Portal de Periódicos da CAPES

Chemical and hormonal determinants of vascular calcification in vitro

2006; Elsevier BV; Volume: 69; Issue: 8 Linguagem: Inglês

10.1038/sj.ki.5000297

1523-1755

Koba A. Lomashvili, Puneet Garg, W. Charles O’Neill,

Bone health and osteoporosis research

Vascular calcification is a complex process that is dependent not only on the physicochemical effects of Ca, PO4, and pH, but also on smooth muscle factors that may be regulated by these ions as well as by 1,25-dihydroxyvitamin D3 (calcitriol) and parathyroid hormone (PTH). These minerals and hormones were tested in a model of medial calcification in rat aorta maintained in culture for 9 days. Calcification was quantitated as incorporation of 45Ca, alkaline phosphatase activity was measured in aortic homogenates, and osteopontin production was measured from immunoblots of culture medium. At 1.8 mM Ca (1.46 mM free), calcification occurred at or above 2.8 mM PO4. At 3.8 mM PO4, calcification occurred at or above 1.10 mM free [Ca]. At a constant [Ca] × [PO4], calcification varied directly with [Ca] and inversely with [PO4]. Calcification was directly related to pH between 7.19 and 7.50 but not altered by PTH or calcitriol. Alkaline phosphatase activity and osteopontin production were increased by Ca, PO4, calcitriol, and PTH. We conclude that calcification of rat aorta in vitro requires elevation of both [Ca] and [PO4], and that [Ca] rather than [PO4] or the product of the two is the dominant determinant. The induction of alkaline phosphatase and osteopontin indicates that Ca and PO4 have effects in addition to simple physicochemical actions. Although PTH and calcitriol did not increase calcification in vivo, they have effects on smooth muscle that could influence calcification in vivo. Calcification is enhanced by alkalinity within the range produced during hemodialysis. Vascular calcification is a complex process that is dependent not only on the physicochemical effects of Ca, PO4, and pH, but also on smooth muscle factors that may be regulated by these ions as well as by 1,25-dihydroxyvitamin D3 (calcitriol) and parathyroid hormone (PTH). These minerals and hormones were tested in a model of medial calcification in rat aorta maintained in culture for 9 days. Calcification was quantitated as incorporation of 45Ca, alkaline phosphatase activity was measured in aortic homogenates, and osteopontin production was measured from immunoblots of culture medium. At 1.8 mM Ca (1.46 mM free), calcification occurred at or above 2.8 mM PO4. At 3.8 mM PO4, calcification occurred at or above 1.10 mM free [Ca]. At a constant [Ca] × [PO4], calcification varied directly with [Ca] and inversely with [PO4]. Calcification was directly related to pH between 7.19 and 7.50 but not altered by PTH or calcitriol. Alkaline phosphatase activity and osteopontin production were increased by Ca, PO4, calcitriol, and PTH. We conclude that calcification of rat aorta in vitro requires elevation of both [Ca] and [PO4], and that [Ca] rather than [PO4] or the product of the two is the dominant determinant. The induction of alkaline phosphatase and osteopontin indicates that Ca and PO4 have effects in addition to simple physicochemical actions. Although PTH and calcitriol did not increase calcification in vivo, they have effects on smooth muscle that could influence calcification in vivo. Calcification is enhanced by alkalinity within the range produced during hemodialysis. Vascular calcification is extremely common in patients with advanced renal failure or end-stage renal disease,1.Goodman W.G. Goldin J. Kuizon B.D. et al.Coronary-artery calcification in young adults with end-stage renal disease who are undergoing dialysis.New Engl J Med. 2000; 342: 1478-1483Crossref PubMed Scopus (2331) Google Scholar and may contribute to their elevated risk of cardiovascular disease. Although calcification can be associated with atherosclerosis in these patients, the deposition of hydroxyapatite is primarily medial rather than intimal, and is not associated with inflammation.2.Ibels L.S. Alfrey A.C. Huffer W.E. et al.Arterial calcification and pathology in uremic patients undergoing dialysis.Am J Med. 1979; 66: 790-796Abstract Full Text PDF PubMed Scopus (281) Google Scholar, 3.Ejerblad S. Ericsson J.L.E. Erikson I. Arterial lesions of the radial artery in uraemic patients.Acta Chir Scand. 1979; 145: 415-428PubMed Google Scholar The pathophysiology of medial calcification is complex, involving not only physicochemical factors but also biologic actions in smooth muscle. Physicochemical considerations are the basis for the clinical use of the calcium–phosphorus product (Ca × P). Although the serum Ca × P is associated with soft tissue calcification,4.Milliner D.S. Zinsmeister A.R. Lieberman E. et al.Soft tissue calcification in pediatric patients with end-stage renal disease.Kidney Int. 1990; 38: 931-936Abstract Full Text PDF PubMed Scopus (264) Google Scholar coronary artery calcification,1.Goodman W.G. Goldin J. Kuizon B.D. et al.Coronary-artery calcification in young adults with end-stage renal disease who are undergoing dialysis.New Engl J Med. 2000; 342: 1478-1483Crossref PubMed Scopus (2331) Google Scholar and cardiovascular mortality5.Ganesh S.K. Stack A.G. Levin N.W. et al.Association of elevated serum PO4, Ca × PO4 product, and parathyroid hormone with cardiac mortality risk in chronic hemodialysis patients.J Am Soc Nephrol. 2001; 12: 2131-2138Crossref PubMed Scopus (1432) Google Scholar in hemodialysis patients, there are no data showing a causative role in vascular calcification. The principal biologic effect of smooth muscle is the production of inhibitors of calcification. Humans lacking an enzyme that produces extracellular pyrophosphate, a known inhibitor of hydroxyapatite formation,6.Russell R.G.G. Bisaz S. Fleisch H. Pyrophosphate and diphosphates in calcium metabolism and their possible role in renal failure.Arch Intern Med. 1969; 124: 571-575Crossref PubMed Scopus (44) Google Scholar, 7.Meyer J.L. Can biological calcification occur in the presence of pyrophosphate?.Arch Biochem Biophys. 1984; 231: 1-8Crossref PubMed Scopus (166) Google Scholar, 8.Francis M.D. Russell R.G.G. Fleisch H. Diphosphonates inhibit formation of calcium phosphate crystals in vitro and pathologic calcification in vivo.Science. 1969; 165: 1264-1266Crossref PubMed Scopus (465) Google Scholar develop severe medial vascular calcification in childhood,9.Terkeltaub R.A. Inorganic pyrophosphate generation and disposition in pathology.Am J Physiol: Cell Physiol. 2001; 281: C1-C11PubMed Google Scholar, 10.Rutsch F. Vaingankar S. Johnson K. et al.PC-1 nucleotide triphosphate pyrophosphohydrolase deficiency in idiopathic infantile arterial calcification.Am J Pathol. 2001; 158: 543-554Abstract Full Text Full Text PDF PubMed Scopus (227) Google Scholar indicating that calcification can occur at normal calcium and phosphate levels in the absence of inhibition. We have recently confirmed the inhibitory role of smooth muscle pyrophosphate in rat aortas in vitro.11.Lomashvili K.A. Cobbs S. Hennigar R.A. et al.Phosphate-induced vascular calcification: role of pyrophosphate and osteopontin.J Am Soc Nephrol. 2004; 15: 1392-1401Crossref PubMed Scopus (239) Google Scholar Protein inhibitors have been described as well. Mice lacking matrix Gla protein and rats treated with warfarin to prevent its γ-carboxylation develop severe medial calcification,12.Luo G. Ducy P. McKee M.D. et al.Spontaneous calcification of arteries and cartilage in mice lacking matrix Gla protein.Nature. 1997; 386: 78-81Crossref PubMed Scopus (1645) Google Scholar, 13.Price P.A. Faus S.A. Williamson M.K. Warfarin causes rapid calcification of the elastic lamella in rat arteries and heart valves.Arterioscler Thromb Vasc Biol. 1998; 18: 1400-1407Crossref PubMed Scopus (444) Google Scholar although humans lacking this protein (Keutel syndrome) do not develop prominent calcification.14.Munroe P.B. Olgunturk R.O. Fryns J.P. et al.Mutations in the gene encoding the human matrix Gla protein cause Keutel syndrome.Nat Genet. 1999; 21: 142-144Crossref PubMed Scopus (322) Google Scholar Mice deficient in osteoprotegerin also develop medial calcification,15.Bucay N. Sarosi I. Dunstan C. et al.Osteoprotegerin-deficient mice develop early onset osteoporosis and arterial calcification.Genes Dev. 1998; 12: 1260-1268Crossref PubMed Scopus (2025) Google Scholar and osteopontin inhibits calcification in cultured smooth muscle cells.16.Wada T. McKee M.D. Steitz S. et al.Calcification of vascular smooth muscle cell cultures. Inhibition by osteopontin.Circ Res. 1999; 84: 166-178Crossref PubMed Scopus (387) Google Scholar, 17.Jono S. Peinado C. Giachelli C.M. Phosphorylation of osteopontin is required for inhibition of vascular smooth muscle cell calcification.J Biol Chem. 2000; 275: 20197-20203Crossref PubMed Scopus (268) Google Scholar Although deficiency of osteopontin does not lead to vascular calcification in mice, it does worsen calcification in mice lacking matrix Gla protein.18.Speer M.Y. McKee M.D. Guldberg R.E. 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: 1047-1055Crossref PubMed Scopus (266) Google Scholar An additional biologic action that may promote calcification is osteogenic differentiation of smooth muscle.19.Moe S.M. Chen N.X. Pathophysiology of vascular calcification in chronic kidney disease.Circ Res. 2004; 95: 560-567Crossref PubMed Scopus (392) Google Scholar Recent data suggest that biologic effects of smooth muscle on calcification may be governed by systemic mineral metabolism. In cultured vascular smooth muscle cells, both Ca and PO4 increase alkaline phosphatase activity (which hydrolyzes pyrophosphate) and osteopontin production,16.Wada T. McKee M.D. Steitz S. et al.Calcification of vascular smooth muscle cell cultures. Inhibition by osteopontin.Circ Res. 1999; 84: 166-178Crossref PubMed Scopus (387) Google Scholar, 20.Chen N.X. O'Neill K.D. Moe S.M. Phosphorus and uremic serum up-regulate osteopontin expression in vascular smooth muscle cells.Kidney Int. 2002; 62: 1724-1731Abstract Full Text Full Text PDF PubMed Scopus (286) Google Scholar and phosphate induces osteoblastic differentiation factors such as osteocalcin and Cbfa-1.21.Jono S. McKee M.D. Murry C.E. et al.Phosphate regulation of vascular smooth muscle cell calcification.Circ Res. 2000; 87: E10-E17Crossref PubMed Google Scholar High doses of vitamin D3 induce medial vascular calcification in rats,22.Irving J.T. Schibler D. Fleisch H. Effect of condensed phosphates on vitamin D-induced aortic calcification in rats.Proc Soc Exper Biol Med. 1966; 122: 852-856Crossref PubMed Scopus (11) Google Scholar and calcitriol enhances calcium deposition in cultured smooth muscle cells.23.Jono S. Nishizawa Y. Shioi A. et al.1,25-dihydrozyvitamin D3 increases in vitro vascular calcification by modulating secretion of endogenous parathyroid hormone-related peptide.Circulation. 1998; 98: 1302-1306Crossref PubMed Scopus (249) Google Scholar Parathyroid hormone (PTH) at high concentrations inhibits calcification of vascular smooth muscle cells in culture.24.Jono S. Nishizawa Y. Morii H. Parathyroid hormone-related peptide as a local regulator of vascular calcification.Arterioscl Thromb VascBiol. 1997; 17: 1135-1142Crossref PubMed Scopus (139) Google Scholar In vitro studies of vascular calcification have been performed almost exclusively in cultured smooth muscle cells, but this model is limited by the substantial phenotypic changes these cells undergo in culture and the fact that they lack elastin, the site of medial calcification in vivo.25.Vyavahare N. Ogle M. Schoen F.J. et al.Elastin calcification and its prevention with aluminum chloride pretreatment.AmJ Pathol. 1999; 155: 973-982Abstract Full Text Full Text PDF PubMed Scopus (78) Google Scholar Thus, it is not clear that these cells accurately reflect the pathophysiology of medial vascular calcification and there are no data showing a direct role for Ca, PO4, calcitriol, or PTH in the calcification of intact vessels. To address this, we recently developed an in vitro model of vascular calcification in rat aortas during long-term culture under conditions that maintain viability and normal histology and prevent apoptosis.11.Lomashvili K.A. Cobbs S. Hennigar R.A. et al.Phosphate-induced vascular calcification: role of pyrophosphate and osteopontin.J Am Soc Nephrol. 2004; 15: 1392-1401Crossref PubMed Scopus (239) Google Scholar In the presence of high calcium and phosphate concentrations along with alkaline phosphatase to remove inhibitory pyrophosphate, medial deposits of hydroxyapatite develop along elastic lamina in a pattern similar to that observed in vessels from patients with chronic renal failure.11.Lomashvili K.A. Cobbs S. Hennigar R.A. et al.Phosphate-induced vascular calcification: role of pyrophosphate and osteopontin.J Am Soc Nephrol. 2004; 15: 1392-1401Crossref PubMed Scopus (239) Google Scholar Using this model, we examined the effect of Ca, PO4, pH, calcitriol, and PTH on the rate of calcification and the induction of proteins related to vascular calcification. Figure 1 shows a von Kossa stain of an aortic ring cultured for 9 days in the presence of alkaline phosphatase and a high phosphate concentration. This degree of calcification is observed in cultured aortas when the calcium content is roughly 500 nmol/mg dry weight or greater, and contents below 50 nmol/mg are usually not visible on Von Kossa staining. To determine the dependence of calcification on the concentration of phosphate, incorporation of 45Ca was measured at different phosphate concentrations with the calcium concentration maintained constant at 1.8 mM, the concentration normally present in Dulbecco's modified Eagle's medium (DMEM) (Figure 2). There was a very small incorporation of 45Ca into aortas cultured in 0.9 mM PO4, the concentration normally present in DMEM. We have previously shown that this incorporation occurs in the first day and represents equilibration with Ca normally present in the vessel wall.11.Lomashvili K.A. Cobbs S. Hennigar R.A. et al.Phosphate-induced vascular calcification: role of pyrophosphate and osteopontin.J Am Soc Nephrol. 2004; 15: 1392-1401Crossref PubMed Scopus (239) Google Scholar This incorporation began to increase when [PO4] reached 2.7 mM and increased substantially thereafter. The measured concentrations of ionized Ca at the lowest and highest [PO4] were 1.46 and 1.21 mM. The dependence of calcification on calcium concentration was assessed by varying the concentration of calcium at a constant phosphate concentration of 3.8 mM (Figure 3). Because the concentration of Ca is already high in DMEM, a medium similar to DMEM was reformulated from components to yield a reduced [Ca] of 1.33 mM. No calcification occurred at a [Ca] of 1.33 or 1.49 mM but calcification increased substantially starting at 1.65 mM (1.10 mM free [Ca]). Calcification did not occur at physiologic concentrations of PO4 (0.9 mM) or Ca (0.99 mM free) despite maximal concentrations of the other ion. To determine whether the effects of Ca and PO4 were governed by the product of their concentrations, the concentrations were varied inversely to maintain a constant product of 6.84 mmol2/l2. As shown in Figure 4, calcification varied directly with [Ca] and inversely with [PO4], with only basal calcium incorporation at the highest [PO4] and lowest [Ca] and extensive calcification at the highest [Ca] and lowest [PO4].Figure 2Effect of phosphate concentration on calcification of rat aortas in culture. Calcium concentration was 1.8 mM. Values are the means of 3–8 individual aortic segments. Error bars: s.e. *P<0.001 vs 0.9 mM phosphate.View Large Image Figure ViewerDownload (PPT)Figure 3Effect of calcium concentration on calcification of rat aortas in culture. Phosphate concentration was 3.8 mM. Values are the means of 4–6 individual aortic segments. Error bars: s.e. *P<0.001 vs 1.33 mM calcium.View Large Image Figure ViewerDownload (PPT)Figure 4Calcification at varying [Ca] and [PO4] with Ca × P kept constant at 6.84 mmol2/l2. Values are the means of 7–15 individual aortic segments. Error bars: s.e. *P<0.001 vs 1.33 mM calcium and 5.14 mM phosphate.View Large Image Figure ViewerDownload (PPT) Because vascular calcification in uremia may be dependent on vessel injury and circulating factors other than Ca and PO4, these experiments were repeated in the presence of uremic plasma or after vessel injury. The same dependence on Ca rather than PO4 at a constant product was observed in vessels cultured with 10% pooled, predialysis plasma from hemodialysis patients (Figure 5), or in vessels that had been frozen and thawed prior to culture (Figure 6). The threshold [Ca] and [PO4] for calcification were 36 and 44% higher in the presence of plasma or serum (not shown). This was due in part to binding of Ca to protein since ionized Ca concentrations were 15% lower.Figure 6Calcification of injured aortas at varying [Ca] and [PO4] but constant Ca × P in the presence of 10% serum without added alkaline phosphatase. Aortas were frozen and thawed five times prior to culture. Left-hand bars: Ca × P=8.61 mmol2/l2. Right-hand bars: Ca × P=9.35 mmol2/l2. Values are the means of four individual aortic segments. Error bars: s.e. *P<0.003, **P<0.001 vs highest phosphate concentration.View Large Image Figure ViewerDownload (PPT) In addition to Ca and PO4 concentrations, solubility of hydroxyapatite is also dependent on pH. To determine whether aortic calcification varied within the pH range that occurs in vivo, the pH of the culture medium was varied from 7.2 to 7.5 (Figure 7). Calcification was reduced in acidic medium and was substantially increased at pH 7.5. Greater degrees of alkalinity could not be tested because of precipitation of calcium from the medium. A phosphate concentration of 2.8 mM was used in these studies to ensure that an intermediate degree of calcification occurred at normal pH (7.4). To determine whether intermittent periods of alkalosis equivalent to those that occur during hemodialysis could also increase calcification, pH of the medium was increased from 7.4 to 7.5 for 5 h every other day. As shown in the right-hand bars, calcification increased 2.5-fold compared to medium changes performed at a constant pH of 7.4. Calcium incorporation into rat aortas was not altered by calcitriol or rat PTH 1–34 (Figure 8), each at a concentration of 100 nM. Several concentrations of phosphate were used so that either augmentation or inhibition of calcification could be detected. To assess responsiveness to PTH, cyclic 3′,5′ adenosine monophosphate was measured in aortas before and after exposure to PTH for 5 min. In fresh aortas, PTH increased cyclic 3′,5′ adenosine monophosphate content (normalized to wet weight) 77% from 5.84±0.33 to 10.3±1.37 pmol/mg (P<0.01). Basal cAMP content was lower after 9 days in culture (2.15±0.08 pmol/mg), but still responded to PTH with an increase of 53% to 3.29±0.35 pmol/mg (P<0.05). In addition to calcification, the induction of proteins related to calcification was also investigated. Figure 9 shows the activity of alkaline phosphatase in homogenates of aorta after 9 days in culture without added alkaline phosphatase. Activity was increased two- to three-fold when PO4, PTH, or calcitriol were added to the medium. The effect of PTH and PO4 were not additive (not shown). The effect of calcium was studied separately since reformulated DMEM was used. An increase in [Ca] from 1.3 mM back to 1.8 mM more than tripled alkaline phosphatase activity. The reformulated medium with 1.8 mM Ca yielded a higher alkaline phosphatase activity than commercially prepared DMEM with the same Ca concentration. Osteopontin production, measured by immunoblotting medium from the last 3 days of culture (without added alkaline phosphatase), was also increased by PO4, Ca, PTH, or calcitriol (Figure 10).Figure 10Effect of Ca, PO4, PTH, and calcitriol on osteopontin production by rat aorta. Aortas were incubated in DMEM medium without added alkaline phosphatase under the specified conditions for 9 days. Medium was changed after 6 days and then collected at the end of the culture for measurement of osteopontin by immunoblotting. Control DMEM contained 1.8 mM Ca and 0.9 mM PO4. PTH or calcitriol were added at 100 nM. To test the effect of Ca, DMEM was reformulated from components to contain 1.3 mM Ca and then additional Ca was added to yield 1.8 mM. Results are means±s.e. of three separate experiments. *P<0.001.View Large Image Figure ViewerDownload (PPT) This is the first study to examine the effects of calcium, phosphate, and relevant hormones on arterial calcification in vitro. Although effects have been studied in smooth muscle cells in culture,16.Wada T. McKee M.D. Steitz S. et al.Calcification of vascular smooth muscle cell cultures. Inhibition by osteopontin.Circ Res. 1999; 84: 166-178Crossref PubMed Scopus (387) Google Scholar, 20.Chen N.X. O'Neill K.D. Moe S.M. Phosphorus and uremic serum up-regulate osteopontin expression in vascular smooth muscle cells.Kidney Int. 2002; 62: 1724-1731Abstract Full Text Full Text PDF PubMed Scopus (286) Google Scholar, 21.Jono S. McKee M.D. Murry C.E. et al.Phosphate regulation of vascular smooth muscle cell calcification.Circ Res. 2000; 87: E10-E17Crossref PubMed Google Scholar these cells undergo phenotypic changes and lack a normal extracellular matrix including elastin, the principal site of medial calcification in vivo.25.Vyavahare N. Ogle M. Schoen F.J. et al.Elastin calcification and its prevention with aluminum chloride pretreatment.AmJ Pathol. 1999; 155: 973-982Abstract Full Text Full Text PDF PubMed Scopus (78) Google Scholar Since aortas from small mammals do not contain vasa vasorum,26.Wolinsky H. Glagov S. Nature of species differences in the medial distribution of aortic vasa vasorum in mammals.Circ Res. 1967; 20: 409-421Crossref PubMed Scopus (243) Google Scholar they rely entirely on diffusion for nutrition and thus are amenable to culture. We have previously shown that aortic structure and viability are maintained for at least 9 days in culture.11.Lomashvili K.A. Cobbs S. Hennigar R.A. et al.Phosphate-induced vascular calcification: role of pyrophosphate and osteopontin.J Am Soc Nephrol. 2004; 15: 1392-1401Crossref PubMed Scopus (239) Google Scholar Calcification of rat aorta in culture increased directly with the concentration of calcium or phosphate, but at concentrations greater than those commonly observed in vivo. In general, calcification required PO4 concentrations of 2.8 mM or greater and Ca concentrations of 1.8 mM (1.30 mM free Ca) or greater. This is equivalent to a serum phosphorus concentration in humans of 8.4 mg/dl and, based on the assumption that 43% of the calcium in serum is ionized,27.Aurbach G.D. Marx S.J. Spiegel A.M. Parathyroid hormone, calcitonin, and the calciferols.in: Wilson J.D. Foster D.W. Williams Textbook of Endocrinology. 7th edn. WB Saunders, Philadelphia1985: 1137-1217Google Scholar a serum Ca concentration of 12.6 mg/dl. The thresholds for calcification were higher in the presence of serum or in injured aortas but the patterns were unchanged. To reflect the dual roles of PO4 and Ca, the product of the two concentrations is often used clinically to assess the risk of vascular calcification. However, the assumption that the two ions affect calcification equally has never been tested. When [Ca] and [PO4] were varied inversely to maintain a constant product, calcification of rat aorta in culture varied substantially, demonstrating that calcification is not a function of the product but rather depends on the individual concentrations. Of note, calcification varied directly with [Ca] despite a decreasing [PO4]. The dominance of Ca is not unexpected based on the stoichiometry of calcium phosphate. The same dependence on [Ca] rather than [PO4] was observed in the presence of plasma from patients with end-stage renal disease and in injured vessels. While exposure to uremic plasma in vitro does not necessarily mimic uremia, these results indicate that the dependence of calcification on Ca and PO4 is not altered by circulating factors. The formation of hydroxyapatite is complex and cannot be described by a simple solubility product.28.Neuman W.F. Neuman M.W. The Chemical Dynamics of Bone Mineral. University of Chicago Press, Chicago1958Google Scholar The initial event appears to be formation of CaHPO4·2H2O (brushite), which is not stable at physiologic pH but can be converted to hydroxyapatite, the only stable form of calcium phosphate at physiologic pH. Owing to the extremely low solubility of hydroxyapatite, normal human serum is supersaturated with respect to hydroxyapatite. However, hydroxyapatite is not formed directly and derives instead from brushite, and the product of the activities of Ca2+ and HPO42- in human serum and DMEM is less than the solubility constant for brushite under physiologic conditions (2.2 × 10-7 M). All the media with higher phosphate concentrations exceed this and also exceed the precipitation point for brushite in simple salt solution at pH 7.4, which is 50% greater. Despite this, no spontaneous precipitation occurred, indicating that the precipitation point is higher in DMEM. In media with varying calcium and phosphate concentrations, aortic calcification varied widely despite solubility products that were very similar (8.5–8.6 × 10-7). Thus, medial vascular calcification cannot be explained solely by solubility considerations. Our data also indicate that calcium and phosphate have biologic effects on vascular smooth muscle that could influence calcification in addition to physicochemical effects on hydroxyapatite formation. In cultured aortas, both alkaline phosphatase activity and osteopontin production were increased by maximal concentrations of Ca or PO4, which lends credence to similar findings in cultured vascular smooth muscle cells.16.Wada T. McKee M.D. Steitz S. et al.Calcification of vascular smooth muscle cell cultures. Inhibition by osteopontin.Circ Res. 1999; 84: 166-178Crossref PubMed Scopus (387) Google Scholar, 20.Chen N.X. O'Neill K.D. Moe S.M. Phosphorus and uremic serum up-regulate osteopontin expression in vascular smooth muscle cells.Kidney Int. 2002; 62: 1724-1731Abstract Full Text Full Text PDF PubMed Scopus (286) Google Scholar Since both proteins can be markers of osteoblastic differentiation, this supports the finding that phosphate induces osteoblastic differentiation factors such as osteocalcin and Cbfa-1 in smooth muscle cells in culture.21.Jono S. McKee M.D. Murry C.E. et al.Phosphate regulation of vascular smooth muscle cell calcification.Circ Res. 2000; 87: E10-E17Crossref PubMed Google Scholar The relative dependence of vascular calcification on Ca and PO4 and the absence of an effect of PTH or calcitriol observed in vitro cannot be extrapolated to clinical practice since the mechanism of vascular calcification in end-stage renal disease is not established and is a slower process that may not be entirely recreated in our culture system. We have previously shown that calcification of aortas is apparent by 6 days in culture and increases rapidly thereafter.11.Lomashvili K.A. Cobbs S. Hennigar R.A. et al.Phosphate-induced vascular calcification: role of pyrophosphate and osteopontin.J Am Soc Nephrol. 2004; 15: 1392-1401Crossref PubMed Scopus (239) Google Scholar Since calcification is a continuous process, the amount present at 9 days actually represents a rate rather than a fixed amount. Thus, the absence of 42Ca after 9 days could also be due to a very slow rate of calcification. Calcification in vitro occurs only in the presence of supranormal levels of alkaline phosphatase, necessary to remove inhibitory pyrophosphate.11.Lomashvili K.A. Cobbs S. Hennigar R.A. et al.Phosphate-induced vascular calcification: role of pyrophosphate and osteopontin.J Am Soc Nephrol. 2004; 15: 1392-1401Crossref PubMed Scopus (239) Google Scholar Cultured smooth muscle cells have a much higher alkaline phosphatase level than intact vessels,16.Wada T. McKee M.D. Steitz S. et al.Calcification of vascular smooth muscle cell cultures. Inhibition by osteopontin.Circ Res. 1999; 84: 166-178Crossref PubMed Scopus (387) Google Scholar, 11.Lomashvili K.A. Cobbs S. Hennigar R.A. et al.Phosphate-induced vascular calcification: role of pyrophosphate and osteopontin.J Am Soc Nephrol. 2004; 15: 1392-1401Crossref PubMed Scopus (239) Google Scholar but exogenous enxyme must be added to rat aortas. The concentrations of Ca and PO4 required for calcification in vitro are higher than the circulating levels in most end-stage renal disease patients. Depending on which ion was varied, the threshold Ca × P for calcification in vitro was equivalent to values of 102 to 122 mg2/dl2 in human serum. It is possible that the high concentrations of calcium and phosphate are necessary to overcome active inhibition of calcification by vascular smooth muscle.11.Lomashvili K.A. Cobbs S. Hennigar R.A. et al.Phosphate-induced vascular calcification: role of pyrophosphate and osteopontin.J Am Soc Nephrol. 2004; 15: 1392-1401Crossref PubMed Scopus (239) Google Scholar, 12.Luo G. Ducy P. McKee M.D. et al.Spontaneous calcification of arteries and cartilage in mice lacking matrix Gla protein.Nature. 1997; 386: 78-81Crossref PubMed Scopus (1645) Google Scholar, 13.Price P.A. Faus S.A. Williamson M.K. Warfarin causes rapid calcification of the elastic lamella in rat arteries and heart valves.Arterioscler Thromb Vasc Biol. 1998; 18: 1400-1407Crossref PubMed Scopus (444) Google Scholar, 14.Munroe P.B. Olgunturk R.O. Fryns J.P. et al.Mutations in the gene encoding the human matrix Gla protein cause Keutel syndrome.Nat Genet. 1999; 21: 142-144Crossref PubMed Scopus (322) Google Scholar, 15.Bucay N. Sarosi I. Dunstan C. et al.Osteoprotegerin-deficient mice develop early onset osteoporosis and arterial calcification.Genes Dev. 1998; 12: 1260-1268Crossref PubMed Scopus (2025) Google Scholar, 16.Wada T. McKee M.D. Steitz S. et al.Calcification of vascular smooth muscle cell cultures. Inhibition by osteopontin.Circ Res. 1999; 84: 166-178Crossref PubMed Scopus (387) Google Scholar, 17.Jono S. Peinado C. Giachelli C.M. Phosphorylation of osteopontin is required for inhibition of vascular smooth muscle cell calcification.J Biol Chem. 2000; 275: 20197-20203Crossref PubMed Scopus (268) Google Scholar Alternatively, since calcification in vivo is a slower process, it could occur at lower concentrations of Ca and PO4. Despite the limitations of these in vitro data, they are the only data on vessel calcification and the results suggest that further studies are needed to determine whether the calcium–phosphorus product is an appropriate parameter for gauging the risk of vascular calcification in vivo. The solubility of calcium phosphate is dependent on pH, but the effect of pH on vascular calcification has not previously been examined. The increased calcification of rat aortas in alkaline medium may have important implications because of the transient alkalemia during hemodialysis. Typical dialysates contain 35–40 mM bicarbonate and result in a blood pH of 7.50–7.52 after dialysis,29.Ahmad S. Pagel M. Vizzo J. et al.Effect of the normalization of acid-base balance on postdialysis plasma bicarbonate.Trans Am Soc Artif Intern Organs. 1980; 26: 318-321PubMed Google Scholar, 30.Williams A.J. Dittmer I.D. McArley A. et al.High bicarbonate dialysate in haemodialysis patients: effects on acidosis and nutritional status.Nephrol Dialysis Transpl. 1997; 12: 2633-2637Crossref PubMed Scopus (74) Google Scholar which is equal to or even greater than the degree of alkalinity shown to increase calcification in cultured aortas (pH 7.5). Even transient increases in medium pH akin to hemodialysis every other day resulted in increased calcification, raising the possibility that hemodialysis could actually worsen vascular calcification. The fact that coronary artery calcification correlates very strongly with the number of years of hemodialysis1.Goodman W.G. Goldin J. Kuizon B.D. et al.Coronary-artery calcification in young adults with end-stage renal disease who are undergoing dialysis.New Engl J Med. 2000; 342: 1478-1483Crossref PubMed Scopus (2331) Google Scholar is consistent with this, but could also be related to the duration of end-stage renal disease or to other therapies such as calcium-based phosphate binders.1.Goodman W.G. Goldin J. Kuizon B.D. et al.Coronary-artery calcification in young adults with end-stage renal disease who are undergoing dialysis.New Engl J Med. 2000; 342: 1478-1483Crossref PubMed Scopus (2331) Google Scholar, 31.London G.M. Marty C. Marchais S.J. et al.Arterial calcifications and bone histomorphometry in end-stage renal disease.J Am Soc Nephrol. 2004; 15: 1943-1951Crossref PubMed Scopus (488) Google Scholar Transient alkalemia is avoided in peritoneal dialysis, but whether this is associated with less vascular calcification is unknown. Although calcitriol has been reported to increase calcification of smooth muscle cells in culture,23.Jono S. Nishizawa Y. Shioi A. et al.1,25-dihydrozyvitamin D3 increases in vitro vascular calcification by modulating secretion of endogenous parathyroid hormone-related peptide.Circulation. 1998; 98: 1302-1306Crossref PubMed Scopus (249) Google Scholar calcitriol did not affect calcification in intact aortas despite a concentration that was 1000-fold greater than levels in human serum.27.Aurbach G.D. Marx S.J. Spiegel A.M. Parathyroid hormone, calcitonin, and the calciferols.in: Wilson J.D. Foster D.W. Williams Textbook of Endocrinology. 7th edn. WB Saunders, Philadelphia1985: 1137-1217Google Scholar The fact that osteopontin was upregulated indicates that the aortas were responding to calcitriol, which is consistent with the known transcriptional regulation of osteopontin by calcitriol in bone cells.32.Khoury R. Ridall A.L. Norman A.W. et al.Target gene activation by 1,25-dihydroxyvitamin D3 in osteosarcoma cells is independent of calcium influx.Endocrinology. 1994; 135: 2446-2453Crossref PubMed Scopus (28) Google Scholar Since high doses of vitamin D3 produce vascular calcification in rats,22.Irving J.T. Schibler D. Fleisch H. Effect of condensed phosphates on vitamin D-induced aortic calcification in rats.Proc Soc Exper Biol Med. 1966; 122: 852-856Crossref PubMed Scopus (11) Google Scholar our results suggest that this is not a direct effect of calcitriol on vascular smooth muscle. We cannot rule out the possibility that calcitriol may act synergistically with other factors in vivo. PTH also had no effect on aortic calcification in culture despite the fact that the smooth muscle remained responsive to PTH during culture. This result in intact smooth muscle differs from the inhibition by PTH of calcification in cultured vascular smooth muscle cells.24.Jono S. Nishizawa Y. Morii H. Parathyroid hormone-related peptide as a local regulator of vascular calcification.Arterioscl Thromb VascBiol. 1997; 17: 1135-1142Crossref PubMed Scopus (139) Google Scholar We cannot rule out the possibility that longer periods of culture are required to observe effects of calcitriol or PTH on calcification, but the duration of treatment was longer than that for previous studies demonstrating effects of these hormones on cultured smooth muscle cells. The effects noted in cultured cells but not in cultured aortas may be the result of a calcification process in cultured cells that differs from medial calcification in intact vessels. The increase in aortic alkaline phosphatase activity and osteopontin production observed with Ca, PO4, calcitriol, and PTH may have important implications for vascular calcification. Alkaline phosphatase hydrolyzes pyrophosphate, an important inhibitor of vascular calcification,11.Lomashvili K.A. Cobbs S. Hennigar R.A. et al.Phosphate-induced vascular calcification: role of pyrophosphate and osteopontin.J Am Soc Nephrol. 2004; 15: 1392-1401Crossref PubMed Scopus (239) Google Scholar, 10.Rutsch F. Vaingankar S. Johnson K. et al.PC-1 nucleotide triphosphate pyrophosphohydrolase deficiency in idiopathic infantile arterial calcification.Am J Pathol. 2001; 158: 543-554Abstract Full Text Full Text PDF PubMed Scopus (227) Google Scholar and controls its level in vivo.33.Rachow J.W. Ryan L.M. Inorganic pyrophosphate metabolism in arthritis.Rheumatic Dis Clin N Am. 1988; 14: 289-302PubMed Google Scholar Osteopontin, on the other hand, is an inhibitor of calcification under certain conditions when it is phosphorylated.17.Jono S. Peinado C. Giachelli C.M. Phosphorylation of osteopontin is required for inhibition of vascular smooth muscle cell calcification.J Biol Chem. 2000; 275: 20197-20203Crossref PubMed Scopus (268) Google Scholar However, this effect might be nullified by the upregulation of alkaline phosphatase, which dephosphorylates osteopontin. The upregulation of these proteins cannot explain the calcification observed in aortic culture since excess alkaline phosphatase was added. However, this upregulation could contribute to calcification in vivo. In summary, this study demonstrates that calcification of rat aortas in vitro requires elevation of both Ca and PO4 concentrations, but is not strictly governed by the product of the two. PTH and calcitriol do not directly affect calcification but, along with Ca and PO4, have effects on osteopontin and alkaline phosphatase that could affect calcification. Calcification is also enhanced by alkalemia in the range observed after hemodialysis. Aortic culture was performed as previously described.11.Lomashvili K.A. Cobbs S. Hennigar R.A. et al.Phosphate-induced vascular calcification: role of pyrophosphate and osteopontin.J Am Soc Nephrol. 2004; 15: 1392-1401Crossref PubMed Scopus (239) Google Scholar Briefly, aortas were removed under sterile conditions from male Sprague–Dawley rats weighing 150–250 g and most of the adventitia was removed by careful dissection. Adventitia was not completely removed as this led to smooth muscle injury. The vessels were cut into 3–4 mm rings and placed in DMEM medium (Mediatech, Herndon, VA, USA) without serum at 37°C in a 5% CO2 incubator with medium changes every 3 days. Unless otherwise specified, calf intestinal alkaline phosphatase (Promega, Madison, WI, USA) was added at a final concentration of 3.75 U/ml. DMEM medium with a reduced [Ca] of 1.3 mM instead of 1.8 mM was made from components (50 × minimal essential medium amino acids, 100 × minimal essential medium nonessential amino acids, and 100 × minimal essential medium vitamins) purchased from Gibco-BRL (Rockville, MD, USA). Medium pH was altered by adding HCl or NaHCO3. Ionized concentrations of Ca and phosphate were determined using CaBuffer software based on a [Mg] of 0.88 mM, a [glycine] of 0.4 mM, an ionic strength of 0.3 M, and a pH of 7.4. Measurements with an ion-selective probe gave ionized Ca values that were 17% lower in standard medium and 26% lower in high-phosphate medium, and the calculated values were corrected accordingly. Activities of Ca2+ and HPO42- were calculated from published activity coefficients,28.Neuman W.F. Neuman M.W. The Chemical Dynamics of Bone Mineral. University of Chicago Press, Chicago1958Google Scholar and the proportion of phosphate present as HPO4-2 was assumed to be 81%.28.Neuman W.F. Neuman M.W. The Chemical Dynamics of Bone Mineral. University of Chicago Press, Chicago1958Google Scholar Measurement of tracer 45Ca was performed on all media before and after centrifugation to ensure that all the calcium was soluble. Medium 45Ca was measured at the end of culture periods to ensure that calcium levels were maintained. Rat PTH 1–34 and 1,25-dihydroxyvitamin D3 (calcitriol) were obtained from Sigma Chemicals (St Louis, MO, USA). Tracer 45Ca (0.5 μCi/ml) was added to the culture medium, and after 9 days the aortas were washed five times in physiologic saline. Residual adventitia was then removed and the rings were dried, weighed, and then dissolved in 15% H2O2 and 35% HClO4 for 2 h at 80°C for quantification of radioactivity by liquid scintillation. Immunoblots were performed after separation of culture medium on a 10% SDS polyacrylamide gel and blotting onto polyvinylidine difluoride membranes, using a mouse monoclonal antibody (MPIIIB101) created by M Solursh and A Franzen and obtained from the Developmental Studies Hybridoma Bank developed under the auspices of the NICHD and maintained by the University of Iowa, Department of Biological Sciences (Iowa City, IA, USA). Rat recombinant osteopontin was produced as a hexahistidine fusion protein from cDNA kindly provided by Dr Magnus Hook (Institute of Bioscience and Technology, Texas A&M University). Quantification was by densitometry, using recombinant osteopontin as a standard. Alkaline phosphatase was measured colorimetrically as the hydrolysis of p-nitrophenyl phosphate according to instructions from the supplier (Sigma Diagnostics, St Louis, MO, USA). Aortas were homogenized in radioimmunoprecipitation assay buffer (10 mM Tris, pH 7.4; 2.5 mM EDTA; 50 mM NaF; 1 mM Na4P2O7·10H2O; 1% Triton X-100; 10% glycerol; 1% deoxycholate, 1 μg/ml aprotinin, 0.18 mg/ml PMSF, 0.18 mg/ml orthovanadate, 1% Triton X-100 in 0.9% saline) on ice and centrifuged in a microfuge at maximum speed for 5 min. Supernatant was removed for assay. Aortas were incubated with PTH for 5 min and then frozen in liquid nitrogen. After grinding the aortas in 100 mM HCl at room temperature, cyclic AMP was measured by an enzyme-linked immunoassay (Assay Design Incorporated; Ann Arbor, MI, USA) as described by the manufacturer. Data are presented as mean±s.e. Significance was tested by two-tailed Student's t-test. For multiple comparisons, the P-value was multiplied by the number of comparisons. For skewed data, a logarithmic transformation was applied prior to statistical analysis. This work was supported by grants from the National Institutes of Health (RO1HL47449), The Genzyme Corporation, and the Genzyme Renal Innovations Program.