Red blood cell engineering.

Red blood cells as vehicles for the introduction of novel therapeutics, immunomodulatory agents, and diagnostic imaging probes into the human body.

Human red cells have a lifespan of 120 days and contain no DNA; any genes introduced into red cell precursors will no longer be present in the enucleated red cells introduced into a recipient. A large DARPA- supported project, in collaboration with Prof. Hidde Ploegh, involves the generation in culture of both murine and human red blood cells that have on their surface monoclonal antibodies that inactivate a variety of toxic substances, or receptors that can bind and remove unwanted materials from the blood. New techniques allow the covalent linkage of 5000 – 10,000 molecules of any desired molecule to mature mouse or human erythrocytes. Red cells that have on their surface any of several covalently linked foreign or proteins induce immune tolerance and T cell anergy rather than an immune reaction, offering promise for novel treatments of several autoimmune disorders.

Red blood cells possess many unique characteristics that make them attractive candidates for in vivo delivery of natural and synthetic payloads. They have a long circulatory half-life (~120 days in humans and ~50 days in mice), and old or damaged RBCs are removed and degraded by cells of the reticuloendothelial system. They are biocompatible and have a large surface area of ~ 140 µm2 with a favorable surface to volume ratio. Importantly, they contain no DNA; any genes introduced into red cell precursors will no longer be present in the enucleated red cells introduced into a recipient.

A large DARPA- supported project, in collaboration with Prof. Hidde Ploegh, involves the generation in culture of both murine and human red blood cells that have on their surface monoclonal antibodies that inactivate a variety of toxic substances, or receptors that can bind and remove unwanted materials from the blood. Gene- modified red cells can be targeted to specific sites in the vasculature where they can deliver drugs or reagents or serve as imaging modalities.

Engineered red blood cells that bind and neutralize toxic proteins

Single domain antibodies (VHH) are the antigen- binding domains of the unique functional heavy- chain- only camelid antibodies. These VHHs have equivalent binding activities to their cognate antigens compared to conventional IgGs and these VHHs are more stable and easier to make. VHHs that can target botulinum neurotoxin (BoNT) have been generated and studied intensively by our collaborator Dr. Charles B. Shoemaker of Tufts University Veterinary School; these can neutralize botulinum neurotoxins in cell culture and in vivo.

Nai-Jia Huang and Nova Pishesha generated genes encoding four chimeric proteins, fusing the cDNAs encoding a heterodimeric VHH for BoNT/A or the VHH for BoNT/B, with either the N- terminus of glycophorin A or the C- terminus of Kell. Both types of red cells expressing the VHH BoNT/A chimeras effectively neutralized BoNT/A in an in vitro nerve- protection assay; in this assay primary neuronal cells are incubated with the toxin and the cleavage of the SNAP-25 protein by the toxin is quantified. They calculated that it should be possible to protect mice from a 1 LD50 toxin dose by administering as few as 2000 engineered red cell – VNA conjugates.

More recently they generated and tested the in vitro neutralization ability of VHH BoNT/A -engineered human red blood cells. As few as 5 x106 human red cells expressing GPA-VNA/A and 106 cells expressing Kell-VNA/A effectively neutralized the botulinum toxin.

In collaboration with the Shoemaker laboratory, Nai-Jia and Nova transplanted murine hematopoietic stem/progenitor cells expressing the VHH BoNT/A glycophorin and the Kell - VHH BoNT/A chimeric proteins into irradiated CD1 mice to produce normal red blood cells bearing these chimeras. These mice, together with control transplanted mice, were challenged with BoNT/A. All control mice died after challenge with 10 LD50 of BoNT/A. In contrast, all mice bearing red cells expressing either GPA-VNA/A or Kell-VNA/A survived successive challenges, at one - week intervals, of up to 1,000 LD50 doses of BoNT/A.

Nai-Jia and Nova next transfused naïve mice with blood expressing GPA- VHH BoNT/A chimeric proteins and found these mice can be protected from 10 LD50 of BoNT/A for up to 28 days. They also showed that NOD/SCID mice transfused with human red cells made in an in vitro culture and expressing the GPA-VNA/A chimeric protein can be protected from a 10LD50 BoNT/A challenge. This work demonstrates that engineered human RBCs expressing VHHs can provide prolonged prophylactic protection in vivo against bacterial toxins, and illustrates the potentially broad translatability of our engineered RBC strategies for an array of human applications. In particular, we are beginning a collaboration with scientists at the United States Army Medical Research Institute of Chemical Defense (USAMRICD), the nation's leading science and technology laboratory, to develop engineered human red cells that can neutralize certain phosphorus- containing nerve gasses.

Engineered red blood cells that induce immune tolerance

Nova Pishesha, together with Rhogerry Dhesycka, Harun Sugito, and Valentino Sudaryo, in a collaborative project with Hidde Ploegh’s laboratory, is currently exploring the possibility of applying the aforementioned methods to generate red cells that have on their surface any of several covalently linked foreign or proteins. They aim to use these red cells to induce immune tolerance rather than an immune reaction, offering promise for novel treatments of several autoimmune disorders.

Their current work has used the transpeptidase sortase to covalently attach several peptides to both genetically engineered and unmodified red blood cells (RBCs). Transfusion of RBCs expressing these antigens can induce antigen-specific tolerance; this approach blunts the contribution to immunity by major subsets of immune effector cells (B cells, CD4+ and CD8+ T cells) in an antigen-specific manner. Building on the encouraging results of this T cell adoptive transfer model, they have adapted the system to inhibit the development of experimental autoimmune encephalomyelitis, a mouse model of multiple sclerosis. Additionally, modified red blood cells can maintain normoglycemia in a mouse model of Type 1 diabetes. They are currently expanding the applications of this approach into additional immune-mediated disease models.