• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Kidney Int. Author manuscript; available in PMC Dec 29, 2009.
Published in final edited form as:
PMCID: PMC2799244
NIHMSID: NIHMS164945

Phosphate feeding induces arterial medial calcification in uremic mice: role of serum phosphorus, FGF-23 and osteopontin

Abstract

Arterial medial calcification (AMC) is a hallmark of vascular disease in chronic kidney disease patients, and is a strong prognostic marker for cardiovascular and all-cause mortality. This study was aimed at determining the role of phosphate feeding and severity of uremia on AMC in a high phosphate fed (HP) mouse model of renal insufficiency. Severe (SU) and moderate (MU) uremia was achieved by varying the degree of renal ablation in calcification-prone DBA/2 mice. SU/HP and MU/HP mice developed extensive AMC whereas normal phosphate fed (NP) uremic mice and sham controls did not. AMC in the SU/HP mice was associated with statistically significant hyperphosphatemia. In contrast, MU/HP mice were not hyperphosphatemic, but a significant rise in serum fibroblast growth factor 23 (FGF-23) and osteopontin (OPN) levels in this group was observed. Further, FGF-23 and OPN were significantly correlated with AMC. Histochemistry revealed widespread AMC with no evidence of atherosclerotic disease. At early stages of calcification, upregulation of osteochondrogenic markers, Runx2 and OPN, and downregulation of smooth muscle cell (SMC) marker, SM22α, in medial SMCs was observed. In extensively calcified regions, medial SMC drop out was evident. These studies support important roles for phosphate loading and degree of renal insufficiency in mediating AMC in mice, and suggest potential roles for FGF-23 and OPN as markers and/or inducers of AMC.

INTRODUCTION

Chronic kidney disease (CKD) is a worldwide public health problem with high cost, poor outcome and increasing prevalence. Cardiovascular disease is the leading cause of death among end stage renal disease (ESRD) patients [1] and their risk of cardiovascular mortality is 10 - 100 times greater than in healthy individuals [2]. Vascular calcification, the pathological deposition of calcium phosphate salts in cardiovascular tissues, is highly correlated with cardiovascular mortality in both pediatric and adult CKD patients [3-5]. Vascular calcification can occur in the arterial intimal layer in association with atherosclerosis, or in the arterial medial layer independent of atherosclerotic disease.

Arterial medial calcification (AMC), is a hallmark of CKD and has emerged as a strong prognostic marker for all-cause mortality in dialysis patients [4, 6]. At the artery level, AMC results in increased wall stiffness, decreased compliance, and increased pulse wave velocity and systolic pressure. Over time, these altered mechanical and hemodynamic properties lead to left ventricular hypertrophy, decreased coronary perfusion and heart failure [3, 4]. At the arteriole level, AMC causes the ischemic skin lesions characteristic of calcific uremic arteriolopathy (CUA), a condition almost exclusively observed in dialysis patients and associated with an extremely high mortality rate [7, 8].

The mechanisms regulating vascular calcification are still under investigation. However, it is currently accepted that the process is complex, involving the active participation of a number of calcification inducers and inhibitors [9]. The interaction between the regulators, as well as their roles in the initiation and/or progression of the calcification process, is still unknown. Phosphate has emerged as a major promoter of vascular calcification and is increasingly adopting a central role in the calcification process. Hyperphosphatemia is a persistent, prevalent problem in ESRD patients, and epidemiological studies have revealed that elevated serum phosphorus is a strong predictor of morbidity and cardiovascular mortality in dialysis patients [10, 11]. Also, growing evidence indicates that hyperphosphatemia is correlated with calcification of the coronaries, peripheral arteries and cardiac valves as well as with CUA in ESRD patients [12-14]. Effective control of hyperphosphatemia without increasing total calcium load has been correlated with attenuated progression of vascular calcification in prevalent dialysis patients [15, 16], and decreased mortality in incident dialysis patients [17]. Moreover, even relatively small elevations in serum phosphate in the high normal range (3.5-4.5 mg /dl) have been correlated with increased risk of cardiovascular and all cause mortality in both CKD patients [11] and nonuremic patients with coronary disease [18]. Thus, it is important to determine whether phosphate load, even in the absence of outright hyperphosphatemia, is an important driver of vascular calcification.

To date, mechanistic and genetic studies of uremia-induced AMC have been hampered by lack of a robust mouse model in which AMC occurs in the absence of atherosclerotic disease, but nevertheless mimics the extensive AMC observed in patients with ESRD [19]. In this study, we describe a uremic mouse model of extensive, widespread AMC that was not associated with atherosclerotic disease, and was dependent on the degree of renal insufficiency and phosphate feeding. AMC was associated with early osteochondrogenic phenotype change and late medial cell loss, and correlated with serum FGF-23 and OPN levels.

RESULTS

AMC and serum biochemistry in severely uremic, high phosphate fed mice

To determine the effect of renal insufficiency and phosphate loading on AMC, severe uremia was induced in mice using a two-step severe renal ablation surgery, and mice were fed a normal or high phosphate diet as described in the Materials and Methods. Figure 1 shows the aortic calcium content in the severely uremic groups and their sham controls at termination. High levels of aortic calcium were detected in the SU/HP group (18.2 ± 7.3 μg/mg dry wt). In contrast, very little aortic calcium was detected in the SU/NP group (1.1 ± 0.3 μg/mg dry wt) and this was not significantly different from sham controls (1.0 ± 0.2 μg/mg for the Sham/NP and 1.0 ± 0.3 μg/mg for the Sham/HP). These data suggest that uremia alone, in the absence of a dietary phosphate load, was not sufficient to induce AMC in these mice.

Figure 1
Aortic calcium content in severely uremic mice and sham controls under normal and high phosphate feeding conditions. Dotted lines indicate mean values for each group. (P<0.05, ANOVA); Sham/NP, n=5; Sham/HP, n=6; SU/NP, n=4; SU/HP, n=7.

Mineralization of the arteries was visualized in whole mounts and cross sections by Alizarin Red staining for calcium. Mineral deposits in the SU/HP mice were extensive and widespread and included all portions of the arterial tree (Fig 2A). The aortic arch, upper and lower thoracic aorta, and abdominal aorta were consistently affected. The mid-portion of the thoracic aorta was a site that was consistently spared of calcification. In addition, smaller arteries, such as the iliac, renal and mesenteric vessels and their derivatives, appeared to be more severely calcified compared to the large, central arteries, suggesting that these smaller diameter peripheral arteries may be highly susceptible to calcification (Fig 2A, and data not shown). Representative Alizarin Red stained cross sections for the abdominal aorta, carotid, iliac, and mesenteric arteries are shown in Fig 2 B-E. In these vessels, calcification was most often circumferential and was restricted to the tunica media with no evidence of endothelial or adventitial involvement (Fig 2 F, G). In agreement with aortic calcium quantitation (Fig 1), no calcification was histologically detectable in the Sham/NP, Sham/HP, or SU/NP groups (Fig 2 J-M). In addition to vessels, we observed that some other soft tissues were also calcified in this model, including myocardium and lungs (data not shown).

Figure 2
Histological analyses of AMC in uremic mouse arteries. A, Alizarin Red staining of vascular tree whole mount. B-M, Representative micrographs of Alizarin Red stained paraffin sections of arteries in the different study groups. B, abdominal aorta; C, carotid; ...

As shown in Table 1, mice with severe uremia had BUN values about three fold higher than nonuremic, sham controls (70 ± 4 and 62 ± 4 mg/dl for SU/NP and SU/HP, respectively versus 22 ± 1 and 21 ± 1 mg/dl for Sham/NP and Sham/HP, respectively). However, as shown in Fig 3A, only the mice in the SU/HP group developed significant hyperphosphatemia. As shown in Fig 3B, while a statistically significant increase in serum calcium levels was observed in the SU/NP group compared to all other groups, these mice did not develop aortic calcification (Fig 1) and linear regression analysis revealed no correlation between serum calcium levels and aortic calcification (R=0.28). Calcium X phosphorus products (Ca X P) in each group (Fig 3C) mirrored the trends observed with serum phosphorus (Fig 3A) revealing a significant increase in the product only in the SU/HP group. Furthermore, linear regression analysis revealed only a weak correlation between serum phosphorus levels and aortic calcification in SU mice (R= 0.61; p<0.05) as shown in Fig 3D.

Figure 3Figure 3Figure 3Figure 3
Serum phosphorus (A), calcium (B), and calcium X phosphorus (C) linear regression between serum phosphorus and aortic calcium (D) in severely uremic mice and sham controls under normal and high phosphate feeding conditions. Data represent the mean ± ...
Table 1
Serum biochemical data and body weights in the severely uremic study.

A statistically significant increase in serum PTH levels was present in SU/NP and SU/HP mice compared to sham controls (Table 1). Consistent with previous studies in uremic mice fed a normal phosphate diet [20], serum ALP levels were elevated in the SU/NP as well as in SU/HP groups compared to sham controls (Table 1). These data suggest increased bone remodeling in the uremic mice, and indeed, in other studies we have observed increased cortical porosity and increased % trabecular bone volume (BV/TV) in uremic, high phosphate fed mice compared to sham controls (data not shown). However, no arterial calcification was observed in the SU/NP group despite elevated PTH and ALP.

AMC and serum biochemistry in moderately uremic, high phosphate fed mice

To investigate the effects of the degree of renal insufficiency on AMC, mice were made moderately uremic by less severe renal ablation, and fed a high phosphate diet as described in the Materials and Methods. Figure 4 shows the aortic calcium content in the moderately uremic groups and sham controls at termination. A significant increase in aortic calcium content was detected in the MU/HP group (5.9 ± 2.7 μg/mg dry wt) compared to the MU/NP group (0.7 ± 0.09 μg/mg dry wt) and the sham controls (0.7 ± 0.1 μg/mg for the Sham/NP and 0.9 ± 0.1 μg/mg for the Sham/HP). Furthermore, a three fold decrease in aortic calcification was observed in the MU/HP group (5.9 ± 2.7 μg/mg dry wt; Fig 4) compared to the SU/HP group (18.2 ± 7.2 μg/mg dry wt; Fig 1) even though the MU/HP group were treated with the HP diet for at least 9 weeks longer than the SU/HP mice.

Figure 4
Aortic calcium content in moderately uremic mice and sham controls under normal and high phosphate feeding conditions. Dotted lines indicate mean values for each group. (P<0.05, ANOVA) ; Sham/NP, n=7; Sham/HP, n=7; MU/NP, n=10; MU/HP, n=10.

Histologically, AMC was also widespread in the MU/HP mice, but notably less severe than in the SU/HP mice. As shown for the aorta, foci of mineralization were randomly scattered along the medial layer, associated with either the elastic lamellae (Fig.2, K) or the extracellular matrix between lamellae (Fig 2, L). Of interest, fragmentation of the elastic lamella was noted in several sections (data not shown). Again, no mineralization was observed in the adventitia or endothelial layers.

As shown in Table 2, mice in the MU/NP and MU/HP groups had BUN values of 44 ± 1 and 45 ± 2 mg/dl respectively, while sham controls had BUN values of 30 ± 2 and 22 ± 1 mg/dl for Sham/NP and Sham/HP, respectively. In contrast with findings in severely uremic mice, Fig 5A showed that there was no increase in serum phosphorus levels in the MU/HP mice (8.3 ± 0.4 mg/dl) compared to the sham controls (7.7 ± 0.5 and 7.9 ± 0.4 mg/dl for Sham/NP and Sham/HP). This was despite the presence of a significant increase in AMC in the MU/HP mice compared to the other groups (Fig. 4), indicating that AMC could occur in the absence of hyperphosphatemia in moderately uremic mice. Interestingly, we observed a significant decrease in serum phosphorus levels concurrent with a significant rise in serum calcium levels and a significant drop in serum PTH levels (Fig 5A, B and Table 2) in the MU/NP group compared to sham controls. These observations are consistent with the possibility that the rise in serum calcium levels triggered a feedback downregulation of PTH followed by a decrease in phosphorus (and calcium) uptake in the intestines. Noteworthy is that linear regression analysis revealed no correlation between either serum PTH (R= 0.06),or calcium (R= 0.06) and aortic calcification in the moderately uremic mice. As shown in Fig 5C, Ca X P level were not significantly elevated in MU/HP compared to sham controls, but were decreased in the MU/NP group mirroring the calcium results (Fig 5B). Interestingly, as observed for severely uremic mice, there was little correlation between aortic calcium content and serum phosphate levels (R=0.464), p<0.05 (Fig 5D) in the MU (HP/NP) mice.

Figure 5Figure 5Figure 5Figure 5
Serum phosphorus (A), calcium (B), and calcium X phosphorus (C) linear regression between serum phosphorus and aortic calcium (D) in moderately uremic mice and sham controls under normal and high phosphate feeding conditions. Data represent the mean ± ...
Table 2
Serum biochemical data and body weights in the moderately uremic study

In an attempt to understand the mechanism for the lack of hyperphosphatemia in the MU/HP group, despite the presence of the dietary phosphate load and the clear induction of AMC, we analyzed serum FGF-23 levels in these mice. FGF-23 has been identified recently as important regulator of phosphate homeostasis. As shown in Figure 6A, phosphate feeding alone was not sufficient to induce fasting serum FGF-23 in sham controls. On the other hand, a dramatic, statistically significant increase in fasting serum FGF-23 levels was observed in the MU/HP mice compared to all other groups (Figure 6A). Furthermore, as shown in figure 6B, linear regression analysis revealed a very strong correlation between serum FGF-23 levels and aortic calcification (R=0.901; P<0.0001) in MU (HP/NP) and Sham (HP/NP) groups. Interestingly, when aortic calcium content of only uremic groups (MU: HP/NP) were correlated with serum FGF-23, the R value (0.905) still remained significant (p<0.05).

Figure 6Figure 6Figure 6Figure 6Figure 6
Serum FGF-23 (A) correlation between serum FGF-23 and aortic calcium (B) serum osteopontin (C) correlation between serum OPN and aortic calcium (D) correlation between serum OPN and FGF-23 (E) levels in moderately uremic mice and sham controls under normal ...

To determine whether elevated FGF-23 levels led to increased excretion of phosphate in MU/HP mice and could thus help explain normal serum phosphorus in these mice despite a phosphate feeding, we performed the following experiment. Fasting serum and spot urines were collected from sham mice fed either a high or low phosphate (normal serum phosphate and FGF-23 conditions) and moderately uremic (MU) mice fed the high phosphate for 12 weeks (MU/HP; BUN=59.0 ± 2.5 mg/dL; serum phosphate=10.9 ± 0.6 mg/dL; serum FGF-23=820 ±- 125 pg/ml). We then used serum phosphate and creatinine levels and urine phosphate and creatinine levels to calculate the fractional excretion of phosphate (FEP) in these mice. We found that FEP was somewhat lower in sham mice fed the normal phosphate diet vs the high phosphate diet, but this was not statistically significant. (0.06 ± 0.01 vs 0.26 ± 0.13; p=0.3). On the other hand, FEP was significantly higher in MU/HP mice compared to either the sham mice fed a normal phosphate diet (0.55 ± 0.04 vs 0.06 ± 0.01; p=.0002), or to the sham mice fed a high phosphate diet (0.55 ± 0.04 vs 0.26 ± 0.13; p=0.06). These data are consistent with the elevated FGF-23 levels in MU/HP mice leading to increased fractional excretion of phosphate in the uremic mice, thus accounting for maintenance of serum phosphate levels despite high phosphate feeding. Similar observations in non-uremic transgenic mice chronically overexpressing FGF-23 have been observed [21].

Finally, we examined levels of the circulating calcification inhibitor, OPN, to determine whether it might play a role in AMC in this mouse model. As shown in Figure 6C, a striking statistically significant increase in serum OPN levels was observed in the MU/HP group compared to MU/NP, Sham/HP and Sham/NP groups. Furthermore, as shown in figure 6D, regression analysis revealed a strong positive correlation between serum OPN levels and aortic calcification (R= 0.925; P< 0.0001) in MU and sham groups. As in case of FGF-23, this correlation did not change significantly (R = 0.923, p<0.05) even when shams were excluded from the correlation calculations suggesting that the correlation between FGF-23 and aortic calcium content was independent of the uremic state of the mice. Interestingly, for the first time, a positive correlation was also observed between serum FGF-23 and OPN levels (R= 0.95; P< 0.0001) as shown in Fig 6E. This likely reflects the observation that FGF-23 and OPN are both phosphate sensitive genes that are likely to be co-regulated by phosphate loading [22, 23].

Immunohistochemisty of Calcified Arteries

In order to gain insights into the etiology of AMC in uremic, high phosphate fed mice, we analyzed arteries for immunochemical evidence of two potential mechanisms of vascular calcification: osteochondrogenic phenotype change and inflammation. Osteochondrogenic phenotype change was assessed by examining the expression of smooth muscle and osteochondrogenic lineage markers. Fig 7A shows a representative focus of early mineralization along the elastic lamella in the aorta of a uremic, high phosphate-fed mouse, as indicated by Alizarin Red staining (Fig 7A, arrow). Upregulation of the osteochondrogenic marker, OPN, was detected both in association with the mineralized elastic lamina (Fig 7B, arrowhead), as well as in medial smooth muscle cells (Fig 7B, arrows). In addition, downregulation of the smooth muscle cell lineage marker, SM22α, was observed in the cells surrounding the calcified area (Fig 7C). The loss of SM22α expression was not due to cellular deficiency, as counterstaining with methyl green (Fig 7C) and adjacent H&E stained sections (Fig 7D) demonstrated nuclei in the calcified area. On the other hand, adjacent aortic sections (Fig 4 E-H) from a noncalcified uremic mouse fed a normal phosphate diet had no Alizarin Red positive staining (Fig 7E), no OPN expression (Fig 7F), strong SM22α expression (Fig 7G), and normal medial structure as determined by H& E staining (Fig 7H). Interestingly, strong Runx2/cbfa1 staining was detected in the nuclei of medial smooth muscle cells in uremic, high phosphate fed mice prior to mineralization, consistent with its possible role as an initiating event in SMC phenotype change and calcification (Fig 7I). In contrast, no Runx2/cbfa1 staining was detected in the noncalcified artery in Fig 7J. No evidence of SMC phenotype change was observed in any of the SU/NP, MU/NP, Sham/NP and Sham/HP mouse arteries analyzed (data not shown). Finally, no collagen II staining (indicative of chondrogenic differentiation) was detected in any specimens examined (data not shown).

Figure 7
Histochemical staining of mildly calcified and noncalcified arteries. A - D are adjacent sections capturing an early mineralization focus in the aorta of a uremic, high phosphate fed mouse. A, Alizarin Red positive foci of elastin calcification (arrowhead); ...

While early foci of calcification were associated with an osteogenic phenotype change, more heavily calcified regions, presumably representing later stages of the calcification process, showed evidence of medial cell loss. Figure 8 shows representative stained sections of a severely calcified aorta from a SU/HP mouse. H&E staining demonstrated extensive, circumferential calcification as indicated by basophilic staining in the media (Fig 8A, C) and Alizarin Red staining (Fig 8B). SM22 staining was absent in the calcified areas (Fig 8D), but in contrast to early calcification sites where loss of SM22 staining occurred due to phenotypic change (Fig 7), loss of SM22 in the highly calcified specimens appeared to be due to a deficit in medial cells in these regions. This was as evidenced by lack of nuclear staining in adjacent H&E sections (Fig 8A and 8C) and methyl green counterstaining (Fig 8D). Thus, cell loss was evident late in the calcification process.

Figure 8
Immunochemical staining of a severely calcified artery showing evidence of SMC loss. A, H&E staining (scale bar = 200 μm), boxed area is shown at higher magnification in B-D; B, Alizarin Red staining (scale bar = 30μm); C, H&E ...

To determine whether inflammation contributed to the calcification process, we investigated the expression of BM8, a macrophage specific surface marker, in representative sections from all mouse groups. We did not observe any macrophages associated with either calcified or non-calcified arterial media in any of the groups (data not shown). Only occasionally, in extensively calcified regions, did we detect a few BM8 positive cells in the underlying adventitia (data not shown), suggesting that inflammation is unlikely to be driving AMC in this uremic mouse model. Furthermore, no intimal or neointimal accumulation of macrophages or any other cell was detected in any of the groups, indicating that atherosclerotic lesion formation was not associated with calcification in any of the vessels (Fig 8A, C, and data not shown).

DISCUSSION

In this study, we describe a uremic mouse model of robust AMC in the absence of atherosclerosis and associated inflammation. Uremic mice fed a normal phosphate diet did not develop AMC, while those fed a high phosphate diet did. Although uremia in the absence of phosphate feeding did not induce vascular calcification, the degree of renal insufficiency was an important modulator of the extent of AMC in phosphate fed mice. In severely uremic mice, high phosphate feeding induced hyperphosphatemia. On the other hand, moderately uremic mice fed a high phosphate diet developed AMC under normophosphatemic conditions. In MU mice, FGF-23 and OPN levels but not serum phosphate levels were highly correlated with AMC. Finally, we observed an osteogenic SMC phenotypic change associated with early stages of calcification, and provide evidence that late stages of calcification were associated with medial SMC loss.

Under severe uremic conditions, hyperphosphatemia was detected and a positive correlation between serum phosphorus concentrations and AMC was revealed in uremic mice. Elevated serum phosphorus levels in association with uremia and ectopic calcification have also been observed in rats [24, 25] as well as LDLR null [26, 27] and apolipoprotein E-deficient mice [25, 27, 28]. These findings agree with the growing clinical evidence that serum phosphorus levels above normal are a risk factor for vascular calcification even at early stages of CKD [13, 14].

On the other hand, the MU/HP group did not show hyperphosphatemia despite having significant AMC. However, it should be noted that while MU/HP mice did not have higher serum phosphorus than non-uremic mice, they did have higher serum phosphorus than the MU/NP mice. While we attempted to avoid post-prandial variations in serum phosphorus levels by fasting the animals, it is possible that time averaged serum phosphorus may be elevated in the MU/HP mice. Nonetheless, our studies are in agreement with clinical data, since hyperphosphatemia typically manifests itself only in the later stages of the renal disease while the risk for vascular calcification starts in the early stage [11, 29]. Of interest, several recent studies have correlated even relatively small elevations in serum phosphate in the high normal range (3.5-4.5 mg /dl) with increased risk of cardiovascular and all cause mortality in chronic kidney disease patients [11, 30]. Together, these data suggest that phosphate burden, even in the absence of overt hyperphosphatemia, plays an important role in vascular calcification that can contribute to cardiovascular and all cause mortality in CKD patients.

A likely explanation for normal serum phosphate levels in the presence of a dietary phosphate burden in the MU/HP mice was the striking elevation of serum FGF-23 levels observed in this group compared to the MU/NP, Sh/NP, and Sh/HP groups. FGF-23 is a hormone produced by osteocytes that acts to suppress phosphate reabsorption in the kidney thus causing increased phosphate excretion [31]. In support of this possibility, we found that the fractional excretion of phosphate (FEP) in the urine was significantly higher in MU/HP mice compared to Sh/NP or Sh/HP mice. Thus, our data support the idea that elevated FGF-23 levels in MU/HP lead to increased FEP in the uremic mice, thus accounting for maintenance of serum phosphate levels despite high phosphate feeding. Similar observations in non-uremic transgenic mice chronically overexpressing FGF-23 have been observed [21]. In those studies, transgenic mice expressing high levels of FGF23 (17,600 RU/ml) had higher FEP compared to wild type mice with undetectable serum FGF23 levels (0.72 vs 0.14; p<.001). As expected, overexpression of FGF-23 led to hypophosphatemia in these non uremic mice fed a chow diet due to chronic high FEP.

Our findings suggest that in early stages of CKD, mimicked by the moderate uremia conditions in our study, serum phosphate levels are largely maintained in the normal range due to successful protection by FGF-23. Interestingly, we detected a strong correlation between serum FGF-23 levels and aortic calcification in uremic mice. On one hand, this finding was unexpected, since FGF-23 deficiency causes hyperphosphatemia and vascular calcification in both humans and mice [21, 32-34]. However, high serum FGF-23 levels are reported in CKD and hemodialysis patients [21, 35, 36], a population typically having extensive cardiovascular calcification [5, 37], and a recent study has shown FGF-23 levels are independently associated with mortality in dialysis patients [38]. Our studies showing that FGF-23 levels are highly correlated with AMC might provide biological plausibility for these important clinical findings.

Our findings also bring up the interesting possibility that FGF-23 might serve as a marker of vascular calcification, which would be particularly useful in the earlier stages of the disease when serum phosphate levels are in the normal range. Alternatively, FGF-23 may play some as yet undefined role in promoting AMC. However, we cannot rule out the possibility that elevated FGF23 levels may simply reflect a reaction to the phosphate loading that the mice were subjected to in the uremic state. More work is needed to distinguish between these possibilities.

Our studies also indicated that the degree of uremia was an important modulator of the extent of AMC in mice. Severe uremic conditions that increased levels of serum PTH (~ 3-4 fold), ALP (~ 3 fold), calcium (~10%), and cholesterol (~ 50% ), and that were comparable to changes in PTH and ALP typically observed in patients with stage 4-5 CKD compared to nonuremic controls [39], were not sufficient to induce AMC in the absence of a dietary phosphate load in the SU/NP group. On the other hand, in the presence of the dietary phosphate load, both moderately and severely uremic mice developed AMC. To our knowledge, this is the first animal model demonstrating that the severity of uremia, though not sufficient to initiate AMC, played a significant role in the acceleration of the calcification process.

These studies are the first to show that serum OPN levels are induced by uremia and high phosphate feeding in mice. OPN is a secreted phosphoprotein that is normally synthesized in bone and kidney [40], but whose expression is highly upregulated in calcified tissues, including arteries [6, 41], and this study, Fig 7. In vitro, OPN levels were induced by elevated phosphate levels in both osteoblasts [42] and SMC [43-45] indicating that OPN is a phosphate-responsive gene. Furthermore, we identified a strong positive correlation between serum OPN and AMC. Since OPN has been shown to be a potent inhibitor of mineralization [46-49], a possible interpretation of this unexpected finding is that serum OPN was induced as a protective mechanism that was overwhelmed in uremic, high phosphate fed mice. Alternatively, since OPN phosphorylation has been shown to be required for its anticalcific activity [47, 50], it is possible that circulating OPN was inactive due to underphosphorylation. Consistent with this possibility, ALP has been shown to dephosphorylate OPN [48], and was elevated in the serum of uremic mice. Unfortunately, there are currently no antibodies available that allow us to distinguish between phosphorylation states of OPN. Lastly, it is possible that serum OPN might promote AMC via its proinflammatory properties [40]. However, this is considered unlikely, as inflammation was not associated with AMC in this model.

Finally, we addressed possible mechanisms involved in uremic AMC in the mouse. Prior to mineralization, we observed elevated Runx2 levels in the media of uremic, high phosphate fed mice compared to nonuremic mice. At the earliest sites of calcium deposition, AMC was associated with increased OPN expression and decreased SM22α expression, indicating that an SMC phenotypic change occurred concomitant with AMC. In contrast to other vascular calcification models where cartilage-like cells have been observed [51], the SMC phenotype change in the present study appeared to be predominantly entrained to an osteogenic rather than an osteochondrogenic lineage as collagen II was not expressed in any of the calcified arteries. Whether SMC phenotype is required for AMC is currently unknown. In addition, we observed a loss of SMC in severely calcified vessels. This is interesting, since cell death has been implicated as an inducer of vascular calcification [52, 53]. Whether cell loss occurred prior to or as a result of calcification could not be determined from the present study, and is currently under investigation. Finally, our results suggest that inflammation was not a major inducer of AMC in uremic, high phosphate fed mice. These findings are consistent with clinical data, since AMC in humans, unlike atherosclerotic calcifications, are typically not associated with inflammation [54].

In conclusion, these studies suggest important roles for phosphate loading and degree of renal insufficiency in mediating AMC in mice. Furthermore, serum FGF-23 and OPN were highly correlated with AMC even when serum phosphorus levels were still in the normal range, suggesting potential roles for these molecules as markers and/or etiological mediators of AMC. Finally, the development of a mouse model for robust arterial medial calcification following renal ablation opens the door for future mechanistic studies using knockout mice bred on the DBA/2 calcification susceptible background.

METHODS

Animals and Diets

Female DBA/2 mice were purchased from Taconic (Germantown, NY) and Charles River laboratories (Wilmington, MA) and maintained in a specific pathogen-free environment. DBA/2 mice were chosen based on previous studies demonstrating that these mice are highly susceptible to ectopic calcification, in contrast to C57BL/6 that are resistant [55]. Female mice were used since they showed higher susceptibility to calcification than males (our unpublished data, and [56]). HP and NP diets were purchased from Dyets Inc. (Bethlehem, PA). The HP diet contained 0.9% phosphate, 0.6% calcium and the NP diet contained 0.5% phosphate and 0.6% calcium. Mice had access to food and water ad libitum and were maintained in compliance with the NIH guide for the Care and Use of Laboratory Animals. The University of Washington Animal Care Committee approved the study protocol.

Surgical procedure

Uremia was induced in 20-21 week old mice following the two-step surgical procedure for partial renal ablation described by Gagnon and Gallimore [57]. Briefly, during surgery 1, the right kidney was exposed, decapsulated, and electrocauterized. By manipulating the extent of electrocautery, mice with moderate (MU; ~50% renal ablation) or severe (SU; ~75% renal ablation) uremia were obtained. Following a two week recovery period, left total nephrectomy was performed (surgery 2). Control mice underwent sham surgeries, where dorsal incisions were made, the kidneys surfaced then reinserted into the abdominal cavity. Within 72 hours of post surgery 2, mice were put on either the NP or the HP diets. At termination, the aortas were collected for calcium quantitation and histological/immunohistochemical analyses. Carotid, iliac, external iliac, and mesenteric arteries and hearts and lungs were collected for histological and immunohistochemical analyses.

Study groups

A. Severe uremia study: 1) Sham/NP: sham operated mice fed the NP diet, 2) Sham/HP: sham operated mice fed the HP diet. 3) SU/NP: mice that underwent severe renal ablation and were fed the NP diet; and 4) SU/HP: mice that underwent severe renal ablation and were fed the HP diet. Saphenous blood was collected at week 14 from mice that had been fasted for 3 hrs. Due to weakness and significant percent loss of body weight in the severely uremic group, all four groups were terminated between weeks 14-15. B. Moderate uremia study: 1) Sham/NP, 2) Sham/HP, 3) MU/NP: moderately uremic mice: mice that underwent moderate renal ablation and were fed the NP diet; and 4) MU/HP: mice that underwent moderate renal ablation and were fed the HP diet. Saphenous blood was collected at week 23 from mice that had been fasted for 3 hrs, followed by the termination of all four groups at week 24.

Serum chemistry

Serum chemistries were analyzed by standard autoanalyzer laboratory methods performed at Phoenix Central Laboratory (Everett, WA). Serum PTH, FGF-23 and OPN levels were determined using mouse intact PTH-ALPCO ELISA (ALPCO; Salem, NH), full length FGF-23 ELISA (Kinos Inc., Japan), and OPN DY441 ELISA (R&D Systems), respectively.

Quantitative Biochemical Analysis of Aortic Calcium

Dissected aortic tissues were frozen, lyophilized and decalcified with 0.6 mM HCl at 37°C for 24 hours. After centrifugation, the calcium content of the supernatant was determined colorimetrically with the o-cresolphthalein complexone reagent using the TECO calcium diagnostic kit (TECO Diagnostics, Anaheim, CA) as previously described [46]. Aortic calcium content was normalized to the dry weight of the tissue and expressed as μg Ca/mg dry weight.

Fractional Excretion of Phosphate (FEP)

Fasting serum and spot urines were collected from unmanipulated sham mice fed either a high or low phosphate diet, and moderately uremic (MU) mice fed the HP diet for 12 weeks. Serum creatinine was measured using the Quantichrom Creatinine assay kit (BioAssay Systems, Hayward, CA). Serum phosphorus, urine phosphorus and creatinine were measured by autoanalyzer as described above. Fractional excretion of phosphate was calculated using the following equation: FEC = serum creatinine × urine phosphate/urine creatinine x serum phosphate.

Histology and Immunohistochemisty

Tissues were fixed in methyl Carnoy’s fixative (3:1, methanol:acetic acid) overnight, then transferred to 70% ethanol and embedded in paraffin using an automated tissue processor, and 5 μm sections were prepared. Alizarin Red (0.5%, pH 9.0; Sigma-Aldrich) staining was used to detect calcification while hematoxylin and eosin (H&E) staining was performed for basic histological analysis. For immunohistochemical staining, sections were deparaffinized, rehydrated, then blocked first with 0.3% H2O2, then with avidin and biotin blocking reagents (Vector Laboratories, Burlingame, CA) and finally with PBS containing 0.25-0.5% bovine serum albumin and 4% serum (same species in which the secondary antibody was raised) [46]. Polyclonal goat anti-mouse OPN antibody (AF808, R&D Systems, Minneapolis, MN); monoclonal rat anti-human Runx2/cbfa1 antibody (MAB2006, R&D systems, Minneapolis, MN); polyclonal goat anti-mouse smooth muscle SM 22α antibody (Abcam Inc., Cambridge, MA); and monoclonal rat anti-murine pan macrophage antibody (BM-8; Accurate Chemical & Scientific Corp., Westbury, NY) were utilized. After incubation at room temperature for 1 hour, the sections were incubated with biotinylated secondary antibody (Vector Laboratories) before streptavidin-conjugated peroxidase staining using 3,3′ diaminobenzidine as chromogen. Counterstaining was performed with 2% methyl green, followed by dehydration and mounting. Whole mount Alizarin Red staining of the full length aorta was performed as previously described [58].

Statistical Analysis

Statistical analyses were performed using the SPSS v16.0 (Chicago, IL) . Group means and variances were determined using ANOVA unless otherwise noted, and values are presented as means ± SEM. Posthoc, pair-wise comparisons of group means were performed using the Fisher’s least significant difference test (LSD), and differences between groups were considered statistically significant at a p value of < 0.05. The Pearson correlation coefficient (R) and P-values were used to determine associations. A positive correlation between two variables, shown as R > 0.5, was considered statistically significant at a P value < 0.05.

ACKNOWLEDGMENTS

This research was supported by NIH grants HL62329 and HL081785 (CMG) and NIH training grant T32 DE07023-29 fellowship (MME).

Footnotes

DISCLOSURES: none

REFERENCES

1. Bloembergen WE. Cardiac disease in chronic uremia: epidemiology. Adv Ren Replace Ther. 1997;4:185–193. [PubMed]
2. Foley RN, Parfrey PS. Cardiovascular disease and mortality in ESRD. J Nephrol. 1998;11:239–245. [PubMed]
3. Blacher J, Guerin AP, Pannier B, et al. Arterial calcifications, arterial stiffness, and cardiovascular risk in end-stage renal disease. Hypertension. 2001;38:938–942. [PubMed]
4. London GM, Guerin AP, Marchais SJ, et al. Arterial media calcification in end-stage renal disease: impact on all-cause and cardiovascular mortality. Nephrol Dial Transplant. 2003;18:1731–1740. [PubMed]
5. Raggi P, Boulay A, Chasan-Taber S, et al. Cardiac calcification in adult hemodialysis patients. A link between end-stage renal disease and cardiovascular disease? J Am Coll Cardiol. 2002;39:695–701. [PubMed]
6. Moe SM, O’Neill KD, Duan D, et al. Medial artery calcification in ESRD patients is associated with deposition of bone matrix proteins. Kidney Int. 2002;61:638–647. [PubMed]
7. Coates T, Kirkland GS, Dymock RB, et al. Cutaneous necrosis from calcific uremic arteriolopathy. Am J Kidney Dis. 1998;32:384–391. [PubMed]
8. Ahmed S, O’Neill KD, Hood AF, et al. Calciphylaxis is associated with hyperphosphatemia and increased osteopontin expression by vascular smooth muscle cells. Am J Kidney Dis. 2001;37:1267–1276. [PubMed]
9. El-Abbadi M, Giachelli CM. Mechanisms of vascular calcification. Adv Chronic Kidney Dis. 2007;14:54–66. [PubMed]
10. Block GA, Hulbert-Shearon TE. Association of serum phosphorus and calcium X phosphate product with mortality risk in chronic hemodialysis patients: a national study. American Journal of Kidney Disease. 1998;31:607–617. [PubMed]
11. Kestenbaum B, Sampson JN, Rudser KD, et al. Serum phosphate levels and mortality risk among people with chronic kidney disease. J Am Soc Nephrol. 2005;16:520–528. [PubMed]
12. Oh J, Wunsch R, Turzer M, et al. Advanced coronary and carotid arteriopathy in young adults with childhood-onset chronic renal failure. Circulation. 2002;106:100–105. [PubMed]
13. Goodman WG, Goldin J, Kuizon BD, et al. Coronary-artery calcification in young adults with end-stage renal disease who are undergoing dialysis. N Engl J Med. 2000;342:1478–1483. [PubMed]
14. Mazhar AR, Johnson RJ, Gillen D, et al. Risk factors and mortality associated with calciphylaxis in end-stage renal disease. Kidney Int. 2001;60:324–332. [PubMed]
15. Chertow GM, Burke SK, Raggi P. Sevelamer attenuates the progression of coronary and aortic calcification in hemodialysis patients. Kidney Int. 2002;62:245–252. [PubMed]
16. Raggi P, Ali O. Phosphorus restriction and control of coronary calcification as assessed by electron beam tomography. Curr Opin Nephrol Hypertens. 2002;11:391–395. [PubMed]
17. Block GA, Raggi P, Bellasi A, et al. Mortality effect of coronary calcification and phosphate binder choice in incident hemodialysis patients. Kidney Int. 2007;71:438–441. [PubMed]
18. Tonelli M, Sacks F, Pfeffer M, et al. Relation between serum phosphate level and cardiovascular event rate in people with coronary disease. Circulation. 2005;112:2627–2633. [PubMed]
19. Ibels LS, Alfrey AC, Huffer WE, et al. Arterial calcification and pathology in uremic patients undergoing dialysis. Am J Med. 1979;66:790–796. [PubMed]
20. Gonzalez EA, Lund RJ, Martin KJ, et al. Treatment of a murine model of high-turnover renal osteodystrophy by exogenous BMP-7. Kidney Int. 2002;61:1322–1331. [PubMed]
21. Larsson T, Marsell R, Schipani E, et al. Transgenic mice expressing fibroblast growth factor 23 under the control of the alpha1(I) collagen promoter exhibit growth retardation, osteomalacia, and disturbed phosphate homeostasis. Endocrinology. 2004;145:3087–3094. [PubMed]
22. Foster BL, Nociti FH, Jr., Swanson EC, et al. Regulation of cementoblast gene expression by inorganic phosphate in vitro. Calcif Tissue Int. 2006;78:103–112. [PubMed]
23. Lorenz-Depiereux B, Bastepe M, Benet-Pages A, et al. DMP1 mutations in autosomal recessive hypophosphatemia implicate a bone matrix protein in the regulation of phosphate homeostasis. Nat Genet. 2006;38:1248–1250. [PubMed]
24. Mizobuchi M, Ogata H, Hatamura I, et al. Up-regulation of Cbfa1 and Pit-1 in calcified artery of uraemic rats with severe hyperphosphataemia and secondary hyperparathyroidism. Nephrol Dial Transplant. 2006;21:911–916. [PubMed]
25. Cozzolino M, Staniforth ME, Liapis H, et al. Sevelamer hydrochloride attenuates kidney and cardiovascular calcifications in long-term experimental uremia. Kidney Int. 2003;64:1653–1661. [PubMed]
26. Davies MR, Lund RJ, Hruska KA. BMP-7 Is an Efficacious Treatment of Vascular Calcification in a Murine Model of Atherosclerosis and Chronic Renal Failure. J Am Soc Nephrol. 2003;14:1559–1567. [PubMed]
27. Mathew S, Lund RJ, Strebeck F, et al. Reversal of the adynamic bone disorder and decreased vascular calcification in chronic kidney disease by sevelamer carbonate therapy. J Am Soc Nephrol. 2007;18:122–130. [PubMed]
28. Massy ZA, Ivanovski O, Nguyen-Khoa T, et al. Uremia accelerates both atherosclerosis and arterial calcification in apolipoprotein E knockout mice. J Am Soc Nephrol. 2005;16:109–116. [PubMed]
29. Ritz E, McClellan WM. Overview: increased cardiovascular risk in patients with minor renal dysfunction: an emerging issue with far-reaching consequences. J Am Soc Nephrol. 2004;15:513–516. [PubMed]
30. Menon V, Greene T, Pereira AA, et al. Relationship of phosphorus and calcium-phosphorus product with mortality in CKD. Am J Kidney Dis. 2005;46:455–463. [PubMed]
31. Stubbs J, Liu S, Quarles LD. Role of fibroblast growth factor 23 in phosphate homeostasis and pathogenesis of disordered mineral metabolism in chronic kidney disease. Semin Dial. 2007;20:302–308. [PubMed]
32. Benet-Pages A, Orlik P, Strom TM, et al. An FGF23 missense mutation causes familial tumoral calcinosis with hyperphosphatemia. Hum Mol Genet. 2005;14:385–390. [PubMed]
33. Shimada T, Kakitani M, Yamazaki Y, et al. Targeted ablation of Fgf23 demonstrates an essential physiological role of FGF23 in phosphate and vitamin D metabolism. J Clin Invest. 2004;113:561–568. [PMC free article] [PubMed]
34. Sitara D, Razzaque MS, Hesse M, et al. Homozygous ablation of fibroblast growth factor-23 results in hyperphosphatemia and impaired skeletogenesis, and reverses hypophosphatemia in Phex-deficient mice. Matrix Biol. 2004;23:421–432. [PMC free article] [PubMed]
35. Torres P Urena, Friedlander G, de Vernejoul MC, et al. Bone mass does not correlate with the serum fibroblast growth factor 23 in hemodialysis patients. Kidney Int. 2008;73:102–107. [PubMed]
36. Emmett M. What does serum fibroblast growth factor 23 do in hemodialysis patients? Kidney Int. 2008;73:3–5. [PubMed]
37. Block GA, Spiegel DM, Ehrlich J, et al. Effects of sevelamer and calcium on coronary artery calcification in patients new to hemodialysis. Kidney Int. 2005;68:1815–1824. [PubMed]
38. Gutierrez OM, Mannstadt M, Isakova T, et al. Fibroblast growth factor 23 and mortality among patients undergoing hemodialysis. N Engl J Med. 2008;359:584–592. [PMC free article] [PubMed]
39. Rix M, Andreassen H, Eskildsen P, et al. Bone mineral density and biochemical markers of bone turnover in patients with predialysis chronic renal failure. Kidney Int. 1999;56:1084–1093. [PubMed]
40. Scatena M, Liaw L, Giachelli CM. Osteopontin: a multifunctional molecule regulating chronic inflammation and vascular disease. Arterioscler Thromb Vasc Biol. 2007;27:2302–2309. [PubMed]
41. Giachelli CM, Bae N, Almeida M, et al. Osteopontin is elevated during neointima formation in rat arteries and is a novel component of human atherosclerotic plaques. J Clin Invest. 1993;92:1686–1696. [PMC free article] [PubMed]
42. Beck GR, Jr., Zerler B, Moran E. Phosphate is a specific signal for induction of osteopontin gene expression. Proc Natl Acad Sci U S A. 2000;97:8352–8357. [PMC free article] [PubMed]
43. Steitz SA, Speer MY, Curinga G, et al. Smooth muscle cell phenotypic transition associated with calcification: upregulation of Cbfa1 and downregulation of smooth muscle lineage markers. Circ Res. 2001;89:1147–1154. [PubMed]
44. Speer MY, Chien YC, Quan M, et al. Smooth muscle cells deficient in osteopontin have enhanced susceptibility to calcification in vitro. Cardiovasc Res. 2005;66:324–333. [PubMed]
45. Chen NX, O’Neill KD, Duan D, et al. Phosphorus and uremic serum up-regulate osteopontin expression in vascular smooth muscle cells. Kidney International. 2002;62:1724–1731. [PubMed]
46. Speer MY, McKee MD, Guldberg RE, et al. Inactivation of the osteopontin gene enhances vascular calcification of matrix Gla protein-deficient mice: evidence for osteopontin as an inducible inhibitor of vascular calcification in vivo. J Exp Med. 2002;196:1047–1055. [PMC free article] [PubMed]
47. Ohri R, Tung E, Rajachar R, et al. Mitigation of Ectopic Calcification in Osteopontin-Deficient Mice by Exogenous Osteopontin. Calcif Tissue Int. 2005 [PubMed]
48. Jono S, Peinado C, Giachelli CM. Phosphorylation of osteopontin is required for inhibition of vascular smooth muscle cell calcification. J Biol Chem. 2000;275:20197–20203. [PubMed]
49. Wada T, McKee MD, Steitz S, et al. Calcification of vascular smooth muscle cell cultures inhibition by osteopontin. Circulation Research. 1999;84:166–178. [PubMed]
50. Jono S, McKee MD, Murry CE, et al. Phosphate regulation of vascular smooth muscle cell calcification. Circ Res. 2000;87:E10–17. [PubMed]
51. Bennett BJ, Scatena M, Kirk EA, et al. Osteoprotegerin inactivation accelerates advanced atherosclerotic lesion progression and calcification in older ApoE-/-mice. Arterioscler Thromb Vasc Biol. 2006;26:2117–2124. [PubMed]
52. Proudfoot D, Skepper JN, Hegyi L, et al. Apoptosis regulates human vascular calcification in vitro: evidence for initiation of vascular calcification by apoptotic bodies. Circ Res. 2000;87:1055–1062. [PubMed]
53. Clarke MC, Littlewood TD, Figg N, et al. Chronic apoptosis of vascular smooth muscle cells accelerates atherosclerosis and promotes calcification and medial degeneration. Circ Res. 2008;102:1529–1538. [PubMed]
54. Shanahan CM, Cary NR, Salisbury JR, et al. Medial localization of mineralization-regulating proteins in association with Monckeberg’s sclerosis: evidence for smooth muscle cell-mediated vascular calcification. Circulation. 1999;100:2168–2176. [PubMed]
55. Ivandic BT, Utz HF, Kaczmarek PM, et al. New Dyscalc loci for myocardial cell necrosis and calcification (dystrophic cardiac calcinosis) in mice. Physiol Genomics. 2001;6:137–144. [PubMed]
56. Qiao JH, Fishbein MC, Demer LL, et al. Genetic determination of cartilaginous metaplasia in mouse aorta. Arterioscler Thromb Vasc Biol. 1995;15:2265–2272. [PubMed]
57. Gagnon RF, Gallimore B. Characterization of a mouse model of chronic uremia. Urol Res. 1988;16:119–126. [PubMed]
58. Price PA, Chan WS, Jolson DM, et al. The elastic lamellae of devitalized arteries calcify when incubated in serum: evidence for a serum calcification factor. Arterioscler Thromb Vasc Biol. 2006;26:1079–1085. [PubMed]
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

  • Compound
    Compound
    PubChem Compound links
  • MedGen
    MedGen
    Related information in MedGen
  • PubMed
    PubMed
    PubMed citations for these articles
  • Substance
    Substance
    PubChem Substance links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...