MCGILL UNIVERSITY

CALCIUM OXALATE CRYSTAL FORMATION IN HUMAN URINE AND IDENTIFICATION OF MINERAL-BINDING PROTEINS Dy QUYNH DUNG SARAH NGUYEN

A THESIS SUBMITTED TO THE FACULTY OF GRADUATE STUDIES AND RESEARCH IN PARTIAL FULFILMENT OF THE REQUIREMENTS OF THE DEGREE OF MASTER OF SCIENCE

FACULTY OF DENTISTRY

©2001

1+1

National Library of Canada

Bibliothèque nationale du Canada

Acquisitions and Bibliographie Services

Acquisitions et services bibliographiques

395 Wellington Street Ottawa ON K1A ON4 canada

395, rue Wellington Ottawa ON K1A ON4 canada Your file votre réfétrlnœ Our 61e Notre réfétrlnœ

The author bas granted a nonexclusive licence allowing the National Library of Canada to reproduce, loan, distribute or sell copies of this thesis in microform, paper or electronic formats.

L'auteur a accordé une licence non exclusive permettant à la Bibliothèque nationale du Canada de reproduire, prêter, distribuer ou vendre des copies de cette thèse sous la forme de microfiche/film, de reproduction sur papier ou sur format électronique.

The author retains ownership ofthe copyright in this thesis. Neither the thesis nor substantial extracts frOID it may be printed or otherwise reproduced without the author's pemnsslOn.

L'auteur conserve la propriété du droit d'auteur qui protège cette thèse. Ni la thèse ni des extraits substantiels de celle-ci ne doivent être imprimés ou autrement reproduits sans son autorisation.

0-612-78932-2

Canada

TABLE OF CONTENTS List of Abbreviations

.IV

List of Figures

VII

Abstract and Résumé

.IX

Introduction and Literature Review Kidney stones Kidney stone composition Kidney stone formation Crystallization Supersaturation Nucleation Growth Aggregation Calcium·oxalate crystals Kidney stone matrix Urinary proteins Osteopontin Urinary Prothrombin Fragment 1 Albumin Tamm-Horsfall Protein Urine from stone formers versus non-stone formers Urine ofmales versus females

7 9 11 12 13 14 15 17 18

Rationale and objectives ofthis research project

20

Materials and Methods The precipitation of calcium oxalate (CO) crystals from human urine Part 1: The effects ofurine manipulation Osteopontin: Further characterization Thrombin digestion Osteopontin associated with calcium oxalate crystals versus calcium phosphate crystals precipitatedfrom male urine Part II: Gender differences The precipitation of calcium oxalate (CO) crystals from rat urine SnS-PAGE of crystal associated proteins Gel staining methods Double staining: Stains-Ali / Ag nitrate Si/ver Staining Western blotting Light and fluorescence microscopy Scanning Electron Microscopy (SEM)

1 2 2 3 4 4 4

6 6

22 23 23 26 26

27 27 28 29 29 29 30 31 34 .35

II

Immunohistochemistry ofkidney stones Paraffin embedding Hematoxylin and eosin staining Immunohistochemicallocalization ofmatrix proteins Preparation ofpure inorganic calcium oxalate dihydrate (Weddelite) crystals Hydroxyapatite (HAP) beads Poly-L-Aspartic acid (poly-Asp/PA) Fluorescein isothiocyanate labeling ofpoly-Asp Inhibition of calcium oxalate dihydrate growth Competitive peptide/protein binding assays

.35 36 36 .36 37 38 .38 .38 39 .40

Results , Precipitation of calcium oxalate crystals from urine Immunohistochemical staining of calcium oxalate kidney stones for osteopontin ; Calcium oxalate crystals Hydroxyapatite beads

42 .43

Discussion Precipitation of calcium oxalate crystals from urine Immunohistochemical staining of calcium oxalate kidney stones for osteopontin The use of synthetic calcium oxalate crystals for peptide/protein-binding analysis The use ofBioRad hydroxyapatite ceramide beads for peptide/protein-binding analysis

86 87

10 1

Conclusions and summary

104

"

65 68 ,.78

96 96

References

108

Acknowledgements

118

III

LIST OF ABBREVIATIONS

oc

degrees Celsius

Abs

absorbance

Ag

silver

APS

ammonium persulfate

BSA

bovine serum albumin

Ca

calcium

CaCh

calcium cWoride

CaP

calcium phosphate

CO

calcium oxalate

COD

calcium oxalate dihydrate

COM

calcium oxalate monohydrate

COT

calcium oxalate trihydrate

CMP

crystal rnatrix protein

ddH 20

double deionized water

EDS

Electron Dispersive Spectroscopy

EDTA

ethylenediaminetetracetic acid

FACS

Fluorescence Activated Cell Sorter

FITC

fluoresceinisothiocyanate

FLM

fluorescence light microscopy

g

gram

GAG

glycosaminoglycan

Glu

glutamic acid

IV

H&E

hematoxylin and eosin

HAP

hydroxyapatite

HCL

hydrochloric aeid

HRP

horseradish peroxidase

HSA

human serum albumin

KCL

potassium chloride

kDa

kilodalton

KH2P04

potassium dihydrogen orthophosphate

1

liter

LM

light microscopy

M

moles per liter

MEB

microscopie éléctronique à balayage

mg

milligrams

mM

millimoles per liter

MW

molecular weight

NaCI

sodium chloride

Na2HP04

sodium phosphate dibasic, anhydrous

NaN3

sodium azide

NaOx

sodium oxalate

NaP

sodium phosphate

OC

oxalate de calcium

OPN

osteopontin

PBS

phosphate buffer saline

v

PMSF

phenylmethylsufanylflouride

poly-Asp/PA poly-L-aspartic acid RBC

red blood cell

RGD

arginine-glycine-aspartate (Arg-Gly-Asp)

RNA

ribonuc1eic acid

rpm

revolutions per minute

SDS

sodium dodecyl sulfate

SDS-PAGE

sodium dodecyl sulfate polyacrylamide gel electrophoresis

SEM

scanning electron microscopy

SS

supersaturation

TEMED

N,N,N',Nil -tetramethylethylenediamine

THP

Tamm-Horsfall protein

Tween20

polyoxyethylene sorbitan monolaurate

J.tg

microgram, IxlO-6

J.tl

microliter

UMM

urinary macromolecules

UPTFI

urinary prothrombin fragment 1

V

volt

w/v

weight per volume

VI

LIST OF FIGURES

Figure 1.

Mode! for the initial intratubular events in the formation and aggregation of calcium oxalate and calcium phosphate crystals leading to kidney stone formation (page 5)

Figure 2.

Important factors for the development of a calcium stone (page 7)

Figure 3.

SDS-PAGE ofmale urine samples (page 48)

Figure 4.

Western blots ofmale urine samples (page 49)

Figure 5.

SEM of crystals from male urine (page 50)

Figure 6.

SDS-PAGE offemale urine samples (page 51)

Figure 7.

SDS-PAGE ofCMP digested with thrornbin (page 52)

Figure 8.

Western blot ofCMP frorn CO and CaP crystals (page 53)

Figure 9.

SEM and X-ray rnicroanalysis of CO and CaP crystals (page 54)

Figure 10.

SDS-PAGE ofmale and female urine samples (page 55)

Figure 11.

Western blots ofmale and female urine samples (page 56)

Figure 12.

Western blots ofmale and female concentrated urine samples (page 57)

Figure 13.

SDS-PAGE ofmale and female CMP samples (page 58)

Figure 14.

Western blots ofmale and female CMP samples (page 59)

Figure 15.

SDS-PAGE ofthe supernatants obtained frorn male and female urine samples (page 60)

Figure 16.

Western blots ofthe supernatants obtained frorn male and female urine samples (page 61)

Figure 17.

SEM of crystals frorn male and female urine (page 62)

VII

Figure 18.

SDS-PAGE and Western blots ofrat urine samples (page 63)

Figure 19.

SEM of crystals from rat urine (page 64)

Figure 20.

LM ofH&E-stained kidney stone sections (page 66)

Figure 21.

LM ofkidney stone sections stained for OPN (page 67)

Figure 22.

LM ofsynthetic COD crystals (page 71)

Figure 23.

SEM of synthetic COD crystals (page 72)

Figure 24.

SDS-PAGE ofpoly-Asp (page 73)

Figure 25.

LM of COD grown with poly-Asp (page 74)

Figure 26.

SEM of COD grown with poly-Asp (page 75)

Figure 27.

LM of COD grown with HSA (page 76)

Figure 28.

SEM of COD grown with HSA (page 77)

Figure 29.

LM ofHAP beads (page 80)

Figure 30.

SEM ofHAP beads (page 81)

Figure 31.

FLM ofHAP beads and poly-Asp (page 82)

Figure 32.

FACS analysis ofHAP beads and poly-Asp (page 83)

Figure 33.

FLM ofHAP beads and HSA (page 84)

Figure 34.

FACS analysis ofHAP beads and HSA (page 85)

VIII

ABSTRACT AND RÉSUMÉ

IX

ABSTRACT Urolithiasis occurs in 20% of males and 5-10% offemales, and 75% ofkidney stones contain calcium oxalate (CO) mineraI. To analyze mineral-binding proteins and to make gender comparisons, using the model of Doyle et al. (Clin Chem, 37: 1589-1594, 1991), CO crystals were generated in whole and centrifuged urine samples and then washed with water or sodium hydroxide. Crystals and mineral-binding proteins were analyzed by SDS-PAGE, Western blotting and electron microscopy (SEM). Regardless ofurine or crystal treatment, osteopontin and UPTF 1 proteins were consistently present in the samples, whereas THP and albumin were partially removed. SEM showed larger crystals precipitated from female than from male urine. Western blotting demonstrated more albumin bound to crystals from females. In other experiments, CO crystals were grown in the presence of poly-L-aspartic acid (PA) and albumin. SEM demonstrated that these proteins affected CO crystallization. Competitive protein-binding assays and fluorescence activated cell sorter analysis after binding of PA and albumin to hydroxyapatite indicated that PA binds hydroxyapatite with a stronger affinity than albumin.

RÉSUMÉ La lithiase rénale se produit chez 20% des hommes et 5-10% des femmes et 75% des pierres rénales contiennent de l'oxalate de calcium (OC) comme minéral principal. En utilisant le modèle de Doyle et al. (Clin Chem, 37: 1589-1594, 1991) pour analyser les protéines et pour comparer les deux sexes, des cristaux OC ont été précipité de l'urine centrifugé et non- centrifugé. Les protéines liées aux cristaux ont été analysé par l'éléctrophorèse (SDS-PAGE), Western et par la microscopie électronique à balayage (MEB). Les protéines d'ostéopontine et la prothombine urinaire (fragment 1) n'ont pas été influencé, tandis que la protéine de Tamm-Horsfall et l'albumine ont été éliminé des échantillons d'urine traité. Le MEB a démontré que des plus gros cristaux ont précipité de l'urine des femmes que de l'urine des hommes, et les Westerns ont démontré qu'il y a plus d'albumine liée aux cristaux des femmes. Dans d'autres expériences, la croissance des cristaux OC en présence de l'acide Poly-L-Aspartique (PA) et de l'albumine, analysée par MEB, a démontré un effet marquant sur la cristallisation de OC par ces protéines.

Les analyses de FACS de la liaison de PA et d'albumine à l'apathite ont indiqué que la PA se lie à l'apathite avec une affinité plus importante que l'albumine.

x

INTRODUCTION AND LITERATURE REVIEW

INTRODUCTION AND LITERATURE REVIEW

Kidney stones

Kidney stone disease affects approximately 10-12% of individuals in the industrialized world [4, 91]. Clinical protocols exist to identi:fy the risk factors leading to renal stone formation, however many cases remain idiopathic, with no identifiable biochemical or anatomical markers [92]. The risk ofrecurrence is high, ranging from 6070% after 10 years, thus placing increased importance on prevention of secondary stone formation [4, 91, 93]. Increased incidence ofurolithiasis is associated with male gender (a group comprising two-thirds of stone formers), increasing age (up until the age of 65), low urine volume, hereditary factors and disorders, as weIl as other kidney disorders and geographic factors. Other fàctors which can influence the rate of stone disease are dietary intake, hypercalciuria, hyperuricosuria, hyperoxaluria, hypocitraturia, and acidosis [34]. The formation ofcalcium oxalate stones, in particular, can he caused by primary hyperthyroidism, idiopathic hypercalciuria, low urine citrate level, hyperoxaluria, or hyperuricosuria, and in the case ofcalcium phosphate stones, renal tubular acidosis [21].

Kidney stone composition

About 80% of renal stones are composed of calcium oxalate (CO) and calcium phosphate (CaP) [34,90] and about 70% ofthese stones are calcium oxalate [87]. The

2

remaining stones are composed ofuric acid or mixed urie acid and calcium (10%), struvite stones (10%), and cystine stones (l %) [34]. Approximately two-thirds ofkidney stones contain more than one type of crysta~ and CO mixed with CaP is the most common combination encountered [45]. Calcium oxalate kidney stone disease is a common clinical problem occurring at an annual rate of 1 in 1,000 people. There is as yet no completely corrective therapy for idiopathie calcium oxalate urolithiasis [69]. As most other stone types have medical conditions associated with them and only 10% of CO stones have an identifiable pathology, the remainder ofthe discussion on kidney stones will he limited to the pathophysiology of idiopathie calcium oxalate stone formers [21,34].

Kidney stone formation There are three main contributing issues relevant to kidney stone formation: supersaturation ofurine with respect to ions (such as calcium and oxalate), crystal nucleation, growth and aggregation, and the presence of inhibitors and promoters of crystallization [83]. Stone formation is a complex process, involving the formation of a crystalline (minerai) and a non-crystalline (organic) phase [34, 90]. The crystalline phase

will he further discussed in the section entitled "Crystallization" and the non-crystalline phase in the section entitled "Kidney Stone Matrix".

3

Crystallization

Supersaturation One ofthe important factors for CO crystallization is supersaturation (SS) ofthe urine with respect to calcium and oxalate. This is defined by Asplin et al. (1997) as the ratio ofthe concentration ofthe dissolved salt divided by the solubility ofthat salt in urine, at body temperature. As the SS increases, a level is attained at which a solid phase formation is possible, and this is referred to as the upper limit ofthe metastable range. Below this point, supersaturation allows the growth ofpreformed crystals but not their de

novo formation. The typical concentrations in human urine are 2-8 mM for calcium and 0.2-0.5 mM for oxalate [2]. Normal human urine is in significant excess of calcium ions with a Ca:Ox ratio in the range of6:1 to 10:1. An even wider range is seen in the urine of stone patients [86].

Nucleation The interaction between mineraI phases and macromolecules in vivo is a complex phenomenon, influenced by physiological factors such as inorganic ion concentration and pH [18]. Although human urine is generally insufficiently supersaturated with respect to calcium oxalate to induce nuc1eation of CO crystals, the dissolved calcium oxalate can form nuc1ei when its supersaturation reaches 7-11 times its solubility [21, 45]. This type of formation is known as homogeneous nuc1eation. More commonly, nuc1ei are formed on existing surfaces in a process referred to as heterogeneous nucleation. Surfaces in the kidney that can serve as sites for heterogeneous nuc1eation are epitheliallinings, cellular debris, urinary casts (such as aggregations of Tamm-Horsfall protein [THP]) and other

4

crystals. Anything that increases the rate of formation ofheterogeneous nuclei in tubular fluid or urine would lower the supersaturation at which crystals :fust form. Hyperuricosuria, which promotes the formation of CO crystallization for example, may have an effect by producing urate or uric acid seeds that could serve as sites for heterogeneous nucleation [21,37,88]. Another type ofmineral that could serve as sites of CO nucleation is CaP. Studies have shown that more than 70% of oxalate-rich stones had CaP within or near their central core [38]. Calcium phosphate crystals that remain in the nephron or in the renal collecting ducts could thereby act as possible promoters of CO nucleation. Under normal conditions in the kidney, the pH is around 6.75 in the proximal tubule and 6.45 in the distal tubule [20, 38, 45]. Calcium phosphate crystals would most likely form under these conditions (at a pH higher than 6.2), whereas CO would form in the collecting duct (at a pH between 5.0-6.2) [20, 38, 45]. Theoretically, CaP formed in the nephron could partly or completely dissolve in the collecting duct when the pH is low. This would cause an increased local concentration of calcium and thus an increased supersaturation with respect to CO, promoting CO precipitation on preformed CaP crystals [39]. Proximal tubule Loop ofHenle

Distal tubule prox part

.....

--.

•••

pH

•

••

6.45

6

caz ca2• • , Caz'

Caz, Caz,

Cah

• 55 Jlm). At 10 Jlg/ml HSA (d), the average size ofthe crystals obtained decreased, and this size change remained steady over 20 and 50 Jlg/ml HSA [panels (e) and (f)]. At a concentration of 100 Jlg/ml (g), the number of small crystals «2.5 Jlm) increased with a fraction ofthe crystals remaining at their original size. The highest protein concentration ofHSA tested was 200 Jlg/ml (h), and aIl of the crystals produced at this concentration were smaller than 2.5 Jlm. These results demonstrate that in the presence ofHSA, COD crystal growth is affected at concentrations as low as 10 Jlg/ml. At higher concentrations (>100 Jlg/ml) ofHSA, growth of the crystals appears to be secondary to nucleation of new crystals, yie1ding very numerous smaller crystals, aIl of the COD morphology. There was no apparent difference in the overall amount of

69

crystals formed in the presence ofHSA by visual assessment ofthe pellets obtained after centrifugation for the collection ofthe crystals, however, no attempt was made to quantify the number of crystals produced.

70

Figure 22 LM OF SYNTHETIC COD CRYSTALS

-

lOJ.lm

c

o

0

o

Light micrographs of inorganic COD crystals synthesized in solution by the addition of CaCh and NaOx (100: 1 Ca: Ox).

71

Figure 23

a)

SEM OF SYNTHETIC COD CRYSTALS

-

21Jl1l

b)

Scanning electron micrograph of synthetic COD crystals: a) washed with ddH20 and b) washed with 0.1 M NaOH.

72

Figure 24

SnS-PAGE OFPOLy-ASp

• 14

1

2

3

1

2

3

kDa-1 a)

b)

SnS-PAGE profile ofPoly-L-Aspartic acid stained with a) silver nitrate (20% gel) and with b) Stains-Ali (180/0 gel). Legend: lane 1: 20 Ilg Poly-L-Aspartic acid lane 2: 50 Ilg Poly-L-Aspartic acid lane 3: 100 Ilg Poly-L-Aspartic acid

73

Figure 25 LM OF COD GROWN WITH POLy-Asp 0.05 JlM

tÔDtrol (

a)

b) 0.1 JlM

c)

0.25 JlM

d) 0.5 JlM

1 JlM

e)

t)

g)

h)

Light micrographs of COD crystals grown in the presence of Poly-LAspartic acid (PA). a) no PA added (control), b) 0.05 /lM PA, c) 0.1 /lM PA, d) 0.25 /lM PA, e) 0.5 /lM PA, f) 1 /lM PA, g), h) 2 /lM PA.

74

Figure 26 SEM OF COD GROWNWITHPOLy-Asp

a)

b)

c)

.

.

d)

e)

.

..

f)

g)

.

"

h)

Scanning electron micrographs of COD crystals grown in the presence of Poly-L-Aspartic acid (PA). a) no PA added (control), b) 0.05 /lM PA, c) 0.1 /lM PA, d) 0.25 /lM PA, e) 0.5 /lM PA, t) 1 /lM PA, g), h) 2 /lM PA. 75

Figure 27 LM OF COD GROWN WITH H8A Ûntrol

IJ.lWml

()J

b)

a)

10 J.1~/ml

5 J.1Wml

c)

d) 20 J.lWml

50 J.lWml

Q

e)

200 J.l~/ml

10Q.J1Wlb.l

g)

h)

Light micrographs of COD crystals grown in the presence of albumin. a) no HSA added (control), b) 1 Ilg/ml HSA, c) 5 Ilg/ml HSA, d) 10 Ilg/ml HSA, e) 20 Ilg/ml HSA, f) 50 Ilg/ml HSA, g) 100 ug/ml HSA h) 200 Ilg/ml HSA.

76

Figure 28 SEM OF COD GROWN WITH HSA

a)

b)

c)

d) - ..

e)

t)

g)

h)

Scanning electron micrographs of COD crystals grown in the presence ofalhumin. a) no HSA added (control), b) 1 ~glm1 HSA, c) 5 ~glml HSA, d) 10 ~glm1 HSA, e) 20 ~glm1 HSA, f) 50 ~glml HSA, g) 100 ~glml HSA h) 200 ~glml HSA.

77

Results: Hydroxyapatite (HAP) beads Given the variability in size, number and shape ofthe synthetic CO crystals, and also because we were interested in exploring protein interactions with the CaP phase of kidney stones and other mineralized tissues, commercially available HAP beads were used in competitive binding assays involving FITC-Iabeled PA and protein. Although there was sorne variability in the size ofthe beads, the particulate population was significantly more homogeneous than the COD crystals. The beads were larger than COD, averaging 20 /lm in diameter. Light micrographs ofthe beads are shown in Figure 29, and scanning e1ectron micrographs ofthe same beads are depicted in Figure 30. The surface ofthe beads, as seen in Figure 30a, was porous, and the porosity extended weIl into the interior as observed after crushing ofthe beads in a microcentrifuge tube using a pipette tip (Figure 30b).

Competitive peptidelprotein-binding assays In order to study the binding characteristics ofFITC-labeled PA to HAP beads, competitive binding studies were performed in the presence of IX, 10X, 100X, 200X and 500X unlabe1ed PA and HSA. The results for each protein were analyzed by fluorescence microscopy and by FACS. Fluorescence micrographs (Figure 31) are shown in the cases where the fluorescence was detectable by our imaging system. Competition assays between labeled and unlabeled PA demonstrated that aIl of the labeled peptide could he displaced by excess (200X) unlabeled PA (Figure 32g). This

indicates that the extra FITC group bound to the labeled PA does not have a significant effect on its binding affinity to HAP. In these experiments, compared to the maximal

78

fluorescence as seen in the control samples (c), there was a graduaI decrease in the fluorescence in the presence of increasing amounts of excess unlabeled PA. The results obtained in the competition assays with HSA are shown in Figures 33 and 34. Differences in fluorescent labeling ofthe beads with FITC-Iabeled PA suggest that HSA binds to HAP with less affinity than PA. The maximum excess HSA tested (SOOX) inhibited the binding oflabeled PA to approximately the same degree as IX of unlabeled PA, representing a SOO-fold difference in binding affinity ofHSA compared to PA. There was a graduaI decrease in fluorescence accompanied by an increasing amount ofexcess HSA, indicating a slight inhibition in the binding oflabeled PA to the HAP beads by HSA. From the results obtained here, it is unlikely that any amount of excess HSA would completely inhibit the binding of PA to HAP.

79

Figure 29

LM OF HAP BEADS

o

- - 20/.lm

Light micrographs of BioRad hydroxyapatite ceramide type II beads.

80

Figure 30

SEM OF HAP BEADS

a) -

b)

lOpm

_

- - 5JU1l

2JU1l

Scanning electron micrographs of BioRad hydroxyapatite beads: a) whole beads, b) crushed beads demonstrating the porosity of the surface of the beads.

81

Figure 31

FLM OF HAP BEADS AND POLy-ASp

b)

a)

o

o 1)

control

0

("'j

IX PA

2)

Fluorescence light micrographs of BioRad HAP beads: competition assay using FITC-Iabeled Poly-L-Asp and unlabeled Poly-L-Asp. Panell: FITC-Iabeled PA binding to beads alone, panel 2: IX competition with unlabeled peptide. a) fluorescence, b) bright field.

82

Figure 32

FACS ANALYSISOF HAP BEADS AND POLy-ASp

OC! N

......

b)

a)

e)

d)

c)

~ (I.J

ë Q;l > w

~, 10 4 FL1-H

Histogram of FACS analysis: Fluorescence obtained after the competitive binding of FITC-labeled and unlabeled Poly-L-Aspartic acid to BioRad HAP ceramide beads. The y-axis depicts the number of events that have corresponding relative fluorescence energy on the xaxis. Legend: a) autof1uorescence (HAP beads alone) b) IX Poly-L-Asp c) FITC-Iabe1ed Poly-L-Asp d) IX Poly-L-Asp and FITC labe1ed-Poly-L-Asp e) IOX Poly-L-Asp and FITC labeled-Poly-L-Asp f) IOOX Poly-L-Asp and FITC labeled-Poly-L-Asp g) 200X Poly-L-Asp and FITC labeled-Poly-L-Asp h) 500X Poly-L-Asp and FITC labeled Poly-L-Asp

83

Figure 33

FLM OF HAP BEADS AND HSA a)

b) control

1) lXHSA

2) lOXHSA

3) lOOXHSA

4)

Fluorescence light micrographs of BioRad HAP beads: competition assay using USA and FITC-labeled Poly-L-Asp. Panell: FITC-Iabe1ed PA binding to beads alone, panel 2: FITC-Iabeled PA plus IX HSA competition, panel 3: FITC-Iabeled PA plus IOX HSA, panel 4: FITC-Iabeled PA plus IOOX competition with HSA. a) fluorescence, b) bright field

84

Figure 34

F ACS ANALYSIS OF HAP BEADS AND HSA

co N

..... a)

10 4 FL1-H

Histogram of FACS analysis: Fluorescence obtained aCter the competitive binding of FITC-Iabeled Poly-L-Aspartic acid and albumin to BioRad HAP ceramide beads. The y-axis depicts the number of events that have corresponding relative fluorescence energy on the xaxis. Legend: a) autofluorescence (HAP ceramide beads) b) IXHSA c) FITC labeled Poly-L-Asp d) IX HSA and FITC labeled-Poly-L-Asp e) IOX HSA and FITC labeled-Poly-L-Asp f) lOOX HSA and FITC labeled-Poly-L-Asp g) 200X HSA and FITC labeled-Poly-L-Asp h) 500X HSA and FITC labeled-Poly-L-Asp

DISCUSSION

DISCUSSION

The precipitation of calcium oxalate crystals (CO) from human urine The study ofthe numerous factors known to contribute to idiopathic kidney stone disease remains a perplexing one. Although great progress has been made in recent years to identify possible pathogenic components which could play a role in urolithiasis, the multifactorial nature ofthis disease suggests that a combination of factors are likely responsible. Physiological, internaI factors such as urine chemistry and type and abundance of urinary proteins, and in particular the proteins osteopontin, urinary prothrombin fragment 1, albumin and Tamm-HorsfaIl protein, have aIl received much attention in the field as candidates for regulating urolithiasis. Other factors that likely contribute to kidney stone disease are fluid intake and diet, which are often required to be modified in stone-formers after diagnosis ofthe disease in order to alter the urinary concentrations of certain molecules known to affect crystallization [22]. The presence and concentrations of ions are major determinants of nephrolithiasis. As human urine is supersaturated with respect to calcium and oxalate, these ions combine to produce a near insoluble salt in the form of crystals that are destined, in the non-stone former, to he excreted [63]. There are approximately 7,200 crystals present per milliliter ofurine in the non-stone former, which adds up to a daily 7

excretion of lxl0 crystals [52]. Not only do these ions contribute to the actual formation of crystals, but it has been shown that oxalate Can cause proliferation of, and or injury to, renal epithelial ceIls [48, 52]. This would likely result in an increase in crystal retention

87

and aggregation, as exposed regions of damaged cell membranes would serve as sites for crystal attachment [52]. Factors present in urine that serve as inhibitors ofthese crystallization processes likely regulate crystal growth during the process of stone formation. Macromolecules that have been identified as inhibitors of CO crystallization include OPN, UPTF1, HSA, THP, nephrocaIcin, uronic acid-rich protein, GAGs and lithostathine [88], and we speculate that these molecules are either absent or aberrant in the urine of stone formers. The physiological response to mineraIs in the body is to coat with protein, which usually is believed to act in an inhibitory manner [61]. Most biological fluid systems are complex and involve multiple organic components, which act not only by themselves as monomer molecules, but often have the propensity to form large macromolecular assemblies that may have similar or different functions. Macromolecules can act to inhibit crystallization in numerous ways. Initially, these proteins could prevent stone development by binding ions and by forming small mineraI nuclei. Although the promotion of nucleation as a means of decreasing stone formation seems counterintuitive, the binding of calcium and oxalate by proteins would lower their relative urinary supersaturation, thus decreasing their availability for the growth ofpreformed larger nuclei [17]. In the latter case, the deposition ofnew mineraI on pre-existing nuclei would increase their size, thus increasing the potential to remain in the renal tubule. In the non-pathogenic scenario, the newly formed nuclei would be washed away in the tubular fluid for eventuaI excretion in the urine. In the case of stone formers, however, these nuclei remain in the kidney, serving as potential sites for new mineraI deposition and/or aggregation of other crystals. Inhibitory proteins could

88

potentially act at this stage by coating the crystal surfaces, thus decreasing the sites available for aggregation or cell attachment. If the cell surface anions responsible for mineraI binding, such as sialic acid, GAG-containing proteins and groups of anionic amino acids (such as glutamic and aspartic acid) succeed in anchoring the crystal to the renal epithelium, two outcomes are possible: 1) the crystals remain on the cell surface where they can potentially initiate stone formation, or 2) they are internalized via membrane-lined vacuoles ofthe phagocytic/endosomal system to be dissolved within the cell or carried into the interstitium [57]. The precise factors regulating these processes are currently unknown, however, it is clear that proteins play a definite role in the mechanisms leading to, or preventing, kidney stone formation. The characterization of proteins regulating crystallization may be key to understanding the pathogenic mechanisms leading to stone formation. In this study, we confirmed the association ofOPN, UPTF1, HSA and THP with CO crystals precipitated from the urine of non-stone formers. Particular interest was centered on identifYing difIerences between male and female non-stone formers, in the hopes of discerning the potential factor(s) leading to the higher incidence ofkidney stones in males. Although several other reports have compared genders [3, 7, 14], this is the first study to do so in a multiple, yet individual, manner, identifYing OPN, UPTF1, HSA and THP in the urine, and crystal matrix, as well as looking at the morphology ofthe crystals obtained from the same subjects. The role ofOPN and HSA in the mechanisms involved in crystallization was further elucidated, confirming their importance in the modulation of crystal growth. This study also provided insight into the mineral-binding behavior ofthese two proteins. Prior to proceeding with the study, however, it was essential to establish the particular

89

effects that urine handIing/processing has on urinary proteins. A priori, it was decided that the particular urine processing conditions that had the least affect on the proteins thought to be involved in urolithiasis would therefore he used throughout the remainder of this study.

The effects of urine manipulation In the past decade, there has been much controversy over the effects ofurine manipulation on urinary proteins [5, 47, 47, 60]. Factors such as the centrifugation of urine as weIl as the washing conditions of the crystals precipitated from urine yielded inconsistent results in the literature; therefore, we sought to determine the effects ofthese manipulations in a controlled and reproducible manner. This is the first systematic comparison ofthe different urine preparation conditions on multiple samples from both genders. The urine samples from six male non-stone formers were halved and then centrifuged, or left as whole samples. Four banding regions were identified in the urine at 95, 67, 40 and 18 kDa in aIl the samples processed, and only the 18 kDa region was unaccounted for with the antibodies used in this study. Due to the similar electrophoretic mobilities, we speculate that this band likely represents nephrocalcin [18,88], however, further studies would he required in order to verify this statement. The remaining protein bands are similar to those found previously in human urine [3, 5, 26, 60, 64], confirming the abundance ofHSA and THP, and the much lower levels ofOPN and UPTFI in the urine. The effects ofurine centrifugation are to remove most of the THP as weIl as partially removing HSA from the urine. The removal ofthese proteins by centrifugation

90

is most likely due to the potential formation of aggregated forms, which would tend to make them settle due to their higher molecular weight. Centrifugation does not have a significant effect on crystal-bound proteins, as the banding pattern of CMP obtained from uncentrifuged urine is identical to those from centrifuged urine. The CMP samples obtained from whole urine OOd an electrophoretic banding pattern in three regions, between 67-43 kDa, 31 kDa and 18 kDa, similar to those found previously for CMPs using similar methods and obtained from human urine [3, 5, 26, 60, 64]. By Western blotting, OPN and UPTFI were found to be much more abundant than HSA and THP, although these proteins were nevertheless found to be part ofthe matrix. Unlike previous studies [3, 26, 79] showing a greater binding ofUPTFl than OPN to the crystals precipitated from urine, the results obtained in this study did not confirm those findings; we detected OPN at levels similar to UPTFI in aIl our samples. Although the levels of HSA and THP detected in the matrix of crystals were slightly variable, these two proteins are consistent components ofthe crystal matrix in the majority ofthe samples used in this study. This suggests that the lack ofthese proteins in the matrices of crystals in other studies [3, 26, 64, 65] is perOOps attributable to differences in the levels ofthese proteins on an individual basis, or perhaps is attributable to the method ofurine processing used in those studies. Washing of CO crystals with NaOH instead of ddH2 0 removed the crystal-bound THP and most of the HSA; however, no significant effect on the crystal-bound OPN and UPTFI was observed. It is has been noted tOOt centrifugation and filtration ofthe urine prior to processing causes an almost complete loss of Tamm-Horsfall protein and sorne reduction in the concentration of albumin and other urinary proteins. As these are the

91

two most abundant proteins found in urine and they are removed by NaOH washing, it has been speculated that these proteins are non-selectively bound to the crystals [64]. In our work, both types of CO crystals, COM and COD, were seen in all four conditions of crystal preparation. There was no difference in size or morphology of the crystals obtained in our samples, suggesting that there is little or no effect of centrifugation or washing conditions on the crystals precipitated from urine. To verny the effects ofurine storage conditions on urinary proteins, the effect of the temperature at which urine samples are collected and stored as well as the effects of protease inhibitors were compared to freshly voided samples. No significant differences were detected in any ofthe five-day-old urine samples compared to a freshly voided sample from the same female control subject. The storage of urine at room temperature versus at 4°C did not appear to have an effect on the proteins, nor did storage temperature with or without added protease inhibitors. This indicated that the urinary proteins remain unaffected by temperature within the time frame and by the detection methods used in this study, although further confirmation by Western blotting with the antibodies used in this study would be required.

Osteopontin: Further characterization Thrombin digestion Osteopontin, a protein accumulating in significant amounts with the precipitation of calcium oxalate in human urine, is susceptible to proteolytic cleavage by thrombin - a physiological processing step shown to be ofphysiological relevance in terms of cell adhesion [30], but ofunknown significance in terms ofits effects on crystallization.

92

Importantly, the cleaved forms of OPN can escape detection by antibodies possibly due to epitopes at the cleavage site or an alteration in conformation ofthe OPN molecule itselfby proteolytic cleavage; the precise function ofthese processed/cleaved forms has yet to be determined [51]. To verify that the bands with electrophoretic mobilities around 67 kDa seen in CMP preparations were indeed OPN and not HSA, HSA, a CMP sample, as weIl as urinary OPN were aIl treated with thrombin. After digestion with thrombin, HSA remained at 67 kDa whereas in the CMP sample, the bands found between 43-67 kDa are replaced by bands at 31 kDa indicating the cleavage of the molecule as seen in the urinary OPN sample. A faint band present at 67 kDa in the thrombin-digested CMP sample either would suggest the presence of albumin or undigested OPN. This indicates that the majority ofthe bands found in the CMP preparation represent the intact form of OPN, thus confirming its abundance in the matrix, while the remaining bands belong to HSA, thus confirming the presence ofHSA in CO crystals.

Osteopontin in calcium oxalate crystals versus calcium phosphate crystals The results obtained by Western blotting using anti-OPN antibodies suggest that the OPN isoforms are equally bound/incorporated into the CO crystals, whereas CaP crystals preferentially bind the higher and lower molecular-weight forms ofOPN. The phosphorylation ofthis protein has been previously shown to have an effect on its inhibitory activity for crystal growth in both CO and CaP [88], however, the specifie effects ofpost-translational modifications on mineral binding are not yet known. Since most ofthe urinary isoforms ofOPN are found in the so-called crystal matrix, with all of them binding to the mineral phase, we believe that at least a portion of the binding sites

93

involved in crystal-protein interactions are likely related to the primary amino acid sequence of the molecule.

Gender differences In this study, a significant amount of intra-gender variability in the patterns was observed, though there were no significant qualitative inter-gender differences with respect to urine. Similar patterns for OPN, UPTFI and THP were noted in the male and female samples, though intra-gender variability appeared to he present for HSA orny. The presence ofOPN in the urine was confirmed in samples concentrated ten-fold, demonstrating a large variability between individuals for this protein as detected by the antibodies used. Overall, the intra-gender variability was greater in the patterns observed in the female samples than in the male samples. There are qualitatively more proteins bound to the CMP precipitated from female urines than male urines with an increased crystal binding ofHSA and THP in females, once again, confirming their existence in the matrix of crystals precipitated from human urine. The presence ofthese proteins in the supernatant, however, demonstrates that HSA and THP are not completely bound to the crystals, suggesting the non-selective binding ofthese proteins to CO crystals. Numerous different forms ofHSA were identified in the crystals obtained from female urine that were absent from those obtained from male urine. This may be one of the key differences hetween the genders. We suspect tOOt albumin modulates the preferential formation of COD over COM crystals, which would lead to the decreased formation ofaggregates hecause of the greater repulsive charges between COD crystals

94

[14]. Further studies are required, however, to determine the specific role that albumin plays in the greater inhibitory capacity of female urine on stone formation. Although the formation of COD over COM crystals would he beneficial, this would be counteracted by the fact that larger crystals were identified in samples obtained from female urine. The tendency for females to form larger crystals has heen documented [14], however, we suspect that this is due to factors other than HSA, as supported by previous [14] as weIl as CUITent evidence (in this study) that HSA acts to promote nucleation leading to crystallization of smaller particles.

Precipitation of co crystals from rat urine In looking at another mammalian species for comparative purposes, the samples precipitated from rat urine, compared to human urine, appeared to contain qualitatively more protein, and in particular, low molecular-weight proteins. From the results obtained in this study, there did not appear to he many proteins incorporated into the crystals precipitated from rat urine. Using the antibodies against the anti-N-, and anti-C-terminal ends ofOPN, bands were detected in the rat urine samples hetween 43-67 kDa and only a very faint band, if any at aIl, at 43 kDa in the CMP preparation. Crystals precipitated from rat urine were in the form ofhoth COM and COD crystals, with the latter being rather "steIlate" in form. This crystal morphology was not typically seen in samples precipitated from human urine, though the crystals were approximately the same overall size as those seen in human female samples. The difference in crystal morphology observed is most likely due to the difference in the CMP bound to human crystals versus those in rats, thus influencing crystal structure.

95

Immunohistochemical staining of calcium oxalate kidney stones for osteopontin These results, obtained by hematoxylin and eosin staining of the calcium. oxalate stones, demonstrated the layered arrangement of matrix in calcium. oxalate stones. The concentric lamellar immunohistochemical staining pattern ofOPN confirms that the protein is most like1y deposited in layers with the mineraI phase ofthe stones. Both the concentric lamellae ofmatrix and the interlamellar substances (radial striations) label intensely for OPN, and these findings are consistent with previous studies on kidney stones [44, 61].

The use of synthetic calcium oxalate crystals for peptide/protein-binding analysis Growth ofinorganic COD crysta/s Calcium. oxalate dihydrate crystals grown in the absence of organic material yielded crystals ranging in size from 1-15 Jlm, as typically observed in previous studies [86]. The shape ofthe crystals was variable, with the majority ofthe crystals being ofthe typical bipyramidal morphology characteristic of COD, and the remaining crystals having rounded edges or anvil-shaped morphologies characteristic of COM. Although it had previously been shown that crystal-washing conditions have an effect on the crystal surface by the removal ofprotein as small etched pits and cavities at the crystal surfaces [58], SEM analysis of crystals obtained from urine washed with NaOH in this study, however, did not show any evidence ofsuch protein removal. Surfaces of the crystals obtained were smooth, even after the same treatment with 0.1 M NaOH used by Ryall et

96

al. [58]. To determine whether this washing condition directly affected the mineraI phase ofthe crystal surface, synthetic CO crystals grown in solution devoid ofproteins were washed using the same concentration ofNaOH. We were unable to reproduce the etching and pitting observed on the surface ofthe crystals grown in the presence or absence ofprotein, however, as there was no effect observed on the surface ofthe crystals grown inorganically, this suggests that this particular washing condition does not directly affect the mineraI phase ofthe crystal.

Crystal growth assay in the presence ofprotein As other studies generally utilized systems that yielded COM crystals, we opted to use one that would preferentially yield COD crystals [1, 14], giving crystals ofwelldefined geometries and clearly identifiable crystallographic faces for future imaging studies on the interactions ofproteins with crystal surfaces. To begin these studies, crystals were grown in the presence ofthe peptide PA, a well-characterlzed modulator of crystal growth that represents an amino acid domain common to many acidic mineralized tissue proteins, to determine the effects ofthis peptide on crystal growth in our system. In comparison to the crystals grown without added peptide, low concentrations of PA had little or no effect on crystal growth, yielding crystals with approximately the same size or smaller than the control crystals. Differences were observed starting at a concentration of 0.1 JlM PA and greater, where a more elongated version ofthe COD crystal were seen, unlike in the control samples. This particular crystal shape was the only type formed at 0.25 JlM PA, with the majority ofthem being smaller than 5 Jlm. At increasing concentrations ofPA, the crystals formed were more elongated, the width being smaller

97

towards the center than at the lengths ofthe crystals, distinguishing them from the typical plate-like COM crystals. As these crystals were grown in a controlled, chemically defined in vitro system, it is appropriate to assume that these are CO crystals. Although this particular shape ofcrystal has not yet been reported in the literature, a similar, yet less elongated dumbbell morphology has been referred to by McKee et al. [61], to describe crystal ghosts representing the organic component associated with the mineraI phase of small crystalline particles as weIl as larger kidney stones. Such crystal forms imply inhibition of crystal growth in the central regions ofthe elongated forms, presumably from the direct inhibitory binding of PA at these sites. The particular significance ofthis crystal morphology is unknown, however it can be speculated that such rounded edges would he beneficial in decreasing intratubular crystal-attachment by reducing renal epithelial cell damage potentially caused by the rough edges of a typical bipyramidal-shaped COD crystal. Crystal-crystal associations, however, would most likely remain the same, as the rounded ends of the crystals would have the same propensity to interlock molecularly with the central regions of an adjacent crystal of the same morphology, thus joining and stacking them in the same manner as in the original crystal forms. By visual inspection, the size ofthe crystal-containing pellet obtained after centrifuging the crystals during collection was markedly reduced by the addition of the higher concentrations of PA, suggesting that it acts not only on crystal growth but also to inhibit the in vitro nucleation of CO crystals. To test the effects of a urinary protein on crystal growth in our system, HSA was used, and at relatively low concentrations ofHSA, the crystals produced were as large or

98

even larger than the control crystals (>5 fJm). This would he expected, as HSA was previously shown to he a promoter of the nucleation of CO crystals [17]; H8A binding to the mineraI would increase the relative size ofthe crystal compared to the control crystaIs grown without added protein. At 10 fJg/ml ofH8A, the average size of the crystaIs obtained decreased, and this size change remained steady over 20 and 50 fJg/ml of added HSA. At a concentration of 100 fJg/ml, the number ofsmall crystals (100 fJg/ml) ofHSA, growth ofthe crystals appears to he secondary to nucleation of new crystals, yielding very numerous smaller crystals, all of COD morphology. There was no apparent difference in the overall amount of crystaIs formed in the presence ofHSA by visual assessment ofthe pellets obtained after centrifugation for the collection of the crystals. This is expected, as increasing concentrations of added HSA yielded smaller yet more numerous crystals, whereas lower concentrations of added HSA yielded larger less numerous crystals, thus producing the same overall particulate volume. The precise quantification ofthe numher and size ofcrystals produced was not possible due to constraints ofthe instruments available to us. The results obtained in this study on the effects ofH8A on crystal structure are consistent with previous findings [17]. The increased formation of COD crystals over COM crystals in the presence ofH8A is attributed not to the chemical transformation of COM crystals into COD, but to the inhibition of COM crystallization leading to a

99

relatively higher frequency of COD crystals [17]. As previously mentioned, the formation ofsmaller crystals in the presence ofH8A is significant in decreasing the rate of stone formation, as smaller crystals would be less prone to occluding the tubular lumen. Even ifpolymerized forms ofH8A have been shown to promote aggregation [17], this would be counterbalanced by the formation of smaller crystals. The paradoxical nature ofthe action ofthis protein, as well as others, on crystallization underlines the notion that not one, but many factors acting in concert, are responsible for determining whether a single crystal will develop into a stone. ProteinmineraI interactions have been previously studied [11, 17,42,53,62,86], demonstrating several possible mechanisms for the regulation of crystallization by proteins. Many of these protein-mineral interactions are higWy complex, involving specific groups on the proteins, such as the carboxylate groups on OPN [42], as well as the secondary structure ofthe protein, such as the repetitive homologous helical domains ofH8A [17]. A protein with the required repetitive structure could act as a nucleator by providing a template for the appropriate positioning of ions to form the :tirst crystallattice. Other proteins could then act on crystallization by favoring new mineral deposition on one crystal plane over another, or by blocking the sites ofnew mineraI deposition altogether. This particular inhibition could result in a decreased rate ofgrowth on crystal surfaces that would normally grow rapidly, conversely promoting growth on surfaces that would not usually grow - thus modulating crystal structure. Aggregation and retention of crystals are potentially affected in the same manner; the sites for crystal-cell or crystal-crystal associations would be exposed or blocked depending on the nature ofthe macromolecules present in the surrounding fluid.

100

The use ofBioRad hydroxyapatite (HAP) ceramide beads for peptide/proteinbinding analysis As calcium phosphate is predicted to play an important role as a nuc1eator of calcium oxalate, and it is conveniently conunercially available in well-characterized forms, it was used to simulate the CaP mineraI component in protein-binding assays. The homogeneity of the bead population was also an advantage compared to the synthetic COD crystals, as the size and shape ofthe latter crystals produced in this study were highly variable. The HAP beads are larger than the COD crystals that we prepared, averaging 20 Ilm in diameter, and demonstrated significant porosity. Thus while having a smaller surface area per unit weight, the high level of porosity permitted significant additional binding of our reagents used in the protein-binding studies - a feature advantageous for fluorescent imaging ofthe beads and for quantitative studies (see below).

Competitive protein-binding assays In order to study the binding characteristics ofFITC-labeled Poly-L-Aspartic acid (PA) on HAP beads, competitive binding studies were performed in the presence of IX, 10X, 100X, 200X and 500X unlabeled PA and albumin (HSA). The results for each protein were analyzed by fluorescence light microscopy and by FACS. Competition assays between labeled and unlabeled PA demonstrated that virtually all ofthe labeled peptide could be displaced by excess (200X) unlabeled PA. This verified that the addition of an FITC group bound to the PA does not have a significant effect on its

101

binding affinity to HAP, as previously shown [33]. There was a graduaI decrease in the fluorescence compared to the maximal fluorescence as seen in the control samples by the different amounts of excess PA added, again confirming the findings of previous studies of PA binding to HAP [33], and showing that fluorescently labeled PA can be used to model the interactions ofthis primary amino acid sequence (as is found in OPN) with mineraI. In terros ofthe specificity ofbinding, the results ofthis study show tOOt the binding of PA to HAP is highly specific, since very high concentrations (SOO-foId) of HSA were required to inhibit FITC-Iabeled PA binding to the same degree as I-fold unlabeled PA. There was a graduaI decrease in fluorescence ofthe beads with increasing amounts ofHSA, indicating a slight inhibition in the binding oflabeled PA to the HAP beads, likely representing nonspecific interactions ofFITC-labeled PA with the beads. It is unlikely that anyamount of excess HSA would compIete1y inhibit the binding of PA to HAP indicating tOOt aithough HSA has been shown to have sorne eifects on crystallization, its ability to bind to CaP is much less than that ofPA. The results from this and other studies point to HSA OOving roles in urine more important than binding to minerai surfaces.

Future studies It wouid be informative to repeat the first part ofthis study using samples from stone patients in order to assess the differences if any, between urine samples obtained from stone-formers and the results observed in this study for the urine of non-stone formers. Further studies characterizing the urinary albumin from males and females may

102

also elucidate if the differences between the genders is in part due to albumin. With respect to protein-crystal interactions, detailed studies of CaP crystals should result in a better understanding ofthe mineral-binding properties ofOPN, as well as UPTFl, HSA and THP. Although our data is consistent with the PA region of OPN playing an important role in binding and modulating calcium-based minerais in urine, additional studies are required using intact, whole OPN in this and other crystal growth mode! systems. It would also likely be informative to perform these experiments using UPTFI and THP. Using a different system ofinorganic COD production might yield a crystal population stable and homogeneous enough to repeat the protein-binding experiments in order to compare binding capacities ofboth types ofminerals studied here.

103

CONCLUSIONS AND SUMMARY

CONCLUSIONS AND SUMMARY

The effect ofurine manipulation, such as centrifugation, was verified by snsPAGE, Western blotting and SEM of the crystals precipitated from these urines. Centrifugation has an effect on Tamm-Horsfall protein and albumin, as these proteins were found in the pellet remaining after centrifugation. The washing conditions of crystals obtained from human urine (by NaOH) had a significant effect on proteins that are non-selectively bound to the crystals precipitated. We were able to identify these proteins as THP and HSA. There was no effect ofthese urinary manipulations on the types of crystals obtained from the urines, as both COM and COD crystals were found in samples from uncentrifuged as weIl as centrifuged urines washed in both ddH20 and NaOH. From the results obtained in this study, the incorporation ofproteins into crystals precipitated from urine is a selective process. OPN and UPTFl, which are abundantly bound to the crystals, are usually found in urine at very low concentrations, whereas two ofthe most abundant proteins in urine, THP and HSA are only found in scant amounts in the crystals. Ifthe inclusion process were non-selective, one would expect the opposite tendency to he true. Conditions ofurine collection as weIl as storage temperatures do not appear to have effect on urinary proteins. The addition ofprotease inhibitors does not have an added effect on the proteins, indicating that there was no protein degradation due to proteases in the urine. This was tested in female urine, known to he high in protease

105

activity. Further confirmation, by Western blotting, using the antibodies studied here are required in order to confirm these findings. As both OPN and HSA have electrophoretic mobilities of approximately 67 kDa, the identification of OPN bands in the CMP was achieved by treatment with thrombin, as OPN contains a thrombin c1eavage site that splits the molecule almost into equal halves. This demonstrated that the majority ofthe intensity ofthe bands in this region was OPN, as only a very faint band remained at 67 kDa after treatment with thrombin, indicating the presence ofHSA. Studies in precipitated CaP crystals demonstrated a different incorporation pattern ofOPN into this crystal type. By Western blotting, we showed that the different isoforms of OPN in urine are incorporated equa11y into CO crystals. Different isoforms OPN are preferentially bound to CaP, suggesting that post-translational modifications ofthe protein are important in the binding to this type of crystal. With respect to the differences between genders, there are no significant differences between the urines from male and female control subjects, however, the crystals precipitated from female urine contain qualitative1y more protein than those precipitated from male urine. There seems to be more albumin as well as THP bound to crystals from female urine. By SEM, crystals precipitated from female urine were, on average, larger than those precipitated from male urine. Both genders produced COM and COD crystals. Rat urine is abundant in proteins as seen by SDS-PAGE. The OPN bound to the crystals precipitated from rat urine contain the same bands between 43-67 kDa as found in human samples. The crystals are COM and COD and are approximately the same size

106

as those precipitated from human urine. The COD crystals are consistently of the stellate-shaped morphology. There is no difference in the shape ofthe COM crystals compared to human COM crystals. Staining ofkidney stone sections obtained from two female patients confirmed the findings of other studies that showed that the mineral is deposited in lamellae between layers ofmatrix. The presence ofOPN was also confirmed in these layers. Inorganic COD have the same morphology as COD precipitated from urine as shown by SEM. Poly-Asp decreases crystal production and appears to favor the formation of a modified form of COD in a calcium to oxalate ratio which would normally produce bipyramidal-shaped COD crystals. The significance ofthis finding is not yet known. Albumin appears to have an effect on the nucleation of COD crystals, producing numerous small COD crystals. This finding is important because COD is the form of crystals found most commonly in non-stone formers. Protein binding to HAP beads was performed using PA and H8A. The binding of FITC-Iabeled PA was inhibited by excess unlabeled PA but not so efficiently by H8A. This finding indicates that PA binds to HAP to a greater degree than H8A. In summary, the results obtained here demonstrate that the model used to characterize urine in this study is a valid one, producing consistent findings. We have shown that the process of incorporation ofproteins into calcium oxalate is a selective one and that differences exist between the crystals precipitated from the urines ofmales and females. Growth of calcium oxalate crystals in the presence of peptides/proteins indicates that proteins influence crystal structure. Protein-mineral binding studies demonstrated that different proteins bind to mineraI with different affinities.

107

REFERENCES

REFERENCES 1. Asplin JR, Arsenault D, Parks JH, Coe FL, Hoyer JR: Contribution ofhuman uropontin to inhibition ofcalcium oxalate crystal1ization. Kidney Int. 53:194-199, 1998 2. Asplin JR, Parks JH, Coe FL: Dependence ofupper limit ofmetastability on supersaturation in nephrolithiasis. Kidney Int. 52:1602-1608, 1997 3. Atmani F, G1enton PA, Khan SR: Identification ofproteins extracted from calcium oxalate and calcium phosphate crysta1s induced in the urine ofhea1thy and stone forming subjects. Urol.Res. 26:201-207, 1998 4. Atmani F, Khan SR: Effects of an extract from Herniaria hirsuta on calcium oxalate crystallization in vitro. RJ.U. Int. 85:621-625,2000 5. Atmani F, Opalko FJ, Khan SR: Association ofurinary macromo1ecu1es with calcium oxalate crysta1s induced in vitro in normal human and rat urine. Urol.Res. 24:45-50, 1996 6. Bartho10mew RS, Rebell0 PF: Ca1ciiun oxalate crysta1s in the aqueous. Am.J.Ophthalmol. 88:1026-1028, 1979 7. Bautista DS, Denstedt J, Chambers AF, Harris JF: Low-mo1ecu1ar-weight variants of osteopontin generated by serine proteinases in urine of patients with kidney stones. ICell Biochem. 61:402-409,1996 8. Bautista DS, Xuan JW, Hota C, Chambers AF, Harris JF: Inhibition of Arg-GlyAsp (RGD)-mediated cell adhesion to osteopontin by a monoclonal antibody against osteopontin. IBiol.Chem. 269:23280-23285, 1994 9. Bezeaud A, Guillin MC: Quantitation of prothrombin activation products in human urine. Br.J.Haematol. 58:597-606, 1984 10. Blum H, Beier H, Gross HJ: Improved silver-staining ofplant proteins, RNA and DNA in polyacrylamide gels. Electrophoresis 8:93-99, 1987 Il. Boskey AL, Maresca M, U11rlch W, Doty SB, Butler WT, Prince CW: Osteopontin-hydroxyapatite interactions in vitro: inhibition ofhydroxyapatite formation and growth in a ge1atin-gel. Bone Miner. 22:147-59, 1993

109

12. Boyce WH: Sorne observations on the ultrastructure of "idiopathie" human renal calculi., in Urolithiasis, Physical Aspects., Washington, D.C., National Academy of Science, 1972, pp 97-114 13. Brown LF, Berse B, Van de Water L, Papadopoulos-Sergiou A, Perruzzi CA, Manseau EJ, Dvorak. HF, Senger DR: Expression and distribution of osteopontin in human tissues: widespread association with luminal epithelial surfaces. Mol. Biol. Cell 3: 1169-80, 1992 14. Buchholz NP, Kim DS, Grover PK, Ryall RL: Calcium oxalate crystallization in urine ofhealthy men and women: a comparative study. Scanning Microsc. 10:435-442, 1996 15. Burnette WN: Western blotting: electrophoretic transfer ofproteins from sodiumdodecyl sulfate-polyacrilamide gels to unmodified nitrocellulose and radiographie detection with antibody and radioiodinated protein A. Anal.Biochem. 112:195203, 1981 16. Burns JR, Finlayson B: Changes in calcium oxalate crystal, morphology as a function of concentration. Invest. Urol. 18: 174-177, 1980 17. Cerini C, Geider S, Dussol B, Hennequin C, Daudon M, Veesler S, Nitsche S, Boistelle R, Berthezene P, Dupuy P, Vazi A, Berland Y, Dagorn JC, Verdier JM: Nuc1eation of calcium oxalate crystals by albumin: involvement in the prevention ofstone formation. Kidney Int. 55:1776-1786, 1999 18. Chang LC, Lin HS, Chen WC: The reappraisal of nephrocalcin--its role in the inhibition ofcalcium oxalate crystal growth and interaction with divalent metal ions. Urol.Res. 29:89-93,2001 19. Clark RH, Campbell AA, Klumb LA, Long CJ, Stayton PS: Protein e1ectrostatic surface distribution can determine whether calcium oxalate crystal growth is promoted or inhibited. Calcif.Tissue Int. 64:516-521, 1999 20. Coe F, Parks JH: Defenses of an unstable compromise: crystallization inhibitors and the kidney's role in mineraI regulation. Kidney Int. 38:625-631, 1990 21. Coe FL, Parks JH, Asplin JR: The pathogenesis and treatment ofkidney stones. N.Engl.J.Med. 327:1141-1152, 1992 22. Curhan GC: Epidemiologie evidence for the role of oxalate in idiopathie nephrolithiasis. J. Endourol. 13:629-31, 1999

110

23. Denhardt DT, Guo X: Osteopontin: A protein with diverse functions. FASEB 1. 7:1475-1482, 1993 24. Denhardt DT, Noda M, O'Regan AW, Pavlin D, Berman JS: Osteopontin as a means to cope with environmental insults: regulation ofinflammation, tissue remodeling, and cell survival. J.Clin.Invest 107:1055-1061,2001 25. Doyle IR, Marshall VR, Dawson CJ, RyaU RL: Calcium oxalate crystal matrix extract: the most potent macromolecular inhibitor of crystal growth and aggregation yet tested in undiluted human urine in vitro. Urol.Res. 23:53-62, 1995 26. Doyle IR, RyaU RL, Marshall VR: Inclusion ofproteins into calcium oxalate crystals precipitated from human urine: a highly selective phenomenon. Clin. Chem. 37:1589-94, 1991 27. Durrbaum D, Rodgers AL, Sturrock ED: A study of crystal matrix extract and urinary prothrombin fragment 1 from a stone-prone and stone-free population. Urol.Res. 29:83-88, 2001 28. Fisher LW, Torchia DA, Fohr B, Young MF, Fedarko NS: Flexible structures of SIBLING proteins, bone sialoprotein, and osteopontin. Biochem.Biophys.Res.Commun. 280:460-465, 2001 29. Fisher L, Stubbs J, Young M: Antisera and cDNA probes to human and certain animal model bone matrix noncollagenous proteins. Acta Orthop. Scand. Suppl. 266:61-65, 1995 30. Flores ME, Heinegard D, Reinholt FP, Andersson G: Bone sialoprotein coated on glass and plastic surfaces is recognized by different (33 integrins. Exp.Cell Res. 227:40-46, 1996 31. Gang X, Ueki K, Kon S, Maeda M, Naruse T, Nojima Y: Reduced urinary excretion of intact osteopontin in patients with IgA nephropathy. Am.J.Kidney Dis. 37:374-379, 2001 32. Goldberg HA, Warner KJ: The staining ofacidic proteins on polyacrylamide gels: enhanced sensitivity and stability of "Stains-aU" staining in combination with silver nitrate. Anal.Biochem. 251:227-233, 1997 33. Goldberg, H. A. and Warner, K. J. Binding ofbone sialoprotein to hydroxyapatite. 2000. Unpublished work.

111

34. Goldfarb DS, Coe FL: Prevention of recurrent nephrolithiasis. Am. Fam. Physician 60:2269-76, 1999 35. Grover PK, Stapleton AM, Miyazawa K, RyaII RL: Simple, sensitive and accurate method for the quantification ofprothrombin mRNA by using competitive PCR. Biochem.J. 356:111-120,2001 36. Hedgepeth R, Yang L, Resnick M, Marengo S: Expression ofPIOteins that Inhibit Calcium Oxalate Crystallization ln Vitro in the Urine and Stone-Forming Individuals. Am. J. Kid. Dis. 37:104-112,2001 37. Hess B, Kok DJ: Kidney Stones: Medical and Surgical Management, chap. 1, in Kidney Stones: Medical and Surgical Management, edited by Coe FL, Favus MJ, Pak CYC, Parks JH, Preminger GM, Philadelphia, Lippencott-Raven Publishers, 1996, p 3 38. Hojgaard l, Fornander AM, Nilsson MA, Tiselius HG: Crystallization during volume reduction of solutions with a composition corresponding to that in the collecting duct: the influence ofhydroxyapatite seed crystals and urinary macromolecules. UIOl.Res. 27:417-425, 1999 39. Hojgaard l, Tiselius HG: Crystallization in the nephron. Urol.Res. 27:397-403, 1999 40. Hotta H, Kon S, Katagiri YU, Tosa N, Tsukamoto T, Chambers A, Uede T: Detection ofVarious Epitopes ofMurine Osteopontin by Monoclonal Antibodies. Biochem. Biophys. Res. Comm. 257:6-11, 1999 41. Hoyer JR, Asplin JR, Otvos L: Phosphorylated osteopontin peptides suppress crystaIlization by inhibiting the growth of calcium oxalate crystals. Kidney Int. 60:77-82,2001 42. Hunter GK, Hauschka PV, Poole AR, Rosenberg LC, Goldberg HA: Nucleation and inhibition ofhydroxyapatite formation by mineralized tissue proteins. Biochem.J. 317:59-64, 1996 43. Iguchi M, Takamura C, Umekawa T, Kurita T, Kohri K: Inhibitory effects of female sex hormones on urinary stone formation in rats. Kidney Int. 56:479-485, 1999 44. Khan SR: AnimaI models ofkidney stone formation: an analysis. World J.UIOI. 15:236-243, 1997

112

45. Khan SR: Calcium phosphate calcium oxalate crystal association in urinary stones: Implications for heterogeneous nucleation of calcium oxalate. J.Vrol. 157:376-383, 1997 46. Khan SR, Atmani F, Glenton P, Hou Z, Talham DR, Khurshid M: Lipids and membranes in the organic matrix ofurinary calcifie crystals and stones. Calcif.Tissue Int. 59:357-365, 1996 47. Khan SR, Maslamani SA, Atmani F, Glenton PA, Opalko FJ, Thamilselvan S, Hammett-Stabler C: Membranes and their constituents as promoters of calcium oxalate crystal formation in human urine. Calcif.Tissue Int. 66:90-96, 2000 48. Khan SR, Thamilselvan S: Nephrolithiasis: a consequence ofrenal epithelial cell exposure to oxalate and calcium oxalate crystals. Mol.Vrol. 4:305-312, 2000 49. Kitamura T, Pak CYC: Tamm and Horsfall glycoprotein does not promote spontaneous precipitation and crystal growth of calcium oxalate in vitro. J.VIOI. 127:1024-1026, 1982 50. Kohri K, Nomura S, Kitamura Y, Nagata T, Yoshioka K, Iguchi M, Yamate T, Vmekawa T, Suzuki Y, Sinohara H, Kurita T: Structure and expression ofthe rnRNA encoding urinary stone protein (osteopontin). J.BioI.Chem. 268:1518015184, 1993 51. Kon S, MaedaM, Segawa T, Hagiwara Y, Horikoshi Y, Chikuma S, TanakaK, Rashid MM, Inobe M, Chambers AF, Vede T: Antibodies to different peptides in osteopontin reveal complexities in the various secreted forms. J.Cell Biochem. 77:487-498,2000 52. Koul HK, Koul S, Fu S, Santosham V, Seikhon A, Menon M: Oxalate: from crystal formation to crystal retention. J.Am.Soc.Nephrol. 10 Suppl 14:S417-S421, 1999 53. Kuboki Y, Takita H, Fujisawa R, Yamaguchi H, Yamada H, Tazaki M, Ohnuma Y, Mizuno M, Tsuzaki M: Histidinoalanine-containing phosphoprotein in bone. Calcif.Tissue Int. 48:196-203, 1991 54. Laemmli UK: Cleavage of structural proteins during the assembly ofthe head of bacteriophage T4. Nature 227:680-685, 1970 55. Lee YH, Huang WC, Chiang H, Chen MT, Huang JK, Chang LS: Determinant role oftestosterone in the pathogenesis ofurolithiasis in rats. J.VIOI. 147:11341138, 1992

113

56. Lepage L, Tawashi R: Growth and characterization of calcium oxalate dihydrate crystals (weddellite). J.Pharm.Scï. 71:1059-1062, 1982 57. Lieske JC, Deganello S: Nucleation, adhesion, and intemalization of calciumcontaining urinary crystals by renal cells. IAm.Soc.Nephroi. 10 Suppl 14:S422S429, 1999 58. Lyons RR, Fleming DE, Doyle IR, Evans NA, Dean CJ, Marshall VR: Intracrystalline proteins and the hidden ultrastructure of calcium oxalate urinary crystals: implications for kidney stone formation. J.Struct.Bioi. 134:5-14,2001 59. Mandel NS, Mandel GS, Hasegawa AT: The effect ofsome urinary stone inhibitors on membrane interaction potentials of stone crystals. IUroi. 138:557562, 1987 60. Maslamani S, Glenton PA, Khan SR: Changes in urine macromolecular composition during processing. J.Uroi. 164:230-236,2000 61. McKee MD, Nanci A, Khan SR: Ultrastructural immunodetection of osteopontin and osteocalcin as major matrix components ofrenal calculi. J.Bone Miner.Res. 10:1913-1929, 1995 62. Moreno EC, Kresak M, Hay DI: Adsorption of molecules ofbiological interest onto hydroxyapatite. Calcif.Tissue Int. 36:48-59, 1984 63. Morozumi M, Ogawa Y: Impact of dietary calcium and oxalate ratio on urinary stone formation in rats. MoI.Droi. 4:313-320, 2000 64. Morse RM, Resnick MI: A new approach to the study ofurinary macromolecules as a participant in calcium oxalate crystallization. IUroi. 139:869-873, 1988 65. Morse RM, Resnick MI: Urinary stone matrix. J.Droi. 139:602-606, 1988 66. Nishio S, Hatanaka M, Takeda H, Aoki K, Iseda T, Iwata H, Yokoyama M: Calcium phosphate crystal-associated proteins: alpha2-HS-glycoprotein, prothrombin FI, and osteopontin. MoI.Uroi. 4:383-390, 2000 67. Nishio S, Hatanaka M, Takeda H, Iseda T, Iwata H, Yokoyama M: Analysis of urinary concentrations of calcium phosphate crystal- associated proteins: alpha2HS-glycoprotein, prothrombin FI, and osteopontin. J.Am.Soc.Nephroi. 10 Suppl 14:S394-S396, 1999

114

68. OIdberg A, Franzen A, Heinegard D: Cloning and sequence analysis of rat bone sialoprotein (osteopontin) cDNA reveaIs an Arg-Gly-Asp cell binding sequence. Proc.NatI.Acad.Sci.USA 83:8819-8813, 1986 69. Pak CY: Etiology and treatment ofurolithiasis. Am.J.Kidney Dis. 18:624-637, 1991 70. Rittling SR, Denhardt DT: Osteopontin function in pathology: Iessons from osteopontin-deficient mice. Exp.Nephrol. 7:103-113, 1999 71. Romero MC, Nocera S, Nesse AB: Decreased Tamm-Horsfall protein in lithiasic patients. Clin.Biochem. 30:63-67, 1997 72. Rose GA, Sulaiman S: Tamm-Horsfall mucoproteins promote calcium oxalate crystal formation in urine: Quantitative studies. J. Urol. 127:177-179, 1982 73. Ryall R, Hibberd CM, Marshall VR: A method for measuring inhibitory activity in whole urine. Urol.Res. 3:285-289, 1985 74. Ryall RL, Fleming DE, Grover PK, Chauvet M, Dean CJ, Marshall VR: The hole truth: intracrystalline proteins and calcium oxalate kidney stones. Mol.Urol. 4:391-402,2000 75. Ryall RL, Harnett RM, Hibberd CM, Mazzachi BC, Mazzachi RD, Marshall VR: Urinary risk factors in calcium oxalate stone disease: comparison of men and women. Br.J.Urol. 60:480-488, 1987 76. Salih E, Zhou H, Glirncher M: Phosphorylation ofPurified Bovine Bone Sialoprotein and Osteopontin by Protein Kinases. J. Biol. Chem. 271:1689716905, 1996 77. Sarada B, Satyanarayana U: Influence ofsex and age in the risk ofurolithiasis--a biochemical evaluation in Indian subjects. Ann.Clin.Biochem. 28 ( Pt 4):365-367, 1991 78. Shïraga H, Min W, VanDusen WJ, Clayman MD, Miner D, Terrell CH, Sherbotie JR, Foreman JW, Przysiecki C, Neilson EG, Hoyer JR: Inhibition of calcium oxalate crystal growth in vitro by uropontin: Another member ofthe aspartic acidrich protein superfamily. Proc.Natl.Acad.Sci.USA 89:426-430, 1992 79. Stapleton AM, Dawson CJ, Grover PK, Hohmann A, Comacchio R, Boswarva V, Tang Y, Ryall RL: Further evidence linking urolithiasis and blood coagulation:

115

urinary prothrombin fragment 1 is present in stone matrix. Kidney Int. 49:880888, 1996 80. Stapleton AM, Ryall RL: Blood coagulation proteins and urolithiasis are linked: crystal matrix protein is the FI activation peptide ofhuman prothrombin. Br.J.Urol. 75:712-719, 1995 81. Tardive! S, Medetognon J, Randoux C, Kebede M, Drueke T, Daudon M, Rennequin C, Lacour B: Alpha-1-microglobulin: inhibitory effect on calcium oxalate crystallization in vitro and decreased urinary concentration in calcium oxalate stone formers. Urol.Res. 27:243-249, 1999 82. Tawada T, Fujita K, Sakakura T, Shibutani T, Nagata T, Iguchi M, Kohri K: Distribution of osteopontin and calprotectin as matrix protein in calciumcontaining stone. Urol.Res. 27:238-242, 1999 83. Trewick AL, Rumsby G: Isoelectric focusing ofnative urinary uromodulin (Tamm-Rorsfall protein) shows no physicochemical differences between stone formers and non-stone formers. Urol.Res. 27:250-254, 1999 84. Verkoelen CF, van der Boom BG, Kok DJ, Romijn JC: Sialic acid and crystal binding. KidneyInt. 57:1072-1082,2000 85. Verkoelen CF, van der Boom BG, Schroder FR, Romijn JC: Cell cultures and nephrolithiasis. World J.Drol. 15:229-235, 1997 86. Wesson JA, Worcester E: Formation ofhydrated calcium oxalates in the presence ofpoly-L- aspartic acid. Scanning Microsc. 10:415-423, 1996 87. 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 88. Worcester EM: Inhibitors of stone formation. Semin.Nephrol. 16:474-486, 1996 89. Yasui T, Sato M, Fujita K, Tozawa K, Nomura S, Kohri K: Effects of citrate on renal stone formation and osteopontin expression in a rat urolithiasis model.

Urol.Res. 29:50-56, 2001 90. Yoshida 0, Okada Y: Epidemiology ofurolithiasis in Japan: a chronological and geographical study. Urol.Int. 45:104-111, 1990

116

91. Baumann JM: Stone prevention: why so little progress? Urol. Res. 26: 77-81, 1998 92. Watts RWE: Urinary stone disease (urolithiasis). In: Weatherall DJ, Kedingham JGG, Warrell DA (eds) Oxford textbook ofmedicine, 3 rd edn. Oxford Medical Publications, Oxford, 1996, pp 3251-3257 93. Finlayson B: Symposium in renallithiasis. Renallithiasis in review. Urol. Clin. North Am. 1: 181-212, 1974 94. Boyce WH: Organic matrix ofhuman urinary concretions. Am. J. Med.: 45: 673, 1968 95. Spector AR, Gray A, Prien Jr. EL: Kidney stone matrix. Differences in acidic protein composition. Invest. Urol.: 13: 387, 1976 96. Hoyer IR: Uropontin in urinary calcium stone formation. Miner. Electrolyte Metab.: 20: 385-392, 1994 97. Pegoraro AA, Peracha W, Hasnain M, Ranginwala N, Shaykh M, Singh AK, Arruda JAL, Dunea G: Evaluation of a new fluorescent dye method to measure urinary albumin in lie ofurinary total protein. Am. 1. Kid. Dis.: 35: 739-744, 2000 98. Newman DJ, Pugia MJ, Lott JA, Wallace IF, Hiar, AM: Urinary protein and albumin excretion corrected by creatinine and specific gravity. Clin. Chim. Acta: 294: 139-155,2000

117

ACKNOWLEDGEMENTS

ACKNOWLEDGEMENTS

.:. 1 would like to thank Dr. Marc D. McKee and Dr. Denise Arsenault for giving me the tremendous opportunity to do this work. Thank-you for your guidance and for believing in me. •:. Roxana Atanasiu, thank-you for teaching and guiding me through the fust part ofmy Masters as well as for your contribution to Figure 8 (sample preparation and WB). •:. Douglas Vandor, you are an inspiration to do bigger and better things. •:. Thank-you Isabelle Turgeon for YOur friendship and advice, for ordering all ofthose things 1 needed 1ast minute, and for proofreading my abstracts. •:. Caroline Tanguay: thank-you for your help, for all those hours oftalking, looking at pictures, your continuous support and for proofreading my abstracts. •:. Mari Kaartinen, thank-you for all your help, your insights and your company in and outside the office. 1 will always he anti-chicken. •:. Thank-you SherifEI-Madaawy, for your help with the kidney stone sections. •:. To Helen Campbell, many thanks for your assistance and for teaching me to use the SEM and X-ray microanalysis. •:. Jaime Sanchez-Dardon: thank-you for your assistance in using the FACscan. •:. 1 would also like to thank Dr. C.E. Smith for keeping me up-to-date with the literature

and for his insight on anomalous gels and dentistry. •:. Dad, Mom, Grandma, Golda and Tommy, thanks for heing so patient with me; 1 do not know where 1 would be without you. 1 love you aU very much. •:. Thank-you John W. Graham for providing excellent conversation, heing more than pleasant company and for proofreading my thesis. •:. And last but not least, thanks to aU my friends, especiaUy Eva Lee, Helen Fong, Kristen Itagawa, Patricia Tellis, Steve Villeneuve, and Toshiro Nguyen for being

there for me throughout my thesis and making life that much sweeter.

119