|
|
||||||||


Atherosclerosis Research Center, Division of Cardiology, Department of Medicine, and the Burns and Allen Research Institute, Cedars-Sinai Medical Center and UCLA School of Medicine, Los Angeles, California, USA;
Division of Endocrinology, Department of Internal Medicine, Mayo Clinic and Mayo Foundation, Rochester, Minnesota, USA; and
Department of Development, Amgen, Incorporated, Thousand Oaks, California, USA
1Correspondence: Atherosclerosis Research Center, Davis Research Bldg., Room 1062, Cedars-Sinai Medical Center, 8700 Beverly Blvd., Los Angeles, CA 90048-1865, USA. E-mail: rajavashisth{at}cshs.org
ABSTRACT
Atherosclerotic arteries frequently become calcified, and these calcium deposits are associated with a high risk of adverse clinical events. Descriptive studies suggest calcification is an organized and regulated process with many similarities to osteogenesis, yet the mechanism and its relationship to atherosclerosis remain largely unknown. In bone development and homeostasis, mineral deposition by osteoblasts and mineral resorption by osteoclasts are delicately balanced such that there is no overall gain or loss in bone mass. We hypothesize that there exists in arteries a mechanism that similarly balances mineral deposition with resorption. We propose that the cellular mediators of arterial mineral resorption are osteoclast-like cells (OLCs) derived from hematopoietic precursors of the mononuclear phagocytic lineage. In arterial microenvironments, mononuclear precursors are induced to differentiate toward OLCs by macrophage-colony stimulating factor and receptor activator of NF-
B ligand, both of which are necessary and sufficient for osteoclastogenesis and mineral resorption in bone. OLCs may participate in normal mineral homeostasis within the arterial wall or, alternatively, may be recruited to specific sites within developing plaque. Net calcium deposition occurs as a result of focal perturbation of the balance between the activity of osteoblast-like cells and OLCs. Our proposed mechanism thus views arterial mineral deposition not so much as an active pathological process, but as a localized failure of protective mechanisms that actively oppose mineral deposition within the disordered metabolic milieu of developing atherosclerotic plaque.Doherty, T. M., Uzui, H., Fitzpatrick, L. A., Tripathi, P. V., Dunstan, C. R., Asotra, K., Rajavashisth, T. B. Rationale for the role of osteoclast-like cells in arterial calcification.
Key Words: atherosclerosis mineral resorption mononuclear phagocytes OLCs
CALCIUM MINERAL DEPOSITION frequently accompanies atherosclerosis, begins when the disease is still an early stage of development, and is associated with a high likelihood of subsequent adverse events such as myocardial infarction and coronary death (1
2
3
4
5)
. Although previously conceptualized as a passive precipitation, accumulating evidence now suggests that arterial calcification is the result of organized, regulated processes with many similarities to osteogenesis (3
, 6
7
8)
. In vitro studies have shown that the artery wall contains cells that retain the capacity to differentiate into osteoblast-like cells (9
10
11
12
13
14
15
16
17
18
19)
. These cells are referred to as calcifying vascular cells (CVCs) and, under selective stimuli, have the potential to spontaneously form mineralized nodules and express bone morphogenetic protein (BMP) -2, osteocalcin, osteopontin, osteonectin, alkaline phosphatase, and collagen type I (6
, 7
, 20
, 21)
. CVCs are sparsely present in normal intima and media but become more numerous in atherosclerotic areas. CVCs respond to factors such as TGF-ß1, 25-hydroxycholesterol, and BMP-2 by dramatically accelerating nodule formation and expression of osteoblast-related genes (18
19
20
21)
.
Mineral deposition in bone is accomplished by osteoblasts; the opposing processmineral resorptionis mediated by osteoclasts (22
, 23)
. Under normal homeostatic conditions, the activities of these two processes are coupled, so there is no net change in bone mass. Bone pathologies such as osteoporosis and osteopetrosis are characterized by an uncoupling of mineral deposition and resorption in which one of these processes predominates (24)
. Demonstration of the existence of osteoblast-like cells in atherosclerotic arteries and pluripotent cells capable of maturing along an osteoblast lineage in normal arteries (25
, 26)
led us to whether there was an opposing process arteries similar to that in bone mediated by an arterial cell that could differentiate into a mineral resorbing cell similar to an osteoclast.
HYPOTHESIS
We hypothesize that mineralization in atherosclerotic arteries results from an uncoupling of two opposing processes: mineral deposition by osteoblast-like cells and mineral resorption by osteoclast-like cells (OLCs). OLCs may have a homeostatic function in normal arteries and participate in regulating extracellular mineral concentrations. Alternatively, OLCs might only be recruited to sites of developing plaque. We further hypothesize that the primary determinant of net mineral deposition in diseased arteries is inhibition of mineral resorption by OLCs rather than mineral deposition of osteoblast-like cells. Reduced mineral resorption might be secondary to decreased maturation, survival, and/or function of OLCs within developing calcified vascular lesion. Here we present several lines of evidence that lend support to our hypotheses, provide the basis for a mechanistic test, and raise questions concerning ontogenetic cell fate and its determinants in pathobiologic microenvironments.
OSTEOPETROTIC MICE EXHIBIT AUGMENTED ARTERIAL CALCIFICATION
Mutations in mice affecting either maturation or function of osteoclasts result in osteopetrosis (22
, 23
, 27)
. Studies in some of these mice have shown that osteopetrosis is accompanied by arterial calcification. For example, atherosclerosis-prone mice that exhibit osteopetrosis because of a point mutation in the gene encoding macrophage-colony stimulating factor (M-CSF) develop extensive arterial calcification (28)
, suggesting that the development of calcification in atherosclerotic lesions may be in part the result of decreased vascular osteoclastic activity. Supporting this conclusion are results from mice lacking carbonic anhydrase (CA) II. Osteoclasts lacking CA II are incapable of acidifying the resorptive microenvironment (23
, 29
, 30)
, and mice deficient in CA II are osteopetrotic and develop arterial calcification (31
, 32)
. Collectively, findings from these studies suggest that reduced numbers of functional osteoclasts enable mineral deposition by arterial osteoblast-like cells to occur unopposed. Paradoxically, however, mice that lack osteoprotegrin (OPG), the soluble decoy receptor for and receptor activator of NF-
B ligand (RANKL), exhibit not only osteoporosis but also arterial calcification and the presence of tartrate-resistant acid phosphatase-(TRAP) and cathepsin K-positive vascular OLCs (33)
, some of which are multinucleated (Fig. 1
). Intravenous injection of recombinant OPG protein or transgenic overexpression of OPG in OPG-deficient mice rescues the osteoporotic phenotype but only the latter treatment prevents arterial calcification (34)
. However, in a rat model of arterial calcification induced by warfarin, vitamin D, and vitamin K, treatment with OPG reduced or inhibited the development of arterial calcification (35)
. The explanation for these divergent findings is not known, but there are several possibilities. First, OPG might play a role in regulation of expression of RANKL and RANK in arterial tissues that is different from that in bone microenvironments. RANKL is a type II transmembrane protein, but can be proteolytically cleaved to form a soluble cytokine by TNF-
-converting enzyme (36)
. Activated murine T lymphocytes secrete a soluble and active form of RANKL into culture medium (37)
. It has recently been shown that OPG deficiency stimulates osteoblast-mediated maturation of osteoclasts in cocultures of osteoblasts and bone marrow precursors derived from OPG-deficient mice, and this was inhibited by the addition of OPG to cocultures (37
, 38)
. RANKL is not expressed in the arteries of wild-type mice but is expressed in the arteries of OPG-deficient mice (34)
, suggesting that OPG may regulate RANKL and possibly RANK expression in arteries in a manner that is distinct from that in bone marrow. We propose that OPG deficiency in arterial tissues affects RANKL and/or RANK expression, diminishes RANKL effects, and alters expression of and/or signaling by M-CSF and its receptor, c-Fms.
|
Alternatively, arterial calcification in OPG knockout mice could be related to BMP-2 signaling, which may be different in bone marrow than in arteries. The role of BMPs in osteoblast development has been investigated extensively, but recent evidence suggests that BMPs may play an important role in osteoclast differentiation and function as well. BMP-2 dramatically increased osteoclast formation in bone marrow-derived macrophages in the presence of RANKL and M-CSF, which was inhibited by soluble decoy BMP receptors (39)
. BMP-2 also enhanced osteoclast survival. Thus, BMP-2 appears to potentiate the effects of RANKL and M-CSF on osteoclast development. It has been shown that BMP-2 is expressed in atherosclerotic arteries (19)
; however, it is unlikely that BMP-2 is expressed in normal arteries. It is therefore possible that the absence of BMP-2 results in a relative deficiency of OLCs in arteries vs. osteoblast-like cells.
It may be that atherosclerosis induces the migration of OLC precursors from circulating mononuclear phagocytic cells (MPCs); in the absence of plaque, no precursors would be expected to be present in arteries. Nevertheless, it has been suggested that pluripotent osteoblast-like cells are present even in normal arteries. OPG deficiency in this context might induce increased activation or maturation of these osteoblast-like precursors, which can then form calcium mineral deposits unopposed since there are no OLCs in the nonatherosclerotic arteries of OPG-deficient mice. Further studies are clearly needed to determine the molecular mechanisms regulating arterial calcification in OPG knockout mice.
Supporting the role of OLCs in arterial calcification are recent findings from our laboratory showing that human atherosclerotic coronary arteries with calcium deposits exhibit colocalization of TRAP- and cathepsin K-positive phagocytic cells with areas of mineralization as determined by von Kossa staining (unpublished observations).
ARTERY WALL CELLS EXPRESS MOLECULES REQUIRED FOR THE MATURATION AND FUNCTION OF OLCs
M-CSF- and RANKL-mediated cell signaling are both necessary and sufficient for osteoclast maturation from precursors (40
41
42
43)
, as demonstrated by the lack of osteoclast development in mice with null a mutation of either gene (37
, 44
, 45)
. All cells of the normal and atherosclerotic vessel wall, including endothelial cells, smooth muscle cells, and monocyte-macrophages, express M-CSF (46
47
48)
. Human vascular endothelial cells express RANKL and OPG (49)
. RANKL protein is expressed in small blood vessels of the skin (50)
, OPG is expressed in normal human and mouse arteries (33
, 34
, 51)
, and OPG, RANK, and RANKL transcripts, normally expressed by osteoblastic stromal cells and osteoclast precursors (52
, 53)
, are found in cells associated with calcified arterial lesions of OPG-deficient mice. Circulating osteoclast precursors, MPCs, and B cells (but not T cells) express c-fms (54)
, the cell-surface receptor for M-CSF (55
, 56)
, and RANK, the cell-surface receptor for RANKL (57)
.
We hypothesize that both M-CSF and RANKL mediate the differentiation of a subpopulation of MPCs within developing atherosclerotic lesions toward an osteoclast-like phenotype. A schematic representation of our hypothesis is shown in Fig. 2
. We propose that maturation of MPCs to OLCs requires the presence of CVCs, which express multiple M-CSF isoforms and RANKL. Secreted, membrane-bound, or matrix-associated M-CSF isoforms and RANKL bind to their specific receptors (c-fms and RANK, respectively) expressed on the surface of MPCs, thereby providing essential and sufficient signals for maturation of MPCs to OLCs.M-CSF- and RANKL-dependent cell signaling may affect processes related to cell survival and function such as mineral resorption in arteries. Although both M-CSF and RANKL signaling are necessary for osteoclast development, lack of M-CSF alone is sufficient to cause diminution in osteoclast numbers and consequent osteopetrosis and arterial calcification (28
, 58)
.
|
CELLULAR POTENTIAL FOR OLCs EXISTS IN ARTERIES
Pluripotent cells exist in arteries, and these could conceivably give rise to OLCs. However, for several reasons it seems more likely that OLCs originate from circulating hematopoietic precursors that migrate into the arterial wall. First, MPCs are abundantly present in plaque during all stages of atherogenesis (59
, 60)
. Second, MPCs and OLCs are both hematopoietic in origin and are closely related to one another (23
, 40
, 61)
. Third, compelling evidence indicates that osteoclast precursors reside not just in bone marrow, but also spleen and peripheral blood subfractions (62)
, and recent evidence shows the presence of circulating marrow stromal-derived cells in atherosclerotic neointima (63)
. Finally, it has been shown that monocyte (59
, 60)
and osteoclast (64)
precursors are both recruited by endothelial cells from circulating blood.
These findings suggest that pluripotent OLC precursors are present in arteries. Several lines of evidence from tissue culture experiments support the hypothesis that arterial MPC precursors can differentiate into OLCs that can resorb arterial mineral deposits. First, circulating monocytic and extraskeletal fibroblastic cells can be induced to differentiate into osteoclasts in tissue culture (65
66
67)
. Second, coculture of human blood monocytes with spindle-shaped cells from bone tumors results in differentiation of monocytes into TRAP-positive, multinucleated, giant OLCs that express the osteoclast-associated calcitonin receptor antigen and show bone resorption activity (68)
. These effects did not require direct cell-to-cell contact, since they persisted even with interposition of barriers between the two cell types (68)
. Last, we have recently observed that coculture of human monocytes with confluent cultures of CVCs with preformed calcified mineral deposits leads to mineral resorption and reduction in the size of mineralized acellular nodules (unpublished data). Collectively, these findings strongly suggest that the cellular prerequisites for mineral resorption may exist in arteries.
Figure 3
illustrates our proposed mechanism for the maturation of circulating hematopoietic precursors to osteoclasts in bone and OLCs in atherosclerotic plaque. Studies of osteopetrotic mouse mutants provide compelling evidence that a common myeloid lineage gives rise to both mononuclear-macrophage lineage and the osteoclast lineage (22
, 23
, 27
, 28)
. Results from OPG-deficient mice showing deposits of calcium in arteries coinciding with multinucleated cells that are TRAP- and cathepsin K-positive, but negative for the macrophage antigen F4/80, are consistent with a divergence of differentiation of MPCs away from the macrophage lineage and toward an osteoclast lineage under the influence of RANKL and M-CSF signaling (33
, 34)
. Taken together, these findings support our hypothesis that peripheral MPCs can differentiate along an osteoclast lineage and that OLCs may be an important cellular mediator of mineral resorption in arteries in both normal homeostasis and pathological conditions. Future studies should shed light on the validity of these hypotheses.
|
ACKNOWLEDGMENTS
We thank Prediman K. Shah and Douglas J. Wilkin for critically reviewing the manuscript. This work was supported by grants HL 51980 and HL 58555 from the National Heart, Lung and Blood Institute, National Institutes of Health (to T.B.R.) and 1S10 RR13717 from the National Center for Research Resources, National Institutes of Health.
Received for publication December 6, 2001. Accepted for publication January 9, 2002.
REFERENCES
B ligand) mRNA and protein in skeletal and extraskeletal tissues. Bone 25,525-534[Medline]
This article has been cited by other articles:
![]() |
S. Mathew, R. J. Lund, L. R. Chaudhary, T. Geurs, and K. A. Hruska Vitamin D Receptor Activators Can Protect against Vascular Calcification J. Am. Soc. Nephrol., August 1, 2008; 19(8): 1509 - 1519. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Mathew, K. S. Tustison, T. Sugatani, L. R. Chaudhary, L. Rifas, and K. A. Hruska The Mechanism of Phosphorus as a Cardiovascular Risk Factor in CKD J. Am. Soc. Nephrol., June 1, 2008; 19(6): 1092 - 1105. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. G. Seipelt, C. L. Backer, C. Mavroudis, V. Stellmach, M. Cornwell, I. M. Seipelt, F. A. Schoendube, and S. E. Crawford Osteopontin expression and adventitial angiogenesis induced by local vascular endothelial growth factor 165 reduces experimental aortic calcification J. Thorac. Cardiovasc. Surg., April 1, 2005; 129(4): 773 - 781. [Abstract] [Full Text] [PDF] |
||||
![]() |
P.E. Norman and J.T. Powell Vitamin D, Shedding Light on the Development of Disease in Peripheral Arteries Arterioscler. Thromb. Vasc. Biol., January 1, 2005; 25(1): 39 - 46. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Collin-Osdoby Regulation of Vascular Calcification by Osteoclast Regulatory Factors RANKL and Osteoprotegerin Circ. Res., November 26, 2004; 95(11): 1046 - 1057. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. D. Stout and J. Suttles Functional plasticity of macrophages: reversible adaptation to changing microenvironments J. Leukoc. Biol., September 1, 2004; 76(3): 509 - 513. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. M. Doherty, L. A. Fitzpatrick, D. Inoue, J.-H. Qiao, M. C. Fishbein, R. C. Detrano, P. K. Shah, and T. B. Rajavashisth Molecular, Endocrine, and Genetic Mechanisms of Arterial Calcification Endocr. Rev., August 1, 2004; 25(4): 629 - 672. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Subbarao, V. R. Jala, S. Mathis, J. Suttles, W. Zacharias, J. Ahamed, H. Ali, M. T. Tseng, and B. Haribabu Role of Leukotriene B4 Receptors in the Development of Atherosclerosis: Potential Mechanisms Arterioscler. Thromb. Vasc. Biol., February 1, 2004; 24(2): 369 - 375. [Abstract] [Full Text] |
||||
![]() |
T. M. Doherty, K. Asotra, L. A. Fitzpatrick, J.-H. Qiao, D. J. Wilkin, R. C. Detrano, C. R. Dunstan, P. K. Shah, and T. B. Rajavashisth Calcification in atherosclerosis: Bone biology and chronic inflammation at the arterial crossroads PNAS, September 30, 2003; 100(20): 11201 - 11206. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. A. Fitzpatrick, R. T. Turner, and E. R. Ritman Endochondral Bone Formation in the Heart: A Possible Mechanism of Coronary Calcification Endocrinology, June 1, 2003; 144(6): 2214 - 2219. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. L. Tyson, J. L. Reynolds, R. McNair, Q. Zhang, P. L. Weissberg, and C. M. Shanahan Osteo/Chondrocytic Transcription Factors and Their Target Genes Exhibit Distinct Patterns of Expression in Human Arterial Calcification Arterioscler. Thromb. Vasc. Biol., March 1, 2003; 23(3): 489 - 494. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |