Reactions used in the Platelet-Plasma model.

Rxn 1. Bach et al.[61] report a decrease in KH from 14.9 to 0.58 nM as % phosphatidylserine (PS) increases from 0 to 40.
O'Brien et al.[62] report a kassoc of 3.14×105 M−1 s−1 and a kdiss of 6.29×10−4 s−1, which yields a Kd of ∼2 nM. With our choice of η, Kd is comparable to this value.
Rxn 2. Shaw et al.[63] report a decrease in Kd from ∼60 pM to ∼10 pM as % PS increases from 10–40. on TF liposomes and a decrease from ∼90 pM to ∼10 pM as % PS increases from 10–70 on TF nanodiscs.
Rxn 6. Shaw et al.[63] also report an ∼20× decrease in Km from ∼400 nM to 20 nM for X activation as % PS increases from 10 to 40. Baugh et al.[64] report an experimental Km of 238 nM on 25% PS vescicles.
Rxn 7. The product inhibition of TF∶VIIa by Xa is dependent on local Xa concentrations. Given that Xa and X binding to a PSPC bilayer increases hyperbolically with Kds of 53.9 and 34.2 nM respectively [65] and that the TF∶VIIa∶X complex is strengthened with increasing PS content (see 6), it is reasonable to assume that the TF∶VIIa∶Xa complex is also strengthened with increasing PS levels.
Rxn 8. Beals et al.[66] report that the Kd for bovine IX binding to lipid surface at optimum [Ca2 + ] decreases from 4.9 to 1.7 µM as % PS increases from 20 to 50. Given that the formation of TF∶VIIa is favoured with increasing PS content (See 2) and analogous to the increased strength of the TF∶VIIa∶X complex with increased PS (see 6), we expect that the formation of TF∶VIIa∶IX will also increase with increasing PS content.
Rxn 11. Jenkins et al.[67] report a decrease in Kd from 351 nM to 4 nM on PCPS vesicles. Neuenschander and Jesty [68] report a Kd of 74 pM on activated platelet surfaces as opposed to 550 pM on equimolar PSPC vesicles. A Kd of 10 pM on the activated platelet was required to fit the shape of the IXa titration in Figure 9B.
Rxn 12. Rawala Sheikh et al.[69] report a decrease in Km from 45µM to 160nM from using unactivated to activated platelets respectively.
Rxn 13. Fay et al.[70] report a Kd value of ∼260 nM for this interaction at pH 7.4 in the absence of phospholipids. Fay et al.[71] report that this interaction is stabilized by the presence of phospholipid. In the Platelet-Plasma model this dissociation constant changes from 270 nM to 2.7 nM as the platelet activates.
Rxn 17. Lindhout et al.[72] report a decrease in Kd from 3.3 nM in solution to 30 pM using 10µM 40% PS.
Rxn 18. Rosing et al.[73] report a decrease in Km from 34.5 to 0.21 µM using 7.5 µM phospholipids.
Rxn 20. Huang et al.[74] report a decrease in Ki from 85.2 to 65.2 pM on using phospholipids.
Rxn 21. Given that the TF∶VIIa∶Xa (see 7) and the Xa∶TFPI (see 20) complexes are strengthened on phospholipids, we expect the stabilization of the TF∶VIIa∶Xa∶TFPI complex with the exposure of phospholipids as the platelet activates.
Baugh et al.[64] report an off rate 3.6×10−4 s−1 for Xa unbinding Xa∶TFPI, and on rates experimentally determined to be 9.0×105 M−1s−1, or numerically estimated to range between 6.8×105 and 1.35×106 M−1s−1. Their data imply that these constants are comparable to those for the binding of TF∶VIIa∶Xa to TFPI (ie a Kd between 2.66×10−10 and 5.29×10−10 M). The original constants for this reaction in the Hockin-Mann model were fitted empirically, but their choice of constants results in a far stronger complex than can be reasonably expected from literature. Hence we have increased Kd by two orders (η = 100) of magnitude from their reported value.
Rxn 28. Experimentally determined by fitting initial velocities of AMC release to standard Michelis-menton kinetics.
Rxn 29.XII activation was coarse grained by assuming a first order dependence on XII concentration and estimating a rate of production (5×10−4s−1) that would resolve the disparity between the Hockin-Mann model prediction and the experimentally observed control.
Rxn 30. Kinetics of XIIa autoactivation (in the presence of negatively charged dextran sulfate) was from Tankersley et al.[75]. Griep et al.[76] showed that the autoactivation (and Kallikrein activation, See 32) of XII is strongly promoted by negatively charged sulfatides.
Rxn 31. Kinetics of Pre-Kallikrein activation by β-XIIa (in the presence of dextran sulfate) was from Tankersley et al.[75]. Pre-Kallikrein activation by XIIa was shown to be facilitated by negatively charged phosphoinositides [77].
Rxn 32. Kinetics of XII activation by Kallikrein (in the presence of dextran sulfate) was from Tankersley et.al[75]. Walsh and Griffin [78] showed that this reaction is sped up by the presence of activated platelets.
Rxn 33. Kinetics of second order Kallikrein autoactivation was from Tans et al.[79].
Rxn 34. The pseudo first order rate constant for the inhibition of Kallikrein in plasma (by C1 inhibitor, α2-macroglobulin and ATIII) was obtained from Van-Der-Graaf et al.[80].
Rxn 35. Hojima et al.[17] report a Ki of 24 nM for the inhibition of XIIa by CTI.
Rxns 36 and 37. Kinetics of XIIa inhibition by C1inhibitor and ATIII were from Pixley et al.[81]. C1 inhibitor is the primary inhibitor of XIIa. ATIII inhibition (although minor) was considering for consistency with other inhibitory reactions.
Rxn 38. Rate constants (in solution) for this reaction are from Gailani et al.[82]. Some controversy exists over the physiological surface for this reaction. Oliver et al.[83] showed that this reaction happens physiologically on the activated platelet surface. However several seminal papers
by Baglia-Walsh et al. in the laboratory of Peter N. Walsh which originally proposed that this mechanism happens on the active platelet have subsequently been retracted. We therefore chose not to include a dependence of this reaction on ε.
Rxn 39. Rate constants (in solution) for this reaction are from Gailani et al.[82]. Walsh and Griffin [78], showed that this reaction is sped up by the presence of activated platelets.
Rxn 40. Several authors describe this mechanism of XI auto-activation (See for example [45], [82]). However, following the retraction of (Baglia et al. JBC 2000) we are not aware of an experimental report of the kinetics of this reaction. Kramoroff et al.[46]estimate the second order rate constant of this reaction to be 3.19 µM−1s−1 by optimizing an ODE model of the intrinsic cascade to experimental measurements of APTT. They consider either XI autoactivation or XI activation by thrombin (but not both possibilities) as plausible mechanisms for XI activation (in addition to activation by XIIa), thus their estimated value is likely an overestimate. We utilized a value 4 fold lower than the value they report for this constant, since we consider thrombin activation of XI in addition to autoactivation. This was in keeping with the experimental titration of XIa (Figure 9C) where we have noticed strong sensitivity to even minute amounts of XIa.
Rxns 41–44. Rate constants are for inhibition of XIa in plasma are from Wuillemein et al.[84].
Rxn 45. Rate constants (in solution) for this reaction are from Walsh et al.[85]. Gailani et al.[86] propose a mechanism by which this reaction could happen on the platelet surface facilitated by the dimeric form of factor XI.
Rxn 46. Rawala - Sheikh et al.[69] report a reduction in Km from 45 µM to 390 nM from unactivated to thrombin activated platelets. In later publications from the same lab, Scandura and Walsh [87] report a Km of 16nM and a kcat of 5.1×10−4 for the activation of X by IXa alone on SFLLRN activated platelets in a model where platelet bound IXa interacts with zymogen X, and Wilkinson et al.[88] report a Km of 6.4 nM and a kcat of 7.0×10−4.
Rxn 47. Rate constants were obtained from Leipold et al.[89] using catalytic efficiencies reported in Lollar et al.[90]. Activation of VIII by Xa, unlike activation by thrombin (reaction 10) was reported to be markedly dependent on the presence of either phospholipid or active platelet surface [68].
Rxn 48. Rate constants for this reaction were from Komiyama et al.[91]. Unlike the activation of X by VIIa alone (see 49) the Km for this reaction was reported to be relatively constant over a wide range of added PCPS concentrations, thus unlike other unbinding reactions in the model there was no dynamic change in reaction rate with platelet activation.
Rxn 49. Rate constants for this reaction were from Komiyama et al.[91]. The authors report a decrease in Km from 1.48 to 0.25 µM with PCPS levels increasing from 1.4 to 21 µM.
Rxns 50–57. Kinetics of fibrin polymerization are taken from Naski et al.[47].
Reactions 1–27 comprise the original Hockin-Mann Model.
For reactions 1–27, parameter values were from the Hockin-Mann model when a reference is not cited.
Except for reaction 13, ε wherever applicable operates on the off rate usually defined as k−1. For reaction 13, ε operates on k1 which is the actual unbinding rate of the VIIIa complex. The notation for this reaction is kept consistent with its description in the Hockin-Mann Model.
On-Rates were assumed to be diffusion limited (with a k1 of 1.0×108 M−1 s−1) [20], and the corresponding off-rate was calculated from Km using .