51
Vol. 5, No. 1
Endoglin (cd105) and S100A13
as markers of active angiogenesis
in endometriosis
Soren Hayrabedyan, Stanimir Kyurkchiev, Ivan Kehayov1
Department of Molecular Immunology,
Institute of Biology and Immunology of Reproduction,
Bulgarian Academy of Sciences, Sofia, Bulgaria
Received: 10 July 2004; accepted: 5 February 2005
SUMMARY
The aim of the present study was to evaluate the expression of the neo-angiogenic marker endoglin and its localization in tissues of normal and endometriotic patients as well as to compare it with one new angiogenic marker
candidate – S100A13. Human recombinant S100A13 and endoglin 35mer
synthetic peptide of the intracellular domain were used for the production
of rabbit polyclonal antisera. The antisera were characterized for specificity,
using immunoenzyme assay (ELISA), Western blot and immunohistochemistry. Formalin-fixed, paraffin-embedded tissue sections from normal endometrium, adenomyosis, ovarian endometriosis, eutopic endometrium from
different endometriotic specimens were tested by immunohistochemistry. No
endoglin specific staining was observed on the microvessels of the normal
endometrium. In adenomyosis and ovarian endometriosis, the expression
pattern was different – endoglin was expressed in all microvessels, with an
Corresponding author: Ivan Kehayov, Department of Molecular Immunology, Institute of Biology
and Immunology of Reproduction, Bulgarian Academy of Sciences, Sofia 1113, 73, Tzarigradsko
shosse blvd., Bulgaria; [email protected]
1
Copyright © 2005 by the Society for Biology of Reproduction
52
Angiogenic markers in endometriosis
even stronger expression in the myometrial compartment. Weak endoglinpositive staining was detected in the microvessels of eutopic endometrium
specimens from different endometriosis cases. In comparison to endoglin,
S100A13 exhibited a moderate expression in endometrial glands of normal endometrium, but strong expression in endometriotic specimens. No
S100A13 extensive staining of the microvessels was observed in normal
endometrium, while in endometriosis, it exhibited very intense staining in
microvascular endothelia and less intense in the perivascular area of middle
to large-sized vessels. This study for the first time shows over-expression
of S100A13 in endometriosis. These data show that the expression of endoglin and S100A13 corresponds to the activation of the endothelial cells
in the process of endometriotic angiogenesis, suggesting a beneficial role
for these two molecules as markers for actively progressing endometriotic
process. Reproductive Biology 2005 5(1): 51-67.
Key words: endoglin, CD105, S100A13, endothelial marker, angiogenic
status, microvascular density marker
INTRODUCTION
Endometriosis is characterized by development of endometriotic stroma and
glands at sites different from the uterine endometrium [27]. Angiogenesis is
considered to be a pivotal process in the pathogenesis of endometriosis and
is required for establishment, development and maintenance of endometriotic lesions as well as for their further propagation. Although endometriosis
seems to be a unified pathology, it has many different forms. Adenomyosis
and ovarian endometriosis are two of these. It has been currently considered
by some authors [7], that these two subgroups have distinct pathogeneses
and possibly different angiogenesis mechanisms.
Endoglin (Eng, CD105) is a transmembrane homodimer glycoprotein
(190 kDa, [19]) known as a modulator of Transforming Growth Factor Receptor I (TGF-β RI, [21]). Studies on gene knock-out mice [20] revealed that
Eng is a vital factor for the angiogenesis early in development. Endoglin is
expressed in human endothelial cells [8], macrophages [17], stromal cells
Hayrabedyan et al.
53
[25, 29], and vascular smooth muscle cells (VSMCs; [1, 5, 22]). Over-expression of Eng is characteristic in activated endothelial cells (ECs; [30]),
but not quiescent ECs [15, 31]. Since it is used for microvascular density
evaluation and ECs counting in tumours [2, 3], Eng is of primary interest
as an angiogenic marker in endometriosis as well.
S100A13 is a newly discovered members of the S100 protein family,
which is characterized by specificity to different forms of cancer [9, 11, 18].
S100A13 was reported to be co-expressed with the fibroblast growth factor
1 (FGF-1) in brain tumours [16] demonstrating a perivascular distribution.
Translocation of S100A13 was observed in endothelial cells in response to an
increase in intracellular calcium levels or to angiotensin II, and the process
was dependent on the classic Golgi-endoplasmic reticulum pathway [10].
This implication of S100A13 as involved in multiple signalling pathways in
the EC suggests for its future use as a possible marker of EC activation.
The aims of this study were to: 1. evaluate the expression of the neoangiogenic marker Eng and its localization in tissues of normal and endometriotic patients; 2. compare the expression pattern of Eng with expression
pattern of S100A13 – a possible candidate for angiogenic marker.
MATERIALS AND METHODS
A synthetic 35-mer peptide (corresponding to a portion of the intracellular
domain of the Eng molecule, and referred to as Eng-peptide) and human
recombinant S100A13 protein (hrS100A13) used for the production of
anti-endoglin and anti-S100A13 polyclonal sera respectively, were kindly
donated by the Maine Medical Centre Research Institute (USA).
Formalin-fixed paraffin embedded (FFPE) tissue sections from cases of
adenomyosis (n=20), ovarian endometriosis (n=20), and eutopic endometrium of women with endometriosis (n=5) and from normal endometrium
(n=5) were provided by the Pathology Department of the University Hospital, University of Pleven, Bulgaria. All endometriosis cases were classified according to the criteria of ASF (American Society for Fertility) and
histologically graded after Noyes classification [23] by two pathologists.
54
Angiogenic markers in endometriosis
Antisera production and characteristics
Chinchilla rabbits (males, 2.5 kg) were injected subcutaneously either with
Eng-peptide (150 µg) or hrS100A13 (100 µg) every two weeks for three
months. The polyclonal sera were subsequently collected and tested for
specificity by ELISA and Western blot.
A. Antisera specificity assessed by immunoenzyme method (ELISA)
Indirect ELISA method was used to test the produced polyclonal antisera
[9]. Briefly, the Eng-peptide (1 µg/ml) or hrS100A13 protein (1 µg/ml) were
coated on 96 well microtiter plates (Costar, USA) in carbonate buffer (pH
9.4) during overnight incubation (4oC). Bovine serum albumin (1% BSA)
was used to block the plates, and then serially diluted antisera – either antiEng-peptide (starting at 1/800) or anti-S100A13 (starting at 1/200) were
distributed in each coated well. The assay was performed in duplicates.
An anti-rabbit IgG antiserum labelled with peroxidase (Sigma Co, USA),
diluted 1/1000 in blocking buffer, was added to each well and kept for 1 hour at
room temperature. The enzyme reaction was developed using ortho-phenylenediamine and the coloured product intensity was read using an ELISA reader (LKB,
Sweden) at a wavelength of 492 nm. As an inhibition ELISA test for specificity,
the same dilutions of anti-Eng-peptide or anti-hrS100A13 antibodies pre-incubated
with endoglin peptide (1 µg/ml) and hrS100A13 (1 µg/ml) protein, respectively,
were applied in parallel wells on the same plate and incubated at RT for 2 hours.
Additionally, the sera were tested against a panel of various human proteins (tetanus toxoid, normal human serum, human serum albumin, IL-1α, FGF-1, S100A1,
S100A6, S100B, and calmodulin) used as antigens, by indirect ELISA.
B. Antisera specificity assessed by Western blot analysis
Electrophoresis in a 15% (w/v) polyacrylamide gel, in the presence of sodium
dodecyl sulphate (SDS) under reducing conditions, was performed following
Hayrabedyan et al.
55
the well-established protocol of Laemmli (1970). Samples of endoglin peptide
and hrS100A13 were added to serial dilutions (with an initial concentration
of 5 μg/ml) and transferred to a “HyBond” nitrocellulose membrane (Sigma
Co, USA) for 1 hour at 0.8V/cm2 using a Miniblotter apparatus (PharmaciaAmersham, Sweden). The membrane was then blocked with 3% (w/v) nonfat dry milk and incubated in anti-Eng antibodies (1/200) or anti-hrS100A13
(1/400), respectively for 2 hours at room temperature. Each membrane was
further treated with anti-rabbit IgG serum conjugated to alkaline phosphatase
(Sigma Co, USA) diluted 1/1000 in blocking buffer. The reaction was developed using 100 μl NBT, 100 μl BCIP in 5 ml alkaline phosphatase buffer (0.1
M Tris, 0.1 M NaCl, 0.005 M MgCl2, pH 5.0).
Immunohistochemistry
Immunohistochemistry was used to determine the tissue localization of the Eng
and S100A13 and to evaluate the intensity of their expression. Sections were
deparaffinized in xylene, sequentially rehydrated in gradients of ethanol, and
treated with 1.2% (v/v) hydrogen peroxide in methanol solution for 30 min at
room temperature to quench the endogenous peroxidase. In the case of Engpeptide, antigen retrieval was performed using 1mM citric acid buffer, pH 6.2
[28]. No antigen retrieval was necessary for S100A13. All sections were then
blocked with 3% BSA and 0.1% Tween 20 for 1 hour at room temperature.
Anti-Eng-peptide and anti-S100A13 sera diluted in blocking buffer 1:800 and
1:100, respectively, were incubated on sections overnight at RT. The reaction was revealed using the streptavidin-biotin-peroxidase technique. Briefly,
sections were incubated (1 h, RT) with biotin conjugated donkey anti-rabbit
IgG (SAPU, diluted 1/500). This was followed by incubation (1 hour, room
temperature) with streptavidin-conjugated horseradish-peroxidase (SAPU,
diluted 1/500). Then, the sections were incubated with 3, 3-diaminobenzidine
(0.05 diaminobenzidine in 0.05 M Tris buffer, pH 7.6 and 0.01% hydrogen
peroxide) and counterstained with Mayer’s hematoxylin.
All bright-field observations were carried out on “Olympus BX-40” (Japan) microscope system, equipped with “Sony CCD” (Japan) camera, and
56
Angiogenic markers in endometriosis
Fig. 1. ELISA and immunoblot characterization of anti-Eng-peptide and
anti-S100A13 polyclonal sera. A. ELISA titration curve of anti-Eng-peptide
serum. Dotted line represents the dose-dependent inhibition curve, which resulted after preincubation with synthetic Endoglin peptide. B. Immunoblot of
synthetic Endoglin-peptide, detected as a single band with a m.m. of 7 kDa,
demonstrating concentration-dependent intensity. C. ELISA titration curve of
anti-S100A13 serum with dose-dependent characteristic. Dotted line represents
the dose-dependent inhibition curve, which resulted after preincubation with
S100A13 protein. D. Immunoblot of S100A13 protein, detected as a single
band with a m.m. of 14 kDa.
Hayrabedyan et al.
57
digital image capture board mounted in Windows PC. “Adobe Photoshop
ver.7” (Canada) software was used for image acquisition and processing.
Serial sections from each sample with no first antibody applied, were used
as negative controls.
RESULTS
A. Antisera specificity assessment
Rabbit anti-Eng-peptide and anti-S100A13 polyclonal sera were tested by
ELISA and Western blotting. Several tests were applied in order to confirm
the specificity of the reaction. First, an indirect ELISA assay was performed
and dose-dependent curve was made (fig. 1 A, C) Secondly, the antibodies were pre-incubated with free protein/peptide that caused significant
inhibition of their reactivity against solid phase coated antigen. Thirdly,
an additional ELISA test was done to check for cross-reactivity against the
panel of human proteins but no reaction was observed. These results strongly
demonstrated the specificity of the studied sera against the antigens.
The anti-sera specificity was also confirmed by Western blotting as described above. Using anti-Eng serum, Eng-peptide was detected as a single
band with a m.m. of 7 kDa (fig. 1B). When polyclonal rabbit anti-S100A13
serum was applied, S100A13 protein migrated as a single band with m.m.
of 14 kDa (fig. 1D). The band staining intensity had a dose-dependent character and corresponded to molecular mass of studied antigens. The position
of Eng bands corresponded to a m.m. of 7 kDa, rather than m.m. of 190
kDa as it should be for the whole endoglin homodimer molecule, since the
synthetic 35-mer Eng-peptide was used instead.
B. Endoglin and S100A13 expression in endometriosis
Using immunohistochemistry, we investigated the expression of endoglin
and S100A13 in three groups of specimens – normal secretory endometrium,
58
Angiogenic markers in endometriosis
Fig. 2. Secretory phase normal endometrium microvessels were negative for endoglin (white dotted arrows), while microvessels in the myometrium were regarded as
internal positive controls (black solid arrow). G-glands, S-stroma, M-myometrium,
C-capillaries; magnification x 100.
endometriotic lesions (adenomyosis and ovarian endometriosis) and eutopic
endometrium from women with endometriosis.
Endoglin was not expressed in the capillaries of the normal endometrium.
The borderline microvessels of the uterus exhibited only subtle Eng expression, while myometrium had stronger endoglin expression (fig. 2). Eng was
detected in the middle to large-sized vessels of the normal endometrium
specimens. It was expressed predominantly in the pericyte component,
rather than the endothelial component in the largest arterial vessels of the
uterine myometrium.
The immunohistochemical studies of specimens from adenomyosis
and ovarian endometriosis revealed a distinctive expression of Eng. Endoglin was expressed in all microvessels, with positively stained endothelial
cells. Multiple single capillaries were stained with different intensity in
adenomyosis, scattered within the endometriotic stroma. An even stronger
- compared to the endometriotic lesion - vascular staining intensity was
observed in the surrounding myometrial compartment (fig. 3). The stained
Hayrabedyan et al.
59
Fig. 3. Microvessels surrounding endometriotic glands in adenomyosis, demonstrated strong endoglin expression (black solid arrows). S-stroma, V-vessels;
magnification x 200.
microvessels in ovarian endometriosis formed capillary vascular networks
rather than separate microvessels (fig. 4). In some cases, Eng positive microvessels were seen in close vicinity to endometriotic glands and within
the glands themselves. Positive immunostaining in middle to large blood
vessels dislocated within the myometrium and ovary was not in endothelial
but rather in the VSMCs.
Endoglin expression was examined only in the endometrial compartment of women with endometriosis. The endometriomata was not under
investigation. In this group the microvascular cells within the endometrium
were mostly positive, but with lower staining intensity than those in the
endometriotic lesions (fig. 5). There was significant visual difference between the expression levels of Eng in normal endometrium and in eutopic
endometrium of women with endometriosis.
Immunohistochemistry with anti-S100A13 serum in normal endometrium demonstrated moderate staining in the epithelial cells of the
endometrial glands and insignificant stromal staining background. Only
60
Angiogenic markers in endometriosis
Fig. 4. Endoglin stained moderately the microvessels network (white solid
arrows) infiltrating the stroma of ovarian endometrioma. G-glands, C-capillaries;
magnification x 200.
Fig. 5. Positive endoglin staining of microvessels (white solid arrows) in the endometrial compartment of eutopic proliferative endometrium from woman with
endometriosis. G-glands, C-capillaries; magnification x 200.
Hayrabedyan et al.
61
Fig. 6. Immunolocalization of S100A13 in endometrial glands in normal
endometrium. Only subtle staining of the microvessels (black doted arrows) was
observed. The glandular epithelium (large hollow arrows) demonstrated cytoplasmic staining, most intense on the membranes, especially at the apical side
– towards the glandular lumen. G-glands, S-stroma, M-myometrium, V-vessels;
magnification x 100.
some of the surrounding stromal cells stained stronger for S100A13.
The positive staining was cytoplasmic with the most prominent membrane staining near the glandular lumen. Microvessels were only subtly
stained, especially in comparison to the staining intensity of the glands
(fig. 6).
S100A13 was over-expressed in some of the endometriotic specimens,
primarily in glandular epithelia. Surprisingly, it demonstrated predominantly
a perivascular expression pattern in other specimens. S100A13 was expressed in the large vessels as well as in the perivascular space. Microvessels
were with strong positive staining for S100A13, mainly in the endothelial
cells’ cytoplasm, while in the perivascular area, the staining was weaker
(fig. 7). The main difference in S100A13 expression in adenomyosis and
ovarian endometriosis was in staining intensity. The antigen was expressed
stronger in ovarian endometriosis specimens.
62
Angiogenic markers in endometriosis
Fig. 7. S100A13 was expressed in adenomyosis on midle- to large-sized arteries, in the para-vascular space of arterioles (black solid arrows) as well as in the
endothelial cells of some capillaries (black thin arrows). G-glands, S-stroma, Aarteries, a-arterioles, V-veins, C-capillaries; magnification x 200.
DISCUSSION
Endometriosis has been considered an angiogenic disease in recent years. The
establishment, propagation, and sustaining of endometriotic implants depend
largely on proper angiogenic support. The molecules investigated in this study,
endoglin and S100A13, are angiogenesis-related. They have an expression
level corresponding to the state of activated endothelia. Although the exact
mechanism of their involvement in endometriotic angiogenesis process is
currently unknown, they both represent not only marker molecules, but also
active participants in this process. S100A13 expression is coupled with the expression of FGF-1, a proven angiogenic and growth factor, as well as with the
expression of IL-1α, a cytokine also known to be pro-angiogenic. It has been
established that Eng is an active participant in angiogenesis and its expression
is highly restricted only to activated endothelial cells. [4, 6, 14, 26]
From a morphological point of view, both factors demonstrated distinct
staining patterns in normal and pathologic conditions, suggesting their pos-
Hayrabedyan et al.
63
sible pathogenic role in endometriotic angiogenesis. It should be noted that
when sections from non-endometriotic tissue were tested, no Eng staining
of microvessels was observed. This fact seems to be of utmost importance
considering Eng is assumed to have expression predominantly in endothelial
cells of microvessels of tumours and other pathologies. On the other hand,
the large vessels expressed Eng, as observed also by Zhang et al. [32]. According to them, the Eng staining in normal endometrium, throughout the
entire menstrual cycle, is confined primarily to the arterioles and arteries,
with only weak positive staining of the veins and no staining of the capillary microvessels. The arterial staining is most intensive during the early
secretory phase, followed by the early proliferative phase. We observed
similar spatial pattern of Eng expression on normal endometrium, but this
expression during the various menstruation cycle phases was not explored.
In contrast to Zhang et al, there was no positive staining for the Eng microvessels even in normal secretory endometrium in our study. Our results
demonstrated negative staining of the eutopic endometrium capillaries in
non-endometriosis patients and positive staining of eutopic endometrium
in endometriosis patients. This observation has been currently stated only
by Kim et al [12].
In this study, we evaluated the immunohistochemical expression of Eng
and S100A13 in two main endometriosis subgroups – adenomyosis and
ovarian endometriosis. Adenomyosis is characterised by the localization of
endometrial glands and stroma within the uterine myometrium. When the
same lesion types are localized within the ovarian tissue, forming frequently
so-called ovarian “chocolate” cysts, the condition is denoted as ovarian
endometriosis. The selection of these two groups was made because of
the statements [7] that both these endometriosis subgroups have distinct
pathogeneses.
The expression of Eng in the endometriotic lesions was investigated,
and strong positive staining of the microvessels was observed. The intensity of staining between the two endometriotic subgroups was visually
indistinguishable. The only difference was that single stained microvessels
were observed in adenomyosis while these microvessels formed capillary
networks in ovarian endometriosis. The specific anti-Eng-peptide serum
64
Angiogenic markers in endometriosis
stained the cytoplasm of endothelial cells, corresponding to the intracellular
localization of the selected peptide, recognized by the antibodies. The Eng
expression in endometriotic endothelial cells was stronger, compared to
normal endometrium where its expression was weaker which substantiate
data published earlier [12]. We observed Eng expression in vascular smooth
muscle cells that has been published by others [5, 24], with the assumption
that myometrial vessels are useful as an internal positive control for Eng
immunostaining [12].
It has been demonstrated that the expression of Eng is upregulated
in endometriosis and it may be a better marker for evaluation of microvessel density in comparison to CD34 and von Wielebrand factor
(vWF; [3, 32]). The demonstration of positive Eng expression within the
endometriomata, along with the data of its participation in the TGF-β
pathway, suggests that Eng has a definite role in angiogenesis in endometriosis. Angiogenesis is a pre-requisite for development of endometriotic
lesions which implies that Eng is a better marker for both microvascular
density evaluation and comparison between distinct endometriotic loci
[3, 32].
The immunostaining of S100A13 demonstrated broader distribution
pattern. It was expressed in endometrial glands, single stromal cells, the
VSMCs and in the perivascular space. S100A13 has the ability to stain both
large vessels and microvessels, giving a total vascular marking. This makes
S100A13 preferable to vWF, which is reported to stain the ECs within the
large vessels, while staining is weak or even absent within the capillaries
[32]. Furthermore, it was differentially expressed in the microvessels located
in the normal eutopic endometrium compared to endometriotic stromal
compartment, exhibiting stronger staining pattern in the endometriotic
microvessels. Another widely known and accepted endothelial cell marker
– CD34, stains non-endothelial supportive cells around the endometrial
glands in endometrium, forming their basal membrane [13, 32], suggesting
that no vascular-restricted marker exists. S100A13, despite its staining heterogeneity, proves to be useful tool in detecting the activated microvessels.
This suggests its use as a universal angiogenic marker, at least as a marker
of activated endometriosis.
Hayrabedyan et al.
65
In this paper we confirm, what had only been stated by Kim et al. [12],
the positive expression of endoglin not only in endometriotic lesions, but
also in eutopic endometrium of women with endometriosis. In our study,
S100A13 overexpression was demonstrated for the first time in endometriotic tissues, compared to normal endometrium.
In conclusions, both endoglin and S100A13 stained the activated microvessel endothelia in endometriosis. S100A13 displayed positive expression in normal endometrial microvessels while Eng did not. Only S100A13,
but not Eng, was expressed in epithelia, with the highest staining intensity
in endometriosis. Further studies to test the possibility to use endoglin and
S100A13 as markers for monitoring the clinical status and the effect of
treatment of patients with endometriosis are currently in progress.
ACKNOWLEDGEMENTS
This work was supported by Bulgarian National Science Fund (Grant
No. K-1201/2000).
REFERENCES
1. Adam PJ, Clesham GJ, Weissberg PL 1998 Expression of endoglin mRNA and protein
in human vascular smooth muscle cells. Biochemical and Biophysical Research Communications 247 33-7.
2. Bredow S, Lewin M, Hofmann B, Marecos E, Weissleder R 2000 Imaging of tumour
neovasculature by targeting the TGF-beta binding receptor endoglin. European Journal
of Cancer 36 675-81.
3. Brewer CA, Setterdahl JJ, Li MJ, Johnston JM, Mann JL, McAsey ME 2000 Endoglin expression as a measure of microvessel density in cervical cancer. Obstetrics and
Gynecology 96 224-8.
4. Charpin-Taranger C, Dales JP, Garcia S, Andrac-Meyer L, Ramuz O, CarpentierMeunier S, Bonnier P 2003 The immunohistochemical expression of CD105 is a
marker for high metastatic risk and worse prognosis in breast cancers. Bulletin De L’
Academie Nationale De Medecin (Paris) 187 1129-45; discussion 1145-6.
5. Conley BA, Smith JD, Guerrero-Esteo M, Bernabeu C, Vary CP 2000 Endoglin, a
TGF-beta receptor-associated protein, is expressed by smooth muscle cells in human
atherosclerotic plaques. Atherosclerosis 153 323-35.
66
Angiogenic markers in endometriosis
6. Dales JP, Garcia S, Andrac L, Carpentier S, Ramuz O, Lavaut MN, Allasia C, Bonnier P, Charpin C 2004 Prognostic significance of angiogenesis evaluated by CD105
expression compared to CD31 in 905 breast carcinomas: correlation with long-term
patient outcome. International Journal of Oncology 24 1197-204.
7. Donnez J, Smoes P, Gillerot S, Casans-Roux F, Nissole M 1998 Vascular endothelial
growth factor (VEGF) in endometriosis. Human Reproduction 13 1686-1690.
8. Gougos A, Letarte M 1988 Identification of a human endothelial cell antigen with
monoclonal antibody 44G4 produced against a pre-B leukemic cell line. Journal of
Immunology 141 1925-33.
9. Harlow E, Lane D 1988 Antibodies: A Laboratory Manual. In Antibodies: A Laboratory Manual, pp 553-612. Cold Spring Harbor Laboratory Press, New York
10. Hsieh HL, Schafer BW, Cox JA, Heizmann CW 2002 S100A13 and S100A6 exhibit
distinct translocation pathways in endothelial cells. Journal of Cell Science 115 314958.
11. Hsieh HL, Schafer BW, Sasaki N, Heizmann CW 2003 Expression analysis of S100
proteins and RAGE in human tumors using tissue microarrays. Biochemical and
Biophysical Research Communications 307 375-81.
12. Kim SH, Choi YM, Chae HD, Kim KR, Kim CH, Kang BM 2001 Increased expression of endoglin in the eutopic endometrium of women with endometriosis. Fertility
and Sterility 76 918-22.
13. Kohnen G, Campbell S, Jeffers MD, Cameron IT 2000 Spatially regulated differentiation
of endometrial vascular smooth muscle cells. Human Reproduction 15 284-292.
14. Korn T, Muller R, Kontermann RE 2004 Bispecific single-chain diabody-mediated
killing of endoglin-positive endothelial cells by cytotoxic T lymphocytes. Journal of
Immunotherapy 27 99-106.
15. Krupinski J, Kaluza J, Kumar P, Kumar S, Wang JM 1994 Role of angiogenesis in
patients with cerebral ischemic stroke. Stroke 25 1794-8.
16. Landriscina M, Schinzari G, Cassano A, Leonardo GD, Quirino M, Pozzo C, Scerrati
M, Barone C 2002 S100A13, a new marker of angiogenesis in brain tumors. Annals
of Oncology 13 165 - 166.
17. Lastres P, Bellon T, Cabanas C, Sanchez-Madrid F, Acevedo A, Gougos A, Letarte M,
Bernabeu C 1992 Regulated expression on human macrophages of endoglin, an ArgGly-Asp-containing surface antigen. European Journal of Immunology 22 393-7.
18. Lee WY, Su WC, Lin PW, Guo HR, Chang TW, Chen HH 2004 Expression of S100A4
and Met: potential predictors for metastasis and survival in early-stage breast cancer.
Oncology 66 429-38.
19. Letamendia A, Lastres P, Botella LM, Raab U, Langa C, Velasco B, Attisano L,
Bernabeu C 1998 Role of Endoglin in Cellular Responses to Transforming Growth
Factor-beta. A comparative study with betaglycan. Journal of Biological Chemistry
273 33011-33019.
20. Li DY, Sorensen LK, Brooke BS, Urness LD, Davis EC, Taylor DG, Boak BB, Wendel
DP 1999 Defective angiogenesis in mice lacking endoglin. Science 284 1534-7
21. Lopez-Casillas F, Wrana JL, Massague J 1993 Betaglycan presents ligand to the TGF
beta signaling receptor. Cell 73 1435-44.
Hayrabedyan et al.
67
22. Ma X, Labinaz M, Goldstein J, Miller H, Keon WJ, Letarte M, O’Brien E 2000 Endoglin is overexpressed after arterial injury and is required for transforming growth
factor-beta-induced inhibition of smooth muscle cell migration. Arteriosclerosis,
Thrombosis, and Vascular Biology 20 2546-52.
23. Noyes RW, Hertwig AT, Rock J 1950 Dating the endometrial biopsy. Fertility and
Sterility 1 3-25.
24. Paul M, Mazurek U, Witek A, Graniczka M, Wilczok T 2001 Histone H3 gene expression level as markers of hyperplasia simplex of endometrium. Ginekologia Polska 72
1434-8.
25. Robledo MM, Hidalgo A, Lastres P, Arroyo AG, Bernabeu C, Sanchez-Madrid F,
Teixido J 1996 Characterization of TGF-beta 1-binding proteins in human bone marrow stromal cells. British Journal of Haematology 93 507-14.
26. Salvesen HB, Gulluoglu MG, Stefansson I, Akslen LA 2003 Significance of CD 105
expression for tumour angiogenesis and prognosis in endometrial carcinomas. Acta
Pathologica, Microbiologica, et Immunologica Scandinavica. Section A, Pathology
111 1011-8.
27. Sampson JA 1927 Peritoneal endometriosis due to menstrual dissemination of endometrial tissue into the peritoneal cavity. American Journal of Obstetrics and Gynecology
14 422-469.
28. Shi SR, Cote RJ, Taylor CR 2001 Antigen Retrieval Techniques: Current Perspectives.
Journal of Histochemistry and Cytochemistry 49 931-937.
29. St-Jacques S, Cymerman U, Pece N, Letarte M 1994 Molecular characterization and in
situ localization of murine endoglin reveal that it is a transforming growth factor-beta
binding protein of endothelial and stromal cells. Endocrinology 134 2645-57.
30. Wang JM, Kumar S, Pye D, Haboubi N, al-Nakib L 1994 Breast carcinoma: comparative study of tumor vasculature using two endothelial cell markers. Journal of National
Cancer Institute 86 386-8.
31. Westphal JR, Willems HW, Schalkwijk CJ, Ruiter DJ, de Waal RM 1993 A new 180kDa dermal endothelial cell activation antigen: in vitro and in situ characteristics.
Journal of Investigative Dermatology 100 27-34.
32. Zhang EG, Smith SK, Charnock-Jones DS 2002 Expression of CD105 (endoglin) in
arteriolar endothelial cells of human endometrium throughout the menstrual cycle.
Reproduction 124 703-11.