ARTICLE
Auteur(s) : Kenichi Tanabe1,
Yasuyuki Amoh1, Maho Kanoh1, Hiroshi
Takasu1, Naohiko Sakai2, Yuichi
Sato3, Kensei Katsuoka1
1Department of Dermatology, Kitasato University
School of Medicine, 1-15-1 Kitasato, Sagamihara, Kanagawa
228-8555, Japan
2Department of Plastic Surgery, Kitasato University
School of Medicine
3Department of Molecular Diagnostics, Kitasato
University School of Allied Health Sciences, Sagamihara, Kanagawa,
Japan
accepté le 9 Decembre 2009
Nestin is a neural stem cell marker protein that is also
expressed in the bulge area stem cell compartment of the hair
follicle. Nestin-expressing hair follicle stem cells give rise to
the outer root sheath and the nestin-expressing interfollicular
vascular network. Nestin-expressing stem cells isolated from the
hair follicle bulge region are negative for the keratinocyte marker
keratin 15 (K15), and are able to differentiate into numerous
cell types in vitro, including neurons, glia, keratinocytes, smooth
muscle cells and melanocytes. This primitive state demonstrated by
the nestin-positive stem cells is compatible with their
pluripotency [1-9]. Moreover, as various types of cutaneous tumors,
including melanoma, originate from the hair follicle and/or
epidermal stem cells [10], nestin expression is thought to be
important in tumorigenesis.
In the present study, we examined nestin expression in malignant
melanoma and investigated the relationship between nestin
expression levels and the prognosis of patients with malignant
melanomas.
Materials and methods
Patients and tumor collection
Formalin-fixed and paraffin-embedded tissues were obtained from
78 (35 male and 43 female) malignant melanoma
patients who were observed over a period of at least 5 years
and who were surgically resected at the Department of Dermatology
or Plastic Surgery of Kitasato University Hospital in Japan. The
patients (4 to 85 years old; median, 56.6 years old)
between 1988 and 2004 were retrospectively reviewed. And
we observed the five year or more passages (the longest is
21 years) of all the cases, excluding the patients who died
within five years. All patients underwent subsequent tumor
excision. The tumor samples analyzed were obtained from central
areas of the primary tumor. Paraffin blocks representing the most
typical tumor tissue (mainly central portions of primary tumors)
were selected, and 4-μm thick sections were prepared for
immunohistochemistry. The samples were histologically analyzed
according to the pathological staging criteria of the American
Joint Committee on Cancer (AJCC) and subdivided into 6 cases
of stage 0, 18 cases of stage I, 33 cases of stage II,
13 cases of stage III and 8 cases of stage IV. The
majority of patients in stages II to IV, 51 out of 54 (without
three patients in Stage IV), received chemotherapy (DAV-F:
intravenous (i.v.) dacarbazine, 160 mg/m2/day on
days 1-5; nimustine, 80 mg/m2/day i.v. on day 1;
vincristine, 0.8 mg/m2/day i.v. on day 1; and
β-feron, 3 million Units s.c. on days 1-5). The survival time
was calculated as the date of surgery to the date of death.
Informed consent was obtained from all patients prior to admission
into the study. This study was approved by the Ethics Committee of
Kitasato University School of Medicine.
Immunohistochemical analysis
Indirect immunohistochemical staining was performed using the
following primary antibodies: polyclonal anti-nestin (1:20, IBL,
Gunma, Japan), monoclonal anti-melanosome (clone HMB-45, undiluted,
Dako, Glostrup, Denmark) and polyclonal anti-S100 (1:400,
Dako). Bound primary antibodies were detected using the dextran
polymer reagent (ChemMate Envision, Dako). Following
deparaffinization, rehydration and elimination of endogenous
peroxidase activity by treatment with 3% hydrogen peroxide for
5 min at room temperature (RT), sections were incubated with
5% fetal calf serum in PBS for 5 min at RT to block
non-specific protein binding. The sections were incubated with each
antibody for 60 min and then with ChemMate Envision for
30 min at RT. Finally, the color reaction was developed with
8-amino-9-ethylcarbazole solution (Dako) and the sections
counterstained with Mayer's hematoxylin (Wako Pure Chemical, Osaka,
Japan). S100 protein staining was considered positive when
definite expression was observed in the nucleus and/or cytoplasm of
> 5% of tumor cells, while HMB-45 and nestin staining was
considered positive when definite expression was observed in the
cytoplasm of tumor cells. The slide of Nestin staining was
evaluated according to staining extent and intensity. Staining
extension was assessed by the percentage of stained cells and
scored semi-quantitatively, using a 0 to 4 scale for
expression: 0 = no expression; 1+ = 1-25%, 2+ = 26-50%, 3+ =
51-75%, 4+ = 76-100%. Staining intensity was categorized into three
groups by comparing the staining intensity of tumor cells with
vascular endothelial cells: 1+ = weaker than endothelial cells; 2+
= same as endothelial cells; and 3+ = stronger than endothelial
cells. By adding the staining extensity and intensity scores, the
combined scores were calculated. The combined scores were then
divided into 4 groups: Negative = combined score 0; weak
staining = combined score 2, moderate staining = combined scores
3-4; strong staining = combine scores 5-7 [10].
Analysis of clinical parameters
The overall survival of the patients among the 6 groups was
analyzed using the Kaplan-Meier method, Z-test and Fisher's exact
test. Values < 0.05 were considered statistically
significant.
Results
Immunohistochemical analysis
The expression of S100 protein was detected in the nucleus and
cytoplasm of tumor cells in all cases. The expression of
HMB-45 protein was also observed in the cytoplasm of tumor
cells. The expression of HMB-45 was detected in 5 out of
6 (83.3%) stage 0 tumors, 17 out of 18 (94.4%)
stage I tumors, 29 out of 33 (87.9%) stage II tumors,
10 out of 13 (76.9%) stage III tumors, and 8 out of
8 (100%) stage IV tumors. Nestin expression was observed in
the cytoplasm of tumor cells in 59.0% of malignant melanomas (n =
78). Positivity rates were 0 out of 6 (0%) stage
0 tumors, 6 out of 18 (33.3%) stage I tumors,
23 out of 33 (69.7%) stage II tumors, 10 out of
13 (76.9%) stage III tumors and 7 of 8 (87.5%) stage
IV tumors. The frequency and combined scores of nestin expression
of groups are shown in table 1. Average
combined score was 0 in stage 0 tumors, 1.5 in stage
I tumors, 3.18 in stage II tumors, 3.46 in stage III
tumors and 4.57 in stage IV tumors. Average combined score
rose gradually as the stage progressed. In addition, some vascular
endothelial cells, fibroblasts and peripheral nerve cells were also
nestin positive. Figure
1 indicates the nestin- and HMB-45-positive melanoma cells
in the junctional area (figure 1A), dermis (figure 1B), and deep
dermis (figure
1C).
Table 1 The frequency and combined scores (all
patients)
|
Stage 0
|
Stage I
|
Stage II
|
Stage III
|
Stage IV
|
|
Nestin expression
|
|
|
Positive
|
0 -
|
6 (33%)
|
23 (70%)
|
9 (69%)
|
7 (88%)
|
|
Negative
|
6 (100%)
|
12 (67%)
|
10 (30%)
|
4 (31%)
|
1 (13%)
|
|
Extension scores
|
|
|
0
|
6 (100%)
|
12 (67%)
|
10 (30%)
|
3 (23%)
|
1 (13%)
|
|
1+
|
0 -
|
1 (6%)
|
5 (15%)
|
1 (8%)
|
1 (13%)
|
|
2+
|
0 -
|
4 (22%)
|
2 (6%)
|
3 (23%)
|
0 -
|
|
3+
|
0 -
|
0 -
|
9 (27%)
|
4 (31%)
|
3 (38%)
|
|
4+
|
0 -
|
1 (6%)
|
7 (21%)
|
2 (15%)
|
3 (38%)
|
|
Intensity scores
|
|
|
1+
|
0 -
|
1 (6%)
|
8 (24%)
|
4 (31%)
|
0 -
|
|
2+
|
0 -
|
2 (11%)
|
8 (24%)
|
2 (15%)
|
5 (63%)
|
|
3+
|
0 -
|
3 (17%)
|
7 (21%)
|
4 (31%)
|
2 (25%)
|
|
Combined scores
|
|
|
0 (negative)
|
6 (100%)
|
12 (67%)
|
10 (30%)
|
3 (23%)
|
1 (13%)
|
|
2 (weak)
|
0 -
|
1 (6%)
|
3 (9%)
|
1 (8%)
|
0 -
|
|
3-4 (moderate)
|
0 -
|
2 (11%)
|
6 (18%)
|
3 (23%)
|
1 (13%)
|
|
5-7 (strong)
|
0 -
|
3 (17%)
|
14 (42%)
|
5 (38%)
|
6 (75%)
|
|
Average scores (0-7)
|
0 -
|
1.5
|
3.18
|
3.46
|
4.57
|
Prognostic relevance
Follow-up of patients ranged from 5 to 180 months. During
this time, 29.5% of the patients died. The survival curves for each
of the different tumor stages are presented in figure 2, and the 5-year
survival rates of nestin-positive and nestin-negative patients are
shown in figure
3. The 5-year survival rates were 6 out of
6 (100%) for stage 0 tumors, 16 out of
18 (88.9%) for stage I tumors, 28 out of 33 (84.8%)
for stage II tumors, 8 out of 13 (61.5%) for stage III
tumors and 1 out of 8 (12.5%) for stage IV tumors. The
difference in survival rate among the stages was significant (figure 2). Comparison
of the survival rate between nestin-positive patients and
nestin-negative patients for each stage also showed significant
differences in all cases (P < 0.001 Fisher's exact test).
That is, the 5-year survival rate was 100% in all patients
exhibiting nestin-negative tumors (stage 0: 6 cases, stage I:
12 cases, stage II: 10 cases, stage III: 3 cases and
stage IV: 1 case). In stage I tumors, nestin-positive patients
demonstrated a decreased 5-year survival rate (66.7%) compared to
the nestin-negative patients (100%, P < 0.05, Z-test).
Nestin-positive patients also exhibited a decreased 5-year survival
rate (78.3%) in patients with stage II tumors compared to the
nestin-negative patients with stage II tumors (100%, P <
0.05 Z-test) (figure 3). And the
frequency and combined scores of nestin expression of groups,
divided 5-year survival groups and death groups, are shown in table 2. In each stage, average combined
scores in 5-year survival groups were higher than in death
groups.
Table 2 The frequency and combined scores (divided 5
year survival groups and death groups)
|
Stage 0
|
Stage I
|
Stage II
|
Stage III
|
Stage IV
|
|
Survive
|
Death
|
Survive
|
Death
|
Survive
|
Death
|
Survive
|
Death
|
Survive
|
Death
|
|
Nestin expression
|
|
|
Positive
|
0
|
0
|
4
|
2
|
18
|
5
|
4
|
5
|
0
|
7
|
|
Negative
|
6
|
0
|
12
|
0
|
10
|
0
|
4
|
0
|
1
|
0
|
|
Extension scores
|
|
|
0
|
6
|
0
|
12
|
0
|
10
|
0
|
4
|
0
|
1
|
0
|
|
1+
|
0
|
0
|
1
|
0
|
4
|
1
|
1
|
0
|
0
|
1
|
|
2+
|
0
|
0
|
2
|
2
|
2
|
0
|
0
|
3
|
0
|
0
|
|
3+
|
0
|
0
|
0
|
0
|
7
|
2
|
3
|
1
|
0
|
3
|
|
4+
|
0
|
0
|
1
|
0
|
5
|
2
|
1
|
1
|
0
|
3
|
|
Intensity scores
|
|
|
1+
|
0
|
0
|
1
|
0
|
6
|
2
|
3
|
1
|
1
|
0
|
|
2+
|
0
|
0
|
1
|
1
|
7
|
1
|
1
|
1
|
0
|
5
|
|
3+
|
0
|
0
|
2
|
1
|
5
|
2
|
1
|
3
|
0
|
2
|
|
Combined scores
|
|
|
0 (negative)
|
6
|
0
|
12
|
0
|
10
|
0
|
3
|
0
|
1
|
0
|
|
2 (weak)
|
0
|
0
|
1
|
0
|
2
|
1
|
1
|
0
|
0
|
0
|
|
3-4 (moderate)
|
0
|
0
|
1
|
1
|
6
|
0
|
1
|
2
|
0
|
1
|
|
5-7 (strong)
|
0
|
0
|
2
|
1
|
10
|
4
|
2
|
3
|
0
|
6
|
|
Average scores (0-7)
|
0
|
0
|
1.13
|
4.5
|
2.85
|
5
|
2.5
|
5
|
0
|
5.43
|
Discussion
Nestin is a neural stem cell marker protein that is also expressed
in hair follicle stem cells in the bulge region [5-9]. The
nestin-expressing hair follicle stem cells give rise to the outer
root sheath and the nestin-expressing interfollicular vascular
network in nestin-GFP transgenic mice [6]. We have recently
demonstrated that nestin-expressing stem cells isolated from the
hair follicle stem cell region in mice that were negative for the
keratinocyte marker K15 were able to differentiate into
neurons, glia, keratinocytes, smooth muscle cells and melanocytes
in vitro [5-9]. In addition to being K15-negative, the pluripotent
nestin-expressing stem cells are also positive for the stem cell
marker CD34, demonstrating their relatively undifferentiated state.
This primitive state of the nestin-expressing stem cells appears to
be compatible with their pluripotency [6-9]. Recently, numerous
studies have suggested that various forms of cutaneous tumors,
including melanoma, originate from the hair follicle and epidermal
stem cells. Brychtova et al. [11] demonstrated via
immunohistochemical analysis that nestin is expressed in malignant
melanoma and melanocytic nevi. They also demonstrated that nestin
immunoreactivity was significantly increased in malignant
melanomas, and that the precise levels of nestin correlated with
the clinical stage of the tumor. Moreover, immunohistochemical
analysis demonstrated nestin-positive cells in 35 of
42 (83.3%) nodular melanomas, 10 of 32 (31.3%)
superficial spreading melanomas, 10 of 12 (83.3%)
metastatic melanomas, 2 of 10 (20.0%) dysplastic nevi and
20 of 43 (46.5%) nevomelanocytic nevi [12]. Thus, the
expression levels of nestin significantly correlate with the
aggressiveness of malignant melanoma [13]. In addition, a
significantly greater percentage of CD166-, CD133- and
nestin-positive tumor cells were identified in malignant melanomas
compared to nevomelanocytic nevi, while all cases of metastatic
melanoma expressed at least one stem cell marker. However,
statistical significance for nestin expression was only detected
between the primary and metastatic melanomas [13]. Moreover, nestin
expression has not been observed in HMB-45-negative melanotic and
amelanotic malignant melanomas [12].
As the precise diagnosis of HMB-45-negative malignant melanoma
is clinically most important, we investigated the relationship
between nestin expression and melanoma class in the present study.
We found that nestin was a useful marker for the diagnosis of
HMB-45-negative malignant melanoma. Tumor cells in epidermal
lesions that also expressed melanin failed to express nestin. This
result correlated with previous studies demonstrating that tumor
cells proliferating into the dermis, especially within the invasive
front, lacked melanin and expressed nestin [12]. Xu et al.
evaluated the reactivity of a panel of antibodies against markers
associated with melanoma and melanocytic differentiation in
HMB-45-negative, non-desmoplastic melanomas. They concluded that
melanoma antigen 1 (MAGE-1), melanocyte-specific transcription
factor (MITF), tyrosinase and Melan-A served as useful markers for
the diagnosis of malignant melanotic lesions when HMB-45 is
not present [14]. In contrast, we reported the presence of Melan-A
in only 6 out of 10 cases of dermal lesions in nodular
malignant melanoma, and only 2 out of 5 cases of dermal
lesions of amelanotic malignant melanoma. In addition, MITF
immunoreactivity was only observed in 5 out of 10 cases
of dermal lesions of nodular malignant melanoma and 4 out of
5 cases of dermal lesions of amelanotic malignant melanoma,
while MAGE-1 immunoreactivity was only observed in 1 case
of dermal lesion, 10 cases of nodular malignant melanoma and
5 cases of amelanotic malignant melanoma. Thus, nestin may be
a useful marker of HMB-45-negative melanoma cells in dermal lesions
of melanotic and amelanotic nodular malignant melanomas [12].
Recently, Flammiger et al. described that the intermediate
filament protein and stem cell marker nestin, as well as the
lineage restricted transcription factors BRN2, SOX9, and SOX10, are
expressed in melanoma cell lines of all progression stages as well
as in melanoma tissues, and that SOX9 and SOX10 but not
BRN2 can alter nestin expression in melanoma [15]. And Bakos
et al. demonstrated that nestin and SOX9 expression are
increased, respectively, in ulcerated melanomas and advanced-stage
melanoma, and may be markers of tumor aggressiveness [16]. In
addition, Yang et al. suggested that the expression of nestin
may play an important role in the development of some neoplasms,
such as GIST and angiosarcoma [17].
In the present study, we investigated the relationship between
nestin expression and patient outcome in malignant melanomas. We
found that nestin, HMB- 45 and S100 protein were detected
in 56.5%, 88.4% and 100% of malignant melanomas, respectively. The
5-year survival rates of stages I and II nestin-positive cases was
significantly decreased compared to the nestin-negative cases (p
< 0.05). In addition, the 5-year survival rates were 100% in
nestin-negative malignant melanomas at all stages of tumor
development. We conclude that nestin expression may be a predictor
of poor prognosis in patients with malignant melanoma.
Acknowledgements
This work was partially supported by Grants-in-Aid for the All
Kitasato Project Study 2007 and 2008, and the Parents’ Association
Grand of Kitasato University, School of Medicine. We are also
grateful to Ms Masako Ishii for providing excellent technical
assistance. Conflict of interest: none.
References
1 Hoffman RM. The hair follicle as a gene therapy target. Nat
Biotechnol 2000; 18: 20-1.
2 Oshima H, Rochat A, Kedzia C, Kobayashi K,
Barrandon Y. Morphogenesis and Renewal of Hair Follicles from
Adult Multipotent Stem Cells. Cell 2001; 104: 233-45.
3 Cotsarelis G, Sun T-T, Lavker RM.
Label-retaining cells reside in the bulge area of pilosebaceous
unit: implications for follicular stem cells, hair cycle, and skin
carcinogenesis. Cell 1990; 61: 1329-37.
4 Taylor G, Lehrer MS, Jensen PJ, Sun TT,
Lavker RM. Involvement of Follicular Stem Cells in Forming Not
Only the Follicle but Also the Epidermis. Cell 2000; 102:
451-61.
5 Li L, Mignone J, Yang M, et al. Nestin
expression in hair follicle sheath progenitor cells. Proc Natl Acad
Sci USA 2003; 100: 9958-61.
6 Amoh Y, Li L, Yang M, et al. Nascent blood
vessels in the skin arise from nestin-expressing hair-follicle
cells. Proc Natl Acad Sci USA 2004; 101: 13291-5.
7 Amoh Y, Li L, Yang M, et al. Multipotent
nestin-positive, keratin-negative hair-follicle bulge stem cells
can form neurons. Proc Natl Acad Sci USA 2005; 102: 5530-4.
8 Amoh Y, Li L, Campillo R, et al. Implanted
hair follicle stem cells form Schwann cells which support repair of
severed peripheral nerves. Proc Natl Acad Sci USA 2005; 102:
17734-8.
9 Amoh Y, Li L, Katsuoka K, Hoffman RM.
Chemotherapy targets the hair-follicle vascular network but not the
stem cells. J Invest Dermatol 2007; 127: 11-5.
10 Levent Y, Mehmet K, Oguz A, Bedri K.
Fascin expression in melanocytic lesions of the skin. Eur J
Dermatol 2009; 19: 445-50.
11 Brychtova S, Fiuraskova M, Hlobilková A,
Brychta T, Hirnak J. Nestin expression in cutaneous
melanomas and melanocytic nevi. J Cutan Pathol 2007; 34: 370-5.
12 Kanoh M, Amoh Y, Tanabe K, Maejima H,
Takasu H, Katsuoka K. Nestin is expressed in HMB-45
negative melanoma cells in dermal lesion of nodular malignant
melanoma. Eur J Derm 2008; 18: 518-23.
13 Klein WM, Wu BP, Zhao S, Wu H,
Klein-Szanto AJ, Tahan SR. Increased expression of stem
cell markers in malignant melanoma. Mod Pathol 2007; 20: 102-7.
14 Xu X, Chu AY, Pasha TL, Elder DE,
Zhang PJ. Immunoprofile of MITF, Tyrosinase, Melan-A, and
MAGE-1 in HMB-45 negative Melanomas. Am J Surg Pathol 2002; 26:
82-7.
15 Flammiger A, Besch R, Cook AL, Maier T,
Sturm RA, Berking C. SOX9 and SOX10 but Bot BRN2 Are
Required for Nestin Expression in Human Melanoma Cells. J Invest
Dermatol 2009; 129: 945-53.
16 Bakos RM, Maier T, Besch R, Mestel DS, Ruzicka T, Sturm RA,
Berking C. Sturm and Carola Berking. Nestin and SOX9 and SOX10
transcription factors are coexpressed in melanoma. Exp Dermatol
2009 Oct 21 (Epub ahead of print).
17 Yang XH, Wu QL, Yu XB, et al. Nestin
expression and malignant grade for different tumors. J Clin Pathol
2008; 61: 467-73.
|
Printable version
Version PDF
