Selected publications

• Ameri K, Luong R, Zhang H, Powell A A, Montgomery K D, Espinosa I, Bouley D M, Harris A L, and Jeffrey S S (2010) Circulating tumour cells demonstrate an altered response to hypoxia and an aggressive phenotype. Br J Cancer, 102(3):561-9.

• Azam F, Mehta S, and Harris A L (2010) Mechanisms of resistance to antiangiogenesis therapy. Eur J Cancer.

• Buffa F M, Harris A L, West C M, and Miller C J (2010) Large meta-analysis of multiple cancers reveals a common, compact and highly prognostic hypoxia metagene. Br J Cancer, 102(2):428-35.

• Jubb Adrian M, Soilleux Elizabeth J, Turley Helen, Steers Graham, Parker Andrew, Low Irene, Blades Jennifer, Li Ji-Liang, Allen Paul, Leek Russell, Noguera-Troise Irene, Gatter Kevin C, Thurston Gavin, and Harris Adrian L (2010) Expression of vascular notch ligand delta-like 4 and inflammatory markers in breast cancer. Am J Pathol, 176(4):2019-28.

• Catto James WF, Miah Saiful, Owen Helen C, Bryant Helen, Myers Katie, Dudziec Ewa, Larre Stephane, Milo Marta, Rehman Ishtiaq, Rosario Derek J, Di Martino Erica, Knowles Margaret A, Meuth Mark, Harris Adrian L, and Hamdy Freddie C (2009) Distinct microRNA alterations characterize high- and low-grade bladder cancer. Cancer Res, 69(21):8472-81.

Adrian Harris

Our group works on the role of hypoxia in tumour biology and angiogenesis. The laboratory aims to apply its research findings to common clinical cancers, to understand their behaviour and develop new treatments. Particular tumours of interest include: breast cancer, bladder cancer, head and neck cancer and renal cancer. We have demonstrated that high angiogenesis is associated with poor outcome in most tumour types and that pathways regulated by hypoxia, including hypoxia inducible factor 1alpha (HIF-1alpha) and hypoxia inducible factor 2alpha (HIF-2alpha) have key effects on tumour growth. This has led to the development of new treatment approaches for common cancers.

RESEARCH PROGRAMMES

Angiogenesis mediated by notch signalling

We showed that an endothelial specific notch ligand (delta-like 4, Dll4) was regulated by HIF-1alpha and was markedly upregulated in tumour vessels (Fig). In experimental models, upregulation of DLL4 induced tumour growth and enhanced perfusion of tumours. Surprisingly, this was associated with a reduction in the number of blood vessels but increase in their diameter. Thus regulation of notch signalling is likely to be an important new approach to regulating tumour angiogenesis. We will investigate genes induced by delta-like 4 in human endothelium.

Those most upregulated and relating to cell signalling and cell surface proteins will be investigated in in vitro models of 3D vessel growth and zebra fish embryos with fluorescently tagged vessel markers. This will allow us to select genes most critically involved in angiogenesis and development of antibodies or small molecule inhibitors against them. The interactions of this pathway with vascular endothelial growth factor (VEGF) and how inhibition of DLL4 affects responses to chemotherapy and radiotherapy is being studied.

Hypoxic cell death – autophagy and pH regulation

Cancer cells can survive hypoxic conditions that occur during tumour growth. These conditions are exacerbated by antian-giogenesis therapy, so it is important to understand the pathways involved, as they are likely to mediate resistance to such therapy. Autophagy is one mechanism and we have recently found a mechanism upregulating this pathway mediated by ATF4.

Since tumours generate an acidic microenvironment, they need survival mechanisms to cope with this. We showed that hypoxia upregulates an extracellular membrane bound carbonic anhydrase, CA9. This is associated with an aggressive phenotype. We are analysing its role in regulation of pH in 3 dimensional spheroid models of tumour growth, and how it interacts with growth factor pathways.

Many mitochondrial genes are markedly suppressed by hypoxia, but we have found some are induced and critical to cell survival. These will be investigated by knockdown and analysis of interacting proteins. Many metabolism changes are induced and we are investigating which are most important for survival

Anoxic response pathways –genetic screens in vitro and in vivo

More severe or prolonged hypoxia leads to activation of HIF independent pathways via the unfolded protein response (PR). We have shown that ATF4 has a critical role in cell survival under this stress and are analysing the mechanisms responsible. Using siRNA screens we are isolating new signalling pathways interacting with the UPR and ATF4.

New pathways regulated by HIF-1alpha and HIF-2alpha-microRNAs

We have found a population of miRNAs induced by hypoxia, common to many cancer cell lines. Analysing their expression in human tumours indicates they are associated with a poor prognosis. Most are regulated by HIF-1alpha and we are using a combination of bioinformatics approaches and gene array analysis to elucidate their function.

We have used a novel bioinformatics method to describe the in vivo hypoxia transcriptome in tumours and we are using this to classify patients’ tumours for therapy trials and define new hypoxia regulated genes.

Clinical application of novel antiangiogenesis approaches and personalised medicine

By using profile genes induced by hypoxia, a hypoxia metagene, we can classify patients prospectively before entering into trials of anti-angiogenic drugs to relate this to response to therapy and mechanisms of resistance. We are using a strategy of neoadjuvant therapy for patients with window-of-opportunity studies using novel agents blocking angiogenesis metabolism to analyse both the imaging changes on PET scanning or MRI revert gene array and proteomic changes to better classify responders and resistance mechanisms based on baseline profiles and dynamic changes. This should lead to much more cost affective and therapeutically effective management for patients, basing novel targeted agents on baseline profiles and the response to early treatment effects.

Fig:  the human glioblastoma cell line U87 was transfected with the notch ligand Dll4 [Dll4] or empty vector [EV] and grown as a xenograft. The tumour vessels changed markedly with notch signalling, becoming larger, less branched, better perfused and stimulating tumour growth. Blood vessels shown using an antibody to CD31.

Spheroids and role of membrane Carbonic Anhydrase IX [CAIX] in necrosis [N].

Human colon cancer cells are grown in 3D, forming spheroids, which mimic early tumour growth. The centres are hypoxic and we investigated how the rim cells survive and grow. The cells in the centre die [N]. The enzyme CAIX excretes acid from cells, which builds up in hypoxia. So spheroids on the left induce CAIX in cells near the hypoxic centre [black arrow, top left]. This results in a larger necrotic core, but overall growth is maintained by the cells with CAIX, bottom left.

However by knocking down the CAIX in the same cells, there is less acid released and less necrosis, top right. Only a few cells are left that can make CAIX. They cannot grow well and overall the spheroids are smaller and grow more poorly.

Human umbilical vein endothelial cells forming vascular sprouts in 3D growth.

Human endothelial cells are grown attached to beads in a fibrin gel, which supports formation of vascular sprouts and tubes, mimicking early angiogenesis. They can then be used to investigate mechanisms of angiogenesis by genetically modifying the endothelial cells or using drugs.