|Home | About | Journals | Submit | Contact Us | Français|
Vascular calcification is associated with significant cardiovascular morbidity and mortality in patients with chronic kidney disease (CKD). Factors unique to CKD patients, such as hyperphosphatemia, predispose these patients to early and progressive vascular calcification. Hyperphosphatemia appears to be involved in a number of mechanisms that trigger and advance progression of vascular calcification including (1) transition of vascular smooth muscle cells (VSMC) from a contractile to an osteochondrogenic phenotype and mineralization of VSMC matrix through sodium-dependent phosphate cotransporters; (2) induction of apoptosis of VSMC; (3) inhibition of monocyte/macrophage differentiation into osteoclast-like cells; (4) elevation of fibroblast growth factor 23 levels; and (5) decreases in klotho expression. Whether vascular calcification can be prevented or reversed with strategies aimed at maintaining phosphate homeostasis is currently unknown. The current review discusses these mechanisms in-depth, exploring the interplay among vascular calcification promoters, inhibitors and substrate that affect phosphorus handling leading to vascular calcification in individuals with CKD.
Cardiovascular disease (CVD) is the most common cause of mortality in patients with chronic kidney disease (CKD) and may be due in part to excess vascular calcification. In CKD patients, the presence and extent of arterial calcification independently predicts CVD and mortality.1 The hallmark of vascular calcification is calcium phosphate deposition, which can occur in the blood vessels, myocardium, and cardiac valves.2 Calcification occurs at two distinct sites, the intimal and medial layers of the vasculature. Intimal deposition is associated with atherosclerotic plaques, while medial deposition is associated with vascular stiffening and arteriosclerosis.3,4 Although both forms of calcification occur in patients with CKD, medial calcification appears to be more common.2 Vascular calcification is associated with adverse outcomes including myocardial infarction, stroke, and all-cause and CVD mortality in both the general and CKD population.5–8 However, vascular calcification occurs decades earlier in CKD patients than in the general population.9 A major area of interest concerns the reasons behind the development and accelerated progression of this abnormal calcification in patients with kidney disease.
Calcification is a highly regulated process. Under normal conditions, cell-mediated processes control inducers and inhibitors responsible for mineralization, and the process of vascular calcification is avoided;2 however, an imbalance of these regulators results in the initiation of calcification. The key factors responsible for calcification appear to differ across disease states. In the general population, the development of vascular calcification is mainly associated with age and atherosclerotic risk factors, while in the CKD population, vascular calcification is associated with additional non-traditional factors that may be unique to CKD and thus predispose these patients to early and more accelerated calcification. Traditional and non-traditional risk factors for CVD and vascular calcification in patients with kidney disease are shown in Boxes 1 and and2,2, respectively.10 One such factor is serum phosphorus, which has been linked to vascular calcification in several studies and is emerging as a key regulator of calcification in the CKD population.9,11,12 This review focuses on the potential mechanisms by which phosphorus may trigger and/or advance progression of accelerated vascular calcification.
The apparent participation of serum phosphorus in vascular calcification stems from studies of patients with kidney disease and work detailing genetic syndromes that result in hyperphosphatemia. Mutations in the genes for fibroblast growth factor 23 (FGF-23) and uridine diphosphate (UDP) N-acetyl-α-D-galactosamine result in familial tumor calcinosis, which is characterized by elevated serum phosphorus and vascular calcifications.13,14 This syndrome of hyperphosphatemia and vascular calcification has also been observed in mice with a targeted deletion of the FGF-23 gene15 or klotho16, a gene required for FGF-23 function. Hyperphosphatemia is also associated with prevalence and progression of vascular calcification in dialysis patients.9 Similar findings were reported by Raggi et al;12 in a study of 205 adult patients on maintenance dialysis, they noted that the extent of coronary artery calcification was more pronounced in the patients with higher serum phosphorus levels. The association between higher serum phosphorus levels and vascular calcification has also been reported in patients with CKD stage 3–4. In 439 participants from the Multi-Ethnic Study of Atherosclerosis (MESA) with moderate CKD (mean estimated GFR of 51 ml/min/1.73m2) and no clinical CVD, each 1-mg/dL increase in serum phosphorus was associated with a 21% (p=0.002), 33% (p=0.001), 25% (p=0.16) and 62% (p=0.007) greater prevalence of coronary artery, thoracic, aortic and mitral valve calcification, respectively, after adjusting for demographics and estimated GFR.17 Recently, serum phosphorus levels have also been linked to vascular calcification in the general population. In 3015 individuals from the CARDIA (Coronary Artery Risk Development in Young Adults) study18, participants with serum phosphorus levels greater than 3.9 mg/dL had a 52% greater risk of coronary artery calcification 15 years later versus participants with a serum phosphorus level less than 3.3 mg/dL.
Interestingly, higher serum phosphorus level is also linked to vascular stiffness in patients with and without CKD. In patients with moderate CKD, a serum phosphorus level greater than 4.0 mg/dL is associated with a 4-fold greater risk of high ankle-brachial index, a hallmark of peripheral arterial stiffness, compared to a phosphorus level less than 3.0 mg/dL.19 In a study of over 500 participants with normal kidney function from NHANES (the National Health and Nutrition Examination Survey), individuals with serum phosphorus levels of 3.7–5.0 mg/dL had almost a five-fold increased risk of high ankle-brachial index compared to those with phosphorus levels of 3.1–3.4 mg/dL.20 Thus, even serum phosphorus levels within the normal laboratory range are associated with vascular calcification and stiffness, suggesting that phosphate load and/or abnormal regulation of serum phosphorus and not just overt hyperphosphatemia is a key mediator of vascular calcification. Furthermore, higher serum phosphorus concentrations are associated with death and cardiovascular events in patients with and without kidney disease.21–25
If phosphorus is a major player in vascular calcification, then methods to reduce serum phosphorus levels may prevent or at least slow progression of calcification. In uremic rats fed high-phosphate diets, treatment with sevelamer or calcium carbonate decreased myocardial and hepatic calcification.26 In dialysis patients, two randomized trials have shown that reducing serum phosphorus levels with sevelamer slows the progression of vascular calcification.27,28 Similar findings were reported in a trial involving patients with stage 3–4 CKD.29 However, two other recent randomized trials did not find any decrease in vascular calcification with the use of sevelamer or calcium-based binders.30,31 Thus, additional large randomized trials using agents that reduce serum phosphorus levels are needed to prove a causal relationship. To date, no randomized trials have compared phosphate binding agents with placebo in order to evaluate whether targeting phosphate metabolism reduces or prevents vascular calcification and/or cardiovascular events.
The many factors hypothesized to play a role in vascular calcification are summarized in Table 1; those that influence phosphorus handling are discussed in this review. There are multiple mechanisms by which abnormal phosphorus handling may contribute to the initiation and/or progression of vascular calcification (Fig 1).
To determine whether phosphorus directly affects vascular calcification, several in vitro studies have been done using human vascular smooth muscle cells (VSMC). When VSMC are exposed to high levels of inorganic phosphate (> 2.4 mM), consistent with levels seen with hyperphosphatemia, calcification is induced in the extracellular matrix surrounding the VSMC.32 This calcification has features similar to calcification that occurs in bone, including matrix vesicles and bioapatite.33 Furthermore, phosphate directly induces phenotypic changes in VSMC, causing them to transform from a contractile phenotype into an osteochondrogenic phenotype. When VSMC are exposed to elevated phosphate concentrations in vitro, there is increased gene transcription of messenger RNAs encoding proteins involved in matrix mineralization and bone formation, such as osteocalcin and CBFA1/RUNX2 (core-binding factor subunit 1α/runt-related transcription factor 2) and simultaneous down-regulation of transcription factors for smooth muscle cells.34,35 Notably, these changes also occur in human and animal models of calcification. For example, calcified inferior epigastric arteries from dialysis patients express CBFA1/RUNX2 and osteopontin in both the media and intima layers.36 Giachelli et al32 has hypothesized that this phenotypic change in VSMC may serve to repair or adapt to a mineralizing environment, given the increased expression of several mineral-regulating molecules in the VSMC. The mechanisms responsible for controlling this phosphate-induced phenotypic change are currently unknown. However, it is known that this phosphate-induced phenotypic change in VSMC is dependent on the activity of sodium-dependent phosphate cotransporters (Na/Pi cotransporters). Type III Na/Pi cotransporters, specifically PiT-1, have been characterized in VSMC.2In vitro and animal studies demonstrate that phosphate and the PiT-1 cotransporter on VSMC play key roles in controlling vascular calcification. In vitro treatment of VSMC with an inhibitor of the Na/Pi cotransporter results in a decrease in phosphate uptake, calcification, and osteochondrogenic phenotype.35,37,38 Furthermore, the expression of osteogenic differentiation markers CBFA1/RUNX2 and osteopontin are blocked in VSMC that have been transduced with PiT-1 knockdown cells in vitro.39 Interestingly, different disease states and cytokines appear to increase the expression of the Na/Pi cotransporter.2 Mizobuchi et al.40 showed in vivo that PiT-1 mRNA levels are elevated in calcified aortas of rats with kidney disease and severe secondary hyperparathyroidism, whereas no calcification was found in aortas of control animals. Additionally, PDGF (platelet-derived growth factor) induces PiT-1 expression and calcification in cultured VSMC.41 This increased expression of the Na/Pi cotransporter in response to disease states may result in an increased susceptibility to calcification even at serum phosphorus levels within the normal laboratory range. It is currently unknown whether Na/Pi cotransporters play a role in vascular calcification in human disease but this data demonstrates that PiT-1 phosphate transport is required for calcification of human VSMC in vitro.
Studies have found a link between apoptosis of VSMC and vascular calcification; accordingly, it has been suggested that apoptosis is a key regulator of VSMC calcification.42–44 Matrix vesicles appear to play a role in apoptosis-induced vascular calcification. Matrix vesicles are generated by budding from chondrocytes and osteoblasts and hold within all of the factors needed for initiation of calcification. Matrix vesicles or at least structures very similar to matrix vesicles have been identified in VSMC calcification.33 The matrix vesicles derived from VSMC may be remnants of apoptotic cells. For example, in advanced carotid artery atherosclerotic plaques, Kockx et al.43 found that the matrix vesicles were derived from VSMC and they contained the proapoptotic protein BAX (BCL2-associated X protein). Furthermore, studies have suggested that cell death leads to the production of matrix vesicles.42,45 Thus, it is these matrix vesicles/apoptotic bodies that initiate calcification because they have the capacity to concentrate and crystallize calcium.44 Evidence supports the role of apoptosis in calcification, with studies finding that apoptosis in VSMC occurs before the onset of calcification.44 Furthermore, when apoptosis is inhibited in VSMC by a caspase inhibitor there is a significant reduction in calcification.44 However, Giachelli et al32 found that calcification of VSMC does not appear to require apoptosis for the initiation of calcification; nonetheless, apoptosis appears to accelerate the calcification process.
Phosphate has been shown to induce apoptosis in terminally differentiated chondrocytes. The role of phosphate in apoptosis appears to be associated with chondrocyte maturation and extracellular matrix mineralization.46 Exposure to elevated phosphorus concentration causes an increase in intracellular phosphate which triggers chondrocyte apoptosis.47 It appears that Na/Pi cotransporters play a role in the increase in intracellular phosphate as competitive inhibitors of the Na/Pi transporter block apoptosis in chondrocytes.47 How increased phosphorus concentrations results in apoptosis is unclear, but this process may be related to disruptions in normal mitochondrial energy metabolism.47 Phosphate has also been shown to induce apoptosis in human aortic smooth muscle cells.48 When human aortic smooth muscle cells were exposed to high phosphate (similar to levels seen with hyperphosphatemia), calcification and apoptosis were induced in a dose and time dependent manner.48 Interestingly, the use of HMG-CoA (β-hydroxy-β-methylglutaryl-CoA) reductase inhibitors (statins) blocks phosphate-induced calcification in human aortic smooth muscle cells.48 The inhibitory effect of statins reflected prevention of apoptosis rather than inhibition of Na/Pi cotransporters. This may be the reason for the decrease in vascular calcification seen with the use of statins in clinical trials.49,50 More studies are needed to confirm that phosphate-induced apoptosis plays a role in vascular calcification in humans.
In bone formation there is a delicate balance between mineral deposition by osteoblasts and mineral resorption by osteoclasts. Disruptions in this balance lead to disease states such as osteoporosis. It has been hypothesized that a similar balance between mineral deposition and resorption exists in the vasculature and disruptions result in calcification of the vessel walls.51 As discussed earlier, the calcified vascular wall contains VSMC that have an osteoblast-like phenotype. However, the calcified wall also contains monocytes and macrophages that are able to differentiate into cells with an osteoclast-like phenotype.52 It is thought that these osteoclast-like cells may be important regulators in arterial homeostasis and extracellular mineralization.52 Thus, an imbalance that favors the osteoblast-like phenotype may both initiate and cause progression of calcification. Recent studies have found that uremic toxins favor the osteoblast-like phenotype.51 Elevated phosphorus concentrations to levels seen in advanced kidney disease significantly decrease the differentiation of monocytes/macrophages into osteoclast-like cells in vitro by downregulating RANK (receptor activator of nuclear factor –κB) ligand-induced signaling.53 This reduction in osteoclast activity may be an additional way that elevated phosphorus levels results in vascular calcification. Additional studies are needed to evaluate whether activating these osteoclast-like cells in the vessel wall prevents or reverses vascular calcification.
FGF-23 is a hormone secreted by osteocytes. It is the main regulator of serum phosphorus and plays a key role in vitamin D metabolism and secondary hyperparathyroidism.54 FGF-23 results in renal phosphate excretion through inhibition of proximal tubule renal phosphate reabsorption and also inhibits 1-α- hydroxylase activity, resulting in decreasing calcitriol and decreased phosphate reabsorption in the intestine.54 FGF-23 weakly binds its receptor (FGFR1c) and thus requires a co-factor, klotho, for activity. Given the requirement for klotho, FGF-23 is presumed to be active only in tissues that manufacture klotho (distal tubule of kidney, parathyroid glands, brain). FGF-23 levels increase early in the course of kidney disease to maintain serum phosphorus within the normal range, which is why phosphorus levels do not become noticeably elevated until the glomerular filtration rate falls to < 20 ml/min/1.73 m2. In addition, FGF-23 suppresses parathyroid hormone (PTH) secretion, PTH gene expression, and parathyroid cell proliferation, thereby closing the loop in the bone-parathyroid axis.55,56 In advanced CKD there is down-regulation of the parathyroid klotho-FGFR1c, and FGF-23 fails to decrease PTH expression and parathyroid cell proliferation.56 Interestingly, FGF-23 is stimulated not only by increased phosphate but also by PTH,55 and recent data have demonstrated that FGF-23/klotho signaling is not necessary for the phosphaturic and anabolic actions of PTH.57 Thus, PTH still exerts its functions even in the presence of extremely high serum FGF-23 levels.
Several recent studies have linked higher FGF-23 levels with death and cardiovascular events in patients with and without kidney disease, although the mechanism for this relationship remains unclear.58–62 In moderately uremic mice fed high-phosphate diets, serum FGF-23 levels rather than serum phosphorus levels were associated with extensive arterial-medial calcifications.63 In dialysis patients, FGF-23 has been independently linked to coronary artery, aortic and peripheral vascular calcification.64–66 These data suggest that another mechanism by which phosphate affects vascular calcification may be via phosphorus-mediated elevation of FGF-23.
It is unclear if elevated FGF-23 levels directly results in vascular calcification or if FGF-23 inhibits calcification, as patients with inactivating mutations of FGF-23 have an increase rather than a decrease in vascular calcifications. Furthermore, a recent study in dialysis patients evaluating changes in aortic arch calcification scores found that FGF-23 levels were higher in regressors than in non-progressors and progressors.67 FGF-23 was also negatively associated with changes in aortic arch calcification scores during the 5 year follow-up period (β value = −0.001, p=0.0115).67 Thus, extremely high FGF-23 levels may result in inhibition of calcification in the vessel wall in dialysis patients. The mechanism by which FGF-23 effects vascular calcification is unknown, but it is possible that FGF-23 has both direct and indirect actions on vascular calcification. It is unknown if Klotho is found on VSMC but VSMC do contain FGF receptors.61 Even though FGF-23 typically requires Klotho for its activity, it has been hypothesized that the extremely high levels of FGF-23 seen in dialysis patients may induce Klotho-independent functions.66 Further large, prospective studies are needed to specify whether FGF-23 may be a marker or a potential mechanism for vascular calcification.
Klotho is a single-pass transmembrane protein that functions as a co-receptor for FGF-23.54 Klotho is also secreted into the blood, urine and cerebrospinal fluid.68 CKD appears to be a state of Klotho deficiency, and renal Klotho expression is decreased in human nephrectomy samples from end-stage kidneys.69 Furthermore, secreted Klotho is also decreased in the blood and urine in people with CKD.68 In humans, the decrease in urine Klotho becomes apparent in early stages of CKD and continues as kidney disease progresses.68
Klotho deficiency in mice results in hyperphosphatemia and ectopic calcification.16 Hu et al68 showed in an experimental model of CKD that transgenic mice overexpressing Klotho have preserved Klotho levels, better kidney function, enhanced phosphaturia and less ectopic calcification than wild-type mice with CKD. Thus, data suggests that Klotho may be an inhibitor of vascular calcification. The anticalcification effect of Klotho is likely due to both direct and indirect actions of Klotho. Indirectly, Klotho protects against calcification through its actions on phosphorus and vitamin D metabolism. One of the key roles of Klotho is to promote renal phosphate excretion through its interaction with FGF-23. This control of serum phosphorus results in reduced calcification. Overexpression of Klotho also decreases kidney disease progression.68 Preservation of kidney function results in less disturbances in mineral metabolism and other risk factors for vascular calcification. Recent data by Hu et al.68 further suggest that Klotho has a direct effect on the vasculature. In an experimental model of CKD, animals lacking Klotho had increased expression of the Na/Pi cotransporter and the osteogenic transcription factor CBFA1/RUNX2 in VSMC, suggesting that decreased Klotho drives calcification.68 Furthermore, when Klotho is added to VSMC in vitro, it reduces phosphate uptake by suppressing the activity of the Na/Pi cotransporter and prevents the phenotypic change of VSMC to an osteochondrogenic phenotype. It is unclear how Klotho mediates its effects on VSMC. Further studies are needed to confirm the role of Klotho in vascular calcification.
Vascular calcification is associated with significant morbidity and mortality in patients with CKD. An elevated serum phosphorus level is a risk factor for death, cardiovascular events and vascular calcification in CKD patients. Hyperphosphatemia contributes to several mechanisms that trigger or advance the progression of vascular calcification and is emerging as a key regulator of calcification in patients with kidney disease. The question remains as to whether strategies that reduce serum phosphorus prevent or reverse vascular calcification.
Support: This work is supported by National Institute of Diabetes and Digestive and Kidney Disease grants R01 DK081473 and R01 DK078112 and K23DK087859-01A1.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Financial Disclosure: The authors declare that they have no relevant financial interests.