Results
Growth and Cyst Formation of Endometriotic Lesions
High-resolution ultrasound imaging showed that uterine tissue samples, which were sutured to the abdominal wall of vehicle-treated and resveratrol-treated animals, exhibited a comparable initial size of ~1 mm³, providing standardized conditions for further analyses of growth and cyst formation in developing endometriotic lesions (Fig. 1C). Throughout the further time course of the experiment, the size of lesions from the control group progressively increased, finally presenting with a volume of ~3 mm³ at Day 28 (Fig. 1A, C and D). This was associated with the growth of the stromal tissue fraction (Fig. 1E and F) and endometrial cysts (Fig. 1G and H).
(Enlarge Image)
Figure 1.
(A and B) High-resolution ultrasound imaging of developing endometriotic lesions (borders marked by a red broken line, cysts marked by a yellow broken line) 28 days after transplantation of uterine tissue samples to the abdominal wall of BALB/c mice. The animals received either vehicle (A) or 40 mg/kg resveratrol (B). Scale bars: 1 mm. (C–H) Overall lesion volume (C, mm³), lesion growth (D, %), stromal tissue volume (E, mm³), stromal tissue growth (F, %), cyst volume (G, mm³) and fraction of cyst-containing lesions (H, %) of BALB/c mice that received either vehicle (control, white bars; n = 10) or 40 mg/kg resveratrol (black bars; n = 10) throughout an observation period of 28 days. Mean ± SEM; *P < 0.05 versus control (unpaired Student's t-test; Mann–Whitney rank sum test); P < 0.05 versus Day 0 (ANOVA followed by Dunnett post hoc test).
In contrast, resveratrol treatment markedly suppressed the development of endometriotic lesions. In fact, at Day 28 they still exhibited a lesion volume, which did not differ from that of Day 0 (Fig. 1B–D). This was due to the regression of the stromal tissue. Accordingly, the stromal tissue volume and the growth of these lesions were significantly reduced between Days 14–28 when compared with controls (Fig. 1E and F). In addition, although the fraction of cyst-containing lesions was comparable in both experimental groups (Fig. 1H), endometrial cysts in resveratrol-treated lesions presented with lower volumes until the end of the experiment (Fig. 1G). In general, there was no resveratrol-treated or control animal that exhibited complete regression of a lesion.
Histological analyses at Day 28 proved that the transplanted uterine tissue samples of both groups had developed to typical endometriotic lesions consisting of cyst-like dilated endometrial glands and vascularized stroma, independently of their localization at the peritoneal wall or in the mesentery (Fig. 2A–D). However, peritoneal and mesenteric lesions of resveratrol-treated animals exhibited a markedly decreased size when compared with vehicle-treated controls (Fig. 2A–D). This was confirmed by quantitative measurements of lesion sizes by means of a caliper (Fig. 2E). More detailed histomorphometric analyses revealed that the reduced lesion sizes in resveratrol-treated animals were caused by a lower stromal tissue volume and cyst volume (Table I). Of interest, volumes measured by histology were markedly underestimated when compared with the ultrasound and caliper measurements, because the volume of endometriotic lesions drastically decreases during the processing of tissue samples for histology (Laschke et al., 2010).
(Enlarge Image)
Figure 2.
(A–D) Hematoxylin–eosin-stained cross sections of endometriotic lesions (borders marked by broken line) at Day 28 after surgical induction by fixation of uterine tissue samples to the peritoneal wall (A and B) or the intestinal mesentery (C and D) of BALB/c mice that received vehicle (control; A and C) or 40 mg/kg resveratrol (B and D). The lesions show a typical histomorphology with endometrial glands surrounded by stromal tissue. Note that resveratrol-treated lesions exhibit a markedly reduced size when compared with controls. Scale bars: 400 µm. (E) The size (mm²) of peritoneal and mesenteric endometriotic lesions of control (white bars; n = 10) and resveratol-treated BALB/c mice (black bars; n = 10), as assessed by caliper measurement at Day 28. Mean ± SEM; *P < 0.05 versus control (unpaired Student's t-test).
Apoptosis and Cell Proliferation in Endometriotic Lesions
Because we did not detect any cleaved caspase-3-positive apoptotic cells in endometriotic lesions of both groups at Day 28 (data not shown), we speculated that the observed differences in lesion size were primarily caused by a reduced cell proliferation in resveratrol-treated animals. To test this hypothesis, tissue sections were additionally stained for the proliferation markers PCNA and Ki67. By this, we found that treatment with resveratrol resulted in a significantly reduced number of PCNA-positive stromal cells in peritoneal and mesenteric lesions when compared with controls (Fig. 3A–E), whereas PCNA expression in glandular cells did not show marked differences between the two groups (Fig. 3A–D and F).
(Enlarge Image)
Figure 3.
(A–D) Immunohistochemical staining of endometriotic lesions at Day 28 after surgical induction by fixation of uterine tissue samples to the peritoneal wall (A and B) or the intestinal mesentery (C and D) of BALB/c mice, which received vehicle (control; A and C) or 40 mg/kg resveratrol (B and D). Sections were stained with an antibody against PCNA for the detection of proliferating cells in the endometrial stroma (arrows) and glands (arrowheads). Note that resveratrol-treated lesions exhibit a reduced number of PCNA-positive stromal cells (arrows) when compared with controls. Scale bars: 50 µm. (E and F) PCNA-positive (%) stromal (E) and glandular (F) cells of peritoneal and mesenteric endometriotic lesions of control (white bars; n = 10) and resveratrol-treated BALB/c mice (black bars; n = 10). Mean ± SEM; *P < 0.05 versus control (unpaired Student's t-test; Mann–Whitney rank sum test).
As expected, numbers of Ki67-positive cells in endometriotic lesions were substantially lower (Fig. 4A–F), because Ki67 is known to be a more specific marker of proliferation than PCNA (Kordek et al., 1996). Ki67 staining of peritoneal and mesenteric lesions revealed a comparable number of positive stromal cells in both groups (Fig. 4A–E). In contrast, numbers of Ki67-positive glandular cells were significantly decreased in resveratrol-treated animals when compared with controls (Fig. 4A–D and F).
(Enlarge Image)
Figure 4.
(A–D) Immunohistochemical cross sections of endometriotic lesions at Day 28 after surgical induction by fixation of uterine tissue samples to the peritoneal wall (A and B) or the intestinal mesentery (C and D) of BALB/c mice that received vehicle (control; A and C) or 40 mg/kg resveratrol (B and D). Sections were stained with an antibody against Ki67 for the detection of proliferating cells in the endometrial stroma (arrows) and glands (arrowheads). Note that resveratrol-treated lesions exhibit a reduced number of Ki67-positive glandular cells (arrowheads) when compared with controls. Scale bars: 50 µm. (E and F) Ki67-positive (%) stromal (E) and glandular (F) cells of peritoneal and mesenteric endometriotic lesions of control (white bars; n = 10) and resveratrol-treated BALB/c mice (black bars; n = 10). Mean ± SEM; *P < 0.05 versus control (unpaired Student's t-test).
Vascularization of Endometriotic Lesions
Treatment with resveratrol effectively inhibited angiogenesis in peritoneal and mesenteric endometriotic lesions. This was indicated by a significantly reduced density of CD31-positive microvessels at Day 28 (Fig. 5A–E). More detailed immunohistochemical analyses showed that this was caused by a reduced proliferating activity of the microvascular endothelium in resveratrol-treated lesions, exhibiting less PCNA-positive endothelial cells when compared with controls (Fig. 6A–E).
(Enlarge Image)
Figure 5.
(A–D) Immunofluorescent detection of microvessels (arrows) within endometriotic lesions at Day 28 after surgical induction by fixation of uterine tissue samples to the peritoneal wall (A and B) or the intestinal mesentery (C and D) of BALB/c mice that received vehicle (control; A and C) or 40 mg/kg resveratrol (B and D). Sections were stained with Hoechst 33 342 to identify cell nuclei (blue) and an antibody against CD31 for the detection of the microvascular endothelium (red). Note that resveratrol-treated lesions exhibit a reduced number of microvessels when compared with controls. Scale bars: 50 µm. (E) Microvessel density (mm²) of peritoneal and mesenteric endometriotic lesions of control (white bars; n = 10) and resveratrol-treated BALB/c mice (black bars; n = 10). Mean ± SEM; *P < 0.05 versus control (unpaired Student's t-test).
(Enlarge Image)
Figure 6.
(A–D) Immunofluorescent sections of endometriotic lesions at Day 28 after surgical induction by fixation of uterine tissue samples to the peritoneal wall (A and B) or the intestinal mesentery (C and D) of BALB/c mice that received vehicle (control; A and C) or 40 mg/kg resveratrol (B and D). Sections were stained with Hoechst 33 342 to identify cell nuclei (blue), an antibody against CD31 for the detection of the microvascular endothelium (green) and an antibody against the proliferation marker PCNA (red). Note that microvessels in resveratrol-treated lesions exhibit a reduced number of PCNA/CD31-positive proliferating endothelial cells (arrows) when compared with controls. Scale bars: 25 µm. (E) PCNA/CD31-positive cells (%) in peritoneal and mesenteric endometriotic lesions of control (white bars; n = 10) and resveratrol-treated BALB/c mice (black bars; n = 10). Mean ± SEM; *P < 0.05 versus control (unpaired Student's t-test).