Substantial Dysregulation of miRNA Passenger Strands Underlies the Vascular Response to Injury

Abstract

Vascular smooth muscle cell (VSMC) dedifferentiation is a common feature of vascular disorders leading to pro-migratory and proliferative phenotypes, a process induced through growth factor and cytokine signaling cascades. Recently, many studies have demonstrated that small non-coding RNAs (miRNAs) can induce phenotypic effects on VSMCs in response to vessel injury. However, most studies have focused on the contribution of individual miRNAs. Our study aimed to conduct a detailed and unbiased analysis of both guide and passenger miRNA expression in vascular cells in vitro and disease models in vivo. We analyzed 100 miRNA stem loops by TaqMan Low Density Array (TLDA) from primary VSMCs in vitro. Intriguingly, we found that a larger proportion of the passenger strands was significantly dysregulated compared to the guide strands after exposure to pathological stimuli, such as platelet-derived growth factor (PDGF) and IL-1α. Similar findings were observed in response to injury in porcine vein grafts and stent models in vivo. In these studies, we reveal that the miRNA passenger strands are predominantly dysregulated in response to vascular injury.

Keywords: biomarker; miRNA expression and regulation; passenger miRNA; porcine vein graft and stent models; smooth muscle cells; vascular injury.

Figure 1

Low passenger strands expression compared to the guide strands in VSMCs and in vivo. One hundred hairpins were analyzed by TaqMan Low Density Array (TLDA) in human VSMCs (average of n = 3), the porcine models of vein graft failure and in stent restenosis (average of n = 6 for both the porcine models, except for the 7 days DES (drug eluting stent) condition, where n = 5). The level of expression (relative expression, %) was classified into three groups: high (Ct < 23), medium (Ct between 23 and 28) and low (Ct > 28) expressions. These groups are represented as a % in each pie chart. (A) Each pie chart represents the global expression of the guide and passenger strands in human VSMCs. The heat map illustrates the relative abundance of the number of miRNA hairpins that were consistently expressed across three independent patients and represents the level of expression of the guide versus passenger strands. (B) Each pie chart represents the global expression of the guide and passenger strands in the porcine model of in stent restenosis and vein graft failure. (C) Each pie chart represents the expression of the guide versus passenger strands within each independent hairpin.

Figure 2

High passenger strand dysregulation compared to the guide strand dysregulation in VSMCs and in vivo. The graphs show the percentage of dysregulated miRNAs for each form in the three independent models. (A) The relative dysregulation of each strand of the miRNA hairpin was quantified after stimulation by platelet-derived growth factor (PDGF) and IL-1α in human VSMCs (the statistics have been made between the passenger and guide strands within each condition; unpaired t-test; ** p < 0.01 for passenger vs guide strands). The heat map illustrates the relative dysregulation (in relative quotient (RQ)) of a number of miRNA hairpins after stimulation by PDGF and IL-1α across three independent patient samples (average of n = 3). (B) The dysregulation of the guide and passenger strands was quantified at 7 days following stenting (BMS (bare-metal stent) or DES (drug eluting stent)) or at 7 days following vein grafting in vivo (n = 6 for both the porcine models, except for the 7 days in the DES condition where n = 5) (the statistics have been made between the passenger and guide strands within each condition; unpaired t-test; * p < 0.05 and ** p < 0.01; and vs. the corresponding control (unstented control animals or saphenous), # p < 0.05 and ### p < 0.001).