functions in the cell, including regulation of glycogen, sugar, and lipid metabolism.

Irregular activation of PKA has been implicated in breast and liver cancer, heart disease, and diabetes. It is also associated with multiple endocrine disorders including Cushing’s disease. So it’s important for scientists to understand the mechanism by which PKA is regulated properly.

Boras points out that PKA is one of the first kinases to be discovered and credits Taylor’s pioneering studies of PKA for establishing the molecular basis for protein kinase structure, function, and regulation.

Delving into the biochemistry underlying this research, PKA consists of two catalytic (C)-subunits and a regulatory (R)-subunit dimer. The C-subunits each have binding domains for cAMP. Each R-subunit has two tandem binding domains, A and B.

In the absence of cAMP, the two C-subunits combine with the R-subunit dimer to form an inactive complex. Allosteric activation of PKA is then achieved by binding of cAMP to the complex, which alters the interactions between the R- and C-subunits and unleashes the catalytic activity.

The research team developed well over a dozen Markov State Models to test competing theories about how cAMP binding to the R-subunit leads to activation. Each combination of binding partners (C, R, and cAMP) was considered to be a Markov state, and those states represented various molecular conformations affecting the transitions to and from each state. The transitions between states represented binding events. In essence, the models showed how the C-R interactions are essential to explain PKA activity.

All models were implemented in the Virtual Cell computational environment, programmed into MATLAB code, and used with an existing optimization algorithm. Then the computations were run in parallel on 12 cores of the NBCR Linux cluster.

The researchers found that binding of cAMP to the R-subunit’s B domain plays an essential role in promoting the release of the first C-subunit. However, all four sites need to be bound to enable the release of the second C-subunit, which leads to full activation.

This modeling work is changing how scientists think about PKA activation. “Common wisdom over the last 20 years,” says Boras, “has held that the release happened in the four locations all at once. But now we know that’s not true. It turns out that PKA activation is a cascade of release events.”

Susan Taylor says that recent structures of full-length PKA holoenzymes have provided the first glimpse of PKA as a physiological macromolecular signaling complex where activation is mediated by allosteric binding of cAMP to the R-subunit. “Britton’s computational analysis of this process,” she says, “supports a model where cAMP docks initially to the R-subunit’s B domain, which initiates the process of catalytic activity. It not only supports a conformational selection model for allostery but helps explain why acrodysostosis mutations hinder cAMP binding and produce such severe clinical consequences.” Acrodysostosis is a rare congenital syndrome that produces physical malformations, including shortening of the interphalangeal joints of the hands and feet, physical abnormalities of the face, and intellectual disability.

These results may prove useful in drug design work that targets PKA and other cAMP-binding proteins. But Boras offers some caution: “People think that, once we know the shape of a protein, we can target it. But it’s much more complicated than that. None of these proteins is stationary. They’re all fluid and always in motion. Being able to determine a protein’s most likely conformations can be more important than knowing the amino acid sequence.” He adds that, precisely because PKA is ubiquitous, it’s extremely difficult to target. For example, if you want to target PKA in just the heart, you need to be able to do so without affecting PKA that’s working well in other parts of the body.

Having isolated and solved the problem that led him from the whole cell to its biochemical constituents, Boras is now working back in the other direction across scales from smaller to larger, adapting his model into a whole-cell model to better understand how heart muscle cells work. “While this kind of scientific process may seem circular,” says Boras, “sometimes it’s the best, or maybe only, way to get where you want to go.”


  1. Boras, Britton W., Alexandr Kornev, Susan S. Taylor, and Andrew D. McCulloch, Using Markov State Models to Develop a Mechanistic Understanding of Protein Kinase A-RIα Activation in Response to cAMP Binding, Journal of Biological Chemistry, November 2014. (This citation will be updated with complete information upon publication.)