Fully Human Anti-Hepcidin Antibodies Mobilize Iron and Treat Inflammatory Anemia in a Mouse Model.

Abstract 626

Iron maldistribution has been associated with multiple diseases including atherosclerosis, diabetes, neurodegenerative disorders and the anemia of inflammation (AI). Hepcidin is a 25aa peptide that is the central mediator of iron homeostasis. It is produced in response to inflammation and is implicated in limiting the response to erythropoietin (EPO) even in the presence of adequate body iron stores. To further explore the role of hepcidin in inflammatory anemia, a novel EPO-resistant model of AI was developed in mice and the effects of a high affinity fully human anti-hepcidin antibody (12B9m; 2pM affinity) were examined.

Since antibody 12B9m had low affinity for mouse hepcidin (10-20nM), human hepcidin (hHepc) knock-in mice were created. Strains were generated with either the mouse hepc1 gene or both hepc1 and 2 genes replaced with the single hHepc gene (HAMP). To induce inflammation and establish AI, mice were injected intraperitoneally with heat-killed Brucella abortus (BA: 3×10⁸ particles/mouse): a procedure adapted from an inflammatory fatigue model. BA treatment led to increased serum levels of IL-6 (3818 ± 548 pg/ml versus <10 pg/ml in controls, p<0.0001) and hepcidin (216 ± 9 ng/ml versus 99 ± 6 ng/ml, p<0.0001) by 6 hours. Hypoferremia reached a nadir 3 days after BA treatment (total serum iron was 201 ±10 mg/dL versus 370 ± 19 mg/dL in controls, p<0.0001). Mice developed anemia, with hemoglobin (Hb) reaching a nadir 10-14 days after BA and returning to normal by day 28. Treatment with EPO (300 mg/kg 8 days after BA) had no effect on hemoglobin by day 14 (7.6 ± 0.4 g/dL in controls and 7.6 ± 0.5 g/dL with EPO; NS).

Treatment with antibody 12B9m (5mg/mouse; n=6) increased serum iron at day 3 (907 ± 47 mg/dL versus 201 ± 10 mg/dL, p<0.0001) in AI mice, and restored response to EPO (day 14 Hb of 10.0 ± 0.4 g/dL versus 7.6 ± 0.5 g/dL for control antibody treatment; p<0.001, n=5). Early iron-dependent red cell parameters such as reticulocyte MCV were also increased (71.5 ± 5.2 fL versus 56.7 ±1.9 fL; p<0.05, n=5/group), presumably due to the increased iron availability. In a larger study, 12B9m treatment alone (without EPO co-administration) led to a significant increase in Hb (10.0 ± 0.5 g/dL versus 8.0 ± 0.3 g/dL for control antibody treatment; p<0.001, n=14/group).

Treatment with 12B9m did not affect inflammatory cytokine induction (e.g., IL-6 levels at 6 hours post-BA of 4996 ± 487 pg/ml versus 3818 ± 548 pg/ml in the control antibody group: NS), indicating that neutralization of hepcidin alone was sufficient to allow an effective response to EPO.

These data indicated that antibodies which neutralize hepcidin may treat AI by increasing serum iron availability for erythropoiesis. The ability to manipulate iron uptake and redistribution in vivo may also offer promise in other diseases where iron maldistribution is involved.