Pronethalol decreases RBPJκ to reduce Sox2 in cerebral arteriovenous malformation

Cerebral arteriovenous malformations (AVMs) are common vascular malformations that tend to rupture and cause hemorrhagic strokes.¹ Disruptions in the integrity of the vascular endothelium and the endothelial cell (EC) differentiation give rise to the formation of cerebral AVMs.^1–4 In previous studies, we reported that endothelial–mesenchymal transitions (EndMTs) contributed to cerebral AVMs in that ECs lost their identity and gained mesenchymal plasticity to cause lumen disorder.⁵ We showed that unwanted induction of Sry-box 2 (Sox2) signaling was responsible for the EndMTs in cerebral AVMs, and identified the beta-adrenergic antagonist pronethalol as an inhibitor of Sox2 expression that stabilized EC differentiation and lumen formation, thereby limiting the cerebral AVMs.⁵ We also showed that depletion of the beta-adrenergic receptors had no effect on the Sox2 expression, which suggested that the beta antagonists exerted their effect through other pathways.⁵ However, the nature of these pathways remains unclear. Here, we hypothesize that the beta-adrenergic antagonists directly inhibit the Notch-associated transcription factor recombination signal binding protein for immunoglobulin kappa J (RBPJκ), a key component of Notch signaling, to limit Sox2 induction and improve cerebral AVMs.

The methods are described as follows. Human brain microvascular endothelial cells (HBMECs) were obtained from ScienCell Research Laboratories (Carlsbad, CA, USA) and cultured as per the manufacturer’s protocol.⁵ Transient transfections of HBMECs with siRNA (Silencer®¹ Pre-designed siRNA; Applied Biosystems, Foster City, CA, USA) were optimized and performed as previously described.⁶ Silencer®¹ Pre-designed siRNAs were acquired for Notch1, Jagged1 and 2, and RBPJκ. The siRNA sequences were as follows: UCGUCUACCUGGAGAUUGAtt (sense), UCAAUCUCC AGGUAGACGAtg (antisense) for Notch1; CACCUGUGACUGUAACAAAtt (sense), UUUGUUACAGUCACAGGUGaa (antisense) for Jagged1; CAACGACUGCCUUCCCGAUtt (sense), AUCGGGAAGGCAGUCGUUGgg (antisense) for Jagged2 CAAACAGCGAGGGAAGUUAtt (sense), UAACUUCCCUCGCUGUUUGtg (antisense) for RBPJκ.

Real-time polymerase chain reaction (PCR) analysis was performed as previously described.⁷ Glyceraldehyde 3-phosphate dehydrogenase (Gapdh) was used as a control gene.⁷ Primers and probes for the mouse genes Sox2, Notch1, Jagged 1 and 2, and RBPJκ were obtained from Applied Biosystems as part of Taqman®¹ Gene Expression Assays. Specific antibodies were used to perform ChIP in order to enrich the genomic DNA as previously described.⁸ We used specific antibodies for RBPJκ (ab25949; Abcam, Cambridge, MA). The primers for real-time PCR were designed as previously reported.⁷

We reported that bone morphogenetic protein 6 (BMP6) activated activin receptor-like kinase 3 (ALK3) to induce Notch1 and Jagged1 and Jagged2 in cerebral ECs, which in turn enhanced Notch signaling and induced Sox2.⁹ To determine if the beta-adrenergic antagonists suppress Sox2 expression by regulating BMP and Notch signaling, we repeated the experiments HBMECs, where matrix Gla protein (MGP) was depleted by using the gene-editing tool of clustered regularly interspaced short palindromic repeats and its associated protein 9 (CRISPR/Cas9).⁵ Such cells were previously shown to be a cell model that mimicked the abnormal ECs in cerebral AVMs⁵ and were referred to as MGP CRISPR cells. We treated the MGP CRISPR cells with BMP6, and found that BMP6 induced Sox2, Notch1, Jagged1, and Jagged2 (Figure 1A). We further performed gene knockdown to reduce RBPJκ and found that this knockdown abolished the Sox2 induction (Figure 1B). We previously found that pronethalol was the most efficient among the beta-adrenergic antagonists in improving cerebral AVMs by suppressing mesenchymal markers and lumen-associated genes through Sox2 inhibition.⁵ Therefore, we chose pronethalol for treating the MGP CRISPR cells and to examine the expression of Notch1, Jagged1, Jagged2, and RBPJκ. The results revealed that pronethalol significantly reduced the expression of RBPJκ but not that of Notch1, Jagged1, or Jagged2 (Figure 1C). We examined the RBPJκ DNA-binding to the Sox2 promoter region by using chromatin immunoprecipitation (ChIP) assays.¹⁰ The results showed highly enriched DNA-binding around the RBPJκ binding site in the Sox2 promoter of the MGP CRISPR cells as compared to the control cells (Figure 1D). Excess RBPJκ DNA-binding was abolished by the pronethalol treatment in the MGP CRISPR cells (Figure 1D). Together, the results suggest that the beta-adrenergic antagonists reduce RBPJκ, thereby limiting excess BMP/Notch signaling and Sox2 induction (Figure 1E).

Figure 1.

Pronethalol decreases RBPJκ to reduce Sox2 expression. (A) Expression of Sox2, Notch1, Jagged1, and Jagged2 in HMECs and MGP CRISPR cells treated with or without BMP6 (n = 6). (B) Sox2 expression in HBMECs after treatment with BMP6 and transfection with specific siRNAs to Notch1, Jagged1 and Jagged2, or RBPJκ (n = 5). (C) Expression of Sox2, Notch1, Jagged1, and Jagged2 in HMECs and MGP CRISPR cells treated with or without pronethalol as shown by real-time PCR and immunoblotting (n = 6). (D) Enrichment of RBPJκ DNA-binding in the promoter of Sox2 genes as examined by ChIP assays (n = 6). (E) How pronethalol decreases RBPJκ to suppress Sox2 expression.

*p < 0.05; ***p < 0.0001.

BMP6, bone morphogenetic protein; ChIP, chromatin immunoprecipitation; CRISPR, clustered regularly interspaced short palindromic repeats; Ctr, control; HBMECs, human brain microvascular endothelial cells; HMECs, human microvascular endothelial cells; IgG, immunoglobulin; MGP, matrix Gla protein; PCR, polymerase chain reaction; RBPJκ, recombination signal binding protein for immunoglobulin kappa J; SCR, Scrambled siRNA

We have reported a range of Sox2 inhibition for different beta-adrenergic antagonists, including nebivolol, acebutolol, atenolol, bisoprolol, carvedilol, labetalol, propranolol, metoprolol, nadolol, and sotalol.⁵ Our current results suggest that the beta-adrenergic antagonists inhibit the induction of Sox2 through a reduction of RBPJκ, which provides a novel role for the beta-adrenergic antagonists in the crosstalk between the BMP and Notch signaling pathways.

Our previous results show that pronethalol limits the formation of cerebral AVMs,⁵ whereas this study links the pronethalol treatment with the regulation of RBPJκ and proposes that the inhibition of RBPJκ may prevent ill-fated EC transitions in cerebral AVMs. Pronethalol is limited to clinical use because of neurological side effects in clinical trials and moderately potent carcinogenesis in mice.¹¹ However, other beta-adrenergic antagonists have been tested, including propranolol, which was shown to improve AVMs in patients who orally received 20 mg three times per day for 40 days.¹² This suggests that a similar mechanism involving RBPJκ may underlie the beneficial effect of propranolol on cerebral AVMs. Future studies may identify new strategies for the treatment of cerebral AVMs by focusing on the regulation of RBPJκ in ECs.