Targeting HOX transcription factors in prostate cancer
© Morgan et al.; licensee BioMed Central Ltd. 2014
Received: 16 October 2013
Accepted: 30 January 2014
Published: 5 February 2014
Skip to main content
© Morgan et al.; licensee BioMed Central Ltd. 2014
Received: 16 October 2013
Accepted: 30 January 2014
Published: 5 February 2014
The HOX genes are a family of transcription factors that help to determine cell and tissue identity during early development, and which are also over-expressed in a number of malignancies where they have been shown to promote cell proliferation and survival. The purpose of this study was to evaluate the expression of HOX genes in prostate cancer and to establish whether prostate cancer cells are sensitive to killing by HXR9, an inhibitor of HOX function.
HOX function was inhibited using the HXR9 peptide. HOX gene expression was assessed by RNA extraction from cells or tissues followed by quantitative PCR, and siRNA was used to block the expression of the HOX target gene, cFos. In vivo modelling involved a mouse flank tumour induced by inoculation with LNCaP cells.
In this study we show that the expression of HOX genes in prostate tumours is greatly increased with respect to normal prostate tissue. Targeting the interaction between HOX proteins and their PBX cofactor induces apoptosis in the prostate cancer derived cell lines PC3, DU145 and LNCaP, through a mechanism that involves a rapid increase in the expression of cFos, an oncogenic transcription factor. Furthermore, disrupting HOX/PBX binding using the HXR9 antagonist blocks the growth of LNCaP tumours in a xenograft model over an extended period.
Many HOX genes are highly over-expressed in prostate cancer, and prostate cancer cells are sensitive to killing by HXR9 both in vitro and in vivo. The HOX genes are therefore a potential therapeutic target in prostate cancer.
Prostate cancer is the most prevalent male malignancy with just under one million new cases worldwide each year . Treatment pathways for this disease are relatively well defined and include surgery, radiotherapy and/or hormonal therapy. While the majority of patients with early stage disease are cured, 10-15% patient still develop locally recurrent or metastatic disease and have a significantly reduced survival rate . Despite the general adoption of docetaxel chemotherapy agents and novel agents such as abiraterone , there is still an urgent need to develop effective new treatments, and therefore it is necessary to explore new target proteins and intracellular signalling pathways.
Recently, considerable interest has been shown in genes that play key roles in defining the identity of cells and tissues in early development and which therefore also have important regulatory roles in cell proliferation and survival. One group of genes that fit into this category are the HOX family of transcription factors . HOX proteins are characterised in part by a highly conserved homeodomain that mediates DNA binding, together with a defined set of co-factors that modify their function including members of PBX family [5–7]. The pro-proliferative and anti-apoptotic roles of some HOX genes in development make them potential oncogenes, and indeed there are numerous reports of HOX overexpression in a range of malignancies, including prostate cancer [4, 8–11]. Although definitive oncogenic roles for some HOX genes have been described, in general studies on the function of individual HOX genes in cancer have been complicated by the high levels of sequence identity and functional redundancy exhibited by most members [12, 13]. This functional redundancy in particular has made the results of conventional knock-down studies (using for example siRNA) hard to interpret. As an alternative approach we developed a peptide, HXR9 that acts as a competitive antagonist of the interaction between HOX proteins and their PBX co-factor. This interaction is mediated by a conserved hexapeptide sequence shared by the majority of HOX proteins, and HXR9 can globally repress HOX function through mimicking this peptide [14–22]. In this study we show that prostate tumours have a highly dysregulated pattern of HOX expression and that HXR9 induces apoptosis in prostate cancer derived cell lines through a mechanism that involves a rapid increase in expression of the cFos gene. Furthermore, HXR9 can block prostate tumour growth in vivo for an extended period, suggesting that HXR9 or its derivatives might represent a possible therapeutic option for locally recurrent prostate cancer.
The cell lines used in this study were DU145 (derived from a prostate carcinoma brain metastasis) , PC3 (derived from a prostate adenocarcinoma bone metastasis) , LNCaP (derived from a prostate carcinoma lymph node metastasis), and WPMY-1 (derived from normal prostate stroma and immortalised with SV40 Large T antigen) . They were obtained from the ATCC through LGC Standards Ltd (UK), and were cultured according to the instructions on the LGC Standards website.
HXR9: WYPWMKKHHRRRRRRRRR (2700.06 Da)
CXR9: WYPAMKKHHRRRRRRRRR (2604.14 Da)
Total RNA from prostate tumours and normal prostate tissue was obtained from OriGene Technologies Ltd, Rockville, USA. Six normal prostate tissue samples (median age of donor 56 years, range 52–71 years), and 17 prostate tumour samples (median age of donor 60 years, range 48–73 years) were included in the analysis. Of the prostate tumour samples, 5 were Gleason grade 6, 8 were Gleason grade 7, 1 was Gleason grade 8, and 3 were Gleason grade 9. Reverse transcription and QPCR were performed as described below.
Total RNA was isolated from cells using the RNeasy Plus Mini Kit (Qiagen) by following the manufacturer’s protocol. The RNA was denatured by heating to 65ºC for 5 minutes. cDNA was synthesized from RNA using the Cloned AMV First Strand Synthesis Kit (Invitrogen) according to the manufacturer’s instructions.
Quantitative PCR was done using the Stratagene MX3005P real-time PCR machine and the Brilliant SYBR Green QPCR Master Mix (Stratagene). Oligonucleotide primers were designed to facilitate the unique amplification of β-actin, c-Fos, and each HOX gene. The expression of each gene was calculated using the ΔΔCt method.
All animal experiments were conducted in accordance with the United Kingdom Co-ordinating Committee on Cancer Research (UKCCCR) guidelines for the Welfare of Animals in Experimental Neoplasia . The experimental protocol was approved by the University of Surrey Animal Welfare Ethical Review Board, and by the UK Home Office (licence number 70/7347).
Athymic nude mice were kept in positive pressure isolators in 12 hour light/dark cycles and food and water were available ad libitum. Mice were inoculated subcutaneously with a suspension of 2.5 × 106 LNCaP cells in culture media (100 μl). Once tumours reached volumes of approximately 100 mm3, mice received an initial, single intratumoural dose of 100 mg/Kg CXR9 or HXR9 dissolved in 0.1 ml PBS, with subsequent dosing when or if the tumour reoccurred. The HXR9 and CXR9 treatment groups contained 9 and 8 mice, respectively. The mice were monitored carefully for signs of distress, including behavioural changes and weight loss.
The cytotoxicity of HXR9 and CXR9 was determined for all three cancer-derived lines and the non-malignant line WPMY-1 using an MTS assay. This revealed that all three of the lines were sensitive to killing by HXR9, whilst WPMY-1 cells were significantly less sensitive (Figure 2e, f).
To further explore whether HXR9 induces cell death primarily through apoptosis, we also used a FACS based analysis for changes in the cell membrane that are characteristic of process and which can be detected by fluorescently labelled Anexxin. The assay also utilises a fluorescent DNA label (7AAD) to measure the membrane integrity of cells, thus allowing cells to be divided into those undergoing early or late stage apoptosis depending on the relative binding of the two labels (Figure 3b). All of the cell lines tested had significantly increased levels of apoptosis after a two hour treatment with 60 μM HXR9, compared to CXR9 treated cells. Apoptosis was considerably higher in the prostate cancer derived cell lines PC3, DU145 and LNCaP (10.6, 8.2 and 7.9 fold, respectively) than in WPMY-1 (3.8 fold).
To provide further confirmation that cells were undergoing apoptosis, HXR9 and CXR9 treated cells were also treated with 50 μM Z-VAD, a caspase inhibitor that blocks the apoptotic cascade. This caused a significant reduction in the proportion of cells undergoing apoptosis (Figure 3b), with the exception of WPMY-1 cells.
In order to establish whether increased cFos levels were directly responsible for inducing cell death, we used a siRNA knock down strategy to reduce cFos expression in HXR9 treated cells. DU145 cells were transfected with a random control siRNA (rnd siRNA), or one of two different siRNAs designed against the cFos sequence (siRNA1/2). Pre-treatment of DU145 cells with either of the cFos siRNAs was sufficient to block the increase in cFos expression upon subsequent treatment with HXR9, both at the mRNA and protein level (Figure 4b). This also resulted in a significant increase in cell survival (from 12% in cells treated with HXR9 only, to 37% and 56% in cells pre-treated with cFos siRNA1 and cFos siRNA2, respectively; Figure 4c).
In this study we have shown that HOX genes are highly deregulated in prostate tumours and in prostate cancer derived cell lines, which concurs with the findings of a number of previous studies [9–11]. It reveals that there is a very high level of deregulation with the majority of HOX genes being highly expressed in tumours but not in normal prostate tissue. This global increase in HOX expression makes it difficult to study those aspects of HOX function that are redundant throughout this highly conserved group. Here we have used HXR9, an inhibitor of the interaction between HOX proteins and their common cofactor, PBX, to target a large subset of HOX proteins (i.e. members of paralogue groups 1-9) . HXR9 causes apoptosis in all three of the prostate derived cells line studied, but only to a far lesser degree in a non-malignant cell line derived from prostate stroma (WPMY-1).
Disruption of HOX/PBX regulated transcription would be expected to cause changes in the expression of numerous target genes, and indeed previous studies have shown this to be the case. However only one of these targets – cFos – has been shown to be directly relevant to the induction of apoptosis by HXR9 . It was previously shown that cFos up-regulation mediated the HXR9-induced apoptosis in melanoma B16F10 cells, and here we show that a similar mechanism exists in the prostate cancer-derived cell lines DU145, PC3 and LNCaP, as siRNA knock-down of cFos can partially rescue each of these cell lines from HXR9 –induced cell death. Although cFos is classically considered to be an oncogene, there are now a number of reports of it acting as a pro-apoptotic gene [19, 28–31]. Our observation that HXR9 results in a rapid and very large increase in cFos expression indicates that the HOX/PBX dimer acts as a repressor of this gene. Whilst this could be a direct result of HOX/PBX binding to its regulatory sequences, a recent study showed that it could also be due to the increased transcription of the oncogenic microRNAs miR221 and mir222, which in turn repress cFos expression .
The prevalence of HOX over expression in prostate cancer combined with the novel therapeutic mechanism exploited by HXR9 suggest that it could be a therapeutic approach where there is small volume, well defined disease. Local delivery of HXR9 into a range of tumours in mice has not resulted in a local inflammatory response [18, 20, 21]. Therefore delivery of HXR9 directly into the restricted confines of a primary or locally recurrent prostate cancer is feasible, and would not be limited due to the risk of prostatitis. Indeed a number of studies have evaluated intraprostatic gene therapy and oncolytic viral therapy and have reported no dose-limiting toxicity. These approaches utilised current imaging technology to achieve the precise delivery of reagents to small volume, well defined disease [32–34]. The application of HXR9 may be as a primary focal therapy, or where standard treatments approaches have failed, for example in cases of local recurrence following radical radiotherapy. The latter group of men currently receive ablative therapy which has low efficacy and is associated with significant toxicities . In contrast to current treatments such as cryotherapy, the lack of inflammatory response associated with HXR9 treatment would potentially allow multiple, sequential intratumoral delivery.
The HOX genes are highly dysregulated, and generally over-expressed in prostate cancer. Targeting the interaction between HOX proteins and their PBX co-factor is a potential therapeutic strategy in this malignancy.
The authors gratefully acknowledge the support of the Prostate Project charity (UK).
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.