Of Genes in PEFigure 1. Bioinformatic analysis of the gene expression microarray results. (A) Volcano plots of genes with differential expression in pathological versus normal placentas. The x axis represents the log2 of the fold change, and the y axis represents the -log10 of the p value from a student’s t-test. So the red points in the plot represent the differentially expressed genes with statistical significance (with a fold change . = 1.5, and p value ,0.05). (B) Functional annotation analysis of genes that were significantly differentially expressed between preeclamptic and normal placentas. GO, Gene Ontology; BP, Biological Process. doi:10.1371/journal.pone.0059753.gDNA Methylation Analysis of LEP and SH3PXD2AIt is quite clear that DNA methylation has been a potentially important mechanism to regulate gene expression. In order to explore whether the different expressions of LEP and SH3PXD2A were influenced by the effect of DNA methylation, and to have a better understanding of the mechanism underlying the occurrence of PE, we made further investigation of DNA methylation of the relevant genes. We analyzed the methylation patterns of LEP and SH3PXD2A in 32 placentas (16 pathological versus 16 normal placentas) using the high-throughput MALDI-TOF MS assay (Sequenom). Thegene maps of LEP locus and SH3PXD2A locus were shown in Figure 3. The analyzed amplicons in the study comprised the CpG island (CGI) Acid Yellow 23 
web region in both genes. For LEP, three amplicons were designed to cover the whole CGI region, in which 62 CpG sites (40 units) per sample were amenable to be analyzed (Figure 3a). For SH3PXD2A which contains 6 CGIs, its upstream 4 CGIs (CGI71, CGI74, CGI18 and CGI34) were analyzed in the study and 88 CpG sites (54 units) per sample were able to be detected (Figure 3b). In the LEP gene, most of the CpG sites in amplicon 1 and 2 are at a high degree of methylation (average methylation level .0.Upregulation and Hypomethylation of Genes in PEFigure 2. Validation the mRNA expression of LEP and SH3PXD2A in preeclamptic (n = 7) versus normal (n = 6) placentas. (A) Expression of LEP mRNA measured by qRT-PCR. 
The difference between preeclamptic placentas and normal controls is highly significant (p = 0.003). **p,0.01. Each bar represents the average relative expression compared with GAPDH. The average mRNA level of LEP and SH3PXD2A in healthy controls was GSK -3203591 price defined as 1. (B) Expression of SH3PXD2A mRNA quantified by qRT-PCR. The expression of SH3PXD2A is significantly elevated in placentas from pregnancies with PE (p = 0.024). *p,0.05. Each bar represents the average relative expression compared with GAPDH. The average mRNA level of SH3PXD2A in healthy controls was defined as 1. doi:10.1371/journal.pone.0059753.gexcept unit 3), while the CpG sites in amplicon 3 around TSS show a low degree of methylation. Interestingly, we found that CpG dinucleotides situated around TSS [such as CpG sites determined to bind Sp1 (unit 28, average methylation = 0.193, 0.284 in preeclamptic vs normal placentas respectively, p = 1.5761024), LP1 (unit 29, average methylation = 0.163, 0.208 in preeclamptic vs normal placentas respectively, p = 0.023) and CEBPa (unit 31, average methylation = 0.591, 0.689 in preeclamptic vs normal placentas respectively, p = 0.031)and CpG sites in the position of TSS (unit 34, average methylation = 0.145, 0.198 in preeclamptic vs normal placentas respectively, p = 0.001)] were significantly hypomethyalted in the placentas from pregnancie.Of Genes in PEFigure 1. Bioinformatic analysis of the gene expression microarray results. (A) Volcano plots of genes with differential expression in pathological versus normal placentas. The x axis represents the log2 of the fold change, and the y axis represents the -log10 of the p value from a student’s t-test. So the red points in the plot represent the differentially expressed genes with statistical significance (with a fold change . = 1.5, and p value ,0.05). (B) Functional annotation analysis of genes that were significantly differentially expressed between preeclamptic and normal placentas. GO, Gene Ontology; BP, Biological Process. doi:10.1371/journal.pone.0059753.gDNA Methylation Analysis of LEP and SH3PXD2AIt is quite clear that DNA methylation has been a potentially important mechanism to regulate gene expression. In order to explore whether the different expressions of LEP and SH3PXD2A were influenced by the effect of DNA methylation, and to have a better understanding of the mechanism underlying the occurrence of PE, we made further investigation of DNA methylation of the relevant genes. We analyzed the methylation patterns of LEP and SH3PXD2A in 32 placentas (16 pathological versus 16 normal placentas) using the high-throughput MALDI-TOF MS assay (Sequenom). Thegene maps of LEP locus and SH3PXD2A locus were shown in Figure 3. The analyzed amplicons in the study comprised the CpG island (CGI) region in both genes. For LEP, three amplicons were designed to cover the whole CGI region, in which 62 CpG sites (40 units) per sample were amenable to be analyzed (Figure 3a). For SH3PXD2A which contains 6 CGIs, its upstream 4 CGIs (CGI71, CGI74, CGI18 and CGI34) were analyzed in the study and 88 CpG sites (54 units) per sample were able to be detected (Figure 3b). In the LEP gene, most of the CpG sites in amplicon 1 and 2 are at a high degree of methylation (average methylation level .0.Upregulation and Hypomethylation of Genes in PEFigure 2. Validation the mRNA expression of LEP and SH3PXD2A in preeclamptic (n = 7) versus normal (n = 6) placentas. (A) Expression of LEP mRNA measured by qRT-PCR. The difference between preeclamptic placentas and normal controls is highly significant (p = 0.003). **p,0.01. Each bar represents the average relative expression compared with GAPDH. The average mRNA level of LEP and SH3PXD2A in healthy controls was defined as 1. (B) Expression of SH3PXD2A mRNA quantified by qRT-PCR. The expression of SH3PXD2A is significantly elevated in placentas from pregnancies with PE (p = 0.024). *p,0.05. Each bar represents the average relative expression compared with GAPDH. The average mRNA level of SH3PXD2A in healthy controls was defined as 1. doi:10.1371/journal.pone.0059753.gexcept unit 3), while the CpG sites in amplicon 3 around TSS show a low degree of methylation. Interestingly, we found that CpG dinucleotides situated around TSS [such as CpG sites determined to bind Sp1 (unit 28, average methylation = 0.193, 0.284 in preeclamptic vs normal placentas respectively, p = 1.5761024), LP1 (unit 29, average methylation = 0.163, 0.208 in preeclamptic vs normal placentas respectively, p = 0.023) and CEBPa (unit 31, average methylation = 0.591, 0.689 in preeclamptic vs normal placentas respectively, p = 0.031)and CpG sites in the position of TSS (unit 34, average methylation = 0.145, 0.198 in preeclamptic vs normal placentas respectively, p = 0.001)] were significantly hypomethyalted in the placentas from pregnancie.
