IGF2 and H19 Genomic Imprinting Locus (Chr11p15.5)

The Insulin-Like Growth Factor 2 (IGF2) and H19 gene cluster inside the distal subtelomere of human chromosome 11p15.5 (NC_000011.10:2,016,000–2,155,000, spanning ~140 kb within a larger 1-Mb imprinted domain) represents the paradigm of mammalian genomic imprinting—an epigenetic mechanism in which autosomal genes are expressed monoallelically in a parent-of-origin-specific manner. Although every diploid somatic cell inherits one maternal and one paternal copy of Chr11p15.5 with identical DNA sequences, the paternal allele selectively transcribes IGF2 (encoding a 67-amino-acid, 7.5-kDa secreted mitogenic hormone that stimulates fetal and placental proliferation via the IGF1R and IR-A receptor tyrosine kinases) while silencing H19. Conversely, the maternal allele silences IGF2 and selectively transcribes H19—a capped, spliced, and polyadenylated 2.3-kb long non-coding RNA (lncRNA) whose first exon hosts the growth-suppressing microRNA miR-675 (miR-675-3p and miR-675-5p, which downregulate IGF1R, SMAD1/5, and placental trophoblast expansion).

Discovered in 1991 by Marisa Bartolomei, Shirley Tilghman, Thomas DeChiara, and Efstratiadis, the reciprocal monoallelic expression of IGF2 and H19 is controlled by a single ~2.4-to-5.0-kb intergenic differentially methylated region (DMR) located 2 to 4 kb upstream of the H19 transcription start site, designated Imprinting Control Region 1 (ICR1 or IC1, also known as the H19-DMR). Parent-of-origin DNA methylation established at IC1 during male versus female gametogenesis acts as a binary molecular switch for the architectural zinc-finger protein CTCF (CCCTC-binding factor), reconfiguring three-dimensional (3D) chromatin enhancer–promoter loops across the Chr11p15.5 domain and explaining how opposing epimutations cause the mirror-image human growth disorders Beckwith–Wiedemann syndrome (BWS) and Silver–Russell syndrome (SRS).

Seven CTCF Binding Sites, G-Quadruplex Density, and Gametic Methylation

The human Chr11p15.5 imprinting cluster is organized into two adjacent topological domains (Domain 1 and Domain 2):

• Telomeric Domain 1 (IC1 / IGF2–H19): Governed by IC1 (H19-DMR), which is methylated on the paternal chromosome (established in prospermatogonia during fetal testis development by DNMT3A/DNMT3L and protected after fertilization by ZFP57/ZNF445–KAP1) and unmethylated on the maternal chromosome (protected in growing oocytes by OCT4/SOX2 binding). Human IC1 comprises two repetitive blocks (A-repeats and B-repeats) containing seven conserved CTCF target sites (CTS1–CTS7, with CTS6 serving as the primary diagnostic site) embedded within a GC-rich (59% GC) subtelomeric landscape that harbors 605 canonical G-quadruplex (G4) motifs across the IGF2–H19 locus (36.9 G4s/kb inside H19, where non-B DNA quadruplexes interact with DNMT1 and CTCF to stabilize epigenetic memory).
• Centromeric Domain 2 (IC2 / KCNQ1–CDKN1C): Located ~700 kb centromeric to IC1 and governed by IC2 (KvDMR1 inside Intron 10 of KCNQ1), which exhibits the opposite gametic imprint: methylated on the maternal allele (allowing maternal expression of the cyclin-dependent kinase inhibitor CDKN1C / p57KIP2) and unmethylated on the paternal allele (which transcribes the antisense long non-coding RNA KCNQ1OT1 to silence paternal CDKN1C in cis).

Allele-Specific Cohesin Loop Extrusion and Shared Downstream Enhancers

Both IGF2 (located ~90 kb upstream of H19) and H19 depend on the same set of powerful tissue-specific mesodermal and endodermal enhancers located 10 to 120 kb downstream (3′) of the H19 gene. IC1 sits between IGF2 and these downstream enhancers, operating as a methylation-sensitive CTCF chromatin insulator:

1. Maternal Chromosome (Unmethylated IC1 → CTCF Insulator Active): Because the CpG dinucleotides inside CTS1–CTS7 at maternal IC1 are unmethylated, the 11-zinc-finger architectural protein CTCF binds with high affinity to maternal IC1 and halts ATP-dependent cohesin (SMC1/SMC3/RAD21) loop extrusion. This creates an insulated 3D chromatin domain (sub-TAD) that sequesters the IGF2 promoters into a silent loop (interacting with a centrally located differentially methylated region DMR0/CCD) and topologically blocks the downstream enhancers from contacting IGF2. Instead, the downstream enhancers loop to the maternal H19 promoter, driving maternal H19 and miR-675 transcription (growth suppression).
2. Paternal Chromosome (Methylated IC1 → CTCF Blocked, Enhancer Bypass): On the paternal chromosome, 5-methylcytosine (5mC) at the CpG dinucleotides within CTS1–CTS7 sterically blocks CTCF DNA binding (specifically cytosine methylation at position 2 and 12 of the CTCF core consensus). With no CTCF/cohesin roadblock at paternal IC1—and as CpG methylation spreads across the paternal H19 promoter to silence H19—cohesin loop extrusion proceeds unimpeded past IC1, bringing the 3′ downstream enhancers into direct 3D contact with the paternal IGF2 promoters (P0, P1, P2, P3, P4), driving high-level paternal IGF2 expression (fetal growth promotion).

Haig's Parental-Conflict Hypothesis of Genomic Imprinting

Why did placental mammals (eutherians and marsupials, but not egg-laying monotremes or birds) evolve parent-of-origin genomic imprinting at IGF2 and H19? In 1991, David Haig and Graham Moore formulated the kinship (parental-conflict) theory of genomic imprinting. In any viviparous species where females mate with multiple males across their reproductive lifespan, a fetus's paternally derived alleles are not assured of being shared with the mother's future offspring sired by other males; natural selection on paternal alleles therefore favors maximizing extraction of maternal placental nutrients to grow the current fetus as large as possible (paternally expressed growth-promoting *IGF2*).

Conversely, the mother shares an equal 50% genetic relatedness (r = 0.5) with the current fetus and with all of her future offspring; selection on maternally derived alleles therefore favors conserving maternal metabolic resources and restraining fetal/placental overgrowth to ensure maternal survival and equal provisioning across litters (maternally expressed growth-suppressing *H19* / *miR-675*, *CDKN1C*, and the cation-independent mannose-6-phosphate / **IGF2R** scavenger receptor on Chr6q25.3, which internalizes and degrades excess IGF2 protein in lysosomes).

Beckwith–Wiedemann Syndrome, Silver–Russell Syndrome, and Wilms Tumor

Clinical human genetics at Chr11p15.5 provides a direct quantitative verification of the insulator model and Haig's kinship theory:

• Beckwith–Wiedemann Syndrome (BWS, OMIM #130650, ~1 in 10,000 live births): Caused by any genetic or epigenetic mechanism that increases IGF2 dosage and/or decreases CDKN1C dosage, including IC2 loss of methylation (~50%), segmental paternal uniparental disomy (pUPD11, ~20%), IC1 gain of methylation (GoM, ~5–10%, often caused by microdeletions removing CTCF binding sites CTS4–CTS6), or maternal CDKN1C point mutations (~5%). Notably, molecular subtype guides cancer surveillance: BWS patients with IC1 GoM or pUPD11 (which double IGF2 expression in metanephric renal blastema) face a 15% to 28% risk of Wilms tumor and hepatoblastoma, whereas IC2 LoM patients face <1–3% tumor risk (predominantly hepatoblastoma/rhabdomyosarcoma rather than Wilms tumor).
• Silver–Russell Syndrome (SRS, OMIM #180860, ~1 in 30,000–100,000 live births): The mirror image of BWS. In ~40–60% of SRS patients, loss of methylation (LoM) at paternal IC1 permits CTCF to bind both chromosomes, insulating both alleles of IGF2 from the downstream enhancers (biallelic IGF2 silencing + 2× H19 expression) and causing intrauterine growth restriction (birth weight and length >2 SD below the mean), relative preservation of cranial circumference (sparing brain growth, where IGF2 is expressed biallelically from the choroid plexus/leptomeninges via promoter P1), triangular facies, and clinodactyly.

Why does Silver–Russell syndrome cause marked body growth restriction while sparing brain and head circumference (relative macrocephaly)?

Throughout fetal liver, skeletal muscle, cartilage, and placenta, IGF2 is transcribed selectively from its paternal promoters P2, P3, and P4 under the control of the IC1 CTCF insulator. When IC1 loses methylation in Silver–Russell syndrome (IC1 LoM), somatic IGF2 production drops, restricting body weight and skeletal length. Inside the human central nervous system (choroid plexus and leptomeninges), however, IGF2 is transcribed biallelically from both parental chromosomes via the upstream Promoter P1 (and P0), which operates independently of the IC1 insulator—preserving normal fetal brain growth inside a small body (relative macrocephaly).

Why does a maternal microdeletion removing CTCF binding sites inside IC1 cause Beckwith–Wiedemann overgrowth rather than Silver–Russell syndrome?

On a normal maternal chromosome, unmethylated IC1 requires intact CTCF binding sites (specifically CTS1–CTS6) to recruit CTCF/cohesin, form the chromatin insulator loop, and block the downstream enhancers from activating maternal IGF2. If a small 1.4-to-2.2-kb microdeletion removes key CTCF target sites (such as CTS4–CTS6) on the maternal chromosome, CTCF can no longer bind cleanly, the residual maternal IC1 becomes hypermethylated (IC1 GoM), the insulator collapses, and the downstream enhancers activate maternal IGF2—causing biallelic IGF2 overexpression and Beckwith–Wiedemann syndrome (with high Wilms tumor penetrance), whereas the identical microdeletion inherited on the paternal chromosome (where IC1 is already methylated and CTCF-free) is clinically silent!

How does MS-MLPA testing simultaneously diagnose both DNA copy-number deletions and parent-of-origin methylation defects at Chr11p15.5?

Methylation-Specific Multiplex Ligation-Dependent Probe Amplification (MS-MLPA, ME030 panel) hybridizes oligonucleotide probe pairs across Chr11p15.5 (IC1 and IC2) and divides the patient DNA sample into two reactions: an undigested reaction (which measures genomic copy number, detecting duplications or microdeletions) and a reaction digested with the methylation-sensitive restriction endonuclease HhaI (GCGC), which cleaves unmethylated DNA hybrids while sparing methylated CpGs—quantifying the methylation index (normal = 0.50 across both IC1 and IC2).

1. Bartolomei, M. S., Zemel, S., & Tilghman, S. M. (1991). "Parental imprinting of the mouse H19 gene." Nature, 351(6322), 153–155. https://doi.org/10.1038/351153a0
2. Bell, A. C., & Felsenfeld, G. (2000). "Methylation of a CTCF-dependent boundary controls imprinted expression of the Igf2 gene." Nature, 405(6785), 482–485. https://doi.org/10.1038/35013100
3. Hark, A. T., Schoenherr, C. J., Katz, D. J., Ingram, R. S., Levorse, J. M., & Tilghman, S. M. (2000). "CTCF mediates methylation-sensitive enhancer-blocking activity at the H19/Igf2 locus." Nature, 405(6785), 486–489. https://doi.org/10.1038/35013106
4. Brioude, F., Kalish, J. M., Mussa, A., et al. (2018). "Clinical and molecular diagnosis, screening and management of Beckwith–Wiedemann syndrome: an international consensus statement." Nature Reviews Endocrinology, 14(4), 229–249. https://doi.org/10.1038/nrendo.2017.166