Anti tumour agents from natural products - new approaches to targeting ultra potent analogues of the duocarmycins

Molecules that kill cells that divide, known as cytotoxic drugs, are still the backbone of clinically used chemotherapeutic agents against cancer. The problem is that these compounds are very toxic. They kill all dividing cells in the body and so lead to horrendous and debilitating side effects in patients that are already very ill. In the last twenty years, the focus for the design of new agents has switched to two main approaches in order to try to curtail these side effects - either targeting tumour specific pathways or through targeting of cytotoxic agents so that they only exert their effects at the desired site of action. Both approaches have met with some success. In the former case, the antibody called trastuzumab (trade name herceptin) is well known and is used in the treatment of breast cancer. It targets a receptor called Her2, which is highly over-expressed on the surface of some tumour cells and therefore the agent is selective for those tumours. Similarly, the kinase inhibitor imatinib mesylate (gleevec or glivec) targets a protein in chronic myeloid leukaemia and is now used in the treatment of this disease.
These compounds kill cells in a selective way at the protein level. What if we could target ultrapotent drugs at the level of genes, so before any protein has been produced? We have recently developed methodology to rapidly generate analogues of the estremely potent natural products the duocarmycins using solid phase synthesis and are investigating the development of antibody drug conjugates (ADCs) and protein targeting based upon these molecules. However, there are other potential targeting methodologies that we now aim to investigate. This involves the selective targeting of particular genes via the incorporation of duocarmycins into selective molecules that bind to particular sequences of double stranded DNA and have the potential to turn off particular genes. These are called distamycin analogues or "hairpin polyamides". While simply attaching a duocarmycin analogue to a gene selective agent by a linker has been studied, the incorporation of the duocarmycin into the polyamide has only become a realistic proposition with the development of our approach. The ability to turn off particular genes through selective DNA alkylation will be a powerful approach to the development of targeted agents and will help in the identification of new molecules with potential in the treatment of disease, including cancer.

From a non-academic point of view, ultimately the main beneficiaries of this research will be cancer patients. Cancer remains one of the main killers of people in the world and, where surgery and radiotherapy fail, chemotherapy must be brought into play. At the moment, and in spite of many new, more selective drugs on the market, the side effects associated with chemotherapy are extensive and life threatening. Classical chemotherapy is based upon the ability to target cells that are dividing uncontrollably, which gives a small therapeutic window when cells that naturally divide are taken into account and are also destroyed by these agents. More selective agents have come along, such as imatinib and trastuzumab, that target particular pathways but these are restricted to tumours that show the phenotype that is required for the drug to be active - a translocation in the case of imatinib and the over expression of Her2 for trastuzumab. This means we are constantly striving to find new treatments for cancers that are intractable to such drugs and for which novel aberrant pathways have been identified but for which no small molecule exists. Targeting at the DNA level is one way to go about generating such selectivity. If a gene is over-expressed, then we could turn off that gene and potentially stop the growth of the tumour. In parallel, it will be possible to target the agent to the tumour (through, for example, conjugation to an antibody) and this would be even more effective.
The impact will also be in the pharmaceutical industry, who are extremely interested in duocarmycins in themselves as ADCs, but who may be enthused by the notion of new agents, developed and validated in an academic setting, that they can then adopt. Gene targeting is largely limited to methods such as siRNA, miRNA and CRISPR-cas and all of these methodologies have their drawbacks from a therapeutic point of view (and their controversies). The idea of a relatively small molecule that can target particular sequences will be attractive to the pharmaceutical concerns.