Structure-function analysis of vascular tethering molecules using atomic force microscope

By studying the structure-function relationship of vascular adhesion molecules, such as L-selectin, GPIb-VWF, and ADAMTS13, which are involved in regulation of leukocyte and platelet adhesion to the vascular wall, we quantify the effects of three specific point mutations on L-selectin on its interaction kinetics with 2-GSP-6 and 6-sulfo-sLex, we characterize the kinetics of GPIbalpha as it interacts with the VWF A1 domain, and we characterize the kinetics of ADAMTS13 as it interacts with the VWF A1A2A3 tri-domain and we characterize its cleavage effects on A1A2A3. The overall project goal is to study how mechanical force regulates the binding kinetics of these proteins.;Selectin-ligand interactions (bonds) mediate the way leukocytes roll on vascular surfaces. The molecular basis for differential ligand recognition by selectins is poorly understood. Atomic force microscopy is used to compare the kinetics of wild-type L-selectin with the kinetics of three mutants of L-selectin interacting with 2-GSP-6; these mutants are a synthetic glycosulfopeptide modeled after the binding site of PSGL-1, and 6-sulfo-sLex, and a synthetic glycan prototypical of PNAd. Rather than first prolong (catch) and then shorten (slip) bond lifetimes, increasing force monotonically shortened the lifetimes of L-selectin MutI (A108H+H110A) and MutIA (A108H) bonds with 2-GSP-6. MutIB (H110A) exhibited an augmented catch bond. L-selectin also formed catch-slip transitional bonds with 6-sulfo-sLex. In sharp contrast, MutI, MutIA and MutIB had no effect on the bond lifetimes. These results distinguish molecular mechanisms for L-selectin to bind to PSGL-1 and PNAd.;Although catch bonds have been observed for selectins interacting with their ligands, it is still not clear whether other cell adhesion molecules also exhibit catch bond behavior. The interaction between glycoprotein Ib (GPIb) and the von Willebrand Factor (VWF) mediates platelet translocation at the vascular vessel damage sites, which plays a critical role in initiating platelet adhesion and thrombus formation. Similar to L-selectin-mediated tethering and rolling of leukocytes, translocation of platelets on VWF requires a shear threshold, suggesting a possible catch bond at work there. We characterized the kinetics of GPIbalpha interacting with the VWF A1 domain, confirming that the catch bond existed. Two type 2B VWD A1 mutants eliminated the catch bond and gave longer low force lifetimes. The prolonged lifetimes at low force resulted in more agglutination of platelets with A1 coated microspheres in flow. Three type 2M VWD A1 mutants showed shifted catch-slip transitional bonds that exhibited shorter lifetimes at low force but longer lifetimes at high force level. A2A3 domains affected the GPIbalpha-A1 catch bond quantitatively. Type III collagen's capturing of A1 or A1A2A3 also quantitatively shifted their bond lifetimes with GPIbalpha, indicating that A1 could have different conformational states.;During the process of hemostasis, the size of prothrombotic ULVWF affects the affinity of VWF to platelets bearing GPIbalpha on the membrane. Seven years ago, ADAMTS13 was identified and characterized as a multi-domain metalloprotease that can cleave at the Tyr1605-Met1606 bond of VWF, thus regulating the size of ULVWF. We studied how force regulated the binding and cleavage of ADAMTS13 on VWF. The full length ADAMTS13 molecule formed catch-slip transitional bonds with A1A2A3 while CUB domains (CUB1&2) only formed slip bonds, suggesting that shear force may play a role in facilitating the enzyme's binding to its substrate. By utilizing the analysis of two force drop events, we found the cleavage effects could only be observed after the catastrophic structural change of A1A2A3. The putative uncoupling of A1 from the A2 domain could only have 14nm contour length increment and would not favor cleavage before A2 unfolding. The putative unfolding of the A2 domain would have much longer contour length increment capacity, depending on how many ss-sheets would be pulled out of the A2 domain. Unfolding the A2 domain exposed the ADAMTS13 cleavage site and favored the cleavage. Two protocols using different stretching molecules (GPIbalpha and CR1) and A1A2A3 immobilization methods (physical adsorption and anti-His capturing A3) revealed that the cleavage effects diminished with increases in stretching force. Regardless of single bond kinetics, time-to-unfold exhibited catch bond behavior for both stretching protocols, suggesting that catch bonds could also be observed during the domain internal structural change. (Abstract shortened by UMI.)...