Altered efficacy of AT1R-targeted treatment after spontaneous cancer cell-AT1R upregulation
© Ager et al; licensee BioMed Central Ltd. 2011
Received: 6 December 2010
Accepted: 26 June 2011
Published: 26 June 2011
Skip to main content
© Ager et al; licensee BioMed Central Ltd. 2011
Received: 6 December 2010
Accepted: 26 June 2011
Published: 26 June 2011
Targeting of the renin angiotensin system (RAS) reduces tumour growth in experimental models of cancer. We aimed to establish if combined targeting of the 'classical' and 'alternative' arms of the RAS could result in synergistic inhibition of colorectal cancer (CRC) liver metastases.
Immediately following induction of CRC liver metastases through intrasplenic injection of murine CRC cells, treatment with irbesartan (AT1R blocker; 50 mg/kg/day s.c.), captopril (ACE inhibitor; 750 mg/kg/day i.p.), CGP42112A (AT2R agonist; 0.6 μg/kg/hr i.p.), and/or ANG-(1-7) (24 μg/kg/hr i.p.) began and continued for 21 days. Liver to body weight ratio and/or stereology were used as a measure of tumour burden. Immunohistochemistry was used to determine AT1R and VEGF expression as well as proliferation (Ki67), apoptosis (active caspase 3) and angiogenesis (CD34).
Combined RAS therapies failed to improve upon single arm therapies. However, while irbesartan previously inhibited tumour growth in this model, in the current experiments irbesartan failed to affect tumour burden. Subsequent analysis showed a cancer-cell specific upregulation of the angiotensin II type I receptor (AT1R) in irbesartan-insensitive compared to irbesartan-sensitive tumours. The upregulation of AT1R was associated with an increase in proliferation and VEGF expression by cancer cells. While animals bearing irbesartan-sensitive tumours showed a marked decrease in the number of proliferating cells in the liver and VEGF-expressing infiltrating cells in the tumour following AT1R treatment, these were unchanged by treatment in animals bearing irbesartan-insensitive (high AT1R expressing) tumours.
Although the results do not support increased efficacy of combined treatment, they provide intriguing evidence of the importance of RAS expression in determining patient response and tumour growth potential and suggest that components of the RAS could be used as biomarkers to aid in patient selection.
Metastasis to the liver is the leading cause of death in patients with colorectal cancer (CRC). We previously showed that targeting of the renin angiotensin system (RAS) with either an angiotensin (ANG) II type I receptor (AT1R) blocker (irbesartan) or an angiotensin converting enzyme (ACE) inhibitor (captopril) could inhibit tumour growth in an orthotopic syngeneic mouse model of CRC liver metastases [2, 3]. ACE is responsible for converting inactive ANG I into the key active peptide of the classical RAS, ANG II. The AT1R mediates proliferative, proinflammatory, and angiogenic effects of ANG II [4, 5].
The RAS also has an 'alternative' pathway which counteracts many of the actions induced by ANG II-AT1R signalling. The alternative ANG II receptor (the AT2R) generally exerts actions antagonistic to the AT1R including inhibition of proliferation and promotion of apoptosis . ACE2, a homologue of ACE, generates a second RAS peptide, ANG-(1-7), directly from ANG II. ANG-(1-7), through its specific receptor MasR, also appears to counteract many of the actions induced by the classical AT1R/ANGII RAS pathway . Activation of the alternative ANG II receptor, the AT2R, has been shown to inhibit tumour growth (although to lesser extent then either irbesartan or captopril). ANG-(1-7) can also be infused to reduce tumour growth in several experimental cancer models [8, 9]. Two independent Phase I clinical trials are examining ANG-(1-7)  and AT1R blockade  in the treatment of various solid tumours.
Given the counter-regulatory actions of the classical and alternative RAS pathways we hypothesized that combining inhibition of the classical RAS (AT1R blockade or ACE inhibition) with activation of the alternative RAS (ANG-(1-7) infusion or AT2R activation) would synergistically inhibit tumour growth.
The mouse colorectal cancer (MoCR) cell line used for in vivo experiments was harvested from a dimethylhydrazine-induced colon carcinoma in a CBA mouse at a stage known to metastasise to the liver . Liver metastases were induced as described previously [3, 12]. Briefly, 25000 MoCR cells were injected into the spleen of 6 to 8 week old male CBA mice and, after 3 minutes, the spleen removed to confine metastases to the liver. A minimum of 5 animals per group were used, in treatments inducing fewer tumours sample size was increased to 10. All experiments were approved by the Austin Health Animal Ethics Committee. Liver samples were collected and fixed in fresh 4% PFA.
In vivo treatments included ANG-(1-7) (Auspep, 2588; 24 μg/kg/hr), CGP42112A (AT2R agonist, Sigma-Aldrich, C160; 0.6 μg/kg/hr), and/or telmisartan (AT1R blocker, Sigma-Aldrich, T8949; 12.5 μg/kg/hr) infusion (Alzet® osmotic mini pumps 1004) or s.c. daily injections of irbesartan (AT1R antagonist, Bristol-Myers Squibb) at 50 mg/kg. Captopril was given as daily i.p. injections of 750 mg/kg (Sigma-Aldrich, 21751). Doses were based on previously published studies [3, 5, 13–15]. The solubilising agent (saline or methyl cellulose) provided a control. Treatments continued from the time of tumour induction to tissue collection at day 21.
AT1R (rabbit polyclonal against human, Santa Cruz, sc-1173), proliferation (Ki67; rat monoclonal anti-mouse, Thermoscientific, RM-9106-S1), apoptosis (active caspase 3; rabbit polyclonal anti-human, R&DSystems AF835), angiogenesis (CD34, neovascularisation marker; monoclonal rat anti-mouse, Abd Serotec MCA18256), and VEGF (rabbit polyclonal anti-human, CalBiochem, PC315) were assessed in PFA-fixed paraffin embedded tissues. Specificity of AT1R and VEGF antibodies was confirmed by western blot (data not shown). AT1R was used at a concentration of 0.5 μg/ml, Ki67 at 1:100 dilution (dilution provided with manufacturer's datasheet), active caspase-3 at 1.0 μg/ml, CD34 at 0.1 μg/ml, and VEGF at 1.5 μg/ml. Non-immunized rabbit IgG (Santa Cruz, sc-2027) at an equivalent concentration to the primary target antibody was used as a negative control. Endogenous peroxidases were blocked with 3% H2O2 and non-specific binding inhibited with 10% normal goat serum (Zymed, 01-6201). Slides were incubated with primary antibodies at 37°C for 1 hour and then 4°C overnight. Slides were then incubated with the secondary antibody (Dako Envision+ Goat anti-rabbit HRP secondary 4011 for AT1R, Ki67, caspase 3, and VEGF and the Rat on Mouse AP-polymer kit (Biocare Medical; RT518H) for CD34 for 1 hour at 37°C before visualisation with DAB or, for CD34, Vulcan fast red (Applied Medical FR805H). Slides were counterstained with Mayer's haematoxylin.
Images of stained tumours were taken using digital light microscope (Nikon Coolscope®, Nikon Corporation, Japan) at between 40x and 400x magnification (with a scale bar for size calibration) and were analysed using Image-Pro plus (version 5). Depending on the animal and its tumour load, between 10 and 30 images across 1 to 5 tumours were taken for analysis. AT1R staining was assessed in control tissues, while VEGF, Ki67, caspase3, and CD34 were all examined in control and irbesartan treated tissues. Where possible, the number of positive cells out of the total number of cells was used to measure changes in protein levels/content. A scoring system was used to quantify differences in the strength and abundance of AT1R and VEGF staining in tumours. The intensity of immune-reactive staining was evaluated subjectively (by two independent researchers) using the intensity of immunoreactive colour development as an measure of relative protein content. Strong staining of many/most cells was given a grade of 5 reducing to no staining (0). Only areas of viable tumour were considered in analysis. A subpopulation of intense VEGF expressing infiltrating cells was also counted and is expressed as the number of positive cells per area tumour. Ki67 and active caspase3 were assessed as both the area of proliferating tumour (4x magnification) and the number of proliferating cells per area tumour or liver (20x magnification). CD34 was assessed as the length of positively stained endothelium per tumour area and provide an indication of the angiogenic potential of these tumours.
Quantitative data are presented as mean ± SEM or boxplots showing the minimum value, first quartile, median, third quartile and maximum value. Statistical analyses were conducted using SPSS (Statistical Package for the Social Sciences, version 17, USA) or Microsoft Excel (2003). Normally distributed data were assessed by ANOVA. Mann-Whitney U test were used for non-parametric data sets. A probability (P) value of less than 0.05 was considered as statistically significant.
ACE inhibition with captopril was found to retain its anti-tumour activity (Figure 1A). Captopril-treated animals had significantly lower liver-to-body weight ratios (0.312 ± 0.010) compared to the control group with (0.139 ± 0.011) (P = 2.907 x 10-8, t-test equal variance). The level of tumour inhibition achieved by captopril treatment in the current experiment is comparable to our previously published studies [2, 3].
While irbesartan was previously found to inhibit tumour growth in this mouse model of CRC liver metastases, in the current experiment, it failed to reduce liver tumour burden despite the same experimental protocols being employed. We hypothesized that this difference was likely due to either to the animals used, which although of the same strain, sex, and age were obtained 2 year apart, or to the cell line, which again may have acquired changes over the intervening years between experiments.
Immunohistochemistry was used to assess the relative level and extent of AT1R expression in CRC-derived tumours growing in the liver from both studies. While no obvious difference was seen in the liver surrounding tumours, tumours from the current study showed a marked upregulation of AT1R expression. AT1R staining in cancer cells was found to be higher in the irbesartan-insensitive animals compared to the irbesartan-sensitive animals (Additional file 1 and Figure 3). Pairwise comparisons between tumours of different sizes all showed higher AT1R expression in irbesartan-insensitive tumours (AT1RHI) compared to irbesartan-sensitive tumours (AT1RLOW) (small tumours: 3 vs 2, P = 0.002; medium tumours score: 4 vs 2, P < 0.0001; large tumours: 4 vs 2, P < 0.0001 respectively, Mann-Whitney U test).
In contrast to the increase in cancer-cell proliferation by AT1RHI tumours, the liver surrounding these tumours had fewer (P ≤ 0.0327, t-test) proliferating cells compared to animals bearing AT1RLOW tumours (Figure 4C). Moreover, there was a marked reduction (P ≤ 0.0002, t-test) in the number of these proliferating cells with irbesartan treatment in the AT1RLOW, but not the AT1RHI tumour-bearing animals. It is not clear what type of cell is represented, but both non-paranchymal and paranchymal cells were present in the proliferating population and based on morphology and location relative to the sinusoids, these cells are likely to be hepatic stellate cells or metastasizing tumour cells. Indeed, in our previous study we showed that irbesartan could reduce the number of metastatic lesions .
While cancer cell-associated VEGF was higher in the AT1RHI tumours (P ≤ 0.0123, t-test), the number of intense-staining infiltrating cells was highest in the AT1RLOW tumours and this was decreased by irbesartan treatment (P = 1.109 x 10-5) (Figure 6B).
Blockade of the classical RAS through AT1R blockade or ACE inhibition reduces tumour growth in several experimental mouse models of cancer [2, 3, 16, 17]. Conversely activation of the alternative RAS, through ANG-(1-7) infusion  or AT2R activation , can also reduce tumour growth. We sought to determine if greater inhibition of tumour growth could be achieved through dual targeting of the RAS - inhibition of the classical RAS with simultaneous activation of the alternative RAS. While, we were unable to show any benefit for combined RAS treatments, we found that an upregulation of the AT1R by the cancer cell line rendered them insensitive to AT1R blockade. These same cells, however, remained sensitive to ACE inhibition, suggesting significant differences in the molecular mechanisms by which these agents inhibit tumour growth.
The anti-tumour effects of AT1R blockers have been well documented [3, 18, 19]. However, in our experiment, irbesartan, when used alone or in combination with ANG-(1-7) infusion, failed to decrease tumour growth. We also tested the potent AT1R blocker telmisartan, which has been utilised as an anti-cancer agent in other experimental models where it was shown to be effective [16, 20, 21], but again we found no benefit of treatment compared to control animals. The lack of an effect of AT1R blockade also contradicts a previous study conducted in our laboratory using the same experimental protocols .
To further investigate this finding we performed immunohistochemical analysis on tumours from both studies (for which tissues had been collected in the same experimental manner). AT1R expression by cancer cells from animals insensitive to AT1R blockade showed a marked increase in immunohistochemical staining compared to tissues collected from our previous irbesartan sensitive animals. These results suggest that cancer cells could be rendered insensitive to AT1R blockade through an upregulation of the AT1R. Moreover, despite strong AT1R expression by liver macrophages described here and elsewhere  and the importance of macrophages during tumour progression , our results suggest that cancer cell AT1R expression can overcome any macrophage-mediated antitumour activity associated with AT1R blockade.
The AT1R is also known to regulate angiogenesis, migration, and apoptosis all of which could contribute to regulating tumour growth [24, 25]. Thus, it is likely that the upregulation of AT1R is, at least in part, responsible for the resistance of these cells to AT1R blockade. In support of this we found that high AT1R-expressing tumours had increased cancer cell proliferation, higher levels of cancer cell-associated VEGF and tumour-associated angiogenesis. Interestingly, irbesartan lead to a slight increase in VEGF levels in both tumour types. This is in contrast to some studies of breast cancer [26, 27], but supports our previous study which showed that, in these tumours, activation of the AT2R increased VEGF expression . Clere et al. (2010) similarly found that in fibrosarcoma, the AT2R could promote cancer cell VEGF production . AT2R expression has also been documented in blood vessels of human pituitary adenomas  and both the AT1R and AT2R stimulate VEGF secretion by rat pituitary tumour cells . Thus, it would appear that in at least some cancer-associated circumstances the AT2R can also be proangiogeneic. However, angiogenesis (as measured by CD34 staining) was decreased by irbesartan and this appeared to be associated with a decrease in the number of infiltrating VEGF-expressing cells.
We also report an increase in apoptosis and a decrease in angiogenic vessel formation associated with irbesartan, and while still present in AT1RHI tumours, these effects were less than that seen for AT1RLOW tumours. Additionally, while it is apparent that cancer-cell AT1R expression has a direct effect on cancer cell proliferation, VEGF-expression, and apoptosis, we also saw marked changes in the surrounding liver and in cells infiltrating the tumours. In particular, irbesartan sensitive tumours (AT1RLOW) when untreated (i.e. control animals) were associated with higher numbers of VEGF-infiltrating cells in tumours and higher numbers of proliferating cells in the liver surrounding tumours. Both of these phenomena were reduced by irbesartan treatment. These results would suggest that while tumour resistance was conferred by higher cancer cell-AT1R expression (and that this was associated with certain growth advantages such as increased proliferation and reduced treatment-associated apoptosis), the non-parenchymal cells in the liver may be important in determining the response of low AT1R-expressing tumours to irbesartan.
In contrast to AT1R blockade, ACE inhibition by captopril maintained its effectiveness, consistent to the previous findings . This suggests that ACE inhibitors, other than blocking the production of ANG II which subsequently binds AT1R, also generate anti-tumour effects through non-ANG II mediated pathways. Indeed this has been documented in several studies where matrix metalloproteinases and vascular endothelial growth factor expression have been recognised as non-RAS dependent targets of ACE inhibition [31–33].
This research initially set out to determine if AT1R blockade or ACE inhibition in combination with either ANG-(1-7) or AT2R activation could lead to synergistic inhibition of CRCLM. While we failed to show any benefit of combined targeting of the RAS, our results were fortuitous in providing insight into both the importance of differences between the mechanisms of action of ACE inhibition and AT1R blockade as well as the potential of RAS expression as a biomarker.
angiotensin converting enzyme
angiotensin II type 1 receptor
angiotensin II type 2 receptor
vascular endothelial growth factor
renin angiotensin system
This work was supported by Cancer Australia's Priority-driven Collaborative Cancer Research Scheme and was co-funded by Cancer Australia and Cure Cancer Australia Foundation. We would also like to thank the CanToo fund raisers. Initial funding was provided by the Cancer Council of Victoria. Dr Ager is supported by an NHMRC Post-doctoral Training Award. Ms Shu-wen Wen and Dr Jaclyn Neo are supported by an Australian Rotary Health Research Fund PhD Scholarship.
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.