Endometrial tissues were obtained from a total of 61 patients who had undergone gynecological operations in their reproductive ages. Thirty-two patients had laparoscopic and histological evidence of advanced stage endometriosis, while 29 patients with proven fertility without the disease served as controls. The number of 30 patients for each group was initially chosen to detect a difference of 4.0 in the secretory phase between the two groups, assuming that a standard deviation is 4.0 and about half of the subjects are in the secretory phase, with a power of 0.8 for p<0.05. In the endometriosis group, the extent of disease was staged according to the American Society for Reproductive Medicine (1997) and we divided the patients into 2 subgroups; those who presented with infertility and those who presented with problems other than infertility. Endometrial samples were obtained either by using a curette after pelviscopy or by opening the uterine cavity immediately after hysterectomy. The review board for human research of Seoul National University Hospital approved this project, and written informed consent for use of the endometrium was obtained from each woman. All sections were dated by a gynecological pathologist according to the histological criteria of Noyes (7) and were classified as early proliferative (days 4-7), mid proliferative (days 8-10), late proliferative (days 11-14), early secretory (days 15-18), mid secretory (days 19-22) or late secretory (days 23-28) phases.

All patients in the study had regular menstrual cycles and had not received any hormonal therapy during the previous 6 months. The clinical characteristics, the menstrual phases of each group, and the stages of endometriosis are summarized in Table 1. Patients with a previous history of infertility and histological or sonographic evidence of adenomyosis or submucous myoma, which might be associated with infertility, were excluded from the control group. The indications for surgery amongst the control group were benign ovarian cyst (n=14), carcinoma in situ of the uterine cervix (n=10), and uterine intramural leiomyoma (n=5). There were no significant differences in age between the two groups, but the average number of deliveries was significantly lower in the endometriosis group (p<0.05).

Endometrial tissues used for mRNA extraction were snapfrozen in liquid nitrogen and stored at -70℃ until analyzed. The tissues for immunohistochemical staining were immersed in OCT compound (Tissue Tek, Elkhart, IN, U.S.A.) and stored at -70℃ until analyzed. Total RNA was isolated using Trizol reagent (Life Technologies Inc, Rockville, MD, U.S.A.) according to the manufacturer's protocol and stored at -70℃. RT was performed using the Promega Reverse Transcription System (Promega, Madison, WI, U.S.A.). Real-time PCR was used for relative gene expression using an ABI Prism 7700 Sequence Detection System Thermal Cycler (Applied Biosystems, Foster City, CA, U.S.A.). Primers and probes were designed using Primer Express 1.5 Software and synthesized commercially (Applied Biosystems). The primer and probe sequences for angiogenin are shown in Table 2, and commercially synthesized primer and probe were used for glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (Human GAPDH Endogenous Control, Applied Biosystems).

The optimal concentrations of all primers and probes were determined according to the optimization protocol provided by Applied Biosystems. A stock placental cDNA pool was prepared as a standard and used for all real-time assays. PCR reaction mixes for each standard and samples were prepared in separate tubes, using TaqMan Universal PCR master mix, primers, probe and cDNA. All samples were assayed in triplicate, and 25 µL aliquots were transferred to wells of a MicroAmp optical 96-well reaction plate (Applied Biosystems). The thermocycler parameters were 50℃ for 2 min and 95℃ for 10 min, followed by 40 cycles of 95℃ for 15 sec and 60℃ for 1 min. A GAPDH pre-designed assay from Applied Biosystems was used to measure the expression of the housekeeping gene GAPDH in each sample using the same standard preparation. Angiogenin gene expression was normalized with GAPDH gene expression in each sample and the ratio between angiogenin and GAPDH was expressed as angiogenin mRNA level in all samples.

We selected 2 endometrial samples in each menstrual phase for the endometriosis and control group. Immunohistochemical staining was performed as previously described by Koga et al. (6). Briefly, 6 µm cryostat sections cut from frozen tissues were mounted on poly-L-lysine-treated slides, fixed in cold acetone at -20℃ for 10 min, and treated with 3% hydrogen peroxide for 10 min to eliminate endogenous peroxidases. The sections were incubated with mouse monoclonal antibody to human angiogenin (1:25; Biogenesis, Poole, U.K.) overnight at 4℃. Control slides were incubated with nonimmune mouse IgG, at a concentration adjusted to that of the primary antibody. The sections were incubated with biotinylated horse antimouse IgG, followed by avidin peroxidase and diaminobenzidine, using the Vectastain Elite ABC kit (Vector Laboratories, Inc., Burlingame, CA, U.S.A.). All sections were counterstained with hematoxylin. As the primary purpose of immunohistochemical staining was to localize the area of angiogenin expression in the control and endometriosis group, we did not perform semiquantitative comparison between the two groups. Assessments of immunostaining were based on agreement among three independent observers who were blinded as to which sections were from patients with endometriosis and which were from control patients.

Comparisons of age, number of deliveries, and angiogenin mRNA level between two groups were performed using the Mann-Whitney U test for nonparametric data. Angiogenin mRNA levels within the endometriosis or control group were compared among the 5 or 6 different menstrual phases using the Kruskal-Wallis test, followed by the Mann-Whitney U test with Bonferonni correction, if significant. Angiogenin mRNA levels within the control group were compared among the 3 different subgroups according to the diagnosis using the Kruskal-Wallis test. Results with p<0.05 were considered significant.