|Home | About | Journals | Submit | Contact Us | Français|
Mutations in components of the Wnt pathway contribute to the aberrant activation of the β‐catenin/T‐cell factor‐4 (TCF‐4) complex and initiate most colorectal cancers.1 Upon its activation, the β‐catenin/TCF‐4 complex induces the expression of target genes such as c‐MYC, cyclin‐D1 and EPHB receptors.2 EphB receptors are receptor tyrosine kinases (RTKs) that play an important role in the coordination of cell proliferation, migration and compartmentalisation along intestinal crypts.3,4 The main effectors of these functions in the intestine are believed to be EphB2 and EphB3 receptors.5 Although EphB receptors are targets of β‐catenin/TCF‐4, Batlle et al. reported the unexpected downregulation of EphB2 receptor expression at the colorectal “adenoma–carcinoma” transition, despite the evident nuclear localisation of β‐catenin.6 Intriguingly, EphB receptors were found to play a suppressive role in colorectal tumorigenesis as their loss promotes tumour progression in mice.6 The authors proposed the existence of a secondary, yet unknown, silencing mechanism that underlies the downregulation of EphB receptors, thus allowing colorectal tumour progression.6 Such a mechanism, we suggest, could be instigated by alterations in the tumour microenvironment. Hypoxia, the state of low oxygen tension in tissues, is a hallmark of the microenvironment in solid tumours.7 Hypoxia is believed to represent a key regulatory factor for tumour growth, survival and aggressiveness.8 The cellular response to hypoxia is orchestrated by a master transcription factor known as hypoxia inducible factor‐1 (HIF‐1). We have recently shown that during hypoxia HIF‐1 competes with TCF‐4 for the direct binding to β‐catenin in colon cancer cells, thereby resulting in the inhibition of β‐catenin/TCF‐4‐mediated transcription with no obvious effect on β‐catenin levels or nuclear localisation.9 Having shown that hypoxia inhibits β‐catenin/TCF‐4 activity in colon cancer cells in vitro and since EphB2 is a target of β‐catenin/TCF‐4; here we hypothesised that hypoxia may contribute to the silencing of EphB2 receptor expression during colorectal tumour progression in vivo.
To test this hypothesis, we initially examined the expression of EphB2 in a panel of normal and cancer tissue from the colon. Consistent with the previous report,6 we found that while EphB2 is expressed at the base of the normal colonic crypt where the Wnt pathway is known to signal (fig 1A1A,, top left panel), the expression of EphB2 was reduced in colorectal carcinoma tissue (fig 1A1A,, bottom left panel). Next, serial sections were stained to assess the expression of the hypoxia‐specific marker carbonic anhydrase‐IX (CA‐IX). To this end, CA‐IX was hardly detectable in the epithelia of normal colonic crypts (fig 1A1A,, top right panel); but was highly, although heterogeneously, expressed in colorectal carcinomas (fig 1A1A,, bottom right panel). Since the occurrence of hypoxia could vary within tumours, depending on the extent of vascularisation, one may expect a heterogeneity in CA‐IX and EphB2 expression. Indeed, in some cancer tissue we observed a patchy staining pattern for both EphB2 and CA‐IX with tumours containing areas that stained largely positive adjacent to areas with downregulated EphB or CA‐IX (fig 1B1B).). Remarkably, in these cases, the staining patterns of EphB2 and CA‐IX were complementary (ie, areas weakly positive for EphB2 had low CA‐IX (fig 1B1B,, red arrows), and regions with strong CA‐IX staining had reduced levels of EphB2 (fig 1B1B,, black arrows)). It is important to note that the heterogeneous staining of EphB2 was also observed by Batlle et al.6 Given that CA‐IX is a hypoxia‐specific marker, these findings suggest a possible role for hypoxia in silencing EphB2 receptor expression. To gain further insight into the potential role of hypoxia in the downregulation of EphB2 receptor expression, we obtained hypoxic large‐bowel tissue from patients with mesenteric ischaemia to examine EphB2 and CA‐IX expression. Interestingly, while the expression of CA‐IX was markedly induced in hypoxic colon tissues indicating a state of hypoxia (fig 1C1C,, bottom right panel), EphB2 receptor expression was lost from the bottom of the hypoxic colonic crypts (fig 1C1C;; top right panel, green arrows), where it is normally expressed (fig 1C1C;; top left panel, green arrow). Taken together, these findings suggest that hypoxia may reduce β‐catenin/TCF‐4 activity in vivo where it results in the downregulation of EphB2 expression from the bottom of hypoxic colonic crypts as well as the hypoxic regions in advanced colorectal carcinomas.
In light of this study, we provide a possible explanation for the loss of EphB receptor expression in advanced colorectal cancers, despite the constitutive accumulation of nuclear β‐catenin. As we have previously demonstrated, the inhibition of TCF‐4 activity during hypoxia occurs as a consequence of direct dynamic competition between TCF‐4 and HIF‐1α for binding to nuclear β‐catenin.9 The current in vivo findings suggest hypoxia as a possible silencing mechanism that may underlie, at least in part, the reduction in EphB2 receptor expression at the advanced stages of colorectal tumorigenesis. Hypoxia is thought to arise during tumorigenesis of solid tissues and it contributes to the tumorigenic process through the promotion of cell survival, stimulation of angiogenesis, and enhancement of invasion and metastasis.8,10 The current investigation highlights the critical role of the microenvironment in transcriptional regulation, whereby an extrinsic fine‐tuning mechanism is instigated to favour those events that particularly promote tumour progression.
This work was funded by a programme grant from Cancer Research UK, a scholarship from the Algerian Government and the Citrina foundation. Tissues were obtained from the archives of the Department of Histopathology, Bristol Royal Infirmary, UK. This was approved by the local research ethics committee.
Competing interests: None.