Pharmaceutical compositions comprising bioelastomer fusion proteins

This application is a continuation-in-part of commonly owned, copending application Ser. No. 09/273,025, filed Mar. 19, 1999, the disclosure of which is incorporated by reference herein in its entirety.

The present invention concerns bioelastic polymers and methods of use thereof, particularly for immunoassays, for the production of regenerable biosensors, for the localized delivery of compounds in vivo, and for the patterning of molecules on substrates.

Numerous bioelastic polymers, also known as bioelastomers or elastin-like proteins (“ELPs”) are known. Examples are described in D. Urry et al., A Simple Method for the Purification of a Bioelastic Polymer, PCT Application WO 96/32406. Such compounds are proteins or peptides, typically polypeptides, that exhibit an inverse temperature transition: that is, the compounds condense at a higher temperature range in an aqueous system on raising the temperature of the compounds through their transition temperature (T_I). Bioelastic polymers are soluble in water at a sufficiently low temperature, but hydrophobically fold and associate to form a separate phase as the temperature is raised through a particular temperature range.

Immunoassays are commonly used to detect analytes such as enzymes, hormones, drugs, and other molecules of interest in complex biological mixtures. By definition, an immunoassay relies on the specific binding of an antigen by an antibody, but assays using other biological high affinity binding partners (e.g., ligand-cell surface receptor, inhibitor-enzyme, etc.) can also be employed in an analogous manner. A number of different immunoassay formats have been developed to detect and/or quantitate the levels of analytes; for a review of immunoassays, see PCT Patent Application WO 86/06492.

Ligand-binding proteins such as receptors and antibodies, currently used in biosensors, can detect specific analytes (ligands) with high sensitivity in the presence of potential interference\'s in complex mixtures. The high affinity of protein-analyte interactions is the basis of their exquisite sensitivity. However, high affinity is generally accompanied by an extremely slow dissociation rate (off rate) of the protein-ligand complex. Therefore, in practice, most biosensors are “one shot” devices; dosimeters rather than continuous sensors or alternatively, sensors with very slow response times. In order to use biosensors for the semi-continuous, in situ monitoring of analytes, or for subsequent rounds of sensing in batch mode, the sensor must be regenerated for reuse in an expedient time frame.

There are two possible approaches to sensor regeneration. When the receptor is covalently coupled to the sensor surface, free receptor can be regenerated by displacing the bound analyte. Unfortunately, methods to gently and reversibly regenerate analyte-free receptor do not currently exist: most current methods disrupt noncovalent interactions between analyte and receptor by partially denaturing the receptor using drastic changes in the protein-ligand environment such as low pH (<3), or high chaotrope concentration, conditions which often irreversibly denature the protein after a few rounds of regeneration.

If the receptor is not covalently attached to the substrate, a second method for surface regeneration is feasible where the surface itself can be regenerated by removing the analyte-bound receptor from the surface. The potential advantage is that the analyte “sees” fresh receptor in every round of sensing, which can decrease drift in the sensor response and maintain high affinity and homogenous binding kinetics. This approach to sensor regeneration is difficult to broadly implement because noncovalent methods to immobilize proteins on surfaces typically involves their physical adsorption, which is typically irreversible, and subsequent stripping of adsorbed protein with detergents or chaotropes is frequently incomplete. In order to noncovalently and reversibly bind a receptor to the surface, methods must be found to reversibly control the physico-chemical properties of the receptor such that the adsorption-desorption process can be triggered reversibly.

The targeted delivery of drugs to solid tumors is a complex problem because of the impediments to drug delivery that are posed by tumor heterogeneity. Cancer cells typically occupy less than half of the total tumor volume. Approximately 1-10% is contributed by tumor vasculature, and the rest is occupied by a collagen-rich interstitium. The major impediments to drug delivery arise from heterogeneous distribution of blood vessels, combined with aberrant branching and tortuosity, which results in uneven and slowed blood flow. The leakiness of tumor vessels combined with the absence of a functional lymphatic system results in an elevated interstitial pressure, which retards the convective transport of high MW (>2000 Da) drugs. R. Jain, Sci. Am. 271: 58-65 (1994). The heterogeneity of antigen and receptor expression in tumors is an additional problem in affinity-targeted delivery of drugs to solid tumors.

Front-line therapies for different tumors include surgery, chemotherapy, and radiation. The infiltrative nature of many solid tumors often prevents complete surgical resection because of the high risk of compromising function, thereby necessitating postoperative chemotherapy and/or radiotherapy. However, chemotherapy, particularly when delivered systemically is of limited effectiveness due to inadequate drug delivery, systemic toxicity, and a markedly variable biological sensitivity. External beam irradiation, while useful for many types of tumors, is also limited by dose limiting toxicity to healthy tissue.

Two other treatment modalities that have been suggested for the treatment of solid tumors, are hyperthermia [S. Field and J. Hand, An Introduction to the Practical Aspects of Clinical Hyperthermia (Taylor and Francis, London 1990)] and targeted radiotherapy [C. Hoefnagel., Int. J. Biol. Markers 8: 172 (1993); M. Gaze, Phys. Med. Biol. 41: 1895 (1993)]. The use of local hyperthermia as a therapeutic modality for sold tumors is motivated by the increased thermal sensitivity of tumor vasculature compared to normal vasculature. Hyperthermia, at temperatures between 40 and 42° C., is known to increase tumor blood flow and vascular permeability. Because hyperthermia sensitizes cells to radiation, it has been combined with radiation therapy to increase tumor cytotoxicity [M. Hauck et al., in Handbook of Targeted Delivery of Imaging Agents, pp. 335-361 (V. Torchilin Ed. 1995)].

The limitations of current therapeutic approaches for the management of solid tumors provide a compelling need for the development of improved modalities for the targeted delivery of therapeutics.

A first aspect of the present invention is a method of binding a compound of interest in an aqueous solution. The method comprises the steps of:

(a) providing a conjugate comprising a bioelastic compound and a binding compound, wherein the binding compound specifically binds the compound of interest, and wherein the bioelastic compound has a transition temperature below which the bioelastic compound is soluble in the solution, and above which the bioelastic compound is insoluble in the solution;

(b) contacting the conjugate to the compound of interest in an aqueous solution so that the compound binds thereto, with the contacting step carried out at a temperature below the transition temperature of the bioelastic compound; and then