Human Reproduction Vol.20, No.6 pp. 1695–1701, 2005 doi:10.

1093/humrep/deh794
Advance Access publication March 3, 2005

Interleukin 1a and tissue-lytic matrix metalloproteinase-1

are elevated in ectopic endometrium of patients with
endometriosis

G.Hudelist1,2,7, H.Lass3, J.Keckstein2, I.Walter4, F.Wieser5, R.Wenzl5, R.Mueller6,

K.Czerwenka6, E.Kubista1 and C.F.Singer1
1
Department of Obstetrics and Gynecology, Division of Special Gynecology, Vienna Medical University, 1090 Vienna, 2Department
of Obstetrics and Gynaecology, LKH Villach, 9500 Villach, 3Department of Obstetrics and Gynecology, Wilhelminenspital, 1060
Vienna, 4Department of Histology/Embryology, University of Veterinary Medicine, 1200 Vienna, 5Department of Obstetrics and
Gynecology, Division of Endocrinology and Reproductive Medicine, 6Department of Clinical Pathology, Division of
Gynaecopathology, Vienna Medical University, 1090 Vienna, Austria
7
To whom correspondence should be addressed at: Department of Obstetrics and Gynecology, LKH Villach, Nikolaigasse 43,

Downloaded from http://humrep.oxfordjournals.org/ by guest on January 1, 2015

A –9500 Villach, Austria. E-mail: gernot_hudelist@yahoo.de

BACKGROUND: Matrix metalloproteinases (MMP) play an essential role in tissue remodelling and menstruation
and appear to be regulated by cytokines such as interleukin-1a (IL-1a). In order to investigate their role in the
pathogenesis of endometriosis, the aim of the present study was to compare the protein localization of matrix
metalloproteinase-1 (MMP-1) and of its main stimulatory cytokine IL-1a in eutopic and dystopic endometrium of
patients with endometriosis. METHODS: MMP-1 and IL-1a protein localization was analysed retrospectively in
paired paraffin-embedded tissue biopsies obtained simultaneously from the endometrial cavity and from endo-
metrial lesions of 37 patients with peritoneal or ovarian endometriosis and in cycling endometria from 37 women
without endmetriosis. Protein localization was demonstrated by immunohistochemistry; antibody specifity was con-
firmed by western blot analysis. RESULTS: MMP-1 and IL-1a protein staining in women suffering from endo-
metriosis was significantly more pronounced in endometriotic lesions than in eutopic endometrium. This held true
for both epithelial MMP-1 and IL-1a staining (P < 0.006 and P < 0.001), and for stromal MMP-1 and IL-1a stain-
ing (P < 0.001 and P < 0.001). Furthermore, stromal MMP-1 and IL-1a were significantly co-expressed in dystopic
endometriotic tissue (P 5 0.045). Endometrial MMP-1 and IL-1a protein expression pattern in eutopic endo-
metrium from women suffering from endometriosis, however, did not differ significantly from the pattern seen in
healthy women. CONCLUSIONS: The increased expression of both matrix-degrading MMP-1 and its major stimu-
latory cytokine IL-1a in endometriotic lesions and the selective co-expression in the stroma of endometriotic foci
clearly suggests their involvement in the pathogenic mechanisms leading to local invasion and tissue destruction.

Key words: endometriosis/IL-1a/MMP-1

Introduction Although the association between endometriosis and infer-

Ten to fifteen per cent of women in reproductive age and up
to 50% of infertile patients are affected by endometriosis, a
disease that is characterized by the dystopic location of endo-
metrial cells in tissues other than the uterine cavity (Goldman
and Cramer, 1990). Although endometriosis is considered to
be a benign gynaecological disorder, several pathophysio-
logical aspects of endometrial lesions resemble the behaviour
of malignant tissue: similar to carcinomas, endometriotic
cells are able to invade surrounding tissue and to locally
destruct the extracellular matrix (ECM), thereby leading to
classical symptoms such as pelvic pain, dyspareunia or
infertility.
q The Author 2005. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved.
For Permissions, please email: journals.permissions@oupjournals.org
tility has led to extensive research efforts, the origin of endo-
metriotic cells remains elusive. The most widely accepted
theory on the aetiology of endometriosis is retrograde
menstruation (Brosens and Brosens, 2000). However, besides
the fact that this phenomenon has also been described in
healthy women and can thus be viewed as a physiological
process (Halme et al., 1984; Chung et al., 2002), several
lines of evidence indicate that endometriotic lesions are bio-
logically different from normal uterine endometrium (Kressin
et al., 2001; Sillem et al., 2001; Chung et al., 2002;
Mizumoto et al., 2002). One of the most prominent pheno-
typic features of endometriotic lesions is the expression of
1695
G.Hudelist et al.

specific matrix metalloproteinases (MMP). Constituting a Table I. Patient characteristics

group of matrix-degrading zinc enzymes, MMP are not only
Patient no. Age Cycle phase Localization of endometriotic tissue
known to play a pivotal role in the initiation of menstrual (years)
bleeding, but have also been shown to contribute to implan-
tation and further invasion of seeded endometriotic explants 1 45 Secretory Fallopian tube
2 51 Unknown Ovary
(Singer et al., 1997; Henriet et al., 2002). Chung et al. (2001, 3 40 Proliferative Peritoneum
2002); Ria et al. (2002) and Wenzl and Heinzl (1998) have 4 26 Secretory Ovary
detected higher levels of MMP-2 and -9 expression in euto- 5 30 Secretory Peritoneum (vesical)
6 41 Secretory Peritoneum (fossa ovarica)
pic and in ectopic endometria of patients with endometriosis 7 55 Proliferative Peritoneum (cavum Douglas)
than in corresponding healthy controls. Furthermore, Cox 8 52 Proliferative Peritoneum (cavum Douglas)
et al. (2001) observed elevated levels of MMP-3 in ectopic 9 33 Secretory Peritoneum (cavum Douglas)
10 28 Proliferative Peritoneum (fossa ovarica)
endometrial explants compared to normal uterine-derived 11 45 Proliferative Peritoneum (fossa ovarica)
endometrium in a rat model of endometriosis. 12 54 Proliferative Ovary
Some reports also indicate a possible function for MMP-1 13 30 Proliferative Sacro-uterine ligament
14 36 Secretory Peritoneum (cavum Douglas)
in the development and invasion of endometriosis. During 15 39 Proliferative Peritoneum (vesical)
the normal menstrual cycle, MMP-1 expression and activity 16 33 Secretory Peritoneum (fossa ovarica)
is highly regulated and is confined to a few days before and 17 22 Secretory Sacro-uterine ligament
18 45 Secretory Ovary
during menstruation, where it is thought to act as the princi- 19 29 Proliferative Peritoneum (fossa ovarica)
pal tissue-degrading proteolytic enzyme (Kokorine et al., 20 29 Proliferative Ovary

Downloaded from http://humrep.oxfordjournals.org/ by guest on January 1, 2015

1997; Singer et al., 1997). Kokorine et al. (1997) demon- 21 30 Proliferative Peritoneum (vesical)
22 52 Unknown Ovary
strated a direct correlation between the expression of MMP-1 23 46 Secretory Ovary (endometrioma)
and the activity of endometriotic foci. In agreement with this 24 47 Proliferative Peritoneum (vesical)
finding, Sillem et al. (2001) observed elevated levels of 25 49 Unknown Ovary
26 46 Proliferative Sacro-uterine ligament
MMP-1 and MMP-3 in culture supernatants of uterine endo- 27 33 Proliferative Peritoneum (cavum Douglas)
metrial cells of patients suffering from endometriosis when 28 36 Proliferative Peritoneum (fossa ovarica)
compared to normal uterine endometrium. 29 49 Proliferative Ovary (endometrioma)
30 44 Proliferative Peritoneum (vesical)
While the release of interleukin-1a (IL-1a) appears to be 31 29 Proliferative Ovary
the principal inducer of endometrial MMP-1 expression 32 52 Proliferative Ovary
during menstruation (Rawdanowicz et al., 1994; Singer et al., 33 43 Secretory Ovary
34 33 Secretory Ovary (endometrioma)
1997; Keller et al., 2000), the mechanisms leading to the 35 44 Secretory Peritoneum (cavum Douglas)
elevation and abnormal expression of MMP-1 and other 36 32 Proliferative Sacro-uterine ligament
MMP in endometriotic tissue are still poorly understood. 37 28 Secretory Peritoneum (fossa ovarica)
Recent studies strongly suggest a link between the expression
of members of the interleukin family and the altered to patients with a history of pelvic inflammatory disease or malig-
nancy, adenomyosis uteri, intake of GnRH agonists or exposure to
secretion of MMP in endometriotic lesions. Although other
steroids within the last 6 months prior to surgery were excluded.
members of this cytokine family have been shown to be The study was approved by the local IRB and only consenting
elevated in peritoneal fluids of women with endometriosis patients were admitted into the study. Ectopic endometriotic tissues
(Taketani et al., 1992; Calhaz-Jorge et al., 2003; Song et al., were obtained during laparoscopic or open surgical procedures for
2003) as well as in endometriotic tissue samples (Bergqvist severe endometriosis or during hysterectomy for non-malignant
et al., 2001; Chegini et al., 2003), little information exists on lesions such as fibroids. Eutopic endometrial biopsies were either
the localization of interleukin-1a (IL-1a) in endometrial tis- obtained by curettage immediately before the laparoscopic pro-
sue of patients with endometriosis. In order to investigate the cedure, or were obtained by the pathologist immediately after
role of IL-1a and MMP-1 in the proteolytic activity and removal of the organ in patients undergoing hysterectomy for
growth of endometriotic lesions, we have analysed protein reasons unrelated to endometrial pathology. Endometrial tissue from
37 healthy controls (confirmed by laparoscopic surgery) was also
staining of both components in paired eutopic and ectopic
collected. Specimens were embedded in paraffin and in each endo-
endometrial tissue of women with endometriosis and in
metrial sample the cycle phase was determined.
healthy controls.

Immunohistochemical staining of IL-1a and MMP-1

Materials and methods
Patients and tissue samples
Paired eutopic and ectopic endometrial tissue biopsies were obtained
from 37 women with endometriosis of the pelvic cavity and from 37
patients who underwent laparoscopic surgery for diseases other than
endometriosis. Characteristics of patients with endometriosis are
listed in Table I. Patients suffering from endometriosis who were
enrolled in the study were pre-menopausal and aged between 22 and
53 years with a median age of 36.3 years. Specimens corresponding
1696
Antibody specifity and immunohistochemical detection of IL-1a in
paraffin-embedded tissue has already been described elsewhere
(Singer et al., 2002). In brief, 5 mm paraffin sections were deparaf-
finized and endogenous peroxidase was inactivated by a 15 min
treatment with methanol containing 0.6% hydrogen peroxide.
Additional antigen retrieval for MMP-1 staining was achieved by
boiling the sections in 10 mmol/l of citrate buffer (pH 6.0) for
30 min in a microwave pressure cooker. After a short rinsing step in
tap water, non-specific binding was blocked with goat serum for
30 min and anti-IL-1a antibody (monoclonal mouse antibody; R&D
 Elevated levels of IL 1a and MMP-1 in ectopic endometriosis

Systems, USA) and anti-MMP-1 antibody (rabbit polyclonal anti- Results

body Ab-6; NeoMarkers, USA) were added at concentrations of
1:50. Formalin-fixed, paraffin-embedded human placenta was used
as positive control for MMP-1 and IL-1a whereas primary
antibodies were omitted in negative controls. All slides were incu-
bated at 4 8C overnight, washed in PBS for 5 min, and incubated
with peroxidase-labelled dextran polymers conjugated to anti-
mouse/anti-rabbit immunoglobulins (EnVision; Dako Corporation,
USA). Sections were subsequently washed with PBS and subjected
either to diaminobenzidine (DAB) or to 3-amino-9-ethylcarbazole
(AEC) substrate for 10 min. Following this, sections were rinsed in
distilled water for 5 min. All slides were counterstained with
Mayer’s haematoxylin, dehydrated, and mounted with DPX mount-
ing medium (Fluka, Switzerland).
Determination of anti-MMP-1 antibody (Ab-6) specificity
Lysates of HFL-1 cells (which express MMP-1) and of HT-1080
cells (which express MMP-9 but not MMP-1) were used to perform
western blot analysis. Lysates were separated on a 4 – 20% Tris –
MMP-1 and IL1-a protein staining in eutopic and ectopic
endometrium
Table II shows staining intensities of MMP-1 and IL-1a in
eutopic and ectopic endometria of patients with endometrio-
sis: ectopic endometriotic glands exhibited strong MMP-1
staining in four cases (11%), moderate signals in eight cases
(22%), weak intensities in 11 samples (29%), and lacked
MMP-1 protein in 14 specimens (38%). In ectopic endo-
metriotic stroma, MMP-1 was strongly expressed in four
samples (11%), moderately in eight (22%), weakly in 16
(43%), and was absent in nine samples (24%). In contrast,
glands in uterine endometrium of patients with endometriosis
lacked strong MMP-1 staining. Moderate intensities were
only observed in five cases (14%) and weak staining was
seen in nine specimens (24%). Eutopic glandular tissue did
not express MMP-1 protein in 23 cases (62%). Similarly,
strong or moderate MMP-1 staining was absent in stromal

Downloaded from http://humrep.oxfordjournals.org/ by guest on January 1, 2015

glycine polyacrylamide electrophoretic gel (PAGE; EC6025BOX, components. A weak stromal MMP-1 staining was only
Invitrogen) at a constant voltage of 200 V for 30 min and transferred observed in three samples (8%), and remained undetectable
to a nitrocellulose membrane (LC2001; Novex) by electroblot ana- in all of the other 34 cases (92%). When paired eutopic and
lysis (20 V for 1 h). Standards (Amersham) were used as molecular ectopic tissues were compared, glandular presence of MMP-1
weight markers. Membranes were then immersed in blocking sol-
in uterine samples significantly correlated with the presence
ution [1 £ Tris-buffered saline (TBS) þ5% bovine serum albumin
of MMP-1 in glands of ectopic lesions (P ¼ 0.001, Kendall’s
(BSA)] for 1 h at room temperature and incubated with 1 mg/ml of
rabbit polyclonal anti MMP-1 Ab-6 (RB1536PO; NeoMarkers) in tau test). Additionally, ectopic lesions in patients with endo-
1 £ TBS þ1% BSA at room temperature for 2 h. After three metriosis expressed significantly higher levels of MMP-1
washes in 1 £ TBS þ 0.5% Tween 20 for 15 min, the membranes when compared to corresponding eutopic tissue, regarding
were incubated in alkaline phosphatase-coupled goat anti-rabbit both glandular and stromal components (P ¼ 0.006 and
antibody (Cat no. ALI4405; BioSource International) for 1 h at P , 0.001 respectively, Figure 1 and 2).
room temperature. Following another washing step, the membranes When the staining intensities of IL-1a in uterine and ectopic
were incubated in 4-nitroblue tetrazolium chloride/5-bromo-4- endometrium of patients with endometriosis were analysed
chloro-3-indolyl phosphate (NBT/BCIP; Roche Diagnostic GmbH, (Table II), ectopic glands exhibited strong staining in three
Austria) developer solution for 10 min. Membranes were developed cases (8%), moderate in six cases (16%), weak in 19 samples
and finally rinsed in bi-distilled water. (52%), and no staining in nine samples (24%). In ectopic endo-
metriotic stroma, IL-1a lacked strong staining in any of the
Immunostaining quantification samples and was moderately detected in two tissues (5%),
A semiquantitative scoring system (immunoreactive score; IRS) weakly in 12 (33%), and absent in 23 samples (62%). Weak
according to Remmele and Schicketanz (1993) was used to allow stromal IL-1a protein was only detected in one uterine endo-
for a reproducible evaluation of protein staining levels in ductal and metrial sample (3%) but was absent in all the other 36 biopsies
stromal components of immunohistochemically stained tissue sec- obtained from endometrial curettages (97%). None of eutopic
tions. Tissue sections were scored independently by two experienced glandular components strongly expressed IL-1a (0%).
pathologists. The IRS was calculated according to the following
formula: IRS ¼ staining intensity (0– 3) £ percentage of positive Table II. Matrix metalloproteinase-1 (MMP-1) and interleukin-1a (IL-1a)
cells or nuclei (0, , 10%; 1, 10 – 25%; 2, 26 –50%; 3, 51 – 75%; 4, protein staining intensities in ectopic and uterine endometrium of patients
with endometriosis
76 – 100%). Possible scores ranged from a minimum of 0 to a maxi-
mum of 12. Scores of 0 – 2 points were considered as negative (0); Lesion Lesion Eutopic Eutopic
3 – 5 points as weak staining (þ); 6 – 8 points as intermediate (þ þ); gland stroma gland stroma
(n ¼ 37) (n ¼ 37) (n ¼ 37) (n ¼ 37)
and 9 – 12 points as strong staining (þþ þ).
MMP-1
Strong 4 (11) 4 (11) 0 (0) 0 (0)
Statistical analysis Intermediate 8 (22) 8 (22) 5 (14) 0 (0)
SSPS version 10 (Statistical Package for Social Sciences; SSPS Inc., Weak 11 (29) 16 (43) 9 (24) 3 (8)
Absent 14 (38) 9 (24) 23 (62) 34 (92)
USA) was used for statistical analysis. Distribution of MMP-1 and
IL-1a
IL-1a in eutopic and ectopic endometrial tissue was compared by Strong 3 (8) 0 (0) 0 (0) 0 (0)
using Kendall’s tau-coefficient (categorical predictor). Wilcoxon’s Intermediate 6 (16) 2 (5) 2 (5) 0 (0)
signed rank test was used for the comparison of quantitative Weak 19 (52) 12 (33) 12 (33) 1 (3)
Absent 9 (24) 23 (63) 23 (63) 36 (97)
differences in the staining of MMP-1 and IL-1a between eutopic
and ectopic endometria. P , 0.05 was considered statistically Values in parentheses are percentages.
significant. Protein levels were assessed by using a semi-quantitative scoring system.

1697
G.Hudelist et al.

Downloaded from http://humrep.oxfordjournals.org/ by guest on January 1, 2015

Figure 1. (A) Uterine endometrium from a patient with endometriosis (magnification £100), and (B) negative for interleukin-1a (IL-1a)
after staining with an anti-IL-1a-specific antibody (£200). In contrast, IL-1a is weakly expressed in stromal and strongly expressed epithelial
components (C: £100; D: £200) of ectopic endometriotic tissue (peritoneal lesion) deriving from the same patient.

Figure 2. (A) Uterine endometrium from a patient with endometriosis (magnification £100), and (B) negative for matrix metalloproteinase-1
(MMP-1) after staining with an anti-MMP-1-specific antibody (£ 200). In contrast, MMP-1 is weakly expressed in stromal and strongly
expressed in epithelial components (C: £ 100; D: £ 200) of ectopic endometriotic tissue (peritoneal lesion) deriving from the same patient.

Moderate staining was observed in two samples (5%), weak
staining in 12 samples (32%), whereas 23 cases failed to
express stromal IL-1a (63%). Glandular localization of IL-1a
protein staining in uterine endometrium significantly correlated
with glandular IL-1a staining of corresponding ectopic lesions
(P , 0.001, Kendall’s tau). Similar to MMP-1, uterine glandu-
lar and stromal endometrial cells expressed significantly lower
1698
levels of IL-1a when compared to ectopic endometrial tissue
(P , 0.001 and P , 0.001 respectively, Kendall’s tau). In
addition, stromal IL-1a protein expression was significantly co-
expressed with stromal MMP-1 in dystopic endometriotic tissue
(P ¼ 0.045). However, no such correlation was found for the
epithelial or epithelial–stromal expression pattern of these two
compounds.
 Elevated levels of IL 1a and MMP-1 in ectopic endometriosis

Histological dating confirmed that 19 patients were in the

proliferative phase and 14 were in the secretory phase. In
four patients, the phase of the menstrual cycle could not be
determined. In our patient collective, the presence of IL-1a
and MMP-1 was not influenced by the endometrial phase in
eutopic and ectopic endometrial samples. Furthermore, the
location of the endometriotic lesions (ovarian endometriosis
versus peritoneal lesions) did not yield significant differences
in MMP-1 or IL-1a expression. We therefore evaluated the
results of immunohistochemical detection of both parameters
irrespective of endometriotic location and menstrual cycle.
We then looked at the IL-1a protein staining in endometria
from women without endometriosis undergoing comparable
cycle phases (Table III). We found an intermediate and weak
epithelial expression in only three cases each (8% and 8%
respectively) and a weak stromal expression in only one case Figure 3. Western blot demonstrating the specifity of anti-MMP-1
(3%), which is similar to the expression pattern we observed antibody (NeoMarkers Ab-6; 57 kDa, lane 2). To exclude cross-reac-
tivity with MMP-9, cell lysate of MMP-9 expressing cell lines (HT-
in the endometrium of endometriotic patients (P ¼ 0.162 1080) was loaded on lane 3 lacking reactivity with anti-MMP-1
and P ¼ 0.368, Kendall’s tau). By analogy, in comparable Ab6. Lane 1: molecular weight markers (97, 66, 45, 30 and 20 kDa).

Downloaded from http://humrep.oxfordjournals.org/ by guest on January 1, 2015

women without endometriosis, epithelial MMP-1 expression
was detected in 12 cases (one case of moderate expression,
histochemistry specifically detects MMP-1 (lane 2, band at
11 cases of weak expression, 3% and 30%) and stromal
57 kDa) but does not recognize MMP-9 (lane 3, band at
expression in 19 cases (two cases of moderate, 17 cases of
92 kDa).
weak expression, 5% and 46%) which is again similar to the
distribution pattern we observed in the endometrium of endo-
metriotic patients (P ¼ 0.077 and P ¼ 0.309 Kendall’s tau).
Stromal MMP-1 in uterine endometrium of healthy controls Discussion
was found to be elevated when compared to eutopic endome- It is now well established that MMP play an essential role in
trium of patients with endometriosis (P , 0.001 respect- tissue remodelling and menstruation. In regularly cycling
ively). No further quantitative differences could be found endometrium, the expression of MMP-1 is induced by the
between healthy controls and uterine endometrium of patients peri-menstrual fall of ovarian steroids and appears to be
with endometriosis. Twenty-two patients were in the prolif- mediated by interleukins, most importantly by IL-1a.
erative phase and 15 were in the secretory phase. Again, stat- Rawdanowicz et al. (1994) were able to stimulate the
istical analysis of the two subgroups revealed that neither the secretion of MMP-1, -3 and -9 in cultured human endo-
presence of IL-1a nor the expression of MMP-1 was influ- metrial cells by adding IL-1a to culture media. Using endo-
enced by the endometrial phase. metrial cells and tissue explants, Singer et al. (1997)
demonstrated a correlation between the epithelial release of
Specifity of anti-MMP-1 antibody Ab-6 by western IL-1a and the stromal production of MMP-1 which could be
blot analysis suppressed by anti-IL-1a antibodies and sex steroids. Simi-
Figure 3 shows the results of western blot analysis performed larly, in vitro data by Keller et al. (2000) provide evidence
on MMP-1-expressing HFL-1 cells (lane 2) and MMP-9- for the IL-1a-mediated stimulation of MMP-3, which could
expressing HT-1080 cells (lane 3). Lane 1 shows molecular be blocked by progesterone in a dose-dependent manner.
weight markers. The anti-MMP-1 antibody used for immuno- Besides a possible paracrine function of IL-1a in the stimu-
lation of MMP, a study conducted on dermal fibroblasts
(Yamamoto et al., 2000) also suggested a mechanism by
Table III. Matrix metalloproteinase-1 (MMP-1) and interleukin-1a (IL-1a)
protein staining intensities in normal endometrium of women without which expression of MMP-1 is mediated by IL-1a via an
endometriosis autocrine loop.
Glandular (n ¼ 37) Stromal (n ¼ 37) In the present study, we hypothesized that the peri-men-
strual rise of these two compounds also could play a role
MMP-1 in the implantation and growth of ectopic endometrium.
Strong 0 (0) 0 (0)
Intermediate 1 (3) 2 (5) Indeed, we observed that ectopic endometrial tissue
Weak 11 (30) 17 (46) obtained from patients with endometriosis expressed signifi-
Absent 25 (68) 18 (49) cantly higher amounts of IL-1a and MMP-1 compared to
IL-1a
Strong 0 (0) 0 (0) corresponding eutopic tissue deriving from the same
Intermediate 3 (8) 0 (0) patients. Interestingly, the presence of MMP-1 was not
Weak 3 (8) 1 (3) exclusively confined to stromal components but could
Absent 31 (84) 36 (97)
also be detected in eutopic and ectopic epithelial cells.
Protein levels were assessed by using a semi-quantitative scoring system. This is in contrast to studies by Kokorine et al. (1996)
1699
G.Hudelist et al.
and Salamonsen and Woolley (1996) who detected MMP-1
expression exclusively in stromal components of normal
endometrial tissue. However, both groups only examined
the presence of MMP-1 in eutopic endometrium but did
not extend their investigations to endometriotic lesions.
Epithelial expression of MMP-1 was also shown by
Mizumoto et al. (2002) using immunohistochemical
techniques, thereby supporting our observation that
MMP-1 is secreted by both stromal and epithelial endo-
metrial cells.
Although the present work does not provide evidence for a
mechanistic link between IL-1a and MMP-1, we found
IL-1a to be significantly co-expressed with MMP-1 in the
stroma of dystopic endometriotic tissue. This observation is
in accordance with previous studies indicating that the
secretion of MMP-1 in endometriotic tissue could be
cytokine-mediated, possibly via an IL-1a autocrine loop
(Yamamoto et al., 2000). Our findings that the presence of
eutopic glandular MMP-1 and IL-1a is correlated with the
presence of endometriotic glandular IL-1a and MMP-1 pro-
tein suggests a biochemical similarity of eutopic and ectopic
endometrial tissue. We therefore aimed to investigate paired
eutopic and ectopic endometrial samples from patients with
endometriosis, in contrast to previous works investigating
endometrial tissues obtained from patients with endometriosis
and healthy controls. Nevertheless, we also looked at the
expression of IL-1a and MMP-1 in endometrial samples of
healthy controls. We found MMP-1 levels to be elevated in
uterine stroma of women without disease when compared to
uterine stroma of patients with endometriosis during both the
secretory and proliferative phases. This somewhat contradicts
previous studies demonstrating that significant endometrial
MMP secretion is confined to the comparatively short peri-
menstrual phase. Although the reason for and significance of
the decreased uterine MMP-1 expression in women with
endometriosis are unclear, it is quite possible that it is the
side-effect of a physiological systemic feedback regulation
by which the body attempts to down-regulate elevated MMP
in endometriotic foci.
Some in vitro studies have already pointed out a possible
correlation between the altered expression of MMP-1 and IL-
1a and the ability of endometrial cells to infiltrate their local
environment. Recently, Wolber et al. (2003) found a signifi-
cant increase of MMP-1 mRNA concentrations of endo-
metrial explants after culture on a model of endometriosis
using chorioallantoic membranes of chick embryos (CAM).
Apart from several studies reporting on altered cytokine
levels in endometriotic tissue, our own laboratory was able to
detect a direct correlation between the expression of IL-1a
and the invasive properties of malignant endometrial epi-
thelium (Singer et al., 2002). To our knowledge, this is the
first report describing elevated levels of IL-1a and MMP-1
protein in endometriotic lesions when compared to the
corresponding eutopic endometrium from patients with endo-
metriosis. We therefore hypothesize that the increased
expression of MMP-1 and its stimulatory cytokine IL-1a in
endometriotic explants is involved in the implantation and
local tissue invasion of endometriotic lesions.
1700
References
Bergqvist A, Bruse C, Carlberg M and Carlstrom K (2001) Interleukin 1beta,
interleukin-6, and tumor necrosis factor-alpha in endometriotic tissue and
in endometrium. Fertil Steril 75,489– 495.
Brosens IA and Brosens JJ (2000) Endometriosis. Eur J Obstet Gynecol
Reprod Biol 90,159–164.
Calhaz-Jorge C, Costa AP, Santos MC and Palma-Carlos ML (2003) Perito-
neal fluid concentrations of interleukin-8 in patients with endometriosis
depend on the severity of the disorder and are higher in the luteal phase.
Hum Reprod 18,593–597.
Chegini N, Roberts M and Ripps B (2003) Differential expression of inter-
leukins (IL)-13 and IL-15 in ectopic and eutopic endometrium of women
with endometriosis and normal fertile women. Am J Reprod Immunol
49,75–83.
Chung HW, Wen Y, Chun SH, Nezhat C, Woo BH and Lake Polan M
(2001) Matrix metalloproteinase-9 and tissue inhibitor of metalloprotei-
nase-3 mRNA expression in ectopic and eutopic endometrium in women
with endometriosis: a rationale for endometriotic invasiveness. Fertil Steril
75,152–159.
Chung HW, Lee JY, Moon HS, Hur SE, Park MH, Wen Y and Polan ML
(2002) Matrix metalloproteinase-2, membranous type 1 matrix metallopro-
teinase, and tissue inhibitor of metalloproteinase-2 expression in ectopic
and eutopic endometrium. Fertil Steril 78,787–795.
Downloaded from http://humrep.oxfordjournals.org/ by guest on January 1, 2015
Cox KE, Piva M and Sharpe-Timms KL (2001) Differential regulation of
matrix metalloproteinase-3 gene expression in endometriotic lesions com-
pared with endometrium. Biol Reprod 65,1297–1303.
Goldman MB and Cramer DW (1990) The epidemiology of endometriosis.
Prog Clin Biol Res 323,15–31.
Halme J, Hammond MG, Hulka JF, Raj SG and Talbert LM (1984) Retro-
grade menstruation in healthy women and in patients with endometriosis.
Obstet Gynecol 64,151–154.
Henriet P, Cornet PB, Lemoine P, Galant C, Singer CF, Courtoy PJ,
Eeckhout Y, Marbaix E (2002) Circulating ovarian steroids and endo-
metrial matrix metalloproteinases (MMPs). Ann NY Acad Sci
955,119– 138. discussion 157 –118, 396 –406.
Keller NR, Sierra-Rivera E, Eisenberg E and Osteen KG (2000) Progesterone
exposure prevents matrix metalloproteinase-3 (MMP-3) stimulation by
interleukin-1alpha in human endometrial stromal cells. J Clin Endocrinol
Metab 85,1611–1619.
Kokorine I, Marbaix E, Henriet P, Okada Y, Donnez J, Eeckhout Y and
Courtoy PJ (1996) Focal cellular origin and regulation of interstitial col-
lagenase (matrix metalloproteinase-1) are related to menstrual breakdown
in the human endometrium. J Cell Sci 109(Pt 8),2151–2160.
Kokorine I, Nisolle M, Donnez J, Eeckhout Y, Courtoy PJ and Marbaix E
(1997) Expression of interstitial collagenase (matrix metalloproteinase-1)
is related to the activity of human endometriotic lesions. Fertil Steril
68,246–251.
Kressin P, Wolber EM, Wodrich H, Meyhofer-Malik A, Buchweitz O,
Diedrich K and Malik E (2001) Vascular endothelial growth factor mRNA
in eutopic and ectopic endometrium. Fertil Steril 76,1220–1224.
Mizumoto H, Saito T, Ashihara K, Nishimura M, Takehara M, Tanaka R,
Ito E and Kudo R (2002) Expression of matrix metalloproteinases in ovar-
ian endometriomas: immunohistochemical study and enzyme immunoas-
say. Life Sci 71,259– 273.
Rawdanowicz TJ, Hampton AL, Nagase H, Woolley DE and Salamonsen LA
(1994) Matrix metalloproteinase production by cultured human endo-
metrial stromal cells: identification of interstitial collagenase, gelatinase-A,
gelatinase-B, and stromelysin-1 and their differential regulation by inter-
leukin-1 alpha and tumor necrosis factor-alpha. J Clin Endocrinol Metab
79,530–536.
Remmele W and Schicketanz KH (1993) Immunohistochemical determi-
nation of estrogen and progesterone receptor content in human breast can-
cer. Computer-assisted image analysis (QIC score) vs subjective grading
(IRS). Pathol Res Pract 189,862–866.
Ria R, Loverro G, Vacca A, Ribatti D, Cormio G, Roccaro AM and
Selvaggi L (2002) Angiogenesis extent and expression of matrix metallo-
proteinase-2 and -9 agree with progression of ovarian endometriomas. Eur
J Clin Invest 32,199– 206.
Salamonsen LA and Woolley DE (1996) Matrix metalloproteinases in normal
menstruation. Hum Reprod 11(Suppl 2),124–133.
Sillem M, Prifti S, Koch A, Neher M, Jauckus J and Runnebaum B (2001)
Regulation of matrix metalloproteinases and their inhibitors in uterine
endometrial cells of patients with and without endometriosis. Eur J Obstet
Gynecol Reprod Biol 95,167– 174.

Singer CF, Marbaix E, Kokorine I, Lemoine P, Donnez J, Eeckhout Y and
Courtoy PJ (1997) Paracrine stimulation of interstitial collagenase (MMP-
1) in the human endometrium by interleukin 1alpha and its dual block by
ovarian steroids. Proc Natl Acad Sci USA 94,10341–10345.
Singer CF, Kronsteiner N, Marton E, Walter I, Kubista M, Czerwenka K,
Schreiber M, Tschugguel W, Wieser F and Kubista E (2002) IL-1alpha
gene expression in human endometrial cancer is independent of ovarian
steroid receptor expression. Eur J Cancer 38(Suppl 6),S76–S77.
Song M, Karabina SA, Kavtaradze N, Murphy AA and Parthasarathy S
(2003) Presence of endometrial epithelial cells in the peritoneal cavity and
the mesothelial inflammatory response. Fertil Steril 79(Suppl 1),789–794.
Taketani Y, Kuo TM and Mizuno M (1992) Comparison of cytokine levels
and embryo toxicity in peritoneal fluid in infertile women with untreated
or treated endometriosis. Am J Obstet Gynecol 167,265–270.
Elevated levels of IL 1a and MMP-1 in ectopic endometriosis
Wenzl RJ and Heinzl H (1998) Localization of matrix metalloproteinase-2 in
uterine endometrium and ectopic implants. Gynecol Obstet Invest 45,
253 –257.
Wolber EM, Kressin P, Meyhofer-Malik A, Diedrich K and Malik E (2003)
Differential induction of matrix metalloproteinase 1 and 2 in ectopic endo-
metrium. Reprod Biomed Online 6,238–243.
Yamamoto T, Eckes B, Mauch C, Hartmann K and Krieg T (2000) Mono-
cyte chemoattractant protein-1 enhances gene expression and synthesis of
matrix metalloproteinase-1 in human fibroblasts by an autocrine IL-1
alpha loop. J Immunol 164,6174–6179.
Submitted on June 29, 2004; resubmitted on December 9, 2004; accepted on
January 18, 2005
Downloaded from http://humrep.oxfordjournals.org/ by guest on January 1, 2015
1701