A Not-so-Little Role for Lp(a) in the Development of Calcific Aortic Valve Disease

Alterations in lipid metabolism and inflammatory processes are well established as potential risk factors in the development and progression of cardiovascular disease.¹ However, with complications ranging from valve dysfunction to arrhythmia to myocardial infarction and stroke, the underlying mechanisms may be as varied as cardiovascular disease itself. On the other hand, the reoccurrence of common molecular and cellular pathways identified in the collective body of cardiovascular research could suggest shared initiators or mechanistic nodes between seemingly divergent processes, including lipid metabolism and inflammation. One area where this may hold true is cardiovascular calcification, in which dysregulated mineral metabolism in cardiovascular tissues leads to increased morbidity and mortality.

Calcification of soft tissues results from the deposition of calcium, largely in the form of hydroxyapatite in the vascular wall and/or valve leaflets. Previously thought to be a passive degenerative process, cardiovascular calcification has become increasingly apparent to be an active process initiated by many triggers. Recent studies have demonstrated variation in the gene LPA, which determines the plasma concentration of lipoprotein(a) (Lp(a); pronounced “L P little a”) to be associated with calcific aortic valve disease (CAVD).^2,3 Lp(a) consists of a LDL-like particle in which apolipoprotein(a) is covalently bound to apolipoprotein B. Additionally, Lp(a) is a genetic risk factor for atherosclerotic events.⁴ As in atherosclerosis, calcifications in CAVD localize to areas with lipoprotein accumulation and inflammatory cell infiltration, suggesting a shared disease process.⁵ However, some noticeable differences do exist, including increased mechanical stresses and calcification involved valve obstruction in CAVD as opposed to microcalcifications leading atherosclerosis plaque rupture.⁶

In the current issue of Circulation, Bouchareb and Mahmut et al.⁷ propose a highly plausible mechanistic pathway through which Lp(a) and valve interstitial cell (VICs)-derived autotaxin (ATX) may induce valve calcification by regulating inflammation induced bone morphogenetic protein (BMP). This study connects lipid metabolism to inflammation and valve calcification, and in doing so identifies a pathway that may help lead to the development of CAVD therapeutics, an area with high unmet clinical need. Mathieu and colleagues had recently reported⁸ that lipoprotein-associated phospholipase A2 (Lp-PLA2), an enzyme that utilizes oxidized phospholipids carried by Lp(a) to generate lysophosphatidylcholine (LPC), is both highly expressed in CAVD and to plays a role in the mineralization of VICs. The CAVD functional role of ATX, a key enzyme involved in the conversion of LPC to the signaling phospholipid, lysophosphatidic acid⁹ (LPA) has yet to be reported. ATX is a member of the ectophosphodiesterase/nucleotide phosphohydrolase (ENPP) family. It is notable that to varying extents ENPPs can hydrolyze ATP to generate pyrophosphate,^{10, 11} a known inhibitor of the bone and vascular smooth muscle calcification. However, in vitro analysis of ENPP substrate hydrolysis suggest that ATX is a poor nucleotide pyrophosphatase/phosphodiesterase, and unique among the ENPPs in acting as a phospholipase.¹¹ As such, ATX phospholipase activity converting LPC to LPA, particularly in the context of elevated Lp(a), may play a greater role in CAVD development.

LPA is a potent extracellular signaling molecule with a diverse array of physiologic and pathologic actions including: induction of the mitogenic RAS-extracellular signal-regulated kinase pathway, the phosphoinositide 3-kinase (PI3K)-AKT cell survival pathway, Rho- and RAC-mediated cytoskeletal remodeling and cell migration, cell proliferation, vascular and neural development, phospholipase C activation leading to calcium mobilization, fibrosis, lymphocyte homing, and cytokine production.¹² ATX acts locally, and signals through LPA generation and six LPA guanine-nucleotide-binding protein-coupled receptors (LPAR1-6), located on the surface of a wide variety of cells. Lp(a) can bind to a number of receptors including LDLR, LRPs, VLDLR, and SR-BI, although the extent to which it acts as a ligand can vary widely.¹³ Given the role of these and other cell surface receptors in cellular metabolism and trafficking, examination of the involvement of intracellular sorting processes including those affecting membrane composition and events such as endocytosis and exocytosis, may provide novel CAVD insight and should be examined in future studies.

In the current study by Bouchareb and Mahmut et al.,⁷ increased LPA, ATX lysophospholipase activity and protein abundance were found in mineralized human aortic valves. Increased ATX levels in valves were associated with oxidized lipids, increased remodeling score, and measurements of inflammation. As pointed out by the authors, one limitation of this study was that ATX was examined in human aortic valves with advanced end-stage pathology. As such a causative role of Lp(a) and VIC-derived ATX in the valve calcification process could not be clearly assigned from the human tissue studies alone. The authors hypothesized that LPC may induce valve calcification through NF-κB activation leading to IL-6 production and BMP2 signaling in VICs. To test whether NF-κB activation, IL-6, and BMP2 were involved in LPC mediated mineralization, Bouchareb and Mahmut et al.⁷ treated VICs with LPC, mineralizing medium (a combination of inorganic phosphate, insulin, and ascorbic acid), and inhibitors of BMP, NF-κB, LPARs, or silencing RNAs directed against ATX or IL-6. Disrupting any of these components strongly reduced/inhibited LPC enhanced mineralization. The human tissue and cell culture data were partially confirmed in this study with the use of an animal model of CAVD, LDLR^−/−/ApoB^100/100/IGFII transgenic mice, in which ATX was found to be increased in the aortic leaflets. Additionally, lysophosphatidic acid-treated mice showed a 1.7-fold increase in aortic valve leaflet calcification, along with an increase in BMP2. These results indicate that VICs and Lp(a)-derived ATX and oxidized lipids lead to increased LPA that acts to induce inflammation via LPARs, NF-κB activation, and ultimately results in IL-6 regulation of BMP mediated valve calcification.

Approximately 50,000 valve replacements are performed annually in the United States for patients with severe aortic stenosis.⁵ Aside from valve replacement, there are currently no treatments that prevent or slow the progression of valve disease, which is responsible for more than 22,000 deaths each year in the United States.¹⁴ Bouchareb and Mahmut et al.⁷ correctly concluded that inhibition of ATX or blocking LPARs as potential novel CAVD therapies warrant future investigation. Some caution should be taken in this approach, as bone defects have been reported in LPA receptor modified mice.¹⁵ Similarly, assessment of targeting Lp(a) itself also seems warranted, although whether lowering Lp(a) levels can reduce the rate of incidence or progression of aortic valve disease remains to be determined. However, given that a common variant in Lp(a) was reported to increase the risk of developing aortic stenosis by more than 50%,² Lp(a) targeting therapies should be explored further in a CAVD context. Of potential interest in this area is the recent development of PCSK9 based therapeutics, which significantly reduce major adverse cardiovascular events (death from coronary heart disease, nonfatal myocardial infarction, fatal or nonfatal ischemic stroke, and unstable angina requiring hospitalization),¹⁶ in addition to reducing Lp(a) levels.¹⁷ Further analysis of whether reduction in Lp(a) levels via PCSK9 inhibition plays a role in reduced cardiovascular events;, particularly in relation to CAVD, may prove to be of importance (Figure 1). Outside of PCSK9 inhibition, niacin and the cholesteryl ester transfer protein inhibitor, anacetrapib, have been shown to reduce Lp(a) levels.¹⁸ Although whether these or similar compounds would act therapeutically in CAVD, or more specifically a CAVD at-risk subpopulation with elevated Lp(a), is unknown. HMG-CoA reductase inhibitors (statins), are known to act on both lipid metabolism and inflammation, but have largely shown a lack of therapeutic benefit in CAVD.¹⁹ However, it is worth pointing out that HMG-CoA reductase inhibitors have been reported to lower LDL cholesterol without reducing Lp(a),²⁰ and while there are some conflicting reports showing both elevated and decreased Lp(a), several statin studies show no major changes. This result may not be too unexpected given that Lp(a) is reportedly a relatively poor ligand for LDLR,¹³ a receptor that serves as one of the major means of action on lipid metabolism following statin administration.

Figure 1.

Potential role of Lp(a) and PCSK9 in CAVD. Lp(a) may get taken up and metabolized by cells via Lp(a) receptors (receptors for which Lp(a) is a ligand). However, in the presence of PCSK9 some of these receptors may be internalized and degraded instead of recycled back to the cell surface (A). As such more Lp(a) derived oxidized lipids may be converted to LPA and be taken up by VICs via LPA receptors. This may lead to the production of inflammation related cytokines (e.g., IL6) through NF-κB nuclear localization that in turn increase BMP and/or other calcification inducing processes leading to valve calcification. PCSK9 inhibitors may act to block PCSK9 interaction with Lp(a) receptors, resulting in Lp(a) receptors being recycled back to the cell surface where they can take up more Lp(a) (B). Dashed arrows indicate multiple steps in the pathway.

In summary, the work of Bouchareb and Mahmut et al.⁷ builds on previous studies to identify a mechanistic pathway through which Lp(a) and ATX may be driving aortic valve calcification, and in doing such presents an important area of research worthy of additional investigation.