Imprinted Gene Regulation in the Placenta and Fetal Development

The development of placenta into the various trophoblast
subtypes helps to determine the success of a pregnancy outcome...


Unique Patterns of Gene Expression in the Placenta
Various mRNA transcripts such as human chorionic gonadotropin β subunit (hCGB), corticotropin-releasing hormone (CRH) and human placental lactogen (PL) have been characterized in maternal plasma [3]. The expression pattern of each gene is dependent on the gestational age. Transcripts such as fetal-derived γ-globin increase in maternal circulation after elective pregnancy termination [4]. Placental transcripts of clinical significance have been characterized through microarray analysis to generate noninvasive fetal gene profiles [5] including the increased CRH levels in preeclampsia [6] and the decreased HS3ST3A1 mRNA expression in pre-eclamptic placental tissue. The imprinted gene regulation of placenta and fetal development during different gestational stages is summarized in Table 1. Additionally, identified common gene pathways involved in imprinted gene regulation of placental and fetal development are also shown. In mammals, genetic imprinting is a significant process where many genes undergoes epigenetic modification and imprinted specifically in the placenta [7][8][9][10][11][12][13][14][15][16][17][18]. Placental imprinting appears to be a continuous process that occurs throughout the pregnancy [19]. Data from the entire genome methylation from placental tissue during the first and third trimester indicates a methylation-induced down regulation for a number of tumor related genes as a normal placentation process [20]. In this review, it is reported imprinted genes are expressed in a temporal process in the course of normal human placenta development. The study has also demonstrated differential gene expression of PHLDLA2 and IGF2 between first and third trimester of placental tissue. Additionally, gene expression in the placenta is also affected by metabolic conditions such as maternal obesity and gestational diabetes mellitus (GDM) by affecting the energy sensing, which modulates the maternal body mass index (BMI) and GDM on birth weight. of BCRP compared to that of P-gp in primary term trophoblast cells [28]. There is a remarkable reduction in P-gp levels at term labor but the relative expression of P-gp and BCRP during early stages of pregnancy is yet to be understood. There is also limited understanding on the regulation of placental ABCB1/Pgp and ABCG2/BCRP but a number of regulatory elements and transcription factors of ABCB1 and ABCG2 gene promoters have been reported to be significance in gene activation.

Placental RNA Quality and Quantity
Evaluation of the quality and quantity RNA is essential for analysis of gene expression; this is reflected in degraded samples which might influence the interpretation of RNA expression levels.
Study by Monk et al., [19] on imprinted genes specific to human placenta taken from first and third trimester placental tissue along with maternal and third trimester paternal blood samples was largely by biallelic gene expression (cyclooxygenase 2 and 5B (COX2, COX5B) and cytochrome P450 (CYP) 2D1 and −2D7 isoforms (CYP2D1, CYP2D7)) throughout the gestational period. Validation of these genes by PCR reported a specific regulation of COX2 and COX5B in cases with smoking history regardless of gestational age. Studies report that there are epigenetic modifications that may directly influence the size; morphology as well as transport capacity of the placenta [20]. Samples of Placental tissue from preterm and term deliveries alleged to be from preterm labor presented a significant increase in expression of TNFα and IL6, with decreased expression of IFNγ [29]. The placental epigenetic status can be influenced by environmental factors that may in turn affect the fetal growth and development [21].

Gene Annotations
For gene annotations, Protein Analysis Through Evolutionary Relationships (PANTHER) [30] software system has been used to analyze gene sequence and relate them to their particular biological processes and molecular roles (http://www.pantherdb.org). The signaling pathways involved during normal placental development have been identified using Bonferroni correction for multiple testing and expression data analysis tool [9]. Analysis of Pathways of differential gene expression between first and third trimester of the placentas reported angiogenesis to be extremely active in first trimester.

Placental Gene Imprinting
Fetal growth and development of the placenta are regulated by imprinted genes in mammalian species and are thought to have co evolved with placentation ( Table 2). In majority of mammals, autosomal genes are expressed co-dominantly from the two parental chromosomes. The process of monoallelic expression is attained through epigenetic asymmetry between parental alleles such as maternal gene expression specific to the placenta and are imprinted regardless of their elevated level of expression in decidua as in the case of Tissue Factor Pathway Inhibitor 2 (Tfpi2) [31]. As a result, false-positive maternal expression due to decidua contamination can be distinguished from actual imprinting by making use of in situ staining or from backcrosses whereby homozygous embryos are formed in a heterozygous mother. In such cases, the allele detected can be absent in the embryo's genome which signify maternal contamination [32]. Placenta contains several specific imprinted transcripts, which are mainly found in the large imprinted domains as shown in Table 2. For instance, two clusters of maternally expressed genes specific to the placenta are located on the distal chromosome seven and proximal chromosome seventeen, both of which are regulated by maternally methylated regions, the KvDMR1 and Airn DMRs, respectively [33,34]. In these DMRs, there are promoters for long ncRNAs. The Airn non coding RNA (ncRNA) transcript silences two adjacent genes, solute carrier family 22 member 2 (SLC22A2) and SLC22A3 in the Igf2r domain [35] through recruitment of histone methyltransferase enzymes such as G9a to the paternal allele [36], and deposit of repressive histone mark on lysine 9 of histone H3 (H3K9me2) [37]. Additionally, Airn cause direct imprinting of Igf2r gene since transcription via the Igf2r promoter on the paternal allele is adequate for silencing, probably by dislocating transcription factors important for Igf2r expression [9]. The long paternal expression of ncRNA KCNQ1OT1 on chromosome seven recruit G9a and together with H3K27me3 histone methyltransferase enhancer of zeste homolog 2 (Ezh2), results to the paternal silencing of eight flanking genes within the placenta [38]. Interestingly, within the KCNQ1OT1 loci, imprinting is not conserved in human placenta due to lack of allelic repressive histone modifications. The imprinting of SLC22A2 and Insulin-like growth factor 2 receptor (IGF2R) is polymorphic in humans [39].
The role of these genes in placenta-related complications and intrauterine growth restriction should further be investigated along with aberrant genes as biomarkers in complicated pregnancies, in order to help in the diagnosis of at-risk pregnancies during early gestation [40].

Placental Gene Regulation
Certain genes in the placenta are under strict epigenetic regulation and therefore prone to genomic imprinting. The Based on various studies, the computation of standardized mean difference (mu) of the six genes exhibit a differential expression in the initial meta-signature two of the twelve genes (VIM and HSD17B1) were amongst the 688 genes that were differentially expressed, but they were left out during the leave-one-out-analysis placenta, SOD1 is one of significantly down regulated genes [45].
Hypoxia/ischemia compromise pre-eclamptic placentas and hypoxia-inducible factor-1 alpha (HIF-1α) mainly mediate gene expression. Over 10% of the meta-signature genes that are up regulated are directly targeted by HIF-1α CREB binding protein (CREBBP)/EP300, which is a key transcriptional co activator of HIF-1α, therefore, low CREBBP/EP300 level may reduce the placenta's ability to respond to shortage of oxygen, which worsen the pre-eclamptic state [43]. A study from previous work indicated the concentration of tumor necrosis factor (TNF-α), interleukin 1 alpha (IL-1β) and IL-6 in serum was higher in GDM group compared to the control group. Forkhead box O1 (FoxO1) expression was detected in adipose tissue of both the placenta and fetus. When compared to the control group, the gene and protein expression of FoxO1 & TNF-α was higher in the GDM group in both tissues [46].
Another study also reported a positive co-relation between FoxO1 expression in the placenta with homeostatic model assessment of Insulin Resistance (HOMA-IR) and TNF-α. TNF-α gene stimulation increases FoxO1 expression in trophoblast cell cultures. Deletion of FoxO1 in the cells reduces TNF-α-induced expression of IL-6 and IL-1β pro-inflammatory cytokines [47]. These findings indicate that FoxO1 plays as a pro-inflammatory factor in GDM as well as in IR by interacting with TNF-α, a pro-inflammatory cytokine that helps to modulate the acute phase reaction and first discovered in placenta and amnion [37]. The insulin-like growth factor-binding protein (IGFBP) also known as pregnancy-associated plasma protein A2 (PAPPA2), protease that is highly expressed in the placenta is up regulated in pre-eclampsia; HELLP (Haemolytic anaemia, Elevated Liver enzymes, and Low Platelet count) syndrome [48].
High expression of PAPPA2 results to abnormal placental development and its up regulation may be meant to compensate for placental pathology. Oxidative stress and hypoxia are among other conditions that affect the expression of PAPPA2 in preeclamptic placenta; TNF-α and prostaglandin E2 (PGE2) results to the up regulation of Pappalysin 2 (PAPPA2) [49]. Hypoxia, which is common in pre-eclamptic placenta results to PAPPA2 expression.
These findings confirm the hypothesis of up regulation of PAPPA2 as a result of placental pathology as opposed to the assumption that high PAPPA2 levels cause preeclampsia [50].

Placental DNA Methylation
Extensive research has been done on the effect of imprinted methylation on a number of placental abnormalities stated in the DNA methyltransferase 1o (DNMT1o) model, initially, DNA methylation determined on fifteen imprinted gametic differentially methylated domains (gDMDs) is thrice during the last half of gestation period [52]. The mean methylation fraction across twelve non-redundant gDMD EpiTYPER amplicons for the wild-type as well as mutant specimens at the respective time points, methylation was less in DNMT1o-deficient placental tissues at embryonic 12.5 day (E12.5), E15.5 and E17.5. At E12.5 it showed a significant reduction in the average methylation for all gDMDs (wild-type and mutant placentas) [53]. The average gDMD methylation was 0.283 for a total of 23 E15.5 DNMT1o-deficient placentas, which was significantly lower as compared to the wild-type 0.382 [54,55]. In one study, Meta-analysis was carried out in placenta by RNA microarray in 116 pre-eclamptic and 139 normotensive pregnancies using statistical and standard bioinformatics procedures [43], where pathway analysis of the expression signature in genes interactions were deduced as well as differentially expressed genes resulting to 388-gene meta-signatures of pre-eclamptic placenta. The analysis indicated the role of hypoxia/HIF1A pathway in the expression of pre-eclamptic gene profile which was consistence with previous reports [56]. Analysis of protein interaction networks showed that CREBBP/EP300 is a new element key to pre-eclamptic placental transcriptome and there is a high incidence of preeclampsia in pregnant women carrying fetus with a mutation in CREBBP/EP300 which normally lead to Rubinstein-Taybi Syndrome [57]. From the 388-gene-preeclampsia meta-signature reported, important information can be generated on the role of these genes in the placental tissue such as CREBBP/EP300 and the related pathways, which can be utilized as functional molecules or biomarkers in preeclampsia. Understanding the molecular basis of preeclampsia may help in development of therapeutic measures to alleviate placental pathologies [58].

Placental Gene Deletion
Achaete-Scute Family BHLH Transcription Factor 2 (Ascl2) gene deletion in the mouse Kcnq1 cluster results to fetal lethality because of the limited development of placenta labyrinth and the resulting in build-up of trophoblast giant cells (TGCs) at E10.5 [59]. Deletion of Cdkn1c or Phlda2 genes found within the Kcnq1 cluster, lead to placental overgrowth [60,61] as well as transgenic over-expression of Cdkn1c or Phlda2 that leads to impaired placental growth [62][63][64]. The growth and development of the placenta also relies on Igf2, which is a significant part of H19 gene imprinting cluster; as a result, deletion of Igf2 leads to growth restriction of both the placenta and the fetus while Igf2 overexpression lead to overgrowth of the placenta as well as the fetus [65]. In addition, when other imprinted genes are deleted that are not located within the Kcnq1 or H19 clusters, there is also abnormalities in placental phenotypes as in the case of deletion of Igf2r, Growth-factor receptor bound   [70]. Decidual expression of Rac1 helps to regulate TGCs proliferation and differentiation at the maternal-fetal interface and allows for correct placenta formation and development [71].

Placenta Gene Mutations
Mutations such as E1A binding protein p300 (EP300) and histone acetyltransferase action may result to RSTS [73]. A number of preeclampsia cases result to a birth of a child with RSTS [74]. As a result, the fetal genotype is implicated with RSTS and affects the placenta function which leads to high incidences of preeclampsia cases [75]. These findings can be incorporated in the meta-analysis data in non-RSTS mothers to indicate that low CEBBP/EP300 levels impair placenta functioning which increases the chances of the mother developing preeclampsia [76].

Conclusion
There are limited longitudinal studies on placental gene expression in human partly because of ethical constraints.
However, studies in mice indicate that transition occur during midgestation and same cellular subsets may express different genes without necessarily causing changes in the placental morphology in humans, microarray data from basal plate biopsy from the maternal fetal interface indicate remarkable changes from mid-gestation to term. Plethora of studies on differential gene expression profile are available both for healthy and impaired placentas, particularly in preeclampsia. However, there are limited numbers of studies on differential gene expression profile during normal gestational development of placenta in human. Only a single microarray study has addressed this gap. This review aimed to explore the molecular pathways as well as physiological changes that occur in the course of gestation that may affect the placental structure and function.
We therefore hypothesized that the molecular reorganization and phenotypic variations are needed for normal placental development and can be deduced from the level of genes expressed. More studies are therefore needed to ascertain this.