Background
Colorectal cancer (CRC) is the third most common cancer worldwide [
1]. The majority of CRC related deaths are associated with liver metastasis [
2,
3]. Surgical removal of colorectal liver metastases (CRLM) supported by systemic chemotherapy provides the best possibility for cure in a percentage of patients and even among these, 30% - 60% will develop tumor recurrence in the liver or in other organs [
4,
5]. Recent studies have demonstrated involvement of the renin angiotensin system (RAS) in cancer progression, including CRLM [
6,
7]. There is also evidence of crosstalk between the Kallikrein Kinin System KKS and RAS pathways. The angiotensin-converting enzyme (ACE) catalyses both the production of angiotensin II and the degradation of bradykinin, suggesting a cross-regulation between the two systems [
8], however the effects of the Kallikrein Kinin System (KKS) on CRLM have not been as well studied.
The role of kinin receptors in cancer has been investigated [
9‐
11]. Many cancer tissues display higher expression of B1R and/or B2R compared to their normal counterparts [
9]; ‘kinin’ in humans and most mammal contexts refers to bradykinin (BK). Signalling by BK through either B1R, B2R or both can promote angiogenesis in different experimental models by promoting vascular cell proliferation and survival, and by increasing vascular permeability [
12,
13]. Selective antagonists of B1R or B2R have shown anti-proliferative, anti-inflammatory, anti-angiogenic and anti-migratory properties [
9,
12,
14]. Kinin receptors have been reported in mouse and human colon carcinoma cell lines [
15,
16]. BK treatment of the CRC cell line SW-480 results in mitogenic activation [
16]. Zelawski et al. found a higher B1R expression in human tubular adenomas, a benign tumor that can become colonic carcinoma, suggesting that kinins may contribute to cellular transformation [
17].
While the literature support the role of the Kallikrein Kinin System (KKS) signalling in tumor aggressiveness and progression, additional studies are needed in relevant animal models that closely represent the clinical tumor progression. We have developed and characterised a mouse model of CRLM in a fully immunocompetent host which we used in the present study to investigate the effects of kinin receptor inhibition on the CRLM tumors. Additionally we investigated the direct effects of KKS inhibitors on cultured CRC cell proliferation and migration.
Discussion
The KKS is a complex multifunctional signalling cascade and its role in cancer remains unclear. Early reports on the likely role of the KKS in cancer suggested enhanced vascular permeability and upregulation of nitric oxide synthase and prostaglandin [
23,
24]. This led to the suggestion that KKS family members may be novel biomarkers for cancer [
25‐
28]. More recent studies further support the role of the KKS in cancer development [
7,
29] and a number of KKS generated proteins confirmed to exhibit pro-inflammatory properties [
30]. Chronic inflammation is tightly associated with cancer development and progression [
31].
In this study B1R or B2R blockade led to reduction in tumor progression compared to untreated controls, however this reduction did not reach significance for B2R blockade. Blockade of B1R resulted in significant viable tumor reduction and significant increase in apoptosis suggesting that the reduction in tumor progression is effected, at least in part, through tumor cell apoptosis. In contrast there was no reduction in the percentage of viable tumor or increase in apoptosis associated with B2R blockade. It has been shown that B1R and B2R can activate several intracellular signalling pathways including NF-κB [
32,
33]. Indeed the p53 and NF-κB pathways are, arguably, the two major cellular stress response pathways leading to pro-apoptosis or anti-apoptosis pathways respectively. Webster et al. demonstrated that during injury or stress there is crosstalk between NF-κB and p53; they showed that these two transcription factors can modulate each other’s functions depending on their relative levels present. Following injury, the ability of NF-kB to supress p53 can determine the cell fate [
34]. Although both B1R and B2R have been shown to induce NF-kB activation; the patterns of signalling are different in terms of duration and intensity [
35]. It is possible that B1R activation of NF-kB leads to anti-apoptotic properties, therefore inhibition of B1R demonstrated a significant increase in tumor apoptosis. Another possible explanation for the reduced tumor burden following B1R blockade could be due to a reduction in chronic inflammation normally seen in cancer as a result of increased expression and signalling of the B1R. In contrast B2R is constitutively expressed and hence assumed to be mostly responsible for effecting normal physiological functions.
Our results show that blockade of either the B1R or the B2R resulted in significant reduction in tumor vascular density and this may be one of the mechanisms by which tumor progression is retarded. B1R and B2R activation have been implicated in angiogenesis and neovascularization in other studies. Kinins induce EC proliferation in cell cultures through the B1-cAMP pathway. B1R stimulation has been shown to induce neovascularization in the rabbit cornea [
36], while B2R stimulation can activate the mitogen-activated protein kinase pathways (MAPK) and P13K/AKT, contributing to proliferation and angiogenesis [
37‐
39]. Other studies have shown tumor suppression and angiogenic inhibition following treatment with B2R inhibition [
40] or in kininogen knock out mice [
12]. Moreover, Morbidelli et al. demonstrated the ability of BK to stimulate EC proliferation via activation of B1R or indirect activation of B2R, further supporting a role of kinin receptors in angiogenesis and tumor development [
41].
In agreement with Wang et al. [
22] we also detected B1R and B2R expression in both MoCR and SW480 colon cancer cells lines suggesting that receptor blockade may also have direct effects on the tumor cells. There are several studies which suggest that activation of B1R and B2R leads to stimulation of tumor cell proliferation and migration [
37,
42,
43]. Unlike Barki-Harrington et al. who found increased proliferation following stimulation of B1R and B2R [
42], our in vitro study failed to show increased proliferation with B1R agonist (desArg9-Bradykinin (DABK)), however, the involvement of B1R in proliferation cannot be completely ruled out, since treatment with a B1R antagonist (SSR240612) demonstrated significant reduction in tumor proliferation. McLean et al. found that the synthesis of DABK is significantly upregulated during inflammation [
44]; it is possible that MoCR cells may have maximal DABK secretion therefore external addition of the DABK agonist does not further increase cell proliferation. Although DABK failed to increase tumor proliferation, our current study did find that BK treatment (B2R agonist) resulted in increased cell proliferation and the use of a B2R antagonist (FR173657) significantly decreased proliferation, further demonstrating that kinins do play a role in tumor proliferation. Further stimulation of kinin receptors was shown to increase tumor cell migration. Wang et al. reported that treatment with BK stimulated B2R and ERK1/2 leading to increased IL-6 production, ultimately increasing the invasiveness of colorectal cancer cells [
22]., Ehrenfeld et al. found that activation of the B1R increased secretion of the metalloproteases (MMPs)-2 and − 9 by breast cancer cells [
11]. Previous studies have shown that MMP2 activates integrin αvβ3 resulting in cellular invasion [
45]; similarly MMP-9 has been shown to enhance metastatic capacity through activation of αvβ3 in breast cancer cell lines [
46] and αvβ6 integrin activation has been shown to promote invasion of squamous cell carcinoma cells [
45]. In agreement, using invasion assays, we observed significant decrease in tumor invasion following treatment with both B1R and B2R antagonists, further confirming a role of bradykinin receptors in CRLM migration.