The C-terminal Domain

Since it is difficult to work with transmembrane proteins (difficult to crystallize) the authors of this paper have decided to simply analyze the ~800 amino acid CTD instead. This was done by expressing residues 341 to 1056 in Spodoptera frugiperda (insect cells), growing crystals in 50mM Ca²⁺ solution and using x-ray crystallography to obtain a 3.3 Å image which was then refined to R/R_free = 0.25/0.28 to obtain the 3.0 Å model (where R = R-factor and R_free = free R-factor). 

With this electron density map and by using the already elucidated structure of the prokaryotic MthK K⁺ channel homologue, they were able to determine that the CTD is composed of two tandem (regulator of K⁺ conductance) RCK domains, RCK1 and RCK2. 

Fig. 3: Structure of the C-terminal domain. Red: RCK1; Green: RCK2; Yellow: Calcium ion.

Each RCK is bi-lobular and is composed of a large N-terminal lobe, which contains a Rossman fold, followed by a helix-turn-helix connector and the small C-terminal lobe.

Fig 4: Detailed image showing the 2 lobes of RCK1. Green: C-terminal lobe (small), Red: N-terminal lobe (large - contains Rossman Fold) & Blue: Helix-turn-helix motif.

The RCK1 and RCK2 domains are folded tightly against each other in a similar way that two separate identical RCK domains dimerize to form the "flexible interface" in MthK K⁺ channels. This similarity extends to the use of the helix-turn-helix connectors as clasps to attach the two lobes of each RCK domain together. However, a difference can be seen between the two structures in that the interface between the two RCK domains is larger in the BK K⁺ channel than in the MthK K⁺ channel. However, signature features are still conserved. This homology is used later on to superimpose the human BK channel CTD tetramer onto the MthK channel.

In fact, the RCK1-RCK2 interface hides a surface area of 8500 Ų from the surrounding solvent and has a shape complimenatrity of index of 0.68! A value similar to a typical Antigen-Antibody interface!

Fig 5: Representation of the clasp-like interface between the helix-turn-helix motifs in RCK 1 and RCK 2. Red:

αS αT helices of the RCK2 domain; Green: αF αG helices of the RCK1 domain; Yellow: Hydrophobic residues.

Even though the structure is complete, there are still two linker regions (one between RCK1 and RCK2 and the other between the αQ to βO in RCK2) that could not be elucidated due to their unstructured nature. However, it has been shown that both these regions are nonessential to the function of the CTD regions because large chunks of these regions can be deleted without adverse effects. However, these linkers have been shown to need a minimum length and might be implicated in higher level forms of regulation because they both contain consensus sequences for binding SH3 domains.

Fig 6: [click to enlarge] Domain topology of the RCK1 of the human BK channel. Represents the amino acid sequence and the labeled secondary structure elements that create the tertiary structure. Dotted line means that the region is unstructured (as described above). (Adapted from Peng's Yuan, et. al's paper).

Fig 7: [click to enlarge] Domain topology of the RCK2 of the human BK channel. Represents the amino acid sequence and the labeled secondary structure elements that create the tertiary structure. Dotted line means that the region is unstructured (as described above). (Adapted from Peng's Yuan, et. al's paper).

Another structure similarity between the E. coli MthK and the human BK is a salt bridge that is conserved between K448 and D481. Furthermore, it is thought that the two RCK domains in human BK undergo a calcium induced conformational change across the flexible interface in a similar manner to the RCK dimers of the MthK channel.

Fig 8: Diagram to highlight the salt bridge interaction between D481 and K448 in RCK1.