Early Detection and Diagnosis
Consistent expression of the stem cell renewal factor BMI-1 in primary and metastatic melanoma
Article first published online: 27 JUN 2007
Copyright © 2007 Wiley-Liss, Inc.
International Journal of Cancer
Volume 121, Issue 8, pages 1764–1770, 15 October 2007
How to Cite
Mihic-Probst, D., Kuster, A., Kilgus, S., Bode-Lesniewska, B., Ingold-Heppner, B., Leung, C., Storz, M., Seifert, B., Marino, S., Schraml, P., Dummer, R. and Moch, H. (2007), Consistent expression of the stem cell renewal factor BMI-1 in primary and metastatic melanoma. Int. J. Cancer, 121: 1764–1770. doi: 10.1002/ijc.22891
- Issue published online: 14 AUG 2007
- Article first published online: 27 JUN 2007
- Manuscript Accepted: 28 MAR 2007
- Manuscript Received: 29 DEC 2006
- UBS AG Switzerland
- dermal nevi;
- stem cells;
Stem cell-like cells have recently been identified in melanoma cell lines, but their relevance for melanoma pathogenesis is controversial. To characterize the stem cell signature of melanoma, expression of stem cell markers BMI-1 and nestin was studied in 64 cutaneous melanomas, 165 melanoma metastases as well as 53 melanoma cell lines. Stem cell renewal factor BMI-1 is a transcriptional repressor of the Ink4a/Arf locus encoding p16ink4a and p14Arf. Increased nuclear BMI-1 expression was detectable in 41 of 64 (64%) primary melanomas, 117 of 165 melanoma metastases (71%) and 15 of 53 (28%) melanoma cell lines. High nestin expression was observed in 14 of 56 primary melanomas (25%), 84 of 165 melanoma metastases (50%) and 21 of 53 melanoma cell lines (40%). There was a significant correlation between BMI-1 and nestin expression in cell lines (p = 0.001) and metastases (p = 0.02). These data indicate that cells in primary melanomas and their metastases may have stem cell properties. Cell lines obtained from melanoma metastases showed a significant higher BMI-1 expression compared to cell lines from primary melanoma (p = 0.001). Further, primary melanoma lacking lymphatic metastases at presentation (pN0, n = 40) was less frequently BMI-1 positive than melanomas presenting with lymphatic metastases (pN1; n = 24; 52% versus 83%; p = 0.01). Therefore, BMI-1 expression appears to induce a metastatic tendency. Because BMI-1 functions as a transcriptional repressor of the Ink4a/Arf locus, p16ink4a and p14Arf expression was also analyzed. A high BMI-1/low p16ink4a expression pattern was a significant predictor of metastasis by means of logistic regression analysis (p = 0.005). This suggests that BMI-1 mediated repression of p16ink4a may contribute to an increased aggressive behavior of stem cell-like melanoma cells. © 2007 Wiley-Liss, Inc.
B-cell-specific Moloney murine leukaemia virus Integration site 1 (BMI-1) was first identified by retroviral insertional mutagenesis when assessing collaborating oncogenes in Eμ-myc transgenic mice.1, 2 The mouse BMI-1 gene encodes a nuclear protein of 45–47 kDa consisting of 324 amino acids.3 Its 3.6 kb cDNA is composed of 10 exons extended over 10 kb of genomic DNA. The human BMI-1 protein structurally and with respect to its amino acid composition is identical to the mouse BMI-1 protein and was the first member of the Polycomb gene family identified in mammals.4 Low BMI-1 expression is observed in almost all tissues, with higher expression in the brain, spinal cord, kidney, lung and gonads.3, 5 In addition, BMI-1 is highly expressed in embryonic stem cells and the placenta.1
BMI-1 is a transcriptional repressor of the Ink4a/Arf locus,6 which encodes 2 separate tumor suppressor genes, namely p16ink4a and p14Arf (p19Arf in mouse), which are read in alternating reading frames. p16ink4a is a cyclin-dependent kinase inhibitor, which blocks cell cycle entry by regulating the phosphorylation state of pRb (retinoblastoma protein). p16ink4a inhibits the kinase active sites of the cyclin D-dependent kinases Cdk4 and Cdk6 and prevents their associations with cyclins.7 p16ink4a normally is downregulated upon mitogen stimulation, thus promoting cell cycle progression. p16ink4a knock out mice develop normally, but have increased cancer susceptibility and develop multiple malignancies.8, 9 p14Arf promotes cell cycle arrest and apoptosis via its regulation of p53 stability stages. p14Arf effectively prevents the degradation of the tumor suppressor protein p53, which is required for cell cycle arrest.10, 11
In addition to their tumor suppressor activities, p16ink4a and 14Arf are modulators of senescence.12, 13 Senescence is a permanent growth arrest triggered by the accumulation of cell division, leading to telomeric loss. By impacting on p16ink4a and p14Arf simultaneously, BMI-1 plays a pivotal role in the regulation of the cell cycle progression, arrest and senescence. Since BMI-1 regulates the Ink4a/Arf locus, which is often deleted or silenced in tumors, it is not surprising that BMI-1 overexpression has been found in various cancer, such as colorectal cancer,14 nonsmall cell lung cancer,15 hepatocellular carcinoma,16 prostate cancer17 and neuroblastoma.18
BMI-1 has been recently shown to play a crucial role in self-renewal of stem cells.19, 20 Stem cells are slowly proliferating cells that give rise to more rapidly proliferating cells that eventually fully differentiate into mature tissue components. Cells with stem cell-like features have been identified in a number of human malignancies21-25 including melanoma.26, 27. Expression of the neural stem cell marker nestin has been demonstrated in melanoma28 and melanoma cell lines.26 Nestin is an intermediate filament expressed in neural crestic cells and proliferating neuronal progenitor cells, but not in the adult brain.29, 30
Overexpression of BMI-1 in stem cells could lead to uncontrolled growth and cancer.31 Absence of BMI-1 in hematopoietic stem cells leads to an accelerated differentiation, while overexpression promotes its self-renewal.31-33 The impaired hematopoietic stem cell self-renewal capacity is caused in part by the derepression of p16ink4a and p14Arf senescence pathways. Thus, BMI-1 is, at least in part, required for the maintenance of adult stem cells in some tissues because of its function in repressing senescence and cell death. Recently, we have shown the relevance of BMI-1 expression for medulloblastoma, a tumor derived from progenitor cells of the external granular layer of the cerebellum.19 Since melanomas and benign nevi are derived from neuroectodermal cells in the skin, we aimed to assess whether BMI-1 is also expressed by melanoma cells.
Consequently, we analyzed benign nevi, primary melanomas, melanoma metastases and cell lines, and demonstrate frequent expression of the stem cell markers BMI-1 and nestin. Moreover, we show that high BMI-1 expression is associated with an increased risk of metastatic disease. Besides being of prognostic relevance, these data may provide new insights in melanoma tumorigenesis.
Material and methods
Formalin-fixed, paraffin-embedded tissue of 64 primary cutaneous melanomas and 20 skin nevi (4 junctional, 10 dermoepidermal and 6 dermal) were immunohistochemically analyzed for BMI-1 expression. The 64 patients with primary cutaneous melanomas were previously reported in a sentinel lymph node (SLN) study.34 The patient age ranged from 20 to 75 years. Clinical follow-up was available in 54 of 64 patients (mean clinical follow up, 34 ± 12 month). There were 35 nodular, 13 superficial spreading, 1 lentiginous, 7 acral lentiginous, 7 not otherwise specified and 1 desmoplastic malignant melanoma. All melanomas had a Breslow tumor thickness ≥1.0 mm. Thirty four of the 64 patients had a Breslow tumor thickness from 1 to 2 mm, while tumor thickness was more than 2 mm in 30 patients. Forty three of 64 patients (67%) had a pN0 (sn; i-) SLN stage category. Nineteen of 64 patients had a pN1 or pN2 stage after lymphadenectomy. Five patients with negative SLN developed clinically obvious lymph node (n = 4) or organ (n = 1) metastases during follow-up. Tumor-matched samples of primary melanoma and metastasis were available in 8 patients. Eight of 54 (15%) patients died of metastatic disease during clinical follow-up. All except 1 had positive SLNs at the time of primary diagnosis.
To screen for BMI-1 expression in melanoma metastases, paraffin-embedded tissue of 165 melanoma metastases of different anatomic sites were identified in the archives of the Institute for Surgical Pathology Zurich. They consist of 116 brain, 24 lymph node, 7 soft tissue, 5 gastrointestinal, 4 skin, 4 lung, 2 liver, 1 parotis, 1 bone and 1 adrenal metastases.
In addition, cell lines derived from 53 melanomas (22 primary cutaneous melanomas, 26 metastases and 5 of unknown sites) were grown. The cell lines from metastases were from 12 skin, 6 lymph node, 3 cerebral and 1 gastric metastases. In 4 patients the origin was unknown. The melanoma cells were obtained from surgical specimen of melanoma patients included in a clinical trial. All patients provided written informed consent on the use of their tissue samples for the establishment of melanoma cell cultures.
Melanoma cell cultures
Melanoma cells were released from washed and fragmented tissue sections by means of serial incubation with dispase and collagenase as previously described.35 Outgrowing melanoma cells were routinely cultivated as monolayers in RPMI-1640 supplemented with 10% heat inactivated fetal calf serum (Seromed, Berlin, Germany), 5 mM glutamine (Biochrom KG, Berlin, Germany), 1 mM sodium pyruvate (Gibco™ Invitrogen AG, Basel, Switzerland) and 1% antibiotic mixture (Gibco) at 37°C and in 5% CO2 atmosphere. The melanoma cells proliferated in culture with a median growth time of 66 days (range 4–287 days). Cultures were within 3 passages at processing. Melanoma cell line UKRRV-Mel2 (kindly provided by Prof. Dirk Schadendorf, German Cancer Research Center, Skin Cancer Unit, Heidelberg, Germany), freshly isolated human fibroblasts and Jurkat lymphoma cell line (American Type Culture Collection, Manassas, VA) were used as controls.
Tissue microarray construction from metastases and cell lines
Paraffin-embedded tissue of 165 metastases was used for the preparation of a melanoma metastases tissue microarray. A morphologically representative region of the paraffin “donor” blocks was chosen. The representative region was taken with a core tissue biopsy (diameter, 0.6 mm; height, 3–4mm) and precisely arrayed into a new “recipient” paraffin block using a customer-built instrument.36 After the block construction was completed, 4.0-μm sections of the resulting tumor tissue microarray block were cut for further analysis.
For the cell line array, 2 drops of melanoma cell culture suspension was added to a Cytolyt-solution, when the cell count reached 20 × 106 cells/ml. Cells were centrifuged for 5 min at 2,000 rpm. After removing the supernatant, 4 drops of plasma were added to the pellet and the solution was mixed. One drop of thrombin was added. After 5 min the coagulated material was encapsulated for fixation in formalin 4% and embedded in paraffin. Cell cylinders of 0.6 mm in diameter were punched from the “cell line” tissue block and inserted into a recipient paraffin block by utilizing a custom-made precision instrument as described earlier. About 4.0 μm sections of the resulting cell line microarray were further processed as described earlier. An experienced cytologist (SK) reviewed all 53 cell lines and confirmed the presence of melanoma cells.
BMI-1 immunohistochemistry was performed on paraffin sections of formalin-fixed tissues, using a Ventana Benchmark automated staining system (Ventana Medical Systems, Tucson, Arizona) as recently described.19 For antigen retrieval, slides were heated with a cell conditioner 1 (standard procedure). Endogenous biotin was blocked with the appropriate kit. Primary monoclonal mouse antibody against BMI-1 (Upstate, Lake Placid, NY, clone F6, dilution 1:50) was then applied and revealed with the iVIEW DAB detection kit, yielding a brown reaction product. The signal was enhanced with the Ventana amplification kit. Slides were counterstained with hematoxylin prior to glass coverslipping.
Nestin and p14Arf immunohistochemistry was performed as described earlier. Primary monoclonal mouse antibodies against nestin (Chemicon, MAB5326, dilution 1:100) and p14Arf (Labvision Coporation, RB-9402-P1, dilution 1:25) were used. p16ink4a immunohistochemistry was performed as recently described.37, 38 Primary monoclonal mouse antibody against p16ink4a (Neomarkers, Freemont, CA, Clone 16P04, dilution 1:600) was used.
BMI-1, p16ink4a, p14Arf and nestin expression in primary melanoma were compared between different patient groups using the Mann-Whitney test. Correlations between BMI-1, p16ink4a, p14Arf, nestin and Breslow tumor thickness were analyzed using Spearman's rank correlation. Differences in tumor-specific survival between groups were calculated by log rank test. A logistic regression was performed to evaluate the predictive power of BMI-1 and p16ink4a expression in primary malignant melanoma for lymph node metastasis. p-Values below 0.05 were considered as significant. SPSS 12.0.1 for windows (SPSS) was used for statistical analyses.
BMI-1, p16ink4a, p14Arf and nestin expression in benign nevi and melanomas
Using the primary monoclonal mouse antibody against BMI-1 (clone F6) according to our recently reported staining procedure,19 spermatogonia and lymphocytes showed nuclear BMI-1 expression with a strong staining intensity. Therefore, spermatogonia in testes were used as external positive control. Lymphocytes in investigated benign nevi and melanomas served as positive internal control. Almost all benign nevi and melanomas showed BMI-1 expression in single cells or a faint nuclear staining, which was regarded as unspecific. To exclude the possibility of unspecific staining and to clearly define melanomas with increased BMI-1 expression, a high cutoff level was used to define tumors with increased BMI-1 expression. High BMI-1 expression was defined as the same staining intensity as lymphocytes in at least 50% of the tumor cells. Other patterns were categorized as low BMI-1 expression.
All 20 benign nevi (100%), 41 of 64 (64%) primary cutaneous melanomas and 117 of 165 melanoma metastases (71%) showed high BMI-1 expression (Fig. 1). There was no difference in BMI-1 expression between primary melanomas and metastases. Fifteen of 53 (28%) melanoma cell lines showed high BMI-1 expression. The remaining cell lines showed BMI-1 expression in less than 50% of the cells (Fig. 2). Cell lines obtained from melanoma metastases showed a significant higher BMI-1 expression than cell lines from primary melanomas (10/26 (38%) vs. 3/22 (15%), p = 0.001). There was no significant difference in BMI-1 expression between different anatomic sites (skin, lymph node, cerebral) of metastatic melanoma cell lines.
A significant difference in BMI-1 expression has been observed in primary malignant melanoma of patients with (n = 24) and without (n = 40) lymph node metastases. Primary melanomas with metastases showed more frequently a high BMI-1 expression compared with primary melanomas without metastases (83% versus 52%; p = 0.01) (Table I). Five of 8 matched tumor samples showed a similar high BMI-1 expression in the primary melanoma as well as in the corresponding metastases. Two patients showed low BMI-1 expression in the primary melanoma compared to high expression in the corresponding metastases. Only one patient showed high BMI-1 expression in the primary melanoma but not in its metastases. In our patient collective, Breslow tumor thickness did neither correlate with BMI-1 expression nor with presence of lymph node metastasis. There was no difference in BMI-1 expression between different histological melanoma types.
p16ink4a expression patterns and frequencies have recently been described in the primary melanomas of this cohort.37 In this study we used the same semiquantitative scoring system for p16ink4a as for BMI-1. For p16ink4a, nuclear and cytoplasmic staining was recorded because cytoplasmic staining is biologically relevant in malignant melanoma.37, 39-41 High p16ink4a expression was defined as nuclear and cytoplasmic or exclusive cytoplasmic positivity in at least 50% of tumor cells. In brief, 7 of 64 (11%) primary melanomas showed high nuclear and cytoplasmic—and 22 of 64 (34%) primary melanomas showed exclusive high cytoplasmic p16ink4a expression. High p16ink4a expression was associated with absence of lymph node metastases and increased patient survival. Herein we describe the frequency of p16ink4a expression in melanoma metastases and melanoma cell lines. One hundred twenty six of 165 metastases (76%) showed low p16ink4a expression. There was high nuclear and cytoplasmic p16ink4a in 8 of 165 (5%) metastases. Exclusive high p16ink4a cytoplasmic expression was detectable in 29 of 165 (18%) metastases. Fifty of 53 (94%) cell lines showed low p16ink4a expression. Three of 53 (6%) cell lines showed exclusively high cytoplasmic p16ink4a positivity. There was no statistical significant correlation between BMI-1 and p16ink4a expression, neither in primary melanomas nor in metastases or cell lines.
For p14Arf, a dot-like nuclear or cytoplasmic immunostaining was detected. Tumors with a dot-like nuclear and/or cytoplasmic p14Arf expression were regarded as p14Arf positive (Fig. 2). Ten of 64 (16%) primary melanomas, 32 of 165 metastases (19%) and 13 of 53 (25%) cell lines were p14Arf positive. All 20 nevi were p14Arf negative. Neither nuclear nor cytoplasmic p14Arf expression was associated with BMI-1-, p16ink4a-expression or metastatic disease.
For nestin, a cytoplasmic immunostaining was detectable. Fifty six of 64 primary melanomas were available for nestin evaluation. Using the 50% cutoff to define tumors with high nestin expression, 14 of 56 (25%) primary melanomas, 84 of 165 metastases (50%) and 21 of 53 cell lines (40%) showed increased nestin expression (Fig. 2). There was a significant correlation between BMI-1 and nestin expression in cell lines (p = 0.001) and melanoma metastases (p = 0.02), but not in primary melanoma. Nestin expression was not associated with lymph node metastases.
BMI-1 and melanoma progression
Tumor thickness (p = 0.002), LN status (p < 0.0001) and p16ink4a expression (p = 0.001) were associated with tumor-specific survival but there was no association between BMI-1 and nestin expression and survival. Univariate logistic regression analysis in primary malignant melanoma demonstrated that high BMI-1 and low p16ink4a expression are predictors of metastatic disease (p = 0.02 and 0.04; respectively; Table II). To analyze the correlation between BMI-1-, p16ink4a-expression and metastatic disease, 3 categories were defined. Category 1: low BMI-1/high p16ink4a expression; Category 2: high BMI-1/high p16ink4a or low BMI-1/low p16ink4a expression; Category 3: high BMI-1/low p16ink4a expression. There were 15 melanomas in Category 1, 27 in Category 2 and 22 in Category 3. Interestingly, primary melanomas in Category 3 had a higher probability for metastases than all other categories. The statistical significance of combined BMI-1 and p16ink4a analysis was even stronger (p = 0.005) than for BMI-1 alone (p = 0.03). In multivariate logistic regression analysis, high BMI-1 expression was an independent predictor of metastatic disease, whereas p16ink4a negativity did not reach statistical significance.
|n||Univariate OR||p-value||Multivariate OR||p-value|
|p16ink4a low vs. high1||35/29||3.0 (1.0–8.6)2||0.04||2.7 (0.89–8.1)||0.08|
|BMI-1 high vs. Low1||41/23||4.5 (1.3–15.6)||0.02||4.1 (1.2–14.6)||0.03|
|p16ink4a low/BMI-1 high vs. others1||22/42||3.2 (1.4–7.3)||0.005|
Here we demonstrate that a significant percentage of human melanomas has stem cell properties and that these tumors are associated with higher incidence of metastases. Different molecular mechanisms are involved in maintenance of the self-renewal capacity of stem cells, which also can contribute to tumorigenesis. The most prominent example is the Polycomb group gene BMI-1, whose role in control of self-renewal of hematopoetic and neural stem cells and its overexpression in leukemia and brain tumors, most notably medulloblastomas,19 is well known. BMI-1 protein expression was already reported in various human cancers.14, 15, 18, 31, 42-48. Using an immunohistochemical approach, we describe for the first time BMI-1 expression in human malignant melanomas. Immunohistochemically, weak nuclear expression with low staining intensity was observed in many different human tissues.1 Therefore, we used very stringent criteria to define increased BMI-1 expression levels in melanocytic lesions. Increased BMI-1-expression was given, if at least 50% of melanocytic cells were positive with a staining intensity similar to lymphocytes and spermatogonia.
BMI-1 is highly expressed in specific neuroepithelial cells. Therefore, it was not surprising that all benign nevi and 60–70% of melanomas showed high BMI-1 expression levels. Melanocytes are neuroectodermal cells, and hence, markers of neuroectodermal differentiation can be found in almost all melanomas. Importantly, high BMI-1 expression levels may serve as an argument for stem cell properties of melanoma cells. BMI-1 is highly expressed in both hematopoietic and neural stem cells. In the brain, BMI-1 is expressed in the subventricular zone, the source of neural stem cells, as well as in cortical neural stem cells and progenitor cells.20, 49 The expression of the stem cell marker nestin is also consistent with a particular high percentage of melanoma cells with stem cell characteristics in a large subgroup of human melanoma. Nestin expression showed high intratumoral heterogeneity with positive and negative melanoma cells within the same tumor. There was a significant correlation between BMI-1 and nestin expression in melanoma metastases and melanoma cell lines, but not in primary melanoma. Importantly, melanoma metastases are more frequently positive for nestin than for primary melanomas, which is in line with a previous study.28 It is tempting to speculate that melanoma cells with stem cell properties have an increased metastatic behavior and those stem cells or stem cell-like cells are therefore enriched in melanoma metastases.
Specific stem cell populations may account for tumor formation21-23, 50, 51 and stem cell renewal factors like BMI-1 have been implicated in the generation of different tumors. Recent analyses indicate a tumorigenic subpopulation with stem cell properties in melanoma. Two recent studies26, 27 described subpopulations with stem cell properties in melanomas and melanoma cell lines. These cells maintained stem cell characteristics after cloning and were able to self-renew and differentiate into melanogenic, adipogenic, chondrogenic and osteogenic lineages.27 Immature melanocytic are present in normal human epidermis26 and melanocytic stem cells have been identified in the murine hair follicle.52 The high risk of metastatic disease of malignant melanoma, often with delayed onset, is frequently explained by more primitive stem cell biology.
Importantly, high BMI-1 expression levels in melanoma were associated with lymph node metastases. Our finding is consistent with observations of BMI-1 expression in breast cancer: Kim et al. showed a positive correlation between BMI-1 overexpression in invasive ductal breast cancer and presence of axillary lymph node metastases.44 A prognostic relevance of BMI-1 expression was also shown for non-Hodgkin B-cell lymphomas and the myelodysplastic syndrome.45, 48 Highly aggressive non-Hodgkin B-cell lymphomas had significantly higher BMI-1 levels compared to low-grade malignant B-cell non-Hodgkin lymphoma.48 Recently, a correlation between BMI-1 expression and poor prognosis was also described for nasopharyngeal carcinomas.47
Interestingly, Raaphorst et al. reported increasing BMI-1 expression levels during progression from ductal carcinoma in situ to invasive cancer.46 A stepwise progression model with accumulation of genetic alterations has been proposed for breast cancer.53 Therefore, neoexpression of BMI-1 in less-differentiated tumors could explain the association with metastases in breast cancer. This is in contrast to melanoma, because all nevi and the majority of melanomas are BMI-1 positive. One might hypothesize that specific molecular mechanisms are responsible for the increased metastatic behavior of BMI-1 positive tumors.
Our data show that a high BMI-1/low p16ink4a phenotype is strongly linked to lymph node metastasis, although there was no direct correlation between BMI-1 and p16ink4a expression in primary cutaneous melanoma. This provides indirect evidence that the interaction between BMI-1 and p16 ink4a influences the metastatic behavior of melanoma cells. Others and we37, 54, 55 have previously shown the strong prognostic relevance of p16ink4a in melanoma. p16ink4a is inactivated by mutation, promotor methylation or loss of heterozygosity in familial and sporadic melanoma.54, 56 Jacob et al. showed a downstream effect of BMI-1 on p16ink4a.57 Therefore, BMI-1 may play a role in the regulation of the cell cycle progression by influencing p16ink4a, which might lead to increased proliferation of melanoma cells.
In summary, we demonstrate a frequent expression of the stem cell renewal marker BMI-1 in malignant melanomas. The BMI-1 expression level may be relevant for the metastatic behavior of melanoma cells and might have a biologically important downstream effect on p16ink4a.
We are grateful to Mr. N. Wey for photographic reproductions.
- 40Electron microscopy evidence that cytoplasmic localization of the p16(INK4A) “nuclear” cyclin-dependent kinase inhibitor (CKI) in tumor cells is specific and not an artifact. A study in non-small cell lung carcinomas. Biotech Histochem 2004; 79: 5–10., , , , , , , , , .
- 52Dominant role of the niche in melanocyte stem-cell fate determination. Nature 2002; 416854–60., , , , , , , , , .