Marshall L. Stoller, Maxwell V. Meng-Urinary Stone Disease

Подождите немного. Документ загружается.

Chapter 10 / Modulators of Crystallization 175

175

From: Current Clinical Urology, Urinary Stone Disease:

A Practical Guide to Medical and Surgical Management

Edited by: M. L. Stoller and M. V. Meng © Humana Press Inc., Totowa, NJ

Modulators of Crystallization

of Stone Salts

Saeed R. Khan, PhD and Dirk J. Kok, PhD

CONTENTS

INTRODUCTION

STONE MATRIX

MORPHOLOGY

CRYSTALLIZATION MODULATORS IN URINE

SUMMARY

REFERENCES

Key Words: Nephrolithiasis; crystallization; saturation.

INTRODUCTION

As a direct consequence of the renal function of water preservation urine becomes

supersaturated with slightly soluble salts like calcium oxalates and calcium phosphates

(1). When the supersaturation is high enough and lasts long enough, crystals will form.

Crystallization involves crystal nucleation, growth, and aggregation (2–5). Most crys-

tals are, however, discharged during urination without causing any discomfort (6,7). But

in some individuals, a number of crystals remain in the urinary space either by adhering

to the urothelium or by accreting mass through aggregation (8,9). Under appropriate

conditions, retained crystals evolve into stone nidi thus establishing a base for stone

growth (10). Apparently there are mechanisms to ensure crystals can be passed harm-

lessly. It has become clear that defense mechanisms act at all levels, from supersaturation

to nucleation, crystal growth, crystal aggregation, crystal structure, crystal habit, crystal

surface properties, and crystal interactions with the epithelial lining of the renal tubules

and urinary tract. In fact there also appears to be a mechanism for removal of material

that passed the previous defenses. Crystals that are retained in the renal tubules can enter

the renal interstitium where they are attacked by macrophages (11–13) that are attracted

176 Kahn and Kok

by the chemokines such as MCP-1 (14) produced by renal epithelial cells as well as the

crystal-associated compounds like osteopontin (11,15).

Urinary compounds that can associate with the crystals mediate these mechanisms in

several ways. First, urine supersaturation can be decreased by lowering the salt ion

concentrations or by making them unavailable for crystallization. For example, citrate

will bind calcium, magnesium binds oxalate, and increasing total urine ionic strength

directly increases the amount of calcium oxalate or -phosphate that can be held in solu-

tion. Thus a mainstay of therapy has been to reduce the supersaturation by dilution

(drinking advice), by decreasing excretion of stone ions (e.g., restricted oxalate intake),

or by increasing excretion of chelators (alkali therapy to increase citrate excretion).

However, the fact that people who do not form stones also periodically have supersatu-

rated urine and crystalluria has prompted investigators to look at the processes beyond

supersaturation and at the role these compounds play in urine.

The stones can contain a variety of crystals including calcium oxalate (CaOx), cal-

cium phosphate (CaP), uric acid, struvite, and cystine (2–4,16). CaP is the most common

crystal in both the urine and stones whereas CaOx is the major crystal in most stones. The

stones, particularly those containing CaOx or uric acid, have a highly compacted struc-

ture (Fig. 1). Their outer surfaces appear smooth at low magnification but reveal the

presence of individual tabular or platelike CaOx monohydrate (COM) crystals at higher

magnifications (Fig. 2). Crystal habits are generally not evident on surfaces exposed by

cutting or fragmenting the stone. Such surfaces are typically stratified with radial stria-

tions and concentric laminations or layers, with radial striations being the predominant

feature. Some of the striations run through many laminations whereas the others are

limited to only one. Many of them converge to a point at the base of a lamination mimick-

ing the arrangement of petals in a flower. These points are suggested to be the nucleation

Fig. 1. SEM of the fractured surface of calcium oxalate monohydrate stone. The stone is compact

and shows distinct concentric laminations and radial striations. Concentric laminations are the

result of outward growth of the stone. Radial striations are the result of a plate-like habit of

calcium oxalate monohyydrate crystals.

Chapter 10 / Modulators of Crystallization 177

sites of crystals. The laminations are approx 50–60 µm thick and in many stones can be

easily separated from each other exposing the underlying surfaces. These surfaces show

the same structure as the outer surface of the stone with protruding tips of the tabular

crystals of COM frequently covered with amorphous to flaky matrix material. Many

stones have a well-defined nucleus. The stone nucleus appears granular and nonstratified

and is generally occupied by spherulitic or amorphous CaP and/or aggregates of dumb-

bell-shaped twinned COM crystals. CaP is frequently seen filling the space between

CaOx crystals as well as the concentric laminations.

Like other products of crystallization in biological systems (17), stones are a compos-

ite of crystals and organic materials (18–22). The latter are often referred to as matrix and

comprise macromolecules generally present in the urine. These macromolecules play a

significant role in the development of kidney stones. Some promote crystal formation,

growth, and aggregation whereas others inhibit these processes. Still others promote or

inhibit crystal retention within the urinary space. Activity of a macromolecule is often

controlled by conditions prevalent in the urine at the time of crystallization. The same

macromolecule can promote or inhibit a process. For example macromolecules in solu-

tion behave differently than when they are attached or adsorbed to a surface. This chapter

will summarize our current understanding of the involvement of urinary macromol-

ecules in various phases of crystallization in the urine and the development of urinary

stones. Special emphasis will be given to crystallization of CaOx, because it is the major

crystal in most stones and, as such, has been investigated extensively. We will start with

a discussion of the stone matrix, emphasizing crystal/matrix associations in urinary

crystals and stones, and then examine some of the urinary macromolecules and other

compounds that constitute stone matrix and are known to influence crystallization events

in the urinary tract. Although these compounds can both stimulate and inhibit several

crystallization processes, they are usually referred to as crystallization inhibitors or

modulators.

Fig. 2. Outer surface of calcium oxalate monohydrate stone. Edges of tabular crystals are clearly

visible.

178 Kahn and Kok

STONE MATRIX

The organic matrix of most urinary stones accounts for 2–3% of their total dry

weight, the rare matrix stones with 65% matrix contents notwithstanding. Boyce and

co-workers defined and established the importance of stone matrix in the formation of

kidney stones postulating an active architectonic role for matrix (18–20). In their view

the matrix acts as a ground substance and controls crystallization within its bounds. An

opposite hypothesis was advanced by Vermeulen and co-workers, who viewed matrix

and its ubiquitous presence as a mere coincidence because stones form as a result of

crystallization in urine in the presence of large macromolecules (23,24). According to

them the matrix is adventitiously acquired primarily by physical adsorption of urinary

mucoproteins on crystal surfaces. The theory of physical adsorption of urinary proteins

on surfaces of CaOx crystals was tested by solution depletion method (25) and exami-

nation of crystals by transmission electron microscopy after incubation in protein so-

lutions (26). Results showed proteins have a strong affinity for CaOx crystals.

Adsorption of anionic proteins was found sensitive to calcium ion concentration,

whereas cationic protein adsorption depended on the oxalate ion concentration with

temperature and pH playing only a minor role. Proteins formed a discontinuous coat

around the crystals ranging in thickness from 10 to 20 nm. It has been suggested that

newly formed crystals with a macromolecular coat are less likely to dissolve during the

routine urinary ionic and pH changes and therein may lie the importance of matrix in

stone formation (22).

MORPHOLOGY

Stone matrix is extremely insoluble. As a result, chemical identification of all the

constituent macromolecules has been difficult. However this property has provided the

opportunity to microscopically examine the mostly EDTA-insoluble matrix and iden-

tify matrix components using a variety of stains and more recently specific antibodies.

Morphological examination of decalcified as well as intact stones has provided evi-

dence that their matrix is pervasive (Fig. 3), distributed throughout the stones, and has

both amorphous as well as fibrous components (20–22). The crystal matrix association

is so intimate that dipyramidal habit of CaOx dihydrate (COD), monoclinic, or plate-

like habit of COM and spherulitic habit of CaP stayed intact even after the total removal

of their mineral contents (22,27). Scanning electron microscopic examination of the

decalcified COM stone revealed the matrix to be organized in concentrically arranged

layers. Individual layers were 2–5 µm thick, much thinner than the laminations seen

in an intact stone, and had a uniform thickness throughout the circumference. Succes-

sive layers appeared as leaves or rings of an onion bulb with very little or no space

between them. The layers were made of loosely or tightly matted fibers and contained

empty columns representing crystal ghosts, which were presumably formed by disso-

lution of tabular COM crystals. These crystal ghosts were often arranged radially in

relation to the stone center reminiscent of the radially arranged crystals in the intact

stones. Examination of these concentrically arranged layers by transmission electron

microscopy confirmed that they contained radially arranged columnar crystal ghosts

surrounded by an amorphous electron dense coat and embedded in a fibrous matrix. An

electron dense material was also present inside the crystal ghosts. Cellular degradation

products including degenerating nuclei, mitochondria, endoplasmic reticulum, and

membrane fragments as well as vesicles occupied the intercrystalline spaces (22).

Chapter 10 / Modulators of Crystallization 179

Histochemically the layered matrix was sudanophilic (28), and stained positive

with periodic acid schiff (PAS) and colloidal iron (29–31) indicating the presence of

lipids, proteins and glycosaminoglycans (GAG). Immunohistochemical examination

of decalcified stones using antibodies against osteopontin (OPN) and calprotectin

showed them to stain both the stone center and concentric laminations (32). Ultrastruc-

tural immunodetection using specific antibodies showed the OPN and osteocalcin as

major components of the matrix of human kidney stones as well as CaOx crystals and

stones experimentally induced in male rats (11). OPN was detected both inside the

crystals as well as on their surfaces. Ultrastructural examination of decalcified stones

also showed the crystal associated matrix to stain positive with malachite green indi-

cating the presence of phospholipids (33).

Chemical Composition

Organic matrix of urinary stones contains lipids, GAG carbohydrates, and proteins,

with proteins comprising approx 64% of the matrix. Table 1 lists the compounds iden-

tified in the stone matrices; and as to be expected most of them are proteinaceous in

nature. Many other proteins have also been detected but not identified. Initially lipids

were not recognized as constituent of stone matrix (20) even though detected as an

osmiophilic substance during histochemical examination of decalcified stones. As dis-

cussed later in the chapter, recent studies have provided compelling evidence that lipids

are an important component of stone matrix and play a significant role in stone forma-

tion.

It is now generally understood that matrix components play a significant role in stone

formation and that stone matrix is derived from the incorporation of urinary macromol-

ecules in the growing stones. All macromolecules present in the urine can become part

Fig. 3. Fractured surface of demineralized calcium oxalate monohydrate stone. Demineralization

was accomplished by treating the stone pieces with EDTA. Once the stone became translucent,

it was dried and fractured to expose the interior. Microanalysis of the stones showed no calcium.

A comparison with Fig. 1 indicates the ubiquitous nature of the organic matrix. Increase in

concentric laminations and loss of radial striations suggests that the organic matrix became

incorporated during the outward growth of the stones.

180 Kahn and Kok

of the stone matrix but only some of them are there because they participated in crystal-

lization and stone formation. This appreciation has led investigators to study crystalli-

zation in vitro, using freshly collected urine and determine the macromolecules that

become a part of the crystal matrix (34,35). As we discuss later, crystal matrix also

contains lipids, GAGs, and proteins.

LYCOSAMINOGLYCANS (GAGS)

GAGs can account for up to 20% of the stone matrix (36). Heparan sulfate (HS) and

hyaluronic acid (HA) are the two major GAGs in the matrix of both stones and CaOx

crystals precipitated in the urine (37,38). The most abundant urinary GAG, chondroitin

sulfate (CS) has not been detected in the matrices of CaOx crystals or stones indicating

selectivity in incorporation (39). Keratan sulfate and dermatan sulfate are occasionally

present in trace amounts.

ROTEINS

More than twenty individual proteins have been detected in the matrix of various types

of stones. Although most of them have been identified (Table 1), some still remain

nameless (40) and a few await confirmation of their identification (41). Here we will

Table 1

Macromolecules Detected in CaOx/CaP Stones and/or Crystals

Macromolecules References

• Proteins

Albumin 18, 42

α-1-microglobulin 34

α-1-acid glycoprotein 260

α and γ-globulins 18, 42

α-1-antitrypsin 299

Apolipoprotein A1 260

β-2-microglobulin 260

Calprotectin (Calgranulin) 32, 300

Haemoglobin 298

Inter-α-inhibitor 43, 192, 193

Nephrocalcin 156–158

Neutrophil elastase 298

Osteopontin (Uropontin) 218

Porin 40

Renal lithostatin 260

Retinol binding protein 260

Superoxide dismutase 40

Tamm-Horsfall protein 42

Transferin 257

Urinary prothrombin fragment-1 237–239

• Glycosaminoglycans

Heparan sulfate 36–39

Hyaluronic acid

• Lipids

Phospholipids, cholesterol, glycolipids 28, 33, 280

Chapter 10 / Modulators of Crystallization 181

discuss in greater details some of the proteins investigated for their perceived importance

in stone formation. Human serum albumin (HAS), α and γ globulins and Tamm-Horsfall

protein (THP) were the first proteins identified in stone matrix (42). Albumin is a major

component of the matrix of all type of stones including CaOx, uric acid, struvite, and

cystine. Albumin is also found in the matrix of CaOx and CaP crystals precipitated from

the human urine and it is more pronounced in crystals induced in stone formers urine

(43,44). Both CaOx and CaP crystals are known to adsorb HAS.

THP is not always detected in stones and even then in only minor quantities, 0.002–

1.04 mg/g (w/w) of stone (45). We found THP in matrices of both the CaOx and CaP

crystals induced in human urine and as a major component of CaP crystal matrix. We

also discovered that THP associated with CaOx crystals could be easily removed by

washing the crystals with sodium hydroxide solution (43,44,46), indicating that THP

interacts with crystal surfaces. Ultrastructural investigations of human CaOx urinary

stones (unpublished results) and CaOx nephroliths induced in an animal model also

demonstrated that THP interacts primarily with the crystal surfaces and does not appear

to be occluded in the crystalline mass (47,48). This may explain THP’s scanty presence

in the stone matrix.

Of the other stone matrix proteins listed in Table 1, OPN, α-1-microglobulin, urinary

prothrombin fragment-1 (UPFT-1), and light and heavy chains of inter-α-inhibitor have

been identified in the matrix of CaOx and CaP crystals precipitated from the urine of

normal and stone forming individuals (43,44). Ultrastructurally OPN is the major com-

ponent of the matrix of CaOx stones (11,49). More OPN is present in CaOx monohydrate

stones (800 µg /100 mg stone) than in COD stones (10 µg /100 mg stone).

IPIDS

Lipids are an integral part of the organic matrices of all mineralized tissues and

pathological calcifications (50–52). Even though they account for a relatively small

proportion of the matrix; 7–14% in bone, 2–6% in dentin, 12–22% in newly mineralized

enamel, approx 9.6% in submandibular salivary gland calculi, and 10.2% in supragingival

calculi (50–55), lipids play a significant role in the calcification process. They promote

crystal nucleation, modulate growth and aggregation and become incorporated in the

growing calcifications.

The matrix of all stones investigated to date including struvite, uric acid, CaOx, and

CaP contains lipids (28,33). The protein to lipid ratio is, however, higher in the matrix

of struvite and uric acid stones than the matrix of CaOx and CaP stones (Table 2). Even

though there are no significant differences in various types of lipids determined, matrix

of struvite stones contain more cholesterol, cholesterol ester, and triglycerides than the

other three types of stones. Analysis by one dimensional thin layer chromatography

separated and identified various phospholipids and glycolipids including sphingomyelin

(SM), phosphatidyl choline (PC), phosphatidyl ethanolamine (PE), and cardiolipin (CL)

and trace amounts of phosphatidylserine (PS) in matrices of all stones. Occasionally

stone matrix also contains phosphatidyl inositol (PI), lyso-PC, lyso-phosphatidic acid

(PA) and lyso-PE. Glycolipids detected in all stones include gangliosides,

D-sphin-

gosine, and glucocerebrosides. In addition, struvite stone matrix contains sulfatides and

digalacto diglycerides whereas matrices of CaOx and CaP stones contain cerebrosides

1 and 2 and digalacto-diglycerides.

The matrices of all stones comprise both complexed and noncomplexed lipids. The

amount of complexed lipids is highest in CaP stones and lowest in the uric acid stones.

182 Kahn and Kok

Table 2

Stone Matrix Composition (% of EDTA Soluble Matrix) and Lipid Constituents (mg/g stone)

Stone type Protein Lipid Total cholesterol Triglycerides Glycolipids Phospholipid

Struvite (

= 5) 74% 26% 1.53

0.72 10.71 ±

9.17 0.13 ±

0.05 0.06

0.04

Calcium oxalate (n

= 5) 20% 80% 0.64 ±

0.27 1.64 ± 0.60 0.16 ±

0.06 0.05

0.03

Calcium phosphate (

= 3) 33% 67% 0.76

0.50 1.45 ± 0.13 0.17 ±

0.10 0.05

0.02

Uric acid (n

= 5) 75% 25% 0.20 ±

0.07 1.60 ± 0.34 0.09 ±

0.03 0.03

0.01

182

Chapter 10 / Modulators of Crystallization 183

Both complexed and noncomplexed lipids contain cholesterol, triglycerides, phospho-

lipids and gangliosides.

Both CaOx and CaP crystals induced in the urine contain lipids (33). However, there

are no significant differences in either the nature of lipid constituents or the lipid amounts

per gram of crystals between the two types of calcific crystals (Table 3). Glucocere-

brosides are the most common glycolipids, whereas SM is the most common phospho-

lipid. Gangliosides are the second most common glycolipid and PC and PE the most

common phospholipids. Determinations of lipids in the urine before and after experi-

mental induction of CaOx crystals show that the formation of crystals depletes the urine

of its phospholipids indicating its incorporation in the crystal matrix. Almost all urinary

phospholipids become incorporated in the formation of crystals (33).

CRYSTALLIZATION MODULATORS IN URINE

In urine, three classes of crystallization modulators can be recognized; low-molecular

weight (MW) compounds like citrate and pyrophosphate; (glyco)-proteins; and high-

MW nonprotein compounds such as glycosaminoglycans, acid mucopolysaccharides,

and various types of lipids. All have some influence on crystallization and urolithiasis,

either through direct involvement in the various processes such as crystal nucleation,

growth, aggregation, and retention or by influencing the urinary environment. From

crystallization experiments with diluted and undiluted whole urine, it appears that in

nonstone formers the concerted actions of these compound ensure that: (1) the crystals

formed remain unaggregated and small enough to be excreted (56,57), (2) the crystals

have a reduced affinity for epithelial cells (58,60), and (3) the crystals if needed are easily

recognized and removed by macrophages (61).

Which inhibitors play a major role in this? The first approach to answering this

question has been to identify individual compounds in the urine and test their “inhibi-

tory” effectiveness in crystallization and cell-culture systems. The next problem has

been to translate these data to the whole urine situation where singular inhibitors may

also co-operate or compete with each other and where restrictions posed by the kidney

and urinary tract itself (flow-rates, residence time, and changing urine composition)

affect their inhibitory and stimulatory powers. We will discuss the actions of individual

modulators in inorganic solutions as well as in the urinary conditions.

Low-MW Compounds

PYROPHOSPHATE AND ITS ANALOGS

Citrate and pyrophosphate are two major representatives of the low MW compounds

that have an effect on crystallization. Pyrophosphate is present in urine at micromolar

concentrations (Table 4) and effective as inhibitor at those same concentrations. When

Table 3

Lipids Present in the Matrix of Crystals Produced

in Normal Human Urine (mg/total crystal/24 h)

Crystal type Cholesterol Triglycerides Glycolipids Phospholipids

Calcium oxalate 0.37 ± 0.09 3.11 ± 2.9 1.15 ± 1.20 0.09 ± 0.04

Calcium phosphate 0.27 ± 0.10 2.51 ± 2.4 1.35 ± 1.04 0.02 ± 0.01

184 Kahn and Kok

tested as a single compound in a seeded crystal growth system, it inhibits CaOx crystal

growth by 50% at a concentration of 16–20 µM (62–65), (Table 5) which is within its

urine concentration range (Table 6). In addition, pyrophosphate under nonurine condi-

tions inhibits the growth of CaPs (66,67) and causes preferential formation of COD

(68). The effectiveness of pyrophosphate suggested that it may significantly contribute

to the crystal growth inhibitory potential of urine. At its urine concentration range (15–

100 µM) pyrophosphate inhibits CaOx monohydrate crystal growth up to 80% when it

is tested in vitro. These data also prompted interest in the effect of nonbiodegradable

pyrophosphate analogs, bisphosphonates on crystallization, as these can be given thera-

peutically. It was found that bisphosphonates yield 50% COM crystal growth inhibition

at concentrations ranging from 1 to 20 µM (69). Some bisphosphonates inhibit crystal

growth at lower concentrations then pyrophosphate.

Pyrophosphate and several analogs inhibit crystallization of hydroxyapatite and

brushite. Pyrophosphate was most effective in inhibiting hydroxyapatite precipitation

whereas phytate most effectively blocked brushite precipitation (66). COD is preferen-

tially formed in the presence of citrate 1.0  10

–3

mol/L and pyrophosphate 2.0  10

–4

mol/L. Nearly pure COD produced with pyrophosphate was stable over 7 d, whereas

that with citrate underwent partial transformation within 48 h. An additive effect of

citrate and pyrophosphate was found on the stability of COD. It was concluded that a

pyrophosphate concentration above a critical point was sufficient to prevent solution-

mediated transformation of COD, and this critical point might be lowered to the physi-

ological range with the presence of citrate (68).

Further studies revealed that the effectivity of the bisphosphonates depends on how

well they can bind to the calcium ions on the crystal surface. This is regulated by the

concerted action of the two phosphonate groups and the rest group attached to the same

carbon atom of the molecule. A hydroxyl group or amino-group as rest group yields

the best inhibitors (69). The growth inhibitory power further depends on the protona-

tion state of the bisphosphonate, thus on the pH and the pKa-values of the bisphosphon-

Table 4

Inhibitor Characteristics

Urinary excretion, mg/24-h avg (range)

Compound MW Control Stone former

Total GAGs, mg/24 h 23–28 (0–50) 23 (0–53)

Chondroitin sulfate A, C

5–20 kDa 14 (8–18) 14 (4–19)

Chondroitin sulfate C

Keratan sulfate

5 kDa 2 (0.5–7) 2 (0.5–7)

Dermatan sulfate (Chs-B)

1 (0.5–2) 1 (0.5–2)

Heparan sulfate

5 (2–14) 5 (3–14)

Hyaluronic acid

up to 10

3 (1.8–7.5) 3 (1–8)

Pentosanpoly sulfate

4–7 kDa 1–16

Pyrophosphate (µM) 178 Da 30–70 (15–100) 36–50 (8–94)

Based on measurements of separate GAGs and on reported percentages of the total GAGs.

12% >100 kDa, 64% 10–100, 24% <10 kDa

Excretion rate at a 400 mg/24-h oral dose.

(Data from References 62, 71–75, 97, 107, 108, 124, 126, 128, 130, 301–308.)