Formation of cyanogen iodide by lactoperoxidase

Article abstract of DOI:10.1016/j.jinorgbio.2015.11.005

The haem protein lactoperoxidase (LPO) is an important component of the anti-microbial immune defence in external secretions and is also applied as preservative in food, oral care and cosmetic products. Upon oxidation of SCN^- and I^- by the LPO-hydrogen peroxide system, oxidised species are formed with bacteriostatic and/or bactericidal activity. Here we describe the formation of the inter(pseudo)halogen cyanogen iodide (ICN) by LPO. This product is formed when both, thiocyanate and iodide, are present together in the reaction mixture. Using ¹³C nuclear magnetic resonance spectroscopy and gas chromatography-mass spectrometry we could identify this inter(pseudo)halogen after applying iodide in slight excess over thiocyanate. The formation of ICN is based on the reaction of oxidised iodine species with thiocyanate. Further, we could demonstrate that ICN is also formed by the related haem enzyme myeloperoxidase and, in lower amounts, in the enzyme-free system. As I^- is not competitive for SCN^- under physiologically relevant conditions, the formation of ICN is not expected in secretions but may be relevant for LPO-containing products.

Full text of DOI:10.1016/j.jinorgbio.2015.11.005

Journal of Inorganic Biochemistry 154 (2016) 35–41
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry
journal homepage: www.elsevier.com/locate/jinorgbio
Formation of cyanogen iodide by lactoperoxidase
a,
a
b
a
Denise Schlorke ⁎, Jörg Flemmig , Claudia Birkemeyer , Jürgen Arnhold
a
Institute for Medical Physics and Biophysics, Medical Faculty, University of Leipzig, Leipzig, Germany
Institute of Analytical Chemistry, Department of Chemistry and Mineralogy, University of Leipzig, Leipzig, Germany
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 8 July 2015
Received in revised form 23 September 2015
Accepted 2 November 2015
Available online 3 November 2015
The haem protein lactoperoxidase (LPO) is an important component of the anti-microbial immune defence in ex-
ternal secretions and is also applied as preservative in food, oral care and cosmetic products. Upon oxidation of
SCN and I by the LPO–hydrogen peroxide system, oxidised species are formed with bacteriostatic and/or bac-
−
−
tericidal activity. Here we describe the formation of the inter(pseudo)halogen cyanogen iodide (ICN) by LPO. This
1
3
product is formed when both, thiocyanate and iodide, are present together in the reaction mixture. Using C nu-
clear magnetic resonance spectroscopy and gas chromatography–mass spectrometry we could identify this
inter(pseudo)halogen after applying iodide in slight excess over thiocyanate. The formation of ICN is based on
the reaction of oxidised iodine species with thiocyanate. Further, we could demonstrate that ICN is also formed
Keywords:
Cyanogen iodide
Iodide oxidation
Lactoperoxidase
Myeloperoxidase
Thiocyanate oxidation
−
by the related haem enzyme myeloperoxidase and, in lower amounts, in the enzyme-free system. As I is not
competitive for SCN− under physiologically relevant conditions, the formation of ICN is not expected in secre-
tions but may be relevant for LPO-containing products.
©
2015 Elsevier Inc. All rights reserved.
1
. Introduction
is in the low micromolar range [9]. Thus, a significant contribution of io-
dide to immune defence in secretory epithelia is unlikely.
Primarily isolated from bovine milk, the haem protein lacto-
Due to the low price and easy availability, the anti-microbial LPO
system is widely applied as preservative in food as well as in oral care,
cosmetic and other products [2,11]. In these applications, LPO is used to-
gether with a hydrogen peroxide generating system and thiocyanate
peroxidase (LPO) is a crucial component of immune defence against
pathogens [1]. Chemically and immunologically very similar proteins,
here also referred to as lactoperoxidase, are found in secretions from
human mammary, salivary, and lacrimal glands, and secretory glands
of the upper airways, where they exert an anti-microbial activity to-
gether with hydrogen peroxide and thiocyanate [2]. Hydrogen peroxide
activates ferric LPO to the oxo-ferryl species Compound I that oxidises
−
and/or iodide. The oxidation of I by Compound I of LPO results in the
formation of hypoiodous acid (HOI), with a pK
a
of 10.6 [12] and other
−
−
−
oxidised iodine species such as I
, I
2 3
,
2
OI, and I OH [13]. In contrast
to hypothiocyanite, oxidised iodine products are directed against a
broader range of molecules: They target peptides and proteins with
sulfhydryl and thioether moieties and incorporate iodine into tyrosine
−
thiocyanate to hypothiocyanite ( OSCN) by abstracting two electrons.
8
−1 −1
The rate of this oxidation is 2 × 10 M
s
at pH 7.0 and 15 °C [3].
This species is in equilibrium with hypothiocyanous acid (HOSCN)
with a pK value of 5.3 [4]. The weak oxidant HOSCN/ OSCN is able to
a
residues and oxidise NAD(P)H [12,14,15]. While the LPO–H
2 2
O system
−
exhibits a bacteriostatic activity against Escherichia coli in the sole
−
−
permeate through cell membranes and penetrates into microorganisms
of biofilms, where it preferentially oxidises low molecular weight thiols
and inactivates enzymes with functional sulfhydryl groups [5,6]. This ef-
ficient anti-microbial activity is further supported by the high yield of
presence of SCN , this system acts bactericidal when I is used instead
[16,17]. A further study demonstrated the antiviral activity of the LPO–
−
H O
2
2
–I system against the respiratory viral pathogens adenovirus and
−
respiratory syncytial virus whereas the LPO–H
2 2
O –SCN was ineffec-
−
SCN in secretions. In saliva, thiocyanate levels are around 0.5–4 mM,
tive [18].
which strongly depend on diet, smoking habit, and time of day [7–9].
The stronger bactericidal effect of the LPO–H O system in the pres-
2 2
−
−
−
In upper airway fluids, about 400 μM SCN is present [10]. The LPO–
ence of I as compared to SCN is also demonstrated by experiments
where both ions were applied together: Increasing amounts of thiocya-
nate reduced or abolished the bactericidal effect of I− [19–21]. In
−
8
−1 −1
H O
2 2
system also oxidises I with a high rate of 1.2 × 10 M
s
at
pH 7.0 and 15 °C [3]. However, in secretions the iodide concentration
contrast, the addition of 2.5 mM potassium iodide to a mixture of
−
LPO–H
2
O
2
–saliva (containing 0.3–2.5 mM SCN ) strongly reduced the
Abbreviations: GC–MS, gas chromatography–mass spectrometry; HS, headspace; LPO,
lactoperoxidase; MPO, myeloperoxidase; NMR, nuclear magnetic resonance.
Corresponding author at: Institute for Medical Physics and Biophysics, Medical
Faculty, University of Leipzig, Härtelstr. 16-18, 04107 Leipzig, Germany.
viability of Fusobacterium nucleatum (60 ± 22%) compared to LPO–
–saliva alone (89 ± 2%), but was only slightly more effective
than an enzyme-free mixture (KI–H –saliva, 66 ± 18%) [22]. A sup-
plementation with iodide was also more effective in the case of Candida
2 2
H O
⁎
2 2
O
E-mail address: denise.schlorke@medizin.uni-leipzig.de (D. Schlorke).
http://dx.doi.org/10.1016/j.jinorgbio.2015.11.005
0162-0134/© 2015 Elsevier Inc. All rights reserved.

36
D. Schlorke et al. / Journal of Inorganic Biochemistry 154 (2016) 35–41
albicans, Escherichia coli and Staphylococcus aureus as compared to the
which was set to −3.1 ppm. The Bruker software Topspin™ 2.1 was
used for data analysis.
−
2 2
sole LPO–H O –SCN system [23].
It remains unknown whether the results of the mixed application of
SCN− and I with the LPO–H
−
O
system can solely be explained by
2.4. Headspace gas chromatography–mass spectrometry
2
2
competitive effects between these ions for Compound I, or whether in
addition to hypothiocyanite and oxidised iodine species some yet
unknown inter(pseudo)halogens are formed that may contribute to
the cytotoxic effect. Therefore, we analysed in the present work
Headspace (HS) GC–MS analysis was carried out on a Shimadzu QP-
2010 GC-EI-quadrupole-MS equipped with a DB5-MS capillary column
(J&W Fisher, 30 m, 0.25 mM id, 0.25 μM film) connected to an HS-20
headspace injector (all modules from Shimadzu, Kyoto, Japan). Head-
space injection was done in loop mode (0.5 min load time and 1 min in-
jection time) with helium as carrier gas at a column flow of 0.95 mL/min
and a split of 10:1. The HS oven was set to 60 °C, sample and transfer line
to 160 °C with a pressurising gas pressure of 90 kPa and 1 min
pressurising time. The GC programme started at 35 °C, held for 1 min
and ramped with 10 °C/min to 75 °C. The electron impact ion source op-
erated at 200 °C and 70 eV scanning from m/z 45–300.
−
−
13
2 2
the species formed by the LPO–H O –SCN /I system by C nuclear
magnetic resonance (NMR) spectroscopy and gas-chromatography–
mass spectrometry (GC–MS). We could identify and characterise the
inter(pseudo)halogen cyanogen iodide (ICN) as a yet unknown
lactoperoxidase product. This product was also formed by applying
the haem protein myeloperoxidase (MPO) or an enzyme-free system.
2
. Materials and methods
2
.1. Materials
3. Results
2 2
3.1. Analysis of the LPO–H O
–SCN–I− reaction by C NMR spectroscopy
13
Lactoperoxidase from bovine milk was purchased from Sigma-
Aldrich (Deisenhofen, Germany). 800 μM aliquots of LPO were prepared
in 100 mM potassium phosphate buffer and stored at −25 °C.
Enzyme purity was 0.8 (absorbance ratio 412 nm/280 nm). Its con-
Mixed preparations of SCN− and I− together with the LPO–H
O sys-
2 2
tem exhibits stronger and broader anti-microbial effects than the sole
−
1
−1
centration was determined using ε412 = 112,000 M
Human myeloperoxidase was obtained from Planta Natural Products
Vienna, Austria) at a purity of 99.5% (according to the absorbance
ratio 430 nm/280 nm). Its concentration was determined using ε430
cm
[3].
application of thiocyanate [19–21,23]. In order to analyse whether a
13
new oxidation product is formed in this system, we used C NMR spec-
troscopy to identify the nature of prospective 13C-labelled products. C-
13
(
=
2 2
labelled thiocyanate (133.4 ppm) was oxidised by the LPO–H O system
−
1
−1
13
13
−
9
1,000 M
cm
per haem [24]. C-enriched ( C 95–99%) and
C– N-enriched ( C 99%, N 98%) potassium thiocyanate was
to hypothiocyanite ( OSCN) with a chemical shift of 126.8 ppm
(Fig. 1A). There was also a small peak for cyanate at 128.2 ppm. The for-
1
3
15
13
15
−
−
−
obtained from Cambridge Isotope Laboratories (Tewksbury, USA)
at a purity of 95%. Unlabelled cyanogen iodide (97.5% purity)
was purchased from Acros Organics (Geel, Belgium). Deuterium
oxide, hydrogen peroxide as a 30% solution, potassium iodide (99% pu-
rity) and all other chemicals were purchased from Sigma-Aldrich
mation of OSCN and OCN by the LPO–H
2
O
2
–SCN system agrees
−
well with literature data [26]. Chemical shift data given for OSCN/
HOSCN range from 126.5 ppm to 127.7 ppm and depend apparently
on pH [26–28].
In the additional presence of iodide (Fig. 1B-F), the peak for −OSCN
completely vanished and a new, singlet signal appeared with a chemical
shift of approximately 51.0 ppm. With increasing iodide concentration,
the position of this signal shifted slightly downfield from 51.0 ppm
(
Deisenhofen, Germany). The concentration of a stock solution of hy-
drogen peroxide was spectroscopically determined immediately prior
1
−1
to use (ε230 = 74 M cm , [25]).
−
−
− −
(
SCN :I = 1:2) to 53.2 ppm (SCN :I = 1:8). In Fig. 1G an overlay
2
.2. Peroxidase-dependent and -independent oxidation of SCN− and I−
of the spectra shown in (B–F) is given, illustrating again the dependence
−
−
of the position of the new product on the applied SCN /I ratio.
The direct comparison of the spectra indicates also different product
yields with increasing iodide. In fact, as shown in Fig. 2A, in case of a 1:1
KSCN (40 mM) was mixed with KI in different ratios (1:1–1:8) in
0
.1 M phosphate buffer (pH 7.0), at 22 °C in the presence or absence
−
−
−
of 4 μM LPO. Ten portions of hydrogen peroxide (final concentration
0 mM) were added every 30 s to initiate (pseudo-)halide oxidation. Af-
ratio for SCN /I neither OSCN nor the signal at 51–53 ppm was de-
tected, whereas a maximum intensity of the 51–53 ppm signal was ob-
served at a 1:2 ratio. At higher ratios, its generation decreased. This
signal did not change its intensity over 2 h upon storage of samples at
6 °C (data not shown), indicating a considerable stability of the corre-
sponding new product under the chosen experimental conditions. By
1
terwards, the samples were measured immediately. In experiments
concerning the pH dependence of the reaction, 100 mM phosphate buff-
er (pH 6.0–8.0) or 200 mM/100 mM citrate-phosphate buffer (pH 5.0)
was used.
In samples in which LPO should be removed, the solution was fil-
tered through a filter with a cut-off of 30 kDa (VWR International,
Darmstadt, Germany) by centrifugation (8 min, 15,000 ×g). In selected
13
comparing the integral intensities of signals in C NMR spectra, and
−
using the signal at 133.6 ppm for SCN (40 mM) as reference, the prod-
uct at 51–53 ppm has a concentration of 5.6 ± 0.4 mM (n = 3) at a 1:2
−
−
experiments, the LPO–H
2
O
2
-reaction was performed in the sole pres-
ratio between SCN and I .
−
−
−
−
ence of SCN or I . After filtration, either I (to oxidised SCN ) or
We further analysed which influence the pH value has on the forma-
tion of the 51–53 ppm product (Fig. 2B). A decrease in pH to 5.0 and 6.0
reduced the amount of detectable product as compared to pH 7.0 while
an increase to 8.0 slightly increased the product formation. The higher
values observed under more alkaline conditions may indicate a more ef-
ficient product formation or its higher stability.
−
−
SCN (to oxidised I ) was added to yield a final ratio of 1:2 between
−
−
SCN and I .
In similar experiments, MPO was used instead of LPO. Selected con-
trol experiments were also performed in the absence of peroxidases. If
13
13 15
mentioned in the text, C- or
C N-labelled thiocyanate was used.
13
2 2
3.2. Formation of the product at 51–53 ppm by the LPO–H O -
–SCN–I−
2
.3. C NMR measurements
system is based on iodide oxidation
13C NMR measurements were performed at a Bruker Avance
III 600 MHz spectrometer (Bruker Biospin GmbH, Rheinstetten,
Germany) equipped with a 5 mM broad-band probe. 10% D O was
used as frequency lock. All measurements were performed at 6 °C.
Chemical shifts were referenced to trimethylsilyl propanoic acid,
Next, we analysed whether the oxidation of SCN− or I− is required
for the formation of the new product. The lactoperoxidase reaction
was run first with S13CN . After removing the enzyme by ultrafiltration,
2
−
−
13
the double amount of I was added. In the C NMR spectrum, only the

D. Schlorke et al. / Journal of Inorganic Biochemistry 154 (2016) 35–41
37
1
3
–SCN− reaction products in dependence on the I− concentration. 40 mM S13CN− was mixed with 0 mM (A), 40 mM (B), 80 mM (C), 160 mM (D),
40 mM (E) or 320 mM (F) I− (corresponding to ratios of 1:1 (B) up to 1:8 (F)) and 4 μM LPO in 100 mM phosphate buffer, pH 7.0 at 22 °C. (Pseudo)halide oxidation was started by H
O .
2 2
Ten portions of H O (final concentration 10 mM) were added every 30 s. C NMR measurements were conducted at 6 °C. (G) is an overlay of the spectra shown in (C–F). Representative
2 2
2 2
Fig. 1. C NMR spectra of the LPO–H O
2
13
spectra of three independent experiments are given.
signals for −OSCN (126.8 ppm) and −OCN (128.2 ppm) were detected
3.3. Formation of the product at 51–53 ppm by myeloperoxidase and in the
enzyme-free system
−
(
Fig. 3A). Contrary, when two portions of I were applied at first togeth-
13
−
2 2
er with LPO and H O , followed by addition of one portion S CN , the
product with the peak at 51.0 ppm was formed (Fig. 3B). Thus, this prod-
uct is derived from the interaction of oxidised iodine species with thiocy-
anate and a role of hypothiocyanite in this reaction can be excluded.
Instead of LPO we next tested whether the haem protein MPO per-
forms the similar reaction. This enzyme is also well known to oxidise
both, I and SCN [24]. As with LPO, the MPO–H O system did not
2 2
−
−

38
D. Schlorke et al. / Journal of Inorganic Biochemistry 154 (2016) 35–41
Fig. 2. Dependence of the intensity of the newly formed product at 51–53 ppm on the SCN−/I− ratio (A) and pH (B). (A) Samples were prepared as given in Fig. 1. In (B), the LPO-driven
reaction was performed at pH values from 5.0 to 8.0 at a ratio of 1:2 SCN− to I . Citrate-phosphate-buffer (200 mM/100 mM) at pH 5.0, or a 100 mM phosphate buffer at pH 6.0 to 8.0 was
applied. Mean and standard deviation of three independent experiments are given.
−
produce the new product at a 1:1-ratio of SCN and I− (Fig. 4A). How-
−
Moreover, the 13C NMR chemical shift of ICN was determined to be
50.39 ppm [31]. Hence, we analysed commercially available unlabelled
ICN by 13C NMR and actually found a small peak with a chemical shift at
50.1 ppm for the pure compound (Fig. 5). Interestingly, we observed the
ever, using a ratio of 1:2 between S13CN and I , a comparable strong
−
−
signal at 51.0 ppm was detected as observed for LPO. Again, its intensity
was reduced at a 1:4-ratio. Alternatively, this peak also appeared when
−
−
first I was oxidised by the MPO–H
O
2 2
system followed by addition of
same pattern of downfield shifting upon addition of I as in experi-
13
−
13
−
−
S
CN
,
but not, when first S CN and then I were added (data not
ments with LPO: additions of KI in two portions shifted the position of
−
shown). These results are also in line with those observed for LPO.
Similar experiments were also performed in the absence of LPO or
this peak to 51.1 ppm (ICN/I ratio 2.2:1) and further to 51.8 ppm
−
(ICN/I ratio 1.1:1). The only difference to the LPO system was a broad-
−
13
−
MPO. The addition of H
O
2 2
to a mixture of I and S CN caused also
ening of the peak. This is apparently caused by using unlabelled ICN.
the formation of this product between 50.6–53.2 ppm for all tested
−
−
SCN /I ratios (1:1–1:8). However, in comparison to the enzyme sys-
3
.5. Identification of ICN by gas chromatography mass spectrometry as LPO
−
−
tem, the product was already formed when a 1:1-ratio of SCN and I
was applied, but the yield of this product was generally lower
Fig. 4B). At a 1:2-ratio, its amount was by 73% lower in the enzyme
free-system than in the LPO-dependent reaction. As in the enzyme-
dependent reactions, we observed the highest yield at a SCN /I ratio
of 1:2. At higher concentrations the production was reduced, yet, with-
out concentration dependence.
product
(
In order to confirm the formation of ICN as reaction product of the
−
LPO–H
Fig. 6). Similar to C NMR spectroscopy measurements, we did not
observe any product formation when equimolar concentrations of
2 2
O –SCN–I system, we performed headspace GC–MS analysis
−
−
13
(
−
−
SCN and I were applied (Fig. 6A). However, when we altered the
−
−
Applying a rough quantitation comparing the integral intensities in
NMR spectra, the product at 51–53 ppm has a concentration of 7.7 ±
SCN /I ratio to 1:2 (Fig. 6B), we detected a product whose identity
was approved as ICN by comparing the retention time and mass
spectrum with an authentic standard (not shown). A further proof
0
2 2
.3 mM (n = 3) for the MPO–H O system at a 1:2 ratio between
−
−
was provided when we applied S13CN (Fig. 6C) and S
−
13 15 −
C N
SCN and I . At this anion ratio, 1.5 ± 0.4 mM of this product were pro-
duced in the enzyme-free system.
+
•
(
Fig. 6D). In these corresponding spectra, the molecular ion of ICN
detected at an m/z-value of 152.9 shifted to m/z 153.9 and m/z 154.9
due to enhanced molecular masses of 1 u and 2 u, respectively, for the
labelled molecular ions. The peaks at m/z 76.5 (Fig. 6B), 77.0 (Fig. 6C)
and 77.5 (Fig. 6D) are suggested as the corresponding, doubly charged
molecular ions (ICN2 ), confirmed by their isotopic pattern (not
1
3
3
.4. Identification of the product at 51–53 ppm by C NMR spectroscopy
Upon interaction of SCN− with
I , the formation of the
2
+
inter(pseudo)halogen cyanogen iodide (ICN) was described [29,30].
Fig. 3. Effects of the sequential addition of SCN and I− on the formation of the novel product. (A) C NMR spectra were taken after the addition of I to the ultrafiltrate of the LPO–H
S
−
13
−
2 2
O –
13
−
13
−
2 2
O
–I system. In both cases, a ratio of 1:2 between SCN− and I was applied. Representative spectra of three
−
−
CN system. In (B), S CN was added to the ultrafiltrate of the LPO–H
independent experiments are given.

D. Schlorke et al. / Journal of Inorganic Biochemistry 154 (2016) 35–41
39
Fig. 4. Formation of the new product in the presence of MPO and in the enzyme-free system. (A) 40 mM S13CN− was mixed with 40–160 mM I− and 4 μM MPO in 100 mM phosphate
2 2
buffer, pH 7.0 at 22 °C. Upon addition of H (final concentration 10 mM), the reaction was started. The intensity of the signal at 51–53 ppm is given in arbitrary units. (B) H O
2
O
2
−
13
−
(
10 mM) was incubated with I (40–320 mM) and S CN (40 mM). Mean and standard deviation of three independent experiments are given.
shown). Furthermore, we could observe a fragmentation pattern with
following equation was given for the reaction in neutral and slightly
basic media:
+
+
an I in all spectra at m/z 127.0 as well as m/z 139.0 for IC (m/z
1
40.0 for I1 C ) derived from the cleavage of nitrogen. The latter species
3
+
−
13
−
−
2−
þ 7I þ 8 Hþ þ ICN:
−
formed for SCN samples at m/z 139.0 and at m/z 140.0 for the S CN
4I þ SCN þ 4H O→SO
2
2
4
1
3 15 −
and S
C N samples.
Taken together, the data comprising 13C NMR spectra and headspace
Under strong acidic conditions, a protonated complex (HI
2
SCN) be-
GC–MS analysis reveal that cyanogen iodide is indeed a reaction prod-
−
tween I
2
and SCN is formed, from which the inter(pseudo)halogen
−
−
uct formed by LPO and MPO when SCN and I are both present in
the reaction mixture with H
ISCN is derived [34,35]. However, under our neutral experimental con-
ditions, we did not detect any other interhalogen species than ICN.
2 2
O .
13
The slight downfield shift of the C NMR signal for ICN at increasing
4
. Discussion
The anti-microbial lactoperoxidase–hydrogen peroxide–I /SCN−
I− concentrations can be explained by the reversible formation of the
−
−
2
complex ion I –CN resulting from the reaction between I and ICN
−
[36]. In our LPO experiments, the ICN signal shifted from 51.0 to
53.2 ppm with increasing iodide. Applying commercial ICN, this range
was 50.1 (without iodide addition) to at least 52.9 ppm (1:2 ratio of
ICN to KI, data not shown). The low value for pure ICN of 50.1 ppm is ap-
parently caused by the fact that no iodide was present in this sample. As
all LPO (and MPO) samples contained a certain amount of iodide, it is
not surprising that chemical shifts were not observed lower than
51.0 ppm under these conditions. In ICN, iodine is slightly positively
charged as the electron pair of the iodine-cyanide bond is pulled closer
to the cyanide group [37]. This property favours the formation of the
system is able to form the inter(pseudo)halogen species cyanogen
iodide ICN as revealed by 13C NMR spectroscopy and confirmed by
headspace GC–MS. Thereby, an oxidised iodine species, most likely
−
hypoiodous acid (HOI) or I
2
, reacts with SCN as shown by sequential
−
addition of (pseudo-)halide ions. In contrast, the addition of I to pre-
formed hypothiocyanite did not generate ICN. The formation of ICN
was also observed by replacing LPO by MPO and in the absence of
these haem enzymes. In the latter case, however, a considerably lower
amount of ICN was formed. Thus, while hydrogen peroxide itself is
−
−
also able to oxidise I [32] and, consequently, leads to the formation
above mentioned complex with I .
−
13
of ICN in the presence of SCN , the LPO–H
2
O
2
system (and also the
2 2
In C NMR experiments with the LPO–H O system, the highest
−
− −
MPO–H O
2 2
system) strongly accelerates the I oxidation, resulting in
yield of ICN was observed applying a SCN /I ratio of 1:2. The decrease
−
−
a more efficient ICN formation if SCN is present.
In the chemical literature, the reaction between iodine and SCN has
been known for more than a hundred years [29,30,33], whereby the
in ICN formation at higher I concentrations might have several rea-
−
−
sons. Increasing I concentrations can affect binding of hydrogen perox-
ide to the active site of LPO and can disturb, thus, LPO activity similarly
Fig. 5. 13C NMR overlay spectrum of unlabelled ICN in the presence of KI. The addition of I to a mixture of unlabelled ICN and S13CN− caused a downfield shift of the peak for ICN from
−
1
3
13
−
13
−
5
1.0 ppm to 51.8 ppm. C NMR spectra were taken at 6 °C under the following conditions: 178 mM ICN + 10 mM S CN (light grey spectrum); 165 mM ICN + 9.3 mM S CN + 74 mM
13
−
KI (dark grey spectrum); and 153 mM ICN + 8.6 mM S CN + 138 mM KI (black spectrum). All measurements were done in 100 mM phosphate buffer pH 7.0. Selected spectra of three
independent experiments are shown.

40
D. Schlorke et al. / Journal of Inorganic Biochemistry 154 (2016) 35–41
2 2
O
–SCN–I− system. Mass spectra at 3.55 min retention time in the corresponding chromatograms are given for the
13 13 15
Fig. 6. Headspace GC–MS analysis of the reaction products of the LPO–H
reaction of 4 μM LPO with 10 mM H
and 40 mM I− in the presence of (A) 40 mM SCN , (B) 80 mM SCN−, (C) 80 mM S CN and (D) 80 mM S
three independent experiments are shown.
−
−
−
O
2 2
C N . Selected chromatograms of
as has been shown for SCN− and H
complex chemistry of oxidised iodine species [39]. Increased I concen-
O
[38]. Another aspect concerns the
toxic reactions of ICN are very scarce. ICN induces a rapid and
irreversible swelling of the gram-negative bacterium Pseudomonas
aeruginosa and modifies sulfhydryls in these cells [45,46]. Cultures of
the marine diatom Nitzschia cf pellucida produce cyanogen bromide
and cyanogen iodide in a haloperoxidase-dependent reaction and use
them to control the growth of competing microalgae [47]. Thus,
at present we can only speculate about any microbicidal activity of
ICN in LPO applications. Due to its low boiling point, ICN is volatile [44]
and can exert its activity also in close neighbourhood of its production
place.
Cyanogen chloride (ClCN) and cyanogen bromide (BrCN) are related
inter(pseudo)halogens to ICN. Their formation has been observed after
reaction between the corresponding hypohalous acid and cyanide [31,
48]. Cyanogen chloride is also formed upon exposure of HOCl to cyano-
cobalamin [49]. However, the roles of ClCN and BrCN in haem-
dependent oxidation reactions and in the anti-microbial system have
to be determined as well.
2
2
−
−
−
trations may favour the formation of I
3 2
from I and I and decrease,
thus, the concentration of active oxidised iodine intermediates.
−
2
Although the formation of ICN from I and SCN has been well
known for more than 100 years, its production was never described be-
fore in connection with haem peroxidases. The formation of ICN by LPO
and MPO is not expected under physiologically relevant conditions as
iodide concentration is usually rather low in biological fluids [9,40].
−
Considering millimolar concentrations of SCN in saliva [7–9] or around
−
4
00 μM SCN in upper airway secretions [10], thiocyanate is preferen-
2 2
tially oxidised by the LPO–H O system with hypothiocyanite as the
−
dominating product. Thus, I is not competitive as LPO substrate for
−
SCN under these conditions [41]. Although MPO is normally not re-
leased into secretions, this enzyme can occur in these fluids under
inflammatory conditions [42]. In contrast to LPO, MPO oxidises
also chloride and bromide with significant rate [24]. Considering
−
(
pseudo-)halide concentrations and reaction rates, SCN is the preferred
−
−
target for MPO in saliva, while in plasma both Cl and SCN will be
oxidised to a comparable extent by the enzyme [43].
Acknowledgements
However, in numerous applications of LPO as anti-microbial agent,
conditions exist favouring the formation of ICN. In our experiments,
the most intense formation of ICN was found at a 1:2-ratio between
The authors gratefully acknowledge the financial support provided
by the German Federal Ministry of Education and Research (BMBF
315883). The GC–MS analysis was partially funded by the European
Fund for Regional Structure Development, EFRE (“Europe funds
Saxony”, EFRE 3901-2014).
1
−
−
SCN and I . This agrees well with the second order rate constants of
−
−
LPO Compound I with SCN and I . At 15 °C and pH 7.0, the rate of
−
−
LPO Compound I with SCN is about 1.7 times higher than for I [3]. An-
−
−
other factor is the competition between SCN and I for the binding
near the haem. As iodide oxidation is the prerequisite for ICN formation,
this product will be formed under conditions where the binding of I
References
−
[
1] H. Theorell, Å. Åkeson, A. Kemi, Mineral Geol. 17 (1943) 1–6.
[2] K.D. Kussendrager, A.C.M. van Hooijdonk, Br. J. Nutr. 84 (2000) 19–25.
3] P.G. Furtmüller, W. Jantschko, G. Regelsberger, C. Jakopitsch, J. Arnhold, C. Obinger,
−
dominates over the binding of SCN .
[
Furthermore, at a given ratio between both ions, we found an in-
creased production of ICN by LPO with increasing pH. At pH values
lower than 7.0, the yield of ICN was reduced. We suggest that this
Biochemistry 41 (2002) 11895–11900.
[4] E.L. Thomas, Biochemistry 20 (1981) 3273–3280.
[
[
[
5] T.J. Barrett, C.L. Hawkins, Chem. Res. Toxicol. 25 (2012) 263–273.
6] C.L. Hawkins, Free Radic. Res. 43 (2009) 1147–1158.
7] M.A. Fragoso, V. Fernandez, R. Forteza, S.H. Randell, M. Salathe, G.E. Conner, J. Phys-
iol. 561 (2004) 183–194.
[8] C.P. Schultz, M.K. Ahmed, C. Dawes, H.H. Mantsch, Anal. Biochem. 240 (1996) 7–12.
9] J. Tenovuo, K.K. Makinen, J. Dent. Res. 55 (1976) 661–663.
10] C. Wijkstrom-Frei, S. El-Chemaly, R. Ali-Rachedi, C. Gerson, M.A. Cobas, R. Forteza,
M. Salathe, G.E. Conner, Am. J. Respir. Cell Mol. Biol. 29 (2003) 206–212.
11] E. Seifu, E.M. Buys, E.F. Donkin, Trends Food Sci. Technol. 16 (2005) 137–154.
12] F. Bafort, O. Parisi, J.P. Perraudin, M.H. Jijakli, Enzym. Res. 2014 (2014) 517164.
13] M. Huwiler, H. Kohler, Eur. J. Biochem. 141 (1984) 69–74.
might be mainly caused by the pH dependence of the reaction between
−
HOI or I
2
and SCN [34,35] since the (pseudo-)halogenating activity of
−
2 2
LPO depends on the concentration ratio between SCN and H O [38]
[
and should have an optimum activity around pH 5 under our experi-
mental conditions, where this ratio is b10.
It is generally believed that ICN is a very toxic agent [44]. However, it
remains unknown to what extent ICN will contribute to the anti-
microbial activity of LPO-derived oxidants. Mechanistic data about
[
[
[
[
[14] S.J. Klebanoff, J. Exp. Med. 126 (1967) 1063–1078.

D. Schlorke et al. / Journal of Inorganic Biochemistry 154 (2016) 35–41
41
[
[
[
[
15] E.L. Thomas, T.M. Aune, Antimicrob. Agents Chemother. 13 (1978) 1006–1010.
16] E.L. Thomas, T.M. Aune, Antimicrob. Agents Chemother. 13 (1978) 261–265.
17] E.L. Thomas, T.M. Aune, Infect. Immun. 20 (1978) 456–463.
18] A.J. Fischer, N.J. Lennemann, S. Krishnamurthy, P. Pocza, L. Durairaj, J.L. Launspach,
B.A. Rhein, C. Wohlford-Lenane, D. Lorentzen, B. Banfi, P.B. Mccray, Am. J. Respir.
Cell Mol. Biol. 45 (2011) 874–881.
[33] A. Thiel, Ber. Dtsch. Chem. Ges. 35 (1902) 2766–2768.
[34] R.O. Griffith, A. McKeown, Trans. Faraday Soc. 31 (1935) 868–875.
[35] V.K. Yatsimirsky, V.L. Budarin, V.Y. Tanchuk, Ukr. Khim. Zh. 58 (1992) 958–962.
[36] D.M. Yost, W.E. Stone, J. Am. Chem. Soc. 55 (1933) 1889–1895.
[37] M. Møller, Det Kgl Danske Vid Selskab Math-fys Meddelelser 14 (1936) 1–37.
[38] R. Wever, W.M. Kast, J.H. Kasinoedin, R. Boelens, Biochim. Biophys. Acta 709 (1982)
[
[
[
19] M. Ahariz, P. Courtois, Clin. Cosmet. Investig. Dent. 2 (2010) 69–78.
20] L.W.T. Fweja, M.J. Lewis, A.S. Grandison, J. Dairy Sci. 91 (2008) 2566–2574.
21] R. Ihalin, V. Loimaranta, M. Lenander-Lumikari, J. Tenovuo, J. Periodontal Res. 33
212–219.
[39] W. Gottardi, Arch. Pharm. 332 (1999) 151–157.
[40] J. Holzbecher, D.E. Ryan, Clin. Biochem. 13 (1980) 277–278.
[41] J.D. Chandler, B.J. Day, Free Radic. Res. 49 (2015) 695–710.
[42] R. Ihalin, V. Loimaranta, J. Tenovuo, Arch. Biochem. Biophys. 445 (2006) 261–268.
[43] C.J. van Dalen, M.W. Whitehouse, C.C. Winterbourn, A.J. Kettle, Biochem. J. 327
(1997) 487–492.
[44] B. Bak, A. Hillebert, Org. Synth. 32 (1952) 29–31.
[45] F. Bernheim, Proc. Soc. Exp. Biol. Med. 138 (1971) 444–447.
[46] F. Bernheim, Microbios 4 (1971) 87–92.
[47] B. Vanelslander, C. Paul, J. Grueneberg, E.K. Prince, J. Gillard, K. Sabbe, G. Pohnert, W.
Vyverman, Proc. Natl. Acad. Sci. U. S. A. 109 (2012) 2412–2417.
[48] C.M. Gerritsen, D.W. Margerum, Inorg. Chem. 29 (1990) 2757–2762.
[49] H.M. Abu-Soud, D. Maitra, J. Byun, C.E. Souza, J. Banerjee, G.M. Saed, M.P. Diamond,
P.R. Andreana, S. Pennathur, Free Radic. Biol. Med. 52 (2012) 616–625.
(
1998) 421–427.
22] R. Ihalin, J. Nuutila, V. Loimaranta, M. Lenander, J. Tenovuo, E.M. Lilius, Anaerobe 9
2003) 23–30.
[
(
[
[
[
[
[
23] E.H. Bosch, H. Van Doorne, S. De Vries, J. Appl. Microbiol. 89 (2000) 215–224.
24] P.G. Furtmüller, U. Burner, C. Obinger, Biochemistry 37 (1998) 17923–17930.
25] R.F. Beers, I.W. Sizer, J. Biol. Chem. 195 (1952) 133–140.
26] P. Nagy, S.S. Alguindigue, M.T. Ashby, Biochemistry 45 (2006) 12610–12616.
27] M. Arlandson, T. Decker, V.A. Roongta, L. Bonilla, K.H. Mayo, J.C. MacPherson, S.L.
Hazen, A. Slungaard, J. Biol. Chem. 276 (2001) 215–224.
[
[
[
[
[
28] J.R. Pollock, H.M. Goff, Biochim. Biophys. Acta 1159 (1992) 279–285.
29] C. Meineke, Z. Anorg. Chem. 2 (1892) 157–164.
30] E. Rupp, A. Schied, Ber. Dtsch. Chem. Ges. 35 (1902) 2191–2195.
31] C.M. Gerritsen, M. Gazda, D.W. Margerum, Inorg. Chem. 32 (1993) 5739–5748.
32] G. Schmitz, Phys. Chem. Chem. Phys. 3 (2001) 4741–4746.