ABSTRACT. Hyperphosphatemia is thought to underlie medial vascularcalcification in advanced renal failure, but calcification canoccur in other conditions in the absence of hyperphosphatemia,indicating that additional factors are important. To identifythese factors, a model of medial calcification in rat aortain vitro was developed. Aortic rings from rats were incubatedin serum-free medium for 9 d, and calcification was measuredas incorporation of 45Ca and confirmed by histology and x-raydiffraction. No calcification occurred in normal vessels despiteelevated free Ca2+ and PO43– concentrations of 1.8 mMand 3.8 mM, respectively, but mechanical injury resulted inextensive calcification in the media. Co-incubation studiesrevealed that normal aortas produced a soluble inhibitor ofcalcification in injured vessels that was destroyed by alkalinephosphatase. Culture of normal aortas with alkaline phosphataseresulted in calcification of the elastic lamina identified ashydroxyapatite by x-ray diffraction. This effect of alkalinephosphatase was not due to dephosphorylation of osteopontin(OPN), and calcification was not increased in aortas from OPN-deficientmice. The inhibitor was identified as pyrophosphate on the basisof the calcification induced in aortas cultured with inorganicpyrophosphatase, the inhibition of calcification in injuredaortas by pyrophosphate, and the production of inhibitory levelsof pyrophosphate by normal aortas. No calcification occurredunder any conditions at a normal PO43– concentration.It is concluded that elevated concentrations of Ca2+ and PO43–are not sufficient for medial vascular calcification becauseof inhibition by pyrophosphate. Alkaline phosphatase can promotecalcification by hydrolyzing pyrophosphate, but OPN is not anendogenous inhibitor of calcification in rat aorta.
Arterial calcification is common in patients with advanced renalfailure and ESRD and is thought to contribute to their increasedcardiovascular mortality (1). Two distinct forms of calcificationare recognized (2,3). Intimal calcification occurs in atheromatousdisease and is associated with inflammatory cells (3), whereasmedial calcification occurs in the matrix between smooth musclecells in the absence of atherosclerosis and inflammatory cells(2,4). Medial calcification commonly occurs in advanced renalfailure (4,5), where it is thought to result from plasma concentrationsof Ca2+ and PO43– that exceed the solubility product forcalcium phosphate. However, medial calcification is also seenin diabetes and with aging in the presence of normal serum Ca2+and PO43– concentrations (6), indicating that hyperphosphatemiais not required for medial calcification.
Considerable data suggest that vascular calcification is a spontaneousevent, even at normal calcium and phosphate concentrations,that is prevented by inhibitory factors within the vessel wall.Several proteins have been implicated in this process. Micedeficient in matrix Gla protein (MGP) develop rapid and severemedial calcification (7), and a similar phenotype is seen inrats that are treated with warfarin to inhibit -carboxylationof MGP (8). Osteopontin (OPN), which is abundant at sites ofmedial calcification (2,9), inhibits hydroxyapatite crystallizationin vitro (10) and calcification in cultured vascular smoothmuscle cells (11,12). Although deficiency of OPN in mice doesnot lead to vascular calcification, it does accelerate calcificationin MGP-deficient mice (13). Osteoprotegerin (14,15) and fetuin(16,17) are additional proteins implicated in the inhibitionof ectopic calcification.
Elucidation of the pathophysiology of medial calcification hasbeen hampered by the lack of an appropriate in vitro model.Cultures of vascular smooth muscle cells lack the architectureand matrix of a normal vessel. The rapid conversion of thesecells to a proliferative, secretory phenotype and the use ofgrowth factors are also problematic. To address this, we developeda model in intact vessels during long-term culture that allowedus to examine calcification both histologically and quantitatively.Using this model, we identified specific properties of vesselsand medium that influence medial calcification.
Aortic Culture
Aortas (from the arch to the renal arteries) were removed ina sterile manner from male Sprague-Dawley rats that weighed150 to 300 g. After most of the adventitia was gently removedby careful dissection, the vessels were cut into 2- to 3-mmrings and placed in culture medium for up to 9 d. We previouslyused this technique to demonstrate aldosterone-responsivenessof smooth muscle potassium fluxes for up to 7 d (18), indicatingthat viability is maintained in culture. Entire removal of adventitiabefore culture resulted in variable degrees of calcification,presumably as a result of smooth muscle injury. Some aortaswere purposely injured by rubbing the abluminal surface 30 timeswith a cotton swab. In some vessels, selective removal of endotheliumwas accomplished by twirling a loop of 3-0 nylon monofilamentsuture in the lumen (19).
Culture Media
Aortic segments were placed into DMEM (Mediatech, Herndon, VA)with penicillin and streptomycin but no other additives andmaintained at 37°C in a 5% CO2 atmosphere with medium changesevery 3 d. Serum was not added to culture medium because itcaused outgrowth and proliferation of smooth muscle cells (notshown) and others have reported adverse effects of serum oncultured vessels (20). DMEM contains 1.8 mM Ca2+ and 0.9 mMPO43–. The [PO43–] was increased to 3.8 mM by adding3 vol of 100 mM NaH2PO4 (unbuffered) to 100 vol of DMEM thathad been equilibrated with 5% CO2. The high-PO43– mediumwas maintained in 5% CO2 to prevent alkalinization and possibleprecipitation of calcium phosphate. The absence of precipitationwas confirmed by measuring 45Ca2+ concentration before and aftercentrifugation. Final free concentrations of Ca2+ in the low-and high-PO43– media were 1.76 and 1.63 mM as calculatedby CaBuffer software on the basis of the PO43– concentrationsand the concentration of Mg2+ (0.81 mM) and glycine (0.4 mM)in DMEM. Assuming that 43% of total Ca in human serum is ionized(21), these concentrations are equivalent to total calcium concentrationsin human serum of 16.4 and 15.2 mg/dl. The serum phosphorusconcentrations that correspond to the normal and high-phosphateDMEM are 2.7 and 11.4 mg/dl. Some media were supplemented with7.5 U/ml calf intestinal alkaline phosphatase (Promega, Madison,WI) or 1.2 to 12 U/ml inorganic pyrophosphatase from Bakers’yeast (Sigma Diagnostics, St. Louis, MO). Conditioned mediumwas prepared by incubating four to five rings in 500 µlof normal DMEM for 3 d. Some conditioned medium was either filteredthrough a 10,000 molecular weight cut-off filter (Centricon,Amicon Corp.) or dialyzed against DMEM overnight at 100:1 withone change of the dialysate using a Spectra/Por membrane, 6000to 8000 molecular weight cut-off (Spectrum Medical Industries,Los Angeles, CA) before use. Phosphate was added just beforethe use of the medium in culture.
Calcification Assay
Approximately 0.3 µCi/ml 45Ca (DuPont-NEN, Boston, MA)was added to the culture medium, and at the end of the culturethe aortic rings were washed five times in a HEPES-buffered(pH 7.4) physiologic salt solution that contained 1.8 mM Ca2+and 0.9 mM PO43–. The rings were then dried in an oven,weighed, and then dissolved in equal volumes of 70% H2O2 and60% HClO4, and radioactivity was measured by liquid scintillation.Results are expressed as nanomoles calcium per milligram oftissue.
Tissue Analyses
For histologic analysis, samples were placed in formalin andprocessed for paraffin embedding. Hematoxylin and eosin staining,von Kossa staining (silver nitrate plus nuclear fast red), andtoluidine blue staining were according to standard protocols.For viability staining, aortic segments were incubated with0.5 mg/ml methylthiazoletetrazolium (MTT) in DMEM for 3 h at37°C and washed in physiologic saline three times and thenembedded for frozen sectioning. For x-ray diffraction, a smallpiece of dried aorta was mounted on the end of a glass fiber,and various rotation frames were taken with a Bruker D8 x-raydiffractometer at 23°C using a SMART 1000 CCD detector andmonochromatized CuK radiation. The frames were processed, analyzed,and compared with standard samples by using GADDS-NT V4.0 software(Bruker AXS, Madison, WI).
Alkaline Phosphatase
Alkaline phosphatase was measured colorimetrically as the hydrolysisof p-nitrophenyl phosphate according to instructions from thesupplier (Sigma Diagnostics). Aortas were homogenized with 1%Triton X-100 in 0.9% saline on ice and centrifuged in a microfugeat maximum speed for 5 min. Supernatant was removed for assay.
OPN
Immunoblots were performed after separation on a 10% SDS polyacrylamidegel and blotting onto PVDF membranes, using a mouse monoclonalantibody (MPIIIB101) created by M. Solursh and A. Franzen andobtained from the Developmental Studies Hybridoma Bank developedunder the auspices of the National Institute of Child Healthand Human Development and maintained by the University of Iowa,Department of Biologic Sciences (Iowa City, IA). Electrophoresisunder nondenaturing conditions was performed on 6% polyacrylamidewithout SDS and mercaptoethanol. Rat recombinant OPN was producedas a hexahistidine fusion protein from cDNA (provided by Dr.Magnus Hook, Institute of Bioscience and Technology, Texas A&MUniversity). For analysis of OPN mRNA, total RNA was extractedfrom aortas by the phenol-chloroform method, separated on a1% agarose gel, and probed with rat OPN cDNA (obtained fromDr. Robert Taylor, Emory University). OPN-deficient mice wereprovided by Dr. Lucy Liaw (Maine Medical Center Research Institute).
Pyrophosphate
Measurement of pyrophosphate was performed enzymatically (22).Briefly, culture medium was incubated with [14C]uridinediphosphoglucose(UDPG; DuPont-NEN) and UDPG pyrophosphorylase, phosphoglucomutase,glucose-6-phosphate dehydrogenase, and NADP. Production of [14C]6-phosphogluconicacid was measured after removal of [14C]UDPG with activatedcharcoal. Analysis of pyrophosphate hydrolysis was by thin layerchromatography using [32P]pyrophosphate (23).
Statistical Analyses
Data are presented as means ± SE. Significance was determinedby t test.
The histology of aortic rings maintained in DMEM for 9 d isshown in Figure 1. Hematoxylin and eosin staining (Figure 1A)revealed normal-appearing smooth muscle with intact endothelium.For assessing cell viability, vessels were incubated with MTT,a technique commonly used to assess cell viability (24). Thiscompound is converted by mitochondria to an insoluble compoundthat precipitates and thus stains viable cells. As shown inFigure 1B, all of the smooth muscle cells between the elasticlamina were stained, and this did not differ from fresh aortas(Figure 1C), indicating that all of the smooth muscle cellswere viable after 9 d in culture. Identical results were obtainedin vessels cultured in high-phosphate medium, and electrophoresisof DNA from 9-d cultures revealed no fragmentation of DNA indicativeof apoptosis (not shown). By comparison, there was no stainingof aortas that were heated at 65°C for 10 min (Figure 1D).Despite the high concentrations of calcium and phosphate inthe medium, no calcification on was observed by von Kossa stainingafter 9 d (Figure 2A), and staining remained negative in aortasthat were cultured for up to 21 d (not shown). However, extensivestaining was visible after 9 d in aortas that were purposelyinjured by rubbing the abluminal surface before culture (Figure 2B).With this injury, 20% of smooth muscle cells did not stainwith MTT (not shown), and higher magnification revealed calcificationof both cells and elastic laminae (Figure 2C). This was notdue to loss of endothelium since selective removal of endotheliumbefore culture did not result in calcification (not shown).
Figure 1. Histology of rat aortas after culture for 9 d in standard medium. (A) Hematoxylin and eosin stain. (B) Cultured aorta stained with methylthiazoletetrazolium (MTT; 10-µ section). (C) Freshly isolated aorta stained with MTT (10-µ section). (D) Aorta heated at 65°C for 10 min before staining with MTT (10-µ section). L, vessel lumen. Magnifications: x400 in A, x200 in B and C, x300 in D.
Figure 2. Calcification of cultured rat aortas detected by von Kossa staining. There was no staining of normal aortas after 9 d in culture (A) but diffuse staining of injured aortas (B). Higher magnification (C) revealed diffuse staining of the cells as well as linear staining of the elastic lamina between cells (arrows). Magnifications: x40 in A and B, x400 in C.
Calcification was quantitated as incorporation of 45Ca fromthe medium (Figure 3). Because some calcification of residualadventitia occurred, it was removed before 45Ca content wasmeasured so that only medial calcification was measured. Toensure that all adventitia was removed, this was performed witha dissecting microscope and confirmed by histology (not shown).Consistent with the negative von Kossa stain, there was verylittle incorporation of 45Ca into normal aortas cultured inhigh-phosphate DMEM. All of this was incorporated in the firstday, and cultures up to 21 d showed no additional incorporation(not shown). In contrast, calcium incorporation into injuredvessels showed a steady increase over time (Figure 3A). Basalincorporation into uninjured aortas decreased slightly but significantlywhen [PO43–] was decreased to 0.9 mM (Figure 3B). Additionof 10 mM -glycerophosphate instead of phosphate, which inducescalcification in cultures of smooth muscle cells (25), did notincrease Ca incorporation in normal aortas. Calcium incorporationincreased >100-fold in injured aortas but only when the phosphateconcentration was elevated or -glycerophosphate was present.The slight increase in calcium incorporation in injured aortascompared with uninjured aortas in 0.9 mM [PO43–] was NS.
Figure 3. Incorporation of 45Ca into aortic rings. (A) Time course in injured aortas (•) and uninjured aortas () in DMEM containing 1.8 mM Ca2+ and 3.8 mM PO43–. Results are the means of at least five (injured) or four (uninjured) aortic rings. (B) Calcification under different conditions. Aortic rings were incubated for 9 d in DMEM containing 1.8 mM Ca2+ and either 0.9 or 3.8 mM PO43– as indicated. The concentration of -glycerophosphate was 10 mM. Aortas were injured by rubbing the abluminal surface. Numbers at bottom of bars indicate the number of aortic segments studied. *P < 0.001 versus 0.9 mM PO43–; **P < 0.001 versus uninjured aorta in 3.8 mM PO43–.
Calcification of injured aortas was substantially reduced whenthey were co-incubated with uninjured aortas or incubated inconditioned medium from uninjured aortas (Figure 4), suggestingthat a soluble inhibitor of calcification is released by normalaorta. There was no significant calcification of uninjured aortaswhen they were co-incubated with injured aortas or incubatedwith conditioned medium from injured aortas. Inhibition of calcificationby conditioned medium was reversed by adding alkaline phosphatase,suggesting that the inhibitor was a phosphorylated compound.Addition of alkaline phosphatase during culture also inducedcalcium incorporation into uninjured aortas (Figure 5A), indicatingthat this inhibitor also prevents calcification under normalconditions. X-ray diffraction patterns from alkaline phosphatase–treatedaorta and bone are shown in Figure 5B. The location and intensityof the rings, particularly the inner two, are identical andconsistent with hydroxyapatite. However, these data do not ruleout the presence of additional forms of calcium phosphate.
Figure 4. Effect of co-incubation and conditioned medium on aortic calcification in high-PO43– medium. For control incubations of normal or injured aortas, single rings were incubated in 500 µl of medium. For co-incubations, single injured or uninjured rings were cultured with four uninjured or injured rings, respectively. Single rings were incubated in 500 µl of conditioned medium prepared by incubating four to five rings in 500 µl of normal DMEM for 3 d. Culture of injured aortas with uninjured aortas or medium from uninjured aortas prevented calcification, and this soluble factor was destroyed by addition of 7.5 U/ml alkaline phosphatase. Results are means of at least five aortic rings. *P < 0.001 versus normal medium.
Figure 5. Induction of calcification by alkaline phosphatase. (A) Time course. Normal aortas were incubated in DMEM containing 3.8 mM PO43– and 7.5 units/ml calf intestinal alkaline phosphatase with medium changes every 3 d. Adventitia was removed before measuring 45Ca incorporation. Results are means of at least six aortic segments. (B) X-ray diffraction after 9 d.
Staining of aortas cultured with alkaline phosphatase revealedmedial calcification (Figure 6A), and under higher power, thevon Kossa stain showed linear staining of the elastic laminabetween the smooth muscle cells (Figure 6B). To identify theinitial site of calcification, we examined earlier stages ofcalcification by culturing aortas for 6 d or in a lower phosphateconcentration (2.8 mM). When silver nitrate staining was repeatedafter incubation of alkaline phosphatase–cultured aortaswith MTT (Figure 6C), calcification was clearly visible in theelastic lamina between viable cells. Localization of the stainingto elastic lamina is also apparent after staining with toluidineblue (Figure 6D).
Figure 6. Histology of normal aortas cultured with alkaline phosphatase as described in the legend of Figure 5. (A) von Kossa stain showing scattered foci of calcification as well as aggregates. (B) Enlargement of a x400 image of a von Kossa stain showing linear calcification (arrows) localized to the elastic laminae between cells. (C) Silver nitrate stain performed after incubation with MTT (9-d culture, 2.8 mM PO4). The calcified elastic laminae (stained brown, arrows) are surrounded by viable cells (stained black). (D) Toluidine with MTT blue stain of elastic laminae shows areas of dark staining consistent with calcification in the laminae, without any calcification of the cells between the laminae (6-d culture, 3.8 mM PO4). Magnifications: x200 in A, x400 in B through D.
OPN production was investigated to determine whether it wasresponsible for the inhibition of calcification and the effectof alkaline phosphatase. OPN mRNA was not detected in freshlyisolated aortas but appeared during culture in normal DMEM (Figure 7A),suggesting transcriptional induction. Likewise, there wasvery little OPN production (measured in the medium) by freshlyisolated aortas, but this increased substantially during culture(Figure 7B). OPN was also produced by injured aortas (Figure 7C),and the rate, as judged by densitometry, was similar tothat for uninjured aortas (uninjured: 734 ± 104 ng/mgper d; injured: 741 ± 119 ng/mg per d). On nondenaturingpolyacrylamide gels (Figure 7D), OPN in culture medium migratedfaster than rat recombinant OPN, and this faster migration waseliminated by pretreatment with alkaline phosphatase, indicatingthat all of the OPN produced by normal and injured aorta wasphosphorylated.
Figure 7. Osteopontin (OPN) production in cultured rat aorta. (A) Northern blot of mRNA from cultured rat aorta using probe against rat OPN probes and mouse glyceraldehyde-3-phosphate dehydrogenase. (B) Daily OPN production during culture of normal aortas. Culture medium was collected and replaced daily, and 20 µl of medium was analyzed by SDS-PAGE and immunoblotting. Numbers indicate the days on which the medium was collected. Left lane, 40 ng of recombinant rat osteopontin. (C) OPN production in injured and uninjured aorta during 3 d of culture. Left lane, 40 ng of rat recombinant OPN; inj, injured aortas; unlabeled lanes, uninjured aortas. By densitometry, OPN production per weight did not differ between injured and uninjured aortas. (D) Phosphorylation state of OPN in culture medium. PAGE was performed under nondenaturing conditions on normal DMEM from injured aortas (lane 1) or uninjured aortas (lane 2) cultured for 3 d. Lane 3, medium from uninjured aortas treated with 7.5 U/ml calf intestinal alkaline phosphatase for 2 h at 37°C; lane 4, recombinant rat OPN. All results are representative of at least three experiments. The results show that rat aorta produces phosphorylated OPN, and this is unchanged by injury.
Inhibition of calcification by conditioned medium persistedafter filtration to exclude proteins >10 kD in size and wasremoved by dialysis using a membrane with a size exclusion of6 to 8 kD (Figure 8A). Immunoblotting revealed that OPN, whichis 32 kD, was retained by the dialysis membrane and not presentin the filtrate (Figure 8B). This was not consistent with OPNas the inhibitor of calcification in conditioned medium.
Figure 8. Role of OPN in rat aortic calcification. (A) Aortas were incubated in DMEM or conditioned medium from normal aortas, both containing 3.8 mM PO43–, for 9 d as described in Figure 4. Conditioned medium was either filtered through a 10,000 molecular weight cut-off filter (Centricon, Amicon Corp.) or dialyzed against DMEM before use. Results are means of at least 10 aortic rings. *P < 0.001 versus injured aorta in normal medium. (B) Immunoblot of OPN after filtration and dialysis. Lane 1, recombinant rat OPN; lane 2, conditioned medium. OPN is retained after dialysis (lane 3) and removed by filtration (lane 4). Thus, calcification is independent of OPN content in the medium.
Pyrophosphate was also investigated as a possible inhibitorof calcification. It is not present in DMEM, but its concentrationafter 3 d of culture was 0.44 ± 0.03 µM (one ringin 500 µl of medium), indicating that it was producedby aortas. These measurements were made in normal DMEM to avoidsequestration of pyrophosphate in calcium phosphate deposits(26). Elimination of pyrophosphate by adding inorganic pyrophosphataseinduced calcification of normal aortas (Figure 9A). This concentrationof pyrophosphatase resulted in complete disappearance of [32P]pyrophosphatefrom the culture medium (Figure 9B) but did not dephosphorylateOPN (Figure 9C). Von Kossa staining (Figure 9, D and E) revealedextensive calcification of the media, again with calcificationof elastin fibers.
Figure 9. Inhibition of vascular calcification by pyrophosphate. (A) Incorporation of calcium in aortas incubated for 9 d in DMEM containing 3.8 mM PO43– with or without 1.25 units/ml inorganic pyrophosphatase added daily. Results are means of at least 10 aortic rings. (B) Autoradiogram of thin-layer chromatography showing complete hydrolysis of pyrophosphate. Lane 1, [32P]pyrophosphate; lane 2, medium from aortas cultured with [32P]pyrophosphate for 24 h; lane 3, medium from aortas cultured with [32P]pyrophosphate and 1.25 U/ml inorganic pyrophosphatase for 24 h; Pi, orthophosphate; Ppi, pyrophosphate; Orig, origin. (C) Immunoblot showing that pyrophosphatase does not dephosphorylate OPN. Lane 1, conditioned medium from normal aorta. Treatment with alkaline phosphatase (lane 2) retards migration, indicating dephosphorylation. This does not occur after treatment with pyrophosphatase (lane 3). (D) Von Kossa stain of aorta cultured with pyrophosphatase for 9 d. Higher magnification (E) shows both diffuse and focal calcification. 1.25 U/ml pyrophosphatase. Magnifications: x20 in D, x400 in E.
Addition of pyrophosphate prevented calcification in injuredaortas (Figure 10), confirming that pyrophosphate inhibits medialcalcification. There was no inhibition with 2.5 µM butalmost complete inhibition with 10 µM pyrophosphate. Onthe basis of the rate of hydrolysis of [32P]pyrophosphate inaortic cultures (not shown), the estimated concentrations 3d after adding 5, 10, and 30 µM pyrophosphate were 1.8,3.1, and 7.9 µM, respectively. Thus, pyrophosphate isactually a more potent inhibitor of calcification than indicatedin the Figure 10. The appearance rate of pyrophosphate in culturemedium was substantially reduced in injured aortas (36 ±4 pmol/mg per d, n = 12, versus 145 ± 8 pmol/mg per din normal aortas, n = 22; P < 0.001). Alkaline phosphataseactivity after 9 d of culture was significantly increased ininjured aortas (1.16 ± 0.17 units/mg, n = 15 versus 0.43+ 0.04 units/mg, n = 12, in uninjured aorta; P < 0.001).The activity in cultured, uninjured aorta was identical to thatin freshly isolated aorta (0.44 ± 0.05 units/mg, n =10). No alkaline phosphatase activity was detected in the culturemedium of normal or injured aortas.
Figure 10. Suppression of calcification in injured aortas by pyrophosphate. Injured aortas were incubated for 6 d in DMEM containing 3.8 mM PO43– and varying concentrations of pyrophosphate. Results are means of at least four aortic rings.
The relative roles of pyrophosphate and OPN in inhibiting calcificationwere also examined in aortas from OPN-deficient mice. Theseaortas did not exhibit any greater calcification than aortasfrom wild-type mice after 9 d in high-PO43– medium (Figure 11),and they calcified to the same extent in response to injuryor culture with alkaline phosphatase or pyrophosphatase.
Figure 11. Calcification of aortas from OPN-deficient mice. Aortas were cultured for 9 d in high-PO43– DMEM with or without 7.5 units/ml alkaline phosphatase 1.25 units/ml pyrophosphatase and medium changes every 3 d. Results are the means of at least six aortic rings.
This study demonstrates that medial calcification can be inducedin intact rat aorta cultured with alkaline phosphatase or inorganicpyrophosphatase. The calcification is in the form of hydroxyapatite,requires a high PO43– concentration, and is histologicallysimilar to the calcification observed in vessels from uremicpatients (5,27) and rats with chronic renal failure (28). Thatthere was no histologic evidence of cell death and that thecalcification occurred in the matrix between cells rather thanwithin cells argue against dystrophic calcification. Rat aortascultured without these enzymes and not subjected to injury exhibitedno calcification in the high-PO43– medium, even up to21 d in culture. The small, initial incorporation of 45Ca undernormal conditions presumably represents equilibration with intracellularCa and Ca normally bound to extracellular matrix because itdid not increase over time. Concentrations of both Ca2+ andPO43– are elevated in high-PO43– medium comparedwith human serum and, on the basis of free concentrations, wouldbe equivalent to a total calcium-phosphorus product in humanserum of 180 mg2/dl2, which is well above generally acceptedclinical thresholds. Thus, a supraphysiologic elevation of thecalcium-phosphorus product is not sufficient to produce medialcalcification in vitro in this model. Vascular calcificationis a chronic process in vivo, and we cannot rule out the possibilitythat longer culture times are required to observe calcificationof normal vessels in vitro. However, the absence of any increasein 45Ca deposition over 3 wk argues against this. It is alsopossible that plasma factors not present in culture or mechanicaleffects such as blood flow and pulsatile vessel distension couldcontribute to calcification in vivo.
The absence of calcification was due to inhibitory activityin normal aortas, and this inhibition could be explained bythe release of pyrophosphate from smooth muscle. Alkaline phosphataseand inorganic pyrophosphatase induced calcification of normalaortas, and pyrophosphate inhibited calcification of injuredaortas. Pyrophosphate inhibits hydroxyapatite formation in vitro(26,29,30), and exogenous pyrophosphate inhibits aortic calcificationin rats that are given large doses of vitamin D3 (31,32). Diphosphonates,which are analogs of pyrophosphate, exhibit the same properties(33,34). It is likely that the inhibition by endogenous pyrophosphatedemonstrated in cultured rat aortas also occurs in vivo becausethe concentration that maximally inhibited calcification ininjured aortas (3 µM) is similar to that reported fornormal human plasma (35–38). Furthermore, deficiency ofPC-1, an ecto-ATPase that produces pyrophosphate, results inreduced plasma pyrophosphate levels and extensive arterial calcificationin humans (39), which can be prevented with diphosphonate therapy.Mice lacking ANK, a putative pyrophosphate transporter, exhibitreduced pyrophosphate production and extensive ectopic calcification(40), although not in vessels. Serum was not used in the aorticculture because it causes smooth muscle proliferation and outgrowthand can have detrimental effects on smooth muscle (20). Thus,it is possible that there are additional circulating inhibitorsin vivo, such as fetuin (16,17), that would not be apparentin culture.
Whether alterations in pyrophosphate production or clearanceplay a role in the medial vascular calcification of advancedrenal failure is unknown. Measurements performed more than twodecades ago revealed slightly elevated plasma levels of pyrophosphatein patients who underwent hemodialysis (33), but dialytic clearancecould be greater with current membranes and dialysis delivery.Hydrolysis of pyrophosphate by alkaline phosphatase may alsoplay a role. The amount of this enzyme used to produce calcificationin rat aorta was only severalfold higher than normal serum activity,and elevated serum activity of alkaline phosphatase is associatedwith calcific uremic arteriolopathy (41,42). That plasma pyrophosphateconcentration is increased in hypophosphatasia, a genetic deficiencyof alkaline phosphatase, indicates that this enzyme does influencepyrophosphate metabolism in vivo (35).
The calcification that occurred in injured aortas was associatedwith reduced levels of pyrophosphate in the medium and was preventedby exogenous pyrophosphate, suggesting a deficiency of pyrophosphate.This deficiency could be due to decreased production becausethere is loss of viable cells. However, this loss (as judgedfrom the MTT staining) was far less than the reduction in mediumpyrophosphate. Alternatively, the deficiency of pyrophosphatecould have resulted from the increased alkaline phosphataseactivity in injured aortas. Medial calcification occurs afterballoon injury of rabbit aortas (43), indicating that injuryinduces calcification in vivo as well.
A surprising finding was that OPN did not correlate with calcificationin rat aortas, which is in distinct contrast to its inhibitionof calcification in cultures of rat aortic smooth muscle cells(11,12). Although inhibition of calcification was not testeddirectly by adding phosphorylated OPN to the cultures, phosphorylatedOPN was produced by cultured aortas and levels present in culturemedium (4 µg/ml after 3 d) were equivalent to concentrationsthat fully inhibited calcification in cultured cells (12). Nodephosphorylated OPN was detected, consistent with the low activityof alkaline phosphatase in aortas. Removal of OPN did not affectthe ability of conditioned medium to inhibit calcification ininjured aortas, and the degree of calcification produced byalkaline phosphatase, which dephosphorylates OPN but also destroyspyrophosphate, was no greater than that produced by inorganicpyrophosphatase, which does not dephosphorylate OPN. However,we cannot rule out the possibility that other unknown factorsthat influence calcification are also affected by these maneuvers.Last, calcification was not increased in aortas from OPN-deficientmice. It is important to note that although OPN was producedin culture, there was very little OPN mRNA or OPN productionin freshly isolated aortas. This is consistent with previousstudies (2,44) and indicates that OPN is not available to inhibitcalcification of normal vessels in vivo. These data suggestthat OPN is not an endogenous inhibitor of calcification innormal vessels and are consistent with the lack of vascularcalcification in OPN-deficient mice in vivo (45,46). OPN becomesabundant in calcified vessels (2,9), and OPN deficiency doesaccelerate vascular calcification in mice lacking matrix Glaprotein (13), suggesting that OPN may modulate calcificationat a more advanced stage.
Cultured aortas provide an opportunity to study calcificationof intact vessels in vitro. The normal histology and the normalsmooth muscle potassium fluxes and sensitivity to agonists (18)that we have previously demonstrated indicate that vessel structureand viability are maintained in culture. Furthermore, the calcificationobserved in rat aortas that were treated with alkaline phosphataseor pyrophosphatase was histologically similar to that observedin uremic human vessels (5) and occurred, at least initially,in the elastic laminae. Calcification of aortas in culture differedfrom that described in cultures of smooth muscle cells in severalrespects. -Glycerophosphate, which induces calcification incultured cells, was without effect in normal aorta. This isprobably due to the 100-fold greater activity of alkaline phosphatasein cultured cells than in aorta (11), which would increase theavailability of PO43– from -glycerophosphate and eliminatethe inhibitory effect of pyrophosphate. OPN inhibits calcificationin cultured cells (11,12), but in cultured aorta, the same concentrationof OPN (present endogenously) does not prevent calcification.Because of these differences and because cultured cells lackthe elastic laminae that are the initial site of calcification,we believe that aorta cultured with alkaline phosphatase orpyrophosphatase more accurately reflects medial vascular calcificationin vivo and will be a useful model for future studies.
Acknowledgments
Supported by a research grant (HL47449) and a major instrumentationgrant (S10-RR13673) from the National Institutes of Health andan unrestricted grant from Genzyme Corporation.
Goodman WG, Goldin J, Kuizon BD, Yoon C, Gales B, Sider D, Wang Y, Chung J, Emerick A, Greaser L, Elashoff RM, Salusky IB: Coronary-artery calcification in young adults with end-stage renal disease who are undergoing dialysis. N Engl J Med 342: 1478–1483, 2000[Abstract/Free Full Text]
Shanahan CM, Cary NRB, Salisbury JR, Proudfoot D, Weissberg PL, Edmonds ME: Medial localization of mineralization-regulating proteins in association with Monckeberg’s sclerosis: Evidence for smooth muscle cell-mediated vascular calcification. Circulation 100: 2168–2176, 1999[Abstract/Free Full Text]
Jeziorska M, McCollum C, Woolley DE: Calcification in atherosclerotic plaque of human carotid arteries: Associations with mast cells and macrophages. J Pathol 185: 10–17, 1998[CrossRef][Medline]
Ibels LS, Alfrey AC, Huffer WE, Craswell PW, Anderson JT, Weil R: Arterial calcification and pathology in uremic patients undergoing dialysis. Am J Med 66: 790–796, 1979[CrossRef][Medline]
Ejerblad S, Ericsson JLE, Erikson I: Arterial lesions of the radial artery in uraemic patients. Acta Chir Scand 145: 415–428, 1979[Medline]
Everhart JE, Pettitt DJ, Knowler WC, Rose FA, Bennett PH: Medial arterial calcification and its association with mortality and complications of diabetes. Diabetologia 31: 16–23, 1988[Medline]
Luo G, Ducy P, McKee MD, Pinero GJ, Loyer E, Behringer RR, Karsenty G: Spontaneous calcification of arteries and cartilage in mice lacking matrix Gla protein. Nature 386: 78–81, 1997[CrossRef][Medline]
Price PA, Faus SA, Williamson MK: Warfarin causes rapid calcification of the elastic lamella in rat arteries and heart valves. Arterioscler Thromb Vasc Biol 18: 1400–1407, 1998[Abstract/Free Full Text]
Ahmed S, O’Neill KD, Hood AF, Evan AP, Moe SM: Calciphylaxis Is associated with hyperphosphatemia and increased osteopontin expression by vascular smooth muscle cells. Am J Kidney Dis 37: 1267–1276, 2001[Medline]
Denhardt DT, Guo X: Osteopontin: A protein with diverse functions. FASEB J 7: 1475–1482, 1993[Abstract]
Wada T, McKee MD, Steitz S, Giachelli CM: Calcification of vascular smooth muscle cell cultures. Inhibition by osteopontin. Circ Res 84: 166–178, 1999[Abstract/Free Full Text]
Jono S, Peinado C, Giachelli CM: Phosphorylation of osteopontin is required for inhibition of vascular smooth muscle cell calcification. J Biol Chem 275: 20197–20203, 2000[Abstract/Free Full Text]
Speer MY, McKee MD, Guldberg RE, Liaw L, Yang H-Y, Tung E, Karsenty G, Giachelli CM: 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 196: 1047–1055, 2002[Abstract/Free Full Text]
Min H, Morony S, Sarosi I, Dunstan CR, Caparelli C, Scully S, Van G, Kaufman S, Kostenuik PJ, Lacey DL, Boyle WJ, Simonet WS: Osteoprotegerin reverses osteoporosis by inhibiting endosteal osteoclasts and prevents vascular calcification by blocking a process resembling osteoclastogenesis. J Exp Med 192: 463–474, 2000[Abstract/Free Full Text]
Bucay N, Sarosi I, Dunstan C, Moroney S, Tarpley J, Capparelli C, Scully S, Tan HL, Xu W, Lacey DL, Boyle WJ, Simonet WS: Osteoprotegerin-deficient mice develop early onset osteoporosis and arterial calcification. Genes Dev 12: 1260–1268, 1998[Abstract/Free Full Text]
Jahnen-Dechent W, Schinke T, Trindl A, Muller-Esterl W, Sablitzky F, Kaiser S, Blessing M: Cloning and targeted deletion of the mouse fetuin gene. J Biol Chem 272: 31496–31503, 1997[Abstract/Free Full Text]
Schinke T, Amendt C, Trindl A, Poschke O, Muller-Esterl W, Jahnen-Dechent W: The serum protein a2-HS glycoprotein/fetuin inhibits apatite formation in vitro and in mineralizing calvaria cells. J Biol Chem 271: 20789–20796, 1996[Abstract/Free Full Text]
Jiang G, Cobbs S, Klein JD, O’Neill WC: Aldosterone regulates the Na-K-2Cl cotransporter in vascular smooth muscle. Hypertension 41: 1131–1135, 2003[Abstract/Free Full Text]
Fingerle J, Au YP, Clowes AW, Reidy MA: Intimal lesion formation in rat carotid arteries after endothelial denudation in absence of medial injury. Arteriosclerosis 10: 1082–1087, 1990[Abstract/Free Full Text]
Zeidan A, Nordstrom I, Dreja K, Malmqvist U, Hellstrand P: Stretch-dependent modulation of contractility and growth in smooth muscle of rat portal vein. Circ Res 87: 228–234, 2000[Abstract/Free Full Text]
Aurbach GD, Marx SJ, Spiegel AM: Parathyroid hormone, calcitonin, and the calciferols. In: Williams Textbook of Endocrinology, 7th Ed., edited by Wilson JD, Foster DW, Philadelphia, W.B. Saunders, 1985, pp 1137–1217,
Cheung CP, Suhadolnik RJ: Analysis of inorganic pyrophosphate at the picomole level. Anal Biochem 83: 61–63, 1977[CrossRef][Medline]
Grobben B, Anciaux K, Roymans D, Stefan C, Bollen M, Esmans EL, Slegers H: An ecto-nucleotide pyrophosphatase is one of the main enzymes involved in the extracellular metabolism of ATP in rat C6 Glioma. J Neurochem 72: 826–834, 1999[CrossRef][Medline]
Loo DT, Rillema JR: Measurement of cell death. Methods Cell Biol 57: 251–264, 1998[Medline]
Shioi A, Nishizawa Y, Jono S, Koyama H, Hosoi M, Morii H: b-Glycerophosphate accelerates calcification in cultured bovine vascular smooth muscle cells. Arterioscler Thromb Vasc Biol 15: 2003–2009, 1995[Abstract/Free Full Text]
Meyer JL: Can biological calcification occur in the presence of pyrophosphate? Arch Biochem Biophys 231: 1–8, 1984[CrossRef][Medline]
Sakata N, Noma A, Yamamoto Y, Okamoto K, Meng J, Takebayashi S, Nagai R, Horiuchi S: Modification of elastin by pentosidine is associated with the calcification of aortic media in patients with end-stage renal disease. Nephrol Dial Transplant 18: 1601–1609, 2003[Abstract/Free Full Text]
Katsura K, Kusek JW, Hirata Y, Tsunemi K, Nagano N, Burke SK, Fukushima N: Sevelamer hydrochloride prevents ectopic calcification and renal osteodystrophy in chronic renal failure rats. Kidney Int 64: 441–450, 2003[CrossRef][Medline]
Fleisch H, Neuman WF: Mechanism of calcification: role of collagen, polyphosphates, and phosphatase. Am J Physiol 200: 1296–1300, 1961[Abstract/Free Full Text]
Fleisch H, Russell GG, Straumann F: Effect of pyrophosphate on hydroxyapatite and its implications in calcium homeostasis. Nature 212: 901–903, 1966
Fleisch H, Schibler D, Maerki J, Frossard I: Inhibition of aortic calcification by means of pyrophosphate and polyphosphates. Nature 207: 1300–1301, 1965[Medline]
Schibler D, Russell GG, Fleisch H: Inhibition by pyrophosphate and polyphosphate of aortic calcification induced by vitamin D3 in rats. Clin Sci 35: 363–372, 1968[Medline]
Russell RGG, Bisaz S, Fleisch H: Pyrophosphate and diphosphates in calcium metabolism and their possible role in renal failure. Arch Intern Med 124: 571–575, 1969[CrossRef][Medline]
Francis MD, Russell RGG, Fleisch H: Diphosphonates inhibit formation of calcium phosphate crystals in vitro and pathologic calcification in vivo. Science 165: 1264–1266, 1969[Abstract/Free Full Text]
Russell RGG, Bisaz S, Donath A, Morgan DB, Fleisch H: Inorganic pyrophosphate in plasma in normal persons and in patients with hypophosphatasia, osteogenesis imperfecta, and other disorders of bone. J Clin Invest 50: 961–969, 1971
Rachow JW, Ryan LM: Inorganic pyrophosphate metabolism in arthritis. Rheum Dis Clin North Am 14: 289–302, 1988[Medline]
Whyte MP: Hypophosphatasia and the role of alkaline phosphatase in skeletal mineralization. Endocr Rev 15: 439–461, 1994[CrossRef][Medline]
Fleisch H, Bisaz S: Mechanism of calcification: inhibitory role of pyrophosphate. Nature 195: 911, 1962
Rutsch F, Vaingankar S, Johnson K, Goldfine I, Maddux B, Schauerte P, Kalhoff H, Sano K, Boisvert WA, Superti-Furga A, Terkeltaub RA: PC-1 nucleotide triphosphate pyrophosphohydrolase deficiency in idiopathic infantile arterial calcification. Am J Pathol 158: 543–554, 2001[Abstract/Free Full Text]
Ho AM, Johnson MD, Kingsley DM: Role of the mouse ank gene in control of tissue calcification and arthritis. Science 289: 265–270, 2000[Abstract/Free Full Text]
Schwarz G, Stork H: The role of pyrophosphate and alkaline phosphatase in skin calcification (calciphylaxis). Experientia 23: 295–296, 1967[CrossRef][Medline]
Mazhar AR, Johnson RJ, Gillen D, Stivelman JC, Ryan MJ, Davis CL, Stehman-Breen C: Risk factors and mortality associated with calciphylaxis in end-stage renal disease. Kidney Int 60: 324–332, 2001[CrossRef][Medline]
Gadeau A-P, Chaulet H, Daret D, Kockx M, Daniel-Lamaziere J-M, Desgranges C: Time course of osteopontin, osteocalcin, and osteonectin accumulation and calcification after acute vessel wall injury. J Histochem Cytochem 49: 79–86, 2001[Abstract/Free Full Text]
Hao H, Hirota S, Tsukamoto Y, Imakita M, Ishibashi-Ueda H, Yutani C: Alterations of bone matrix protein mRNA expression in rat aorta in vitro. Arterioscler Thromb Vasc Biol 15: 1474–1480, 1995[Abstract/Free Full Text]
Liaw L, Birk DE, Ballas CB, Aagenaes O, Whitsitt JS, Davidson JM, Hogan BLM: Altered wound healing in mice lacking a functional osteopontin gene (spp1). J Clin Invest 101: 1468–1478, 1998[Medline]
Rittling SR, Matsumoto HN, McKee MD, Nanci A, An XR, Novick KE, Kowalski AJ, Noda M, Denhardt DT: Mice lacking osteopontin show normal development and bone structure but display altered osteoclast formation in vitro. J Bone Miner Res 13: 1101–1111, 1998[CrossRef][Medline]
Received for publication September 25, 2003.
Accepted for publication March 4, 2004.
This article has been cited by other articles:
A. Covic, P. Kothawala, M. Bernal, S. Robbins, A. Chalian, and D. Goldsmith Systematic review of the evidence underlying the association between mineral metabolism disturbances and risk of all-cause mortality, cardiovascular mortality and cardiovascular events in chronic kidney disease
Nephrol. Dial. Transplant.,
November 11, 2008;
(2008)
gfn613v1.
[Abstract][Full Text][PDF]
R. C. Shroff, R. McNair, N. Figg, J. N. Skepper, L. Schurgers, A. Gupta, M. Hiorns, A. E. Donald, J. Deanfield, L. Rees, et al. Dialysis Accelerates Medial Vascular Calcification in Part by Triggering Smooth Muscle Cell Apoptosis
Circulation,
October 21, 2008;
118(17):
1748 - 1757.
[Abstract][Full Text][PDF]
L. L. Demer and Y. Tintut Vascular Calcification: Pathobiology of a Multifaceted Disease
Circulation,
June 3, 2008;
117(22):
2938 - 2948.
[Full Text][PDF]
K. A. Johnson, M. Polewski, and R. A. Terkeltaub Transglutaminase 2 Is Central to Induction of the Arterial Calcification Program by Smooth Muscle Cells
Circ. Res.,
March 14, 2008;
102(5):
529 - 537.
[Abstract][Full Text][PDF]
K. A. Johnson, W. Yao, N. E. Lane, P. Naquet, and R. A. Terkeltaub Vanin-1 Pantetheinase Drives Increased Chondrogenic Potential of Mesenchymal Precursors in ank/ank Mice
Am. J. Pathol.,
February 1, 2008;
172(2):
440 - 453.
[Abstract][Full Text][PDF]
I. Nikolov, N. Joki, T. Drueke, and Z. Massy Beyond phosphate--role of uraemic toxins in cardiovascular calcification
Nephrol. Dial. Transplant.,
December 1, 2006;
21(12):
3354 - 3357.
[Full Text][PDF]
R. C. Johnson, J. A. Leopold, and J. Loscalzo Vascular Calcification: Pathobiological Mechanisms and Clinical Implications
Circ. Res.,
November 10, 2006;
99(10):
1044 - 1059.
[Abstract][Full Text][PDF]
W. C. O'Neill Pyrophosphate, Alkaline Phosphatase, and Vascular Calcification
Circ. Res.,
July 21, 2006;
99(2):
e2 - e2.
[Full Text][PDF]
X. Qin, M. A. Corriere, L. M. Matrisian, and R. J. Guzman Matrix Metalloproteinase Inhibition Attenuates Aortic Calcification
Arterioscler. Thromb. Vasc. Biol.,
July 1, 2006;
26(7):
1510 - 1516.
[Abstract][Full Text][PDF]
M. Ketteler, G. Schlieper, and J. Floege Calcification and Cardiovascular Health: New Insights Into an Old Phenomenon
Hypertension,
June 1, 2006;
47(6):
1027 - 1034.
[Full Text][PDF]
C. M. Shanahan Vascular calcification--a matter of damage limitation?
Nephrol. Dial. Transplant.,
May 1, 2006;
21(5):
1166 - 1169.
[Full Text][PDF]
P. A. Price, W. Si Chan, D. M. Jolson, and M. K. Williamson The Elastic Lamellae of Devitalized Arteries Calcify When Incubated in Serum: Evidence for a Serum Calcification Factor
Arterioscler. Thromb. Vasc. Biol.,
May 1, 2006;
26(5):
1079 - 1085.
[Abstract][Full Text][PDF]
J. L. Reynolds, J. N. Skepper, R. McNair, T. Kasama, K. Gupta, P. L. Weissberg, W. Jahnen-Dechent, and C. M. Shanahan Multifunctional Roles for Serum Protein Fetuin-A in Inhibition of Human Vascular Smooth Muscle Cell Calcification
J. Am. Soc. Nephrol.,
October 1, 2005;
16(10):
2920 - 2930.
[Abstract][Full Text][PDF]
K. A. Lomashvili, W. Khawandi, and W. C. O'Neill Reduced Plasma Pyrophosphate Levels in Hemodialysis Patients
J. Am. Soc. Nephrol.,
August 1, 2005;
16(8):
2495 - 2500.
[Abstract][Full Text][PDF]
C. Henley, M. Colloton, R. C. Cattley, E. Shatzen, D. A. Towler, D. Lacey, and D. Martin 1,25-Dihydroxyvitamin D3 but not cinacalcet HCl (Sensipar(R)/Mimpara(R)) treatment mediates aortic calcification in a rat model of secondary hyperparathyroidism
Nephrol. Dial. Transplant.,
July 1, 2005;
20(7):
1370 - 1377.
[Abstract][Full Text][PDF]
K. Johnson, M. Polewski, D. van Etten, and R. Terkeltaub Chondrogenesis Mediated by PPi Depletion Promotes Spontaneous Aortic Calcification in NPP1-/- Mice
Arterioscler. Thromb. Vasc. Biol.,
April 1, 2005;
25(4):
686 - 691.
[Abstract][Full Text][PDF]
Top
Abstract
Introduction
-carboxylation of MGP (8). Osteopontin (OPN), which is abundant at sites of medial calcification (2,9), inhibits hydroxyapatite crystallization in vitro (10) and calcification in cultured vascular smooth muscle cells (11,12). Although deficiency of OPN in mice does not lead to vascular calcification, it does accelerate calcification in MGP-deficient mice (13). Osteoprotegerin (14,15) and fetuin (16,17) are additional proteins implicated in the inhibition of ectopic calcification. 
radiation. The frames were processed, analyzed, and compared with standard samples by using GADDS-NT V4.0 software (Bruker AXS, Madison, WI). 
20% of smooth muscle cells did not stain with MTT (not shown), and higher magnification revealed calcification of both cells and elastic laminae (Figure 2C). This was not due to loss of endothelium since selective removal of endothelium before culture did not result in calcification (not shown). 
-glycerophosphate instead of phosphate, which induces calcification in cultures of smooth muscle cells (25), did not increase Ca incorporation in normal aortas. Calcium incorporation increased >100-fold in injured aortas but only when the phosphate concentration was elevated or 
) in DMEM containing 1.8 mM Ca2+ and 3.8 mM PO43–. Results are the means of at least five (injured) or four (uninjured) aortic rings. (B) Calcification under different conditions. Aortic rings were incubated for 9 d in DMEM containing 1.8 mM Ca2+ and either 0.9 or 3.8 mM PO43– as indicated. The concentration of 
