Defining the role of aberrant O-linked glycosylation in breast cancer

Proteins are encoded by genes contained within the DNA of the cell and the pattern of proteins expressed is known as the proteome. However, the vast majority of proteins found on the cell surface and many found within the cell carry sugars known as glycans. These sugars, and sugars carried on other molecules expressed by the cell are known as the glycome. A single protein can be modified by different glycans, which will affect its function and how it interacts with other components in the same cell and on different cells. Thus the glycome plays a key role in many biological processes and can be more complex then the proteome.
Glycans play a role in many diseases and in cancer the glycans attached to proteins are often different to those attached to the same protein expressed by normal cells. As this can alter the way proteins and therefore cells function, changes in the glycosylation observed in cancer can influence tumour development and progression.
We have been studying one particular type of glycosylation, O-linked, and changes in this type of glycosylation occurs in >90% of breast cancers. We hypothesise that such a common and consistent change plays a role in the development and progression of breast cancer. We have found that the changes in these glycans attached to proteins that occur in breast cancer are due to changes in the expression of enzymes (known as glycosyltransferases) that catalyse the addition of the glycans to the protein.
Two particular glycoproteins known as MUC1 and osteopontin are associated with breast cancer and the glycans carried on these proteins are changed when they are expressed by cancer compared to normal cells. MUC1 is a large glycoprotein which can interact with smaller surface proteins for example EGFR, and affect how these molecules signal to the nucleus and so control gene expression

In this project we aim to establish the mechanism by which a particular glycosyltransferase that is over-expressed in breast cancer can influence the development of mammary cancer. To achieve this aim we will use a murine model developed by us where over-expression of the glycosyltransferase causes mammary tumours to develop earlier than the controls. We will also investigate the role changes in the glycans of MUC1 and osteopontin have on the progression of breast cancer. We will investigate how the O-linked glycans on MUC1 alter its interaction with EGFR and what the consequences are for gene transcription. The O-linked glycans attached to active and inactive osteopontin will be determined. By manipulating the expression of glycosyltransferases we will change the glycans attached to osteopontin and determine the influence on its ability to stimulate metastasis.

The data coming from this project should give an insight into some of the mechanisms controlling cancer development and progression and therefore identify new potential therapeutic targets for the treatment of cancer.

Aberrant O-linked glycosylation is seen in many cancers and in 90% of breast carcinomas. Breast cancers and especially those expressing the estrogen receptor overexpress two sialyltransferases, ST6GalNAc-II and ST3Gal-I. This results in these tumours carrying core 1 rather than the core 2 glycans observed in the normal mammary gland. We will:
1.Determine how over-expression of ST3Gal-I contributes to tumorigenesis. Expressing this sialyltransferase in the mammary gland results in earlier development of tumours when mice transgenic for ST3Gal-I are crossed with a model of mammary cancer (PyMT mice). We will determine if expression of ST3Gal-I in the virgin gland alters their development by microscopically comparing morphology to control glands. Mammary glands from ST3Gal-I mice x PyMT mice will then be examined for morphology, glycan expression and immune infiltration using histochemistry, applying the proximity ligation assay when necessary. Changes in signalling will be analysed by co-immunprecipitations and Western blots looking at pathways activated in PyMT tumours, many of which are also activated in human cancers.
2.Investigate how the O-linked glycosylation of two glycoproteins influences tumour growth. Using a breast cancer line engineered to express MUC1 carrying core 2 or core 1 glycans, we will study the effect of these glycans on its interaction with EGFR after EGF stimulation using immunofluorescence microscopy. The transcriptome of the cells lines carrying MUC1 with core 1 and core 2 glycans will also be established. Osteopontin (OPN) acts as an instigator stimulating growth of tumour cells. We will determine the glycosyltransferase expression (by qRT-PCR) and glycomic profile (by mass spectrometry) of cells expressing OPN that can stimulate cell growth and engineer the glycosylation of another cell type by shRNA and/or transfection of glycosylatransferases to mimic this glycosylation. The activity of OPN will be tested in murine models.

As well as producing new knowledge by understanding the relevance of an important aspect of malignancy, this research project has the potential to contribute to the treatment options offered to cancer patients thereby enhancing their quality of life.
The beneficiaries of this research fall into the following categories:
1. Academics within King's College and Imperial College, and in the international community could benefit from the research. This has been discussed in detail in the academic beneficiary section.
2. In the longer term cancer patients: The aim of the project is to understand how changes in O-linked glycosylation can contribute to the development and progression of breast cancer. This may impact on designing of small molecule inhibitors and antigens to be used as immunogens for treatment of breast cancer. Moreover, if specific glycoforms of OPN are demonstrated to affect its function glycoform specific monoclonal antibodies could be developed to target this glycoform of OPN that may function to inhibit metastasis. Moreover, as aberrant O-linked glycosylation has been observed in many other cancers, the possible target groups goes beyond breast cancer patients.
3. Clinicians may benefit from having additional therapeutic options to recommend to their patients.
4. The pharmaceutical industry: any IP arising from the research could be licensed to Pharma.

Timescale: Academics should benefit from the research being conducting during the 3 years of the project duration. The benefit to patients, clinicians and Pharma is more long term although the development of monoclonal antibodies to osteopontin that are glycoform specific is expected to be possible within the next 4-5years. If these antibodies are shown to influence metastasis in an animal model they will be either licensed to a pharmaceutical company for pre-clinical testing or this may continue "in house".