by ecancer reporter Clare Sansom

A key feature of the IIAR conference series is the large number of specialist symposia held during the meetings, organised by individuals and interest groups.

One such symposium held on during the ninth meeting on Tuesday 7 October 2014, concerned the relationship between homeotic or HOX genes and cancer.

This was organised by Richard Morgan and Claire Aukim-Hastie of the University of Surrey, Guildford, UK.

Morgan began the symposium by giving a brief introduction to the function of the HOX genes, which encode transcription factors, and their role in cancer development.

His first slide featured an image known to all advanced students of biology: an Antennapedia fly mutant showing a pair of legs growing in the place of antennae: this mutation is caused by a single mutation in one such gene.

This illustrates the most well-known function of these genes: their involvement in defining the roles and identities of cells and tissues during development. However, they are also involved in regulating cell proliferation and survival, which are among the “hallmarks” of cancer.

Humans have a total of 39 HOX genes, classified into four groups, and many of these have been implicated in the development of one or more types of cancer.

Morgan’s own talk described some of his recent work with HOX genes as targets and biomarkers in malignant mesothelioma, an asbestos-related tumour that most often affects the outer lining of the lungs.

Some of the HOX genes in the human genome seem to function largely as oncogenes, others as tumour suppressors, and the role of others is still unclear.

Morgan’s work has focused on the interaction between these genes and the PBX proteins, which act as co-factors; blocking the dimerisation of a HOX protein and its co-factor prevents its activity.

His group have designed a six-residue peptide now named HXR9 that disrupts the interaction between HOX and PBX proteins and prevents HOX activity.

A second peptide, named CXR9 (the control peptide), differs from the first at one amino acid position only but does not prevent dimerisation.

Mesothelioma cells express high levels of two HOX genes, HOXA4 and HOXA5; when HXR9 is added to block HOX activity, mesothelioma cells but not normal mesothelial cells undergo apoptosis.

Keith Hunter from the University of Sheffield described the role of HOX genes in head and neck squamous cell carcinoma (HNSCC).

About 600,000 people worldwide are diagnosed with this tumour each year, and it causes 350,000 deaths.

The molecular basis of the disease is very heterogeneous, with many genes implicated in at least some cases, but only p53 is almost universally mutated and only one targeted therapy, an EFGR inhibitor, is available for these patients.

Hunter and his colleagues identified differentially expressed genes from a panel of HOX and HOX-related genes in cell cultures from normal oral keratinocytes, pre-malignant tissue, primary HNSCC tissue and metastases.

Several of these genes were differentially expressed between tumour and normal tissue, and two, HOXB9 and HOXD10, were highly expressed in pre-malignant and primary tumour tissue but not in metastases or normal cells.

These two transcription factors appear to promote cell proliferation and migration, but decrease invasion.

Genes directly regulated by HOXD10 include AMOTp80 and a micro-RNA, miR146a, and knockdown of these genes inhibits the proliferative phenotype in HNSCC cells.

Hunter concluded by suggesting that the protein AMOTp80 might be a valid therapeutic target for this tumour type.

Manuel Penichet from the University of California Los Angeles, CA, USA, described his group’s analysis of the expression and role of HOX genes in multiple myeloma.

This cancer of monoclonal plasma cells in the bone marrow is currently responsible for about 75,000 deaths a year worldwide.

Penichet and his colleagues tested the expression levels of HOX genes and the cytotoxic effect of the “anti-HOX” peptide HXR9 in malignant B-cell lines including four multiple myeloma lines; one of these, U266, is a “bullet-proof” myeloma line that is particularly resistant to apoptosis.

Several HOX genes were differentially expressed in one or more of these cell lines, and HXR9 induced apoptosis in all lines, indicating that they were all dependent in some way on HOX expression.

An antibody-avidin fusion protein, ch128.1.Av, which is specific for the human transferrin receptor 1 (CD71) is also toxic to malignant B cells as it deprives them of iron.

The cytotoxic effect of HXR9 in myeloma cells was found to be enhanced if it was combined with this protein, suggesting that both HOX proteins and the transferrin receptor might be valid drug targets for multiple myeloma.

Acute myeloid leukaemia (AML) is a genetically heterogeneous disease, and its treatment, which consists largely of standard chemotherapy drugs, has scarcely changed in the last 30 years.

Altered expression of HOX genes is, however, a hallmark of this disease, and Alexander Thompson of Queen’s University Belfast, Northern Ireland, UK, described the development of a HOX based screen for candidate drugs for this condition.

Many cytogenetically normal AML cells – that is, cells with no chromosomal translocations – express high levels of genes in the HOXA cluster, and patients with this type of AML and low expression of HOXA genes tend to have a better prognosis than similar patients with high HOXA expression.

Acute myeloid leukaemia can be established in a mouse model by over-expressing HOXA9 and another homeobox protein, Meis1.

A methylcellulose screen of this mouse model and cell lines taken from 351 cytogenetically normal AML patients, plus mouse and human controls, identified several small molecules already registered by the FDA that were able to change the HOX gene expression profile of leukaemia cells close to that of normal myeloid cells.

Thompson described entinostat, a histone deacetylase inhibitor that is undergoing clinical trials for the treatment of Hodgkin’s lymphoma, breast and lung cancer, as the “most interesting” of these and suggested that direct targeting of over-expressed HOX proteins might be an appropriate strategy for treating AML and other blood cancers.

Other talks in the session described the role of HOX genes in breast and ovarian cancer and malignant melanoma, and discussed their role in pluripotent stem cell differentiation.