Printer Friendly

Effects of LDL, cholesterol, and their oxidized forms on the precipitation kinetics of calcium phosphates.

Vascular calcification, as a hallmark of clinically significant atherosclerotic lesions (1), may begin as early as the second decade of life just after fatty streak formation (2). The heterogeneous deposits in calcified lesions are mainly calcium apatite (17%) and carbonate (9%) (3). Proposed starting sites for calcification include matrix vesicles derived from cells and lipoproteins in plasma or originally internalized by cells (4) as well as cholesterol (CH) (3) in tissue sections of lesions (5,6). Some serum proteins (7), glycoprotein, and matrix Gla protein (8) inhibit calcification, and a deficiency of such proteins may be involved in the pathogenesis of vascular calcifications. A suggested mechanism for the inhibition of mineral growth involves the adsorption of serum proteins to the growing hydroxyapatite (HAP) crystals. On the other hand, LDL, CH, and their oxidized forms promote calcification and are known cardiovascular risk factors (9,10), although little is known about whether they are involved directly in the formation of calcium phosphates.

Because the various risk factors involved in vascular calcification are not necessarily present in one pathologic stage of lesion development, vascular calcification may proceed through different mechanisms. For example, vascular calcification may initiate with matrix vesicle formation and mineralization, following a process similar to that in bone (11). Campbell and Campbell (12) and Steitz et al. (13) demonstrated the association of smooth cell phenotypic transition with vascular calcification, indicating that calcification in advanced atherosclerotic lesions may be an organized, regulated process similar to bone formation and may even have purpose (14). On the other hand, the temporal and spatial association of CH with HAP in the early stages of atherosclerosis (5) and the structural compatibility between the (001) crystalline planes of both HAP and CH monohydrate (15) imply that a physicochemical mechanism is also possible under certain circumstances.

The role of CH monohydrate in HAP formation was anticipated by Craven (15) in 1976. On the basis of crystal structure analysis of both CH monohydrate and HAP, Craven proposed that, with the intervention of a transition layer of hydrogen-bonded water molecules, a microcrystal of either structure may serve as a nucleus for growth of the other. In 1990, Hirsch et al. (16) reported the precipitation of CH on HAP seeds that were either added or formed in situ in solutions containing ethanol. The close spatial relationship between the two kinds of crystals was further confirmed by Hirsch et al. (5) and Sarig et al. (17) with confocal fluorescence microscopy. They demonstrated the association of CH with the surface of HAP seeds and the incorporation of CH within calcium phosphate-CH agglomerates produced in vitro, although the conditions under which they carried out their experiments were physiologically impossible because of the presence of large amounts of ethanol in their solutions. Most importantly, they found that CH was within apatite particles isolated from human atherosclerotic lesions, confirming that calcification around a CH nucleus can happen in vivo (17).

In addition to CH, CH ester [e.g., cholesterol linoleate (CHE)], oxidized CH in the form of cholestane-3[beta],5[alpha],6[beta]-triol (OX-CH), lipids, LDL, and its oxidized form (Ox-LDL) are often found in atherosclerotic lesions (4,18-20). The CH that accumulates in atherosclerotic lesions originates primarily in plasma lipoproteins, including LDL (9,21). To gain insight into the roles of LDL, CH, and their oxidized forms during the early stages of vascular calcification, we investigated the effects of these substances on the precipitation of calcium phosphates in aqueous solutions under constant physiologic pH and 37 [degrees]C. We found that CH is the only one of the investigated compounds that accelerated the nucleation phase in vitro. This finding, which is in agreement with the results reported by Hirsch et al. (5) and Sarig et al. (17), provides a connection between the in vitro and in vivo studies. Because no CH-induced precipitation of calcium phosphates has been reported for bone formation, this finding may also reveal one of the main differences between cardiovascular calcification and bone formation.

Materials and Methods


CH (C3045), OX-CH (C2523), CHE (C0289), L-[alpha]-phosphatidylcholine (PC; P5638), Ca[Cl.sub.2] x 2 [H.sup.2]O (C7902), Na[N.sub.3] (S2002), bovine albumin fraction V (A9674), and thiobarbituric acid (TBA; 5500) were all purchased from Sigma. 1,1,3,3-Tetraethoxypropane was obtained from Merck. Folin-phenol reagent for protein content determinations was obtained from Ding-Guo Co. Other reagents were of analytical grade. The water used for preparing solutions was deionized and freshly doubly distilled.

For the calcium chloride stock solution, Ca[Cl.sub.2] x 2[H.sub.2]O and KCI were dissolved in an appropriate volume of water. The solution pH was adjusted to 7.4 with NaOH on a pH meter (ZD-2 digital autotitrator), and the solution was filtered twice before use through 0.45 /,tm nylon syringe filters (Beijing Chemicals). The mean (SD) final concentrations of calcium in the filtrate were 69.3 (0.6) mmol/L for experiments with LDL and Ox-LDL and 34.5 (0.1) mmol/L for experiments with CH, OX-CH, CHE, and PC, as measured by inductively coupled plasma atomic emission spectrophotometry (ICP-AES; Leeman Labs Inc.). The filtrate contained 0.150 mol/L KCI to maintain the ionic strength and 2.0 mmol/L Na[N.sub.3] to prevent microbial growth.

For the phosphate stock solution, the desired amounts of K[H.sub.2]P[O.sub.4], KCl, and Na[N.sub.3] were dissolved in water. The solution pH was adjusted to 7.4 with NaOH on a pH meter (ZD-2), and the solution was filtered twice before use through 0.45 [micro]m nylon syringe filters. The final concentrations of KCI and Na[N.sub.3] were 0.150 mol/L and 2.0 mmol/L, respectively, as in the calcium chloride stock solution. The phosphate concentrations of the solutions were 46.9 (0.4) mmol/L for experiments with LDL and Ox-LDL and 23.5 (0.2) mmol/L for experiments with cholesterol, OX-CH, CHE, and PC, as measured by ICP-AES.

The KOH solution was prepared under a nitrogen atmosphere from washed KOH pellets and contained 2.0 mmol/L Na[N.sub.3]. The exact concentration was 50.20 mmol/L as titrated with potassium hydrogen phthalate. The solution was filtered twice before use through 0.45 [micro]m nylon syringe filters.

We prepared the phosphate-buffered saline (PBS) for dialysis of protein solutions by dissolving 8.0001 g of NaCl, 0.2001 g of KCI, 3.5804 g of [Na.sub.2]HP[O.sub.4] x 12 [H.sub.2]O, 0.2015 g of K[H.sub.2]P[O.sub.4] and 0.1999 g of Na[N.sub.3] in water, titrating the solution with KOH solution to pH 7.4, and bringing the final volume up to 1.0 L with water.

The dialysis tubing ([M.sub.r] 8000-14 000; Union Carbide) used throughout the experiments was boiled in solution containing 20 g/L NaHC[O.sub.3] and 1.0 mmol/L disodium EDTA for 10 min, followed by washing vigorously with water. The dialysis tubing was dipped in water for more than 5 h, and the water was changed once every hour during the period. The tubing was then boiled in water for 10 min and kept at <4 [degrees]C in water containing 2.0 mmol/L Na[N.sub.3] before use.

LDL was provided by the Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences. The concentrations of the protein ranged from 3 to 4 g/L, and the solutions, buffered at pH 7.4, contained 9 g/kg NaCl, 0.3 g/kg Na[N.sub.3], 0.10 g/kg EDTA, and 20.0 mmol/L Tris. To remove the EDTA from the protein solution, 1.7-mL aliquots of protein solution were dialyzed ([M.sub.r] 8000-14 000 cutoff dialysis tubing) in the dark against 500 mL PBS at 4 [degrees]C for 24 h (two changes of 500 mL of PBS).

Aliquots of the dialysate were used for preparation of Ox-LDL by the method of Liu et al. (22). To the protein solution, [Cu.sup.2+] was added to a final concentration of 10 [micro]mol/L. The mixture was then sealed and incubated at 37 [degrees]C for various times to achieve oxidation. The reaction was ended by the addition of EDTA up to 200 [micro]mol/L. Aliquots of the solution were used to determine the extent of oxidation (see below), and the remaining solution was dialyzed as above to remove [Cu.sup.2+] and EDTA. The protein content was determined with Folin-phenol reagent as described by Lowry et al. (23), using bovine serum albumin as the calibrator.

The extent of oxidation of the protein was measured as the amount of TBA-reactive substances (TBARS) according to the published procedure (24, 25). Solutions containing LDL or Ox-LDL (containing 25 [micro]g of protein) were mixed with 1.5 mL of 6.7 g/L TBA and 1.5 mL of 200 mL/L [Cl.sub.3]CCOOH. After heating at 100 [degrees]C for 30 min, the product was assayed fluorescently on a spectrofluorometer (RF-5301PC; Shimadzu) with excitation at 515 run and emission at 535 run. The amount of TBARS was estimated by comparison of the produced fluorescence intensity of the sample solution (F) with that of the standard substance, tetramethoxypropane ([F.sup.S]) according to the formula: 40OX F/[F.sup.S] (nmol/mg of protein).

To prepare suspensions containing CH, OX-CH, or CHE, we dissolved the appropriate amount of the substances (15.0 or 30.0 mg of CH, 15.0 mg of OX-CH, 30.0 mg of CHE, 30.0 mg of PC, and 30.0 mg of PC plus 15.0 mg of CH) in CH[Cl.sub.3] in a 100-mL round-bottomed flask. The solvent was then removed by use of a rotary evaporator. We added 10 mL of 0.150 mol/L KCl to the flask under nitrogen, and then sonicated the solution for 30-40 min after 30 min of hydration. The distribution of the particle sizes in the suspensions was measured on a LA-300 particle size analyzer (HORIBA).


The pH-stat technique as described by Hunter et al. (26) was used for the kinetic study of calcium phosphate precipitation. Stock solutions of Ca[Cl.sub.2] and K[H.sub.2]P[O.sub.4] were diluted separately to 16.67 times of the original volume with 150 mmol/L KCl and warmed to 37 [degrees]C in a water bath. Equal volumes (25 mL) of the two solutions were then mixed with protein solution (<1 mL) containing either LDL or Ox-LDL in a water-jacketed Pyrex cell. The total volume of the mixture was ~50 mL. For other experiments, we mixed 100 mL each of diluted Ca[Cl.sub.2] and K[H.sub.2]P[O.sub.4] solutions and added 10 mL of a solution containing CH, OX-CH, or CHE to obtain the desired concentration. The total volume was 210 mL in these cases. The temperature was kept constant at 37.0 [+ or -] 0.2 [degrees]C. For experiments with the protein components, we mixed 20.0 mL of diluted Ca[Cl.sub.2] and 20.0 mL of diluted K[H.sub.2]P[O.sub.4] with 10 mL of solution containing one or two of the components. We rapidly titrated the mixture to pH 7.4 while it was being stirred magnetically at a constant speed and purged with water vapor-saturated nitrogen gas ([N.sub.2] [greater than or equal to] 99.995%) at a constant rate, after which we immediately began timing the reaction. The pH of the solution was then kept constant by the addition of 50 mmol/L KOH under the control of a pH autotitrator (ZD-2) that was equipped with a combinational glass electrode (E201-C9). The volume of KOH solution consumed in maintaining the pH constant was recorded at intervals of 5-60 min. A typical kinetic curve of calcium phosphate precipitation is composed of two parts: a nucleation phase followed by a crystal growth phase. The former refers to the period from time zero to the point at which the curve begins to rise rapidly, and the latter contains the rest of the curve. Both phases could be estimated from the kinetic curve. All kinetic data presented here are the means (SD) of three independent experiments.

At the end of the experiments with CH, OX-CH, or CHE, the deposits were collected by passing the suspension through a 0.22 [micro]m nylon syringe filter and washing the filter three times with ice-cold water. The deposits containing mixed PC/CH vesicles, as well as those containing PC vesicles, were collected by centrifugation at 20 000g for 30 min on a centrifuge (LD4-2A) and washing three times with ice-cold water, followed by drying at 50 [degrees]C.


The x-ray diffraction (XRD) analysis was carried out on a DMAX/2000 with monochromated radiation (Cu K[alpha]; [lambda] = 0.154 run) at a rate of 4 degrees/min. Scanning electron microscope analyses were performed on an S-250MK3 scanning electron microscope equipped with an energy-dispersion x-ray (EDX), and transmission electron microscope analyses were performed on a H-9000NAR transmission electron microscope (Hitachi) equipped with an EDX.

To determine the calcium and phosphate content in vesicles, we dissolved parts of the precipitates in 0.10 mol/L HCl. The lipid component was extracted with CH[Cl.sub.3]. The aqueous phase was separated and analyzed for calcium and phosphorus by ICP-AES.



The three experiments displayed in Fig. 1 are for a control (i.e., without LDL; ??), a sample with low LDL content (14.8 mg/L protein; ??), and a sample with approximately threefold higher LDL (43.1 mg/L protein; ??). The initial concentrations were 2.08 (0.02) mmol/L for calcium and 1.41 (0.01) mmol/L for phosphorus as determined by ICP-AES. The concentration of the titrant KOH was 52.29 mmol/L, as calibrated by potassium hydrogen phthalate. As soon as calcium phosphate precipitates formed, the protons released by [H.sub.2]P[O.sub.4.sup.-] and HP[O.sub.4.sup.2-] were autotitrated by KOH. By recording the consumed volume of KOH along the time course, we could plot a kinetic curve. As shown in Fig. 1, not only did the presence of LDL at the two concentrations (14.8 and 43.1 mg/L protein) produce no obvious differences (P <0.05, t-test) in the kinetics, but the two curves in the presence of LDL and the one in the absence of LDL almost overlapped.

Oxidative modification of LDL remarkably changed the effect of the protein on the kinetics of calcium phosphate precipitation. The control in Fig. 2 is the same as in Fig. 1. As shown in Fig. 2, in the presence of the protein (43.1 mg/L protein) that had been oxidized previously by incubation with [Cu.sup.2+] for 12 h, the nucleation phase was prolonged (*) compared with that of the control (??). At a lower protein concentration (29.5 mg/L protein; ??), the change in the nucleation phase was smaller, although the protein had been oxidized more extensively, as indicated by the data in Table 1. When we compared the three systems (Fig. 2, ??, ??, and *), we observed no significant changes in the crystal rapid growth phase. However, the crystal rapid growth phase did change significantly in the presence of more highly oxidized protein (oxidative modification for 24 h) at a higher protein concentration (43.1 mg/L protein); the portion of that curve (Fig. 2, ??) showing the crystal rapid growth phase is less steep than the corresponding portions of the other curves.




The extent of oxidation of the protein increased with increasing incubation time with [Cu.sup.2+] at 37 [degrees]C. As listed in Table 1, the concentration of TBARS produced [53.2 - 12.8 = 40.4 nmol x [L.sup.-1] x [(mg protein).sup.-1]] after 24 h of oxidation was more than two times higher than that after 12 h of oxidation [28.3 - 12.8 = 15.5 nmol x [L.sup.-1] x [(mg protein).sup.-1]].


Addition of CH microcrystals accelerated the nucleation phase of the precipitation kinetics. The initial concentrations were 1.82 (0.01) mmol/L for calcium and 1.23 (0.01) mmol/L for phosphorus. The concentration of the titrant KOH was 50.24 mmol/L. In the presence of CH as a suspension at a concentration of 71.4 mg/L (Fig. 3A, ??), the nucleation time decreased to 300 min, compared with 390 min for the control system (Fig. 3A, ??). At a higher CH concentration (143 mg/L), the nucleation phase was even shorter (270 min; Fig. 3A, ??).

After CH was distributed into phospholipid vesicles, its promotive effect on the nucleation of calcium phosphates was overwhelmed by the inhibitive effect of PC molecules. As shown in Fig. 3B, without CH, the phospholipid vesicles alone (Fig. 3B, ??) lengthened the nucleation phase to 600 min, more than 150% of that of the control system (390 min; Fig. 3B, ??). In the presence of CH/PC mixed vesicles (Fig. 3B, ??), the nucleation phase was 480 min, indicating that the promotive effect of CH partially counteracted the inhibitive effect of the phospholipid molecules.


OX-CH is one of the main forms of oxidized CH in vivo. As shown in Fig. 4, the oxidized CH made the nucleation phase (420 min; ??) longer than that of the control system (390 min; ??), an effect that was quite different from the nonoxidized ones (Figs. 3A and 4, ??). Similarly, the presence of CHE also lengthened the nucleation phase (630 min; Fig. 4, *).


In the suspensions, the middle values of the particle diameters of CH and OX-CH were 40.64 and 20.78 [micro]m, respectively, as shown in Fig. 5. The specific surface area values of particles in the two systems were 2005.1 and 3635.3 [cm.sup.2]/[cm.sup.3], respectively.



Summarized in Table 2 are the XRD data for the final products obtained in various reaction systems. The symbols 2[theta] and d are the diffraction angle and distance between crystal faces, respectively. The weakest reflections occur for 2[theta] >25 degrees, although reflections also exist for lower 20 values (27). These data are listed downward in the order of the relative diffraction intensity. Compared with the standard values ([d.sub.s]) of the defective HAP [DHAP; JCPDS card no. 460905 for [Ca.sub.9](HP[O.sub.4])[(P[O.sub.4]).sub.5](OH); calcium/phosphorus = 9/6 = 1.50], it seems that these precipitates are principally of DHAP nature, also containing octacalcium phosphate [OCP; [Ca.sub.8][H.sub.2][(P[O.sub.4]).sub.6] x 5 [H.sub.2]O; calcium/phosphorus = 8/6 = 1.33] and HAP [[Ca.sub.10][(P[O.sub.4]).sub.6][(OH).sub.2]; calcium/phosphorus = 10/6 = 1.67].

At the end of the experiments with CH microcrystals as seeds, the precipitates were separated and analyzed by XRD. As shown in Table 3, the d values in the 2[theta] range 13.100-23.520 were characteristic of the CH standard (JCPDS card no. 70742 for [C.sub.27][H.sub.46]O), indicating that CH underwent no chemical changes that led to mineral formation.

The molar ratios of calcium/phosphorus in precipitates were calculated from the ICP-AES analysis as described at the end of the section on characterization of precipitates and are listed in Table 4. The highest and the lowest calcium/phosphorus molar ratios occurred in the system containing CH microcrystals (calcium/phosphorus = 1.490) and phospholipid/CH mixed vesicles (calcium/phosphorus = 1.356), respectively.

The scanning electron microscopic results for the precipitate sections are shown in Fig. 6. The precipitates formed in the presence of CH had an organic core (Fig. 6B, arrow), whereas the amounts of calcium (11%) and phosphorus (17%) are much less than in the "shell" (calcium, 66%; phosphorus, 34%) as determined by the EDX analysis. This phenomenon is distinct from those in other systems (Fig. 6, A, C and D).


The EDX analysis for the precipitate shown in Fig. 6B also indicated that the calcium/phosphorus molar ratios are 1.42, 1.47, and 1.53, respectively, on the outer surface, in the middle of the mineral layer, and at the interface between the mineral layer and the organic core. The crystal characteristics based on XRD analysis of the precipitate at the interface are listed in Table 5. From a comparison of the d values and relative intensities of the diffraction rings, it can be deduced that the crystal structures of the precipitates are very similar to that of HAP [JCPDS card no. 721243 for [Ca.sub.10][(P[O.sub.4]).sub.6][(OH).sub.2]].


Vascular calcification can be affected (or even directed) by various biological factors. Many proteins involved in bone formation have also been found in vascular calcification. However, the possibility cannot be excluded that the vascular calcification possesses some idiosyncratic characteristics different from that of bone formation.

Clinical, epidemiologic, and genetic studies have convincingly demonstrated that LDL promotes atherosclerosis or, specifically, promotes the development of the early fatty-streak lesion (28). Because vascular calcification can also occur in this stage (2), one may expect LDL to play a role in vascular calcification. As is shown in Fig. 1, in the control system, the nucleation phase lasted 120 min before the rapid growth phase began, and no significant difference was observed for systems containing LDL at the two protein concentrations (14.8 and 43.1 mg/L protein). These findings seem to suggest that LDL neither prohibits nor promotes the formation of calcium phosphates.

Whether a protein can influence calcium phosphate precipitation depends on the proteins amino acid sequence and conformation. Studies on bone sialoprotein have suggested that one or both of the proteins glutamic acid-rich sequences are involved in promoting the nucleation of HAP (29), whereas the phosphate and carboxylate groups of osteopontin, possibly including the conserved sequence of contiguous aspartic acid residues, may be required for the proteins inhibitive effect on HAP formation (26). However, the promotive and inhibitive effects of carboxylate and phosphate groups, respectively, seem not applicable to the effects of LDL.


After oxidative modification, LDL was transformed to Ox-LDL. As indicated by the data in Table 1, Ox-LDL that had been produced by oxidation of LDL for 24 h produced twice as many TBARS as Ox-LDL that had been produced by 12 h of oxidation. The oxidized form of the protein contains lecithin and oxidized CH (e.g., triglycerides), in addition to Ox-apoLDL. The concentrations of cholesteryl ester and oxysterols are lower and higher, respectively, in Ox-LDL than in LDL. The increased negative charges on the surface of these particles endow Ox-LDL with higher calcium-binding capacity, an effect that reduces the concentration of [Ca.sup.2+] in the solution. As shown in Fig. 2, both the nucleation and crystal rapid growth phases can be inhibited by Ox-LDL. It is noteworthy that at the same concentration, the difference in oxidative content produced a difference in the crystal rapid growth phase (Fig. 2, ?? and *), whereas the difference in the concentration of the oxidized protein, regardless of the extent of oxidation, made a difference only in the length of nucleation phase (Fig. 2, ??, ??, and *). More intensive oxidization of LDL inhibited both the nucleation and crystal growth phases (Fig. 2, ??). This fact indicates that binding of calcium ions on Ox-LDL significantly diminished the ionic concentration.

As shown in Table 1, even nonoxidized LDL contains a small amount of oxidized material. One may expect that a small shift would be observed in the curve in Fig. 1, at least for the highest protein content (43.1 mg/L protein). However, this tendency was not observed under experimental conditions. One explanation is that the small amount of oxidized material in nonoxidized LDL was contained inside the core of the LDL or buried in the lipid monolayer, whereas in Ox-LDL the conformation, as well as the composition, of the protein changed.

CH, OX-CH, and CHE, being independent vascular risk factors, respectively, may also be taken as the components of LDL and Ox-LDL. To understand which component of LDL and Ox-LDL is responsible for their effects, we also studied these substances. As seen in Figs. 3 and 4, it was the nucleation stage that was affected significantly. As soon as ion clusters in the supersaturated solutions grew to the critical size, nuclei formed. Afterward, the rapid growth stage commenced and was insensitive to the effects of these protein component substances.

CH is the only one of the investigated additives that accelerated the nucleation of calcium phosphates (Fig. 3A). This finding supports the hypothesis of enhanced interaction between some crystal surfaces of CH and those of certain calcium phosphates. Scanning electron microscopic analysis on the section of a precipitate particle showed that there is an organic core inside the particle (Fig. 6B), and the composition of the core is, as indicated by the XRD data in Table 3, CH that did not undergo any chemical change under the experimental conditions.

Around the core is an inorganic shell. At physiologic conditions there are four well-defined calcium phosphates: dicalcium phosphate dihydrate (DCPD; CaHP[O.sub.4] x 2[H.sub.2]O), OCP, [beta]-dicalcium phosphate [[beta]-TCP; [Ca.sub.3][(P[O.sub.4]).sub.2]], and HAP (30). In addition, there is also a less-defined calcium phosphate: calcium-deficient HAP (30). Among these minerals, DCPD, OCP, and [beta]-TCP have been proposed as the precursors in calcium phosphate precipitation (31). At pH 7.4 and 37 [degrees]C, the molar ratio of HP[O.sub.4.sup.2-] to [H.sub.2]P[O.sub.4.sup.-] is ~5:1 [calculated with the secondary dissociation constants of phosphoric acid, p[K.sub.a2] = 6.7 at 37 [degrees]C and an ionic strength of 150 mmol/L (32)]. Because the initial concentrations are 1.82 mmol/L for [Ca.sup.2+] and 1.23 mmol/L for phosphorus, we have the concentration product Ca x HP[O.sub.4.sup.2-] = 1.82 x 1.23 x 5/6 x [10.sup.-6] = 1.87 x [10.sup.-6] [(mol/L).sup.2]. This value is only slightly larger than the solubility product [1.8 x [10.sup.-6] [(mol/L).sup.2]] of DCPD at 37 [degrees]C and an ionic strength of 150 mmol/L (33). Because a small amount of [Ca.sup.2+] forms a complex, CaHP[O.sub.4](aq) (apparent dissociation constant [~10.sup.-1.86]) (34), it is unlikely that DCPD would precipitate spontaneously. In addition, the calcium/phosphorus ratio values for minerals with cholesterol (1.49 in Table 4; 1.42, 1.47, and 1.53 around the organic core by EDX analysis; see the section on product analysis) are different from that of DCPD (calcium/phosphorus = 1.0). The formation of [beta]-TCP seems less possible at ambient temperatures (31), although its calcium/phosphorus ratio (1.5) is close to the measured values. Even on the [beta]-TCP seed crystals, it was OCP that formed, instead of [beta]-TCP (35). Hence, [beta]-TCP can also be excluded.

Several results suggest that the mineral layer around the cholesterol core is DHAP [JCPDS card no. 460905 for [Ca.sub.9](HP[O.sub.4])[(P[O.sub.4]).sub.5](OH); calcium/phosphorus = 9/6 = 1.50]. Listed in Table 2 are the XRD data for the final products obtained in various reaction systems. For the inorganic species separated in the experiments with CH, the 2[theta] values at 25.94 degrees (crystal face 002) and 31.70 degrees (crystal face 211; Table 2, column 11) are characteristic of HAP. The crystal face characteristics corresponding to the diffraction rings produced by the precipitates at the inorganic-organic interface are very close to those of HAP (Table 5). The calcium/phosphorus molar ratios of the inorganic compound in contact with the organic core were 1.50 and 1.53, respectively, by scanning electron microscopy-EDX and transmission electron microscopy-EDX analysis, which is consistent with the composition of DHAP. Therefore, the HAP nature of the precipitates at the inorganic-organic interface is supported by both the crystallographic and composition similarities between the precipitates and HAP. However, it could as well be that the mineral is HAP interlayered with OCP (27).

Craven (15) reported that crystals of both HAP and CH monohydrate usually develop (001) faces. The hydroxyl ions of HAP are in planes parallel to (001) with O-H bonds favorably oriented for hydrogen bonding at the (001) crystal surface. A fit of the (001) planes shows close superposition of hydrogen bonding groups from the two crystal structures. Unlike the cocrystallization of CH and calcium phosphate done by Hirsch et al. (16), our results (Fig. 6B and Table 5) provide the first experimental evidence for the nucleation of HAP on CH seeds in vitro in aqueous solution at pH 7.4, 37 [degrees]C, and an ionic strength of 150 mol/L. This finding, as a counterpart of those reported by Hirsch et al. (5) and Sarig et al. (17), provides a connection between the in vitro and in vivo results. Furthermore, because there has been experimental confirmation for both CH precipitation on HAP seeds (16) and HAP precipitation on CH seeds (present study), Craven's (15) notion that "a microcrystal of either structure may serve as a nucleus for growth of the other" has been substantiated.

The calcium/phosphorus molar ratios of the precipitates, 1.47 in the middle of the inorganic layer and 1.42 on the outer surface, are between those of HAP (calcium/ phosphorus = 1.67) and OCP (calcium/phosphorus = 1.33). These results are consistent with the findings of Nancollas and Tomazi (36) that the rapidly formed mineral phase on HAP seeds is OCP, based on the similarities in part of their crystalline structure (37). The calcification seems to start with slow precipitation of (DHAP) HAP on microcrystal nuclei of CH monohydrate, followed by rapid deposition of OCP on HAP. The slowness of the nucleation should be the result of the structural compatibility between the crystals of CH monohydrate and HAP. Because the rapid growth phase is ~9 h and the duration of the experiment was 14 h, the transformation of kinetically favored OCP to more stable HAP should not be complete.

Because calcifiable vesicles may have a role in plaque calcification (38), much work has been done on phospholipid vesicles as substitutes of matrix vesicles in simulative biomineralization research. Both promotive and inhibitive roles have been reported for various phospholipid vesicles. Our results show that phospholipid vesicles (Fig. 3B, ??), contrary to the promotive effect of CH microcrystals, made the nucleation phase even longer. This effect of lipid molecules was partially counteracted by CH distributed in the vesicles (Fig. 3B, ??). Because lipids and CH, from destroyed cells or accumulated in the vascular intima directly from blood plasma, may form calcified sites in vivo (6), our findings may be of physiologic and pathologic importance.

The calcification-inducing effect has been reported for vesicles composed of phosphatidylserine (39, 40). This has been attributed to the negative charges carried by the phospholipid molecule at pH 7.4. The attraction of the negative carboxylic groups of phosphatidylserine to calcium cations may finally lead to the formation of the heterotrimetric phosphatidylserine-[Ca.sup.2+]-P[O.sub.4.sup.3-] complex that initiates nucleation on vesicle surface by reducing the activation energy. This concentration effect facilitates the nucleation and growth of calcium phosphates. Vogel (41) reported that electrically neutral lipids, e.g., phosphatidyl glycerol, did not induce calcium phosphate precipitation. Under our experimental conditions, PC exhibited an inhibitive effect. It seems that only the calcium cations were attracted to the phospholipid molecules, diminishing the calcium concentration while leaving the phosphate anions in solution, thus producing a longer nucleation phase.


As some of the products of LDL oxidation, both CHE and OX-CH may be present in blood vessels. Because Ox-LDL inhibits the nucleation stage of calcium phosphate precipitation (Fig. 2), one may expect that the components of Ox-LDL operate similarly. Indeed, as indicated by the nucleation times shown in Fig. 4, the inhibitive effect of the less polar CHE is stronger than that of OX-CH on the precipitation. The lack of promotive effect of the two kinds of substances on nucleation may be attributed to the absence of stereochemical matching, implying that the two substances do not serve as nucleation sites in calcium phosphate precipitation. The specific reason for the differences in the inhibitory capacities of CHE and OX-CH remains unknown. It is noteworthy that CH acted as a promoter for calcification, whereas OX-CH acted as an inhibitor, although the specific surface area of the former is smaller than that of the latter (Fig. 5; a larger particle size corresponds to a smaller specific surface area). These observations show the importance of structural compatibility in vascular calcification.

In conclusion, we investigated the physicochemical effects of LDL, CH, and their oxidized forms. LDL is not involved directly in the precipitation of calcium phosphates. Ox-LDL is inhibitive to both nucleation and crystal growth, although it promotes atherosclerosis biologically. Because of the structural compatibility in the crystalline form, CH microcrystals serve as seeds for inducing the precipitation of HAP, whereas PC, OX-CH, and CHE exhibit inhibitive effects on the nucleation of calcium phosphates.

This project was funded by the Natural Sciences Foundation of China (Grants 20031010 and 20271005). T.L.Z. was also supported by Peking University (985 Project) and the State Key Laboratory of Natural and Biomimetic Drugs.

Received July 11, 2003; accepted September 3, 2003.

DOI: 10.1373/clinchem.2003.024513


(1.) Parhami E, Tintut Y, Patel JK, Mody N, Hemmat A, Demer LL. Regulation of vascular calcification in atherosclerosis. Z Kardiol 2001;90:11127-30.

(2.) Stary HC. The sequence of cell and matrix changes in atherosclerotic lesions of coronary arteries in the first forty years of life. Eur Heart J 1990;11(Suppl E):3-19.

(3.) Schmid K, McSharry W0, Pameijer CH, Binette JP. Chemical and physicochemical studies on the mineral deposits of the human atherosclerotic aorta. Atherosclerosis 1980;37:199-210.

(4.) Stary HC, Chandler AB, Dinsmore RE, Fuster V, Glagov S, Insull W Jr, et al. A definition of advanced types of atherosclerotic lesions and a histological classification of atherosclerosis. Circulation 1995;92:1355-74.

(5.) Hirsch D, Azoury R, Sarig S, Kruth HS, Colocalization of cholesterol and hydroxyapatite in human atherosclerotic lesions. Calcif Tissue Int 1993;52:94-8.

(6.) Bobryshev YV, Lord RSA, Warren BA. Calcified deposit formation in intimal thickenings of the human aorta. Atherosclerosis 1995; 118:9-21.

(7.) Garnett J, Dieppe P. The effects of serum and human albumin on calcium hydroxyapatite crystal growth. Biochem J 1990;266: 863-8.

(8.) Ketteler M, Vermeer C, Wanner C, Westenfeld R, Jahnen Dechent W, Floege J. Novel insights into uremic vascular calcification: role of matrix Gla protein and [alpha]-2-Heremans Schmid glycoprotein/ fetuin. Blood Purif 2002;20:473-6.

(9.) Steinberg DS. Oxidative modification of LDL and atherosclerosis. Circulation 1997;95:1062-71.

(10.) Jialal I, Devaraj S. Low-density lipoprotein oxidation, antioxidants, and atherosclerosis: a clinical biochemistry perspective. Clin Chem 1996;42:498-506.

(11.) Tintut Y, Demer LL. Recent advances in multifactorial regulation of vascular calcification. Curr Opin Lipidol 2001;12:555-60.

(12.) Campbell GR, Campbell JH. Vascular smooth muscle and arterial calcification. Z Kardiol 2000;89(Suppl 2):54-62.

(13.) Steitz SA, Speer MY, Curinga G, Yang HY, Haynes P, Aebersold R, et al. Smooth muscle cell phenotypic transition associated with calcification. Circ Res 2001;89:1147-54.

(14.) Doherty TM, Detrano RC. Coronary arterial calcification as an active process: a new perspective on an old problem. Calcified Tissue Int 1994; 54:224-30.

(15.) Craven BM. Crystal structure of cholesterol monohydrate. Nature 1976;260:727-9.

(16.) Hirsch D, Azoury R, Sarig S, Co-crystallization of cholesterol and calcium phosphate as related to atherosclerosis. J Crystal Growth 1990;104:759-65.

(17.) Sarig S, Weiss TA, Katz J, Kahana F, Azoury R, Okon E, Kruth HS. Detection of cholesterol associated with calcium mineral using confocal fluorescence microscopy. Lab Invest 1994;71:782-7.

(18.) Goldstein JL, Brown MS. The low density lipoprotein pathway and its relation to atherosclerosis. Annu Rev Biochem 1977;46:897930.

(19.) Boyd HC, Gown AM, Wolfbauer G, Chait A. Direct evidence for a protein recognized by a monoclonal antibody against oxidatively modified LDL in atherosclerotic lesions from a Watanabe heritable hyperlipidemic rabbit. Am J Pathol 1989;135:815-25.

(20.) Esterbauer H, Gebicki J, Puhl H, Jurgens G. The role of lipid peroxidation and antioxidants in oxidative modification of LDL. Free Radic Biol Med 1992;13:341-90.

(21.) Newman HAI, Zilversmit DB. Quantitative aspects of cholesterol flux in rabbit atheromatous lesions. J Biol Chem 1962;237: 2078-84.

(22.) Liu SX, Zhou M, Chen Y. Comparative study of oxidative modification and malondialdehyde modification of low density lipoprotein. Acta Biochim Biophys Sinica 1992;24:569-74.

(23.) Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the folin phenol reagent. J Biol Chem 1951;193:265-75.

(24.) Yagi K. A simple fluorometric assay for lipoperoxide in blood plasma. Biochem Med 1976;15:212-6.

(25.) Strinbrecher UP, Parthasarathy S, Leake DS, Witztum JL, Steiberg D. Modification of low density lipoprotein by endothelial cells involves lipid peroxidation and degradation of low density lipoprotein phospholipids. Proc Natl Acad Sci U S A 1984;81:3883-7.

(26.) Hunter GK, Kyle CL, Goldberg HA. Modulation of crystal formation by bone phosphoproteins: structural specificity of the osteopontin-mediated inhibition of hydroxyapatite formation. Biochem J 1994; 300:723-8.

(27.) Elliott JC, Studies in inorganic chemistry 18. Structure and chemistry of the apatites and other calcium orthophosphates. Amsterdam: Elsevier, 1994:404pp.

(28.) Usui S, Kakuuchi H, Okamoto M, Mizukami Y, Okazaki M. Differential reactivity of two homogeneous LDL-cholesterol methods to LDL and VLDL subfractions, as demonstrated by ultracentrifugation and HPLC. Clin Chem 2002;48:1946-54.

(29.) Hunter GK, Goldberg HA. Modulation of crystal formation by bone phosphoproteins: role of glutamic acid-rich sequences in the nucleation of hydroxyapatite by bone sialoprotein. Biochem J 1994;302:175-9.

(30.) Lundager-Madsen HE. Heterogeneous nucleation of calcium phosphates. I. Kinetics of nucleation of tetracalcium monohydrogen phosphate on crystals of calcium monohydrogen phosphate dihydrate. Acta Chem Scand 1970;24:1677-86.

(31.) Heughebaert JC, Nancollas GH. Kinetics of crystallization of octacalcium phosphate. J Phys Chem 1984;88:2478-81.

(32.) Duffield JR, Edwards K, Evans DA, Morrish DM, Vobe RA, Williams DR. Low molecular mass aluminum complex speciation in biofluids. J Coord Chem 1991;23:277-90.

(33.) Moreno EC, Gregory TM, Brown WE. Solubility of CaHP[O.sub.4] x 2[H.sub.2]O and the formation of ion pairs in the system Ca(OH)[.sub.2] - [H.sub.3]P[O.sub.4] - [H.sub.2]O at 37.5 [degrees]C. J Res Natl Bur Stand Sect A 1966;70:545-52.

(34.) Gosselin RE, Coghlam ER. The stability of complexes between calcium and orthophosphate, polymeric phosphate, and phytate. Arch Biochem Biophys 1953;45:301-11.

(35.) Nancollas GH. In vitro studies of calcium phosphate crystallization. In: Mann S, Webb J, Williams RJP, eds. Biomineralization: chemical and biochemical perspectives. Weinheim: VCH, 1989: 180-2.

(36.) Nancollas GH, Tomazi B. Growth of calcium phosphate on hydroxyapatite crystals. Effect of supersaturation and ionic medium. J Phys Chem 1974;78:2218-25.

(37.) Brown WE, Smith JP, Lehr JR, Frazier AW. Octacalcium phosphate and hydroxyapatite. Nature 1962;196:1048-55.

(38.) Hsu HHT, Camacho NP, Sun F, Tawfik O, Aono H. Isolation of calcifiable vesicles from aortas of rabbits fed with high cholesterol diets. Atherosclerosis 2000;153:337-48.

(39.) Wuthier RE. Lipid composition of isolated epiphyseal cartilage cells membranes and matrix vesicles. Biochim Biophys Acta 1975;409:128-43.

(40.) Nancollas GH, Tsortos A, Zieba A. The nucleation and growth of calcium phosphate crystals at protein and phosphatidylserine liposome surfaces. Scanning Microsc Int 1996;10:499-508.

(41.) Vogel JJ. Calcium phosphate solid phase induction by dioleoylphosphatidate liposomes. J Colloid Interface Sci 1986;111: 152-9.


[1] Department of Chemical Biology, Peking University School of Pharmaceutical Sciences, 38 Xueyuan Rd., Beijing 100083, Peoples Republic of China.

[2] State Key Laboratory of Natural and Biomimetic Drugs. Beijing 100083, Peoples Republic of China.

[3] Nonstandard abbreviations: CH, cholesterol; HAP, hydroxyapatite; CHE, cholesteryl linoleate; OX-CH, cholestane-3[beta],5[alpha],6[beta]-triol; Ox-LDL, oxidized LDL; PC, L-a-phosphatidylcholine; TBA, thiobarbituric acid; ICP-AES, inductively coupled plasma atomic emission spectrophotometry; PBS, phosphate-buffered saline; TBARS, thiobarbituric acid-reactive species; XRD, x-ray diffraction; EDX, energy-dispersion x-ray; DHAP, defective hydroxyapatite; OCP, octacalcium phosphate; DCPD, dicalcium phosphate dihydrate; and [beta]-TCP, [beta]-dicalcium phosphate.

* Author for correspondence. Fax 86-10-62015584; e-mail tlzhang@
Table 1. Oxidative content of LDL (n = 3).

Incubation Mean (SD) TBARS,
period, h nmol x [L.sup.-1) x (mg protein)[sup.-1]

0 12.8 (0.3)
12 28.3 (0.8)
24 53.2 (0.2)

Table 2. Comparison of the XRD data for the precipitates
obtained in the experiments with DHAP (JCPDS card no. 460905).

 Control OX-CH

2 [theta], d, 2 [theta], d,
degrees [Angstrom] degrees [Angstrom]

25.96 3.429 25.92 3.435
28.92 3.085 28.80 3.098
31.90 2.803 31.78 2.813
32.24 2.774 32.12 2.784
34.08 2.629 34.00 2.635
39.88 2.259 39.92 2.256
46.60 1.947 46.68 1.944
49.50 1.840 49.52 1.839
50.62 1.802 50.32 1.812
53.44 1.713 53.28 1.718


2 [theta], d, 2 [theta], d,
degrees [Angstrom] degrees [Angstrom]

25.96 3.429 25.96 3.429
28.98 3.079 28.88 3.089
31.78 2.813 31.74 2.817
32.28 2.771 32.16 2.781
34.08 2.629 34.02 2.633
39.72 2.267 39.86 2.260
46.62 1.947 46.70 1.944
49.38 1.844 49.96 1.841
50.52 1.805
53.22 1.720

 CH/PC vesicles CH

2 [theta], d, 2 [theta], d,
degrees [Angstrom] degrees [Angstrom]

26.04 3.419 25.94 3.432
29.02 3.074 28.98 3.079
31.88 2.805 31.70 2.820
32.34 2.766 32.16 2.781
34.14 2.624 34.08 2.629
39.64 2.272 39.74 2.266
46.76 1.941 46.60 1.947
49.66 1.834 49.56 1.838
 50.34 1.811
53.38 1.715


hkl (a) d [sub.s]

002 3.441
210 3.090
211 2.819
112 2.781
202 2.632
310 2.268
222 1.946
213 1.842
321 1.809
004 1.720

(a) hkl, Millerindices.

Table 3. Comparison of the XRD data (2 [theta] <25 degrees) for
the precipitates formed in the presence of CH with those
for the CH standard (JCPDS card no. 70742).

 CH precipitates CH standard

2 [theta], degrees d, [Angstrom] hkl (a) d [sub.s]

13.100 6.7527 2,1,0 6.7500
14.160 6.2495 2,2.0 6.2400
15.360 5.7638 -2,-1,1 5.7400
16.920 5.2358 1,-5.1 5.2300
17.380 5.0982 0,1,2 5.0900
18.120 4.8916 1,-1,2 4.9000
22.70 3.9140 1,8,0 3.9100
23.520 3.7794 0,8,1 3.7800

(a) hkl, Miller indices.

Table 4. Mean (SD) calcium/phosphorus molar ratios in precipitates
(n = 3).


 Control CH OX-CH

Calcium/Phosphorus ratio 1.458 (0.009) 1.490 (0.001) 1.418 (0.005)

 CHE PC CH/PC vesicles

Calcium/Phosphorus ratio 1.480 (0.002) 1.40 (0.02) 1.36 (0.01)

Table 5. Comparison of the XRD data of the precipitates at the
interface with those for the HAP standard [JCPDS card no. 721243 for


R, (a) mm d, (a) Intensity

5.15 4.06 2nd strongest
6.10 3.43 3rd strongest
7.45 2.81 Strongest
8.95 2.34 Weak
10.30 2.03 Weak
11.30 1.85 Weak

 HAP standard

R, (a) mm hkl d, [Angstrom] [l.sub.s]

5.15 200 4.084 65
6.10 002 3.440 378
7.45 211 2.817 999
8.95 220 2.358 2
10.30 400 2.042 7
11.30 213 1.841 294

(a) R, radius of the relative diffraction rings; hkl, Miller
indices; [l.sub.s], diffraction intensity of the standard.

(b) d = 20.915/R.
COPYRIGHT 2003 American Association for Clinical Chemistry, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2003 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Lipids, Lipoproteins, and Cardiovascular Risk Factors
Author:Wang, He-Ping; Feng, Xiao-Jing; Gou, Bao-Di; Zhang, Tian-Lan; Xu, Shan-Jin; Wang, Kui
Publication:Clinical Chemistry
Date:Dec 1, 2003
Previous Article:Prognostic value of combination of cardiac troponin T and B-type natriuretic peptide after initiation of treatment in patients with chronic heart...
Next Article:Amino-terminal form of parathyroid hormone (PTH) with immunologic similarities to hPTH(1-84) is overproduced in primary and secondary...

Related Articles
Dyslipidemia in school-children / Okul cocuklarinda dislipidemi.
Corgenix Receives Third United States Patent on AtherOx Technology.
L-carnitine decreases oxidized LDl in diabetic patients.
Analytical performance and diagnostic accuracy of immunometric assays for the measurement of circulating Oxidized LDL.
Estimating LDL ApoB: infomania or clinical advance?
Significance of autoantibodies to oxidatively modified LDL in plasma of children with down syndrome.
Comparison of a new method for the direct and simultaneous assessment of LDL- and HDL-cholesterol with ultracentrifugation and established methods.
In search of the ideal measure of high-density lipoprotein.
Evolution of methods for measurement of HDL-cholesterol: from ultracentrifugation to homogeneous assays.
Consider six markers for CV risk assessment.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters