Matrix Metalloproteinase-7 Degradation of Fetuin Blocks ... · PDF fileMatrix Metalloproteinase-7 Degradation of Fetuin Blocks Fetuin-Mediated Inhibition of Mineralization ... Electron
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Matrix Metalloproteinase-7 Degradation of Fetuin Blocks Fetuin-Mediated Inhibition of Mineralization
by
Reyhaneh Rezaei
A thesis submitted in conformity with the requirements for the degree of Master of Science
Discipline of Periodontology, Faculty of Dentistry University of Toronto
calcium solution (CaCl2, 50 mM; Hepes, 50 mM; NaCl, 150 mM; pH 7.4) and 80 µl of
buffer solution (HEPES, 50 mM; NaCl, 150 mM; pH 7.4). The control groups were
arranged into columns in the plates. Control groups included phosphate solution only,
calcium solution only or buffer solution only. The control groups were analyzed at the
same time as test groups (phosphate, calcium and buffer solutions). After incubations (up
to 4 hours at room temperature), the supernatant was removed, leaving the nascent
hydroxyapatite crystals bound to the bottom of the well. Electron microscopy (see below)
was used to examine the crystalline structures that were formed at the bottom of the
wells.
Presumptive crystals of HA (and with this type of analysis, apatite) were stained with
75 µl alizarin Red S (0.5% alizarin Red S; pH 4.2; 5 minutes of staining). The unbound
alizarin red solution was removed by pipetting. To dissolve the crystal-bound alizarin red,
cetylpyridinium chloride solution (100 µl; 100 mM) was added to each well and
incubated at room temperature for 30 minutes. The relative abundance of hydroxyapatite
crystals formed in each well was estimated from the amount of Alizarin Red S stain
!
!28!
remaining in the cetylpyridinium chloride solution, which was measured by absorbance at
540 nm with a spectrophotometer. These experiments, each with 8 replicates, were
repeated three times and the mean values were used for analyses.
6.2. Fetuin-mediated Inhibition of Mineralization In vitro
Various concentrations of bovine fetuin (0.5, 1.0, 1.4, 2.0 μM) were added to the
mineralization solutions to assess the inhibitory effect of fetuin on mineralization and to
determine whether MMP degradation of fetuin would affect mineralization in vitro. To
study the kinetics of mineralization in the presence or absence of fetuin, mineralization
was quantified at 30 minutes time intervals. The experiment was repeated with 8
replicates during each trial; mean values were used to plot time-courses over 4 hours.
6.3. Crystal Species Formed in Mineralization Assays
The nature of the crystals formed in the assays was assessed by embedding crystalline
material removed from the bottom of dishes and examination by electron microscopy.
Briefly, the dried mineral was scraped from the bottom of culture wells, placed into
electron microscopy block molds and embedded in Quetol-Spurr resin. Sections (100 nm
thick) were cut on an RMC MT6000 ultramicrotome, placed on formvar coated-grids,
imaged with a FEI Tecnai 20 transmission electron microscope and the images were
stored electronically.
To examine the type(s) of crystal species formed in the presence or absence of fetuin,
electron diffraction analysis was performed. The d-spacing values of the samples were
compared with those of authentic hydroxyapatite crystal standards.
!
!29!
6.4. Effect of MMP Cleavage of Fetuin on its Ability to Inhibit
Mineralization
Bovine and human fetuin were treated with activated human MMP-7 or activated mouse
MMP-3 at an enzyme to substrate ratio of 1:60. Incubations of bovine and human fetuin
with MMPs were conducted overnight to ensure optimal digestion of fetuin by the MMP
under test. MMP-mediated fetuin degradation fragments were assessed by SDS-PAGE;
gels were stained with Coomassie blue to identify protein bands. MMP-treated bovine or
human fetuin at 0 or 2 μM was incubated with the mineralizing solutions and
mineralization quantified as described above.
6.5. Statistical Analyses
For all data shown, experiments were repeated 2-3 times and were conducted on different
days. For each individual experiment, at least 5 replicates were analyzed. For continuous
variables, means and standard deviations were computed. For the binding experiments,
means, standard errors were computed. Comparisons between groups were assessed with
analysis of variance and individual group differences were analyzed post hoc with
Tukey’s test. The type I error threshold for estimation of statistical significance was set at
p<0.05.
!
!30!
7. Results
7.1. Cell-free In vitro Mineralization
The calcium phosphate and buffer control groups demonstrated absorbance
measurements of 0.5 and 0.4 absorbance units respectively at the end of 4 hour
incubations. The test groups in which all ions were included in the test solutions showed
~2.5 absorbance units (Figure 5A), suggesting that mineralization had occurred under
these experimental conditions.
In the test group in which all components of the assay were added sequentially, there
was measureable mineralization. In preliminary trials of the test group, assessment of
mineralization by spectrophotometric absorbance (540 nm) of mineral-bound alizarin red
showed that there was a time-dependent (0-4 hours) increase of absorbance from ~0.5 to
2.5 absorbance units, which peaked at about 3.5 hours (Figure 5B).
7.2. Fetuin Inhibition of Mineralization In vitro
The inhibitory effects of fetuin on HA mineralization in vitro were assessed. When the
mineralization assays were conducted with fetuin added to the incubation medium, fetuin
exerted a concentration-dependent inhibitory effect on mineralization (0-2 µΜ). When
the mineralization assays were conducted with bovine fetuin (2 μM), absorbance values
were reduced ~5-fold (p<0.001) in comparison to the same assay conducted in the
absence of fetuin (4 hours incubations; Figure 6A).
The mineralization assays were conducted at various concentrations of fetuin (0.5,
1.0, 1.4, 2.0 µM) and absorbance was measured after 3.5 hours. This time point was
!
!31!
identified as the time at which maximal mineralization occurred under control conditions.
In brief, the data demonstrate that fetuin has a concentration-dependent inhibitory effect
on mineralization (Figure 6B).
7.3. Crystal Species Formed in Mineralization Assays
Transmission electron microscopic imaging of mineralized samples conducted in the
presence or absence of fetuin at various concentrations showed needle-shaped or rod-
shaped crystal species, which are consistent with the appearance of HA crystals 126,127
(Figure 7A).
Electron diffraction analysis of the hydroxyapatite mineralization showed that the d-
spacing and intensities of the pattern of the HA formed in the in vitro assay matched the
published standards (Figure 7B). These data showed that even fetuin inhibited
mineralization but there was no evidence for total inhibition. Further, in the presence of
fetuin, the mineral that was formed was HA.
7.4. Effect of MMP Cleavage on Fetuin Inhibition of Mineralization
Bovine and human fetuin were treated with MMP-7 and MMP-3 and the digestion
products examined by SDS-PAGE. After 24 hours of MMP-7 digestion of human or
bovine fetuin, small fetuin fragments (~17 kDa) were observed (Figure 8A&B).
When mineralization assays were conducted with human or bovine fetuin that had
been degraded by MMP-3 (24 hours at E:S=1:60), the measurements of mineralization
were not substantially different compared with intact fetuin (p>0.2). In contrast, for
human fetuin that was degraded by MMP-7, its ability to inhibit mineralization was
!
!32!
reduced by at least 3.4-fold more compared to intact fetuin (p<0.001). When bovine
fetuin was degraded by MMP-7, there was also a 4-fold increase of mineralization
compared to intact bovine fetuin (p<0.001). Notably, the overall impact of MMP-7 on the
ability of fetuin to inhibit mineralization was more marked with bovine fetuin than with
human fetuin. These data also show that the smaller fetuin digestion fragments (~17kDa)
generated by MMP-7 do not have the capacity to inhibit mineral formation unlike the
native molecule (Figure 8C&D).
!
!33!
8. Discussion
In this study, a simple mineralization assay was used based on the spontaneous and rapid
formation of HA crystals in vitro, an approach that enabled study of the effect of intact
and degraded fetuin on mineralization. The data show that fetuin reduces mineralization
by inhibiting nucleation processes; since it was added simultaneously with the
mineralizing solutions that contained supersaturated concentrations of calcium and
phosphate ions. The rationale for this experimental approach is that fetuin physiologically
inhibits calcification by preventing nucleation of crystal formation107. This inhibitory
effect cannot be assessed after the peak of mineralization (3.5 hrs.) since there is a natural
and significant reduction of mineral formation after this duration of incubation, possibly
due to loss of mineral ions. The reduction of mineralization after 3.5 hours could also be
due to a decrease in pH after consumption of the OH- as a result of formation of HA
crystals128. One of the challenges with the conduct of the mineralization assays in this
study was maintaining a constant pH for the duration of the mineralization assay. Despite
these pitfalls, the mineralization assay data showed increased mineralization for the
duration of the assay (3.5 hours) in the absence of fetuin. This trend was the same in
assays that included fetuin and which also exhibited a decrease in the absorbance
measurements.
Although mineralization was reduced in the presence of fetuin, the nature of the
crystals that were formed did not apparently change, indicating that fetuin inhibited
precipitation of apatite crystals, which is in accord with current ideas on fetuin regulation
of mineralization and with other data showing that in biological systems, apatite is the
!
!34!
first crystal to emerge during mineralization, despite the possibility that other minerals
might form129.
Notably, the in vitro calcification experiments presented in this thesis are an
approximation of what occurs in vivo, which involves the interaction of other regulatory
molecules119. The interaction of fetuin with other proteins and enzymes in serum needs to
be considered to study its effect on calcification of atheromas and a possible association
with periodontitis. Further, only one type of mineralization assay was conducted here;
other types of assays may provide additional insights107.
My main findings are that while intact bovine and human fetuin inhibit HA
formation in vitro, after digestion of fetuin by MMP-7, the inhibitory effect on
mineralization was greatly reduced. In contrast, MMP-3-digested fetuin inhibited
mineralization similar to that of native fetuin. Therefore, MMP-7, a prominent enzyme in
periodontal lesions130 may contribute to the inhibition of vascular mineralization that is
seen in patients with periodontitis131,132.
Periodontitis and atherosclerotic vascular disease are important pathological
disorders due to their high prevalence54. Further , periodontitis and calcifying atheromas
have been associated in epidemiological studies 133,134, suggesting the possibility of a
common underlying mechanism that link these two diseases. The endothelial dysfunction
observed in patients with periodontitis is comparable to that seen in patients with
hypertension78,135. While there is some evidence for the involvement of periodontal
pathogens in the formation of atheromas and damage to endothelial cells 136, the
underlying mechanisms that link periodontitis and the formation of calcifying atheromas
are not defined 54. Notably, enzymes released into the circulation such as MMP-7 may
!
!35!
contribute to the pathogenesis of atherosclerosis 137 but currently we are uncertain
whether there are marked increases of MMP-7 in serum of patients with periodontitis. In
this context, MMP-3 and MMP-7 levels are increased in serum of patients with gastric
cancer 138 and MMP-9 levels are increased in the serum of patients with periodontitis 139.
Based on the results obtained here, MMP-7 can digest fetuin and disturb its
inhibitory effects on mineralization in vitro. If serum levels of MMP-7 are found to be
higher in patients with periodontitis and can cleave fetuin in vivo, this notion could
explain in part the association between periodontitis and calcified atheroma formation.
Further, measurement of serum levels of MMP-7 could be considered as a potentially
useful tool to assess the risk of developing calcifying atheroma in patients with
periodontitis.
! !
!
!36!
Conclusions
My data support the notion that fetuin, which protects against vascular calcification,
can be degraded by proteolytic enzymes that are increased in periodontitis and other
inflammatory diseases. If this main finding can be confirmed in other, more in-depth
studies, it may explain the epidemiological association between periodontitis and
vascular calcification.
!
!37!
Table
Table 1-Definition of sensitivity, specificity, relative risk, biomarker.
Term Definition
Sensitivity The proportion of people with disease who will have a positive test result140
Specificity The proportion of people without the disease who will have a negative test result140
Relative Risk A measure of association between exposure to a particular factor and risk of a certain outcome. It is defined as the ratio of risk in the exposed and unexposed groups141.
Biomarker A molecule that is measured objectively and evaluated as an indicator of normal biologic processes, pathogenic processes, and pharmacologic responses to a therapeutic intervention142
amino acid sequence of the catalytic domain is related toMMP-1. MMPs generally consist of a prodomain, a catalyticdomain, a hinge region, and a hemopexin domain (see Figure1). They are either secreted from the cell or anchored to theplasma membrane. On the basis of substrate specificity,sequence similarity, and domain organization, vertebrateMMPs can be divided into six groups (see Figure 1 and Table1), as described below. An extended version of Table 1,including MMP substrates, is available in the online datasupplement (available at http://www.circresaha.org).
CollagenasesMMP-1, MMP-8, MMP-13, and MMP-18 (Xenopus) are inthis group. The key feature of these enzymes is their ability tocleave interstitial collagens I, II, and III at a specific sitethree-fourths from the N-terminus. Collagenases can alsodigest a number of other ECM and non-ECM molecules.
GelatinasesGelatinase A (MMP-2) and gelatinase B (MMP-9) belong tothis group. They readily digest the denatured collagens,gelatins. These enzymes have three repeats of a type IIfibronectin domain inserted in the catalytic domain, whichbind to gelatin, collagens, and laminin.12 MMP-2, but notMMP-9, digests type I, II, and III collagens.13,14 AlthoughMMP-2 null mice develop without any apparent abnormali-ty,15 mutations in human MMP-2 resulting in the absence ofactive enzyme are linked with an autosomal recessive form ofmulticentric osteolysis, a rare genetic disorder that causesdestruction and resorption of the affected bones.16 Thissuggests that MMP-2 in humans is important forosteogenesis.16
StromelysinsStromelysin 1 (MMP-3) and stromelysin 2 (MMP-10) bothhave similar substrate specificities, but MMP-3 has a proteo-lytic efficiency higher than that of MMP-10 in general.Besides digesting ECM components, MMP-3 activates anumber of proMMPs, and its action on a partially processedproMMP-1 is critical for the generation of fully activeMMP-1.17 MMP-11 is called stromelysin 3, but it is usually
grouped with “other MMPs” because the sequence andsubstrate specificity diverge from those of MMP-3.
MatrilysinsThe matrilysins are characterized by the lack of a hemopexindomain. Matrilysin 1 (MMP-7) and matrilysin 2 (MMP-26),18also called endometase,19 are in this group. Besides ECMcomponents, MMP-7 processes cell surface molecules suchas pro–!-defensin, Fas-ligand, pro–tumor necrosis factor(TNF)-!, and E-cadherin. Matrilysin 2 (MMP-26) also di-gests a number of ECM components.
Membrane-Type MMPsThere are six membrane-type MMPs (MT-MMPs): four aretype I transmembrane proteins (MMP-14, MMP-15, MMP-16, and MMP-24), and two are glycosylphosphatidylinositol(GPI) anchored proteins (MMP-17 and MMP-25). With theexception of MT4-MMP, they are all capable of activatingproMMP-2. These enzymes can also digest a number of ECMmolecules, and MT1-MMP has collagenolytic activity ontype I, II, and III collagens.20 MT1-MMP null mice exhibitskeletal abnormalities during postnatal development that aremost likely due to lack of collagenolytic activity.21 MT1-MMP also plays an important role in angiogenesis.22 MT5-MMP is brain specific and is mainly expressed in thecerebellum.23 MT6-MMP (MMP-25) is expressed almostexclusively in peripheral blood leukocytes and in anaplasticastrocytomas and glioblastomas but not in meningiomas.24,25
Other MMPsSeven MMPs are not classified in the above categories.Metalloelastase (MMP-12) is mainly expressed in macro-phages26 and is essential for macrophage migration.27 Besideselastin, it digests a number of other proteins.MMP-19 was identified by cDNA cloning from liver28 and
as a T-cell–derived autoantigen from patients with rheuma-toid arthritis (RASI).29Enamelysin (MMP-20), which digests amelogenin, is pri-
marily located within newly formed tooth enamel. Ameloge-nin imperfecta, a genetic disorder caused by defective enamelformation, is due to mutations at MMP-20 cleavage sites.30
Figure 1. Domain structure of MMPs. Thedomain organization of MMPs is as indicated: S,signal peptide; Pro, propeptide; Cat, catalyticdomain; Zn, active-site zinc; Hpx, hemopexindomain; Fn, fibronectin domain; V, vitronectininsert; I, type I transmembrane domain; II, type IItransmembrane domain; G, GPI anchor; Cp,cytoplasmic domain; Ca, cysteine array region;and Ig, IgG-like domain. A furin cleavage site isdepicted as a black band between propeptideand catalytic domain.
828 Circulation Research May 2, 2003
!
!39!
Figure 2-Human Ahsg domain stucture89–Human Ahsg consists of 3 domains; cystatin domain1, cystatin domain2 and domain3
share a function with fatty acid binding proteins, a family ofabundantly expressed 14-kDa to 15-kDa proteins. Like fetuin,fatty acid binding proteins reversibly bind hydrophobic li-gands, including saturated and unsaturated long-chain fattyacids, and other lipids with high affinity.40
Because of their rich complex glycosylation pattern, fe-tuins serve as model substances for lectin and glycoproteinresearch. Lectin binding should always be seriously consid-ered when fetuin binding to cells and to the extracellularmatrix is studied. On a practical note, the strong binding ofpertussis toxin to terminal sialic acid residues in fetuin formthe basis of a Food and Drug Administration-approvedpertussis toxin test.41,42 Fetuin-A sequestration of lectinsproved a major complication in experimental cytotoxic ther-apy using cancer cell–specific antibodies coupled to theRicinus communis agglutinin ricin.43 The immunotoxins wererapidly cleared by the asialoglycoprotein receptor and causedliver toxicity.44
In a search for natural transforming growth factor-beta(TGF-!) receptor antagonists, a sequence homology wasfound between TGF-! receptor type II and fetuin-A.45 TheTGF-! receptor II homology 1 domain from fetuin boundpreferentially to bone morphogenetic protein (BMP)-2. Full-length fetuin-A bound directly to TGF-!1 and TGF-!2 andwith greater affinity to the TGF-!–related BMP-2, BMP-4,and BMP-6. Finally, and likely impinging on fetuin’s role inmineralization biology, fetuin or neutralizing anti-TGF-!antibodies blocked osteogenesis and deposition of calcium-containing matrix in mineralizing cell cultures.46 An alteredbone phenotype in fetuin-A–deficient mice (Ahsg!/!) wasexplained accordingly in terms of failure to block TGF-!–dependent signaling in osteoblastic cells.47 Tumorigenesisexperiments using Ahsg!/! mice further supported the hy-pothesis that fetuin-A is an antagonist of TGF-! in vivo, inthat it inhibited intestinal tumor progression.48
Antiinflammatory Role of Fetuin-AFetuin-A, one of the most abundant fetal plasma proteins, wasfound to be essential for the inhibition of the proinflammatorycytokine tumor necrosis factor production by spermine and itssynthetic analogues.49,50 Accordingly, the strong fetal expres-sion of fetuin and spermine have been associated with thetolerance of the fetus, “nature’s transplant,” by mothers.51
The strong antiinflammatory effects of fetuin were verified invivo using several models of inflammation, includinglipopolysaccharide-induced miscarriage in rats,52 carrageenaninjection,53 cerebral ischemic injury in rodents,54 and cecalligation and puncture in mice.55 In all cases fetuin-A wasassociated with reduced inflammatory response and increasedsurvival, and administering additional fetuin generally im-proved outcome.
Thus fetuin-A generally may be regarded as antiinflam-matory.56 The antiinflammatory property of serum "2-HSglycoprotein/human fetuin-A was further supported by thedemonstration that fetuin-A is a potent and specificcrystal-bound inhibitor of neutrophil stimulation by hy-droxyapatite crystals.57 Calcium phosphate crystals induceproinflammatory cytokine secretion through the NLRP3inflammasome in monocytes/macrophages,58 – 60 cell deathin human vascular smooth muscle cells,61 and cell activa-tion in chondrocytes.62– 64 Antiapoptotic activity offetuin-A has been observed in smooth muscle cells65 anddampening of the cell-specific responses would generallybe expected to alleviate the detrimental consequences oflocal inflammation, cell death, and cartilage degradation.The proven protective function of fetuin-A in many animalmodels of inflammation,52–55 the inhibition of proinflam-matory compounds,50,51,66 – 68 and the inhibition of crystal-induced neutrophil activation57 collectively suggest thatfetuin-A may generally protect during pathological miner-alization as well.69
Figure 2. Cartoon of human fetuin-A/"2-HS glycoprotein showing cystatin-likedomains 1 and 2 and a third unrelateddomain in green, yellow, and blue shad-ing, respectively. The cotranslational andposttranslational modifications disulfidebridges (C-C in yellow) Ser-phosphoryla-tion sites (S in red), protease-sensitivesites (R-K, dibasic tryptic cleavage site;L-L, chymotryptic cleavage site; R-T,furin-sensitive cleavage site; all in green),allelic variants (T230/T238; M230/S238)depicted as orange asterisks, and Asn-linked complex N-glycosylation sites, andSer/Thr O-glycosylation sites depicted asblue symbols, respectively. Data takenfrom.6,7,13,96
Jahnen-Dechent et al Fetuin-A Regulation of Calcified Matrix Metabolism 1497
at University of Toronto on October 20, 2011http://circres.ahajournals.org/Downloaded from
!
!40!
Figure 3- Fetuin interaction with BCP. This figure illustrates the interaction of fetuin with mineral particles: the green-colored structure is the cystatin-like D1 domain; the yellow structure is the D2 domain; the blue structure is the D3 domain; the red structure refers to basic calcium phosphate 89.
respectively, all accelerate the particle ripening process,demonstrating that calciprotein particle maturation followsthe Arrhenius law. Secondary calciprotein particles are sta-bilized by a compact outer fetuin-A monolayer against furthergrowth (Figure 4E) for up to 30 hours at body temperature,which is ample time for clearing of calciprotein particles fromcirculation.
The formation and maturation of calciprotein particles canbe followed by optical monitoring110 of mineralizing solu-tions. Supplemental Figure I (available online at http://www.circresaha.org) shows dynamic light scatter diagramsillustrating that the transformation from primary to secondarycalciprotein particles is fetuin-A concentration-dependent.Higher fetuin-A concentrations cause prolonged stability ofboth calciprotein monomers and calciprotein particles, andthus a right-shift in the transformation curves. A clinical testcould use the kinetic parameters of the precipitation reaction,thus measuring the overall calcification risk in biologicalfluids, eg, in the blood of dialysis patients.
Protein–mineral complexes are soluble precursors of phys-iological mineralization of bones and teeth as well. Recentresearch has shown that biomineralization starts with amor-phous mineral–protein complexes containing fetuin-A.123–126
Price et al127–129 have shown that fetuin-A is criticallyrequired to direct mineralization to the interior of syntheticmatrices that have size exclusion characteristics similar tothose of collagen. Fetuin-A does so by selectively inhibitingmineral growth outside of these matrices. This mineralizationby inhibitor exclusion is likely redundant, because fetuin-A–deficient mouse bone is mineralized perfectly well. The micenevertheless show a bone phenotype of stunted femur length,indicating premature growth plate mineralization and dis-turbed osteogenic signaling.47
Nanobacteria: A Red Herring From MarsNanobacteria initially described nanoscopic life forms de-tected by electron microscopy in rock sediments.130 Similarentities were discovered in meterorites from Mars131 and,finally, in cell culture.132 Nanobacteria attracted a lot ofscientific and economical attention as a causal agent of majordiseases, including vascular calcification, kidney stones, andcancer.132–136 In the year 2000, Cisar et al137 determined thatsimple mixtures of phospholipids and calcium phosphatecrystals closely resembled nanobacteria in that they showedlife-like growth and replication. Nucleic acid sequencespreviously thought to be diagnostic markers of nanobacteriawere in fact diagnostic of common laboratory contami-nants.137 Two groups of researchers finally solved the riddleof pathological mineralizing nanobacteria.120,138 A series ofexperiments on the origin of putative nanobacteria showedthat calcium phosphate together with proteins and furthernonmineral compounds formed nanoparticles that resemblednanobacteria in shape and behavior.139 Minerals containingnanoparticles had a high binding capacity for charged mole-cules, including ions, carbohydrates, lipids, and nucleic acids.Depending on the exact composition, mixtures of mineral andnonmineral compounds sustained either crystallization ofhydroxyapatite mineral or the formation of complex protein–mineral complexes.122,140 It was found that the main proteincomponent of the nanobacteria was fetuin-A.120 Besidesfetuin-A, serum albumin and apolipoproteins were also iden-tified;121 hence, the term nanobacteria was exchanged for themore apt term calcifying nanoparticles, virtually identicalwith the protein–mineral complexes, calciprotein particles, orfetuin–mineral complexes described before.
Calciprotein Particles Metabolism and ClearingFigure 5 illustrates the putative metabolism of calciproteinparticles, soluble protein–mineral complexes, which are now
Figure 4. Formation of primary and sec-ondary calciprotein particles (CPP) fromfetuin-A monomer and amorphous mineralprecursors. A, Hydroxyapatite precursorviewed along the [001] axis. The dottedlines confine the hexagonal unit cell. Thecircle represents a Ca9(PO4)6 Posner clus-ter as a building block of the apatite struc-ture. B, A three-dimensional view of fivePosner clusters juxtaposed to the acidic!-sheet of the computer-modeled ami-noterminal fetuin-A domain D1. Thedomain surface is plotted semitransparentwith negative charge indicated in red andpositive charge indicated in blue. C, Acomputer-generated homology modelstructure illustrating the fetuin-A–mineralinterface. The fetuin-A protein structurewas modeled after published structures forcystatin-like domains 1 (green) and 2 (yel-low).176 Domain 3 (blue) was modeled afterthe Escherichia coli protein malonyl-CoA:acyl carrier protein transacylase(1MLA).177 Fetuin-A does not influence the
formation of mineral nuclei. However, fetuin-A does prevent the rapid growth and aggregation of nuclei and, thus, mineral precipita-tion.112 D, Primary CPP are spherical and rather unstructured agglomerates of mineral (clusters) and fetuin-A, whereas (E) secondaryCPP consist of a crystalline mineral core covered by fetuin-A. In conclusion, fetuin-A effectively shields the mineral phase, leading tostable CPP that can be transported and cleared, as shown in Figure 5. The computer graphics were generated using the softwarepackages VESTA,178 VMD,179 and APBS.180
1500 Circulation Research June 10, 2011
at University of Toronto on October 20, 2011http://circres.ahajournals.org/Downloaded from
!
!41!
Figure 4-Bragg’s formula147- 2d sinθ=nλ. λ: wavelength of the incident wave, d: spacing between the planes in the lattice, θ: the angle between the incident beam and the scattering planes
!
!42!
Figure 5A-Alizarin red staining of minerals and cetylpyridinum extraction for absorbance measurement. Non-tissue culture treated well plates were used to add the ions sequentially. The alizarin red was added to stain the mineral particles. Alizarin red was extracted using cetylpyridinium chloride. Absorbance measurements were done at 540 nm after 30 minutes. (24 well plate dish was used here for illustration purposes)
Phosphate Calcium Buffer Buffer All All
!
!43!
Figure 5B-Mineralization time course over 4 hours. Absorbance was measured at 540 nm every 30 minutes. The mean values+SEM were plotted versus time to obtain the time course of mineralization in the absence of fetuin. The peak value was obtained at 3.5 hours.
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
Abs
orba
nce
(540
nm)
Time (hrs)
Time Course with SEM - All Trials: 0uM Fetuin
!
!44!
Figure 6A-Mineralization time course over 4 hours with fetuin at 2 µM. Absorbance was measured at 540 nm every 30 minutes in the presence of fetuin at 2µM concentration. The mean values+SEM were plotted versus time to obtain the time course mineralization in the presence of fetuin and compare to the results from the control group of no fetuin. Same trend was noted with significantly smaller values in the presence of fetuin
!!
Fetuin Inhibition of Mineralization
7
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
Abs
orba
nce
(540
nm)
Time (hrs)
Optimize the Concentration of Fetuin to Inhibit Mineralization in vitro
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0 0.5 1 1.5 2 2.5 3 3.5 4
Ab
so
rba
nc
e (
54
0n
m)
Time (hrs)
Time Course for Mineralization
0µM
2µM
!
!45!
Figure 6B-Mineralization assessment at 3.5 hours in the presence of fetuin at various concentrations. In vitro mineralization assay was conducted in the presence of bovine fetuin at 0.5, 1.0, 1.4, 2.0 µM concentration. The mean+SEM values at 3.5 hours were used to plot the bar graph. As fetuin concentration increased, the absorbance measurement values decreased (dose-dependent).
Inhibition of Mineralization by Fetuin (3.5 hr incubation)
Abs
orba
nce
mea
sure
men
t
(540
nm
)
Bovine Fetuin Concentration (µM)
8
0.000
0.500
1.000
1.500
2.000
2.500
0uM 0.5uM 1uM 1.4uM 2uM
!
!46!
Figure 7A-TEM analysis of crystal species in the presence and absence of fetuin. TEM imaging was used to analyze the crystal species formed in the absence (control) and presence of fetuin at various concentrations. The needle shape of the mineral species suggests the hydroxyapatite nature of the crystal species
.
TEM Analysis Demonstrating Effects of Fetuin on Formation of Apatite Crystals
in Mineralization Assay
Control Fetuin- 0.5 µM Fetuin- 1.0 µM
Fetuin- 1.4 µM
Fetuin- 2.0 µM
!
!47!
Figure 7B-Electron diffraction analysis of crystal species. Electron diffraction analysis was also used to further confirm the presence of mineral species and their characteristics. The d-spacing values of the samples were compared to the gold standard database. The result confirmed the presence of hydroxyapatite formation Electron Diffraction Analysis Demonstrating Formation of
Apatite Crystals in Mineralization Assay System
!
!48!
Figure 8A-SDS-PAGE analysis of bovine fetuin in the presence of MMP-7 and MMP-3. Bovine fetuin was treated with MMP-7 and MMP-3 at a substrate to enzyme ration of 1:60 and was incubated overnight. The digestion of fetuin with MMP-7 seemed more pronounced than the digestion with MMP-3. MMP-7 and MMP-3 Cleavage of Fetuin
Native Fetuin MMP-7 MMP-3 95
MMP-Treated Bovine Fetuin
170 130
72
55 43 34
26
17
10
10
!
!49!
Figure 8B-SDS-PAGE analysis of human AHSG in the presence of MMP-7 and MMP-3Human AHSG was incubated with MMP-7 and MMP-3 1:60 ratio. MMP-7-mediated cleavage of human AHSG was largely complete after 24 hours but was less advanced with MMP-3.
Mr
1 H
R
2 H
R
4 H
R
8 H
R
24 H
R
0 H
R
2 H
R
4 H
R
8 H
R
24 H
R
1 H
R
MMP-7 MMP-3
72 kDa
34 kDa
26 kDa
55 kDa
14 kDa
Und
iges
ted
fetu
in
Dig
est 1
hr
Dig
est 2
4 hr
s
55 kDa
34 kDa
14 kDa
Dig
est +
Inhi
bito
r 1 h
r
Dig
est +
Inhi
bito
r 24
hrs
Fig. 2
A
B
43 kDa
92 kDa
MMP-Treated Human AHSG
!
!50!
Figure 8C-Mineralization in the presence of intact fetuin or MMP-7 and MMP-3 treated bovine fetuin. In vitro mineralization assays were conducted in the presence of no fetuin, 2 µM fetuin, MMP-3 degraded or MMP-7-degraded fetuin. The mineralization was significantly inhibited in the presence of intact fetuin and MMP-3-degraded fetuin, which is consistent with the incomplete digestion of fetuin by MMP-3.
Impact of MMP-treated Bovine Fetuin on Mineralization (2 µM; 4 hours)
Figure 8D-Mineralization in the presence of intact fetuin, MMP-7 and MMP-3- treated human fetuin. In vitro mineralization assays were conducted in the presence of no human fetuin, 2 µM fetuin, MMP-3-degraded fetuin and MMP-7-degraded fetuin. Mineralization was inhibited in the presence of intact fetuin and MMP-3-treated AHSG, and to a lesser extent by MMP-7-degraded fetuin.
0.000
0.500
1.000
1.500
2.000
2.500
3.000
3.500
4.000
No Fetuin 2uM Fetuin MMP3 Degraded Fetuin MMP7 Degraded Fetuin
Abs
orba
nce
Mea
sure
men
t (5
40 n
m)
Treatment
Impact of MMP-treated Human Fetuin on Mineralization (2 µM; 4 hours)
13
!
!52!
Copyright Acknowledgements
Portions of this work is in press in the Journal of Periodontal Research in 2012.