Archives of Oral Biology 85 (2018) 7078

Contents lists available at ScienceDirect

Archives of Oral Biology

journal homepage: www.elsevier.com/locate/archoralbio

Reactivation of peroxidase activity in human saliva samples by polyphenols T

Jana Gau, Jrgen Arnhold, Jrg Flemmig
Institute for Medical Physics and Biophysics, Medical Faculty, Leipzig University, Hrtelstrae 16-18, 04107 Leipzig, Germany

A R T I C L E I N F O A B S T R A C T

Keywords: Objectives: The enzyme lactoperoxidase (LPO), which is released into several body uids like saliva, is an es-
Inammation sential part to maintain the oral bacterial homeostasis by catalysing the oxidation of thiocyanate (SCN) to
Saliva hypo-thiocyanite (OSCN). The formation of unreactive redox intermediates (like Compound II) leads to a
Lactoperoxidase decreased pseudo-halogenating enzyme activity, which is associated with a higher risk for oral infections.
Hypothiocyanite
According to former studies with bovine LPO selected avonoids were tested in respect to their potential to
Polyphenols
Salivary peroxidase
reactivate the enzymatic activity in a more physiological, human salivary system.
Design: Saliva samples from healthy donors were collected and characterized by using several gel staining
methods and immunoblotting. Afterwards kinetic measurements were performed by applying the TNB-assay to
evaluate the pseudo-halogenating salivary peroxidase (SAPX) activity. The measurements were performed in the
presence of excess H2O2 to simulate pro-inammatory conditions. Moreover selected avonoids or an ethanolic
extract of Tormentillae rhizoma were applied to test their regenerating eect on the LPO-derived OSCN pro-
duction.
Results: Despite the complex protein composition of the collected saliva samples, an SAPX-derived pseudo-ha-
logenating activity could be identied. The OSCN regenerating eects of the tested polyphenols were com-
pletely comparable to previous in vitro experiments with bovine LPO. Thus, we could show that phenolic sub-
stances are suitable to regenerate the peroxidase activity in human saliva samples after H2O2-induced
inactivation.
Conclusion: The studies provide new insights into the eect of pharmaceutical relevant polyphenols on salivary
peroxidase activity and thus, suggest this enzyme as a new target for the prevention and therapy of oral in-
ammatory diseases.

1. Introduction
Lactoperoxidase (LPO) is a heme containing protein, which belongs
to the immunological relevant chordata peroxidases (Zamocky et al.,
2015). In the oral cavity LPO-analogue salivary peroxidase (SAPX) is
released into the primary saliva secretion via acinus cells of the salivary
gland (Ihalin, Loimaranta, & Tenovuo, 2006). In the presence of thio-
cyanate (SCN) and hydrogen peroxide (H2O2) LPO/SAPX catalyzes
the formation of antimicrobial hypothiocyanite (OSCN) that inhibits
microbial growth by entering bacterial biolms and reacting with free
thiol groups (Day, 2012, 2015; Day, 2012, 2015; Furtmuller et al.,
2002; Hawkins, 2009). SCN, which is mainly obtained by food, enters
the epithelium from the blood stream via sodium-iodide-symporter and
is released by dierent ion channels like CFTR (cystic brosis trans-
membrane conductance regulator) or calcium-activated chloride
channel (Chandler & Day, 2012). H2O2 is formed after spontaneous or
enzymatic (superoxide dismutase) conversion of superoxide anion
(O2%), which is released by membrane-associated enzymes like
NADPH-oxidase (NOX) or dual oxidase (DUOX-2) (Cross & Segal, 2004;
Donko, Peter, Sum, Leto, & Geiszt, 2005; Harper, Xu, McManus,
Heidersbach, & Eiserich, 2006). Furthermore H2O2 could enter the
lumen after release from several pathogens like Lactobacilli or Strepto-
cocci during oral infections or neutrophil granulocytes in case of oral
bleeding or exudation (Chandler & Day, 2015; Fabian, Hermann, Beck,
Fejerdy, & Fabian, 2012; Hoogendoorn & Moorer, 1973).
As illustrated in Fig. 1, the catalytic cycle of LPO/SAPX can be

Abbreviations: ABTS, 2,2, azinobis(3-ethylbenzothiazoline-6-sulfonic acid); ANOVA, analysis of variance; bLPO, bovine lactoperoxidase; CFTR, cystic brosis transmembrane con-
ductance regulator; DMSO, dimethyl sulfoxide; DUOX-2, dual oxidase; DTNB, 5,5-dithiobis-(2-nitrobenzoic acid); ECL, electrochemiluminescence; IgG, immunoglobulin G; LPO, lac-
toperoxidase; MPW, millipore water; NOX, NADPH-oxidase; PAS, periodic acid-Schis reagent; PRP, proline-rich proteins; PVDF, polyvinylidene uoride; SAPX, salivary peroxidase;
TBS, tris-buered saline; TMB, 3,3,5,5-tetramethylbenzidine; TNB, 5-thio-2-nitrobenzoic acid

Corresponding author.
E-mail addresses: jana.gau@medizin.uni-leipzig.de (J. Gau), juergen.arnhold@medizin.uni-leipzig.de (J. Arnhold),
joerg.emmig@medizin.uni-leipzig.de, joerg.emmig@uni-leipzig.de (J. Flemmig).

http://dx.doi.org/10.1016/j.archoralbio.2017.09.037
Received 22 December 2016; Received in revised form 8 September 2017; Accepted 30 September 2017
0003-9969/ 2017 Elsevier Ltd. All rights reserved.
J. Gau et al.
divided into a OSCN-active halogenation and an OSCN-inactive
peroxidase cycle (Furtmuller et al., 2006). Both reactivities require the
primary activation of the ferric enzyme to Compound I by two-elec-
tronic H2O2 reduction (Furtmuller et al., 2006). In the halogenation
cycle this activated enzymatic state two-electronically oxidizes SCN to
OSCN under the regeneration of the native enzyme (Furtmuller et al.,
2002, 2006). Alternatively the native enzyme can be regenerated via
two consecutive one-electron reactions whereby Compound II is formed
as an enzymatic redox intermediate (Furtmuller et al., 2006). In the
absence of the natural substrate SCN the halogenating cycle can be
also disrupted by the spontaneous transition of Compound I to Com-
pound I* that can also lead to the formation of Compound II in the
presence of one-electron donating substances (Furtmuller et al., 2002).
As the resolution of Compound II is the rate-determining step in the
peroxidase cycle this redox intermediate can accumulate in the absence
of suitable substrates, leading to a disturbed formation of OSCN
(Flemmig, Rusch, Czerwinska, Rauwald, & Arnhold, 2014). The pseudo-
halogenating LPO/SAPX activity is known to be an important part of
the oral bacterial homeostasis and thus was shown to be an important
indicator for oral health and hygiene (Koch & Strand, 1979;
Tenovuo & Anttonen,1980; Tenovuo, Mansson-Rahemtulla, Pruitt, &
Arnold, 1981). In fact, an impaired salivary peroxidase activity, as e.g.
observed at oral inammation or xerostomia, is associated with higher
risk of caries, gingivitis or periodontitis (Hugoson, Koch,
Thilander, & Hoogendoorn, 1974; Lagerlof & Oliveby, 1994; van
Steenberghe, Van den Eynde, Jacobs, & Quirynen, 1994).
The accumulation of Compound II is not an irreversible process and
can be overcome in the presence of several naturally occurring sub-
stances. In former studies especially 3,4-dihydroxylated polyphenols
like avonoids were shown to be potent regenerators of the pseudo-
halogenating activity of bovine LPO after H2O2-mediated enzyme in-
activation (Gau, Furtmuller, Obinger, Arnhold, & Flemmig, 2015; Gau
et al., 2016). These aspects indicate SAPX as an interesting new target
for naturally occurring polyphenolic substances as well as for those
plant extracts, which are known for their antibacterial properties and
are already used to treat slight oral infections (ESCOP Scientic
Committee, 2013; Hnsel & Sticher, 1965). Yet the stated previous
studies were only performed in the presence of commercially available
bovine LPO. Therefore in the current study we extended our peroxidase
activity measurements to enriched human saliva although the amino
acid sequence of human SAPX and bovine LPO is quite similar (at least
85 %) (Sharma et al., 2013). After establishing the test system selected
polyphenolic substances were tested in order to get information about
the transferability of the data obtained with bovine LPO to the human
saliva system.
These results show a suitable method to address the pseudo-halo-
genating peroxidase activity in human saliva samples and also provide
options to inuence the salivary OSCN-formation in the presence of
suitable Compound II-resolving aromatic peroxidase substrates. Thus
71
Archives of Oral Biology 85 (2018) 7078
Fig. 1. Catalytic cycle of LPO. In the halogenation cycle (OSCN-active
cycle) the native enzyme is oxidized to Compound I in the presence of
H2O2. The reconstitution of the native enzyme can either occur via the two
electronic oxidation of SCN to OSCN or by two consecutive one elec-
tronic reductions via the formation of Compound II in the so-called per-
oxidase cycle. As the resolution of Compound II is the rate determining
step to reform the native enzyme only in the presence of suitable one
electron donors the OSCN-active cycle can be continued. Furthermore a
spontaneous transition from Compound I to Compound I* is possible,
which can be also transferred to Compound II by one electron-donating
substrates. AH and A% indicate oxidizable substances and the substance
radicals, respectively.
this study may also give new insights into the mode of action of
pharmaceutical relevant substances like natural anti-inammatory oral
care products.
2. Materials and methods
2.1. Materials
The chemical reagents and biochemicals used in this study were
obtained at the highest grade available from following sources:
Coomassie brilliant blue G250; laemmli sample buer; 2-mer-
aptothenanol, tris/glycine (TG buer) or tris/glycine/SDS buer (TGS
buer) and 420% precast polyacrylamide gel were purchased from
Bio-Rad Laboratories GmbH, Munich, Germany. Luteolin and quercetin
were acquired from Biopurify Phytochemicals Ltd., Chengdu, China.
Tormentil (Potentilla erecta) rhizome was bought from Alfred Galke
GmbH, Gittelde, Germany. Ethanol, methanol and Tween 20 were
purchased from Carl Roth GmbH + Co. KG, Karlsruhe, Germany.
Rabbit anti-LPO monoclonal antibody was bought from Abcam,
Cambridge, USA. Gout anti-rabbit secondary antibody (HRP con-
jugated) was obtained from Dianova GmbH, Hamburg, Germany.
Hydrochloric acid and acetic acid were purchased from Merck
Chemicals GmbH, Darmstadt, Germany. Paran pellets were acquired
from Ivoclar Vivadent GmbH, Ellwangen, Germany. A prestained pro-
tein standard (Spectra multicolor broad range protein ladder) was ob-
tained from Thermo Fisher Scientic, Waltham USA. The electro-
chemiluminescence reagent (ECL reagent) was purchased from GE
Healthcare Europe GmbH, Freiburg, Germany. All other chemicals and
enzymes were bought from Sigma Aldrich, Germany.
2.2. Preparation of human saliva
For electrophoretic experiments human saliva samples (10100 ml
per sample) were obtained from three healthy donors (indicated with
AC) 2 h after the last meal. In the same way four healthy donors (in-
dicated with 14) are compared in case of kinetic measurements. For
both experimental settings the salivation was stimulated by chewing
paran gum. Both female and male donors at the age of 2433 were
included, none of them being smokers. After collecting, the samples
were centrifuged at 25,000g for 15 min at 4 C. Afterwards the su-
pernatants were dialyzed against phosphate buer (10 mM, pH 7.4) for
24 h with a molecular weight cut o of 30 kDa. In case of kinetic
measurements the samples were always freshly prepared. We took care
that the sample volumes after collection were similar to be able to
compare the results of the peroxidase activity measurements. For gel
electrophoretic experiments the dialyzed samples were afterwards
lyophilized and stored at 20 C for a maximum of 7 days.
The whole protein content of the saliva samples was determined at
280 nm by using a NanoDrop ND-1000-spectrophotometer, peQlab
J. Gau et al.
Biotechnology GmbH, Erlangen, Germany. It was not possible to de-
termine the concentration of SAPX in the saliva samples at 412 nm
(Furtmuller et al., 2002) UVvis spectroscopically since the whole
protein content was too high, thus overlapping the SAPX-specic Soret
peak.
As the SAPX activity highly depends on the presence of its natural
substrate SCN it was necessary to eliminate this ion from the samples via
dialysis (against 10 mM phosphate buer, 24 h at 4 C). Afterwards 2 mM
SCN was added to the samples, following the conditions used in former
experiments with bovine LPO (bLPO). The eciency of the SCN removal
was checked by thiocyanate quantication with 20 mM iron(III)-chloride
(FeCl3 * 6H2O in 1 M hydrochloric acid) (Thomas, Bates, & Jeerson,
1980). Briey, in the presence of residual thiocyanate a red complex is
formed which can be quantied spectroscopically at 450 nm (determined
extinction coecient: 450 = 1589.33 M1 cm1).
2.3. Gel electrophoresis
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-
PAGE) was performed under non-reducing or reducing conditions (ab-
sence or presence of 2-mercaptoethanol in accordance to the method of
Laemmli (1970). Briey, the lyophilized samples were dissolved in
phosphate buer (10 mM, pH 7.4). 20 l of sample volume (corre-
sponding to 38 g whole protein content determined via NanoDrop at
280 nm), 5 l of bovine LPO (corresponding to 4.2 g protein) and a
multicolor broad range protein ladder were applied on the gel. For
protein detection the gels were stained with coomassie brilliant blue
G250 solution and afterwards de-stained with a mixture of 10% ethanol
and 2% phosphoric acid.
2.4. In gel staining for peroxidase activity
After protein separation the gels were stained for peroxidase activity
with three parts of 6.3 mM 3,3,5,5-tetramethylbenzidine (TMB) in
methanol mixed with seven parts of 0.25 M sodium acetate, pH 5.0, for
1 h. The reaction was started by adding 30 mM H2O2 (Thomas,
Ryan, & Levin, 1976).
2.5. In gel staining for glycoproteins
As saliva is known to contain a high amount of glycoproteins, in gel
staining with periodic acid-Schis reagent (PAS) was performed
(Rahim & Yaacob, 1992). Briey, after protein separation the gels were
stored in trichloroacetic acid for 1 h and, after rinsing with Millipore
water (MPW), pivoted in a 1 % periodic acid solution for 45 min
Afterwards the gels were transferred into PAS solution (containing one
part 1 % periodic acid and ve parts Schis reagent) and stored for
20 min at 4 C. Afterwards excess PAS was removed and the gels were
stored in Schis reagent overnight before de-staining the gels in 10 %
acetic acid solution.
2.6. Immunoblotting
Human salivary proteins and bovine LPO were fractionated by SDS-
PAGE under reducing conditions and afterwards electro-transferred to a
polyvinylidene uoride (PVDF) blotting membrane. The blots were
blocked for 3.5 h at room temperature with 5 % bovine serum albumin
in tris-buered saline (TBS, pH 7.4), containing 0.05 % Tween 20.
Afterwards the blots were incubated over night with anti-LPO antibody
(diluted 1:5000) in TBS containing 0.1% Tween 20. After three washing
steps (in each case for 15 min) the blots were incubated with horse-
radish peroxidase-conjugated gout anti-rabbit IgG (diluted 1:50,000) in
TBS containing 0.1 % Tween 20 for 1 h. Again the blots were washed
three times for 10 min with TBS. Anti-LPO immune reactive bands were
visualized by using the ECL reagent with X-ray lms in accordance to
the manufacturers instructions.
72
Archives of Oral Biology 85 (2018) 7078
2.7. Kinetic measurements and determination of kinetic parameters
All kinetic measurements with human saliva were performed in
phosphate buer (pH 7.4) at 37 C by using a multiplate reader (Tecan
Innite 200 Pro, Mnnerdorf, Switzerland). The nal saliva con-
centration ranged between 025 %. While SCN and Cl were used at
nal concentrations of 50 mM and 2 mM respectively, the H2O2 con-
centration was varied between 20 and 100 M. According to former
studies (Gau et al., 2015) the peroxidase-mediated OSCN-formation
was followed by detecting the formation of Ellmans reagent (5,5-di-
thiobis-(2-nitrobenzoic acid), DTNB) from 5-thio-2-nitrobenzoic acid
(TNB) at 412 nm (412 = 14,100 M1 cm1 (Eyer et al., 2003)). Fur-
ther investigations on LPO activity also used for example 2,2-azinobis
(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) which is not ap-
plicable for kinetic measurements or 2,7-diacetylchlorouorescin
(Hannig et al., 2008; Clausen, Skibsted & Stagsted, 2008). As TNB al-
lows the discrimination between halogenating and peroxidase activity
by a fast reaction in the presence of OSCN (Gau et al., 2016), it was
the rst-choice reagent for this study.
The tested avonoids were dissolved in dimethyl sulfoxide DMSO
and further diluted with phosphate buer to yield nal concentrations
between 0.0120 M. Thereby the DMSO concentration did not exceed
1 % (v/v). The tested plant extract was analyzed and prepared as de-
scribed in previous studies (Gau et al., 2017). The nal extract con-
centrations ranged between 0.1 and 1000 g/ml. Kinetic parameters
(Michaelis-Menten constant (KM), maximal rate (Vmax), turnover
number (kcat), specicity constant (KM/kcat), half maximal eective
concentration (EC50)) were determined for the tested compounds as
described before (Gau et al., 2015). The eect of avonoids or the
mentioned plant extract on the OSCN-formation rate is related to the
percentage of the prepared saliva (the dialyzed solution corresponds to
100%), which was added to the reaction mixture (in total 250 l). It
was not possible to perform kinetic measurements in the whole saliva
samples because of the fast reaction of TNB after adding H2O2, which
cannot be followed with the multiplate reader.
2.8. Statistical analysis
All measurements were performed at least in triplicates. The sta-
tistical analysis of the results was performed by one- and two-way
analysis of variance (ANOVA) followed by a Tukey or Dunnetts multi
comparison test. The data was signicantly drawn from a normally
distribution, which was tested by the Shapiro-Wilk normality test. The
indicated asterisks correspond to the level of signicance: (*)
p < 0.05; (**) p < 0.01; (***) p < 0.001.
2.9. Ethics
As conrmed by the Ethical committee at the Medical Faculty,
Leipzig University (IRB00001750) no ethical approval was necessary
for the study.
3. Results
3.1. Characterization of human saliva samples
The prepared and lyophilized saliva samples were analyzed by dif-
ferent gel staining methods. As we mainly addressed the question
whether the LPO-related salivary peroxidase is traceable, it was not the
scope of this study to qualify the saliva samples. Thus, the obtained
results just give hints for the protein properties and content in the ob-
tained saliva samples. The results for the gel staining with coomassie
brilliant blue, PAS and TMB are shown in Figs. 2 and 3 respectively.
Coomassie brilliant blue interacts with the basic side chains of amino
acids, which is directly connected with the protein amount of the
sample. Accordingly for the reference bLPO a single clear band is visible
J. Gau et al.
in the range between 70 and 80 kDa, which correspond to the described
molecular weight ( 78.5 kDa (Booth, Kimura, Lee, Ikeda-
Saito, & Caughey, 1989)). In contrast for all 3 saliva donors an almost
similar protein pattern occurs whereby the band at about 50 kDa is
most obviously. Additional protein bands are visible in the range of 15
30 kDa, as well as 260 kDa and 70100 kDa. Due to the high protein
content, a corresponding SAPX band was not detectable. As the saliva
samples were dialyzed against a 30 kDa cut-o lter the protein bands
in the low molecular range are somewhat surprising and may be ex-
plained by the formation of protein fragments during the sample pre-
paration of the gel electrophoresis.
The migration of proteins into the gel depends on their level of
glycosylation. Thus, especially glycoproteins are responsible for
smeared and blurred protein bands after coomassie staining (Sarni-
Manchado, Canals-Bosch, Mazerolles, & Cheynier, 2008). In fact the
results of the PAS staining (Fig. 2), that is used to trace glycosylated
proteins, show clear, violet bands that indicate a high amount of gly-
coproteins in the saliva samples. Especially between 50 and 100 kDa
and above 140 kDa the violet dyeing was visible.
For the detection of peroxidase activity the gels were incubated with
TMB and hydrogen peroxide. For every donor green bands are visible in
the range of the reference bLPO, whereby Donor A shows the most
intensive color.
3.2. Western blot analysis
For the direct detection of LPO-related salivary peroxidase im-
munoblot assays were performed. As shown in Fig. 4 for all three do-
nors an immune-reactive band at the same height as the reference LPO
was found, indicating the presence of a related enzyme. Furthermore
for every donor at least 2 additional bands of higher molecular weight
were detectable, which are most visible for Donor B and C. At lower
molecular ( 78,000 kDa) weights no further bands were traceable.
73
Archives of Oral Biology 85 (2018) 7078
Fig. 2. Gel electrophoretic analysis of dierent human saliva samples.
Saliva samples were collected from 3 dierent donors; bLPO was used as a
reference. Protein bands were visualized either with Coomassie brilliant
blue (A) or PAS (B). After staining with coomassie brilliant blue intensive
blue protein bands were detected in the range of 50 kDa. With the PAS
staining procedure clear violet bands were visualized between 70 and
100 kDa and at higher molecular weight (above 140 kDa). (For inter-
pretation of the references to colour in this gure legend, the reader is
referred to the web version of this article.)
3.3. Detection of the hypothiocyanite formation rate in human saliva
samples
Kinetic measurements were performed in the presence of dierent
saliva amounts and H2O2 concentrations. Note that all the described
results are SCN-dependent, which was checked by several control
measurements in the absence of pseudo-halide. The single addition of
Cl or H2O2 did not lead to a TNB oxidation (data not shown). This is a
quite important aspect to exclude other SCN and OSCN-independent
eects of the saliva samples that may induce TNB degradation.
The inuence of dierent saliva amounts on the OSCN-formation
rate (in the presence of 20 M H2O2) is shown in Fig. 5A for 4 donors
(indicated by dierent gray-colored bars) while Fig. 5B indicates the
averaged data. Depending on the saliva concentration for every donor a
signicant increase of the pseudo-halogenating activity was obtained.
Still, upon application of 30% instead of 5% saliva an averaged increase
of the OSCN-formation rate by factor six was observed. In contrast, by
comparing 15% and 30 % saliva only a small averaged increase in the
pseudo-halogenating activity was detectable. By comparing the
medium OSCN-formation rate in the presence of 5 % saliva
(0.044 M s1; 2 mM SCN, 20 M H2O2) with former studies on bLPO
a peroxidase concentration of about 5 nM can be assumed. These
conditions provided the basis for further experiments with dierent
H2O2 concentrations.
The dependence of the relative OSCN-formation rate on the H2O2
concentration is indicated in Fig. 6A for all 4 donors (indicated by
dierent gray-colored bars) and shows a signicant decrease of the
pseudo-halogenating activity at higher H2O2 concentrations. Fig. 6B
indicates the averaged data for the donor-dependent eect shown in
Fig. 6A. According to former experiments with bLPO 20 M H2O2 were
used as a comparative system for the active enzyme (Gau et al.,
2015). The increase of the H2O2 concentration from 20 to 80 M led to
a reduction of the relative OSCN-formation rate by the factor 18. A
concentration of 100 M H2O2 caused (especially for donor 2 and 4) an
J. Gau et al.
Fig. 3. In gel peroxidase staining for 3 dierent human saliva samples. After gel elec-
trophoretic protein separation under non-reducing conditions the gels were incubated
with TMP and H2O2. Peroxidase activity could be visualized by green bands for every
donor in the molecular weight range of bLPO. (For interpretation of the references to
colour in this gure legend, the reader is referred to the web version of this article.)
almost complete inhibition of the determined OSCN-formation. In
former studies with the bLPO system the average relative OSCN-for-
mation rate in the presence of 80 M H2O2 (2.9 104 s1; 2 mM
SCN, 20 M H2O2) was again quite comparable to 5 nM bLPO under
equal conditions (2.3 104 s1) (Gau et al., 2016). The decreased
OSCN-formation at 80 M was afterwards used as a model system for
a reduced enzyme activity, which occurs during inammatory pro-
cesses. Under these conditions selected avonoids were tested to re-
generate the pseudo-halogenating activity. Certain members of this
substance class turned out as excellent polyphenolic substrates for the
re-generation of the pseudo-halogenating bLPO activity after H2O2-
derived enzyme inactivation (Gau et al., 2015, 2016).
3.4. Eect of avonoids on the peroxidase activity
In the current study the eect of the avonoids luteolin, quercetin and
()-epicatechin on the human salivary peroxidase was tested. The results
are shown in Fig. 7 (black rectangles for the salivary system). Thereby the
averaged eect values for 4 dierent saliva donors are shown, which were
calculated by setting the initial activity for every donor to 1 and calculating
the X-fold increase of the OSCN formation rate in the presence of several
avonoid amounts. For comparison also the results obtained with bLPO are
74
Archives of Oral Biology 85 (2018) 7078
Fig. 4. Western blot analysis for 3 dierent human saliva samples. Anti-LPO immune
reactive bands were detectable for every donor, whereby bLPO was used as a reference.
While for bLPO only one band was detectable at least two further bands were traceable
for the human samples.
shown (white rectangles in Fig. 7). In case of the salivary system the pre-
sence of luteolin, ()-epicatechin and quercetin (Fig. 5AC) led to an in-
creased OSCN-formation rate, which is almost comparable to the bLPO
system. Thereby ()-epicatechin and luteolin exhibit almost similar activity
regenerating eects for the human system (14- and 12-fold increase, re-
spectively. For ()-epicatechin these eects were comparable to those of
the avanol on bLPO (15-fold increase) while luteolin exhibited a higher
eect on bLPO (44-fold). In contrast quercetin exhibited the strongest eect
(50-fold increase) on the peroxidase activity of the salivary system while for
bLPO a considerably weaker eect (20-fold increase) was observed. For
every activating compound a saturation eect was detectable after reaching
a maximal value, which is quite similar between the dierent substances: 1
M for luteolin, 2 M for ()-epicatechin and 2 M for quercetin.
The eect of (+)-pinocembrin (Fig. 7D), which was shown to be an
(competitive (Gau et al., 2016)) inhibitor of the enzymatic activity of
bLPO was also transferable to the salivary system. The initial value was
reduced about 80 % for bLPO and 70 % for the saliva samples. Thus, as
already observed for bLPO, the OSCN-forming activity in human
saliva can be eciently regenerated upon application of 34-dihy-
droxylated avonoids while in the absence of the stated B-ring hydro-
xylation inhibitory eects on the pseudo-halogenating activity of saliva
samples were observed.
3.5. Eect of an ethanolic plant extract from tormentil rhizome on the
peroxidase activity
Besides single polyphenolic compounds also a complex plant extract
J. Gau et al. Archives of Oral Biology 85 (2018) 7078

Fig. 5. Dependence of the OSCN-formation rate on the amount of added saliva. In (A) for 4 dierent donors (shown by black, dark gray, gray and white bars) the OSCN-formation rate
clearly increased in dependence on the saliva concentration. Thereby pseudo-halogenating activity varied signicantly between the dierent donors. In (B) the average eect is shown.
Mean and standard deviation of n = 3 independent measurements of 4 dierent saliva samples are given. p < 0.05 (*), p < 0.01 (**), p < 0.001 (***) referred to 0 % saliva.

from tormentil rhizome was tested regarding its ability to regenerate
the pseudo-halogenating activity of the salivary peroxidase system. The
results that are shown in Fig. 8 indicate an equal concentration-de-
pendent eect of the extract towards the human (black rectangles) and
the bovine system (white rectangles). After an increase of the OSCN-
formation rate up to an extract concentration of 100 g/ml (5-fold in-
crease) for the salivary system and 75 g/ml (8-fold increase) for the
bovine enzyme the pseudo-halogenating activity signicantly decreased.
The calculated EC50-value ranged between 5.65 g/ml for the human
samples and 3.15 g/ml for the animal enzyme.
4. Discussion
4.1. Properties of human saliva samples
The obtained saliva samples were characterized by using dierent
gel staining methods like PAS staining to detect glycosylated proteins,
which occur in higher amounts in human saliva (Humphrey &
Williamson, 2001). As it was not the scope of this study to qualify the
saliva samples only presumptions can be made about their protein
content. Prolin-rich proteins (PRP) are the most abundant proteins in
saliva (Carlson, 1993). While the molecular weight of acidic and basic
PRPs ranges between 10 40 kDa high glycosylated PRPs exhibit a
molecular weight about 6090 kDa (Fabian, Beck, Fejerdy, Hermann,
& Fabian, 2015; Sarni-Manchado et al., 2008). Since PAS-positive pro-
tein bands for PRPs are described in the literature for 5797 kDa (Sarni-
Manchado et al., 2008) it can be guessed that the corresponding bands
of our PAS stained gels belongs to these proteins. High molecular
compounds (above 140 kDa), which were stained with coomassie
brilliant blue as well as PAS may represent mucins (like MUC7 and
MUC5B) that are the main components of the mucosa coat
(Fabian et al., 2015; Takehara, Yanagishita, Podyma-Inoue, &
Kawaguchi, 2013). Despite the complexity of the saliva samples the
peroxidase activity was traceable exclusively in the range of the re-
ference bLPO band: H2O2 led to the oxidation of TMB and the formation
of a green-blue two-electronic oxidation product (Josephy, Eling, &
Mason, 1982).
Furthermore the immunoblot analysis could provide an additional
evidence for the occurrence of a SAPX-analogue peroxidase in our
saliva samples. Besides one anti-LPO immune reactive band with a si-
milar molecular weight like the reference bLPO at least two further
immune reactive bands appeared at higher molecular weight. It is
possible that these additional bands correspond to covalent linked en-
zyme aggregates that are described for salivary peroxidase in literature
(Makinen, Tenovuo, & Scheinin, 1976; Ozdemir, Aygul, & Kfrevioglu,
2001). Due to genetic polymorphisms, 3 dierent types of salivary
peroxidase (designated as SAPX 1, 2, 3) are mentioned by Azen (1977).
SAPX 2 as well as SAPX 3 are bigger than SAPX 1 and dissociate in the
smaller, unmodied SAPX 1 form in the presence of 2-mercaptoethanol.
The heterogeneity of the human SAPX molecule and the existence of
high and low molecular forms are also described by Tenovuo et al.
(Tenovuo, 1981). As there are controversial data about the appearance
of these bands under reducing and non-reducing conditions (Mansson-
Rahemtulla, Rahemtulla, Baldone, Pruitt, & Hjerpe, 1988) we cannot
completely exclude antibody cross reactivity due to the high content of
foreign proteins in the prepared saliva samples.

Fig. 6. Dependence of the relative OSCN-formation rate on the H2O2 concentration. In (A) for 4 dierent donors (shown by black, dark gray, gray and white bars) the relative OSCN-
formation rate (related to the H2O2 concentration) clearly decreased in dependence on the amount of H2O2. At 100 M for donor 24 almost no pseudo-halogenating activity was
detectable. Thereby the individual activity varied signicantly within the group of donors. In (B) the average eect is shown. Mean and standard deviation of n = 3 independent
measurements of 4 dierent saliva samples are given. p < 0.05 (*), p < 0.01 (**), p < 0.001 (***) referred to 20 M H2O2.

75
J. Gau et al. Archives of Oral Biology 85 (2018) 7078

Fig. 7. Inuence of several avonoids on the OSCN-formation rate of the human peroxidase system compared to the bovine enzyme. (A)(C) indicate the eect (X-fold increase if the

OSCN-formation rate) of the enzyme regenerating substances luteolin, ()-epicatechin and quercetin on the salivary system (black rectangles) and bLPO (white rectangles).
Concentration-dependent the pseudo-halogenating activity of both enzymatic systems increased signicantly while at higher avonoid concentrations (up to 12 M) saturation is
reached. In the presence of (+)-pinocembrin (D) the OSCN-formation rate for the human and animal enzyme considerable decreased. Mean and standard deviation of n = 3 in-
dependent measurements of 4 dierent saliva samples are given. p < 0.05 (*), p < 0.01 (**), p < 0.001 (***) referred to the tested substances for the salivary system.

4.2. Pseudo-halogenating activity of human saliva samples
In the kinetic measurements a clear dependence of the OSCN-
formation rate of the human saliva system on the saliva amount and
H2O2 concentration was shown. The observed inhibitory eect of excess
H2O2 on the OSCN formation was comparable to the results obtained
with bLPO (Gau et al., 2015) and suggest the accumulation of SCN-
inactive peroxidase redox intermediates like Compound II. Control
measurements with chloride (in the absence of SCN; data not shown)
were performed to exclude the eect of other peroxidases like
myeloperoxidase, which can enter the saliva by secretion from invading
neutrophils at inammatory sites (Hannig et al., 2008; Pruitt, Kamau,
Miller, Mansson-Rahemtulla, & Rahemtulla, 1990).
The high similarity between bLPO and SAPX regarding the com-
position of the active centre and the catalytic triad further support the
assumption of a comparable mechanism of H2O2-derived activity in-
hibition (Sharma et al., 2013). The observed high donor-dependent
eects may be caused by dierent enzyme concentrations and enzyme
activities. The estimated enzyme concentrations (1.515.3 nM) dier
from the literature data ( 20 nM (Thomas, Jeerson, Joyner,

Fig. 8. Inuence of an ethanolic tormentil plant extract on the OSCN-

formation rate of the human peroxidase system compared to the bovine
enzyme. Concentration-dependent the OSCN-formation rate for the
salivary system (black rectangles) as well as for the bovine LPO (white
rectangles) signicantly increased in the presence of the tormentil rhizome
extract. After reaching a maximum the eect strongly decreased to the
initial value in case of both systems. Mean and standard deviation of n = 3
independent measurements of 4 dierent saliva samples are given.
p < 0.05 (*), p < 0.01 (**), p < 0.001 (***) referred to the 0 g/ml
extract for the salivary system.

76
J. Gau et al.
Cook, & King, 1994)) and could be explained by a high donor variability
regarding the peroxidase content in saliva or by enzyme inactivation
during sample collection and processing.
Independent of the donor specic initial enzyme activity, the a-
vonoids ()-epicatechin, luteolin and quercetin, which are potent re-
activators for the enzymatic activity of bLPO (Gau et al., 2016), could
be shown to also eciently re-activate the OSCN-formation rate of the
salivary system after the H2O2-mediated inactivation of the pseudo-
halogenating activity. Former studies already indicated the importance
of a 3,4-dihydroxylated avonoid B-ring for ecient enzyme re-
activation. Especially the 4-hydroxyl group could be shown as the
structural element with the highest oxidation probability and the
strongest interacting capacity with the catalytic triad in the heme cavity
of bLPO (Gau et al., 2017). In contrast to the bLPO system in the current
study on SAPX quercetin appeared as the most potent enzyme re-gen-
erator. Due to the chemical properties (oxo group in position C4; hy-
droxyl group in position C3, C2eC3 double bond) of quercetin, which
lead to a good stabilisation of oxidation products (Galati, Moridani,
Chan, & O'Brien, 2001), it is known to be a potent one-electron donor
for bLPO compound I and II that is because of its high reaction rates
also able to compete with the natural substrate SCN for bLPO
Compound I (Gau et al., 2016). Although the human and the bovine
enzyme are quite comparable to date no data exists about the reaction
rate constants for the catalytic cycle of the human enzyme. Slight dif-
ferences could result in changed reaction rates of the salivary perox-
idase with the phenolic compounds, which lead to a dierent reactivity
prole as compared to the animal enzyme. As already described for
bLPO, the avanone (+)-pinocembrin was shown to inhibit the
OSCN-formation of the salivary peroxidase system. In contrast to the
activating compounds the avonoid B-ring of (+)-pinocembrin is un-
substituted. Docking studies with crystallographic data for bLPO
showed that (+)-pinocembrin shows a completely inverted orientation
near the enzymatic active centre as compared to the B-ring dihy-
droxylated compounds (Gau et al., 2016). Since the results for kinetic
measurements with human saliva are quite comparable it can be as-
sumed that equal binding and reactivity mechanisms are responsible for
the inhibitory eects.
In addition to the testing of single polyphenolic compounds we
extended our investigations on the regenerating eect of a complex
plant extract from tormentil rhizome. The European Scientic Cooperative
on Phytotherapy (ESCOP), which publishes evidence-based monographs
about leading herbal medicinal products, proposed tormentil rhizome
for the treatment of slight oral infections (ESCOP Scientic Committee,
2013). Therefore this plant extract was chosen, providing link between
the pseudo-halogenating SAPX activity in saliva, its reactivation by
single polyphenolic compounds and a suitable pharmaceutical appli-
cation. As it was not the scope of this study to qualify and evaluate
single components of the extract we mainly concentrated on the whole
eect, which is likely based on the interaction between several single
components. A lot of data exists in the literature about the (poly-)
phenolic composition of tormentil rhizome (ESCOP Scientic
Committee, 2013; Herrmann & Enge, 1957). Several monomeric and
polymeric (in case of condensed tannins) avonoids as well as cinnamic
acid- and benzoic acid derivatives are described as leading ingredients
(Ahn, 1974; Geiger & Rimpler, 1990; Geiger, Scholz, & Rimpler, 1994;
Vennat, Pouget, Pourrat, & Pourrat, 1992) that were also already shown
to be potent re-activators of the enzymatic activity of bLPO after H2O2-
mediated inactivation (Gau et al., 2015, 2016). Thus these compounds
are most likely involved in the reactivation of salivary peroxidase.
Depending on the tannin or polyphenol concentration, respectively the
pseudo-halogenating activity signicantly decreases. This could be also
shown by other studies in case of comparatively higher concentrations
of polyphenol (Flemmig et al., 2014; Hannig et al., 2008). Thereby the
one-electron reaction of Compound I to inactive Compound II is pro-
moted, which can be compensated by Compound I regeneration of these
substances (Gau et al., 2016). In case of increased tannin concentrations
77
Archives of Oral Biology 85 (2018) 7078
also precipitating eects has to be taken into account (Adamczyk,
Salminen, Smolander, & Kitunen, 2012).
5. Conclusion
In summary, we identied the bLPO-analogue salivary peroxidase in
partial puried saliva samples with the help of dierent gel staining
methods and western blot analysis. Furthermore a test system was es-
tablished to identify pseudo-halogenating activity and perform kinetic
measurements. The salivary peroxidase plays an important role for the
maintenance of the bacterial homeostasis in the oral cavity
(Koch & Strand, 1979; Tenovuo & Anttonen, 1980; Tenovuo et al.,
1981). Infectious conditions or a decreased salivary ow lead to a re-
duced pseudo-halogenating activity in saliva and support the risk of
inammatory diseases like caries or gingivitis (Hugoson et al., 1974;
Lagerlof & Oliveby, 1994; van Steenberghe et al., 1994). Based on
former studies where natural polyphenolic substances were system-
atized for their regenerating eect of the OSCN-formation of bLPO
after H2O2-induced enzyme inactivation, selected kinetic measurements
were repeated with the human saliva test system. While the in vitro
experiments were performed under very articial conditions the addi-
tional testing of partial puried saliva samples allow further in-
vestigations with a more physiological setting. It was shown that the
results are almost transferable between the isolated bovine LPO and the
human salivary model and that polyphenols (like avonoids) are also
potent regenerators of the salivary pseudo-halogenating activity. This
provides new insights into the known mode of action of anti-microbial-
acting avonoids and denes salivary peroxidase as a new target for
traditionally used phyto-pharmaceutical mouth rinses for the treatment
and prevention of oral infections. Furthermore polyphenols from green
tea (like ()-epigallocatechine gallate, which was also shown as an
OSCN-regenerating compound (Gau et al., 2017)) inhibit the growth
of halitosis-causing bacteria that could be mediated or supported by the
enhancement of the SAPX-related pseudo-halogenating activity of these
substances. Thus, also the application of the SAPX-system against oral
malodour is possible (Zeng, Wu & Pika, 2010).
Conict of interest
The authors have no conict of interest to disclose.
Submission declaration
The data presented here have not been published previously and are
currently not under consideration for publication elsewhere.
Acknowledgment
Jana Gau was supported by a state stipend of the free state of
Saxony (grant LAU-R-N-03-2-0714) provided from the Saxon Ministry
of Science and Fine Arts (SMWK).
References
Adamczyk, B., Salminen, J.-P., Smolander, A., & Kitunen, V. (2012). Precipitation of
proteins by tannins: Eects of concentration: Protein/tannin ratio and pH.
International Journal of Food Science & Technology, 47, 875878.
Ahn, B.-Z. (1974). Catechingerbstoe aus der Tormentilla-Wurzel. Archiv der Pharmazie,
304, 241250.
Azen, E. A. (1977). Salivary peroxidase (SAPX): Genetic modication and relationship to
the proline-rich (Pr) and acidic (Pa) proteins. Biochemical Genetics, 15, 929.
Booth, K. S., Kimura, S., Lee, H. C., Ikeda-Saito, M., & Caughey, W. S. (1989). Bovine
myeloperoxidase and lactoperoxidase each contain a high anity site for calcium.
Biochemical and Biophysical Research Communications, 160, 897902.
Carlson, D. M. (1993). Salivary proline-rich proteins: Biochemistry, molecular biology:
And regulation of expression. Critical Reviews in Oral Biology and Medicine, 4,
495502.
Chandler, J. D., & Day, B. J. (2012). Thiocyanate: A potentially useful therapeutic agent
with host defense and antioxidant properties. Biochemical Pharmacology, 84,
J. Gau et al.
13811387.
Chandler, J. D., & Day, B. J. (2015). Biochemical mechanisms and therapeutic potential of
pseudohalide thiocyanate in human health. Free Radical Research, 49, 695710.
Clausen, M. R., Skibsted, L. H., & Stagsted, J. (2008). Inhibition of lactoperoxidase-cat-
alyzed 2,2'-azino-bis(3-ethylbenzthiazoline-6-sulfonic acid) (ABTS) and tyrosine
oxidation by tyrosine-containing random amino acid copolymers. Journal of
Agricultural and Food Chemistry, 56, 86928698.
Cross, A. R., & Segal, A. W. (2004). The NADPH oxidase of professional phagocyte-
sprototype of the NOX electron transport chain systems. Biochimica et Biophysica
Acta, 1657, 122.
Donko, A., Peter, Z., Sum, A., Leto, T., & Geiszt, M. (2005). Dual oxidases. Philosophical
transactions of the Royal Society of London. Series B, Biological Sciences, 360,
23012308.
ESCOP Scientic Committeee (2013). Tormentillae rhizoma. ESCOP monograph.
Eyer, P., Worek, F., Kiderlen, D., Sinko, G., Stuglin, A., Simeon-Rudolf, V., et al. (2003).
Molar absorption coecients for the reduced Ellman reagent: Reassessment.
Analytical Biochemistry, 312, 224227.
Fabian, T. K., Hermann, P., Beck, A., Fejerdy, P., & Fabian, G. (2012). Salivary defense
proteins: Their network and role in innate and acquired oral immunity. International
Journal of Molecular Sciences, 13, 42954320.
Fabian, T. K., Beck, A., Fejerdy, P., Hermann, P., & Fabian, G. (2015). Molecular me-
chanisms of taste recognition: Considerations about the role of saliva. International
Journal of Molecular Sciences, 16, 59455974.
Flemmig, J., Rusch, D., Czerwinska, M. E., Rauwald, H. W., & Arnhold, J. (2014).
Components of a standardised olive leaf dry extract (Ph:Eur.) promote hypothio-
cyanite production by lactoperoxidase. Archives of Biochemistry and Biophysics, 549,
1725.
Furtmuller, P. G., Jantschko, W., Regelsberger, G., Jakopitsch, C., Arnhold, J., & Obinger,
C. (2002). Reaction of lactoperoxidase compound I with halides and thiocyanate.
Biochemistry, 41, 1189511900.
Furtmuller, P. G., Zederbauer, M., Jantschko, W., Helm, J., Bogner, M., Jakopitsch, C.,
et al. (2006). Active site structure and catalytic mechanisms of human peroxidases.
Archives of Biochemistry and Biophysics, 445, 199213.
Galati, G., Moridani, M. Y., Chan, T. S., & O'Brien, P. J. (2001). Peroxidative metabolism
of apigenin and naringenin versus luteolin and quercetin: Glutathione oxidation and
conjugation. Free Radical Biology & Medicine, 30, 370382.
Gau, J., Furtmuller, P. G., Obinger, C., Arnhold, J., & Flemmig, J. (2015). Enhancing
hypothiocyanite production by lactoperoxidase mechanism and chemical properties
of promotors. Biochemistry and Biophysics Reports, 4, 257267.
Gau, J., Furtmuller, P. G., Obinger, C., Prevost, M., Van, A. P., Arnhold, J., et al. (2016).
Flavonoids as promoters of the (pseudo-)halogenating activity of lactoperoxidase and
myeloperoxidase. Free Radical Biology & Medicine, 97, 307319.
Gau, J., Prevost, M., Van Antwerpen, P., Sarosi, M.-B., Rodewald, S., Arnhold, J., et al.
(2017). Tannins and tannin-related derivatives enhance the (pseudo-)halogenating
activity of lactoperoxidase. Journal of Natural Products, 80, 13281338.
Geiger, C., & Rimpler, H. (1990). Ellagitannins from tormentillae rhizoma and alchem-
illae herba. Planta Medica, 56, 585586.
Geiger, C., Scholz, E., & Rimpler, H. (1994). Ellagitannins from Alchemilla xanthochlora
and Potentilla erecta. Planta Medica, 60, 384385.
Hnsel, R., & Sticher, O. (1965). Pharmakognosie Phytopharmazie (9th ed.). Heidelberg:
Springer.
Hannig, C., Spitzmller, B., Knausenberger, S., Hoth-Hannig, W., Hellwig, E., & Hannig,
M. (2008). Detection and activity of peroxidase in the in situ formed enamel pellicle.
Archives of Oral Biology, 53, 849858.
Harper, R. W., Xu, C., McManus, M., Heidersbach, A., & Eiserich, J. P. (2006). Duox2
exhibits potent heme peroxidase activity in human respiratory tract epithelium. FEBS
Letters, 580, 51505154.
Hawkins, C. L. (2009). The role of hypothiocyanous acid (HOSCN) in biological systems.
Free Radical Research, 43, 11471158.
Herrmann, K., & Enge, W. (1957). berr den Gerbsto der Rhizoma Tormentillae. Archiv
Der Pharmazie Und Berichte Der Deutschen Pharmazeutischen Gesellschaft, 290,
276280.
Hoogendoorn, H., & Moorer, W. R. (1973). Lactoperoxidase in the prevention of plaque
accumulation: Gingivitis and dental caries. I. Eect on oral streptococci and lacto-
bacilli. Odontologisk Revy, 24, 355366.
Hugoson, A., Koch, G., Thilander, H., & Hoogendoorn, H. (1974). Lactoperoxidase in the
prevention of plaque accumulation, gingivitis and dental caries. 3. Eect of
78
Archives of Oral Biology 85 (2018) 7078
mouthrinses with amyloglucosidase and glucoseoxidase in the model system of ex-
perimental gingivitis and caries in man. Odontologisk Revy, 25, 6980.
Humphrey, S. P., & Williamson, R. T. (2001). A review of saliva: Normal composition,
ow and function. Journal of Prosthetic Dentistry, 85, 162169.
Ihalin, R., Loimaranta, V., & Tenovuo, J. (2006). Origin, structure: And biological ac-
tivities of peroxidases in human saliva. Archives of Biochemistry and Biophysics, 445,
261268.
Josephy, P. D., Eling, T., & Mason, R. P. (1982). The horseradish peroxidase-catalyzed
oxidation of 3,5,3',5'-tetramethylbenzidine: Free radical and charge-transfer complex
intermediates. The Journal of Biological Chemistry, 257, 36693675.
Koch, G., & Strand, G. (1979). Eect of an enzyme dentifrice on caries: A two-year clinical
pilot study. Swedish Dental Journal, 3, 913.
Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of
bacteriophage T4. Nature, 227, 680685.
Lagerlof, F., & Oliveby, A. (1994). Caries-protective factors in saliva. Advances in Dental
Research, 8, 229238.
Makinen, K. K., Tenovuo, J., & Scheinin, A. (1976). Turku sugar studies XII. The eect of
the diet on oral peroxidases, redox potential and the concentration of ionized
uorine, iodine and thiocyanate. Acta Odontologica Scandinavica, 34, 353369.
Mansson-Rahemtulla, B., Rahemtulla, F., Baldone, D. C., Pruitt, K. M., & Hjerpe, A.
(1988). Purication and characterization of human salivary peroxidase. Biochemistry,
27, 233239.
Ozdemir, H., Aygul, I., & Kfrevioglu, O. I. (2001). Purication of lactoperoxidase from
bovine milk and investigation of the kinetic properties. Preparative Biochemistry and
Biotechnology, 31, 125134.
Pruitt, K. M., Kamau, D. N., Miller, K., Mansson-Rahemtulla, B., & Rahemtulla, F. (1990).
Quantitative, standardized assays for determining the concentrations of bovine lac-
toperoxidase, human salivary peroxidase, and human myeloperoxidase. Analytical
Biochemistry, 191, 278286.
Rahim, Z. H., & Yaacob, H. B. (1992). Electrophoretic detection of salivary alpha-amylase
activity. The Journal of Nihon University School of Dentistry, 34, 273277.
Sarni-Manchado, P., Canals-Bosch, J. M., Mazerolles, G., & Cheynier, V. (2008). Inuence
of the glycosylation of human salivary proline-rich proteins on their interactions with
condensed tannins. Journal of Agricultural and Food Chemistry, 56, 95639569.
Sharma, S., Singh, A. K., Kaushik, S., Sinha, M., Singh, R. P., Sharma, P., et al. (2013).
Lactoperoxidase: Structural insights into the function: Ligand binding and inhibition.
International Journal of Biochemistry and Molecular Biology, 4, 108128.
Takehara, S., Yanagishita, M., Podyma-Inoue, K. A., & Kawaguchi, Y. (2013). Degradation
of MUC7 and MUC5B in human saliva. PLoS One, 8, e69059.
Tenovuo, J., & Anttonen, T. (1980). Peroxidase-catalyzed hypothiocyanite production in
human salivary sediment in relation to oral health. Caries Research, 14, 269275.
Tenovuo, J., Mansson-Rahemtulla, B., Pruitt, K. M., & Arnold, R. (1981). Inhibition of
dental plaque acid production by the salivary lactoperoxidase antimicrobial system.
Infection and Immunity, 34, 208214.
Tenovuo, J. (1981). Dierent molecular forms of human salivary lactoperoxidase.
Archives of Oral Biology, 26, 10511055.
Thomas, P. E., Ryan, D., & Levin, W. (1976). An improved staining procedure for the
detection of the peroxidase activity of cytochrome P-450 on sodium dodecyl sulfate
polyacrylamide gels. Analytical Biochemistry, 75, 168176.
Thomas, E. L., Bates, K. P., & Jeerson, M. M. (1980). Hypothiocyanite ion: Detection of
the antimicrobial agent in human saliva. Journal of Dental Research, 59, 14661472.
Thomas, E. L., Jeerson, M. M., Joyner, R. E., Cook, G. S., & King, C. C. (1994). Leukocyte
myeloperoxidase and salivary lactoperoxidase: Identication and quantitation in
human mixed saliva. Journal of Dental Research, 73, 544555.
van Steenberghe, D., Van den Eynde, E., Jacobs, R., & Quirynen, M. (1994). Eect of a
lactoperoxidase containing toothpaste in radiation-induced xerostomia. International
Dental Journal, 44, 133138.
Vennat, B., Pouget, M. P., Pourrat, A., & Pourrat, H. (1992). Identication and quanti-
cation of proanthocyanidins in root extract of Potentilla tormentilla (Rosaceae).
Journal de Pharmacie de Belgique, 47, 485493.
Zamocky, M., Hofbauer, S., Schaner, I., Gasselhuber, B., Nicolussi, A., Soudi, M., et al.
(2015). Independent evolution of four heme peroxidase superfamilies. Archives of
Biochemistry and Biophysics, 574, 108119.
Zeng, Q. C., Wu, A. Z., & Pika, J. (2010). The eect of green tea extract on the removal of
sulfur-containing oral malodor volatiles in vitro and its potential application in
chewing gum. Journal Breath Research, 4, 18.