Thrombotic thrombocytopenic purpura (TTP) is a life threatening illness due to a deficiency of the VWF-cleaving protease, ADAMTS13. The ADAMTS13 protein is composed of a propeptide, followed by a typical zinc metalloprotease domain. The C-terminal 2/3 of the molecule contains disintegrin-like, cystine-rich, and spacer domains, as well as a total of eight TSP1 motifs and two CUB domains. The function of this C-terminal portion of the molecule and its composite motifs is unknown, though TSP1 and CUB domains of other proteins have been shown to mediate protein-protein interactions. To further explore the interaction between ADAMTS13 and VWF, we cloned full length human cDNAs for both ADAMTS13 and VWF into the mammalian expression vector pcDNA3.1. These constructs were transiently transfected into 293T cells and COS cells respectively, and conditioned media collected for analysis. Using an anti-myc antibody, myc-tagged VWF co-immunoprecipitated (co-IP) with ADAMTS13, as demonstrated by western blot analysis using antisera raised against a C-terminal peptide derived from the predicted ADAMTS13 sequence. This direct interaction required partial denaturation of VWF in 1M urea, with no co-IP observed in the absence of urea. To map the segment within ADAMTS13 responsible for VWF binding, we cloned a series of overlapping ADAMTS13 fragments into the bacterial expression vector, Pet44b. Fusion proteins were purified by binding of the included His-tag to Ni-NTA beads and incubated with recombinant myc-VWF in the presence of 1M urea. Association with VWF was analyzed by co-IP with anti-myc followed by western blot analysis using an antibody to the C-terminal HSV-tag present in each fusion protein. The CUB2 (Glu1298- Thr1427) fusion protein co-IP’d with full-length VWF and also demonstrated concentration-dependent competition with full-length ADAMTS13 for VWF binding. In summary, we have demonstrated a direct protein-protein interaction between VWF and ADAMTS13. Binding requires partial denaturation of VWF and appears to be mediated primarily through contacts with the ADAMTS13 CUB2 domain. This interaction may account for the previously observed co-purification of VWF and ADAMTS13 from human plasma. Furthermore, the requirement for 1M urea suggests that this interaction may only occur physiologically under conditions of high shear. Though others have shown that the C-terminal domains of ADAMTS13, including CUB2, are not required for VWF cleavage in vitro, our data, together with several C-terminal mutations previously reported in TTP patients, suggest that interactions between VWF and the ADAMTS13 CUB2 domain may be important in vivo.