We hypothesized that circulating ECs (CECs) and/or peripheral blood mononuclear cells (PBMC) may define a subset of differentially regulated biomarkers in IPAH and DaPAH that may lead to earlier diagnosis, better means for assessing responses to therapy, and possibly identify novel targets for therapeutic interventions. CECs are a valuable source of material for studying diseases characterized by EC dysfunction. However, no clear methodology existed for isolating clinically relevant numbers of CECs. This project used flow cytometry to develop a methodology for isolating relevant numbers of viable CECs from healthy and PAH subjects. In addition, peripheral blood (PB) specimens were obtained for CECs and PBMCs for microarrays from healthy and PAH subjects. Plasma was also saved for application to cultured microvascular cells. A subset of subjects had a right heart catheterization to assess pulmonary pressures and obtain pulmonary blood specimens. The protocol started enrolling in 2006 and closed to enrollment in 2009 (n=31). Available data suggested there was no trend towards CEC enrichment in pulmonary vein blood compared to PB for healthy (4.4 vs 4.8 CEC/ml) and PAH (2.4 vs 3.0 CEC/ml) subjects. There was a trend towards CEC enrichment in pulmonary artery blood compared to PB for healthy (13.8 vs 4.8 CEC/ml) and PAH (3.3 vs 3.0 CEC/ml) subjects. In 2010-11, total RNA was processed from PBMCs for genome-wide expression analysis. An abstract on PBMC differential gene expression patterns in PAH was presented at the ATS meeting (AJRCCM 183:A5511, 2011). Patterns reflected treatment related signatures and underlying disease pathophysiology. In 2011-13, using PBMC expression profiles from 10 PAH with 10 matched healthy subjects we identified >230 differentially regulated genes at a 20% false discovery rate (FDR). Ingenuity Pathway Analysis identified gene signatures for inflammation, cell-to-cell signaling & interaction, cytoskeletal rearrangement, cellular movement, hemostasis and cell death. In vitro data from our collaborating lab showed spironolactone (SP) suppresses phorbol 12-myristate 13-acetate-induced (PMA; an AP-1 activator) inflammatory gene transcription in primary human PAECs. To explore the effect of SP on PAH-associated vascular inflammation we conducted a promoter level analysis of the upregulated genes we identified in PAH subjects. Bioinformatics software identified AP-1 as a key transcriptional regulator. Experiments using PBMCs isolated from healthy subjects and stimulated with PMA demonstrate that SP suppresses AP-1 inducible, PAH-associated genes in a dose-dependent manner. Similarly, in PBMCs from healthy and PAH subjects, SP strongly suppressed the basal expression of genes upregulated in PBMCs of PAH patients. In 2011-12 we continued to develop a bioassay assessing global transcriptomic changes induced by plasma from PAH compared to healthy subjects using microarrays. Exposure of human PAECs to plasma from 5 PAH subjects compared to 5 matched controls, identified >300 differentially expressed transcripts at a 10% FDR. We also explored (2012-13) the gene expression changes in cultured human PAECs induced by plasma from PAH subjects and found that 20% of the signature overlapped with gene expression changes induced following BMPR-2 gene silencing in PAECs. Importantly, >90% of the overlap was directionally discordant, suggesting circulating factors may work to counter-regulate genotypic and phenotypic abnormalities that drive PAH. In 2013-14, we expanded our PBMC expression findings, by starting to collect data from all other published human PAH PBMC genome-wide expression profiling studies for a meta-analysis. In 2014-15 we completed the necessary steps of data collection, annotation and aggregation of all published human PAH PBMC expression profiling studies. Meta-analysis of all studies comparing gene expression profiles from IPAH and DaPAH to healthy subjects identified 579 and 1186 differentially expressed transcripts, respectively, at a 1% FDR. Interestingly, comparing gene expression profiles from IPAH and DaPAH patients yielded no differentially expressed transcripts at a 1% or 5% FDR. Defining a robust genomic signature across multiple PBMC studies may highlight previously unrecognized gene expression patterns. In 2015-16, we added data from a recently published blood transcriptomic study conducted in chronic respiratory disease patients (PAH n=8) compared to healthy controls (n=28). With the updated data set, we completed further downstream analyses. Differentially expressed genes in PAH previously identified in the literature demonstrated only modest reproducibility, defined as a FDR 10% by meta-analysis. Different bioinformatic approaches consistently identified inflammatory signaling and regulators of cell proliferation as overrepresented pathways among the shared genomic signature in IPAH and DaPAH (abstract at ATS; AJRCCM 193: A4619, 2016). In 2016-17, >30 differentially expressed genes in the combined PAH cohort were selected for validation using samples from this protocol. Importantly, >900 genes identified by meta-analysis were not previously reported in the 7 published studies. Functional analysis of the 1269 differentially expressed genes derived from the combined PAH cohort identified several inflammatory signaling pathways as significantly overrepresented. Among these EIF2 and mTOR signaling have been linked to PAH pathogenesis demonstrating the validity of our meta-analytic approach. We also found that in circulating immune cells interferon plays a prominent role in the inflammatory component of PAH pathobiology. Next, we performed a gene set analysis using data from all 22753 genes included in the meta-analysis without using arbitrary gene selection criteria. The analysis supported the prominence of interferon signaling in circulating immune cells of PAH patients. Lastly, we conducted a promoter-level analysis to determine if there was enrichment for specific transcription factor binding sites in upregulated genes from the combined PAH cohort. The upregulated genes were compared to a group of genes that were not differentially regulated between PAH and healthy subjects. The results show highly significant enrichment of binding sites for the interferon regulatory factor family of transcription factors among upregulated PAH genes. Project data was presented at the 2017 ACC Annual Scientific Session as an invited talk entitled Genomics: Meta-Analysis of Blood Expression Profiling Studies in PAH. In 2017-18, bioinformatics analyses were finalized and a manuscript reporting the findings of PBMC gene expression meta-analysis is in preparation. The gene signature identified through meta-analysis will be explored in other cohorts to determine if these genes are impacted by PAH therapies. For example, the PAH Biobank has samples suitable for gene expression profiling of treatment naive PAH patients compared to healthy controls and compared to each other following initiation of PAH therapy. Additionally, in our ongoing Spironolactone Randomized Interventional Trial in PAH (12-CC-0211) PBMCs are serially collected for RNA and protein isolation. Genome-wide expression profiles in these subjects will be analyzed using gene set enrichment of the PAH gene signature identified in the meta-analysis.
| Tropea, Margaret M; Harper, Bonnie J A; Graninger, Grace M et al. (2014) Isolation of a circulating CD45-, CD34dim cell population and validation of their endothelial phenotype. Thromb Haemost 112:770-80 |
| Desai, Aditi; Glaser, Alexander; Liu, Delong et al. (2009) Microarray-based characterization of a colony assay used to investigate endothelial progenitor cells and relevance to endothelial function in humans. Arterioscler Thromb Vasc Biol 29:121-7 |
