Clinicol Science and Molecular Medicine (1977) 52, 143-148.
Epitaxial relationships in uolithiasis: the brushite-whewellite system J. L. MEYER, J. H. BERGERT
AND
L. H. SMITH
Mayo Clinic and Mayo Foundation, Rochester, Minnesota, U.S.A.
(Received 16 March 1976; accepted 12 August 1976)
human urinary stones was suggested in 1968 by Lonsdale, from the close dimensional similarity of many of the crystal faces of the common urinary stone components, which is apparently required for epitaxially induced crystal growth. Although many calculi contain a mixture of insoluble organic and inorganic compounds (Prien, 1949), little experimental evidence has been presented to support (or deny) the existence of epitaxial relationships among the various crystalline components of human urinary stones. We have studied the circumstances, in vivo, in which crystals of one urinary stone component are able to nucleate epitaxially another crystalline phase from its otherwise stable, supersaturated solution. In systems where this occurs (at moderately low supersaturations and with greatly increased rates) epitaxially induced nucleation and crystal growth in urine seem most likely. The results may lead to therapeutic regimens that will decrease the most efficient crystalline nucleatingagents from stone-forming urine. Previously the hydroxyapatite-whewellite (calcium oxalate monohydrate) (Meyer, Bergert &Smith, 1975) and uric acid-whewellite (Meyer, Bergert & Smith, 1976) crystal growth systems have been considered. We now present experimental data on the whewellite-brushite (calcium monohydrogen phosphate dihydrate) system.
S-Y 1. Whewellite (calcium oxalate monohydrate) crystals were found to induce epitaxially the heterogeneous nucleation of brushite (calcium monohydrogen phosphate dihydrate) from its metastable supersaturated solution in approximately one-quarter of the time required for spontaneous precipitation in the absence of added nucleating agents. Scanning electronmicroscope observation of the crystalline phase showed brushite crystals originating from the whewellite seed crystals. 2. Crystal growth, upon nucleation, proceeded rapidly, and the metastable solutions quickly approached saturation. 3. Brushite crystals also induced the precipitation of calcium oxalate crystals in about onequarter of the time required for spontaneous precipitation; however, the rate of crystal growth was considerably slower. In support of the chemical data, scanning electron micrographs showed few crystals of calcium oxalate nucleated on the surface of the brushite seed. 4. The results provide some insight into the cause of stones containing calcium oxalate or calcium phosphate (or both), which form in the normally acid environment of human urine.
Key words: brushite, epitaxy, urolithiasis, whewellite.
Materials and methods
Introduction Acontribution of epitaxy to the growth of mixed
Materials Reagent-grade chemicals were used without further purification. All solutions were prepared
Correspondence: Dr Lynwood H. Smith, Mayo Clinic, 200 First Street SW, Rochester, MN 55901,
U.S.A. C
143
144
J. L. Meyer, J. H. Bergert and L. H . Smith
with deionized, distilled, carbonate-free water to a constant ionic strength (0.15). Sodium chloride was used as the supporting electrolyte. Seed crystals of calcium oxalate (Meyer & Smith, 1975) and brushite (Marshall & Nancollas, 1969) were prepared as described in the literature. The crystals were stored in slurry form at 37°C before use. All experiments were performed at 37°C. Growth of brushite on whewellite seed Stable supersaturated solutions of calcium phosphate were prepared by the dropwise addition of a 0.075 mol/l phosphate solution (0.01 mol/l Na2HP04; 0.065 mol/l KH2P04; pH 6.0) to a calcium chloride solution (5-7 mmol/l). Each solution was prepared with a calcium: phosphate molar ratio of 1:1. The pH of the solution was adjusted to 6.00 before the introduction of calcium oxalate seed crystals, and maintained within kO.01 pH unit during the course of the crystal-growth experiments by means of pH stat delivery of sodium hydroxide solution (Methrohm, Combititrator 3-D). Induction periods in which no calcium or phosphate precipitated from solution always followed the addition of seed material. The dissolution of seed material is rapid and approaches saturation within 10 min (Meyer & Smith, 1975) with negligible loss of crystal mass. The rate of crystal growth after nucleation was noted by withdrawing aliquots at intervals, filtering out the crystalline phase by 0.22 ,urn Millipore filtration, and analysing the filtrate for calcium and phosphate. Crystal growth rate was also monitored from the base required to maintain constant pH, since proton release is a direct measure of the phosphate precipitated as brushite. The presence of brushite to the exclusion of other calcium phosphate phases was confirmed by microscopic and petrographic observation of the crystalline phase. The free ionic concentrations of all major species in solution were calculated from the modified Debye-Huckel equation (Davies, 1962) and the formation constants for calcium oxalate (Nancollas & Gardner, 1974) and calcium phosphate (Chughtai, Marshall & Nancollas, 1968) ion pairs.
oxalate were prepared by the dropwise addition of potassium oxalate solution (0.01 mol/l) to a solution of calcium chloride ( 0 . 5 4 7 mmolll), 1 :1 calcium: phosphate molar ratios being present in solution. Total concentrations of calcium and oxalate were adjusted for each of the supersaturations used so that approximately equimolar concentrations of free ionic calcium and oxalate ion were present in the crystal growth solution. Before addition of oxalate ion, the phosphate concentration was brought to 0.018 mol/l with crystalline KH2P04. This resultant solution was slightly undersaturated with respect to brushite when the pH was adjusted to 6.5 with sodium hydroxide. An aliquot of the brushite crystal slurry was added and equilibrated for 30 min, before filtration through a 0.22 jm Millipore filter, the oxalate stock solution then being added to produce the metastable calcium oxalate solution. An aliquot of the brushite seed suspension was then added, and the rate of crystal growth of calcium oxalate was measured by withdrawing samples at intervals, filtering through the 0.22 pm Millipore filters, and analysing the filtrate for calcium by atomic absorption. Phosphate analyses were performed on selected samples to ensure that changes in calcium concentration were not due to calcium phosphate crystal growth or dissolution. Free ionic concentrations were calculated as described above. Scanning electron microscopy
Sample preparation for both epitaxial crystalgrowth systems was similar. Aliquots were removed from the crystal-growth experimental systems, and the crystalline phase was separated by immediate filtration through 0.2 pm Nucleopore filters. The crystals were sparingly washed with deionized distilled water to remove excess solution. The filters were dried, mounted on aluminium specimen stages, and vacuum-coated with gold-palladium and carbon. The scanning electron microscope (ETEC Autoscan) was operated at 5-20 kV and a specimen tilt of 45"; the crystals were observed at magnifications up to x20000. Measurements of surface area
Growth of whewellite on brushite seed Stable supersaturated solutions of calcium
A dynamic flow surface area analyser (Quantasorb, Quantachrome Corp.) was used
Epitaxy in urolithiasis to measure the surface areas of the dried crystals used as the seeding material in the heterogeneous crystal-growth experiments. Three partial pressures of the adsorbate (nitrogen in helium) were used, the Brunauer, Emmett & Teller (1938) equation being utilized. Results The supersaturation ratio, S, was previously defined as [Ca' +]/[Ca' +Irn, in which [Ca' + I and [Ca'+], are the free ionic concentrations of calcium present initially in solutions and at saturation respectively (Meyer et al., 1975). For a saturated solution, S = 1; S< 1 and S > 1 indicate undersaturation and supersaturation respectively.
Nucleation of brushite by whewellite crystals The heterogeneous nucleation of brushite from its metastable supersaturated solution, caused by the introduction of whewellite
'r
I
0
I
100
I
I
200
1
I
300
I 4 00
Time ( m i n )
FIG.1. Epitaxially induced growth of brushite crystals on whewellite seeds at S = 1.77. Surface area of seed is 9.4 cm'/ml. Samples for scanning electron micrographs (Fig. 2, Fig. 3 and Fig. 4) were taken at points indicated by arrows.
crystals, is shown in Fig. 1. The change in the total calcium concentration with time gives chemical-kinetic evidence that the seed crystals induced the crystal growth of calcium phosphate after a 4 h induction period. Spontaneous precipitation in the absence of foreign nucleating
145
materials under these conditions occurred in 15 h. The solution phosphate concentration (not presented) fell in parallel to that of calcium, resulting in a new crystallinephase with a molar 1:1 calcium: phosphate stoicheiometry. To confirm the chemical evidence that calcium phosphate was being epitaxially nucleated by the calcium oxalate, seed samples were withdrawn from the crystal-growth system, at times indicated by the arrows in Fig. 1, and observed with scanningelectron microscopyand petrographic microscopy. The morphology of the crystalline material, isolated 30 min after its introduction, is shown in Figs. 2(A) and 2(B). The crystals appear identical in all respects with those of the seed material, and no extraneous phases are present. Crystals isolated shortly after the induction period exhibit a second phase originating from the calcium oxalate seed crystals (Fig. 3A). Higher magnifications (Fig. 3B) show that the new crystalline phase is always in contact with seed crystals. Crystals isolated near the end of the rapid growth process are shown in Figs. 4(A) and 4(B). The calcium phosphate crystals appear larger than those in Figs. 3(A) and 3(B)respectively,even though the original magnifications are one-half. Highermagnification micrographs (Fig. 4B) again show the morphologically distinct whewellite crystals in physical contact with the calcium phosphate phase. The optical properties as well as the morphology of the new phase are compatible with those of brushite. Experiments were also performed at other supersaturations(Fig. 5). The free ionic calcium concentration is plotted with the time-axis for each of the three experiments set so that spontaneous precipitation without foreign seed material occurs at the same point. Induction periods for spontaneous precipitation without added nucleating agents for S = 1.77, 1.63 and 1.50 were found to be 15, 36 and 100 h respectively. When equivalent amounts of whewellite crystals were introduced into each of the three metastable calcium phosphate solutions, the induction periods for nucleation were reduced to 4,9 and 25 h for the solutions with S = 1-77, 1.63 and 1.50 respectively. In each instance the time required for heterogeneous nucleation is approximately one-quarter of that needed for spontaneous precipitation. The induction periods for epitaxially induced heterogeneous nucleation did not depend upon the amount of
*--.--*-.
J. L. Meyer, J. H. Bergert and L. H. Smith
146
-0-
-*-*-*-
f - m - -m-+-m--m--m&-&+-4--CC4-tC
Saturation
S = 1.630 CaGOI,HIO r e e d ; I n o seed S = 1-50 A CaG04,&0 sned; A no seed
I 0
I
I
I
1
3
6
9
12
1
I
I
I
0
9
18
27
L
I
I
I
I
0
20
40
60
80
' 0 15
I& A
Time(h)
FIG.5. Epitaxially induced growth of brushite crystals on whewellite seeds at different supersaturations. Abscissae are normalized so that spontaneous precipitation in the absence of foreign seed material occurs at the same point on the graph. Seed surface area was 1.6 cm2/ml.
metastability of calcium oxalate solutions (Fig. 6) shows that at S = 1.90 the metastability of the calcium oxalate is maintained for about 24 h. The introduction of brushite crystals produced a measurable fall in calcium ion concentration after an induction period of about 6 h, with a slow approach to saturation, only 27% of the potential calcium oxalate crystal growth occurring within 24 h. Examination of the crystals removedafter 7 h (Fig. 7A)indicated that only a
seed material introduced. Experiments (not shown) performed at the three supersaturations with various amounts of seed material had no effect on the induction periods (e.g. six times the whewellite seed in Fig. 1 resulted in approximately the same induction period at S = 1.77). Nucleation of whewellite by brushite crystals
The influence of brushite crystals on the
I 0
I
1
I
2
I
3
I
4
I
5
I
6
7
I
'
%
Time (h)
FIG.6. Epitaxially induced growth of calcium oxalate crystals on brushite seeds (0).The broken line is obtained in the absence of seed. S = 1.90; surface area of seeds was 16.53 cm2/ml.
Epitaxy in urolithiasis
FIG.2. Scanning electron micrographs (A, ~ 2 4 0 0 :B, x 12 000)o f whewellite crystals isolated from the experiment depicted in Fig. 1 after 30 min. (Fucinp p . 146)
J . L. Meyer, J . H . Bergert and L. H . Smith
FIG.3. Scanning electron micrographs (A, 2400; B, 12 000) of crystals isolated from the experiment depicted in Fig. 1 after 240 rnin, at which point there is a measurable fall in calcium and phosphate concentration. Crystals of brushite appear to originate from the whewellite seed. ~
Epitaxy in urolithiasis
FIG.4. Scanning electron micrographs (A, x 1350; B, x 6750) of crystals isolated from the experiment depicted in Fig. I after 260 min. The rapid crystal-growth process is nearly complete and the brushite crystals appear much larger.
J . L . M e y e r , J . H . Bergert mid L. H . Smith
FIG.7. Scanning electron micrographs ( A . 2600: B. 6800) o f crystals isolated from the experiiment depicte:d in Fig. 6 after 7 h. A few crystals. apparently whewellite. are present o n the brushite seed crystal,>.
Epitaxy in urolithiasis
few had nucleated on the brushite seed crystals. Higher magnifications (Fig. 7B) revealed that the new crystalline phase had a morphology similar to that of whewellite. This was supported by petrographic examination. The induction period for heterogeneous nucleation, as in the system described above, did not depend on the number of brushite seed crystals added.
Discussion Hydroxyapatite has been shown previously to nucleate epitaxially whewellite crystals from a supersaturated solution of calcium oxalate but whewellite crystals were unable to nucleate hydroxyapatite crystals (Meyer et al., 1975). Those experiments were performed at pH 7.4, whereas the average urinary pH is closer to 6, at which brushite, a more acidic compound, usually precipitates from solution (Pak, Eanes & Ruskin, 1971). The justification for performing experiments at pH 7.4 is that apatite and not brushite is the most common form of calcium phosphate found in either ‘pure’ or ‘mixed’ urinary calculi (Prien, 1949). In contrast to the pH 7.4 system, our results suggest that at pH 6.0 whewellite crystals are able to nucleate a calcium phosphate phase from solution. Chemical, morphological and petrographic evidence suggested that the material was brushite. An induction period for the epitaxially induced heterogeneous nucleation was found for each supersaturation that was approximately one-quarter of the time required for spontaneous precipitation without added nucleating agents. We conclude that, at an appropriately high supersaturation, whewellite would nucleate brushite crystals in less time than is required for the urine to traverse the urinary tract. Upon nucleation, a rapid approach to saturation via crystal growth is observed. That crystal growth and not further nucleation accounts for most of the loss of calcium and phosphate from solution is illustrated in micrographs showing brushite crystals that have greatly increased in size as compared with those at the onset of nucleation. No extraneous phase suggestive of calcium phosphate is evident during the induction period. During the crystal-growth process the brushite crystals appear to project from clusters of whewellite seed crystals. High-magnification micrographs always show the morphologically distinct
147
whewellite crystals at the apparent origin of the radially directed brushite crystals, suggesting that these crystals act as the nidi for the calcium phosphate crystals. As the time required for epitaxial nucleation did not depend on the amount of seed material, the driving force for the nucleation may be the statistical probability of a critical nucleus forming at an appropriate site on the host crystal. This would depend primarily upon the supersaturation(free energy) of the solution and not upon the number of sites (Walton, 1967). Brushite seed crystals also appear to induce epitaxially the nucleation of whewellite from its stable supersaturated solution. Although the induction periods are not as well defined as for whewellite-induced brushite formation, the presence of brushite crystals reduces the time required for spontaneousprecipitation by about fourfold. Thus the same amount of lattice mismatch appears to exist between the host crystal and the nucleating phase in each of the two epitaxial systems. The rate of crystal growth of calcium oxalate on nuclei induced by brushite seed is slow compared with the rate of calcium phosphate growth on nuclei induced by whewellite seeds. This suggests that few sites are available on the brushite crystals that produce nuclei of calcium oxalate, at the moderately low supersaturations studied here. Micrographs of the brushite-seeded experiments confirm the existence of relatively few calcium oxalate crystals on the surface of the brushite crystals. Our results for the brushitewhewellite epitaxial crystal-growth system may provide some insight into the cause of certain stones containing calcium oxalate or calcium phosphate, or both. Although brushite is a rare component of urinary calculi, it may still be the initially precipitated calcium phosphate phase, which later hydrolyses irreversibly to the thermodynamically more stable hydroxyapatite (Brown, Pate1 & Chow, 1975). This could occur during the normal diurnal fluctuations in pH (Elliot, Sharp & Lewis, 1959). The presence of either whewellite or brushite crystals in acidic urine would promote the precipitation of the other phase from its supersaturated solution under conditions where spontaneous nucleation would not occur. This would result either in the formation of ‘pure’ calculi with a foreign nucleus or the common mixed calcium phos-
148
J. L. Meyer, J. H. Bergert and L. H . Smith
phatealcium oxalate stones. The apparent potential for rapid crystal growth of brushite on nuclei induced by whewellite, coupled with a crystal morphology (e.g. Fig. 3A, B) that could easily result in entanglement within the urinary tract, provides a setting for the initiation and subsequent growth of a urinary calculus. On the other hand, it appears that the presence of brushite crystals in a urine that is metastable with respect to calcium oxalate would result in only a relatively slow crystal growth of calcium oxalate. Hydroxyapatite seems to be much more effective in inducing calcium oxalate crystal growth (Meyer et al., 1975), and it would appear that this latter phase of calcium phosphate would be more important in the initiation and growth of the mixed calcium phosphate-calcium oxalate stones. In addition, it is commonly found intracellularly as well as extracellularly in kidneys of patients who form calcium oxalate stones (Randall, 1937; Malek & Boyce, 1973).
Acknowledgment This investigation was supported in part by Research Grant AM-17717 from the National Institutes of Health, Public Health Service.
Journal of the American Chemical Society, 60, 309319.
CHUCHTAI,A,, MARSHALL, R. & NANCOLLAS, G.H. (1968) Complexes in calcium phosphate solutions. Journal of Physical Chemistry, 72,208-21 1. DAVID, C.W. (1962) Ion Association. Butterworths, London. ELLIOT,J.S., SHARP,R.F. & LEWIS,L. (1959) Urinary pH. Journal of Urology, 81, 339-343. LONSDALE, K. (1968) Epitaxy as a growth factor i n urinary calculi and gallstones. Nature (London), 217, 56-58.
MALEK,R.S. & BOYCE,W.H. (1973) Intranephronic calculosis: its significance and relationship to matrix in nephrolithiasis. Journal of Urology, 109, 551-555. MARSHALL, R.W. & NANCOLLAS, G.H. (1969) The kinetics of crystal growth of dicalcium phosphate dihydrate. Journal of Physical Chemistry, 73, 38383844.
MEYER,J.L., BERGERT, J.H. & SMITH,L.H. (1975) Epitaxial relationships in urolithiasis: the calcium oxalate monohydrate-hydroxyapatite system. Clinical Science and Molecular Medicine, 49, 369-374. MEYER, J.L., BERCERT, J.H. & SMITH,L.H. (1976) The epitaxially induced crystal growth of calcium oxalate by crystalline uric acid. Inuestigariue Urology (In press). MEYER, J.L. & SMITH,L.H. (1975) Growth of calcium oxalate crystals. I. A model for urinary stone growth. Investigative Urology, 13, 31-35. NANCOLLAS, G.H. & GARDNER, G.L. (1974) Kinetics of crystal growth of calcium oxalate monohydrate. Journal of Crystal Growth, 21, 267-216. PAK,C.Y.C., EANES,E.D. & RUSKIN,B. (1971) Spontaneous precipitation of brushite in urine: evidence that brushite is the nidus of renal stones originating as calcium phosphate. Proceedings of the National Academy of Sciences of the United States of America, 68, 1456-1460.
References BROWN,W.E., PATEL,P.R. & CHOW, L.C. (1975) Formation of CaHP04.2H20 from enamel mineral and its relationship to caries mechanism. Journal o/ Dental Research, 5 4 , 4 7 5 4 8 1. BRUNAUER, S., EMMETT,P.H. & TELLER, E. (1938) Adsorption of gases in multimolecular layers
PRIEN,E.L. (1949) Studies in urolithiasis. 11. Relationships between pathogenesis, structure and composition of calculi. Journal of Urology, 61, 821-836. RANDALL, A. (1937) The origin and growth of renal calculi. Annals of Surgery, 105, 1009-1027. WALTON, A.G. (1967) The Formation and Properties of Precipitates, p. 7. Interscience Publishers, New York.
Epitaxial relationships in uolithiasis: the brushite-whewellite system J. L. MEYER, J. H. BERGERT
AND
L. H. SMITH
Mayo Clinic and Mayo Foundation, Rochester, Minnesota, U.S.A.
(Received 16 March 1976; accepted 12 August 1976)
human urinary stones was suggested in 1968 by Lonsdale, from the close dimensional similarity of many of the crystal faces of the common urinary stone components, which is apparently required for epitaxially induced crystal growth. Although many calculi contain a mixture of insoluble organic and inorganic compounds (Prien, 1949), little experimental evidence has been presented to support (or deny) the existence of epitaxial relationships among the various crystalline components of human urinary stones. We have studied the circumstances, in vivo, in which crystals of one urinary stone component are able to nucleate epitaxially another crystalline phase from its otherwise stable, supersaturated solution. In systems where this occurs (at moderately low supersaturations and with greatly increased rates) epitaxially induced nucleation and crystal growth in urine seem most likely. The results may lead to therapeutic regimens that will decrease the most efficient crystalline nucleatingagents from stone-forming urine. Previously the hydroxyapatite-whewellite (calcium oxalate monohydrate) (Meyer, Bergert &Smith, 1975) and uric acid-whewellite (Meyer, Bergert & Smith, 1976) crystal growth systems have been considered. We now present experimental data on the whewellite-brushite (calcium monohydrogen phosphate dihydrate) system.
S-Y 1. Whewellite (calcium oxalate monohydrate) crystals were found to induce epitaxially the heterogeneous nucleation of brushite (calcium monohydrogen phosphate dihydrate) from its metastable supersaturated solution in approximately one-quarter of the time required for spontaneous precipitation in the absence of added nucleating agents. Scanning electronmicroscope observation of the crystalline phase showed brushite crystals originating from the whewellite seed crystals. 2. Crystal growth, upon nucleation, proceeded rapidly, and the metastable solutions quickly approached saturation. 3. Brushite crystals also induced the precipitation of calcium oxalate crystals in about onequarter of the time required for spontaneous precipitation; however, the rate of crystal growth was considerably slower. In support of the chemical data, scanning electron micrographs showed few crystals of calcium oxalate nucleated on the surface of the brushite seed. 4. The results provide some insight into the cause of stones containing calcium oxalate or calcium phosphate (or both), which form in the normally acid environment of human urine.
Key words: brushite, epitaxy, urolithiasis, whewellite.
Materials and methods
Introduction Acontribution of epitaxy to the growth of mixed
Materials Reagent-grade chemicals were used without further purification. All solutions were prepared
Correspondence: Dr Lynwood H. Smith, Mayo Clinic, 200 First Street SW, Rochester, MN 55901,
U.S.A. C
143
144
J. L. Meyer, J. H. Bergert and L. H . Smith
with deionized, distilled, carbonate-free water to a constant ionic strength (0.15). Sodium chloride was used as the supporting electrolyte. Seed crystals of calcium oxalate (Meyer & Smith, 1975) and brushite (Marshall & Nancollas, 1969) were prepared as described in the literature. The crystals were stored in slurry form at 37°C before use. All experiments were performed at 37°C. Growth of brushite on whewellite seed Stable supersaturated solutions of calcium phosphate were prepared by the dropwise addition of a 0.075 mol/l phosphate solution (0.01 mol/l Na2HP04; 0.065 mol/l KH2P04; pH 6.0) to a calcium chloride solution (5-7 mmol/l). Each solution was prepared with a calcium: phosphate molar ratio of 1:1. The pH of the solution was adjusted to 6.00 before the introduction of calcium oxalate seed crystals, and maintained within kO.01 pH unit during the course of the crystal-growth experiments by means of pH stat delivery of sodium hydroxide solution (Methrohm, Combititrator 3-D). Induction periods in which no calcium or phosphate precipitated from solution always followed the addition of seed material. The dissolution of seed material is rapid and approaches saturation within 10 min (Meyer & Smith, 1975) with negligible loss of crystal mass. The rate of crystal growth after nucleation was noted by withdrawing aliquots at intervals, filtering out the crystalline phase by 0.22 ,urn Millipore filtration, and analysing the filtrate for calcium and phosphate. Crystal growth rate was also monitored from the base required to maintain constant pH, since proton release is a direct measure of the phosphate precipitated as brushite. The presence of brushite to the exclusion of other calcium phosphate phases was confirmed by microscopic and petrographic observation of the crystalline phase. The free ionic concentrations of all major species in solution were calculated from the modified Debye-Huckel equation (Davies, 1962) and the formation constants for calcium oxalate (Nancollas & Gardner, 1974) and calcium phosphate (Chughtai, Marshall & Nancollas, 1968) ion pairs.
oxalate were prepared by the dropwise addition of potassium oxalate solution (0.01 mol/l) to a solution of calcium chloride ( 0 . 5 4 7 mmolll), 1 :1 calcium: phosphate molar ratios being present in solution. Total concentrations of calcium and oxalate were adjusted for each of the supersaturations used so that approximately equimolar concentrations of free ionic calcium and oxalate ion were present in the crystal growth solution. Before addition of oxalate ion, the phosphate concentration was brought to 0.018 mol/l with crystalline KH2P04. This resultant solution was slightly undersaturated with respect to brushite when the pH was adjusted to 6.5 with sodium hydroxide. An aliquot of the brushite crystal slurry was added and equilibrated for 30 min, before filtration through a 0.22 jm Millipore filter, the oxalate stock solution then being added to produce the metastable calcium oxalate solution. An aliquot of the brushite seed suspension was then added, and the rate of crystal growth of calcium oxalate was measured by withdrawing samples at intervals, filtering through the 0.22 pm Millipore filters, and analysing the filtrate for calcium by atomic absorption. Phosphate analyses were performed on selected samples to ensure that changes in calcium concentration were not due to calcium phosphate crystal growth or dissolution. Free ionic concentrations were calculated as described above. Scanning electron microscopy
Sample preparation for both epitaxial crystalgrowth systems was similar. Aliquots were removed from the crystal-growth experimental systems, and the crystalline phase was separated by immediate filtration through 0.2 pm Nucleopore filters. The crystals were sparingly washed with deionized distilled water to remove excess solution. The filters were dried, mounted on aluminium specimen stages, and vacuum-coated with gold-palladium and carbon. The scanning electron microscope (ETEC Autoscan) was operated at 5-20 kV and a specimen tilt of 45"; the crystals were observed at magnifications up to x20000. Measurements of surface area
Growth of whewellite on brushite seed Stable supersaturated solutions of calcium
A dynamic flow surface area analyser (Quantasorb, Quantachrome Corp.) was used
Epitaxy in urolithiasis to measure the surface areas of the dried crystals used as the seeding material in the heterogeneous crystal-growth experiments. Three partial pressures of the adsorbate (nitrogen in helium) were used, the Brunauer, Emmett & Teller (1938) equation being utilized. Results The supersaturation ratio, S, was previously defined as [Ca' +]/[Ca' +Irn, in which [Ca' + I and [Ca'+], are the free ionic concentrations of calcium present initially in solutions and at saturation respectively (Meyer et al., 1975). For a saturated solution, S = 1; S< 1 and S > 1 indicate undersaturation and supersaturation respectively.
Nucleation of brushite by whewellite crystals The heterogeneous nucleation of brushite from its metastable supersaturated solution, caused by the introduction of whewellite
'r
I
0
I
100
I
I
200
1
I
300
I 4 00
Time ( m i n )
FIG.1. Epitaxially induced growth of brushite crystals on whewellite seeds at S = 1.77. Surface area of seed is 9.4 cm'/ml. Samples for scanning electron micrographs (Fig. 2, Fig. 3 and Fig. 4) were taken at points indicated by arrows.
crystals, is shown in Fig. 1. The change in the total calcium concentration with time gives chemical-kinetic evidence that the seed crystals induced the crystal growth of calcium phosphate after a 4 h induction period. Spontaneous precipitation in the absence of foreign nucleating
145
materials under these conditions occurred in 15 h. The solution phosphate concentration (not presented) fell in parallel to that of calcium, resulting in a new crystallinephase with a molar 1:1 calcium: phosphate stoicheiometry. To confirm the chemical evidence that calcium phosphate was being epitaxially nucleated by the calcium oxalate, seed samples were withdrawn from the crystal-growth system, at times indicated by the arrows in Fig. 1, and observed with scanningelectron microscopyand petrographic microscopy. The morphology of the crystalline material, isolated 30 min after its introduction, is shown in Figs. 2(A) and 2(B). The crystals appear identical in all respects with those of the seed material, and no extraneous phases are present. Crystals isolated shortly after the induction period exhibit a second phase originating from the calcium oxalate seed crystals (Fig. 3A). Higher magnifications (Fig. 3B) show that the new crystalline phase is always in contact with seed crystals. Crystals isolated near the end of the rapid growth process are shown in Figs. 4(A) and 4(B). The calcium phosphate crystals appear larger than those in Figs. 3(A) and 3(B)respectively,even though the original magnifications are one-half. Highermagnification micrographs (Fig. 4B) again show the morphologically distinct whewellite crystals in physical contact with the calcium phosphate phase. The optical properties as well as the morphology of the new phase are compatible with those of brushite. Experiments were also performed at other supersaturations(Fig. 5). The free ionic calcium concentration is plotted with the time-axis for each of the three experiments set so that spontaneous precipitation without foreign seed material occurs at the same point. Induction periods for spontaneous precipitation without added nucleating agents for S = 1.77, 1.63 and 1.50 were found to be 15, 36 and 100 h respectively. When equivalent amounts of whewellite crystals were introduced into each of the three metastable calcium phosphate solutions, the induction periods for nucleation were reduced to 4,9 and 25 h for the solutions with S = 1-77, 1.63 and 1.50 respectively. In each instance the time required for heterogeneous nucleation is approximately one-quarter of that needed for spontaneous precipitation. The induction periods for epitaxially induced heterogeneous nucleation did not depend upon the amount of
*--.--*-.
J. L. Meyer, J. H. Bergert and L. H. Smith
146
-0-
-*-*-*-
f - m - -m-+-m--m--m&-&+-4--CC4-tC
Saturation
S = 1.630 CaGOI,HIO r e e d ; I n o seed S = 1-50 A CaG04,&0 sned; A no seed
I 0
I
I
I
1
3
6
9
12
1
I
I
I
0
9
18
27
L
I
I
I
I
0
20
40
60
80
' 0 15
I& A
Time(h)
FIG.5. Epitaxially induced growth of brushite crystals on whewellite seeds at different supersaturations. Abscissae are normalized so that spontaneous precipitation in the absence of foreign seed material occurs at the same point on the graph. Seed surface area was 1.6 cm2/ml.
metastability of calcium oxalate solutions (Fig. 6) shows that at S = 1.90 the metastability of the calcium oxalate is maintained for about 24 h. The introduction of brushite crystals produced a measurable fall in calcium ion concentration after an induction period of about 6 h, with a slow approach to saturation, only 27% of the potential calcium oxalate crystal growth occurring within 24 h. Examination of the crystals removedafter 7 h (Fig. 7A)indicated that only a
seed material introduced. Experiments (not shown) performed at the three supersaturations with various amounts of seed material had no effect on the induction periods (e.g. six times the whewellite seed in Fig. 1 resulted in approximately the same induction period at S = 1.77). Nucleation of whewellite by brushite crystals
The influence of brushite crystals on the
I 0
I
1
I
2
I
3
I
4
I
5
I
6
7
I
'
%
Time (h)
FIG.6. Epitaxially induced growth of calcium oxalate crystals on brushite seeds (0).The broken line is obtained in the absence of seed. S = 1.90; surface area of seeds was 16.53 cm2/ml.
Epitaxy in urolithiasis
FIG.2. Scanning electron micrographs (A, ~ 2 4 0 0 :B, x 12 000)o f whewellite crystals isolated from the experiment depicted in Fig. 1 after 30 min. (Fucinp p . 146)
J . L. Meyer, J . H . Bergert and L. H . Smith
FIG.3. Scanning electron micrographs (A, 2400; B, 12 000) of crystals isolated from the experiment depicted in Fig. 1 after 240 rnin, at which point there is a measurable fall in calcium and phosphate concentration. Crystals of brushite appear to originate from the whewellite seed. ~
Epitaxy in urolithiasis
FIG.4. Scanning electron micrographs (A, x 1350; B, x 6750) of crystals isolated from the experiment depicted in Fig. I after 260 min. The rapid crystal-growth process is nearly complete and the brushite crystals appear much larger.
J . L . M e y e r , J . H . Bergert mid L. H . Smith
FIG.7. Scanning electron micrographs ( A . 2600: B. 6800) o f crystals isolated from the experiiment depicte:d in Fig. 6 after 7 h. A few crystals. apparently whewellite. are present o n the brushite seed crystal,>.
Epitaxy in urolithiasis
few had nucleated on the brushite seed crystals. Higher magnifications (Fig. 7B) revealed that the new crystalline phase had a morphology similar to that of whewellite. This was supported by petrographic examination. The induction period for heterogeneous nucleation, as in the system described above, did not depend on the number of brushite seed crystals added.
Discussion Hydroxyapatite has been shown previously to nucleate epitaxially whewellite crystals from a supersaturated solution of calcium oxalate but whewellite crystals were unable to nucleate hydroxyapatite crystals (Meyer et al., 1975). Those experiments were performed at pH 7.4, whereas the average urinary pH is closer to 6, at which brushite, a more acidic compound, usually precipitates from solution (Pak, Eanes & Ruskin, 1971). The justification for performing experiments at pH 7.4 is that apatite and not brushite is the most common form of calcium phosphate found in either ‘pure’ or ‘mixed’ urinary calculi (Prien, 1949). In contrast to the pH 7.4 system, our results suggest that at pH 6.0 whewellite crystals are able to nucleate a calcium phosphate phase from solution. Chemical, morphological and petrographic evidence suggested that the material was brushite. An induction period for the epitaxially induced heterogeneous nucleation was found for each supersaturation that was approximately one-quarter of the time required for spontaneous precipitation without added nucleating agents. We conclude that, at an appropriately high supersaturation, whewellite would nucleate brushite crystals in less time than is required for the urine to traverse the urinary tract. Upon nucleation, a rapid approach to saturation via crystal growth is observed. That crystal growth and not further nucleation accounts for most of the loss of calcium and phosphate from solution is illustrated in micrographs showing brushite crystals that have greatly increased in size as compared with those at the onset of nucleation. No extraneous phase suggestive of calcium phosphate is evident during the induction period. During the crystal-growth process the brushite crystals appear to project from clusters of whewellite seed crystals. High-magnification micrographs always show the morphologically distinct
147
whewellite crystals at the apparent origin of the radially directed brushite crystals, suggesting that these crystals act as the nidi for the calcium phosphate crystals. As the time required for epitaxial nucleation did not depend on the amount of seed material, the driving force for the nucleation may be the statistical probability of a critical nucleus forming at an appropriate site on the host crystal. This would depend primarily upon the supersaturation(free energy) of the solution and not upon the number of sites (Walton, 1967). Brushite seed crystals also appear to induce epitaxially the nucleation of whewellite from its stable supersaturated solution. Although the induction periods are not as well defined as for whewellite-induced brushite formation, the presence of brushite crystals reduces the time required for spontaneousprecipitation by about fourfold. Thus the same amount of lattice mismatch appears to exist between the host crystal and the nucleating phase in each of the two epitaxial systems. The rate of crystal growth of calcium oxalate on nuclei induced by brushite seed is slow compared with the rate of calcium phosphate growth on nuclei induced by whewellite seeds. This suggests that few sites are available on the brushite crystals that produce nuclei of calcium oxalate, at the moderately low supersaturations studied here. Micrographs of the brushite-seeded experiments confirm the existence of relatively few calcium oxalate crystals on the surface of the brushite crystals. Our results for the brushitewhewellite epitaxial crystal-growth system may provide some insight into the cause of certain stones containing calcium oxalate or calcium phosphate, or both. Although brushite is a rare component of urinary calculi, it may still be the initially precipitated calcium phosphate phase, which later hydrolyses irreversibly to the thermodynamically more stable hydroxyapatite (Brown, Pate1 & Chow, 1975). This could occur during the normal diurnal fluctuations in pH (Elliot, Sharp & Lewis, 1959). The presence of either whewellite or brushite crystals in acidic urine would promote the precipitation of the other phase from its supersaturated solution under conditions where spontaneous nucleation would not occur. This would result either in the formation of ‘pure’ calculi with a foreign nucleus or the common mixed calcium phos-
148
J. L. Meyer, J. H. Bergert and L. H . Smith
phatealcium oxalate stones. The apparent potential for rapid crystal growth of brushite on nuclei induced by whewellite, coupled with a crystal morphology (e.g. Fig. 3A, B) that could easily result in entanglement within the urinary tract, provides a setting for the initiation and subsequent growth of a urinary calculus. On the other hand, it appears that the presence of brushite crystals in a urine that is metastable with respect to calcium oxalate would result in only a relatively slow crystal growth of calcium oxalate. Hydroxyapatite seems to be much more effective in inducing calcium oxalate crystal growth (Meyer et al., 1975), and it would appear that this latter phase of calcium phosphate would be more important in the initiation and growth of the mixed calcium phosphate-calcium oxalate stones. In addition, it is commonly found intracellularly as well as extracellularly in kidneys of patients who form calcium oxalate stones (Randall, 1937; Malek & Boyce, 1973).
Acknowledgment This investigation was supported in part by Research Grant AM-17717 from the National Institutes of Health, Public Health Service.
Journal of the American Chemical Society, 60, 309319.
CHUCHTAI,A,, MARSHALL, R. & NANCOLLAS, G.H. (1968) Complexes in calcium phosphate solutions. Journal of Physical Chemistry, 72,208-21 1. DAVID, C.W. (1962) Ion Association. Butterworths, London. ELLIOT,J.S., SHARP,R.F. & LEWIS,L. (1959) Urinary pH. Journal of Urology, 81, 339-343. LONSDALE, K. (1968) Epitaxy as a growth factor i n urinary calculi and gallstones. Nature (London), 217, 56-58.
MALEK,R.S. & BOYCE,W.H. (1973) Intranephronic calculosis: its significance and relationship to matrix in nephrolithiasis. Journal of Urology, 109, 551-555. MARSHALL, R.W. & NANCOLLAS, G.H. (1969) The kinetics of crystal growth of dicalcium phosphate dihydrate. Journal of Physical Chemistry, 73, 38383844.
MEYER,J.L., BERGERT, J.H. & SMITH,L.H. (1975) Epitaxial relationships in urolithiasis: the calcium oxalate monohydrate-hydroxyapatite system. Clinical Science and Molecular Medicine, 49, 369-374. MEYER, J.L., BERCERT, J.H. & SMITH,L.H. (1976) The epitaxially induced crystal growth of calcium oxalate by crystalline uric acid. Inuestigariue Urology (In press). MEYER, J.L. & SMITH,L.H. (1975) Growth of calcium oxalate crystals. I. A model for urinary stone growth. Investigative Urology, 13, 31-35. NANCOLLAS, G.H. & GARDNER, G.L. (1974) Kinetics of crystal growth of calcium oxalate monohydrate. Journal of Crystal Growth, 21, 267-216. PAK,C.Y.C., EANES,E.D. & RUSKIN,B. (1971) Spontaneous precipitation of brushite in urine: evidence that brushite is the nidus of renal stones originating as calcium phosphate. Proceedings of the National Academy of Sciences of the United States of America, 68, 1456-1460.
References BROWN,W.E., PATEL,P.R. & CHOW, L.C. (1975) Formation of CaHP04.2H20 from enamel mineral and its relationship to caries mechanism. Journal o/ Dental Research, 5 4 , 4 7 5 4 8 1. BRUNAUER, S., EMMETT,P.H. & TELLER, E. (1938) Adsorption of gases in multimolecular layers
PRIEN,E.L. (1949) Studies in urolithiasis. 11. Relationships between pathogenesis, structure and composition of calculi. Journal of Urology, 61, 821-836. RANDALL, A. (1937) The origin and growth of renal calculi. Annals of Surgery, 105, 1009-1027. WALTON, A.G. (1967) The Formation and Properties of Precipitates, p. 7. Interscience Publishers, New York.
