Identification of candidate genes in scleroderma-related pulmonary arterial hypertension

Abstract

We hypothesize that pulmonary arterial hypertension (PAH)-associated genes identified by expression profiling of peripheral blood mononuclear cells (PBMCs) from patients with idiopathic pulmonary arterial hypertension (IPAH) can also be identified in PBMCs from scleroderma patients with PAH (PAH-SSc). Gene expression profiles of PBMCs collected from IPAH (n = 9), PAH-SSc (n = 10) patients, and healthy controls (n = 5) were generated using HG_U133A_2.0 GeneChips and were processed by the RMA/GCOS_1.4/SAM_1.21 data analysis pipeline. Disease severity in consecutive patients was assessed by functional status and hemodynamic measurements. The expression profiles were analyzed using PAH severity-stratification, and identified candidate genes were validated with real-time polymerase chain reaction (PCR). Transcriptomics of PBMCs from IPAH patients was highly comparable with that of PMBCs from PAH-SSc patients. The PBMC gene expression patterns significantly correlate with right atrium pressure (RA) and cardiac index (CI), which are known predictors of survival in PAH. Array stratification by RA and CI identified 364 PAH-associated candidate genes. Gene ontology (GO) analysis revealed significant (Z(score) > 1.96) alterations in angiogenesis genes according to PAH severity: matrix metalloproteinase 9 (MMP9) and vascular endothelial growth factor (VEGF) were significantly upregulated in mild as compared with severe PAH and healthy controls, as confirmed by real-time PCR. These data demonstrate that PBMCs from patients with PAH-SSc carry distinct transcriptional expression. Furthermore, our findings suggest an association between angiogenesis-related gene expression and severity of PAH in PAH-SSc patients. Deciphering the role of genes involved in vascular remodeling and PAH development may reveal new treatment targets for this devastating disorder.

Figure 1. Hierarchical clustering of genes significantly affected by IPAH or PAH-SSc

The 287 PAH-associated genes were combined and clustered using MeV as described in Materials and Methods. Each column represents a PAH affected patient and each row represents the expression pattern of a specific gene throughout all patients. Hierarchical clustering, which was conducted using uncentered Pearson correlation (complete linkage), identified 6 major clusters of which one cluster (third from the bottom) demonstrated clear expression differences between severe and mild patients in both diseases. The specific group of genes in the neighboring cluster (third from the top) demonstrated severity- and disease-independent upregulation. Both groups of genes are highlighted with blue rectangles, and most representative genes within each group are listed on the right. Red color indicates up-regulation and green color indicates down-regulation of gene expression relative to healthy controls, with color intensity corresponding to the fold-change amplitude (fold-change scale shown on the left).

Figure 2. Association between hemodynamic measurements and genomic changes in PAH-SSc patients

The gene expression profiles of PAH-SSc patients were sorted by severity in ascending order using each hemodynamic (CI, RA, PA, and PVRI) or functional (WHO functional class and 6 MWD) measurement as a severity indicator. The SAM analysis (100 permutations) was performed comparing severity-polar patient groups (i.e., groups of patients deemed have severe versus mild disease). After each of 7 cycles (x axes) the less polar patients (middle profile) was removed and SAM analysis repeated. The number of compared profiles depicted on the x axes (mild vs severe) and groups that were compared for each cycle are presented as circle under the x axes. An open circle represents the patient with the most mild PAH and solid circle represents patient with the most severe PAH. The gray scale intensity corresponds to the PAH severity in SS patients. The beginning comparison was between 5 mild and 5 severe profiles, and final comparison was between 2 mild and 2 severe profiles. The lowest FDR point for each hemodynamic or functional measurement is labeled with corresponding abbreviation.

Figure 3. Real time PCR expression of genes associated with PAH severity

The relative MMP9 and VEGF message abundance was detected by real time RT-PCR using total PBMC RNA. Results are represented as mean ± SEM. Statistical significance (P < 0.05) of the gene expression difference between patients and healthy volunteers was derived with two-tailed t-test of normalized to three different internal controls threshold cycle (Ct) values.