Mol. Hum. Reprod. Advance Access originally published online on February 8, 2007
Molecular Human Reproduction 2007 13(4):251-263; doi:10.1093/molehr/gal116
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Genome-wide expression profiling of placentas in the p57Kip2 model of pre-eclampsia
Genetics Department, Standford University, Stanford, CA, USA
1 To whom correspondence should be addressed at: Genetics Department, Stanford University, Stanford, CA 94062, USA. E-mail: jbaker{at}stanford.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
Pre-eclampsia affects 6–10% of pregnancies and is one of the primary causes of premature birth. It is widely accepted that inappropriate placental development, combined with environmental factors, plays a major role in disease pathogenesis. The p57Kip2 mouse is the only mouse model of pre-eclampsia that recapitulates the full spectrum of symptoms of the human disease, including placental abnormalities, hypertension, proteinuria and premature labour. In addition, pregnant females expressing wild-type levels of p57Kip2 develop pre-eclampsia when carrying fetuses that lack p57Kip2 expression. This demonstrates that either the fetus or the placenta causes the disease. Here, taking advantage of the unique genetics of the p57Kip2 mouse, we have used full genome expression profiling to define the placental aspect of the p57Kip2 phenotype at a molecular level and to conduct an unbiased search for factors involved in pre-eclampsia pathogenesis. During this analysis, we found that although mutant embryos demonstrate altered placental architecture and have histological changes indicative of reduced utero-placental blood flow, the p57Kip2 pregnant females do not demonstrate hypertension or renal pathology. This suggests a model in which placental abnormalities cause pre-eclampsia only given other environmental variables. On the basis of this model, we expect that misregulation of molecular factors, while not able to cause a full spectrum of disease symptoms in this context, still occurs in these p57Kip2 mutant mice. Our studies suggest a role for environmental factors in the p57Kip2 pre-eclampsia phenotype and have identified several candidates for pre-eclampsia predisposition in this model, including known regulators of blood pressure, inflammation and apoptosis.
Key words: gene expression profiling/mouse model of human disease/p57Kip2/placenta/pre-eclampsia
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
Pre-eclampsia, a disease characterized by hypertension and proteinuria, affects 6–10% of pregnancies and is thought to be caused by abnormal placentation (Roberts and Lain, 2002; Cross, 2003; Cecil et al., 2004; Creasy et al., 2004; Maynard et al., 2005). Current models suggest that improper placental development creates reduced placental perfusion, which subsequently causes the overproduction of toxic molecules or, alternatively, the underproduction of protective molecules by the placenta (Roberts and Lain, 2002; Cross, 2003). This model has gained strong support from the high incidence of placental abnormalities, especially placental vascular abnormalities, found in pre-eclampsic women (Roberts and Lain, 2002; Cross, 2003; Creasy et al., 2004) and from cases in which the placenta rather than the fetus has proven to be required for disease (Elmer et al., 1993; Shembrey and Noble, 1995). Although an abnormal placenta is clearly a high risk factor for pre-eclampsia, environmental factors also play roles in the frequency and onset of the disease. Obesity, hypertension, age, smoking, altitude and even month of conception have significant effects on the incidence of pre-eclampsia, emphasizing the role of maternal and environmental risk factors in disease pathogenesis (Lindqvist and Marsal, 1999; Palmer et al., 1999; Zhang et al., 1999; Cunningham and Williams, 2001; Phillips et al., 2004).
The contribution of both genetic abnormalities and environmental factors to pre-eclampsia has greatly hampered attempts to define the disease at a molecular level in humans. Furthermore, genetic studies are complicated by the possible contribution of both maternal and feto-placental factors to disease pathogenesis, as well as by significant heterogeneity within the syndrome itself. One recent study successfully identified STOX1, a ‘maternal effect gene’ located on 10q22, whose loss of function in the placenta is associated with pre-eclampsia in a large cohort of Dutch females (van Dijk et al., 2005). However, it is unlikely that mutations at this locus are responsible for the majority of human cases, as linkage analyses in other populations have mapped susceptibility to over 10 other loci (Lachmeijer et al., 2002). Case–control comparisons at the RNA and protein levels have also been difficult to interpret, since most studies rely on blood or tissue samples taken well after symptoms of pre-eclampsia have begun. Although many factors are misregulated in pre-eclampsic women compared with controls, distinguishing potentially causative changes from the mass of downstream, secondary changes that occur as a result of widespread inflammation and clotting dysregulation can be incredibly difficult. Given these challenges, many basic questions about the pathogenesis of pre-eclampsia may be best addressed in more tractable mouse model systems.
Although several mouse models mimic aspects of human pre-eclampsia, only the p57Kip2 mutant mouse faithfully recapitulates the full spectrum of symptoms. One of the most unusual aspects of the p57Kip2 model is that these mice demonstrate both abnormal placental development as well as premature labour. p57Kip2 females also exhibit increased blood pressure (BP), proteinuria and glomerular lesions, all of which are associated with the human syndrome (Takahashi et al., 2000a; Kanayama et al., 2002). In addition, p57Kip2 is paternally imprinted (Hatada and Mukai, 1995), meaning that only the maternal allele is expressed. Heterozygous females that inherit a mutant allele from their fathers, referred to here as ‘phenotypically normal heterozygotes,’ express p57Kip2 at wild-type levels. These females develop symptoms of pre-eclampsia when carrying a litter in which half of the embryos lack p57Kip2 expression. Since the female herself has wild-type levels of p57Kip2, it is clear that disease is caused by a feto-placental defect rather than a maternal defect.
We have used the unique genetics of the p57Kip2 mouse model in combination with full genome microarrays to define the p57Kip2 placental phenotype at a molecular level and to search for placental factors that are misregulated during pre-eclampsic pregnancies. This study has three unique advantages we can use to identify causative factors of pre-eclampsia. First, the p57Kip2 model allows us to disentangle the maternal and feto-placental contributions to disease because phenotypically normal heterozygote mothers develop symptoms of pre-eclampsia simply from carrying a litter that includes mutant embryos. Thus, we know that the pre-eclampsic effects are caused by feto-placental abnormalities rather than maternal abnormalities. Second, the use of a mouse model system gives us the ability to look early during pregnancy, prior to overt presentation of symptoms. This allows a focus on changes that may be causative rather than secondary. Third, the use of full genome microarray technology allows for an unbiased search for factors that are up- or down-regulated during development of the p57Kip2 mutant placentas. Therefore, to investigate the molecular basis of pre-eclampsia, we compared expression levels of over 30,000 genes in normal and p57Kip2 mutant sibling placentas beginning at embryonic day 12 (e12.0), prior to the reported onset of symptoms, and continuing to e14.5 and e17.0, just prior to initiation of preterm labour. This study has molecularly defined the p57Kip2 placental phenotype and has identified candidate genes that may be involved in disease pathogenesis in the p57Kip2 model.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
Mice
The p57Kip2 knockout mice used in this study were a kind gift from the group of K. Nakayama at Kyushu University. C57BL/6 mice from Charles River Laboratories (Wilmington, MA, USA) were used as breeding partners to expand the colony. Genotypes were determined by PCR of tail DNA samples, using primers to the mutant and wild-type p57Kip2 alleles as previously described (Takahashi et al., 2000b). Since p57Kip2 is paternally imprinted, there is no expression from the paternal allele. Heterozygotes that inherit a mutant allele from their mothers lack p57Kip2 expression and have a genotype notated as +P/–M, where the ‘M’ and ‘P’ superscripts refer to the maternal (expressed) and paternal (non-expressed) alleles, respectively. Heterozygotes that inherit a mutant allele from their fathers have wild-type p57Kip2 levels, with a genotype notated as +M/–P. For all experiments, p57Kip2 +M/–P females were bred to p57Kip2 +M/–P males. This breeding scheme generated a total of seven pregnancies, three of which were used for microarray studies. Embryos from all pregnancies were staged according to Theiler's morphological criteria and are referred to here by the corresponding embryonic day (Theiler, 1989).
Feto-placental dissection and genotyping
Pregnant female mice were sacrificed at e12.0, e14.5 and e17.0. At each stage, the entire litter of fetuses and placentas was dissected. At e12.0, the decidua was removed from fetally derived portions of the placenta, whereas at e14.5 and e17.0 the decidua was left intact. Placental samples were flash frozen in liquid nitrogen immediately after dissection and were stored at –80°C until RNA isolation. For several litters, the placenta was cut in half, with one half flash frozen and the second fixed overnight in 4% paraformaldehyde. Samples of fetal tissue were used to determine feto-placental genotypes. Genotyping was conducted using PCR, with primers for mutant and wild-type p57Kip2 alleles as previously described (Takahashi et al., 2000b). Expression of p57Kip2 was determined by RT–PCR analysis of placental tissue samples, again using primers for wild-type p57Kip2 transcript. Following genotyping and RT–PCR analysis, seven placentas from each timepoint were selected for microarray analysis: four wild-type and three mutant placentas at e12.0 and e17.0 and three wild-type plus four mutant placentas at e14.5.
Array sample preparation and hybridization
RNA was isolated from each placental sample individually. Each placenta was homogenized in Trizol with a rotor homogenizer. RNA was isolated by chloroform extraction from the Trizol sample, purified using a Qiagen (Valencia, CA, USA) RNeasy column and then ethanol precipitated using glycogen as a carrier. For each sample, 10 µg of starting RNA was used to prepare labelled cRNA according to Affymetrix (Santa Clara, CA, USA) protocols. Briefly, RNA was reverse transcribed and the resulting double-stranded cDNA was purified by phenol–chloroform extraction followed by ethanol precipitation. The cDNA was used as a template for biotin labelling in vitro transcription. The resulting cRNA was purified using an RNeasy column and then quantitated. For each sample, 20 µg of cRNA was fragmented using Potassium Acetate (KOAc) RNA fragmentation buffer and then hybridized to an Affymetrix mouse 430 2.0 microarray using standard hybridization procedures (www.affymetrix.com). The hybridization was conducted by the Stanford Protein and Nucleic Acid Biotechnology Facility. All arrays from each timepoint were processed and hybridized together to avoid possible batching artifacts.
Microarray data analysis
In order to control for differences in overall strength of hybridization, all arrays were normalized to the same median intensity using the dCHIP invariant set method (Li and Wong, 2001). Modelling was performed using first the dCHIP perfect match (PM) modelling algorithm, and second using the dCHIP Perfect Match/Mismatch (PM/MM) modelling algorithm. Although there are differences in the expression values generated by the two methods, the results of our multi-stage mutant versus wild type comparison were similar to the two sets, as shown in Table I. While Table I includes data generated from both PM and PM/MM modelling algorithms, all subsequent analyses presented in this article are based on PM modelling only. Prior to comparative analysis, gene lists were filtered to eliminate probe sets that are not expressed above background levels. This was done by including only those probe sets for which at least one array had an absolute expression value over 200 and at least one array was called ‘present’ by the dCHIP software.
|
Identification of differences between mutant and wild-type placentas at each timepoint, and across all timepoints for the multi-stage comparison, was conducted using statistical analysis of microarrays (SAM) (Tusher et al., 2001). Each group includes a minimum of three samples, providing the statistical power to detect consistent changes over a broad range of fold changes. SAM identifies up- and down-regulated genes and, importantly, uses a permutation-based approach to calculate a false discovery rate (FDR) for each analysis. For each gene, SAM generates a fold change and a q-value. The q-value for a given gene represents the percentage of genes predicted by SAM to be false positive when the stringency (delta value) is set to include that gene. SAM was used to generate lists of up- and down-regulated genes, and a 10% FDR threshold was set by only considering those genes with a q-value <10. Therefore, 1 out of every 10 of the identified up- and down-regulated genes is expected to be a false positive. A fold-change threshold of 1.50 was used for the stringent, global comparison, whereas a threshold of 1.25 was used for the stage-specific analyses. Because the SAM q-value calculation uses a random permutation-based approach, the q-values calculated for a given gene will vary somewhat between individual analyses. The gene lists presented here were generated from a representative, if somewhat conservative, SAM comparative analysis. The full, unfiltered data set is freely available on the lab website (http://www.stanford.edu/group/bakerlab/).
BP measurements
Systolic BP measurements were measured with a tail cuff using the Visitech BP-2000 (Apex, NC, USA). BP measurements were recorded only once the mice appeared calm, and any measurement interrupted by movement was eliminated from the analysis. Each BP measurement is reported as the mean of a minimum of five valid measurements taken in one session. With the exception of one mouse, all mice were trained on the BP machine in the 2–4 weeks preceding pregnancy, with each mouse undergoing a minimum of five training sessions. One mouse underwent only a single training session prior to pregnancy, but measurements taken throughout pregnancy demonstrated high consistency, so are included in Figure 2. To avoid cuff-to-cuff variability, all measurements for an individual mouse were taken using the same BP cuff. To account for the cuff-to-cuff and mouse-to-mouse baseline variability, BP values for all mice were scaled to the same mean pre-pregnancy systolic BP. Specifically, the average pre-pregnancy systolic BP was calculated for each mouse, and the average of these pre-pregnancy values was taken across all mice. For each mouse, the ratio of the group average pre-pregnancy BP to the individual mouse average pre-pregnancy BP was used to scale all systolic BP readings.
|
Tissue samples and histology
Kidney samples were taken from pregnant p57Kip2 +M/–P females sacrificed at e12.0, e16.5 or e17.0. Kidneys were fixed overnight in 4% paraformaldehyde, dehydrated in methanol and stored at –80°C until embedded in paraffin wax. Sections were stained with haematoxylin and eosin, PAS or trichrome. Haematoxylin and eosin and PAS stains were used for assessment of overall kidney morphology, and trichrome was used to detect fibrin deposition. Embedding, sectioning and staining were performed by the Stanford Pathology Department Histology Services. Additional control kidneys were taken from a p57Kip2 +/+ (Rb+/–) mouse at e15.0 and processed as above. Placenta samples were taken from pregnant p57Kip2 +M/–P females sacrificed at e16.5 or e17.0. Placentas were cut in half, fixed overnight in 4% paraformaldehyde, dehydrated in ethanol and stored at –80°C until embedded in paraffin wax. Embedding, sectioning and haematoxylin and eosin staining were performed by the Stanford Pathology Department Histology Services. Additional wild-type placental samples were taken at e17.0 and e10.0 for use in RNA in-situ hybridization. These samples were fixed for 3–4 h in 4% paraformaldehype, infused with sucrose and embedded in OCT prior to sectioning.
RNA in situ hybridization
A digoxigenin labelled riboprobe was generated from a linearized plasmid for p57Kip2 (gift of Dr Mark Krasnow's laboratory at Stanford University). In-situ hybridization for p57Kip2 was conducted at 55°C overnight on e17.0 and e10.0 wild-type placental sections using standard procedures. Digoxigenin signal was detected using a Roche (Indianapolis, IN, USA) anti-digoxigenin alkaline phosphatase conjugated antibody. Staining was performed for 4–6 h using Roche (Indianapolis) NBT and BCIP and couterstaining was done with Biomeda (Foster City, CA, USA) nuclear fast red.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
Imprinting in the p57Kip2 mouse allows isolation of mutant and wild-type sibling placentas from phenotypically normal +M/–P females
Our goal is to use p57Kip2 mutant and sibling wild-type placentas to define the molecular changes that occur in mutant placentas and to identify placental gene expression changes that may predispose to pre-eclampsia. Since p57Kip2 is paternally imprinted, there is no expression from the paternal allele (Hatada and Mukai, 1995). Heterozygote pups that inherit a mutant allele from their mothers have a genotype notated as +P/–M, with the superscript ‘P’ indicating the non-expressed paternally inherited allele and the superscript ‘M’ indicating the expressed maternally inherited allele. These pups die soon after birth. Heterozygote pups that inherit a mutant allele from their fathers (+M/–P), however, are phenotypically normal since they have a wild-type maternal expressed allele (Takahashi et al., 2000b). In this study, we mated phenotypically normal p57Kip2 +M/–P female mice to phenotypically normal +M/–P male mice, as previously described (Kanayama et al., 2002). The resulting pregnant females carry
25% +/+ embryos, 25% +M/–P embryos, 25% +P/–M embryos and 25% –/– embryos. It was previously demonstrated that p57Kip2 +M/–P females display the symptoms of pre-eclampsia, including increased BP beginning at e10.0–e12.0, increases in urinary protein excretion by e13.5, clotting factor dysregulation, increased endothelin 1 levels and renal lesions by e17.5 (Kanayama et al., 2002). Therefore, we chose three timepoints that span the course of disease: e12.0, at the onset of BP changes; e14.5, when BP has risen to mid-range levels; e17.0, when BP peaks and renal lesions have been reported. At each of these timepoints, we dissected a full litter of placentas and fetuses. Fetal tissue samples were used to determine feto-placental genotypes. As expected, +/+, +/– and –/– genotypes were obtained at approximately Mendelian ratios (Figure 1A). To determine the number of placentas lacking p57Kip2 expression (–/– and +P/–M) versus the number with wild-type levels of p57Kip2 (+/+ and +M/–P), we isolated placental RNA and conducted RT–PCR on each of the samples using primers for wild-type p57Kip2 transcript. As expected, approximately half of the placentas in each litter showed wild-type levels of p57Kip2 expression (+/+ and +M/–P), whereas half lacked the p57Kip2 expression entirely (–/– and +P/–M) (Figure 1B).
|
Phenotypic heterogeneity among p57Kip2 mice suggests a role for environmental factors in disease pathogenesis
To evaluate the spectrum of pre-eclampsia symptoms demonstrated by p57Kip2 mice, we systematically analysed the pregnant p57Kip2 +M/–P females for BP changes and renal abnormalities. The mice used in this study were previously reported to have placental abnormalities as well as dramatic BP increases and renal pathology in response to pregnancy (Kanayama et al., 2002). Our p57Kip2 female mice carried litter sizes ranging from five to nine, with RT–PCR and microarray experiments confirming that approximately half of each litter lacked expression of p57Kip2. Although the mutant placentas are visibly abnormal by e16.5–e17.0, in our hands, the pregnant p57Kip2 females do not exhibit consistent BP increases nor do they demonstrate altered renal histology. The BP of p57Kip2 +M/–P female mice carrying non-expressing p57Kip2 mutant embryos showed a wide range of variability, with no increase over the course of pregnancy (Figure 2). To further investigate the BP discrepancy, we followed two pregnant p57Kip2 heterozygous females throughout pregnancy and for several days after they gave birth at e19.5. Although Kanayama et al. (2002) saw BP rise 35 mmHg during pregnancy and then fall dramatically to baseline levels within 2 days after parturition, our p57Kip2 females had slight but inconsistent BP increases during pregnancy and no decline in the 3–5 days following parturition (Figure 2E and F). Furthermore, kidney samples were taken from three p57Kip2 +M/–P female mice mated to p57Kip2 +M/–P males and sacrificed at e16.5–e17.0. No differences were observed between these mice and a p57Kip2 +M/–P female sacrificed at e12.0 or a p57Kip2 +/+ female sacrificed at midgestation (Figure 3). Kanayama et al. (2002) described diffuse glomerular enlargement as well as an increase in subendothelial and mesangial deposits in the glomerular capillaries of p57Kip2 heterozygous females. We used a panel of haematoxylin and eosin, PAS and trichrome stains, yet did not observe any such abnormalities in glomeruli of these pregnant females (Figure 3). Although we cannot completely rule out a strain-specific effect, we received the p57Kip2 mice directly from the Kanayama lab and followed their published breeding protocol, using C57BL/6 as the parent strain. Therefore, given that these mice demonstrated and continue to demonstrate symptoms of pre-eclampsia in other laboratories (K. Takahashi and N. Kanayama, 2005 personal communication), the phenotypic heterogeneity we observe suggests that environmental factors may play an important role in disease pathogenesis in p57Kip2 mice. These factors could include diet, water, housing conditions, air filtration, temperature, seasonality and overall level of stress.
|
p57Kip2 mutant placental histology is consistent with a role in pre-eclampsia pathogenesis
In order to characterize the placental phenotype in our mice, we compared +/+ and –/– placentas from p57Kip2 +M/–P mothers at e17.0. In addition to gross morphological defects, the mutant placentas demonstrate a broad spectrum of histological changes compared with wild-type sibling placentas (Figure 4). Consistent with results reported by Takahashi et al. (2000a) and Kanayama et al. (2002), our p57Kip2 mutant placentas demonstrate a clear disruption of normal architecture as well as abnormal fibrin deposition. Rather than generalized placentamegaly, however, we see a marked decrease in the thickness of the labyrinth layer, a common finding in human pre-eclampsic placentas (Kraus et al., 2004), and evidence of extensive calcifications, indicative of decreased utero-placental blood flow (Sternberg et al., 2004). In addition, we see evidence of infarction, fibrin extravasation and fibrinoid necrosis throughout the labyrinth of mutant placentas. Increased numbers of nucleated RBCs are also identified (Figure 4, inset), again suggestive of pre-eclampsic or hypoxic placental tissue (Kraus et al., 2004); degenerating cells are seen both in the decidua and the spongiotrophoblast. Expression of p57Kip2 has previously been reported in the labyrinth and spongiotrophoblast layers of wild-type placentas at e17.5 (Takahashi et al., 2000a). Given the broad range of changes demonstrated by our p57Kip2 mutant placentas, we used in-situ hybridization to confirm this finding and to investigate the expression of p57Kip2 earlier, at e10.0. We generated an antisense digoxigenin labelled riboprobe from mouse p57Kip2 cDNA and hybridized to e17.0 and e10.0 wild-type placenta sections. As shown in Figure 5, at e17.0, there is intense staining in the labyrinth as well as in clusters of cells within the spongiotrophoblast layer, consistent with results previously reported by Takahashi et al. (2000a). At e10.0, there is staining of the labyrinth and spongiotrophoblast layers, as well as the overlying trophoblast giant cells. No staining was seen using control probes that are not predicted to have placental expression (data not shown).
|
|
p57Kip2 mutant placentas demonstrate consistent expression differences from wild-type placentas
Our goal is to define the placental aspect of the p57Kip2 phenotype at a molecular level and to identify molecules whose misregulation may be involved in pre-eclampsia pathogenesis in the p57Kip2 mouse. Our p57Kip2 mice demonstrate significant disruption of placental architecture and a broad spectrum of histological changes suggestive of decreased utero-placental blood flow, yet do not exhibit the full spectrum of pre-eclampsia symptoms. This suggests a model in which placental abnormalities cause pre-eclampsia only given other environmental variables. According to this model, we would expect that misregulation of molecular factors, although not able to cause a full spectrum of symptoms in this environmental context, still occurs in these p57Kip2 mutant mice. Thus, this study specifically targets molecules whose misregulation may predispose towards pre-eclampsia. To identify such molecules, we used Affymetrix full genome mouse microarrays (430 2.0) to compare expression levels of over 30 000 genes between mutant placentas and wild-type sibling placentas. In a first stringent analysis, we grouped the data generated from all placental samples (e12.0, e14.5, e17.5) to identify genes consistently up- and down-regulated throughout gestation in the mutant placentas (Table I). Normalization, modelling and baseline filtering were conducted using dCHIP software (Li and Wong, 2001) and comparisons were made using SAM (Tusher et al., 2001), as described in Materials and methods. As shown in Table I, the q-value for a given gene represents the percentage of genes predicted by SAM to be false positive when the stringency is set to include that gene. In this case, we wanted a FDR of <10% and so included only those genes with a q-value of <10.
Using this stringent approach and an average fold-change threshold of 1.50, we identified five probe sets that are consistently down-regulated and seven probe sets that are consistently up-regulated in mutant placentas when compared with wild type (Table I). Among the up-regulated genes are two genes involved in inflammation. The first, hepcidin antimicrobial peptide, is a negative regulator of iron absorption, which is produced in response to inflammation or iron overload (Ganz, 2003). This gene appears more than once in the list, indicating that it is represented by more than one probe set on the array. The second, toll-like receptor 4, is a pattern-recognition receptor involved in innate immunity, whose activation leads to increased expression of genes coding for proinflammatory, proapoptotic and antimicrobial peptides (Akira et al., 2001; Kiechl et al., 2003). The down-regulated genes include resistance to inhibitors of cholinesterase 8 homologue (Ric-8, also known as synembryn) and inner centromere protein (Incenp), both genes with potential connections to cell cycle dysregulation caused by loss of p57Kip2. Ric-8 is a guanine nucleotide exchange factor involved in asymmetric cell division in Caenorhabditis elegans and Drosophila, with a specific role demonstrated in mitotic spindle alignment (Miller and Rand, 2000; Afshar et al., 2004; Couwenbergs et al., 2004; Wang et al., 2005). Incenp is a chromosomal passenger protein involved in chromosome segregation and required for the metaphase–anaphase transition (Cutts et al., 1999; Goto et al., 2006; Kaitna et al., 2000).
Mutant placentas demonstrate stage-specific misregulation of molecules involved in BP regulation, inflammation and apoptosis
Although grouping the data as discussed above provides stringent criteria to find molecules misregulated throughout all stages of development in the mutant placentas, we were also interested in identifying more subtle stage-specific differences between mutant and wild-type placentas. The numbers of genes up- and down-regulated in mutants compared with wild type, using an FDR threshold of 10% and fold-change boundaries of 3.0, 2.0, 1.5 and 1.25, are shown in Table II. These genes provide a molecular profile of the p57Kip2 placental phenotype. In accordance with the increasing severity of the morphological phenotype over time, the molecular abnormalities become more significant as development progresses. Given the large number of significant changes that occur at later timepoints, we used Gene Ontology terms to characterize down-regulated genes at e14.5 and both up- and down-regulated genes at e17.0. We found that the up- and down-regulated lists contained genes involved in BP regulation, inflammation and apoptosis, all processes that are implicated in pathogenesis of pre-eclampsia (Figure 6). Genes annotated to each of these categories are shown in Table III, along with fold changes seen in mutant placentas compared with wild type. Of particular interest among the genes up-regulated at e17.0 are the vasoconstrictor molecule endothelin 2, the proinflammatory molecules chemokine (C-X-C motif) ligand 1 and macrophage migration-inhibitory factor and the proapoptotic molecules Bcl2-modifying factor and BCL2-associated transcription factor 1. We also used the NIH DAVID tool (Dennis et al., 2003) to look for over-representation of specific Biological Process GO terms within the lists. We found slight over-representation (P < 0.05) of the Biological Process GO terms ‘steroid metabolism’, ‘mitotic cell cycle’ and ‘cell cycle’ in the genes up-regulated at e17.0, of the term ‘morphogenesis’ in genes down-regulated at e17.0 and of terms related to polysaccharide, carbohydrate and alcohol metabolism in genes down-regulated at e14.5. Given the role of p57Kip2 in cell cycle regulation, we were not surprised to find cell-cycle-related genes represented in all 3 of the gene lists: 15 in the e17.0 up-regulated category, 4 in the e17.0 down-regulated category and 3 in the e14.5 down-regulated category.
|
|
|
The top 10 up- and down-regulated probe sets found at each stage in the stage-specific comparison are shown in Tables IV and V. Again, the expected FDR is <10%, since all q-values are <10. Using this FDR threshold and a fold-change threshold of 1.25, four probe sets besides p57Kip2 itself are significantly down-regulated at all three timepoints. Two of these (1438949_at and 1438950_x_at) represent ric-8 (resistance to inhibitors of cholinesterase 8 homologue). These probe sets are shown in Table V for e12.0 and e14.5. At e17.0, the same two probe sets have fold changes of 1.88 (1438949_at) and 1.76 (1438950_x_at), but do not make the list of top 10 fold changes. The other two probe sets are 1423093_at (CTD phosphatase, subunit 1) and 1428214_a (translocase of outer mitochondrial membrane 7 homologue). These probe sets have fold changes of 1.42 and 1.37, respectively, at e14.5, and fold changes of 1.69 and 1.52 at e17.5. A full list of all significant changes is available in the supplementary material.
|
|
Flt1, Ren1 and Vegfa are expressed at wild-type levels in p57Kip2 mutant placentas
In looking for molecules that might predispose to pre-eclampsia, we were interested not only in general processes (BP regulation, inflammation and apoptosis), but also in specific previously identified candidate genes. Although numerous genes have been found to be up- or down-regulated in pre-eclampsic mothers compared with controls, most of these changes are likely to be secondary rather than causative. Therefore, we focused specifically on three candidates, Flt1, Ren1 and Vegf, that have been shown to cause pre-eclampsic symptoms when injected into wild-type rodents or overexpressed in transgenic mice (Takimoto et al., 1996; Maynard et al., 2003; Murakami et al., 2005). None of the three genes was identified by SAM as having significantly altered expression in mutant placentas, using a 10% FDR and a fold-change cutoff of 1.25. By plotting expression of Ren1 (one probe set), Flt1 (four probe sets) and Vegfa (two probe sets) over time, we see no trend towards increased expression in mutant placentas (Figure 7). Previous studies of the p57Kip2 mouse in a disease context have demonstrated a 2.5-fold increase in Vegfa mRNA levels and a 9-fold increase in Vegf 164 protein levels in mutant p57Kip2 placentas at e17.5 (Matsuura et al., 2002). Although these data indicate that pre-eclampsia predisposition in this model is not due to misregulation of Ren1, Flt1 or Vegfa, the discrepancy in Vegfa levels between disease and non-disease contexts suggests a role for environmental factors in misregulation of Vegfa in mutant placentas.
|
p57Kip2 mutant placentas maintain wild-type levels of important pregnancy-related hormones
Given the dramatic changes that occur in p57Kip2 mutant placentas by late gestation (Figure 4; Takahashi et al., 2000a; Kanayama et al., 2002), we were interested in whether these placentas were able to maintain wild-type function. As one way of assessing function, we looked at hormones that are secreted by the placenta during pregnancy. As shown earlier in Table V, we see that there are significant expression differences found between mutant and wild-type placentas for some hormones. However, looking over time, we see that overall p57Kip2 mutant placentas are able to maintain approximately wild-type expression profiles for many pregnancy-related hormones (Figure 8). These include well-studied hormones known to be crucial to maintenance of pregnancy, including placental lactogen 1 (csh1) and placental lactogen 2 (csh2), as well as newly discovered prolactin gene family members whose functions have not yet been elucidated (Wiemers et al., 2003). Production of wild-type hormone levels suggests that p57Kip2 mutant placentas maintain at least baseline functionality despite a significant disruption of placental architecture and a broad spectrum of histological changes suggestive of decreased utero-placental blood flow.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
Environmental factors in the p57Kip2 pre-eclampsia phenotype
Given that factors as diverse as obesity, age, smoking, altitude and even month of conception have been shown to affect pre-eclampsia risk in humans (Lindqvist and Marsal, 1999; Palmer et al., 1999; Zhang et al., 1999; Cunningham and Williams, 2001; Phillips et al., 2004), it is not surprising that the p57Kip2 pre-eclampsia phenotype is sensitive to environmental variables. In fact, the importance of environmental variables on induction of pre-eclampsia-like symptoms in rodents is well documented: in one study, cold stress stimulation of rats was sufficient to cause a pre-eclampsia-like syndrome including hypertension and proteinuria (Kanayama et al., 1997), whereas in a follow-up study, it was found that fasting stress alone, cold stress alone or a combination of the two induced biochemical and histological changes similar to those seen in human pre-eclampsia (Khatun et al., 1999). Possibilities for environmental differences affecting pre-eclampsia phenotype in the p57Kip2 mice include diet, water, housing conditions, air filtration, temperature, seasonality and overall level of stress.
Given a model in which additional environmental stressors are required for the disease phenotype in p57Kip2 mice, we would expect that misregulation of placental factors, although not able to cause full pre-eclampsia in this environmental context, still occurs in these p57Kip2 mutant mice. Therefore, this study targets molecules that can initiate symptoms of pre-eclampsia given a predisposing environmental background. This has consequences for our interpretation of changes observed at different stages of gestation. In a disease context, any causative factor made by mutant placentas is likely to be misregulated early, near the onset of pre-eclampsia symptoms, as observed by Kanayama et al. (2002) at approximately e12.0. However, in an environmental context that does not lead to a clinical phenotype, misregulation of predisposing factors could occur early, as predicted in the disease context, or could be delayed. In our search for predisposing factors, therefore, we are interested in changes at later timepoints (e14.5 and e17.0) as well as changes occurring early (e12.0).
Potential roles for genes involved in BP regulation, inflammation and apoptosis in pre-eclampsia predisposition
In this study, we have found that p57Kip2 mutant placentas demonstrate significant up- or down-regulation of transcripts coding for a variety of molecules involved in BP regulation, inflammation and apoptosis. Each of these processes has been implicated in pathogenesis of pre-eclampsia (Roberts and Lain, 2002; Creasy et al., 2004; Neale and Mor, 2005; Redman and Sargent, 2005). Two genes involved in BP regulation were found to be up-regulated in mutant placentas at e17.0. Of these, endothelin 2 is of particular interest as the endothelin 2 protein has been demonstrated to cause BP increases when injected into rats (Saida et al., 1989). One attractive possibility is that low-level endothelin 2 up-regulation, although not sufficient to cause increased BP in this context, is aggravated in the context of additional environmental stressors.
Of the up-regulated molecules involved in inflammation, two are secreted proinflammatory molecules and one is a proinflammatory transcription factor. The first secreted molecule, chemokine (C-X-C motif) ligand 1 (Cxcl1, also known as N51/KC), has been shown to act in vivo as a neutrophil chemoattractant (Lira et al., 1994; Tani et al., 1996). The second, macrophage migration-inhibitory factor (Mif), is a proinflammatory cytokine that is rapidly released in response to microbial products or other inflammatory stimuli; of interest, it has also been shown to cause up-regulation of toll-like receptor 4 (Roger et al., 2001; Calandra and Roger, 2003). Nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor, zeta (Nfkbiz or Mail), is a transcription factor that is induced in response to toll-like receptor ligands and IL-1 and is an important part of the inflammatory response (Yamamoto et al., 2004; Motoyama et al., 2005). Additionally, it has been shown that at least part of the Nfkbiz induction seen in response to LPS is mediated by Tlr4 (Kitamura et al., 2002). Toll-like receptor 4 itself, although not found to be up-regulated using PM modelling, was found to be consistently up-regulated across all timepoints using PM/MM modelling (Table I). Up-regulation of these proinflammatory molecules in mutant placentas suggests a possible role for placental inflammatory mediators in predisposition to pre-eclampsia in the p57Kip2 model.
The potential role of apoptosis in pre-eclampsia is 2-fold, with both mechanisms resulting in exaggerated maternal immune responses to fetoplacental tissues (Neale and Mor, 2005). First, increased apoptosis in the placenta could lead to increased release of apoptotic material from the trophoblast into the maternal circulation. Second, decreased apoptotic susceptibility of activated maternal T cells at the maternal–fetal interface could prevent normal deletion of immune cells that recognize fetal antigens. Four of the seven proteins encoded by genes up-regulated in p57Kip2 mutant placentas, Bcl2-modifying factor (Bmf) (Puthalakath et al., 2001), BCL2-associated transcription factor 1 (Bclaf1) (Kasof et al., 1999), Cd27-binding protein (Siva) (Yoon et al., 1999) and phorbol-12-myristate-13-acetate-induced protein 1 (Pmaip or Noxa) (Oda et al., 2000) have pro-apoptotic activity, suggesting increased apoptosis in p57Kip2 mutant trophoblast as one possible mechanism leading to pre-eclampsia predisposition. One protein, heat shock protein 1B, is anti-apoptoic, and the other two are electronically annotated or have inconsistent annotation. Of the six proteins encoded by genes down-regulated in p57Kip2 mutant placentas, one, microphthalmia-associated transcription factor (Widlund et al., 2002), is anti-apoptotic and two, IGFPB-3 and Granzyme B, are proapoptotic (Pardo et al., 2004; Lee et al., 2005). The others are electronically annotated or have inconsistent annotation.
Misregulation of genes involved in the cell cycle
p57Kip2 is a member of the CIP/KIP family of cyclin-dependent kinase inhibitors and has been shown to inhibit a variety of cyclin-dependent kinase/cyclin complexes (Lee et al., 1995; Matsuoka et al., 1995; Lee and Yang, 2001). In p57Kip2 mutants, therefore, we expect to see some misregulation of the cell cycle. Two of the most consistently down-regulated genes in our data set are Ric-8 (also known as synembryn) and inner centromere protein (Incenp). Although it is unlikely that these genes play a direct role in pre-eclampsia predisposition, their consistent down-regulation in mutant placentas strongly suggests a link to p57Kip2. Both of these genes have connections to the cell cycle. Incenp is a chromosomal passenger protein involved in chromosome segregation and cytokinesis, whose activity is required for the metaphase–anaphase transition (Cutts et al., 1999; Kaitna et al., 2000; Goto et al., 2006). Ric-8 is a guanine nucleotide exchange factor involved in asymmetric cell division in Caenorhabditis elegans and Drosophila, with a specific role demonstrated in mitotic spindle alignment (Miller and Rand, 2000; Afshar et al., 2004; Couwenbergs et al., 2004; Wang et al., 2005). In the mouse, ric-8 is found in the developing nervous system (Tonissoo et al., 2003); homozygous ric-8 mutant mice die early in embryonic development, and heterozygous mice exhibit decreased anxiety and impaired spatial memory (Tonissoo et al., 2006). Further supporting the possibility of cell cycle changes in the p57Kip2 mutants is the relatively high number of cell-cycle-related genes found to be significantly altered in our stage-specific comparisons. We identified 15 genes annotated with the Gene Ontology process term ‘cell cycle’ in the e17.0 up-regulated category, 4 in the e17.0 down-regulated category and 3 in the e14.5 down-regulated category.
|
Conclusion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
Pre-eclampsia affects
6–10% of pregnancies (Cecil et al., 2004) and is a leading cause of maternal mortality worldwide (Roberts et al., 2003). It is responsible for 15% of premature births and causes a 5-fold increase in perinatal mortality (Meis et al., 1998; Roberts et al., 2003). The underlying causes of pre-eclampsia have proven incredibly difficult to elucidate. Here, taking advantage of the unique genetics of the p57Kip2 mouse, we have used full genome microarrays to define the placental aspect of the p57Kip2 phenotype at a molecular level and to conduct an unbiased search for factors that may pre-dispose to pre-eclampsia. These studies suggest a role for environmental factors in the p57Kip2 pre-eclampsia phenotype and have identified several candidates for pre-eclampsia predisposition in this model.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
We would like to thank Dr Andrew Patterson in the Stanford University Department of Anesthesia for use of the BP 2000 for mouse BP measurement and Rani Agrawal for her help with the BP 2000. We also appreciate the assistance of Dr Neeraja Kambham of the Stanford University Pathology Department with histopathological evaluation of kidney and placental samples. Additional thanks to Eric Chiao and all the members of the Baker lab for support and advice throughout the project. The p57Kip2 mice used in this study were a kind gift from the group of K. Nakayama at Kyushu University. This work was supported by the March of Dimes. K. K. support provided by Stanford University Medical Scientist Training Program.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionConclusionAcknowledgementsReferences |
|---|
Afshar K, Willard FS, Colombo K, Johnston CA, McCudden CR, Siderovski DP, Gonczy P. (2004) RIC-8 is required for GPR-1/2-dependent Galpha function during asymmetric division of C. elegans embryos. Cell 119:219–230.[CrossRef][Web of Science][Medline]
Akira S, Takeda K, Kaisho T. (2001) Toll-like receptors: critical proteins linking innate and acquired immunity. Nat Immunol 2:675–680.[CrossRef][Web of Science][Medline]
Calandra T and Roger T. (2003) Macrophage migration inhibitory factor: a regulator of innate immunity. Nat Rev Immunol 3:791–800.[CrossRef][Web of Science][Medline]
Cecil RL, Goldman L, Ausiello DA. (2004) Cecil Textbook of Medicine 22nd edn (Saunders, Philadelphia, PA).
Couwenbergs C, Spilker AC, Gotta M. (2004) Control of embryonic spindle positioning and Galpha activity by C. elegans RIC-8. Curr Biol 14:1871–1876.[CrossRef][Web of Science][Medline]
Creasy RK, Resnik R, Iams JD. (2004) Maternal-fetal Medicine 5th edn (W.B. Saunders Co., Philadelphia, PA).
Cross JC. (2003) The genetics of pre-eclampsia: a feto-placental or maternal problem? Clin Genet 64:96–103.[CrossRef][Web of Science][Medline]
Cunningham FG and Williams JW. (2001) Williams Obstetrics 21st edn (McGraw-Hill, New York).
Cutts SM, Fowler KJ, Kile BT, Hii LL, O'Dowd RA, Hudson DF, Saffery R, Kalitsis P, Earle E, Choo KH. (1999) Defective chromosome segregation, microtubule bundling and nuclear bridging in inner centromere protein gene (Incenp)-disrupted mice. Hum Mol Genet 8:1145–1155.
Dennis G, Sherman BT, Hosack DA, Yang J, Gao W, Lane HC, Lempicki RA. (2003) DAVID: Database for Annotation, Visualization, and Integrated Discovery. Genome Biol 4:P3.[CrossRef][Medline]
Elmer DB, Granai CO, Ball HG, Curry SL. (1993) Persistence of gestational trophoblastic disease for longer than 1 year following evacuation of hydatidiform mole. Obstet Gynecol 81:888–890.[Web of Science][Medline]
Ganz T. (2003) Hepcidin, a key regulator of iron metabolism and mediator of anemia of inflammation. Blood 102:783–788.
Goto H, Kiyono T, Tomono Y, Kawajiri A, Urano T, Furukawa K, Nigg EA, Inagaki M. (2006) Complex formation of Plk1 and INCENP required for metaphase-anaphase transition. Nat Cell Biol 8:180–187.[CrossRef][Web of Science][Medline]
Hatada I and Mukai T. (1995) Genomic imprinting of p57KIP2, a cyclin-dependent kinase inhibitor, in mouse. Nat Genet 11:204–206.[CrossRef][Web of Science][Medline]
Kaitna S, Mendoza M, Jantsch-Plunger V, Glotzer M. (2000) Incenp and an aurora-like kinase form a complex essential for chromosome segregation and efficient completion of cytokinesis. Curr Biol 10:1172–1181.[CrossRef][Web of Science][Medline]
Kanayama N, Takahashi K, Matsuura T, Sugimura M, Kobayashi T, Moniwa N, Tomita M, Nakayama K. (2002) Deficiency in p57Kip2 expression induces preeclampsia-like symptoms in mice. Mol Hum Reprod 8:1129–1135.
Kanayama N, Tsujimura R, She L, Maehara K, Terao T. (1997) Cold-induced stress stimulates the sympathetic nervous system, causing hypertension and proteinuria in rats. J Hypertens 15:383–389.[CrossRef][Web of Science][Medline]
Kasof GM, Goyal L, White E. (1999) Btf, a novel death-promoting transcriptional repressor that interacts with Bcl-2-related proteins. Mol Cell Biol 19:4390–4404.
Khatun S, Kanayama N, Belayet HM, Masui M, Sugimura M, Kobayashi T, Terao T. (1999) Induction of preeclampsia like phenomena by stimulation of sympathetic nerve with cold and fasting stress. Eur J Obstet Gynecol Reprod Biol 86:89–97.[CrossRef][Web of Science][Medline]
Kiechl S, Wiedermann CJ, Willeit J. (2003) Toll-like receptor 4 and atherogenesis. Ann Med 35:164–171.[CrossRef][Web of Science][Medline]
Kitamura H, Kanehira K, Shiina T, Morimatsu M, Jung BD, Akashi S, Saito M. (2002) Bacterial lipopolysaccharide induces mRNA expression of an IkappaB MAIL through toll-like receptor 4. J Vet Med Sci 64:419–422.[CrossRef][Web of Science][Medline]
Kraus FT, Redline RW, Gersell DJ, Nelson DM, Kicke JM. (2004) Placental Pathology(American Registry of Pathology, Washington, D.C.).
Lachmeijer AM, Dekker GA, Pals G, Aarnoudse JG, ten Kate LP, Arngrimsson R. (2002) Searching for preeclampsia genes: the current position. Eur J Obstet Gynecol Reprod Biol 105:94–113.[CrossRef][Web of Science][Medline]
Lee KW, Ma L, Yan X, Liu B, Zhang XK, Cohen P. (2005) Rapid apoptosis induction by IGFBP-3 involves an insulin-like growth factor-independent nucleomitochondrial translocation of RXRalpha/Nur77. J Biol Chem 280:16942–16948.
Lee MH, Reynisdottir I, Massague J. (1995) Cloning of p57KIP2, a cyclin-dependent kinase inhibitor with unique domain structure and tissue distribution. Genes Dev 9:639–649.
Lee MH and Yang HY. (2001) Negative regulators of cyclin-dependent kinases and their roles in cancers. Cell Mol Life Sci 58:1907–1922.[CrossRef][Web of Science][Medline]
Li C and Wong WH. (2001) Model-based analysis of oligonucleotide arrays: expression index computation and outlier detection. Proc Natl Acad Sci U S A 98:31–36.
Lindqvist PG and Marsal K. (1999) Moderate smoking during pregnancy is associated with a reduced risk of preeclampsia. Acta Obstet Gynecol Scand 78:693–697.[CrossRef][Web of Science][Medline]
Lira SA, Zalamea P, Heinrich JN, Fuentes ME, Carrasco D, Lewin AC, Barton DS, Durham S, Bravo R. (1994) Expression of the chemokine N51/KC in the thymus and epidermis of transgenic mice results in marked infiltration of a single class of inflammatory cells. J Exp Med 180:2039–2048.
Matsuoka S, Edwards MC, Bai C, Parker S, Zhang P, Baldini A, Harper JW, Elledge SJ. (1995) p57KIP2, a structurally distinct member of the p21CIP1 Cdk inhibitor family, is a candidate tumor suppressor gene. Genes Dev 9:650–662.
Matsuura T, Takahashi K, Nakayama K, Kobayashi T, Choi-Miura NH, Tomita M, Kanayama N. (2002) Increased expression of vascular endothelial growth factor in placentas of p57(Kip2) null embryos. FEBS Lett 532:283–288.[CrossRef][Web of Science][Medline]
Maynard SE, Min JY, Merchan J, Lim KH, Li J, Mondal S, Libermann TA, Morgan JP, Sellke FW, Stillman IE, et al. (2003) Excess placental soluble fms-like tyrosine kinase 1 (sFlt1) may contribute to endothelial dysfunction, hypertension, and proteinuria in preeclampsia. J Clin Invest 111:649–658.[CrossRef][Web of Science][Medline]
Maynard SE, Venkatesha S, Thadhani R, Karumanchi SA. (2005) Soluble Fms-like tyrosine kinase 1 and endothelial dysfunction in the pathogenesis of preeclampsia. Pediatr Res 57:1R–7R.
Meis PJ, Goldenberg RL, Mercer BM, Iams JD, Moawad AH, Miodovnik M, Menard MK, Caritis SN, Thurnau GR, Bottoms SF, et al. (1998) The preterm prediction study: risk factors for indicated preterm births. Maternal-Fetal Medicine Units Network of the National Institute of Child Health and Human Development. Am J Obstet Gynecol 178:562–567.[CrossRef][Web of Science][Medline]
Miller KG and Rand JB. (2000) A role for RIC-8 (Synembryn) and GOA-1 (G(o)alpha) in regulating a subset of centrosome movements during early embryogenesis in Caenorhabditis elegans. Genetics 156:1649–1660.
Motoyama M, Yamazaki S, Eto-Kimura A, Takeshige K, Muta T. (2005) Positive and negative regulation of nuclear factor-kappaB-mediated transcription by IkappaB-zeta, an inducible nuclear protein. J Biol Chem 280:7444–7451.
Murakami Y, Kobayashi T, Omatsu K, Suzuki M, Ohashi R, Matsuura T, Sugimura M, Kanayama N. (2005) Exogenous vascular endothelial growth factor can induce preeclampsia-like symptoms in pregnant mice. Semin Thromb Hemost 31:307–313.[CrossRef][Web of Science][Medline]
Neale DM and Mor G. (2005) The role of Fas mediated apoptosis in preeclampsia. J Perinat Med 33:471–477.[CrossRef][Web of Science][Medline]
Oda E, Ohki R, Murasawa H, Nemoto J, Shibue T, Yamashita T, Tokino T, Taniguchi T, Tanaka N. (2000) Noxa, a BH3-only member of the Bcl-2 family and candidate mediator of p53-induced apoptosis. Science 288:1053–1058.
Palmer SK, Moore LG, Young D, Cregger B, Berman JC, Zamudio S. (1999) Altered blood pressure course during normal pregnancy and increased preeclampsia at high altitude (3100 meters) in Colorado. Am J Obstet Gynecol 180:1161–1168.[CrossRef][Web of Science][Medline]
Pardo J, Bosque A, Brehm R, Wallich R, Naval J, Mullbacher A, Anel A, Simon MM. (2004) Apoptotic pathways are selectively activated by granzyme A and/or granzyme B in CTL-mediated target cell lysis. J Cell Biol 167:457–468.
Phillips JK, Bernstein IM, Mongeon JA, Badger GJ. (2004) Seasonal variation in preeclampsia based on timing of conception. Obstet Gynecol 104:1015–1020.[CrossRef][Web of Science][Medline]
Puthalakath H, Villunger A, O'Reilly LA, Beaumont JG, Coultas L, Cheney RE, Huang DC, Strasser A. (2001) Bmf: a proapoptotic BH3-only protein regulated by interaction with the myosin V actin motor complex, activated by anoikis. Science 293:1829–1832.
Redman CW and Sargent IL. (2005) Latest advances in understanding preeclampsia. Science 308:1592–1594.
Roberts JM and Lain KY. (2002) Recent Insights into the pathogenesis of pre-eclampsia. Placenta 23:359–372.[CrossRef][Web of Science][Medline]
Roberts JM, Pearson G, Cutler J, Lindheimer M. (2003) Summary of the NHLBI Working Group on Research on Hypertension During Pregnancy. Hypertension 41:437–445.
Roger T, David J, Glauser MP, Calandra T. (2001) MIF regulates innate immune responses through modulation of Toll-like receptor 4. Nature 414:920–924.[CrossRef][Medline]
Saida K, Mitsui Y, Ishida N. (1989) A novel peptide, vasoactive intestinal contractor, of a new (endothelin) peptide family. Molecular cloning, expression, and biological activity. J Biol Chem 264:14613–14616.
Shembrey MA and Noble AD. (1995) An instructive case of abdominal pregnancy. Aust N Z J Obstet Gynaecol 35:220–221.[Web of Science][Medline]
Sternberg SS, Mills SE, Carter D. (2004) Sternberg's Diagnostic Surgical Pathology 4th edn (Lippincott Williams & Wilkins, Philadelphia).
Takahashi K, Kobayashi T, Kanayama N. (2000a) p57(Kip2) regulates the proper development of labyrinthine and spongiotrophoblasts. Mol Hum Reprod 6:1019–1025.
Takahashi K, Nakayama K, Nakayama K. (2000b) Mice lacking a CDK inhibitor, p57Kip2, exhibit skeletal abnormalities and growth retardation. J Biochem (Tokyo) 127:73–83.
Takimoto E, Ishida J, Sugiyama F, Horiguchi H, Murakami K, Fukamizu A. (1996) Hypertension induced in pregnant mice by placental renin and maternal angiotensinogen. Science 274:995–998.
Tani M, Fuentes ME, Peterson JW, Trapp BD, Durham SK, Loy JK, Bravo R, Ransohoff RM, Lira SA. (1996) Neutrophil infiltration, glial reaction, and neurological disease in transgenic mice expressing the chemokine N51/KC in oligodendrocytes. J Clin Invest 98:529–539.[Web of Science][Medline]
Theiler K. (1989) The House Mouse: Atlas of Embryonic Development(Springer-Verlag, New York).
Tonissoo T, Koks S, Meier R, Raud S, Plaas M, Vasar E, Karis A. (2006) Heterozygous mice with Ric-8 mutation exhibit impaired spatial memory and decreased anxiety. Behav Brain Res 167:42–48.[CrossRef][Web of Science][Medline]
Tonissoo T, Meier R, Talts K, Plaas M, Karis A. (2003) Expression of ric-8 (synembryn) gene in the nervous system of developing and adult mouse. Gene Expr Patterns 3:591–594.[CrossRef][Medline]
Tusher VG, Tibshirani R, Chu G. (2001) Significance analysis of microarrays applied to the ionizing radiation response. Proc Natl Acad Sci U S A 98:5116–5121.
van Dijk M, Mulders J, Poutsma A, Konst AA, Lachmeijer AM, Dekker GA, Blankenstein MA, Oudejans CB. (2005) Maternal segregation of the Dutch preeclampsia locus at 10q22 with a new member of the winged helix gene family. Nat Genet 37:514–519.[CrossRef][Web of Science][Medline]
Wang H, Ng KH, Qian H, Siderovski DP, Chia W, Yu F. (2005) Ric-8 controls Drosophila neural progenitor asymmetric division by regulating heterotrimeric G proteins. Nat Cell Biol 7:1091–1098.[CrossRef][Web of Science][Medline]
Widlund HR, Horstmann MA, Price ER, Cui J, Lessnick SL, Wu M, He X, Fisher DE. (2002) Beta-catenin-induced melanoma growth requires the downstream target Microphthalmia-associated transcription factor. J Cell Biol 158:1079–1087.
Wiemers DO, Shao LJ, Ain R, Dai G, Soares MJ. (2003) The mouse prolactin gene family locus. Endocrinology 144:313–325.
Yamamoto M, Yamazaki S, Uematsu S, Sato S, Hemmi H, Hoshino K, Kaisho T, Kuwata H, Takeuchi O, Takeshige K, et al. (2004) Regulation of Toll/IL-1-receptor-mediated gene expression by the inducible nuclear protein IkappaBzeta. Nature 430:218–222.[CrossRef][Medline]
Yoon Y, Ao Z, Cheng Y, Schlossman SF, Prasad KV. (1999) Murine Siva-1 and Siva-2, alternate splice forms of the mouse Siva gene, both bind to CD27 but differentially transduce apoptosis. Oncogene 18:7174–7179.[CrossRef][Web of Science][Medline]
Zhang J, Klebanoff MA, Levine RJ, Puri M, Moyer P. (1999) The puzzling association between smoking and hypertension during pregnancy. Am J Obstet Gynecol 181:1407–1413.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. Falcao, C. Solomon, C. Monat, J. Berube, J. Gutkowska, and J. L. Lavoie Impact of diet and stress on the development of preeclampsia-like symptoms in p57kip2 mice Am J Physiol Heart Circ Physiol, January 1, 2009; 296(1): H119 - H126. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||