| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Pregnancy |
-Induced Apoptosis and Integrin Switching in Human Extravillous Trophoblast Cell Line1
Department of Obstetrics and Gynecology,3 Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan
Department of Obstetrics and Gynecology,4 Kanto-Rosai Hospital, Kanagawa, Japan
Department of Obstetrics and Gynecology,5 Saitama Medical Center, Saitama Medical School, Saitama, Japan
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
) induced apoptosis in TCL1 cells but not in JEG3 cells derived from choriocarcinoma while the addition of interleukin-1ß, leukemia inhibitory factor, or transforming growth factor had no effect on TCL1 cells. This apoptosis was suppressed when TCL1 cells were seeded on fibronectin (Fn), collagen type I (C1), collagen type IV (C4), or laminin (Ln). Wortmannin, a specific PI3 kinase inhibitor, inhibited this suppression. Spreading assays and adhesion blocking assays indicated that TCL1 cells express integrin-5 and -6 and ß1 and ß4 subunits. Adhesion on Fn is mediated by 5ß1, and adhesion on C1, C4, or Ln is mediated by 6ß1 integrins. TNF suppressed 6 integrin expression and enhanced 1 integrin expression in a dose-dependent manner. In addition, aggregation of ß1 subunits on C4 was detected after addition of TNF. Taken together, these results suggest that TNF and ECM, through activation of PI3 kinase mediated by ß1 integrin signaling, might collaboratively regulate differentiation of trophoblast cells through integrin signaling in establishing and maintaining successful pregnancy.
apoptosis, extracellular matrices, integrin switching, invasive trophoblast, tumor necrosis factor alpha
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Integrins comprise a large family of cell surface receptors that recognize a variety of extracellular matrix (ECM) components. Integrins mediate cell adhesion and migration, orchestrate organization of the actin-based cytoskeleton, and activate signal transduction pathways including mitogen-activated protein (MAP) kinase and phosphatidylinositol 3 (PI3) kinase pathways [4]. Loss of adhesion to ECM causes apoptosis, known as anoikis, in many types of cells [5–7]. In addition, integrins also facilitate an exit from the cell cycle and provide a signal for differentiation [8] in cooperation with stimulation by extrinsic factors. Therefore, integrins function principally as an essential cue in cell survival and differentiation. Integrins can also attach to different ECM [4]. During normal trophoblast invasion, the ECM undergoes a transition from laminin (Ln) to fibronectin (Fn) and collagen type IV (C4). In addition, integrin 6ß4 is downregulated and integrins 5ß1 and 1ß1 are upregulated in differentiating and invasive CTs [9, 10]. These data suggest that alterations of integrin expression by EVT may occur in response to signals occurring after adhesion to the ECM [11–14]. The significance for normal placentation of this CT differentiation is highlighted by the fact that, in preeclampsia, in which both interstitial and endovascular invasion are abnormally shallow, CTs show significant defects in differentiation [11–14]. Furthermore, CTs in preeclampsia increase expression of the 5ß1 but fail both to upregulate 1ß1 and to downregulate 6ß4 [10, 14]. The 1 and 5 subunits seemed to be involved in the regulation of invasiveness in vitro [10]. Thus, they appear arrested in their differentiation and express an ECM receptor phenotype that may not be optimal for invasion [15]. Taken these evidences together, it is likely that expression of these integrin subunits are involved in regulation of normal EVT differentiation and are essential for normal pregnancy. However, the mechanism to regulate this integrin subunit conversion is still unclear.
A variety of cytokines have influences on the behavior of trophoblast cells [16]. Some cytokines stimulate the proliferation of trophoblast cells while others are associated with the conversion to an invasive phenotype. Cytokines derived from endometrial cells seem to be particularly involved in the differentiation of EVT [17]. Among placental cytokines, tumor necrosis factor alpha (TNF) is known to induce apoptosis in many tissues [18, 19]. In preeclampsia, TNF levels are apparently elevated in maternal and cord blood, and higher expression in placenta of this cytokine has been reported [20, 21]. Thus, TNF may regulate both invasiveness and viability in trophoblast cells during pregnancy. However, there have been no reports regarding the relationship between cytokines, including TNF, and integrin expression in trophoblasts. TCL1 cells are an immortalized cell line established from the chorionic membrane and express specific markers of EVT [22–24]. In order to clarify the essential molecules that regulate the differentiation of EVT, we examined the effects of cytokines and the ECM on a cell line (TCL1) derived from human EVT.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
ECM used in this study were as follows: bovine collagen type I was purchased from Becton Dickinson Labware (Bedford, MA); human placental collagen type IV, laminin, and bovine fibronectin were all purchased from Sigma Chemical Co. (St. Louis, MO).
TNF was purchased from Chemicon International Inc. (Temecula, CA). Leukemia inhibitory factor (LIF), transforming growth factor ß (TGFß), and interleukin (IL)-1ß were purchased from Sigma Chemical Co. Wortmannin and PD98059 were purchased from Wako (Tokyo, Japan).
Mouse mAbs against integrin 5, ß1, and ß4 (CD104) subunits and rat mAb against integrin 6 (CD49f) were all obtained from Chemicon International Inc. Goat anti-rat IgG and anti-mouse IgG were used as secondary antibodies (Biosource International Inc., Camarillo, CA).
The antibodies used in immunoblotting were as follows: rabbit anti-integrin 1 subunit polyclonal antibody, mouse anti-human CD49f monoclonal antibody, and mouse anti-human integrin ß1 monoclonal antibody were obtained from Chemicon Inc., horseradish peroxidase (HRP)-conjugated anti-mouse IgG from American Qualex (San Clemente, CA), and HRP-conjugated anti-rabbit IgG from MBL (Tokyo, Japan).
The mouse antihuman integrin ß1 monoclonal antibody, K20, was used as the primary antibody for immunofluorescence (Chemicon Inc.). FITC-conjugated anti-mouse IgG (Molecular Probes Inc., Eugene, OR) was used as the secondary antibody.
Cell Line and Cell Culture
TCL1 cells were established from mixed primary cultures of cells, isolated from the chorionic membrane of placenta, obtained from elective caesarean sections performed before term [22]. The primary cultures contained 8%–10% of EVT. The isolated cells were immortalized by retroviral expression of SV40 large antigen, and single-cell cloning techniques were employed, which revealed that the cells with an epithelial morphology were the only type present in the population after 6 mo of culture. These cells did not show tumorigenicity either in vitro or in vivo. The cloned population expresses hCG, ßhCG, and CSF1 but does not express markers for decidualized endometrial cells, macrophages, and natural killer cells. TCL1 cells were positive for cytokeratin, while none of the cells were stained for vimentin. In addition, TCL1 cells constitutively express gelatinase-A, but gelatinase-B enzyme was expressed only when cultured in the presence of ECM, which is a specifically restricted phenotype of the invading cytotrophoblast [22–24]. JEG3 [25] cells were establish from human choriocarcinoma. Cells were cultured in RPMI1640 (Nipro, Tokyo, Japan) supplemented with 10% fetal calf serum (FCS; Gibco Invitorogen Corporation, Carlsbad, CA) or conditioned serum in a humidified atmosphere containing 5% CO2 at 37°C.
Detection of DNA Fragmentation
TCL1 cells were harvested and incubated in lysis buffer (0.5 mM Tris-HCl, pH 7.5, 20 mM EDTA, pH 8.0, 0.5% Triton X-100). Total cellular DNA was isolated by phenol extraction, incubated for 30 min in the presence of RNase (10 mg/ml) as described previously [26], and electrophoresed in a 2.0% agarose gel in buffer (0.04 M Tris acetate, 1 mM EDTA). The gel was visualized using ethidium bromide.
Detection of Apoptotic Nuclei
Coverslips used for ECM coating were purchased from Matsunami (Kishiwada, Japan) and were precleaned using aqua regia. Coverslips were then coated with 20 µg/ml of ECM for 1 h at room temperature. The coverslips were incubated with 10 mg/ml of BSA in PBS overnight at 4°C. After rinse, 2 x 105 TCL1 cells were seeded on coverslips coated with ECM and incubated overnight at 37°C. After washing twice with PBS, TCL1 cells incubated for an indicated time (range 0–24 h) with TNF (at the indicated concentrations (range 0–100 pg/ml) were added to complete the media containing 10% FCS. The cells were fixed with 100% methanol and stained with Hoechst 338502 (Molecular Probes). After washing twice, the cells were mounted onto slide glasses with shielding buffer. The cells were analyzed using a fluorescent microscope (Leica, Solms, Germany). Fragmented nuclei were counted using a microscope (100x). More than 500 cells (including both apoptotic and normal cells) were examined.
Adhesion Assays of TCL1 Binding to ECM and Integrin Adhesion Receptors
Adhesion assays of TCL1 binding to ECM were performed as described previously [27]. In brief, TCL1 cells were detached with trypsin-EDTA and then allowed to recover from the trypsinization in Dulbecco modified eagle medium (DMEM) containing 10% FCS with 25 µg/ml cyclohexamide for 20 min. After the cells were washed twice with serum-free medium containing 25 µg/ml cyclohexamide, 5 x 105 TCL1 cells were incubated in suspension with serum-free DMEM containing 25 µg/ml cyclohexamide and anti-integrin mAbs as described above (5 µg/ml) for an additional 20 min. Then the cells were seeded on coverslips coated with collagen type I, type IV, laminin, or fibronectin and incubated for 1 h in the same suspension as described above. Finally, cells that had adhered to coverslips were manually counted using a microscope (100x) in at least five different fields.
Spreading Assays of TCL1 Binding to ECM and Integrin Adhesion Receptors
Spreading assays of TCL1 binding to ECM were performed as described previously [24]. In brief, TCL1 cells were detached with trypsin-EDTA and then allowed to recover from the trypsinization in DMEM containing 10% FCS for 20 min. After the cells were washed twice with serum-free medium, 5 x 105 cells were incubated in suspension with serum-free DMEM for an additional 20 min. The cells were then seeded on coverslips coated with ECM or anti-integrin antibodies and incubated for 6 h in the same suspension as described above. Finally, spread cells were manually counted using a microscope (100x). More than 500 cells (including both attached and spread cells) were examined.
SDS-PAGE and Immunoblotting
Cells were lysed in buffer containing 62.5 mM Tris-HCl (pH 6.8), 100 mM dithiothreitol, 2%(w/v) SDS, and 10% glycerol. Cellular proteins were electrophoresed in an SDS 5–20% gradient polyacrylamide slab gel together with prestained molecular weight markers (BioRad Laboratories, Hercules, CA), transferred onto Immobilon-P (Millipore, Bedford, MA), and then analyzed by immunoblotting as described previously [28].
Immunofluorescence
Exponentially growing TCL1 cells (2 x 105) were seeded on coverslips coated with collagen type 1. After incubation overnight, the cells were incubated with 100 pg/ml TNF for 24 h. The cells were then fixed with 4% paraformaldehyde for 10 min and permeabilized with PBS containing 0.5% Triton-X. After blocking with 3% BSA for 30 min, the cells were incubated with an anti-ß1 antibody (K20) overnight at 4°C followed by incubation with FITC-conjugated secondary antibody for 45 min at room temperature. After washing twice, the cells were mounted onto slide glasses with shielding buffer. The cells were analyzed using a confocal fluorescent-microscope (Leica).
Statistical Analysis
Statistical analysis was performed using a Mann-Whitney test in GraphPad Prism (GraphPad Software, Inc., San Diego, CA).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Induces Apoptosis in Human Immortalized Trophoblast Cells
To analyze the effect of cytokines on cellular proliferation in human immortalized trophoblast cells, we incubated TCL1 cells, a cell line established from human preterm placenta that expresses a specifically restricted phenotype of the invading cytotrophoblast [21] with cytokines. As shown in Figure 1a, the relative numbers of cells incubated with 0, 20, and 100 pg/ml of TNF for 24 h were 1.95 ± 0.21, 1.84 ± 0.11, and 0.92 ± 0.09, respectively. At 48 h of incubation, there were 2.63 ± 0.11, 1.88 ± 0.18, and 0.82 ± 0.20 cells, respectively. These data show that growth of TCL1 cells is suppressed when incubated with 20 pg/ml of TNF. When incubated with 100 pg/ml of TNF, the TCL1 cells did not grow at all whereas TNF could not suppress growth of JEG3 cells derived from human choriocarcinoma. Incubation with 10 ng/ml of IL-1ß, 100 ng/ml of LIF, or 10 ng/ml of TGFß did not affect TCL1 cell growth.
|
To address whether TNF induced apoptotic cell death in TCL1 cells, we extracted cellular DNA and performed gel electrophoresis. As shown in Figure 1b, DNA fragmentation was detected after 6 h of incubation with 100 pg/ml of TNF whereas this oligonucleosomal fragmentation was not observed in cells incubated with 20 pg/ml of TNF.
After 6 h of incubation, we observed that TCL1 cells rounded up and detached from the dishes. To evaluate the time course of TNF-induced apoptosis in TCL1 cells, nuclear staining with Hoechst was performed in TCL1 cells incubated with 100 pg/ml of TNF (Fig. 2a). After 6 h of incubation, nuclear fragmentation was apparent in about 30% of TCL1 cells. After 8 h of incubation, 60% of TCL1 cells showed fragmented nuclei. Taken together, our data suggest that TNF plays an important role in proliferation and viability in TCL1 cells.
|
Extracellular Matrix Components Suppress TNF-Induced Apoptosis
To investigate whether ECM could affect this TNF-induced apoptosis in trophoblastic cells, we seeded TCL1 cells on coverslips coated with either collagen type I (C1), collagen type IV (C4), or poly-L-lysine (PL) and incubated them with TNF. As shown in Figure 3a, TCL1 cells attached and spread when seeded on C1 (not shown) and C4, while TCL1 cells attached but did not spread on PL-coated coverslips (Fig. 3a, upper panel).
|
We then incubated TCL1 cells seeded on ECM-coated coverslips with 100 pg/ml of TNF. After 8 h of incubation, 48.3% ± 5.1% of TCL1 cells seeded on PL-coated coverslips showed the typical phenotype of apoptotic cell death, including cellular shrinkage and nuclear fragmentation, while only 8.0% ± 3.2% and 7.5% ± 2.7% of TCL1 cells seeded on C1 and C4 showed apoptotic cell death (P < 0.01) (Fig. 3a, lower panel; Fig. 3b). The time course of apoptotic cell death on PL-coated coverslips was almost identical to that of TCL1 cells grown on culture dishes (Figs. 2b and 3b).
Next we investigated the suppressive effect of other ECM components on TNF-induced apoptosis in TCL1 cells (Fig. 3c). At 8 h of incubation, the percentage of apoptotic cells when seeded on C1, C4, Ln, Fn, or PL was 7.7% ± 5.6%, 9.1% ± 1.7%, 15.2% ± 6.1%, 14.6% ± 7.0%, and 37.6% ± 6.5%, respectively. Among ECM components examined, TNF-induced apoptosis was significantly suppressed when cells were seeded on C1, C4, Ln, and Fn but not on PL (P < 0.01).
To clarify the signaling pathway of this suppressive activity, we investigated the effect of kinase inhibitors. As shown in Figure 3d, TCL1 cells were seeded on C4 and incubated for 8 h with 100 pg/ml of TNF with 200 nM of wortmannin, a specific PI3 kinase inhibitor, or 25 µM of PD98059, a drug that specifically inhibits MAP kinase. The percentage of TNF-induced apoptotic cells grown on C4 and PL was 9.4% ± 5.5% and 28.0% ± 6.9%, respectively. The percentage of TNF-induced apoptotic cells grown on C4 when incubated with PD98059 or wortmannin was 8.5% ± 3.7% and 23.3% ± 5.5%, respectively, indicating that PD98509 significantly suppressed TNF-induced apoptosis but wortmannin did not (P < 0.01).
Thus, these data suggest that Ln, C1, C4, and Fn suppress TNF-induced apoptosis in TCL1 cells and that the PI3 kinase pathway is involved in this suppressive effect of type I collagen.
TCL1 Cells Express 5, 6, ß1, and ß4 Integrin Subunits
To address whether this survival signal against TNF-induced apoptosis is mediated via integrin adhesion receptors, we determined the integrin subunits expressed on TCL1 cells using spreading assays with ECM-coated coverslips (Table 1, upper column). The percentage of spread cells seeded on C1, C4, Ln, Fn, vitronectin (Vn), and PL was 45.8%, 39.7%, 41.0%, 38.7%, 6.7%, and 3.5%, respectively. These results are consisted with the apoptosis-suppression assay in Figure 3a.
|
To determine the integrin subunits expressed on TCL1 cells, we performed spreading assays using antibodies against and ß integrin subunits (Table 1, lower column). The percentage of spread cells seeded on 1, 2, 3, 5, 6, ß1, ß4, and IgG was 2.9%, 2.5%, 2.7%, 3.8%, 13.3%, 16.5%, 15.2%, and 16.7%, respectively, indicating that TCL1 cells adhere to integrin 5-, 6-, ß1-, and ß4-coated coverslips but not to 1-, 2-, and 3-coated coverslips (P < 0.01). These results suggest that exponentially growing TCL1 cells express integrin 5, 6, ß1, and ß4 subunits.
To clarify which subunits are actually involved in adhesion to ECM, we performed adhesion-blocking assays in TCL1 cells as described in Materials and Methods. As shown in Figure 4a, the number of adhered cells to Ln was 104.2 ± 27.1. The number of adhered cells when incubated with anti-ß4, -6ß4, -6, -ß1, and -6ß1 antibodies was 72.8 ± 18.2, 54.2 ± 20.4, 28.6 ± 17.3, 9.2 ± 5.1, and 1.8 ± 2.0, respectively. Anti-ß1 antibodies significantly blocked adhesion to Ln compared with anti-ß4 antibodies (P < 0.01). The number of adhered cells to C1 was 112.4 ± 16.2. The number of adhered cells when incubated with anti-6, -ß1, and -6ß1 antibody was 66.6 ± 28.8, 2.4 ± 2.2, and 0.4 ± 0.5, respectively (Fig. 4b). The number of adhered cells to C4 was 135.0 ± 19.7. The number of adhered cells when incubated with anti-6, -ß1, and -6ß1 antibodies was 79.8 ± 8.8, 21.8 ± 5.2, and 2.5 ± 2.3, respectively (Fig. 4c). The number of adhered cells to Fn was 122.8 ± 30.8. The number of adhered cells when incubated with anti-5, -ß1, -5ß1 antibodies and RGD peptides was 64.8 ± 22.9, 26.2 ± 9.5, 34.2 ± 8.7, and 44.4 ± 14.1, respectively (Fig. 4d).
|
These results suggest that TCL1 cells adhere to C1, C4, and Ln via the 6ß1 integrin adhesion receptor and to Fn via the 5ß1 receptor. These results are in contrast with reports based on immunohistological data suggesting that 6ß4 is the dominant receptor mediating adhesion between Ln and invading CT [9, 10].
TNF-Induced Integrin Switching in TCL1 Cells
Our results suggest that signaling via integrin adhesion receptor molecules plays an important role as a survival signal to escape from TNF-induced apoptosis in TCL1 cells. To examine the effect of TNF on integrin expression, we performed Western blotting analysis using anti-integrin subunit antibodies in TCL1 cells incubated with TNF. As the spreading assay showed, integrins 5, 6, ß1, and ß4 but not 1 were expressed in exponentially growing TCL1 cells. Surprisingly, after incubation with high concentrations of TNF, TCL1 cells ceased to express 6 whereas 1 expression was induced after 24 h of incubation (Fig. 5a). We could not detect 5 integrin subunits in immunoblotting of TCL1 cells.
|
To confirm the effect of TNF on integrin function, we stained TCL1 cells seeded on C1 with an anti-ß1 integrin subunit monoclonal antibody, K20. As shown in Figure 5b, aggregation of the integrin ß1 subunits was seen after incubation with TNF for 24 h. On the contrary, exponentially growing TCL1 cells did not show these aggregations even when seeded on C1. TNF appeared to strengthen integrin adhesive functions.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
6ß4 integrin subunits and, as they invade, they express 1ß1 subunits. The distribution of particular ECM components also varies throughout the decidual tissue. Ln is rich at superficial areas that face the villous trophoblast. On the contrary, C4 and Fn are more abundant at deeper sites [9, 10]. In fact, EVT alters their integrin expression in parallel with the varied ECM distribution during invasion. Thus, it has been suggested that alteration of integrin expression during the differentiation of EVT may be responsible for mediating distinct signals to EVT upon adhesion to ECM. In addition, it has been suggested that abnormal EVT invasion and aberrant integrin expression are often seen in abnormal pregnancies, especially in preeclampsia. CTs in preeclampsia increase expression of the 5ß1 but fail both to upregulate 1ß1 and to downregulate 6ß4 [10, 14]. This aberrant integrin expression may contribute to the inability of EVT to invade the uterus in these conditions [11–14]. Therefore, it is possible that regulation of integrin subunit expression is involved in regulation of normal EVT differentiation and is essential for normal pregnancy. However, the mechanism that induces alterations in integrin expression in these cells is still unclear.
Here we have demonstrated that TNF downregulates 6 and upregulates 1 subunit expression in immortalized human trophoblast cells. Moreover, TNF induces aggregation of ß1 integrin. These results indicate that TNF regulates integrin expression and function in TCL1 cells. CTs, when grown in Matrigel, a mixture of extracellular matrix components and growth factors, express an integrin repertoire similar to that expressed by CTs within the uterus. Moreover, CTs can be induced to switch their complement of integrins by exposure to a mixture of purified Ln and C4 [10]. Certain integrin signals may also facilitate exit from the cell cycle and provide signals for differentiation [4]. However, integrin signals alone may only be permissive for differentiation. For example, differentiated functions of primary mammary epithelial cells require both adhesion to basement membrane and exposure to lactogenic hormones [29]. In the present experiment, TNF could induce integrin subunit conversion without ECM. It remains unclear, however, whether ECM alone can induce integrin conversion. It is also possible that CTs provide a soluble, autocrine factor that induces subunit conversion in the presence of ECM. CTs can produce many cytokines, including TNF [30].
The other issue is that TCL1 cells were established from the EVT in the third trimester. The EVT invasion with integrin subunit conversion is observed already in the first trimester [10]. During the second trimester of pregnancy, modulation of trophoblast integrin expression during the invasion process was nearly identical to the pattern observed during the first trimester. Although EVT was stained like those of the second trimester, there was one notable difference between the second trimester and term staining patterns. Staining for the 6 subunit was once again prominent in the third trimester, whereas staining for ß4 was not detected [11]. It should be noted that TCL1 cells predominantly expressed the 6ß4 subunit. Thus, TCL1 cells have an integrin repertoire different from those of the third trimester. TCL1 cells constitutively express gelatinase-A, but only express the gelatinase-B enzyme when cultured in the presence of ECM, like BeWo cells derived from choriocarcinoma, which has invasive properties [22]. The expression of the gelatinase-B enzymes is restricted to discrete cell types in the first trimester, at which time trophoblast cells display a highly invasive phenotype [22–24]. BeWo, TCL1, and primary trophoblasts secret CSF1 and express its receptor [22]. The CSF autocrine loop is seen in many trophoblast cell lines that show invasive properties [31, 32]. In addition, TCL1 cells show a tubelike formation on Matrigel, indicating that TCL1 cells have an ability to differentiate to endothelium (Fukushima, unpublished manuscript). More than this evidence, the pattern of integrin subunit conversion that TCL1 showed is consisted with those of EVT in histochemical studies. Therefore, it could be assumed that TCL1 cells provide at least a potential model to elucidate functional differentiation in EVT cells in vitro. As TNF did not induce apoptosis in choriocarcinoma cells, it has remained unclear whether our results are specific for this particular cell line or are a general phenomenon in EVT. Although TCL1 cells are very similar to EVT, our experiments should be verified in the cells from the first trimester and also in vivo.
Many studies have suggested that programmed cell death plays an important role in pregnancy [13, 33]. Indeed, there are many reports of abnormal apoptosis observed in many abnormal conditions of pregnancy such as miscarriage, preeclampsia, and fetal growth restriction [34, 35]. TNF is a proinflammatory cytokine with a wide range of biological effects and plays a crucial role in immune and inflammatory processes as well as in the pathogenesis of many human diseases [18, 19]. As well as induction of programmed cell death, TNF is known to control gene transcription and cellular differentiation through NF-
B activation [36]. We have demonstrated that TNF induces growth suppression at lower concentrations and apoptosis at higher concentrations in TCL1 cells, established from human EVT. We have previously reported that certain extrinsic circulating factors are cytotoxic to TCL1 cells in preeclampsia [37]. It is plausible that TNF is a candidate for such a cytotoxic factor in preeclampsia. However, many cytokines and growth factors in addition to TNF are abundant at the feto-maternal interface; thus, they could also be involved in the pathogenesis of preeclampsia.
There has been, however, much suggestive evidence that TNF is involved in preeclampsia. Serum TNF levels are elevated in preeclamptic patients [20] and expression of placental TNF is increased in preeclampsia [21]. Therefore, it is possible that abnormal levels of TNF in placenta induce impaired trophoblast integrin expression, which consequently leads to the failure of trophoblast differentiation. TNF is also abundant in decidual tissue and reportedly induces apoptosis in human trophoblast cells [38]. Decidual cells, fibroblasts, inflammatory cells, and trophoblast cells can secrete TNF and express TNF receptors [17, 39]. TNF might also act on these cells to induce changes in ECM secretion. Switching from 6 to 1 subunit expression may help cells adapt to this environmental change. Our results may provide a basis for biological communication between the maternal microenvironment and fetal trophoblast cells. As an immunological defending response, maternal cells may secrete TNF. Although this would induce elimination of some of the trophoblasts by apoptosis, at the same time, ECM could serve as a survival factor for trophoblasts through integrin signaling if they could convert their phenotype successfully. Because integrins can transduce signals through the cell membrane in either direction, inside-out and outside-in signaling [4, 40], they may be well suited to serve an important role in feto-maternal reciprocal signaling.
In summary, in this article, we have shown that TNF induces apoptosis and integrin switching in a human EVT cell line. In addition, this integrin switching seems to be involved in responding to survival signals from ECM because ß1 subunit aggregation was induced. These results suggest that both TNF and ECM might collaboratively regulate biological behavior of trophoblast cells through integrin signaling in establishing and maintaining successful pregnancy.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: Shingo Miyamoto, Department of Obstetrics and Gynecology, Graduate School of Medical Sciences, Kyushu University, Maidashi 3-1-1, Higashi-ku, Fukuoka 812-8582, Japan. FAX: 81-92-642-5414; smiya{at}gynob.med.kyushu-u.ac.jp
Received: 29 August 2002.
First decision: 1 October 2002.
Accepted: 5 December 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
levels before the clinical manifestations of preeclampsia. Am J Reprod Immunol 1997 38:89-93, interleukin 1ß and interleukin 10 is increased in preeclampsia. Am J Obstet Gynecol 1999 181:915-920[CrossRef][Medline]This article has been cited by other articles:
|
|
 |
|
  E. S. Taglauer, A. S. Trikhacheva, J. G. Slusser, and M. G. Petroff Expression and Function of PDCD1 at the Human Maternal-Fetal Interface Biol Reprod, September 1, 2008; 79(3): 562 - 569. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Fukushima, S. Miyamoto, K. Tsukimori, H. Kobayashi, H. Seki, S. Takeda, E. Kensuke, K. Ohtani, M. Shibuya, and H. Nakano Tumor Necrosis Factor and Vascular Endothelial Growth Factor Induce Endothelial Integrin Repertories, Regulating Endovascular Differentiation and Apoptosis in a Human Extravillous Trophoblast Cell Line Biol Reprod, July 1, 2005; 73(1): 172 - 179. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, control; 
, 20 pg/ml of TNF
, 100 pg/ml of TNF
, 10 ng/ml of TGFß; 
, 100 ng/ml of LIF, 
, 10 ng/ml of IL-1ß for TCL1 cells; 
, JEG3cells. b) DNA fragmentation. Asynchronously growing TCL1 cells were seeded at 1 x 106/100-mm dish at 37°C. After 1 day of incubation, the medium was replaced with fresh, complete medium containing 10% FBS and the indicated concentrations of TNF
H HaeIII digested
