Retinoids – dissolving the calcification paradox

Calcification in the human body is ideally reserved for bones and teeth, however pathologic calcification can arise as an adaptive response to insults such as inflammation, injury, and trauma¹. Ectopic calcification can form in nearly all soft tissues throughout the body^2–6, however the primary manifestation occurs in cardiovascular tissues. Calcification can be found in the microvessels of the skin, as in calciphylaxis⁷, in the necrotic core of atherosclerotic plaques⁸, in the smooth muscle cell-rich medial layer of vessels along damaged elastic lamina^{9, 10}, as well as in the leaflets of the heart valves¹¹. Coronary artery calcium score is now the metric most commonly used to assess the risk for major adverse cardiovascular events in both symptomatic and non-symptomatic patients¹².

Cardiovascular calcification was originally assumed to be a passive degenerative consequence of aging until evidence of true ossification was observed in atherosclerotic plaques by Demer and Bostrom¹³. Since that seminal discovery many mechanisms involving active bone remodeling processes and cellular phenotypic switching have been shown to contribute to cardiovascular calcification^{14, 15}. Further elucidating this dynamic process has been the identification of mutations causing rare genetic diseases that present with ectopic calcification in the vessels and valves^16–21. While these diseases have identified novel pathways driving calcification, they have also identified the limitations of using mouse models to study these phenotypes, as mice do not fully recapitulate the human disease states^22–25.

Vitamin A (also called retinol) is an essential dietary component that can be found in the circulation or stored in the liver and adipose tissue in the form of retinyl esters²⁶. Retinoic acids (the most prevalent being all-trans retinoid acid, ATRA) are the active metabolites derived from retinol and retinyl esters and are the ligand responsible for the activation of the nuclear receptor retinoic acid receptors (RAR). Once active, RAR can function as transcriptional activators or repressors, depending on the co-factors bound, and RAR are essential for proper mammalian skeletal and limb development^{27, 28}. RAR signaling in ectopic calcification is less clear. Using mice as the model systems, one study found ATRA prevented heterotopic calcification²⁹, while another identified that it promoted valve calcification³⁰.

Considering these divergent findings for the role of ATRA in murine vascular calcification models, and the known limitations of these models to fully recapitulate human vascular calcification phenotypes, in the current issue of Arteriosclerosis, Thrombosis, and Vascular Biology, Rogers and Chen et al investigated the role of ATRA in vascular calcification utilizing human tissues and primary cells³¹. They identified that supplementation with ATRA, the retinoic acid isomer 9-cis-RA, as well as an acyclic synthetic retinoid, peretinoin (generic name, NIK-333 Kowa Company, Ltd), reduced calcification and osteogenic differentiation of primary human coronary artery smooth muscle cells (CASMCs) and valve interstitial cells (VICs), while siRNA-mediated knockdown or treatment with the RAR inhibitor, AGN 193109, promoted mineralization.

The enzyme tissue non-specific alkaline phosphatase (TNAP) generates inorganic phosphate which is a critical building block in the formation of ectopic calcification. This group previously found that a key step in the nucleation and propagation of calcification in the extracellular matrix is the accumulation of extracellular vesicles that have been loaded with TNAP protein³². In the present study they showed that in the presence of ATRA, TNAP transcription was reduced as was TNAP activity on extracellular vesicles. Regarding osteogenic transition of vascular cells, expression of osteogenic master regulator RUNX2 was decreased when ATRA or peretinoin was administered during in vitro osteogenic experiments, while transcription of the endogenous calcification inhibitor, MGP, was increased, illustrating that RAR signaling inhibits osteogenic transcriptional programs in CASMCs and VICs. Maintaining proper homeostasis between the bone and the vasculature is critical, as elegantly presented by Thompson and Towler³³; thus to leverage their findings toward the development of therapies, Rogers and Chen el al compared the effects of ATRA and peretinoin in the calcification of osteoblasts derived from human femur bone. They found that while the endogenous compound ATRA reduced calcification of these bone-forming cells, the synthetic acyclic retinoid peretinoin did not. To date, cyclic retinoids ATRA and 9-cis retinoic acid have not been readily utilized for therapy in part due to their ability to increase levels of the lipoprotein lipase inhibitor apolipoprotein CIII and apolipoprotein A1, but repress expression of some metabolic proteins, such as some cytochrome P450 family members. Importantly, while these effects of cyclic retinoids were confirmed in this study in human liver HepG2 cells, peretinoin did not induce similar effects.

Together this study not only illustrates the anti-calcific effects of retinoic acid signaling in human cardiovascular cells, it further identifies that the synthetic acyclic retinoid, peretinoin, can specifically block calcification in a vascular cell, while sparing osteoblasts of this anti-calcific effect (see Figure). Currently, there is no treatment available to prevent or reverse vascular calcification. As peretinoin has been found to be safe for use in humans, perhaps future studies will be designed to test its effects on vascular calcification in humans^{34, 35}.

Both all-trans retinoic acid (ATRA) and the acyclic synthetic retinoid peretinoin inhibited calcification by decreasing expression of pro-osteogenic genes TNAP and RUNX2 and normalized expression of genes downregulated during osteogenesis (i.e. MGP, SOX9), in vascular smooth muscle cells and valve interstitial cells. ATRA also decreased the expression of osteogenic genes in osteoblasts, however peretinoin did not. These findings suggest that this compound has the potential to specifically target vascular calcification without affecting bone cells.