Since the pioneering studies on biotin-(strept)avidin, mechanical unbinding of a number of protein-biomolecule pairs has been reported (both macromolecules and small ligands, see references in Table 8.2). ttese include interactions between proteins (adhesion proteins, cytoskeletal and motor proteins, GroEL chaperonin, and denatured proteins), receptor-ligand pairs, antigen-Ig pairs, protein-DNA, and protein-lipid. tte mechanical properties of a selection of these protein-biomole-cule pairs are presented in Table 8.2. Studies on the interaction between proteins, antigen-Ig pairs, and receptor-ligand pairs are the most extended works (Zlatanova et al. 2000; Hinterdorfer 2002; Weisel et al. 2003), while interactions between DNA and proteins have received less attention.
As we have seen, a mechanical force causes bond rupture and, intuitively larger forces just increase bond failure rates. Ms is the characteristic behavior of "slip bonds." However, the so-called catch bonds would strengthen under force and their existence has recently been demonstrated experimentally using SMFS in the P-selec-tin adhesion bond. Ms finding might be physiologically relevant since the biphasic transition between catch and slip bonds provides an interesting mechanical switch for the regulation of leukocyte rolling on selectins, which initially increases and then decreases as the wall shear stress increases (Table 8.2; Marshall et al. 2003). Catch bonds also seem to regulate flow-dependent bacterial adhesion.
Recently, AFM has been used to study the energetics of extracting single membrane proteins from their lipidic environment. For example, several studies have analyzed the anchoring forces of bacteriorhodopsin to its native "purple membrane" (a well-characterized seven-transmembrane a-helical protein from Halobacterium salinarum) and the mechanical stability of an a-helix in a hydrophobic environment (reviewed by Frederix et al. 2003; Janovjak et al. 2006). In these experiments, which use a strategy that departs considerably from the classical receptor-ligand methods, single-molecule identification was achieved by AFM imaging (before and after the mechanical pulling), while desorption from the mica substrate was controlled by repeating the experiment on a double layered membrane, which produced similar results. tte contour length of the force peaks was compared with the expected lengths of the loop and transmembrane regions of the protein. In this way, the a-helices were found to detach from the membrane two by two (except for the A a-helix) in the expected order, with forces ranging from 100 to 200 pN (at 0.04 nm ms_1). tte forces needed to simultaneously extract and unfold the A a-helix were found to be similar to those needed to either unfold other a-helices or dissociate them from the membrane once already unfolded (87,99, and 105 pN respectively, at 0.04 nm ms_1). It should be noted that the 99 pN value represents the force needed to unfold (and partially extract) an a-helix in a hydrophobic environment, not in an aqueous solution (Frederix et al. 2003 and references therein). Hence, the bacteriorhodpsin system is a good example of the complexity of this type of experiment and it offers a unique opportunity to get a handle on the elusive structure of membrane proteins. We have not included this information in Table 8.1 as there is not conclusive evidence attributing these force values exclusively to the unfolding of an a-helix in the membrane. Indeed, we cannot rule out the contribution of unbinding interactions with lipids or other proteins in the membrane.
Recently, Kienberger et al. (2005) genetically engineered the bacteriorhodopsin system in order to estimate the force range of intermolecular vs. intramolecular interactions in an antibody-antigen system, ttey found that the intermolecular antibody/antigen unbinding force was significantly lower (126 vs. 204 pN at the same loading rate) than the force required to mechanically extract and unfold the helix pairs from the membrane.
Was this article helpful?