Wnt Signaling in Cardiovascular Disease: Opportunities and Challenges

Abstract

Purpose of review

Cardiometabolic diseases increasingly afflict our aging, dysmetabolic citizenry. Complex signals regulating low density lipoprotein receptor-related protein (LRP) and frizzled protein family members – the plasma membrane receptors for the cadre of Wnt polypeptide morphogens – contribute to the control of cardiovascular homeostasis.

Recent findings

Both canonical (β-catenin-dependent) and noncanonical (β-catenin-independent) Wnt signaling programs control vascular smooth muscle cell (VSM) phenotypic modulation in cardiometabolic disease. LRP6 limits VSM proliferation, reduces arteriosclerotic transcriptional reprogramming, and preserves insulin sensitivity while LRP5 restrains foam cell formation. Adipose, skeletal muscle, macrophages and VSM have emerged as important sources of circulating Wnt ligands that are dynamically regulated during the prediabetes-diabetes transition with cardiometabolic consequences. Platelets release Dkk1, an LRP5/LRP6 inhibitor that induces endothelial inflammation and the prosclerotic endothelial-mesenchymal transition. By contrast, inhibitory secreted frizzled-related proteins shape the Wnt signaling milieu to limit myocardial inflammation with ischemia-reperfusion injury. VSM sclerostin, an inhibitor of canonical Wnt signaling in bone, restrains remodeling that predisposes to aneurysm formation, and is down-regulated in aneurysmal vessels by epigenetic methylation.

Summary

Components of the Wnt signaling cascade represent novel targets for pharmacological intervention in cardiometabolic disease. Conversely, strategies targeting the Wnt signaling cascade for other therapeutic purposes will have cardiovascular consequences that must be delineated to establish clinically useful pharmacokinetic – pharmacodynamic relationships.

Introduction

1.0. Wnt signaling and lipoprotein receptor-related proteins: An overview

1.1. Lipoprotein receptor-related receptors, Wnt signaling and molecular genetics of cardiometabolic disease

Lipoprotein receptors facilitate diverse signaling and transport systems well beyond their namesake. Lipoprotein levels and ratios were identified as key predictors of cardiovascular disease as early as the 1950s, but it was unknown how these lipoproteins were regulated to convey cardiometabolic risk [1]. Early clues came from patients with familial hypercholesterolemia (FH), who exhibited early coronary atherosclerosis and elevated serum LDL cholesterol regardless of diet [2]. While normal fibroblasts bound and internalized LDL, shutting off cholesterol synthesis within the cell, fibroblasts isolated from FH homozygotes neither bound nor internalized LDL, and did not reduce cholesterol synthesis [3]. Elucidating the molecular genetics of FH in cardiometabolic disease revealed the fundamentals of receptor-mediated endocytosis [4].

Since the early studies of the LDL receptor (LDLR), many other related receptors that bind lipoproteins have been identified – and once again, their detailed study is yielding fundamental insights[5]. While investigation initially focused upon cholesterol transport, the roles of the LDL receptor-related proteins in embryonic development, Wnt signaling, and adult homeostasis have increasingly occupied centre stage. Low density lipoprotein receptor-related proteins (LRPs) are a group of transmembrane receptors involved in apolipoprotein binding, Wnt signaling, and human diseases [6, 7]. LRP - 5 and - 6 (LRP5 and LRP6) are two such receptors with specific patient mutations leading to metabolic dysregulation, elevated serum LDL and triglycerides (TGs), atherosclerosis, and high or low bone mass phenotypes [8–10]. LRP6 can form a complex with LDLR, facilitating receptor-mediated endocytosis of LDL particles [10, 11], and also directly bind apolipoprotein E-containing lipoproteins. However, LRP5 and LRP6 also heterodimerize with the Frizzled (Fzd) family of G protein-coupled receptors and bind Wnt ligands via 4 cysteine rich extracellular [12] to propagate canonical Wnt signals (see below) [13, 14]. While they share similar patterns of expression, significant sequence homology and substantial functional redundancy in development and adult bone, the Lrp5 and Lrp6 genes are not functionally identical [15–18]. Homozygous deletion of Lrp6 is embryonic lethal, while mice lacking Lrp5 survive through adulthood [19]. Although Wnt ligands provide important morphogenetic signals during embryonic development, more recent work has demonstrated significance during adult homeostasis [20].

Nineteen different Wnt ligands and 10 members of the Fzd family of transmembrane G-protein-coupled receptors have been identified in humans [21]. Wnt - bound Fzd receptors competitively interact with co-receptors such as LRP5, LRP6, ROR2 and receptor-like tyrosine kinase (RYK) – and downstream signals vary dependent upon these Wnt-regulated interactions [18, 22–24]. Wnt ligands have classically been categorized as canonical versus noncanonical based upon beta-catenin activation (see below), though this categorization has limitations (see below). The task of assigning specific functions to any one Wnt, Wnt receptor or co-receptor becomes daunting in the absence of genetic evidence [25]. However, human and murine molecular genetics has revealed a uniquely important role for LRP6 in cardiometabolic disease [26–28]. Wnt signaling exemplifies context dependence in biology, relevant to personalized medicine, with identical inputs eliciting opposing outputs in different cell types and/or metabolic contexts [29]. Combinatorial complexity is necessary for development and homeostasis, but stymies attempts to rigidly assign a Wnt “code” in simple terms with respect to specific ligands. Nevertheless, it is both historically and practically helpful to organize receptor signaling regulated by Wnts into canonical vs. noncanonical programs when considering the cardiometabolic actions of LRP6 and other Wnt modulators (Figure 1). We provide a highly abridged overview to orient the reader.

Figure 1. A highly abridged overview of canonical and noncanonical Wnt signaling.

Frizzled (Fzd) Wnt receptors form ligand-dependent heterodimers with the LDL receptor related proteins LRP5 and LRP6 to convey canonical signals conveyed by nuclear β-catenin. Since GSK3 also inhibits Taz and mTOR, canonical Wnt activation is often associated with upregulation of these pathways. Inhibitory proteins of the sclerostin and Dickkopf (Dkk) families directly bind to LRP5/6. Other Wnt ligand-dependent Fzd receptor complexes signal via pathways that do not require LRPs and β-catenin; G proteins of the Rho/Rac/cdc42 and Gαq families, and intracellular calcium and cytoskeletal signals, utilize transcription factors of the NFAT, USF, and Jun/AP1 to convey these noncanonical signals. Dishevelled (Dvl) platform proteins contribute to both canonical and noncanonical programs. The ROR/Ryk family of receptor tyrosine kinases (TK) can signal within the Fzd cell polarity pathway or via 14-3-3 platform proteins. The secreted frizzled related proteins (SFRPs) bind Wnt ligands as faux (decoy) receptors, and inhibit both canonical and noncanonical signals. Green asterix, phosphorylation increased with Wnt signaling; yellow asterisk, phosphorylation decreased with Wnt signaling.

1.2. Canonical vs. noncanonical Wnt signals

Wnt ligands are polypeptides of approximately 350 residues, and are hydrophobic owing to post-translational serine fatty acid O-acylation with palmitoleic acid. Fatty acylation is essential for secretion, frizzled co-receptor binding and signaling [30–33]. Wnt ligands generate diverse signals in target cells, broadly categorized into canonical and noncanonical signaling modes. Canonical signaling denotes Wnt activation of the transcriptional co-activator, β-catenin, ultimately leading to the upregulation of target genes by the family of TCF/LEF (T cell factor/lymphoid enhancer factor) transcription factors [34]. However, β-catenin positively and negatively regulates a broad range of nuclear transcriptional responses. Under basal conditions, the kinase GSK3 phosphorylates β-catenin and fosters binding to the proteins Axin and APC, thereby assembling a β-catenin degradation complex [35, 36]. This state retains suppression of target genes by the LEF/TCF family of transcription factors and the transcriptional repressor, Groucho [21]. As canonical programs are activated, co-receptors LRP5/LRP6 and Fzd [37] recruit and polymerize Dishevelled signaling platform proteins on the Fzd receptor. This complex then directs GSK3 inhibition via phosphorylation and sequestration[38] which prevents β-catenin degradation [39, 40]. GSK3 sequestration liberates β-catenin, allowing its translocation to the nucleus (Figure 1). Nuclear β-catenin displaces Groucho repressors from target promoters, activating canonical genomic targets [37]. Since GSK3 also inhibits mTOR (mammalian target of rapamycin) activation [41] and TAZ (transcriptional co-activator with PDZ-binding motif) [42], activation of canonical Wnt signaling also upregulates these signaling pathways (Figure 1).

Noncanonical signaling encompasses the impressively broad range of β-catenin-independent responses elicited by Wnt ligands [43]. This category includes the planar cell polarity (PCP) pathway controlling cell orientation and cytoskeletal function, the Wnt/calcium pathway controlling calcium release from the endoplasmic reticulum, the Jun N-terminal kinase and mTOR signaling relays, and a few instances in which Wnt ligands have been shown to affect cells independent of β-catenin by other means [44–46] including the TAZ pathway mentioned above [47]. There is significant overlap and regulatory cross-talk between Wnt these signaling pathways [48].

The Wnt/calcium pathway is particularly relevant to adult metabolic and cardiovascular health [43, 49, 50], since calcium-regulated transcription factors control vascular smooth muscle (VSM) phenotypic modulation[51]. In this pathway, Wnt ligands engage Fzd co-receptor complexes without LRP5 or LRP6, leading to small G-protein activation, plasma membrane phospholipase-mediated liberation of inositol-1,4,5-trisphosphate, and calcium release from the endoplasmic reticulum [52–54]. Increased cytosolic calcium levels stimulate Ca2+/calmodulin dependent enzymes, including calmodulin-dependent protein kinase II (CamKII) and calcineurin [55]. The Ser/Thr phosphatase calcineurin dephosphorylates the cytoplasmic pool of nuclear factor associated with T-cells (NFAT), a family of transcription regulators controlling inflammation and immunity [56] and VSM gene expression [51]. De-phosphorylation enables nuclear import of NFAT and activation of its target genes [57]. Of note, calcium-dependent CamKII activation triggers a separate cascade that phosphorylates and inactivates TCF/LEF transcription [58]. Thus, CamKII reciprocally inhibits canonical β-catenin-LEF/TCF activation with the upregulation of noncanonical Wnt/calcium signals [59].

The ROR2/ROR1/Ryk family of receptor tyrosine kinases also bind Wnt ligands to form heterodimers with Fzd family members in the planar cell polarity pathway [48, 60], or alternatively homodimerize to elicit G-protein mediated cell migration [61] and other responses via tyrosine phosphorylation of 14-3-3 platforms[62] (Figure 1). However, Aaronson and colleagues highlighted that LRP5/6 and ROR/Ryk family members fundamentally compete for common Fzd co-receptors to support canonical vs. noncanonical signals, respectively [48]. Thus, LRP5 and LRP6 fine-tune the relative activation of canonical and noncanonical Wnt signals in a cell-autonomous fashion (Figure 2; and see below).

Figure 2. LRP6 finely tunes the balance of signals via canonical and noncanonical pathways in cardiovascular disease.

Mani and colleagues first identified hypomorphic LRP6 alleles exert dominant effects that convey precocious atherosclerosis with metabolic syndrome. Knockin LRP6 (R611C) mice and conditional knockout of LRP6 in VSM demonstrate that LRP6 stabilizes the VSM phenotype by inhibiting noncanonical Wnt signals. Fzd8 and Fzd9/Fzd10 appear to mediate osteofibrogenic signals. Canonical LRP6 signals also enhance insulin receptor expression and signaling in peripheral tissues and suppress hepatic lipoprotein biogenesis. See text Section 2 for discussion and details.

2.0. Wnt Signaling in Cardiometabolic Disease

2.1. Wnt signals as regulators of the dysmetabolic milieu driving cardiovascular disease

Given the evolutionary relationships between the LRPs and the LDL receptor, it is not surprising that the LRP/Wnt signaling cascade would play a role in metabolic homeostasis. In 2007, Mani and colleagues reported on a private mutation (rare and present in one family), LRP6 (R611C), that caused precocious coronary artery disease with metabolic syndrome and osteoporosis in a family of Iranian ancestry [28]. While this autosomal dominant variant does impair LDL cholesterol clearance, it also hampers canonical Wnt signaling required for TCF7L2-dependent transcriptional support of insulin receptor expression in peripheral tissues [63]. As a proof-of principle, Mani’s group went on to show that augmenting canonical Wnt signaling programs in LRP6 (611C/611C) homozygous mutant mice with recombinant Wnt3a injections reversed the combined dyslipidemia [64]. These studies revealed that hepatic de novo lipogenesis and apoB-containing lipoprotein secretion is held in check by canonical Wnt signals via LRP6 [64]. Since the insulin resistance and dyslipidemia of metabolic syndrome are clear contributors to cardiovascular disease risk [65], LRP6-dependent Wnt signaling tone globally mitigates cardiometabolic risk [28].

Likewise, LRP5 also plays a role in cardiometabolic risk. Badimon’s group demonstrated that LRP5 inhibits aortic macrophage infiltration and inflammatory cytokine production in mice fed diets that induce hypercholesterolemia [66]. The mechanism is not completely understood, but may relate to LRP5-dependent promotion of an anti-inflammatory macrophage phenotype [67].

Recently, Kozinski et al showed that high fat diabetogenic diets alter the ratio of circulating Wnt3a and Wnt4; specifically, upon progression to frank diabetes, Wnt3a levels fall while Wnt4 levels rise[68], eliciting a program that results in dysfunctional pancreatic islet cell function. While sources of circulating Wnts have yet to be robustly established in vivo, adipose and skeletal muscle[68] and VSM[27] can all produce Wnt4. In addition to inhibiting Wnt3a-induced pancreatic beta-cell insulin secretion[68], Wnt4 stimulates VSM proliferation and promote arterial intimal thickening[69] as relevant to the cardiovascular disease of diabesity. Importantly, in very recent data, George and colleagues have identified matricellular proteins of the CCN family (Cysteine-rich protein 61/Connective tissue growth factor/Nephroblastoma overexpressed gene) as down-stream mediators of Wnt-initiated VSM migration during neointima formation[70, 71].

2.2. Wnt signals as mediators of cardiovascular disease in response to the dysmetabolic milieu

LRP6 is also highly expressed in VSM, and recent studies have focused upon Wnt signaling in this cell type in cardiovascular disease[27, 72]. In addition to providing structure, ductility, and contractility, VSMs participate in local signal relays via endocrine/paracrine cues. VSM has prodigious capacity for phenotypic modulation[73]. Early descriptions focused upon contractile and synthetic phenotypes, the latter characterized by myofibroblast-like, high-level fibrillar collagen production. Subsequent studies established capacity for VSM osteochondral trans-differentiation that contributes to vascular calcification [73], and the ability to adopt macrophage-like foam cell phenotypes with cholesterol loading[74, 75]. More recently, Majesky et al showed that with de-differentiation VSM can generate adventitial Sca1+ vascular progenitors – a population that also has osteochondrogenic potential [76] – and the Yamanaka factor Klf4 is one important factor for this phenotypic plasticity [77]. Thus, inhibiting VSM plasticity with stabilization of the mature contractile phenotype holds great promise for mitigating arteriosclerosis of conduit vessels in response to dysmetabolic states.

VSM phenotypic modulation normally follows mechanical vessel injury and inflammation [78–80]. As a normal function of wound repair and innate immunity, Klf4 and NFAT are activated to promote phenotypic plasticity[81–83], losing features of contractile VSM and enabling trans-differentiation to resemble other cell types in the mesenchymal lineage [84, 85]. NFAT family members are Rel-domain transcription factors that dimerize with leucine zipper proteins and other Rel-domain proteins to control gene expression[86]. NFATs are critical components of noncanonical Wnt/calcium signals that regulate stem cell phenotypes [87] and VSM differentiation[51]. While the specific members of the NFAT family that mediate phenotypic switching in VSM have yet to be determined[51], NFATc1[88–91], NFATc3 and NFATc4[92] are likely to be important.

In studies of LRP6-deficient VSM, our laboratory identified that noncanonical Wnt signaling programs involving NFAT and USF1 were upregulated, activating arteriosclerotic osteogenic programs in vivo and in vitro [27]. This VSM-to-osteogenic phenotypic modulation was inhibited by a chemical antagonist that blocked cdc42/rac1 G-protein signaling downstream of Fzd10 but upstream of USF1 and NFAT activation by protein arginine methylation relays[27]. Mani’s group simultaneously identified that LRP6 (R611C) was hypomorphic for inhibition of VSM plasticity, arising in part due to enhanced Sp1 transcriptional activation of platelet-derived growth factor signaling[26]. Since (a) β-catenin/TCF7L2 interactions sustain the native, contractile VSM phenotype[26] including transgelin/SM22 expression[93]; and (b) noncanonical Wnt signaling inhibits beta-catenin actions[94], the data from our lab[27] converges with that of Mani’s [26] to establish the cell-autonomous role for LRP6 in maintaining the mature VSM phenotype via restraint of noncanonical programs (Figure 2). Of note, similar regulatory circuits likely function during development [97]. Thus, augmenting VSM noncanonical Wnt signals results in VSM plasticity, de-differentiation and an arteriosclerotic tissue phenotype [27] – thereby increasing myocardial workload, impairing Windkessel physiology necessary for smooth distal tissue perfusion and causing cardiovascular dysfunction[98, 99].

A number of inhibitors of Wnt signaling exist that function to shape cardiovascular health and disease. Secreted frizzled-related proteins (SFRPs) contain a cysteine-rich domain homologous to the Wnt binding sites of the membrane-bound Fzd G-protein coupled receptors [100]. Thus, SFRPs can limit Wnt signaling by binding and sequestering Wnt ligands as faux receptors, and inactivation of specific SFRP genes can enable constitutive Wnt actions [101–103]. This family of Wnt antagonists reduces endothelial and VSM cell proliferation in vitro and in vivo [104], with therapeutically relevant cardiovascular consequences. Overexpression of SFRP1 reduces myocardial infarction size in mice and improves cardiac function [105]. Similarly, SFRP5 inhibits myocardial inflammation and injury in a preclinical ischemia/reperfusion model[106]. Importantly, while SFRP family members can inhibit both canonical and noncanonical signaling, it appears that SFRP5 preferentially inhibits noncanonical signaling in proinflammatory cellular contexts [64, 107]. Of note, however, SFRP2 exhibits a more nuanced cardiovascular response, promoting myocardial stem cell survival and repair with ischemia[108], likely by shaping canonical [109] and planar cell polarity[110] signals that mitigate fibrosis[109, 111]. Additionally, during development, SFRP2 can bias noncanonical Wnt signaling via ROR2 in lieu of Fzd7 by stabilizing Wnt5a-ROR2 complexes[112]. By contrast, SFRP1 appears to be a pure antagonist[113]. A better understanding of how the SFRPs differentially shape the balance, duration, and extent of Wnt ligand is needed.

Two other types of Wnt signaling inhibitors, Dickkopf and sclerostin, antagonize function by binding to the LRP co-receptors. During development, Dickkopf and sclerostin family members play roles overlapping yet quite distinct from the SFRPs[114, 115]. The vertebrate Dickkopf (Dkk) proteins Dkk1, Dkk2, and Dkk4 are LRP5/LRP6 ligands that antagonize Wnt binding and activation of canonical programs[116]. Dkk1, Dkk2, Dkk3, and Dkk4 are canonical antagonists; however, Dkk3 can also indirectly activate canonical programs[117] and Dkk4 can indirectly activate noncanonical c-Jun signaling[118]. The precise molecular mechanisms whereby Dkk3 and Dkk4 elicits these latter surprising actions are as yet unknown, but may involve Dkk actions through their Kremen receptors that regulate LRP5/6 receptor trafficking[119]. Dkk1 is readily measurable in the circulation, is released by activated platelets [120] and induces secretion of inflammatory cytokines by adjacent endothelium[121]. As such, Dkk1 can promote a prosclerotic endothelial-mesenchymal transition[122]. Circulating levels of Dkk1 are elevated patients afflicted with acute ischemic stroke[123] and symptomatic aortic stenosis[124], and may ultimately prove to be a clinically useful biomarker.

Sclerostin is another canonical Wnt signaling inhibitor that binds LRP4, LRP5, and LRP6[125]. Production by osteocytes in bone participates in a negative feedback loop – a servo mechanism that restrains excessive canonical Wnt responses in the skeleton [125]. Unlike the Dkks, however, sclerostin does not appear to compete for Wnt ligand binding. Intriguingly, sclerostin is also expressed in aortic VSM[126], is upregulated during arteriosclerotic calcification[126], and limits vascular remodeling that predispose to aneurysm formation[127]. Epigenomic methylation of the sclerostin gene down-regulates its expression in those VSM residing within areas of human aortic aneurysms, and mice transgenic for sclerostin are resistant to angiotensin-induced aneurysm[127]. These latter data suggest local regulation of sclerostin actions with arterial remodeling, and are particularly important to consider since anti-sclerostin antibodies with prolonged pharmacokinetics are in development to treat osteoporosis[128, 129].

Oxidized LDL (oxLDL) upregulates Wnt ligands in multiple cell types including cells of the monocyte/macrophage lineage[130]. Recent data suggests that upregulation of Abca1 by Wnt5a in RAW264.7 myeloid cells following oxLDL treatment reduces lipid accumulation via enhanced reverse cholesterol transport[131]. Similarly, Boucher, Herz, and colleagues have shown that Wnt5a upregulates Abcg1 and inhibits 3-hydroxy-3-methyl-glutaryl-coenzyme A reductase and synthetase in mouse embryonic fibroblasts, thereby limiting the intracellular accumulation of cholesterol[132]. Boucher has hypothesized that while Wnt5a induction may initially provide an adaptive mechanism to limit intracellular cholesterol accumulation downstream of LRP1, with time the increased Wnt5a tone may promote cardiovascular calcification[132, 133]. This intriguing notion adds to accumulating data indicating that atheroma formation (atherosis) and arterial sclerosis (fibrosis, calcification, arterial stiffness) must be independently assessed to fully capture the impact of Wnt signaling and its modulation in atherosclerotic disease[134–136].

3.0 The cardiometabolic opportunities and challenges for pharmacotherapies targeting the Wnt signaling cascade

As is evident, human and murine molecular genetics converge to indicate the important role for Wnt/LRP signaling in cardiometabolic health [26–28]. Wnt signaling plays a major role in the progression of heart disease, in terms of both metabolic alterations (insulin sensitivity) and cardiovascular remodeling and structural changes (fibrosis, sclerosis, atheroma formation, smooth muscle cell proliferation, hypertrophy) [137]. While there is great interest in identifying small-molecule modulators of Wnt signaling to treat multiple diseases, no compound has been identified with sufficient efficacy and specificity for use in humans [138–140]. However, a select few molecules have shown promise by targeting Wnt secretion or turnover of the β-catenin destruction complex. For example, GNF-6231 is an inhibitor of porcupine, the endoplasmic reticulum protein that is required for Wnt palmitoylation. GNF-6231 inhibits the secretion of both canonical and noncanonical Wnt ligands, and transient porcupine inhibitioin limits pro-fibrotic myocardial injury and enhances recovery in preclinical models of myocardial infarction[141]. Similarly, antagonists of tankyrase – a poly ADP-ribose polymerase that reduces the canonical pathway inhibitor Axin (Figure 1) – increase Axin levels, reduce canonical Wnt signaling and mechanical injury-induced neointima formation[142]. These data indicate that some novel small molecules in development for cancer therapies might be repurposed for certain cardiovascular diseases.

However, it becomes apparent that some of the most promising Wnt pathway modulators may biologicals – viz., recombinant proteins or neutralizing antibodies [143]. The SFRPs or engineered mimetics are particularly attractive, since they possess intrinsic capacities to productively shape canonical and noncanonical programs in response to complex ligand milieus[106, 110]. In certain settings, e.g. dilated cardiomyopathy [117], specific ligands such as Dkk3 with novel signaling profiles could hold therapeutic promise as well. However, the extent to which Dkk actions depend solely upon modulation of Wnt/LRP signaling versus actions via its Kremen receptors [116] has yet to be determined. Because Wnt ligands can heterodimerize to create unique biological responses[144], any therapies implementing recombinant Wnt proteins will ultimately have to consider this complexity – along with any untoward impact on occult malignancy. The extracellular domains of LRPs themselves are also biologically active, and enzymatic cleavage of LRP6 extracellular domain yields a constitutively active intracellular canonical signal [145, 146]. The released intracellular domain is constitutively active independent of Fzd co-receptors, binding to GSK3 in the destruction complex and relieving β-catenin inhibition [145]. On the cell surface, LRP6 extracellular domain inhibits noncanonical Wnt signaling [97], potentially forming complexes with Fzd proteins that preclude association with noncanonical coreceptors like ROR2. The soluble LRP extracellular domain also binds DKK1 and potentially mitigates its actions [147].

Antibodies directed against specific extracellular components of the Wnt regulatory cascade are in clinical development. Inhibitors of both Dkk1 and sclerostin have been developed to promote fracture repair and treat osteoporosis, respectively [148]. However, the prolonged pharmacokinetics afforded by inhibitory antibodies may exert unexpected and untoward responses given the dynamic, homeostatic interactions between Wnt activators and inhibitors. For example, prolonged exposure to Dkk1 stabilizes LRP6 protein accumulation until Kremen2 engages Dkk1[149], and continuous Dkk1 exposure can result in “rebound” canonical signals in vitro [150]. Conversely, disease-dependent anatomical differences in VSM sclerostin epigenetic silencing may significantly shape the arterial dose-response relationship to antibody-mediated sclerostin inhibition [127, 136]. The extent to which this occurs in vivo has yet to be established, but points to how either sustained activation or inhibition of Wnt signaling programs may elicit time-dependent yet mechanism-based “toxicities” due to poorly established pharmacokinetic-pharmacodynamic relationships. Thus, the cardiovascular consequences must be studied in detail for therapeutic approaches modulating Wnt signaling as pharmacotherapy for any indication – cancer, skeletal health, regenerative medicine, etc. – in addition to strategies targeting these pathways for cardiometabolic benefit.

4.0 Conclusion

Human and murine molecular genetics clearly identify the contributions of Wnt/LRP signaling to cardiometabolic health and homeostasis and cardiovascular disease. Successful modulation of Wnt pathways for any therapeutic indication will require a better understanding of how ligand-receptor complexes regulate context-dependent intracellular relays – including mediators other than β-catenin – within the cardiovascular system.

Summary.

• Dysregulated Wnt signaling is a particularly damaging feature of cardiovascular inflammation, altering cellular plasticity, intracellular cholesterol accumulation, and osteofibrotic responses to metabolic or mechanical injury.

• Human and murine molecular genetics have firmly established the important role of the Wnt co-receptor LRP6 in the biology of cardiometabolic health and disease.

• Several components of the overall Wnt signaling cascade are attractive targets for therapeutic intervention, achieved either by small molecule inhibitors or by biologicals that mimic or modulate components of the extracellular regulatory machinery.

• However, like most endocrine systems, Wnt signaling can be either deleterious or beneficial; this is dependent upon cell type, metabolic context, and stage-specific contributions of canonical vs. noncanonical relays to disease biology.

• Thus, the pharmacokinetic-pharmacodynamic responses of any Wnt pathway therapeutic – including duration and intensity of modulation – must be carefully determined to optimize clinical benefits while mitigating mechanism-based deleterious responses.