pH-dependent association of factor VIII chains: enhancement of affinity at physiological pH by Cu2+

Abstract

Reconstitution of factor VIII from isolated heavy chain (HC) and light chain (LC) shows pH-dependence. In the presence of Ca2+, up to 80% of native factor VIII activity was recovered over a wide range of pH. In contrast, affinity of HC and LC was maximal at pH 6.5-6.75 (Kd approximately 4 nM), whereas a Kd approximately 20 nM was observed at physiological pH (7.25). The effect of Cu2+ (0.5 microM total Cu2+) on maximal activity regenerated was negligible at pH 6.25-8.0. However, this level of Cu2+ increased the inter-chain affinity by approximately 5-fold at pH 7.25. This effect resulted from an approximately 1.5-fold increased association rate constant (k(on)) and an approximately 3-fold reduced dissociation rate constant (k(off)). High affinity (Kd=5.3 fM) of the factor VIII heterodimer for Cu2+ was estimated by increases in cofactor activity. No significant increase in inter-chain affinity was observed when either isolated chain was reacted with Cu2+ followed by addition of the complementary chain. Together, these results suggest that the protonation state of specific residues modulates inter-chain affinity. Furthermore, copper ion contributes to the maintenance of the heterodimer at physiologic pH by a mechanism consistent with bridging the two chains.

Figure 1

Titration of factor VIII chains in the absence (open circles) or presence (closed circles) of Cu²⁺. HC (10 nM) was reacted with 0-300 nM LC at pH 6.0, 6.25, 6.5, 6.75, 7.0, 7.25, 7.5, 7.75, and 8.0 for 6 hours at 23°C in 20 mM MES (pH 6.0-6.5), MOPS (pH 6.75), or HEPES (pH 7.0-8.0) plus 0.3 M KCl, 0.01% Tween 20, 0.01% BSA, 25 mM CaCl₂, pH 7.25. The figure shows results obtained at pH 6.0 (A), 6.5 (B), 7.25 (C), and 8.0 (D). Reconstituted activity was determined by factor Xa generation assay and lines were drawn from the curve fit as described in Materials and Methods. Error bars are drawn from the values of the standard errors of three separate determinations.

Figure 2

A) Dissociation constant (K_d) and B) Activity at chain saturation (V_max) at various pH values. The K_d and V_max for the factor VIII chain association as shown in Figure 1 were estimated by nonlinear least squares regression as described in Materials and Methods and plotted as a function of pH. Error bars are drawn from the values of asymptotic errors of nonlinear least squares regression.

Figure 3

Association kinetics of factor VIII HC and LC. Factor VIII chain association in the absence or presence of Cu²⁺. HC (10 nM) or LC (200 nM) was separately incubated for 4 hours at 23°C in 20 mM HEPES, 0.3 M KCl, 0.01% Tween 20, 0.01% BSA, 25 mM CaCl₂, pH 7.25 in the absence or presence of 0.5 μM Cu²⁺. Reactions were initiated by mixing LC and HC solutions. Aliquots were taken at indicated times and factor VIII activity was measured using the factor Xa generation assay. Curves were drawn from the fitted data according to single exponential as described in Materials and Methods. Error bars are drawn from the values for the standard errors of three separate determinations.

Figure 4

Titration of factor VIII reconstitution by Cu²⁺. HC (10 nM) and LC (15 nM) in 20 mM HEPES, 0.3 M KCl, 0.01% Tween 20, 0.01% BSA, 25 mM CaCl₂, pH 7.25 the presence of 0-2.8 pM free Cu²⁺ was incubated for 6 hrs at 23°C. The indicated free Cu²⁺ concentrations were calculated as described in Materials and Methods. Reconstituted activity was determined by factor Xa generation assay and lines were drawn from the curve fit as described in Materials and Methods. Error bars are drawn from the values for the standard errors of three separate determinations.

Figure 5

Effect of pre-incubation of HC or LC with Cu²⁺. Cu²⁺ (1 μM) was added to 1 μM HC or 1 μM LC in 20 mM HEPES, 0.3 M KCl, 0.01% Tween 20, 0.01% BSA, pH 7.25 for 1 hour, followed by dialysis into 20 mM HEPES, 0.3 M KCl, 0.01% Tween 20, pH 7.25. A) 10 nM HC or Cu²⁺-treated HC was reacted with 0–80 nM LC. B) 10 nM LC or Cu²⁺-treated LC was reacted with 0-80 nM HC. C) 10 nM Cu²⁺-treated HC was reacted with 0-80 nM Cu²⁺-treated HC. Reactions were run for 6 hours at 23°C in 20 mM HEPES, 0.3 M KCl, 0.01% Tween 20, 0.01% BSA, 25 mM CaCl₂, pH 7.25. Reconstituted activity was determined by factor Xa generation and lines were drawn from the curve fit as described in Materials and Methods. Error bars are drawn from the values for the standard errors of three separate determinations. Values for Vmax (nM /min/nM HC) and K_d (nM) respectively were 74.7 ± 4.2 and 16.4 ± 2.7 (HC/LC), 74.6 ± 2.8 and 16.4 ± 1.8 (Cu²⁺-HC/LC), 69.8 ± 2.6 and 15.2 ± 1.6 (LC/HC), 72.4 ± 1.8 and 17.7 ± 1.2 (Cu²⁺-LC/HC), and 75.1 ± 3.0 and 14.7 ± 1.8 (Cu²⁺-HC/Cu²⁺-LC).