The angiotensin receptor blocker, Losartan, inhibits mammary tumor development and progression to invasive carcinoma

Abstract

Drugs that target the Renin-Angiotensin System (RAS) have recently come into focus for their potential utility as cancer treatments. The use of Angiotensin Receptor Blockers (ARBs) and Angiotensin-Converting Enzyme (ACE) Inhibitors (ACEIs) to manage hypertension in cancer patients is correlated with improved survival outcomes for renal, prostate, breast and small cell lung cancer. Previous studies demonstrate that the Angiotensin Receptor Type I (AT1R) is linked to breast cancer pathogenesis, with unbiased analysis of gene-expression studies identifying significant up-regulation of AGTR1, the gene encoding AT1R in ER+ve/HER2-ve tumors correlating with poor prognosis. However, there is no evidence, so far, of the functional contribution of AT1R to breast tumorigenesis. We explored the potential therapeutic benefit of ARB in a carcinogen-induced mouse model of breast cancer and clarified the mechanisms associated with its success.Mammary tumors were induced with 7,12-dimethylbenz[α]antracene (DMBA) and medroxyprogesterone acetate (MPA) in female wild type mice and the effects of the ARB, Losartan treatment assessed in a preventative setting (n = 15 per group). Tumor histopathology was characterised by immunohistochemistry, real-time qPCR to detect gene expression signatures, and tumor cytokine levels measured with quantitative bioplex assays. AT1R was detected with radiolabelled ligand binding assays in fresh frozen tumor samples.We showed that therapeutic inhibition of AT1R, with Losartan, resulted in a significant reduction in tumor burden; and no mammary tumor incidence in 20% of animals. We observed a significant reduction in tumor progression from DCIS to invasive cancer with Losartan treatment. This was associated with reduced tumor cell proliferation and a significant reduction in IL-6, pSTAT3 and TNFα levels. Analysis of tumor immune cell infiltrates, however, demonstrated no significant differences in the recruitment of lymphocytes or tumour-associated macrophages in Losartan or vehicle-treated mammary tumors.Analysis of AT1R expression with radiolabelled ligand binding assays in human breast cancer biopsies showed high AT1R levels in 30% of invasive ductal carcinomas analysed. Furthermore, analysis of the TCGA database identified that high AT1R expression to be associated with luminal breast cancer subtype.Our in vivo data and analysis of human invasive ductal carcinoma samples identify the AT1R is a potential therapeutic target in breast cancer, with the availability of a range of well-tolerated inhibitors currently used in clinics. We describe a novel signalling pathway critical in breast tumorigenesis, that may provide new therapeutic avenues to complement current treatments.

Keywords: angiotensin receptor; interleukin 6; invasive ductal carcinoma; luminal breast cancer; tumor necrosis factor.

Figure 1. Inhibition of AT1R with Losartan decreases mammary tumor onset and progression associated with MPA and DMBA treatment

Nulliparous six-week-old female mice were implanted subcutaneously with MPA pellets and treated orally with DMBA. Losartan (600 mg/L) was administered in drinking water. a. Onset of mammary tumors in age-matched wildtype MPA/DMBA-treated mice control-treated (3% sucrose) or with AT₁R blocker, Losartan. Data for panels are expressed as percentage of mice free of tumors (palpable and histological assessment) after last DMBA treatment (P = 0.004, Mantel Cox Log-Rank Test). b. Tumor latency, development of palpable tumors days post-DMBA administration, data point represents individual animal. Immunohistochemistry quantification of c. Ki67 and d. Cleaved Caspase 3 in control and Losartan-treated animals. Each data point represents individual tumor analysed; and immunostain is presented as a percentage of total tumour area including tumour-associated stroma e. Representative Haematoxylin/Eosin (H&E) stained histological sections of tumors from age-matched control and Losartan treated animals. f. Representative histological sections of mammary tumor stages identified including Ductal Carcinoma in situ (DCIS), mixed DCIS and invasive ductal carcinoma (DCIS/IDC) and IDC; H&E stained (top panel) and immunostained with a differentiation marker for cancer-associated fibroblasts, α Smooth Muscle Actin antibody (αSMA, bottom panel). Intact myoepithelial layer identified by αSMA is indicated (arrowheads) in well-differentiated DCIS, while loss of myoepithelial αSMA and gain of αSMA expression in the stroma (arrowheads) is indicated in DCIS/IDC and the poorly differentiated squamous IDC. g. Tumor histopathology quantified as % of tumors assessed. Difference in incidence of invasive tumors between the 2 groups is indicated (P = 0.03, Fisher's Exact Test). Tumor-free animals are only observed in the Losartan group (green). h. Autoradiograph showing binding of I¹²⁵-[Sar¹, Ile⁸] Angiotensin II (I¹²⁵AngII) to a section (unmagnified) of the inguinal mammary containing a DCIS lesion demonstrating strong AT₁R expression. Insert: control tissue co-incubated with unlabeled AngII and I¹²⁵AngII. H&E stained serial section of DCIS lesion at 20X magnification. Data are presented as mean +SEM, n = 10-15 animals per group. *P < 0.05, Mann Whitney Test unless otherwise stated.

Figure 2. Inhibition of AT1R with Losartan decreases tumor cytokine production

a. qRT-PCR expression of AT₁R-regulated genes, with Losartan treatment, in mammary glands derived from mice 2 weeks after last DMBA-administration. Data is represented as fold change of mean gene expression in control samples (mean+SEM, n = 5 animals per group, unpaired Student's T-Test). b. Multiplex quantitation of cytokines in tumor lysates (mean+SEM, Control tumors n = 12 and Losartan-treated tumors n = 6). c. Representative immunostaining for E-cadherin, an EMT marker, in tumor sections from age-matched control and Losartan-treated animals. d. Quantitation of membrane-specific E-cadherin immunostaining intensity with ImageScope; score of 3 given to strongest stain (black arrows) and 1 given to weak/absence of stain (red arrows). e. Phospho-STAT3 localisation in age-matched tumors and the quantitation of pSTAT3 immunostain f. and with AlphaLISA in tumor protein lysates g. (mean+SEM, Control tumors n = 12 and Losartan-treated tumors n = 6). h. Representative α Smooth Muscle Actin (αSMA) immunostained sections from age-matched tumors highlighting greater proportion of αSMA positive CAFs in the tumor stroma in control animals compared to Losartan-treated tumors. Alteration in αSMA expression pattern in different tumor cell types is shown. In Losartan-treated tumors a strong myoepithelial staining in differentiated tumours (Losartan, black arrow) and the lack of αSMA in the CAFs (Losartan, red arrow). In control, vehicle treated animals CAF-specific αSMA expression is observed in the poorly differentiated, advanced tumors highlighted (Control, black arrows). i. Quantitation of αSMA positive cancer-associated fibroblasts (CAFs) presented as percentage αSMA-positive CAFs in the total tumor stroma (Control tumors n = 9 and Losartan-treated tumors n = 6). Representative tumor images are from serial sections taken from tumors 50 days after last DMBA administration. Data are presented as mean+SEM, *P < 0.05, *P < 0.01, Mann-Whitney Test unless otherwise stated.

Figure 2. Inhibition of AT1R with Losartan decreases tumor cytokine production

a. qRT-PCR expression of AT₁R-regulated genes, with Losartan treatment, in mammary glands derived from mice 2 weeks after last DMBA-administration. Data is represented as fold change of mean gene expression in control samples (mean+SEM, n = 5 animals per group, unpaired Student's T-Test). b. Multiplex quantitation of cytokines in tumor lysates (mean+SEM, Control tumors n = 12 and Losartan-treated tumors n = 6). c. Representative immunostaining for E-cadherin, an EMT marker, in tumor sections from age-matched control and Losartan-treated animals. d. Quantitation of membrane-specific E-cadherin immunostaining intensity with ImageScope; score of 3 given to strongest stain (black arrows) and 1 given to weak/absence of stain (red arrows). e. Phospho-STAT3 localisation in age-matched tumors and the quantitation of pSTAT3 immunostain f. and with AlphaLISA in tumor protein lysates g. (mean+SEM, Control tumors n = 12 and Losartan-treated tumors n = 6). h. Representative α Smooth Muscle Actin (αSMA) immunostained sections from age-matched tumors highlighting greater proportion of αSMA positive CAFs in the tumor stroma in control animals compared to Losartan-treated tumors. Alteration in αSMA expression pattern in different tumor cell types is shown. In Losartan-treated tumors a strong myoepithelial staining in differentiated tumours (Losartan, black arrow) and the lack of αSMA in the CAFs (Losartan, red arrow). In control, vehicle treated animals CAF-specific αSMA expression is observed in the poorly differentiated, advanced tumors highlighted (Control, black arrows). i. Quantitation of αSMA positive cancer-associated fibroblasts (CAFs) presented as percentage αSMA-positive CAFs in the total tumor stroma (Control tumors n = 9 and Losartan-treated tumors n = 6). Representative tumor images are from serial sections taken from tumors 50 days after last DMBA administration. Data are presented as mean+SEM, *P < 0.05, *P < 0.01, Mann-Whitney Test unless otherwise stated.

Figure 3. Losartan treatment increases peripheral blood vessel diameter and reduces VEGFA levels in tumors

a. Vessel diameter was measured for the blood vessels located at the tumor circumference, b. blood vessel density and c. VEGFA, a marker for angiogenesis, measured by ELISA in tumor lysates derived from control and Losartan-treated animals. Quantitation of immunohistochemistry stain of d. CD3, a marker for T cells, e. B220, a marker of B cells and f. the macrophage marker F4/80, in mammary tumour tissue of the two treatment groups. Data are presented as percentage of positively stained areas of the total tumour area including tumour-associated stromal regions; mean+SEM, P < 0.05, Mann Whitney Test.

Figure 4. Expression of AT1R is increased in invasive breast cancer

a. Binding of ¹²⁵I-[Sar¹, Ile⁸] AngII to serial sections of a Grade II, ER^+ve invasive tumor. Radioligand binding is displaced by unlabelled AngII or Losartan (AT₁R blocker), but not PD123319 (AT₂R blocker) indicating expression of AT₁R. b. Left-hand panel: H&E-stained breast carcinoma sections; Right-hand panel: Corresponding ¹²⁵I-[Sar¹, Ile⁸] AngII binding in serial sections. c. Mean intensity of bound ¹²⁵I-[Sar¹⁸] AngII /total area in tumor subtypes and matched normal tissue, n = 10; data points are combined from matched normal for each sub-type, n = 30. d. AGTR1 mRNA expression in TCGA breast invasive carcinoma, subtyped by the PAM50 molecular signature and e. P values highlighting statistically significant differences in AGTR1 expression between the PAM50 tumor subtypes (Mann Whitney Test). TCGA data was retrieved from cBioPortal.