Activity of a Two-Domain Antifreeze Protein is Not
Dependent on Linker Sequence
Nolan Holland 1, Yoshiyuki Nishimiya 2, Sakae Tsuda 2 and Frank D Sönnichsen 1*
1 Case Western Reserve University
2 AIST Sapporo
* To whom correspondence should be addressed. E-mail: fds{at}cwru.edu.
Submitted on July 13, 2006
Revised on September 7, 2006
Accepted on 18 September 2006
 |
Abstract |
|---|
The reported NMR structure of RD3, a naturally occurring two-domain antifreeze protein, suggests that the two nearly identical domains are oriented to allow simultaneous binding of their active regions to the ice surface (1). It is implied that the nine-residues linking the two domains play a role in this alignment, but this has not been established. We have designed and expressed a modified form of RD3 that replaces the nine-residue linker with a generic sequence of one serine and eight glycine residues to test the importance of the linker amino acid sequence. The modified linker is shown to have significantly different characteristics compared to the original linker. Heteronuclear NOE experiments show that the new linker residues have more mobility than the linker residues in the native protein. Further, NMR data show that the folding of the C-terminal domain is somewhat perturbed by the altered linker. Finally, distributions of residual dipolar couplings indicate that the two domains tumble and move independently of each other. Nevertheless, the thermal hysteresis activity of the modified protein is indistinguishable from that of native RD3, proving that increased activity of the two-domain AFP is not dependent on structure of the linker.
Key Words:
NMR, Type III AFP, dynamics, residual dipolar coupling, structure, thermal hysteresis
