Genomic imprinting alterations have been shown to be associated with assisted reproductive technologies (ARTs) in animals. At present, data obtained in humans are inconclusive; however, some epidemiological studies have demonstrated an increased incidence of imprinting disorders in children conceived by ARTs. In the present study, we focused on the effect of ARTs [IVF and intracytoplasmic sperm injection (ICSI)] on the epigenetic reprogramming of the maternally methylated imprinting control region KvDMR1 in clinically normal children. Qualitative and quantitative methylation at KvDMR1 were assessed by the methylation-specific PCR approach and by the methylation-sensitive enzymatic digestion associated with real-time PCR method, respectively. DNA was obtained from peripheral blood of 12/18 and umbilical cord blood and placenta of 6/18 children conceived by IVF or ICSI. The methylation patterns observed in this group were compared with the patterns observed in 30 clinically normal naturally conceived children (negative controls) and in 3 naturally conceived Beckwith–Wiedemann syndrome patients (positive controls). Hypomethylation at KvDMR1 was observed in 3/18 clinically normal children conceived by ARTs (2 conceived by IVF and 1 by ICSI). A discordant methylation pattern was observed in the three corresponding dizygotic twins. Our findings corroborate the hypothesis of vulnerability of maternal imprinting to ARTs. Furthermore, the discordant methylation at KvDMR1 observed between dizygotic twins could be consequent to one of the following possibilities: (i) a differential vulnerability of maternal imprints among different embryos; or (ii) epimutations that occurred during gametogenesis resulting in the production of oocytes without the correct primary imprint at KvDMR1.

Polymorphisms in the endothelial nitric oxide synthase (eNOS) gene have been inconsistently associated with preeclampsia. We compared genotype and haplotype frequencies of three eNOS gene polymorphisms in normotensive and hypertensive pregnancies. Genotypes and haplotypes for eNOS polymorphisms (T-786C, Glu298Asp and intron 4 b/a) were determined in 326 pregnant women (110 healthy pregnancies, 103 gestational hypertensives and 113 preeclamptic). No differences were observed in the frequencies of genotypes and alleles of the three polymorphisms among the groups (all p > 0.05). However, the haplotype 'T Glu a' was more common in healthy pregnancies than in gestational hypertensives or preeclamptic (20 vs 6 and 6%, respectively; p < 0.0032). Conversely, the haplotype 'C Glu a' was more common in gestational hypertensives and preeclamptic than in healthy pregnancies (17 vs 17 and 5%; p = 0.0061). These findings suggest a contribution of eNOS haplotypes to the development of hypertensive disorders of pregnancy that is obscured when specific eNOS genotypes alone are considered.

X-chromosome inactivation (XCI) is the epigenetic transcriptional silencing of an X-chromosome during the early stages of embryonic development in female eutherian mammals. XCI assures monoallelic expression in each cell and compensation for dosage-sensitive X-linked genes between females (XX) and males (XY). DNA methylation at the carbon-5 position of the cytosine pyrimidine ring in the context of a CpG dinucleotide sequence (5^meCpG) in promoter regions is a key epigenetic marker for transcriptional gene silencing. Using computational analysis, we revealed an extragenic tandem GAAA repeat 230-bp from the landmark CpG island of the human X-linked retinitis pigmentosa 2 RP2 promoter whose 5^meCpG status correlates with XCI. We used this RP2 onshore tandem GAAA repeat to develop an allele-specific 5^meCpG-based PCR assay that is highly concordant with the human androgen receptor (AR) exonic tandem CAG repeat-based standard HUMARA assay in discriminating active (Xa) from inactive (Xi) X-chromosomes. The RP2 onshore tandem GAAA repeat contains neutral features that are lacking in the AR disease-linked tandem CAG repeat, is highly polymorphic (heterozygosity rates approximately 0.8) and shows minimal variation in the Xa/Xi ratio. The combined informativeness of RP2/AR is approximately 0.97, and this assay excels at determining the 5^meCpG status of alleles at the Xp (RP2) and Xq (AR) chromosome arms in a single reaction. These findings are relevant and directly translatable to nonhuman primate models of XCI in which the AR CAG-repeat is monomorphic. We conducted the RP2 onshore tandem GAAA repeat assay in the naturally occurring chimeric New World monkey marmoset (Callitrichidae) and found it to be informative. The RP2 onshore tandem GAAA repeat will facilitate studies on the variable phenotypic expression of dominant and recessive X-linked diseases, epigenetic changes in twins, the physiology of aging hematopoiesis, the pathogenesis of age-related hematopoietic malignancies and the clonality of cancers in human and nonhuman primates.

Vascular endothelial growth factor (VEGF) is relevant for normal pregnancy, and abnormalities in VEGF functions are associated with hypertensive disorders of pregnancy. Because there are few studies on how VEGF genetic polymorphisms affect susceptibility to pre-eclampsia (PE), and no studies on how they affect susceptibility to gestational hypertension (GH), we compared VEGF genotype and haplotype distributions in normotensive and hypertensive pregnancies. Genotypes and haplotypes for VEGF polymorphisms (C-2578A, G-1154A and G-634C) were determined in 303 pregnant women (108 healthy pregnant, HP; 101 with GH and 94 with PE). When white and non-white pregnant women were considered together, no significant differences were found in the distributions of VEGF genotypes or haplotypes (P > 0.05) in the three groups. However, with only white subjects, significant differences were found in genotypes distributions for two (C-2578A and G-634C) VEGF polymorphisms (both P < 0.05) between the HP and the PE groups. Importantly, the haplotype including the variants C-2578, G-1154 and C-634, which is associated with higher VEGF gene expression, was less common in the PE group compared with the HP group (4% versus 16%; P = 0.0047). However, we found no significant differences in VEGF haplotypes distributions when the HP and GH groups were compared (P > 0.05). These findings suggest a protective effect for the ‘C-2578, G-1154 and C-634’ haplotype against the development of PE, but no major effects of VEGF gene variants on susceptibility to GH.