**4. miRNA and osteogenesis in the vascular wall**

Cardiovascular calcification is an active, cell-regulated process. Various studies provide evidence of phenotypic transition or transition/dedifferentiation of mature SMCs or VICs into an osteogenic phenotype — a key feature in cardiovascular calcification. In medial calcification, SMCs undergo dedifferentiation from a contractile to a pro-atherogenic synthestic phenotype, lose the expression of their marker genes, acquire osteogenic markers, and deposit a mineral‐ ized bone-like matrix. In valvular calcification, VICs can undergo the transition to osteoblastlike bone-forming cells [58]. Recently, a novel concept emerged of circulating cells harboring osteogenic potential that can home to atherosclerotic lesions and contribute to intimal calcifi‐ cation [59, 60]. Comparing the sources of cells that contribute to atherosclerotic intimal calcification revealed that SMCs are the major contributors that reprogram its lineage towards osteochondrogenesis/blastogenesis; circulating bone marrow-derived cells, however, also contribute to early osteochondrogenic differentiation in atherosclerotic vessels [61]. The master transcription factors, including Runx2/Cbfa1, Msx2, and Osterix, designate cells for osteoblast lineages through the induction of downstream genes such as alkaline phosphatase, osteopon‐ tin, and osteocalcin. Here we summarize miRNAs involved in SMC differentiation, as well as in osteogenesis, with targets involved in cardiovascular calcification.

The SMC phenotype is dependent on the miR-143/145 cluster [62-64]. Circulating miR-145 levels are reduced in CAD patients [14]. miR-145 is one of the most recognized arterial miRNAs [65]. Inhibition of miR-143/145 promotes a phenotypic switch to the synthetic, pro-atherogenic SMC state [62], including the inhibition of SMC marker-like alpha-smooth muscle actin and smooth muscle myosin heavy chain [66] — both diminished in osteogenic SMCs [67]. miR-145 modulates SMC differentiation by targeting Krüppel-like factor 4 (KLF4) [63]. KLF4 mediates high phosphate-induced conversion of SMCs into osteogenic cells [68]. Conversely, miR-145 deficient mice [69] and overexpression of miR-145 [66] both reduce neointima formation in vascular injury.

Similar to miR-145, miR-133 has a potent inhibitory role on the vascular SMC phenotypic switch [70]. Runx2, a cell-fate determining osteoblastic transcription factor, is a target of miR-133 [71]. Runx2 acts as a critical regulator of osteogenic lineage and a modulator of bonerelated genes [72]. Runx2 is essential and sufficient for regulating osteogenesis in SMC and VIC [73, 74, 75, 76]. Discovered in the bone biology field, a program of miRNAs controls Runx2 expression to prevent skeletal disorders [77]. Three of these miRNAs (miR-133a, miR-135a, and miR-218) are altered in cardiovascular diseases associated with vascular calcification [14, 17, 20, 28]. Klotho mutant mice, which display vascular calcification due to hyperphosphate‐ mia and through a Runx2-dependent mechanism [78], show overexpression of miR-135a (together with miR-762, miR-714, and miR-712) in the aortic media, which causes SMC calcification by disruption of Ca2+ transporters and increasing intracellular Ca2+ concentrations [79]. More recently, miR-204, another candidate of the Runx2-cluster, was found to contribute to SMC calcification *in vitro* and *in vivo* [80]. Downstream targets of Runx2 are bone-specific genes like osteopontin, osterix and osteocalcin, all present in the cardiovascular osteogenic cell phenotype [2, 81]. We recently demonstrated that miR-125b, which inhibits osteoblast differentiation [82] regulates the transition of SMCs into osteoblast-like cells partially by targeting the transcription factor osterix, providing the first miR-dependent mechanism in the progression of vascular calcification [83]. Additionally, miRNA-processing enzymes essential for SMC function [84] — were reduced in calcified SMCs [83].

calcification by inhibiting the BMP signalling pathway or the renin–angiotensin system,

In summary, a set of circulating miRNAs (consisting of miR-21, miR-27, miR-34a, miR-126, miR-146a, miR-155, and miR-210) is dysregulated in various pro-inflammatory diseases and may represent a miRNA signature for cardiovascular calcification. Of note, systemic and local inflammation paradoxically affects cardiovascular calcification and bone loss, which supports the concept of inflammation-dependent cardiovascular calcification previously proposed by

Cardiovascular calcification is an active, cell-regulated process. Various studies provide evidence of phenotypic transition or transition/dedifferentiation of mature SMCs or VICs into an osteogenic phenotype — a key feature in cardiovascular calcification. In medial calcification, SMCs undergo dedifferentiation from a contractile to a pro-atherogenic synthestic phenotype, lose the expression of their marker genes, acquire osteogenic markers, and deposit a mineral‐ ized bone-like matrix. In valvular calcification, VICs can undergo the transition to osteoblastlike bone-forming cells [58]. Recently, a novel concept emerged of circulating cells harboring osteogenic potential that can home to atherosclerotic lesions and contribute to intimal calcifi‐ cation [59, 60]. Comparing the sources of cells that contribute to atherosclerotic intimal calcification revealed that SMCs are the major contributors that reprogram its lineage towards osteochondrogenesis/blastogenesis; circulating bone marrow-derived cells, however, also contribute to early osteochondrogenic differentiation in atherosclerotic vessels [61]. The master transcription factors, including Runx2/Cbfa1, Msx2, and Osterix, designate cells for osteoblast lineages through the induction of downstream genes such as alkaline phosphatase, osteopon‐ tin, and osteocalcin. Here we summarize miRNAs involved in SMC differentiation, as well as

The SMC phenotype is dependent on the miR-143/145 cluster [62-64]. Circulating miR-145 levels are reduced in CAD patients [14]. miR-145 is one of the most recognized arterial miRNAs [65]. Inhibition of miR-143/145 promotes a phenotypic switch to the synthetic, pro-atherogenic SMC state [62], including the inhibition of SMC marker-like alpha-smooth muscle actin and smooth muscle myosin heavy chain [66] — both diminished in osteogenic SMCs [67]. miR-145 modulates SMC differentiation by targeting Krüppel-like factor 4 (KLF4) [63]. KLF4 mediates high phosphate-induced conversion of SMCs into osteogenic cells [68]. Conversely, miR-145 deficient mice [69] and overexpression of miR-145 [66] both reduce neointima formation in

Similar to miR-145, miR-133 has a potent inhibitory role on the vascular SMC phenotypic switch [70]. Runx2, a cell-fate determining osteoblastic transcription factor, is a target of miR-133 [71]. Runx2 acts as a critical regulator of osteogenic lineage and a modulator of bonerelated genes [72]. Runx2 is essential and sufficient for regulating osteogenesis in SMC and VIC [73, 74, 75, 76]. Discovered in the bone biology field, a program of miRNAs controls Runx2

making it a promising anti-calcification therapeutic target.

**4. miRNA and osteogenesis in the vascular wall**

in osteogenesis, with targets involved in cardiovascular calcification.

our group and others [13, 40, 55-57].

130 Calcific Aortic Valve Disease

vascular injury.

Another potent regulator of vascular and valvular calcification is the BMP signaling pathway (reviewed in detail elsewhere [85]). BMP2 and BMP4 are potent osteogenic differentiation factors detected in calcified valve and atherosclerotic lesions [86-88]. BMPs elicit their effects through activation of receptor complex composed of type I and type II receptors and activate receptor-type–dependent and ligand-dependent Smad transcription factors, which modulate the expression of Runx2 [85]. MiR-26a, miR-135, and miR-155 were previously reported as Smad-regulating miRNAs related to osteoblastogenesis; they functionally repress osteoblast differentiation by targeting Smad1 and Smad5, respectively [54]. miR-155 is one of the circulating miRNAs that is decreased in CAD [14] and CRD [37] (discussed earlier). miR-26a was repressed in aortic valve leaflets of patients with aortic stenosis, and human aortic valvular interstitial cells showed decreased mRNA levels of BMP2 and Smad1 when treated with miR-26a mimic [44]. The same group found lower expression of miR-30b, which targets Smad1 and Smad3. Another group reported deceased miR-141 levels together with increased BMP2 levels in bicuspid versus tricuspid aortic valve leaflets, and showed *in vitro* that miR-141 represses the VIC response to calcification, in part through BMP2-dependent calcification [45]. Itoh *et al*. identified miR-141 as a pre-osteoblast differentiation-related miRNA, which modulated the BMP2-induced pre-osteoblast differentiation by direct translational repression of Dlx5, a transcription factor for osterix [89].

Activation of canonical wingless-type (WNT) signaling is crucial for osteoblast function [90] and for the programming of valvular and vascular cells during cardiovascular calcification [85]. Activation of the Wnt/β-catenin signaling pathway occurs in human calcified aortic valve stenosis [91], in LDL receptor (LDLR)-deficient mice [92, 93], and in osteogenic SMCs *in vitro* [94]. Dickkopf (Dkk)1 is an extracellular antagonist of the canonical Wnt signaling that plays a crucial role in bone remodeling by binding to and inactivating signaling from LDLR-related protein 5/6 [95, 96]. Dkk-1 may also play a role in vascular calcification. In CRD patients, Dkk1 serum levels correlated negatively with arterial stiffness [97]. Dkk-1 prevents warfarininduced activation of β-catenin, and osteogenic transdifferentiation of SMCs [98] and TNF αinduced induction of alkaline phosphatase activity [92]. Remarkably, two miRNAs targeting bone Dkk-1 (miR-335-5p, miR-29a) increase with age [99, 100] — a risk for cardiovascular calcification. miR-335-5p directly targets and represses Wnt inhibitor Dkk-1, thereby enhanc‐ ing Wnt signaling and promoting osteoblast differentiation [101]. To date, no publications exist regarding the role of miR-335-5p in the cardiovascular system. Yet, the age-dependent increase of miR-335 in rat renal tissue inhibited the expression and function of the enzymes implicated in oxidative stress defense [99]. Likewise, miR-29a potentiates osteoblastogenesis by modu‐ lating Wnt signaling. Canonical Wnt signaling induces miR-29a expression, which negatively targets regulators of Wnt signaling, including Dkk-1, sFRP2, Kremen, and osteonectin [102, 103]. miR-29 increased age-dependently in mouse aortic tissue and associated with reduced extracellular matrix components, such as collagen and elastin [100]. Elastolysis accelerates arterial and aortic valve calcification [40]. Furthermore, MMP-2, another target of miR-29 [104], was shown to promote arterial calcification in CRD [105] and valvular calcification [106].

The contribution of osteoclasts to cardiovascular calcification is still controversial [59]. The observation of osteoclast-like cells in calcified atherosclerotic lesions suggested this bonerelated cell is active in the vessel wall. The evidence was strengthened recently by Sun et al., who demonstrated the functional role of SMC-derived Runx2 promoting infiltration of macrophages into the calcified lesion to form osteoclast-like cells — suggesting that the development of vascular calcification is coupled with the formation of osteoclast-like cells, paralleling the bone remodeling process [74]. The receptor activator of the nuclear factor-kappa B (NF-kappa B) ligand (RANKL)/osteoprotegerin (OPG) system controls proper osteoclasto‐ genesis, and actes as a biomarker for CAD [107, 108]. *In silico* analysis revealed RANKL as a target of miR-126 [109], which is decreased in the plasma of CAD [14] and T2D [33] patients. miR-146a, highly expressed in atherosclerotic arteries [16], inhibits osteoclastogenesis [46]. The number of tartrate-resistant acid phosphatase-positive multinucleated cells was significantly reduced by miR-146a in a dose-dependent manner [46]. Furthermore, miR-155, which is decreased in plasma of CRD [37] and CAD [14] patients, was shown to inhibit osteoclast function [110].

Taken together, osteogenic processes in both bone and the cardiovascular system are tightly controlled by miRNAs (Figure 2). Further studies are needed to elucidate whether interplay of miRNAs could explain the bone-vascular axis "calcification paradox," or whether they act independently in the calcifying vessel and bone.

blebbing of the plasma membrane, and can calcify; and apoptotic bodies, 50-5000 nm in

**Figure 2. Potential and established miRNAs contributing to osteogenic regulation of vascular calcification.** Bold, established miRNAs in vascular calcification; underlined, dysregulated in cardiovascular disease (circulating or tis‐

Role of MicroRNAs in Cardiovascular Calcification

http://dx.doi.org/10.5772/55326

133

The majority of miRNAs are independent of vesicles [111] and co-purify with the Ago2 complex [112, 113]. But in CAD patients, most plasma miRNAs associate with extracellular vesicles, and only a small amount are found in extracellular vesicle-free plasma [114]. A celltype-specific miRNA release and different export systems are implicated, as the miRNA release pattern within vesicles is different from that associated with Ago2 complexes [112]. Thus, cells can select miRNA and pre-miRNA for controlled cellular release [115, 116]. miRNA profiles of extracellular vesicles are different from their maternal cell profiles, indicating an active mechanism of selective miRNA packing from cells into vesicles [114]. We have limited knowledge about miRNA secretion. Blockade of sphingomyelinase inhibits exosome genera‐ tion and miRNA secretion, and intercellular miRNA transfer implicates a ceramide-dependent mechanism [117, 118]. Ago2–miRNA complexes may be passively produced by dead cells, released by live cells, or actively transported though cell-membrane–associated channels or

Extracellular vesicles use miRNA to mediate intercellular communication over long distances or on a local tissue level [120]. Endothelial apoptotic bodies can convey miR-126 to athero‐ sclerotic lesions, which demonstrate uniquely paracrine-signaling function for miRNA during

diameter, are released from fragmented apoptotic cells.

sue); gray, predicted miRNA binding sites.

receptors [119].
