Mar 2, 2010
Direction of evolution in the bio-therapeutics market diverges with an emphasis on downsizing.
The biotherapeutics market grows larger each year, with no signs of attenuation. In 2008, the top biologics, Enbrel, Remicade, and Rituxan, each recorded over $5 billion in sales. The primary indications for approved protein therapeutics to date have been various inflammatory diseases, such as rheumatoid arthritis and psoriasis, as well as different forms of cancer. Antibodies designed to treat cancer, such as Rituxan and Herceptin, have achieved remarkable sales despite significant limitations to their efficacy. Studies of therapeutic distribution in tumors by molecular imaging and mathematical modeling have shown that these antibodies fail to effectively target the entire tumor, leaving ample room for improvements. While the biotech industry’s response has been focused on the development of antibody-drug conjugates, several academic groups have been working with antibody fragments and alternative scaffolds. These are smaller proteins that achieve comparable affinity and specificity using molecular structures one-tenth the size of full antibodies. Antibody-drug conjugates are antibodies that have been functionalized with cytotoxic agents like chemotherapeutics. They can be more effective than antibodies alone but they fail to address the tumor transport limitations that have hindered efficacy. Furthermore, smaller scaffolds can be developed just as easily as conjugates. As these smaller biologics begin to reach the clinic, it’s clear that the future of protein therapeutics lies in downsizing.
Fibronectins and DARPins are two different scaffolds to which some have pinned their hopes as the ideal small protein therapeutics. Protein scaffolds are structured polypeptides with sets of amino acids available for diversification and selection of different variants with desired therapeutic properties. The tenth Type III domain of human fibronectin is a small part of the cell adhesion protein; it is a particularly suitable scaffold, given its remarkable thermal stability and tolerance of sequence variability. The ability to withstand fluctuations in temperature allows more flexible storage options, and sequence variation is necessary for optimizing therapeutic applications. DARPins are related to the ankyrin family of proteins and use repeated alpha helices to confer structure and mediate binding to therapeutic antigens. These different solutions to the small binding protein problem have each been engineered in different settings using directed evolution techniques that link genotype and phenotype, including phage, yeast, and ribosome display1-4.
Recent work by Peter Milovnik and Davide Ferrari in Andreas Pluckthun’s lab at the University of Zurich serves as a pertinent example of how well-adapted protein engineering technology can be used to generate small binding proteins for unique targets5. The group targets the rat neurotensin receptor 1 (NTR1) using a stabilized, soluble form of the G-protein coupled receptor. NTR1 is implicated in a variety of diseases, including schizophrenia, drug addiction, Parkinson’s disease, and cancer, motivating research for a neurotensin analog that is permeable to the blood-brain barrier and resistant to degradation by proteases. Evolving ligand analogs for NTR1 and other membrane proteins is difficult as soluble protein is generally needed during the selection process, and soluble forms of membrane proteins are often hard to synthesize. To circumvent this complication, NTR1 was subjected to directed evolution to find a stable soluble form6. A DARPin binding NTR1 was selected using ribosome display of a library of DARPins containing two or three ankyrin repeats randomized in the interaction residues, achieving nanomolar affinities for the membrane protein7. Their findings demonstrate the power of protein engineering techniques currently in use and the breadth of their applications. Further, by targeting the less well-known NTR1, they kick the current trend in the biotechnology industry: favoring validated targets for cancer such as epidermal growth factor receptor or carcinoembryonic antigen. Perhaps this merely highlights the differences in flexibility between commercial and academic research. Risk-averse corporations will tend to avoid larger technological leaps that academics might strive for without hesitation. Regardless, the ability to engineer small binding proteins for affinity to tricky domains will make them even more valuable in the therapeutic market.
One of the drawbacks to smaller protein scaffolds is that they are rapidly cleared from circulation, requiring them to be administered more frequently. While they can in some cases avoid proteolytic degradation or thermal denaturation, small proteins are cleared in a matter of minutes, whereas antibodies can circulate for days. It is also possible to design binders with higher affinities so that they are retained in the target tissue for longer periods, even in the absence of a driving plasma concentration. In the Wittrup lab at MIT, fibronectin domains were engineered by yeast display to achieve picomolar affinities8. Hundreds of millions of fibronectin variants were screened to find lysozyme binders. While lysozyme is a much simpler and less therapeutically relevant target than NTR1, this work demonstrates the ability of directed evolution techniques to discover small binding proteins with affinities far greater than those of naturally occurring antibodies.
A decade or so has passed since the early discoveries of the aforementioned small protein binders, and in that time they have been applied to various protein engineering efforts, in some cases leading to commercialization. In 2004, the Swiss company Molecular Partners AG drafted a business plan centered on the clinical development of DARPins. Their plan won first prize in a competition organized by McKinsey & Co. and the Swiss technical university ETH. Molecular Partners has now teamed up with Centocor for pre-clinical work on DARPin technology for two targets of inflammatory disease. Commercial development of the fibronectin domain has progressed even more substantially. Located in Waltham, MA, Adnexus is the owner of numerous patents covering the discovery and development of fibronectin-based therapeutics that they call Adnectins. Their lead candidate is currently in Phase II clinical trials for the treatment of the brain cancer glioblastoma multiforme. Adnexus is now a subsidiary of Bristol-Myers Squibb, following the 2007 purchase that came on the heels of the announcement of positive Phase I data for their leading drug candidate.
As these companies strive to pave the way for moving small binding proteins into the clinic, there still exists a niche for the researcher-entrepreneur. If we follow the sequential development of research from antibodies to antibody fragments to small binding proteins; the next logical step is oligopeptides – peptides containing between two and twenty amino acids. Given the small number of amino acids that substantially contribute to the affinity and specificity of a particular protein-protein interaction, it is not inconceivable that binding might be just as easily achieved in a more minimalist setting. Continuing to shrink the peptide backbone for binding might eventually lead us down the slippery slope to small molecule drugs, where Aileron Therapeutics is an example of one company that has begun to slip. Their “stapled peptide” technology, spawned in the lab of Paramjit Arora at New York University, shows membrane permeability and plasma half-lives more consistent with small-molecule drugs than with protein therapeutics. Very small peptides with certain sequences and structures can obtain lipid membrane permeability, thereby making them almost functionally equivalent to small molecules. A peptide with promiscuous permeability loses specificity beyond its binding target differentiation. At this point, we approach the grey area between biologics and pharmaceuticals. Somewhere in between there ought to exist a polypeptide format that can satisfy our pharmacokinetic and pharmacodynamic dreams.
- Koide, A., J. Wojcik, et al. (2009). Accelerating phage-display library selection by reversible and site-specific biotinylation. Protein Eng Des Sel 22, 685-690.
- Lipovsek, D., S. M. Lippow, et al. (2007). Evolution of an interloop disulfide bond in high-affinity antibody mimics based on fibronectin type III domain and selected by yeast surface display: molecular convergence with single-domain camelid and shark antibodies. J Mol Biol 368, 1024-1041.
- Steiner, D., P. Forrer, et al. (2008). Efficient selection of DARPins with sub-nanomolar affinities using SRP phage display. J Mol Biol 382, 1211-1227.
- Zahnd, C., E. Wyler, et al. (2007). A designed ankyrin repeat protein evolved to picomolar affinity to Her2. J Mol Biol 369, 1015-1028.
- Milovnik, P., D. Ferrari, et al. (2009). Selection and characterization of DARPins specific for the neurotensin receptor 1. Protein Eng Des Sel 22, 357-66.
- Sarkar, C. A., I. Dodevski, et al. (2008). Directed evolution of a G protein-coupled receptor for expression, stability, and binding selectivity. Proc Natl Acad Sci USA 105, 14808-13.
- Hanes, J. and A. Pluckthun (1997). In vitro selection and evolution of functional proteins by using ribosome display. Proc Natl Acad Sci USA 94, 4937-42.