- Email: [email protected]
Altered expression of a cell-cycle suppressor gene, Tob-1, in endometriotic cells by cDNA array analyses
FERTILITY AND STERILITY威 VOL. 78, NO. 4, OCTOBER 2002 Copyright ©2002 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A.
Altered expression of a cell-cycle suppressor gene, Tob-1, in endometriotic cells by cDNA array analyses Dan I. Lebovic, M.D.,a Russell A. Baldocchi, Ph.D.,b Michael D. Mueller, M.D.,c and Robert N. Taylor, M.D., Ph.D.d University of California, San Francisco School of Medicine, San Francisco, California
Received October 9, 2001; revised and accepted January 23, 2002. Supported by the following NIH grants and fellowships: HD08517 (D.I.L.) and HD37321 (D.I.L., R.N.T.), through the Specialized Cooperative Centers Program in Reproductive Research. Reprint requests: Robert N. Taylor, M.D., Ph.D., Center for Reproductive Sciences, Department of Obstetrics, Gynecology and Reproductive Sciences, University of California, San Francisco, HSW 1656, Box 0556, San Francisco, California 94143-0556 (FAX: 415-753-3271; E-mail: rtaylor@socrates. ucsf.edu). a Reproductive Endocrinology Division, Department of Obstetrics and Gynecology, University of Michigan, Ann Arbor, Michigan. b Comprehensive Cancer Center, University of California, San Francisco, San Francisco, California. c Endometriosis Center, Universita¨ts-Frauenklinik, Berne, Switzerland. d Center for Reproductive Sciences, Department of Obstetrics, Gynecology and Reproductive Sciences, University of California, San Francisco, San Francisco, California. 0015-0282/02/$22.00 PII S0015-0282(02)03319-8
Objective: Interleukin (IL)-1, a product of activated peritoneal macrophages, is a central cytokine coordinating neovascularization and monocyte chemotaxis in endometriotic implants. To evaluate the effects of this cytokine on normal endometrial stromal cells and endometriotic stromal cells we performed cDNA expression array analyses before and after exposure to IL-1. Design: Nested case-control study of women with and without laparoscopic evidence of endometriosis. Setting: Reproductive endocrinology clinic at a university hospital. Patient(s): Endometriosis and normal endometrial biopsies from eight patients were used to prepare stromal cell cultures from which mRNA was extracted. Intervention(s): None. Main Outcome Measure(s): Commercially available expression arrays (Atlas Human cDNA Expression Array, Clontech, representing 597 individual genes) were used to screen for mRNAs whose expression was affected by 12 hours of exposure to IL-1 (10 ng/mL). Northern blotting and subsequent quantitative densitometric evaluation was done to confirm steady-state levels of Tob-1 mRNA transcripts. Result(s): Array analyses revealed a cell-cycle regulatory gene, Tob-1, which was differentially expressed by the two cell types after incubation with IL-1. Tob-1 was reduced 48% in endometriotic stromal cells exposed to IL-1, but there was only a 16% reduction in normal endometrial stromal cells. Replicate Northern analyses (n ⫽ 4) showed that exposure to IL-1 for 12 hours resulted in a 25% ⫾ 5% diminution of Tob-1 mRNA in endometriotic stromal cells. In contrast, no significant decrease (⬍3%) was observed in IL-1 exposed normal endometrial stromal cells. Conclusion(s): Tob-1, a cell-cycle inhibitor gene is differentially responsive to IL-1 in endometriotic stromal cells compared to normal endometrial stromal cells. IL-1 down-regulated Tob-1 in endometriotic stromal cells, but had no significant effect on normal endometrial stromal cells. Our results suggest that IL-1 promotes growth of endometriotic lesions through inhibition of Tob-1. These findings are the first to associate IL-1 with an alteration of cell-cycle gene expression in cells derived from endometriotic implants. (Fertil Steril威 2002;78:849 –54. ©2002 by American Society for Reproductive Medicine.) Key Words: Endometriosis, Tob-1, apoptosis, interleukin-1, cDNA array
Ectopic endometrial glandular and stromal cells define endometriotic lesions. How they get there and why they endure are the subjects of ongoing debate. Once present, however, it seems that these cells are shielded from, or are stimulated by, the associated inflammatory response (1) and proliferate and invade. Apoptosis is a normal phenomenon in the human endometrium (2, 3). One hypothesis for the survival of ectopic endometrium is an interference with this apoptotic process (4). Transducer of ErbB (Tob-1) is a cell-cycle reg-
ulatory protein that inhibits the erythroblastosis virus-related human proto-oncogene ErbB-2 protein, a receptor tyrosine kinase, and is associated with antiproliferative activity (5). Tob-1 may in fact function as a tumor suppressor gene. Expression of Tob-1 mRNA is observed in various cell types, although it has not been examined in human endometrial tissue. Our previous work identified the cytokine interleukin (IL)-1, a product of peritoneal macrophages, as a central cytokine coordinating neovascularization and monocyte chemo849
taxis in endometriotic implants. These effects appeared to be mediated by vascular endothelial growth factor (VEGF), IL-6, and regulated on activation, normal T-cell expressed and secreted (RANTES) gene expression, respectively (6). Because IL-1 has been shown to have mitogenic properties in other cell types, we sought a mechanism whereby the exaggerated immune response seen with endometriosis may be linked to the enhanced survival of endometriotic lesions through IL-1 induction of cell-cycle regulatory proteins. We performed cDNA microarray analyses of normal endometrial stromal cells and endometriotic stromal cells after exposure to IL-1. Attention was directed toward genes involved in proliferative and apoptotic processes. One gene in particular, Tob-1, which encodes a growth inhibitory protein, was identified. We found that this mRNA was differentially expressed in normal endometrial stromal cells compared to endometriotic stromal cells.
MATERIALS AND METHODS Sources of Tissues Tissue specimens were obtained from patients undergoing laparoscopy or laparotomy after providing written informed consent under a study protocol approved by the University of California, San Francisco, Institutional Review Board. Healthy ovulatory women, who had not received hormones or GnRH agonist therapy for at least 6 months before surgery, were recruited as control patients providing endometrium. The four control patients had final pathologic diagnoses of uterine leiomyomata but no clinical or pathological evidence of endometriosis. Women with ovarian endometriosis were selected as case subjects and staged intraoperatively according to the revised American Society for Reproductive Medicine classification (7). Endometrioma biopsies were collected under sterile conditions and transported to the laboratory on ice in minimal essential medicine (MEM)-␣ with 10% fetal bovine serum (FBS). All samples were examined histologically and considered endometriotic lesions when the epithelium and stroma were seen. The demographics were without significant difference between the control and case subjects in terms of age or race. None of the women used hormonal therapy within 6 months before the acquisition of tissue.
Interleukin-1 Stimulation Cultures of normal endometrial stromal cells and endometriotic stromal cells were plated in 10-cm culture dishes (Becton Dickinson, Lincoln Park, NJ) and allowed to grow to confluence in 10% FBS-supplemented media. Before the addition of cytokine, the medium was changed to a low serum medium (MEM-␣ supplemented with 2.5% FBS, antibiotics, nucleosides, and nonessential amino acids). Pilot dose–response experiments showed a maximum stimulation of IL-6 mRNA and protein after treatment with recombinant human IL-1 (10 ng/mL ⫽ 0.6 nM; Sigma Chemical Co., St. Louis, MO). Pilot experiments also showed that 85% of maximal protein accumulation was reached after 12 hours.
Array Hybridization Atlas arrays representing 597 known genes were used according to the manufacturer’s protocol. Briefly, the hybridization probes were produced by reverse-transcribing 10 g of total RNA (TRIzol-extracted) while incorporating [32P]-␣-dATP. This reaction was primed with a cocktail of gene-specific primers, one for each of the 597 genes queried by the array. To compare pairs of treatments, the complex probes were hybridized to pairs of filters overnight at 68°C, then washed finally with 0.1⫻ saline sodium citrate and 0.5% sodium dodecyl sulfate.
Scanning and Normalization of Hybridized Blots After wrapping in sheets of cellophane, the washed blots were placed onto detection screens for 4 –16 hours, then scanned by a Molecular Dynamics PhosphorImager 445SI (Storm-Molecular Dynamics, Menlo Park, CA). Images of the blots were segmented and spot intensities were quantified using ImageQuant (Storm-Molecular Genetics) software. Background intensities were taken from regions adjacent to spots and then subtracted from intensities for each spot. Relative expression was calculated from ratios of background-subtracted intensities of corresponding spots, taken from pairs of filters. Normalization of these ratios was performed on the basis of the interblot intensity ratios of the population of spots that gave substantial signal-to-background ratios, the average for which was adjusted to a value of 1.
Human Endometrial Cell Cultures
Preparation of Total RNA and Northern Analysis
Primary cell cultures were prepared from endometrial and endometrioma biopsies, as we have described previously (8). Glandular epithelial cells were separated from stromal cells and debris by filtration through narrow gauge sieves. Normal endometrial stromal cells and endometriotic stromal cells were subcultured to eliminate contamination by macrophages or other leukocytes. Extensive characterization of the stromal cultures prepared using this protocol confirmed that they were more than 95% pure and retained functional markers of their endometrial origin in vitro (8, 9).
Total RNA was extracted from independent cell cultures using the TRIzol reagent kit (GIBCO BRL, Gaithersburg, MD). Total RNA (10 g) was subjected to electrophoresis on agarose gels and blotted by capillary transfer onto nylon membranes (Schleicher & Schuell, Keene, NH). The membranes were hybridized with a 32P-labeled Tob-1 complementary DNA (cDNA) probe synthesized by random primer extension (Clontech, Palo Alto, CA). The template for the Tob-1 probe was a 165-bp fragment of Tob-1 cDNA (5). The integrity and relative amount of RNA loaded into each lane
850
Lebovic et al.
Altered cell-cycle gene in endometriosis
Vol. 78, No. 4, October 2002
were confirmed using a 240-bp 32P-labeled glyceraldehyde3-phosphate dehydrogenase cDNA probe as a constitutively expressed marker. The IL-6 mRNA, a positive control for IL-1 action, was detected on the blots as described previously (6). Data were analyzed as ratios of the density of the hybridization signals of Tob-1 or IL-6 to glyceraldehyde-3phosphate dehydrogenase mRNA, as determined by a phosphorImager (Storm-Molecular Dynamics).
Statistical Analysis All experiments were repeated a minimum of three times and analyzed by unpaired or paired t tests as appropriate. Results are presented as mean ⫾ SEM. Significant differences were accepted when two-tailed analyses yielded P⬍.05 (10).
FIGURE 1 Clontech Atlas expression array analysis of endometrial stromal cells treated with interleukin-1 (IL-1). Poly-adenylated RNA was isolated from variously treated cells (detailed in Methods), labeled by reverse transcribing with [32P]dATP, and hybridized to filters spotted with cloned cDNA PCR products. Differential amounts of specific transcripts were determined by measuring normalized spot intensity variations. Arrows indicate spots that represent the Tob-1 transcript, the abundance of which varied as a function of both treatment with IL-1 and stromal cell type. The difference in Tob-1 expression method by this assay is indicated on the figure. NSC ⫽ normal stromal cells; ESC ⫽ endometriotic stromal cells.
RESULTS Array Expression Profile of Tob-1 mRNA in Normal Endometrial Stromal Cells and Endometriotic Stromal Cells Atlas human cDNA expression arrays contain polymerase chain reaction (PCR) products from cDNA clones of 597 genes, comprised of oncogenes, tumor suppressors, cellcycle regulators, and genes from various other functional classes. Of these, only some spots gave signals substantially above background (⬎10% above) on both filters within a comparison. Such spots enabled the reliable evaluation of expression differences using these arrays. Thus, in the normal endometrial stromal cells ⫾ IL-1 comparison, 118 or 19.8% of the 597 spots met these criteria, whereas in the endometriotic stromal cells comparison, 126 or 21.1% of the spots had adequate intensity. There are a total of 67 cell-cycle regulatory genes available for comparison on these arrays. Of these, 9 genes in the endometriotic stromal cells comparison and 12 in the normal endometrial stromal cells comparison were represented by spots having significant signal in both the IL-1-treated and control RNA samples, therefore a valid expression ratio could be determined (meeting the above criteria), and only one of these genes was reduced by IL-1–Tob-1. Tob-1 expression was reduced 48% by IL-1 treatment of endometriotic stromal cells, whereas the IL-1 effect on normal endometrial stromal cells (16% inhibition) was significantly less (Fig. 1). Therefore, we chose to study the effect on Tob-1 more extensively by Northern analysis. In addition, the arrays showed that expression of the IL-6 cytokine gene was stimulated more than threefold in endometriotic stromal cells after exposure to IL-1 treatment, relative to normal endometrial stromal cells. This observation previously was established by Northern blots (6) and affords a further indication that the array comparison method is robust and reproducible. FERTILITY & STERILITY威
Lebovic. Altered cell-cycle gene in endometriosis. Fertil Steril 2002.
Northern Analysis: Tob-1 mRNA Expression in Endometriotic Stromal Cells Four independent normal endometrial stromal cell and endometriotic stromal cell preparations were evaluated for the expression of Tob-1 mRNA after IL-1 stimulation. Northern hybridization was used to identify and quantify Tob-1 mRNA transcripts (Fig. 2). The Tob-1 probe detected a single transcript of 2.4 kb, in agreement with other reports (5). This probe also showed low level cross-hybridization with 28S and 18S rRNA bands on the total RNA blots, but these could be easily distinguished from the Tob-1 mRNA signal. As an internal control for RNA quantity and integrity, 851
FIGURE 2 Northern analyses demonstrating the induction of Tob-1 mRNA. Representative Northern blots demonstrating the induction of Tob-1 mRNA in interleukin-1 (IL-1)-treated (10 ng/mL) human normal stromal cells (NSC) and endometriotic stromal cells (ESC). Tob-1 transcripts of 2.4-kb were detected. The positions of 28S and 18S ribosomal RNA on the original gels are marked. The integrity and amount of total RNA loaded were confirmed by subsequent hybridization of the blot with a glyceraldehyde3-phosphate dehydrogenase (GAPDH) probe (1.2-kb).
Lebovic. Altered cell-cycle gene in endometriosis. Fertil Steril 2002.
the blots were reprobed to quantify mRNA representing the constitutive glyceraldehyde-3-phosphate dehydrogenase gene (1.2 kb), which was used to normalize the phosphorImaging data. Relative to untreated endometriotic stromal cells, incubation for 12 hours in the presence of IL-1 (10 ng/mL) resulted in a 25% ⫾ 5% diminution of steady-state Tob-1 mRNA (P⬍.05). In contrast, no significant Tob-1 mRNA decrease, ⬍3%, was observed in IL-1-exposed normal endometrial stromal cells (Fig. 3).
DISCUSSION Endometriosis is a common disease among reproductiveaged women. Most investigators agree that retrograde men852
Lebovic et al.
Altered cell-cycle gene in endometriosis
struation plays a key role in the pathogenesis of endometriosis, but the exact mechanisms are still unknown. Retrograde menstruation is a frequent physiologic occurrence in women with patent tubes (11) and viable endometrial cells can be recovered from the peritoneal fluid during menstruation (12). However, only about 15% of women develop endometriosis, suggesting that the endometrial cells of women with endometriosis possess inherent mechanisms that allow lesion persistence in the abdominal cavity. Gebel et al. (13) showed that spontaneous apoptosis was significantly lower in ectopic compared to eutopic endometrium. Furthermore, Meresman et al. (14) found a decreased number of apoptotic cells in eutopic endometrium from Vol. 78, No. 4, October 2002
FIGURE 3 PhosphorImaging quantification of Northern analyses for Tob-1 mRNA. The Tob-1 mRNA levels in interleukin-1 (IL-1)-treated normal stromal cells (NSC) and endometriotic stromal cells (ESC) are shown overlapping (■) the percentage in untreated controls (䊐). Interleukin-1 inhibited Tob-1 mRNA levels ⬍3% and 25% ⫾ 5% in NSC and ESC, respectively.
Lebovic. Altered cell-cycle gene in endometriosis. Fertil Steril 2002.
women with endometriosis compared to normal women. They also reported increased expression of a proto-oncogene that blocks cell death, B-cell lymphoma/leukemia-2 (Bcl-2), in proliferative eutopic endometrium from women with endometriosis. Bax, a Bcl-2 inhibitor, was absent in proliferative endometrium, although it predominates in secretory endometrium. The implication is that Bcl-2 is an important antiapoptotic factor in eutopic endometrium from patients with endometriosis, rendering their tissue less susceptible to apoptosis than endometrium from controls. Macrophage-derived IL-1 plays several important contributory roles in the establishment and maintenance of endometriotic lesions (1). It is a stimulus for angiogenesis and monocyte chemotaxis in endometriotic cells, through VEGF, IL-6, and RANTES expression. IL-1 induces mitogenesis of U373 cells, rat pulmonary myofibroblasts, airway epithelial cells, and rat aortic smooth muscle cells (15–18). Therefore, we sought a relationship between IL-1 and the survival of endometriotic tissue. Of the total of 126 array genes that showed measurable signal intensity when looking at IL-1-induced changes in endometriotic stromal cells relative to normal endometrial stromal cells, only 23 genes, or 18%, demonstrated a significant, more than twofold, difference in expression due to IL-1 treatment. Two of these are linked directly to cellcycle control: Tob-1 (reduced in endometriotic stromal cells) and cyclin D1 (increased in endometriotic stromal cells). In the current study we focused on the cell cycle inhibitor Tob-1, which demonstrated a differential response to IL-1 in endometriotic stromal cells relative to normal endometrial stromal cells. Absolute differences were not assessed in this study; rather, the magnitude of effect using IL-1 on Tob-1 mRNA gene expression was the primary aim of this study. However, the basal levels of Tob-1 mRNA were observed to FERTILITY & STERILITY威
be higher in normal endometrial stromal cells than endometriotic stromal cells by Northern analysis (Fig. 2). Tob-1 is a newly identified tumor suppressor that interferes with protein tyrosine kinase receptors such as ErbB-2, and is thus considered antiproliferative (5). The Tob-1 gene is located on chromosome 17q21. Functional loss of this chromosome has been implicated in endometriotic lesions (19). Tob-1 inhibits cell growth by sequestering cells in G0/G1 phases of the cycle. Tob-1 was down-regulated 48% in endometriotic stromal cells exposed to IL-1, but normal endometrial stromal cells only showed a 16% diminution by array analyses. Validation of mRNA expression was performed by replicate Northern analyses showing that exposure to IL-1 for 12 hours resulted in a 25% ⫾ 5% diminution of Tob-1 mRNA in endometriotic stromal cells (P⬍.05). In contrast, no significant decrease (⬍3%) was observed in IL-1-exposed normal endometrial stromal cells. Unfortunately, Tob-1 cell-cycle suppression has not been sufficiently studied in other cell systems to allow for an accurate assessment of the impact of a 25% reduction in Tob-1 mRNA expression. Tzachanis et al. (20) used Tob-1 antisense oligonuclelotides to induce human T-cell proliferation and cytokine production. The decreased Tob-1 production was determined by immunoblotting, but there was no quantification of Tob-1 mRNA or protein suppression. Furthermore, although not compelling or direct evidence, it is interesting to note that Guardavaccaro et al. (21) have found that PC3, a member of the Tob-1 gene family, directly caused a reduction in cyclin D1 levels. This mirrors our own observations, in that IL-1 reduces Tob-1 and increases cyclin D1 expression in endometriotic stromal cells relative to normal endometrial stromal cells. The reciprocal relationship in the expression of these two genes is consistent with the idea that 853
even a modest (25%) reduction in Tob-1 expression may impact regulation of the G1/S phase transition. In conclusion, we have demonstrated that IL-1 plays a role in the control of Tob-1 expression in endometriotic cells. The suppression of Tob-1 in endometriotic stromal cells by IL-1 provides a mechanism for the continued/ enhanced growth potential of endometriotic lesions. Although cell-cycle control is a complex process, Tob-1 has been singled out due to the significant differences seen on array and Northern analyses. IL-1 appears to exert a plethora of effects on the persistence and propagation of endometriosis through chemotaxis, angiogenesis, and proliferation.
7. 8. 9.
10. 11. 12. 13. 14.
15. Acknowledgments: The authors thank the clinical staff of the UCSF Department of Obstetrics, Gynecology and Reproductive Sciences for their generous contributions to the study.
16. 17.
References 1. Lebovic DI, Mueller MD, Taylor RN. Immunobiology of endometriosis. Fertil Steril 2001;75:1–10. 2. Kokawa K, Shikone T, Nakano R. Apoptosis in the human uterine endometrium during the menstrual cycle. J Clin Endocrinol Metab 1996;81:4144 –7. 3. Hopwood D, Levison DA. Atrophy and apoptosis in the cyclical human endometrium. J Pathol 1976;119:159 – 66. 4. Braun DP, Dmowski WP. Endometriosis: abnormal endometrium and dysfunctional immune response. Curr Opin Obstet Gynecol 1998;10: 365–9. 5. Matsuda S, Kawamura-Tsuzuku J, Ohsugi M, Yoshida M, Emi M, Nakamura Y, et al. Tob1, a novel protein that interacts with p185erbB2, is associated with anti-proliferative activity. Oncogene 1996;12:705– 13. 6. Lebovic DI, Bentzien F, Chao VA, Garrett EN, Meng YG, Taylor RN.
854
Lebovic et al.
Altered cell-cycle gene in endometriosis
18.
19. 20.
21.
Induction of an angiogenic phenotype in endometriotic stromal cell cultures by interleukin-1. Mol Hum Reprod 2000;6:269 –75. Revised American Society for Reproductive Medicine classification of endometriosis: 1996. Fertil Steril 1997;67:817–21. Ryan I, Schriock E, Taylor R. Isolation, characterization, and comparison of human endometrial and endometriosis cells in vitro. J Clin Endocrinol Metab 1994;78:642–9. Brandenberger AW, Lebovic DI, Tee MK, Ryan IP, Tseng JF, Jaffe RB, et al. Estrogen receptor (ER)-␣ and ER- isoforms in normal endometrial and endometriosis-derived stromal cells. Mol Hum Reprod 1999; 5:651–5. Glantz SA. Primer of biostatistics. New York: McGraw-Hill, 1992. Halme J, Hammond M, Hulka J, Raj S, Talbert L. Retrograde menstruation in healthy women and in patients with endometriosis. Obstet Gynecol 1984;64:151– 4. Kruitwagen RF, Poels LG, Willemsen WN, DeRonge IJ, Jap PH, Thomas CM, et al. Endometrial epithelial cells in peritoneal fluid during the early follicular phase. Fertil Steril 1991;55:297–303. Gebel HM, Braun DP, Tambur A, Frame D, Rana N, Dmowski WP. Spontaneous apoptosis of endometrial tissue is impaired in women with endometriosis. Fertil Steril 1998;69:1042–7. Meresman GF, Vighi S, Buquet RA, Contreras-Ortiz O, Tesone M, Rumi LS. Apoptosis and expression of Bcl-2 and Bax in eutopic endometrium from women with endometriosis. Fertil Steril 2000;74: 760 – 6. Bourcier T, Dockter M, Hassid A. Synergistic interaction of interleukin-1 beta and growth factors in primary cultures of rat aortic smooth muscle cells. J Cell Physiol 1995;164:644 –57. Murlas CG, Sharma AC, Gulati A, Najmabadi F. Interleukin-1 beta increases airway epithelial cell mitogenesis partly by stimulating endothelin-1 production. Lung 1997;175:117–26. Utal AK, Stopka AL, Coleman PD. Interleukin-1 beta stimulates mitogen-activated protein kinase in U373 astrocytoma cells without the production of lipid second messengers. Neurochem Res 1998;23:235– 42. Boyle JE, Lindroos PM, Rice AB, Zhang L, Zeldin DC, Bonner JC. Prostaglandin-E2 counteracts interleukin-1beta-stimulated upregulation of platelet-derived growth factor alpha-receptor on rat pulmonary myofibroblasts. Am J Respir Cell Mol Biol 1999;20:433– 40. Kosugi Y, Elias S, Malinak LR, Nagata J, Isaka K, Takayama M, et al. Increased heterogeneity of chromosome 17 aneuploidy in endometriosis. Am J Obstet Gynecol 1999;180:792–7. Tzachanis D, Freeman GJ, Hirano N, van Puijenbroek AA, Delfs MW, Berezovskaya A, et al. Tob1 is a negative regulator of activation that is expressed in anergic and quiescent T cells. Nat Immunol 2001;2:1174 – 82. Guardavaccaro D, Corrente G, Covone F, Micheli L, D’Agnano I, Starace G, et al. Arrest of G(1)-S progression by the p53-inducible gene PC3 is Rb dependent and relies on the inhibition of cyclin D1 transcription. Mol Cell Biol 2000;20:1797– 815.
Vol. 78, No. 4, October 2002
Altered expression of a cell-cycle suppressor gene, Tob-1, in endometriotic cells by cDNA array analyses Dan I. Lebovic, M.D.,a Russell A. Baldocchi, Ph.D.,b Michael D. Mueller, M.D.,c and Robert N. Taylor, M.D., Ph.D.d University of California, San Francisco School of Medicine, San Francisco, California
Received October 9, 2001; revised and accepted January 23, 2002. Supported by the following NIH grants and fellowships: HD08517 (D.I.L.) and HD37321 (D.I.L., R.N.T.), through the Specialized Cooperative Centers Program in Reproductive Research. Reprint requests: Robert N. Taylor, M.D., Ph.D., Center for Reproductive Sciences, Department of Obstetrics, Gynecology and Reproductive Sciences, University of California, San Francisco, HSW 1656, Box 0556, San Francisco, California 94143-0556 (FAX: 415-753-3271; E-mail: rtaylor@socrates. ucsf.edu). a Reproductive Endocrinology Division, Department of Obstetrics and Gynecology, University of Michigan, Ann Arbor, Michigan. b Comprehensive Cancer Center, University of California, San Francisco, San Francisco, California. c Endometriosis Center, Universita¨ts-Frauenklinik, Berne, Switzerland. d Center for Reproductive Sciences, Department of Obstetrics, Gynecology and Reproductive Sciences, University of California, San Francisco, San Francisco, California. 0015-0282/02/$22.00 PII S0015-0282(02)03319-8
Objective: Interleukin (IL)-1, a product of activated peritoneal macrophages, is a central cytokine coordinating neovascularization and monocyte chemotaxis in endometriotic implants. To evaluate the effects of this cytokine on normal endometrial stromal cells and endometriotic stromal cells we performed cDNA expression array analyses before and after exposure to IL-1. Design: Nested case-control study of women with and without laparoscopic evidence of endometriosis. Setting: Reproductive endocrinology clinic at a university hospital. Patient(s): Endometriosis and normal endometrial biopsies from eight patients were used to prepare stromal cell cultures from which mRNA was extracted. Intervention(s): None. Main Outcome Measure(s): Commercially available expression arrays (Atlas Human cDNA Expression Array, Clontech, representing 597 individual genes) were used to screen for mRNAs whose expression was affected by 12 hours of exposure to IL-1 (10 ng/mL). Northern blotting and subsequent quantitative densitometric evaluation was done to confirm steady-state levels of Tob-1 mRNA transcripts. Result(s): Array analyses revealed a cell-cycle regulatory gene, Tob-1, which was differentially expressed by the two cell types after incubation with IL-1. Tob-1 was reduced 48% in endometriotic stromal cells exposed to IL-1, but there was only a 16% reduction in normal endometrial stromal cells. Replicate Northern analyses (n ⫽ 4) showed that exposure to IL-1 for 12 hours resulted in a 25% ⫾ 5% diminution of Tob-1 mRNA in endometriotic stromal cells. In contrast, no significant decrease (⬍3%) was observed in IL-1 exposed normal endometrial stromal cells. Conclusion(s): Tob-1, a cell-cycle inhibitor gene is differentially responsive to IL-1 in endometriotic stromal cells compared to normal endometrial stromal cells. IL-1 down-regulated Tob-1 in endometriotic stromal cells, but had no significant effect on normal endometrial stromal cells. Our results suggest that IL-1 promotes growth of endometriotic lesions through inhibition of Tob-1. These findings are the first to associate IL-1 with an alteration of cell-cycle gene expression in cells derived from endometriotic implants. (Fertil Steril威 2002;78:849 –54. ©2002 by American Society for Reproductive Medicine.) Key Words: Endometriosis, Tob-1, apoptosis, interleukin-1, cDNA array
Ectopic endometrial glandular and stromal cells define endometriotic lesions. How they get there and why they endure are the subjects of ongoing debate. Once present, however, it seems that these cells are shielded from, or are stimulated by, the associated inflammatory response (1) and proliferate and invade. Apoptosis is a normal phenomenon in the human endometrium (2, 3). One hypothesis for the survival of ectopic endometrium is an interference with this apoptotic process (4). Transducer of ErbB (Tob-1) is a cell-cycle reg-
ulatory protein that inhibits the erythroblastosis virus-related human proto-oncogene ErbB-2 protein, a receptor tyrosine kinase, and is associated with antiproliferative activity (5). Tob-1 may in fact function as a tumor suppressor gene. Expression of Tob-1 mRNA is observed in various cell types, although it has not been examined in human endometrial tissue. Our previous work identified the cytokine interleukin (IL)-1, a product of peritoneal macrophages, as a central cytokine coordinating neovascularization and monocyte chemo849
taxis in endometriotic implants. These effects appeared to be mediated by vascular endothelial growth factor (VEGF), IL-6, and regulated on activation, normal T-cell expressed and secreted (RANTES) gene expression, respectively (6). Because IL-1 has been shown to have mitogenic properties in other cell types, we sought a mechanism whereby the exaggerated immune response seen with endometriosis may be linked to the enhanced survival of endometriotic lesions through IL-1 induction of cell-cycle regulatory proteins. We performed cDNA microarray analyses of normal endometrial stromal cells and endometriotic stromal cells after exposure to IL-1. Attention was directed toward genes involved in proliferative and apoptotic processes. One gene in particular, Tob-1, which encodes a growth inhibitory protein, was identified. We found that this mRNA was differentially expressed in normal endometrial stromal cells compared to endometriotic stromal cells.
MATERIALS AND METHODS Sources of Tissues Tissue specimens were obtained from patients undergoing laparoscopy or laparotomy after providing written informed consent under a study protocol approved by the University of California, San Francisco, Institutional Review Board. Healthy ovulatory women, who had not received hormones or GnRH agonist therapy for at least 6 months before surgery, were recruited as control patients providing endometrium. The four control patients had final pathologic diagnoses of uterine leiomyomata but no clinical or pathological evidence of endometriosis. Women with ovarian endometriosis were selected as case subjects and staged intraoperatively according to the revised American Society for Reproductive Medicine classification (7). Endometrioma biopsies were collected under sterile conditions and transported to the laboratory on ice in minimal essential medicine (MEM)-␣ with 10% fetal bovine serum (FBS). All samples were examined histologically and considered endometriotic lesions when the epithelium and stroma were seen. The demographics were without significant difference between the control and case subjects in terms of age or race. None of the women used hormonal therapy within 6 months before the acquisition of tissue.
Interleukin-1 Stimulation Cultures of normal endometrial stromal cells and endometriotic stromal cells were plated in 10-cm culture dishes (Becton Dickinson, Lincoln Park, NJ) and allowed to grow to confluence in 10% FBS-supplemented media. Before the addition of cytokine, the medium was changed to a low serum medium (MEM-␣ supplemented with 2.5% FBS, antibiotics, nucleosides, and nonessential amino acids). Pilot dose–response experiments showed a maximum stimulation of IL-6 mRNA and protein after treatment with recombinant human IL-1 (10 ng/mL ⫽ 0.6 nM; Sigma Chemical Co., St. Louis, MO). Pilot experiments also showed that 85% of maximal protein accumulation was reached after 12 hours.
Array Hybridization Atlas arrays representing 597 known genes were used according to the manufacturer’s protocol. Briefly, the hybridization probes were produced by reverse-transcribing 10 g of total RNA (TRIzol-extracted) while incorporating [32P]-␣-dATP. This reaction was primed with a cocktail of gene-specific primers, one for each of the 597 genes queried by the array. To compare pairs of treatments, the complex probes were hybridized to pairs of filters overnight at 68°C, then washed finally with 0.1⫻ saline sodium citrate and 0.5% sodium dodecyl sulfate.
Scanning and Normalization of Hybridized Blots After wrapping in sheets of cellophane, the washed blots were placed onto detection screens for 4 –16 hours, then scanned by a Molecular Dynamics PhosphorImager 445SI (Storm-Molecular Dynamics, Menlo Park, CA). Images of the blots were segmented and spot intensities were quantified using ImageQuant (Storm-Molecular Genetics) software. Background intensities were taken from regions adjacent to spots and then subtracted from intensities for each spot. Relative expression was calculated from ratios of background-subtracted intensities of corresponding spots, taken from pairs of filters. Normalization of these ratios was performed on the basis of the interblot intensity ratios of the population of spots that gave substantial signal-to-background ratios, the average for which was adjusted to a value of 1.
Human Endometrial Cell Cultures
Preparation of Total RNA and Northern Analysis
Primary cell cultures were prepared from endometrial and endometrioma biopsies, as we have described previously (8). Glandular epithelial cells were separated from stromal cells and debris by filtration through narrow gauge sieves. Normal endometrial stromal cells and endometriotic stromal cells were subcultured to eliminate contamination by macrophages or other leukocytes. Extensive characterization of the stromal cultures prepared using this protocol confirmed that they were more than 95% pure and retained functional markers of their endometrial origin in vitro (8, 9).
Total RNA was extracted from independent cell cultures using the TRIzol reagent kit (GIBCO BRL, Gaithersburg, MD). Total RNA (10 g) was subjected to electrophoresis on agarose gels and blotted by capillary transfer onto nylon membranes (Schleicher & Schuell, Keene, NH). The membranes were hybridized with a 32P-labeled Tob-1 complementary DNA (cDNA) probe synthesized by random primer extension (Clontech, Palo Alto, CA). The template for the Tob-1 probe was a 165-bp fragment of Tob-1 cDNA (5). The integrity and relative amount of RNA loaded into each lane
850
Lebovic et al.
Altered cell-cycle gene in endometriosis
Vol. 78, No. 4, October 2002
were confirmed using a 240-bp 32P-labeled glyceraldehyde3-phosphate dehydrogenase cDNA probe as a constitutively expressed marker. The IL-6 mRNA, a positive control for IL-1 action, was detected on the blots as described previously (6). Data were analyzed as ratios of the density of the hybridization signals of Tob-1 or IL-6 to glyceraldehyde-3phosphate dehydrogenase mRNA, as determined by a phosphorImager (Storm-Molecular Dynamics).
Statistical Analysis All experiments were repeated a minimum of three times and analyzed by unpaired or paired t tests as appropriate. Results are presented as mean ⫾ SEM. Significant differences were accepted when two-tailed analyses yielded P⬍.05 (10).
FIGURE 1 Clontech Atlas expression array analysis of endometrial stromal cells treated with interleukin-1 (IL-1). Poly-adenylated RNA was isolated from variously treated cells (detailed in Methods), labeled by reverse transcribing with [32P]dATP, and hybridized to filters spotted with cloned cDNA PCR products. Differential amounts of specific transcripts were determined by measuring normalized spot intensity variations. Arrows indicate spots that represent the Tob-1 transcript, the abundance of which varied as a function of both treatment with IL-1 and stromal cell type. The difference in Tob-1 expression method by this assay is indicated on the figure. NSC ⫽ normal stromal cells; ESC ⫽ endometriotic stromal cells.
RESULTS Array Expression Profile of Tob-1 mRNA in Normal Endometrial Stromal Cells and Endometriotic Stromal Cells Atlas human cDNA expression arrays contain polymerase chain reaction (PCR) products from cDNA clones of 597 genes, comprised of oncogenes, tumor suppressors, cellcycle regulators, and genes from various other functional classes. Of these, only some spots gave signals substantially above background (⬎10% above) on both filters within a comparison. Such spots enabled the reliable evaluation of expression differences using these arrays. Thus, in the normal endometrial stromal cells ⫾ IL-1 comparison, 118 or 19.8% of the 597 spots met these criteria, whereas in the endometriotic stromal cells comparison, 126 or 21.1% of the spots had adequate intensity. There are a total of 67 cell-cycle regulatory genes available for comparison on these arrays. Of these, 9 genes in the endometriotic stromal cells comparison and 12 in the normal endometrial stromal cells comparison were represented by spots having significant signal in both the IL-1-treated and control RNA samples, therefore a valid expression ratio could be determined (meeting the above criteria), and only one of these genes was reduced by IL-1–Tob-1. Tob-1 expression was reduced 48% by IL-1 treatment of endometriotic stromal cells, whereas the IL-1 effect on normal endometrial stromal cells (16% inhibition) was significantly less (Fig. 1). Therefore, we chose to study the effect on Tob-1 more extensively by Northern analysis. In addition, the arrays showed that expression of the IL-6 cytokine gene was stimulated more than threefold in endometriotic stromal cells after exposure to IL-1 treatment, relative to normal endometrial stromal cells. This observation previously was established by Northern blots (6) and affords a further indication that the array comparison method is robust and reproducible. FERTILITY & STERILITY威
Lebovic. Altered cell-cycle gene in endometriosis. Fertil Steril 2002.
Northern Analysis: Tob-1 mRNA Expression in Endometriotic Stromal Cells Four independent normal endometrial stromal cell and endometriotic stromal cell preparations were evaluated for the expression of Tob-1 mRNA after IL-1 stimulation. Northern hybridization was used to identify and quantify Tob-1 mRNA transcripts (Fig. 2). The Tob-1 probe detected a single transcript of 2.4 kb, in agreement with other reports (5). This probe also showed low level cross-hybridization with 28S and 18S rRNA bands on the total RNA blots, but these could be easily distinguished from the Tob-1 mRNA signal. As an internal control for RNA quantity and integrity, 851
FIGURE 2 Northern analyses demonstrating the induction of Tob-1 mRNA. Representative Northern blots demonstrating the induction of Tob-1 mRNA in interleukin-1 (IL-1)-treated (10 ng/mL) human normal stromal cells (NSC) and endometriotic stromal cells (ESC). Tob-1 transcripts of 2.4-kb were detected. The positions of 28S and 18S ribosomal RNA on the original gels are marked. The integrity and amount of total RNA loaded were confirmed by subsequent hybridization of the blot with a glyceraldehyde3-phosphate dehydrogenase (GAPDH) probe (1.2-kb).
Lebovic. Altered cell-cycle gene in endometriosis. Fertil Steril 2002.
the blots were reprobed to quantify mRNA representing the constitutive glyceraldehyde-3-phosphate dehydrogenase gene (1.2 kb), which was used to normalize the phosphorImaging data. Relative to untreated endometriotic stromal cells, incubation for 12 hours in the presence of IL-1 (10 ng/mL) resulted in a 25% ⫾ 5% diminution of steady-state Tob-1 mRNA (P⬍.05). In contrast, no significant Tob-1 mRNA decrease, ⬍3%, was observed in IL-1-exposed normal endometrial stromal cells (Fig. 3).
DISCUSSION Endometriosis is a common disease among reproductiveaged women. Most investigators agree that retrograde men852
Lebovic et al.
Altered cell-cycle gene in endometriosis
struation plays a key role in the pathogenesis of endometriosis, but the exact mechanisms are still unknown. Retrograde menstruation is a frequent physiologic occurrence in women with patent tubes (11) and viable endometrial cells can be recovered from the peritoneal fluid during menstruation (12). However, only about 15% of women develop endometriosis, suggesting that the endometrial cells of women with endometriosis possess inherent mechanisms that allow lesion persistence in the abdominal cavity. Gebel et al. (13) showed that spontaneous apoptosis was significantly lower in ectopic compared to eutopic endometrium. Furthermore, Meresman et al. (14) found a decreased number of apoptotic cells in eutopic endometrium from Vol. 78, No. 4, October 2002
FIGURE 3 PhosphorImaging quantification of Northern analyses for Tob-1 mRNA. The Tob-1 mRNA levels in interleukin-1 (IL-1)-treated normal stromal cells (NSC) and endometriotic stromal cells (ESC) are shown overlapping (■) the percentage in untreated controls (䊐). Interleukin-1 inhibited Tob-1 mRNA levels ⬍3% and 25% ⫾ 5% in NSC and ESC, respectively.
Lebovic. Altered cell-cycle gene in endometriosis. Fertil Steril 2002.
women with endometriosis compared to normal women. They also reported increased expression of a proto-oncogene that blocks cell death, B-cell lymphoma/leukemia-2 (Bcl-2), in proliferative eutopic endometrium from women with endometriosis. Bax, a Bcl-2 inhibitor, was absent in proliferative endometrium, although it predominates in secretory endometrium. The implication is that Bcl-2 is an important antiapoptotic factor in eutopic endometrium from patients with endometriosis, rendering their tissue less susceptible to apoptosis than endometrium from controls. Macrophage-derived IL-1 plays several important contributory roles in the establishment and maintenance of endometriotic lesions (1). It is a stimulus for angiogenesis and monocyte chemotaxis in endometriotic cells, through VEGF, IL-6, and RANTES expression. IL-1 induces mitogenesis of U373 cells, rat pulmonary myofibroblasts, airway epithelial cells, and rat aortic smooth muscle cells (15–18). Therefore, we sought a relationship between IL-1 and the survival of endometriotic tissue. Of the total of 126 array genes that showed measurable signal intensity when looking at IL-1-induced changes in endometriotic stromal cells relative to normal endometrial stromal cells, only 23 genes, or 18%, demonstrated a significant, more than twofold, difference in expression due to IL-1 treatment. Two of these are linked directly to cellcycle control: Tob-1 (reduced in endometriotic stromal cells) and cyclin D1 (increased in endometriotic stromal cells). In the current study we focused on the cell cycle inhibitor Tob-1, which demonstrated a differential response to IL-1 in endometriotic stromal cells relative to normal endometrial stromal cells. Absolute differences were not assessed in this study; rather, the magnitude of effect using IL-1 on Tob-1 mRNA gene expression was the primary aim of this study. However, the basal levels of Tob-1 mRNA were observed to FERTILITY & STERILITY威
be higher in normal endometrial stromal cells than endometriotic stromal cells by Northern analysis (Fig. 2). Tob-1 is a newly identified tumor suppressor that interferes with protein tyrosine kinase receptors such as ErbB-2, and is thus considered antiproliferative (5). The Tob-1 gene is located on chromosome 17q21. Functional loss of this chromosome has been implicated in endometriotic lesions (19). Tob-1 inhibits cell growth by sequestering cells in G0/G1 phases of the cycle. Tob-1 was down-regulated 48% in endometriotic stromal cells exposed to IL-1, but normal endometrial stromal cells only showed a 16% diminution by array analyses. Validation of mRNA expression was performed by replicate Northern analyses showing that exposure to IL-1 for 12 hours resulted in a 25% ⫾ 5% diminution of Tob-1 mRNA in endometriotic stromal cells (P⬍.05). In contrast, no significant decrease (⬍3%) was observed in IL-1-exposed normal endometrial stromal cells. Unfortunately, Tob-1 cell-cycle suppression has not been sufficiently studied in other cell systems to allow for an accurate assessment of the impact of a 25% reduction in Tob-1 mRNA expression. Tzachanis et al. (20) used Tob-1 antisense oligonuclelotides to induce human T-cell proliferation and cytokine production. The decreased Tob-1 production was determined by immunoblotting, but there was no quantification of Tob-1 mRNA or protein suppression. Furthermore, although not compelling or direct evidence, it is interesting to note that Guardavaccaro et al. (21) have found that PC3, a member of the Tob-1 gene family, directly caused a reduction in cyclin D1 levels. This mirrors our own observations, in that IL-1 reduces Tob-1 and increases cyclin D1 expression in endometriotic stromal cells relative to normal endometrial stromal cells. The reciprocal relationship in the expression of these two genes is consistent with the idea that 853
even a modest (25%) reduction in Tob-1 expression may impact regulation of the G1/S phase transition. In conclusion, we have demonstrated that IL-1 plays a role in the control of Tob-1 expression in endometriotic cells. The suppression of Tob-1 in endometriotic stromal cells by IL-1 provides a mechanism for the continued/ enhanced growth potential of endometriotic lesions. Although cell-cycle control is a complex process, Tob-1 has been singled out due to the significant differences seen on array and Northern analyses. IL-1 appears to exert a plethora of effects on the persistence and propagation of endometriosis through chemotaxis, angiogenesis, and proliferation.
7. 8. 9.
10. 11. 12. 13. 14.
15. Acknowledgments: The authors thank the clinical staff of the UCSF Department of Obstetrics, Gynecology and Reproductive Sciences for their generous contributions to the study.
16. 17.
References 1. Lebovic DI, Mueller MD, Taylor RN. Immunobiology of endometriosis. Fertil Steril 2001;75:1–10. 2. Kokawa K, Shikone T, Nakano R. Apoptosis in the human uterine endometrium during the menstrual cycle. J Clin Endocrinol Metab 1996;81:4144 –7. 3. Hopwood D, Levison DA. Atrophy and apoptosis in the cyclical human endometrium. J Pathol 1976;119:159 – 66. 4. Braun DP, Dmowski WP. Endometriosis: abnormal endometrium and dysfunctional immune response. Curr Opin Obstet Gynecol 1998;10: 365–9. 5. Matsuda S, Kawamura-Tsuzuku J, Ohsugi M, Yoshida M, Emi M, Nakamura Y, et al. Tob1, a novel protein that interacts with p185erbB2, is associated with anti-proliferative activity. Oncogene 1996;12:705– 13. 6. Lebovic DI, Bentzien F, Chao VA, Garrett EN, Meng YG, Taylor RN.
854
Lebovic et al.
Altered cell-cycle gene in endometriosis
18.
19. 20.
21.
Induction of an angiogenic phenotype in endometriotic stromal cell cultures by interleukin-1. Mol Hum Reprod 2000;6:269 –75. Revised American Society for Reproductive Medicine classification of endometriosis: 1996. Fertil Steril 1997;67:817–21. Ryan I, Schriock E, Taylor R. Isolation, characterization, and comparison of human endometrial and endometriosis cells in vitro. J Clin Endocrinol Metab 1994;78:642–9. Brandenberger AW, Lebovic DI, Tee MK, Ryan IP, Tseng JF, Jaffe RB, et al. Estrogen receptor (ER)-␣ and ER- isoforms in normal endometrial and endometriosis-derived stromal cells. Mol Hum Reprod 1999; 5:651–5. Glantz SA. Primer of biostatistics. New York: McGraw-Hill, 1992. Halme J, Hammond M, Hulka J, Raj S, Talbert L. Retrograde menstruation in healthy women and in patients with endometriosis. Obstet Gynecol 1984;64:151– 4. Kruitwagen RF, Poels LG, Willemsen WN, DeRonge IJ, Jap PH, Thomas CM, et al. Endometrial epithelial cells in peritoneal fluid during the early follicular phase. Fertil Steril 1991;55:297–303. Gebel HM, Braun DP, Tambur A, Frame D, Rana N, Dmowski WP. Spontaneous apoptosis of endometrial tissue is impaired in women with endometriosis. Fertil Steril 1998;69:1042–7. Meresman GF, Vighi S, Buquet RA, Contreras-Ortiz O, Tesone M, Rumi LS. Apoptosis and expression of Bcl-2 and Bax in eutopic endometrium from women with endometriosis. Fertil Steril 2000;74: 760 – 6. Bourcier T, Dockter M, Hassid A. Synergistic interaction of interleukin-1 beta and growth factors in primary cultures of rat aortic smooth muscle cells. J Cell Physiol 1995;164:644 –57. Murlas CG, Sharma AC, Gulati A, Najmabadi F. Interleukin-1 beta increases airway epithelial cell mitogenesis partly by stimulating endothelin-1 production. Lung 1997;175:117–26. Utal AK, Stopka AL, Coleman PD. Interleukin-1 beta stimulates mitogen-activated protein kinase in U373 astrocytoma cells without the production of lipid second messengers. Neurochem Res 1998;23:235– 42. Boyle JE, Lindroos PM, Rice AB, Zhang L, Zeldin DC, Bonner JC. Prostaglandin-E2 counteracts interleukin-1beta-stimulated upregulation of platelet-derived growth factor alpha-receptor on rat pulmonary myofibroblasts. Am J Respir Cell Mol Biol 1999;20:433– 40. Kosugi Y, Elias S, Malinak LR, Nagata J, Isaka K, Takayama M, et al. Increased heterogeneity of chromosome 17 aneuploidy in endometriosis. Am J Obstet Gynecol 1999;180:792–7. Tzachanis D, Freeman GJ, Hirano N, van Puijenbroek AA, Delfs MW, Berezovskaya A, et al. Tob1 is a negative regulator of activation that is expressed in anergic and quiescent T cells. Nat Immunol 2001;2:1174 – 82. Guardavaccaro D, Corrente G, Covone F, Micheli L, D’Agnano I, Starace G, et al. Arrest of G(1)-S progression by the p53-inducible gene PC3 is Rb dependent and relies on the inhibition of cyclin D1 transcription. Mol Cell Biol 2000;20:1797– 815.
Vol. 78, No. 4, October 2002