Morphological changes in mesothelial cells induced by shed menstrual endometrium in vitro are not primarily due to apoptosis or necrosis

Ayse Y. Demir Weusten1,2,4, Patrick G. Groothuis1,3, Gerard A.J. Dunselman1,2, Anton F.P.M. de Goeij1,3, Jan Willem Arends1,3 and Johannes L.H. Evers1,2

1 Research Institute Growth and Development (GROW), Departments of 2 Obstetrics and Gynaecology and 3 Pathology, Academisch Ziekenhuis and Maastricht University, Maastricht, The Netherlands


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
In a previous study on the pathogenesis of endometriosis, we observed that constituents of menstrual effluent induce morphological alterations in human mesothelial cells. In this study, we investigated whether these alterations were associated with apoptosis or necrosis or were the result of cellular remodelling. After overnight incubation of confluent monolayers of human omental mesothelial cells (HOMEC) with conditioned media prepared from menstrual effluent shed anterogradely, severe alterations in morphology were observed. Typical polygonal mesothelial cell cultures at confluency acquired elongated spindle morphology, resulting in gaps between the cells. In contrast, mesothelial cells from the control groups receiving culture medium only, retained a normal morphology. Immunofluorescence staining revealed that cytokeratin, vimentin and actin filaments were still present, homogeneously distributed in the cell cytoplasm following changes in morphology. To evaluate whether the morphological alterations were associated with apoptosis and/or necrosis, the cells were stained with the M30 CytoDeath antibody or annexin V with propidium iodide and analysed using flow cytometry. The results showed that only a small percentage (1–7%) of the affected HOMEC were undergoing apoptosis or necrosis. We conclude that the profoundly altered morphology of HOMEC is a result of cellular remodelling and that the role of apoptosis and necrosis is negligible. Soluble paracrine factors released by cells isolated from menstrual effluent shed anterogradely may induce a reorganization of the cytoskeleton. As a result, the underlying basement membrane will be exposed and the mesothelium may no longer prevent implantation of endometrium shed retrogradely into the peritoneum, thus facilitating the development of endometriosis.

Key words: apoptosis/endometriosis/mesothelium/morphology/necrosis


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Although endometriosis is one of the most frequently encountered gynaecological disorders, its pathogenesis is poorly understood. Viable endometrial fragments in retrograde menstrual effluent (Kruitwagen et al., 1991Go) spilling into the abdominal cavity, are suspected to implant on the peritoneum (Sampson, 1940Go). Alternatively, endometrial cells may induce the mesothelial cells to differentiate into endometrium-like tissue (Levander and Norman, 1955Go).

In previous studies, we have shown that cells obtained from proliferative endometrium and menstrual endometrium do not adhere to the intact mesothelial monolayer of the peritoneum. Denudation of the mesothelium allows endometrial cells to attach to the underlying basement membrane (Groothuis et al., 1999Go; Koks et al., 1999Go). Recently, we observed that cells isolated from menstrual effluent as well as conditioned medium prepared from menstrual effluent are able to induce morphological alterations in cultured mesothelial cells (Koks et al., 2000Go). These changes in morphology resulted in exposure of the underlying basement membrane. Based on these observations, we speculated that by altering the morphology of mesothelial cells, constituents of menstrual effluent might create their own adhesion sites. Morphological alterations, including cell shrinkage, condensation and change into a round morphology, have also been associated with cells undergoing either apoptosis or necrosis (Majno and Joris, 1995Go).

In this study, we investigated whether the morphological alterations in human omental mesothelial cells (HOMEC) induced by the refluxed endometrial cells are the result of the induction of apoptosis or necrosis or, alternatively, are the result of cellular remodelling. Changes in the cytoskeleton were evaluated using immunofluorescent staining of cytokeratin and actin filaments. Apoptosis was evaluated with the DNA fragmentation assay as well as flow cytometry using the M30 CytoDeath antibody (Leers et al., 1999Go) and annexin V binding (Martin et al., 1995Go; Darzynkiewicz et al., 1997Go). The M30 CytoDeath antibody binds to a caspase-cleaved formalin-resistant epitope of cytokeratin 18, which is exposed during early apoptosis (Leers et al., 1999Go). Annexin V binds to phosphatidylserine (PS) that is translocated from the inner to the outer leaflet of the plasma membrane during early apoptosis (Darzynkiewicz et al., 1997Go). During necrosis the cell membrane is degraded, whereas during apoptosis the cell membrane remains intact (Darzynkiewicz et al., 1997Go). Therefore, to distinguish apoptotic from necrotic cells, cells were incubated with annexin V combined with propidium iodide (PI), which only incorporates in nucleic acids after membrane damage (Darzynkiewicz et al., 1997Go).


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Tissue collection and preparation of conditioned media
Human omental tissue was obtained from female patients (n = 9) undergoing abdominal surgery for benign indications. All patients gave written informed consent. The institutional research review board and the medical ethics committee approved the study protocol.

A total of 54 samples of menstrual effluent shed anterogradely were collected by 11 healthy volunteers using a menstrual cup (Koks et al., 1997Go). The volunteers had no history of endometriosis and were not using oral contraception. Menstrual effluent was collected for 2–3 h during the first 3 days of menstruation. When a donor delivered menstrual effluent more than once to the laboratory during the same cycle, the conditioned media obtained were pooled and considered as one sample of conditioned medium. Thus, a total of 29 conditioned media were prepared from 54 menstrual effluents (Table IGo).


View this table:
[in this window]
[in a new window]
 
Table I. Overview of the conditioned media (CM) used in the experiments: samples of menstrual effluent (MS), light microscopy (LM), immunohistochemistry and immunofluorescence (IHC/IF), flow cytometric analysis using M30 CytoDeath antibody (M30) and Annexin V binding (Ann V)
 
After collection, the menstrual effluent was immediately transferred to the laboratory in a sterile plastic tube. The effluent (5–10 ml) was diluted with Dulbecco's modified Eagle's medium (DMEM)/Ham's F-12 (1:5) glutamine (Life Science Technologies, Breda, The Netherlands) and mixed with a Pasteur pipette until a homogeneous suspension was achieved. To remove red blood cells, this suspension was layered on a Ficoll-Paque gradient (Sigma-Aldrich Chemie B.V., Zwijndrecht, The Netherlands) and centrifuged at 1167 g for 5 min. All cell suspensions were centrifuged in a Hettich centrifuge. The interface containing endometrial and inflammatory cells was collected, washed twice in DMEM/Ham's F-12 and resuspended in control medium consisting of DMEM/Ham's F-12 supplemented with 10% fetal calf serum (FCS), 100 IU/ml penicillin, 100 µg/ml streptomycin and 2 mmol/l L-glutamine (all from Life Technologies). The suspension was plated in a T25 tissue culture flask (Costar; Corning Inc., Corning, NY, USA). After culturing for 24 h at 37°C and 5% CO2, the culture supernatant was collected, centrifuged at 1680 g for 10 min in a microfuge and stored at –80°C.

Isolation of human omental mesothelial cells (HOMEC)
A published procedure (Kern et al., 1983Go) was used after minor modifications. The tissue was minced and incubated in 2 mg/ml collagenase (ICN Biomedicals B.V., Zoetermeer, The Netherlands) in DMEM/Ham's F-12 at 37°C for 15–20 min. After digestion, the suspension was filtered through a 400 µm stainless steel sieve (Sigma-Aldrich). The filtrate was centrifuged (30 g for 10 min), after which the floating fat cells were removed. Following a second centrifugation step (605 g for 5 min), the supernatant was discarded and the pellet resuspended in DMEM/Ham's F-12. This suspension was subsequently filtered through a 100 µm nylon mesh filter (Micronic, Lelystad, The Netherlands) and a 10 µm polyamide filter (Stokvis & Smits, IJmuiden, The Netherlands). Mesothelial cells retained on the 10 µm filter were back-washed with DMEM/Ham's F-12, pelleted once more and resuspended in MEM (D-valine; Life Technologies) supplemented with 20% FCS, growth medium supplement (GMS-A) (insulin, selenium and transferrin), 100 IU/ml penicillin and 100 µg/ml streptomycin. High serum concentration as well as the presence of D-valine instead of L-valine are known to selectively inhibit fibroblast growth in vitro (Gilbert and Migeon, 1975Go). In addition, the cells were subjected to differential plating. The cell suspension was transferred to a T-25 flask and incubated at 37°C. Omental fibroblasts adhered within 30 min, whereas most mesothelial cells did not. After incubating for 30 min, the supernatant with the non-adherent mesothelial cells was collected and transferred to a new T25 culture flask. After reaching confluency the mesothelial cells were detached by trypsin/EDTA treatment and were passaged with a split ratio of 1:3. Confluent or subconfluent HOMEC monolayers were used after the second or third passage.

To perform immunohistochemistry, HOMEC were plated on multi-chamber glass slides (ICN Biomedicals). To study the effect of conditioned media on mesothelial cell morphology, cells were plated in 24-well plates. For flow cytometric analysis, HOMEC were plated in 6-well plates.

Treatments
Three treatment groups were included in all experiments. (i) Negative controls were human omental mesothelial cells (HOMEC) cultured for 3–4 days in control medium consisting of DMEM/Ham's F12 supplemented with 10% FCS, 100 IU/ml penicillin, 100 µg/ml streptomycin and 2 mmol/l L-glutamine (Gibco, Life Technologies). (ii) Positive controls; after an adaptation period of 2–3 days in control medium, HOMEC were cultured overnight in control medium supplemented with etoposide (200 µmol/l; Omnilabo International B.V., Breda, The Netherlands). Etoposide inhibits DNA topoisomerase II-alpha and induces apoptosis in various cell types (Bortner et al., 1995Go). (iii) For the experimental group, HOMEC were cultured overnight in conditioned medium.

Characterization of HOMEC morphology
Immunohistochemistry was applied to evaluate the purity of the mesothelial cell cultures. Mesothelial cells were stained with monoclonal antibodies against the epithelial cell marker cytokeratin (RCK106, 1:10; Ramaekers et al., 1990), the stromal cell marker vimentin (1:100, ICN Biomedicals) and the endothelial cell marker CD34 (1:200; Beckton Dickinson, San Jose, CA, USA). In brief, cells were rinsed in Tris-buffered saline (TBS, pH 7.4) and fixed in methanol for 1 min at –20°C, followed by an acetone dip. The cells were rinsed three times in TBS and incubated with the primary antibodies for 1 h at room temperature. After incubation, cells were rinsed three times in TBS and incubated for 1 h at room temperature with biotinylated sheep anti-mouse IgG (1:250; Amersham Nederland B.V., Den Bosch, The Netherlands). Cells were washed again three times in TBS and incubated with a streptavidin-biotin-peroxidase complex (1:1000; StreptABC kit, DAKO A/S, Glastrup, Denmark). Antibody binding was visualized using 3'-3'-diaminobenzidine and hydrogen peroxide. The slides were counterstained with haematoxylin.

After overnight incubation in the presence of control medium, etoposide or conditioned medium, morphological changes of HOMEC were characterized by conventional light microscopy and immunofluorescence. An indirect FITC-conjugated second antibody method using the antibody against vimentin, a pan-cytokeratin antibody against cytokeratins 5, 6, 8, 17, 19 (1:100, MNF116, DAKO A/S), and a direct rhodamine-conjugated phalloidin method for fibrillar actin (Friedman et al., 1984Go) were used to study changes in the cytoskeleton.

DNA ladder assay
After overnight incubation in the presence of 100 µmol/l etoposide, mesothelial cells were collected by scraping and pooled with the detached cells in the culture medium. After centrifugation, the pellet was resuspended in PBS containing proteinase K (60 µg/ml; Boehringer, Mannheim, Germany) and RNase A (100 µg/ml, Boehringer) and incubated at 55°C. After 1 h, fresh proteinase K (60 µg/ml) was added and the suspension was incubated for another hour. Phenol/chloroform/isoamylalcohol (25:24:1) was added, and the mixture was vortexed. After centrifugation for 20 min, the top water phase was collected and 500 µl of chloroform was added. Following vortexing and centrifugation, the DNA in the water phase was collected and precipitated with 3 mol/l sodium acetate (pH 5.2) and 100% absolute ethanol overnight at –20°C. The precipitated DNA was washed with 70% ethanol and centrifuged again. The pellet was then resuspended in TBS buffer and visualized with Gel Star® (FMC Bioproducts, Rockland, ME, USA) after 1.5% agarose gel electrophoresis. The same procedure was applied to U-937 cells that grow in suspension and are known to show a DNA ladder pattern after treatment with etoposide (Bortner et al., 1995Go).

Flow cytometry
For flow cytometric analysis, a FACSort equipped with a single Argon ion laser (Beckton Dickinson, Sunnyvale, CA, USA) was used. Data analysis was performed with CELLQuest software (Beckton Dickinson). Data were gated on pulse-processed propidium iodide signals to exclude doublets and larger aggregates as a standard procedure. Ten thousand cells per sample were analysed.

Annexin V
Mesothelial cells were trypsinized and combined with the detached cells in the culture supernatant and incubated with FITC-labelled annexin V (1 µg/ml; Nexins Research B.V., Hoeven, The Netherlands). Propidium iodide (PI, 5 µg/ml) was added to the binding buffer (10 mmol/l HEPES-NaOH, pH 7.4, 150 mmol/l NaCl, 5 mmol/l KCl, 1 mmol/l MgCl2, 1.8 mmol/l CaCl2) together with annexin V. After incubating for 10 min on ice, the cells were washed three times with binding buffer and the FITC signal was quantified on the flow cytometer. Twelve conditioned media were tested in two independent experiments.

M30 CytoDeath antibody
After trypsinization, the attached and detached HOMEC fractions in the culture supernatant were combined and fixed in methanol at –20°C. After 30 min, the cells were washed twice with PBS containing 0.1% Tween 20. Non-specific binding was blocked with PBS containing 1% BSA and 0.1% Tween 20 at room temperature. After 10 min, the blocking buffer was removed and the cells were incubated in 100 µl of M30 CytoDeath antibody (1:100; Boehringer) at room temperature for 60 min. To visualize M30 CytoDeath antibody, a FITC conjugated second antibody was used and the FITC signal was evaluated. Seventeen conditioned media prepared from endometrium from menstrual biopsies were tested in three independent experiments.

Statistical analysis
Analysis of the data obtained by flow cytometry was performed using two-way analysis of variance, using conditioned medium (fixed effect) and experiment (random effect) as independent variables and the logarithm of the observed cell counts as dependent variable. The use of the logarithm was considered necessary due to the heterogeneous variances observed in the data. Variance components were estimated using the type I sums of squares. To test the homogeneity of variances in the groups, the test of Levene (Brown and Forsythe, 1974Go) was used, after removal of the groups containing only single observations.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Characterization of HOMEC cultures
Mesothelial cells isolated from human omentum had a cobblestone appearance typical for these cells (Figure 1AGo). When cultured on glass, the cells stained positive for pan-cytokeratin and vimentin, but were negative for CD34 (Figure 1B–DGo).



View larger version (115K):
[in this window]
[in a new window]
 
Figure 1. (A) The typical cobblestone appearance of a confluent monolayer of human mesothelial cells (HOMEC) grown on a plastic surface. Subconfluent monolayers of HOMEC grown on a glass surface stained positive (brown) with antibodies against pan-cytokeratin (B) and vimentin (C) and negative for CD34 (D). Scale bars represent 100 µm (A) and 40 µm (B, D).

 
Effects of conditioned media on HOMEC morphology
After 4–6 h of incubation with conditioned medium, the HOMEC started to acquire a spindle-like morphology. Following overnight culture the effect became more severe, ultimately resulting in exposure of the plastic culture surface (Figure 2Go). Morphological changes were induced by all of the conditioned media that were used for experiments. The degree of severity observed varied between conditioned media from different donors and from different cycles of the same donor (Figure 2Go).



View larger version (115K):
[in this window]
[in a new window]
 
Figure 2. Light micrographs of HOMEC after overnight co-incubation with conditioned medium. Compared to the control cells (A), HOMEC morphology was severely altered (B). The effects vary when using conditioned medium prepared from menstrual samples collected by the same donor but in a different cycle (C). Scale bars represent 100 µm.

 
After the alterations in mesothelial cell morphology, the cytokeratin and actin filaments were still present, but their staining pattern appeared to be affected (Figure 3Go). In control cultures, cytokeratin filaments were concentrated around the nucleus (Figure 3AGo), whereas actin filaments were more concentrated in foci at the periphery of cells (Figure 3BGo). Upon incubation with conditioned media, both cytokeratin and actin filaments appeared to be distributed equally in the cytoplasm (Figure 3C, DGo), with concentration of cytokeratin filaments in the fibroblast-like extensions (Figure 3DGo). The distribution of vimentin was similar to that observed for cytokeratin (results not shown).



View larger version (68K):
[in this window]
[in a new window]
 
Figure 3. Fluorescence micrographs of HOMEC stained with pan-cytokeratin (A, C) and rhodamine-conjugated phalloidin (B, D), showing the staining patterns before (A, B) and after (C, D) incubation with conditioned medium. Scale bar represents 40 µm.

 
DNA ladder assay
U-937 cells displayed a DNA ladder upon incubation with etoposide, whereas HOMEC did not show this ladder (Figure 4Go).



View larger version (87K):
[in this window]
[in a new window]
 
Figure 4. Results of the DNA ladder assay: lanes 1 and 8 are 100 bp ladders; lanes 2 and 9 are 1 kb ladders; lanes 3–7 and lanes 10–14 are increasing amounts of genomic DNA isolated from mesothelial cells and U-937 cells respectively after overnight incubation with 100 µmol/l etoposide.

 
Flow cytometry
Figure 5Go illustrates the effects of both the etoposide and conditioned media on mesothelial cell morphology, just prior to flow cytometric analysis. In both cases severe morphological alterations occurred. However, etoposide-induced cells appeared to be more rounded up, whereas cells incubated with conditioned media had a more spindle-like appearance. Figure 6Go shows some representative results of flow cytometric analyses. Figure 7Go summarizes the data from all independent experiments.



View larger version (102K):
[in this window]
[in a new window]
 
Figure 5. Light micrograph showing cultures of HOMEC before (A) and after overnight incubation with 200 µmol/l etoposide (B) or conditioned medium (C). Scale bars represent 250 µm.

 


View larger version (35K):
[in this window]
[in a new window]
 
Figure 6. Flow cytometric measurements of HOMEC labelled with M30 CytoDeath antibody (A, C, E, G) and annexin V (B, D, F, H). (A, B) control cultures, (C, D) 200 µmol/l etoposide, (EH) conditioned media.

 


View larger version (11K):
[in this window]
[in a new window]
 
Figure 7. Summary of the results of the flow cytometry experiments. The M30 data represent three independent experiments in which 17 conditioned media were tested. The annexin V data represent the data of two independent experiments in which 12 conditioned media were tested. Bars represent overall mean ± SD.

 
Figures 6A, B and 7Go show the negative controls for M30 CytoDeath antibody (1.3 ± 0.04%) and annexin V binding (1.2 ± 0.5%). After incubation with etoposide, the number of M30 positive cells (Figure 6CGo) and annexin V positive cells (Figure 6DGo) increased dramatically. In addition, a large fraction of the annexin V positive fraction stained positive for PI due to secondary necrosis (Figure 6DGo). When cultured in conditioned media, 1.9% (± 0.9) of the cells stained positive with the M30 CytoDeath antibody (Figures 6E, G and 7GoGo), and 2.8% (± 2.2) of the cells stained positive for annexin V (Figures 6F, H and 7GoGo).

For the M30 positive and annexin V+/PI+ (necrotic) cell fraction measurements, Levene's test of equality of error variances yielded no significance (P = 0.84 and 0.82 respectively). Subsequent analysis revealed that culturing with conditioned media resulted in a small but significant increase in M30-positive cells (P = 0.039). For the annexin V+/PI+ (necrotic) cell fractions no significant effect of the medium was observed. For the annexin V+/PI (apoptotic) cell fractions, Levene's test of equality of error variances did yield highly significant results (P = 0.003), thus complicating further analyses. However, since the observed P-value for the effect of the medium was clearly non-significant (P = 0.61), it is still acceptable to assume that there was no effect of the medium on the annexin V+/PI (apoptotic) cell fractions. When the analysis was performed on the original data, without the log transformation, similar results were obtained.


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Previously we have observed that components of the menstrual effluent induce severe morphological alterations in mesothelial cells. These alterations were not observed in endometrial stroma cells or in ECC-1 cells, an endometrial carcinoma cell line (Koks et al., 2000Go). In the present study, we show that these changes are not the result of mesothelial cells undergoing apoptosis or necrosis.

Flow cytometric analysis revealed that only between 1 and 7% of the HOMEC are apoptotic or necrotic after overnight incubation with conditioned media, whereas the morphology of almost all mesothelial cells was affected. The results of the DNA fragmentation assay showed that mesothelial cells are not subject to internucleosomal DNA cleavage, leading to the typical DNA ladder, and could therefore not be used to evaluate apoptosis. Thus, apoptosis in mesothelial cells is more likely to be associated with cleavage into large fragments or single-strand DNA cleavage (Bortner et al., 1995Go).

Light microscopy observations showed that the staining pattern of cytokeratin, vimentin and actin filaments in mesothelial cells appeared to be affected after incubation with conditioned media. Apparently, cellular remodelling rather than the initiation of apoptosis or necrosis is responsible for the observed alterations in cell morphology.

From oncology studies, it is known that addition of tumour ascites fluid (Kimura et al., 1985Go; Akedo et al., 1986Go) and conditioned medium to cultured mesothelial cells (Niedbala et al., 1985Go), as well as the intraperitoneal injection of tumour ascites fluid or cells (Kiyasu et al., 1981Go; Kishikawa et al., 1995Go), also change the morphology of mesothelial cells. These changes, including disruption of intercellular junctions, retraction and exfoliation of mesothelial cells, lead to the exposure of the underlying extracellular matrix and subsequent attachment and proliferation of tumour cells. Based on these and other observations it has been suggested that infiltration of cancer cells into the peritoneum is also prevented by mesothelial cell monolayers (Leighton et al., 1959Go). Furthermore, others (Yashiro et al., 1996Go) incubated peritoneal cell monolayers with serum-free conditioned medium prepared from eutopic peritoneal fibroblasts, and observed that mesothelial cells also rounded up or exhibited a fibroblast-like morphology. From their results, Yashiro et al. (Yashiro et al., 1996Go) suggested that soluble factors, such as hepatocyte growth factor, affect the morphology of mesothelial cells in monolayer culture, so that the resulting environment may become prone to the peritoneal dissemination of cancer cells. It is clear that the mesothelial lining is very susceptible to alterations inflicted by local or ectopic cells, presumably through the production of paracrine factors.

The interaction between cells isolated from endometrium and mesothelial cell monolayer cultures has been studied before (Sharpe et al., 1992Go; Wild et al., 1994Go; Witz et al., 1999Go). None of these studies reported distinct morphological changes. It has to be taken into account, however, that the endometrium was collected in the proliferative and secretory phase of the menstrual cycle, rather than the menstrual phase. The findings from these studies are in accordance with previous observations that conditioned medium prepared from cyclic endometrium does not affect mesothelial cell morphology (Koks et al., 2000Go).

We have shown that paracrine factors released by menstrual tissue induce changes in cell shape and exposure of extracellular matrix (Koks et al., 2000Go). Several mechanisms may potentially be responsible for this. For instance, the paracrine factors may bind to cell surface receptors and initiate a signalling cascade that eventually results in cellular remodelling. Growth factor receptors are often tyrosine kinases, and after activation by their ligand they are able to induce changes in cellular morphology by altering the phosphorylation status of focal adhesion associated proteins (Burridge et al., 1988Go; Nelson and Fry, 1997Go). These focal adhesion complexes are the sites of actin filaments and changes in the distribution of focal adhesion points would therefore result in a redistribution of the actin cytoskeleton (Gumbiner, 1996Go). These issues are currently under investigation in our laboratory.

We conclude that cells isolated from spontaneously shed menstrual effluent, as opposed to cells obtained from endometrial biopsies, are potentially harmful to the mesothelium. Menstrual effluent shed retrogradely may behave aggressively and damage the mesothelial lining upon entering the abdominal cavity, thus creating adhesion sites. The anatomical distribution of peritoneal endometriosis supports this contention, since most endometriotic lesions are found in the proximity of the Fallopian tube ostia, i.e. on the ovaries and uterine ligaments. It will be of clinical importance to identify the factors responsible for the disruptive effect on the mesothelium. Prevention or neutralization of this effect would aid in reducing the chance of endometriosis development.


    Acknowledgments
 
The authors wish to thank Dr J.J.A.M.Weusten (Organon Technika) for the statistical analysis of the flow cytometry data, Professor Dr F.C.S.Ramaekers (Department of Molecular and Cellular Biology, University Maastricht) for donating the RCK106 antibody and the rhodamine-conjugated phalloidin and Dr Roel Kuijer (Department of Biomaterials, University Maastricht) for supplying etoposide.


    Notes
 
4 To whom correspondence should be addressed at: Department of Obstetrics and Gynaecology, Academic Hospital Maastricht, Postbus 5800, 6202 AZ Maastricht, The Netherlands. E-mail: demirweusten{at}yahoo.com Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Akedo, H., Shinkai, K., Mukai, M. et al. (1986) Interaction of rat ascites hepatoma cells with cultured mesothelial cell monolayers: a model for tumor invasion. Cancer Res., 16, 2416–2422.

Bortner, C.D., Oldenburg, N.B.E. and Cidlowski, J.A. (1995) The role of DNA fragmentation in apoptosis. Trends Cell. Biol., 5, 21–26.[ISI]

Brown, M.B. and Forsythe, A.B. (1974) Robust tests for the equality of variances. J. Am. Statist. Assoc., 69, 364–367.[ISI]

Burridge K., Fath K., Kelly T. et al. (1988). Focal adhesions: transmembrane junctions between the extracellular matrix and the cytoskeleton. Annu. Rev. Cell Biol., 4, 487–525.[ISI]

Darzynkiewicz, Z., Juan, G., Li, X. et al. (1997) Cytometry in cell necrobiology: analysis of apoptosis and accidental cell death (necrosis). Cytometry, 27, 1–20.[ISI][Medline]

Friedman, E., Verderame, M., Winawer, S. and Pollack, R. (1984) Actin cytoskeletal organization loss in the benign-to-malignant tumor transition in cultured human colonic epithelial cells. Cancer Res., 44, 3040–3050.[Abstract]

Gilbert, S.F. and Migeon, B.R. (1975) D-Valine as a selective agent for normal human and rodent epithelial cells in culture. Cell, 5, 11–17.[ISI][Medline]

Groothuis, P.G., Koks, C.A.M., de Goeij, A.F.P.M. et al. (1999) Adhesion of human endometrial fragments to peritoneum in vitro. Fertil. Steril., 71, 1119–1124.[ISI][Medline]

Gumbiner, B.M. (1996) Cell adhesion: the molecular basis of tissue architecture and morphogenesis. Cell, 84, 345–357.[ISI][Medline]

Kern, P.A., Knedler, A. and Eckel, R.H. (1983) Isolation and culture of microvascular endothelium from human adipose tissue. J. Clin. Invest., 71, 1822–1829.[ISI][Medline]

Kimura, A., Koga, S., Kudaoh, H. and Iitsuka, Y. (1985) Peritoneal mesothelial cell injury factors in rat cancerous ascites. Cancer Res., 45, 4330–4333.[Abstract]

Kishikawa, T., Sakamato, M., Ino, Y. et al. (1995) Two distinct patterns of peritoneal involvement shown by in vitro and in vivo ovarian cancer dissemination models. Invasion Metastasis, 15, 11–21.[ISI][Medline]

Kiyasu, Y., Kaneshima, S. and Koga, S. (1981) Morphogenesis of peritoneal metastasis in human gastric cancer. Cancer Res., 41, 1236–1239.[Abstract]

Koks, C.A.M., Dunselman, G.A.J., de Goeij, A.F.P.M. et al. (1997) Evaluation of a menstrual cup to collect shed endometriun for in vitro studies. Fertil. Steril., 68, 560–564.[ISI][Medline]

Koks, C.A.M., Groothuis, P.G., Dunselman, G.A.J. et al. (1999) Adhesion of shed menstrual tissue in an in-vitro model using amnion and peritoneum: a light and electron microscopic study. Hum. Reprod., 14, 816–822.[Abstract/Free Full Text]

Koks, C.A.M., Demir Weusten A.Y., Groothuis, P.G. et al. (2000) Menstruum induces changes in mesothelial cell morphology. Obstet. Gynecol. Invest., in press.

Kruitwagen, R.F., Poels, L.G., Willemsen, W.N. et al. (1991) Endometrial epithelial cells in peritoneal fluid during the early follicular phase. Fertil. Steril., 55, 297–303.[ISI][Medline]

Leers, M.P., Kolgen, W., Bjorkland, V. et al. (1999) Immunocytochemical detection and mapping of a cytokeratin 18 neo-epitope exposed during early apoptosis. J. Pathol., 187, 567–572.[ISI][Medline]

Leighton, J., Kalla, R.L., Kline, I. and Belkin, M. (1959) Pathogenesis of tumor invasion. 1. Interaction between normal tissues and `transformed' cells in tissue culture. Cancer Res., 19, 23–27.[ISI][Medline]

Levander, G. and Norman, P. (1955) The pathogenesis of endometriosis. An experimental study. Acta Obstet. Gynecol. Scand., 34, 366–398.

Majno, G. and Joris, G. (1995) Apoptosis, oncosis and necrosis. An overview of cell death. Am. J. Pathol., 146, 3–15.[Abstract]

Martin, S.J., Reutelingsperger, C.P., McGahon, A.J. et al. (1995) Early redistribution of plasma membrane phosphatidylserine is a general feature of apoptosis regardless of the initiating stimulus: inhibition by overexpression of Bcl-2 and Abl. J. Exp. Med., 182, 1545–1556.[Abstract]

Nelson, J.M. and Fry, D.W. (1997) Cytoskeletal and morphological changes associated with the specific suppression of the epidermal growth factor receptor tyrosine kinase activity in A431 human epidermoid carcinoma. Exp. Cell. Res., 233, 383–390.[ISI][Medline]

Niedbala, M.J., Crickard, K. and Bernacki, R.J. (1985) Interactions of human ovarian tumor cells with human mesothelial cells grown on extracellular matrix. Exp. Cell. Res., 160, 499–513.[ISI][Medline]

Ramaekers, F., Van Niekerk, C., Poels, L. et al. (1990) Use of monoclonal antibodies to keratin 7 in the differential diagnosis of adenocarcinomas. Am. J. Pathol., 136, 641–655.[Abstract]

Sampson, J.A. (1940) The development of the implantation theory for the origin of peritoneal endometriosis. Am. J. Obstet. Gynecol., 40, 549–557.

Sharpe, K.L., Zimmer, R.L., Khan, R.S. and Penney, L.L. (1992) Proliferative and morphogenic changes induced by the coculture of rat uterine and peritoneal cells: a cell culture model for endometriosis. Fertil. Steril., 58, 1220–1229.[ISI][Medline]

Wild, R.A., Zhang, R.J. and Medders, D. (1994) Whole endometrial fragments form characteristics of in vivo endometriosis in a mesothelial cell co-culture system: an in vitro model for the study of the histogenesis of endometriosis. J. Soc. Gynecol. Investig., 1, 65–68.[ISI][Medline]

Witz, C.A., Monotoya-Rodriguez, I.A. and Schenken, R.S. (1999) Whole explants of peritoneum and endometrium: a novel model of the early endometriosis lesion. Fertil. Steril., 71, 56–60.[ISI][Medline]

Yashiro, M., Chung, Y.S., Inoue, T. et al. (1996) Hepatocyte growth factor (HGF) produced by peritoneal fibroblasts may affect mesothelial cell morphology and promote peritoneal dissemination. Int. J. Cancer, 67, 289–293.[ISI][Medline]

Submitted on January 4, 2000; accepted on April 11, 2000.