DNA methylation and gene expression differences in children conceived in vitro vs. in vivo Carmen Sapienza Fels Institute for Cancer Research and Molecular Biology Temple University School of Medicine Philadelphia, PA 19140
Jan 03, 2016
DNA methylation and gene expression differences in children conceived
in vitro vs. in vivo
Carmen Sapienza
Fels Institute for Cancer Research and Molecular BiologyTemple University School of Medicine
Philadelphia, PA 19140
ART vs. in vivo conception: • Neural tube defects - 3X (Ericson and Kallen, 2001)• “Major” birth defects - 2X (Hansen et al., 2002)• Respiratory disease - 3.5X (Koivurova et al., 2003)• Low birth weight - 2.6X (Schieve et al., 2002)
ART and imprinting defects:• Beckwith-Wiedemann syndrome - 5.4 - 8.1X• Angelman syndrome - ?X (3/3 imp.def.)
• Animal models demonstrate “LOI”, “LOS”
EnvironmentGenes
Changes in Epigenotype
Long-Term, Somatically HeritableChanges in Gene Expression
“Genes + Environment” Phenotype
Effectors of cellular stressresponses
Do Children Conceived in vitro Differ from Children Conceived
in vivo in “Epigenotype” ?
Experimental Designs for Small Effects• Compare many individuals at few “likely suspect”
loci • Compare few individuals at many loci
– Placenta and cord blood DNA - 10 ART children versus 12 Control children
– Methylation at 1,536 CpGs in promoters of more than 700 genes (highly biased selection)
• 205 CpGs in DMRs of all genes known or suspected to be imprinted/expressed from only one allele
– Use DNA methylation differences to select candidate genes for analysis of mRNA levels in a larger number of individuals
“GoldenGate” assay (Illumina, Inc.): fraction of methyl C at a CpG sitehybridization/synthesis/ligation/PCR in bisulfite treated/untreated DNA
Data Filtering for Candidate Genes:
Reduction of false positives: require 2 CpGs differ in the same gene
Reduction of false negatives: reduce stringency CpGs differ at P<0.08
Result: 78 genes “different” in cord blood 40 genes “different” in placenta
EnvironmentGenes
Changes in Epigenotype
Long-Term, Somatically HeritableChanges in Gene Expression
“Genes + Environment” Phenotype
Effectors of cellular stressresponses
Absolute range of difference small…
Absolute range of difference moderate…
PYY in Placenta
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.0 0.2 0.4 0.6 0.8 1.0
cg03384000
cg22028686
ARTControl
Absolute range of difference large…
EnvironmentGenes
Changes in Epigenotype
Long-Term, Somatically HeritableChanges in Gene Expression
“Genes + Environment” Phenotype
Effectors of cellular stressresponses
Assume cis-acting epigenetic differences linked to steady-state mRNA levels
Do Children Conceived in vitro Differ in “Epigenotype” from Children Conceived in vivo?
As a group, yes. But most individuals are within the
“normal” range.
Associated Phenotype?
CEBPA, MEST, NNAT and SERPINF1 all involved in adipocyte
differentiation/development or insulin signaling
Link between low birthweight – later obesity/hypertension type 2 diabetes?
Conclusions • in vitro conception is associated with
significant DNA methylation differences in 6-10% of genes assayed
• in vitro conception-associated methylation differences are associated with significant differences in steady state mRNA levels of 20-40% of affected (highly selected) genes
• the level of expression of individual genes in some in vitro children is more than two standard deviations from the in vivo mean
Many Questions…
• Infertility, per se, ART or both? • If infertility - genetic, environmental, male
factor, female factor?• If ART – gonadotropins, ICSI, IVF, culture?• Are differences between groups (preexisting
or alleged “changes”) random or are some genes more susceptible than others? (useful in predicting phenotype/treatment?)
Acknowledgments
Fels Institute for Cancer Research,Temple University School of MedicineNahid Turan, Ph.D. (Temple)Sunita Katari, M.D. (Temple)Oluwatoyin Erinle, B.A. (Temple)Halleh Asadpour, B.S. (Temple)Margaret Brannigan, B.A. (Temple)Noa Holzman (Cornell)Leigh Gerson, B.S. (Temple)Sidd Kosaraju (Duke)
Division of Reproductive Endocrinology and Infertility,University of PennsylvaniaChristos Coutifaris, M.D., Ph.D. (Penn)Raffi Chalian, M.D. (Penn) Michael Foster, B.A. (Penn)
Research supported by NIH R01 HD048730