- Email: [email protected]
CHANGES IN URINE MACROMOLECULAR COMPOSITION DURING PROCESSING
0022-5347/00/1641-0230/0 THE JOURNAL OF UROLOGY® Copyright © 2000 by AMERICAN UROLOGICAL ASSOCIATION, INC.®
Vol. 164, 230 –236, July 2000 Printed in U.S.A.
CHANGES IN URINE MACROMOLECULAR COMPOSITION DURING PROCESSING SUZANNE MASLAMANI, PATRICIA A. GLENTON
AND
SAEED R. KHAN*
From the Department of Pathology, College of Medicine, University of Florida, Gainesville, Florida
ABSTRACT
Purpose: To determine the urinary crystallization inhibitory activity, urine is generally centrifuged and/or filtered. These preparative procedures may result in a total or partial removal of many macromolecular constituents implicated in crystallization. The main purpose of this study was to investigate the changes in urinary macromolecular composition following centrifugation and filtration. Materials and Methods: Twenty-four hour urine samples were collected from human volunteers. Each was divided into 4 aliquots; one was filtered, the other was centrifuged, another was centrifuged and filtered. The control sample was neither filtered nor centrifuged. Total protein and lipid contents of each sample were determined. Proteins were analyzed using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). Western blot analysis was performed using antibodies against osteopontin (OPN), prothrombin (PT) related proteins, inter-␣-inhibitor (I␣I) related proteins, Tamm-Horsfall protein (THP), and albumin (ALB). The effect of processing on incorporation of urinary proteins in crystal matrices was also examined. Calcium oxalate crystals were produced in processed and unprocessed urine samples by the addition of sodium oxalate. Crystals were harvested, de-mineralized and their proteins analyzed by SDS-PAGE and Western blotting. Results: Processing reduced the amounts of both proteins and lipids in the urine. Previously we identified phospholipids in the matrix of calcium oxalate crystals as well as the filtrate and retentate removed during filtration and centrifugation. Phospholipids have a high affinity for calcium-containing crystals. In the case of proteins, those with high molecular weights appeared to be clearly affected by filtration and centrifugation. Processing also appeared to influence the incorporation of proteins in the crystals. The matrix of crystals produced in processed urine contained less THP than those made in unprocessed urine, apparently a result of the loss of this higher molecular weight protein during processing. Incorporation of PT-related proteins, particularly fragment 1, was increased. Conclusions: We propose that selective inclusion of macromolecules is a result of an increase in available binding sites on crystal surfaces because of the removal of certain calcium binding substances such as phospholipids and proteins. Removal of larger macromolecules from the milieu may also provide a better access to the crystal surfaces. KEY WORDS: calcium oxalate, urolithiasis, osteopontin, bikunin, Tamm-Horsfall protein, prothrombin
Two forces mainly control crystallization in human urine: urinary supersaturation with respect to the stone salts and presence of crystallization modulators.1– 4 Most macromolecular modulators are either glycoproteins or glycosaminoglycans and can promote or inhibit crystal nucleation, growth and/or aggregation. To define the role of modulators in the formation of crystals and stones it is customary to investigate crystallization in vitro in the presence or absence of individual macromolecules, employing either aqueous inorganic solutions or the urine itself.3–5 Since inorganic solutions do not replicate the complex urinary environment, use of urine as the medium is generally encouraged. However, use of urine requires many processing steps such as filtration, centrifugation and dilution,3, 5–9 which can drastically change the milieu10 and affect crystallization. This study was undertaken to investigate the effect of urine processing on its
protein contents with particular reference to osteopontin (OPN), prothrombin (PT) related proteins, inter-␣-inhibitor (I␣I) related proteins, Tamm-Horsfall protein (THP), and albumin (ALB). Twenty-four hour urine specimens were collected from male and female human volunteers. Each specimen was divided into 4 samples: one remained unprocessed, the second was filtered, the third was centrifuged and the fourth was centrifuged and filtered. Total protein content of each sample were determined using a biochemical assay. Proteins were analyzed using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and Western blot analysis. The effect of processing on incorporation of urinary proteins in crystal matrices was also examined. Calcium oxalate crystals were produced in processed and unprocessed urine samples by addition of sodium oxalate. Crystals were harvested, demineralized and their proteins analyzed by SDS-PAGE and Western blotting.10
Accepted for publication January 19, 2000. * Requests for reprints: Department of Pathology, PO Box 100275, College of Medicine, University of Florida, Gainesville, Florida 32610. Supported by NIH grant #RO1 DK41434. MATERIALS AND METHODS This work was part of a thesis presented to the graduate school of Urine collection and processing. 24-hour urine samples the University of Florida, for the partial fulfillment of the requirewere collected from five males and four females. During ments for the degree of Master of Science. 230
231
URINE MACROMOLECULAR COMPOSITION
collection, the specimens were maintained at room temperature: approximately 24C. One ml. of 20% sodium azide (Fisher Scientific, Norcross, GA), an anti-bacterial agent, was placed in collection bottles prior to collection. The volume and pH of each specimen were measured and 250 ml. aliquots of urine were placed in Nalgene bottles marked “NT” (no processing), “F” (filtered), “C” (centrifuged), and “C⫹F” (centrifuged and filtered). Specimens were filtered using 0.22 m. sodium acetate filters (Fisher Scientific, Norcross, GA) and centrifuged at 10,000 ⫻g at 20C for 20 minutes using the J-14 rotor of the J2 to 21 centrifuge (Beckman Instruments, Westbury, NY). Ten ml. of urine from each sample was concentrated to 1 ml. using speed vacuum model RC 10.10 (Jouan Inc., Winchester, VA) and then lyophilized (Flexi-Dry lyophilizer from FTS Systems, Stone Ridge, NY). 500 L. of 0.05 M Tris buffer pH 7.4 was added to each tube. These specimens were kept in a freezer at ⫺20C until the SDS-PAGE and western blotting procedures were performed. Preparation of crystal matrix. The urine samples were allowed to warm to 37C in a shaking water bath (Fisher Scientific, Norcross, GA) and divided in equal halves. Calcium oxalate crystals were induced by the addition of 15 ml./l. of 0.1 M sodium oxalate and incubated for 3 hours. At the end of incubation period, the urine specimens were centrifuged at 10,000 ⫻g for 25 minutes at 20C in the J2 to 21 centrifuge (Beckman Instruments, Westbury, NY). The supernatant was aspirated and the pellet placed in a microcentrifuge tube and washed three times with 2 ml. of distilled water. The contents of the tubes were dried for 24 hours in the Flexi-Dry lyophilizer. The crystals were demineralized by treatment with 5 ml. of 0.25 M ethylene diamine tetraacetic acid (EDTA) pH 8.0 at 4C for 3 days with continuous stirring. The extract was centrifuged at 10,000 ⫻g for 5 minutes. Supernatant was dialyzed against water for 24 hours at 4C using dialysis tubing with a 6 to 8 kDa cut-off (Spectrum Medical Industries, CA). Protein assay. Protein concentration in the urine and crystal matrix extract was determined using Bio-Rad’s Dye reagents (Bio-Rad Laboratories, Hercules, CA). The protein standards (Sigma-Aldrich; St. Louis, MO) were prepared using bovine serum albumin to provide a linear range of 0 g./ml. to 10 g./ml. The assay was performed in duplicate. Urine samples were diluted 1:2 with water prior to analysis. Fifty L. of each standard were mixed with 2.5 ml. of a dilute 1:4 mixture of Bio-Rad’s Dye Reagent Concentrate. The tubes were vortexed and allowed to incubate for 5 minutes. Absorbance was measured at 595 nm using a Pharmacia LKB Ultrospec 3 UV Visible Light (Hoffer Pharmacia Biotech, Piscataway, NJ). Lipid assay. To demonstrate the loss of lipids by filtration and centrifugation we isolated lipids from the urine. The methods are described in detail in earlier publications.11 In brief, 400 ml. of processed or unprocessed (native) urine was mixed with 1.2 l. of 2:1 chloroform: methanol. The mixture was shaken and placed on an end-over-end mixer for 24 hours at 4C. Samples were then centrifuged at 7000 r.p.m. to achieve phase separation. The top portion was removed and combined as the aqueous layer; the middle layer was recovered as the interface; and the lower phase was collected and pooled as the organic layer. After evaporation to a smaller volume, the organic sample was Folch-washed twice, pooling the respective phases with the previous ones. The organic phase was then lyophilized and weighed. SDS-PAGE analysis. Separation of the proteins was accomplished using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). The samples were prepared by adding 6 L. of sample buffer (2.5 ml. 0.5 M Tris base pH 6.8 ⫹ 4 ml. 10% SDS ⫹ 2 ml. glycerol ⫹ 0.002 bromophenol blue ⫹ 0.31 gm. dithiothreitol ⫹ distilled water to a volume of 10 ml.) to 6 L. of each sample from the whole urine and the crystal matrix extract specimens. Ten L. of this mixture
was applied to a designated lane on a 10 to 20% SDS-PAGE. The gels were electrophoresed at 70 V for approximately 2 to 2.5 hours with standard Gel Running Buffer (Tris base 3 gm./l., glycine 14 gm./l., SDS 1 gm./l.). The gels were stained with Coomassie blue to detect protein bands. Western blotting. Individual proteins were identified by Western blot analysis using specific antibodies. After the SDS-PAGE electrophoresis, the gel was removed and placed in a small amount of Blotting Buffer pH 8.3 to equilibrate for 15 minutes. Blotting Buffer was made with 3.03 gm. Tris (Fisher Biotech BP152 to 1) and 14.4 gm. glycine (Fisher Biotech BP381–1) in 500 ml. distilled water, 200 ml. methanol (Fisher Biotech A452–1), and 300 ml. distilled water (no pH adjustment was performed). The proteins were transferred to a nitrocellulose membrane (BioBlot-NC Cat. #8801, Cambridge, MA) using the Bio-Rad Power Pac 200 set at 100V for 15 minutes, then 50V for 45 minutes. The cassette positions within the electrode holder were exchanged with each other and then the blotting continued for 15 minutes at 100V and 50V for the remaining 45 minutes. When blotting was complete, the nitrocellulose membranes were removed and placed in weigh boats with 10 ml. Blocking Buffer for at least 30 minutes. Blocking Buffer was made with 1 gm. powdered milk and 20 ml. Tris buffered saline with Tween (TBST). Tris buffered saline with Tween was made with 1.546 gm. Tris HCl, 2.922 gm. NaCl, 0.5 ml. Tween-20 (Bio-Rad Cat. # 170-6531). The pH was adjusted to 8.0 using HCl or NaOH. When blocking was complete, the primary antibody for the particular protein to be examined was added to the membrane at a starting dilution of 1:1000. Immunochemical staining was performed at room temperature by using different polyclonal antibodies. Polyclonal antibodies used at 1:1000 dilution were as follows: THP, OPN, ALB, prothrombin, and I␣I. THP antibody was made in our laboratory by injecting rat THP into rabbit as antigen (Kel Farm Laboratory, FL) and OPN antibody was a gift from Dr. E. M. Worcester (Department of Medicine, Medical College of Wisconsin). PT antibody (ICS Biomedical Inc., CA) reacts with prothrombin fragment-1, fragment 1⫹2 as well as prothrombin. I␣I antibody (Accurate Chemical Scientific Corporation, NY) reacts with all three chains of I␣I, heavy chains H1, H2, and the light chain, bikunin. Then the membranes were incubated with the secondary antibodies marked with alkaline phosphates (Hyclone Laboratories. UT) and used at a dilution of 1:5000. Finally, the membranes were allowed to develop for 5 minutes by using a substrate kit (Bio Rad Laboratories, CA.). Quantification and statistical analysis. The gels were read using a Pharmacia LKB Imagemaster DTS densitometer (Piscataway, NJ). The optical density (OD) of each band was computed and then the sum of the optical densities in each lane (representing a particular preparative treatment) was determined. The calculated optical densities in each treatment lane were multiplied by the 24-hour urine output of that particular individual to get an OD/24 hours value for each treatment. The OD for each individual’s Filtered, Centrifuged, and Centrifuged ⫹ Filtered group was compared with their own No Treatment OD/24 hours calculation. A one-tailed paired t test with a significance of p ⬍0.05 was used to analyze the data. RESULTS
As shown in table 1, filtration and centrifugation had significant effect on urinary concentrations of proteins and lipids. Both filtration and centrifugation resulted in a reduction in the protein and lipid contents of the urine, filtration being more effective in removing them from the urine than centrifugation. Processing of the urine also had a significant effect on its metastable limit with respect to crystallization of cal-
232
URINE MACROMOLECULAR COMPOSITION
TABLE 1. Effect of various preparative procedures on urinary concentration of proteins, lipids (per unit volume of urine) and metastable limit of the urine with respect to crystallization of CaOx No Treatment
Filtration
Centrifugation
0.67 ⫾ 0.34a Metastable Limit (mM/l. of oxalate added) 0.58 ⫾ 0.33 0.67 ⫾ 0.34a Protein (Bradford assay g./ml., n ⫽ 9) 515.8 ⫾ 128.4 436.4 ⫾ 100.1a 456.4 ⫾ 115.5a Protein (SDS-PAGE, OD/24 hour, n ⫽ 4) 1988.0 ⫾ 585.8 1323.3 ⫾ 455.4a 1986.5 ⫾ 1069.2b Total Lipids (mg./l., n ⫽ 5) 296.6 ⫾ 50.8 203.2 ⫾ 47.6a 249.8 ⫾ 81.4 (a Significantly different from unprocessed, b significantly different from filtered, c significantly different from centrifuged).
cium oxalate. An ultrastructural examination of the pellet following centrifugation and retentate after filtration (fig. 1) showed them to contain amorphous material, fibrous substances and membranous fragments and vesicles. SDS-PAGE analysis of urine revealed the presence of 10 to 15 bands with approximate molecular weights ranging between 17 kDa to 150 kDa. Some samples showed bands even at 220 kDa. Major bands were recognized at approximate molecular weights of 108, 80, 67, 58, 48, 40, 35, 28, and 22 kDa (fig. 2, A). Western blot analysis using specific antibodies recognized osteopontin (not illustrated) at approximately 22, and 28 kDa, both bands being diffuse and non-descript. A band at 80 was also positive for OPN in some samples. Five bands at approximately 85, 70, 49, 40 and 28 kDa were recognized as albumin with 70, 49 and 40 kDa bands being most noticeable and common in all samples (fig. 3, A). Bands at approximate molecular weights of 80, 48, and 35 were positive for prothrombin related proteins (3, B). These bands had similar appearance and were detected in almost all samples. Three bands at 107, 68 and 50 kDa were positive for
FIG. 1. Ultrastructure of retentate on filter illustrating fibrous, amorphous and membranous substances.
Centrifugation and Filtration 0.68 ⫾ 0.38a 385.8 ⫾ 72.8abc 1465.7 ⫾ 777.8ac 202.5 ⫾ 45.6ac
FIG. 2. SDS-PAGE analysis. Lane 1: Molecular weight markers, Lane 2: Unprocessed urine, Lane 3: Filtered urine, Lane 4: Centrifuged urine, Lane 5: Centrifuged and filtered urine. A, urine. Comparison of lane 2 with 3 and 5 indicates absence of high molecular band in filtered urine. Centrifugation (Compare Lane 4 with 2 and 3) resulted in weakening of high molecular weight band. B, crystal matrix. Comparison of lane 2 with the others indicates that crystals made in processed urine have lesser number of bands. Bands also appear lighter except for band around 35 kDa, which is more intense in crystals made in processed urine.
Tamm-Horsfall protein, 107 and 68 kDa bands being common and most conspicuous (3C). I␣I was recognized at approximate molecular weights of 123, 74, 68, 62, 48, 40 and 30 kDa. Bands at 48 and 40 kDa were prominent and common to all samples (3D). Urinary protein profile was negatively impacted by various preparative procedures (fig. 2, A). A comparison of bands in Lane 2 with those in Lanes 3, 4 and 5 illustrates that high molecular weight proteins were at least partially removed by the preparative techniques. The bands representing these protein fractions were either totally absent or lighter in appearance in filtered urine (Lane 3) than in the unprocessed urine (Lane 2) samples. Centrifugation (Lane 4) also resulted in removal of proteins but to a lesser degree. The effect of combined treatment of centrifugation and filtration (Lane 5) appeared to be similar to filtration. A comparison of band densities of the Western blots of urinary proteins (table 2) also showed that centrifugation and filtration both resulted in reduction of albumin, THP and I␣I. SDS-PAGE analysis of the matrix proteins isolated from calcium oxalate crystals is illustrated in fig. 2, B. Seven distinct bands from estimated molecular weights of 28-kDa to120 kDa were recognized in proteins isolated from the crystals induced in unprocessed urine. Crystals produced in processed urine showed fewer and weaker bands. However a band at around 35 kDa was much more pronounced and dense in crystals produced in processed urine particularly those produced in filtered urine. Western blot analysis (fig. 4) identified OPN, ALB, THP, PT- and I␣I - related proteins in the organic matrix of calcium oxalate crystals produced in both the processed and unprocessed urine samples. A single band at 28 kDa was recognized, as OPN (not illustrated). Albumin was detected at 70 and 48 kDa (fig. 4, A). Up-to three bands stained positive for PT -related proteins. Every sample showed a dense band near 35 kDa (fig. 4, B). Interestingly, the crystals
233
URINE MACROMOLECULAR COMPOSITION
FIG. 3. Western blots of urine. Lanes are same as in fig. 2. A, albumin. All lanes have approximately same number of bands with similar intensities. B, prothrombin. All lanes have 3 major bands. C, Tamm-Horsfall Protein. Band intensity is lower in filtered urine (Compare Lane 2 with 3 and 5). D, inter-␣-inhibitor. Number of bands and their intensity appear the same.
made in processed urine showed more and denser PT-positive bands than the crystals made in unprocessed urine. Crystals made in unprocessed urine showed many THP positive bands (fig. 4, C), the band at 107 kDa being the most prominent. Crystals from processed urine also showed this band but they were very weak. All samples showed a I␣I - related protein positive band at 68 kDa (fig. 4, D). A few samples showed another band at 48 kDa and a faint band was sometime seen at 40 kDa. DISCUSSION
Commonly, urine is filtered and/or centrifuged prior to testing crystallization metastable limits or investigating the activity of individual crystallization inhibitors.5–9 Our preliminary studies indicated the possibility that filtration and centrifugation may remove some of the urinary modulators of crystallization.10 To further explore this issue we studied specific urinary crystallization modulators before and after filtration and centrifugation. Biochemical quantification showed that filtration results in significant reductions in the amounts of both the proteins and lipids in the urine. Centrifugation also results in reduction of these macromolecules but to a lesser extent. Ultrastructural examination showed that both the retentate following filtration and pellet following centrifugation contain fibrous material and membrane bound vesicles and fragments. Apparently fibrous material represents the separated proteins while membranous vesi-
FIG. 4. Western blots of crystal matrix. Lanes are same as in fig. 2. A, albumin. Two bands are identifiable. Band intensity is higher in matrix of crystals made in unprocessed urine. B, prothrombin. Crystals made in unprocessed urine have lighter bands in matrix while those made in processed urine show very intense bands particularly at 35 kDa. C, Tamm-Horsfall Protein. More bands with higher density are seen in matrix of crystals made in unprocessed urine (Lane 2) than those made in processed urine (Lanes 3 to 5). In addition they contain high molecular bands. D, inter-␣-inhibitor. At least two bands are identifiable in each sample.
cles are the source of extracted lipids. Removal of these macromolecules could have significant effect on urine’s crystallization potential since most urinary proteins are recognized as inhibitors of various crystallization processes,3, 4 while membranes and their lipids are shown to promote crystal nucleation.11, 12 Changes in urine’s metastable limit following filtration and centrifugation support this assertion. SDS-PAGE and Western blot analysis were employed to identify the proteins that were lost or had their concentration reduced during preparative techniques. We particularly examined the effect of processing on five proteins, THP, OPN, ALB, PT related proteins and I␣I related proteins, which have been shown to play significant role in calcium oxalate crystallization and nephrolithiasis. Salient features of these proteins are described in table 3. Coomassie revealed filtration resulted in a total loss of some bands representing high molecular weight proteins. Some other bands appeared weaker in processed urine. Western blot analysis using specific antibodies showed that processing had no apparent qualitative effect on the five bands that stained positive for albumin. But an analysis of the optical density of protein bands on western blot revealed significant reduction in urinary albumin concentration following filtration and centrifugation. Prothrombin antibody stained 3 main bands representing prothrombin at approximately 80 kDa, Fragment
TABLE 2. Effect of various preparative procedures on selected proteins No Treatment Albumin (n ⫽ 4) 5921 ⫾ 1982 Tamm-Horsfall Protein (n ⫽ 4) 5150 ⫾ 3267 Inter-␣-Inhibitor (n ⫽ 5) 5142 ⫾ 4068 Result of Western Blot Analysis (Optical Density/24 hour of urine, cantly different from centrifuged).
Filtration
Centrifugation
Centrifugation and Filtration
5903 ⫾ 2896 5119 ⫾ 1140 3922 ⫾ 1892a 2449 ⫾ 1896 4587 ⫾ 3672b 2780 ⫾ 1966a, b 2057 ⫾ 952 4203 ⫾ 3762 2717 ⫾ 2017 a significantly different from unprocessed, b significantly different from filtered, c signifi-
234
URINE MACROMOLECULAR COMPOSITION TABLE 3. Urinary proteins with potential to modulate crystallization Protein Name
Tamm-Horsfall Protein4, 6, 13, 15, 16 Osteopontin4, 6, 14–17 Human Serum Albumin4, 6, 18 Urinary Prothrombin Fragment 14, 6, 19–23 Inter-␣-Inhibitor4, 6, 24, 25 H1 (Heavy Chain 1) H2 (Heavy Chain 2) HI-30 (Bikunin)26–30
MW (kDa)
Origin
Unique Features
Role in Crystallization
Presence in Urine
Presence in Stone
80–100
TAL of the Kidney
12% ASP
Promoter, Inhibitor of Aggregation
Yes 20–100 mg/day
Yes
42–80
TDL, TAL of the Kidney Plasma
RGD Sequence
Yes 2.4–3.7 mg/day
Yes
Yes 1.6–34.2 mg/day
Yes
Plasma ?, TAL of the Kidney Plasma?
10 GLA Residues
Inhibitor of Nucleation, Growth, Aggregation Facilitates Binding of Other Proteins to Crystals Inhibitor of Crystal Growth and Aggregation
Yes 13.4 nM/day
Yes
Yes 2–10 mg/day Yes Yes 5.01 g/ml
Yes Yes Yes
68 31–35 85 71 30–35
Binds to Crystals
Sulfated GAGs
Plasma ?, Kidney
Inhibitor of Crystal Nucleation, Growth, and Aggregation Location In Kidneys Based On Studies In Rats; Urinary Excretion Rate In Normal Humans; Role In Crystallization Based Mostly On CaOx Crystal Formation; ASP, Aspartic acid; GLA, g-Carboxyglutamic acid; RGD, Arginine-Glycine-Aspartic Acid; PT, Proximal Tubule; TAL, Thin Ascending Limb Of The Loop of Henle; TDL, Thick Descending Limb Of The Loop of Henle.
F1⫹2 at 48 kDa and Fragment F1 at 35 kDa. None appeared affected by filtration or centrifugation. THP was recognized as 3 bands at 107 kDa, 68 kDa and 50 kDa, with 107 kDa band being the most prominent. The fact that MWR of THP is slightly different from expected single band between 80 to 100 and that the 3 bands stained positive, can be due to the buffer or the SDS-PAGE system as well as the possibility of enzymatic degradation of protein in the urine. However the most important factor is that this detection system is an antibody which is highly specific for THP.15 Comparison of band intensity between unprocessed and processed urine showed that the 107 kDa band was definitely affected by the filtration. Other THP positive bands also appeared lighter. The I␣I family of proteins consists of I␣I (⬎172 kDa), pre-␣inhibitor (115 kDa), H1 (86 kDa), H2 (71 kDa), H3 (65 kDa) and bikunin (38 kDa). The I␣I antibody we used can interact with all of them. All urinary samples contained H1, H2, H3 and bikunin with bikunin being the most common. Some sample also contained pre-␣I. Filtration or centrifugation had no effect on I␣I-positive bands ranging between 74 kDa to 28 kDa. However the band representing higher molecular weight pre-␣I disappeared after processing. We have also shown a significant reduction in various urinary lipids by filtration or centrifugation of the urine.31 Apparently, these procedures remove many urinary substances including various lipids, albumin, THP, and high molecular weight members of the I␣I family. A number of other studies have also shown that centrifugation and filtration of urine remove THP and albumin.32–33 Analysis of crystal matrix provided the evidence that urine processing also affects the incorporation of proteins in crystals. The matrix of crystals produced in unprocessed urine contained many more proteins than those produced in processed urine. High molecular weight proteins were lost, as suggested by the absence of high molecular weight bands. The concentration of a few other proteins appeared lower in the matrix of crystals made in processed urine, as indicated by weaker bands. Western blot analysis showed a substantial reduction in THP inclusion because of its obvious loss during filtration and centrifugation. The amount of albumin included also appeared to be reduced perhaps because of its removal during processing. There was a distinct increase in the incorporation of PT-related proteins, particularly prothrombin fragment-1, when crystals were produced in processed urine. A 68 kDa I␣I-positive band also showed increased intensity in the matrix of these crystals. It has been suggested that incorporation of organic material in the crystal matrix is a highly selective process.34 –36 Proteins such as albumin and THP are excluded while prothrombin fragment-1 is selectively included. One report concluded that “centrifugation and filtration of the urine before crystallization had little effect on the qualitative protein
contents of the crystal extracts” and that “Tamm-Horsfall glycoprotein was notably absent from the crystal extracts (matrix)”.34 Interestingly, electrophoretograms presented in this study clearly showed that processing of urine resulted in a loss of many proteins including albumin and THP, and that both THP and albumin were present in the matrix of crystals produced in whole urine. Only the matrix of crystals made in the ultra-filtered urine showed a single band, which at the time was designated crystal matrix protein and later identified as urinary prothrombin-fragment-1.20 –22 Our results presented here and elsewhere10 clearly demonstrate the incorporation of both THP and albumin in the crystal matrix when experiments were carried out in unprocessed urine. We have previously described the presence of lipids in the organic matrix of both calcium oxalate and calcium phosphate crystals produced in unprocessed urine in vitro. In addition almost all the urinary phospholipids were assimilated in the crystal matrix whereas only a fraction of urinary neutral and glycolipids became incorporated.12 The presence of various phospholipids, OPN, THP and ALB, in crystal matrices should not be surprising since all of them can interact with crystal surfaces, have been shown to be involved in calcium oxalate crystallization as modulators of nucleation, growth or aggregation, and are important constituents of stone matrix (table 3). Ultrastructural studies of calcium oxalate human urinary stones37 as well as deposits in rat kidneys38 have shown both OPN and THP to be intimately associated with crystals. OPN becomes involved very early in crystal formation and is even located inside the crystals. THP appears to promote aggregation by binding to the surfaces of crystals and connecting them together. Obviously the traditional processing of urine involving centrifugation and filtration removes some macromolecules, resulting in their decrease or absence from the organic matrices of crystals made in these urines. Removal of certain macromolecules through filtering or centrifugation would increase the likelihood of interaction between growing crystals and the residual macromolecules, thereby increasing the probability of their binding to crystal surfaces and eventual incorporation in crystal matrix. Substances eliminated during processing may share binding sites on crystal surfaces with those that are left behind. For example many lipids and proteins can bind to crystals, particularly calcium-containing crystals. They may also hinder the movement of other macromolecules to the vicinity of crystal surfaces. Thus in the experiments where urine was processed before crystal induction, the selective inclusion of some macromolecules may actually be the result of exclusion of others. Activity of crystallization modulators is generally measured in simple inorganic solutions and on rare occasions in undiluted (whole) or diluted urine.2, 5, 6 It is justifiably argued that the inhibitory potential of a macromolecule should
URINE MACROMOLECULAR COMPOSITION
be tested in whole urine since it is the only medium representative of the physiological state in which crystals and stones form. We must, however, recognize that kidney stones form in urine when it is still in the renal tubules and has a different constitution than the excreted bladder urine.39, 40 Even the “whole” urine is almost always processed, which as we have shown here, results in the removal of many substances with a direct role in crystallization. Removal of these modulators may erroneously exaggerate the role of the remaining macromolecules. In the case of diluted urine, activity of the inhibitors may change because of their lowered concentrations.2 Apparently every methodology used has its limitations, and results obtained using any methods need to be carefully evaluated with respect to their advantages and disadvantages. The main purpose of our investigation was to determine the effect of preparative procedures on the protein contents of urine. Both filtration and centrifugation remove proteins and lipids from the urine. The removal may influence crystallization activity in the urine because both proteins and lipids are known to modulate crystallization of calcium phosphate as well as calcium oxalate. Such a removal may also be responsible for the reported exclusion of proteins such as THP and albumin and the disproportionate inclusion of others, such as prothrombin fragment-1, in the matrix of crystals produced in vitro in processed urine. We thank Professor Raymond L. Hackett for reviewing the manuscript and Mrs. Patricia Khan for reviewing the manuscript and preparation of figures.
REFERENCES
1. Finlayson, B.: Physicochemical aspects of urolithiasis. Kidney Int, 13: 344, 1978 2. Hess, B. and Kok, D. J.: Nucleation, growth, and aggregation of stone forming crystals. In: Kidney Stones: Medical and Surgical Management. Edited by F. L. Coe, M. J. Favus, C. Y. C. Pak, J. H. Parks and G. M. Preminger. Philadelphia: Lippincottt-Raven Publishers, pp 3–32, 1996 3. Ryall, R. L. and Stapleton, A. M. F.: Urinary macromolecules in calcium oxalate stone and crystal matrix: good, bad, or indifferent. In: Calcium Oxalate in Biological Systems. Edited by S. R. Khan. Boca Raton: CRC Press, pp 265–290, 1995 4. Khan, S. R.: Interactions between stone-forming calcific crystals and macromolecules. Urol Int, 59: 59, 1997 5. Kavanagh, J. P.: Calcium oxalate crystallization in vitro. In: Calcium Oxalate in Biological Systems. Edited by S. R. Khan. Boca Raton: CRC Press, pp 1–21, 1995 6. Ryall, R. L., Hibberd, C. M. and Marshall, V. R.: A method for studying inhibitory activity in whole urine. Urol Res, 13: 285, 1985 7. Asplin, J. R., Bushinsky, D. A., Singharetnam, W., Parks, J. H. and Coe, F. L.: Relationship between supersaturation and crystal inhibition in hypercalciuric rats. Kidney Int, 51: 640, 1997 8. Baumann, J. M. and Wacker, M.: The direct measurement to inhibitory capacity to crystal growth of calcium oxalate in undiluted urine and other inhibitor containing solutions. Urol Res, 8: 171, 1980 9. Pak, C. Y. C. and Holt, K.: Nucleation and growth of brushite and calcium oxalate in urine of stone formers. Metabolism, 25: 665, 1976 10. Atmani, F., Glenton, P. A. and Khan, S. R.: Identification of proteins extracted from calcium oxalate and calcium phosphate crystals induced in the urine of healthy and stone forming subjects. Urol Res, 26: 1, 1998 11. Khan, S. R.: Heterogeneous nucleation of calcium oxalate crystals in mammalian urine. Scan Microsc, 9: 597, 1995 12. Khan, S. R., Atmani, F., Glenton, P. A., Hou, Z-C., Talham, D. R. and Khurshid, M.: Lipids and membranes in the organic matrix of urinary calcific crystals and stones. Calcif Tissue Int, 59: 357, 1996 13. Hess, B.: Tamm-Horsfall glycoprotein and calcium nephrolithiasis. Min Electr Metab, 20: 393, 1994
235
14. Hoyer, J.: Uropontin in urinary calcium stone formation. Min Electr Metab, 20: 385, 1994 15. Gokhale, J. A., Glenton, P. A. and Khan, S. R.: Biochemical and quantitative analysis of Tamm Horsfall protein in rats. Urol Res, 25: 347, 1997 16. Gokhale, J. A., Glenton, P. A. and Khan, S. R.: Localization of Tamm Horsfall protein and osteopontin in a rat nephrolithiasis model. Nephron, 73: 456, 1996 17. Shiraga, H., Min, W., Vandusen, W., Clayman, M. D., Miner, D., Terrell, C. H., Sherbotie, J. R., Foreman, J. W., Przysiecki, C., Neilson, E. G. and Hoyer, J. R.: Inhibition of calcium oxalate crystal growth in vitro by uropontin: another member of the aspartic acid-rich protein superfamily. Proc Natl Acad Sci USA, 89: 426, 1992 18. Dussol, B., Geider, S., Lilova, A., Le´onetti, F., Dupuy, P., Daudon, M., Berlan, Y., Dagorn, J. and Verdier, J.: Analysis of the soluble organic matrix of five morphologically different kidney stones. Evidence for a specific role of albumin in the constitution of the stone protein matrix. Urol Res, 23: 45, 1995 19. Doyle, I. R., Marshall, V. R., Dawson, C. J. and Ryall, R. L.: Calcium oxalate crystal matrix extract: the most potent macromolecular inhibitor of crystal growth and aggregation yet tested in human undiluted human urine in vitro. Urol Res, 23: 53, 1995 20. Stapleton, A. M. F. and Ryall, R. L.: Blood coagulation proteins and urolithiasis are linked: crystal matrix protein is the F1 activation peptide of human prothrombin. Br J Urol, 75: 712, 1995 21. Suzuki, K., Moriyama, M., Nakajima, C., Kawamura, K., Miyazawa, K., Tsugawa, R., Kikuchi, N. and Nagata, K.: Isolation and partial characterization of crystal matrix protein as a potent inhibitor of calcium oxalate crystal aggregation: evidence of activation peptide of human prothrombin. Urol Res, 22: 45, 1994 22. Stapleton, A. M. F., Dawson, C. J., Grover, P. K., Hohmann, A., Comacchio,R., Boswarva, V., Tang, Y. and Ryall, R. L.: Further evidence linking urolithiasis and blood coagulation: urinary prothrombin fragment-1 is present in stone matrix. Kidney Int, 49: 880, 1996 23. Stapleton, A. M. F., Seymour, A. E., Brennan, J. S., Doyle, I. R., Marshall, V. R. and Ryall, R. L.: Immuno-histochemical distribution and quantification of crystal matrix protein. Kidney Int, 44: 817, 1993 24. Enghild, J. J., Thogersen, I. B., Pizzo, S. V. and Salvesen, G.: Analysis of inter-␣-trypsin inhibitor and a novel trypsin inhibitor, pre-␣-trypsin inhibitor, from human plasma. Polypeptide chain stoichiometry and assembly by glycan. J Biol Chem, 264: 15975,1989 25. Salier, J. P., Rouet, P., Raguenez, G. and Daveau, M.: The inter-␣-inhibitor family: from structure to regulation. Biocehem J, 315: 1, 1997 26. Atmani, F., Mizon, J. and Khan, S. R.: Identification of uronic acid rich protein as urinary bikunin, the light chain of inter␣-inhibitor. Eur J Biochem, 236: 984, 1996 27. Marengo, S. R., Resnick, M. I, Yang, L. and Chung, J-Y.: Differential expression of urinary inter-␣-trypsin inhibitor trimers and dimers in normal compared to active calcium oxalate stone forming men. J Urol, 159: 1444, 1998 28. Iida, S., Peck, A. B., Johnson-Tardieu, J., Moriyama, M., Glenton, P. A., Byer, K. J. and Khan, S. R.: Temporal changes in mRNA expression for bikunin in the kidneys of rats during calcium oxalate nephrolithiasis. J Am Soc Nephrol, 10: 986, 1999 29. Medetognon-Benissan, J., Tardivel, S., Hennequin, C., Daudon, M., Drueke, T. and Lacour, B.: Inhibitory effect of bikunin on calcium oxalate crystallization in vitro and urinary bikunin decrease in renal stone formers. Urol Res, 27: 69, 1999 30. Atmani, F. and Khan, S. R.: Role of bikunin in the inhibition of calcium oxalate crystallization. J Am Soc Nephrol, 10: S385, 1999 31. Khan, S. R., Maslamani, S. A., Atmani, F., Glenton, P. A., Opalko, F. J., Thamilselvan, S. and Hammett-Stabler, C.: Membranes and their constituents as promoters of calcium oxalate crystal formation in human urine. Calcif Tissue Int, 66: 90, 2000 32. Wiggins, R. C.: Uromucoid (Tamm-Horsfall glycoprotein) forms different polymeric arrangements on a filter surface under different physicochemical conditions. Clin Chim Acta, 162: 329, 1987
236
URINE MACROMOLECULAR COMPOSITION
33. Doyle, I. R., Ryall, R. L. and Marshall, V. R.: The effect of low speed centrifugation and millipore filtration on the urinary protein contents. In: Urolithiasis. Edited by V. R. Walker, R. A. L. Sutton, E. C. B. Cameron, C. Y. C. Pak and W. G. Robertson. New York: Plenum Press, pp 593–594, 1989 34. Doyle, I. R., Ryall, R. L. and Marshall, V. R.: Inclusion of proteins into calcium oxalate crystals precipitated from human urine: a highly selective phenomenon. Clin Chem, 37: 1589, 1991 35. Morse, R. M. and Resnick, M. I.: A new approach to the study of urinary macromolecules as a participant in calcium oxalate crystallization. J Urol, 139: 869, 1988 36. Morse, R. M. and Resnick, M. I.: A study of the incorporation of urinary macromolecules onto crystals of different mineral compositions. J Urol, 141: 641, 1989
37. McKee, M. C., Nanci, A. and Khan S. R.: Ultrastructural immunodetection of osteopontin and osteocalcin as major matrix components of renal calculi. J Bone Mineral Res, 10: 1913– 1929, 1995 38. Gokhale, J. A., McKee, M. C. and Khan, S. R.: Immunocytochemical localization of Tamm-Horsfall protein in the kidneys of normal and nephrolithic rats. Urol Res, 24: 201, 1996 39. Kok, D. J.: Crystallization and stone formation inside the nephron. Scann Microsc, 10: 471, 1996 40. Martin, X., Opgenroth, T. J., Werness, P. G., Rundquist, R. T., Romero, J. C. and Smith, L. H.: The contribution of the bladder to calcium oxalate crystal growth inhibition in voided urine. In: Urolithiasis and Related Clinical Research. Edited by P. O. Schwille, L. H. Smith, W. G. Robertson and W. Vahlensieck. New York: Plenum Press, pp 871– 873, 1985
Vol. 164, 230 –236, July 2000 Printed in U.S.A.
CHANGES IN URINE MACROMOLECULAR COMPOSITION DURING PROCESSING SUZANNE MASLAMANI, PATRICIA A. GLENTON
AND
SAEED R. KHAN*
From the Department of Pathology, College of Medicine, University of Florida, Gainesville, Florida
ABSTRACT
Purpose: To determine the urinary crystallization inhibitory activity, urine is generally centrifuged and/or filtered. These preparative procedures may result in a total or partial removal of many macromolecular constituents implicated in crystallization. The main purpose of this study was to investigate the changes in urinary macromolecular composition following centrifugation and filtration. Materials and Methods: Twenty-four hour urine samples were collected from human volunteers. Each was divided into 4 aliquots; one was filtered, the other was centrifuged, another was centrifuged and filtered. The control sample was neither filtered nor centrifuged. Total protein and lipid contents of each sample were determined. Proteins were analyzed using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). Western blot analysis was performed using antibodies against osteopontin (OPN), prothrombin (PT) related proteins, inter-␣-inhibitor (I␣I) related proteins, Tamm-Horsfall protein (THP), and albumin (ALB). The effect of processing on incorporation of urinary proteins in crystal matrices was also examined. Calcium oxalate crystals were produced in processed and unprocessed urine samples by the addition of sodium oxalate. Crystals were harvested, de-mineralized and their proteins analyzed by SDS-PAGE and Western blotting. Results: Processing reduced the amounts of both proteins and lipids in the urine. Previously we identified phospholipids in the matrix of calcium oxalate crystals as well as the filtrate and retentate removed during filtration and centrifugation. Phospholipids have a high affinity for calcium-containing crystals. In the case of proteins, those with high molecular weights appeared to be clearly affected by filtration and centrifugation. Processing also appeared to influence the incorporation of proteins in the crystals. The matrix of crystals produced in processed urine contained less THP than those made in unprocessed urine, apparently a result of the loss of this higher molecular weight protein during processing. Incorporation of PT-related proteins, particularly fragment 1, was increased. Conclusions: We propose that selective inclusion of macromolecules is a result of an increase in available binding sites on crystal surfaces because of the removal of certain calcium binding substances such as phospholipids and proteins. Removal of larger macromolecules from the milieu may also provide a better access to the crystal surfaces. KEY WORDS: calcium oxalate, urolithiasis, osteopontin, bikunin, Tamm-Horsfall protein, prothrombin
Two forces mainly control crystallization in human urine: urinary supersaturation with respect to the stone salts and presence of crystallization modulators.1– 4 Most macromolecular modulators are either glycoproteins or glycosaminoglycans and can promote or inhibit crystal nucleation, growth and/or aggregation. To define the role of modulators in the formation of crystals and stones it is customary to investigate crystallization in vitro in the presence or absence of individual macromolecules, employing either aqueous inorganic solutions or the urine itself.3–5 Since inorganic solutions do not replicate the complex urinary environment, use of urine as the medium is generally encouraged. However, use of urine requires many processing steps such as filtration, centrifugation and dilution,3, 5–9 which can drastically change the milieu10 and affect crystallization. This study was undertaken to investigate the effect of urine processing on its
protein contents with particular reference to osteopontin (OPN), prothrombin (PT) related proteins, inter-␣-inhibitor (I␣I) related proteins, Tamm-Horsfall protein (THP), and albumin (ALB). Twenty-four hour urine specimens were collected from male and female human volunteers. Each specimen was divided into 4 samples: one remained unprocessed, the second was filtered, the third was centrifuged and the fourth was centrifuged and filtered. Total protein content of each sample were determined using a biochemical assay. Proteins were analyzed using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and Western blot analysis. The effect of processing on incorporation of urinary proteins in crystal matrices was also examined. Calcium oxalate crystals were produced in processed and unprocessed urine samples by addition of sodium oxalate. Crystals were harvested, demineralized and their proteins analyzed by SDS-PAGE and Western blotting.10
Accepted for publication January 19, 2000. * Requests for reprints: Department of Pathology, PO Box 100275, College of Medicine, University of Florida, Gainesville, Florida 32610. Supported by NIH grant #RO1 DK41434. MATERIALS AND METHODS This work was part of a thesis presented to the graduate school of Urine collection and processing. 24-hour urine samples the University of Florida, for the partial fulfillment of the requirewere collected from five males and four females. During ments for the degree of Master of Science. 230
231
URINE MACROMOLECULAR COMPOSITION
collection, the specimens were maintained at room temperature: approximately 24C. One ml. of 20% sodium azide (Fisher Scientific, Norcross, GA), an anti-bacterial agent, was placed in collection bottles prior to collection. The volume and pH of each specimen were measured and 250 ml. aliquots of urine were placed in Nalgene bottles marked “NT” (no processing), “F” (filtered), “C” (centrifuged), and “C⫹F” (centrifuged and filtered). Specimens were filtered using 0.22 m. sodium acetate filters (Fisher Scientific, Norcross, GA) and centrifuged at 10,000 ⫻g at 20C for 20 minutes using the J-14 rotor of the J2 to 21 centrifuge (Beckman Instruments, Westbury, NY). Ten ml. of urine from each sample was concentrated to 1 ml. using speed vacuum model RC 10.10 (Jouan Inc., Winchester, VA) and then lyophilized (Flexi-Dry lyophilizer from FTS Systems, Stone Ridge, NY). 500 L. of 0.05 M Tris buffer pH 7.4 was added to each tube. These specimens were kept in a freezer at ⫺20C until the SDS-PAGE and western blotting procedures were performed. Preparation of crystal matrix. The urine samples were allowed to warm to 37C in a shaking water bath (Fisher Scientific, Norcross, GA) and divided in equal halves. Calcium oxalate crystals were induced by the addition of 15 ml./l. of 0.1 M sodium oxalate and incubated for 3 hours. At the end of incubation period, the urine specimens were centrifuged at 10,000 ⫻g for 25 minutes at 20C in the J2 to 21 centrifuge (Beckman Instruments, Westbury, NY). The supernatant was aspirated and the pellet placed in a microcentrifuge tube and washed three times with 2 ml. of distilled water. The contents of the tubes were dried for 24 hours in the Flexi-Dry lyophilizer. The crystals were demineralized by treatment with 5 ml. of 0.25 M ethylene diamine tetraacetic acid (EDTA) pH 8.0 at 4C for 3 days with continuous stirring. The extract was centrifuged at 10,000 ⫻g for 5 minutes. Supernatant was dialyzed against water for 24 hours at 4C using dialysis tubing with a 6 to 8 kDa cut-off (Spectrum Medical Industries, CA). Protein assay. Protein concentration in the urine and crystal matrix extract was determined using Bio-Rad’s Dye reagents (Bio-Rad Laboratories, Hercules, CA). The protein standards (Sigma-Aldrich; St. Louis, MO) were prepared using bovine serum albumin to provide a linear range of 0 g./ml. to 10 g./ml. The assay was performed in duplicate. Urine samples were diluted 1:2 with water prior to analysis. Fifty L. of each standard were mixed with 2.5 ml. of a dilute 1:4 mixture of Bio-Rad’s Dye Reagent Concentrate. The tubes were vortexed and allowed to incubate for 5 minutes. Absorbance was measured at 595 nm using a Pharmacia LKB Ultrospec 3 UV Visible Light (Hoffer Pharmacia Biotech, Piscataway, NJ). Lipid assay. To demonstrate the loss of lipids by filtration and centrifugation we isolated lipids from the urine. The methods are described in detail in earlier publications.11 In brief, 400 ml. of processed or unprocessed (native) urine was mixed with 1.2 l. of 2:1 chloroform: methanol. The mixture was shaken and placed on an end-over-end mixer for 24 hours at 4C. Samples were then centrifuged at 7000 r.p.m. to achieve phase separation. The top portion was removed and combined as the aqueous layer; the middle layer was recovered as the interface; and the lower phase was collected and pooled as the organic layer. After evaporation to a smaller volume, the organic sample was Folch-washed twice, pooling the respective phases with the previous ones. The organic phase was then lyophilized and weighed. SDS-PAGE analysis. Separation of the proteins was accomplished using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). The samples were prepared by adding 6 L. of sample buffer (2.5 ml. 0.5 M Tris base pH 6.8 ⫹ 4 ml. 10% SDS ⫹ 2 ml. glycerol ⫹ 0.002 bromophenol blue ⫹ 0.31 gm. dithiothreitol ⫹ distilled water to a volume of 10 ml.) to 6 L. of each sample from the whole urine and the crystal matrix extract specimens. Ten L. of this mixture
was applied to a designated lane on a 10 to 20% SDS-PAGE. The gels were electrophoresed at 70 V for approximately 2 to 2.5 hours with standard Gel Running Buffer (Tris base 3 gm./l., glycine 14 gm./l., SDS 1 gm./l.). The gels were stained with Coomassie blue to detect protein bands. Western blotting. Individual proteins were identified by Western blot analysis using specific antibodies. After the SDS-PAGE electrophoresis, the gel was removed and placed in a small amount of Blotting Buffer pH 8.3 to equilibrate for 15 minutes. Blotting Buffer was made with 3.03 gm. Tris (Fisher Biotech BP152 to 1) and 14.4 gm. glycine (Fisher Biotech BP381–1) in 500 ml. distilled water, 200 ml. methanol (Fisher Biotech A452–1), and 300 ml. distilled water (no pH adjustment was performed). The proteins were transferred to a nitrocellulose membrane (BioBlot-NC Cat. #8801, Cambridge, MA) using the Bio-Rad Power Pac 200 set at 100V for 15 minutes, then 50V for 45 minutes. The cassette positions within the electrode holder were exchanged with each other and then the blotting continued for 15 minutes at 100V and 50V for the remaining 45 minutes. When blotting was complete, the nitrocellulose membranes were removed and placed in weigh boats with 10 ml. Blocking Buffer for at least 30 minutes. Blocking Buffer was made with 1 gm. powdered milk and 20 ml. Tris buffered saline with Tween (TBST). Tris buffered saline with Tween was made with 1.546 gm. Tris HCl, 2.922 gm. NaCl, 0.5 ml. Tween-20 (Bio-Rad Cat. # 170-6531). The pH was adjusted to 8.0 using HCl or NaOH. When blocking was complete, the primary antibody for the particular protein to be examined was added to the membrane at a starting dilution of 1:1000. Immunochemical staining was performed at room temperature by using different polyclonal antibodies. Polyclonal antibodies used at 1:1000 dilution were as follows: THP, OPN, ALB, prothrombin, and I␣I. THP antibody was made in our laboratory by injecting rat THP into rabbit as antigen (Kel Farm Laboratory, FL) and OPN antibody was a gift from Dr. E. M. Worcester (Department of Medicine, Medical College of Wisconsin). PT antibody (ICS Biomedical Inc., CA) reacts with prothrombin fragment-1, fragment 1⫹2 as well as prothrombin. I␣I antibody (Accurate Chemical Scientific Corporation, NY) reacts with all three chains of I␣I, heavy chains H1, H2, and the light chain, bikunin. Then the membranes were incubated with the secondary antibodies marked with alkaline phosphates (Hyclone Laboratories. UT) and used at a dilution of 1:5000. Finally, the membranes were allowed to develop for 5 minutes by using a substrate kit (Bio Rad Laboratories, CA.). Quantification and statistical analysis. The gels were read using a Pharmacia LKB Imagemaster DTS densitometer (Piscataway, NJ). The optical density (OD) of each band was computed and then the sum of the optical densities in each lane (representing a particular preparative treatment) was determined. The calculated optical densities in each treatment lane were multiplied by the 24-hour urine output of that particular individual to get an OD/24 hours value for each treatment. The OD for each individual’s Filtered, Centrifuged, and Centrifuged ⫹ Filtered group was compared with their own No Treatment OD/24 hours calculation. A one-tailed paired t test with a significance of p ⬍0.05 was used to analyze the data. RESULTS
As shown in table 1, filtration and centrifugation had significant effect on urinary concentrations of proteins and lipids. Both filtration and centrifugation resulted in a reduction in the protein and lipid contents of the urine, filtration being more effective in removing them from the urine than centrifugation. Processing of the urine also had a significant effect on its metastable limit with respect to crystallization of cal-
232
URINE MACROMOLECULAR COMPOSITION
TABLE 1. Effect of various preparative procedures on urinary concentration of proteins, lipids (per unit volume of urine) and metastable limit of the urine with respect to crystallization of CaOx No Treatment
Filtration
Centrifugation
0.67 ⫾ 0.34a Metastable Limit (mM/l. of oxalate added) 0.58 ⫾ 0.33 0.67 ⫾ 0.34a Protein (Bradford assay g./ml., n ⫽ 9) 515.8 ⫾ 128.4 436.4 ⫾ 100.1a 456.4 ⫾ 115.5a Protein (SDS-PAGE, OD/24 hour, n ⫽ 4) 1988.0 ⫾ 585.8 1323.3 ⫾ 455.4a 1986.5 ⫾ 1069.2b Total Lipids (mg./l., n ⫽ 5) 296.6 ⫾ 50.8 203.2 ⫾ 47.6a 249.8 ⫾ 81.4 (a Significantly different from unprocessed, b significantly different from filtered, c significantly different from centrifuged).
cium oxalate. An ultrastructural examination of the pellet following centrifugation and retentate after filtration (fig. 1) showed them to contain amorphous material, fibrous substances and membranous fragments and vesicles. SDS-PAGE analysis of urine revealed the presence of 10 to 15 bands with approximate molecular weights ranging between 17 kDa to 150 kDa. Some samples showed bands even at 220 kDa. Major bands were recognized at approximate molecular weights of 108, 80, 67, 58, 48, 40, 35, 28, and 22 kDa (fig. 2, A). Western blot analysis using specific antibodies recognized osteopontin (not illustrated) at approximately 22, and 28 kDa, both bands being diffuse and non-descript. A band at 80 was also positive for OPN in some samples. Five bands at approximately 85, 70, 49, 40 and 28 kDa were recognized as albumin with 70, 49 and 40 kDa bands being most noticeable and common in all samples (fig. 3, A). Bands at approximate molecular weights of 80, 48, and 35 were positive for prothrombin related proteins (3, B). These bands had similar appearance and were detected in almost all samples. Three bands at 107, 68 and 50 kDa were positive for
FIG. 1. Ultrastructure of retentate on filter illustrating fibrous, amorphous and membranous substances.
Centrifugation and Filtration 0.68 ⫾ 0.38a 385.8 ⫾ 72.8abc 1465.7 ⫾ 777.8ac 202.5 ⫾ 45.6ac
FIG. 2. SDS-PAGE analysis. Lane 1: Molecular weight markers, Lane 2: Unprocessed urine, Lane 3: Filtered urine, Lane 4: Centrifuged urine, Lane 5: Centrifuged and filtered urine. A, urine. Comparison of lane 2 with 3 and 5 indicates absence of high molecular band in filtered urine. Centrifugation (Compare Lane 4 with 2 and 3) resulted in weakening of high molecular weight band. B, crystal matrix. Comparison of lane 2 with the others indicates that crystals made in processed urine have lesser number of bands. Bands also appear lighter except for band around 35 kDa, which is more intense in crystals made in processed urine.
Tamm-Horsfall protein, 107 and 68 kDa bands being common and most conspicuous (3C). I␣I was recognized at approximate molecular weights of 123, 74, 68, 62, 48, 40 and 30 kDa. Bands at 48 and 40 kDa were prominent and common to all samples (3D). Urinary protein profile was negatively impacted by various preparative procedures (fig. 2, A). A comparison of bands in Lane 2 with those in Lanes 3, 4 and 5 illustrates that high molecular weight proteins were at least partially removed by the preparative techniques. The bands representing these protein fractions were either totally absent or lighter in appearance in filtered urine (Lane 3) than in the unprocessed urine (Lane 2) samples. Centrifugation (Lane 4) also resulted in removal of proteins but to a lesser degree. The effect of combined treatment of centrifugation and filtration (Lane 5) appeared to be similar to filtration. A comparison of band densities of the Western blots of urinary proteins (table 2) also showed that centrifugation and filtration both resulted in reduction of albumin, THP and I␣I. SDS-PAGE analysis of the matrix proteins isolated from calcium oxalate crystals is illustrated in fig. 2, B. Seven distinct bands from estimated molecular weights of 28-kDa to120 kDa were recognized in proteins isolated from the crystals induced in unprocessed urine. Crystals produced in processed urine showed fewer and weaker bands. However a band at around 35 kDa was much more pronounced and dense in crystals produced in processed urine particularly those produced in filtered urine. Western blot analysis (fig. 4) identified OPN, ALB, THP, PT- and I␣I - related proteins in the organic matrix of calcium oxalate crystals produced in both the processed and unprocessed urine samples. A single band at 28 kDa was recognized, as OPN (not illustrated). Albumin was detected at 70 and 48 kDa (fig. 4, A). Up-to three bands stained positive for PT -related proteins. Every sample showed a dense band near 35 kDa (fig. 4, B). Interestingly, the crystals
233
URINE MACROMOLECULAR COMPOSITION
FIG. 3. Western blots of urine. Lanes are same as in fig. 2. A, albumin. All lanes have approximately same number of bands with similar intensities. B, prothrombin. All lanes have 3 major bands. C, Tamm-Horsfall Protein. Band intensity is lower in filtered urine (Compare Lane 2 with 3 and 5). D, inter-␣-inhibitor. Number of bands and their intensity appear the same.
made in processed urine showed more and denser PT-positive bands than the crystals made in unprocessed urine. Crystals made in unprocessed urine showed many THP positive bands (fig. 4, C), the band at 107 kDa being the most prominent. Crystals from processed urine also showed this band but they were very weak. All samples showed a I␣I - related protein positive band at 68 kDa (fig. 4, D). A few samples showed another band at 48 kDa and a faint band was sometime seen at 40 kDa. DISCUSSION
Commonly, urine is filtered and/or centrifuged prior to testing crystallization metastable limits or investigating the activity of individual crystallization inhibitors.5–9 Our preliminary studies indicated the possibility that filtration and centrifugation may remove some of the urinary modulators of crystallization.10 To further explore this issue we studied specific urinary crystallization modulators before and after filtration and centrifugation. Biochemical quantification showed that filtration results in significant reductions in the amounts of both the proteins and lipids in the urine. Centrifugation also results in reduction of these macromolecules but to a lesser extent. Ultrastructural examination showed that both the retentate following filtration and pellet following centrifugation contain fibrous material and membrane bound vesicles and fragments. Apparently fibrous material represents the separated proteins while membranous vesi-
FIG. 4. Western blots of crystal matrix. Lanes are same as in fig. 2. A, albumin. Two bands are identifiable. Band intensity is higher in matrix of crystals made in unprocessed urine. B, prothrombin. Crystals made in unprocessed urine have lighter bands in matrix while those made in processed urine show very intense bands particularly at 35 kDa. C, Tamm-Horsfall Protein. More bands with higher density are seen in matrix of crystals made in unprocessed urine (Lane 2) than those made in processed urine (Lanes 3 to 5). In addition they contain high molecular bands. D, inter-␣-inhibitor. At least two bands are identifiable in each sample.
cles are the source of extracted lipids. Removal of these macromolecules could have significant effect on urine’s crystallization potential since most urinary proteins are recognized as inhibitors of various crystallization processes,3, 4 while membranes and their lipids are shown to promote crystal nucleation.11, 12 Changes in urine’s metastable limit following filtration and centrifugation support this assertion. SDS-PAGE and Western blot analysis were employed to identify the proteins that were lost or had their concentration reduced during preparative techniques. We particularly examined the effect of processing on five proteins, THP, OPN, ALB, PT related proteins and I␣I related proteins, which have been shown to play significant role in calcium oxalate crystallization and nephrolithiasis. Salient features of these proteins are described in table 3. Coomassie revealed filtration resulted in a total loss of some bands representing high molecular weight proteins. Some other bands appeared weaker in processed urine. Western blot analysis using specific antibodies showed that processing had no apparent qualitative effect on the five bands that stained positive for albumin. But an analysis of the optical density of protein bands on western blot revealed significant reduction in urinary albumin concentration following filtration and centrifugation. Prothrombin antibody stained 3 main bands representing prothrombin at approximately 80 kDa, Fragment
TABLE 2. Effect of various preparative procedures on selected proteins No Treatment Albumin (n ⫽ 4) 5921 ⫾ 1982 Tamm-Horsfall Protein (n ⫽ 4) 5150 ⫾ 3267 Inter-␣-Inhibitor (n ⫽ 5) 5142 ⫾ 4068 Result of Western Blot Analysis (Optical Density/24 hour of urine, cantly different from centrifuged).
Filtration
Centrifugation
Centrifugation and Filtration
5903 ⫾ 2896 5119 ⫾ 1140 3922 ⫾ 1892a 2449 ⫾ 1896 4587 ⫾ 3672b 2780 ⫾ 1966a, b 2057 ⫾ 952 4203 ⫾ 3762 2717 ⫾ 2017 a significantly different from unprocessed, b significantly different from filtered, c signifi-
234
URINE MACROMOLECULAR COMPOSITION TABLE 3. Urinary proteins with potential to modulate crystallization Protein Name
Tamm-Horsfall Protein4, 6, 13, 15, 16 Osteopontin4, 6, 14–17 Human Serum Albumin4, 6, 18 Urinary Prothrombin Fragment 14, 6, 19–23 Inter-␣-Inhibitor4, 6, 24, 25 H1 (Heavy Chain 1) H2 (Heavy Chain 2) HI-30 (Bikunin)26–30
MW (kDa)
Origin
Unique Features
Role in Crystallization
Presence in Urine
Presence in Stone
80–100
TAL of the Kidney
12% ASP
Promoter, Inhibitor of Aggregation
Yes 20–100 mg/day
Yes
42–80
TDL, TAL of the Kidney Plasma
RGD Sequence
Yes 2.4–3.7 mg/day
Yes
Yes 1.6–34.2 mg/day
Yes
Plasma ?, TAL of the Kidney Plasma?
10 GLA Residues
Inhibitor of Nucleation, Growth, Aggregation Facilitates Binding of Other Proteins to Crystals Inhibitor of Crystal Growth and Aggregation
Yes 13.4 nM/day
Yes
Yes 2–10 mg/day Yes Yes 5.01 g/ml
Yes Yes Yes
68 31–35 85 71 30–35
Binds to Crystals
Sulfated GAGs
Plasma ?, Kidney
Inhibitor of Crystal Nucleation, Growth, and Aggregation Location In Kidneys Based On Studies In Rats; Urinary Excretion Rate In Normal Humans; Role In Crystallization Based Mostly On CaOx Crystal Formation; ASP, Aspartic acid; GLA, g-Carboxyglutamic acid; RGD, Arginine-Glycine-Aspartic Acid; PT, Proximal Tubule; TAL, Thin Ascending Limb Of The Loop of Henle; TDL, Thick Descending Limb Of The Loop of Henle.
F1⫹2 at 48 kDa and Fragment F1 at 35 kDa. None appeared affected by filtration or centrifugation. THP was recognized as 3 bands at 107 kDa, 68 kDa and 50 kDa, with 107 kDa band being the most prominent. The fact that MWR of THP is slightly different from expected single band between 80 to 100 and that the 3 bands stained positive, can be due to the buffer or the SDS-PAGE system as well as the possibility of enzymatic degradation of protein in the urine. However the most important factor is that this detection system is an antibody which is highly specific for THP.15 Comparison of band intensity between unprocessed and processed urine showed that the 107 kDa band was definitely affected by the filtration. Other THP positive bands also appeared lighter. The I␣I family of proteins consists of I␣I (⬎172 kDa), pre-␣inhibitor (115 kDa), H1 (86 kDa), H2 (71 kDa), H3 (65 kDa) and bikunin (38 kDa). The I␣I antibody we used can interact with all of them. All urinary samples contained H1, H2, H3 and bikunin with bikunin being the most common. Some sample also contained pre-␣I. Filtration or centrifugation had no effect on I␣I-positive bands ranging between 74 kDa to 28 kDa. However the band representing higher molecular weight pre-␣I disappeared after processing. We have also shown a significant reduction in various urinary lipids by filtration or centrifugation of the urine.31 Apparently, these procedures remove many urinary substances including various lipids, albumin, THP, and high molecular weight members of the I␣I family. A number of other studies have also shown that centrifugation and filtration of urine remove THP and albumin.32–33 Analysis of crystal matrix provided the evidence that urine processing also affects the incorporation of proteins in crystals. The matrix of crystals produced in unprocessed urine contained many more proteins than those produced in processed urine. High molecular weight proteins were lost, as suggested by the absence of high molecular weight bands. The concentration of a few other proteins appeared lower in the matrix of crystals made in processed urine, as indicated by weaker bands. Western blot analysis showed a substantial reduction in THP inclusion because of its obvious loss during filtration and centrifugation. The amount of albumin included also appeared to be reduced perhaps because of its removal during processing. There was a distinct increase in the incorporation of PT-related proteins, particularly prothrombin fragment-1, when crystals were produced in processed urine. A 68 kDa I␣I-positive band also showed increased intensity in the matrix of these crystals. It has been suggested that incorporation of organic material in the crystal matrix is a highly selective process.34 –36 Proteins such as albumin and THP are excluded while prothrombin fragment-1 is selectively included. One report concluded that “centrifugation and filtration of the urine before crystallization had little effect on the qualitative protein
contents of the crystal extracts” and that “Tamm-Horsfall glycoprotein was notably absent from the crystal extracts (matrix)”.34 Interestingly, electrophoretograms presented in this study clearly showed that processing of urine resulted in a loss of many proteins including albumin and THP, and that both THP and albumin were present in the matrix of crystals produced in whole urine. Only the matrix of crystals made in the ultra-filtered urine showed a single band, which at the time was designated crystal matrix protein and later identified as urinary prothrombin-fragment-1.20 –22 Our results presented here and elsewhere10 clearly demonstrate the incorporation of both THP and albumin in the crystal matrix when experiments were carried out in unprocessed urine. We have previously described the presence of lipids in the organic matrix of both calcium oxalate and calcium phosphate crystals produced in unprocessed urine in vitro. In addition almost all the urinary phospholipids were assimilated in the crystal matrix whereas only a fraction of urinary neutral and glycolipids became incorporated.12 The presence of various phospholipids, OPN, THP and ALB, in crystal matrices should not be surprising since all of them can interact with crystal surfaces, have been shown to be involved in calcium oxalate crystallization as modulators of nucleation, growth or aggregation, and are important constituents of stone matrix (table 3). Ultrastructural studies of calcium oxalate human urinary stones37 as well as deposits in rat kidneys38 have shown both OPN and THP to be intimately associated with crystals. OPN becomes involved very early in crystal formation and is even located inside the crystals. THP appears to promote aggregation by binding to the surfaces of crystals and connecting them together. Obviously the traditional processing of urine involving centrifugation and filtration removes some macromolecules, resulting in their decrease or absence from the organic matrices of crystals made in these urines. Removal of certain macromolecules through filtering or centrifugation would increase the likelihood of interaction between growing crystals and the residual macromolecules, thereby increasing the probability of their binding to crystal surfaces and eventual incorporation in crystal matrix. Substances eliminated during processing may share binding sites on crystal surfaces with those that are left behind. For example many lipids and proteins can bind to crystals, particularly calcium-containing crystals. They may also hinder the movement of other macromolecules to the vicinity of crystal surfaces. Thus in the experiments where urine was processed before crystal induction, the selective inclusion of some macromolecules may actually be the result of exclusion of others. Activity of crystallization modulators is generally measured in simple inorganic solutions and on rare occasions in undiluted (whole) or diluted urine.2, 5, 6 It is justifiably argued that the inhibitory potential of a macromolecule should
URINE MACROMOLECULAR COMPOSITION
be tested in whole urine since it is the only medium representative of the physiological state in which crystals and stones form. We must, however, recognize that kidney stones form in urine when it is still in the renal tubules and has a different constitution than the excreted bladder urine.39, 40 Even the “whole” urine is almost always processed, which as we have shown here, results in the removal of many substances with a direct role in crystallization. Removal of these modulators may erroneously exaggerate the role of the remaining macromolecules. In the case of diluted urine, activity of the inhibitors may change because of their lowered concentrations.2 Apparently every methodology used has its limitations, and results obtained using any methods need to be carefully evaluated with respect to their advantages and disadvantages. The main purpose of our investigation was to determine the effect of preparative procedures on the protein contents of urine. Both filtration and centrifugation remove proteins and lipids from the urine. The removal may influence crystallization activity in the urine because both proteins and lipids are known to modulate crystallization of calcium phosphate as well as calcium oxalate. Such a removal may also be responsible for the reported exclusion of proteins such as THP and albumin and the disproportionate inclusion of others, such as prothrombin fragment-1, in the matrix of crystals produced in vitro in processed urine. We thank Professor Raymond L. Hackett for reviewing the manuscript and Mrs. Patricia Khan for reviewing the manuscript and preparation of figures.
REFERENCES
1. Finlayson, B.: Physicochemical aspects of urolithiasis. Kidney Int, 13: 344, 1978 2. Hess, B. and Kok, D. J.: Nucleation, growth, and aggregation of stone forming crystals. In: Kidney Stones: Medical and Surgical Management. Edited by F. L. Coe, M. J. Favus, C. Y. C. Pak, J. H. Parks and G. M. Preminger. Philadelphia: Lippincottt-Raven Publishers, pp 3–32, 1996 3. Ryall, R. L. and Stapleton, A. M. F.: Urinary macromolecules in calcium oxalate stone and crystal matrix: good, bad, or indifferent. In: Calcium Oxalate in Biological Systems. Edited by S. R. Khan. Boca Raton: CRC Press, pp 265–290, 1995 4. Khan, S. R.: Interactions between stone-forming calcific crystals and macromolecules. Urol Int, 59: 59, 1997 5. Kavanagh, J. P.: Calcium oxalate crystallization in vitro. In: Calcium Oxalate in Biological Systems. Edited by S. R. Khan. Boca Raton: CRC Press, pp 1–21, 1995 6. Ryall, R. L., Hibberd, C. M. and Marshall, V. R.: A method for studying inhibitory activity in whole urine. Urol Res, 13: 285, 1985 7. Asplin, J. R., Bushinsky, D. A., Singharetnam, W., Parks, J. H. and Coe, F. L.: Relationship between supersaturation and crystal inhibition in hypercalciuric rats. Kidney Int, 51: 640, 1997 8. Baumann, J. M. and Wacker, M.: The direct measurement to inhibitory capacity to crystal growth of calcium oxalate in undiluted urine and other inhibitor containing solutions. Urol Res, 8: 171, 1980 9. Pak, C. Y. C. and Holt, K.: Nucleation and growth of brushite and calcium oxalate in urine of stone formers. Metabolism, 25: 665, 1976 10. Atmani, F., Glenton, P. A. and Khan, S. R.: Identification of proteins extracted from calcium oxalate and calcium phosphate crystals induced in the urine of healthy and stone forming subjects. Urol Res, 26: 1, 1998 11. Khan, S. R.: Heterogeneous nucleation of calcium oxalate crystals in mammalian urine. Scan Microsc, 9: 597, 1995 12. Khan, S. R., Atmani, F., Glenton, P. A., Hou, Z-C., Talham, D. R. and Khurshid, M.: Lipids and membranes in the organic matrix of urinary calcific crystals and stones. Calcif Tissue Int, 59: 357, 1996 13. Hess, B.: Tamm-Horsfall glycoprotein and calcium nephrolithiasis. Min Electr Metab, 20: 393, 1994
235
14. Hoyer, J.: Uropontin in urinary calcium stone formation. Min Electr Metab, 20: 385, 1994 15. Gokhale, J. A., Glenton, P. A. and Khan, S. R.: Biochemical and quantitative analysis of Tamm Horsfall protein in rats. Urol Res, 25: 347, 1997 16. Gokhale, J. A., Glenton, P. A. and Khan, S. R.: Localization of Tamm Horsfall protein and osteopontin in a rat nephrolithiasis model. Nephron, 73: 456, 1996 17. Shiraga, H., Min, W., Vandusen, W., Clayman, M. D., Miner, D., Terrell, C. H., Sherbotie, J. R., Foreman, J. W., Przysiecki, C., Neilson, E. G. and Hoyer, J. R.: Inhibition of calcium oxalate crystal growth in vitro by uropontin: another member of the aspartic acid-rich protein superfamily. Proc Natl Acad Sci USA, 89: 426, 1992 18. Dussol, B., Geider, S., Lilova, A., Le´onetti, F., Dupuy, P., Daudon, M., Berlan, Y., Dagorn, J. and Verdier, J.: Analysis of the soluble organic matrix of five morphologically different kidney stones. Evidence for a specific role of albumin in the constitution of the stone protein matrix. Urol Res, 23: 45, 1995 19. Doyle, I. R., Marshall, V. R., Dawson, C. J. and Ryall, R. L.: Calcium oxalate crystal matrix extract: the most potent macromolecular inhibitor of crystal growth and aggregation yet tested in human undiluted human urine in vitro. Urol Res, 23: 53, 1995 20. Stapleton, A. M. F. and Ryall, R. L.: Blood coagulation proteins and urolithiasis are linked: crystal matrix protein is the F1 activation peptide of human prothrombin. Br J Urol, 75: 712, 1995 21. Suzuki, K., Moriyama, M., Nakajima, C., Kawamura, K., Miyazawa, K., Tsugawa, R., Kikuchi, N. and Nagata, K.: Isolation and partial characterization of crystal matrix protein as a potent inhibitor of calcium oxalate crystal aggregation: evidence of activation peptide of human prothrombin. Urol Res, 22: 45, 1994 22. Stapleton, A. M. F., Dawson, C. J., Grover, P. K., Hohmann, A., Comacchio,R., Boswarva, V., Tang, Y. and Ryall, R. L.: Further evidence linking urolithiasis and blood coagulation: urinary prothrombin fragment-1 is present in stone matrix. Kidney Int, 49: 880, 1996 23. Stapleton, A. M. F., Seymour, A. E., Brennan, J. S., Doyle, I. R., Marshall, V. R. and Ryall, R. L.: Immuno-histochemical distribution and quantification of crystal matrix protein. Kidney Int, 44: 817, 1993 24. Enghild, J. J., Thogersen, I. B., Pizzo, S. V. and Salvesen, G.: Analysis of inter-␣-trypsin inhibitor and a novel trypsin inhibitor, pre-␣-trypsin inhibitor, from human plasma. Polypeptide chain stoichiometry and assembly by glycan. J Biol Chem, 264: 15975,1989 25. Salier, J. P., Rouet, P., Raguenez, G. and Daveau, M.: The inter-␣-inhibitor family: from structure to regulation. Biocehem J, 315: 1, 1997 26. Atmani, F., Mizon, J. and Khan, S. R.: Identification of uronic acid rich protein as urinary bikunin, the light chain of inter␣-inhibitor. Eur J Biochem, 236: 984, 1996 27. Marengo, S. R., Resnick, M. I, Yang, L. and Chung, J-Y.: Differential expression of urinary inter-␣-trypsin inhibitor trimers and dimers in normal compared to active calcium oxalate stone forming men. J Urol, 159: 1444, 1998 28. Iida, S., Peck, A. B., Johnson-Tardieu, J., Moriyama, M., Glenton, P. A., Byer, K. J. and Khan, S. R.: Temporal changes in mRNA expression for bikunin in the kidneys of rats during calcium oxalate nephrolithiasis. J Am Soc Nephrol, 10: 986, 1999 29. Medetognon-Benissan, J., Tardivel, S., Hennequin, C., Daudon, M., Drueke, T. and Lacour, B.: Inhibitory effect of bikunin on calcium oxalate crystallization in vitro and urinary bikunin decrease in renal stone formers. Urol Res, 27: 69, 1999 30. Atmani, F. and Khan, S. R.: Role of bikunin in the inhibition of calcium oxalate crystallization. J Am Soc Nephrol, 10: S385, 1999 31. Khan, S. R., Maslamani, S. A., Atmani, F., Glenton, P. A., Opalko, F. J., Thamilselvan, S. and Hammett-Stabler, C.: Membranes and their constituents as promoters of calcium oxalate crystal formation in human urine. Calcif Tissue Int, 66: 90, 2000 32. Wiggins, R. C.: Uromucoid (Tamm-Horsfall glycoprotein) forms different polymeric arrangements on a filter surface under different physicochemical conditions. Clin Chim Acta, 162: 329, 1987
236
URINE MACROMOLECULAR COMPOSITION
33. Doyle, I. R., Ryall, R. L. and Marshall, V. R.: The effect of low speed centrifugation and millipore filtration on the urinary protein contents. In: Urolithiasis. Edited by V. R. Walker, R. A. L. Sutton, E. C. B. Cameron, C. Y. C. Pak and W. G. Robertson. New York: Plenum Press, pp 593–594, 1989 34. Doyle, I. R., Ryall, R. L. and Marshall, V. R.: Inclusion of proteins into calcium oxalate crystals precipitated from human urine: a highly selective phenomenon. Clin Chem, 37: 1589, 1991 35. Morse, R. M. and Resnick, M. I.: A new approach to the study of urinary macromolecules as a participant in calcium oxalate crystallization. J Urol, 139: 869, 1988 36. Morse, R. M. and Resnick, M. I.: A study of the incorporation of urinary macromolecules onto crystals of different mineral compositions. J Urol, 141: 641, 1989
37. McKee, M. C., Nanci, A. and Khan S. R.: Ultrastructural immunodetection of osteopontin and osteocalcin as major matrix components of renal calculi. J Bone Mineral Res, 10: 1913– 1929, 1995 38. Gokhale, J. A., McKee, M. C. and Khan, S. R.: Immunocytochemical localization of Tamm-Horsfall protein in the kidneys of normal and nephrolithic rats. Urol Res, 24: 201, 1996 39. Kok, D. J.: Crystallization and stone formation inside the nephron. Scann Microsc, 10: 471, 1996 40. Martin, X., Opgenroth, T. J., Werness, P. G., Rundquist, R. T., Romero, J. C. and Smith, L. H.: The contribution of the bladder to calcium oxalate crystal growth inhibition in voided urine. In: Urolithiasis and Related Clinical Research. Edited by P. O. Schwille, L. H. Smith, W. G. Robertson and W. Vahlensieck. New York: Plenum Press, pp 871– 873, 1985