It is well established that mice which do not produce endogenous factor VIII (fVIII−/−) can manifest a robust immune response to exogenous fVIII treatments. They form B-cell and T-cell responses even when they encounter fVIII through traditionally tolerogenic routes (e.g., intravenous or intraperitoneal). In the fVIII−/− mouse, repeated administration of recombinant human fVIII has emerged as a useful model for studying the physiologic response in hemophilic patients iatrogenically immunized to therapeutic factor VIII treatments. While environmental factors likely offer some co-stimulatory signals, nonetheless, the ability to respond effectively in the absence of extrinsic adjuvant begs the questions of what is the “danger signal” required for immune responsiveness to fVIII? We have previously shown that when factor VIII is heat inactivated (56°, 30′), it completely losses function and much of its immunogenicity (

Skupsky and Scott,
). Heated fVIII lacks several of its B-cell epitopes (we did not find a subsequent response to neo-epitopes), but retains its T-cell epitopes. We concluded that fVIII’s immunogenicity is inherently tied to its function. To explore this topic further, we have immunized hemophilic mice with rfVIII and compared the response to mice treated with both rfVIII and Hirudin. Hirudin is the reactive agent found in medicinal leech saliva and its anti-coagulant activity is based on its ability to inhibit thrombin. We found that T cell responses to rfVIII in mice protected with Hirudin are significantly reduced (p<0.05) and the anti-fVIII antibody concentration has decreased by 25%. As a control, we injected a third group of mice i.v. with an equivalent amount of another foreign protein, ovalbumin (OVA) in PBS. As expected, the mice did not respond to this historically tolerogenic treatment. Interestingly, when mice were injected simultaneously with rfVIII and OVA, they did form a humoral response to both the fVIII (200 μg/ml) and the OVA (30μg/ml). This suggests that fVIII may have adjuvant properties remaining to be discovered. Overall, these data suggest that the activation of thrombin provides co-stimulatory signals necessary for the immune response. Activated thrombin does this directly or indirectly through the activation of other blood components, including platelets.

Disclosures: No relevant conflicts of interest to declare.

(Supported by NIH RO1 HL061883, NIH T32 HL007698, and a Predoctoral fellowship from the American Heart Association.)

Author notes

Corresponding author