Ectopic transcription due to inappropriately inherited histone methylation may interfere with the ongoing function of terminally differentiated cells.

Many human neurodevelopmental disorders are caused by de novo mutations in histone modifying enzymes. These patients have craniofacial defects, developmental delay, intellectual disability and behavioral abnormalities, but it remains unclear how the mutations lead to such developmental defects. Here we take advantage of the invariant C. elegans lineage along with a unique double mutant in the H3K4me1/2 demethylase SPR-5/LSD1/KDM1A and the H3K9 methyltransferase MET-2/SETDB1 to address this question. We demonstrate that spr-5; met-2 double mutant worms have a severe chemotaxis defect that is dependent upon the ectopic expression of germline genes in somatic tissues. In addition, by performing single-cell RNAseq, we find that germline genes begin to be ectopically expression widely in spr-5; met-2 embryos. However, surprisingly we found that spr-5; met-2 mutants have no somatic lineage defects prior to the 200-cell stage of embryogenesis. This suggests that the altered chemotaxis behavior may be due to ongoing defect in terminally differentiated cells rather than a defect in development. To test this directly, we used RNAi to shut off the ectopic expression of germline genes in L2 spr-5; met-2 larvae, which have a fully formed nervous system. Remarkably, we find that shutting off the ectopic germline expression rescues normal chemotaxis behavior in the same adult worms that previously had a chemotaxis defect at the L2 stage. This suggests that ongoing ectopic transcription can block normal behavior in a fully intact nervous system. These data raise the possibility that intellectual disability and altered behavior in neurodevelopmental syndromes, caused by mutations in histone modifying enzymes, could be due to ongoing ectopic transcription and may be reversible.

Characterization of long G4-rich enhancer-associated genomic regions engaging in a novel loop:loop 'G4 Kissing' interaction.

Mammalian antibody switch regions (∼1500 bp) are composed of a series of closely neighboring G4-capable sequences. Whereas numerous structural and genome-wide analyses of roles for minimal G4s in transcriptional regulation have been reported, Long G4-capable regions (LG4s)-like those at antibody switch regions-remain virtually unexplored. Using a novel computational approach we have identified 301 LG4s in the human genome and find LG4s prone to mutation and significantly associated with chromosomal rearrangements in malignancy. Strikingly, 217 LG4s overlap annotated enhancers, and we find the promoters regulated by these enhancers markedly enriched in G4-capable sequences suggesting G4s facilitate promoter-enhancer interactions. Finally, and much to our surprise, we also find single-stranded loops of minimal G4s within individual LG4 loci are frequently highly complementary to one another with 178 LG4 loci averaging >35 internal loop:loop complements of >8 bp. As such, we hypothesized (then experimentally confirmed) that G4 loops within individual LG4 loci directly basepair with one another (similar to characterized stem-loop kissing interactions) forming a hitherto undescribed, higher-order, G4-based secondary structure we term a 'G4 Kiss or G4K'. In conclusion, LG4s adopt novel, higher-order, composite G4 structures directly contributing to the inherent instability, regulatory capacity, and maintenance of these conspicuous genomic regions.

Comparative analysis of expression and secretion of placental leptin in mammals.

Increased plasma level of leptin appears to be a ubiquitous feature of pregnant mammals. The mechanisms by which leptin levels are increased may be species specific, however, with some species upregulating adipose leptin production and others expressing leptin in the placenta. Placental leptin expression was examined in representative species of the two most abundant mammalian orders, Rodentia and Chiroptera, and in cultured human choriocarcinoma (BeWo) cells. Leptin mRNA was expressed in BeWo cells and in placentas of Myotis lucifugus (little brown bat), Eptesicus fuscus (big brown bat), and Rattus norvegicus (laboratory rat), but not the common laboratory mouse Mus musculus. cAMP stimulated secretion of leptin from BeWo cells and also stimulated leptin mRNA expression in the cells. In addition to adipose and placental tissue, leptin transcript in M. lucifugus was detectable in heart, spleen, and liver, but not in lung, brain, and kidney. Hepatic expression was also observed in E. fuscus, but not in mice or rats, and did not appear to result from hepatic fat deposition. Leptin cDNA was cloned and sequenced from M. lucifugus placenta and shared up to 95% homology with other mammalian leptin cDNAs. It is concluded that 1) placental leptin expression and secretion are species-specific traits, 2) placental leptin production represents one of three major mechanisms for achieving high circulating maternal leptin levels during pregnancy, the others being upregulation of adipose leptin production and production of circulating leptin-binding proteins, and 3) hepatic leptin expression in pregnant insectivorous bats may be an adaptation resulting from the presence of extremely low amounts of subcutaneous fat during pregnancy and lactation in these species.