Chimeric activators: quantitatively designed protein therapeutics and uses thereof

#The invention relates to chimeric proteins, fusion proteins, and therapeutic proteins.

#Many therapeutic proteins such as interferon alpha (IFNα), TNF and IL-2 act locally in their natural state. Systemic administration of these proteins results in side effects that limit the dose levels that can be used therapeutically. For example, IFNα is useful in treating hepatitis and certain cancers, but causes side effects such as flu-like symptoms and suicidal thoughts. There are needs for targeting mechanisms to permit therapeutic proteins to be delivered locally and to minimize the side effects of systemic administration of therapeutic proteins.

#Several laboratories have performed protein-based targeting of toxins to cancer cells. The laboratory of Ira Pastan has been working on cancer-targeted toxins for many years. In 1987, this group described a chimeric protein consisting of Pseudomonas exotoxin and TGFα. This toxin normally binds to the cell surface, is internalized into the endosome, and then somehow enters the cytoplasm where it inactivates Elongation Factor II. TGFα is an EGF-like ligand that binds to EGF receptor. Over the years, the Pastan group has fused variants of this exotoxin to diverse ligands and antibody V regions that bind to cell surface receptors. Some of these have been in Phase I/II-type clinical trials, but none have shown enough efficacy to be developed further. Thus, in addition to the needs for targeting mechanisms to permit therapeutic proteins to be delivered locally and to minimize the side effects of systemic administration of therapeutic proteins, there is also a need to maintain the efficacy of therapeutic proteins while meeting one or both of these two needs.

#FIG. 1 illustrates a non-limiting embodiment of a general structure of a Chimeric Activator (“mut” indicates that the activity element has a mutation);

#FIG. 2 shows, without wishing to be bound by theory, a proposed mechanism of specific binding of “Chimeric Activators” to target cells;

#FIG. 3 shows a molecular model of an IFNα-EGF Chimeric Activator;

#Aspects of the invention relate to chimeric proteins that include a targeting element connected to an activity element. In some embodiments, the targeting element binds to molecules that are selectively present on target cells. In some embodiments, the activity element is a variant of a naturally occurring protein that activates cells by binding to one or more cell surface receptors. The variant is selected such that it has reduced or no cell activating properties in the absence of the targeting element. The targeting element is selected such that it selectively binds to target cells thereby increasing the local concentration of the variant activity element on the target cells to a level that results in activation of those cells. In some embodiments, cell activation results in cell death or a reduction or absence of cell proliferation. However, it should be appreciated that other forms of cell activation may be targeted to specific cells using methods and compositions described herein.

#An activity element may be a receptor binding protein (or functional portion or domain thereof) that binds to one or more naturally-occurring receptors on a cell surface, thereby mediating signaling to the cell (e.g., via signal transduction). A variant activity element may include one or more naturally occurring and/or engineered mutations that result in reduced binding to one or more (e.g., all) natural receptors that are bound by the wild-type activity element. For example, the binding affinity of the variant activity element for one or more of its receptors may be at least 2 fold lower and preferably at least 5 fold lower or 8 fold lower (e.g., at least 10 fold lower, about 10-50 fold lower, about 50-100 fold lower, about 100-150 fold lower, about 150-200 fold lower, or more than 200 fold lower) than the binding affinity of the wild-type activity element for its natural receptor(s). As a result, in some embodiments the activity element by itself is significantly less active (e.g., substantially inactive) because it cannot bind to its receptor(s). Accordingly, chimeric proteins of the invention are useful to avoid unwanted side effects caused by the activity element binding to its natural receptor on non-target cells. However, the activity element is active on target cells because the targeting element provides the missing binding affinity required for activation.

#A targeting element may be a receptor binding protein, an antibody, or any other natural or engineered protein that binds to a molecule that is preferentially present on target cells of interest relative to non-target cells. According to some embodiments of the invention, the affinity of the targeting element for its target molecule (e.g., receptor) may be similar to the binding affinity of the variant activity element for its receptor(s). However, the affinity of the targeting element for its target molecule is typically higher (e.g., at least 2 fold, at least 5 fold, about 5-10 fold, about 10-50 fold, about 50-100 fold, about 100-500 fold, about 500-1,000 fold, at least 1,000 fold, at least 10,000 fold, or at least 100,000 fold higher) than the binding affinity of the variant activity element for its receptor(s).