Thrombin (T) inactivation by the serpin, heparin cofactor II (HCII), is accelerated by the glycosaminoglycans (GAGs) dermatan sulfate (DS) and heparin (H). Equilibrium binding and thrombin inactivationThe blood clotting proteinase, ␣-thrombin (T), 1 possesses two electropositive sites, exosites I and II (1, 2), that play distinct roles in its covalent inactivation by the serine proteinase inhibitors (serpins), antithrombin (AT), and heparin cofactor II (HCII), in the absence and presence of rate-accelerating glycosaminoglycans (GAGs) (3-8 (9,10). Inactivation by AT is accelerated up to 50,000-fold by high molecular weight heparin, which acts as a template, by binding to thrombin exosite II and AT helices A and D (4, 11). Whereas heparin accelerates thrombin inactivation by both AT and HCII, dermatan sulfate (DS), and other polyanions (5, 12-17) selectively enhance HCII inhibitory action. The HCII mechanism was initially thought to utilize GAGs as templates between exosite II and HCII helix D (5, 12, 13), but later hypothesized to involve allosteric, GAG-induced interaction of an acidic N-terminal domain of HCII with exosite I (6, 7, 18 -21). A synthetic HCII-(54 -75) peptide was shown to compete with hirudin-(54 -65) for thrombin binding (18). Residues 52-75 in HCII contain two hirudin-like repeats, the first of which interacts directly with exosite I (21). The crystal structure of the HCII⅐S195A-thrombin Michaelis complex confirmed this exosite I interaction (22). Deletion of the first HCII acidic repeat (6,19), and mutations in exosite I residues Arg 67 and Arg 73 (20) caused large decreases in the rate constants for the heparin-and DS-catalyzed inactivation, further suggesting the importance of the direct HCII-exosite I interaction. Selective thrombin exosite II mutations did not affect the DS-catalyzed inactivation rate significantly (23,24). Blocking exosite II with HD22, a single-stranded DNA aptamer only had a modest effect on this inactivation rate, and DS with a chain length of 12-20 saccharide units was found to be as effective as higher M r DS in catalyzing thrombin inactivation (25). Whereas some studies report DS binding to thrombin (13,23,26), others postulate that DS does not interact with thrombin (27, 28). These findings, and the assumption that DS template action, like heparin, engages exosite II, led to the proposal of a mechanism in which direct interaction of the HCII N-terminal domain with exosite I is predominant, and GAG template action may be less important, or not even obligatory.In the mechanism of high affinity heparin-catalyzed throm-