Interleukin-6- and tumour necrosis factor {alpha}-mediated expression of hepatocyte growth factor by stromal cells and its involvement in the growth of endometriosis

Khaleque Newaz Khan1,4, Hideaki Masuzaki1, Akira Fujishita1, Michio Kitajima1, Koichi Hiraki1, Ichiro Sekine2, Toshifumi Matsuyama3 and Tadayuki Ishimaru1

1 Department of Obstetrics and Gynecology, 2 Department of Molecular Pathology, Atomic Bomb Disease Institute, Nagasaki University, 3 Division of Cytokine Signaling, Department of Molecular Microbiology and Immunology, Graduate School of Biomedical Sciences, Nagasaki University, Nagasaki, Japan

4 To whom correspondence should be addressed. Department of Obstetrics and Gynecology, Graduate School of Biomedical Sciences, Nagasaki University, 1-7-1 Sakamoto, Nagasaki 852-8501, Japan. E-mail: nemokhan{at}net.nagasaki-u.ac.jp


    Abstract
 Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: We investigated the expression of the hepatocyte growth factor (HGF) gene and protein by the stromal cells derived from women with or without endometriosis and its regulation by interleukin-6 (IL-6) and tumour necrosis factor {alpha} (TNF{alpha}). METHODS: Stromal cells immunoreactive to vimentin were isolated from the eutopic and ectopic endometrium of 18 infertile women with endometriosis and 12 women without endometriosis. The production of HGF in the culture media of basal and IL-6- or TNF{alpha}-stimulated stromal cells was examined with an enzyme-linked immunosorbent assay. The mRNA expression of HGF and its receptor c-Met in the stroma was investigated by RT–PCR. The localization of HGF and c-Met in isolated stromal cells and in intact tissue was examined by immunohistochemistry. The effect of HGF on the growth of stromal cells alone or in combination with IL-6 or TNF{alpha} was examined in a bromodeoxyuridine (BrdU) incorporation study. RESULTS: The production of HGF in the culture medium of stromal cells was significantly increased after single or combined treatment with either IL-6 or TNF{alpha} when compared with non-treated cells. The production of HGF by stromal cells derived from the eutopic endometrium of women with endometriosis was significantly higher than that of cells from women without endometriosis. This effect was paralleled by increased expression of HGF and c-Met mRNA, as demonstrated by RT–PCR. The BrdU incorporation study indicated that the addition of HGF enhanced the growth of endometrial and endometriotic stroma alone or in combination with IL-6 or TNF{alpha}. CONCLUSION: IL-6 and TNF{alpha} are involved in the production of HGF by endometrial stromal cells and may be involved in the growth of endometriosis by an autocrine mechanism.

Key words: cell growth/endometriosis/hepatocyte growth factor/IL-6/TNF{alpha}


    Introduction
 Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 Acknowledgements
 References
 
Endometriosis is characterized by the presence and growth of endometrial-like tissue (gland and stroma) outside the uterus and occurs in 10% of women of reproductive age (Straphy et al., 1982Go). The pathogenesis of endometriosis is still debatable. A line of evidence has already demonstrated that an innate immune system and ovarian steroid hormones participate either alone or in an orchestrated fashion in the regulation of this disease (Akoum et al., 2000Go; Khan et al., 2005aGo).

A recent study (Khan et al., 2005bGo) suggests that activated macrophages are increased in the peritoneal fluid of women with endometriosis. These activated macrophages secrete numerous macromolecules that may contribute to the implantation of endometrial cells and the progression of endometriosis (Eisermann et al., 1988Go; Buyalos et al., 1992Go; Donnez et al., 1998Go). There is no doubt that, besides ovarian steroid hormones, the regulation of the growth or maintenance of endometrial or endometriotic tissues is contributed to by a number of cytokines and growth factors such as interleukin (IL)-6, IL-8, tumour necrosis factor {alpha} (TNF{alpha}), vascular endothelial cell growth factor and hepatocyte growth factor (HGF) (Fujishita et al., 1999Go; Harada et al., 2001Go; Khan et al., 2003Go, 2005a,b). Among these different cytokines and growth factors, HGF has been reported to possess divergent biological activities in the invasion, growth or metaplastic transformation of endometrial tissue and pelvic mesothelium (Khan et al., 2003Go; Ishimaru et al., 2004Go; Yoshida et al., 2004Go).

Several studies have demonstrated that HGF is elevated in the peritoneal fluid of women with endometriosis compared with healthy controls (Osuga et al., 1999Go; Khan et al., 2002Go, 2004aGo). Besides the paracrine mode of action, the activity of HGF in an autocrine fashion has already been demonstrated in isolated endometrial stroma, stroma of endometriotic cysts and macrophages derived from women with endometriosis (Yoshida et al., 2004Go; Khan et al., 2005aGo).

We have recently demonstrated (Khan et al., 2005aGo) that HGF can be regulated in endometrial cells and inflammatory cells by basal and lipopolysaccharide-stimulated macrophages and is involved in the growth of endometriosis. Since pelvic endometriotic tissues are a regurgitated product of uterine endometrium bathed in the peritoneal fluid, which is enriched with different cytokines, and the HGF promoter region retains the responsive element for IL-6 and TNF{alpha} (Zarnegar, 1995)Go, it is possible that the production of HGF may be regulated by IL-6 and TNF{alpha} in endometrial stromal cells. However, studies are limited regarding the regulation of HGF using stromal cells derived from either eutopic or ectopic endometrium of pelvic endometriosis. A single report by Sugawara et al. (1997)Go demonstrated significantly increased secretion of HGF by basal stromal cells derived from the eutopic endometrium of women with endometriosis compared with women without endometriosis.

Therefore, we investigated the production of HGF by endometrial stromal cells in response to IL-6 and TNF{alpha} at both the protein and the transcriptional level. In addition, we examined the immunolocalization of HGF and its receptor, c-Met, in isolated stroma and intact tissues. Finally, we demonstrated the growth-promoting effect of exogenous HGF either alone or in combination with IL-6 or TNF{alpha} on stromal cells derived from the eutopic and ectopic endometrium of women with or without pelvic endometriosis.


    Material and methods
 Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 Acknowledgements
 References
 
Subjects
For the measurement of HGF, IL-6 and TNF{alpha} concentrations in peritoneal fluid (peritoneal fluid), a total of 30 women between 22 and 38 years of age undergoing laparoscopy for pelvic pain, dysmenorrhoea and/or infertility were recruited. A total of 16 women between 23 and 39 years of age who were free of endometriosis by laparoscopy formed the control group. Endometriosis was diagnosed laparoscopically and histologically. The extent of the disease was staged according to the revised classification of the American Society of Reproductive Medicine (r-ASRM) (1997). Neither the study group nor the endometriosis-free group had been on hormonal medication in the 3 months prior to the surgical procedure. The detailed clinical characteristics of women with and without endometriosis who were included in the present Study are described in a separate study (Khan et al., 2002Go).

All women with or without endometriosis had regular menstrual cycles (28–32 days). All induced menstrual cycles were excluded from the current study. For the isolation of stromal cells in primary culture, biopsy samples were collected from either eutopic or ectopic endometrium of 18 women with endometriosis and 12 women without endometriosis. All biopsy specimens and peritoneal fluids were collected in accordance with the guidelines of the Declaration of Helsinki and with the approval by the Nagasaki University Institutional Review Board. Informed consent was obtained from all women.

Isolation and culture of stromal cells
Stroma was collected from the biopsy specimens of eutopic and ectopic endometrium derived from the women with or without endometriosis. The detail procedure for the isolation of stroma is described previously (Osteen et al., 1989Go; Sugawara et al., 1997Go; Khan et al., 2005aGo).

The characteristics of the cultured stromal cells were determined by morphological appearance and immunocytochemical studies using human vimentin monoclonal antibody (stromal cell-specific) at a dilution of 1:20 (V9; Dako, Glostrup, Denmark) and as described recently (Khan et al., 2005aGo). The isolated cells were placed in four-chamber slides (Nunc, Naperville, IL, USA). After 24 h, the slides were washed in phosphate-buffered saline (PBS), fixed with 4% paraformaldehyde for 10 min, and rinsed with PBS. Slides then were incubated in 0.1% Triton X-100 for 5 min and incubated for 3 h at 37°C with the following antibodies to exclude contamination with other cells: monoclonal antibody (mAb) against human cytokeratin (epithelial-cell specific) at a dilution of 1:50 (MNF 116; Dako), mAb against human von Willebrand factor (endothelial-cell specific) at a dilution of 1:50 (Dako), and mAb against CD45 (other leucocytes) at 1:50 dilution (Dako). The specificity of the immunocytochemical staining was confirmed by the deletion of the first antibody. Immunocytochemical staining was performed on at least three different isolated cells with similar results.

HGF assay in the culture media of treated and non-treated stromal cells
The culture media of basal (non-treated) and stimulated (1, 10 and 50 ng/ml each of IL-6 and TNF{alpha}) stromal cells were collected prospectively and assays were performed retrospectively in each of six women with and without endometriosis (three in the proliferative phase and three in the secretory phase of menstruation). The concentrations of HGF in the serum-free culture media were measured in duplicate using a commercially available sandwich enzyme-linked immunosorbent assay (ELISA) (Quantikine; R & D system, Minneapolis, MN, USA) in a blind fashion. The antibody used in HGF determination does not cross-react with other cytokines. The use of anti-HGF antibody did not affect this assay system. The limit of detection was 40.0 pg/ml for HGF. The intra-assay and inter-assay coefficients of variation were both <10% for all these assays. The neutralizing effect on HGF secretion in the culture media was also performed by preincubation for 4 h with each of anti-IL-6 antibody, anti-TNF{alpha} antibody and anti-HGF antibody (10 µg/ml for each), and the cells were then treated with either IL-6 or TNF{alpha} (10 ng/ml for each) for another 24 h without washing the preincubated antibodies.

Immunolocalization of HGF in isolated cells and in intact tissue
In order to immunolocalize HGF and its receptor, c-Met, in vimentin-immunoreactive isolated stromal cells and in intact tissue, we performed immunohistochemistry using antibodies on paraffin-embedded tissue sections of eutopic and ectopic endometrium derived from each of six women with or without endometriosis (three in the proliferative phase and three in the secretory phase). We used a 1:50 dilution of a rabbit polyclonal antibody against a recombinant protein of HGF{alpha} (H-145) (sc-7949; Santa Cruz Biotechnology, Santa Cruz, CA, USA) and a 1:20 dilution of mouse monoclonal antibody raised against NCL-c-Met receptor (clone 8F11; Novocastra Laboratories, Newcastle, UK) of human origin.

Immunohistochemistry was performed in 5 µm thick serial sections of paraffin-embedded tissues as described previously (Khan et al., 2003Go, 2005a). Non-immune mouse immunoglobulin (Ig) G1 antibody (1:50) was used as a negative control. Placental tissue, which is known to exhibit high levels of HGF, was used as a positive control. The immunostaining was quantified by a modified method of quantitative histogram scoring (Q-H score) as described recently (Khan et al., 2003Go; Ishimaru et al., 2004)Go. The Q-H score was calculated using the following equation: Q-H score = Spi (I + 1), where i = 1, 2 or 3 and Pi is the percentage of stained cells for each intensity. The staining intensity was graded as 0 = no staining, 1 = weak, 2 = moderate, and 3 = strong. We calculated the mean Q-H scores of five different fields of one section by light microscopy at moderate magnification (x200).

Determination of HGF and c-Met mRNA by RT-RCR in stromal cells
We used the RT–PCR technique to determine the mRNA levels of HGF and its receptor, c-Met, in basal (non-treated) and stimulated stromal cells (treated with 1, 10 and 50 ng/ml of each of IL-6 and TNF{alpha} either alone or in combination) derived from women with or without endometriosis. Total RNA was isolated from each of 106 stromal cells cultured in 60 mm Petri dishes (Greiner) using monophasic solutions of 40% phenol and Isogen (Molecular Research Center, Tokyo, Japan), according to the manufacturer’s protocol.

Reverse transcription of RNA extracted from cultured endometrial stromal cells and cDNA and PCR amplification were performed as described recently (Khan et al., 2005aGo). The human oligonucleotide primers of HGF, c-Met and {beta}-actin use in the present study, their location in the cDNA molecule and the corresponding GenBank accession numbers have been described previously (Khan et al., 2005aGo).

The amplification reaction was initiated by heat denaturation at 94°C for 1 min, annealing of the primers for 1 min at 59°C and extension for 1 min at 72°C. This was repeated for 32 cycles for each of HGF and c-Met using a PCR apparatus (Takara Biomedicals, Tokyo, Japan). The amplification protocol for {beta}-actin, used as an internal control, was same as above except that the annealing condition for the primer was 62°C for 1 min and the reaction was repeated for 23 cycles. After the final cycle, the temperature was maintained at 72°C for 10 min to allow completion of synthesis of amplification products. Although not shown, a control with no reverse transcription was run with each sample to confirm that PCR products were free of DNA contamination.

Analysis of PCR-amplified products was performed by fractionation over a 1.5% agarose gel followed by ethidium bromide staining of DNA bands. A scanner densitometer was used to determine the ratio of intensity of each band relative to {beta}-actin, and is represented as the relative expression of the target gene. This relative expression is defined by the fold increase of the mRNA band intensity compared with the basal levels (0 control) and normalized to 1, as described recently (Khan et al., 2005aGo). Densitometric analysis of gel bands was performed using the National Institutes of Health image analysis program (Image J1.33).

Cell proliferation assay by bromodeoxyuridine (BrdU) incorporation study
Proliferation of the stromal cells derived from eutopic and ectopic endometrium was determined spectrophotometrically by measuring the incorporation of BrdU into the replicated cells. The detail procedure for the BrdU incorporation assay was described recently (Khan et al., 2005aGo). The absorbance values correlated directly with the amount of DNA synthesis and thereby with the number of proliferating cells in the culture. The cell proliferation study was performed after single or combined treatment with recombinant IL-6, TNF{alpha} or HGF (R & D Systems).

In order to test whether, besides IL-6 and TNF{alpha}, growth-promoting factor activity in the treated culture medium is contributed by HGF, we used antibody to deplete HGF in the conditioned medium. Cells were pretreated with anti-HGF antibody (10 µg/ml; R & D Systems), incubated for 4 h and then again treated with exogenous IL-6, TNF{alpha} or HGF for another 24 h without washing the preincubated antibody; changes in cell growth were then examined.

Statistical analysis
The clinical characteristics of the subjects were evaluated by one-way analysis of variance. The data are expressed as mean ±SEM or mean ± SD. The distribution of each result was initially analysed with the F-test. When the F-test indicated a skewed distribution, we applied non-parametric statistical analysis, such as the Mann–Whitney U-test. When the F-test indicated a normal distribution of the results between groups, we applied parametric statistical analysis, such as Student’s t-test. Since the concentration of HGF in the culture media was not normally distributed, HGF level differences between endometriosis and non-endometriosis were analysed by non-parametric test. Other continuous variables between groups were compared with Student’s t-test. Spearman’s rank order correlation test was used to determine the correlation between pairs of markers. A power calculation was performed to assess the number of patients required in the proliferative and secretory phases to detect a statistically significant (P = 0.05) difference in the means of 500 pg/ml of HGF at a power of 80%. Differences were considered as statistically significant at P < 0.05.


    Results
 Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 Acknowledgements
 References
 
There were no significant differences in clinical characteristics between women with and without endometriosis (data not shown). We and others previously demonstrated that the levels of HGF, IL-6 and TNF{alpha} in peritoneal fluid were significantly higher in women with endometriosis than in women without endometriosis (Harada et al., 2001Go; Khan et al., 2002Go, 2004aGo). In the present study, we further analysed the relationships between HGF and IL-6 and between HGF and TNF{alpha} in the peritoneal fluid of 30 women with endometriosis. We found a significant positive correlation between the levels of HGF and IL-6 (r = 0.476, P = 0.0079; Figure 1A) and between HGF and TNF{alpha} (r = 0.685, P < 0.0001; Figure 1B). We did not find any correlation between HGF and IL-6 or between HGF and TNF a in 16 women without endometriosis (data not shown).



View larger version (18K):
[in this window]
[in a new window]
 
Figure 1. Correlation between the levels of HGF and IL-6 (A) and between HGF and TNF{alpha} (B) in the peritoneal fluid of 30 women with pelvic endometriosis. A significant positive correlation was observed between the peritoneal fluid levels of HGF and IL-6 (r = 0.476, P = 0.0073) and between HGF and TNF{alpha} (r = 0.685, P < 0001). No difference was observed between HGF and IL-6 or between HGF and TNF a in 16 women without endometriosis (data not shown).

 

HGF production by single and combined treatment of IL-6 and TNFa
After an initial time-dependent study, we found that HGF concentration in the culture media after IL-6 and TNFa treatment peaked at 24–48 h (data not shown). Therefore, in all our following studies we cultured all treated-and non-treated cells for a period of 24 h. As shown in Figure 2A, HGF production in the culture medium was more increased in women with endometriosis (4- to 6-fold) than in women without endometriosis (3- to 4-fold). Although a peak increase in the concentration of HGF was noted at 10 ng/ml each of IL-6 and TNFa compared with either at 1 ng/ml or 50 ng/ml, and also when compared with non-treated stromal cells (Figure 2A), a synergistic effect between IL-6 and TNFa (each at 10 ng/ml and 50 ng/ml) on increasing HGF production was equally observed in women with and without endometriosis. However, no significant difference was observed between single treatment and combined treatment with these cytokines in women with endometriosis and women without endometriosis.



View larger version (22K):
[in this window]
[in a new window]
 
Figure 2. Dose-dependent production of HGF in culture media of isolated stromal cells under treatment with IL-6 or TNF{alpha} alone or combined, derived from the eutopic endometrium of six women with pelvic endometriosis and six women without endometriosis (A) (three each in the proliferative phase and three in the secretory phase of menstruation). The asterisk denotes a significant difference from cells of women without endometriosis at the indicated concentrations of IL-6 and TNF{alpha} (*P < 0.05, A). The neutralizing effect on IL-6- or TNF{alpha}- mediated production of HGF after application of antibody against IL-6, TNF{alpha} or HGF demonstrated partial inhibition of HGF production in the culture media in endometriosis (+) group (B). The results are expressed as mean ± SEM of duplicate determinations of six patients. There was no difference in HGF secretion between the phases of menstruation in either of these two groups of women.

 

When we compared the secretion of HGF by basal (non-treated) and treated stromal cells between women with and without endometriosis, we found that basal stromal cells of women with endometriosis secreted more HGF in the culture media than those from women without endometriosis (180.4 ± 14.2 versus 120.1 ± 2.5pg/ml, P = 0.06), although this difference did not reach statistical significance. However, HGF production by the stromal cells of endometriosis was significantly higher than that from non-endometriosis after single treatment with 10 ng/ml of IL-6 (478.2 ± 20.6 versus 321.5 ± 26.2 pg/ml, P < 0.05) or after combined treatment with IL-6 and TNF{alpha} (642.2 ± 55.0 versus 461.4 ± 25.8 pg/ml at 10 ng/ml each, P < 0.05; and 613.5 ± 19.9 versus 493.2 ± 7.2 pg/ml at 50 ng/ml each, P < 0.05; Figure 2A). Although an apparent increase in the secretion of HGF was observed after treatment with 10 or 50 ng/ml of TNF{alpha} and combined treatment with 1 ng/ml of each of IL-6 and TNF{alpha} in women with endometriosis, no significant difference was found when compared with similarly treated cells from non-endometriosis women (Figure 2A).

We also did not find any difference in the secretion of HGF between the proliferative and secretory phases of the menstrual cycle (data not shown). A sample size calculation indicated that 50 patients in each group would be required to detect with 80% power a significant difference (P = 0.05) in the value of HGF between the proliferative and secretory phases of menstruation.

Since culture media of IL-6- and TNF{alpha}-treated stromal cells retain other macromolecules in addition to HGF and in order to confirm that the secreted product in the culture media is HGF after stimulation and that its production is mediated by IL-6 and TNF{alpha}, we extended our experiment by using antibody to deplete IL-6, TNF{alpha} or HGF in the conditioned medium of stromal cells derived from women with endometriosis. This neutralizing effect of the respective antibodies on the production of HGF is shown in Figure 2B. We found that, although not significant, the blocking effect of IL-6, TNF{alpha} or HGF tended to decrease the production of HGF in the culture medium towards the level of production by non-treated stromal cells (Figure 2B). This indicates that, besides other cytokines and growth factors, the treated culture medium contains HGF and that its production is mediated by IL-6 and TNF{alpha}.

Immunolocalization of HGF and c-Met in isolated stromal cells and in intact tissue
We previously demonstrated that immunostaining of HGF and c-Met in the eutopic endometrium of women with endometriosis was more intense than in that of control women (Khan et al., 2003Go). Furthermore, immunoreaction of HGF in the epithelium and stroma of eutopic endometrium derived from women with endometriosis was stronger than that of control women (Khan et al., 2003Go). In the present study, we further confirmed the tissue localization of HGF and its receptor, c-Met, in isolated stromal cells (Figure 3b and c, respectively) and in intact tissue derived from the eutopic endometrium (Figure 3e and 3f, respectively) and ectopic endometrium (Figure 3h and i, respectively) of women with endometriosis. The corresponding negative controls, as shown by the immunoreaction to non-immune mouse IgG, are shown in Figure 3a, d and g. Although the data not shown, this immunoreaction of HGF and c-Met in the isolated stroma and intact tissue and as measured by the Q-H score was found to be stronger than that in similar cells or tissues of women without endometriosis. We did not find any significant variation in the immunoreaction of these markers between the proliferative and secretory phases of menstruation (data not shown). This indicates that HGF is being synthesized by stromal cells and that, after binding with c-Met, HGF may have an autocrine mode of action.



View larger version (76K):
[in this window]
[in a new window]
 
Figure 3. Immunohistochemical localization of HGF (middle panel) and its receptor, c-Met (right panel) in the vimentin-positive isolated stromal cells (b and c), in the intact tissue of eutopic endometrium (e and f) and ectopic endometrium (h and i) of the same women with pelvic endometriosis. The corresponding negative controls, as shown by the immunoreaction to non-immune mouse IgG, are shown in a, d and g (left panel). Besides isolated stromal cells (b and c), HGF and c-Met was found to be immunoreactive in the stromal cells as well as in glandular epithelial cells of both eutopic (e and f) and ectopic endometrium (h andi). Final magnification was adjusted to x50. The light microscope was connected to a camera (Vanox AHBS; Olympus, Tokyo, Japan).

 

mRNA expression of HGF and c-Met by IL-6 and TNFa-treated stromal cells
In order to determine if the regulation of HGF and its receptor, c-Met, also occurs at the transcriptional level, we examined HGF and c-Met mRNA expression in stromal cells derived from the eutopic endometrium of women with or without endometriosis and after single or combined treatment with IL-6 and TNF{alpha} (Figure 4).



View larger version (50K):
[in this window]
[in a new window]
 
Figure 4. Effects of single (A, B) and combined treatment (C) with different concentrations of IL-6 and TNF{alpha} (0–50 ng/ml) on mRNA expression of HGF and its receptor, c-Met, in isolated stromal cells derived from the eutopic endometrium of women without endometriosis (A) or with pelvic endometriosis (B, C) and as detected by RT–PCR. The transcript sizes of HGF and c-Met are shown at the indicated gel bands. The expression of HGF and c-Met mRNA was found to be stronger in women with endometriosis (B,C) than that in women without endometriosis (A). {beta}-Actin was used as the internal control.

 

The RT–PCR of HGF, c-Met and {beta}-actin mRNAs gave rise to bands of 505, 536 and 300 bp, respectively (Figure 4). In women with endometriosis, HGF mRNA expression was found to be stronger in women with endometriosis than in women without endometriosis after single treatment with either IL-6 or TNF{alpha} (Figure 4A and B), and this expression of HGF mRNA was increased more after combined treatment with IL-6 and TNF{alpha} (Figure 4C). However, c-Met mRNA expression was found to increase in a dose-dependent manner, and this was observed equally for women with and without endometriosis (Figure 4A and B).

After densitometric analysis, a 2.0- to 2.5-fold increase in the expression of HGF mRNA in stromal cells was found in women with endometriosis after treatment with 10 ng/ml (P < 0.05) and 50 ng/ml (P < 0.05) of IL-6 or TNF{alpha} (Figure 5A). A low level of expression of HGF mRNA was observed in women without endometriosis after similar treatment with IL-6 and TNF{alpha} (Figure 5A). However, these levels of HGF expression were greater than basal production or after treatment with either of the cytokines alone in women without endometriosis.



View larger version (31K):
[in this window]
[in a new window]
 
Figure 5. Densitometric analysis of individual mRNA bands of HGF and c-Met shown in Figure 4 (A, B and C) after treatment with IL-6 and TNF{alpha} alone or combined. The band intensity was initially normalized to the corresponding band of internal control ({beta}-actin) and is represented by the fold increase (relative expression) relative to 0 control (without treatment with IL-6 or TNF{alpha}) and normalized to 1. *Significantly different from women without endometriosis. For both HGF and c-Met (A and B), *P < 0.05 versus non-endometriosis at the indicated doses of IL-6 and TNF{alpha}. Data are mean ± SEM of three different experiments derived from three separate patients from each group.

 

The differences in the relative expression of HGF and c-Met mRNA between the stromal cells of women with and without endometriosis are shown in Figure 5. Like the increased secretion of HGF in the culture media shown in Figure 2A, mRNA expression of HGF and its receptor, c-Met, by the stromal cells of endometriosis was also significantly higher than that by non-endometriosis stromal cells (Figure 5A and B). The statistical differences between endometriosis and non-endometriosis are as follows: HGF mRNA, at 10 ng/ml and 50 ng/ml of IL-6 (P < 0.05 for both) and at 50 ng/ml of IL-6 + TNF{alpha}, P < 0.05; c-Met mRNA, at 10 ng/ml each of IL-6 and TNF{alpha}, P < 0.05 for both; and at 50 ng/ml of IL-6 + TNF{alpha}, P < 0.05.

The production of HGF at both gene and protein levels under the present stimulation protocol was independent of either the proliferative phase or the secretory phase of the menstrual cycle (data not shown).

Effects of IL-6, TNFa and HGF on stromal cell proliferation
Since exogenous HGF stimulated maximum proliferation of endometrial epithelial cells and stromal cells at a concentration of 10–50 ng/ml, as described recently in a dose-dependent study (Khan et al., 2005aGo), a cell proliferation study of stromal cells was carried out with 50 ng/ml of recombinant HGF in the present study. Again, the effect of exogenous IL-6 and TNF{alpha} on stromal cell proliferation occurred at a concentration of 10 ng/ml, because both IL-6 and TNF{alpha} stimulated peak HGF production in the culture medium at this concentration.

We found that single or combined treatment of stromal cells with exogenous IL-6, TNF{alpha} and HGF was able to stimulate significant proliferation of stromal cells derived from the eutopic endometrium of women with and without endometriosis, as measured by BrdU incorporation (expressed as the fold increase and normalized to 0 control) (Figure 6A). We observed that after single treatment with either IL-6, TNF{alpha} or HGF, stromal cells derived from women without endometriosis were unable to significantly incorporate BrdU when compared with non-treated cells. However, stromal cells from women without endometriosis significantly proliferated after the combined treatment (IL-6 + HGF, P < 0.05 and TNF{alpha} + HGF, P < 0.05; Figure 6A). In contrast, stromal cells derived from women with endometriosis were equally responsive to either single or combined treatment with IL-6, TNF{alpha} and HGF (TNF{alpha}, P < 0.05; HGF, P < 0.05; IL-6 + HGF, P < 0.05; TNF{alpha} + HGF, P < 0.05; Figure 6A) when compared with non-treated cells. After parametric analysis, the differences in BrdU incorporation between endometriosis and non-endometriosis are as follows: HGF, P < 0.05; IL-6 + HGF, P < 0.05; TNF{alpha} + HGF, P < 0.05 (Figure 6A).



View larger version (27K):
[in this window]
[in a new window]
 
Figure 6. Bromodeoxyuridine (BrdU) incorporation (A) and cell growth (B), measured by counting the number of cells (initial plating, 105 cells/ml in each well) after single or combined treatment with IL-6, TNF{alpha} or HGF, and after application of a neutralizing antibody to HGF to measure the changes in stromal cell proliferation and cell growth derived from the eutopic endometrium of women with or without pelvic endometriosis. The exogenous addition of HGF either alone or in combination with IL-6 or TNF{alpha} significantly stimulated stromal cell proliferation and cell growth (A and B). The blocking effect of anti-HGF antibody reversed cell proliferation towards that of non-treated cells (A, B). Data are fold increase in cell proliferation and cell growth relative to 0 control (without treatment) and are mean ± SEM of three different experiments derived from three separate patients. *Significantly different from non-treated cells (*P < 0.05 versus 0 control for A;*P < 0.05, **P < 0.01 versus 0 control for B). Unlike BrdU incorporation, anti-HGF antibody significantly suppressed HGF-and IL-6-mediated cell growth (*P < 0.05, B).

 

Since the BrdU incorporation study represents the simple incorporation of BrdU into the proliferated DNA of these cells and does not reflect the actual cell growth, as indicated by increased cell number, we also examined the cell growth of stroma by cell number (initial plating 105cells/ml per well and expressed as the fold increase normalized to 0 control) under IL-6, TNF{alpha} and HGF stimulation (Figure 6B). We found parallel and significantly increased cell growth under a stimulation dose of 10 ng/ml of TNF{alpha} (P < 0.05), 50 ng/ml of HGF (P < 0.05) and the combined effect of IL-6 and HGF (P < 0.01) or TNF{alpha} and HGF (P < 0.01) for women with endometriosis, and under IL-6 + HGF (P < 0.05) or TNF{alpha} + HGF (P < 0.05) for women without endometriosis (Figure 6B). The differences in cell growth between endometriosis and non-endometriosis are as follows: basal stroma, 34.0 ± 3.4 x 105 versus 23.7±1.9 x 105 cells/ml; P = 0.08; stromal cells treated with HGF, IL-6 + HGF and TNF{alpha} + HGF, P < 0.05 for all (Figure 6B).

Finally, we attempted to confirm that the combined proliferative effect of IL-6 and HGF or TNF{alpha} and HGF on stromal cells was partly contributed by HGF and not exclusively by IL-6 or TNF{alpha}, by using anti-HGF antibody to block the effect of exogenous HGF in the culture medium. We found that proliferation of stromal cells as induced by combined IL-6 and HGF or TNF{alpha} and HGF was partly or significantly inhibited after the depletion of HGF in the culture medium. This was observed in both the BrdU incorporation study (Figure 6A) and by cell growth as measured by the number of stromal cells (Figure 6B). In fact, anti-HGF antibody significantly inhibited cell growth mediated by IL-6 + HGF in women with and without endometriosis (P < 0.05 for both; Figure 6B). Although not shown, anti-HGF antibody had a tendency to suppress increased BrdU incorporation and cell number on TNF{alpha} treatment in the endometriosis (+) group. This indicates that the growth or persistence of endometrial and endometriotic tissues in the pelvic microenvironment is an orchestrated effect of different macromolecules, including HGF.

We also studied the exogenous effects of IL-6, TNF{alpha} and HGF on stromal cells derived from ectopic endometrium of women with endometriosis (Figure 7). We found that BrdU incorporation into the stromal cells of peritoneal lesions (Figure 7A) and cell growth (Figure 7B) were higher in response to different cytokines when compared with similar cells of corresponding eutopic endometrium (2.0- to 4.5-fold and 1.5- to 3.5-fold, respectively). It was interesting to observe that stromal cells derived from eutopic endometrium of endometriosis was not responsive to IL-6 alone in increasing cell proliferation; however, single treatment with IL-6 resulted in significant proliferation of the stromal cells of ectopic endometrium (P < 0.05; Figure 7A and B). Again, depletion of HGF in the culture medium after pretreatment with anti-HGF antibody suppressed the IL-6 + HGF- or TNF{alpha} + HGF-mediated increase in proliferation of stromal cells derived from ectopic endometrium. This further indicates that HGF may contribute to the growth of both eutopic and ectopic endometrial stromal cells.



View larger version (28K):
[in this window]
[in a new window]
 
Figure 7. Bromodeoxyuridine (BrdU) incorporation (A) and cell growth (B) after treatment with HGF alone or combined with IL-6 or TNF{alpha} in stromal cells derived from eutopic and ectopic endometrium of women with pelvic endometriosis. Data are fold increase in cell proliferation indicated by BrdU incorporation (A) or cell growth by cell number (B) relative to 0 control (without treatment). BrdU incorporation and cell growth of endometriotic stromal cells in response to IL-6, TNF{alpha} or HGF showed a greater increase compared with similar cells of eutopic endometrium (A and B). *P < 0.05 and **P < 0.01 versus 0 control for both BrdU incorporation (A) and cell growth (B). As in eutopic stromal cells, addition of anti-HGF antibody tended to reverse HGF+IL-6-mediated or HGF + TNF{alpha}-mediated cell growth (A and B) of ectopic stroma.

 

Like eutopic endometrium, the basal or treated stroma of ectopic endometrium did not show any difference in cell growth of between the phases of the menstrual cycle. When we compared the difference in stromal cell proliferation (both basal and treated) between eutopic and ectopic endometrium, we found the following statistical differences: IL-6, P < 0.05; IL-6 + HGF, P < 0.05 or TNF{alpha} + HGF, P < 0.05 for both BrdU incorporation and cell growth (Figure 7A and B). It was also interesting to observe that IL-6 + HGF- and TNF{alpha} + HGF-mediated cell proliferation and growth of ectopic endometrial stromal cells were more resistant to suppression than those of eutopic endometrium after application of anti-HGF antibody (Figure 7A and B).


    Discussion
 Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 Acknowledgements
 References
 
We demonstrated that, besides regulation of HGF by isolated macrophages under the stimulation by lipopolysaccharide in the pelvic microenvironment (Khan et al., 2005aGo), HGF could also be regulated in endometrial stromal cells in response to other inflammatory mediators, such as IL-6 and TNF{alpha}. We report here that the production of HGF was significantly elevated in the conditioned medium of stromal cells derived from women with endometriosis compared with women without endometriosis under the stimulation of IL-6 and TNF{alpha} either alone or in combination. This interaction between HGF and cytokines at the cellular level was in accordance with their significant correlation in the peritoneal fluid of women with endometriosis in the pelvic environment.

We further demonstrated in the present study that the increased secretion of HGF in media conditioned by stromal cells and its corresponding immunolocalization in isolated stromal cells and in intact tissue paralleled with their increased transcriptional activity at the mRNA level of HGF and its receptor, c-Met, under the similar stimulation. There was a decreasing tendency of HGF secretion in the culture medium after application of antibody against IL-6 or TNF{alpha}. In addition, IL-6- and TNF{alpha}-mediated stromal cell proliferation was also suppressed by anti-HGF antibody. This indicates that the production of HGF is mediated by either IL-6 or TNF{alpha} and that the growth-promoting effect of endometrial and endometriotic stromal cells may also be contributed by HGF itself in addition to IL-6 and TNF{alpha}.

Sugawara et al. (1997)Go previously reported significantly increased secretion of HGF by non-treated eutopic endometrial stromal cells in women with endometriosis compared with that in women without endometriosis. We also found an increasing tendency of HGF secretion by basal (non-treated) endometrial stromal cells derived from women with endometriosis compared with that from women without endometriosis. Although we expected a similarly significant difference between them, this elevation of HGF in the culture medium did not reach statistical significance. There are some reasons for the discrepancy of HGF secretion between these two studies. First, we cultured stromal cells for a period of 24 h instead of 48 h and in serum-free media, whereas Sugawara et al. (1997)Go used 2.5% fetal bovine serum in their culture media. Secondly, the detection limit of HGF was 0.1 ng/ml (100 pg/ml) in their study and <40 pg/ml in our assay system. The intra-assay and inter-assay coefficients of variation were both <10% for our assay. Irrespective of the difference in these two assay systems, we can at least postulate that endometrial stromal cells of endometriosis are biologically more active than those of non-endometriosis.

If we consider that pelvic endometriotic lesions are bathed in the peritoneal fluid, it is very logical to assume that peritoneal fluid enriched with different cytokines and growth factors, including HGF, IL-6 and TNF{alpha}, could stimulate the growth and progression of endometriosis. In fact, different studies have demonstrated that peritoneal fluid of women with endometriosis contains greater concentrations of HGF, IL-6 and TNF{alpha} than that of women without endometriosis (Harada et al., 2001Go; Khan et al., 2002Go). The origin of HGF, IL-6 and TNF{alpha} in the pelvic environment could be endometrial cells or mesothelial cells (Song et al., 2003Go; Ishimaru et al., 2004Go). In addition, the origin of these macromolecules could also be macrophages that have infiltrated into ectopic endometrium and corresponding eutopic endometrium (Khan et al., 2004bGo).

HGF has diverse biological effects, ranging from cell mitosis, angiogenesis, migration and metaplastic transformation to invasion of cells into pelvic mesothelium. Therefore, we believe that the regulation of HGF by IL-6 and TNF{alpha} in endometrial stromal cells has some biological significance. HGF could be involved in the regeneration of endometrium after menstruation by its mitogenic activity, or in the retrograde migration of menstrual debris by its mitogenic activity. Recently, Yoshida et al. (2004) demonstrated that the HGF/c-Met system plays a role in the pathogenesis of endometriosis by its cell invasion activity after regurgitated endometrial cells enter the pelvic microenvironment. In fact, we observed in the present study that HGF, alone or in combination with IL-6 or TNF{alpha}, was able to induce more proliferation of ectopic stromal cells when compared with eutopic stromal cells. This additional cell growth and cellular resistance to growth suppression by antibody against HGF in pelvic endometriotic lesions could be responsible for the persistence or progression of pelvic endometriosis.

The exact mechanism of this additional cell growth by ectopic stromal cells is unknown. However, according to some recent publications ectopic tissue differs from eutopic endometrium by its proliferation rate, steroid hormone levels and markers of apoptosis (McLaren et al., 1997Go; Beliard et al., 2004Go). Therefore, our present findings of a greater increase in cell proliferation by ectopic stromal cells could be explained by increased sensitivity to proliferation and their resistance to apoptosis compared with similar cells of eutopic endometrium. This could promote the dissemination and implantation of these cells to ectopic sites. Despite this biological difference between ectopic and eutopic endometrium, we did not find any influence of the phase of the menstrual cycle in the proliferation of ectopic tissues.

The autocrine and paracrine regulation of HGF between cells of mesenchymal origin and epithelial cells or within the same endometriotic stromal cells has been reported in several studies (Yoshida et al., 2004Go; Khan et al., 2005aGo). We also recently reported mRNA expression of HGF and its receptor, c-Met, in infiltrated macrophages derived from women with endometriosis and we established that there is an autocrine, paracrine or intracrine relationship among endometrial epithelial cells, stromal cells and infiltrated macrophages (Khan et al., 2005aGo). However, information regarding the regulation of HGF by IL-6 and TNF{alpha} in endometrial stromal cells has been lacking.

An interaction between TNF{alpha} and IL-6 with the consequent up-regulation of either the IL-6 or the IL-8 gene and protein and their involvement in the proliferation of endometriotic stromal cells derived from chocolate cyst has been reported (Iwabe et al., 2000Go). However, there is no study on the regulation of HGF by IL-6 and TNF{alpha} in women with pelvic endometriosis. Therefore, we demonstrated for the first time, using endometrial stromal cells derived from women with pelvic endometriosis, that IL-6 and TNF{alpha} either alone or in combination can effectively induce the expression or production of HGF at both gene and protein levels.

In our recent study (Khan et al., 2005aGo) we reported that, as an initial inflammatory mediator, lipopolysaccharide could regulate the production of HGF by macrophages and that this could enhance the proliferation of endometrial epithelial cells, stromal cells and also infiltrated macrophages. We further demonstrated that exogenous HGF, either alone or in combination with IL-6 or TNF{alpha}, was able to significantly induce the proliferation of endometrial stromal cells derived from women with pelvic endometriosis. This mitogenic activity of HGF was higher in women with endometriosis than in women without endometriosis. Besides other cytokines and growth factors, the contribution of HGF in this enhanced proliferation of stromal cells was confirmed by the decreasing tendency of stromal cell proliferation after application of anti-HGF antibody.

We conclude that, in addition to its production by inflammatory cells or other mesenchymal cells, HGF may also be produced by endometrial stromal cells derived from women with pelvic endometriosis; its regulation can be mediated by IL-6 or TNF{alpha}, which are constant components of peritoneal fluid in the pelvic microenvironment; and their orchestrated effect may be involved in the development of endometriosis.


    Acknowledgements
 Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 Acknowledgements
 References
 
We thank Miss Kazumi Hayashida and Miss Kyoko Ishida, Department of Obstetrics and Gynecology, Nagasaki University Graduate School of Biomedical Sciences, Nagasaki, Japan, for their excellent technical assistance.


    References
 Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 Acknowledgements
 References
 
Akoum A, Jolicoeur C and Boucher A (2000) Estradiol amplifies interleukin-1-induced monocyte chemotactic protein-1 expression by ectopic endometrial cells of women with endometriosis. J Clin Endocrinol Metab 85,896–904.[Abstract/Free Full Text]

American Society for Reproductive Medicine (1997) Revised American Society for Reproductive Medicine classification of endometriosis: 1996. Fertil Steril 167,817–821.[CrossRef]

Beliard A, Noel A and Foidart JM (2004) Reduction of apoptosis and proliferation in endometriosis. Fertil Steril 82,80–85.[CrossRef][ISI][Medline]

Buyalos RP, Funari VA, Azziz R, Watson JM and Martinez-Maza O (1992) Elevated interleukin-6 levels in peritoneal fluid of patients with pelvic pathology. Fertil Steril 58,302–306.[ISI][Medline]

Donnez J, Smoes P, Gillerot S, Casanas-Roux F and Nissole M (1998) Vascular endothelial growth factor in endometriosis. Hum Reprod 13,1686–1690.[Abstract]

Eisermann J, Gast MJ, Pineda J, Odem RR and Collins JL (1988) Tumor necrosis factor in peritoneal fluid of women undergoing laparoscopic surgery. Fertil Steril 50,573–579.[ISI][Medline]

Fujishita A, Hasuo A, Khan KN, Masuzaki H, Nakashima H and Ishimaru T (1999) Immunohistochemical study of angiogenic factors in endometrium and endometriosis. Gynecol Obstet Invest 48 (Suppl 1),36–44.

Harada T, Iwabe T and Terakawa N (2001) Role of cytokines in endometriosis. Fertil Steril 76,1–10.[CrossRef][ISI][Medline]

Ishimaru T, Khan KN, Fujishita A, Kitajima M and Masuzaki H (2004) Hepatocyte growth factor may be involved in cellular changes to the peritoneal mesothelium adjacent to pelvic endometriosis. Fertil Steril 81 (Suppl 1), 810–818.[CrossRef]

Iwabe T, Harada T, Tsudo T, Nagano Y, Yoshida S, Tanikawa M and Terakawa N (2000) Tumor necrosis factor-{alpha} promotes proliferation of endometriotic stromal cells by inducing interleukin-8 gene and protein expression. J Clin Endocrinol Metab 85,824–829.[Abstract/Free Full Text]

Khan KN, Masuzaki H, Fujishita A, Hamasaki T, Kitajima M, Hasuo A and Ishimaru T (2002) Association of interleukin-6 and estradiol with hepatocyte growth factor in peritoneal fluid of women with endometriosis. Acta Obstet Gynecol Scand 81,764–771.[CrossRef][ISI][Medline]

Khan KN, Masuzaki H, Fujishita A, Kitajima M, Sekine I and Ishimaru T (2003) Immunoexpression of hepatocyte growth factor and c-Met receptor in eutopic endometrium predicts the activity of ectopic endometrium. Fertil Steril 79,173–181.[CrossRef][ISI][Medline]

Khan KN, Masuzaki H, Fujishita A, Kitajima M, Sekine I and Ishimaru T (2004a) Higher activity by opaque endometriotic lesions than non-opaque lesions. Acta Obstet Gynecol Scand 83,375–382.[CrossRef][ISI][Medline]

Khan KN, Masuzaki H, Fujishita A, Kitajima M, Sekine I and Ishimaru T (2004b) Differential macrophage infiltration in early and advanced endometriosis and adjacent peritoneum. Fertil Steril 81,652–661.[CrossRef][ISI][Medline]

Khan KN, Masuzaki H, Fujishita A, Kitajima M, Kohno T, Sekine I, Matsuyama T and Ishimaru T (2005a) Regulation of hepatocyte growth factor by basal and stimulated macrophages in women with endometriosis. Hum Reprod 20,49–60.[Abstract/Free Full Text]

Khan KN, Masuzaki H, Fujishita A, Kitajima M, Sekine I, Matsuyama T and Ishimaru T (2005b) Estrogen and progesterone receptor expression in macrophages and regulation of hepatocyte growth factor by ovarian steroids in women with endometriosis. Hum Reprod 20,2004–2013.

McLaren J, Prentice A, Charnock-Jones DS, Sharkey AM and Smith SK (1997) Immunolocalization of the apoptosis regulating proteins Bcl-2 and Bax in human endometrium and isolated peritoneal fluid macrophages in endometriosis. Hum Reprod 12,146–152.[CrossRef][ISI][Medline]

Osteen KG, Hill GA, Hargrove JT and Gorstein F (1989) Development of a method to isolate and culture highly purified populations of stromal and epithelial cells from human endometrial biopsy specimens. Fertil Steril 52,965–972.[ISI][Medline]

Osuga Y, Tsutsumi O, Okagaki R, Takai Y, Fujimoto A, Suenaga A et al. (1999) Hepatocyte growth factor concentrations are elevated in peritoneal fluid of women with endometriosis. Hum Reprod 14,1611–1613.[Abstract/Free Full Text]

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.[CrossRef][ISI][Medline]

Straphy JH, Molgaard CA and Coulman CB (1982) Endometriosis and infertility: a laparoscopic study of endometriosis among fertile and infertile women. Fertil Steril 38,667–672.[ISI][Medline]

Sugawara J, Fukaya T, Murakami T, Yoshida H and Yajima A (1997) Increased secretion of hepatocyte growth factor by eutopic endometrial stromal cells in women with endometriosis. Fertil Steril 68,468–472.[CrossRef][ISI][Medline]

Yoshida S, Harada T, Mitsunari M, Iwabe T, Sakamoto Y, Tsukihara S, Iba Y, Horie S and Terakawa N (2004) Hepatocyte growth factor/met system promotes endometrial and endometriotic stromal cell invasion via autocrine and paracrine pathways. J Clin Endocrinol Metab 89,823–832.[Abstract/Free Full Text]

Zarnegar R (1995) Regulation of HGF and HGFR gene expression. Epithelial-mesenchymal Interact Cancer 33–49.

Submitted on December 9, 2004; resubmitted on May 9, 2005; accepted on May 19, 2005.





This Article
Abstract
Full Text (PDF )
All Versions of this Article:
20/10/2715    most recent
dei156v1
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Request Permissions
Google Scholar
Articles by Khan, K. N.
Articles by Ishimaru, T.
PubMed
PubMed Citation
Articles by Khan, K. N.
Articles by Ishimaru, T.