Expression of the gamma 2 chain of laminin-332 in eutopic and ectopic endometrium of patients with endometriosis

Endometriotic lesions (Ec) were removed from women (n = 25, aged 28–50 years, mean age 35.4 ± 4.8 years) undergoing laparoscopy for pain or infertility (Table 1). All the women had normal documented ovulatory cycles as well as normal endocrine parameters and did not receive hormone therapy or take oral contraception for at least 3 months before surgery. Simultaneously, eutopic endometrium was obtained from the same women (EuE+). Normal endometrial tissues (EuE-) were collected from healthy non-menopausal women (n = 27, aged 20–45 years, mean age 39.2 ± 7.9 years) with spontaneous, regular menstrual cycles (26–33 days) who were undergoing laparoscopic surgery for benign gynaecologic indications (tubal ligation, ovarian cystectomy or hysterectomy).

All endometrial biopsy samples were obtained with a Cornier Pipelle suction curette (C.C.D. International, Paris, France), which allows sampling of the functional layer of the endometrium. All samples were classified according to classical histologic criteria [10].

Patients provided informed consent, and the Institutional Review Board of the University of Liège approved the collection and use of human tissue.

For gene expression analysis, endometrial biopsy specimens were collected from healthy volunteers (EuE-: proliferative phase, n = 12; secretory phase, n = 15) and from endometriosis patients (EuE + and Ec, n = 9) (Table 1). After surgical resection, the tissue samples were immediately frozen in liquid nitrogen. The frozen tissues were processed as previously described [11]. The specific primers (Eurogentec, Liège, Belgium) for LAMC2 mRNA were 5′-AAAGCCACGTTGAGTCAGC-3′ (forward) and 5′-TCTTCCACCTGAAAGGACTGAT-3′ (reverse). The specific primers for 28S rRNA were 5′-GTTCACCCACTAATAGGGAACGTGA-3′ (forward) and 5′-GGATTCTGACTTAGAGGCGTTCAGT-3′ (reverse). RT-PCR was performed using 10-ng aliquots of cDNA, Taq polymerase (Takara, Shiga, Japan) and 5 pmol of each primer. The specific PCR products were resolved on 10% polyacrylamide gels (Bio-Rad) and analysed with a luminescent image analyser (LAS-4000, Fujifilm) after GelStar staining (Lonza Rockland, Inc., Rockland, ME). The LAMC2 mRNA levels were expressed as ratios of the 28S rRNA as previously reported [12].

Tissue samples were fixed in 4% formalin for 4–12 hours, embedded in paraffin and cut into 4-μm sections. The sections were mounted on SuperFrost Plus glass slides (Menzel-Gläser, Braunschweig, Germany), dewaxed in xylene, rehydrated and subsequently autoclaved for 11 min at 126°C and 1.4 bar in Target Retrieval Buffer (S2031 for laminin gamma 2 chain; DakoCytomation, Glostrup, Denmark) or treated with proteinase K (S3004 for cytokeratin 7 and S1699 for Ki-67; DakoCytomation, Glostrup, Denmark). Endogenous peroxidases were blocked by treatment with 3% H₂O₂/H₂O for 20 min, and non-specific binding was prevented by incubation in Universal Blocking Reagent (BioGenex, San Ramon, CA, USA) for 3 min. The sections were incubated with the following primary antibodies: laminin gamma 2 chain (Dako, M7262, diluted 1:25), cytokeratin 7 (BD Biosciences, 345779, ready-to-use) and Ki-67 (Dako, M7240, diluted 1:100). The sections were washed in PBS and subsequently incubated for 30 min with EnVision + HRP (K4001, Dako) or biotinylated goat anti-mouse antibodies (Dako E0433, diluted 1:400) followed by incubation with peroxidase-labelled streptavidin for 30 min (Dako P0397, diluted 1:500). Staining was detected with 3,3′-diaminobenzidine (DAB) chromogen. The nuclei were counterstained by incubation with haematoxylin for 2 min. The sections were mounted, examined and photographed. The negative control samples were processed by omitting the primary antibody or by incubating the sections with nonspecific IgG at the same concentration as the primary antibody. Placenta was used as a positive control.

Immunohistochemical staining analysis was semi-quantitative. The intensity of staining was characterised as follows: no staining (0), weak but detectable (1), strong (2) or very strong (3). The percentage of positive glands was graded as follows: no positive glands (0), less than 11% (1), 11-50% (2), 51-80% (3) or greater than 80% (4). The final score was calculated by multiplying the two scores.

The patient groups were compared using the Kruskal-Wallis test, and significant differences were further analysed via pairwise comparisons using the Mann–Whitney test. The results are presented as medians ± quartiles (25^th-75^th percentile). P values < 0.05 were considered statistically significant.

LAMC2 mRNA was detectable in the endometrium of women without endometriosis, and no differences were observed between the proliferative and secretory phases of the menstrual cycle (Figure 1A). When RT-PCR was performed in paired samples of ectopic and eutopic endometrium of women with endometriosis, an up to 3 fold increase of LAMC2 mRNA levels was detected in the ectopic endometrium (Figure 1B, P < 0.05). Similar LAMC2 mRNA levels were observed in the eutopic endometrium of women with and without endometriosis (Figure 1C).

Laminin gamma 2 expression was first investigated in ectopic endometrium (Figure 2A). Endometriotic lesions were confirmed and localised via cytokeratin 7 immunodetection, as illustrated in Figure 2B. To assess the proliferative status of the endometrial glands, Ki-67 immunoreactivity was also examined in serial sections, as illustrated in Figure 2C. Staining with antibody against laminin gamma 2 revealed a cytoplasmic expression pattern exclusively localised in epithelial cells. In endometriotic lesions, at a higher magnification, laminin gamma 2 also appeared in the cytoplasm of epithelial cells, either distributed along the basement membrane or with a more uneven distribution as illustrated in Figure 3A-C. In areas of desquamation, the epithelial cells of the dispersing glands displayed a highly intensified intracytoplasmic immunoreactivity (Figure 3A), whereas laminin gamma 2 was detected in a more linear pattern along the basement membranes of intact glands (Figure 3B). In pluristratified glands, laminin gamma 2 chain immunoreactivity was more irregular (Figure 3C).

Positive staining for the laminin gamma 2 chain was observed in epithelial basement membranes around individual glands and in the basement membranes underlying the endometrial surface epithelium (Figure 3D-F) in the eutopic endometrium of women with endometriosis. In normal eutopic endometrium, a similar cytoplasmic expression pattern was observed in the glandular epithelium; however, in a few cases, stronger expression was observed in the apical region of the epithelium (Figure 3G-I). Laminin gamma 2 was not observed in the stromal cells. There was no significant variation in immunoreactivity between the different menstrual phases.

To compare laminin gamma 2 expression in the ectopic and eutopic endometrium, we evaluated the percentage of positive glands in each tissue (Figure 4A). No differences were noted between eutopic and ectopic endometria from patients with endometriosis (76.4% ± 7.5% versus 88.3% ± 6.2%); however, in patients without endometriosis, the endometrium displayed the lowest percentage of laminin gamma 2 chain-expressing glands (30.1% ± 9.3%, P < 0.001). To better compare the intensity of laminin gamma 2 chain expression in different tissues, a semi-quantitative scoring method was performed using whole tissue sections as described above (see Methods). When we evaluated only the intensity of laminin gamma 2 staining in positive glands from the different tissues, no differences between the groups were observed; however, eutopic endometrium from women without endometriosis exhibited a null score more frequently than eutopic and ectopic endometrium from patients with endometriosis (Figures 3G and 4B). In other words, when glands were positive in EuE-, their intensity was lower than that of positive glands in EuE+. When a global scoring method was applied that considered both the number of positive glands and their intensity, no differences were observed between ectopic and eutopic endometrium in patients with endometriosis. However, endometrium from patients without endometriosis displayed a reduced global expression score compared either with eutopic endometrium from patients with endometriosis; either with ectopic endometrium (Figure 4C, P < 0.005).