Calcium Oxalate Crystal Adherence to Hyaluronan-, Osteopontin-, and CD44-Expressing Injured/Regenerating Tubular Epithelial Cells in Rat Kidneys
Marino Asselman*,
Anja Verhulst,
Marc E. de Broe and
Carl F. Verkoelen*
*Department of Urology, Erasmus Medical Center Rotterdam, Rotterdam, the Netherlands; and Department of Nephrology-Hypertension, University of Antwerp, Antwerp, Belgium
Correspondence to Dr. Marc E. de Broe, University of Antwerp, Department of Nephrology-Hypertension, p/a University Hospital Antwerp, Wilrijkstraat 10, B-2650 Edegem/Antwerpen, Belgium; Phone: +32-3-821-3421; Fax: +32-3-829-0100; E-mail: debroe{at}uia.ua.ac.be
ABSTRACT. Retention of crystals in the kidney is an essentialearly step in renal stone formation. Studies with renal tubularcells in culture indicate that hyaluronan (HA) and osteopontin(OPN) and their mutual cell surface receptor CD44 play an importantrole in calcium oxalate (CaOx) crystal binding during woundhealing. This concept was investigated in vivo by treating ratsfor 1, 4, and 8 d with ethylene glycol (0.5 and 0.75%) in theirdrinking water to induce renal tubular cell damage and CaOxcrystalluria. Tubular injury was morphologically scored on periodicacid-Schiff–stained renal tissue sections and tissue repairassessed by immunohistochemical staining for proliferating cellnuclear antigen. CaOx crystals were visualized in periodic acid-Schiff–stainedsections by polarized light microscopy, and renal calcium depositswere quantified with von Kossa staining. HA was visualized withHA-binding protein and OPN and CD44 immunohistochemically withspecific antibodies and quantified with an image analyzer system.Already after 1 d of treatment, both concentrations of ethyleneglycol induced hyperoxaluria and CaOx crystalluria. At thispoint, there was neither tubular injury nor crystal retentionin the kidney, and expression of HA, OPN, and CD44 was comparableto untreated controls. After 4 and 8 d of ethylene glycol, however,intratubular crystals were found adhered to injured/regenerating(proliferating cell nuclear antigen positive) tubular epithelialcells, expressing HA, OPN, and CD44 at their luminal membrane.In conclusion, the expression of HA, OPN, and CD44 by injured/regeneratingtubular cells seems to play a role in retention of crystalsin the rat kidney.
The development of kidney stones requires formation of crystalsfollowed by their retention in the kidney (1). Crystal retentioncould be caused by adherence of crystals to the epithelial cellslining the renal tubules (2). Although many investigators recognizeda role for renal tubular injury in the pathophysiology of nephrolithiasis,definite proof for this concept and the mechanisms involvedare not yet available (3–7).
Most kidney stones are predominantly composed of precipitatedcalcium salts, the most common of which is calcium oxalate monohydrate(5). Studies with renal tubular epithelial cells in cultureshowed that confluent monolayers of distal tubule/collectingduct–like MDCK-I cells are nonadherent to calcium oxalatemonohydrate crystals (8). In contrast, crystals bind to cellsin subconfluent cultures and in confluent monolayers recoveringfrom mechanical injury (9). The glycosaminoglycan hyaluronan(HA) was identified as a major crystal-binding molecule at thesurface of MDCK-I cells (10) and of human renal tubular cellsin primary culture (11). In addition, it was found that crystal-bindingcells not only expressed HA at their apical surface but alsoosteopontin (OPN) and CD44 (11).
HA is a high molecular mass polysaccharide (>106 Da), composedof linear polymers of a repeating disaccharide structure ofalternating glucuronic acid and N-acetylglucosamine. In thekidney, HA is hardly detectable in the cortex but is abundantlypresent as the main component of the renal inner medullary interstitium(12). HA in the kidney is upregulated during various inflammatorydisease states (13–17). OPN is a glycoprotein and is widelyspread throughout organs and tissues, including the kidney.In the healthy kidney, OPN is confined to the distal parts ofa subset of nephrons. During various types of inflammation,however, renal OPN is severely upregulated in most segmentsof the nephron (13,14,18–20). The transmembrane proteinCD44 is a cell surface receptor for both HA and OPN (21,22)and also is upregulated during inflammation in the kidney (13,14).The present study was conducted to investigate whether in vivorenal tubular injury and HA, OPN, and CD44 expression are involvedin crystal retention.
Experimental Design
Male Wistar rats (300 to 350 g) were obtained from the CentralAnimal Breeding Center (Harlan, Zeist, the Netherlands) anddivided into three groups (n = 9 each) receiving drinking watersupplemented with 0, 0.5, or 0.75% (vol/vol) ethylene glycol(EG) for 1, 4, and 8 d. All animals had free access to standardchow. Twenty-four hours before the indicated times, rats werehoused individually in metabolic cages to collect 24-h urinesamples and to monitor fluid intake. Urine samples were dividedinto portions of 5 ml, one portion of which was acidified with100 µl of 1 M hydrochloric acid and stored at -20°Cuntil analysis. Animals were sedated and killed; kidneys wereextracted and decapsulated; and sagittal slices were immediatelyfixed in either methacarn (60% methanol, 30% chloroform, 10%acetic acid) or Dubosq-Brasil fixative (47% ethanol, 11.7% H2O,23.5% formaldehyde, 17.6% acetic acid, and 4 mM picric acid)for 4 h, rinsed with 70% ethanol, and embedded in low-melting-pointparaffin (52°C; BDH Laboratory Supplies, Poole, UK). Serumspecimens were collected and frozen at -20°C until biochemicalanalysis. The experiments were approved by the local UniversityAnimal Committee and carried out in accordance with the NetherlandsExperiments on Animals Act (1977) and the European Conventionfor the Protection of Vertebrate Animals Used for ExperimentalPurposes (Strasbourg, March 18, 1986).
Urine and Serum Biochemistry
Urinary oxalate was determined in acidified urine portions witha quantitative enzymatic colorimetric assay (Sigma Diagnostics,Dei- senhofen, Germany). For determination of urinary citrate,the ultraviolet method with the test combination of BoehringerMannheim (Darmstadt, Germany) was used. The concentrations ofcalcium in urine and bicarbonate, calcium, and creatinine inserum were determined on a routine autoanalyzer system (Vitros750 XRC). Urine samples were centrifuged at 5000 x g, and sedimentswere inspected by optical and polarized light microscopy (ZeissAxioplan microscope, Oberkochen, Germany).
Tubular Morphology
Methacarn-fixed, paraffin-embedded renal tissue sections (4µm) were stained with periodic acid-Schiff (PAS), andnuclei were counterstained with methyl green. Histologic damagewas evaluated with a morphologic scoring system (Table 1) inproximal tubules (PT), thin limbs of Henle (TLH), distal tubules(DT; including thick ascending limbs [TAL]) and collecting ducts(CD). Tubules were morphologically inspected by a reproducibleprocedure, which comprised a random selection of the first tubularcross-section, followed by shifting the microscopic field overfixed distances according to a standardized pattern (x300 magnification).The cortex, outer stripe of the outer medulla (OSOM), innerstripe of the outer medulla (ISOM), and inner medulla of eachkidney section were evaluated. In the cortex, PT (S1–S2)and DT were evaluated (n = 50 and 25, respectively); in theOSOM, PT (S3) and DT (TAL) were evaluated (n = 50 and 25, respectively);in the ISOM, TLH and DT (TAL) were evaluated (n = 25 each);and in the inner medulla, TLH and CD were evaluated (n = 25each). PT and DT could be distinguished according to at leastone of the following morphologic criteria: topographical localization,tubular size and form, cytoplasmic density and position of thenuclei, and presence or absence of brush border and basolateralcell aspect. Tubules in the cortex and OSOM were scored as PTonly when a brush border could be identified; if not, then theywere scored as DT. In the ISOM and inner medulla, TLH couldbe distinguished from DT (TAL) and CD by tubular size and positionof the nuclei.
Table 1. Scoring system for the evaluation of tubular morphology of PT, TLH, DT (including TAL), and CDa
Proliferation was determined by immunohistochemical stainingfor proliferating cell nuclear antigen (PCNA) using the PC10mAb (DAKO, Glostrup, Denmark) as described previously (23).Routinely, sections were counterstained with PAS and nucleiwere stained with methyl green. In a number of sections, PASstaining was omitted, allowing optimal visualization of crystals.Expression of PCNA was quantified morphometrically with an imageanalyzer system (KS-400 V2.0 image analysis software) in thecortex + OSOM and ISOM + inner medulla, by measuring positivesignals in 25 and 15 randomly chosen microscopic fields, respectively(x250 magnification). Measurements were expressed as fractionalpositive area of the tissue section.
Crystal Retention
During the evaluation of tubular morphology, each tubule wasadditionally inspected by polarized light microscopy for thepresence of crystals. In this way, crystal retention was assessedand sites of crystal location were correlated with tubular morphology.
Calcium deposits were also visualized by von Kossa staining.Deparaffinized Dubosq-Brasil–fixed 4-µm tissue sectionswere incubated in 5% silver nitrate for 45 min. Slides wererinsed in water, incubated in 1% pyrogallic acid for 3 min,rinsed in water, fixed in 5% sodium thiosulfate for 1 min, andcounterstained with hematoxylin and eosin. In each sagittalkidney section, calcium deposits were quantified by countingthe total number of positive stained crystals in the cortex+ OSOM and ISOM + inner medulla.
HA, OPN, and CD44 Expression
Renal tissue sections were stained for HA, OPN, and CD44 asdescribed previously (11). Briefly, methacarn-fixed tissue sectionswere blocked with 1% BSA for HA and with normal horse serumfor OPN and CD44 staining and incubated with primary labels(biotinylated HA-binding protein, Seikagaku, Falmouth, MD; goatanti-human OPN antibody, OP189, C.M. Giachelli, University ofWashington; or mouse anti-human CD44 antibody, Bender MedSystems,Vienna, Austria). For OPN and CD44, sections were subsequentlyincubated with secondary labels, biotinylated horse anti-goatand horse anti-mouse antibodies (Vector Laboratories, Burlingame,CA), respectively. Finally, avidin-biotin peroxidase complex(Vector) and diaminobenzidine were used to detect HA, OPN, andCD44. Sections were counterstained with methyl green. No stainingwas observed when primary labels were omitted.
Expression of HA, OPN, and CD44 was quantified morphometricallywith KS-400 V2.0 image analysis software in the cortex and OSOM,by measuring positive signals in 10 and six randomly chosenmicroscopic fields, respectively (x200 magnification). Measurementswere expressed as fractional positive area of the tissue section.OPN and CD44 were also quantified in the remaining part of themedulla, by analyzing six randomly chosen microscopic fieldsof the ISOM and inner medulla. HA was not quantified in thisregion, because the well known abundant amount of HA in theinterstitium of the inner medulla stains nearly the entire tissue(12).
Statistical Analyses
Data are expressed as mean ± SEM. P < 0.05 was consideredto be significant, using t test. Correlation analysis betweentotal tubular PCNA expression per sagittal kidney section andthe total amount of positive von Kossa renal calcium depositsper sagittal kidney section in individual rats was performedby the nonparametric Spearman’s rank order test. P <0.01 was considered significant (two-tailed). Computations wereperformed using SPSS version 10.
Urine and Serum Biochemistry
In the control group, the average urinary oxalate and calciumexcretion was 0.95 ± 0.05 and 3.24 ± 0.79 mg/24h, respectively (Figure 1), and the urinary sediment did notcontain crystals (Figure 2A). Administration of 0.5 and 0.75%EG to the drinking water induced within 24 h a significant concentration-dependenthyperoxaluria (4.37 ± 1.71 and 7.78 ± 4.46 mg/24h, respectively). In the 0.5% EG group, oxalate gradually increasedfurther to reach its maximum level at day 8, whereas in the0.75% EG group, urinary oxalate reached its maximum level alreadyat day 4. Urinary calcium was decreased at day 1 and was undetectableat day 4 and day 8 (Figure 1). At day 1, calcium oxalate crystalswere observed in urinary sediments of both EG groups (Figure 2B).Oxalate was determined in acidified urine portions, whichdissolves crystals and therefore represents the total amountof oxalate, including oxalate precipitated with calcium. Becausecalcium was determined in urine that was not acidified, it representsthe amount of free calcium ions. Thus, the decreased amountsof urinary calcium apparently resulted from the formation ofCaOx crystals.
Figure 1. Urine- and serum biochemical analysis, urine production, and fluid intake. Statistical analysis by t test. *Significantly different compared with controls at the same point in time (P < 0.05).
Figure 2. Urinary sediment inspected by polarized light microscopy of a control rat (A) and a 0.5% ethylene glycol (EG)-treated rat (B) after 1 d. (A) Some debris was observed, but no urinary crystalline material. (B) Numerous urinary calcium oxalate crystals in the EG-treated rat are clearly visible. Magnification, x400.
In the 0.5% EG group at day 8 and in the 0.75% EG group at days4 and 8, increased diuresis and fluid intake were observed comparedwith controls. Thus, the addition of EG led to polyuria, whichmay be secondary to the osmotic effect of EG excreted in theurine unchanged (24) or simply because these rats drank morewater (Figure 1). There was no increase in serum creatinine,except for a slight increase in the low-dose EG at day 8, indicatingthat renal function was preserved in these rats. These biochemistrydata are in accordance with earlier observations in rats treatedwith EG (25).
EG caused a metabolic acidosis, as can be derived from the concentration-dependentdecrease in urinary citrate after 4 d. This metabolic acidosiswas relatively mild, however, as serum bicarbonate and calciumwere not affected and a compensatory homeostatic response seemedto normalize urinary citrate after continued EG challenge (Figure 1).
Tubular Morphology
The relatively low concentrations of EG in the present studydid not result in frank necrosis but in mild changes in tubularmorphology. At day 1, tubular morphology was comparable to controls,but at days 4 and 8, different degrees of injury/regenerationwere found in tubules (Figure 3). PT of the OSOM (S3) and, toa lesser extent, PT in the cortex (S1–S2) suffered themost morphologic damage, for the most part with loss of brushborder (score 2). Tubules with flattened cells (score 3) werealso observed, predominantly in DT (TAL) of the OSOM. Becausea tubule with flattened cells was scored only as PT if therewas still brush border recognizable, the number of PT with score3 could actually have been higher. In TLH and CD at days 4 and8, increased amounts of luminal cell debris (score 1) were observed,but the majority of these tubules had a normal morphology (score0). Importantly, cell debris in the tubular lumen in the distalnephron could also have been derived from an "upstream" sectionof the nephron, and the tubule concerned could actually havebeen completely healthy.
Figure 3. Evaluation of tubular morphology with the scoring system of Table 1 in proximal tubules (PT), thin limbs of Henle (TLH), distal tubules (DT; including thick ascending limbs [TAL]), and collecting ducts (CD 0 = intact tubule with normal appearance, 1 = tubule with luminal cell debris, 2 = PT: tubule with loss of brush border/DT (TAL) and CD: dilated tubule, 3 = tubule with flattened cells).
At day 1, PCNA staining was comparable to control kidneys, inwhich tubular expression of this protein was sparse (Figure 4A),whereas at days 4 and 8, PCNA staining was clearly upregulatedin tubules in rats that received EG (Figure 4B).
Figure 4. Proliferating cell nuclear antigen (PCNA) staining in a kidney of a control rat (A) and of a rat that received 0.75% EG for 8 d (B through D). In D, periodic acid-Schiff (PAS) counterstaining was omitted. (A) Sparse PCNA staining, mainly present in the interstitium and glomeruli in a control rat. (B) Strongly upregulated PCNA in the tubules of an EG-treated rat. (C and D) Crystals associated with PCNA-positive, flattened cells. Magnifications: x400 in A, x200 in B, x630 in C and D.
Crystal Retention
No crystals were found in control rat kidneys (Figure 5). After1 d of 0.5 and 0.75% EG, no crystals were observed by polarizedlight microscopy, consistent with a marginal amount of positivevon Kossa signals. At days 4 and 8, however, an increasing numberof crystals were found attached to the luminal membrane of tubularepithelial cells, corresponding with a markedly increased numberof positive von Kossa signals (Figure 5). Crystals were smallerin size than the tubular lumen and not seen intracellularlyor in the interstitium.
Figure 5. Quantification of renal calcium deposits on von Kossa–stained tissue sections.
In Figure 6, retention of crystals and tubular morphology isplotted in a graph. In PT, DT (TAL), and CD, crystals were notobserved in histologic normal tubules (score 0) or tubules withluminal cell debris (score 1). Crystals were found adhered totubular cells in PT with loss of brush border and DT (TAL) andCD with dilation of the tubular lumen (score 2), and in tubuleswith flattened cells (score 3). Crystals were not observed inTLH (data not shown). Figure 7 shows a DT with small crystalsat the luminal membrane of flattened tubular epithelial cells(score 3). Furthermore, these flattened cells were PCNA positive,proliferating cells (Figure 4, C and D). Approximately 93% ofthe observed crystals in PCNA-stained renal tissue sectionswere found to be adhered to PCNA-positive cells.
Figure 6. Retention of crystals in relation to tubular morphology. In PT, DT (including TAL), and CD, crystals were not observed in tubules with score 0 (normal appearance) or score 1 (luminal cell debris). Crystals were exclusively retained at the luminal surface of tubular cells in PT, DT, and CD with score 2 (loss of brush border in PT and dilated tubules in DT and CD) and score 3 (tubules with flattened cells).
Figure 7. PAS and methyl green–stained renal tissue section of a rat treated with 0.75% EG for 4 d by optical (A) or polarized (B) light microscopy. (A) A dilated DT is shown with flattened epithelial cells (score 3). (B) In the lumen of this DT, crystals that are smaller than the tubular lumen are clearly visible and located at the luminal side of the injured/regenerating flattened epithelium. Magnification, x1000.
Correlation analysis was performed between the total amountof tubular PCNA expression and the amount of positive von Kossacalcium deposits in sagittal kidney sections of individual rats(Figure 8). Tubular cell regeneration was associated with crystaladhesion/retention (Spearman’s Rho = 0.819; P < 0.0001).
Figure 8. Correlation analysis was performed between the total amount of tubular PCNA expression and the amount of positive von Kossa calcium deposits in sagittal kidney sections of individual rats (Spearman’s Rho = 0.819, P < 0.0001).
HA, OPN, and CD44 Expression
HA in the cortex and OSOM of control rats was scarcely observedin the interstitium and around a few glomerular capsules. After1 d, HA expression in EG-treated rats was comparable to controls(Figure 9). After 4 and 8 d, however, HA was upregulated ina focal pattern throughout the cortex and OSOM of EG-treatedrats (Figure 9), primarily in the interstitium but also at theluminal membrane of tubular cells (Figure 10A). Strikingly,in the majority of cases, crystals were found at the luminalsurface of HA-expressing cells (Figure 10A).
Figure 9. Quantification of renal hyaluronan (HA), osteopontin (OPN), and CD44 expression using computerized image analysis software (KS400). Data are expressed as fractional positive area (%). Statistical analysis by t test. *Significantly increased compared with controls at the same point in time (P < 0.05).
Figure 10. (A) The outer stripe of the outer medulla (OSOM) stained for HA of a rat with 0.75% EG in the drinking water for 8 d. HA is upregulated in the interstitium as well as at the apical membrane of tubular epithelial cells. A crystal is shown close to cells that express HA at their luminal membrane. (B and C) The OSOM stained for OPN after 8 d of 0.75% EG, showing increased typical Golgi apparatus immunostaining in PT and at the apical cell membrane in DT. (C) Crystals visualized with polarized light microscopy. Retained crystals stained for OPN are smaller than the tubular lumen and closely associated with apical OPN-expressing epithelium. (D) The cortex stained for CD44 after 8 d of 0.75% EG. CD44 is upregulated at basolateral and apical tubular cell membranes. Polarized light microscopy shows birefringent crystals in the lumen of a tubule closely associated with the surface of flattened cells positive for CD44. Magnification, x1000.
OPN was expressed in a limited number of distal tubular epithelialcells of control rats. After 1 d of EG, OPN expression was notdifferent from controls, except for a small increase in theOSOM of the moderate-dose EG group (Figure 9). At days 4 and8, however, OPN was significantly upregulated in the kidneysof EG-treated rats (Figure 9). Both PT and DT showed a risein OPN immunostaining (Figure 10, B and C). The classical OPNexpression pattern was observed, with Golgi apparatus immunostainingin PT and at the apical membrane in DT, allowing differentiationbetween PT and DT (previously described after renal ischemia/reperfusion)(19,26). Crystals were retained in apical OPN-expressing DT.Crystals were positive for OPN, which was evident by comparisonbetween optical and polarized light microscopy (Figure 10, B and C).
CD44 was hardly expressed in the rat kidney, except for someinterstitial cells and glomeruli. After 1 d of EG, CD44 expressionwas not significantly different from controls (Figure 9). After4 and 8 d of EG, however, CD44 in the cortex, OSOM, and medullawas markedly upregulated (Figure 9). CD44 expression was prominentin a focal pattern along basolateral and apical tubular cellmembranes (Figure 10D). Crystals were observed frequently closelyat sites where cells expressed CD44 at their luminal membrane(Figure 10D).
In the present study, we investigated the role of HA, OPN, andCD44 in retention of crystals in renal tubules damaged by EG.The EG model is extremely suitable for this purpose since thisagent not only is toxic to the nephron but also generates urinaryCaOx crystals (25,27–30). For investigating the earliestevents involved in crystal retention, these studies were performedshortly (1, 4, and 8 d) after the addition of EG to the drinkingwater and at relatively low concentrations to avoid massivetissue damage. One of the major findings was that there is nocrystal retention in the absence of tubular injury/regenerationbut that crystals are retained as soon as renal tubules areinjured/regenerating. Crystals were found adhered to the luminalsurface of HA-, OPN-, and CD44-expressing injured/regeneratingcells. The results of this study therefore strongly suggestthat crystal retention in the kidney requires tubular epithelialinjury accompanied by luminal expression of HA, OPN, and CD44.
EG, which itself is not nephrotoxic, is metabolized in the liverto several intermediates, including glycoaldehyde, glycolate,glyoxylate, and oxalate (27). It is still a matter of debatewhich of these metabolites are responsible for the damage torenal tubular cells (28,29). Oxalate precipitates as CaOx inthe primary urine as a result of its poor solubility. The nephrotoxiceffect of EG has also been attributed to these crystals (25,31).However, autopsy and renal biopsy studies in humans did notsupport the concept that crystals are the primary cause of EGtoxicity (27). Considering this controversy, it is impossiblein the present study to distinguish between crystals bindingto injured/regenerating cells and the alternative explanationthat crystal deposition causes cell injury (and consequentlycells bind crystals). Crystals did not seem to be retained becauseof their size, because no crystal aggregates occluding the tubularlumen were observed, but crystals were generally smaller (Figures 4, C and D, 7, A and B, and 10, A, B, C, and D).
In urolithiasis research, animals are usually treated with relativelyhigh concentrations of EG for several weeks (25,30,31). Alsoin these studies, crystals are retained in tubules that areclearly injured (25,31,32). In the present study, the shortperiod of treatment (1 to 8 d) with relatively low concentrationsof EG did not result in frank necrosis or cast formation butin a mild form of injury/regeneration. Renal tubular injurywas morphologically scored (see Table 1), and PCNA was usedto assess cell proliferation. PCNA is an auxiliary protein forDNA polymerase and required for both DNA replication and DNArepair. In proliferating cells, it is upregulated in cell nucleimainly during the S-phase (DNA synthesis phase) of the cellcycle (33). Consequently, it is widely used as a marker forproliferating cells. Immunohistochemical PCNA staining is avalidated method for evaluating epithelial regeneration of thekidney after renal damage (23). It is well known that in thekidney, tubular cell injury and regeneration occur concurrently(23). Hence, the flattened, PCNA-positive tubular cells after4 and 8 d of EG treatment (Figure 4B) are dedifferentiated proliferatingcells.
HA is a high molecular mass polysaccharide found in many tissues,where it performs a great range of biologic functions (34,35).In the kidney, HA is normally not expressed at the luminal surfaceof tubular cells and is present only in the medullary interstitium,where it provides structural stability to the tubules and contributesto concentrating the urine (12). HA expression in the renalcortex is upregulated in renal inflammatory diseases such asinterstitial nephritis (13), acute ischemic injury (14,17),autoimmune renal injury (15), and acutely rejecting human kidneygrafts (16). HA accumulates in wounded tissue shortly afterinjury to form loose hydrated matrices that allow cell divisionand migration (36,37). During recent years, we searched extensivelyfor crystal-binding molecules at the surface of renal tubularcells in culture (10,11,38,39). HA was identified as major crystal-bindingsubstance, based on the following results: (1) crystals bindto HA-expressing cells at subconfluence but not to cells inconfluent cultures that do no longer express HA; (2) metaboliclabeling studies showed that the surface of proliferating cellscontains substantially higher levels of radiolabeled HA; (3)crystal binding could be decreased by Streptomyces hyaluronidase,an enzyme that specifically digests HA; and (4) during woundhealing, HA-binding protein binds to migrating and proliferatingflattened cells in damaged areas but not to cells in intactmonolayers (10). The co-localization of intraluminal CaOx crystalswith HA-expressing tubular cells in the present study suggestsfor the first time, to our knowledge, a role for HA in crystalretention in vivo.
The glycoprotein OPN is widely distributed in the body and hasbeen implicated in several physiologic and pathologic processes,including cell adhesion, migration, signaling, inflammation,and biomineralization (18,40,41). In the kidney, the expressionof OPN is upregulated during renal injury (18,19,42), includinginduced by EG (31). The role of OPN under these circumstancesis unclear, but OPN seems to be involved in mediating macrophageaccumulation and interstitial fibrosis (18,43). The role ofOPN in urolithiasis is controversial. OPN has been proposedas an inhibitor of crystal formation and retention but alsoas a promoter of crystal retention (44–46). Recently,crystal retention was studied in OPN knockout and wild-typemice treated with EG (44). Noticeably, after 4 wk of treatmentwith 1% EG, no crystals were retained in wild-type mice, whereassome crystal retention was observed in the kidneys of OPN knockouts(14.3 ± 3 von Kossa signals per sagittal kidney section).From these observations, the authors concluded that OPN protectsthe kidney from crystal formation and retention. However, thecrystal retention inhibitor function of OPN seems to be ineffectivein the present study, because crystals covered with OPN butsmaller in size than the tubular lumen became firmly associatedwith the cell surface (Figure 10, B and C). This is in agreementwith a previous study, in which it was demonstrated that urinaryinhibitors of crystallization were unable to prevent the attachmentof crystals to regenerating renal tubular cells (47). Consequently,possibly because of the differences in species and study designsbetween the present study and the one performed by Wesson etal. (44), the role of OPN remains controversial.
CD44 is a ubiquitous transmembrane glycoprotein that is involvedin many processes including inflammation (48). CD44 serves ascell surface receptor for both HA and OPN (21,22). In fact,the biologic activity of HA and OPN predominantly depends ontheir interaction with CD44 (49–51). In the kidney, theexpression of CD44 is highly upregulated during various renaldisease states (13,14). Hence, it is not surprising that anupregulated expression of CD44 in the renal tubules is accompaniedby an increased expression of its ligands HA and OPN. Reportsof the role of CD44 in renal stone formation are scarce. Incell culture, it was found that CD44 is expressed at the luminalsurface of crystal-binding renal tubular cells but not on thatof cells without affinity for crystals (11). In the presentstudy, crystals were also retained at sites where cells expressedCD44. In our opinion, crystals are not likely to become associatedwith CD44 but rather with HA and/or OPN.
The present study supports in vivo for the first time the conceptthat crystal retention is associated with HA expressed at theluminal surface of injured/regenerating cells. However, we cannotrule out the importance of other crystal-binding molecules,including sialic acid–containing glycoproteins (52), phosphatidylserine (53), collagen (54), and nucleolin-related protein (55).Furthermore, several other macromolecules have been implicatedin CaOx crystallization and retention (56), including inter-inhibitor–related proteins (32) such as bikunin (57) andprothrombin fragment 1 (58). Although in the present study wefocused on HA, OPN, and CD44, the interrelationship betweenthese and other molecules as part of complex cell biologic pathwaysin the pathophysiology of kidney stone disease remains to bedetermined (2).
Although the expression of HA, OPN, and CD44 by injured/regeneratingtubular epithelial cells most likely is aimed to reestablishmentof the epithelial barrier integrity and restoration of renalfunction, a negative side effect could be that it turns a non–crystal-bindingepithelium into a crystal-binding one, thereby setting the stagefor crystal retention (Figure 11). The clinical relevance ofthis concept was recently reinforced by observations in kidneysof preterm neonates, in which the development of nephrocalcinosisis common, showing that HA and OPN are abundantly expressedat the luminal membrane of proliferating tubular cells (unpublishedresults).
Figure 11. Paradigm of crystal retention. A schematic representation of a cross-section of a distal tubule is shown. (A) Under normal conditions, crystals do not bind to renal tubular epithelial cells and are harmlessly excreted in the urine. (B) Crystal retention is preceded by a stressor injuring the epithelium. (C) During the process of tubular epithelial regeneration and repair, flattened epithelial cells express HA, OPN, and CD44 at their apical membrane. HA is a cell surface crystal-binding molecule. Because these regenerating dedifferentiated tubular epithelial cells are susceptible to crystal binding, crystal retention may ensue.
In conclusion, the results obtained in this study support theconcept that the expression of HA, OPN, and CD44 by injured/regeneratingtubular cells is a prerequisite for retention of crystals inthe kidney.
Acknowledgments
This study was supported by the Oxalosis and Hyperoxaluria Foundation(Grant No. 2749036).
We express our gratitude to Dr. C.M. Giachelli (University ofWashington, Seattle, WA) for the generous gift of the OP189antiserum. We thank Simonne Dauwe for excellent technical assistance,as well as the biochemistry laboratory of the Antwerp UniversityHospital (Dr. V. van Hoof and C. Verstraten). We thank Frankvan der Panne and Dirk de Weerdt for contribution of the photographs.
Mandel N: Mechanism of stone formation. Semin Nephrol 16: 364–374, 1996[Medline]
Baggio B, Gambaro G, Ossi E, Favaro S, Borsatti A: Increased urinary excretion of renal enzymes in idiopathic calcium oxalate nephrolithiasis. J Urol 129: 1161–1162, 1983[Medline]
Kumar S, Sigmon D, Miller T, Carpenter B, Khan S, Malhotra R, Scheid C, Menon M: A new model of nephrolithiasis involving tubular dysfunction/injury. J Urol 146: 1384–1389, 1991[Medline]
Verkoelen CF, van der Boom BG, Kok DJ, Houtsmuller AB, Visser P, Schroder FH, Romijn JC: Cell type-specific acquired protection from crystal adherence by renal tubule cells in culture. Kidney Int 55: 1426–1433, 1999[CrossRef][Medline]
Verkoelen CF, van der Boom BG, Houtsmuller AB, Schroder FH, Romijn JC: Increased calcium oxalate monohydrate crystal binding to injured renal tubular epithelial cells in culture. Am J Physiol 274: F958–F965, 1998
Verkoelen CF, Van Der Boom BG, Romijn JC: Identification of hyaluronan as a crystal-binding molecule at the surface of migrating and proliferating MDCK cells. Kidney Int 58: 1045–1054, 2000[CrossRef][Medline]
Verhulst A, Asselman M, Persy VP, Schepers MS, Helbert MF, Verkoelen CF, De Broe ME: Crystal retention capacity of cells in the human nephron: Involvement of CD44 and its ligands hyaluronic acid and osteopontin in the transition of a crystal binding- into a nonadherent epithelium. J Am Soc Nephrol 14: 107–115, 2003[Abstract/Free Full Text]
Knepper MA, Saidel GM, Hascall VC, Dwyer T: Concentration of solutes in the renal inner medulla: Interstitial hyaluronan as a mechano-osmotic transducer. Am J Physiol Renal Physiol 284: F433–F446, 2003[Abstract/Free Full Text]
Sibalic V, Fan X, Loffing J, Wuthrich RP: Upregulated renal tubular CD44, hyaluronan, and osteopontin in kdkd mice with interstitial nephritis. Nephrol Dial Transplant 12: 1344–1353, 1997[Abstract/Free Full Text]
Lewington AJ, Padanilam BJ, Martin DR, Hammerman MR: Expression of CD44 in kidney after acute ischemic injury in rats. Am J Physiol Regul Integr Comp Physiol 278: R247–R254, 2000[Abstract/Free Full Text]
Feusi E, Sun L, Sibalic A, Beck-Schimmer B, Oertli B, Wuthrich RP: Enhanced hyaluronan synthesis in the MRL-Fas(lpr) kidney: Role of cytokines. Nephron 83: 66–73, 1999[CrossRef][Medline]
Wells A, Larsson E, Hanas E, Laurent T, Hallgren R, Tufveson G: Increased hyaluronan in acutely rejecting human kidney grafts. Transplantation 55: 1346–1349, 1993[Medline]
Johnsson C, Tufveson G, Wahlberg J, Hallgren R: Experimentally-induced warm renal ischemia induces cortical accumulation of hyaluronan in the kidney. Kidney Int 50: 1224–1229, 1996[Medline]
Xie Y, Sakatsume M, Nishi S, Narita I, Arakawa M, Gejyo F: Expression, roles, receptors, and regulation of osteopontin in the kidney. Kidney Int 60: 1645–1657, 2001[CrossRef][Medline]
Persy VP, Verstrepen WA, Ysebaert DK, De Greef KE, De Broe ME: Differences in osteopontin up-regulation between proximal and distal tubules after renal ischemia/reperfusion. Kidney Int 56: 601–611, 1999[CrossRef][Medline]
Xie Y, Nishi S, Iguchi S, Imai N, Sakatsume M, Saito A, Ikegame M, Iino N, Shimada H, Ueno M, Kawashima H, Arakawa M, Gejyo F: Expression of osteopontin in gentamicin-induced acute tubular necrosis and its recovery process. Kidney Int 59: 959–974, 2001[CrossRef][Medline]
Aruffo A, Stamenkovic I, Melnick M, Underhill CB, Seed B: CD44 is the principal cell surface receptor for hyaluronate. Cell 61: 1303–1313, 1990[CrossRef][Medline]
Weber GF, Ashkar S, Glimcher MJ, Cantor H: Receptor-ligand interaction between CD44 and osteopontin (Eta-1). Science 271: 509–512, 1996[Abstract]
Nouwen EJ, Verstrepen WA, Buyssens N, Zhu MQ, De Broe ME: Hyperplasia, hypertrophy, and phenotypic alterations in the distal nephron after acute proximal tubular injury in the rat. Lab Invest 70: 479–493, 1994[Medline]
Hewlett TP, Jacobsen D, Collins TD, McMartin KE: Ethylene glycol and glycolate kinetics in rats and dogs. Vet Hum Toxicol 31: 116–120, 1989[Medline]
Khan SR: Animal model of calcium oxalate nephrolithiasis. In: Calcium Oxalate in Biological Systems,edited by Khan SR, Boca Raton, CRC Press, 1995, pp 343–359
Verhulst A, Persy VP, Van Rompay AR, Verstrepen WA, Helbert MF, De Broe ME: Osteopontin synthesis and localization along the human nephron. J Am Soc Nephrol 13: 1210–1218, 2002[Abstract/Free Full Text]
Schrier RW: Ethylene glycol toxicity. In: Diseases of the Kidney and Urinary Tract, 7th ed., Philadelphia, Lippincott, Williams & Wilkins 2001, pp 1316–1326
Poldelski V, Johnson A, Wright S, Rosa VD, Zager RA: Ethylene glycol-mediated tubular injury: Identification of critical metabolites and injury pathways. Am J Kidney Dis 38: 339–348, 2001[Medline]
Roberts JA, Seibold HR: Ethylene glycol toxicity in the monkey. Toxicol Appl Pharmacol 15: 624–631, 1969[CrossRef][Medline]
Lyon ES, Borden TA, Vermeulen CW: Experimental oxalate lithiasis produced with ethylene glycol. Invest Urol 4: 143–151, 1966[Medline]
Khan SR, Johnson JM, Peck AB, Cornelius JG, Glenton PA: Expression of osteopontin in rat kidneys: Induction during ethylene glycol induced calcium oxalate nephrolithiasis. J Urol 168: 1173–1181, 2002[CrossRef][Medline]
Moriyama MT, Glenton PA, Khan SR: Expression of inter-alpha inhibitor related proteins in kidneys and urine of hyperoxaluric rats. J Urol 165: 1687–1692, 2001[CrossRef][Medline]
Tammi MI, Day AJ, Turley EA: Hyaluronan and homeostasis: A balancing act. J Biol Chem 277: 4581–4584, 2002[Free Full Text]
Toole BP, Wight TN, Tammi MI: Hyaluronan-cell interactions in cancer and vascular disease. J Biol Chem 277: 4593–4596, 2002[Free Full Text]
Noble PW: Hyaluronan and its catabolic products in tissue injury and repair. Matrix Biol 21: 25–29, 2002[CrossRef][Medline]
Chen WY, Grant ME, Schor AM, Schor SL: Differences between adult and foetal fibroblasts in the regulation of hyaluronate synthesis: Correlation with migratory activity. J Cell Sci 94: 577–584, 1989[Abstract/Free Full Text]
Verkoelen CF, Van Der Boom BG, Kok DJ, Schroder FH, Romijn JC: Attachment sites for particles in the urinary tract. J Am Soc Nephrol 10 [Suppl 14]: S430–S435, 1999
Verkoelen CF, van der Boom BG, Kok DJ, Romijn JC: Sialic acid and crystal binding. Kidney Int 57: 1072–1082, 2000[CrossRef][Medline]
Hudkins KL, Giachelli CM, Cui Y, Couser WG, Johnson RJ, Alpers CE: Osteopontin expression in fetal and mature human kidney. J Am Soc Nephrol 10: 444–457, 1999[Abstract/Free Full Text]
Mazzali M, Kipari T, Ophascharoensuk V, Wesson JA, Johnson R, Hughes J: Osteopontin—A molecule for all seasons. QJM 95: 3–13, 2002[Free Full Text]
Denhardt DT, Noda M, O’Regan AW, Pavlin D, Berman JS: Osteopontin as a means to cope with environmental insults: Regulation of inflammation, tissue remodeling, and cell survival. J Clin Invest 107: 1055–1061, 2001[Medline]
Persy VP, Verhulst A, Ysebaert DK, De Greef KE, De Broe ME: Reduced postischemic macrophage infiltration and interstitial fibrosis in osteopontin knockout mice. Kidney Int 63: 543–553, 2003[CrossRef][Medline]
Wesson JA, Johnson RJ, Mazzali M, Beshensky AM, Stietz S, Giachelli C, Liaw L, Alpers CE, Couser WG, Kleinman JG, Hughes J: Osteopontin is a critical inhibitor of calcium oxalate crystal formation and retention in renal tubules. J Am Soc Nephrol 14: 139–147, 2003[Abstract/Free Full Text]
Yamate T, Kohri K, Umekawa T, Iguchi M, Kurita T: Osteopontin antisense oligonucleotide inhibits adhesion of calcium oxalate crystals in Madin-Darby canine kidney cell. J Urol 160: 1506–1512, 1998[CrossRef][Medline]
Lieske JC, Hammes MS, Hoyer JR, Toback FG: Renal cell osteopontin production is stimulated by calcium oxalate monohydrate crystals. Kidney Int 51: 679–686, 1997[Medline]
Schepers MS, van der Boom BG, Romijn JC, Schroder FH, Verkoelen CF: Urinary crystallization inhibitors do not prevent crystal binding. J Urol 167: 1844–1847, 2002[CrossRef][Medline]
Pure E, Cuff CA: A crucial role for CD44 in inflammation. Trends Mol Med 7: 213–221, 2001[CrossRef][Medline]
Lesley J, Hyman R, English N, Catterall JB, Turner GA: CD44 in inflammation and metastasis. Glycoconj J 14: 611–622, 1997[CrossRef][Medline]
Lesley J, English NM, Gal I, Mikecz K, Day AJ, Hyman R: Hyaluronan binding properties of a CD44 chimera containing the link module of TSG-6. J Biol Chem 277: 26600–26608, 2002[Abstract/Free Full Text]
Zohar R, Suzuki N, Suzuki K, Arora P, Glogauer M, McCulloch CA, Sodek J: Intracellular osteopontin is an integral component of the CD44-ERM complex involved in cell migration. J Cell Physiol 184: 118–130, 2000[CrossRef][Medline]
Lieske JC, Leonard R, Swift H, Toback FG: Adhesion of calcium oxalate monohydrate crystals to anionic sites on the surface of renal epithelial cells. Am J Physiol 270: F192–F199, 1996
Wiessner JH, Hasegawa AT, Hung LY, Mandel NS: Oxalate-induced exposure of phosphatidylserine on the surface of renal epithelial cells in culture. J Am Soc Nephrol 10 [Suppl 14]: S441–S445, 1999
Kohri K, Kodama M, Ishikawa Y, Katayama Y, Matsuda H, Imanishi M, Takada M, Katoh Y, Kataoka K, Akiyama T, et al.: Immunofluorescent study on the interaction between collagen and calcium oxalate crystals in the renal tubules. Eur Urol 19: 249–252, 1991[Medline]
Sorokina EA, Kleinman JG: Cloning and preliminary characterization of a calcium-binding protein closely related to nucleolin on the apical surface of inner medullary collecting duct cells. J Biol Chem 274: 27491–27496, 1999[Abstract/Free Full Text]
Wesson JA, Worcester EM, Wiessner JH, Mandel NS, Kleinman JG: Control of calcium oxalate crystal structure and cell adherence by urinary macromolecules. Kidney Int 53: 952–957, 1998[CrossRef][Medline]
Iida S, Peck AB, Byer KJ, Khan SR: Expression of bikunin mRNA in renal epithelial cells after oxalate exposure. J Urol 162: 1480–1486, 1999[CrossRef][Medline]
Nishio S, Hatanaka M, Takeda H, Iseda T, Iwata H, Yokoyama M: Analysis of urinary concentrations of calcium phosphate crystal-associated proteins: 2-HS-glycoprotein, prothrombin F1, and osteopontin. J Am Soc Nephrol 10 [Suppl 14]: S394–S396, 1999
Received for publication April 22, 2003.
Accepted for publication September 7, 2003.
This article has been cited by other articles:
C. F. Verkoelen Crystal Retention in Renal Stone Disease: A Crucial Role for the Glycosaminoglycan Hyaluronan?
J. Am. Soc. Nephrol.,
June 1, 2006;
17(6):
1673 - 1687.
[Abstract][Full Text][PDF]
M. L. Green, M. Hatch, and R. W. Freel Ethylene glycol induces hyperoxaluria without metabolic acidosis in rats
Am J Physiol Renal Physiol,
September 1, 2005;
289(3):
F536 - F543.
[Abstract][Full Text][PDF]
Top
Abstract
Introduction
and required for both DNA replication and DNA repair. In proliferating cells, it is upregulated in cell nuclei mainly during the S-phase (DNA synthesis phase) of the cell cycle (33). Consequently, it is widely used as a marker for proliferating cells. Immunohistochemical PCNA staining is a validated method for evaluating epithelial regeneration of the kidney after renal damage (23). It is well known that in the kidney, tubular cell injury and regeneration occur concurrently (23). Hence, the flattened, PCNA-positive tubular cells after 4 and 8 d of EG treatment (Figure 4B) are dedifferentiated proliferating cells. 
inhibitor–related proteins (32) such as bikunin (57) and prothrombin fragment 1 (58). Although in the present study we focused on HA, OPN, and CD44, the interrelationship between these and other molecules as part of complex cell biologic pathways in the pathophysiology of kidney stone disease remains to be determined (2). 
