This article is mentioned in:

Introduction

Bone sarcomas are a heterogenous group of different types of malignant bone tumor characterized by various degrees of mesenchymal differentiation (1). Chondrosarcoma is a rare but aggressive malignant tumor of the cells that produce the cartilage matrix; it is the third most common bone tumor, accounting for >30% of primary bone cancers in the USA, and is resistant to chemotherapy and radiation; therefore, novel therapeutic approaches are urgently required (1–4).

Cancer stem cells (CSCs) have the ability to self-renew and differentiate, which are typical characteristics of any stem cell (5). It has been reported that a tumor contains different stem cell-like populations due to decreased cell differentiation within a tumor, and this heterogenous population includes CSCs (5). DNA damage may render tissue stem cells to become CSCs during the process of differentiation and to accumulate to form different types of tumor (6). Increasing evidence suggests that CSCs exist as a sub-population of quiescent cells within the dominant tumor bulk of heterogeneous tumor cells. These typically dormant cells are resistant to standard antitumor therapies such as chemotherapy and radiation, and they appear capable of self-renewal and differentiation, which suggests that CSCs are responsible for tumor repopulation following removal of the bulk tumor (7–9).

Proline-rich polypeptide-1 (PRP-1), also termed galarmin, is synthesized by the brain neurosecretory cells and comprises 15 amino acids (10). PRP-1 is an mTOR kinase (mTORC1) inhibitor in chondrosarcoma that inhibits cell proliferation (11). Furthermore, PRP-1 has been demonstrated to upregulate several tumor suppressors of the desmosomal protein family, such as desmoglein, and plakoglobin (12), as well as tumor suppressors involved in inflammatory pathways (13) and microRNAs (miRNAs), and to downregulate onco-miRNAs in the human chondrosarcoma JJ012 cell line through inhibition of mTORC1 (3,14). PRP-1 also serves an antiproliferative role via downregulation of the embryonic stem cell marker, miRNA (miR)-302-c and its targets nanog homeobox (NANOG), c-Myc and polycomb protein B lymphoma Mo-MLV insertion region 1 homolog (BMI-1) (3). Pattern recognition receptors of the adaptive immune system, including Toll-like receptor 1 (TLR1), TLR2, TLR6 and secreted mucin 5B, have been recently identified as binding partners for PRP-1, resulting in tumor suppressive effects (4). PRP-1-treated cells have been demonstrated to accumulate in the S phase, thus delaying cell cycle progression (15). The cytostatic, antiproliferative, immunomodulatory and tumor suppressive roles of PRP-1 indicate its potential as a therapeutic agent against human chondrosarcoma cells resistant to radiation and chemotherapy (3,4,10–12,14). The aforementioned properties were further investigated in the present study to determine the ability of PRP-1 to target chondrosarcoma CSCs.

Aldehyde dehydrogenase (ALDH) is an established marker of CSCs in several neoplasms. Cells expressing high levels of ALDH have been isolated in a number of human sarcoma cell lines, including the human chondrosarcoma SW-1353 cell line (16). Identification of normal and malignant stem/progenitor cells by the same marker reinforces the concept that stem and progenitor cells are primary targets of transformation, which further supports the CSC hypothesis. Furthermore, the ability to identify stem/progenitor cells by ALDH1 expression permits analysis of cancer initiation and progression from the normal to the pre-malignant and the malignant states (2,17). A widely accepted method for identifying CSCs is based on detecting the enzymatic activity of ALDH1, a detoxifying enzyme responsible for the oxidation of intracellular aldehydes (16). The self-renewal capacity, proliferative and metastatic roles of ALDH demonstrate its potential as a therapeutic target against human chondrosarcoma (18).

The present study aimed to determine the underlying molecular mechanism by which PRP-1 decreases the CSC population, with particular focus on the mammalian Switch/sucrose non fermentable (SWI/SNF)-complex (19), which can either activate or repress transcription, and to identify additional biomarkers for chondrosarcoma.

Materials and methods

Tissue culture

The human JJ012 chondrosarcoma cells were obtained from Dr Joel Block's Laboratory (Rush University, Chicago, USA). The cells were maintained in complete growth medium containing DMEM + GlutaMAX™ supplemented with F-12 + GlutaMAX™ Nutrient mixture (Ham) (Thermo Fisher Scientific, Inc.), 10% FBS (Sigma-Aldrich; Merck KGaA), 1% penicillin/streptomycin (Thermo Fisher Scientific, Inc.), 25 µg/ml ascorbic acid, 100 ng/ml insulin and 100 nM hydrocortisone (Sigma-Aldrich; Merck KGaA). The cells were incubated at 37°C in a humidified atmosphere with 5% CO2 and periodically checked for mycoplasma. Estrogen negative MDA-MB-231 breast cancer cells (ATCC® CRM-HTB-26) were obtained from ATCC. The medium for this cell line was ATCC-formulated Leibovitz's L-15 Medium (cat. no. 30-2008; ATCC) supplemented with 10% FBS. The MDA-MB-231 breast cancer cells were incubated at 37°C in a humidified atmosphere with 5% CO2. Human B cell lymphoma (HT; cat. no. ACC567), Ly1 (cat. no. ACC722) and Ly3 (cat. no. ACC761) cell lines were obtained from DSMZ. Ly10 cells were a gift from Dr Messner (Ontario Cancer Institute, Canada), and HBL-1 cells were a gift from Dr Vega (MD Anderson Cancer Center, Houston, TX, USA). Melanoma metastatic cell line C81618 was provided by Dr Liu, University of Miami, Miller school of Medicine.

Mycoplasma detection assay

JJ012 chondrosarcoma cells were seeded 1×105 cells/ml in full growth media and treated with 25 µg/ml Plasmocin™ (InvivoGen) for 14 days. Plasmocin™-containing medium was replaced every 3–4 days. MycoAlert® Mycoplasma Detection kit (cat. no. LT07-118; lot. no. 0000678423; Lonza, Inc.) was used according to the manufacturer's instructions. The microplate was placed in the Molecular devices Spectra Max microplate luminometer (Lonza, Inc.) and programmed for Reading A analysis. A total of 100 µl MycoAlert™ substrate was added to each sample and incubated for 10 min, following which the luminometer was programmed for Reading B analysis, and the Reading B/Reading A ratio was calculated.

Cell viability

Cells were cultured under the same conditions as previously described and stained with 0.4% trypan blue stain (Gibco; Thermo Fisher Scientific, Inc.) at a 1:1 ratio. After a 72-h incubation at 37°C in the 5% CO2 incubator, cells were observed using the Luna II Automated Cell Counter™ (Logos Biosystem) according to the manufacturer's protocol to generate cell count and viability data.

PRP-1 treatment

A subset of cultured human JJ012 chondrosarcoma cells were treated with 10 µg/ml PRP-1 as previously described (10) and incubated at 37°C in a humidified atmosphere of 5% CO2 for 24 h prior to performing the Aldefluor™ assay. Cell proliferation was measured in cells seeded at 5×104 cells/ml using the Rapid Cell Proliferation kit (EMD Millipore), according to the manufacturer's instructions.

Aldefluor™ assay and fluorescence-activated cell sorting (FACS)

The Aldefluor™ assay (StemCell Technologies, Inc.) was performed according to the manufacturer's protocol, in order to detect cells with ALDH activity. Briefly, the JJ012 cells were harvested and resuspended in Aldefluor™ assay buffer at a density of 1×106/ml. The cells were incubated with 25 µl DMSO with 25 µl 2N HCl for 15 min, followed by addition of 360 µl assay buffer in order to activate the Aldefluor™ reagent, and incubated for 45 min at 37°C. N,N-Diethylaminobenzaldehyde (DEAB), a specific ALDH inhibitor, was used as the negative control. Following incubation, all tubes were centrifuged for 5 min at 250 × g at 4°C, the supernatant was discarded and the cells were resuspended in Aldefluor™ assay buffer. The cells were then transferred to 5 ml Falcon polystyrene round bottom tubes with cell strainer caps. After labeling, the samples with and without PRP-1 treatment were sorted using a FACSAria II flow cytometer (BD Biosciences) with FACSDiva software (version 6.1.3; BD Biosciences) into ALDHlow and ALDHhigh cells. Data analysis was performed using FlowJo software (version 10; FlowJo, LLC).

Gel electrophoresis and western blotting

Chondrosarcoma JJ012 cells were cultured to 100% confluency. Cells were collected using trypsin, seeded into petri dishes at a density of 1×106 cells/ml and incubated for 24 h at 37°C in a 5% CO2 incubator. After 24 h, cells were washed with ice-cold PBS, and protease inhibitor was added to the CelLytic M cell lysis reagent (Sigma-Aldrich; Merck KGaA) at a 1:100 ratio, and kept in a 5% CO2 incubator for 10 min. Following collection of cells with a rubber scraper and cell membrane lysis with an 18-gauge needle, the cells were centrifuged at 15,000 × g at 4°C for 10 min. The supernatant was collected, and the protein content was measured using a NanoDrop® spectrophotometer. The supernatant was frozen at −80°C until loading into gels (20 µg/lane). Polyacrylamide gel electrophoresis and western blotting reagents were supplied by Lonza Group, Ltd., and the experiments were performed according to the manufacturer's protocols. The reagents used were as follows: Pager Gold Precast Gels (10% Tris-Glycine; Lonza Group, Ltd.); ProSieve Quad Color Protein marker (4.6–300 kD; Lonza Group, Ltd.); 20X reducing agent for ProSieve ProTrack Dual Color Loading buffer (Lonza Group, Ltd.); ProTrack Loading buffer (Lonza Group, Ltd.); ProSieve ProTrack Dual Color Loading buffer EX running buffer (Lonza Group, Ltd.); ProSieve EX Western Blot Transfer buffer (Lonza Group, Ltd.) and Immobilon®-P Polyvinylidene difluoride membranes (Sigma-Aldrich; Merck KGaA).

For the blocking step, Western Blocker solution (Sigma-Aldrich; Merck KGaA) was used. Primary antibodies were diluted in blocking buffer at the ratio of 1:1,000, and the membranes were incubated at 4°C overnight with gentle agitation. The secondary antibodies horseradish peroxidase-conjugated goat anti-rabbit IgG (cat. no. A0545; Sigma-Aldrich; Merck KGaA) and anti-mouse IgG (cat. no. A4416; Sigma-Aldrich; Merck KGaA) were diluted to 1:5,000, and the membranes were incubated for 2 h at room temperature with gentle agitation. Western blot visualization was performed using ECL reagents (GE Healthcare Life Sciences) for 1 h at room temperature.

Antibodies for western blotting

The primary antibodies used for western blot analysis were as follows: CD133 (cat. no. 19898; Abcam), CD144 (VE-cadherin) monoclonal (16B1) PE-Cyanine7 (cat. no. 25-1449-4; Invitrogen; Thermo Fisher Scientific, Inc.), c-Kit (Ab 81) (cat. no. 13508; Santa Cruz Biotechnology, Inc.), rabbit LGR5 (GPR49) (cat. no. 9205024; MyBioSource, Inc.), SOX2 (cat. no. 97959; Abcam), human ALDH1A1 (cat. no. MAB5869; R&D Systems, Inc.), polycomb complex protein oncogene BMI-1 (BMI-1; cat. no. 38295; Abcam), c-Myc (9E10) (cat. no. 40; Santa Cruz Biotechnology, Inc.), CD44 (cat. no. 6124; Abcam), FOXO3 (cat. no. MAB6165; R&D Systems), FOXC1 (cat. no. MAB 6329; R&D Systems), NANOG clone 7F7.1 (cat. no. MABD24; Sigma-Aldrich; Merck KGaA), anti CD34 + mesenchymal stem cell monoclonal antbody (STRO-1; cat. no. 39-8401; Thermo Fisher Scientific, Inc.), CD117 c-Kit (Ab 81) (cat. no. 13508; Santa Cruz Biotechnology, Inc.), BRG1-associated factor 170 (BAF170) or SMARCC2 (cat. no. 9401447; MyBioSource, Inc.), BRG1-associated factor 155 (BAF155) or SMARCC1 (cat. no. 8245507; MyBioSource, Inc.), protein Brahma homolog 1 (BRG1; cat. no. 4081; Abcam), Brahma protein (BRM) or SMARCA 2 (cat. no. 610390; BD Biosciences), ARF tumor suppressor, p14ARF antibody (ARF 4C6/4; cat. no. 53392; Santa Cruz Biotechnology, Inc.), cyclin-dependent kinase inhibitor 2A (CDKN2A) P15/P16 (p15, INK4b; p16, INK4a; cat. no. 377412; Santa Cruz Biotechnology, Inc.), p-topoisomerase IIα (TOP2A) (Ser1469) (cat. no. 13072; Cell Signaling Technology, Inc.), tubulin (cat. no. T5168; Sigma Aldrich; Merck KGaA) and topoisomerase IIα SER1469 [cat. no. 13072 (E); Cell Signaling Technology, Inc.].

Densitometry analysis of western blots

Quantitative analysis and densitometry were obtained using integrated density analysis on ImageJ software 1.52e (National Institutes of Health) to calculate relative optical density (OD) of the previously mentioned proteins compared with the housekeeping protein tubulin; the bulk untreated JJ012 was used as the control.

Apoptosis assay and cell cycle analysis

Staining was performed using the Annexin V Apoptosis Detection kit according to the manufacturer's protocol. Modified Annexin V/propidium iodide (PI) apoptosis assay (cat. no. 88-8007-72; eBioscience; Thermo Fisher Scientific, Inc.) was used for apoptosis experiments with 1×106 cellss/ml. The LRS-II Analyzer was used with DiVa-8 software (BD Biosciences). The antibodies used for flow cytometric analysis were as follows: APC-conjugated human CD133 (cat. no. FAB11331A-025; R&D Systems, Inc.), Alexa Fluor® 647 mouse anti-human BMI-1 clone P51-311 (cat. no. 562637; BD Biosciences), CD144 (VE-cadherin) (16B1) phycoerythrin (PE)-cyanine7 (cat. no. 25-1449-4; Invitrogen; Thermo Fisher Scientific), BV605 mouse anti-human CD117 clone 104D2 (cat. no. 562687; BD Biosciences), PE-CF594 rat anti-human LGR5 (N-terminal) clone 8F2 (cat. no. 563470; BD Biosciences), PE mouse anti-SOX2 clone 245610 (cat. no. 560291; BD Biosciences), BV421 mouse anti-human CD49b (cat. no. 564119; BD Biosciences), CD10 (SN5c) PerCP-eFluor 710 (cat. no. 46-0108-41; Invitrogen; Thermo Fisher Scientific, Inc.), PE-CF594 mouse anti-human CD221 clone 1H7 (cat. no. 562535; BD Biosciences) and Alexa Fluor® 647 mouse anti-human CD271 clone C40-1457 (cat. no. 560326; BD Biosciences).

Aldefluor™ assay and fluorescence-activated cell sorting (FACS)

To measure cells with ALDH activity, the Aldefluor™ assay (cat. no. 01700; StemCell Technologies, Inc.) was performed according to manufacturer's protocol. Briefly, cells were harvested and resuspended in Aldefluor™ assay buffer at a density of 1×106 cells/ml. To activate the Aldefluor™ reagent, first 25 µl DMSO was added and incubated for 15 min with 25 µl 2N HCl, and then 360 µl assay buffer was added to the vial. The cells were incubated with the activated Aldefluor™ reagent for 45 min at 37°C. Diethylaminobenzaldehyde (DEAB), a specific ALDH inhibitor, was added as a negative control. Following incubation, all tubes were centrifuged for 5 min at 250 × g, at 4°C and the supernatant was removed; the cells were then resuspended in Aldefluor™ assay buffer, transferred and strained into a Falcon 5 ml polystyrene round bottom tube with a cell strainer cap (cat. no. 352235; Thermo Fisher Scientific, Inc.). After labeling, the samples were sorted using a Special Order Research Product (SORP) FACSAria II (BD Biosciences) using BD FACSDiva software (version 6.1.3; BD Biosciences) into ALDHlow and ALDHhigh cells with and without PRP-1 treatment. Data analysis was performed using FlowJo software version 10 (FlowJo LLC).

Bromodomain-containing protein 4 (BRD4) inhibitor screening

The cells were seeded at 4,000 cells/well in 384-well plates. BRD4 inhibitor screening was performed by BPS Bioscience Inc., using the BPS Bioscience TR-FRET Assay kit BRD4 according to the manufacturer's instructions.

Statistical analysis

All statistical analyses were performed using GraphPad Prism (version 8.0.2; GraphPad Software, Inc.). Statistical analyses were performed using individual unpaired t-tests for flow cytometry experiments (repeated 10 times). Statistical analyses of relative ODs were completed using one-way ANOVA, followed by Tukey's or Dunnett's post hoc multiple comparison test, and the data were expressed as 95% confidence intervals of the mean difference. All western blots were performed in triplicate. P<0.05 was considered to indicate a statistically significant difference. Data are presented as the mean ± standard error of the mean in all graphs with *P<0.05, **P<0.01 and ***P<0.001.

Results

The inhibitory action of PRP-1 is disease-specific

It has previously been reported that PRP-1 displays disease-specific inhibition of tumor growth; however, in chondrosarcoma, PRP-1 inhibits the stemness signature cluster of miR-302-367 (3). For example, in glioblastoma, miR-302-367 is strongly induced during serum-mediated stemness suppression, which prevents PRP-1 from exerting inhibitory effects on tumor cell proliferation (3). Thus, the present study assessed different types of malignancies, such as lymphoma and melanoma.

PRP-1 failed to inhibit the growth of two major subtypes of diffuse large B-cell lymphoma: Activated B-cell subtype lymphoma 3, 10 (Ly3, Ly10), human B line (HBL1) and germinal center subtype (Ly1 and HT cell lines). The results also demonstrated that proliferation of the metastatic melanoma cell line C81618 was not inhibited by PRP-1. The effect of PRP-1 on cell proliferation was assessed in vitro (data not shown).

ALDH1A1 is be the only biomarker for CSC population in human chondrosarcoma JJ012 cell line

In order to identify additional biomarkers for chondrosarcoma CSCs in the JJ012 cell line, two different approaches were employed, the traditional side population (SP) approach and the enzymatic ALDH1A1 approach. The SP analysis results demonstrated that no SPs for JJ012 cells were detected (data not shown) despite previous reports of successful identification of SP in sarcomas (20). Subsequently, known biomarkers of different types of cancer were assessed using the human chondrosarcoma JJ012 cell line. A number of well-known markers and the Aldefluor assay were employed in order to determine ALDH expression in human chondrosarcoma JJ012 cells, with and without PRP-1 treatment, and the cells were divided into ALDHlow and ALDHhigh populations, respectively.

Human chondrosarcoma JJ012 cells were stained with Aldefluor™ in the presence and absence of the ALDH inhibitor DEAB. The DEAB (+) group was used to determine ALDHlow and ALDHhigh populations (Fig. 1A). JJ012 fractions treated with PRP-1 were labeled as ALDHlow-PRP-1 (21). These gates were used in the analysis of surface staining of well-established CSC markers for sarcoma. Similar to the control MDA-MB-231 breast cancer cells, JJ012 cells expressed CD133, a known CSC marker (22–25). However, no difference was observed in CD133 expression between the ALDHhigh and ALDHlow groups (Fig. 1B). The cells were stained with other surface antibodies that also demonstrated no difference in the expression levels of BMI-1, SOX2, CD144, CD117, CD221, LGR5, CD271, CD10 and CD49b between the ALDHhigh and ALDHlow cells (Fig. 1C). Thus, none of the aforementioned markers could serve as a CSC marker.

Figure 1. Identification of additional biomarkers for human chondrosarcoma. (A-C) Flow cytometry experiments. (A) The DEAB (+) group was used to determine ALDHlow and ALDHhigh populations. (B) CD133 was not different between the two populations. (C) Staining with BMI-1, SOX2, CD144,CD117, CD221, LGR5, CD271, CD10 and CD49b on ALDHhigh and ALDHlow cells verified that none of them was a biomarker for JJ012 human chondrosarcoma cells. (D) Western blot analysis of LGR5, SOX2, CD44, FOX03, FOXC1, BMI-1, c-Myc and NANOG. Only BMI-1, c-Myc and NANOG in ALDHlow+PRP-1 exhibited decreases in protein expression expression compared with that in ALDHlow and ALDHhigh untreated cells. (E) Histograms of the relative OD in ALDH fractions. *P<0.05, **P<0.01 and ***P<0.001. ALDH, aldehyde dehydrogenase; OD, optical density; BMI-1, B lymphoma Mo-MLV insertion region 1 homolog; NANOG, nanog homeobox; LGR5, leucine-rich repeat-containing G protein-coupled receptor 5; FOXO3, forkhead box O3; FOXC1, forkhead box C1; DEAB, N,N-diethylaminobenzaldehyde.

Western blot analysis demonstrated decreased expression levels of BMI, c-Myc and NANOG in the ALDHlow-PRP-1 fraction following treatment with PRP-1 compared with untreated ALDHlow and ALDHhigh fractions (Fig. 1D). However, no differences in the expression levels of CD44, FOX03 and FOXC1 were demonstrated between the fractions. Of note, compared with that in the ALDHlow-untreated group, the expression of SOX2 stemness-associated factor in cancer was upregulated in ALDHhigh-untreated and ALDHlow-PRP-1, and was downregulated in the ALDHlow -untreated fraction (Fig. 1E). Individual one-way ANOVA of the protein markers demonstrated significant differences between ALDHlow-untreated, ALDHhigh-untreated and ALDHlow-PRP-1 cells in the expression levels of BMI-1, NANOG, LGR5, SOX2, c-Myc and FOXC1 (Table SI). No differences of means between the groups were detected for FOXO3. ALDHlow-PRP-1 cells exhibited a significantly lower BMI-1 expression level compared with ALDHhigh-untreated cells. Furthermore, NANOG expression was significantly lower in ALDHlow-PRP-1 cells compared with ALDHlow-untreated and ALDHhigh-untreated cells. ALDHhigh-untreated cells and ALDHlow-PRP-1 cells had significantly lower expression levels of LGR5 compared with ALDHlow-untreated cells (Fig. 1E).

In a JJ012 bulk untreated subpopulation, 4.2% untreated cells were ALDHlow and 42.9% ALDHhigh (Fig. 2A), whereas a PRP-1-treated subpopulation demonstrated a notable change in ALDHlow-PRP-1 (82.6%) and ALDHhigh-PRP-1 (0.6%) cells (Fig. 2B). Fig. 2A demonstrates one of the 10 representative experiments of the untreated JJ012 cells, whereas Fig. 2B shows one of the 10 representative experiments of the PRP-1 treated JJ012 cells; Fig. 2C is the average of the 10 experiments with the SEM represented. The individual results are presented to demonstrate how the gating was performed by the flow cytometry software. The mean value for ALDHlow-untreated cells was 45.03% vs. 87.53% for ALDHlow-PRP-1 treated cells. The mean value for ALDHhigh-untreated was 46.72% vs. 8.27% for ALDHhigh-PRP-1 treated cells (Fig. 2C; Table SII). Subpopulations were collected using flow cytometry and consisted of the following; ALDHhigh cells sorted from untreated JJ012 (ALDHhigh-untreated), ALDHlow cells sorted from untreated JJ012 (ALDHlow-untreated) and ALDHlow cells sorted from PRP-1-treated JJ012 (ALDHlow-PRP-1). ALDHhigh cells from PRP-1-treated JJ012 (ALDHhigh-PRP-1) were difficult to collect as a result of the exceedingly low levels of these CSC populations following treatment with PRP-1; however, this was eventually accomplished following multiple experiments and compilation of the sorting results (Table SII).

Figure 2. Flow cytometry data of JJ012 untreated and PRP-1 treated subpopulations. (A and B) Notable changes were observed between (A) ALDHlow 4.2% and ALDHhigh 42.9% in the untreated populations and (B) 82.6% in ALDHlow-PRP-1 and 0.6% in ALDHhigh-PRP-1. Panel A demonstrates one of the 10 representative experiments of the untreated JJ012 cells, whereas panel B demonstrates one of the 10 representative experiments of the PRP-1 treated JJ012 cells. These were performed to demonstrate how the gating was performed by the flow cytometry software. (C) Histogram of the mean percentages of cells in PRP-1-treated and untreated fractions (n=10). (D) Trypan blue staining of ALDH fractions and JJ012 untreated controls. (E) Histogram of the viability in ALDH-treated and untreated fractions and JJ012 control cells. (F) Western blot of ALDH1A1 expression in ALDHhigh-untreated and ALDHlow-PRP-1. (G) Histogram of relative OD in ALDHhigh-untreated and ALDHlow-PRP-1. *P<0.05, **P<0.01 and ***P<0.001. ALDH, aldehyde dehydrogenase; OD, optical density; PRP-1, proline-rich polypeptide-1.

The cell viability assay demonstrated an increase in non-viable cells in the PRP-1 treated cells compared with the untreated cells (Fig. 2D; Tables SIII and SIV). One-way ANOVA analysis demonstrated significant differences amongst the controls and subpopulations following 72 h PRP-1 treatment between the JJ012 control and PRP-1 treated JJ012 (Tables SIII and SIV; Fig. 1E).

Following subpopulation collection and cell lysis, western blot analysis demonstrated a significant decrease of ALDH1A1 protein expression between subpopulations with and without PRP-1 treatment (Fig. 2F). Densitometry analysis demonstrated a statistically significant difference in ALDH1A1 protein expression between ALDHhigh-untreated cells vs. ALDHlow-PRP-1 cells. When comparing the mean relative OD of ALDH between ALDHhigh-untreated (mean OD, 3.590±0.040; n=2) and ALDHlow-PRP-1 (mean OD, 1.224±0.009; n=2) cells, Student's t-test demonstrated a significant 2.93-fold decrease in ALDH expression in ALDHlow-PRP-1 cells [t(2)=57.23; P=0.0003; Fig. 2G; Table SV].

Epigenetic mechanism of CSC regulation by PRP-1 in chondrosarcoma JJ012 cell line

The epigenetic effect of PRP-1 on CSCs was assessed. In order to further elucidate the molecular mechanism by which PRP-1 decreases CSC-population, the effect of PRP-1 on important epigenetic readers, such as the BD and BET proteins (26). TOP2A is an important enzyme in the regulation of DNA structure and cell proliferation (27–31). However, the results of the present study failed to demonstrate downregulation of TOP2A activity following treatment with PRP-1 (data not shown).

Comparison of BAF170 expression between ALDHhigh-untreated and ALDHlow-PRP-1 cells demonstrated a significant decrease in the mean relative OD in ALDHlow-PRP-1 cells (Fig. 3). In addition, BRG1 expression was significantly decreased in ALDHlow-PRP-1 cells compared with that in ALDHhigh-untreated cells. ALDHlow-PRP-1 cells demonstrated a significant decrease in BRM expression compared with ALDHhigh-untreated cells (Table SVI). No differences were observed between fractions with or without PRP-1 treatment for BAF155 or STRO-1 mesenchymal stem cell factor (Fig. 3A). The downstream targets for BMI-1, such as p14 ARF and p15/p16 did not exhibit any changes at the protein level (Fig. 3B), which suggested that the oncogenic function of BMI-1 in chondrosarcoma was independent from these downstream targets.

Figure 3. (A) Western blot analysis of Switch/sucrose non-fermenting complex key components in ALDHhigh-untreated and ALDHlow-PRP-1 fractions. PRP-1 induced decreases in the protein expression of BAF 155, BRG 1 and BAF170 in ALDHlow-PRP-1 fractions compared with that in ALDHhigh-untreated fractions. Mesenchymal stem cell factor STRO-1 protein levels were not different between the fractions and were not affected by PRP-1. BMI-1 oncogene targets p14ARF, and p15/p16 displayed the same protein levels, and PRP-1 did not cause any changes. (B) Histograms of the relative OD in the western blots. *P<0.05, **P<0.01 and ***P<0.001. ALDH, aldehyde dehydrogenase; OD, optical density; PRP-1, proline-rich polypeptide-1; BRG1, protein Brahma homolog 1; BAF 170, BRM BRG1-associated factor 170; BAF 155, BRG1-associated factor155; BMR, bromodomain protein; p14ARF, ARF tumor suppressor; p15/p16, cyclin-ependent kinase inhibitor 2A.

Discussion

The aforementioned conventional and mesenchymal stem cell markers CD144, CD117, CD221, LGR5, CD271, CD10, BMI, FOXC1 and c-Myc are notable in bone sarcoma stem cells; however, there is no definitive marker for bone chondrosarcoma stem cells other than ALDH1A1 (16). The present study aimed to identify novel specific markers for chondrosarcoma stem cells. The results demonstrated that none of the known markers for other types of sarcoma were identified as chondrosarcoma biomarkers. Thus, the search for potential epigenetic players that decrease the CSC population following PRP-1 treatment remains critical. A previous study has reported that PRP-1 attenuates miR-302 (part of miR-302-367 cluster) expression and targets the embryonic stem cell markers Nanog, polycomb protein BMI-1 and c-Myc (2). A previous study on PRP-1 in the human chondrosarcoma JJ012 cell line investigated the effect of peptides on JJ012 subpopulations, and the outcome indicated that PRP-1 treatment notably decreased ALDHhigh cells from bulk JJ012 population (25).

In the present study, low activity of embryonic stem cell markers was observed in chondrosarcoma JJ012 cells treated with PRP-1, which were identified as ALDHlow-PRP-1-treated following sorting.

The oncogene BMI-1 is a member of the polycomb-group family of proteins, and is frequently overexpressed in different types of tumor to promote carcinogenesis and drive stem cell-like properties (32–43). BMI-1 knockdown using RNA interference was demonstrated to significantly impair osteosarcoma cell viability, in vitro colony formation and in vivo tumorigenesis, as well as to sensitize osteosarcoma cells to cisplatin-induced apoptosis (41). BMI-1 targets Ink4a/ARF, which encodes vital cell cycle inhibitors such as the human product of CDKN2A gene p16 (Ink4a) (17), which helps regulate stem cells. BMI-1 represses the Ink4a/ARF locus and the onset of senescence in human embryonic fibroblasts (43). However, the inhibitory effect of PRP-1 on BMI-1 was demonstrated to be independent of Ink4a/ARF at the protein levels in the present study, as no difference in protein expression was observed with or without PRP-1 treatment.

As PRP-1 has been demonstrated to inhibit Nanog (1), the present study assessed its upstream regulator, BRD4, which is a member of the bromodomain (BD) and extraterminal domain (BET), an important hallmark of cancer and epigenetic regulation and marker for stem cell self-renewal (44). BRD4 binds to acetylated histones at enhancers and promoters via its BDs in order to regulate transcriptional elongation and gene expression programs that have pivotal roles in inflammation and cancer development (44–47). BRD4 JQ1, a BET inhibitor, suppresses c-Myc and inhibits cell proliferation, and induces cell senescence and apoptosis in human chondrosarcoma cells (48). The cytostatic effect of PRP-1 has been demonstrated to be mediated by c-Myc inhibition in human chondrosarcoma cells (11). Over the past decade, the BD, BRD4, BET proteins have emerged as an important class of epigenetic readers (27). However, PRP-1 did not act an inhibitor of BRD4 in the present study.

TOP2A is an important enzyme in the regulation of DNA structure and cell proliferation (28–31). High TOP2A expression has been reported in advanced leiomyosarcoma, high mitotic index and advanced stage tumors (31). TOP2A is highly expressed in soft tissue sarcomas and is an independent predictor of poor prognosis (30). However, the results of the present study failed to demonstrate downregulation of TOP2A activity following treatment with PRP-1.

Gene transcription is dynamically regulated in stemness and cell proliferation (34,35). The chromatin remodeling complexes can affect whether a gene is activated or repressed, therefore the effect of PRP-1 on SWI/SNF remodeling complexes was assessed, which depends on ATP as its energy source, and its subunits BAF170 (SMARCC2), BAF155 (SMARCC1) and BRG1 (SMARCA4) (49–52). The mammalian SWI/SNF-like ATP-dependent chromatin remodeling complex, also termed BAF/BRG/Brahma associated (esBAF) can either activate or repress transcription (51). These factors are recruited by transcription factors to the promoters of target genes, where they can disrupt histone-DNA contacts and allow transcription factors to access their sequence-specific DNA (52). Depending on whether a transcriptional activator or repressor recruits SWI/SNF, transcription can be upregulated or downregulated accordingly (52). The energy from ATP hydrolysis is harnessed to disrupt histone-DNA contacts and move nucleosomes away from the transcription start site or towards it (52). Specifically, alterations to mammalian SWI/SNF (mSWI/SNF or BAF) ATP-dependent chromatin remodeling complexes and polycomb repressive complexes cause disease-specific changes in the chromatin architecture and gene expression across a number of sarcoma subtypes such as Ewing's, synovial sarcoma and chondrosarcoma (48). SWI/SNF is essential for self-renewal of stem cells and is critical for cell differentiation (49). A previous study on chondrosarcoma reported that poorly differentiated types of tumor were demonstrated to exhibit similarities with mesenchymal stem cells at pre-chondrogenic stages, whereas more differentiated types of tumor shared similarities with fully differentiated chondrocytes (53). This suggested that chondrosarcoma progression may be parallel to the deregulated chondrocyte differentiation process of MSCs (53). Future studies will investigate whether the inhibitory effect of PRP-1 on CSC leads to the revival of the differentiation program.

Well-established interactions between esBAF and polycomb complexes have been documented, which can act antagonistically () or synergistically cooperate with polycomb group proteins to support pluripotency (54,55). BRG1 and BRM have the ability to act as either oncogenes or tumor suppressors (54). The results of the present study demonstrated that inhibition of BMI-1 with PRP-1 was accompanied by inhibition of both subunits of the esBAF complex (BRG1 and BRM), which possess oncogenic properties in chondrosarcoma. Depletion of BAF170 (SMARCC2) instead of BAF155 (SMARCC1) resulted in the loss of stem cell properties in hESCs, and this pattern was different from that observed in rodent stem cells (i.e., mESCs and mEpiSCs) (55–58). Deletion of BRG1 (SMARCA4) in embryonic stem cells results in the loss of both self-renewal capacity and pluripotency (57). In the present study, BAF 155 was not affected by PRP-1 treatment, whereas BAF170 protein expression, BRM and BRG1 were all demonstrated to be inhibited following treatment compared with untreated cells Genome-wide studies have indicated that BRG1 and BRM have cooperative and antagonistic interactions in transcription (36). The results of the present study demonstrated that downregulation of BAF 170, BRG and BRM by PRP-1 may act in favor of synergistic regulation of BRG and BRM in chondrosarcoma, and that they may possess oncogenic properties in this microenvironment.

Cytostatic drugs can lead to decreased cell viability (12–14). Autophagy is a tightly regulated catabolic process of cellular self-digestion by which cellular components are targeted to lysosomes for their degradation (59). The present study failed to detect any apoptotic action of PRP-1 (data not shown). In cell cultures, certain drugs that activate autophagy have been associated with programmed cell death, but not apoptosis. Autophagy has been demonstrated to be essential for cell death, especially in cell lines that are resistant to apoptosis-driven chemotherapy, which may benefit from directly destroying tumor cells or sensitizing them to chemotherapy (59,60). Thus, it may be assumed that drastic elimination of chondrosarcoma stem cells by PRP-1 may be mediated by targeting BAF chromatin remodeling complexes and by induction of autophagy. A potential future study will aim to investigate whether this inhibition serves a pivotal role in the process of cancer stem cell decrease mediated by the peptide action.

In conclusion, the results of the present study demonstrated that ALDH1A1 is the only current biomarker in chondrosarcoma CSCs. PRP-1 decreased CSCs across all experiments. As a possible mechanismthe results of the present study indicated that PRP-1 may target chromatin remodeling complexes. Our future efforts will be directed towards understanding the interconnection between CSC maintenance, self-renewal and BAF complexes.

Supplementary Material

Supporting Data

Acknowledgements

The authors would like to thank Patricia Lissette Guevara, M.A from the Flow Cytometry Core Facility, Miller School of Medicine for her assistance in data acquisition, and analysis. The authors would also like to thank Dr Joel Block's Laboratory (Rush University, Chicago, USA) for providing the human JJ012 chondrosarcoma cells; Dr Messner (Ontario Cancer Institute, Canada) for Ly10 cells; and Dr Vega (MD Anderson Cancer Center, Houston, TX, USA) for HBL-1 cells. PRP-1 was provided by professor A. Galoyans laboratory H, Buniatian Institute of Biochemistry of National Academy of Sciences of Armenia. The authors would also like to thank Dr Claude Henry Volmar, Director of the Research Laboratory for the Center of Therapeutic Innovation, University of Miami, Miller School of Medicine for assisting with BRD4 experiments.

Funding

The present study was supported in part by the Ratcliffe Foundation to Miami Center of Orthopaedic Research and Education and the Department of Orthopedics, University of Miami, Miller School of Medicine.

Availability of data and materials

All data generated or analyzed during this study are included in this published article

Authors' contributions

AM conducted most of the experiments and was involved in manuscript writing. AH and AS performed western blot experiments, statistical and image analysis. CG participated in data analysis and certain experiments. SS performed fluorescence-activated cell sorting and flow cytometry data analysis. MB performed lymphoma cell line experiments with PRP-1 and participated in the critical revision of this manuscript. LZJ tested the PRP-1 action in melanoma cell line and contributed to the acquisition, analysis and interpretation of the data. SAC and JP provided substantial contributions to the conception and design and participated in critical revision of the manuscript. JB performed western blotting experiments. KG designed the study, wrote parts of the manuscript and was involved in the critical revision of it. All authors have read and approved the final manuscript.

Ethics approval and consent to participate

Not applicable.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References