Lipoproteins are complex aggregates of lipids and proteins that render lipids suitable for transport in circulation between tissues. Recently, it has been shown that HDL, besides acting as a cellular cholesterol carrier, is also functioning as a miRNA transporter (Fig. 1.1) (Vickers and Remaley 2012; Wagner et al. 2013). Wagner et al. demonstrated circulating miRNAs to be bound to HDL and also, but in a lower amount, to LDL fractions from plasma of healthy subjects as well as in plasma of patients with stable coronary artery disease (CAD) or acute coronary syndromes (ACS) (Wagner et al. 2013). The miRNA levels between patients and healthy subjects were not significantly different. In vitro, miRNAs bound to HDL were not taken up by ECs, SMCs, and peripheral blood mononuclear cells (PBMCs), yet patient-derived HDL transiently modulated miRNA expression in SMCs and PBMCs. Endogenous expression of miR-92a, miR-126, and miR-223 in PBMCs was increased after longer incubation with HDL derived from patients with stable CAD or ACS, in contrast to the effect of HDL from healthy subjects in which the expression of these miRNAs was reduced (Wagner et al. 2013). These results are in line with previous studies showing distinct biological functions for HDL derived of CAD patients versus healthy subjects in several cellular models (Besler et al. 2011). Interestingly, short incubation times of 1 h reduced the expression of miR-126 as a consequence of lower transcription or destabilization of the unprocessed form pri- miR-126 (Wagner et al. 2013). Thus, lipoprotein-bound miRNAs appeared to be not efficiently delivered to endothelial, smooth muscle, and peripheral blood mononuclear cells suggesting that this pool of miRNAs is not well suited to regulate the function of cells in vitro.
In earlier studies, Vickers et al. did not only describe a singular profiling for miRNA-HDL complexes but also revealed a specific signature in patients with familial hypercholesterolemia. The mechanisms by which miRNAs bind to HDL are still not clarified; what it is known is that only mature miRNAs are present and bound through divalent cation bridges (Janas and Yarus 2006; Vickers and Remaley
2012). HDL-miRNA complexes include miR-22, miR-105, and miR-106a being miR-223 the most abundant miRNA. In human familiar hypercholesterolemia, miR-223 is dramatically upregulated; however, it could not be validated as a possible biomarker in circulation for atherosclerosis (Vickers and Remaley 2012; Fichtlscherer et al. 2010). It remains unclear how this miRNAs can display biological functions, as only unprocessed forms of miRNAs are able to bind to the RNA- induced silencing complex. Nevertheless, Vickers et al. demonstrated that indeed HDL is able to transport miR-105 to hepatocytes for altering gene expression (Vickers and Remaley 2012). It was further shown that miR-223 derived from HDL can indeed be transferred to ECs, where it directly targets ICAM-1 as an example of an extracellular miRNA regulating gene expression in cells where it is not transcribed (Wagner et al. 2013). These data together suggest that anti-inflammatory properties of HDL are in part conferred through miR-223 delivery and translational repression of ICAM-1 in ECs. Even though the contribution of
HDL-miRNAs to the total miRNA-pool appears to be rather low, the finding of miRNAs in lipoprotein fractions may represent a new system of intercellular communications.
In conclusion, the identification of this high variety of miRNAs and their carriers in circulation challenges a new generation of potential biomarkers and therapeutic approaches. In this fashion, circulating miRNAs associated with their lipid-based carriers have already emerged as important mediators of novel treatment strategies. Specific miRNAs or miRNA inhibitors combined with particular lipoproteins or packed into vesicles after application into blood circulation may alter the extracellular miRNA signature and thus gain a therapeutic advantage. In atherogenesis, several studies have already demonstrated beneficial and protective effects by delivery of selected miRNAs linked to lipoproteins or engulfed by natural vesicles (Hergenreider et al. 2012; Zernecke et al. 2009).