Drug and Gene Targeting
Efficient Delivery of Agents to Molecular Targets
Over the past 10-15 years, the ability to identify individual molecular mediators of disease has allowed drugs to be designed and developed to be very specific for their molecular targets. This advanced drug discovery for unique molecular targets of disease has not been fully realized, however, due to limits of current technology to deliver the novel drugs selectively to the diseased tissue or organs. Targeting these very specific agents to the site of disease will improve the therapeutic response for the patient by increasing the potency of the drugs to allow efficacy at much lower dosages. Targeting should, therefore, reduce undesired side effects and improve the cost/benefit ratio for the patient.
The selection of targeting scheme will certainly be based on the disease but may also be unique to the extent of disease and genetic profile of each patient. Introduction of targeting to the clinic is in congruence with the trend to individualize treatment for each patient made possible by advances in diagnostic technology to generate a molecular disease profile on patients. Depending on the patient's molecular disease profile, one targeting scheme or unique combination of schemes may be the optimal therapeutic strategy.
GlycoTech's approach to targeting is to utilize our expertise in carbohydrates to design novel targeting schemes based on lectin-carbohydrate interactions. Targeting based on lectin-carbohydrate interactions is advantageous for several reasons. Lectins and carbohydrates are expressed on cell surfaces and are, therefore, accessible to receive a targeted agent. For some diseases, carbohydrates are uniquely expressed or significantly upregulated on the cell surface allowing a targeted agent to be selectively delivered to the diseased cells. Carbohydrates exhibit fast kinetic on-rates with their lectin receptors so targeting can be achieved in environments of convective mass transport such as in the vasculature. The discovery and use of glycoconjugates and glycomimetics will lead to the development of targeting schemes useful in diagnostic imaging, drug delivery, and gene therapy.
Technology
GlycoTech's expertise in carbohydrates coupled with our extensive carbohydrate, neoglycoconjugate, and carbohydrate antibody libraries allows us to probe cell surfaces to identify native carbohydrate structures and, more specifically, the carbohydrate binding epitope that can be a candidate for targeting. Most of our assays are in vitro assays ranging from well-defined ligand binding assays, which include a high throughput robotic screen, to the more complex biological cell-based assays to mimic the in vivo lectin-carbohydrate interactions . The success with this combination of assays has been demonstrated by our discovery of glycomimetics for E-selectin which are 200-fold more active than the native carbohydrate ligand with a 10% reduction in molecular weight.
Due to our success and technology in the selectin area, we modified our biological flow assays to initiate our research in targeting. This approach to use selectins as the model of lectin-carbohydrate interactions may lead to a viable clinical applications for using selectins in targeting but also will develop the tools and validate the concept of using lectin-carbohydrate recognition for targeting.
The benefit of using biological flow assays to study targeting is to allow direct visualization of the lectin-carbohydrate recognition event of the targeted agent, without artifacts of non-specific protein-protein interactions. Direct visualization is conducted in our video microscopy/ digital imaging laboratory depicted below. Our inverted stage microscope has a large working distance and is equipped with bright field, phase contrast, and epifluorescent optics allowing flexibility in selection of assay format and optical mode to obtain the best visualization of targeting events. Assays dependent on direct visualization to record results can be quite subjective by different individuals, so we utilize digital imaging to standardize and quantify the targeting assays.
The targeting flow assay uses a flow loop we designed to allow continuous circulation of the glycoconjugate in solution over the cellular monolayer. This allows targeting to be observed under dynamic conditions as would occur in vivo in the vascular system. The cell monolayer can be observed, usually under epifluorescence, during the course of the assay to monitor the time dependent effects of the glycoconjugate delivery of agents to the cell surface. The schematic of the flow loop apparatus is shown below.
Development of Targeting
The strategy for development of targeting into a viable technology is to conceptualize targeting as comprised of two events for research investigation. The first event is the delivery and binding of the targeted agent to the cell surface. This event of recognition is relevant to applications of diagnostic imaging and drug delivery schemes in which the site of action of the drug is extracellular. The second event, subsequent to the first, is the effect on the cell of the targeted agent. The efficiency of the second event is critical for drugs which need to be internalized to reach their intracellular targets and in gene therapy for which the gene must be taken up by the cell and expression of the therapeutic protein is required for clinical success.
Targeting formats which should be investigated as delivery vehicles include direct coupling of carbohydrate to the drug, polymers and dendimers linked with carbohydrates, and liposomes. Each one of these delivery vehicles may have unique features which are optimum under different targeting schemes. An example of use of a carbohydrate polymer is seen below when we delivered fluorescent phycobiliprotein linked to a multivalent carbohydrate polymer to the surface of activated endothelial cells in the flow loop apparatus. This is also a visualization of the first event of targeting when the delivered agent (phycobiliprotein) is bound to the cell surface through the carbohydrate polymer and is not internalization by the cells. The images are observations of the cells by fluorescent microscopy.
The next example illustrates the second event of targeting in which the delivered agent has an effect on the endothelial cells. In this case, ricin, a potent toxin isolated from castor beans, binds to galactose-terminated oligosaccharides on the endothelial cell surface and is internalized leading to cell death. The cell death is visualized by use of a vital fluorescent dye which makes the cell fluorescent upon cell death. Digital imaging is used to count the number of dead cells, thus generating a killing kinetic profile for the targeted ricin. Similarly, a reporter gene, such as the gene for b-galactosidase (b-gal) or luciferase, could be targeted to the endothelial cells under flow and monitored for the expression of either protein by direct visualization of blue color (b-gal) or generation of fluorescence (luciferase).
The example below demonstrates the feasibility of delivering an agent, in this case a mutant form of diptheria toxin, to the endothelial cell surface by a carbohydrate, sLea, conjugated to the toxin. The toxin, CRM107 is a mutant form of diptheria toxin in which a point mutation has been introduced in the cell binding site to reduce the toxicity 10,000-fold compared to the wild type toxin. Conjugating the carbohydrate, sLea, which recognizes E-selectin on the endothelial cell surface, to CRM107 restores the toxicity as seen in the results of the assay below. Within one hour, most of the endothelial cells have been killed which is in direct contrast to the unconjugated toxin which does not result in any observed killing within one hour of flow. The use of direct coupling of a carbohydrate to an agent to improve the efficiency of delivery would have application in drug, gene, and cell therapies.
The previous examples illustrate our approach to developing successful targeting schemes based on lectin-carbohydrate recognition for use in diagnostic imaging, drug delivery, and gene therapy. Future research will focus on the selection of appropriate carbohydrate mimetic structures for the delivery vehicle. The site density and 3-D presentation of the carbohydrate mimetic coupled with the overall size of the delivery complex (i. e. drug + delivery vehicle) will require optimization. Optimized targeting schemes will, thereby, be a significant technological advance in pharmalogical formulations for use in drug and gene delivery.
References
Alonsovarona, A., Y. Calle, T. Palomares, B. Castro, and E. Barberaguillem (1995), “A simple cell labeling technique by means of lectins linked to fluorochromes for the detection of cells on tissue sections”, Biology of the Cell 83(1): 87-92.
Harris, J. D. and N. R. Lemoine (1996), “Strategies for targeted gene therapy”, Trends in Genetics 12(10): 400-405.
Jacobson, B. S., D. B. Stolz, and J. E. Schnitzer (1996), “Identification of endothelial cell-surface proteins as targets for diagnosis and treatment of disease”, Nature Medicine 2(4): 482-484.
Jain, R. K. (1996), “1995 Whitaker lecture: Delivery of molecules, particles, and cells to solid tumors”, Annals of Biomedical Engineering 24(4): 457-473.
Kiely, J. M., M. I. Cybulsky, F. W. Luscinskas, and M. A. Gimbrone (1995), “Immunoselective targeting of an anti-thrombin agent to the surface of cytokine-activated vascular endothelial cells”, Arteriosclerosis Thrombosis and Vascular Biology 15(8): 1211-1218.
Schumacher, U., E. Adam, S. A. Brooks, and A. J. Leathem (1995), “Lectin-binding properties of human breast cancer cell lines and human milk with particular reference to Helix pomatia agglutinin”, Journal of Histochemistry & Cytochemistry 43(3): 275-281.
Sujathan, K., S. Kannan, P. Remani, K. R. Pillai, B. Chandralekha, N. S. Amma, and M. K. Nair (1996), “Differential expression of jackfruit-lectin-specific glycoconjugates in metastatic adenocarcinoma and reactive mesothelial cells - A diagnostic aid in effusion cytology”, Journal of Cancer Research and Clinical Oncology 122(7): 433-436.
Yamada, N., Y. S. Chung, K. Maeda, T. Sawada, T. Ikehara, H. Nishino, M. Okuno, and M. Sowa (1995), “Increased expression of sialyl Lewis A and sialyl Lewis X in liver metastases of human colorectal carcinoma”, Invasion & Metastasis 15(3-4): 95-102.