Você está na página 1de 7

4193

J ournal of Cell and Tissue Research Vol. 14(2) 4193-4199 (2014)


(Available online at www.tcrjournals.com)
Original Article
ISSN: 0973-0028; E-I SSN: 0974-0910
IMMUNOHISTOCHEMICAL EXPRESSION OF CD 44, CD24 AND ESA
IN CANINE MAMMARY TUMOURS
ANJAN KUMAR, K. R.,
1
SUDHAKAR RAO, G. V.,
2
BALACHANDRAN, C.,
2
MURALI
MANOHAR, B.,
2
DHANAN JAYA RAO, C. G.,
3
UZAMA, S.
3
AND SHAMMI, M.
3
1
Department of Veterinary Pathology, Veterinary College, Shimoga, KVAFSU;
2
Department of Veterinary
Pathology, Madras Veterinary College, Chennai, TANUVAS;
3
Department of Veterinary Surgery and
Radiology, Madras Veterinary College, Chennai, TANUVAS. E. mail: drvet_anjan@hotmail.com
Received: February 4, 2014; Accepted: June 15, 2014
Abstract: Identification of Cancer stem cell lineage subpopulations of tumor cells within the
tumor mass offer to identify targets for therapeutic interventions and a better understanding of
carcinogenesis. The main objective of this study was to evaluate the prognostic value of cancer
stem cell lineage markers i.e., CD44, CD24 and Epithelial specific antigen (ESA) in canine
mammary tumours using triple immunohistochemistry. A total of 40 surgically excised tissue
samples of canine mammary tumours were obtained during the eight month study period. These
tumors were classified according to WHO classification of mammary tumors of dog (2002), as
Carcinoma tubular, Carcinoma Papillary cystic, Squamous cell carcinoma and
Carcinosarcoma with tumor grades I & II respectively. Of 40 samples, 24 selected cases were
subjected to triple immunohistochemistry using antibodies against CD 44, CD 24 and ESA. Two
cases were positive for cancer stem cell phenotype i.e., CD44
+
/CD24
-/low
/ ESA
+
showing colocalized
expression of CD44, epithelial specific antigen (ESA) with membranous staining and absence or
isolated staining for CD24, which correlated with both the cases showing distant metastases and
reduced average survival period of 122 days, indicating poor survivability and distant metastasis.
CD44
+
/CD24
-
/ESA
+
expression was seen in carcinosarcoma (Grade I) indicating an epithelial
mesenchymal transition and could favor metastasis. triple immunohistochemical expression of
cancer stem cell phenotype i.e., CD44
+
/CD24
-
/ESA
+
favored distant metastasis with poor prognosis
and could be used as a prognostic indicator in canine mammary tumors as evident from this
study.
Key words: Canine mammary tumor, CD 44, CD24, ESA
INTRODUCTION
Tumours of the mammary glands are one of the
common neoplasms encountered in female dogs and
ranked second compared to skin neoplasms [1]. The
incidence of mammary carcinoma in canines is three
times that documented in humans and shares several
important epidemiological, morphological, clinicopa-
thological and biochemical features compared to
humans [2]. Close similarity that exist between cancer
genes of dog and human cancer tissue especially of
?
human tumor initiating cells (CD44
+
CD24
-
) makes
the dogs as an excellent tumor models [2-6]. A
population of cancer stem cells (CSC) that have the
exclusive ability to extensively proliferate and form
new tumors can be identified based on marker
expression especially, CD44
+
/ CD24
-/low
ESA
+
which
have more than 50 fold tumourigenic potential and
as few as 200 cells with this phenotype were able to
form tumours in SCID mice consistently, whereas
other lineages failed to form tumors [7]. CSC
population with the phenotype of CD44
+
/ CD24
-/low
4194
J. Cell Tissue Research
correlated with metastatic potential signifying canines
as genuine models for studying Human breast cancer
[8]. Recent data suggest that canine mammary
tumours possessed the ability to generate tumorsph-
eres having stem cell property [9]. Insights into the
CSC marker expression in canine mammary tumours
would help greatly by enhancement of current
perspectives in breast cancer research and to use
canines as suitable animal model for studying breast
cancer. Thus, the primary objective of this study was
to evaluate the prognostic value through co localized
expression of cancer stem cell phenotype CD44
+
/
CD24
-/low
/ ESA
+
in mammary tumours of dogs.
MATERIALS AND METHODS
The current study was conducted at Department of
Veterinary Pathology, Madras Veterinary College,
Chennai, for a period of 14 months from 2008-2009.
Tissues: 40 surgically excised tissue samples were
collected from female dogs aged between 2-17 yrs
(mean age: 9.45 yrs) of various pure and mixed
breeds. Three representative tissue samples (different
locations of tumor) from each case were collected
and fixed in neutral buffered formalin for 48 hrs and
processed to get sections of 3-4m thickness. The
tumours were classified histologically and graded
according to the standered criteria [10,11] respect-
ively. 24 selected cases were subjected to triple
immunohistochemistry based on the area free of
necrosis, hemorrhage, inflammatory cells and certain
areas showing malignant cells with mitotic figures
and invasiveness were considered.
Triple immunohistochemistry of CD 44, CD 24
and ESA using polymer detection system: Tissue
sections were subjected to triple immunostaining using
the following antibodies: mouse monoclonal CD44
(Clone: 156-3C11), CD24 (clone: SN3b) and epithelial
specific antigen (ESA) (clone: VU-1D9), M/s. Lab
Vision, USA. The secondary antibodies included
goat anti-mouse HRP polymer and goat anti-mouse
alkaline phosphatase polymer (Lab Vision, USA).
Chromogens used were DAB, Fast Red (Labvision,
USA). Additional chromogens such as vector blue
and vector vip (Vector laboratories, USA) were also
used. Various chromogen combination was used in
tandem to arrive at the best combination of
chromogens, where the least background noise was
achieved. Denaturing solution was obtained from M/
s Biocare, USA, was used to remove the primary
antibody complex before proceeding into second and
third immunostaining procedure in order to prevent
cross reaction, leaving the colored reaction product
intact. Antigen retrieval was standardized, where all
three antigens were retrieved optimally using 10mM
of Tris-EDTA, ph 9.0 using a microwave oven
(Kenstar # 9925) at 1. Hi-power (> 1000 W) for 5
min - 1 change, 2. 800W for 5min - 1 change 3. 500W
for 25 min - 1 change
Following boiling, the slide rack were brought to room
temperature for 20 minutes and washed in phosphate
buffered saline (PBS). Endogenous peroxidase
activity was blocked by treating with hydrogen
peroxide 3 per cent in methanol for 10 min. Non-
specific staining was eliminated by incubating the
sections ultra V block. The sections were incubated
with the ready to use (RTU) mouse monoclonal CD
44 primary antibody for 60 min at room temperature,
in a humid chamber.
They were then incubated with HRP polymer based
secondary antibody for 30 min at room temperature
(RT). The color was developed at room temperature
with a freshly prepared solution of DAB (3,3' diamine
benzidine) for the first staining sequence [12]. This
was followed by applying denaturing solution (Biocare
# DNS001) for 3 min at RT. Sections were washed
thrice thoroughly with phosphate-buffered saline.
Second staining sequence included Incubation with
mouse monoclonal (1:100) CD24 for 6 hr at RT.
Incubated with HRP polymer based secondary
antibody for 30 min at RT followed with chromogen
Vector VIP in 50 mM Tris HCl pH 7.6 for 10 min.
Followed by denaturing solution (Biocare) for 3
minutes at RT was applied to remove the earlier
applied immune complexes. Slides were washed with
TBS-T before using AP polymer secondary antibo-
dies. The third immunostaining sequence involved
incubation with mouse ESA primary antibody (RTU)
overnight at room temperature at RT in an incubator
followed by alkaline phosphatase polymer based
secondary antibody was incubated for 30 min at RT.
Vector blue in 50 mM Tris HCl pH 8.2 for 15 min
was added as the third chromogen. Sections were
counterstained with 1:10 dilution of haematoxylin. The
slides were dried at 37C for 1h in an incubator and
mounted with Vectamount for permanent mounting.
The criterion for CD24 negative / low staining was
based on absence or partial staining of CD24 molecule
without interference with positive signals for CD44
4195
& ESA. All immunohisochemistry process was
accompanied with human br east ductal
adenocarcinoma as positive controls. Negative
controls included slides without primary antibodies.
RESULTS
Immunohistochemistry: The results of immuno-
histochemistry are presented in Table 1. All 24
samples were subjected to triple immunohistoche-
mistry.
CD44: Out of 24 cases 15 cases were positive for
CD44. The staining pattern was mostly cytoplasmic
and membranous involving the neoplastic epithelial
cells (Fig 1). Myoepithelial cells showed variable,
moderate immunoreactivity in most cases.
CD24: Out of 24 cases subjected to triple immunohist-
ochemistry, 18 cases were positive for CD24. The
staining pattern observed in neoplastic epithelial cells
were apically accentuated membranous staining (Fig.
2) and a few cases showed cytoplasmic staining.
ESA: Out of 24 cases, ESA was positive in 11 cases
and the staining pattern was membranous in the
basolateral surface (Fig. 3) and rest of the cases
showed mainly cytoplasmic staining.
CD44/CD24

/ESA - Stem cell lineage marker: Of
the 24 cases CD24 was positive along with CD44 as
colocalized targets in six cases with the phenotype i.e.,
CD44
+
/CD24
+
/ ESA
-
. ESA expression was also seen
along with CD24 in five cases showing co-localized
pattern of reaction with the phenotype CD44
-
/CD24
+
/
ESA
+
(Fig. 4).
Out of 24 cases subjected to triple immunohistochemi-
stry two cases of carcinosarcoma were positive for
cancer stem cell lineage markers i.e., CD44
+
/CD24
-
/Low
/ESA
+
showing colocalized expression of CD44
(and ESA with membranous staining (Fig.5).
Desquamated intact cell showed strong positive
reaction to CD44 and ESA (Figs. 6-8).
DISCUSSION
CD44: Out of 24 cases, 15 (62.50%) cases were
positive for CD44 expression. Of these 15 cases,
two had thoracic metastasis and out of these two
cases, one had lymph node metastases. Other cases
which had thoracic or lymph node metastasis did not
express CD44. The role of CD44 in human breast
cancer cannot be confidently used as a reliable
prognostic indicator [13] due to the variation in its
ligand, hyaluronan. Hence, when evaluating the role
of CD44, epithelial stromal interactions as well as
ligand and isotype expression profiles should be
considered [14].
All the 15 cases showed intracytoplasmic staining
with one case showing granular cytoplasmic staining
[15] which might be an indication of hyaluronate
uptake required for invasion through extracellular
matrix. Moreover, the problem of reduced expression
was due to the existence of numerous CD44
isoforms, which might have caused the discrepancies
[16] in solely stating that CD44 expression as an
independent prognostic indicator [17] and others not
[18] in human breast cancer. However, this
association was not straight forward as demonstrated
in the current study due to variable expression of
CD44 in canine mammary tumours. So the potential
role of CD44 expression, its variants and its receptor
warrants further investigation.
Out of 15 cases positive for CD44, 13 were in grade
I and two cases were in grade II. These results were
in accordance with findings, that greater CD44
expression was exhibited by well differentiated
(Grade I) tumours compared t o moderately
differentiated (Grade II) tumours [19].
CD24: Out of 24 cases subjected to tr iple
immunohistochemistry, 18 (75%) cases were positive
for CD24. The pattern of expression was apically
accentuated membranous staining and also
cytoplasmic staining was seen. CD24 was abundantly
expressed in carcinomas with membranous staining
intensity remarkably higher in malignant tumours [20].
Out of these, three cases had lung and lymph node
metastases, two cases had lung metastasis and one
case showed lymph node metastasis indicating an
association between CD24 expression and
metastases. CD24 could serve as a marker for human
breast cancer and its expression played an important
role in the metastasis of tumor cells through
interaction with platelet or vascular endothelial cells
[21] and shortened the patient overall survival and
disease free survival with breast cancer [22].
However, six of the cases were negative for CD24
expression, amongst which four showed distant
metastases which could be due to down regulation
of CD 24 in invasive breast cell lines and might be
Anjan Kumar et al.
4196
J. Cell Tissue Research
Fig. 1 Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7 Fig. 8
4197
Table 1: Histopathology, metastases and immunoreactivity of CD44, CD24, ESA, stem cells in mammary tumours of dogs
Carcinoma-PC - Carcioma-Papillary cystic, Sq.c.c - Squamous cell carcinoma
associated with a more aggressive cancer phenotype
[23]. These observations was also supported by other
researchers who stated that only a minority of breast
cancer cells had the ability to form new tumours which
were almost negative for CD24 [7].
Fourteen cases were in grade I and four cases were
in grade II. These results were in agreement with
the research findings mentioning that there was no
significant relationship between histopathological
parameters and overall CD24 or cytoplasmic and
membranous staining [24].
The variation in CD24 immunolocalization from
membranous to cytoplasmic staining differed in this
study, possibly due to its highly heterogeneous expr-
ession within a single case [25]. However, further
investigation is required to elucidate the role of CD24.
ESA: The current research on ESA expression is
first of the kind so far reported in veterinary oncology
literature especially in canine mammary tumours. Out
of 24 cases 11 (45.83%) were positive for ESA
expression which showed similarities with 41.7 to 48
per cent expression in human breast cancer [26].
The expression of ESA was membranous in the
basolateral surface and cytoplasmic in all cases.
However, majority of the cases in this study showed
cytoplasmic expression possibly due to loss of
membranous Ep-CAM expression, which is
associated with a higher extent of tumor budding
characteristics [27,28].
Case No.
TNM
Clinical
staging
Elston
& Ellis
Grading
Histopathology
Metastases
CD44
Subset
cells

CD24
Subset
cells
ESA
Subset
cells
CD44
+
/CD24
-/Low
/ ESA
+
STEM CELL
Lymph
node
Dist ant
113034 IV I Carcinoma-Tubular + + - - - NEGATIVE
113957 III II Carcinoma-Tubular - - - + - NEGATIVE
121720 III I Carcinoma-Tubular - - + + + NEGATIVE
126450 II II Carcinoma-Tubular - - + + - NEGATIVE
122417 III I Carcinoma-Tubular + - + - - NEGATIVE
119972 III II Carcinoma-Tubular - - + + - NEGATIVE
109672 I I Carcinoma-Tubular - - + + - NEGATIVE
117637 III I Carcinoma-Tubular - - - + - NEGATIVE
249 NA I Carcinoma-Tubular NA NA + + - NEGATIVE
115890 IV II Carcinoma-Tubular - + - + + NEGATIVE
120393* III I Carcinoma-PC - - + + - NEGATIVE
118238 III I Carcinoma-PC - - + + - NEGATIVE
114900* II I Carcinoma-PC - - + + - NEGATIVE
126136 IV II Carcinoma-PC + + - + - NEGATIVE
111531 III I Carcinoma-PC + - - + + NEGATIVE
110802 III I Sq.c.c - - + - - NEGATIVE
117818 III I Carcinosarcoma - - + + + NEGATIVE
121923 III I Carcinosarcoma - - + + + NEGATIVE
121498* III I Carcinosarcoma - - + + + NEGATIVE
101156 IV I Carcinosarcoma + + + - + POSITIVE
112921 IV I Carcinosarcoma - + + - + POSITIVE
124264 III II Carcinosarcoma - - - - + NEGATIVE
106388 IV I Carcinosarcoma + + - + + NEGATIVE
248 NA I Carcinosarcoma NA NA - + + NEGATIVE
Anjan Kumar et al.
Fig. 1: Tubular Carcinoma CD44
+
(Brown). Fig 2: Tubular Carcinoma CD24
+
(Fast Red). Fig. 3: Papillary cystic Adenocarcinoma-
Basolateral Staining ESA
+
(Vector Blue) Blue) ESA
+
(DAB). Fig. 4: Colocalized targets CD44
-
(Vector red)CD 24
+
(Vector
Brown). Fig. 5: Carcinosarcoma Colocalized targets CD44
+
(DAB) CD 24
VE
(Fast Red) ESA
+
(Vector Blue). (Vector Blue). Fig.
6: Carcinosarcoma Colocalized targets CD44
+
(Brown) CD24
ve
(Fast Red)

ESA
+
. Fig 7: Carcinosarcoma CD44
+
(Brown) CD24
ve
(Fast Red)

ESA
+
(Vector Blue). Fig 8: Carcinosarcoma ) CD44
+
(Brown) CD24
ve
(Fast Red)

ESA
+
Vector Blue)
Explanation of Figures:
4198
J. Cell Tissue Research
ESA expression was also seen in two of the cases
showing thoracic and lymph node metastases; among
which one case showed thoracic metastasis and one
case with lymph node metastases indicating a
correlation between ESA expression and clinico-
pathological parameters like size and lymphnode
metastasis [26]. There was no correlation between
ESA expression and the tumor grade. Out of the 11
cases, ESA expression was seen in 6 cases which
were in TNM stage III and four in stage IV, indicating
a poor prognosis. However final proof for a direct
cancer promoting role of ESA expression in canine
mammary tumor needs to be investigated.
Cancer stem cell lineage marker / CD44
+
/CD24
-/
Low
/ESA
+
: Out of 24 cases subjected to triple
immunohistochemistry two cases of carcinosarcomas
were positive for stem cell lineage markers i.e., CD44
+
/
CD24
-/Low
/ESA
+
showing co localized expression of
CD44 (membranous / cytoplasmic) and ESA with
membranous staining. Stem cell lineage markers with
the phenotype CD44
+
/CD24
-/low
expression were also
reported in canines with histological feature of
Carcinosarcoma, Cystic papillary Adenocarcinoma and
Tubular Adenocarcinoma respectively [8]. This
expression in Carcinosarcoma was probably attributed
to the induction of an Epithelial Mesenchymal Transition
(EMT) as observed in transformed human mammary
epithelial cells yielding cells with a CD44high/CD24low
antigenic phenotypeprecisely the antigenic phenotype
that has been ascribed to neoplastic mammary stem
cells [29].
However, ESA
+
subset of CD44
+
/CD24
-
cells were
known to be enriched for tumor repopulating capacity,
as many as 200 of these cancer cells with the lineage
of CD44
+
/CD24
-/Low
/ESA
+
consistently formed
tumours when inoculated into SCID/NOD mice and
was concluded that CD44
+
/CD24
-/Low
/ESA
+
lineage
population had more than 50 fold ability to form tumors
[7].
Two positive cases for stem cell markers had thoracic
metastasis (Post surgical survival period 160 days)
and one case had metastasis to liver and all regional
lymph nodes (Post surgical survival period 84 days).
Both the cases died and were in TNM clinical stage
IV with a poor survivability and average survival
period of 122 days.
However the cases in which cells which were
positive for stem cells lineage markers i.e. CD44
+
/
CD24
-/low
/ ESA
+
as evident from this study, favored distant
metastasis which is in agreement with other findings [30].
Cancer cells with CD44
+
/CD24
-/low
/ ESA
+
phenotype should
be subjected to further charac-terization at single cell level,
using in vivo, in vitro and ex vivo models to understand
their pathway in canine mammary tumors that account for
their tumourigenic potential and as possible tumor models
for elucidating targets for therapeutic intervention.
REFERENCES
[1] Nerukar, V.R, Chitale, A., Jalnapurkar, B., Naik, S.N.
and Laliha, U.: J. Comp. Pathol., 101: 389-397
(1989)
[2] Pierrepoint, C.G.: J. Small Anim. Pract., 26: 43-47 (1985)
[3] Schneider, R.: Cancer, 26: 419-426 (1970)
[4] Strandberg, J.D. and Goodman, D.G.: Am. J. Pathol.,
75: 225-228 (1974)
[5] Priester,W.A.: J. Small Anim. Pract., 20: 1-11 (1979)
[6] Uva,P., Aurisicchio, L., Watters, J., Loboda, A.,
Kulkarni, A., Castle, J., Palombo, F., Viti, V., Mesiti,
G., Zappulli, V., Marconato, L., Abramo, F., Ciliberto,
G., Lahm, A., Monica N.L and Rinaldis, E.: BMC
Genomics, 10: 135 (2009).
[7] Al-Hajj, M., Wicha, M.S., Benito-Hernandez, A.,
Morrison, S.J and Clarke, M.F.: Proc. Natl. Acad. Sci.
USA, 100: 3893-3988 (2003).
[8] Anil, A.: Identification of stem cell lineage CD44
+
/
CD24
-
and apoptotic marker in mammary tumours
in dogs. M.V.Sc Thesis, Tamilnadu Veterinary and
Animal Sciences University, Chennai 600 051 (2007)
[9] Cocola, C., Anastasi, P., Astigiano, S., Piscitelli, E.,
Pelucchi, P., Vilardo, L., Bertoli, G., Beccaglia, M.,
Veronesi, M.C., Sanzone, S., Barbieri, O., Reinbold,
R.A., Luvoni, G.C and Zucchi, I.: Reprod. Domest
Anim, 44 (Suppl. 2): 214-217 (2009).
[10] Misdorp, W. : Tumors of the mammary gland.
In: Tumors in Domestic Animals ( Meuten D.J. ed.):
4
th
Edn. Iowa State Press. pp. 575-606 (2002).
[11] Rezaie, A., Tavasoli, A., Bahonar, A. and Mehrazma,
M.: J. Biol. Sci., 9: 333-338 (2009).
[12] Sternberger, L.A. and Joseph, S.A.: J. Histochem.
Cytochem., 27: 1424-1429 (1979).
[13] Herera-Gayol, A. and Jothy, S.: Exp. Mol. Pathol., 66:
149-156 (1999).
[14] Lopez, J.I., Camenisch, T.D., Stevens, M.V., Sands,
B.J., McDonald, J and Schroeder, J.A.: Cancer Res.,
65: 6755-6763 (2005).
[15] Sneath, R.J.S and Mangham. D.C.: Mol. Pathol., 51:
191-200 (1998).
[16] Dall, P., Heider, K.H., Sinn, H.P., Skroch-Angel, P.,
Adolf, G., Kaufmann, M., Herrlich, P. and Ponta, H.:
Int. J. Cancer, 60: 471-477 (1995).
[17] Kaufmann, M., Heider, K.H. and Sinn, H.P.: Lancet,
345: 615-619 (1995).
[18] Sinn,H.P., Heider, K.H and Skroch, P. and Angel.:
Breast cancer Res Treat, 36: 307-313 (1995).
[19] Madrazo, J., Garcia-Fernandez, R.A., Garcia-Iglesias,
M.J., Duran, A.J., Espinosa, J., Perez, C, and
4199
Martinez, C.: Vet. J., 180: 371-376 (2009).
[20] Lim, S.C., and Oh, S.H.: Pathol. Res. Pract., 201: 479-
486 (2005).
[21] Fogel, M., Friederichs, J., Zeller, Y., Husar, M., Smirnov,
A., Roitman, L., Altervogt, P. and Sthoeger, Z.M.:
Cancer Lett, 143: 87-94 (1999).
[22] Kristiansen, G., Winzer, K.J., Mayordomo, E., Bellach,
J., Schluns, K., Denkert, C., Dahl, E., Pilarsky, C.,
Altevogt, P., Guski, H and Dietel, M.: Clin Cancer Res,
9: 4906-4913 (2003).
[23] Schindelmann, S., Windisch, J., Grundmann, R.,
Kreienberg, R., Zeillinger, R. and Deissler, H.: Tumor
Biol, 23: 139-145 (2002)
[24] Bircan, S., Kapucuoglu, N., Baspinar, S., Inan, G. and
Candir, O.: Pathol. Res. Pract., 202: 569-576 (2006)
[25] Kristiansen, G., Pilarsky, C., Pervan, J., Sturzebecher,
B., Stephan, C., Jung, K., Loening, S., Rosenthal, A.
and Dietel, M.: Prostate, 58:183-192 (2004).
[26] Spizzo, G., Went, P., Dirnhofer, S., Obrist, P., Simon, R.,
Spichtin, H., Maurer, R., Metzger, U., Castelberg, B.V.,
Bart, R., Chinskaya, S.S., Kochli, O.R., Haas, P., Mross,
F., Zuber, M., Dietrich, H., Bischoff, S., Mirlacher,
M., Sauter, G. and Gastl, G.: Breast cancer Res. Treat.,
86: 207-213 (2004).
Anjan Kumar et al.
[27] Litvinov, S.V., Balzar, M., Winter, M.J., Bakker, H.A.M.,
Bruijin, I.H.B., Prins F., Fleuren G.J. and Warnaar, S.O..:
J. Cell Biol., 139: 1337-1348 (1997).
[28] Gosens, M.J., Kempen, L.C.V., Van de Velde, C.J., Van
Krieken, J.H.J.M. and Nagtegaal, I.D.: Mod. Pathol,
20: 221-232 (2007).
[29] Mani, A.S., Wenjun, G., Mai, J.L., Eaton, E.Ng.,
Ayyanan, A., Zhou, A.Y., Brooks, M., Reinhard,
F., Zhang,C.C., Shipitsin, M., Campbell, L.L., Polyak,
K., Brisken, C., Yang, Y. and Weinberg, R.A.: Cell,
133(4): 704-715 (2008).
[30] Abraham, B.K., Fritz, P., McCellan, M., Hauptvogel,
P., Athelogou, M. and Brauch, H.: Clin Cancer Res.,
11: 1154-1159 (2005).