Targeting Drug Therapy: One Result of the Molecular Revolution

By Kenneth Kaushansky, M.D.
Dr. Kaushansky is currently a Helen M. Ranney Professor and Chair of the Department of Medicine at the University of California, San Diego.

Those of us who are older recall the first day of our medical school pharmacology course, when the professor stated confidently "although nearly every drug we use today was discovered by serendipity, one day, drugs will be rationally designed based on a thorough understanding of the target organ in health and disease." This dream from the past is now rapidly becoming a reality. Rationally designed drugs are available or on the immediate horizon, representing cytokines and their receptors, small signaling molecules and tyrosine kinases, to mention a few. The focus of this commentary will be on some of the approaches used to identify lead compounds on which successful therapeutics are being developed.

Background
Between 1984 and 1994 nearly every major hematopoietic growth factor we now know to exist was cloned, along with their corresponding receptors. Obviously, this group of molecules, including interleukin(IL)-1, IL-2, erythropoietin (EPO), tumor necrosis factor (TNF)a, granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage (GM)-CSF and IL-11 have spurred tremendous clinical activity when used as protein therapeutics or as targets of neutralizing antibodies or decoy receptors. However, as full length proteins they are expensive to produce and require parenteral administration; major efforts to identify small molecule mimics of these cytokines have attracted much attention.

Cytokine Mimics
Two general approaches to identify cytokine mimics have been utilized and have yielded compounds presently undergoing clinical testing in patients with cytopenias. The first is termed peptide phage display, and employs the capacity of bacteriophage to express on the phage surface a small peptide within a larger coat protein. By inserting a random or near-random mix of nucleotide sequences encoding 10-14 amino acid peptides, a "library" of up to 109 unique phage expressing distinct surface proteins are generated, and then used in a binding assay to identify peptides that interact with the cytokine receptor. After several rounds of selection, a mature structure is identified and tested for biological activity. The approach has been successful, yielding an EPO-mimetic peptide and a thrombopoietin (TOP) mimetic, the latter displaying potency close to that of the full length protein. Such substances can be modified by the addition of polyethylene glycol (AF13948) or by linkage to an immunoglobulin scaffold (AMG531) to improve their pharmacology, and have now entered clinical testing.

The second approach to mimic identification relies on the screening of libraries of small organic molecules in a high throughput biological assay of cytokine activity. Typically, a growth factor responsive cell is genetically modified to express the cytokine receptor of interest, and an indicator of receptor activation. Using robotics, libraries of 106 compounds can be screened: when the indicator cell is properly triggered by an individual compound in the library, a readily identified signal is generated, typically based on fluorescence. This approach has led to the identification of several lead cytokine mimetics, including those for G-CSF (SB247464; SSCL02446-8) and TPO (TM41).

Cytokine Antagonists
Another example of taking a thorough understanding of the molecular basis of disease and generating a novel therapy can be found in the development of cytokine antagonists. Acromegaly, due to overproduction of human growth hormone (hGH), has been treated with counter-regulatory substances with some success, but the development of a direct hGH antagonist remains a desirable goal. The tertiary structure of hGH complexed with its receptor marked a milestone in developing an in-depth understanding of the ways by which cytokines trigger their biological effects. Based on such structures the precise amino acids that lead to high affinity binding could be mapped, and by pertubing those interactions, antagonists generated. These efforts have resulted in pegvisomant, amongst other compounds, which has shown promise in patients with acromegaly who have failed surgical or somatostatin analogue approaches. Given the role of many hematopoietic cells (eosinophils, mast cells, monocytes) and cytokines (TNF and IL-1) in pathological processes this approach to targeted cytokine therapeutics (e.g. IL-5, SCF, and M-CSF antagonists) will remain a topic of much interest.

G-protein coupled receptors - the largest family of targets
Clearly the largest family of signaling mediators of health and disease are the G-protein coupled receptors. In the world of the hematologist, potentially useful targets for drug development include platelet receptors for ADP and thrombin, basophil and mast cell receptors for histamine, and neutrophil and monocyte chemokine receptors that mediate many aspects of inflammation. Conventional drug screening methods have yielded numerous useful pharmacological agents for this class of targets, but a newer approach has employed knowledge of the signaling mechanisms of these receptors to generate novel blocking reagents. When thrombin binds to its receptor, an amino terminal peptide is cleaved, revealing a new amino terminus that can now productively interact with the extracellular region of the remainder of the thrombin receptor. Once this occurs one loop of the intracellular domain of the receptor protein becomes competent to bind and activate its corresponding G-proteins, initiating the platelet activation that follows thrombin binding. A new class of compounds, termed pepducins, have been designed based on the structure of the intracellular loop responsible for engagement of G-proteins and appear to specifically interfere with signaling from one or another thrombin receptor.

Tyrosine Kinases as targets
Tyrosine kinases play a vital role in the growth and development of normal and neoplastic cells. The appreciation that a viral oncogene, v-sis, shares homology with a normal cytokine receptor tyrosine kinase by Russell Doolittle just over 20 years ago catalyzed the current explosion of efforts directed at inhibiting the kinases (EGF-R, HER-2/neu, VEGF-r, c-abl) that support neoplastic cell growth. Although conventional drug discovery methods have been applied to target such kinases, and have yielded some spectacular results (e.g. imatinib mesylate), more recent efforts utilize the tertiary structure of the targeted kinases as the substrate for rationale design of antagonists. At present, rationally designed kinase inhibitors targeting breast, colorectal, lung, and pancreatic cancer are undergoing clinical testing.

It is clear from the foregoing examples that our understanding of the molecular basis of cell growth and development has and will continue to garner new approaches to the therapy of human disease. We live in an exciting era of translational medicine, one that is almost certain to hold many novel approaches to the diagnosis, prognosis, and therapy of our patients.

Two strategies to obtain growth factor or cytokine memetics. A. To screen libraries of small molecules for a potential stimulatory hematopoietic growth factor (HGF) mimic, a reporter cell line is first constructed to express the receptor of interest and easily assayed reporter, in the the example, a stat response element (sre) driven luciferase gene. The cells are distributed into a robotic screening array (a 96 well plate is shown) along with individual test reagents. In wells in which a mimic is present, the receptor activates the endogenous JAK-STAT pathway, which then impacts upon luciferase gene expression, readily detected by the robotics. B. To identify peptides that bind to a cytokine receptor, a random oligonucleotide sequences and encoded peptides. A soluble form of the extracelluglar ligand binding domain of the cytokine receptor of interest is immobolized and the phage library allowed to adsorb, and then washed so that only phage that display a modest or higher affinity for binding to the receptor adhere (shown in the figure as the green gene/peptide bearing phage). The adsorbed phage are then eluted by competition using the authentic cytokine (not shown) and sequenced.

Suggested Reading:

1. Cwirla SE, Balasubramanian P, Duffin DJ, Wagstrom CR, Gates CM, Singer SC, Davis AM, Tansik RL, Mattheakis LC, Boytos CM, Schatz PJ, Baccanari DP, Wrighton NC, Barrett RW, Dower WJ. Peptide agonist of the thrombopoietin receptor as potent as the natural hormone. Science; 276: 1696-1699, 1997.
2. Kusano K, Ebara S, Tachibana K, Nishimura T, Sato S, Kuwaki T, Taniyama T. A potential therapeutic role for small nonpeptidyl compounds that mimic human granulocyte colony-stimulating factor. Blood; Sept 25 [Epub ahead of print] 2003.
3. Covic L, Misra M, Badar J, Singh C, Kuliopulos A. Pepducin-based intervention of thrombin-receptor signaling and systemic platelet activation. Nat Med; 8: 1161-1165, 2002.
4. Traxler P, Furet P. Strategies toward the design of novel and selective protein tyrosine kinase inhibitors. Pharmacol Ther; 82: 195-206, 1999.

Return to Table of Contents