The Quaternary Structure

The authors took advantage of sequence and structural similarities between the voltage- and calcium-activated BK channel and the sodium-activated BK channel chicken homologue Slo2.2 to deduce the tetrameric structure of the CTD domain of the human BK channel. Amino acids 347 to 1201 (the CTD domain) of Slo2.2 were expressed and purified in a vector and it was found that the Slo2.2 CTD forms a very stable tetramer because it migrates as a tetramer on an SDS-PAGE gel. The tetramer even eluted in size exclusion chromatography!

Slo2.2 was crystallized in the presence of 500mM of sodium and this led to an x-ray diffraction pattern of 6.0 angstrom resolution. The results were of space group I422 and contained one monomer in its asymmetric unit. Through the use of a polyalanine model of the BK channel the authors were able to obtain a single solution in molecular replacement (a method used to solve the phase problem in x-ray crystallography) against the Slo2.2 diffraction meaning that the two structures are very similar. This similarity might mean that the human BK CTD also forms a stable tetramer.

Finally, the authors applied the crystallographic symmetry operators to obtain the tetramer structure of the human BK CTD. While this gives the rough details of the quaternary structure, it does not give us atomic resolution because of the low resolution of the Slo2.2 crystal diffraction.

Biological Evidence to support the tetrameric structure.

1) Subunits in this conformation correspond well to the "gating ring" observed in the MthK (ring formed by alternately packing of flexible and assembly interfaces). Differences are that the assembly and flexible interfaces are formed between independent and identical RCK domains in MthK channels. However, in the human BK channel, the flexible interface is formed between the helix-turn-helix motifs between RCK1 and RCK2 [see fig. 5] and the assembly interface is formed between RCK1 of one monomer with the RCK2 of another.

2) Amino acid pairs  that are spatially far apart in the monomer have been shown to interact in double mutant cycles and are close in proximity to each other through the creation of an assembly  interface. [see fig. 6] [see fig. 7]

Fig 11: Assembly interface of the human BK channel (adapted from Peng's Yuan, et. al's paper).

However, there is a difference between the MthK and BK gating ring structures. The calcium binds on the assembly interface on the BK whereas it binds on the flexible interface on the MthK channel.

Fig 12: Tetramer formed by four human BK channel CTD regions (left) and the tetrameric srtucture of the prokaryotic MthK potassium channel (Both images adapted from Peng's Yuan, et. al's paper).

Even though the sequences of the MthK channel, Slo2.2 channel and human BK channel differ greatly in sequence, their tertiary structures are still similar. However, there are still minor differences between them. MthK is more symmetrical as is normal in most proteins in prokaryotic species. Other differences include the fact that the MthK channel binds 8 calcium ions while the human BK channel binds 4 and in the human BK channel they binds near the assembly interface whereas in the prokaryote homologue they bind closer to the flexible interface. These differences are thought to have occurred through evolution.