miRNA-548c: a specific signature in circulating PBMCs from dilated cardiomyopathy patients

Abstract

High fidelity genome-wide expression analysis has strengthened the idea that microRNA (miRNA) signatures in peripheral blood mononuclear cells (PBMCs) can be potentially used to predict the pathology when anatomical samples are inaccessible like the heart. PBMCs from 48 non-failing controls and 44 patients with relatively stable chronic heart failure (ejection fraction of ≤ 40%) associated with dilated cardiomyopathy (DCM) were used for miRNA analysis. Genome-wide miRNA-microarray on PBMCs from chronic heart failure patients identified miRNA signature uniquely characterized by the downregulation of miRNA-548 family members. We have also independently validated downregulation of miRNA-548 family members (miRNA-548c & 548i) using real time-PCR in a large cohort of independent patient samples. Independent in silico Ingenuity Pathway Analysis (IPA) of miRNA-548 targets shows unique enrichment of signaling molecules and pathways associated with cardiovascular disease and hypertrophy. Consistent with specificity of miRNA changes with pathology, PBMCs from breast cancer patients showed no alterations in miRNA-548c expression compared to healthy controls. These studies suggest that miRNA-548 family signature in PBMCs can therefore be used to detect early heart failure. Our studies show that cognate networking of predicted miRNA-548 targets in heart failure can be used as a powerful ancillary tool to predict the ongoing pathology.

Keywords: Biomarker; Canonical signaling networks; Dilated cardiomyopathy; Peripheral blood mononuclear cells (PBMC); microRNA-548.

Figure 1. Quantitative Real Time PCR Analysis

(a) Dendrogram analysis based on real time PCR for 8 miRNAs (miRNA-1, -7, -29, -145, -378, -342, -125, -181) in PBMCs from 48 controls and 44 DCM patients. Expression pattern of 8 miRNAs is shown as a heat map. (b) Log 2 transformation of real time PCR data on 8 miRNAs from PBMCs shows no significant differences between control and DCM patients (data presented as mean ± standard deviation).

Figure 2. Genome wide miRNA-microarray

Heat map showing the expression of miRNAs from PBMCs altered in controls (n=7) and DCM patient samples (n=7). Analysis of the heat map shows a clear difference in the expression of miRNA profile in PBMCs from DCM patients and non-failing controls.

Figure 3. Validation of miRNA-548 family members by real time PCR

(a & b) RNA was isolated from the PBMCs from the control and DCM patients on an independent set of controls and patients with DCM. Reverse transcription was carried out to analyze the miRNA-548 members, miRNA-548c (41 Controls and 37 DCM) (data is presented as fold over internal control – RNU6). Similar analysis was carried out for miRNA-548i.

Figure 4. ROC analysis

(a) Receiver Operating characteristic (ROC) curve for miRNA-548c shows area under the curve to be 85.51% with a significance value of p<0.0001 given that the data generated is from miRNA expression from PBMC. (b) ROC curve for BNP from the plasma, area under the curve is 94.62% with a significance value of p<0.0001. BNP values were measured by ELISA from non failing controls and DCM patients which were used for ROC analysis. (c) ROC curve for the simple linear combination of miRNA-548c with BNP showed increased strength of sensitivity as measured by the area under the curve (95.86%) with a significance value of p<0.0001.

Figure 5. Ingenuity Pathway Analysis (IPA) on predicated targets of miRNA-548

Top network associated with 15 predicated targets of miRNA-548 is shown. Many of the molecules represented in the network are critical molecules in stress signaling indicating that the predicted targets may show a reflection of ongoing pathology in specific tissues.

Figure 6. Evaluation of miRNA-548 in PBMCs from metastatic cancer patients

RNA was isolated from the PBMCs of metastatic cancer patients and controls. The RNA was subjected to real time PCR analysis for miRNA-548c (data is presented as fold over internal control). No appreciable difference in expression pattern of miRNA-548c was observed between PBMCs from controls and cancer patients.