Side Population Does Not Define Stem Cell-Like Cancer Cells in the Adrenocortical Carcinoma Cell Line NCI h295R
Urs D. Lichtenauer, Igor Shapiro, Klaus Geiger, Marcus Quinkler, Martin Fassnacht, Roland Nitschke, Klaus-Dieter Rückauer, and Felix Beuschlein
Institute of Molecular Medicine and Cell Research (U.D.L., I.S., F.B.), Department of Internal Medicine I (K.G.), Life Imaging Center (R.N.), Centre for Systems Biology, and Department of General and Visceral Surgery (K .- D.R.), Albert- Ludwigs-University Freiburg, D-79085 Freiburg, Germany; Clinical Endocrinology (M.Q.), Department of Internal Medicine, Charite Campus Mitte, Charite University Medicine Berlin, D-10117 Berlin, Germany; Department of Internal Medicine I (M.F.), Endocrine and Diabetes Unit, University of Wuerzburg, D-97074 Wuerzburg, Germany; and Medical Clinic (F.B.), University Hospital Innenstadt, Ludwig Maximilians University, D-80336 Munich, Germany
Recent evidence suggests the existence of a stem cell-like sub- population of cells in hematological and solid tumor entities, which determine the malignant phenotype of a given tumor through their proliferative potential and chemotherapy re- sistance. A recently used technique for the isolation of this cell population is through exclusion of the vital dye Hoechst 33342, which defines the so-called side population (SP). Herein we demonstrate the presence of SP cells in a variety of adrenal specimens, including primary cultures of human adrenocor- tical tumors and normal adrenal glands as well as established human and murine adrenocortical cancer cell lines by fluo- rescence-activated cell sorter analysis and confocal micros- copy. On a functional level, SP cells from the human adreno- cortical tumor cell line NCI h295R revealed an expression pattern consistent with a less differentiated phenotype, in-
cluding lower expression of steroidogenic enzymes such as steroid acute regulatory protein (StAR) and side-chain cleav- age enzyme (P450scc) in comparison with non-SP cells. How- ever, proliferation between SP and non-SP cells did not differ (105.6 ± 18.1 vs. 100.0 ± 3.5%). Furthermore, re-sorting and tracing experiments revealed the capacity for both cell types to give rise to the original SP- and non-SP-containing cell population. Similarly to the baseline growth kinetics, no sur- vival benefit was evident in SP cells after treatment with cytotoxic agents commonly used in adrenocortical carcino- mas. Taken together, these findings provide evidence that Hoechst dye exclusion, in contrast to what has been reported for other tumor entities, is not a major tumor stem cell defin- ing marker in adrenocortical NCI h295R tumor cells. (Endo- crinology 149: 1314-1322, 2008)
A N INCREASING body of evidence suggests that a small population of stem cell-like cancer cells define the clinical phenotype of a variety of tumor entities. This con- cept, which has first been established in hematological ma- lignancies such as acute myeloid leukemia, has only recently been adapted for a growing number of solid tumors includ- ing breast cancer (1, 2), neuroblastoma (3), different brain tumors (4, 5), head and neck squamous cell carcinomas (6), colon carcinoma (7), and pancreatic cancer (8). Stem cell-like cancer cells have been proposed to result either from ma- lignant transformation of a normal stem cell or from reac- quisition of stem cell characteristics of a differentiated cell, or a progenitor cell, or both (9). Functional characteristics of stem cell-like cancer cells have been defined as the ability to regrow tumors in xenograft in vivo models, to undergo qual- itative self-renewal, to recapitulate all cell types within an individual tumor, and to possess extensive proliferative ca- pacity (9-11). In contrast to this small cell fraction, the vast majority of cells in a given tumor have been hypothesized to
Abbreviations: FACS, Fluorescence-activated cell sorting; FCS, fetal calf serum; MTT, microculture tetrazolium; SP, side population.
Endocrinology is published monthly by The Endocrine Society (http:// www.endo-society.org), the foremost professional society serving the endocrine community.
represent a mature, nonproliferating population, or at least a population with a more differentiated and less aggressive phenotype. Because therapies specifically targeted to stem cell-like cancer cells could have the potential to eradicate the cellular source of the individual tumor, the pathophysiolog ical concept behind these tumor stem cells has aroused con- siderable clinical interest.
Tumor cells with stem cell characteristics have been iso- lated by cell surface markers such as CD24, CD44, and CD133 as well as on the basis of sphere formation after in vitro cultivation (12). An alternative approach, especially in the absence of known cell markers, is the side population (SP) phenomenon, which has been used to identify and isolate enriched stem cell populations from a variety of tissues in- cluding bone marrow (13), mammary gland (14, 15), skin (16), liver (17), lung (18), skeletal muscle (19), limb (20), heart (21), and brain (22). This technique is based on Hoechst dye 33342 exclusion, which primarily occurs through the activity of membrane pumps encoded by multidrug-resistance gene 1 (MDR1) (23) and breast cancer-resistance gene 1 (ABCG2) (24). As for the identification and isolation of stem cell-like tumor cells, Hoechst dye exclusion has been successfully applied in neuroblastoma tumors and corresponding cell lines (3). Recent studies have extended these findings for primary cultures, for example of gastrointestinal cancers (25,
26), and ovarian cancer (27), defining the SP characteristics as a valuable marker for cells with stem cell characteristics. Intriguingly, the presence of a specific SP has also been demonstrated in established tumor cell lines of different or- igin, such as glioma (28), breast (29), thyroid cancer (30), and melanoma (31) cell lines, arguing for the presence of a func- tionally distinct subpopulation of cells even in a monoclonal cell environment.
Adrenocortical carcinomas are highly malignant tumors that respond poorly to standard chemotherapy treatment regimens (32). The reasons for this clinical phenotype are believed to be based on cellular heterogeneity of the tumor and the presence of multidrug resistance genes, which en- code for pumps that actively expel the cytotoxic substances (33). Because the Hoechst efflux capacity of SP cells is also dependent on the presence of membrane pumps (34), we hypothesized that identification of the SP in adrenocortical tumors could potentially represent a suitable isolation method to evaluate stem cell-like tumor characteristics in adrenocortical malignancies. Thus, this study was initiated to investigate whether stem cell-like tumor cells represented by SP cells are present in adrenal tumors and, if so, whether their drug efflux capabilities are responsible for the known che- motherapy resistance in this endocrine malignancy.
Materials and Methods
Patients
Diagnosis and endocrine work-up of patients with adrenal masses were defined following standard clinical, biochemical, imaging, and histological criteria. All patients gave written informed consent, and the study was approved by the local ethical committee of the University of Freiburg. Ten surgical tissue samples were included from a total of eight patients (one endocrine inactive adenoma, one myolipoma, three hy- perplasias, two adrenocortical carcinomas, and one metastasis of an adrenocortical carcinoma as well as adjacent normal adrenal gland tissue from two patients).
Cell line culture conditions and primary cell culture of adrenal surgical samples
NCI h295R cells were grown at 37 C and 5% CO2 in DMEM/F12 medium (Life Technologies, Inc., Invitrogen, Carlsbad, CA) containing 10% fetal calf serum (FCS) (HyClone, South Logan, UT) and 2 mM L-glutamine (BioWhittaker Cambrex, East Rutherford, NJ), whereas mu- rine Y1 and Y6 cells were maintained in DMEM high glucose containing 7.5% horse serum, 2.5% FCS, and 10 ul/ml penicillin/streptomycin (all Life Technologies) as described previously (35, 36).
For primary cultures from human adrenal tumors and adjacent nor- mal adrenal gland tissue, the samples were cleaned of surrounding fat, connective tissue, and blood vessels. Thereafter, tissue samples were minced into pieces smaller than 0.5 mm using a razor blade. Minced samples were transferred into 50-ml Falcon tubes, spun down at 1000 rpm for 5 min, and rinsed twice with fresh PBS. Digestion was performed with 1 mg collagenase II (Biochrom, Berlin, Germany) per mÌ PBS at 37 C for 50 min in a shaking water bath. Cell suspension was pipetted up and down at least twice during the incubation time. After digestion, pure FCS was added to a minimum concentration of 10% to inactivate the collagenase, followed by a centrifugation step as described above. Cell pellets were resuspended in erythrocyte lysis buffer and incubated for 7 min at room temperature. After another centrifugation step, cells were resuspended in 5-10 ml culture medium depending on the expected cell count (DMEM/F12 with 10% FCS, 3.1 g/liter glucose, 15 mm HEPES, and 10 pl/ml penicillin/streptomycin, all from Life Technologies) and sequentially filtered through a 100- and 70-um nylon mesh. Cells were counted using a Neubauer counting chamber and further processed for fluorescence-activated cell sorting (FACS) as described below.
Hoechst staining and FACS sorting
NCI h295R and Y1 cells were detached from the cell culture flask with trypsin (Sigma Chemical Co., St. Louis, MO), and viable cells were counted with trypan blue, whereas single-cell suspensions from adrenal surgical material were directly used after preparation as described above. Cells were transferred into DMEM high-glucose medium (Life Technologies) containing 2% FCS and 10 mM HEPES and stained with the fluorescent dye Hoechst 33342 (Sigma) at a concentration of 5 µg/ml at 37 C for 90 min as described (37). After the staining procedure, propidium iodide (2 µg/ml; Sigma) was added to the samples for identification and exclusion of dead cells. Cell analysis and sorting were performed on a triple-laser cell sorter [MoFlo; Dako (formerly Cytoma- tion), Fort Collins, CO]. The Hoechst dye was excited at 350 nm using an argon ion laser. Emission wavelengths were detected at 450 nm using a 450/20 bandpass filter and above 675 nm using a 675LP filter. The SP was defined as described (37), including the verification procedure with verapamil (Sigma), with the only difference that special care was taken to exclusively isolate the tip of the SP cell fraction.
Time course and confocal laser microscopy
For the FACS time-course experiments, 50 million NCI h295R cells were harvested and transferred to the staining medium as described above. A small sample underwent the regular staining procedure as a positive control and to calibrate the FACS settings. The circulating water temperature of the FACS sample chamber was raised to 37 C. Once the sample reached this temperature, the first FACS file was saved; Hoechst dye was then added, and subsequent files were saved every 5 min for a total of 120 min. Recalibration and adjustment of the FACS settings were performed before each set of experiments to ensure optimal tech- nical performance and separation of the cell population. To ensure comparability in experiments for which SP quantification was critical, FACS experiments were performed within the same run with identical experimental conditions.
For confocal laser microscopy cell morphometry, 1000 sorted SP and non-SP cells were transferred to a microslide VI chamber (ibidi, München, Germany), and three-dimensional image stacks were re- corded at 405 nm excitation and emission collected with a long-path filter LP420 with a confocal laser scanning microscope (LSM 510 Meta; Zeiss, Jena, Germany). Data were analyzed using the spot detection and anal- ysis functions of the Imaris software (Bitplane AG, Zürich, Switzerland).
For time-course experiments, 450,000 NCI h295R cells were plated on each of two plastic dishes optimized for high-resolution microscopy (p-dish I; ibidi). After 2 d, the cells reached 80% confluency, at which time the regular growth medium was replaced by the staining medium. One of the wells was then placed in a microscopy heating unit (PeCon, Erbach, Germany) that held the temperature of the optical well at 37 C. Hoechst dye was added as described above. After 90 min, the confocal microscope settings were adjusted for optimal recording conditions for the Hoechst fluorescence (two-photon excitation at 760 nm, emission collected from 400-480 nm). Thereafter, the dish with untreated cells was placed within the microscope, followed by the first photograph and addition of Hoechst dye to the medium. Beginning with this incubation, additional photographs were taken every 90 sec for a total of 120 min.
Tracing experiment
NCI h295R cells were separated into SP and non-SP populations as described above with a final number of 250,000 cells for each population. SP cells were stained with the red fluorescence dye PKH26, whereas non-SP cells were labeled green with PKH67 using the General Cell Membrane Labeling Kit (both Sigma). Images of the cells were taken from both populations with a fluorescence microscope (Axiovert 200; Zeiss) before they were mixed in equal proportions and seeded together in a 12-well plate in regular NCI h295R growth medium as described above. Additional fluorescent images were then taken on d 2, 6, and 14.
Microculture tetrazolium (MTT) assay
A total of 2000 SP and non-SP cells were plated separately in triplicate in 96-well plates and were allowed to grow for 14 d at 37 C in a humidified incubator at 5% CO2. To develop the plate, MTT (Sigma) was
dissolved in serum free-medium at a final concentration of 0.5 mg/ml and briefly sonicated. Culture medium was removed from the wells and 100 ul MTT/medium per well was added, followed by an incubation for 2 h at 37 C. Thereafter, the MTT/medium was removed, and 100 pl solubilization (10% SDS, 0.01 M HCI) solution per 100 ul medium was added to stop the coloring reaction and to dissolve the formed formazan crystals. The mixture was kept in the dark and incubated overnight at room temperature. Finally, the plate was shaken at 60 rpm for 5 min to dissolve the precipitate before reading the plate at 555 nm using a microplate reader.
Cell cycle analysis
At least 300,000 SP and non-SP cells were fixed overnight in 4.5 ml of 70% ethanol each and kept at +4 ℃. Thereafter, cells were centrifuged, the supernatant discarded, and the cell pellet resuspended and washed with PBS. After another centrifugation step, cells were resuspended in 300 ul propidium iodide/Triton X-100 staining solution containing 5 µg propidium iodide/ml PBS, 0.1% Triton X-100, and 100 µg/ml freshly prepared RNase A and were incubated at 37 C for 30 min. For FACS analysis, cells were transferred to FACS tubes and placed on ice. FACS analysis was performed using a FACSCalibur (B&D Biosciences, Hei- delberg, Germany) at 488 nm. The data were analyzed using FlowJo version 6.0 and the Watson Pragmatic cell cycle formula.
Real-time PCR
RNA from at least three independent SP and non-SP cell preparations was extracted using the QIAGEN RNA mini kit (QIAGEN, Valencia, CA) following the instructions of the manufacturer. cDNA was tran- scribed using a RT kit (Promega, Mannheim, Germany) and 1.0 µg total RNA.
Real-time PCR was performed using the FastStart DNA MasterPlus SYBR Green I reaction mix in the LightCycler 1.5 (Roche, Mannheim, Germany). The cycling conditions for real-time PCR included a preincu- bation step at 95 C for 10 min, followed by an amplification step that consisted of 40-45 cycles at 95 C for 10 sec, annealing (the temperature was primer dependent as given below) for 6 sec, and an extension step at 72 C, for which the time was calculated by the product length in base pairs divided by 25 (Roche, Indianapolis, IN). Primer sequences as well as prod- uct lengths and annealing temperatures were as follows: P450scc, 5’- GCAAČGTGGAGTCGGTTTAT-3’ and 5’-TCCTCGAAGGACATCTT- GCT-3’ (664 bp; 52 C); steroidogenic acute regulatory enzyme (StAR), 5’-CAGGACAATGGGGACAAAGT-3’ and 5’-ATGAGCGTGTGTAC- CAGTGC-3’ (608 bp; 63 C); IGF-II, 5’-CAAATTACCTGCCCATTCGT-3’ and 5’-GCGTTAAAGGAGTTGAGTTGAG-3’ (357 bp; 58 C); steroidogenic factor 1 (SF-1), 5’-TGCACTGCAGCTGGACCGCCÄGGAGTT-3’ and 5’- AGGGCTCCTGGATCCCTAATGCAAGGA-3’ (390 bp; 58 C); ABCG2, 5’- GAGTGGCTTTCTACCTTGTC-3’ and 5’-CATCACAACATCATCTTG- TACC-3’ (246 bp; 58 C); MDR1, 5’-ATCGTTTGTCTACAGTTCGT-3’ and 5’-TATACTTTCATCCAGAGCCT-3’ (316 bp; 58 C); claudin-1 (CLDN1), 5’-CCGTTGGCATGAAGTGTATG-3’ and 5’-GTTTTGGATAGGGCCTT- GGT-3’ (309 bp; 60 C); zona occludens-1 (TJP1), 5’-AGCCAAGGAAG- GCTTAGAGG-3’ and 5’-ACAACACGGAACACCTCTCC-3’ (270 bp; 61 C); protooncogene MET, 5’-ACTCCCCCTGAAAACCAAAGCC-3’ and 5’- GGCTTACACTTCGGGCACTTAC-3’ (536 bp; 60C); N-cadherin (NCAD), 5’-GACAATGCCCCTCAAGTGTT-3’ and 5’-ACCCACAATCCTGTCCA- CAT-3’ (354 bp; 61 C); and vimentin (VIM). 5’-GCAGGCTCAGATTCAG- GAAC-3’ and 5’-GCTTCAACGGCAAAGTTCTC-3’ (330 bp; 59 C).
Melting curve analysis was performed between 65 and 98 C (0.1 C/sec) to determine the melting temperature of the amplified product and to exclude undesired primer dimers. Furthermore, the products were run on a 1% agarose gel to verify the amplified product. Each sample was run at least in triplicate. Quantification was normalized using the ß-actin as a reference gene. To facilitate overall comparison, expression levels of the particular genes were set at 100% for non-SP cells.
In vitro treatment with cytotoxic agents
Dose-finding experiments were performed on unsorted NCI h295R cells to assess the therapeutic range of etoposide (Medac, Wedel, Ger- many), doxorubicin (Medac), streptozotocin (Pfizer, Quebec, Canada),
cisplatin (Hexal, Holzkirchen, Germany), and mitotane (Bristol-Myer Squibb, Munich, Germany). In a different set of experiments, NCI h295R cells were Hoechst stained and FACS sorted as described above, and the separated SP and non-SP cells were seeded in 96-well plates at a density of 25,000 cells per well in triplicate. After incubation with the predefined dosage range of the individual cytotoxic agent, cell viability was quan- tified by MTT assays as described above.
Statistical analysis
All results are expressed as mean ± SEM. If not stated otherwise, all statistical comparisons were analyzed by ANOVA and Fisher’s protec- tive least significant difference test. Statistical significance is defined as P < 0.05 and is indicated in the figures.
Results
SP cells are present in a variety of adrenocortical tissues and cell lines
FACS analysis after Hoechst dye labeling consistently re- vealed the presence of a SP in single-cell suspension from adrenal surgical tissues including an adrenal adenoma (n = 1, SP fraction 0.1%), a myelolipoma (n = 1, SP fraction 0.98%), adrenocortical carcinomas (n = 2, SP fraction 0.03 and 0.09%), adjacent normal adrenal tissue (n = 2, SP fraction 0.00 and 0.03%), a metastasis of an adrenocortical carcinoma (n = 1, SP fraction 0.02%), and adrenal hyperplasia (n = 3, SP frac- tion 0.03-2.62%). A similar proportion of SP cells was found in normal mouse adrenal glands (data not shown). More- over, a large SP fraction was detectable in both a human and murine adrenocortical cell line (NCI h295R, 2.79 ± 1.13%; Y1, 0.44 ± 0.37%; Y6, 16.3%). For a general overview, the results are summarized in Table 1.
To ensure the specificity of this FACS-derived SP and exclude nonspecific staining effects, coincubation of Hoechst dye with the calcium channel blocker verapamil was per- formed, which resulted in the expected decrease in the SP cell population (e.g. NCI h295R cells, 2.91% untreated vs. 0.11% after verapamil treatment; Fig. 1A). Confocal laser micros- copy morphometry proved a significantly lower mean in- tensity of SP compared with non-SP cells (100 ± 3.11 vs. 81.8 ± 1.82%; P < 0.01) with a comparable mean cell volume (100 ± 2.30 vs. 101.4 ± 1.83%; P = 0.65; Fig. 1B). Moreover, FACS time-course experiments and continuous confocal la- ser microscopy during the Hoechst staining procedure con- firmed a gradual dye uptake with a differential nuclear stain- ing and an optimal visual appearance of the Hoechstlow classified SP cells after 90 min (Fig. 1C). Prolonged Hoechst incubation over 120 min led to a nearly homogeneous Ho- echstPos staining pattern in the confocal microscope (Fig. 1C),
| Tissue | %Side population |
|---|---|
| Adrenal hyperplasia | 0.02-2.62 |
| Adrenal adenoma | 0.1 and 0.98 |
| Adrenocortical carcinoma | 0.03 and 0.09 |
| Metastasis of adrenocortical carcinoma | 0.02 |
| NCI h295R human adrenocortical carcinoma cell line | 2.79 |
| Y1 murine adrenocortical carcinoma cell line | 0.44 |
| Y6 murine adrenocortical carcinoma cell line | 16.3 |
Ten tissue samples from eight patients and three adrenocortical cancer cell lines were analyzed.
A
B
Onon-SP - SP
120
R4
relative to non-SP (%)
100
L
10%
80
0
3
60
40
10º
20
3-dimensional cell reconstruction
0
10
676 ELP
ttP
mean intensity
mean volume
ETS ELP
- Verapamil
+ Verapamil
C
M
-
-
-
-
182
182-
Ya
JE
-
-
.
-
-
.
.
A
5
-
-
.
6
-
1:38
-
2
.
-
-
-
-
6
-
=
8
-
135
150
-
0 min
20 min
90 min
120 min
180 min
so the duration of Hoechst incubation was set to 90 min for all additional experiments.
SP cells from the NCI h295R cell line express lower levels of steroidogenic enzymes and higher levels of multidrug resistance genes
As would be expected from a cell population with stem cell characteristics, semiquantitative RT-PCR analysis revealed lower expression of marker genes specific for differentiated adrenocortical cells including SF-1, P450 side chain cleavage enzyme (SCC), and StAR in SP cells relative to non-SP cells (Fig. 2A). This finding was further substantiated by real-time PCR with lower expression levels found in SP cells (SCC, 59.3 ± 3.1%; P < 0.01, and StAR, 59.0 ± 1.1%; P < 0.01) compared with non-SP cells (SCC, 100 ± 5.9%, and StAR, 100 ± 4.8%; Fig. 2B). In contrast, the expression of IGF-II, a major regulator of adrenocortical development and function, was not different between the two groups (Fig. 2B).
Furthermore, expression levels of the multidrug resistance genes ABCG2 and MDR1, which are considered to be mainly responsible for the Hoechst drug efflux effect, were mea- sured in triplicate in independent samples and were found to be substantially increased in SP (ABCG2, 27089 ± 3283%, P < 0.04, and MDR1, 176.0 ± 23.7%, P<0.03) compared with non-SP cells (ABCG2, 100 ± 47.1%, and MDR1, 100 ± 0.5%; Fig. 2, C and D).
NCI h295R SP cells display similar growth potential and self-renewal capacity in comparison with non-SP cells
In contrast to the assumed phenotypic properties of a stem cell-like tumor cell, after FACS-based cell sorting followed by
separate cultivation and quantification of cellular viability, NCI h295R SP cells displayed a proliferative potential com- parable to that of non-SP cells (105.6 ± 18.1 vs. 100.0 ± 3.5%, P = 0.7; Fig. 3A). To further verify this finding, both popu- lations of cells were labeled with different live-cell stains, which allowed detection of the original cell as well as several daughter cell generations. The resulting red-stained SP cells and green-stained non-SP cells were repooled with equal cell numbers (2500 cells per population per well; Fig. 3B) and separately quantified over the course of the next 14 d. Re- peated cell counting revealed a stable and equal ratio of the original cell populations (SP fraction: d 2, 49.9 ± 2.2%; d 8, 46.1 ±1.8%; d 14, 46.5 ± 2.8%; Fig. 3B) indicating the inability of the SP cells to outgrow the non-SP population.
Based on stem cell physiology, SP cells should give rise to SP and non-SP cells by means of asymmetric cell division, whereas non-SP cells should not have this potential. To fur- ther analyze the SP and non-SP distribution over several generations, NCI h295R cells were repeatedly stained, sorted, restained and re-sorted over four passages. Although the comparable numbers within a given passage consistently demonstrated an enrichment of SP cells after original SP sorting, isolated non-SP cells regularly gave rise to a signif- icant proportion of SP cells (Fig. 4).
SP phenotype in NCI h295R SP is not dependent on cell cycle or epithelial-mesenchymal transition
Because the results from tracing and re-sorting experi- ments suggested a similar growth potential and self-renewal capacity between SP and non-SP NCI h295R cells, we hy- pothesized that phenotypic changes that were associated
A
SP
non
SP
adr H2O
B
non-SP
SP
120
SF-1
390 bp
expression relative to B-actin (%)
100
I
StAR
980 bp
80
60
SCC
947 bp
40
18S
151 bp
20
0
C
D
StAR
SCC
IGF2
35000
ABCG2 expression relative to Bactin (%)
250
MDR1 expression relative to Bactin (%)
30000
200
25000
20000
150
15000
100
10000
5000
50
0
0
non-Sp
SP
non-SP
SP
with Hoechst staining properties could be dependent on the cell cycle of the individual cell. However, FACS analysis of propidium iodine-stained cells revealed no differences in cell cycle distribution (non-SP: G1 phase, 66.9%; S phase, 23.1%; G2 phase, 9.47% vs. SP: G1 phase, 66.8%; S phase, 24.8%; G2 phase, 6.7%; Fig. 5) between SP and non-SP cells.
Another explanation for the free conversion of the pre- sumably more undifferentiated SP cells with a rather mes- enchymal phenotype into the more differentiated non-SP
cells, which possess a more epithelial phenotype, has been proposed to occur by epithelial-mesenchymal transition. However, real-time PCR analysis did not reveal consistent differences between SP and non-SP cells for the expression levels of the epithelial markers CLDN1, TJP1, and MET (non-SP vs. SP, 100.0 ± 1.3 vs. 90.3 ± 1.7%, P <0.01; 100.0 ± 5.9 vs. 100.4 ± 11.1%, P= 0.97; 100.0 ± 2.7 vs. 101.1.3 ± 10.5%, P = 0.92, respectively), and the mesenchymal markers VIM and NCAD (100.0 ± 0.6 vs. 93.7 ± 3.4%, P < 0.01; 100.0 ± 1.1
A
256
R5
non-SP
B
104
R5
nan-SP
192
R4
103
R4
450/20
-
128
450/20
10%
SP
SP
64
101
-
0
0
64
128
192
256
100
675 ELP
100
10!
102
675 ELP
103
104
growth potential relative to non-SP (%)
120
100
100
1
90
SP cell content (%)
80
80
70
60
60
50
40
40
30
20
20
10
0
0
non-SP
SP
day 2
day 8
day 14
Generation/Passage
1
1.09%
II
SP
non-SP
-
8.93%
7.02%
III
SP
non-SP
SP
non-SP
7.36%
6.58%
16.56%
8.50%
SP
non-SP
SP
non-SP
SP
non-SP
SP
non-SP
IV
1-
2.11%
1.87%
5.23%
1.53%
4.20%
0.74%
2.41%
0.81%
vs. 81.2 ± 1.6%, P = 0.03, respectively; Fig. 5 B). Thus, a close association between the SP and non-SP phenotype with a particular mesenchymal or epithelial state in NCI h295R cells seems unlikely.
SP cells in the NCI h295R cell line do not display resistance to cytotoxic agents in comparison with non-SP cells
Chemotherapy resistance represents the key survival ben- efit for stem cell-like tumor cells. However, similar to the finding for cell proliferation under baseline conditions, in- cubation with a variety of chemotherapeutic agents com- monly used in the treatment of adrenocortical cancer (32) did not demonstrate a significantly higher survival of SP cells in comparison with non-SP cells (SP vs. non-SP: etoposide, 100 ug/ml, 30.1 ± 7.6 vs. 36.8 ± 4.1%, P = 0.07; doxorubicin, 1.0 mg/ml, 84.6 ± 13.1 vs. 77.1 ± 19.7%, P = 0.86; streptozotocin, 3 mg/ml, 80.0 ± 6.0 vs. 63.2 ± 12.2%, P = 0.14; cisplatin, 3 µg/ml, 36.0 ± 21.1 vs. 34.8 ± 15.4%, P = 0.91; mitotane, 25 µg/ml, 13.1 ± 1.3 vs. 13.2 ± 3.4%, P = 0.84; Fig. 6). Taken together, these findings indicate an indistinguishable re- sponse of SP and non-SP cells to cytotoxic actions of the applied pharmacological agents.
Discussion
Because of their potential impact in general tumor patho- physiology, stem cell-like tumor cells have gained much attention over the last 2 yr (9, 11). Sparked by the initial publication from Al Hajj and colleagues (1), who first isolated cells with stem cell properties in breast cancers, a number of groups have identified cancer stem cells derived from other
tumors in rapid succession, including ovarian (27) and lung cancers (38), brain tumors (4), and colon carcinomas (7). The prerequisite for the successful isolation of these cell popu- lations was the identification of surface markers that could be used for cell sorting-based approaches.
Cell surface markers in adrenocortical tumor entities have not been identified so far. In a candidate gene approach, we unsuccessfully analyzed the human adrenocortical cancer cell line NCI h295R for known tumor stem cell markers such as prominin (CD133), SCA1 (LY6), and CD24 (data not shown) until we detected a distinct SP in a variety of adrenal tumors. Until 2005, the Hoechst efflux phenomenon of the SP was used solely to isolate a stem cell-enriched population from normal bone marrow, liver, and brain tissue (37). Re- cently, however, Hirschmann-Jax and colleagues (3) were the first to successfully adapt this technique to isolate stem cell- like tumor cells from neuroblastoma cell lines and primary cultures. Similar to the findings achieved with stem cell-like tumor cells that had been isolated on the basis of surface stem cell markers, SP cells were shown to proliferate and possess self-renewal capacity and chemotherapy resistance beyond that of non-SP cells. Since then, SP cells from ovarian (27), brain (39), and thyroid (30) cancers have been demonstrated to exhibit similar stem cell-like functional properties.
Herein we demonstrate the presence and successful iso- lation of SP cells from a variety of adrenocortical tumor specimens, including primary cultures of human adrenocor- tical tumors with different endocrine activities as well as established human and murine adrenocortical cancer cell lines. Because sources of surgical tumor specimens are lim- ited and the presence of functionally distinct SP in permanent
A
256
&
&
non-SP
Courts
B
192
R4
$
450/20
·
128
0
200
400
600
1000
FL2-A
200
SP
160
64
Counts
8
8
?
0
0
64
128
675 ELP
192
256
0
0
200
400
600
1000
FL2-A
800
B
120
non-SP
SP
100
*
T
T
*
expression (% of non-SP)
*
T
80
T
60
40
20
0
CLDN1
TJP
MET
NCAD
VIM
cell lines have been demonstrated for other tumor entities, we chose to further analyze SP cells isolated from the well described human cell line NCI h295R (40), which has been used extensively as an adrenocortical model system in the past (41).
To ensure the specificity of the isolation procedure, the following precautions were implemented. First, we show the expected pharmacological blockage of the Hoechst dye efflux capacity of SP cells using the calcium channel blocker vera- pamil. Second, we demonstrate the appearance of a distinct SP within a detailed time-course experiment based on both FACS analysis as well as visualization by confocal micros- copy. Findings from these time-course experiments were used to specify incubation times for the further experimental settings. Third, special care was taken to exclusively isolate cells from the tip of the SP fraction, because it has been proposed that these cells contain the highest proportion of undifferentiated cells (42, 43). The expected lower Hoechst intensity of the isolated SP cells compared with the non-SP cells was quantified by confocal laser microscope morphom- etry. Fourth, we show that SP cells derived from NCI h295R cells have a distinct expression profile in comparison with cells from the non-SP cells; ABCG2 and MDR1, which are believed to encode for the membrane-associated pumps mainly responsible for the Hoechst 33342 efflux, are sub-
☐ non-SP
SP
viability (% of untreated non-SP)
160
140
Cisplatin
120
100
T
80
60
40
20
0
viability (% of untreated non-SP)
NaCI
1,5ug/ml
2,0 ug/ml
3,0 ug/ml
160
140
Streptozotocin
120
100
80
60
40
20
0
NaCI
2 mg/ml
viability (% of untreated non-SP)
3 mg/ml
140
120
Doxorubicin
100
80
60
40
20
0
viability (% of untreated non-SP)
120
NaCl
0,5 mg/ml
1,0 mg/ml
100
Etoposide
80
60
40
*
*
20
0
viability (% of untreated non-SP)
@120
NaCl
50ug/ml
75 ug/ml
100 ug/ml
100
Mitotane
80
60
40
20
0
NaCI
10 ug/ml
25 ug/ml
stantially more highly expressed in SP cells. Also, because expression of steroidogenic enzymes can be regarded as a marker for differentiated adrenocortical cells, the lower ex- pression levels detected are in line with the expected phe- notype of the isolated SP, which would be that of less dif- ferentiated cells. Overall, these experiments provide ample evidence for the presence of a distinct SP of cells within adrenocortical tumors specimen and cell lines.
Stem cell-like cancer cells have been characterized on the basis of functional properties that include higher prolifera- tive potential with the ability to outgrow other cell popula- tions, self-renewal capacity that is not present in the remain- ing tumor, and the ability to withstand higher levels of cytotoxic agents included in standard chemotherapeutic reg- imens (44-46). In line with this notion, we performed a number of experiments to assess similar functional charac- teristics of the isolated NCI h295R SP cells and define them as potential stem cell-like cancer cells.
In contrast to what has been reported for other tumor cell lines, however, FACS-based cell sorting and separate cul- turing of NCI h295R cells from the SP and non-SP revealed similar growth rates under baseline conditions. These studies were further validated by tracking experiments in which separated cells from the SP and non-SP were marked by means of different dyes that constantly tag the original cell as well as several generations of daughter cells. When the two cell populations were cocultured in an equal distribu- tion, this proportion was maintained over the observation period, indicating parallel and indistinguishable growth ki- netics of the two different populations of cells. To further highlight this finding, iterative re-sorting experiments were carried out. If SP cells would possess significantly higher stem cell capacities than non-SP cells, one would expect not only an enrichment of the SP cells after recurrent re-sorting for the SP phenotype but also the loss of SP cells after re- peated re-sorting for the non-SP phenotype. As expected, a slightly higher proportion of SP cells was evident in subse- quent FACS analyses performed on separated and cultured SP cells in comparison with that of separated and cultured non-SP cells. However, recurrent sorting and culturing of non-SP cells consistently resulted in a cell population that contained an easily detectable proportion of SP cells.
A potential explanation for the findings of the re-sorting and color-tracking experiments could be that a free and un- directed transition of an individual cell between the SP and non-SP phenotype occurs. One possibility would be a con- tinuous cell cycle-dependent change in membrane pump expression or activity, responsible for the SP characteristics. Alternatively, transition of a cell from a more differentiated epithelial-like to a less differentiated mesenchymal-like phe- notype could go in parallel with the observed Hoechst efflux capacities. This epithelial-mesenchymal transition has been considered a prerequisite for tumor infiltration and metas- tasis. Although the original model proposes a unidirectional change toward the less differentiated mesenchymal pheno- type, recent evidence has suggested the presence of incom- plete epithelial-mesenchymal transition as well as a rever- sion of this transition (47). To further explore these possibilities for NCI h295R cells, cell cycle-dependent phe- notypic changes and the expression patterns of proposed
epithelial and mesenchymal markers were investigated. However, cell cycle analysis did not reveal differences be- tween SP and non-SP cells, and there was no evidence for a significant association between Hoechst dye exclusion prop- erties and a specific stage in the cell cycle. Similarly, real-time analysis of a number of epithelial and mesenchymal markers did not indicate a common epithelial or mesenchymal pat- tern for SP or non-SP cells.
Probably the most relevant aspect of the tumor stem cell concept from the clinical perspective is the ability for stem cell-like tumor cells to possess an additional survival benefit in the context of various chemotherapeutic agents. Adreno- cortical carcinomas are characterized by their overall resis- tance to several chemotherapies (32). As such, the targeted therapy of a subpopulation of cells within an adrenocortical carcinoma responsible for this deleterious clinical phenotype would be of major clinical interest. However, although SP cells derived from NCI h295R cells obviously express the postulated pumps responsible for Hoechst efflux, this equip- ment seems to be insufficient to ensure chemotherapy resis- tance associated with a survival benefit over non-SP cells. Hence, the SP of NCI h295R cells seems not to represent a relevant experimental target to explore future stem cell-di- rected therapies.
Taken together, we demonstrate the presence of a distinct SP of cells in adrenocortical tumors and tumor cell lines. Although these findings are in line with those reported for a number of other tumor entities, we provide evidence that SP cells isolated from the NCI h295R adrenocortical cancer cell line are not distinguishable from non-SP cells based on their proliferative potential, self-renewal capacity, or resis- tance to chemotherapeutic agents. As such, the SP charac- teristic is unlikely to determine the malignant phenotype within the NCI h295R cell line. Whether this phenomenon is restricted to this particular cell line or poses a general concept for adrenocortical tumorigenesis needs to be addressed in future studies. However, as Hoechst exclusion is increasingly used for the isolation of stem cell-like tumor cells, it has to be stressed that the mere demonstration of a SP cannot be equated with the presence of a tumor-like stem cell popu- lation in a given sample. In fact, the finding of SP cells has to be scrutinized by detailed functional studies to prove or disprove the stem cell capacity of the SP in each tumor entity.
Acknowledgments
We are indebted to Dirk Engelbert for assistance with the cell cycle analysis and Lilia Spady for excellent technical assistance. Cell sorting was carried out in the Core Facility, Department of Internal Medicine I, Universitaetsklinikum, Freiburg, and confocal microscopy at the Life Imaging Center, University of Freiburg.
Received July 23, 2007. Accepted November 26, 2007.
Address all correspondence and requests for reprints to: Felix Beus- chlein, M.D., Division of Endocrine Research, Department of Medicine Innenstadt, University Hospital Munich, Ziemssenstrasse 1, D-80336 Munich, Germany. E-mail: felix.beuschlein@med.uni-muenchen.de.
This work was supported by Landesstiftung Baden-Württemberg Grant P-LS-ASN/5 (to F.B.).
Disclosure Statement: The authors have nothing to disclose.
References
1. Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF 2003 Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci USA 100:3983-3988
2. Ponti D, Costa A, Zaffaroni N, Pratesi G, Petrangolini G, Coradini D, Pilotti S, Pierotti MA, Daidone MG 2005 Isolation and in vitro propagation of tumorigenic breast cancer cells with stem/progenitor cell properties. Cancer Res 65:5506-5511
3. Hirschmann-Jax C, Foster AE, Wulf GG, Nuchtern JG, Jax TW, Gobel U, Goodell MA, Brenner MK 2004 A distinct “side population” of cells with high drug efflux capacity in human tumor cells. Proc Natl Acad Sci USA 101:14228- 14233
4. Singh SK, Clarke ID, Terasaki M, Bonn VE, Hawkins C, Squire J, Dirks PB 2003 Identification of a cancer stem cell in human brain tumors. Cancer Res 63:5821-5828
5. Singh SK, Hawkins C, Clarke ID, Squire JA, Bayani J, Hide T, Henkelman RM, Cusimano MD, Dirks PB 2004 Identification of human brain tumour initiating cells. Nature 432:396-401
6. Prince ME, Sivanandan R, Kaczorowski A, Wolf GT, Kaplan MJ, Dalerba P, Weissman IL, Clarke MF, Ailles LE 2007 Identification of a subpopulation of cells with cancer stem cell properties in head and neck squamous cell carci- noma. Proc Natl Acad Sci USA 104:973-978
7. Ricci-Vitiani L, Lombardi DG, Pilozzi E, Biffoni M, Todaro M, Peschle C, De Maria R 2007 Identification and expansion of human colon-cancer-initi- ating cells. Nature 445:111-115
8. Li C, Heidt DG, Dalerba P, Burant CF, Zhang L, Adsay V, Wicha M, Clarke MF, Simeone DM 2007 Identification of pancreatic cancer stem cells. Cancer Res 67:1030-1037
9. Reya T, Morrison SJ, Clarke MF, Weissman IL 2001 Stem cells, cancer, and cancer stem cells. Nature 414:105-111
10. Bissell MJ, Labarge MA 2005 Context, tissue plasticity, and cancer: are tumor stem cells also regulated by the microenvironment? Cancer Cell 7:17-23
11. Bjerkvig R, Tysnes BB, Aboody KS, Najbauer J, Terzis AJ 2005 Opinion: the origin of the cancer stem cell: current controversies and new insights. Nat Rev Cancer 5:899-904
12. Jordan CT, Guzman ML, Noble M 2006 Cancer stem cells. N Engl J Med 355:1253-1261
13. Goodell MA 2002 Multipotential stem cells and ‘side population’ cells. Cy- totherapy 4:507-508
14. Welm B, Behbod F, Goodell MA, Rosen JM 2003 Isolation and characteriza- tion of functional mammary gland stem cells. Cell Prolif 36(Suppl 1):17-32
15. Behbod F, Xian W, Shaw CA, Hilsenbeck SG, Tsimelzon A, Rosen JM 2006 Transcriptional profiling of mammary gland side population cells. Stem Cells 24:1065-1074
16. Montanaro F, Liadaki K, Volinski J, Flint A, Kunkel LM 2003 Skeletal muscle engraftment potential of adult mouse skin side population cells. Proc Natl Acad Sci USA 100:9336-9341
17. Wulf GG, Luo KL, Jackson KA, Brenner MK, Goodell MA 2003 Cells of the hepatic side population contribute to liver regeneration and can be replenished with bone marrow stem cells. Haematologica 88:368-378
18. Majka SM, Beutz MA, Hagen M, Izzo AA, Voelkel N, Helm KM 2005 Identification of novel resident pulmonary stem cells: form and function of the lung side population. Stem Cells 23:1073-1081
19. Meeson AP, Hawke TJ, Graham S, Jiang N, Elterman J, Hutcheson K, Dimaio JM, Gallardo TD, Garry DJ 2004 Cellular and molecular regulation of skeletal muscle side population cells. Stem Cells 22:1305-1320
20. Umemoto T, Yamato M, Nishida K, Yang J, Tano Y, Okano T 2006 Limbal epithelial side-population cells have stem cell-like properties, including qui- escent state. Stem Cells 24:86-94
21. Martin CM, Meeson AP, Robertson SM, Hawke TJ, Richardson JA, Bates S, Goetsch SC, Gallardo TD, Garry DJ 2004 Persistent expression of the ATP- binding cassette transporter, Abcg2, identifies cardiac SP cells in the devel- oping and adult heart. Dev Biol 265:262-275
22. Kim M, Morshead CM 2003 Distinct populations of forebrain neural stem and progenitor cells can be isolated using side-population analysis. J Neurosci 23:10703-10709
23. Goodell MA, Brose K, Paradis G, Conner AS, Mulligan RC 1996 Isolation and functional properties of murine hematopoietic stem cells that are replicating in vivo. J Exp Med 183:1797-1806
24. Zhou S, Schuetz JD, Bunting KD, Colapietro AM, Sampath J, Morris JJ, Lagutina I, Grosveld GC, Osawa M, Nakauchi H, Sorrentino BP 2001 The ABC transporter Bcrp1/ABCG2 is expressed in a wide variety of stem cells and
is a molecular determinant of the side-population phenotype. Nat Med 7:1028- 1034
25. Haraguchi N, Inoue H, Tanaka F, Mimori K, Utsunomiya T, Sasaki A, Mori M 2006 Cancer stem cells in human gastrointestinal cancers. Hum Cell 19: 24-29
26. Haraguchi N, Utsunomiya T, Inoue H, Tanaka F, Mimori K, Barnard GF, Mori M 2006 Characterization of a side population of cancer cells from human gastrointestinal system. Stem Cells 24:506-513
27. Szotek PP, Pieretti-Vanmarcke R, Masiakos PT, Dinulescu DM, Connolly D, Foster R, Dombkowski D, Preffer F, Maclaughlin DT, Donahoe PK 2006 Ovarian cancer side population defines cells with stem cell-like characteristics and Mullerian inhibiting substance responsiveness. Proc Natl Acad Sci USA 103:11154-11159
28. Kondo T, Setoguchi T, Taga T 2004 Persistence of a small subpopulation of cancer stem-like cells in the C6 glioma cell line. Proc Natl Acad Sci USA 101:781-786
29. Kruger JA, Kaplan CD, Luo Y, Zhou H, Markowitz D, Xiang R, Reisfeld RA 2006 Characterization of stem cell-like cancer cells in immune-competent mice. Blood 108:3906-3912
30. Mitsutake N, Iwao A, Nagai K, Namba H, Ohtsuru A, Saenko V, Yamashita S 2007 Characterization of side population in thyroid cancer cell lines: cancer stem-like cells are enriched partly but not exclusively. Endocrinology 148: 1797-1803
31. Grichnik JM, Burch JA, Schulteis RD, Shan S, Liu J, Darrow TL, Vervaert CE, Seigler HF 2006 Melanoma, a tumor based on a mutant stem cell? J Invest Dermatol 126:142-153
32. Allolio B, Fassnacht M 2006 Adrenocortical carcinoma: clinical update. J Clin Endocrinol Metab 91:2027-2037
33. Goldstein LJ, Galski H, Fojo A, Willingham M, Lai SL, Gazdar A, Pirker R, Green A, Crist W, Brodeur GM, Lieber M, Cossman J, Gottesman MM. Pastan I 1989 Expression of a multidrug resistance gene in human cancers. J Natl Cancer Inst 81:116-124
34. Montanaro F, Liadaki K, Schienda J, Flint A, Gussoni E, Kunkel LM 2004 Demystifying SP cell purification: viability, yield, and phenotype are defined by isolation parameters. Exp Cell Res 298:144-154
35. Betz MJ, Shapiro I, Fassnacht M, Hahner S, Reincke M, Beuschlein F 2005 Peroxisome proliferator-activated receptor-y agonists suppress adrenocortical tumor cell proliferation and induce differentiation. J Clin Endocrinol Metab 90:3886-3896
36. Zwermann O, Schulte DM, Reincke M, Beuschlein F 2005 ACTH 1-24 in- hibits proliferation of adrenocortical tumors in vivo. Eur J Endocrinol 153: 435-444
37. Goodell MA, Mckinney-Freeman S, Camargo FD 2005 Isolation and char- acterization of side population cells. Methods Mol Biol 290:343-352
38. Kim CF, Jackson EL, Woolfenden AE, Lawrence S, Babar I, Vogel S, Crowley D, Bronson RT, Jacks T 2005 Identification of bronchioalveolar stem cells in normal lung and lung cancer. Cell 121:823-835
39. Singh S, Dirks PB 2007 Brain Tumor Stem Cells: Identification and Concepts. Neurosurg Clin North Am 18:31-38
40. Gazdar AF, Oie HK, Shackleton CH, Chen TR, Triche TJ, Myers CE, Chrou- sos GP, Brennan MF, Stein CA, La Rocca RV 1990 Establishment and char- acterization of a human adrenocortical carcinoma cell line that expresses mul- tiple pathways of steroid biosynthesis. Cancer Res 50:5488-5496
41. Rainey WE, Saner K, Schimmer BP 2004 Adrenocortical cell lines. Mol Cell Endocrinol 228:23-38
42. Camargo FD, Chambers SM, Drew E, McNagny KM, Goodell MA 2006 Hematopoietic stem cells do not engraft with absolute efficiencies. Blood 107:501-507
43. Matsuzaki Y, Kinjo K, Mulligan RC, Okano H 2004 Unexpectedly efficient homing capacity of purified murine hematopoietic stem cells. Immunity 20: 87-93
44. Bunting KD 2002 ABC transporters as phenotypic markers and functional regulators of stem cells. Stem Cells 20:11-20
45. Allen JD, Jackson SC, Schinkel AH 2002 A mutation hot spot in the Bcrp1 (Abcg2) multidrug transporter in mouse cell lines selected for doxorubicin resistance. Cancer Res 62:2294-2299
46. Doyle LA, Ross DD 2003 Multidrug resistance mediated by the breast cancer resistance protein BCRP (ABCG2). Oncogene 22:7340-7358
47. Christiansen JJ, Rajasekaran AK 2006 Reassessing epithelial to mesenchymal transition as a prerequisite for carcinoma invasion and metastasis. Cancer Res 66:8319-8326
Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.