HNRNPA2B1: RNA-Binding Protein That Orchestrates Smooth Muscle Cell Phenotype in Pulmonary Arterial Hypertension

Abstract

Background: RNA-binding proteins are master orchestrators of gene expression regulation. They regulate hundreds of transcripts at once by recognizing specific motifs. Thus, characterizing RNA-binding proteins targets is critical to harvest their full therapeutic potential. However, such investigation has often been restricted to a few RNA-binding protein targets, limiting our understanding of their function. In cancer, the RNA-binding protein HNRNPA2B1 (heterogeneous nuclear ribonucleoprotein A2B1; A2B1) promotes the pro-proliferative/anti-apoptotic phenotype. The same phenotype in pulmonary arterial smooth muscle cells (PASMCs) is responsible for the development of pulmonary arterial hypertension (PAH). However, A2B1 function has never been investigated in PAH.

Method: Through the integration of computational and experimental biology, the authors investigated the role of A2B1 in human PAH-PASMC. Bioinformatics and RNA sequencing allowed them to investigate the transcriptome-wide function of A2B1, and RNA immunoprecipitation and A2B1 silencing experiments allowed them to decipher the intricate molecular mechanism at play. In addition, they performed a preclinical trial in the monocrotaline-induced pulmonary hypertension rat model to investigate the relevance of A2B1 inhibition in mitigating pulmonary hypertension severity.

Results: They found that A2B1 expression and its nuclear localization are increased in human PAH-PASMC. Using bioinformatics, they identified 3 known motifs of A2B1 and all mRNAs carrying them. In PAH-PASMC, they demonstrated the complementary nonredundant function of A2B1 motifs because all motifs are implicated in different aspects of the cell cycle. In addition, they showed that in PAH-PASMC, A2B1 promotes the expression of its targets. A2B1 silencing in PAH-PASMC led to a decrease of all tested mRNAs carrying an A2B1 motif and a concomitant decrease in proliferation and resistance to apoptosis. Last, in vivo A2B1 inhibition in the lungs rescued pulmonary hypertension in rats.

Conclusions: Through the integration of computational and experimental biology, the study revealed the role of A2B1 as a master orchestrator of the PAH-PASMC phenotype and its relevance as a therapeutic target in PAH.

Keywords: RNA-binding proteins; bioinformatics; computational biology; pulmonary arterial smooth muscle cells.

Figure 1.. A2B1 expression is increased in PAH_PASMC.

(A) Localization and quantification of A2B1 protein in PASMC within the lung of healthy and PAH subjects (A2B1 in red, α-SMA in white, Nuclei in blue). Quantification of A2B1 (B) protein and (C) mRNA in primary cultured PASMC isolated from healthy and PAH subjects. (D) Quantification of A2B1 expression from online available PAH-PASMC RNA-sequencing (GSE144274). (E) Localization of A2B1 in primary cultured PASMC assessed by nucleo-cytoplasmic separation showing the increased nuclear expression of A2B1 in PAH-PASMC compared to Ctrl-PASMC. (F) Localization of A2B1 in primary cultured PASMC assessed by immunofluorescence showing an accumulation of A2B1 in the nuclei in PAH-PASMC compared to Ctrl-PASMC. The histogram represents the fluorescence intensity throughout the measured area (dotted white line). Dapi is shown in the blue spectrum and A2B1 in the red spectrum. (G) Alignment of both isoform of A2B1 (HNRNPA2 -NM002137.4 - and HNRNPB1 – NM031243.3) together showing the only sequence mismatch between them. These 36 nucleotides that are absent from the A2 isoform are annotated in the uniprot database to contain a nuclear localization. The alignment was performed using Clustal Omega. (H) mRNA expression of the isoform B1 in PAH-PASMC compared to Ctrl-PASMC.

Figure 2.. A2B1 regulates the cell cycle by promoting expression of its targets in PAH-PASMC.

Panel (A) summarizes our analysis pipeline to identify the mRNA targets of A2B1 in human and its function in PAH-PASMC. (B) To define the putative targets of A2B1 we used the A2B1 motifs sequences (ATtRACT database) and search for them on mRNA sequences (UTRs and CCDS databases). We identified three motifs with enough complexity to be accurately detected and all mRNAs containing these motifs which are the putative targets of A2B1. We also found that these three motifs were enriched in 3’UTRs region. (C) To apply this knowledge in PAH, we identified all putative targets of A2B1 in an RNA-sequencing of primary cultured Ctrl vs PAH PASMC. (D) To define the subset of mRNAs actively regulated by A2B1 in PAH-PASMC, we correlated the expression of A2B1 with its putative targets. From this correlation one clear pattern emerged – a subset of significantly dysregulated putative targets of A2B1 was positively correlated with A2B1 (R > 0.7) in PAH-PASMC – These targets are the A2B1 targets in PAH-PASMC. The density plots shows the percent of putative targets of A2B1 in function of the Spearman correlation coefficient (R) (E) To define the function of A2B1 in PAH-PASMC we performed a gene ontology enrichment analysis using the A2B1 targets for each motif. We found that most of the enriched gene ontology term were related to regulation of the cell cycle regulation. The GOchord graph shows the link between each motif and the enriched gene ontology term associated.

Figure 3.. Each A2B1 RNA recognition motif regulates a unique aspect of the cell cycle.

Mapping of positively correlated A2B1 targets on KEGG pathways: (A) Cell cycle checkpoint, (B) DNA replication and (C) Nuclear pore complex.

Figure 4.. Experimental validation of our bio-informatic findings on primary cultured PAH-PASMC.

Panel (A), (B) and (C) respectively shows for the yellow motif, the blue motif and the red motif, the correlation between A2B1 representative targets and A2B1 in RNA-seq (purple) and RT-qPCR (green) on the left panel. In addition, the right panels shows, respectively, the expression of the representative targets in RNA-seq and RT-qPCR and the results of an RNA-immunoprecipitation (RNA-IP) demonstrating the up-regulation of each target in PAH-PASMC vs Ctrl-PASMC and their binding to A2B1 in PAH-PASMC. (D) quantification of Ki-67-positive PASMC (red) to assess proliferation and (E) quantification of cleaved caspase 3 positive PASMC to assess apoptosis (red) in PASMC Ctrl vs PAH.

Figure 5.. A2B1 silencing down-regulates its predicted targets and restores the pro-proliferative and anti-apoptotic imbalance of PAH-PASMC.

(A) A2B1 and isoform B1 mRNA expression. (B) A2B1 protein expression in PAH-PASMC treated with Si-Scrm or Si-A2B1. (C) Gene set enrichment analysis of RNA-sequencing comparing PAH-PASMC vs Ctrl-PASMC. This analysis revealed that as a set of targets, the targets of A2B1 as well as all the pathways we found to be regulated by A2B1 in Fig.2 were significantly down-regulated when A2B1 was silenced. (D) mRNA expression of A2B1 selected targets demonstrating their down-regulation in PAH-PASMC treated with Si-A2B1. In addition, PAH-PASMC treated with Si-A2B1 demonstrate (E) a decrease in proliferation and (F) an increase in apoptosis.

Figure 6.. A2B1 inhibition in the lungs rescues MCT-induced PH in rats.

Animals were followed by echocardiography for four weeks after MCT injection. (A) Protocol schematic. Black arrow correspond to the time of MCT injection, white lozenge to time of echocardiography and white arrow to time of si-RNA instillation. We measured (B) PAAT, (C) right ventricular thickness and (D) the ratio between main pulmonary artery and the aorta. At the end of the protocol we measured (E) RVSP, (F) LVSP, (G) heart rate, (H) Fulton index and (I) lung weight.

Figure 7.. Pulmonary inhibition of A2B1 reverses vascular remodeling and the pro-proliferative/anti-apoptotic phenotype of SMCs.

Measurement of (A) vascular wall thickness, (B) A2B1 expression in SMCs, (C) SMCs proliferation, (D) apoptosis, (E) MCM6, (E) AURKA and (F) POLR2D in rats.