LBNL Masthead A-Z IndexBerkeley Lab mastheadU.S. Department of Energy logoPhone BookJobsSearch
Structure of the Kinase Domain of CaMKII and Modeling the Holoenzyme Print
Wednesday, 31 May 2006 00:00



The rate and intensity of calcium (Ca2+) currents that oscillate through the plasma membrane around a cell affect such diverse phenomena as fertilization, the cardiac rhythm, and even the formation of memories. How does the cell sense these digital oscillations and transduce them into a cellular signal, such as changes in phosphorylation (addition of a phosphate group to a protein) or gene transcription? A group from the University of California, Berkeley, the Yale University School of Medicine, and Berkeley Lab has combined protein crystallography and small-angle x-ray scattering to give a first glimpse into what this conversion may look like as well as the structural consequences of the conversion.

Experimenting with Synergy

The membranes that separate the inside from the outside of the cells that make up our bodies are neither mechanically rigid nor chemically impenetrable. Nonetheless, they provide an important gate-keeping function by admitting those chemicals that regulate the activity of cellular constituents, such as enzymes that facilitate chemical reactions in the cell. Calcium ions, for example, influence many cellular processes, including fertilization when a sperm fuses with an ovum, the rhythm of the beating heart, and the formation of memories. It turns out that the positively charged calcium enters the cell in waves (oscillating currents), and it is the frequency and the intensity of these waves that are the controlling features. By means of protein crystallography, Rosenberg et al. have obtained atomic-resolution structures for a key portion of the enzyme central to the process. Combining their structure with lower-resolution structural data of the complete enzyme obtained from a technique called small-angle x-ray scattering gave the group a structure-based model for how the cell senses the calcium oscillations and converts them into cellular signals.


The crystal structure of the autoinhibited kinase domain of CaMKII. From top to bottom: The domain organization of CaMKII and two views of the CaMKII dimer, rotated 90° from each other. Molecule A is shown in green and molecule B is shown in blue. The Ca2+/CaM binding residues are shown in magenta and the rest of the regulatory segment is shown in orange. Thr 286 is the residue that is autophosphorylated, leading to Ca2+/CaM independence.

Many years of cell biology and biochemistry have shown that a key player in this process is an enzyme that transfers phosphate groups, the Ca2+/calmodulin-activated protein kinase II (CaMKII). A large protein complex (about 600 kDa), it is made up of twelve individual polypeptide chains (i.e., a dodecamer). Each polypeptide contains a kinase domain, which can phosphorylate downstream targets in the cell, and an “association domain” that links the kinases together in the complete complex (holoenzyme). These two domains are joined by a regulatory segment that binds to the kinase and inhibits its activity, as well as a linker domain of unknown function.

In the absence of Ca2+, the regulatory segment is bound to the kinase, inhibiting its ability to phosphorylate its targets (substrates) including itself (autophosphorylation). When calcium enters the cell, it binds to the small protein calmodulin (CaM), which in turn binds to the regulatory segment, releasing the kinase activity. Underlying the ability of this protein complex to transduce Ca2+ signals is a system, based on the autophosphorylation of the regulatory segment, that allows the kinase activity to be initially dependent on Ca2+/CaM and then to become independent in response to an increasing Ca2+ oscillation frequency, thereby allowing the phosphorylation to persist even after the calcium signal is over.

The group solved the crystal structure of the isolated kinase domain and regulatory segment of CaMKII from data obtained at ALS Beamline 8.2.2. Members were surprised to find that pairs of neighboring kinase domains form dimers where the active site of the kinase domain is blocked by a coiled-coil (two alpha-helix coils coiled together) formed by the regulatory segments. The significance of the dimer structure was suggested by biophysical measurements of Ca2+/CaM binding to the holoenzyme. The measurements showed that the activation of the holoenzyme dodecamer is highly cooperative, thereby suggesting that when Ca2+/CaM binds to the kinase, it releases the activity of more than one kinase domain.


Models of the holoenzyme structure and activation using small-angle x-ray scattering (SAXS) data. Left: A rigid body modeling scheme using data from SAXS experiments resulted in a model of the CaMKII holoenzyme. Right: Ab initio models agree well with the rigid modeling results.

The new structure provides an answer for why this happens: When Ca2+/CaM binds to one inhibiting regulatory segment, it releases its dimer pair at the same time. This insight formed the basis for a model of CaMKII activation by Ca2+/CaM. In its resting state, the CaMKII holoenzyme is a tightly packed and autoinhibited assembly that cannot autophosphorylate. When Ca2+/CaM is added, however, the kinase domains are disrupted and the complex converts into an activated state that is capable of autophosphorylation and Ca2+/CaM-independent activity.

Addition of Ca2+/CaM to the holoenzyme causes a dramatic increase in the radius of gyration of the protein complex.

Because the holoenzyme proved resistant to structural characterization with crystallography, the collaborators turned to small-angle x-ray scattering (SAXS) at the new SIBYLS Beamline 12.3.1. The combination of high-intensity SAXS and crystallography allowed the group to build a model of the holoenzyme structure that was consistent with their crystal structure and their biochemical results. The SAXS model revealed that the kinase holoenzyme is a compact flattened disc about 220 Å in diameter. The addition of Ca2+/CaM to the holoenzyme increases the diameter of the complex by almost 25 Å, suggesting that this addition changes the complex into a much looser association of kinase domains.





Research conducted by O.S. Rosenberg (University of California, Berkeley; Howard Hughes Medical Institute; and Yale University School of Medicine); S. Deindl and R.-J. Sung (University of California, Berkeley, and Howard Hughes Medical Institute); A.C. Nairn (Yale University School of Medicine); and J. Kuriyan (University of California, Berkeley; Howard Hughes Medical Institute; and Berkeley Lab).

Research funding: National Institutes of Health through the Yale School of Medicine Medical Scientist Training Program and the Howard Hughes Medical Institute. Operation of the ALS is supported by the U.S. Department of Energy, Office of Basic Energy Sciences.

Publication about this research: O.S. Rosenberg, S. Deindl, R.-J. Sung, A.C. Nairn, and J. Kuriyan, “Structure of the autoinhibited kinase domain of CaMKII and SAXS analysis of the holoenzyme,” Cell 123, 849 (2005).