Crystal Surface Adhesion Explains the Pathological Activity of Calcium Oxalate Hydrates in Kidney Stone Formation
Xiaoxia Sheng*,
Michael D. Ward* and
Jeffrey A. Wesson
* Department of Chemical Engineering and Materials Science, University of Minnesota, Minneapolis, Minnesota; and Nephrology Division, Department of Veteran Affairs Medical Center and the Medical College of Wisconsin, Milwaukee, Wisconsin
Address correspondence to: Dr. Michael D. Ward, Department of Chemical Engineering and Materials Science, University of Minnesota, 421 Washington Avenue SE, Minneapolis, MN 55455. Phone: 612-625-3062; Fax: 612-626-7805; E-mail:wardx004{at}umn.edu
Renal tubular fluid in the distal nephron of the kidney is supersaturatedwith calcium oxalate (CaOx), which crystallizes in the tubulesas either calcium oxalate monohydrate (COM) or calcium oxalatedihydrate (COD). Kidney stones are aggregates, most commonlycontaining microcrystals of COM as the primary inorganic constituent.Stones also contain small amounts of embedded proteins, whichare thought to play an adhesive role in these aggregates, andthey often are found attached to the tip of renal papilla, presumablythrough adhesive contacts. Voided urine, however, often containsCOD in the form of single micron-sized crystals. This suggeststhat COD formation protects against stone disease because ofits reduced capacity to form stable aggregates and strong adhesioncontacts to renal epithelial cells. Using atomic force microscopyconfigured with tips modified with biologically relevant functionalgroups, we have compared the adhesion strengths of the morphologicallyimportant faces of COM and COD. These measurements provide directexperimental evidence, at the near molecular level, for pooreradhesion at COD crystal faces, which explains the benign characterof COD and has implications for resolving one of the mysteriesof kidney stone formation.
Kidney stone disease, which occurs in >10% of the US population,causes substantial suffering and occasional renal failure, yetthe disease mechanism is poorly understood. Calcium oxalatemonohydrate (COM), the thermodynamically most stable form, isobserved more frequently in clinical stones than calcium oxalatedihydrate (COD), at a ratio of >2:1 (1). Adverse physiologiceffects are correlated with the presence of COM in the urinarytract, and COM retention within the renal tubules of rat kidneysections and in hyperoxaluric patients has been demonstrated.Macromolecules isolated from normal urine (i.e., healthy individuals)inhibit COM crystal growth in vitro, favoring formation of CODover COM, crystallization in normal urine typically affordsCOD, and COM adheres more than COD to renal epithelial cellsin culture (2). Recent experiments using mice that were geneticallyaltered to suppress osteopontin, a urinary protein thought toinhibit cell adhesion, demonstrated that only COM crystals wereretained within the renal tubules (3). Conversely, asymptomaticindividuals often have COD microcrystals in voided urine (4).The evidence strongly suggests that COD is a benign crystalform in the urinary tract, but the origins of its protectiverole are not understood.
Although crystallogenesis is essential to stone formation, calciumoxalate (CaOx) crystal growth rates are sluggish, to the extentthat during typical urine transit times single crystals willnot grow large enough to become lodged in the terminal collectingduct of the kidney (5,6). Consequently, crystal aggregationand attachment (of crystals or aggregates) to renal epithelialcells must represent critical processes in stone formation.Stones contain small amounts of embedded proteins, which arethought to serve as adhesive bridges between crystals in aggregates.Some in vitro studies have suggested that urinary proteins withsubstantial anionic functionalities serve as adhesives for COMaggregation and attachment to epithelial cells (710).These observations, coupled with the predominance of COM instones and substantial amounts of COD microcrystals in voidedurine, suggest that the difference in the pathologic behaviorsof COM and COD is related to the adhesive character of theircrystal faces, with COD less likely to form stable aggregatesor strong adhesion contacts with epithelial cells. The adhesionstrength of the crystal surfaces of these two forms, however,has never been compared.
To address this critical knowledge gap, we have compared theadhesion strength of various crystal faces of COM and COD usingatomic force microscopy (AFM), wherein the force required todetach AFM tips modified with biologically relevant functionalgroups from the crystal surfaces is measured directly. Thisapproach provides direct experimental evidence, at the nearmolecular level, for poorer adhesion at COD crystal faces, whichexplains the benign character of COD and has implications forresolving one of the mysteries of kidney stone formation.
The COM crystals used for adhesion force measurements were grownin vitro according to previously reported methods (1113).The specific method was chosen to maximize the area of a particularcrystal face so that AFM measurements were feasible. COD crystalswith large (100) faces were grown in aqueous solutions containing1.0 mM CaOx, 10 µg/ml poly(acrylic acid), 10 mM HEPESbuffer, and 150 mM sodium chloride (Sigma Aldrich, St. Louis,MO). COD crystals with large (101) faces were grown in 1.0 mMCaOx and 5 µg/ml poly(acrylic acid), in the absence ofbuffer and salt. Characterization of the COM crystal surfacesand adhesion force measurements were executed with a DigitalInstruments Nanoscope IIIa Multimode system (Digital Instruments,Santa Barbara, CA). All measurements were performed in aqueoussolutions saturated with CaOx at approximately pH 7, which ensuredthat the crystal surfaces were stable. The CaOx crystals wereaffixed to the AFM specimen holder in the proper orientationand washed with aqueous solutions saturated with CaOx beforeAFM measurements. The COD crystals were washed thoroughly withthis solution to remove all traces of poly(acrylic acid), whichwas required to promote the formation of COD. AFM images ofthe surfaces of the washed COD crystals were flat and devoidof any features associated with adsorbed poly(acrylic acid),which typically is evident as irregular lumps when adsorbedon CaOx surfaces. Powder x-ray diffraction of COD crystals grownby this method confirmed the existence of COD, and no unusualfeatures were observed that would suggest inclusion of poly(acrylicacid) in the crystals, as has been suggested for proteins andamino acids in COM (14,15). Topographical and lattice imagesof the crystal surfaces were acquired with commercial Si3N4cantilever tips in aqueous solutions containing 0.11 mM CaOx.Adhesion force measurements were performed in the same solutionusing gold-coated Si3N4 tips modified with various thiol molecules.The spring constant of the AFM cantilevers, which is neededto convert the cantilever deflection to an adhesion force, was0.074 ± 0.003 N/m, as measured by a previously reportedmethod (16). The working solution, which was prepared immediatelybefore the measurements and stored in a reservoir connectedto the liquid cell via a Teflon tube, was refreshed at 10-minintervals to ensure uniform solution conditions. A typical adhesionforce measurement with a given modified tip involved acquisitionof 1000 individual force-distance curves recorded at 20 differentlocations on the crystal face. The force-distance curves wereacquired at a rate of 2 Hz and with a loading force of 1 to2 nN. The force-distance curves did not change perceptibly foracquisition rates in the range 0.2 to 2 Hz, nor did they changewith loading forces up to 6 nN. The effects of these parametersoutside these ranges were not examined. The adhesion forceswere determined from the retraction portion of individual curvesusing customized software that automatically calculated thechange in deflection upon detachment of the tip from the crystalsurface. These were tabulated into histograms, and the meanvalues and SD were determined from the normal distribution curves.These were not appreciably different from the arithmetic meanvalues and SD. This procedure accounts for the likely possibilitythat the number of molecules adhering to the surface variesfor individual force curves, while effectively averaging contributionsfrom crystal surface defects and nonuniformities in the tip.
Atomic force microscopy employs an ultrasmall tip, usually siliconor silicon nitride, located at the end of a silicon cantilever,which is brought into contact with a sample surface using piezoelectricactuators (17,18). As the tip is scanned across the sample,the surface topography can be visualized from the vertical motionof the tip, which is deduced from the position of a laser beam,reflected off the back of the cantilever, on a position sensitivephotodiode detector. This enables acquisition of two-dimensionalimages of crystal surfaces, which can reveal features such asterraces and steps that are related to the internal structureof the crystal. AFM also is capable of producing images of thetruncated crystal lattice exposed at specific crystal faces.In another mode, the AFM can be used for direct and quantitativemeasurement of the force required to pull the AFM tip off asurface with which it is in contact. As the cantilever is retractedfrom the surface, it bends away from the surface while the tipremains in contact with the sample. The change in the deflectionof the cantilever upon detachment of the AFM tip from the surfacecan be determined from the position-sensitive photodiode detector.If the force constant of the cantilever is known, the forceof adhesion between the tip and the sample surface can be calculated.Both imaging and force measurements can be performed in a liquidmedium, making this method ideal for examining biologicallyrelevant processes. The force measurement capability has beenused for quantitative measurement of adhesion forces betweenvarious molecular and biomolecular surfaces (19,20), but onlyrecently has it been used to diagnose adhesion events at singlecrystal surfaces (21,22,23,24).
Through adjustment of crystal growth conditions, COM crystalsfor AFM adhesion force measurements were prepared with prominent(100), (12), or (010) faces, or COD crystals with prominent (100) or (101) faces (Figure 1; Miller indices,a three-digit notation used to define specific crystal planes,are indicated). These faces were sufficiently large (5 to 10microns) for positioning of the AFM tip on the surface. TheMiller indices of these crystal faces and the adjoining faceswere assigned with the aid of the crystal modeling program SHAPE(Shape software, Kingsport, TN). Topographic and lattice imagesof the COM faces, acquired with contact mode AFM in 0.11 mMCaOx solutions, indicated that their surfaces were crystalline,with structures identical to their corresponding bulk crystalplanes (13). The COD (100) face revealed large terraces separatedby 6.2 ± 0.2 Å steps, equal to 1/2a and correspondingto a single layer of calcium (Ca2+) and oxalate (Ox2)ions, as expected from the crystal structure. The COD (101)face was less defined, exhibiting ridges rather than terraceswith large areas. Although we have not yet been able to observelattice images for either COD face, the well-formed (100) and(101) facets of COD crystals suggests that, like COM, the compositionsand structures of its crystal surfaces can be deduced reliablyfrom the bulk crystal structures (Figure 2) (25,26).
Figure 1. Scanning electron micrographs of calcium oxalate monohydrate (COM) and calcium oxalate dihydrate (COD) crystals prepared by methods that are tailored to obtain select crystal faces with areas sufficient for atomic force microscopy (AFM) force measurements (Miller indices indicated in white). The schematic representation at the lower right depicts a contact between a calcium oxalate (CaOx) surface and an idealized hemispherical gold-coated AFM tip modified with a monolayer of organosulfur molecules (not drawn to scale). The radius of curvature of the tip and the gold thickness are approximately 75 and 50 nm, respectively, and the adhesion force measurements were performed in aqueous solutions saturated with CaOx.
Figure 2. Structural models of the COM and COD faces depicted in Figure 1, as generated from the reported single crystal structures. Color code: green, calcium ions; red, oxalate oxygen atoms; gray, oxalate carbon atoms. Cambridge Structural Database reference codes CALOXM03 (COM: monoclinic P21/c, a = 6.290 Å, b = 14.5803 Å, c = 10.116 Å, = 109.46°) and WHWLTB (COD: tetragonal I4/m, a = 12.371 Å, c = 7.357 Å).
In the adhesion force measurements performed here, gold-coatedAFM tips were modified with organosulfur molecules adorned withterminal carboxylate or amidinium groups, denoted Au:S(CH2)10COOand Au:S(CH2)2NHC(NH2+)NH2. These tips were chosen because theycan be viewed as mimics of urinary protein segments (e.g., carboxylatemimics aspartate, glutamate; amidinium mimics arginine) thatcould promote adhesion between crystals in aggregates and attachmentto epithelial cell membranes containing embedded proteins. Carboxylategroups on an AFM tip can also be viewed as mimics for Ox2ions on a CaOx surface. In a given experiment, the AFM tip wasbrought into contact with a particular face of a CaOx crystalimmersed in an aqueous solution of 0.11 mM CaOx. The adhesionforce was measured from the change in the cantilever deflectionwhen the AFM tip detached from the crystal surface during retraction.In this configuration, multiple molecules are contacting andseparating from the crystal surface in each measurement. Themean adhesion force was determined from numerous individualforce curves recorded at different locations on a particularcrystal face.
In agreement with our previous observations (21), the mean adhesionforces decreased in the order COM (100) > COM (12) > COM (010) for both the Au:S(CH2)10COOand Au:S(CH2)2NHC(NH2+)NH2 tips. The adhesion forces measuredwith COD crystals, using the same tips under the same conditions,revealed that the adhesion strength of COD (100) was greaterthan that of COD (101), irrespective of the tip used (Figure 3,top and middle panels). Whereas the Au:S(CH2)10COOtip binds more strongly to (COD) (100) than (COM) (12), the Au:S(CH2)2NHC(NH2+)NH2 tip exhibits the oppositeordering. Overall, however, the adhesion forces scale with theCa2+ and Ox2 surface site concentrations deduced fromthe bulk crystal planes of COM and COD, as illustrated in thebottom panel of Figure 3 (the Ca2+ and Ox2 in each crystalplane are nominally equivalent: COM (100) = 0.0542 sites/Å2> COD (100) = 0.0439 sites/Å2 > COM (12) = 0.0429 sites/Å2 > COM (010) = 0.0333 sites/Å2COD (101) = 0.0225 sites/Å2.
Figure 3. (top) Representative histograms for adhesion force measurements with the Au:S(CH2)10COO and the COD (101) and (100) faces, performed in 0.11 mM CaOx solution at pH = 7. Each histogram corresponds to 1000 individual force measurements. (middle) The mean adhesion forces measured for the Au:S(CH2)10COO and Au:S(CH2)2NHC(NH2+)NH2 tips and the prominent COM and COD faces (see Figure 1). In this typical data set, the same tip was used for all the crystal faces. The asterisks (*) denote the most prominent crystal faces of COM and COD. Replicate measurements with different tips and crystals demonstrated that these adhesion force profiles were highly reproducible. The tips did not exhibit any appreciable change in adhesion forces during the measurements that would suggest deterioration of the tip. (bottom) Dependence of the mean adhesion force on the surface concentration of Ca2+ (and Ox2 ions) for the COM and COD faces.
The comparable forces observed on each face for the oppositelycharged Au:S(CH2)10COO and Au:S(CH2)2NHC(NH2+)NH2 tipsinidentical solutionssupports equivalent surface concentrationsof Ca2+ and Ox2 ions and negligible contributions ofcoulombic interactions to the adhesion forces. This argues thatthe adhesion profiles for the Au:S(CH2)10COO tip canbe attributed to specific binding between the carboxylate groupand surface Ca2+ ions, whereas the adhesion profiles for theAu:S(CH2)2NHC(NH2+)NH2 tip can be attributed to charge-assistedhydrogen bonding between the amidinium group and the carboxylatemoiety of a surface oxalate (Figure 4). The N-HH-N distancein the amidinium group (approximately 2.25 to 2.35 Å)and the OO distance of the Ox2 carboxylate group(2.21 to 2.34 Å) are ideally matched for N-HO hydrogenbonding as a heterodimer. The correlation of the adhesion forcewith Ox2 surface concentration suggests that the orientationof the Ox2 ions on the crystal planes does not substantiallyaffect the adhesion to these tip molecules. The mean adhesionforces measured for bare gold tips or tips modified with a simplealkanethiol (Au:S[CH2]10CH3) were approximately 1 nN, irrespectiveof the crystal face. Therefore, any adhesion force above thisvalue can be attributed to specific binding. Notably, the adhesionforce for COD (101) was approximately 1 nN for both tips, suggestingonly minor contributions from specific binding.
Figure 4. Schematic representation of carboxylate and amidinium tips approaching binding sites on a CaOx crystal surface. For purposes of clarity only one molecule on each tip and one type of oxalate orientation are illustrated. The carboxylate probe docks to a surface Ca2+ ion and the amidinium tip binds to a surface Ox2 ion via charge-assisted hydrogen bonding.
Overall, these measurements reveal that the adhesion strengthof the COM and COD faces, using biologically relevant functionalgroups as a probe, decreases in the order COM (100) > COD(100) = COM (12) > COM (010) > COD (101). This ranking establishes a crucial link between the pathologicbehaviors of COM and COD and the adhesion strengths of theirrespective crystal surfaces, and supports the critical rolesof aggregation and crystal adhesion to renal epithelial cellsin stone formation. COM, the pathogenic form, exhibits large(100) faces when grown in urine-like media, and stones oftencontain stacks of COM plates emanating from a central nidusattached to renal cell surfaces. These stacks appear to formby crystal-to-crystal attachment of the large (100) faces. Thus,the most prominent COM face exhibits the largest adhesion strength,an ideal combination for creating robust COM aggregates andstrong attachments to epithelial cell membranes that can persistunder stresses experienced during flow in the renal tubules.Conversely, the protective form, COD, always exhibits large(101) faces when grown in urine-like media and in vivo, andthe exposed area of COD (100) is minimal. Thus, the more adherantface, COD (100), is nearly inaccessible for adhesion contacts.Instead, the most prominent face, COD (101), displays the weakestadhesion strength. This condition would make COD aggregatesand attachments to cell membranes less stable, thereby reducingthe tendency to form stones, as reflected by the large amountsof individual COD microcrystals found in voided urine. Therefore,these adhesion measurements reveal, at a microscopic level,how COD formation can protect against stone disease. Furthermore,the monotonic dependence of the adhesion strength of the crystalfaces with the surface concentration of binding sites suggestsa rather simple explanation for the differences between COMand COD with respect to aggregation and attachment to cells.It is interesting to note that the relative adhesion strengthsof the COM and COD crystal faces are parallel to the face-selectivebinding affinities of these crystals with stearic acid monolayers(27), phospholipid monolayers (28), epithelial cell surfacesin culture (2,29), and nephrocalcin, a urinary glycoprotein(30), which suggests that the adhesion forces and binding affinitiesboth depend on the number of accessible binding sites on thecrystal surfaces. We anticipate that further studies of thekind described here, particularly when performed in the presenceof urinary species thought to regulate stone formation, willfurther identify the key factors responsible for stone diseaseand may ultimately lead to preventative therapies.
Acknowledgments
This work was supported by the Department of Veterans Affairs,the Medical College of Wisconsin, the McKnight Foundation, andthe National Science Foundation (CTS-0323696). The work alsowas supported partially by the MRSEC program of the NationalScience Foundation (DMR-0212302). X. Sheng also acknowledgessupport through a Doctoral Dissertation Fellowship from Universityof Minnesota Graduate School.
Footnotes
Published online ahead of print. Publication date availableat www.jasn.org.
Finlayson B, Reid F: The expectation of free and fixed particles in urinary stone disease.
Invest Urol 15
: 442
448, 1978[Medline]
Kok DJ, Khan SR: Calcium oxalate nephrolithiasis, a free or fixed particle disease.
Kidney Int 46
: 847
854, 1994[Medline]
Govindaraj A, Selvam R: Increased calcium oxalate crystal nucleation and aggregation by peroxidized protein of human kidney stone matrix and renal cells.
Urol Res 29
: 194
198, 2001[Medline]
Grases F, Gil JJ, Conte A: Studies on structure of calcium oxalate monohydrate renal papillary calculi. Mechanism of formation.
Coll Surf 36
: 29
38, 1989
Christmas KG, Gower LB, Khan SR, El-Shall H: Aggregation and dispersion characteristics of calcium oxalate monohydrate: Effect of urinary species.
J Coll Interf Sci 256
: 168
174, 2002[CrossRef][Medline]
Guo S, Ward MD, Wesson JA: Direct visualization of calcium oxalate monohydrate crystallization and dissolution with atomic force microscopy and the role of polymeric additives.
Langmuir 18
: 4284
4291, 2002[CrossRef]
Cody AM, Horner HT: Crystallographic analysis of crystal images in scanning electron micrographs and their application to phytocrystalline studies.
Scanning Electron Microsc 3
: 1451
1460, 1984
Jung T, Sheng X, Choi C, Kim W Wesson JA, Ward MD: Probing crystallization of calcium oxalate monohydrate and the role of macromolecule additives with in situ atomic force microscopy.
Langmuir 20
: 8587
8596, 2004[CrossRef][Medline]
Fleming DE, Van Riessen A, Chauvet MC, Grover PK, Hunter B, Van Bronswijk W, Ryall RL: Intracrystalline proteins and urolithiasis: A synchrotron X-ray diffraction study of calcium oxalate monohydrate.
J Bone Min Res 18
: 1282
1291, 2003[Medline]
Bouropoulos N, Weiner S, Addadi L: calcium oxalate crystals in tomato and tobacco plants: Morphology and in vitro interactions of crystal-associated macromolecules.
Chem Eur J 7
: 1881
1888, 2001[CrossRef]
Cleveland JP, Manne S, Bocek D, Hansma PK: A nondestructive method for determining the spring constant of cantilevers for scanning force microscopy.
Rev Sci Instrum 64
: 403
405, 1993[CrossRef]
Binnig G, Quate CF, Gerber C: Atomic force microscope.
Phys Rev Lett 56
: 930
933, 1986[CrossRef][Medline]
Ward MD: Bulk crystals to surfaces: Combining X-ray diffraction and atomic force microscopy to probe the structure and formation of crystal interfaces.
Chem Rev 101
: 1697
1725, 2001[CrossRef][Medline]
Noy A, Vezenov DV, Lieber CM: Chemical force microscopy: Exploiting chemically-modified tips to quantify adhesion, friction, and functional group distributions in molecular assemblies.
Annu Rev Mater Sci 27
: 381
421, 1997
Takano H, Kenseth JR, Wong S, OBrien JC, Porter MD: Chemical and biochemical analysis using scanning force microscopy.
Chem Rev 99
: 2845
2890, 1999[CrossRef][Medline]
Abendan RS, Swift JA: Surface Characterization of cholesterol monohydrate single crystals by chemical force microscopy.
Langmuir 18
: 4847
4853, 2002[CrossRef]
Danesh A, Davies MC, Hinder SJ Roberts CJ, Tendler SJ, Williams PM, Wilkins MJ: Surface characterization of aspirin crystal planes by dynamic chemical force microscopy.
Anal Chem 72
: 3419
3422, 2000[Medline]
Sheng X, Wesson JA, Ward MD: Adhesion between molecules and calcium oxalate crystals: Critical interactions in kidney stone formation.
J Am Chem Soc 125
: 2854
2855, 2003[CrossRef][Medline]
Sheng X, Jung T, Wesson JA, Ward MD: Adhesion at calcium oxalate cystal surfaces and the effect of urinary constituents.
Proc Nat Acad Sci U S A 102
: 267
272, 2005[Abstract/Free Full Text]
Deganello S: The structure of whewellite, CaC2O4·H2O at 328 K.
Acta Cryst B 37
: 826
829, 1981[CrossRef]
Tazzoli V, Domeneghetti C: The crystal structures of whewellite and weddellite: Re-examination and comparison.
Am Miner 65
: 327
334, 1980[Abstract]
Ouyang J-M, Deng S-P: Controlled and uncontrolled crystallization of calcium oxalate monohydrate in the presence of citric acid.
Dalton Trans 14
: 2846
2851, 2003[CrossRef]
Whipps S, Khan SR OPalko FJ, Backov R, Talham DR: Precipitation of calcium oxalate monohydrate at phospholipid monolayers
J Cryst Growth 192
: 243
249, 1998[CrossRef]
Lieske JC, Toback FG, Deganello SF: Face-selective adhesion of calcium oxalate dihydrate crystals to renal epithelial cells.
Calcif Tissue Int 58
: 195
200, 1996[CrossRef][Medline]
Deganello S: The interaction between nephrocalcin and Tamm-Horsfall proteins with calcium oxalate dihydrate.
Scanning Microscopy 7
: 1111
1118, 1993[Medline]
This article has been cited by other articles:
J. A. Wesson and M. D. Ward Pathological Biomineralization of Kidney Stones
Elements,
December 1, 2007;
3(6):
415 - 421.
[Abstract][Full Text][PDF]