The Association of Endothelin‐1 Signaling with Bone Alkaline Phosphatase Expression and Protumorigenic Activities in Canine Osteosarcoma

Neumann, Z.L. et al. “The Association of Endothelin‐1 Signaling with Bone Alkaline Phosphatase Expression and Protumorigenic Activities in Canine Osteosarcoma.” Journal of Veterinary Internal Medicine 29.6 (2015): 1584–1594. PMC. Web. 28 Jan. 2017.



Canine osteosarcoma (OS) is an aggressive sarcoma characterized by pathologic skeletal resorption and pulmonary metastases. A number of negative prognostic factors, including bone alkaline phosphatase, have been identified in dogs with OS, but the underlying biologic factors responsible for such observations have not been thoroughly investigated. Endothelin‐1‐mediated signaling is active during bone repair, and is responsible for osteoblast migration, survival, proliferation, and bone alkaline phosphatase expression.



The endothelin‐1 signaling axis is active in canine OS cells, and this pathway is utilized by malignant osteoblasts for promoting cellular migration, survival, proliferation, and bone alkaline phosphatase activities.



45 dogs with appendicular OS.



The expressions of endothelin‐1 and endothelin A receptor were studied in OS cell lines and in samples from spontaneously occurring tumors. Activities mediated by endothelin‐1 signaling were investigated by characterizing responses in 3 OS cell lines. In 45 dogs with OS, bone alkaline phosphatase concentrations were correlated with primary tumor osteoproductivity.



Canine OS cells express endothelin‐1 and endothelin A receptor, and this signaling axis mediates OS migration, survival, proliferation, and bone alkaline phosphatase activities. In OS‐bearing dogs, circulating bone alkaline phosphatase activities were positively correlated with primary tumor relative bone mineral densities.


Conclusions and Clinical Importance

Canine OS cells express endothelin‐1 and functional endothelin A receptors, with the potential for a protumorigenic signaling loop. Increases in bone alkaline phosphatase activity are associated with osteoblastic OS lesions, and might be an epiphenomenon of active endothelin‐1 signaling or excessive osteoproduction within the localized bone microenvironment.

Keywords: Cell migration and survival, Osteoblastic response, Prognostic factor


bone alkaline phosphatase
dual energy x‐ray absorptiometry
endothelin receptor type A
relative bone mineral density

Bone alkaline phosphatase (bALP) is an enzyme anchored to the outer plasma membrane leaflet on both normal and malignant osteoblasts, and has been identified as a necessary factor for the initiation of skeletal mineralization.1, 2 Given its essential role in bone homeostasis, serum bALP activity serves as a surrogate of osteoblastic bone formation and can be used as an adjuvant diagnostic test in cancer patients at risk for developing osteoblastic bone metastasis.3, 4, 5 In addition to its diagnostic role in skeletal metastases, increased serum bALP activity is a well‐documented negative prognostic factor in dogs with appendicular osteosarcoma (OS) and is associated with shorter disease‐free intervals and survival times.6, 7, 8 However, the exact reasons for it to serve as a negative prognostic factor in dogs diagnosed with OS remains incompletely defined. Elucidating the biologic mechanisms responsible for its association with a poor prognosis may help to identify novel therapies for the adjuvant treatment of OS in an effort to increase survival and quality of life in these patients.

In humans, and more recently in dogs, a correlation between serum bALP activity and tumor burden has been identified and mechanistically suggests that poor prognosis associated with increased serum bALP activity is simply a function of advanced disease stage, and consequently shorter metastasis‐free survival and overall survival.9, 10, 11 However, this association between serum bALP activity and tumor burden is not always clinically apparent, and a subset of patients with substantial tumor burden does not uniformly have increases in serum bALP activity. Such observations suggest that other biologic mechanisms, in addition to tumor size, likely contribute to an increased serum bALP activity in dogs with OS.

Endothelin‐1 (ET‐1) is a potent peptide, originally discovered to be expressed by vascular endothelial cells, that plays a pivotal role in osteoblastic disease processes, mainly by the mediation of cellular differentiation and migration of bone cells.12, 13 Endothelin‐1 inhibits osteoclasts from resorbing bone, stimulates osteoblast survival and proliferation, and increases expression of osteopontin, osteocalcin, and bALP, resulting in new bone formation.14, 15, 16 These bone biologic effects are mediated by the endothelin A receptor (ETAR) and are attenuated by antagonism of this receptor‐mediator signaling pathway. As such, selective antagonists of ETAR have been developed as an anticancer strategy to block the proliferative and survival effects induced by ET‐1 on prostate cancer cells and associated reparative osteoblasts residing within the bone tumor microenvironment.17, 18, 19

The involvement of ETAR signaling in promoting malignant osteoblast migration, survival, and proliferation, as well as the regulation of bALP activity in OS of dogs has not been well characterized. As such, the purpose of this study was to investigate additional biologic mechanisms, specifically ET‐1 mediated signaling, which might explain the prognostic relevance of serum bALP activity in appendicular OS‐bearing dogs. The objectives of our study were (1) to demonstrate the expression and functionality of the endothelin axis in canine OS cells; (2) to evaluate the protumorigenic properties of ET‐1‐mediated signaling in vitro (3) to determine the effects of ETAR antagonism on bALP expression and activity in vitro; and (4) to evaluate the relationship between serum bALP activity and primary tumor relative bone mineral density (rBMD) measured by means of dual energy x‐ray absorptiometry (DEXA).


Materials and Methods

Cell Lines

Three canine OS cell lines including HMPOS (provided by Dr James Farese, University of Florida), D17 (purchased from American Tissue Culture Collection), and Abrams (provided by Dr Douglas Thamm, Colorado State University) were utilized in this study given their known cellular characteristics.20, 21, 22 The Jurkat (human T lymphocyte cells) and DU145 (human prostate carcinoma) cell lines were purchased from American Tissue Culture Collection and used as positive controls.19, 23 Cells were cultured at 37°C in Dulbecco's Modified Eagle's Medium (DMEM) supplemented with glutamine (2 mmol/L), penicillin (100 IU/mL), streptomycin (100 IU/mL), and 10% fetal bovine serum (FBS) in a humidified atmosphere supplemented with 5% CO2. Cell cultures were maintained in subconfluent monolayers and passaged 2–3 times weekly as necessary.


Reagents and Antibodies

Human recombinant ET‐11 (E7764), anti‐ET‐1 antibody1 (mouse monoclonal; E166), and anti‐ETAR antibody1 (rabbit polyclonal; E3651) were purchased from Sigma‐Aldrich. Atrasentan hydrochloride2 (ABT‐627) was purchased from MedchemExpress. Phosphatidylinositol‐specific phospholipase C3 (P‐6466) was purchased from Life Technologies. Anti‐phosphorylated c‐jun antibody4 (rabbit monoclonal, 3270), anti‐c‐jun antibody4 (rabbit monoclonal, 9165), and anti‐c‐fos antibody4 (rabbit monoclonal; 2250) were purchased from Cell Signaling Technology. Horseradish peroxidase conjugated antimouse and antirabbit secondary antibodies were purchased from GE Healthcare, UK. Anti‐β actin antibody5 (mouse monoclonal; AC‐15) and anti‐ALP antibody5 (mouse monoclonal; ab108337) were purchased from Abcam.

Cell Protein Collection

Cells were grown to 80–100% confluence, themedia was then removed and cells were washed twice with phosphate‐buffered saline (PBS). Cells were trypsinized and centrifuged at 700 g at 4°C for 5 minutes. The resultant cell pellet was homogenized in 1 mL PBS and transferred to a 1.8 mL Eppendorf tube before centrifuging at 6,000 g at for 4°C for 5 minutes. Cell pellets were homogenized with 100–150 μL of Mammalian Protein Extraction Reagent3 (M‐PER) and mixed with fresh protease inhibitor cocktail solution3 for 15 minutes and then centrifuged at 10,000 rpm for 15 minutes at 4°C. The resultant supernatant was assessed for protein concentrations using a standard assay kit3 (Bicinchoninic Acid Protein Assay).

Western Blot Analysis

Endothelin‐1 and ETAR Expression

For each protein investigated, 50 μg samples were electrophoresed on 12% polyacrylamide gel and then transferred to a nitrocellulose membrane. Membranes were blocked with Tris‐Buffered Saline with Tween 20 (TBST) with 5% milk for 1 hour at room temperature. Western blot analysis was performed using ETAR and ET‐1 antibodies at a concentration of 1 : 1,000 in TBST with 5% milk. The membrane then was washed three times with TBST and probed with the secondary antibody diluted 1 : 5,000 in TBST with 5% milk. Blots were developed using ChemiDoc XRS+ molecular imager system6 with Image Lab software6. Band volume analysis was done using ImageJ software.7 Relative protein expressions were adjusted for β‐actin or another reference protein used as a loading control. Results reported were derived from at least 2 independent experiments.

Expressions of c‐fos, c‐jun, and phospho‐c‐jun

The Abrams, D17, and HMPOS cell lines were grown in complete media to 85% confluence. Media was removed and the cells were serum starved for 24 hours. Cells then were exposed to experimental conditions including: (1) serum starved only or serum starved for 24 hours and then stimulated for 1 hour with (2) 10% FBS; (3) DMSO (vehicle); (4) 100 nM ET‐1; (5) 40 μM ABT‐627; or (6) 40 μM ABT‐627 followed by 100 nM ET‐1 for 1 hour (combo). After exposure to experimental conditions, cells were collected in 0% FBS DMEM to avoid changes in the phosphorylation of target proteins. Proteins were electrophoresed and transferred to membranes, and then incubated with either phosphorylated‐c‐jun antibody at 1 : 1,000 in TBST with 5% milk, c‐jun antibody at 1 : 1,000 in TBST with 5% milk, or c‐fos antibody at 1 : 1,000 in TBST with 5% milk. The membrane then was washed three times with TBST and probed with the secondary antibody diluted 1 : 5,000 in TBST with 5% milk and developed using a standard chemiluminescence detection kit.8 The enhanced chemiluminescence (ECL) blots were developed using ChemiDoc XRS+ molecular imager system6 with Image Lab software6. Band volume analysis was done using ImageJ software7. Results reported were derived from at least two independent experiments.



Ten canine appendicular OS tissue blocks were retrieved from the University of Illinois Veterinary Diagnostic Laboratory for immunohistochemical assessment. Slides were deparaffinized in xylene and rehydrated in ethanol. Endogenous peroxidase activity was blocked with 3% hydrogen peroxide in methanol for 15 minutes, and then slides were rinsed twice with wash buffer for 5 minutes. Slides were incubated with preheated 0.1% protease at 37°C for 20 minutes, and then rinsed in wash buffer for 2 minutes. Nonspecific staining was minimized with incubation for 10 minutes with Power Block,9 and then blocking for 15 minutes with avidin and biotin block. Slides were incubated with antihuman primary antibodies against ETAR (1 : 500) or ET‐1 (1 : 2,000) for 1 hour at room temperature. Slides were incubated with a biotinylated secondary antibody for 20 minutes at room temperature, then washed in buffer before incubation for 20 minutes with a streptavidin‐biotinylated horseradish peroxidase complex,10 and developed with 3,3'‐diaminobenzidine (DAB) substrate for 5 minutes. Slides were counterstained with hematoxylin and evaluated by a single investigator (KLW). Negative controls for the samples were processed identically in the absence of the primary antibody.


Colony Forming Assay

The Abrams, D17, and HMPOS cell lines were seeded overnight in 6‐well plates at a concentration of 200 cells per 3 mL of DMEM with 10% FBS. Upon adherence of individual cells, media was discarded and new media containing different experimental conditions were added including DMEM 5% FBS with vehicle (DMSO) or a range of ABT‐627 concentrations (10–40 μM). Cells were allowed to grow undisturbed for 7 (Abrams) or 10 (HMPOS and D17) days. Subsequently, wells containing colonies were gently rinsed with chilled PBS, treated with 3 mL of solution containing 6.0% glutaraldehyde and 0.5% crystal violet for 30 minutes, and then rinsed with tap water. Each of the treatment conditions was performed in 3 separate experiments. The number of visible colonies per experimental condition was quantified 5 times by 3 different individuals blinded to treatment conditions.


Migration Scratch Assay

Qualitative analysis of cell migration was done using the “scratch assay” method.24 The Abrams, HMPOS, and D17 cell lines were grown to 80% confluence in 6‐well plates in DMEM supplemented with 10% FBS. Six experimental conditions for a duration of 24 hours were evaluated including complete media with vehicle (DMSO or H2O), ET‐1 (0.1–100 nM), or ABT‐627 (10–40 μM). A standardized acellular gap was created through cell monolayers using a 200 μL pipette tip in the middle of each well. Images of the acellular gap were captured at time 0 (maximal gap) and 24 hours later for each experimental condition using an inverted microscope with a mounted digital camera. The average width of 5 representative acellular gaps per experiment conditions and cell line were used for quantitative comparisons, with percent residual gap ([(time 0 gap−time 24 gap)/time 0 gap] × 100) serving as the experimental readout. Each of the treatment conditions was performed in duplicate and the assay results are representative of 2 separate experiments. Data were analyzed with ImageJ software7.


Cytologic Detection of bALP Activity

The most robust bALP‐expressing cell line, HMPOS, was plated into 0.5 mL chamber well slides at a concentration of 25,000 cells per well and allowed to adhere overnight in supplemented media. The media then was removed and the cells were exposed to serum‐starved media for 48 hours with (1) media only, (2) DMSO (vehicle), (3) 40 μM ABT‐627, or (4) 40 μM ABT‐627 for 30 minutes followed by 100 nM ET‐1 for 48 hours. The media then was removed and chambers were washed with PBS. The slides were allowed to dry and then were stained with nitroblue tetrazolium chloride/5‐bromo‐4‐chloro‐3‐indolyl phosphate toluidine salt (NBT/BCIP), an ALP substrate, as previously reported.25 Briefly, slides were incubated for precisely 8 minutes at room temperature with sufficient amounts of NBT/BCIP to coat the slide. Slides were rinsed with water, blotted dry, and examined microscopically. Positive staining was indicated by brown to black staining of the cell surface. Three representative images of each chamber were taken and quantitative analysis was performed on normalized pixel intensity measurements which accounted for two variables: (1) differences in cell densities and (2) staining intensities of positive cells among various treated and untreated conditions.


In Vitro bALP Protein Assessments

The most robust bALP‐expressing cell line, HMPOS, was plated at a density of 5 × 105 cells/well in 6‐well plates in 3 mL of 10% FBS DMEM. After cell adherence, media was removed and washed gently with PBS, then cells were exposed to serum‐free media with varying conditions for 48 hours including (1) serum‐starved DMEM only (control), (2) vehicle (DMSO), (3) 40 μM ABT‐627, or (4) 40 μM ABT‐627 for 30 minutes followed by 100 nM ET‐1 (combo) for 48 hours. Media was removed, and cells were washed gently with 1 mL PBS, then scraped for collection. The cell pellet was resuspended in 0.5 mL PBS with 0.5 units of phosphatidylinositol‐specific phospholipase C and incubated for 30 minutes at 37°C. After incubation, cells were pelleted and the supernatant was collected for quantitative analysis with a commercial test kit11 utilizing a murine monoclonal anti‐bALP antibody. To normalize measured bALP activities, protein concentrations were determined in aliquots of each of the cell extracts using bicinchoninic acid (BCA) analysis. This cell layer protein was used as an index of cell number, which could have varied after exposure to the different conditions, and bALP was expressed as normalized activity units per gram of protein (U/L/g). Additionally, cleaved membranous bALP released into cell culture media was assessed by Western blot analysis using 20 μL aliquots of supernatant derived from the various conditions after PI‐PLC incubation. A rabbit monoclonal antihuman/mouse bALP antibody was utilized at a concentration of 1 : 1,000 incubated overnight at 4°C followed by antirabbit secondary antibody at a concentration of 1 : 1,000 for 1 hour at room temperature.


Clinical Population, bALP, rBMD, and ET‐1 Assessment

The study included a subset of dogs with OS that were evaluated at the University of Illinois Cancer Care Clinic between 2003 and 2012. All dogs had a diagnosis of appendicular OS confirmed by either histopathology or cytology with concurrent positive ALP staining. All dogs included in the study had orthogonal radiographs of the primary tumor, 3‐view thoracic radiographs, DEXA scan, and serum collected at the time of diagnosis. Only dogs with untreated primary bone tumors and absence of macroscopic pulmonary metastases were used for associating serum bALP with primary tumor relative bone mineral density.

Serum bALP activity was evaluated with a commercial kit11 utilizing a murine monoclonal anti‐bALP antibody, previously validated for use in dogs.26 At initial presentation, DEXA scans were taken using a dedicated imaging unit12 of the primary tumor and an equivalent anatomic area of the unaffected contralateral limb. Bone mineral density (BMD) of the primary tumor was divided by BMD of the contralateral normal limb, and expressed as a ratio termed relative BMD (rBMD) as previously described.27 For purposes of this study, a subset of 45 dogs was included for analysis and divided dichotomously into two categories (osteolytic or osteoblastic), which correlated with the lower (osteolytic; rBMD < 0.7) and upper (osteoblastic: rBMD > 1.3) rBMD quartiles generated from a larger population of dogs (n = 95). To minimize primary tumor size as a confounding factor, a ratio of rBMD to the longest tumor measurement (RECIST) assessed by digital radiographs was used to identify any associations between normalized rBMD and serum bALP activity. Plasma endothelin concentrations were quantified using a commercially available kit13 previously validated in dogs.28


Statistical Analysis

The distribution of the continuous variable data was evaluated using the Shapiro–Wilk test, skewness, kurtosis, and q–q plots. Data were analyzed with either parametric or nonparametric methods depending upon the achievement of normality assumptions. For normally distributed data, 1‐way ANOVA was used to evaluate for differences among groups with the use of Dunnet's posthoc test. Nonnormally distributed data were analyzed with the Kruskal–Wallis test and Dunn's comparison test. For the comparison of two datasets, student t‐test or Mann–Whitney U‐test was employed for normal or nonnormal distributed datasets, respectively. A Pearson product‐moment correlation coefficient was used to assess the relationship between serum bALP activity and normalized rBMD/RECIST ratio. Statistical calculations were performed using commercial software programs,14 and P < .05 was considered statistically significant for all analyses.



Protein Expression and Functionality of ET‐1 and ETAR in Canine OS

By Western blot analysis, ET‐1 and ETAR were expressed by the 3 immortalized canine OS cell lines utilized in this study (Fig1A). In 10 spontaneously arising primary OS tumors, ET‐1 demonstrated moderate to strong positive cytosolic expression in all (10/10) samples evaluated, but ETAR expression was more variable in both staining intensity and pattern. The majority of primary OS tumors were positive for ETAR staining, either uniformly (4/10) or focally (5/10), but complete absence of ETAR staining was identified in one of 10 samples examined (Fig S1).


Figure 1: (A) Endothelin‐1 (ET‐1) and ETAR expressed in canine OS cell lines by Western blot analysis. Positive control DU145 and Jurkat for ET‐1 and ETAR, respectively. (B–D) ETAR signaling pathway is functional in canine OS cells with upregulation of c‐fos and phospho‐c‐jun following either 10% FBS or ET‐1 (100 nM) stimulation. Similarly, blockade of ETAR is achievable with ABT‐627, a selective ETAR antagonist.


To confirm the functionality of the endothelin intracellular signaling cascade in canine OS cell lines, components of the mitogen‐activated protein kinase (MAPK) pathway, specifically c‐fos and phospho‐c‐jun, were evaluated by Western blot analysis. In serum‐starved HMPOS cells, exposure to either 10% FBS or exogenous ET‐1 (100 nM) for 1 hour produced robust increases in c‐fos and phospho‐c‐jun expression (Fig1B–D). To demonstrate that c‐fos and phospho‐c‐jun upregulation was mediated specifically by ETAR stimulation, HMPOS cells were preincubated with a selective ETAR antagonist (40 μM ABT‐627) for 30 minutes before exogenous ET‐1 (100 nM) incubation. Exposure to ABT‐627 substantively inhibited protein expression of c‐fos and phospho‐c‐jun, and persistent blockade of ETAR signaling was achieved by ABT‐627 in HMPOS cells continuously co‐incubated with exogenous ET‐1. Modulation of c‐fos and phospho‐c‐jun expression also was identified in the D17 and Abrams cell lines, but some divergence in cell signaling responses was evident (Fig S2A,B). Despite comparable ETAR expression by all 3 OS cell lines based on Western blot analysis (Fig1A), response to ETAR stimulation by either FBS or exogenous ET‐1 was variable based upon changes in normalized c‐fos protein expression (Fig S2C), which suggests that the molecular consequences of ETAR stimulation or inhibition could produce heterogeneous biologic responses in OS cells of different origins.


ETAR Antagonism Inhibits OS Cell Survival and Proliferation

To determine the contribution of ETAR‐mediated signaling in malignant osteoblast survival and proliferation, colony‐forming assays were performed utilizing a range of ABT‐627 concentrations (0–40 μM). Continuous exposure to ABT‐627 for 7–10 days during colony formation inhibited cellular survival and proliferation in a dose‐dependent manner across the 3 canine OS cell lines tested (Fig2A,B; Table 1). For the D17 cell line (Fig2A), exposure to ABT‐627 at concentrations ≥30 μM decreased the number of colonies compared to vehicle (DMSO). Similarly, colony formation by the Abrams cell line was attenuated by ABT‐627 at all concentrations ≥20 μM. In the HMPOS cell line, colony formation was impeded by ABT‐627 at concentrations of 20 μM and 40 μM.


Figure 2: Visual reduction (A) in the number of colonies formed over 10 days in the D17 cell line following ETAR signaling blockade by ABT‐627. (B) General dose‐dependent reduction in colony formation as a consequence of ETAR signaling blockade by ABT‐627 in three canine OS cells lines (top, D17; middle, Abrams; bottom, HMPOS). Data expressed as medians and significance defined as **P < .01 and ***P < .001.


Table 2: ET A R antagonism with ABT‐627 reduces colony formation after continuous incubation for 7–10 days.Data expressed as median (range) with significance defined as **P < .01; ***P < .001.


Endothelin‐1 Signaling Mediates Canine OS Cell Migration

In a dose‐dependent manner, the addition of exogenous ET‐1 for 24 hours accelerated the closure of an acellular gap created in confluent cultures of all 3 canine OS cell lines (Fig3A,B; Table 2). The migration of D17 (Fig3A) and Abrams cell lines was uniformly promoted by ET‐1 concentrations ranging between 1.0 and 100 nM. The promigratory effects of ET‐1 were less pronounced in the HMPOS cell line, with only the highest ET‐1 concentration (100 nM) accelerating acellular gap closure after 24 hours of exposure. Converse to the effects exerted by exogenous ET‐1, blockade of ETAR signaling for 24 hours by ABT‐627 inhibited malignant OS cell migration in a dose‐dependent manner (Fig4A,B); Table 3). Based upon residual gap distances after 24 hours, the participation of ETAR signaling in the promotion of cell migration was most pronounced in the Abrams cell line, with concentrations of ABT‐627 ≥ 30 μM completely abrogating malignant OS cell migration (Fig4A). The effect of ABT‐627 to inhibit the migration of D17 and HMPOS cells also was achieved at concentrations ranging from 20–40 μM, however, the degree of inhibition (approximately 60%) was less pronounced compared to that achieved in the Abrams cell line (approximately 100%).


Figure 3: Visual enhancement (A) in cell migration with consequent reduction in residual gap after 24 hours of scratch induction in the D17 cell line following exogenous ET‐1 supplementation. (B) General dose‐dependent augmentation in cell migration induced by a wide dose range of exogenous ET‐1 in 3 canine OS cell lines (top, D17; middle, Abrams; bottom, HMPOS). Data expressed as mean ± SEM and significance defined as *P < .05 and **P < .01.


Figure 4: Visual inhibition (A) in cell migration following exposure to ABT‐627, a selective antagonist of ETAR, in the Abrams OS cell line. (B) Consistent dose‐dependent inhibition of cell migration with ABT‐627 in 3 canine OS cell lines (top, D17; middle, Abrams; bottom, HMPOS). Data expressed as mean ± SEM and significance defined as **P < .01.


Table 2: Endothelin‐1 promotes cell migration and reduces the percent residual gap remaining after 24 hours of cell culture. Data expressed as mean ± SEM with significance defined as *P < .05; **P < .01.


Table 3: ET A R antagonism with ABT‐627 inhibits cell migration and increases percent residual gap remaining after 24 hours of cell culture. Data expressed as mean ± SEM with significance defined as **P < .01.


Endothelin A Receptor Antagonism Decreases bALP Expression and Activity

Modulation of bALP expression and activity as a consequence of ETAR signal blockade was studied in the HMPOS cell line, given its robust expression of bALP (Fig S3A). Three complementary methods to study bALP modulation were employed including cytochemical analysis, enzymatic activity, and protein expression. Exposure of HMPOS cells to ABT‐627 (40 μM) alone or pretreated with ABT‐627 (40 μM) then followed by ET‐1 (100 nM) stimulation resulted in decreased bALP activity as qualitatively appreciated by visual inspection of cytochemical staining intensity. ImageJ software was used to objectively quantify differences in staining intensity among treatment groups by expressing staining data as normalized pixel intensity scores, which account for cell density and positive cell staining intensity  (Fig5A,B). Normalized pixel intensity scores for ABT‐627 alone or followed by ET‐1 stimulation were 1201.4 ± 402.8 pixel intensity/cell or 1295.7 ± 294.1 pixel intensity/cell, respectively. In comparison, HMPOS cells exposed to control or vehicle conditions demonstrated significantly higher bALP‐normalized pixel intensity scores of 4390.9 ± 808.3 pixel intensity/cell or 4460.5 ± 430.4 pixel intensity/cell (P < .05). Analogous to the cytochemical methodology, bALP activity quantified by ELISA and normalized for protein concentration yielded similar findings (Fig5C). Again, HMPOS cells exposed to ABT‐627 alone or in conjunction with ET‐1 demonstrated a marked reduction in normalized bALP activity at 386.4 ± 46.4 U/L/g or 353.1 ± 35.5 U/L/g, respectively, when compared to cells exposed to control or vehicle conditions (4303.7 ± 300.2 U/L/g or 3967.6 ± 172.1 U/L/g, respectively; P < .01). Lastly, cleavable membranous bALP was dramatically decreased in HMPOS cells after exposure to ABT‐627 alone or in conjunction with ET‐1 when compared to control‐ or vehicle‐exposed conditions (Fig S3B,C).


Figure 5: Cytochemical staining intensity representative of alkaline phosphatase activities (A) in the HMPOS cell line following exposure to various conditions including ABT‐627. Marked reduction in cytochemical staining can be observed visually by inverted light microscopy (A; left panel) and digitally rendered pixel intensities (A; right panel), and (B) in the HMPOS cell line following exposure to ABT‐627 alone or in combination with ET‐1. Confirmation that ABT‐627 reduces bone alkaline phosphatase activities in the HMPOS cell line by (C) enzyme‐linked immunoassay. Significance defined as *P < 0.05 and **P < 0.01.


Serum bALP Activity Correlates with Primary Tumor Bone Mineral Density

Primary tumor skeletal characteristics of 95 dogs diagnosed with appendicular OS were analyzed by digital radiography (Fig6A) and DEXA scans. Concurrently, serum bALP activity was measured before to any interventional treatment. Primary tumor relative bone mineral density (rBMD) quantified by DEXA scans was used to arbitrarily categorize skeletal lesions as osteolytic (n = 25), mixed (n = 50), or osteoblastic (n = 20) by segregation of data into quartiles (Fig6B). When evaluating primary tumors categorized as osteoblastic and osteolytic, circulating serum bALP activities were increased in dogs with osteoblastic primary tumor phenotype (median, 26.1 U/L; range, 11.2–86.3 U/L) as compared to dogs with osteolytic lesions (median, 15.7 U/L; range, 4.8–64.3 U/L; P = .005; (Fig6C). Circulating bALP activities associated with a mixed primary tumor phenotype (median, 21.5 U/L; range, 4.7–93.2 U/L) were not different than bALP activities of dogs with primary tumors categorized as either osteolytic or osteoblastic (Fig S4). Given the reported contribution of the primary tumor size to circulating bALP activity,11 primary tumor rBMDs were normalized for tumor length (RECIST) and then compared with circulating bALP activities  (Fig6D). A positive correlation between normalized rBMD and bALP activity was identified (r = 0.38; P = .009). No difference in circulating ET‐1 concentrations was identified between dogs with osteoblastic (1.26 ± 0.1 fmol/mL) versus osteolytic (1.31 ± 0.1 fmol/mL) primary bone tumors (P = .8).


Figure 6: (A) Representative examples of primary OS bone lesions with either osteoblastic or osteolytic radiographic appearances. (B) Classification of 95 dogs with spontaneous appendicular OS with dual energy x‐ray absorptiometry scans into quartiles representing osteolytic (n = 25), mixed (n = 50), and osteoblastic (n = 20) primary tumor relative bone mineral densities. (C) Dogs with osteoblastic primary tumor phenotypes have increased serum bone alkaline phosphatase in comparison with dogs possessing osteolytic bone tumors. (D) Positive association between serum bone alkaline phosphatase and normalized relative bone mineral densities adjusted for primary tumor size (RECIST). Significance defined as **P < 0.01.



With regard to malignant skeletal pathology, ET‐1 serves as a key driver for the development and progression of skeletal metastases, particularly for prostate carcinoma in humans.19, 29 Within the bone microenvironment, carcinoma cells actively release ET‐1 which promotes the successful development of osteoblastic metastases through the induction of reparative osteoblast proliferation and concurrent downregulation of osteoclast activities.18, 19, 29, 30 The role of ET‐1 in OS biology is less well characterized, but nonetheless indicates the capacity of ETAR signaling to support several in vitro protumorigenic responses including invasion and chemoresistance.31, 32, 33, 34 Extending upon these in vitro observations, a limited number of in vivo studies have evaluated the participation of ETAR signaling in the metastatic biology of OS.35, 36 Collectively, ET‐1 and ETAR appear to be involved in critical cellular events that favor OS progression and metastases.

Our findings complement and extend existing knowledge regarding ETAR signaling in OS, and in particular the relevance of ET‐1 in the biology of canine OS.37, 38, 39 As indicated by the protein expression studies, canine OS cell lines and spontaneous tumor samples co‐express both receptor and cognate ligand, findings that substantiate the potential existence of a paracrine or autocrine signaling pathway that could be subverted by malignant osteoblasts. The biologic consequences of ET‐1 in canine OS cells specifically was evaluated in our study with emphasis on the MAPK pathway after stimulation or inhibition of ETAR signaling. Upon stimulation with ET‐1, canine OS cells (specifically HMPOS and D17 cell lines) had the capacity to rapidly upregulate the expression of c‐fos and phospho‐c‐jun, which can heterodimerize and form activator protein‐1 (AP‐1), a transcription factor responsible for multiple cellular responses including differentiation, proliferation, migration, and survival.40, 41 The inhibition of ETAR signaling in canine OS cells was achievable with ABT‐627, which dramatically blocked the expression of c‐fos and phospho‐c‐jun after ET‐1 exposure in the majority of cell lines investigated. These findings support manipulation of ETAR signaling as a potential therapeutic intervention in canine OS.

Recogition of ETAR blockade as a potential therapeutic strategy for canine OS was supported by results derived from both colony forming and migration assays. Canine OS cells exposed to ABT‐627 demonstrated a dose‐dependent decrease in the number of colonies formed, indicative of altered cell survival and proliferative capacities as well as impeded cell migration. Although differences were observed among cell lines with regard to their sensitivity to ET‐1 stimulation or to ABT‐627‐mediated inhibition, a consistent and directional effect elicited by ETAR signaling blockade was conserved across all cell lines tested. However, the observed biologic effects exerted by ABT‐627 might not be solely attributed to ETAR inhibition, because off‐target effects of ABT‐627 likely contributed to some of the cellular behavioral changes identified in our study. Nonetheless, the in vitro findings documented in our study are analogous with the findings derived from murine preclinical models of prostate skeletal metastases, in which ETAR antagonism mitigated reparative osteoblast proliferation and survival induced by metastatic carcinoma cells within the bone microenvironment.42, 43 Collectively, these in vitro and in vivo observations lend credence to the possibility that ETAR blockade might have the potential to slow the growth and progression OS lesions arising within local and metastatic niches.

In addition to cell survival, proliferation, and migration, ETAR blockade attenuated the expression and activity of membranous bALP. Different proteins are expressed along a continuum in association with osteoblast differentiation, with bALP serving as an early marker of pre‐osteoblast maturation, which enzymatically contributes to extracellular matrix maturation.44, 45 Based upon the in vitro findings of our study, canine OS cells responsive to ET‐1‐mediated signaling would have the potential for increased survival and proliferative capacity, cell mobility, and bALP activity. Given this triad of cellular characteristics, the negative prognosis associated with increased bALP activity in dogs with OS could simply be an epiphenomenon of active ETAR signaling, in which OS cells under the influence of ET‐1 would be endowed with “driver” protumorigenic properties and “passenger” bALP expression. The supposition of bALP serving in a “passenger” capacity is supported by recent studies in which the OS cell lines or primary tumor samples derived from dogs with either normal or increased bALP activity failed to show differences in biologic behaviors and gene expression profiles.46, 47

Given the role of ET‐1 in osteoblast survival and proliferation, as well as the participation of bALP in extracellular matrix maturation, circulating bALP activity in dogs with OS could simply be a surrogate for the prevailing bone characteristics within the localized tumor microenvironment. Data derived from our study identified a wide range of rBMD in dogs with OS, which could be phenotypically classified (osteolytic, mixed, osteoblastic) using an adapted methodology employed for the categorization of skeletal metastases in people.48, 49 Dogs with osteoblastic tumor microenvironmental responses had higher bALP activity in comparison to dogs with predominant focal osteolytic characteristics, and a positive correlation between the degree of local tumor microenvironment osteoproduction and circulating bALP activity was identified. These findings suggest that in addition to primary OS size, which has been reported to correlate with serum bALP activity,11 another host/tumor factor that might contribute to serum bALP activity is the net osteoproductivity occurring within the localized tumor microenvironment.

Although our study provides new information pertaining to canine OS biology, several limitations should be acknowledged. First, the number of cell lines utilized in our investigation was limited, and therefore strong conclusions regarding the biologic consequences documented in this investigation associated with ETAR blockade might not be conserved across all OS histologies. Given the heterogeneous nature of OS, responses to ETAR blockade could be variable. This supposition would be supported by our findings because 60% of the spontaneously arising OS samples demonstr ated either no (1/10) or variable expression (5/10) of ETAR. Second, the concentrations of ABT‐627 utilized in vitro to demonstrate the biologic consequences of inhibiting ET‐1‐mediated signaling were relatively high (10–40 μM) and used for prolonged exposure durations (1–24 hours), which are not likely achievable in vivo based upon existing pharmacokinetic data derived from human beings.50, 51 Additionally, although ABT‐627 is a potent and selective ETAR inhibitor, the micromolar concentrations evaluated in our study might have induced off‐target effects, which could have contributed to the altered cellular behaviors documented. As such, the biologic consequences associated with ETAR blockade observed in vitro might not be replicated in dogs with spontaneous OS receiving treatment with ETAR inhibiting strategies. Third, our study did not provide any definitive evidence that increased serum bALP activity could be an epiphenomenon of ET‐1‐mediated signaling. Nonetheless, this speculation remains intriguing and warrants additional future studies. Last, although there proved to be a positive association between serum bALP activity and the osteoproductive tumor microenvironment, it was not possible to associate the contributions of either normal reparative osteoblasts or malignant OS cells with the observed increases in bALP activity. Despite being unable to define the exact cellular origin responsible for bALP liberation, our study identified a novel host/tumor factor that contributes to circulating bALP activity in dogs with OS.

In conclusion, our results showed that canine OS cells express both ET‐1 and ETAR, which are involved in protumorigenic processes including cell survival, proliferation, and migration. These processes were mechanistically associated with downstream changes in the MAPK pathway after ET‐1 stimulation or blockade. Expression of bALP is partially regulated by ETAR‐mediated signaling, and a plausible link between bALP‐associated negative prognosis and the protumorigenic influences of ET‐1 could be proposed. Finally, a new host/tumor factor, specifically osteoproductivity associated with the local tumor microenvironment, contributes to circulating bALP activity in dogs with spontaneously arising OS. Collectively, these findings provide new information and in part elucidate the biologic mechanisms involved in bALP liberation in dogs with OS.


Fig S1. Microscopic evaluation of canine OS primary tumors with hematoxylin and eosin (A‐C), and immunohistochemistry for ETAR (D, E) and endothelin‐1 (F) expressions.


Fig S2. ETAR signaling pathway is functional, but differ in response magnitude in (A) D17 and (B) Abrams canine OS cells. Variable upregulation of c‐fos protein expression (C) in all 3 OS cell lines following either 10% FBS or ET‐1 (100 nM) stimulation.


Fig S3. (A) Relative bone alkaline phosphatase expressions in 3 canine OS cell lines, with HMPOS demonstrating the most robust protein expression. (B‐C) In the HMPOS cell line, membranous cleavage of bone alkaline phosphatase is reduced following exposure to ABT‐627 alone or in combination with endothelin‐1.


Fig S4. Comparison of serum bALP among 3 different osteophenotypes of canine appendicular osteosarcoma. Significance defined as *P < 0.05.



The authors thank the technicians and veterinary oncology residents of the University of Illinois Cancer Care Clinic for their contributions to this study.

Conflict of Interest Declaration: Dr Fan is an associate editor for the Journal of Veterinary Internal Medicine.

Off‐label Antimicrobial Declaration: Authors declare no off‐label use of antimicrobials.



This study was conducted in the Comparative Oncology Research Laboratory and Veterinary Teaching Hospital at the University of Illinois, Urbana, IL 61802.



1Sigma Aldrich, Saint Louis, MO

2MedchemExpress, Monmouth Junction, NJ

3Life Technologies, Grand Island, NY

4Cell Signaling Technology, Danvers, MA

5Abcam, Cambridge, MA

6Bio‐Rad, Hercules, CA

7National Institutes of Health, Bethesda, MD

8GE Healthcare Life Sciences, Pittsburgh, PA

9Thermo Scientific, Schaumburg, IL

10ABC Vector Laboratories, Burlingame, CA

11MicroVue BAP EIA kit, Quidel, San Diego, CA

12QDR‐4500W, Hologic, Bedford, MA

13Endothelin ELISA (1–21), ALPCO, Salem, NH

14GraphPad InStat, Version 3.10, La Jolla, CA



1. Bellows CG, Aubin JE, Heersche JN. Initiation and progression of mineralization of bone nodules formed in vitro: the role of alkaline phosphatase and organic phosphate. Bone Miner 1991;14:27–40. [PubMed]
2. Wennberg C, Hessle L, Lundberg P, et al. Functional characterization of osteoblasts and osteoclasts from alkaline phosphatase knockout mice. J Bone Miner Res 2000;15:1879–1888. [PubMed]
3. Fontana A, Delmas PD. Markers of bone turnover in bone metastases. Cancer 2000;88:2952–2960. [PubMed]
4. Leung KS, Fung KP, Sher AH, et al. Plasma bone‐specific alkaline phosphatase as an indicator of osteoblastic activity. J Bone Joint Surg Br 1993;75:288–292. [PubMed]
5. Morote J, Lorente JA, Encabo G. Prostate carcinoma staging. Clinical utility of bone alkaline phosphatase in addition to prostate specific antigen. Cancer 1996;78:2374–2378. [PubMed]
6. Boerman I, Selvarajah GT, Nielen M, et al. Prognostic factors in canine appendicular osteosarcoma – a meta‐analysis. BMC Vet Res 2012;8:56. [PubMed]
7. Ehrhart N, Dernell WS, Hoffmann WE, et al. Prognostic importance of alkaline phosphatase activity in serum from dogs with appendicular osteosarcoma: 75 cases (1990–1996). J Am Vet Med Assoc 1998;213:1002–1006. [PubMed]
8. Garzotto CK, Berg J, Hoffmann WE, et al. Prognostic significance of serum alkaline phosphatase activity in canine appendicular osteosarcoma. J Vet Intern Med 2000;14:587–592. [PubMed]
9. Bieling P, Rehan N, Winkler P, et al. Tumor size and prognosis in aggressively treated osteosarcoma. J Clin Oncol 1996;14:848–858. [PubMed]
10. Kaste SC, Liu T, Billups CA, et al. Tumor size as a predictor of outcome in pediatric non‐metastatic osteosarcoma of the extremity. Pediatr Blood Cancer 2004;43:723–728. [PubMed]
11. Sternberg RA, Pondenis HC, Yang X, et al. Association between absolute tumor burden and serum bone‐specific alkaline phosphatase in canine appendicular osteosarcoma. J Vet Intern Med 2013;27:955–963. [PubMed]
12. Sasaki T, Hong MH. Localization of endothelin‐1 in the osteoclast. J Electron Microsc 1993;42:193–196. [PubMed]
13. Takuwa Y, Ohue Y, Takuwa N, et al. Endothelin‐1 activates phospholipase C and mobilizes Ca2 + from extra‐ and intracellular pools in osteoblastic cells. Am J Physiol 1989;257:E797–E803. [PubMed]
14. Alam AS, Gallagher A, Shankar V, et al. Endothelin inhibits osteoclastic bone resorption by a direct effect on cell motility: implications for the vascular control of bone resorption. Endocrinology 1992;130:3617–3624. [PubMed]
15. Shioide M, Noda M. Endothelin modulates osteopontin and osteocalcin messenger ribonucleic acid expression in rat osteoblastic osteosarcoma cells. J Cell Biochem 1993;53:176–180. [PubMed]
16. von Schroeder HP, Veillette CJ, Payandeh J, et al. Endothelin‐1 promotes osteoprogenitor proliferation and differentiation in fetal rat calvarial cell cultures. Bone 2003;33:673–684. [PubMed]
17. Nelson JB, Hedican SP, George DJ, et al. Identification of endothelin‐1 in the pathophysiology of metastatic adenocarcinoma of the prostate. Nat Med 1995;1:944–949. [PubMed]
18. Nelson JB, Nguyen SH, Wu‐Wong JR, et al. New bone formation in an osteoblastic tumor model is increased by endothelin‐1 overexpression and decreased by endothelin A receptor blockade. Urology 1999;53:1063–1069. [PubMed]
19. Yin JJ, Mohammad KS, Kakonen SM, et al. A causal role for endothelin‐1 in the pathogenesis of osteoblastic bone metastases. Proc Natl Acad Sci USA 2003;100:10954–10959. [PubMed]
20. Barroga EF, Kadosawa T, Okumura M, et al. Establishment and characterization of the growth and pulmonary metastasis of a highly lung metastasizing cell line from canine osteosarcoma in nude mice. J Vet Med Sci 1999;61:361–367. [PubMed]
21. Riggs JL, McAllister RM, Lennette EH. Immunofluorescent studies of RD‐114 virus replication in cell culture. J Gen Virol 1974;25:21–29. [PubMed]
22. Legare ME, Bush J, Ashley AK, et al. Cellular and phenotypic characterization of canine osteosarcoma cell lines. J Cancer 2011;2:262–270. [PubMed]
23. Coffman L, Mooney C, Lim J, et al. Endothelin receptor‐A is required for the recruitment of antitumor T cells and modulates chemotherapy induction of cancer stem cells. Cancer Biol Ther 2013;14:184–192. [PubMed]
24. Liang CC, Park AY, Guan JL. In vitro scratch assay: a convenient and inexpensive method for analysis of cell migration in vitro. Nat Protoc 2007;2:329–333. [PubMed]
25. Barger A, Graca R, Bailey K, et al. Use of alkaline phosphatase staining to differentiate canine osteosarcoma from other vimentin‐positive tumors. Vet Pathol 2005;42:161–165. [PubMed]
26. Allen LC, Allen MJ, Breur GJ, et al. A comparison of two techniques for the determination of serum bone‐specific alkaline phosphatase activity in dogs. Res Vet Sci 2000;68:231–235. [PubMed]
27. Fan TM, de Lorimier LP, Charney SC, et al. Evaluation of intravenous pamidronate administration in 33 cancer‐bearing dogs with primary or secondary bone involvement. J Vet Intern Med 2005;19:74–80. [PubMed]
28. Prosek R, Sisson DD, Oyama MA, et al. Plasma endothelin‐1 immunoreactivity in normal dogs and dogs with acquired heart disease. J Vet Intern Med 2004;18:840–844. [PubMed]
29. Guise TA, Yin JJ, Mohammad KS. Role of endothelin‐1 in osteoblastic bone metastases. Cancer 2003;97:779–784. [PubMed]
30. Bagnato A, Rosano L. The endothelin axis in cancer. Int J Biochem Cell Biol 2008;40:1443–1451. [PubMed]
31. Felx M, Guyot MC, Isler M, et al. Endothelin‐1 (ET‐1) promotes MMP‐2 and MMP‐9 induction involving the transcription factor NF‐kappaB in human osteosarcoma. Clin Sci 2006;110:645–654. [PubMed]
32. Zhao Y, Liao Q, Zhu Y, et al. Endothelin‐1 promotes osteosarcoma cell invasion and survival against cisplatin‐induced apoptosis. Clin Orthop Relat Res 2011;469:3190–3199. [PubMed]
33. Marion A, Dieudonne FX, Patino‐Garcia A, et al. Calpain‐6 is an endothelin‐1 signaling dependent protective factor in chemoresistant osteosarcoma. Int J Cancer 2012;130:2514–2525. [PubMed]
34. Zhou Y, Zang X, Huang Z, et al. TWIST interacts with endothelin‐1/endothelin A receptor signaling in osteosarcoma cell survival against cisplatin. Oncol Lett 2013;5:857–861. [PubMed]
35. Zang X, Zhou Y, Huang Z, et al. Endothelin‐1 single nucleotide polymorphisms and risk of pulmonary metastatic osteosarcoma. PLoS One 2013;8:e73349. [PubMed]
36. Li Y, Liao Q, Li K, et al. Knockdown of endothelin A receptor expression inhibits osteosarcoma pulmonary metastasis in an orthotopic xenograft mouse model. Mol Med Rep 2012;5:1391–1395. [PubMed]
37. Fukumoto S, Hanazono K, Miyasho T, et al. Serum big endothelin‐1 as a clinical marker for cardiopulmonary and neoplastic diseases in dogs. Life Sci 2014;118:329–332. [PubMed]
38. Shor S, Fadl‐Alla BA, Pondenis HC, et al. Expression of nociceptive ligands in canine osteosarcoma. J Vet Intern Med 2015;29:268–275. [PubMed]
39. Wittenburg LA, Ptitsyn AA, Thamm DH. A systems biology approach to identify molecular pathways altered by HDAC inhibition in osteosarcoma. J Cell Biochem 2012;113:773–783. [PubMed]
40. Curran T, Franza BR Jr. Fos and Jun: the AP‐1 connection. Cell 1988;55:395–397. [PubMed]
41. Karin M, Liu Z, Zandi E. AP‐1 function and regulation. Curr Opin Cell Biol 1997;9:240–246. [PubMed]
42. Carducci MA, Jimeno A. Targeting bone metastasis in prostate cancer with endothelin receptor antagonists. Clin Cancer Res 2006;12:6296s–6300s. [PubMed]
43. Mohammad KS, Guise TA. Mechanisms of osteoblastic metastases: role of endothelin‐1. Clin Orthop Relat Res 2003;(415 Suppl):S67–S74. [PubMed]
44. Owen TA, Aronow M, Shalhoub V, et al. Progressive development of the rat osteoblast phenotype in vitro: reciprocal relationships in expression of genes associated with osteoblast proliferation and differentiation during formation of the bone extracellular matrix. J Cell Physiol 1990;143:420–430. [PubMed]
45. Stein GS, Lian JB, Owen TA. Relationship of cell growth to the regulation of tissue‐specific gene expression during osteoblast differentiation. FASEB J 1990;4:3111–3123. [PubMed]
46. de Sa Rodrigues LC, Holmes KE, Thompson V, et al. Tumourigenic canine osteosarcoma cell lines associated with frizzled‐6 up‐regulation and enhanced side population cell frequency. Vet Comp Oncol 2015. Feb 16. doi: 10.1111/vco.12138. [Epub ahead of print]. [PMC free article] [PubMed]
47. Holmes KE, Thompson V, Piskun CM, et al. Canine osteosarcoma cell lines from patients with differing serum alkaline phosphatase concentrations display no behavioural differences in vitro. Vet Comp Oncol 2015. Feb 2. doi: 10.1111/vco.12132. [Epub ahead of print]. [PMC free article] [PubMed]
48. Chang CH, Tsai CS, Jim YF, et al. Lumbar bone mineral density in prostate cancer patients with bone metastases. Endocr Res 2003;29:177–182. [PubMed]
49. Perk H, Yildiz M, Kosar A, et al. Correlation between BMD and bone scintigraphy in patients with prostate cancer. Urol Oncol 2008;26:250–253. [PubMed]
50. Verhaar MC, Grahn AY, Van Weerdt AW, et al. Pharmacokinetics and pharmacodynamic effects of ABT‐627, an oral ETA selective endothelin antagonist, in humans. Br J Clin Pharmacol 2000;49:562–573. [PubMed]
51. Ryan CW, Vogelzang NJ, Vokes EE, et al. Dose‐ranging study of the safety and pharmacokinetics of atrasentan in patients with refractory malignancies. Clin Cancer Res 2004;10:4406–4411. [PubMed]

Share this article / Teilen Sie diesen Artikel