| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Urology Unit, Department of Surgery, School of Medicine, Flinders University, Bedford Park, South Australia; and 2Department of Applied Chemistry, Curtin University of Technology, Perth, Western Australia, Australia
Submitted 11 November 2007 ; accepted in final form 7 December 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
2 mg/l. These findings were complemented by field emission scanning electron microscopy, which showed that crystal degradation also increased relative to CME concentration. Intracrystalline proteins enhance degradation and dissolution of CaOx crystals and thus may constitute a natural defense against urolithiasis. The findings have significant ramifications in biomineral metabolism and pathogenesis of renal stones. nephrocalcinosis; urolithiasis
CaOx crystals precipitated from human urine contain intracrystalline proteins (17, 48). These create discontinuities within the mineral phase, which increase the nonuniform strain of the crystals and reduce the average size of their component crystallites (17). We have proposed that such crystals would be more vulnerable to attack by urinary proteases before attachment, as well as by intracellular proteases following adhesion of crystals to the tubular epithelium (17, 48). Phagocytosis of crystals attached to the cellular membrane would be succeeded by formation of phagolysosomes, whose internal environment, which is highly acidic and contains a potent cocktail of lysosomal proteases, would be ideally suited to the destruction of crystalline mineral interspersed with protein-filled labyrinths. The proteases would degrade the protein phase, thus providing a conduit for their further infiltration into the mineral, which would increase the area of exposed crystal surface and thus facilitate crystal mineral dissolution in the acidic interior. In a previous study we demonstrated qualitatively that Madin-Darby canine kidney (MDCK-II) cells, which exhibit features of proximal/distal tubule cells (50), caused more extensive degradation of COM crystals generated from centrifuged and filtered urine, than those precipitated from the corresponding ultrafiltered (UF) urine (7). The aims of this investigation were 1) to measure the nonuniform strain and crystallite size of COM crystals precipitated from urine containing increasing amounts of the soluble crystal matrix extract (CME) derived from urinary COM crystals and 2) to compare quantitatively their rates of dissolution after attachment to renal epithelial cells.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Generation of urinary CaOx crystals containing increasing amounts of CME. Twenty-four-hour urine samples collected from six healthy laboratory colleagues were pooled and processed as described earlier (18, 47). This study was reviewed and approved by the Flinders Clinical Research Ethics Committee. The ultrafiltered (10 kDa) sample was divided into five portions, to which CME was added to give final concentrations of 0, 0.5, 1.0, 2.0, and 5.0 mg/l. Radiolabeled and nonradiolabeled urinary COM crystals were then precipitated from each sample (7), filtered (0.22 µm), washed (49), lyophilized, and stored at –70°C. Before use, crystals were sterilized by ethylene oxide and suspended (400 µg/ml) in PBS saturated with CaOx, which we have shown in unpublished work does not alter their binding to or internalization by cells, or their vulnerability to proteolytic degradation. The suspensions were sonicated for 10 min immediately before addition to the cell monolayer.
Unlabeled COM crystals used for microscopy, SDS-PAGE, and Western blotting and synchrotron X-ray diffraction (SXRD) were prepared separately (16).
Field-emission scanning electron microscopy and surface area measurement. Suspensions of the unlabeled COM crystals were examined using a Philips XL30 field-emission scanning electron microscope (FESEM) as described earlier (47, 49). Ten fields were randomly selected for each stub, and the length and breadth of 10 single crystals were measured in each field. Because single COM crystals settle upon the monolayer with one of their large {001} faces in direct contact with the cells, a reasonable approximation of contact surface area was obtained simply from assuming that crystals were rectangular.
SDS-PAGE and Western blotting. The lyophilized samples were mixed with reducing sample buffer, electrophoresed, and the gel was stained with silver as described previously (18, 47, 49). For Western blotting, after electrophoresis, proteins were electrophoretically transferred to nitrocellulose membrane followed by immunoblotting using antibodies detailed previously (18).
SXRD. SXRD patterns were collected and analyzed to calculate nonuniform lattice strain, which describes misalignment of crystalline blocks, and average crystallite size (or coherence length) as described previously (16–17). COM crystals grown slowly in gelatin at room temperature (16–17) were used as a "strain-free" reference.
Cell culture. MDCKII cells were generously provided by Dr. Carl Verkoelen (Erasmus University, Rotterdam, The Netherlands), and high-density quiescent cultures were prepared in 35-mm plastic plates using conditions as described earlier (18).
Measurement of mineral dissolution. The culture medium was removed by aspiration, and the cells were rinsed three times with 2 ml of PBS saturated with CaOx. One milliliter of PBS saturated with CaOx and 1ml of crystal suspension were added to each dish, which was gently agitated. To allow crystal binding and internalization, the dishes were incubated at 37°C in a humidified incubator (5% CO2-95% air) for 60 min. To remove unbound crystals the fluid was aspirated and the cells were washed as above. Three milliliters of growth medium containing 0.5% FCS were added to the plates, which were incubated for up to 48 h at 37°C in a humidified incubator (5% CO2-95% air). At time 0 the medium in one plate was removed and centrifuged, and the level of radioactivity in 600 µl of supernatant was counted in 4 ml of Ready Safe scintillation fluid (Beckman) (18). The attached crystals were then dissolved by adding 1 ml of PBS followed by 0.5 ml of concentrated HCl, and the cells were then removed completely with a scraper. The suspension was centrifuged and the radioactivity in 600 µl of supernatant was counted to give the total counts attached to the plate at time 0. Media from the remaining plates were counted at 2 min and 3, 6, 24, and 48 h, and the percent dissolution of each was calculated from the counts per minute in the supernatant, relative to those in the cells at time 0. The pattern of release of radioactivity into the medium was mirrored exactly by the decline in counts still attached to the cells: at any given time point, the counts released into the medium plus the counts still attached to the cells were consistently 100% of those originally bound to the cells. Nonetheless, results were expressed as radioactivity released into the medium because these data were less variable than the decline in counts attached to the cells.
Preliminary experiments showed that 70% of added crystals attached to the cells. To determine the extent of any dissolution not caused by the cells, 280 µg of each crystal type was incubated in fresh medium alone under identical conditions.
Effect of refreshing the culture medium on crystal dissolution. Because the release of [14C]oxalate into the medium stabilized after 6 h, an experiment was performed using crystals precipitated from ultrafiltered (UF) urine containing 0 and 5 mg CME/l to determine whether the decline in dissolution was caused by saturation of the cell culture medium with CaOx or depletion of nutrients. Dissolution was studied for a total of 10 days, with refreshment of the culture medium every second day. Because crystals could not be quantitatively retained in culture dishes in the absence of cells, the effect of the culture medium on dissolution beyond 48 h could not be tested.
Qualitative assessment of crystal degradation. Dissolution experiments were carried out as described above, but in culture dishes containing sterile 13-mm-diameter Thermanox Plastic coverslips (Nalge Nunc). After 24 h the coverslips were mounted on aluminium stubs, dried overnight at 37°C, coated with platinum, and examined using FESEM as described earlier (47, 49).
Statistical analysis. Each experiment was performed in triplicate, and the data are plotted as means ± SE. The absence of error bars from figures indicates that they are smaller than the symbol used for the means. Statistical comparisons between data sets were performed using Student's t-test, and a level of P < 0.05 was considered statistically significant.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Crystals generated from UF urine in the presence of increasing CME concentrations also consisted principally of COM. CME did not significantly alter the estimated surface areas of the crystals, with values of 52.79, 52.18, 49.0, 51.0, and 48.44 µm2 being observed for crystals in the presence of CME at concentrations of 0, 0.5, 1.0, 2.0, and 5.0 mg/l, respectively.
Figure 1 shows a Western blot stained for PTF1 of the CME of the crystals used in the experiments. PTF1 was absent from crystals grown in UF urine containing no CME (control) but migrated in the remainder as a well-defined band at 31 kDa, whose staining intensity increased sequentially in relation to the final CME concentration. Although not quantitative, these data demonstrate nonetheless that the amount of PTF1 included in CaOx crystals is proportional to the protein's concentration in the medium from which the crystals are precipitated.
|
|
|
|
|
20% of the theoretical maximum after 48 h (Fig. 4), we considered the possibility that this might have resulted from saturation of the culture medium with CaOx or depletion of nutrients. Accordingly, a longer experiment was performed in which the culture medium was replenished every 2 days, using crystals precipitated from UF urine containing 0 and 5 mg/l CME (Fig. 6). [14C]oxalate release approached a plateau of 20% by 1 day. However, each time the culture medium was refreshed, a further increase in dissolution was observed, attaining values at 10 days of 40 and 70% at CME concentrations of 0 and 5 mg/l, respectively. The pattern of release of radioactivity into the medium was mirrored by the decline in counts still attached to the cells (results not shown). In the absence of cells, dissolution of both types of crystals was identical, averaging 8.2% by 24 h.
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
The CME concentrations used in this study were 0.5, 1.0, 2.0, and 5.0 mg/l. These values were selected because previous immunological studies had reported that the normal concentration of PTF1 in human urine ranged from 0.9 to 35.3 nmol/l (6). Since the molecular mass of PTF1 is 30 kDa, those figures convert to from 0.03 to 1.1 mg/l. If we assume that other more abundant urinary proteins, HSA, for example, collectively account for an additional 1–3 mg/l of CME protein, it is reasonable to conclude that the approximate total physiological concentration of CME would be 0.5–4.0 mg/l, which is on the same order as the protein concentrations we used.
We have shown that 48 h after addition of crystals to the cells, the amount of [14C]oxalate released into the culture medium correlated directly with the concentration of CME in the urine from which the crystals had been precipitated. This correlation must have resulted from the action of the cells, since the amount of dissolution occurring in their absence was identical at all CME concentrations. [14C]oxalate release increased in proportion to CME concentration, as did the degree of crystal degradation, both of which must have resulted from digestion of intracrystalline proteins by cellular proteases, because crystals incubated in medium alone were not eroded. Since the amount of protein incorporated into the crystals is directly related to the ambient CME concentration (47), we can conclude that intracrystalline proteins facilitate the decay and dissolution of urinary crystals internalized by cultured renal epithelial cells.
Because the rate of [14C]oxalate release declined after 6 h at every CME concentration, despite the fact that dissolution was clearly incomplete, a further experiment was performed to test the possibility that dissolution was being limited by declining nutrients or supersaturation of the culture medium with dissolved CaOx. The culture medium was refreshed every 48 h for 10 days, and dissolution was quantified by measuring both the accumulation of [14C]oxalate in the culture medium and the fall in radioactivity in the cell monolayer. By day 10, 70% of the crystals precipitated in the presence of CME had dissolved, compared with only 40% of those generated in its absence. Thus the reduction in dissolution rate between 6 and 48 h observed previously, which occurred at all CME concentrations, probably resulted from a decrease in cellular metabolic activity caused by declining nutrients or by supersaturation of the culture medium with CaOx, which was consequently unable to accommodate further solute. Calculation (APPENDIX A) shows that the CaOx crystal dissolution that we observed up to 48 h was about four times greater than would be expected for pure inorganic CaOx in DMEM in a cell-free system at 37°C (5). Prima facie this suggests that the observed decline in dissolution rate between 6 and 48 h resulted from saturation of the culture medium with CaOx, but if that were so, the dissolution time courses at all CME concentrations should have converged toward the same limiting value. Although the curves for the two highest CME concentrations appear to approach convergence, the others do not (Fig. 4). This could perhaps be attributable to the fact that dissolution of COM crystals containing different amounts of CME would have released varying amounts of Ca- or oxalate-binding components in the medium, thereby producing different degrees of saturation (see also below). In any event, irrespective of whether the declining dissolution rate after 6 h resulted from supersaturation of the culture medium or reduced metabolic activity, it is clear that neither would be a factor in vivo.
The fourfold extent of dissolution we found, compared with that predicted from the data of Belliveau and Griffin (5), could have resulted from release into the culture medium of Ca-binding macromolecules that would have increased its limit of metastability. Belliveau and Griffin also measured dissolution in DMEM that did not contain FCS, whose components could have increased CaOx solubility by binding Ca or oxalate ions. Furthermore, dissolution in our study occurred in the presence of active cells, whose acidic phagolysosomal system is ideally equipped to disrupt and dissolve internalized crystals.
Crystals therefore bind to and are eroded and dissolved by renal epithelial cells. However, our findings are also relevant to the disposal of crystals precipitated or transported to other parts of the kidney. Although crystals are passed in the urine of hyperoxaluric rats (53), many are retained within lumina or attached to the epithelium of proximal tubular cells (12–13, 25). Some are later internalized (9, 25) and/or transported to the interstitium (9–19, 25, 53). Given the presence of tubular intracellular crystals in hyperoxaluric and idiopathic stone formers (13, 35) as well as in the renal interstitium of non-stone formers (8) and in patients with primary hyperoxalosis (10, 35), it is clear that any potential role of intracrystalline proteins in CaOx crystals is not confined to tubular epithelial cells. It is well documented that once crystals have been transported to the interstitium they dissolve or are destroyed (10–11, 23), probably by lysosomal enzymes in the macrophages and multinucleated giant cells that engulf them (9, 11–12, 23, 53), since CaOx crystals have been observed inside the phagolysosomes of renal tubular cells of rats with chronic hyperoxaluria (25). CaOx crystals are also dissolved by cultured macrophages (11) and within lysosomal inclusion bodies in BSC-1 cells (30), reinforcing the likelihood that lysosomal enzymes are involved in crystal degradation and eventual destruction.
There is considerable evidence that proteases may play an active role in stone formation and experimental nephrocalcinosis and in crystal-cell attachment. The organic matrix of kidney stones contains neutrophil elastase (45), a peptidase whose activity has been detected in human urine (44). The activities of a range of proteases, including cathepsins B, cathepsin L, and cathepsin D, occur in the rat kidney (27), while the glomerular basement membrane contains significant amounts of acid phosphatase β-galactosidase, β-glucuronidase, and acid protease (51). Activities of N-acetyl-β,D-glucosaminidase (NAG) and β,D-galactosidase are several-fold higher in the proximal tubule than in other segments (40). In mice, cathepsin D is most abundant in the cortical collecting tubule, and cathepsins B and H are located in the proximal convoluted tubule (43). High activities of β-galactosidase, β-hexosaminidase, and cathepsins B, H, L, and D are also present in human kidney and in confluent primary cultures of epithelia from proximal, collecting, and thick ascending limb tubules (20). Kugler et al. (28) noted that the human kidney is well equipped with membrane-bound and lysosomal peptidases, principally in the proximal tubules, whose brush borders contain high levels of aminopeptidase A, aminopeptidase M, 
-glutamyl transferase, and dipeptidylpeptidase IV. Activities of these enzymes also occur right through to the collecting ducts. The same authors also observed cathepsin B and dipeptidylpeptidase activity in the lysosomes of the proximal tubules and the small lysosomes of the distal tubule (28). The activity of cathepsins D (20, 43), B, H, and L (20) have also been demonstrated in collecting tubules.
Activities of the same enzymes, as well as other proteases, have also been detected in urine. Human urine contains type IV collagenase (42), NAG, cathepsins L and B (22, 52), cathepsin H (52), alkaline phosphatase, leucine aminopeptidase, and -glutamyl transpeptidase (22). The activities of NAG and β-galactosidase (3, 21) and also those of 
-glucosidase and -glutamyl transpeptidase (3) are elevated in urine from stone patients, as is the level of an unidentified serine protease that cleaves OPN (4). Similar changes accompany nephrocalcinosis induced in hyperoxaluric rats (24, 25, 56), in which COM crystal formation is associated with significant, proportional increases in urinary oxalate and activity of NAG and the brush-border enzymes alkaline phosphatase, leucine aminopeptidase, and -glutamyl transpeptidase (24, 25, 56).
Release of lysosomal enzymes has also been observed in cultured cells challenged with crystals. Within 4 h, urate crystals are found within intercellular spaces of MDCK cells, whose intracellular lysosomes release significant amounts of -galactosidase and β-glucuronidase (15). A similar increase in β-glucuronidase activity occurs when polymorphonuclear leukocytes are challenged with CaOx crystals (14). Furthermore, damage resulting from treatment of MDCK (19, 23) and LLC-PK1 cells (23) with oxalate or CaOx crystals is accompanied by a significant increase in ambient levels of a number of enzymes, including NAG, into the medium.
Conclusions. We have demonstrated that intracrystalline proteins promote the degradation and dissolution of urinary CaOx crystals in cultured renal epithelial cells in a dose-dependent fashion, thus supporting our hypothesis that intracrystalline proteins may defend against stone formation by facilitating the degradation and destruction of crystals retained within the kidney. These findings have significant ramifications for biomineral metabolism and in the pathogenesis of renal stones.
Appendix A: Calculation of COM Dissolution
Because of the high ionic strength, the equilibrium concentration product of Ca2+ concentration ([Ca2+]) [Ox2–] in DMEM at 25°C is 2.3 x 10–8, which is 10 times higher than the literature solubility product equilibrium constant, Ksp (5). Values of Ksp and equilibrium concentration products are 40–80% higher at 37°C than at 25°C (5). If we assume an increase of 60%, at 37°C in DMEM the equilibrium concentration product would be 3.68 x 10–8 at saturation. [Ca] in the DMEM we used was 1.36 mmol/l, and the 0.5% fetal calf serum would have increased this by a further 0.012 mmol/l, giving a total [Ca] of 1.37 mmol/l in the cell culture medium. Given this relatively high concentration, dissolution of COM in the medium can be assumed to have a negligible effect on the total [Ca]. Thus at saturation, the concentration of oxalate resulting from dissolution of added COM crystals would be on the order of 3.68 x 10–8 (mol/l)2 ÷ 1.37 x 10–3 (mol/l), or 2.686 x 10–5 mol/l. We added 400 µg of COM to each cell culture dish, of which 70% bound to the cells. The volume of medium in each dish was 3 ml, and our experimental data showed that the crystals dissolved by 17%. That is, 3 ml of medium contained 47.6 µg of oxalate, which is a concentration of 10.85 x 10–5 mol/l. Thus the concentration of COM in the medium resulting from dissolution of COM crystals must also be 10.85 x 10–5 mol/l, which is 4.039 times the calculated value of 2.686 x 10–5 mol/l (5).
|
GRANTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  Y.-C. Chien, D. L. Masica, J. J. Gray, S. Nguyen, H. Vali, and M. D. McKee Modulation of Calcium Oxalate Dihydrate Growth by Selective Crystal-face Binding of Phosphorylated Osteopontin and Polyaspartate Peptide Showing Occlusion by Sectoral (Compositional) Zoning J. Biol. Chem., August 28, 2009; 284(35): 23491 - 23501. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. A. Vervaet, A. Verhulst, P. C. D'Haese, and M. E. De Broe Nephrocalcinosis: new insights into mechanisms and consequences Nephrol. Dial. Transplant., July 1, 2009; 24(7): 2030 - 2035. [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Merchant, T. D. Cummins, D. W. Wilkey, S. A. Salyer, D. W. Powell, J. B. Klein, and E. D. Lederer Proteomic analysis of renal calculi indicates an important role for inflammatory processes in calcium stone formation Am J Physiol Renal Physiol, October 1, 2008; 295(4): F1254 - F1258. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
, Mean dissolution occurring in medium alone; 
, 
, 
, 
, and 
: crystals precipitated from UF urine containing CME at final concentrations of 0, 0.5, 1.0, 2.0, and 5.0 mg/l, respectively.
) and 5.0 (
