- Email: [email protected]
O9.4 Laser capture microdissection combined with microarray analysis as tools to identify unique antigens on blood vessels in endometriosis
Oral presentations~Seminar 9. Angiogenesis of the menstrual phase of the endometrium. Remarkably, glandular epithelial cells in the cyclic part of the endometrium also stained strongly positive for fit-4. This staining was largely confined to endometrium in the proliferative phase. Finally, large amounts of VEGF-C mRNA were demonstrated in the glandular epithelium, in the cyclic section of endometrium in the proliferative phase. Our results have demonstrated extensive lymphangiogenesis in endometrium in the proliferative phase. The formation of new lymphatic vessels most likely originates from the pre-existing vessels in the basal layer of the endometrium. The combined expression of fit-4 and VEGF-C mRNA in endometrium in the proliferative phase suggests a role for the VEGF-C/flt-4 pathway in endometrial lymphangiogenesis. Remarkably, the fact that both VEGF-C and its receptor fit-4 were abundantly expressed by glandular epithelial cells indicates that this growth factor/receptor may also be involved in the regeneration of glandular structures in endometrium in the proliferative phase, possibly via an auto- or paracrine mechanism. Our findings indicate that the glandular epithelium is a key player in the (re)construction of the endometrium after menstruation.
I-~
Paradoxical effects of progesterone and oestrogen on angiogenesis in mouse endometrium
J.E. Girling, L.M. Walter, RA.V~ Rogers. Monash University, Clayton,
Australia The process of angiogenesis (formation of new blood vessels from pre-existing vasculature) is fundamental to the regular growth and remodeling of the human endometrium. Abnormal angiogenesis may also contribute to endometriosis, with evidence of increased angiogenesis in eutopic endometrium from women with endometriosis in comparison to women without the disease. We are using recently developed mouse models for fundamental studies of endometrial angiogenesis and arteriogenesis (process whereby capillaries acquire a coat of vascular smooth muscle cells (VSMC)). The aims of this study were to quantify endometrial angiogenesis in pregnant mice and to investigate the role of progesterone in promoting endothelial cell (EC) proliferation in ovariectomized mice; we hypothesized that EC proliferation would increase with increasing plasma progesterone concentrations in early pregnant mice, that progesterone would stimulate EC proliferation in ovariectomized mice, but only following estrogen-priming, and that progesterone-induced EC proliferation would require vascular endothelial growth factor (VEGF). Uteri were collected on days 1 4 of pregnancy when circulating progesterone concentrations were increasing but prior to implantation. Before dissection, mice were injected with BrdU enabling proliferating EC (PEC) to be quantified with CD31/BrdU doubleimmunohistochemistry. There was a significant increase in PEC (KruskalWallis statistic (KW)=17.1, p=0.002) on day 3 of pregnancy (Day 1 2: no proliferation, Day 3:126.6±45.6 PEC/mm 2 (mean±SE)), when plasma progesterone also increased. To determine if this EC proliferation was due to progesterone, an experiment was performed on ovariectomized mice. One group was treated with a single oestradiol injection (100ng) on day eight after ovariectomy, followed by a no-treatment day and 3 consecutive daily injections of progesterone (1 mg). Other groups were treated with either the vehicle, oestradiol or progesterone injections only; all were dissected on day 13 following ovariectomy. Unexpectedly, mice treated with progesteroneonly had the highest amount of EC proliferation (114.7±30.9 PEC/mm2); estrogen priming significantly reduced EC proliferation (44.8± 15.5 PEC/mm 2; KW = 13.8, p=0.008). Additionally, an increase in the amount of Gt-smooth muscle actin (marker of both pericytes and VSMC) associated with vessels was observed (preliminary data only), suggesting that progesterone may also stimulate arteriogenesis. In a further experiment in which VEGF anti-serum was administered concurrently with the progesterone injections, EC proliferation was significantly reduced (progesterone only: 272.1±49.3 PEC/mm2; progesterone plus VEGF anti-serum: 144.8±81.3 PEC/mm 2, p=0.03). We concluded that oestrogen priming is not required for progesterone to stimulate VEGF-dependent endometrial EC proliferation and that it inhibits progesteroneinduced angiogenesis in ovariectomized mice.
I-~
Laser capture microdissection combined with microarray analysis as tools to identify unique antigens on blood vessels in endometriosis
A. Van Langendonckt 1 , C. Punyadeera 2, D. Gonzalez 1 , R. Kamps 2, G. Dunselman 2, J. Donnez 1 , R Groothuis 2. ]Universitd catholique de Louvain,
Brussels, Belgium; 2Maastricht University, Maastricht, The Netherlands Introduction: Rectovaginal endometriotic nodules are well-vascularized and innervated aggregates of fibrous and smooth muscle cells with islands or strands of glands and stroma. Current treatment options are associated with
S19 high morbidity and recurring disease. New options for diagnosis and treatment are warranted. The present study was initiated to search for specific markers on the vasculature in these nodules, which can be employed for the development of improved diagnostic tools and/or treatments. Materials and Methods: Rectovaginal nodules, vaginal tissue, myometrium and endometrium (proliferative, secretory and menstrual) were collected from five patients with endometriosis not receiving hormonal treatment. Biopsies were embedded in Tissue Teck, snap frozen and cut into serial sections. Forty consecutive sections were treated as follows: the first sections were immunostained with anti-CD31 (endothelial marker), anti-CD10 (stromal marker) and anti-Cytokeratin 22 antibodies (epithelial markers) to locate the vessels and the lesion. The subsequent two sections were used for laser capture microdissection (LCM). After microdissection, RNA was extracted and two rounds of linear amplification were performed. The quality of the cRNA was checked and cRNA were pooled as follows: 1) vessels isolated from the fibromuscular tissue of the nodule, 2) vessels from the endometrial stroma of the nodule, 3) vessels isolated from the control tissues (myometrium and endometrium) and 4) the whole tissue controls (vagina and endometrium). After pooling, the cRNA was labelled with biotin and was hybridized in triplicate to Affymetrix chips (HG-U133A2.0). Real time RT-PCR was performed to assess the degree of enrichment of vascular cells isolated by LCM and to validate the findings of the array hybridizations. Results: The RNA isolated from captured vessels was highly enriched for CD31 mRNA (endothelial cells) and s-smooth muscle actin mRNA (pericytes), when compared to surrounding tissue. The quality of the RNA was similar in all cRNA samples included in the pools. Array analysis revealed 105 genes overexpressed more than 10-fold in fibromuscular vessels when compared to the whole tissue control pool; 57 of these genes were also overexpressed more than 10-fold when compared to the vessel control pool. In vessels from endometrial stroma of the nodule, 79 genes were overexpressed more than 10-fold when compared to the whole tissue control pool; 51 of these genes also showed a 10 fold increase in expression when compared to the vessels control pool. Array results were confirmed on pools and individual samples included in the pools for 10 genes overexpressed more than 20 fold by quantitative PCR. Conclusion: We have designed a method to generate good quality cRNA from vessels from rectovaginal endometriotic lesions. Array and PCR results show differential gene expression in vessels isolated from the fibromuscular tissue and the stroma of those nodules, compared to vessels dissected from endometrium and myometrium. Selected markers will be further studied by immunohistochemistry.
I-~
Hypoxia may be an important regulator of disease progression in a mouse model of endometriosis
C.M.B. Becker 1 , T. Cramer 2, T. Funakoshi 1 , N. Rohwer 2, V~ Bernhardt 2, A.M. Treston3, J. Folkman 1 , R.J. D' Amatr. 1Childrens Hospital Boston,
Boston, USA; 2Charitd Campus Yirchow Berlin, Germany; 3EntreMed Inc., Rockville, USA Endometriosis, the presence of ectopic endometrial tissue, remains a poorly understood disease. While various models exist about its underlying pathogenesis, it is now widely accepted that angiogenesis, the growth of new blood vessels, plays an essential role in disease development. The well-characterized angiogenic stimulus, vascular endothelial growth factor (VEGF), has been found to be upregulated both in peritoneal fluid and endometriotic lesions. VEGF production is stimulated by a number of conditions including hypoxia. The transcription factor hypoxia-inducible factor (HIF) 1 alpha, heterodimerizes with HIF-1 beta during hypoxic conditions and subsequently transactivates a number of genes, including VEGF thereby promoting angiogenesis. In the current study, we utilize a well-established mouse model of surgically induced endometriosis to demonstrate an important role for hypoxia during the early stages of disease. Intraperitoneal injections into these mice with the hypoxia probe, pimonidazole hydrochloride (Hypoxyprobe-1), immediately prior to euthanasia demonstrated increased staining intensity in early endometriotic lesions as compared to more advanced lesions. To determine whether HIF-1 alpha expression correlated with disease progression as determined by increased lesion size, we performed immunohistochemistry using a HIF-1 alpha antibody on endometriotic lesions at various stages of disease. We found that HIF-1 alpha expression increases with severity of disease. In addition, Western blot analysis of endometriotic lesions from different stages of disease demonstrated that HIF-1 alpha expression was stable early in disease yet undetectable at later stages. To confirm that HIF- 1 alpha was functional in the early endometriotic lesions, we measured mRNA levels of selected HIF target genes in these tissues. We found that mRNA levels of VEGF, glucose transporter-1 and phosphoglycerate
I-~
Paradoxical effects of progesterone and oestrogen on angiogenesis in mouse endometrium
J.E. Girling, L.M. Walter, RA.V~ Rogers. Monash University, Clayton,
Australia The process of angiogenesis (formation of new blood vessels from pre-existing vasculature) is fundamental to the regular growth and remodeling of the human endometrium. Abnormal angiogenesis may also contribute to endometriosis, with evidence of increased angiogenesis in eutopic endometrium from women with endometriosis in comparison to women without the disease. We are using recently developed mouse models for fundamental studies of endometrial angiogenesis and arteriogenesis (process whereby capillaries acquire a coat of vascular smooth muscle cells (VSMC)). The aims of this study were to quantify endometrial angiogenesis in pregnant mice and to investigate the role of progesterone in promoting endothelial cell (EC) proliferation in ovariectomized mice; we hypothesized that EC proliferation would increase with increasing plasma progesterone concentrations in early pregnant mice, that progesterone would stimulate EC proliferation in ovariectomized mice, but only following estrogen-priming, and that progesterone-induced EC proliferation would require vascular endothelial growth factor (VEGF). Uteri were collected on days 1 4 of pregnancy when circulating progesterone concentrations were increasing but prior to implantation. Before dissection, mice were injected with BrdU enabling proliferating EC (PEC) to be quantified with CD31/BrdU doubleimmunohistochemistry. There was a significant increase in PEC (KruskalWallis statistic (KW)=17.1, p=0.002) on day 3 of pregnancy (Day 1 2: no proliferation, Day 3:126.6±45.6 PEC/mm 2 (mean±SE)), when plasma progesterone also increased. To determine if this EC proliferation was due to progesterone, an experiment was performed on ovariectomized mice. One group was treated with a single oestradiol injection (100ng) on day eight after ovariectomy, followed by a no-treatment day and 3 consecutive daily injections of progesterone (1 mg). Other groups were treated with either the vehicle, oestradiol or progesterone injections only; all were dissected on day 13 following ovariectomy. Unexpectedly, mice treated with progesteroneonly had the highest amount of EC proliferation (114.7±30.9 PEC/mm2); estrogen priming significantly reduced EC proliferation (44.8± 15.5 PEC/mm 2; KW = 13.8, p=0.008). Additionally, an increase in the amount of Gt-smooth muscle actin (marker of both pericytes and VSMC) associated with vessels was observed (preliminary data only), suggesting that progesterone may also stimulate arteriogenesis. In a further experiment in which VEGF anti-serum was administered concurrently with the progesterone injections, EC proliferation was significantly reduced (progesterone only: 272.1±49.3 PEC/mm2; progesterone plus VEGF anti-serum: 144.8±81.3 PEC/mm 2, p=0.03). We concluded that oestrogen priming is not required for progesterone to stimulate VEGF-dependent endometrial EC proliferation and that it inhibits progesteroneinduced angiogenesis in ovariectomized mice.
I-~
Laser capture microdissection combined with microarray analysis as tools to identify unique antigens on blood vessels in endometriosis
A. Van Langendonckt 1 , C. Punyadeera 2, D. Gonzalez 1 , R. Kamps 2, G. Dunselman 2, J. Donnez 1 , R Groothuis 2. ]Universitd catholique de Louvain,
Brussels, Belgium; 2Maastricht University, Maastricht, The Netherlands Introduction: Rectovaginal endometriotic nodules are well-vascularized and innervated aggregates of fibrous and smooth muscle cells with islands or strands of glands and stroma. Current treatment options are associated with
S19 high morbidity and recurring disease. New options for diagnosis and treatment are warranted. The present study was initiated to search for specific markers on the vasculature in these nodules, which can be employed for the development of improved diagnostic tools and/or treatments. Materials and Methods: Rectovaginal nodules, vaginal tissue, myometrium and endometrium (proliferative, secretory and menstrual) were collected from five patients with endometriosis not receiving hormonal treatment. Biopsies were embedded in Tissue Teck, snap frozen and cut into serial sections. Forty consecutive sections were treated as follows: the first sections were immunostained with anti-CD31 (endothelial marker), anti-CD10 (stromal marker) and anti-Cytokeratin 22 antibodies (epithelial markers) to locate the vessels and the lesion. The subsequent two sections were used for laser capture microdissection (LCM). After microdissection, RNA was extracted and two rounds of linear amplification were performed. The quality of the cRNA was checked and cRNA were pooled as follows: 1) vessels isolated from the fibromuscular tissue of the nodule, 2) vessels from the endometrial stroma of the nodule, 3) vessels isolated from the control tissues (myometrium and endometrium) and 4) the whole tissue controls (vagina and endometrium). After pooling, the cRNA was labelled with biotin and was hybridized in triplicate to Affymetrix chips (HG-U133A2.0). Real time RT-PCR was performed to assess the degree of enrichment of vascular cells isolated by LCM and to validate the findings of the array hybridizations. Results: The RNA isolated from captured vessels was highly enriched for CD31 mRNA (endothelial cells) and s-smooth muscle actin mRNA (pericytes), when compared to surrounding tissue. The quality of the RNA was similar in all cRNA samples included in the pools. Array analysis revealed 105 genes overexpressed more than 10-fold in fibromuscular vessels when compared to the whole tissue control pool; 57 of these genes were also overexpressed more than 10-fold when compared to the vessel control pool. In vessels from endometrial stroma of the nodule, 79 genes were overexpressed more than 10-fold when compared to the whole tissue control pool; 51 of these genes also showed a 10 fold increase in expression when compared to the vessels control pool. Array results were confirmed on pools and individual samples included in the pools for 10 genes overexpressed more than 20 fold by quantitative PCR. Conclusion: We have designed a method to generate good quality cRNA from vessels from rectovaginal endometriotic lesions. Array and PCR results show differential gene expression in vessels isolated from the fibromuscular tissue and the stroma of those nodules, compared to vessels dissected from endometrium and myometrium. Selected markers will be further studied by immunohistochemistry.
I-~
Hypoxia may be an important regulator of disease progression in a mouse model of endometriosis
C.M.B. Becker 1 , T. Cramer 2, T. Funakoshi 1 , N. Rohwer 2, V~ Bernhardt 2, A.M. Treston3, J. Folkman 1 , R.J. D' Amatr. 1Childrens Hospital Boston,
Boston, USA; 2Charitd Campus Yirchow Berlin, Germany; 3EntreMed Inc., Rockville, USA Endometriosis, the presence of ectopic endometrial tissue, remains a poorly understood disease. While various models exist about its underlying pathogenesis, it is now widely accepted that angiogenesis, the growth of new blood vessels, plays an essential role in disease development. The well-characterized angiogenic stimulus, vascular endothelial growth factor (VEGF), has been found to be upregulated both in peritoneal fluid and endometriotic lesions. VEGF production is stimulated by a number of conditions including hypoxia. The transcription factor hypoxia-inducible factor (HIF) 1 alpha, heterodimerizes with HIF-1 beta during hypoxic conditions and subsequently transactivates a number of genes, including VEGF thereby promoting angiogenesis. In the current study, we utilize a well-established mouse model of surgically induced endometriosis to demonstrate an important role for hypoxia during the early stages of disease. Intraperitoneal injections into these mice with the hypoxia probe, pimonidazole hydrochloride (Hypoxyprobe-1), immediately prior to euthanasia demonstrated increased staining intensity in early endometriotic lesions as compared to more advanced lesions. To determine whether HIF-1 alpha expression correlated with disease progression as determined by increased lesion size, we performed immunohistochemistry using a HIF-1 alpha antibody on endometriotic lesions at various stages of disease. We found that HIF-1 alpha expression increases with severity of disease. In addition, Western blot analysis of endometriotic lesions from different stages of disease demonstrated that HIF-1 alpha expression was stable early in disease yet undetectable at later stages. To confirm that HIF- 1 alpha was functional in the early endometriotic lesions, we measured mRNA levels of selected HIF target genes in these tissues. We found that mRNA levels of VEGF, glucose transporter-1 and phosphoglycerate