Nuclear Instruments and Methods in Physics Research B 364 (2015) 70–75
Contents lists available at ScienceDirect
Nuclear Instruments and Methods in Physics Research B
journal homepage: www.elsevier.com/locate/nimb
Determination of oxidation state of iron in normal and pathologically
altered human aortic valves
J. Czapla-Masztafiak a,⇑, G.J. Lis b, M. Gajda b, E. Jasek b, U. Czubek c, F. Bolechała d, C. Borca e,
W.M. Kwiatek a
a
Institute of Nuclear Physics PAN, Radzikowskiego 152, 31-342 Kraków, Poland
Department of Histology, Jagiellonian University Medical College, Kopernika 7, 31-034 Kraków, Poland
Department of Coronary Disease, Jagiellonian University Medical College, John Paul II Hospital, Pra˛dnicka 80, 31-202 Kraków, Poland
d
Department of Forensic Medicine, Jagiellonian University Medical College, Grzegórzecka 16, 31-531 Kraków, Poland
e
Swiss Light Source, Paul Scherrer Institute, 5232 Villigen PSI, Switzerland
b
c
a r t i c l e
i n f o
Article history:
Received 15 September 2014
Received in revised form 13 April 2015
Accepted 13 April 2015
Available online 24 April 2015
Keywords:
X-ray absorption spectroscopy
Iron
Aortic valve stenosis
a b s t r a c t
In order to investigate changes in chemical state of iron in normal and pathologically altered human aortic valves X-ray absorption spectroscopy was applied. Since Fe is suspected to play detrimental role in
aortic valve stenosis pathogenesis the oxidation state of this element has been determined. The experimental material consisted of 10 lm sections of valves excised during routine surgery and from autopsies.
The experiment was performed at the MicroXAS beamline of the SLS synchrotron facility in Villigen
(Switzerland). The Fe K-edge XANES spectra obtained from tissue samples were carefully analyzed and
compared with the spectra of reference compounds containing iron in various chemical structures. The
analysis of absorption edge position and shape of the spectra revealed that both chemical forms of iron
are presented in valve tissue but Fe3+ is the predominant form. Small shift of the absorption edge toward
higher energy in the spectra from stenotic valve samples indicates higher content of the Fe3+ form in
pathological tissue. Such a phenomenon suggests the role of Fenton reaction and reactive oxygen species
in the etiology of aortic valve stenosis. The comparison of pre-edge regions of XANES spectra for control
and stenotic valve tissue confirmed no differences in local symmetry or spin state of iron in analyzed
samples.
Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction
According to the report prepared by European Society of
Cardiology in 2003 [1] aortic valve stenosis (AS) is the most common valvular heart disease. AS is characterized by pathological
processes, leading to varying degrees of morphological changes
of the aortic valve leaflets, including fibrosis, accumulation of
lipids, calcification and occasionally bone tissue formation. All
these processes result in valve leaflet thickening, calcification and
inflexibility and clinical symptoms of aortic valve stenosis which
prevalence increase with age [2]. Over the years pathophysiologically the process was described as purely passive degeneration
progressively degrading the fine trilaminar valve leaflet architecture. Subsequently its substantial modification abilities have been
questioned. However, recent years brought emerging epidemiological, histopathological, and experimental evidences indicating
that the process in fact is actively regulated by various cells (e.g.
⇑ Corresponding author.
http://dx.doi.org/10.1016/j.nimb.2015.04.026
0168-583X/Ó 2015 Elsevier B.V. All rights reserved.
endothelial, inflammatory and activated interstitial cells), cytokines produced by these cells, matrix glycoproteins (e.g. matrix
metalloproteinases, bone morphogenetic proteins, tenascin) as
well as blood borne mediators (e.g. renin angiotensin system,
osteoprotegerin and receptor activator of nuclear factor jB ligand)
[3–7]. It is stressed that the histological appearance of stenotic aortic valves and risk factors of the disease resembles atherosclerotic
lesions [8].
Since valve replacement is eventually the only treatment
applicable for aortic valve disease [9], and grafted bioprosthetic
valves seem to undergo similar degeneration as observed in native
ones [5], it is extremely important to understand the mechanisms
involved in the process and to find out medical therapies able to
prevent or slow down its spread.
One of the suggested pathogenic mechanism of AS and other
major cardiovascular diseases is the activity of superoxide and
other reactive oxygen species (ROS) [8]. It was shown that superoxide concentration is increased in human stenotic aortic valves
[10]. Excessive production of ROS can cause damages of DNA,
J. Czapla-Masztafiak et al. / Nuclear Instruments and Methods in Physics Research B 364 (2015) 70–75
proteins and, which is particularly important in the initiation and
progression of AS, in lipid peroxidation. One of the chemical
mechanism that leads to ROS production is Fenton’s reaction,
where the iron ions are involved. Iron is crucial to maintain cellular
function and integrity, however, careful homoeostasis is also critical as both iron-deficient and overload states can lead to pathological changes in human body [11]. The role of iron has, inter alia,
been confirmed in the process of atherosclerosis [11,12].
In order to optimize Fenton reaction low pH values are required.
The rationale for the presence of acidic environment in the aortic
valve degeneration comes from observation that the process is significantly associated with inflammation. Inflammatory conditions
in various tissues are known to create locally acidic environment.
In calcific valve degeneration as well as atherosclerosis high infiltration of macrophages is observed. Activity of these cells, among
others, can acidify microenvironment creating optimal milieu for
Fenton reaction [13]. Recently we found the population of osteoclastic-like cells in stenotic valves [14]. These monocyte-derived
cells typically associated with bone tissue are able to release H+
ions resulting in local acidification and decalcification. What is
more, local intravalvular hypoxia as a consequence of pathological
valve thickening may results in increased anaerobic metabolism
and lactate production and in this way lowering pH in affected
tissues.
There are also evidences for the presence of redox-active iron
intracellularly in macrophageal lysosomes which are vesicles containing a set of acid hydrolases in naturally low pH (4–5) [15].
Since it was documented in atherosclerotic plaques it is reasonable
to suggest that in degenerating valves (a pathology akin to atherosclerosis) similar mechanisms are active as well.
The main objective of the presented study is to compare the
oxidation state and chemical environment of iron in stenotic and
normal human aortic valves using the possibilities offered by Xray absorption near edge spectroscopy (XANES). Similar approach
has been already used in case of, for example, studies of oxidation
state of iron in non-cancerous and cancerous prostate tissue sections [16] and the iron speciation in human nails [17]. Presented
analysis is complementary to the distribution studies of selected
elements by microscopy and SR-lXRF performed on the same samples [18].
2. Materials and methods
The examined material comprised 4 human aortic valves.
Valves were excised during routine surgery and from autopsies.
The samples included 2 calcified stenotic valves and 2 normal
(non-stenotic) valves. The study protocol was approved by the
Bioethical Committee of the Jagiellonian University Medical
College and the patients informed consent was obtained.
Tissue serial sections (10 lm-thick) were cut frozen on cryostat,
mounted on 3 lm-thick Mylar foil, dried and subjected to measurements. Additional sections were processed to histological
hematoxylin and eosin (HE) stained specimens.
Additionally, a set of powder reference compounds (both inorganic and organic), that represented different iron oxidation and
different bonding environment, was examined. Powders were
thoroughly grinded in a mortar with boron nitride (BN) at a mass
ratio of 1:10 in case of inorganic compounds and 1:4 in case of
organic one. The proportion of BN and reference compounds was
chosen in order to obtain good quality fluorescence spectra.
Homogeneous powders were placed into holes of the special plastic holder and placed at the experimental station.
XANES measurements of Fe K-edge were performed at the
MicroXAS beamline of the SLS synchrotron facility (PSI,
Switzerland), in fluorescence mode, in air. A single element Si(Li)
71
fluorescence detector (Ketek) was used. The energy was tuned with
the double crystal monochromator (Si(1 1 1) crystal) and the beam
was focused with Kirkpatrick–Baez mirror system. Full Fe K-edge
XANES spectra were measured on 5 different points on each of
the tissue samples in energy range 7015–7500 eV with acquisition
time of 1 s/energy point. In case of stenotic valves points were chosen so as to be near the focal calcifications, what was verified on
adjacent routinely stained (HE) histological specimens. The beam
was focused to the size of 10 lm 10 lm and 350 lm 350 lm
in case of tissue samples and reference samples, respectively. In
order to obtain good quality data 5 XANES spectra were recorded
on each reference sample and merged. In case of tissue samples
the number of spectra were limited to 3 at each point to avoid possible radiation damage. The photon flux during measurements was
about 2 1011 photons/s/lm2. The collected spectra were carefully examined and no changes in shape nor position were
observed between spectra taken at one point.
All the XANES spectra were analyzed using ATHENA software
package [19]. The background subtraction from the raw XANES
data was performed in ATHENA that determined it by optimizing
the low frequency components of Fourier transform of the data.
Then the data were normalized to the post-edge part. For reference
compounds the self-absorption correction was performed. Despite
the fact that the reference compounds have been mixed with boron
nitride in order to reduce the concentration of Fe atoms in samples,
the obtained XANES spectra shown the influence of self-absorption. For this reason the self-absorption correction algorithm,
offered as one of the features in Athena, was applied. The examples
of such corrections are presented in Fig. 1.
3. Results and discussion
The first step in the analysis of obtained data was the comparison of Fe K-edge spectra for valve samples and reference samples.
The spectra of reference samples with known chemical composition provide information on the shape of XANES spectra and
their position on the energy scale, that depends on the chemical
environment of iron. In the case of iron the energy position of
the absorption edge of the XANES spectrum can be determined
by both the oxidation and spin state of the atom [20,21]. The Fe
K-edge XANES spectra of reference compounds are presented in
Fig. 2.
It can be easily noticed that the shape of individual spectra varies depending on the local chemical environment of the iron in the
analyzed compounds. Moreover the energy shift of absorption
edge between individual spectra due to various oxidation and spin
state of iron (2+ or 3+ oxidation state, low or high spin) is observed.
The comparison of the Fe K-edge XANES spectra obtained for
control samples (K1 and K2), stenotic valve samples (S1 and S2)
and three of the reference compounds is presented in Fig. 3.
The XANES spectra obtained for the tissue samples are located
between the XANES spectra of Fe2+ and Fe3+ reference compounds
but closer to the spectrum of K2Zn[Fe(CN)6] that contains iron on
3+ oxidation state. This indicates that both chemical forms of iron
are presented in valve tissue but Fe3+ is the predominant form,
which is in agreement with the literature [22]. The shape of spectra
from tissue samples is similar to the one obtained for hemin – the
haeme derivative, in which iron has similar chemical environment
to the one in haemoglobin, myoglobin or cytochromes, that play
very important role in biological systems. Hemin contains haeme
structure with trivalent iron coordinated to four nitrogen atoms
and chloride ligand, positioned outside the haeme plane. Next shell
is built of carbon atoms [23]. In contrary to hemin structure, iron in
K2Zn[Fe(CN)6] and Na2[Fe(CN)5NO] is coordinated directly to carbon atoms of cyanide groups and one nitrogen atom in case of
72
J. Czapla-Masztafiak et al. / Nuclear Instruments and Methods in Physics Research B 364 (2015) 70–75
Fig. 1. The example of self-absorption (SA) effect correction performed in ATHENA software for Fe K-edge XANES spectra of FeS (left) and K2Zn[Fe(CN)6] (right) samples.
Fig. 2. Fe K-edge XANES spectra of chosen reference compounds. The numbers in
parentheses are the oxidation states of iron. Vertical line represents E0 value of
Fe2O3.
Na2[Fe(CN)5NO]. Different atomic structure is reflected in different
spectrum shape. In Fig. 3 one can also observe a small shift of the
absorption edge towards higher energies in the spectra from stenotic valve samples in comparison to the position of the absorption
edge of control samples. That may indicate higher content of the
Fe3+ form in pathological tissue.
The intensity of Fe K-edge corresponds to the number of unoccupied p states. This number is associated with coordinating
ligand, bonding strength and angle. Nevertheless the absorption
spectrum of tissue samples is a mixture of different iron K-edge
spectra from the iron-bearing compounds occurring in biological
systems. These compounds have different chemical environment
and the intensity of absorption edge is a weighted average of their
individual spectra. In comparison to the used reference compounds
the intensity of the absorption edge in tissue samples is between
the intensity of FeS and hemin. This stays in agreement with the
fact that both hemin-like structure and iron coordinated to sulfur
atom (e.g. in iron–sulfur proteins) are found in biological systems.
Therefore the intensity of absorption edge in case of tissue samples
would be affected mostly by changes in ratio of different iron-bearing compounds. In case of our samples the intensities of absorption
edge in K1, K2 and S1 spectra are almost the same. S2 spectrum has
slightly lower intensity of the absorption edge. This suggests that
composition of the studied tissues is similar.
In order to analyzed the shift of absorption edge in details the
edge position (E0) was determined from the maximum of the first
derivative of each of the spectra, as it is shown in Fig. 4.
The quality of the calculated first derivative was insufficient to
determine its maximum directly and therefore the obtained curve
was smoothed by using the Savitzky–Golay algorithm, and then fitted by Gaussian function. The procedure was performed in
OrginPro 8 software. The example of the first derivative of the Fe
absorption edge of stenotic valve sample together with smoothed
curve and fitted Gaussian function is presented in Fig. 4.
Absorption edge positions for reference, stenotic and control samples, calculated by presented procedure, are summarized in
Table 1. The error value was based on the energy step used during
experiment and the estimated share from the error of smoothing
and fitting procedure.
The analysis of obtained E0 values showed that the position of
absorption edge differs due to the various oxidation state of studied element (FeS and FeF3), but also due to its spin state (FeS and
Na2[Fe(CN)5NO]) and various chemical environment (hemin and
K2Zn[Fe(CN)6]). The edge positions for samples K1 and K2 differ
although these are both control samples, but obtained from two
different persons. Control samples are usually obtained from
autopsies and medical history is unknown. In some cases very
early stage of pathological conditions, that involve higher content
of oxidized iron, may occur. The higher-energy shift of about
0.5–1 eV of the absorption edge of iron in case of stenotic samples
compared to the control samples was observed. In case of samples
S1, S2 and K2 this difference is evident, while between samples S1,
S2 and K1 the difference is on the limit of statistical significance,
but with the tendency to higher energy shift. As it was mentioned,
one of the explanation of such a shift is an increased content of Fe3+
iron form that is produced, among other, in Fenton reaction – one
of the reactive oxygen species source in biological systems. ROS
damage DNA and proteins and lead to lipid peroxidation. In our
previous studies [18] we have shown that calcifications are
preferentially located in lipid-rich areas and that their presence
stimulates the formation of calcium nodules. All of these results
support the hypothesis that Fenton reaction plays an important
role in the AS etiology.
73
J. Czapla-Masztafiak et al. / Nuclear Instruments and Methods in Physics Research B 364 (2015) 70–75
Fig. 3. (a) Fe K-edge XANES spectra obtained for control valve samples (K1 and K2), stenotic valve samples (S1 and S2) and three chosen reference compounds with iron
oxidation state 2+ and 3+. (b) The absorption edge region of the spectra. (c) The comparison of the absorption edge region obtained for tissue samples and hemin.
in case of the transition between high symmetry and low symmetry structures. When the site of iron is centrosymmetric, the preedge structure is less intense than that of non-centrosymmetric
geometry [24]. In case of aortic valve samples, the pre-edge areas
extracted from XANES spectra were compared in order to check
if there are any changes in chemical environment of iron between
stenotic and control tissues (Fig. 5). The pre-edge region of tissue
samples was compared to the pre-edge of hemin.
Table 1
Calculated positions of absorption edges of reference compounds and tissue samples.
Fig. 4. First derivative of the experimental Fe K-edge XANES spectrum of stenotic
valve sample together with smoothed curve and fitted Gaussian.
Changes in chemical surrounding of an element can also lead to
the shift of XANES spectrum. The analysis of pre-edge feature, that
can be observed in XANES spectra obtained from aortic valves as
well as some of the reference compounds, can provide additional
information about chemical environment of studied atom and its
spin state. This kind of structure is attributed to forbidden
1s ? 3d electronic transitions and its intensity is sensitive to the
changes in local geometry around element of interest, especially
Reference
compounds
Position of absorption edge
(E0) [eV] ± 0.5 eV
Oxidation state
of iron
Spin
state
FeS
7121.9
2+
FeF3
7127.3
3+
Fe2O3
7125.3
3+
Fe2(SO4)3
7127.3
3+
Na2[Fe(CN)5NO]
7127.4
2+
K2Zn[Fe(CN)6]
7127.6
3+
Hemin
7126.0
3+
Highspin
Highspin
Highspin
Highspin
Lowspin
Lowspin
Lowspin
Sample
Position of absorption edge (E0) [eV] ± 0.5 eV
S1
S2
K1
K2
7126.6
7126.5
7126.1
7125.4
74
J. Czapla-Masztafiak et al. / Nuclear Instruments and Methods in Physics Research B 364 (2015) 70–75
chemical surrounding of iron in stenotic and non-stenotic aortic
valve samples.
The results of the analysis Fe K-edge XANES spectra obtained for
control and stenotic valve samples point out that in pathologically
changed tissue the share of iron with 3+ oxidation state is
increased. This kind of change can be related to the occurrence of
Fenton reaction, that leads to increased oxidation stress and tissue
damage. The increased amount of Fe3+ was determined based on
the position of the absorption edge in Fe K-edge XANES spectra.
Based on the comparison of the pre-edge region of XANES spectra
of stenotic and control tissue we concluded that there are no significant changes in spin state or atomic geometry around iron atom
between two types of tissue.
Acknowledgments
Fig. 5. Pre-edge region of Fe K-edge XANES spectra of stenotic (S1 and S2) and
control (K1 and K2) valve samples in comparison to the pre-edge region of hemin.
The extraction was done by fitting the absorption edge with sigmoidal function. This function was then subtracted from the full
XANES spectrum. The intensity, shape and position of pre-edges
in the spectra of all valve samples are similar. Based on that result
it was concluded that there is no significant changes in bonding
configuration and symmetry around Fe atom between pathological
and non-pathological tissue and average spin state, measured by
XANES, is the same. Pre-edge of hemin Fe K-edge spectrum is
shifted towards lower energies in comparison to the pre-edge of
tissue samples spectra. We assume that pre-edge of tissue samples
spectra can be decomposed into more than one component that
correspond to the iron phases present in biological systems.
Pre-edge analysis, next to the analysis of absorption edge position, is also very useful in determining the oxidation state of iron in
studied samples. As it is shown in Petit et al. [24] the shape and
position of pre-edge depend on both the coordination environment
and oxidation state of Fe. Unfortunately, in case of our samples
such detailed analysis, including linear combination fitting, was
impossible because of the large noise signal in our data. The oxidation state was determined only based on absorption edge position
and the pre-edge shape was analyzed only to show that there are
no major differences in this region, that suggests there are no significant changes in spin state or atomic geometry around iron atom
between two types of tissue.
In tissue sample one can observe mixture of two iron phases
Fe2+ and Fe3+ and the changes in the ratio that we can expect in
case of these samples are very subtle. Nevertheless the tendency
of edge position shift to higher energies in case of stenotic tissue
is visible. In case of biological samples changes can be very small
but even such small changes can modify the homoeostasis of
tissue.
4. Conclusions
The problem of diagnosis and treatment of cardiovascular diseases is one of the major topic of research for scientist worldwide,
mainly because of the extending lifetime and increasing number of
AS cases in aging population. The key to better and more effective
therapy is to know the exact mechanism leading to the formation
of the disease. The studies of variations in chemical structure
around important elements in human body may provide information about pathophysiology of the disease. In presented study the
XANES technique was used to establish the oxidation state and
We acknowledge Swiss Light Source at Paul Scherrer Institute
for granting the beamtime in the proposal 20120398. The research
leading to these results has received funding from the European
Community’s Seventh Framework Programme (FP7/2007-2013)
under Grant agreement n.°312284 (CALIPSO).
References
[1] B. Iunga, G. Baron, E.G. Butchart, A prospective survey of patients with valvular
heart disease in Europe: the euro heart survey on valvular heart disease, Eur.
Heart J. 24 (2003) 1231–1243.
[2] C.M. Otto, J. Kuusisto, D.D. Reichenbach, A.M. Gown, K.D. O’Brien,
Characterization of the early lesion of ‘degenerative’ valvular aortic stenosis,
histological and immunohistochemical studies, Circulation 90 (1994) 844–
853.
[3] S.H. Goldbarg, S. Elmariah, M.A. Miller, V. Fuster, Insights into degenerative
aortic valve disease, J. Am. Coll. Cardiol. 50 (2007) 1205–1213.
[4] K.D. O’Brien, Pathogenesis of calcific aortic valve disease: a disease process
comes of age (and a good deal more), Arterioscler. Thromb. Vasc. Biol. 26
(2006) 1721–1728.
[5] G.J. Lis, E. Jasek, M. Gajda, J.A. Litwin, U. Czubek, M. Jasińska, B. Kapelak, J.
Sadowski, Participation of tenascin C in native and homograft aortic valve
degeneration, Adv. Clin. Exp. Med. 20 (2011) 157–164.
[6] G.J. Lis, U. Czubek, M. Jasinska, E. Jasek, A. Loboda, J. Dulak, J. Nessler, J.
Sadowski, J.A. Litwin, Elevated serum osteoprotegerin is associated with
decreased osteoclastic differentiation in stenotic aortic valves, J. Physiol.
Pharmacol. 65 (2014) 377–382.
[7] N.M. Rajamannan, Update on the pathophysiology of aortic stenosis, Eur. Heart
J. Suppl. 10 (2008) E4–E10.
[8] D.D. Heistad, Y. Wakisaka, J. Miller, Y. Chu, R. Pena-Silva, Novel aspects of
oxidative stress in cardiovascular diseases, Circ. J. 73 (2009) 201–207.
[9] J.T. Butcher, G.J. Mahler, L.A. Hockaday, Aortic valve disease and treatment: the
need for naturally engineered solutions, Adv. Drug Deliv. Rev. 63 (2011) 242–
268.
[10] J.D. Miller, Y. Chu, R.M. Brooks, et al., Dysregulation of antioxidant mechanisms
contributes to increased oxidative stress in calcific aortic valvular stenosis in
humans, J. Am. Coll. Cardiol. 52 (2008) 843–850.
[11] G. Ramakrishna, T.W. Rooke, L.T. Cooper, Iron and peripheral arterial disease:
revisiting the iron hypothesis in a different light, Vasc. Med. 8 (2003) 203–210.
[12] B. de Valk, J.J.M. Marx, Iron, atherosclerosis and ischemica heart disease, Arch.
Intern. Med. 159 (1999) 1542–1548.
[13] D.S. Leake, Does an acidic pH explain why low density lipoprotein is oxidised
in atherosclerotic lesions?, Atherosclerosis 129 (1997) 149–157
[14] G. Lis, U. Czubek, M. Jasinska, E. Jasek, A. Loboda, J. Dulak, J. Nessler, J.
Sadowski, J.A. Litwin, Elevated serum osteoprotegerin is associated with
decreased osteoclastic differentiation in stenotic aortic valves, J. Physiol.
Pharmacol. 65 (2014) 377–382.
[15] T. Kurz, J.W. Eaton, U.T. Brunk, The role of lysosomes in iron metabolism and
recycling, Int. J. Biochem. Cell B 43 (2011) 1686–1697.
[16] W.M. Kwiatek, A.L. Hanson, C. Paluszkiewicz, M. Gałka, M. Gajda, T. Cichocki,
Application of SRIXE and XANES to the determination of the oxidation state of
iron in prostate tissue sections, J. Alloy. Compd. 362 (2004) 83–87.
[17] M. Katsikini, F. Pinakidou, E. Mavromati, E.C. Paloura, D. Gioulekas, D.
Grolimund, Fe distribution and speciation in human nails, Nucl. Instr. Meth.
B 268 (2010) 420–424.
[18] G.J. Lis, J. Czapla-Masztafiak, W.M. Kwiatek, et al., Distribution of selected
elements in calcific human aortic valves studied by microscopy combined with
SR-XRF: influence of lipids on progression of calcification, Micron 67 (2014)
141–148.
[19] B. Ravel, M. Newville, ATHENA, ARTEMIS, HEPHAESTUS: data analysis for X-ray
absorption spectroscopy using IFEFFIT, J. Synchrotron Radiat. 12 (2005) 537–
541.
J. Czapla-Masztafiak et al. / Nuclear Instruments and Methods in Physics Research B 364 (2015) 70–75
[20] J.E. Penner-Hahn, Characterization of ‘‘spectroscopically quiet’’ metals in
biology, Coord. Chem. Rev. 249 (2005) 161–177.
[21] M. Khalil, M.A. Marcus, A.L. Smeigh, et al., Picosecond X-ray absorption
spectroscopy of a photoinduced iron(II) spin crossover reaction in solution, J.
Phys. Chem. A 110 (2006) 38–44.
[22] J.L. Pierre, M. Fontecave, R.R. Crichton, Chemistry for an essential biological
process: the reduction of ferric iron, Biometals 15 (2002) 341–346.
75
[23] M. Walczak, K. Ławniczak-Jabłońska, A. Sienkiewicz, M. Czuba, M. Klepka, A.
Graczyk, The local atomic structure of di-alanine amino acid derivative of
protoporphyrin IX, J. Phys. Condens. Mater. 19 (2007) 285214–285225.
[24] P.-E. Petit, F. Farges, M. Wilke, V. Sole, Determination of the iron oxidation
state in Earth materials using XANES pre-edge information, J. Synchrotron
Radiat. 8 (2001) 952–954.