Doxycycline reduces MMP-2 activity and inhibits invasion of 12Z epithelial endometriotic cells as well as MMP-2 and -9 activity in primary endometriotic stromal cells in vitro

This study was conducted on an immortalized epithelial endometriotic cell line (12Z) and on two primary stromal cells freshly isolated from endometriotic tissue as described below. Mycoplasma testing of the cultured cells was carried out on a regular basis during the study and was repeatedly negative. The primary cell study was approved by the institutional review board and individual informed consent was obtained by the participants. Only early passage numbers were used for the experiments (in 12Z: passage 2–5 of the initially received cell line, in primary stromal cells: passage 2–3 of freshly isolated cells).

The 12Z endometriotic cell line was kindly provided by Professor Anna Starzinski-Powitz (Department of Biology, University of Frankfurt, Frankfurt am Main, Germany). The cell line was generated through the in situ electroporation of primary peritoneal epithelial endometriotic cells with SV-40 T-antigen. The characteristics of this cell line have been previously described [18]. 12Z cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM, 21980; Invitrogen, Basel, Switzerland) containing 10% fetal bovine serum (Gibco, LifeTechnologies, Waltham, MA, USA) and 1% antibiotic-antimycotic (containing penicillin, streptomycin and amphotericin B, Gibco) at 37 °C in an atmosphere with 5% CO₂ and 95% humidity. Experiments were performed with a cell confluency of 50–70%.

Primary cells were isolated from histologically confirmed endometriotic tissue of two different patients by the following procedure: Endometriotic tissue samples from endometriomas were collected freshly in the operating room and transferred directly to the cell culture lab for further processing. The diagnosis of endometriosis was confirmed histologically by a pathologist. The freshly collected endometriotic tissue was separated from any connective tissue and then minced in fine slices and incubated for 90 min in a 10 mg/ml collagenase solution at 37 °C (Collagenase C2674, Sigma-Aldrich, Saint Louis, MO, USA). After this digestion procedure, the suspension was passed through a 100-μm mesh filter to separate the cells from any remaining connective tissue. The filtrate was then collected and passed through a second 40-μm mesh filter. The flow through, consisting of the stromal endometriotic cells, was transferred to a plate and cultured under standard culturing conditions in Iscoves modified medium (IMEM, Gibco) supplemented with 10% fetal bovine serum and 1% antibiotic-antimycotic (containing penicillin, streptomycin and amphotericin B, Gibco) in an incubator at 37 °C in an atmosphere of 5% CO₂ and 95% humidity. For the drug treatment experiments cell confluency was 80% at the time point of treatment. Primary epithelial endometriotic cells were not kept in culture since low cell quantity and proliferation rate did not allow an adequate MMP quantification in these cells. As part of a parallel experiment and confirmation of stromal origin of the cells we performed three-dimensional (3D) cell cultures with the primary stromal cells (not shown). The cells aggregated to 3D-cell-conglomerates and we then produced histological slides from the microtissues, which were stained immunohistochemically for pan-cytokeratin, vimentin, and mib-1. Vimentin showed a general strong positive staining and pan-cytokeratin stained ubiquitously negative. Mib-1 stained positive in a few cells. In addition, some brown pigment was found which most likely corresponds to blood residues.

For the drug treatment experiments, 12Z or primary stromal endometriotic cells were seeded in 6-well plates at a density of 3 × 10⁵ cells per well for 12Z and 6 × 10⁵ cells per well for primary stromal cells. The following day, the medium was replaced by serum-free medium after washing the cells once with phosphate-buffered saline. After 24 h the medium was replaced with 2 ml of serum-free medium containing drugs in the indicated doses. Serum-free medium was used to avoid foreign protease activity in the serum as has been done by others [19] and no signs of reduced cell viability were found as described below. Doxycycline (doxycycline hyclate, Sigma-Aldrich) was purchased and stock solutions were diluted in sterile water to a concentration of 1 mg/ml and kept in the dark at − 20 °C. Treatment doses were chosen according to cell culture doses for human cell lines reported in the literature [20]. Progesterone (Sigma-Aldrich) was stored at − 20 °C as a stock solution in DMSO at a concentration of 100 mM. Drugs were freshly diluted from the stock for every treatment. Controls were treated with the vehicle substance in the same concentration.

Cell viability after treatment with doxycycline was assessed in parallel to the drug treatment experiments described above with an automated cell-counting device (Countess®, Invitrogen, LifeTechnologies, Thermo Fisher Scientific Inc., Waltham, MA, USA) after trypan blue staining of the cells to distinguish viable from dead cells. After the 24 h doxycycline treatment, the supernatants were collected and adherent cells were trypsinized and resuspended in the respective supernatants. Ten microliters of the cell suspension were mixed 1:1 with trypan blue staining and filled into the counting chamber. The number and fraction of viable cells was determined for each plate and calculated as percentage of the number of viable cells in the control plate.

Gelatin zymography was used to detect gelatinases in the supernatants of treated cells as has been done by others [19]. Briefly, supernatants of treated cells (a total of 25 μl per well containing v/v 1:1 of supernatant and Laemmli buffer) were loaded on a gelatin (type A, G-8150, Sigma-Aldrich) containing SDS polyacrylamide gels (acrylamide concentration of 8%) followed by electrophoresis for 6 h. SDS and Triton X-100 were then washed out of the gels to allow a renaturation of the MMP proteins. This was followed by incubation for 24 h (if not specified as prolonged incubation of 36 h) in a zinc- and calcium-chloride-containing buffer at 37 °C, allowing gelatin degradation by gelatinases. Gels were stained with Coomassie blue for 2 h and then destained for 45 to 90 min until bands representing gelatin degradation by the gelatinases became visible. Reconstituted lyophilized human pro-MMP-2 and pro-MMP-9 were used as positive controls (contained in Biotrak MMP-2 and MMP-9 activity assays kits, GE Healthcare). The gelatin zymograms were quantified with the Image-J software (imagej.​nih.​gov/​ij).

Levels of the active form of MMP-2 and MMP-9 were measured by an activity assay (Biotrak MMP-2 activity assay and Biotrak MMP-9 activity assay, GE Healthcare) according to the provided manufacturer’s protocol. Briefly, supernatants of treated cells were added to the anti-MMP-2 (respectively anti-MMP-9) precoated microplate wells and incubated overnight at 4 °C in accordance with the manufacturer’s instructions. The following day, the plates were rinsed 4 times with the provided washing buffer and then incubated with the assay buffer in addition to the substrate and detection reagent. The plates were analyzed immediately after the addition of the detection reagent (a dection enzyme that is activated by captured active MMP-2 (respectively MMP-9) through a single proteolytic event, which can be measured using a specific chromogenic peptide substrate) for the baseline measurement and again after 3-6 h for signal recording in a spectrophotometer (Epoch Microplate Spectrophotometer, BioTek, Winooski, VT, USA). Standard curves were obtained for each activity assay and were similar to the standard curve provided by the manufacturer.

The matrigel invasion assay was carried out in 24-well matrigel invasion chambers (BD BioCoat BD Matrigel Invasion Chamber, Franklin Lakes, New Jersey 07417–1880). In brief, 12Z endometriotic cells were seeded in invasion chambers (2.5 × 10⁴ cells per 24-well invasion chamber) in serum-free medium containing the indicated doses of doxycycline. Serum-containing medium (DMEM containing 10% fetal bovine serum) was added to the lower chamber as a chemoattractant and the chambers were placed in an incubator at 37 °C and 5% CO₂ for 22 h. The cells that did not migrate through the matrigel membrane were subsequently removed with a cotton swab and the cells at the bottom of the membrane were fixated, stained with crystal violet and counted manually by bright field microscopy on an inverted microscope (Leica DMI6000B).

Doxycycline treatment of 12Z epithelial endometriotic cells reduced the detected activity of MMP-2 in a dose-dependent manner in the MMP-2 activity assay (Fig. 1a). To exclude cell mortality as a reason for reduced MMP-2 secretion, we performed viability assays. We observed no significant cell mortality up to doses of 20 μg/ml doxycycline, whereas 40 μg/ml doxycycline led to a reduced number of viable cells (65% of the control, Fig. 1b). This means that the reduced levels of MMP-2 up to 20 μg/ml were not due to reduced cell numbers or increased mortality of endometriotic cells. Reduced expression of latent MMP-2 (pro-MMP-2) was found in a zymography assay upon treatment with doxycycline (Fig. 1c and d).

Primary endometriotic stromal cells were obtained from two different patients and extracted from endometriotic tissue issuing from deep-ovarian endometriomas as described above. Analysis of the presence of gelatinases by zymography revealed high levels of latent MMP-2 (pro-MMP-2) and moderate levels of latent MMP-9 (pro-MMP-9) in the supernatants of both primary stromal cell cultures (cell passage number: 2). Doxycycline treatment led to a dose-dependent reduction of MMP-2 and MMP-9 activity in the primary endometriotic cells (Fig. 2a and c) on cell passage number 3. A reduction of pro-MMP-2 and pro-MMP-9 levels in doxycycline-treated primary endometriotic stromal cells (cell passage number 3) was observed in zymography analyses (Fig. 2b, d and e). There were no microscopic signs of increased cell death and same cell counts in primary stromal endometriotic cells after treatment with doxycycline up to 10 μg/ml compared to the control, whereas some signs of increased cell death and reduced cell counts were observable in cells treated with 20 μg/ml and more pronounced with 40 μg/ml as shown in Fig. 2f.

Using matrigel-coated invasion chambers, we observed a reduced invasion of doxycycline treated 12Z endometriotic cells compared with the control. Treatment with doxycycline at a concentration of 1 μg/ml reduced the relative number of invading cells to 65% and 10 μg/ml reduced the invasive fraction to 22% of the control (Fig. 3a and b). EDTA 1 mM was used as a positive control (Fig. 3c).

To assess a possible synergistic effect of doxycycline with progesterone treatment, we treated 12Z with a combination of low-dose doxycycline (1 μg/ml) and progesterone (10 nM). Addition of doxycycline to progesterone further reduced levels of activated MMP-2 compared to progesterone or doxycycline alone (Fig. 4a). A further reduction of pro-MMP-2 levels in progesterone treated 12Z upon simultaneous treatment with doxycycline in a range of 1–40 μg/ml was observed especially at low doxycycline doses in zymography assays (Fig. 4b).