A shielding topology stabilizes the early stage protein-mineral complexes of fetuin-A and calcium phosphate: a time-resolved small-angle X-ray study.

Abstract

We report on the earliest stages of the formation of complexes of calcium phosphate in the presence of the serum protein alpha(2)-HS glycoprotein/fetuin-A termed calciprotein particles (CPPs). Time-resolved small-angle X-ray scattering (TR-SAXS) and stopped-flow analysis were used to monitor the growth of protein mineral particles nucleating from supersaturated salt solutions. It was found that fetuin-A did not influence the formation of mineral nuclei. However, fetuin-A did prevent the aggregation of nuclei and thus mineral precipitation. Hence, fetuin-A shielded spontaneously formed mineral nuclei, leading to stable calciprotein particles in the first stage of mineralization. Fetuin-A is therefore critically required during the earliest stages of the formation of protein-mineral complexes in order to prevent uncontrolled mineralization.

Full text

Page 1

1
FULL PAPERS
DOI: 10.1002/cbic.200((will be filled in by the editorial staff))
Stabilization of Calcium Phosphate Complexes by Fetuin-A
A Shielding Topology Stabilizes the Early Stage
Protein-Mineral Complexes of Fetuin-A and Calcium
Phosphate:
A Time Resolved Small Angle X-Ray Study
Christophe N Rochette[a], Sabine Rosenfeldt[a], Alexander Heiss[b], Theyencheri
Narayanan[c], Matthias Ballauff*[a], Willi Jahnen-Dechent*[b]
We report on the earliest stages of the formation of complexes of
calcium phosphate in presence of the serum protein α2-HS
glycoprotein/fetuin-A termed calciprotein particles (CPP). Time-
resolved small-angle X-ray scattering (TR-SAXS) and stopped-flow
analysis was used to monitor the growth of protein mineral particles
nucleating from super-saturated salt solution. It was found that fetuin-
A did not influence the formation of mineral nuclei. However, fetuin-A

prevented the aggregation of nuclei and thus mineral precipitation.
Hence, fetuin-A shielded spontaneously formed mineral nuclei leading
to stable calciprotein particles in the first stage of mineralization.
Thus, fetuin-A is critically required during the earliest stages of the
formation of protein-mineral complexes in order to prevent
uncontrolled mineralization.
Introduction
Nature controls the crystallization of inorganic material in a
process called biomineralization, resulting in organic-inorganic
hybrid materials with unique material properties. Mineralization
only occurs under well-controlled conditions, otherwise it must be
prevented in a suitable way. The handling of calcium phosphate
in mammalia illustrates an important point: the concentration of
calcium and phosphate ions in mammalian blood is much higher
than predicted by the solubility product for basic calcium
phosphate (1-4). Hence, calcium phosphate is expected to
precipitate all over the human body. However, mineralization is
usually restricted to bones and teeth. Recent work has identified
the serum protein α2-HS glycoprotein/fetuin-A as an important
inhibitor that prevents pathological mineralization in soft tissues
and in the extracellular fluid (2). The mechanism of this inhibition
is related to the formation of the so-called calciprotein particles
(CPP) (3,5-7). Once the solubility product is exceeded by the
concentration of Ca2+ and phosphate, colloidal CPPs are formed
that consist of amorphous calcium phosphate and fetuin-A. In
body fluids, acidic proteins like serum albumin further stabilize the
colloid (7). Studies by transmission electron microscopy (TEM)
and dynamic light scattering revealed that CPPs have diameters
of the order of 25 – 150 nm (3,6,7).
Employing small-angle neutron scattering (SANS) we recently
demonstrated that the initial CPPs transform to mature secondary
CPPs consisting of a core of octacalcium phosphate covered by a
shell of fetuin-A (6). The analysis by SANS, however, required at
least 30 minutes per sample. Hence, SANS did not permit the
study of early stages of CPP formation and critical information
about the early stages of mineral nucleation and the formation of
CPPs in the presence of fetuin-A is still missing. We considered
two alternative mechanisms of mineralization inhibition in the
early stage: i) Fetuin-A could act as a nucleating agent but
forming a large number of nuclei onto which the solid mineral
precipitates; this would effectively decrease the ion product
driving mineral formation, or
spontaneously forming mineral nuclei thus effectively preventing
their further growth and ultimately their precipitation. In the first
case the number N of the particles in the volume V should be
directly proportional to the fetuin-A concentration. At the same
time the size of the primary particles should decrease. If, on the
other hand, fetuin-A acted as a surface active component
stabilizing the particles and preventing further growth by
aggregation, N/V as well as the size of the primary particles
should not change markedly in the presence of fetuin-A.
Here we present a detailed study of the early stage of the CPP
formation using time-resolved small-angle X-ray scattering (TR-
SAXS). Calcium ions and phosphate ions were mixed rapidly in a
stopped-flow apparatus and the growth of mineral particles was
monitored by the increase and angular dependence of the
scattering intensity. As shown previously, the number density N/V
and particle radius R can be extracted from TR-SAXS data (8-10).
Measurements were taken without and with fetuin-A and the role
of this protein in the early stage of mineralization was assessed.
Together with previous data referring to longer time scales (3,6,7)
our present study provides a complete time-resolved mechanism
of the formation, maturation and stability of CPPs.
ii) fetuin-A could shield
[a] C. N. Rochette, Dr. S. Rosenfeldt, Prof. Dr. M. Ballauff
Physikalische Chemie I
University of Bayreuth
95444 Bayreuth (Germany)
Fax: (+49)921-55-2780
E-mail: Matthias.ballauff@uni-bayreuth.de
Dr. A. Heiss, Dr. W. Jahnen-Dechent
Department of Biomedical Engineering-Biointerface Group
RWTH Aachen University
52074 Aachen (Germany)
Dr. Theyencheri Narayanan
European Synchrotron Radiation Facility
38043 Grenoble Cedex (France)
[b]
[c]

Page 2

2

Results and Discussion
In order to study the effect of the glycoprotein on the formation of
calcium phosphate complexes, we have first done measurements
in the absence of fetuin-A. Fig. 1 shows the normalized SAXS
intensities up to t = 1.355 s. Here and in all subsequent scattering
curves the background has been subtracted. Except for a small
minimum around 0.28 nm-1, the decay of the scattering intensity
does not exhibit any pronounced minima or maxima. However, a
pronounced upturn of the scattering intensity is observed at
smallest q vectors (q < 0.15 nm-1). No change in the scattering
intensity is observed after ca. 0.255 s. We found that the
scattering patterns recorded at later time points (145 s after the
mixing process, data not shown) were similar to the ones at 0.27
s. In general, the intensities measured at intermediate and high q
range virtually agree beyond 250 ms and only differ in the region
of small q. Moreover no shift of the minimum was observed
anymore indicating that the formation of particles has completed.
Evidently, the formation of primary particles of calcium phosphate
is very fast. After this stage aggregation sets in leading to the
upturn at small q-values. Obviously, aggregation is beginning at
a very early stage directly after the formation of the primary
particles. A similar observation has been made in a recent study
of the formation of CaCO3 particles (8).
To study the effect of fetuin-A, the measurements were repeated
under the same experimental conditions in presence of 1, 5 and
15 µM fetuin-A, the latter corresponding to the physiological
concentration in plasma. The mineral ion concentrations used
here, i.e. the supersaturation, greatly exceeded their physiological
serum concentrations. We reasoned however, that local
concentrations, e.g. within the skeleton, may be much higher and
should still be effectively controlled by fetuin-A.
0.01
0.1
1
10
0.10.2 0.3 0.40.5
1.355 s
0.255 s
0.035 s
q [nm-1]
I(q) [mm-1]

Figure 1. Evolution of the normalized SAXS intensities in the absence of Fetuin-
A. Later SAXS intensities did not show any evolution of the shape or in intensity.
This demonstrates the very fast kinetic of early calcification leading to the first
equilibrium phase of CaPO4 particles.

The evolution of the SAXS intensities as a function of time is
displayed in Fig. 2 for the sample containing 15 µM fetuin-A. A
slight side maximum is evident. Moreover, the absolute scale and
the upturn of the scattering intensity at lowest q was lower with
than without fetuin-A (Fig. 1A). We conclude that the presence of
fetuin-A led to much smaller mineral particles.

0.01
0.1
1
10
0.10.20.30.40.5
6.415 s
0.695 s
0.255 s
0.035 s
q [nm-1]
I(q) [mm-1]

Figure 2. Evolution of the normalized SAXS intensities with time in presence of
15 µM of Fetuin-A. SAXS intensities measured at longer times did not show any
change in the shape either in intensities. We note here a slower formation of the
primary spherical particles compared to the sample without Fetuin-A (Fig. 1).


Figure 3 and 4 present a typical analysis of SAXS intensities
using eq. (6) at the times indicated. The smeared side maximum
of I(q) indicated polydispersity of the particles. In the region of
smallest q-values we find that data obtained without fetuin-A or at
low concentrations thereof cannot be fitted by assuming a
solution of non-interacting spheres, that is by assuming S(q) ≈ 1
in eq. (6) (see Section modelling). In these cases the deviation at
small q-values indicated aggregation of mineral nuclei as already
discussed in conjunction with Fig. 1 and 2. The resulting
scattering intensities was therefore modeled as a system of
aggregated spheres, that is by use of the structure factor S(q)
given in eq. (4). At elevated fetuin-A concentrations (e.g. 15 µM)
aggregation of primary particles was absent. Hence, under these
conditions the protein molecules seem to cover completely each
spherical particle and thus prevent further aggregation.
0.1
10
1000
0.350.70
q [nm-1]
I(q) [nm-1]

Figure 3. Normalized SAXS intensities 0.23 s after the mixing process. The
figure exhibits from bottom to top the scattering intensities of the samples
containing 0, 1, 5 and 15 µM of Fetuin-A. For the sake of clarity, intensities were
multiplied by a factor of 10 (1µM), 100 (5µM) and 1000 (15µM). For fitting the
sample with high amount of Fetuin-A (15 µM) S(q) was assumed to be unity,
that is, no aggregation took place under these conditions. For all other
concentrations aggregation was taken into account by a S(q)> 1 (see eq.(6);
section modelling).

Page 3

3
0.1
10
1000
0 0.20.40.60.8
x1000
x100
x10
15 µM
5 µM
1 µM
0 µM
q [nm-1]
I(q) [mm-1]

Figure 4. Modelling of the normalized SAXS intensities 0.89 s after the mixing
process. The figure exhibits from bottom to top the scattering intensities of the
samples containing 0, 1, 5 and 15 µM of Fetuin-A. For the sake of clarity,
intensities were multiplied by a factor of 10 (1µM), 100 (5µM) and 1000 (15µM).
For fitting the sample with high amount of Fetuin-A (15 µM) S(q) was assumed
to be unity, that is, no aggregation took place under these conditions. For all
other concentrations aggregation was taken into account by a S(q)> 1 (see
eq.(6); section modelling).
The fits shown in Fig. 3 and 4 allowed us to derive the
parameters number density (number of particles over volume,
N/V) and radius of the primary particles Rpsp as a function of time.
Both quantities were subject to considerable error since the
polydispersity of the primary particles was high (ca. 55% standard
deviation for the data shown in Fig. 1; 35% in the case of the data
shown in Fig.2). However, the data sufficed to observe the
general trends: Fig. 5 and 6 demonstrate that both N/V and Rpsp
increased with time but leveled off after less than 0.1s. The
number density increased by an order of magnitude at this very
early stage. Both the exceedingly fast increase of N/V and Rpsp
suggested that the primary particles form within a fraction of a
second.

5
7
9
11
13
00.2 0.4 0.6 0.81.0
15 µM
5 µM
0 µM
Time [sec]
Rpsp [nm]

Figure 5. Time-evolution of the radius of primary spherical particles without
(squares) and in presence of fetuin-A (triangles; filled circles). Data for 1 µM of
Fetuin-A were intermediate to 0 and 5 µM, but are omitted from the graph for
the sake of clarity. The dashed lines describe the general trend of the evolution
of the radius of primary spherical particles (Rpsp). We note here, within the limit
of error, a growth of the primary particles with the addition of the glycoprotein.
Interestingly, our recent measurements of the hydrodynamic
radius of the calciprotein particles at times beyond one minute
indicated that increased fetuin-A concentrations were associated
with smaller particles (7) suggesting substantial particle
aggregation between one second and several minutes when the
first measurements of our previous study took place.
1010
1011
00.2 0.4
Time [sec]
0.60.8 1.0
0 µM Fetuin-A
5 µM Fetuin-A
15 µM Fetuin-A
N/V [part.cm-3]

Figure 6. Time-evolution of the number density. Data for 1 µM of Fetuin-A were
intermediate to 0 and 5 µM, but are omitted from the graph for the sake of clarity.
The dashed lines describe the general trend of the evolution of the number of
particles per volume. We note here, within the limit of error, a decrease of the
number of particles per volume with the addition of the glycoprotein.
This corroborates the results presented here, namely a strongly
increased colloidal stability of the CPP by fetuin-A: The particles
aggregate much faster if less fetuin-A is in the system.

Moreover, Fig. 5 and 6 show that fetuin-A leads to the formation
of larger primary particles but with a smaller number density.
From 0 up to 5µM fetuin-A no significant differences in growth
parameters could be found. Data obtained in presence of the
physiological concentration of fetuin-A (15µM), however, clearly
showed that the number density N/V of the primary particles had
not increased in presence of the protein. As already discussed in
the Introduction, a nucleation agent would lead to a much
increased of N/V. The number density N/V had even decreased
with increasing fetuin-A concentration. Moreover, we found that
within the experimental error, the rate of formation of calcium
phosphate did not depend on the concentration of fetuin-A in the
solution. This may be directly inferred from plots of N/V·Rpsp
versus time t (data not shown). This quantity was proportional to
the entire amount of material produced as a function of time t.
The error bar of this product was rather high (ca. 20%) but there
was no significant deviation between curves measured for
different fetuin-A concentrations. Hence, we conclude that fetuin-
A is not acting as a nucleating agent. The small changes of N/V
and Rpsp point to a small effect of the protein to the growth
process at a very early stage of CPP formation.

Figure 7 illustrates the aggregation of primary particles in
presence of different amounts of fetuin-A in a quantitative manner.
The quantity S(0) + 1 is the number of primary particles per
aggregate. As demonstrated in Fig. 7, no aggregation was found
in solutions containing 15 µM fetuin-A. This finding is similar to
our recent results on the mineralization of calcium carbonate in
presence of water-soluble block copolymers (8), which likewise
suppressed aggregation of primary mineral particles. The water-
soluble block copolymer was attached to the surface of the
particles thus preventing further aggregation. The same
conclusion may be drawn here as well: Fetuin-A acts as a
surface-active agent that is attached to the surface of the
amorphous particles of calcium phosphate providing colloidal
stability against coagulation.
3

Page 4

4
0
5
10
15
0 0.20.40.60.81.0
0 µM
1 µM
5 µM
15 µM
Time [sec]
S(0) + 1 [∅]

Figure 7. Time-evolution of the average number of primary particles per
aggregate. As noted previously (Fig. 3 and 4), no structure factor was used for
fitting SAXS data of the sample containing 15 µM of Fetuin-A (S(0)+1=1). The
number of primary particles per aggregate decreases with the addition of
glycoprotein, demonstrating the inhibitor effect of the glycoprotein onto
calcification. The dashed lines are representing the general trend of the
evolution of the number of primary particles per aggregate.

Figure 8. Role of fetuin-A in the early stage of mineralization: Calciumphophate
particles are generated once the concentrations of the ions exceed the solubility
product. Virtually no influence of fetuin-A is seen in this early stage that cannot
be studied by the present experimental setup. The primary particles are then
stabilized by the protein by formation of the calciprotein particles. Without fetuin-
A, large aggregates are formed.
Figure 8 summarizes the results obtained by TR-SAXS: The
particles monitored in the very early stage in the TR-SAXS
experiment consist of amorphous calcium phosphate stabilized by
a surface layer of fetuin-A. The surface layer of Fetuin-A protein
positively identified in our recent SANS experiment (5) could not
be resolved by SAXS due to a lack of contrast (see section
Experimental). However, the assumption of a stabilizing surface
layer is in full agreement with previous studies by DLS and SANS
(6,7). We can now specify that “inhibition” of mineralization by
fetuin-A operates through stabilization of the primary particles
against coagulation rather than preventing mineralization
altogether. Once the concentrations of the ions exceed the
solubility product, the primary particles must precipitate. The
process of precipitation is fast (<0.1 s) and fetuin-A has little
influence in this very early stage. However, aggregation of these
primary particles having a size in the colloidal dimension
commences immediately thereafter. At this stage fetuin-A can
intervene by stabilizing the primary particles through formation of
the CPP. These particles are remarkably stable against further
aggregation.
In conclusion, we present a time-resolved synchrotron SAXS
study detailing the formation of calcium phosphate complexes
with the plasma protein fetuin-A. The most significant effect
observed was that fetuin-A stabilized the mineral particles
preventing their subsequent aggregation if the protein is present
in sufficient concentration. The calciprotein particles thus formed
exhibited a remarkable long-term stability (7).
Experimental Section
CPP Preparation
Calcium chloride (CaCl2·2H2O, Roth GmbH, Karlsruhe) and
sodium phosphate (Na3PO4·12H2O, Fluka) were analytical grade.
Solutions of 20 mM Ca2+ and 12 mM PO4
buffered solution (50 mM Tris/HCl, AppliChem, Darmstadt, pH =
7.4). Thus, the concentrations of Ca2+ and PO4
chosen in order to match the experimental conditions used in our
previous SANS study (5). Commercial bovine fetuin-A was
purified as described previously (3). The glycoprotein was diluted
into the calcium and phosphate solutions at concentrations of 0, 1,
5 and 15 µM, respectively.
All TR-SAXS experiments were performed at the high brilliance
undulator beamline ID2 of the ESRF, Grenoble, France (11). The
wavelength λ of radiation was chosen to be 1 Å. The SAXS
patterns were recorded by a high sensitivity fiber-optically
coupled CCD detector (FReLoN). The sample-to-detector
distance was set to 2 m. Data acquisition and counting of the time
t was hardware-triggered within 1 ms before the final mixing
process was initiated. SAXS data were acquired in increments of
170-520 ms with an exposure time of 20-50 ms per frame. By
repeating the experiment with different initial delay times the time
intervals between the readout gap of the detector (ca. 80 ms)
could be interpolated to achieve a better time resolution, when
required. In this way the reaction was thus followed during the
first 8 s after mixing.
The rapid and effective mixing of the CaCl2.2H2O and
Na3PO4.12H2O solutions was accomplished at 37°C by a
stopped-flow device (BioLogic SFM-3). The mixing volumes and
the mixer flow rates were controlled using the software of the
instrument. The total mixer flow rate and the capillary during the
final mixing phase were set to an optimum value of 6.67 ml/s.
After 30 ms of continuous mixing and flowing through the capillary,
the flow of the reagent mixture through the capillary stops and the
sample is left unperturbed. The kinetic time evolved above the
dead time of the device (~ 4 ms, that is the time needed to
transfer the mixture to the point of measurement in the capillary)
after the cessation of the flow. Thus for times t ≤ 35 ms there is a
quasi-steady-state condition due to the continuous flow of the
reaction mixture. The stopped-flow cell filled with the buffer was
taken as the background.


3- were prepared using a
3- solutions were
Modelling
The present analysis of the mineralization by TR-SAXS has been
described in detail previously (8-10). In short, the scattering
intensity I(q) of a system of non-interacting spherical particles can
be calculated by (12),

N
= I(q)
i
i

where the index i refers to the fraction of particles of radius Ri and
with a particle number density Ni/V. The scattering amplitude
Bi(q) is given by

∫
0
(q)
B
V
2
i⋅
∑
(1)
⋅−=
R
m
drr
qr
qr
rqB
2
) sin(
])([4)(
ρρπ
(2)

Page 5

5

where ρ and ρm represent the electron contrast of the particles
studied and of the solvent respectively.
For homogeneous spheres I(q) is given by

⎛
⋅⋅−⋅=
i
V
2
3
22
)(
) cos() sin(
3)()(
∑
⎟⎟
⎠
⎞
⎜⎜
⎝
−
qR
⋅
i
iii
im
i
qR qRqR
V
N
qI
ρρ
(3)

with Vi representing the volume per particle. The assumption of a
spherical shape of the CPP derived from our previous work (3)
showing that the CPPs have a spherical shape in the earliest
stage. Polydispersity was accounted for by assuming a
normalized Gaussian number distribution.
Interparticle interaction as well as aggregation was accounted for
by a structure factor S(q). The effect of S(q) is restricted to
smallest q region where a length scale of interparticle distances is
being probed. Fractal aggregation of the primary particles can be
modeled by the following expression for S(q) (13):

S
qS
D
1
+

where D is the fractal dimension and ζ is a crossover length that
is a measure of the aggregate size. The average number of
primary particles per aggregate is given by 1 + S(0).
Scattering contributions at high q region due to density
fluctuations within the particles may be described by an additive
Ornstein-Zernike term (10):

( )
22
1
q
ξ+

where ξ is the average correlation length, which is of the order of
a few nanometers.
Thus, the overall theoretical intensity Itot(q) is given by:

(
)()()(
IqIqSqI
tot
+=

The electron density of a particle ρ is directly proportional to its
mass density ρT. From the mass density and the chemical
composition, the electron density can be calculated. The density
ρT of the calciprotein particles was determined through
densitometry (DMA 60/602, Paar, Graz, Austria) to 1.72 g/cm3.
WAXS experiments done on particles obtained with and without
Fetuin-A (data not shown) did not exhibit any Bragg-peak
indicative of crystalline phases. Therefore we concluded that the
particles formed in the early stage were amorphous. This is in
agreement with previous work (6) demonstrating that an
amorphous precursor phase of calcium phosphate is formed first.
Similar findings have been reported for the early stage of the
formation of calcium carbonate particles (8-10). The nucleation of
the amorphous phase is a fast process and the formation of
crystalline modifications follows at a later stage.
The chemical composition of amorphous calcium phosphate
(ACP) is not known precisely (14-19). We approximated the
contrast of ACP by taking the average from the density of
crystalline tricalcium phosphate (TCP, Ca3(PO4)2, ρT = 2.89 g/cm3,
ρ = 864 e/nm3) and water (H2O, ρT = 1.00 g/cm3, ρ = 334 e/nm3).
By comparing the density ρT of the calciprotein particles with the
density and the electron contrast of crystalline TCP, the electron
density of ACP was estimated to be ρ = 536 e/nm3. As stated
above, the contrast of the particles is given by the excess
electron density that is the difference of the average electron
density within particles and the corresponding value of the solvent
water. For the CPPs this contrast was very high (202 e/nm3). The
excess electron contrast of the fetuin-A glycoprotein was
determined to be 74 e/nm3 (1.32 g/cm3 and 50.09 kDa by
analytical ultracentrifugation) (5). Therefore, minor differences of
the electron density within the particles by e.g. a preferred
[]
()()
[]
() ζ
q
1
−
ζ
q
ζ
D
D
q
tan1sin ) 0 (
1)(
1
2 / ) 1
−
(
22
−
+=
−
(4)
0
)(
I
qI
fluc
fluc
=
(5)
) )
q
(
fluc
(6)
location of the protein on the surface of the particles will go
unnoticed when measuring intensity. Therefore the CPP was
treated as homogeneous spheres (8).

Acknowledgements
We gratefully acknowledge the European Synchrotron Radiation
Facility (Grenoble, France) for the provision of synchrotron beam
time (SC2029), the German Research Foundation (Deutsche
Forschungsgemeinschaft, priority
Biomineralization”) and the Marie Curie Research and Training
Network (Polyamphi) for financial support.

program “Principles of
Keywords: Dynamic Light Scattering (DLS) · Small Angle
Neutron Scattering (SANS) · Transmission Electron
Microscopy (TEM) · Small Angle X-ray Scattering (SAXS)
[1] Schäfer, C., and Jahnen-Dechent, W. (2007) The Biological And Cellular
Role Of Fetuin Family Proteins In Biomineralization. In: Baeuerlein, E.,
Behrens, P., and Epple, M. (eds). Handbook of Biomineralization, Wiley-
VCH.
C. Schäfer, A. Heiss, A. Schwarz, R. Westenfeld, M. Ketteler, J. Floege,
W. Müller-Esterl, T. Schinke, W. Jahnen-Dechent, J. Clin. Invest. 2003,
112, 357-366.
A. Heiss, A. DuChesne, B. Denecke, J. Grötzinger, K. Yamamoto, T.
Renné, W. Jahnen-Dechent, J. Biol. Chem. 2003, 278, 13333-133411.
P. A. Price, G. R. Thomas, A. W. Pardini, W. F. Figueira, J. M. Caputo,
M. K. Williamson, J. Biol. Chem. 2002, 277, 3926-3934.
A. Heiss, W. Jahnen-Dechent, H. Endo, D. Schwahn, Biointerphases
2007, 2, 16-20.
A. Heiss, D. Schwahn, 2007 Formation and Structure of Calciprotein
Particles: the Calcium Phosphate-Ahsg/Fetuin-A Interface. In: Baeuerlein,
E., Behrens, P., and Epple, M. (eds). Handbook of Biomineralization,
Wiley-VCH.
A. Heiss, T. Eckert, A. Aretz, W. Richtering, W. van Dorp, C. Schäfer, W.
Jahnen-Dechent, J. Biol. Chem. 2008 283, 14815-14825.
J. Bolze, D. Pontoni, M. Ballauff, T. Narayanan, H. Cölfen, J. Colloid.
Interf. Sci. 2004, 277, 84-20.
J. Bolze, B. Peng, N. Dingenouts, T. Narayanan, M. Ballauff, Langmuir
2002, 18, 8364-8369.
[10] D. Pontoni, J. Bolze, N. Dingenouts, T. Narayanan, M. Ballauff, J. Phys.
Chem. B 2003, 107, 5123-5125.
[11] T. Narayanan, O. Diat, P. Bösecke, Nucl. Instrum. Methods Phys. Res. A
2001, 1005, 467-468.
[12] A. Guinier, J. Fournet, 1955 Small Angle Scattering of X-rays, Chapman
& Hall, London .
[13] H. Hoekstra, J. Mewis, T. Narayanan, J. Vermant, Langmuir 2005,
21,11017-11025.
[14] J. D. Termine, A. S. Posner, Archives of Biochemistry and Biophysics
1970, 140, 307-317.
[15] R. E. Wuthier, G. S. Rice, J. E. B. Wallace, R. L. Weaver, R. Z. LeGeros,
E. D. Eanes, Calcif. Tissue Int. 1985, 37, 401-410.
[16] H. E. Lundager Madsen, I. Lopez-Valero, V. Lopez-Acevedo, R. Boistelle,
Journal of Crystal Growth 1986, 75, 429-434.
[17] J. Christoffersen, M. R. Christoffersen, W. Kibalczyc, F. A. Andersen
Journal of Crystal Growth 1989, 94, 767-777.
[18] M. S. A. Johnsson, G. H. Nancollas, Critical Reviews in Oral Biology and
Medicine 1992, 3, 61-82.
[19] J. L. Meyer, E. D. Eanes, Calc Tissue Res. 1978, 25, 59-68.
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
Received: ((will be filled in by the editorial staff))
Published online: ((will be filled in by the editorial staff))

Page 6

6

Page 7

7
Entry for the Table of Contents (Please choose one layout)


FULL PAPERS



Christophe N Rochette, Sabine
Rosenfeldt, Alexander Heiss, Matthias
Ballauff*, Willi Jahnen-Dechent*
Page No. – Page No.
A shielding topology stabilizes the
early stage protein-mineral
complexes of Fetuin-A and calcium
phosphate: A time resolved small
angle x-ray scattering study