The Role Of ERK Signal Transduction In The Invading Process Of Trophoblasts

Successful human placentation requires that extravillous cytotrophoblasts (EVCT) rapidly invade genetically dissimilar maternal tissues during the early stages of pregnancy. EVCT differentiate from cytotrophoblasts (CTB). Although EVCTs behave like cancer cells, in vivo they are only transiently invasive (first trimester) and their invasion is normally limited only to the endometrium and to the proximal third of the myometrium. The invasion of trophoblasts is precisely controlled but errors can happen and might have fatal consequences under some circumstances. For example, shallow invasion is associated with preeclampsia and about fifty percent of the pregnancies with preeclampsia are complicated by intrauterine growth retardation. In contrast, over aggressive invasion (e.g. deeper than the proximal third of the myometrium or beyond the extent of myometrium) is associated with placental site tumors, choriocarcinoma, and placenta accreta. Therefore, it is very important to search for the reasons how the invasion of trophoblasts is controlled and it will help to find the way to prevent and treat diseases associated with abnormality of the invading process.It is very likely that trophoblast differentiation along the invasive pathway is a result of the balanced action of growth factors and cytokines on trophoblasts and surrounding environment. Several cytokines and growth factors are believed to regulate the temporal and spatial invasion of EVCTs in an autocrine way by trophoblastic factors and in a paracrine way by uterine factors. The uterine microenvironment (deciduas, particularly) plays an important role in the control oftrophoblast invasion. These motile and highly invasive EVCTs, which are also referred as intermediate trophoblasts, are found to be cytokeratin positive in the decidua, the intima of uterine blood vessels and the proximal third of the myometrium. The decidua itself provides not only some factors to regulate invasion ability of EVCTs but also extracellular matrix (ECM) to maintain the microenvironment for EVCT growth. Moreover, integrins exist on trophoblasts and the ECM might play a very important role for the interaction of ECM and trophoblasts. But, so far it is not elucidated fully how the integrins affect the invasion of trophoblasts. Focal adhesion plaques (FAP) formed by ECM and intracellular skeleton proteins by integrins are the structure foundation of integrin signal transduction systems. The focal adhesion kinase (FAK) is firstly found in FAP, and play a key function of the integrin signal transduction systems. A key component of signal transduction systems in organisms, extracellular signal regulating protein kinase (ERK) is considered to be involved in many cellular processes and pathogenesis of many severe diseases. According to our study, ERK could play an important role in the regulation of invasion of trophoblasts. Recent research in other labs indicates the FAK may regulate migration and invasion of trophoblasts. Now, it is clear that many integrins could activate ERK either in FAK-dependent or FAK-independent fashion. In some tumor cells, ECM could activate FAK and switch on ERK signal transduction by integrins. Therefore, it is possible that above-mentioned signal transduction pathways might play a role in the regulation of the invasion of trophoblasts and the aim of this thesis research project was to test this hypothesis.PART I: ESTABLISHMENT OF A MODEL (IN VITRO) OF INVASION OF EXTRAVILLOUS CYTOTROPHOBLASTSOBJECTIVETo establish a model of invasion of extravillous cytotrophoblasts (EVCT) in vitro.METHODS1. Isolation and purification of EVCTs. EVCTs were prepared as previously described with minor modifications. Briefly, placental tissue from Human was washed in Ca2+- and Mg2+-free Hanks balanced salt solution supplemented with 100 IU/ml penicillin and 120 \ig/m\ streptomycin. Chorionic villi were dissected, the blood vessels and clots were carefully removed, and then rinsed. The tissue was digested with 0.125% trypsin, and 50 Kunitz/ml Dnase type IV in Hanks solution containing 4.2 mM MgSO4 and 25 mM Hepes, for 35 min at 37°C without agitation. After sedimentation, supernatants were collected and the remaining tissue was rinsed four times with Hanks solution. The supernatants were then pooled and filtered (100 |im pore size) and trypsin activity was neutralized with 10% FCS. Cells were washed once in Hanks and separated by a discontinuous Percoll gradient centrifuge. Furthermore, the EVCT preparation was purified by plating in plastic dishes. After incubation overnight in humidified air containing 5% CO2 at 37°C, villous trophoblasts, macrophages and fibroblasts adhered to the plastic dishes. Floating cells consisted mainly of EVCTs. EVCTs were centrifuged, and then the pellet was resuspended in F12/DMEM-10% FCS and plated in Matrigel-coated culture dishes.2. CK7, TGF|32 and integrin al(31 were detected by cellular immunofluorescence in EVCTs after they were cultured on the Matrigel for 4 hours.3. The growth of EVCTs was determined by CCK-8. 2xlO4/well EVCTs plating in Matrigel-coated 96 well culture plates for 0, 24, 48, and 72 hours, and then the growth of EVCT was determined with CCK-8.4. Transwell invasion system was used to observe the invasion of the EVCTs. EVCTs were plated on Matrigel-coated filters (lxlO5 cells) in 24 Transwell invasion system for 24, 48 and 72 hours.RESULTS1.After 4h of plating, cells were identified as EVCTs by immunofluorescence staining with CK7, TGFP2, integrin al and integrin pi antibody. The positive cells for immunostaining with CK7 and TGFP2 antibody accounted for more than 95% of overall cell counted. Some cells also expressed integrin al and pisubunits. In theseculture conditions, EVCTs rapidly acquired the ability to invade.2.EVCTs were cultured for different time (0, 24, 48 and 72 hours) on Matrigel, and their growth indexes were 100,104+10,102+6 and 99+7 for these 3 different culture time, respectively.3.EVCTs were also cultured in transwell invasion system for 3 different times (24, 48 and 72 hours) and their OD of invasive cells were 0.125+0.008, 0.163+0.010 and 0.157+0.006, respectively.CONCLUSIONEVCT invasion model was successfully established in vitro in our lab, and it could be used to research the invasion mechanisms of trophoblasts in the future.PART II: EFFECTS OF ERK SIGNAL TRANSDUCTION PATHWAY INTROPHOBLAST INVASIONOBJECTIVETo test the effects of ERK signal transduction on the invasion of EVCTs.METHODS1. The growth of EVCT was determined by CCK-8. 2xlO4/well EVCTs plating in Matrigel-coated 96 well culture plates were cultured for 24 or 48 hours with either 0.1%DMSO, different concentration of integrin al and integrin pi combined antibody (0, 0.5, 1, and 5ug/ml), herbimycin A (0, 0.2, 0.5, 1, and 2 ug/ml) or PD98059 (0, 2, 10, 25, and 50uM). CCK-8 was used to determine the growth of EVCTs.2. The immunoprecipitation and/or Western blot were used to determine the phosphorylation of FAK and ERK.A) The EVCTs were cultured on Matrigel for different time (0, 1, 2, 4, and 8 hours) to test the effects of Matrigel on the phosphorylation of FAK and ERK.B) The EVCTs were cultured with different concentration of integrin al and integrin pi combined antibody (0, 0.5, 1, and 5 fig/ml), herbimycin A (0, 0.2, 0.5, 1, and 2 ug/ml), a tyrosine kinase inhibitor or 3) PD98059 (0, 2, 10, 25, and 50 uM), a ERK1/2 specific inhibitor, for 4 hours. Then, total cellular proteins of EVCTs wereextracted. FAK was immunoprecipitated by FAK antibody, and phosphorylation of FAK was determined with anti-FAK phosphorylation specific antibody by Western blot. The phosphorylation of ERK1/2 was directly detected with specific anti-ERKl/2 phosphorylation antibody by Western blot.3.Transwell invasion system was used to observe the invasion of EVCTs. EVCTs were plated on Matrigel-coated filters (lxlO5 cells) in 24 Trans well invasion system with either PBS, 0.1 % DMSO, lug/ml integrin al and integrin pi combined antibody, 1 ng/ml herbimycin A or 25uM PD98059 for 48 hours. Then, the invasion of EVCTs was determined with CCK-8.RESULTSl.The growth of EVCTs was not affected by either 0.1 % DMSO, integrin al and integrin pi combined antibody, herbimycin A or PD98059.2. When EVCTs were cultured on Matrigel, phosphorylation of FAK and ERK time-dependently increased and it reached its plateau after 4 hours.3. The phosphorylation of FAK and ERK in EVCTs dose-dependently decreased with integrin al and integrin pi combined antibody. However, integrin al and integrin pi combined antibody could not completely inhibit phosphorylation of FAK and ERK1/2.4. The phosphorylation of FAK in EVCTs dose-dependently decreased with herbimycin A (between 0 and 1 jxg/kg). With 1 fig/ml of herbimycin A, the phosphorylation of FAK in EVCTs reached its lowest point. The same dose of herbimycin A also partially inhibited phosphorylation of ERK 1/2.5. With PD98059, phosphorylation of ERK1/2 in EVCTs dose-dependently decreased. Phosphorylation of ERK1/2 in EVCT reached its lowest point with 25uM PD98059. But, the same dose of PD98059 did not have any effects on phosphorylation of FAK.6.0.1% DMSO did not have any impact the invasion of EVCTs, its invasion index was 97.6+5.4 (compared with control group, P>0.05). The invasion indexes of EVCTs for 1 ng/ml integrin al and integrin pi combined antibody, 1 ug/ml herbimycin A and 25uM PD98059 group were 61.8+6.9, 47.4+4.0, 32.6+4.7, respectively. The invasion of EVCTs was significantly inhibited by 1 (J-g/ml integrinal and integrin pi combined antibody, 1 |ig/ml herbimycin A and 25uM PD98059.CONCLUSION1. ECM could activate FAK and ERK by integrin alpl in EVCTs.2. The ERK signal transduction pathway could play an important role in the regulation of the invasion of EVCTs.PART III: SCREENING THE RELATED GENES OF ERK SIGNAL TRANSDUCTION PATHWAY IN TROPHOBLAST INVASIONOBJECTIVETo search the related genes of ERK signal transduction pathway in EVCTsMETHODS1. Culture and treatment of EVCTs in vitro. lxlO6/well EVCTs plating in Matrigel-coated 6 well culture plates, treated with 25uM PD98059 or 0.1%DMSO (control), and cultured for 24 hours.2.RNA was isolated from equal number of cells using a commercial reagent RNA TRIzol, according to the manufacturer's instructions. The RNA was then amplified by switching mechanism at 5'-end of RNA transcript technique (SMART). Purification of the amplified cDNA was used by spin-1000 purification column and the purified cDNA was digested by the use of Rsa I or Hae in restriction enzyme.3. Establishment of suppression subtractive hybridization. cDNA (4 ug ) of control group was divided into it into two: a 2 (xg cDNA as a Driver and the other 2 Hg with 0.2 ng 1.90). The good quality of ds cDNA was obtained from small amount of RNA by switching mechanism at 5'-end of RNA transcript technique (SMART) . Most of cDNA fragments (less than 500 bps) were removed by spin-1000 purification column. Electrophorese of 2.5 ul of undigested ds cDNA and 5 ul of Rsa I-digested cDNA on a 1% agarose/EtBr gel in IX TAE buffer was also carried out. It was found that the amplication of cDNA derived from RNA appeared as a smear from 0.5-10 kb. After Rsa I digestion, the average cDNA size is smaller (0.2-2 kb compared with 0.5-10 kb).2. After SSH, the subtractive products were amplified by PCR. The products of PCR were similar to Hae III- digested (pX174, but the only difference was the slightly heavier molecular weight. Because they were ligated adaptors. Mirror orientation selection (MOS) could substantially decrease the number of background molecules.3. A total of 1127 clones were obtained: 635 from the forward subtractive library and 492 from the reverse subtractive library. Among them, 96 clones were randomly chosen from these two libraries. After screening positive clones with the dot blot technology, 16 positive clones might be differentially expressed genes in the forward subtractive library and 14 positive clones in the reverse subtractive library.4. In the forward subtractive library, 3 clones sequencing failed, and the other 13 clones were expressed sequence tag (EST). By homology analysis, 4 ESTs wererelated with golgi SNAP receptor complex member 2, nucleolar protein 5A, IK cytokine, ribosomal protein L34 (RPL34). 1 EST was unknown gene DKFZp667J095. 5 ESTs were highly homologic with upstream regulatory element binding protein 1 (UREB1), eukaryotic translation initiation factor 2C, mitogen-activated protein kinase associated protein 1, and RAS protein activator like 2. 2 ESTs shared some part of sequence, but they were non-related genes. Another 1 EST was not found to share a part of sequence with other ESTs.5. In the reverse subtract!ve library, 1 clone sequencing failed, and the other 13 clones were EST. Among them, 3 ESTs were related with hippocampus abundant transcript 1, mitochondrial GTP-binding protein 1 (GTPBP3) gene, phospholipase A2 (or cDNA FLJ33552 fis, very similar to mitochondrial uncoupling protein4) , 1 EST was related with gene of hypothetical protein LOC253842 and 2 ESTs were highly homologic and very close to the genes for a novel protein similar to phosphoinositide-binding protein and a novel gene for a novel protein similar to oxidoreductase, phospholipase C beta 1. In the remaining 7 ESTs, 5 shared some sequences, but they were not related genes. The last 2 ESTs were not found to be similar to any other genes.CONCLUSIONl.The SSH was successfully established and it could be used to clone the differentially expressed genes.2. The cDNA subtractive library was successfully constructed and it could be useful in test the role of ERK signal transduction in the invasion of trophoblasts.