The histone demethylase PHF8 facilitates alternative splicing of the histocompatibility antigen HLA-G.

Histone3-lysine9 (H3K9) residues not only control gene expression, but also contribute to RNA splicing. Here, the H3K9 histone demethylase PHF8 was investigated in endothelial cells for its involvement in alternative splicing. An angiogenic sprouting assay shows the importance of PHF8 for endothelial cells. Immunoprecipitation reveals that PHF8 interacts with U1 spliceosomal proteins, such as SRPK1 and snRNP70. We identify the histocompatibility antigen HLA-G as a target of PHF8. The inclusion of HLA-G intron 4, with concomitant RNA Polymerase II accumulation at this intron is controlled by PHF8 and H3K9. Soluble HLA-G is generated after PHF8 knockdown, which leads to reduced T-cell proliferation. Collectively, PHF8 knockdown generates the immunosuppressive alternative splice product soluble HLA-G, which is secreted by endothelial cells to elicit a potential inhibitory effect on inflammation.

The histone demethylase Jarid1b mediates angiotensin II-induced endothelial dysfunction by controlling the 3'UTR of soluble epoxide hydrolase.

AIM: The histone demethylase Jarid1b limits gene expression by removing the active methyl mark from histone3 lysine4 at gene promoter regions. A vascular function of Jarid1b is unknown, but a vasoprotective function to inflammatory and hypertrophic stimuli, like angiotensin II (AngII) could be inferred. This hypothesis was tested using Jarid1b knockout mice and the inhibitor PBIT. METHODS: Mice or aortic segments were treated with AngII to induce endothelial dysfunction. Aortae from WT and Jarid1b knockout were studied in organ chambers and endothelium-dependent dilator responses to acetylcholine and endothelium-independent responses to DetaNONOate were recorded after pre-constriction with phenylephrine in the presence or absence of the NO-synthase inhibitor nitro-L-arginine. Molecular mechanisms were investigated with chromatin immunoprecipitation, RNA-Seq, RNA-3'-adaptor-ligation, actinomycin D and RNA-immunoprecipitation. RESULTS: Knockout or inhibition of Jarid1b prevented the development of endothelial dysfunction in response to AngII. This effect was not a consequence of altered nitrite oxide availability but accompanied by a loss of the inflammatory response to AngII. As Jarid1b mainly inhibits gene expression, an indirect effect should account for this observation. AngII induced the soluble epoxide hydrolase (sEH), which degrades anti-inflammatory lipids, and thus promotes inflammation. Knockout or inhibition of Jarid1b prevented the AngII-mediated sEH induction. Mechanistically, Jarid1b maintained the length of the 3'untranslated region of the sEH mRNA, thereby increasing its stability and thus sEH protein expression. Loss of Jarid1b activity therefore resulted in sEH mRNA destabilization. CONCLUSION: Jarid1b contributes to the pro-inflammatory effects of AngII by stabilizing sEH expression. Jarid1b inhibition might be an option for future therapeutics against cardiovascular dysfunction.

The Histone Demethylase PHF8 Is Essential for Endothelial Cell Migration.

Epigenetic marks critically control gene expression and thus the cellular activity state. The functions of many epigenetic modifiers in the vascular system have not yet been studied. We screened for histone modifiers in endothelial cells and observed a fairly high expression of the histone plant homeodomain finger protein 8 (PHF8). Given its high expression, we hypothesize that this histone demethylase is important for endothelial cell function. Overexpression of PHF8 catalyzed the removal of methyl-groups from histone 3 lysine 9 (H3K9) and H4K20, whereas knockdown of the enzyme increased H3K9 methylation. Knockdown of PHF8 by RNAi also attenuated endothelial proliferation and survival. As a functional readout endothelial migration and tube formation was studied. PHF8 siRNA attenuated the capacity for migration and developing of capillary-like structures. Given the impact of PHF8 on cell cycle genes, endothelial E2F transcription factors were screened, which led to the identification of the gene repressor E2F4 to be controlled by PHF8. Importantly, PHF8 maintains E2F4 but not E2F1 expression in endothelial cells. Consistently, chromatin immunoprecipitation revealed that PHF8 reduces the H3K9me2 level at the E2F4 transcriptional start site, demonstrating a direct function of PHF8 in endothelial E2F4 gene regulation. Conclusion: PHF8 by controlling E2F4 expression maintains endothelial function.

Epigenetic Regulation of Angiogenesis by JARID1B-Induced Repression of HOXA5.

OBJECTIVE: Altering endothelial biology through epigenetic modifiers is an attractive novel concept, which is, however, just in its beginnings. We therefore set out to identify chromatin modifiers important for endothelial gene expression and contributing to angiogenesis. APPROACH AND RESULTS: To identify chromatin modifying enzymes in endothelial cells, histone demethylases were screened by microarray and polymerase chain reaction. The histone 3 lysine 4 demethylase JARID1B was identified as a highly expressed enzyme at the mRNA and protein levels. Knockdown of JARID1B by shRNA in human umbilical vein endothelial cells attenuated cell migration, angiogenic sprouting, and tube formation. Similarly, pharmacological inhibition and overexpression of a catalytic inactive JARID1B mutant reduced the angiogenic capacity of human umbilical vein endothelial cells. To identify the in vivo relevance of JARID1B in the vascular system, Jarid1b knockout mice were studied. As global knockout results in increased mortality and developmental defects, tamoxifen-inducible and endothelial-specific knockout mice were generated. Acute knockout of Jarid1b attenuated retinal angiogenesis and endothelial sprout outgrowth from aortic segments. To identify the underlying mechanism, a microarray experiment was performed, which led to the identification of the antiangiogenic transcription factor HOXA5 to be suppressed by JARID1B. Importantly, downregulation or inhibition of JARID1B, but not of JARID1A and JARID1C, induced HOXA5 expression in human umbilical vein endothelial cells. Consistently, chromatin immunoprecipitation revealed that JARID1B occupies and reduces the histone 3 lysine 4 methylation levels at the HOXA5 promoter, demonstrating a direct function of JARID1B in endothelial HOXA5 gene regulation. CONCLUSIONS: JARID1B, by suppressing HOXA5, maintains the endothelial angiogenic capacity in a demethylase-dependent manner.

Intermedin ameliorates atherosclerosis in ApoE null mice by modifying lipid profiles.

Intermedin (IMD) is a recently discovered vasodilator peptide. We studied the role of IMD in the pathogenesis of atherosclerosis by investigating the ability of exogenous IMD to alter lipid profiles and ameliorate the development of atherogenic-diet induced atherosclerosis in ApoE-/- mice. Ten of eight-week-old male C57BL/6J mice were as control. Thirty of eight-week-old male ApoE-/- mice were fed with an atherogenic diet for 18 weeks. After feeding atherogenic diet for 12 weeks, the mice were equally and randomly divided into three groups. Normal saline was given in group A and C57BL/6J mice. Intermedin was given by mini osmotic pumps at the dosage of 100 ng/kg/h and 500 ng/kg/h in group B and group C respectively. After the treatment of IMD for 6 weeks, aortic ultrasonography of group C showed that IMD prevented the progression of atherosclerotic lesions and the increase of wall thickness in the aorta. Oil-red-O staining of the entire aorta and the atherosclerotic aortic root section showed 2 folds decrease atherogenic plaque (p<0.05). Serum lipid profiles were measured, compared with the group A, in group C TC and LDL-C levels were decreased by 86.32% and 89.68%, respectively (both p<0.05), meanwhile, HDL-C level was significantly increased by 74.82% (p<0.05). These data indicate that exogenous administration of IMD could prevent the progression of atherosclerotic plaque. The possible underlying mechanisms may relate to the improvement of lipid profiles.

Disruption of CD38 gene enhances cardiac functions by elevating serum testosterone in the male null mice.

AIMS: Gender-related phenotypes in the cardiovascular system have been observed in various genetically modified mice. Here, we report that cardiac functions are significantly improved only in male CD38-null mice and we explore the potential mechanisms of the sexual dimorphism mediated by CD38 deficiency. MAIN METHODS: Cardiac functions of mice were measured by pressure-volume conductance catheter technique and echocardiography. Serum sex steroids were determined by radioimmunoassay. Relative mRNA levels of myocardial contractile-associated proteins in cardiomyocytes were analyzed by real-time PCR analysis. To clarify the effects of testosterone on the sexual dimorphism, flutamide, an androgen receptor antagonist, was subcutaneously infused into the male null mice for 6 weeks with an osmotic mini-pump. KEY FINDINGS: The myocardial contractility, contraction and relaxation velocities were significantly enhanced only in male CD38-null mice, in which the levels of serum testosterone were markedly elevated. The elevated testosterone levels in the null mice were correlated to an obvious decrease in expression of androgen receptor and dramatic increases in expressions of major genes involved in myocardial contraction, including ryanodine receptor type 2 (RyR2), sarcoplasmic reticular Ca(2+) ATPase (SERCA2) and Na(+)/Ca(2+)-exchanger protein 1 (NCX1), and α myosin heavy chain (α-MHC). More importantly, all of the alternations that were observed in the male null mice were almost completely restored by flutamide administration. SIGNIFICANCE: Elevated serum level of testosterone in the male CD38(-/-) mice enhances cardiac functions through upregulation of major calcium regulatory proteins, which improve our understanding on sex disparities and molecular mechanisms in the incidence and manifestation of heart diseases.

Mouse embryonic fibroblasts from CD38 knockout mice are resistant to oxidative stresses through inhibition of reactive oxygen species production and Ca(2+) overload.

CD38 is a multifunctional enzyme that has both ADP-ribosyl cyclase and cADPR hydrolase activities, being capable of cleaving NAD(+) to cyclic ADP ribose (cADPR) and hydrolyzing cADPR to ADPR. It has been reported that there is markedly a reduction of cADPR and elevation of NAD in many tissues from CD38 knockout (CD38(-/-)) mice. Cyclic ADPR is a potent second messenger for intracellular Ca(2+) mobilization, and NAD is a key cellular metabolite for cellular energetic and a crucial regulator for multiple signaling pathways in cells. We hypothesize that CD38 knockout may have a protective effect in oxidative stresses through elevating NAD and decreasing cADPR. In the present study, we observed that the mouse embryonic fibroblasts (MEFs) from CD38(-/-) mice were significantly resistant to oxidative stress such as H(2)O(2) injury and hypoxia/reoxygenation compared with wild type MEFs (WT MEFs). We further found that production of reactive oxygen species (ROS) and concentrations of intracellular Ca(2+) ([Ca(2+)](i)) in CD38(-/-) MEFs were markedly reduced compared with WT MEFs during hypoxia/reoxygenation. Coincidence with these results, a remarkably lower mRNA level of Nox1, one of the enzymes responsible for ROS generation, was observed in CD38(-/-) MEFs. Furthermore, we found that transcription of Nox1 mRNA in WT MEFs could be elevated by calcium ionophore ionomycin in a dose-dependent manner, indicating that the expression of Nox1 mRNA can be regulated by elevation of intracellular [Ca(2+)]. Therefore we concluded that CD38(-/-) MEFs are resistant to oxidative stresses through inhibiting intracellular Ca(2+) overload and ROS production which may be regulated by Ca(2+)-mediated inhibition of Nox1 expression. Our data should provide an insight for elucidating the roles of CD38 in oxidative stresses and a novel perspective of dealing with the ischemia/reperfusion-related diseases.