QJM Advance Access originally published online on June 13, 2005
QJM 2005 98(7):467-484; doi:10.1093/qjmed/hci077
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Review |
Osteoporosis and atherosclerosis: biological linkages and the emergence of dual-purpose therapies
From the Department of Medicine and Resnick Gerontology Center, Albert Einstein College of Medicine and Montefiore Medical Center, New York, USA
Address correspondence to Dr D. Hamerman, Department of Medicine and Resnick Gerontology Center, Albert Einstein College of Medicine and Montefiore Medical Center, 111 East 210th Street, Bronx, NY 10467, USA. email: hamermandj{at}aol.com
 |
Summary |
|---|
Osteoporosis and atherosclerosis are both widely prevalent in an ageing population, and induce serious morbidities and death. There is growing evidence that in addition to their relationship to ageing, osteoporosis and atherosclerosis are also linked by biological associations. This article reviews their clinical interrelations, discusses the basic biology of bone and the arterial wall, and presents five examples that illustrate their biological linkages. Current therapeutic approaches emerging from these linkages, including statins, bisphosphonates, and the thiazolidinediones, have dual effects on bone and the vasculature. Additional therapies derived from experimental studies that enhance bone density and reduce atherogenesis hold further promise to diminish the morbidity and mortality of osteoporosis and atherosclerosis, with attendant benefits to society.
|
Introduction |
|---|
Osteoporosis and atherosclerosis have profound effects on health outcomes, especially in an ageing population.1,2 Indeed, in attempts to limit the adverse health consequences of these diseases, risk factor assessment and considerations of earlier diagnostic and therapeutic interventions have become the dominant themes in the provision of health care,3–7 even in paediatric and adolescent populations.8,9
Ultimately, health outcomes have multiple determinants, and their causation becomes more complex with age. Factors affecting outcomes include personal health practices, the extent of monitoring and types of intervention provided by physician surveillance of the patient's health status, patient compliance, the person's genetic and environmental background, and the inevitability of ageing-related changes. These ageing changes, of course, begin decades before 65 years, and are initially asymptomatic, but may eventually manifest as disease, disabilities, or death.
Bone health9 evolves over time after adolescence from a peak bone mass to ongoing bone loss after the menopause, and osteoporosis may remain virtually asymptomatic until a fragility fracture occurs decades later. ‘Incipient’ atherosclerosis perhaps describes the state of the arterial vasculature evolving with age from the fatty streak to the stable plaque, but when the unstable atherosclerotic plaque ruptures, acute symptoms and profound cardiac damage occur.
The clinical association of vascular calcification and low bone density becomes increasingly manifest over time, especially in women,10,11 as part of the aging process. Although still not entirely resolved, recent evidence strongly suggests an age-independent causal relationship12 in the sense of shared biological linkages that have great importance for the new insights they provide on disease mechanisms, and potentially, for the unique duality of therapeutic approaches that may diminish their impact on adverse personal health. This article discusses five examples of these biological linkages that point the way to therapeutic approaches relevant to the treatment of both conditions. Due to space limitations, references tend to be derived from reviews, rather than original studies.
|
Clinical interrelations |
|---|
Apart from the toll of fragility fractures themselves, osteoporosis is associated with greater all-cause mortality.13 Low bone mineral density (BMD) appeared to be an independent and better predictor of cardiovascular mortality among older men and women than blood pressure and serum cholesterol.14 In postmenopausal women, low-density lipoprotein (LDL) cholesterol was significantly and inversely correlated with BMD, while low triglycerides were observed with vertebral fractures, leading the authors to propose that lipids and bone mass might have ‘common factors linking osteoporosis and atherosclerosis’.15
Calcification of the aorta12,14,16,17 and the coronary arteries (the ‘skeleton in the atherosclerosis closet’)18 is prevalent in older people, may be the basis for cardiac-related morbidity and mortality,16,19 and is also associated with evidence of bone resorption20 and vertebral fractures.11,12 There have been a number of reviews on vascular calcification and its links with diabetes, osteoporosis, and menopause,21 the mechanisms of arterial calcification in relation to bone biology,22 and the relation of arterial calcification ‘in the face of osteoporosis’.23 McFarlane et al. described an array of associations between osteoporosis and cardiovascular disease, asking the question: ‘brittle bones and boned arteries, is there a link?’.24 Rubin and Silverberg, in considering the nature of the ‘nexus’ between atherosclerosis and osteoporosis, noted that ‘the pace of current research might soon elucidate the molecular link between vascular calcification and bone loss’.25 Calcification in the vascular intima, particularly in the plaque, appears to be the crucial association with atherosclerosis,21,22,26,27 ‘representing a meeting of bone biology with chronic plaque inflammation’26and an ‘active and regulated process akin to bone formation’25 that may be observed in cardiac values as well.28,29
Calcification of the extracellular matrix (ECM) is a complex and multi-determined process30 limited by active inhibition provided by matrix proteins31 or regulated by inhibitors and activators of calcification and bone formation.32–34 The molecular events that link calcification propensity in artery and bone are beginning to be understood,21,22,25–27 and are also part of the broader subject of the expression of regulatory proteins in bone matrix35 and atherosclerotic plaques.21,22,25,33,36 It should be noted that osteoporosis in relation to vascular dysfunction and atherosclerosis has also been considered without invoking calcification, suggesting broader biological associations,37 and implications for other disease processes as well. Thus, Pasceri and Yeh recently described a ‘tale of two diseases’, atherosclerosis and rheumatoid arthritis, in a similar vein.38
I will briefly describe individual aspects of bone and arterial wall biology to provide the background for a discussion of the linkages that occur between them in osteoporosis and atherosclerosis, and then review aspects of dual-purpose therapies.
|
Bone biology |
|---|
Osteoblasts and osteoclasts
This is an abbreviated background of a highly complex process discussed in detailed reviews,39–45 and is necessary to introduce nomenclature, or the ‘cast of characters’.39
Most bone forms from a cartilage model, a process termed endochondral ossification. Osteoblasts (the bone-forming cells) have progenitor cells derived from mesenchyme that are multi-potential, and may produce chondrocytes, osteoblasts, myocytes or adipocytes. Molecular modulation of these mesenchymal cells is thus important in their development, and will ultimately determine their final identity and tissue characteristics. Mesenchymal cells destined to become chondrocytes condense, and under the control of transcription factors of the Sox family, differentiate into type II collagen-producing cells; they proliferate and hypertrophy in the extracellular calcified cartilage matrix they produce.44 The balance between chondrocyte proliferation and hypertrophy is controlled by a negative feedback loop involving Indian hedgehog (Ihh), which induces parathyroid hormone-related peptide (PTHrP), formed in the proliferating chondrocyte, to signal to its receptor (PPR) to suppress differentiation to hypertrophic cells. Fibroblast growth factor receptor (FGFR3) may suppress PTHrP from promoting further Ihh expression.43 Cbfa/Runx2, a transcription factor, is required for hypertrophic chondrocytes to release vascular endothelial growth factor (VEGF). This growth factor induces vascular invasion and entry of differentiating osteoblasts, also under the control of Cbfa/Runx2. Other factors secreted from the hypertrophic chondrocytes, such as bone morphogenetic proteins (BMP) and Wnts appear to act in synergy with Ihh to induce osteogenesis. Cbfa1 continues to be necessary for osteoblast differentiation and function, and controls the expression of other downstream genes that encode osteoblast-specific transcription factors, particularly a zinc-finger-containing transcription factor called osterix.45 A different genetic process controls osteoblast proliferation, namely by a gene (lipoprotein receptor protein) encoding a receptor called LDL-receptor related protein (LRP), which will be discussed later on.
Osteoclasts (bone-resorbing cells) are derived from the monocyte/macrophage cell line. Key issues linking the events in bone and the blood vessel wall are the roles of the tissue-based or circulating descendents of macrophage progenitors. In the arterial wall, macrophages play a key role in the development and progression of the atherosclerotic lesion; in bone, the extent of osteoclastic maturation and resorption of bone is critical. Identifying the molecular events in osteoclast development (osteoclastogenesis) has been one of the most dramatic advances in bone biology in the past decade, with many detailed reviews.46–50 Perhaps the key insight in this respect is the ability of the osteoblast/stromal cell itself to promote or diminish osteoclastogenesis by way of osteoblast release of a number of cytokines that influence the osteoclast, including interleukins (IL-6), and tumour necrosis factor (TNF-
), and especially the elaboration of macrophage colony stimulating factor (M-CSF) and receptor activator of nuclear factor-KB (NF-KB) ligand (RANK L). The CSF binds to its osteoclast receptor C-fms, which associates with and activates phosphatidylinositol-3-kinases (PI-3-Ks), also important in osteoclast differentiation and attachment,51 and in vascular responses to states of insulin resistance.52,53 Osteoblast-released RANK L binds to its osteoclast receptor RANK, and activates the NFkB complex and the transcription factor AP-1.46 Osteoblasts can also release osteoprotegerin (OPG), a member of the TNF receptor family, which acts as a decoy to block RANK L–RANK interaction, thereby diminishing osteoclastogenesis.54–57 The RANK L/OPG system is acted upon by many factors to modulate bone formation or resorption, as will be discussed below.
While the effects of osteoblasts on osteoclast functions have received wide attention, as noted above, osteoclasts also appear to influence osteoblasts. New findings have identified mice with deficient osteoclast CSF-1 receptors where, in the absence of osteoclasts, osteoblastic bone formation was markedly disorganized, with reduced mineralization and spontaneous fractures.58
Matrix proteins and factors shared by bone and the arterial wall
Bone matrix produced by osteoblasts consists of collagen type 1, a number of in situ non-collagenous proteins, and products brought to the matrix from the circulation. The non-collagenous proteins,35 which include osteopontin, osteonectin and osteocalcin, regulate cell functions, bind growth factors, and form a nidus for apatite mineralization.33 As studies of gene enhancement or deletion in mice show, these non-collagenous proteins in bone are particularly important in its structure, and (to the point of this article) many are also identified in the arterial intima and aortic valve, where they are synthesized by vascular cells and regulate calcification and ossification.21,22,26–29,33,36,59,60 Indeed, transcription factors Cbfa, Msx2, and Sox 9, discussed above in bone biology, were found in calcified arterial samples and ‘orchestrate the calcification process’60 by virtue of the ‘osteoblast phenotype’.28 Among the matrix protein components involved in bone and the arterial wall to be discussed in biological linkages in this article are bone morphogenetic protein (BMP-2), Wnts, osteopontin (OPN), matrix Gla proteins (Mgp), osteocalcin (OCN) and osteoprotegerin (OPG).
|
Biology of the arterial wall in atherosclerosis |
|---|
The integrity of the arterial wall rests primarily on the intact endothelial cell.61,62 Bone marrow is the source of progenitor cells that home preferentially to sites of vascular injury.63 The subendothelial extracellular matrix and the smooth muscle cells are additional major players in the integrity of the vessel wall. The activation of endothelial cells by cytokines, chemokines and growth factors induces monocyte-macrophage attachment and proliferation in atherogenesis.64,65 Many of the mediators are present in the arterial wall and in bone. Monocyte chemoattractant protein (MCP-1) is one of the cytokines responsible for direct migration of monocytes into the intima at sites of lesion formation,66 and it is of particular interest that MCP-1 is also expressed by osteoblasts in bone.67 The initiation and expansion of the fatty streak occurs upon entry of circulating monocytes into the arterial intima and their transformation by CSF-1 into macrophages that engorge oxidized LDLs and become foam cells.66,68,69 Inflammation plays a key role in all stages of atherogenesis26,29,66,68–70 and is especially important in that it is manifested clinically by circulating serum markers.6 Part of the inflammatory response is the activation of T cells with release of cytokines,66,71,72 especially of RANK L,50 setting the stage for RANK-dependent activation of NFKB in the macrophage in the vessel wall, as occurs in the osteoclast in bone. Thus, the macrophage is a key player in both tissues. Full recent descriptions of the atherogenic processes that ensue over decades have been presented in great detail elsewhere, and are beyond the scope of this article.63,66,68,69,73
|
Osteoporosis and atherosclerosis: biological linkages |
|---|
1. The 12/15 lipoxygenase system and peroxisome proliferator-activated receptor (PPAR
)The lipoxygenase gene Alox 15 encodes the enzyme 12/15 lipoxygenase (12/15 LO) which is active in both bone and the arterial wall.74,75 Pharmacological inhibition of 12/15 LO enhances bone mass, while its genetic overexpression diminishes bone mass.75 In addition,12/15 LO converts arachidonic and linoleic acids into endogenous ligands for a transcription factor called PPAR
. (Other members of the PPAR family are discussed by Castrillo and Tontonoz).76 Linoleate is the largest reservoir of 12/15 LO substrate, and the most abundant fatty acid in LDLs. 12/15 LO in macrophages oxidizes LDLs that are taken up by macrophages,69 with release of transcription factors and cytokines, T cell activation, and increase in VEGF, promoting endothelial and smooth muscle cell proliferation.77,78 Inhibitors of 12/15 LO were tested in animals, and diminished atherosclerotic lesion formation.79 Oxidized lipids also inhibit osteoblast development in vitro and bone formation in vivo, and serve as substrates for PPAR, which redirects differentiation of mesenchymal progenitor cells from osteoblast precursors to adipocytes.80–82 PPAR also enhances many adipocyte genes70,83 and diminishes Cbfa1/Runx2,84 needed for osteoblast development. All these effects could reduce bone formation and promote atherogenesis. PPAR-related ligands (the thiazolidinediones, TZDs) used to treat patients with type 2 diabetes are insulin-sensitizing, and reduce the development of atherosclerosis,52,85 but by inhibiting OPG expression in human aortic smooth muscle cells86 may promote calcification and increase bone turnover in the arterial wall.29,59,87 The adipogenesis induced by PPAR is suppressed by the osteoblast transcription factor Msx2,88 but Msx2, with BMP-2, promotes vascular calcification,89 and Msx2 also stimulated aortic smooth muscle cell proliferation.90 In contrast to PPAR, which increases progenitor cell expression of adipocytes at the expense of osteoblasts, overexpression in transgenic mice of 
fosB, a member of the AP-1 family of transcription factors,91 downregulates adipocyte differentiation from progenitor cells, and enhances osteoblast differentiation and bone formation.92
2. Bone components shared with the arterial wall and the occurrence of calcification
Osteopontin
Osteopontin (OPN), a major non-collagenous matrix protein of bone, is also present in the arterial wall, and in both sites may bind calcium crystals strongly.93 In bone, phosphorylation of OPN in osteoblasts promotes differentiation.94 Overexpression of OPN diminishes mineral deposition by inhibiting BMP-2, which enhances calcification and bone formation.95 Ovariectomized mice that are also OPN-deficient seem protected from accelerated bone loss,96 perhaps because of altered osteoclast function due to impaired cell surface activity of vß3 integrin and CD44, the hyaluronic acid receptor needed to bind osteoclasts to bone.97,98 In addition, bone resorption induced by PTH seems to be dependent on OPN.99
OPN is also a major component in the ECM of human arteries, but whether it promotes or limits calcification in the arterial wall is not clear. Aortic valve leaflets transplanted into OPN-null mice showed accelerated calcification in the valve, compared to wild-type mice.100 OPN is not constitutively expressed, but arises primarily with vessel injury.101 Inflammatory mediators, such as thrombin, induce cleavage of OPN with fragments that promote T-cell and monocyte/macrophage chemotaxis for endothelial and smooth muscle cells.94 Endothelial dysfunction is further promoted by decreased expression of endothelial cell nitric oxide synthase (eNOS) and reduced formation of nitric oxide (NO),93,102 a regulator of cardiovascular homeostasis.103 In coronary artery disease, OPN has been localized in calcified atherosclerotic plaques in association with high serum levels,104,105 and is up-regulated in dystrophic cardiac calcification in mice.106
It may be possible to modulate mineralization in the OPN system. In a mouse with deficient LDL-receptor-related protein (to be discussed below), where a high-fat diet promoted aortic valve calcification, teriparatide (PTH) induced aortic valve expression of OPN, with high serum OPN levels and decreased valve calcification.107 In cultures of bovine aortic smooth muscle cells, addition of organic phosphate induced mineralization by cells that acquired the osteogenic phenotype and expressed Cbfa1, OPN and osteocalcin.108 In this system, the matrix extracellular phosphoglyceroprotein (MEPE) should be examined, for by being up-regulated it may inhibit BMP-2-mediated mineralization,109 and the role of vascular smooth muscle cells in vascular calcification and atherosclerosis.110
Matrix Gla-proteins (Mgp) and osteocalcin (OCN)
Mgp are part of the family of mineral-binding proteins, including osteocalcin, that contain -carboxylated glutamate residues. The human Gla protein is of particular interest, because vitamin K is a coenzyme for glutamate carboxylase, which converts glutamate to -carboxyglutamate. The Gla residues bind and incorporate calcium into hydroxyapatite crystals.111 A moderate literature has arisen on this subject based on spontaneous or warfarin-induced vitamin K deficiency, whereby undercarboxylated osteocalcin with low mineral binding was demonstrated in the circulation and may signify low BMD and a risk factor for fragility fractures.112–114 Indeed, administration of vitamin K retards femoral neck bone loss in postmenopausal women.115 Jie et al. observed that low serum vitamin K in postmenopausal women was associated with low bone mass and atherosclerotic calcification in the abdominal aorta.16 Undercarboxylated Mgp has also been isolated from calcified atherosclerotic plaques of aging rats.116
A number of experimental studies in mice bear on the effects of genetic deletion of Mgp matrix proteins. In one study,117 mice were generated with a disrupted Mgp allele, and within 2 months, extensive calcification in the aorta and its branches led to their rupture and haemorrhage. No fatty streaks or atherosclerotic plaques were observed; perhaps the time was too short, or other predisposing conditions were not present. The inorganic material in the vessel wall was apatite, the mineral of bone and calcified atherosclerotic lesions. Abnormal calcification was observed in the growth plate, disrupting chondrocyte columns, and leading to short stature, osteopenia, and fractures. Mgp appear to inhibit calcification in bone and arteries, perhaps due to Mgp sequestration of BMP, which, when released upon Mgp deficiency or under-carboxylation, promotes osteogenesis.21,101 Indeed, Wallin et al. point out that many of the forces that induce arterial calcification may act by way of modification of Mgp, such as vitamin K deficiency, as noted, or oxidative stress, and thus by releasing Mgp inhibition, allow BMP-2 to promote mineralization.118 BMP actions are enhanced, and bone formation is increased, by various experimental means that disrupt proteins that relay BMP signals.118–120 Mgp expression induced specifically in vascular smooth muscle cells of Mgp-deficient mice preserved the normal arterial phenotype (in the sense that there was no vascular calcification) but mineralization of cartilages still occurred.121 Speer et al. also described an Mgp mutant mouse, but now crossed with an OPN mutant mouse, and observed decreased survival and significant enhancement of vascular calcification beyond that associated with Mgp deficiency alone, suggesting a role for OPN as an ‘inducible inhibitor of calcification’.101
In another study, osteocalcin-deficient mice were generated,122 and the dominant finding was an increase in bone mass. It was concluded that OCN normally functions to limit osteoblastic bone formation without impairing osteoclastic bone resorption or mineralization. There was no mention of the occurrence of arterial calcification.
Osteoprotegerin (OPG)
Of the components that perhaps best illustrate the bone-arterial wall biological linkages, osteoprotegerin (OPG) has attracted the most attention.123 There are several reasons for this interest.
OPG is part of the system whereby the osteoblast modulates osteoclastogenesis by way of interfering with the binding of RANK L to RANK receptor: mice with augmented or absent OPG expression manifest osteopetrosis or osteopenia, respectively. Even in OPG-deficient mice with osteopenia, osteoblast function was enhanced, with overall greater bone turnover due to heightened osteoclast-induced bone loss, a condition that has been compared clinically to Paget's disease and its variants.124 Whyte et al. reported deletion of a gene encoding OPG in humans, with devastating skeletal deformities thought to be ‘juvenile Paget's disease’,125 but according to Krane, more appropriately termed hyperostosis corticalis deformans juvenilis.126
OPG also appears to be a regulator of calcification in the vessel wall, as mice with deletion of the OPG gene develop arterial calcification as well as osteoporosis with multiple fractures.87 The arteries exhibiting calcification in OPG deleted mice are sites of endogenous OPG expression,123 suggesting that OPG may protect arteries from pathological calcification.127 Indeed, Schoppet et al. noted that ‘OPG could represent the long sought-after molecular link between arterial calcification and bone resorption, which underlies the clinical coincidence of vascular disease and osteoporosis’.128 The mechanism by which OPG regulates calcification in arteries is not known.
OPG injected into adult mice deficient in OPG reversed the osteoporosis phenotype but did not diminish arterial calcification; only the OPG transgene in the OPG deficient mice rescued both arterial calcification and osteoporosis.127 Polymorphisms in the promoter region of the human OPG gene were sought in relation to low BMD; one such change not associated with osteoporosis was identified with arterial vascular morphology that may ‘place these persons at increased risk for cardiovascular disease’.129 Thus, OPG gene polymorphism may represent a potential genetic basis for the clinical association of osteoporosis and arterial calcification in atherosclerosis.130,131
Activation of PPAR inhibits OPG expression in aortic smooth muscle cells, although whether this promotes calcification in the vessel is not clear.86 Kiechl et al. showed a strong positive correlation between high serum OPG levels and advanced atherosclerotic vascular disease and mortality.132 It remains to be determined whether elevated serum OPG levels are a causal risk factor for vascular disease, or a compensatory manifestation. It is even more surprising that high serum levels of OPG were observed in post-menopausal women with osteoporosis compared to age-matched controls; high serum OPG levels might have been expected to occur with high BMD, so this finding is contrary to expectations. It was suggested that high OPG levels may be compensatory for enhanced osteoclastic bone resorption.133 Yet in other studies, a single subcutaneous injection of OPG rapidly and profoundly reduced markers of bone resorption, indicating that OPG may be effective in treatment of osteoporosis with accelerated bone loss.134 From all this, it is evident that advances in basic understanding of OPG action on the arterial wall and bone will provide new insights on clinical and therapeutic aspects of OPG in atherosclerosis and osteoporosis.
At this point it is worthwhile to consider overall the traditional association of age-related atherogenesis, based in part on accumulated adverse life-style practices, and postmenopausal oestrogen deficiency with bone loss, and to extend these aspects further to a causal biological linkage. Inflammation is likely to be one of the unifying processes that influence atherogenesis and bone loss.21,26,27 Many of the inflammatory mediators driving atherogenesis in the arterial wall are known to be in the circulation as markers of cardiovascular risk,6 and could gain access to bone where, with local cytokines, they enhance osteoblastic release of factors that in turn promote osteoclastogenesis. Horowitz has indeed emphasized the roles cytokines in bone play in regulating osteoblast functions, and pointed out that bone-resorbing osteoclasts need to be kept under control, in part by ‘suppressive factors’, of which OPG is a good example.135 In turn, mediators from these bone cell activities would also, by way of the circulation, affect cells in the arterial intima. Also noted above, among other aspects, are vitamin K deficiency with undercarboxylated osteocalcin, or deletion of genes regulating Mgp, and OPG,118 resulting in arterial calcification and bone loss. Many additional associations, of course, influence the arterial-bone system, as will be discussed below.
3. The low-density lipoprotein receptor-related protein (LRP)
The LRP gene family encodes cell surface receptors with over 10 members of this gene family; five of these recognize apolipoprotein E as one of their ligands and others are related to Wnt signaling, important in skeletal development.136,137 While one member of this family, LRP5, appeared to have ‘none of the hallmarks of an interesting gene, it has emerged as an important molecule’.138 Indeed, LRP5 and Wnt signaling constitute a ‘union made for bone’.139 In humans, LRP5 polymorphisms contribute to normal variations in BMD.140 Loss of function mutation in the LRP5 gene provides a link in humans between bone, the eye, and blood vessels, resulting in reduced bone mass, occurrence of fractures, and failure to induce regression of primary vitreal vascularization in the eye, with severe ocular pathology.141 This phenotype is recapitulated in Lrp5-deficient mice, where defects in ocular macrophages that express Lrp5 are required for induction of capillary cell death.136 Human subjects with other forms of LRP5 mutation may also display a familial exudative vitreoretinopathy (FEVR), with premature arrest of retinal angiogenesis/vasculogenesis and low bone mass.142 Another form of mutation in the LRP5 gene promotes a kindred of high bone mass.143–145 Some of the LRP genes may be responsible for osteoblast proliferation and bone matrix deposition by the Cbfa1/Runx2 pathway noted above, activating Wnt proteins that are also responsible for regression of embryonic vasculature in the eye.138,146,147 Further, potential vascular linkage in this gene family relates to cholesterol homeostasis: LRP5 binds apolipopotein E, and dual mutations (Lrp5 and ApoE) in mice are associated with hypercholesterolemia, advanced atherosclerosis, and premature coronary artery disease.148–150 The e4 allele of apolipoprotein E appears to be a risk factor for coronary heart disease,151 calcific aortic stenosis,152 and in some studies, for low bone mass and hip fractures.153
4. Nuclear factor KB (NFKB)
Activation of this transcription factor in the macrophage within the arterial wall and osteoclast in bone may be among the most crucial events that link atherosclerosis and osteoporosis. A family of inhibitors, the IKBs, bind to NFKB and retains it in the cell cytoplasm until activated by a variety of stimuli that include cytokines (especially TNF- and IL-17), oxidants, LDLs, glycation end-products, viruses, and immune stimuli.65,154,155 The inhibitory proteins—the IKBs—are then phosphorylated and polyubiquinated, targeting them for degradation by the proteosome, releasing NFKB, which translocates to the nucleus, binding to and stimulating target genes, with a host of products. (Complete descriptions in references 50, 65, 154, 156, 157). RANK L activation of RANK has been discussed above as one of the key events in osteoclastogenesis, and Collins and Cybulsky have described NF-KB activation and its proatherogenic effects in the vessel wall.65 In human aortic valves, OPG was increased in non-calcified areas while RANK L was high in sites of calcification.158
However, there is also evidence that NF-KB activation late in the inflammatory process may induce anti-inflammatory and anti-atherogenic effects.157 Indeed, one such indication is the activation of NF-KB by TNF- that results in the upregulation of a serine/threonine kinase called Akt (protein kinase B).50,54,159 For the vasculature, this has great significance, perhaps relating to an anti-atherogenic aspect. Akt activates eNOS with the production of NO that preserves endothelial cell integrity, and increases activity of the oxygen radical-scavenging system.102 Physical activity enhances Akt,160 and this may be one basis for the effectiveness of exercise on the vasculature. The activation of eNOS in bone is discussed under statins.
5. Leptin and adiponectin
Adipose tissue has come to be thought of as an endocrine organ, releasing a variety of mediators.161,162 and adiposity signals.163 Leptin and adiponectin, derived from adipose tissue, merit consideration for their actions on bone and the vascular wall. Leptin's potential effect on bone came under consideration when it was observed that obese leptin-resistant (ob/ob) or receptor-deficient (db/db) mice were insulin-resistant, and had a high bone mass in spite of hypogonadism and hypercortisonism, conditions that would presumably predispose to low bone mass.164 However, the structure of trabecular and cortical bone in these mice is complex and heterogeneous rather than uniform, with regional variations of low and high bone mass and marrow adipocyte content.165,166 Leptin administration to ob/ob mice had different effects on bone, depending on the site of injection. Leptin administered by the intracerebroventricular route promoted bone loss apart from leptin action on anorexigenic pathways; leptin appeared to act by way of the sympathetic nervous system,164,167 promoting sympathetic tone and hypertension in rodents,168 representing an interesting (although indirect) link by way of the observed augmentation of bone mass with the use of ß-adrenergic blockers.169,170 Alternatively, studies evaluating the effects of peripheral leptin administration in ob/ob mice observed a stimulating effect on bone growth with a dramatic increase in cortical bone formation (reviewed in reference 171).
There is a growing literature on leptin's relation to fat mass and its roles in fertility, energy homeostasis, and neuroendocrine functions.172,173 High fat mass is associated with high leptin levels. Clinical studies attempting to correlate leptin blood levels with bone mass, however, have been inconclusive, with evidence of a modest relation to bone-specific alkaline phosphatase,174 and a stronger association between leptin levels, bone mass, and adiposity.175 That is, the total fat mass is an important determinant of BMD in postmenopausal women176 (see below). Sader et al. reviewed the clinical situation of leptin resistance or leptin receptor insensitivity, where there is concurrently insulin resistance as well, with obesity and manifestations of the metabolic syndrome, discussed below. Elevated leptin levels indeed reflected cardiovascular morbidity with ventricular hypertrophy, and evidence of ‘autonomic overactivity’ (hence ß-blocker use).177 This situation might be likened to the ‘compensatory’ increase in OPN in heart disease and OPG in postmenopausal osteoporosis, as noted above.
Fat mass is also related to the release of adiponectin,178–182 but in contrast to leptin, the higher the fat mass the lower the circulating adiponectin. This has important implications for both atherosclerosis and BMD. Adiponectin decreases hepatic gluconeogenesis and muscle triglycerides, and increases insulin sensitivity, and thus may have anti-atherogenic and anti-inflammatory effects.53,179,183 In the metabolic syndrome, in diabetics, and in patients at increased risk for coronary heart disease, adiponectin blood levels are low.179,182,184 (The molecular state of circulating adiponectin is discussed by Rajala and Scherer.161) High plasma adiponectin levels are associated with lower risk of myocardial infarction in men.184 Indeed, Chandran et al. considered that adiponectin ‘could provide an answer to the fundamental mechanism for the link between vascular inflammation and atherosclerosis’.179 After induced vascular injury, adiponectin null mice indeed show increased atherosclerotic plaque formation, compared to wild-type mice.185
Body weight and total fat mass are strong predictors of BMD in postmenopausal women, as noted.176,180,181 The relation of adiponectin levels to BMD is of interest in terms of whether there is no apparent correlation180 or indeed an association of lower adiponectin levels and high BMD.181 More studies need to be done in this area. It may be that actual body fat mass per se exerts mechanical forces that enhance bone density. Perhaps low body weight associated with osteopenia is comparable in some respect to ‘weightlessness’ induced in laboratory models, where upregulation of PPAR and diminished Cbfa/Runx2 expression suppressed mesenchymal stem cell differentiation and osteoblastogenesis.186
Thus, in effect, there are strong biological linkages and complex metabolic associations between body weight, fat mass, bone density and atherosclerosis. Ott has pointed out the apparent dilemma or ‘biological tradeoff’, as she calls it, of bone benefiting from high body fat mass while cardiac health is impaired.187
Table 1 presents a summary of the biological linkages between osteoporosis and atherosclerosis discussed in this article.
|
|
Therapies based on biological linkages |
|---|
 TopSummaryIntroductionClinical interrelationsBone biologyBiology of the arterial...Osteoporosis and... Therapies based on biological... The future: combined therapies...Note added in proofReferences |
|---|
If there is validity to the biologic basis for linking aspects of osteoporosis and atherosclerosis, then therapies may concordantly enhance bone density and diminish atherogenesis.
Statins
One such example is the statins. So called ‘blockbuster’ drugs,188 with wide and growing use and demonstrated benefit in curtailing atherogenic-related vascular disease,2 they were also studied in the laboratory for their enhancement of BMP-induced osteoblastic bone formation.189 Statins inhibit the HMG-CoA reductase transformation of HMG-CoA to mevalonate, the first step in the cholesterol biosynthetic pathway. Inhibition of cholesterol synthesis in hepatocytes up-regulates the expression of hepatic LDL receptors, and LDL and its precursors are cleared from the circulation.190 Beyond reducing cholesterol synthesis, statin inhibition of HMG-CoA reductase has profound effects on intermediates lower in the mevalonate pathway—effects that have been termed ‘pleiotropic’ because of their diverse and widespread actions on cells and tissues, including the vasculature and bone: powerful evidence for therapeutic linkages.190,191
The mevalonate pathway leads to the formation of isoprenoid intermediates, important for the post-translational modification of proteins. A prenylated protein readily interacts with cellular membranes. Members of the Ras and Rho GTPase family are major substrates for posttranslational modification by prenylation.192 The key step in the activation of Rho is the attachment of isoprenoid geranylgeraniol, which translocates inactive Rho from the cytosol to the membrane. Statins block geranylgeraniol synthesis, inhibiting Rho membrane translocation and activity. Among the results on the vasculature are improved endothelial function with enhanced eNOS activity, and inhibition of plasminogen activator inhibitor-1 and vascular smooth muscle cell proliferation.192–194 In addition, statins also activate protein kinase Akt, a further inducer of eNOS,195 as discussed above. All these pleiotropic effects are anti-atherogenic. Statins may have anti-inflammatory effects on atherosclerotic vasculature,193,196 and the use of statins has been considered to reduce accelerated atherogenesis and vascular risk in patients with rheumatoid arthritis.197
Statins also act on bone to increase BMP-2 production, leading to osteoblast differentiation and bone formation. eNOS, the isoform most widely expressed in bone,198 is upregulated in osteoblasts and osteocytes. eNOS may mediate the osteogenic effects induced in response to mechanical strain or shear flow. eNOS gene knockout mice are hypertensive and appear to be osteoporotic, apparently due to diminished osteoblastic activity which, unlike the human condition, tends to improve with age.199 Mundy's work stimulated a large number of clinical studies to determine whether statin users, aside from beneficial cardiovascular effects, showed improved BMD and reduced fracture risk. There is generally a conflicting literature here.200–202 Part of the problem, as both Mundy and Edwards et al. have suggested, is that statins in conventional doses are largely taken up by the liver, and blood levels reaching bone would be very low.200,203
Bisphosphonates
Bisphosphonates are the most effective inhibitors of bone resorption, and are widely used for treatment and prevention of osteoporosis.204 Osteoclasts absorb bisphosphonates bound to mineral on bone, and the ingested bisphosphonates limit a number of osteoclast functions. Potent nitrogen-containing bisphosphonates (such as alendronate and risedronate) are not metabolized, and act on the mevalonate pathway lower than the statins, limiting prenylation of proteins normally attached to the cell membrane, thereby inducing osteoclast abnormalities in membrane ruffling and bone resorptive functions.205
The bone mineral to which bisphosphonates adhere concentrates the drug for osteoclast ingestion. Another way to deliver a high concentration of bisphosphonates to tissues is to encapsulate them in liposomes, particularly for phagocytosis by macrophages.206 This approach was taken up by Danenberg et al., who showed that liposomal clodronate or liposomal alendronate depleted circulating monocytes and tissue macrophages, and inhibited in-stent neointimal hyperplasia in hypercholesterolemic rabbits. The potential was considered for therapeutic treatment in patients undergoing coronary catheterization and stent placement.207,208 In animal studies, bisphosphonates in doses that inhibited bone resorption also reduced arterial calcification, and clinical studies in postmenopausal women were suggested.20 Bisphosphonate treatment apparently did not influence the progression of aortic calcification in elderly postmenopausal women.209
PPAR agonists: the thiazolidinediones (TZD)
Patients with type 2 diabetes manifest the metabolic syndrome that includes central obesity, insulin resistance, dyslipidaemia, hypertension, and a pro-inflammatory state, predisposing to atherogenesis and cardiovascular disease.210–212 In addition, subjects with type 2 diabetes who are osteoporotic had higher serum levels of undercarboxylated osteocalcin, perhaps secondary to a defect in -glutamylcarboxylase activity by vitamin K.213 Older women with diabetes appear to have an increased fracture risk as a result of falls, proprioception loss, and vision failure.214 However, bone density in many diabetics may not be low, and indeed may be higher than in age-matched controls.215 The higher BMD observed is not completely accounted for by obesity.216 Perhaps low serum adiponectin in obesity and the metabolic syndrome may be associated with higher bone mass, yet lead to increased insulin resistance and atherogenesis.180,217 PPAR agonists, the TZDs (glitazones), have beneficial effects on glucose homeostasis and lipid metabolism, with reduced release of free fatty acids and insulin-resistance-mediating adipocytokines, such as TNF-, leptin or resistin, and increased production of anti-atherogenic, anti-diabetic adiponectin.52,161,182,218 In experimental studies in animals and human trials, TZDs reduced atherogenesis by diminishing vascular inflammation and vasoconstriction, monocyte chemotaxis, proliferation and migration of smooth muscle cells, and production of adhesion molecules and metalloproteinases.64,85,183,211,219,220 Adiponectin gene expression was increased.178 With respect to the effects of PPAR agonists on bone, the laboratory observation of the shift of progenitor cells from osteoblastogenic to adipogenic80,82 and diminished bone-enhancing components84,86 may reduce bone formation and promote osteopenia; these aspects need to be explored in clinical studies. Nevertheless, it may be that bone density could be maintained by other actions of PPAR that may positively affect bone: specifically, decreasing TNF- formation in adipose tissue. Nanes has written an extensive review of the roles of TNF- on skeletal pathology.221 This cytokine has acquired a central position of interest among its many actions by suppressing oestrogens,222 clearly of importance in postmenopausal osteoporosis, and by acting on an array of systems that limit bone formation by osteoblasts and enhance bone resorption by osteoclasts. Therapies to suppress TNF- may thus benefit bone as well as the vasculature.223–225
|
The future: combined therapies to limit bone loss and atherogenesis? |
|---|
|
TopSummaryIntroductionClinical interrelationsBone biologyBiology of the arterial...Osteoporosis and...Therapies based on biological...The future: combined therapies...Note added in proofReferences |
|---|
Mundy's studies on the bone-enhancing effects of statins really set the stage for the concept of linking therapies that improve bone density and limit progression of atherosclerosis, and this may also be the case for nitrogen-containing bisphosphonates, as noted above. In the course of the studies discussed in this paper, genetic or pharmacological, in mice or in humans, a number of investigators have observed bone density enhancement by (i) pharmacologically inhibiting the enzyme 12/15 LO75 (which also suppressed atherosclerotic lesions79); (ii) diminishing osteocalcin expression;97 (iii) modifying LRP5 in ways predisposing to higher bone mass;143–145 (iv) deleting the Wnt antagonist ‘secreted frizzled-related protein’;226 (v) or using gene therapy with human recombinant osteoprotegerin57 or its subcutaneous injection.134 Dual effects that diminish atherosclerotic lesions and enhance bone density would be particularly desirable. A striking example of such an association is the mouse model of deficient low-density lipoprotein receptor protein, discussed above, where PTH (teriparatide) indeed increased bone mineral density and suppressed aortic valve calcification.107 Bone effects need to be monitored, and indeed may be enhanced, with current use of TZDs or future use of adiponectin to improve the diabetic state and the metabolic syndrome, with known reduction in atherogenesis.179,182,227 Table 2 reviews the range of clinical and experimental therapies discussed here.
|
Many factors influence the biological linkages in humans that regulate osteoporosis and atherosclerosis with calcification: ageing, genetic makeup, body habitus, associated morbidities, lifestyle and dietary practices. Combined therapies now available may enhance bone density and limit atherosclerosis progression. But a greater understanding of the biological linkages, many of them set forth in this paper, may lead to new dual-purpose therapies that may ultimately prevent the adverse outcomes of osteoporosis and atherosclerosis. If this proves to be the case, an ‘era of preventive gerontology’ would indeed be on the way, beyond the current benefits of good personal health practices.228 Improved health, morbidity reduction, and cost curtailment would represent notable benefits to society in the decades ahead.
|
Note added in proof |
|---|
Since submission and final acceptance of this paper, selected additional references—many of them published in 2005—are worth citing here because of their relevance. Molecular mechanisms in bone development were reviewed.229,230 Recent papers discussed diabetic atherosclerosis,231,232 PPAR
in macrophage biology,233 and similarities between adipocytes and macrophages,234 citing aspects of the inflammatory response and the therapeutic effects of TZDs. Clarification of the biological roles of PPAR in osteogenesis and atherosclerosis will influence pharmacological approaches to enhance or inhibit PPAR actions.235 Immune and inflammatory events in bone of ovariectomized mice provided a model for human postmenopausal estrogen deficiency and osteoporosis.236 High serum leptin levels appeared to predict low risk for a fragility fracture, at least partly independent of fat mass.237 Statins inhibited calcification induced by cell lines in vitro, including human vascular smooth muscle cells238 and aortic valve myofibroblasts, while ‘paradoxically’ stimulating calcification in an osteoblastic cell line,239 perhaps evidence for statins as dual-purpose therapies. Calcification in aortic stenosis was ‘surprisingly similar to calcification in the bony skeleton’, and statins may limit progression of valvular stenosis independent of cholesterol lowering.240 In experimental atherosclerosis, bisphosphonates accumulated in macrophages in the vascular wall, and inhibited calcification, lipid accumulation, and fibrosis.241 Yet in a recent study using the Apo E–/– mouse model of atherosclerosis, bisphosphonates induced the ‘surprising’ finding of inflammation and plaque rupture.242 Elevated serum homocysteine levels have earlier been cited as a risk factor for cardiovascular disease, and now more recently, for osteoporotic fractures as well (summarized in reference 243). |
Acknowledgments |
|---|
The author is grateful to Gregory Mundy, PhD, University of Texas Health Science Center, San Antonio, TX for his critical reading of this paper; to Mark Goldberger, MD, and Mark Menegus, MD, Division of Cardiology, Montefiore Medical Center, Bronx, NY, for encouragement and citing references; and to Dawn Bowen-Jenkins for preparation of the manuscript.
|
References |
|---|
1. Braithwaite RS, Col NF, Wong JB. Estimating hip fracture morbidity, mortality and costs. JAGS 2003; 51:364–70.
2. Pepine CJ. Foreword. Statins and the vascular wall: clinical and mechanistic correlates of early benefit. Am J Cardiol 2003; 91(4A):1B.
3. McGill HC, McMahan CA. Starting earlier to prevent heart disease. JAMA 2003; 290:2320–2.
4. Gluckman TJ, Baranowski B, Ashen D, et al. A practical and evidence-based approach to cardiovascular disease reduction. Arch Intern Med 2004; 164:1490–500.
5. Siris ES, Miller PD, Barrett-Connor E, et al. Identification and fracture outcomes of undiagnosed low bone mineral density in postmenopausal women. JAMA 2001; 286: 2815–22.
6. Willerson JT, Ridker PM. Inflammation as a cardiovascular risk factor. Circulation 2004; 109(Suppl II):II-2–II-10.
7. Fruchart JC, Nierman MC, Stoes ESG, Kastelein JJP, Duriez P. New risk factors for atherosclerosis and patient risk factor assessment. Circulation 2004; 109(Suppl):15–19.
8. Slyper AH. What vascular ultrasound testing has revealed about pediatric atherogenesis, and a potential clinical role for ultrasound in pediatric risk assessment. J Clin Endocrinol Metab 2004; 89:3089–95.
9. Hamerman D. Bone health across the generations. A primer for health providers concerned with osteoporosis prevention. Maturitas 2004; 50:1–7.[CrossRef]
10. Kiel DP, Kauppila LI, Cupples LA, et al. Bone loss and the progression of abdominal aortic calcification over a 25 year period: the Framingham Heart Study. Calcif Tissue Int 2001; 68:271–6.[CrossRef][ISI][Medline]
11. Frye MA, Melton LJ 3rd, Bryant SC, et al. Osteoporosis and calcification of the aorta. Bone Miner 1992; 19:185–94.[CrossRef][ISI][Medline]
12. Schulz E, Arfai K, Liu X, Sayre J, Gilsanz V. Aortic calcification and the risk of osteoporosis and fractures. J Clin Endocrinol Metab 2004; 89:4246–53.
13. Kado DM, Browner WS, Blackwell T, Gove R, Cummings SR. Rate of bone loss is associated with mortality in older women: A prospective study. J Bone Miner Res 2000; 15:1974–80.[CrossRef][ISI][Medline]
14. Tankó LB, Bagger YZ, Christiansen. Low bone mineral density in the hip as a marker of advanced atherosclerosis in elderly women. Calcif Tissue Int 2003; 73:15–20.[CrossRef][ISI][Medline]
15. Yamaguchi T, Sugimoto T, Yano S, et al. Plasma lipids and osteoporosis in postmenopausal women. Endocr J 2002; 49:211–17.[ISI][Medline]
16. Jie KSG, Bots ML, Vermeer C, Witteman JCM, Grobbee DE. Vitamin K status and bone mass in women with and without aortic atherosclerosis: A population-based study. Calcif Tissue Int 1996; 59:352–6.[CrossRef][ISI][Medline]
17. Hak AE, Pols HA, van Hemert AM, Hofman A, Witteman JC. Progression of aortic calcification is associated with metacarpal bone loss during menopause: a population-based longitudinal study. Arterioscler Thromb Vasc Biol 2000; 20:1926–31.
18. Demer LL. A skeleton in the atherosclerosis closet. Circulation 1995; 92:2029–32.[ISI][Medline]
19. Thompson GR, Partridge J. Coronary calcification score: the coronary-risk impact factor. Lancet 2004; 363:557–9.[CrossRef][ISI][Medline]
20. Price PA, Faus SA, Williamson MK. Bisphosphonates alendronate and ibandronate inhibit artery calcification at doses comparable to those that inhibit bone resorption. Arterioscler Thromb Vasc Biol 2001; 21:817–24.
21. Abedin M, Tintut Y, Demer LL. Vascular calcification. Mechanisms and clinical ramifications. Arterioscler Thromb Vasc Biol 2004; 24:1161–70.
22. Doherty TM, Fitzpatrick LA, Inoue F, et al. Molecular, endocrine, and genetic mechanisms of arterial calcification. Endocr Rev 2004; 25:629–72.
23. Parhami F, Demer LL. Arterial calcification in the face of osteoporosis in ageing: can we blame oxidized lipids? Curr Opin Lipidol 1997; 8:312–14.[ISI][Medline]
24. McFarlane SI, Muniyappa R, Shin JJ, Bahitiyar G, Sowers JR. Osteoporosis and cardiovascular disease: brittle bones and boned arteries—is there a link? Endocrine 2004; 23: 1–10.[CrossRef][ISI][Medline]
25. Rubin MR, Silverberg SJ. Vascular calcification and osteoporosis–The nature of the nexus. J Clin Endocrinol Metab 2004; 89:4243–5.
26. Doherty TM, Asotra K, Fitzpatrick LA, et al. Calcification in atherosclerosis: bone biology and chronic inflammation at the arterial crossroads. Proc Natl Acad Sci USA 2003; 100:11203–6.
27. Demer LI. Vascular calcification and osteoporosis: inflammatory response to oxidized lipids. Int J Epidemiol 2002; 31:737–41.
28. Rajamannan NM, Subramaniam M, Rickard D, et al. Human aortic valve calcification is associated with an osteoblast phenotype. Circulation 2003; 107:2181–4.[CrossRef][ISI][Medline]
29. Mohler ER III, Gannon F, Reynolds C, Zimmerman R, Keane MG, Kaplan FS. Bone formation and inflammation in cardiac valves. Circulation 2001; 103:1522–30.[ISI][Medline]
30. Dziak R. Role of lipids in osteogenesis: cell signaling and matrix calcification. In: Bonucci E, ed. Calcification in biological systems. Boca Raton, CRC Press, 1992:59–71.
31. Schinke T, McKee MD, Karsenty G. Extracellular matrix calcification: where is the action? Nat Genet 1999; 21: 150–1.[CrossRef][ISI][Medline]
32. Tintut Y, Demer LL. Recent advances in multifactorial regulation of vascular calcification. Curr Opin Lipidol 2001; 12:555–60.[CrossRef][ISI][Medline]
33. Dhore CR, Cleutjens JPM, Lutgens E, et al. Differential expression of bone matrix regulatory proteins in human atherosclerotic plaques. Arterioscler Thromb Vasc Biol 2001; 21:1998–2003.
34. Boskey AL. Bone mineral and matrix. Are they altered in osteoporosis? Orthop Clin NA 1990; 21:19–29.
35. Robey PG. Vertebrate mineralized matrix proteins: Structure and function. Connect Tissue Res 1996; 35:131–6.[Medline]
36. Vattikuti R, Towler DA. Osteogenic regulation of vascular calcification: an early perspective. Am J Physiol Endocrinol Metab 2004; 286: E686–96.
37. Alagiakrishnan K, Juby Y, Hanley D, Tymchak W, Sclater A. Role of vascular factors in osteoporosis. J Gerontol Med Sci 2003; 58A:362–6.
38. Pasceri V, Yeh ETH. A tale of two diseases. Atherosclerosis and rheumatoid arthritis. Circulation 1999; 100:2124–6.[ISI][Medline]
39. Karsenty G, Wagner EF. Reaching a genetic and molecular understanding of skeletal development. Develop Cell 2002; 2:389–406.
40. Karsenty G. Chondrogenesis just ain't what it used to be. J Clin Invest 2001; 107:405–7.[ISI][Medline]
41. Harada S, Rodan GA. Control of osteoblast function and regulation of bone mass. Nature 2003; 423:349–55.[CrossRef][Medline]
42. Chung UI. Essential role of hypertrophic chondrocytes in endochondral bone development. Endocrine J 2004; 51:19–24.
43. Amizuka N, Davidson D, Liu H, et al. Signalling by fibroblast growth factor receptor 3 and parathyroid hormone-related peptide coordi