research report

University of Bradford

2006/7

Institute of Cancer Therapeutics
Director Professor L.H. Patterson

Selective delivery of anti-tumour agents
Dr R.A. Falconer and Prof L.H. Patterson

Matrix metalloproteinases (MMPs) are a family of zinc-dependent endopeptidases that are associated with the tumour vascular development and metastatic process. The primary role of these enzymes is to degrade components of the extracellular matrix in order to facilitate cell migration and invasion. Our objective is to design tumour selective agents that release a potent therapeutic when activated by MMPs which are over-expressed within the tumour microenvironment. We have successfully designed and synthesised a library of peptide conjugates for proof-of-principle studies. Our synthetic strategy is to derivatise the candidate antitumour agents by attachment to an amino acid, with subsequent immobilisation to the solid support through the amino acid side chain. Conventional solid phase peptide synthesis then follows, with resin cleavage producing the target agents in high yield. We are also currently incorporating design features into these agents to enhance their physicochemical (medicine-like) properties.

In parallel, biological studies are on-going that are further validating the MMPs of interest in clinical samples, and identifying tumour models in vitro and in vivo to enable an efficient screening programme to assist in the selection of a lead agent. In collaboration with Dr Jason Gill and Dr Paul Loadman at the Institute of Cancer Therapeutics we are investigating activation and stability of these novel MMP-activated agents. Results to date demonstrate that certain compounds are indeed selectively cleaved in the tumour to release the active drug. Crucially the results also suggest significant stability in plasma and liver, the main site of drug metabolism and inactivation. We have also successfully verified selective (sequence specific) activation of these agents by the relevant recombinant MMP-2, MMP-9 and MMP-14 enzymes.

Targeting the tumour cell glycocalyx

The surface of all cells is decorated by carbohydrate molecules that are crucial to cellular functions as well as the interaction between a cell and its environment. Altered patterns of this decoration are often associated with cancer and can therefore be used as markers to detect tumours and as the basis to develop therapeutic agents. Furthermore, many of these changes drive the growth and migration of cancer cells and prevention of their occurrence has a beneficial effect on cancer patients. We are devising chemical intervention strategies which are targeting several processes which contribute to the synthesis of these carbohydrate-decorated molecules in tumour cells.

1. Polysialyltransferase as a target for the development of anti-metastatic agents
Dr R.A. Falconer, Dr K. Afarinkia and Prof. L. H. Patterson

Polysialic acid (PSA) is a linear a-2,8-linked polymer of up to 200 sialic acid residues which decorates the cell-surface of several tumours, notably small cell lung cancer. Specifically, it decorates the neural cell adhesion molecule (NCAM), and its synthesis is regulated by polysialyltransferases PST (ST8SiaIV) and STX (ST8SiaII). Changes in the expression of PSA are associated with metastasis. We reason that inhibition of polysialyltransferases will interfere with the production of PSA found on the surface of these cancer cells, thereby altering the surface properties of the cell, preventing release and thus inhibiting the metastastic process. This is of considerable interest to the development of new therapies because it is metastatic disease to which patients succumb.

We are designing and synthesising small-molecule inhibitors of PST and STX. Our compound design is informed by molecular modelling of the sialyltransferase structure (in collaboration with Dr Colin Fishwick, University of Leeds). In collaboration with Prof Paul Smith (University of Cardiff) we are investigating the effects of our agents on cell-surface PSA assembly. With Dr Paul Loadman at the Institute of Cancer Therapeutics and Prof Fukuda, (Burnham Institute, USA) we are developing methods to quantify polysialyltransferase-mediated assembly of PSA to enable high throughput screening of the agents in preparation. Promisingly, two of our hits have demonstrated the ability to reduce PSA decoration in vitro.

2. Golgi mannosidase II inhibitors
Dr. K. Afarinkia

Golgi a-mannosidase II (GMII), a class II retaining glycosyl hydrolase of family 38, is a key enzyme in the biosynthesis of hybrid and complex type N-linked cell surface glycoproteins. GMII catalyzes the trimming of two mannose residues, one which is a-1,6 linked and one which is a-1,3-linked, from Man₅GlcNAc₂(Asn) to give Man₃GlcNAc₂(Asn) which is then further manipulated by various glycosyltransferases in the Golgi complex.

GMII is an important target in cancer because its inhibition can lead to both a slowing down of the cell proliferation, and prevention of metastasis through removal of the signals leading to the breakdown of extracellular matrix. Although a number of natural products have been identified from screens to inhibit these enzymes, none has yet been clinically successful. For example, swainsonine, was withdrawn from Phase I trails due to general cytotoxicity resulting from its target non-specificity. We have rationally designed, and have developed a general synthetic route to, a series of aza-linked carbohydrate mimetics which specifically inhibit a-1,6-mannoside cleavage. We have also developed an in vitro assay for the assessment of our compound libraries and are currently in the process of determining their activity and selectivity.

Cytochrome P450s (CYP's) are a superfamily of mixed function oxidases responsible for metabolising drugs and xenobiotics. Metabolism by the CYP 1-3 family is generally viewed as a route to drug detoxification and increased elimination, but they also have the potential for tumour selective drug activation. There is considerable evidence demonstrating expression of a wide range of CYPs in all the major clinically derived solid tumours and even over-expression of selected isoforms including CYP1A1, CYP1B1 and CYP4W1.

Spirocyclopropacyclohexadienone is the reactive unit of several families of natural products that promote cell death via DNA damage, however, they are not tumour selective. We, in collaboration with Dr Mark Searcey, University of East Anglia, have synthesised two libraries of analogues lacking the hydroxyl group crucial to the initiation of DNA covalent binding. Several classes of agent screens have been developed in collaboration with Dr Paul Loadman and Prof Mike C Bibby to identify regioselective hydroxylation by extrahepatic and tumour expressed CYPs and have identified several agents suitable for further development. A hit compound has been identified that possesses CYP1A1 selectivity in vitro and in cell based screens but with no activation in the liver, the primary organ of drug metabolism. In collaboration with Dr Colin Fishwick, University of Leeds, in silico screen models are being developed to identify novel potential CYP-selective drug candidates. We are also developing xenograft tumour over-expressing CYP systems as models to screen in vivo the most promising hit compounds. Dr Friedberg at Dundee University is collaborating to identify a gene therapy approach using high Kcat mutant CYP variants to the library of agents currently under synthesis.

Tumour hypoxia-selective chemotherapeutics
Dr K. Pors and Professor L. H. Patterson

Solid tumours make up more than 90% of all human cancers and can be considerably less oxygenated compared to normal tissues, a phenomenon that is associated with resistance to radiotherapy and chemotherapy in the clinic. The importance of cell killing in those regions of tumours with low oxygen tension (hypoxia) is crucial if cancer is to be effectively treated. As such we are developing agents that are activated by cytochromes P450 associated with the hypoxic tumour environment. Using this approach we have progressed AQ4N, a tumour hypoxia-selective topoisomerase II inhibitor, into on-going worldwide Phase I/II clinical trials undertaken by Kudos Pharma and Novacea. We have now synthesised a library of N-oxides of alkylating anthraquinones that are rationalised to have high affinity for DNA and DNA processing enzymes, but are completely inactive until reduced in hypoxic tumour cells. Screening data shows several hits that alkylate DNA only in hypoxic regions of tumour spheroids, a model system developed by Dr R.M. Phillips of the Institute of Cancer Therapeutics. Studies in vivo are underway in house and in collaboration with Prof Stephanie McKeown, University of Ulster, to identify a lead molecule for full preclinical evaluation.

In September 2005 a proteomic facility was initiated at the Institute of Cancer Therapeutics under the direction of Dr Chris Sutton and Prof Patterson to complement the established molecular biology, chemistry and pharmacology groups. The Proteomics Facility comprises state-of-the art MALDI TOF-TOF mass spectrometry, nano LC, conventional chromatography, electrophoresis and basic bioinformatics capabilities, all of which were fully installed in April 2006. The main aims of the Facility, regarding ICT programmes, are: (1) the characterisation of target proteins from biological systems, which are used to test new drugs candidates, (2) the identification and characterisation of novel targets, and (3) changes in protein expression and activity (pharmacoproteomics) due to drug administration from preclinical and clinical trials.

A number of collaborations have been established: 1. with Dr. Richard Wheelhouse, Department of Pharmacy, UoB, - Ph. D. studentship - to identify a protein target for a novel compound, 2. with Dr. John Kelly, Bruker - Ph. D. studentship - cytochrome P450 profiling and quantitative proteomics applications, 3. with Dr. Kyriacos Kyriacou, Cyprus Institute of Neurology and Genetics -proteomics screening normal and tumour breast tissues (2 year project) and 4. with Dr. Paul Loadman (ICT) and Dr. Malcolm Clench (Sheffield Hallam University) - MALDI imaging to look at drug localisation in tissues.

On the cytochrome P450 programme, a method for identifying the enzymes by SDS PAGE, trypsin digestion and nanoLC has been established. To date the method has been applied to murine liver, where 26 isoforms have been identified. Future experiments will involve identification of CYPs in tumour tissues and transfected cell lines, incorporation of quantitative labelling to enable comparison of samples, and an enrichment step to detect less abundant isoforms. With Dr. Roger Phillips (ICT), differential profiling of proteins expressed in spheroid monolayer cells and necrotic core, which is a model of hypoxia and represents a source of potential new targets for drug development. A preliminary investigation of matrix metalloproteinases expressed in cell lines has been demonstrated by purification of MMP2 by gelatin-sepharose chromatography and subsequent identification by mass spectrometry.