Targeting of Beta Adrenergic Receptors Results in Therapeutic Efficacy against Models of Hemangioendothelioma and Angiosarcoma

Therapeutic targeting of the beta-adrenergic receptors has recently shown remarkable efficacy in the treatment of benign vascular tumors such as infantile hemangiomas. As infantile hemangiomas are reported to express high levels of beta adrenergic receptors, we examined the expression of these receptors on more aggressive vascular tumors such as hemangioendotheliomas and angiosarcomas, revealing beta 1, 2, and 3 receptors were indeed present and therefore aggressive vascular tumors may similarly show increased susceptibility to the inhibitory effects of beta blockade. Using a panel of hemangioendothelioma and angiosarcoma cell lines, we demonstrate that beta adrenergic inhibition blocks cell proliferation and induces apoptosis in a dose dependent manner. Beta blockade is selective for vascular tumor cells over normal endothelial cells and synergistically effective when combined with standard chemotherapeutic or cytotoxic agents. We demonstrate that inhibition of beta adrenergic signaling induces large scale changes in the global gene expression patterns of vascular tumors, including alterations in the expression of established cell cycle and apoptotic regulators. Using in vivo tumor models we demonstrate that beta blockade shows remarkable efficacy as a single agent in reducing the growth of angiosarcoma tumors. In summary, these experiments demonstrate the selective cytotoxicity and tumor suppressive ability of beta adrenergic inhibition on malignant vascular tumors and have laid the groundwork for a promising treatment of angiosarcomas in humans.


Introduction
A wealth of evidence indicates that stressful situations exacerbate tumor progression. Adrenergic processes stimulated by epinephrine and norephinephrine drive the development of tumor growth and metastasis [1,2,3]. Breast cancer models indicate the sympathetic nervous system serves as a neural regulator inducing a metastatic switch, with stress-induced neuroendocrine activation leading to a 30-fold increase in tumor metastasis [4]. Moreover, beta adrenergic receptor antagonists have shown efficacy against melanoma [5,6], breast cancer [7,8,9,10,11,12], and prostate cancer [3], and high beta adrenergic receptor expression in tumors is associated with poor clinical outcome in breast cancer patients [13]. One prospective study revealed a 36% reduction in the risk of melanoma progression for each year of beta blocker treatment [6].
The non-specific beta adrenergic receptor inhibitor propranolol has been utilized as the gold standard treatment in pediatric patients with infantile hemangioma, a benign vascular tumor which affects up to 10% of the Caucasian population [14].
Infantile hemangiomas express high levels of beta adrenergic receptors potentially explaining their sensitively to propranolol, and these receptors are reported to be high in malignant vascular tumors [15]. This suggests the efficacy of propranolol and other beta blockers may extend to aggressive vascular tumors. Several reports indicate positive results from beta blockade in patients with moderately threatening vascular tumors including kaposiform hemangioendothelioma, tufted angiomas, parotid hemangioma, Kasabach-Merritt phenomenon, and drug resistant hepatic hemangiomas [16,17,18,19,20]. It remains to be determined if more malignant vascular tumors such as the rare and understudied metastatic hemangioendotheliomas and angiosarcomas are susceptible to the effects of beta blockade. Such a study is essential given the five year survival rates for angiosarcomas and metastatic hemangioendotheliomas are very poor. Treatment of angiosarcomas typically involves surgery, radiation, and neoadjuvant and/or adjuvant chemotherapy with doxorubicin or taxanes, yet even following aggressive therapy, patient survival in metastatic disease is abysmally low [21,22]. Hemangioendotheliomas are often associated with low mitotic index and are typically more manageable, however approximately 30% of these tumors display aggressive behavior, rapid disease progression, and eventual patient death despite extensive treatment [23,24]. Given the lack of adequate therapies for angiosarcomas and aggressive hemangioendotheliomas, there exists a strong need for the development of treatments against these unique tumor types. Based on the remarkable efficacy of propranolol against benign vascular tumors and the recent report that malignant vascular tumors exhibit high beta adrenergic receptor expression we hypothesized that hemangioendotheliomas and angiosarcomas should be particularly susceptible to propranolol. Thus we examined the effects of propranolol on the oncogenic properties of a panel of malignant vascular tumor lines and the growth of in vivo angiosarcoma tumors. Our data lay the preclinical groundwork for the future clinical use of beta blockade against malignant vascular tumors in humans.

Ethics Statement
Use of de-identified human tissues was approved by the Texas Tech University Health Sciences Center Institutional Review Board for the Protection of Human Subjects (IRB E13029). Waiver of informed consent was approved by IRB under 35 CFR 46.117(c) (1). All mouse experimentation was approved by the Texas Tech University IACUC committee (IACUC #12020) and conducted according to responsible guidelines for the care and use of laboratory animals. Animals were sacrificed using CO2 asphyxiation.

mRNA Expression
For qPCR, RNA from 100 mM propranolol or sham treated SVR cells (24 hours treatment) was purified using the Purelink RNA mini kit (Ambion) and converted to cDNA using the Verso cDNA synthesis kit (Thermo-Scientific). qPCR was performed in triplicate using TaqMan probes for ADRB1 (Invitrogen; human/ mouse #Hs02330048_s1, canine #Cf02691262_g1), ADRB2 (Invitrogen; human/mouse #Hs00240532_s1, canine #Cf02690130_s1), and ADRB3 (Invitrogen; human/mouse #Hs00609046_m1, canine #Cf03022965_s1) or qPCR arrays (SABiosciences; apoptosis array #PAMM-012Z, cell cycle array #PAMM-020Z) on an ABI7900HT real time PCR instrument (Applied Biosystems). Microarrays on SVR cells treated for 24 hours with 100 mM propranolol or sham were performed in triplicate from pooled samples of three biological replicates for each condition as previously reported [26]. Briefly, total RNA was isolated from each replicate using the Purelink RNA Micro kit (Invitrogen), pooled, and amplified and biotin-labeled using Illumina TotalPrep RNA Amplification Kit (Ambion). 750 ng of biotinylated aRNA was then briefly heat-denatured and loaded onto expression arrays to hybridize overnight. Following hybridization, arrays were labeled with Cy3-streptavidin and imaged using the Illumina ISCAN. Intensity values were transferred to Agilent GeneSpring GX microarray analysis software and data was filtered based on the quality of each call. Statistical relevance was determined using ANOVA with a Benjamini Hochberg FDR multiple testing correction (p-value ,0.05). Data were then limited by fold change analysis to statistically relevant data points demonstrating a 2-fold or greater change in expression. Omics pathway analysis was performed with Metacore integrated software suite (GeneGo). Microarray data was publically deposited in Gene Expression Omnibus (Accession # GSE42534).

Immunohistochemistry
Immunohistochemical studies were performed on 5 mm thick, formalin fixed, paraffin-embedded angiosarcoma sections obtained from Hospital Ste Jutine (Montreal, Canada). Sections were deparaffinized, rehydrated, and treated for antigen retrieval using

Cell Proliferation/survival
Cells were plated at approximately 50% confluence and treated with a dose curve of 0 to 200 mM propranolol over a 48 hour time course. Time lapse microscopy was performed using a BioStation CT (Nikon) and the change in cell number over the experimental time course was determined by manually counting cells per vision field. For growth in three dimensional culture to examine the effects of chronic low doses of propranolol on vascular tumor cell survival, cells were seeded onto Alvetex membranes at 2610 5 cells per (12) well, allowed to establish for 48 hours, and subsequently treated with a sub-lethal concentration of propranolol (20 mM) or sham over a period of 6 days, after which the membranes were stained with Hoechst 33342 to label the nuclei to facilitate cell counting. Numerical data presented for all proliferation experiments is the average of at least three biological replicates +/2 the standard deviation. Statistical significance was determined using Student's t-test (P,0.05).

Western Blotting
Cell lysates were collected after 1 or 24 hours treatment (as indicated in the results), subjected to SDS-PAGE, and transferred to nitrocellulose membranes using the Trans-Blot Turbo Transfer System (Biorad). Membranes were incubated with antibodies obtained from Cell Signaling Proteins were detected using Supersignal West Dura Extended Duration Substrate (Thermo Scientific) and digitally captured using a GE Image Quant LAS4000 imaging system.

Cell Cycle Analysis
Cells treated for 24 hours with 50 mM propranolol were trypsinized, fixed in a 70:30 ratio of ethanol:phosphate buffered saline overnight, stained with 200 mg/ml ethidium bromide plus 20 mg/ml RNase A, and incubated overnight. DNA content of each cell treatment condition was analyzed using an Accuri C6 flow cytometer. Quantitative analysis of DNA content was performed using CFlow Plus software (Accuri). Quantification of the percentage of cells in each cell cycle phase represents the average of at least three biological replicates +/2 the standard deviation. Statistical significance was determined using Student's ttest (P,0.05).

Coculture Assays
Primary cultures of human dermal microvascular endothelial cells (HDMVECs, ATCC #CRL-4025) were labeled with Cell Tracker Blue (Invitrogen) and mixed with unlabeled SVR angiosarcoma cells. The coculture was grown in vascular cell basal media (ATCC #PCS-100-030) and supplemented with 0.2% bovine brain extract, 5 ng/ml human epidermal growth factor, 10 mM L-glutamine, 0.75 units/ml heparin sulfate, 1 mg/ ml hydrocortisone, 50 mg/ml ascorbic acid, 2% fetal bovine serum, and pen/strep. Cells were treated 48 hours with 100 mM propranolol or sham and Cell Tracker Blue fluorescence (marking the HDMVECs) and phase contrast (visualizing both HDMVECs and SVR cells) were imaged using a Floid Cell Imaging Station (Life Technologies).

Migration Assay
For scratch assays, confluent cultures of vascular tumor cells were manually scratch wounded with a P-200 micropipette tip, treated with 50 mM propranolol, and wound closure was monitored hourly using a Nikon Biostation CT time lapse imaging robot over an 18 hour period. For the spheroid migration assay, 2000 cells from the Frog and Emma cell lines were cultured in a hanging drop of their growth media (30 ml total volume) for 24 hours and then allowed to adhere to a standard tissue culture substrate in the presence of 50 mM propranolol or sham. Images were obtained 24 hours after attachment using a Nikon TS100 inverted microscope. Data presented for migration experiments is the average of at least three biological replicates +/2 the standard deviation. Statistical significance was determined using Student's ttest (P,0.05).

Computational Cytoskeleton Analysis
Median stress fiber length was calculated using the FiberScore algorithm [27] for actin cytoskeleton immunofluoresence images from propranolol or sham treated malignant vascular tumor cells (obtained as described in Immunofluorescence methods described above). At least 10 images of each condition were used for the analysis.

In vivo Tumor Assays
SVR chorioallantoic membrane (CAM) tumors were grown using the gelatin sponge-chorioallantoic membrane (CAM) assay as previously described [28]. An isotonic saline sham (N = 4 eggs) or 100 mM propranolol (N = 3 eggs) was added daily directly onto the tumors every two days, and tumors were collected 6 days post implantation. Mouse angiosarcoma tumors were generated by subcutaneous injection of 1610 6 SVR cells into the dorsolateral flank of 4 week old female mice as previously described [29]. Mice were treated with sham (isotonic saline; N = 15) or propranolol (10 mg/ml; N = 17) via intraperitoneal injections every 2 days. When the sham tumors reached approximately 1 cm in diameter, the tumors were collected. Numerical data presented for all tumor experiments is the average of at least three biological replicates +/ 2 the standard deviation (for the CAM assays) or the standard error of the mean (for the mouse angiosarcoma tumor assays). Statistical significance was determined using Student's t-test (P,0.05).

Beta Adrenergic Receptors are Expressed in Malignant Vascular Tumors
Beta adrenergic receptors 1, 2, and 3 have been shown to be expressed across a diverse panel of vascular tumors, with the highest expression in malignant vascular tumors [15]. To confirm this, we performed immunohistochemistry for the steady state protein levels of ADRB1, ADRB2, and ADRB3 on four clinically characterized human angiosarcoma tumors, revealing strong signals for ADRB1 and ADRB2 and weaker sample-dependent signals for ADRB3 ( Figure 1A). The clinical data from each angiosarcoma tissue sample is detailed in Table 1. We then examined the steady state mRNA expression levels of ADRB1, ADRB2, and ADRB3 in a panel of malignant vascular tumor lines including 4 primary canine angiosarcoma cell lines (Emma, Frog, Jack, and SB), one mouse angiosarcoma cell line (SVR), and one mouse hemangioendothelioma cell line (EOMA), demonstrating that canine angiosarcoma lines preferentially express ADRB2 and ADRB3 ( Figure 1B), while malignant mouse vascular tumor lines express ADRB1 and ADRB2 ( Figure 1C). As infantile hemangiomas (which strongly express beta adrenergic receptors [15,26]), are clinically susceptible to propranolol, we sought to determine if malignant vascular tumors respond similarly. . We then performed flow cytometric analysis on propidium iodide stained vascular tumor cells which were treated with sham or 50 mM propranolol for 24 hours. We utilized 50 mM propranolol for this experiment since our initial experiments at higher doses (100+ mM propranolol) resulted in a large sub-apoptotic peak for each cell line (data not shown) thus complicating analysis of how beta blockade affects cell cycle progression. Due to the heterogeneous cell populations composing the EOMA line and complicating interpretation of the data with numerous peaks, we excluded this line from cell cycle analysis. Both SVR and Emma cell lines exhibited a significant increase of cells in the G2/M phase following propranolol addition, while treatment of Frog led to an accumulation of cells in the G1 phase ( Figure 2E & F). The cell cycle profile of propranolol treated SB cells was not significantly different from the sham SB cells, corroborating its relative resistance to the treatment.
To identify mechanisms by which propranolol inhibits vascular tumor cell proliferation, we performed a global gene expression microarray on SVR angiosarcoma cells treated with sham or 100 mM propranolol for 24 hours. We identified 84 genes whose expression was statistically altered by more than 3-fold (p,0.05) in response to propranolol and 428 genes exhibiting a significant 2fold or greater expression alteration. The three fold or greater changes are listed in Table 2 and the entire dataset can be publically accessed via the Gene Expression Omnibus (Accession # GSE42534). Genego pathway analysis identified five major cellular processes that were altered in response to propranolol including cell cycle, cell survival, cholesterol metabolism, immune response, and cell movement. We followed up on the changes by testing the expression of a number of genes/proteins involved in cell cycle regulation using qPCR arrays on SVR cells treated for 24 hours with sham or 100 mM propranolol. Quantification of the steady state mRNA expression levels of 84 genes known to be involved in cell cycle regulation identified 25 genes whose expression was significantly altered by more than 1.5 fold (p,0.05) in response to propranolol ( Figure 3A). Of particular interest were the 4.7 and 4.4 fold upregulation of the potent cell cycle inhibitors Cdkn1a and Cdkn1b, respectively. We validated this data via Western and immunofluorescence analysis of Cdkn1a and Cdkn1b (p21 and p27, respectively) in SVR angiosarcoma lines treated for 24 hours with sham or 100 mM propranolol, revealing significant upregulation of both proteins following propranolol treatment ( Figure 3B & C). Further experimentation revealed decreased steady state protein levels for the proliferative marker PCNA and the key cell cycle regulators Cdk4, Cdk6, cyclin D1, and cyclin E1 ( Figure 3C). Our microarray analysis identified a marked 3.2-fold upregulation in Tek expression following 24 hours of propranolol ( Table 2). This gene encodes for the Tie2 receptor which is a major regulator of vessel maturation and may partly explain the selectivity of propranolol for cells of endothelial origin. 24 hour treatment of SVR cells with 100 mM propranolol resulted in a strong upregulation of Tie2 protein levels, and stimulation of the Tie2 receptor with angiopoietin-1 in serum starved SVR cells treated with either sham or 100 mM propranolol led to a propranolol-dependent increase in Tie2 phosphorylation ( Figure 3D). These data suggest that propranolol may be influencing the transcriptional regulation of signaling pathways involved in vessel quiescence and maturation.

Propranolol Induces Apoptosis of Malignant Vascular Tumor Cells
We observed significant cell death in the panel of malignant vascular tumor cell lines at 100 mM propranolol or greater (Figure 2A), thus we investigated the contribution of propranolol to hemangioendothelioma and angiosarcoma cell survival. The cell panel was treated with sham or 100 mM propranolol for 24 hours, co-incubated with membrane impermeable propidium iodide (staining only dead cells) and membrane permeable Hoechst 33342 (staining live and dead cells), and the percent apoptosis was determined by quantifying the portion of propidium iodide stained cells in the total population. Propranolol treatment resulted in a significantly increased apoptosis in each cell line, with no propidium iodide cells detected in any of the sham treatments ( Figure 4A & B). Nuclear modeling of Hoechst 33342 stained SVR cells treated with 100 mM propranolol revealed nuclear morphology alterations indicative of apoptosis ( Figure 4C), and given that pathway analysis of our microarray data revealed a strong apoptotic node, we quantified the steady state mRNA expression levels of 84 genes involved in apoptosis using SVR cells treated for 24 hours with sham or 100 mM propranolol. We identified 18 genes whose expression was significantly altered 1.5 fold or more (p,0.05) ( Figure 4D), indicating that beta blockade regulates the network of genes controlling angiosarcoma cell survival. Furthermore, staining for cleaved-caspase 3 was observed in SVR angiosarcoma cells treated for 24 hours with 100 mM propranolol, but undetectable in sham conditions ( Figure 4E). 1 hour of propranolol treatment in SVR cells resulted in marked increases in the phosphorylation of p38 MAPK ( Figure 4F), suggesting that cells are rapidly responding by initiating the stress response pathway. No changes in the phosphorylation of p42/44   Tek Endothelial-specific receptor tyrosine kinase NM_013690. 1 3.2 or JNK/SAPK were detected following propranolol treatment (data not shown). After 24 hours, Western analysis revealed that 100 mM propranolol treatment of SVR cells resulted in increased p53, cleaved caspase 3 and cleaved PARP ( Figure 4G), indicative of an apoptotic response. Despite propranolol's ability to induce apoptosis at relatively higher concentrations, this effect must be selective for tumor cells, yet spare normal cells. We previously reported that primary cultures of human endothelial cells exhibit reduced proliferation at 50 mM propranolol, however no apoptosis was observed at propranolol concentrations less than 150 mM [26]. To test if propranolol exhibits selectivity against malignant vascular tumor cells, we performed co-culture experiments where unstained SVR cells were grown in the presence of CellTracker Blue dye stained human dermal microvascular endothelial cells (HDMVECs). After 48 hours, sham treatment resulted in a confluent layer of SVR cells intermixed with dispersed HDMVECs cells ( Figure 4H). In contrast, 100 mM propranolol resulted in complete detachment of SVR cells, with only labeled HDMVECs attached to the substrate. Even following chemotherapy and radiation, patients with angiosarcomas and metastatic hemangioendotheliomas experience a high mortality rate. If beta blockade is ever used in a clinical setting, it will likely be used in combination with standard anti-angiosarcoma chemotherapeutics, thus we tested if propranolol is effective as a combination therapeutic against malignant vascular tumors. We treated the cell panel with sham, 100 mM propranolol, chemotherapeutic/cytotoxic agents (cisplatin, busulfan, vincristine, or H 2 O 2 induced free radical damage) alone, or the combination of propranolol and these agents. Either propranolol or the toxic agents led to significant reductions in cell survival after 48 hours in all cell lines, however the combination of propranolol plus the chemotherapeutic/cytotoxic agents resulted in synergistic reductions in survival for all lines except Frog ( Figure 5A). One of the most common reasons for mortality in cancer patients is acquisition of drug resistance following treatment, therefore we tested if our cell panel is capable of developing resistance to propranolol. The cells were treated with lethal doses (100 mM) of propranolol for several days and the few surviving cells were allowed to repopulate the culture. This process was repeated multiple times. Cell survival was measured between naïve cells (1 st treatment) and cells after four rounds of propranolol selection (4 th treatment). Cell survival was statistically no different between naïve cells and experienced cells ( Figure 5B), suggesting that resistance to propranolol may be minimal. This data suggests that propranolol may enhance the efficacy of standard therapies without increasing tumor resistance.

Propranolol Disrupts the Migration of Malignant Vascular Tumors
To test if propranolol affects malignant vascular tumor cell migration, we performed scratch assays on our cell panel. To avoid cell death from high doses of propranolol, we reduced the concentration to 50 mM, for which no net negative decreases in cell proliferation were observed for any line (Figure 2A). At this dosage, we were able to accurately measure migration changes without changes in cell death complicating the data analysis. For all lines tested except SVR, 50 mM propranolol reduced closure of the wound relative to the sham ( Figure 6A). To corroborate these findings, we employed a 3D growth assay where cells were cultured as 3D spheres, plated onto a substrate in the presence of 50 mM propranolol or sham, and allowed to invade into the adjacent surroundings over 24 hours. Similar to our reasoning with scratch wounding migration assays, we used a sublethal dose of propranolol (50 mM) to evaluate migration independent of cell death. Only Emma and Frog assembled into well formed tumor spheroids, thus we proceeded with using these lines. The sham treated tumor spheroids aggressively migrated from the spheroid mass into surrounding areas over the 24 hour period, while 50 mM propranolol abolished all migration ( Figure 6B). To understand the mechanism by which propranolol inhibits cell movement and invasion, we first tested the phosphorylation status of key cytoskeletal regulators in SVR angiosarcoma cells. We have previously shown that propranolol treatment of infantile hemangioma endothelial cells decreased phosphorylation of the actin regulatory protein cofilin [26], however Western blot analysis of SVR cells showed no change in the phosphorylation of cofilin, ezrin/radixin/moesin (ERM), or myosin phosphatase targeting subunit (MYPT) in reponse to propranolol ( Figure 6C). Imaging of actin microfilaments revealed that sham SVR cells exhibited hours treatment with sham or 100 mM propranolol. (F) Western analysis detecting phospho-p38 MAPK and total p38 MAPK protein levels after 1 hour of sham or 100 mM propranolol treatment of SVR angiosarcoma cells. (G) Western blot detection of apoptotic protein levels. (H) Unstained SVR angiosarcoma cells were co-cultured with CellTracker Blue stained HDMVECs. The co-cultures were subjected to sham or 100 mM propranolol and DIC/fluorescent image overlays were obtained after 48 hours treatment. For all experiments, the data is the average +/2 standard deviation for at least three biological replicates. Statistical significant was determined using Students t-test (p,0.05). doi:10.1371/journal.pone.0060021.g004   well developed actin stress fibers and numerous lamellipodia, while sublethal doses of propranolol (50 mM) resulted in dissolution of central actin stress fibers, with only cortical actin remaining along the cell periphery ( Figure 6D). We measured the median stress fiber length of sham or 50 mM propranolol treated malignant vascular tumor cells using the FiberScore algorithm [27], revealing a significantly marked reduction in stress fiber length following propranolol treatment in all but the Emma and EOMA lines ( Figure 6E).

Propranolol Inhibits Angiosarcoma Tumor Growth
To move our findings into more relevant in vivo models, we employed the chorioallantoic membrane (CAM) tumor assay where SVR cells in a gelatin sponge were placed onto the vascularized CAM of 8 day post-fertilization chicken eggs. Tumors were treated with a sham or 100 mM propranolol every two days over a six day time course, after which the tumors were excised, revealing significant reductions in tumor size and weight in propranolol treated tumors ( Figure 7A & B). We expanded our studies into an established mouse angiosarcoma tumor model [29] where the animals were systemically treated with saline sham or 10 mg/kg propranolol every 2 days. Tumor collection occurred when the sham tumors reached approximately 1 cm 3 . Propranolol therapy resulted in significantly reduced angiosarcoma tumor size and weight (357+/258 mg; N = 17, p,0.0001) vs sham (984+/ 292 mg; N = 15, p,0.0001) ( Figure 7C & D). Despite the reduction in tumor size following propranolol administration, tumor sections from both sham and propranolol conditions revealed PCNA expression at the tumor peripheries, indicating active cell division ( Figure 7E). This suggests that future studies should employ combinatorial therapy with propranolol.

Discussion
Malignant vascular tumors such as angiosarcomas and metastatic hemangioendotheliomas are rare solid tumors that readily metastasize due to their intrinsic vascular nature. These tumors are exceptionally difficult to treat and several phase II trials investigating the therapeutic efficacy of novel anti-angiogenic drugs such as Bevacizumab, Sunitinib, and Sorafenib have resulted in a minimal to absent response in vascular tumor patients [30,31,32]. Of the three drugs, Sorafenib exhibited an angiosarcoma subtype-specific response in 14% of patients, however the median progression free survival was only 3.8 months, with 81% of angiosarcoma patients and 100% of hemangioendothelioma patients exhibiting tumor progression after 6 months of treatment [32]. This indicates the strong need to develop more effective therapies against malignant vascular tumors.
Given that we ( Figure 1A-C) and others have demonstrated that malignant vascular tumors express high levels of beta adrenergic receptors [15,26], we hypothesized that beta blockade might demonstrate therapeutic efficacy against these tumors. We reveal that propranolol selectively inhibits proliferation, survival, and migration of a panel of malignant vascular tumor cells, indicating that the oncogenic properties of these tumor types are driven, in part, by beta adrenergic signaling. While propranolol dramatically slows the proliferation rate of all vascular tumor lines tested, four of the five tumor lines exhibited nearly 100% lethality at doses that we have previously reported to be non-toxic for primary cultures of human endothelial cells (Figure 2A) [26].
Indeed, co-culture of HDMVECs and angiosarcoma cells in the presence of propranolol results in selective killing of the angiosarcoma cells ( Figure 4H). These findings indicate that malignant vascular tumor lines are exquisitely sensitive to beta blockade and provide strong support for its use in animal and human models. It should be noted that while beta blockade was effective in reducing the proliferation rate and increasing cellular apoptosis for all cell lines, there were varying levels of resistance between the lines. For instance, the SB line exhibited high resistance to propranolol-induced alterations in proliferation and apoptosis (Figure 2A, Figure 5A & B), while the EOMA hemangioendothelioma cell line consistently showed the highest susceptibility to beta blockade. Future studies should elucidate the molecular profiles characteristic of each of these lines that might serve as putative biomarkers of susceptibility to propranolol.
Our data demonstrate that propranolol blocks proliferation and survival in part through altering levels of key cell cycle and apoptotic regulators (Figures 3 & 4). Our data indicates that propranolol induces a strong upregulation of Tek. Tek is preferentially expressed on endothelial cells, encodes for the Tie2 angiogenic receptor, and strongly promotes vessel regression, quiescence, and stability in a context dependent manner [33]. We suspect that in addition to general cell cycle alterations, propranolol's specificity for vascular tumors may function through Tie2-mediated modulation of vessel stability and maturation, however the Ang/Tie2 network is a very complex signaling pathway interdependent on a number of other proteins, thus future work should address this issue.
Using two in vivo tumor models we demonstrated efficacy for propranolol as an anti-angiosarcoma agent. However, propranolol as a stand-alone therapeutic (like most therapeutics) did not completely abolish tumor cell proliferation as evidenced by positive PCNA staining in propranolol treated tumors. Given our exciting data showing synergistic effects on malignant vascular tumor lines treated with combined propranolol and chemotherapy ( Figure 5A), future emphasis should be placed on determining the optimal strategy for beta blockade against hemangioendotheliomas and angiosarcomas. Indeed, a recent publication reported propranolol potentiates the anti-tumor effects of chemotherapy in preclinical models of breast cancer [10]. Such an endeavor should not prove difficult given that propranolol has been used extensively to treat hypertension, anxiety, and post-traumatic stress syndrome since its Nobel Prize winning discovery in the 1960's, and its toxicity and side effects are well established in pediatric and adult patients.
In this study, we employ a novel application of beta blockade for the successful treatment of malignant vascular tumors using in vitro and in vivo models. These findings are significant because metastatic hemangioendotheliomas and angiosarcomas are very rare aggressive tumor types that are markedly understudied, exceptionally difficult to treat, and result in high patient mortality. These data lay the groundwork for future clinical trials to test the off-label use of the generic drug propranolol for a population of patients in desperate need of effective tumor treatments.