Original Article

Innate Immunity
0(0) 1–10
Periodontal therapy increases neutrophil ! The Author(s) 2019
Article reuse guidelines:
extracellular trap degradation sagepub.com/journals-permissions
DOI: 10.1177/1753425919889392
journals.sagepub.com/home/ini

Carolyn GJ Moonen1 , Kirsten GD Buurma1,

Mouri RJ Faruque1, Maria G Balta1,2, Erol Liefferink1,
Sergio Bizzarro1, Elena A Nicu1,3 and Bruno G Loos1

Abstract
In periodontitis, polymorphonuclear leucocytes (PMNs) are activated. They entrap and eliminate pathogens by releasing
neutrophil extracellular traps (NETs). Abnormal NET degradation is part of a pro-inflammatory status, affecting
co-morbidities such as cardiovascular disease. We aimed to investigate the ex vivo NET degradation capacity of
plasma from periodontitis patients compared to controls (part 1) and to quantify NET degradation before and after
periodontal therapy (part 2). Fresh NETs were obtained by stimulating blood-derived PMNs with phorbol 12-myristate
13-acetate. Plasma samples from untreated periodontitis patients and controls were incubated for 3 h onto freshly
generated NETs (part 1). Similarly, for part 2, NET degradation was studied for 91 patients before and 3, 6 and 12 mo
after non-surgical periodontal therapy with and without adjunctive systemic antibiotics. Finally, NET degradation was
fluorospectrometrically quantified. NET degradation levels did not differ between periodontitis patients and controls,
irrespective of subject-related background characteristics. NET degradation significantly increased from 65.6 ! 1.7%
before periodontal treatment to 75.7 ! 1.2% at 3 mo post periodontal therapy, and this improvement was maintained at
6 and 12 mo, irrespective of systemic usage of antibiotics. Improved NET degradation after periodontitis treatment is
another systemic biomarker reflecting a decreased pro-inflammatory status, which also contributes to an improved
cardiovascular condition.

Keywords
Chronic periodontitis, innate immunology, PMN, polymorphonuclear leucocytes, neutrophils
Date received: 5 September 2019; revised: 20 October 2019; accepted: 29 October 2019

Introduction
in an activated state.15–17 This may be related to the
Periodontitis is a chronic inflammatory disease of the chronic transmigration of oral bacteria from periodon-
tooth-supporting tissues that potentially leads to tooth tal lesions into the blood circulation.1,6 Being the most
loss. Periodontitis has been found to be associated with abundant white blood cells in blood, PMNs have
atherosclerotic cardiovascular disease (ACVD).1–5 Even phagocytic and bacterial killing capacities to neutralise
though the relation between periodontitis and ACVD and remove micro-organisms from the circulation.18
has been broadly reported, the biological mechanisms
and clinical relevance of this interaction are still under
1
investigation.6,7 Nevertheless, previous research has Department of Periodontology, Academic Centre for Dentistry
Amsterdam (ACTA), University of Amsterdam and Vrije Universiteit
shown beneficial effects of non-surgical periodontal ther-
Amsterdam, The Netherlands
apy on several clinical and biochemical parameters of 2
Department of Oral Biology, University of Oslo, Norway
ACVD, including flow-mediated dilatation, intima 3
CMI Dr. Opris M.I., Romania
media thickness, systolic blood pressure, and a decrease
in activated platelets.8–14 Altogether, this may reduce the Corresponding author:
Bruno G Loos, Department of Periodontology, Academic Centre for
risk factors of an ACVD profile of periodontitis patients. Dentistry Amsterdam, Gustav Mahlerlaan 3004, 1081 LA Amsterdam,
In periodontitis, circulatory polymorphonuclear The Netherlands.
leucocytes (PMNs) are present in higher numbers and Email: [email protected]
Creative Commons Non Commercial CC BY-NC: This article is distributed under the terms of the Creative Commons Attribution-
NonCommercial 4.0 License (http://www.creativecommons.org/licenses/by-nc/4.0/) which permits non-commercial use, reproduction and
distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.
sagepub.com/en-us/nam/open-access-at-sage).
2 Innate Immunity 0(0)

Another antimicrobial strategy of PMNs is the release would lead to increased NET degradation levels.
of neutrophil extracellular traps (NETs).19–21 NETs To date, only one study with matched patients and
are web-like structures composed of a core DNA ele- controls (n ¼ 19 pairs) investigated NET degradation
ment, extranuclear histones and de-condensed nuclear in periodontitis.25 The authors reported lower NET
chromatin combined with various antimicrobial com- degradation in periodontitis patients compared to
pounds released from PMN granules that immobilise controls, which was explained by significantly lower
and kill pathogens.22,23 Various stimuli (e.g. phorbol DNaseI plasma concentration levels in the periodonti-
12-myristate 13-acetate (PMA), immune complexes or tis patient group. Additionally, this investigation
bacterial products) induce NET formation through the studied the effect of periodontal treatment with a
activation of the protein kinase C pathway.24 NET 3-mo follow-up on NET degradation, and the NET
release by peripheral blood PMNs in periodontitis degradation was found to be increased after periodon-
patients and healthy controls has been compared, and tal treatment.25 Whether these ‘restored’ NET degrada-
an independent association between elevated NET tion levels remain stable for a longer follow-up period
release and periodontitis was reported.25 was not investigated. The current study was set up in a
It was initially proposed that increased NET pro- larger group of subjects to investigate NET degrada-
duction protects the host from infectious diseases by tion differences further in periodontitis patients and
effective pathogen clearance. However, increased controls (n ¼ 76, part 1) and to investigate NET degra-
NET production or impaired NET degradation can dation at baseline as well as at 3, 6 and 12 mo following
also lead to pathological immune responses.26 non-surgical periodontal treatment (n ¼ 91, part 2).
In healthy individuals, NETs are degraded by multiple
enzymes, in particular plasma DNases, which are
endonucleases secreted by the pancreas. DNaseI
Materials and methods
degrades the phosphodiester linkages of the DNA Study populations
backbone, thereby degrading both single- and double-
stranded DNA.27 In the case of low levels or even in the This study consisted of two parts. In the first part,
absence of DNases, the degradation of NETs is heavily the NET degradation capacity of plasma from peri-
reduced.28 odontitis patients (n ¼ 38) was compared to that of
Accordingly, excessive accumulation of cytotoxic plasma from healthy control subjects (n ¼ 38). In the
NET-associated compounds such as antimicrobial pep- second part, the NET degradation capacity of plasma
tides, auto-immunogenic DNA (citrullinated histones from periodontitis patients (n ¼ 91) before and after
and single-stranded DNA), enzymes (myeloperoxidase non-surgical periodontal treatment was analysed from
and elastase) and entrapped bacteria can amplify samples which were available from the study of
(chronic) immune reactions and potentially triggers Bizzarro et al.38
the presentation of auto-Ags in the host, eventually Patients for the first part of the study were screened
leading to tissue damage.29,30 As such, NETs have for eligibility between June 2014 and December
been suggested as possible players in the development 2016, and patients for the second part of the study
or exacerbation of autoimmune diseases such as were screened in the period 2008–2013. Eligible peri-
rheumatoid arthritis (RA)29 and systemic lupus erythe- odontitis patients for both parts of the study were
matosus (SLE).31–33 Furthermore, NETs can cause asked to participate after their initial visit (intake con-
platelet adhesion, activation and aggregation, and sult) to the Department of Periodontology of the
thereby induce endothelial dysfunction by activation Academic Centre for Dentistry Amsterdam (ACTA),
and damage of endothelial cells.34 These complications Amsterdam, The Netherlands, before the initiation
are risk factors for ACVD events.35,36 Furthermore, of their periodontal treatment. The healthy control
possibly via their negative effects on the endothelial subjects from the first part of the study were dental
lining of coronary arteries, for example, NETs have patients from the educational clinics of ACTA or
been found to play a causative role in the formation laboratory staff. Subjects willing to participate were
of atherosclerotic plaques and venous thrombi.37 Thus, invited for the research. The study protocol for part 1
any condition that can induce increased NETs and/or was reviewed and approved by the Medical Ethical
may result in insufficient degradation of NETs could be Committee of the VU University Medical Centre,
a risk factor for ACVD. Amsterdam, The Netherlands (approval number:
Another mechanism to explain the link between 2014.246, revised on 12 March 2015), and part 2 was
periodontitis and ACVD could be elevated NET approved by the Medical Ethical Committee of the
levels. Since periodontal treatment has shown benefi- Academic Medical Centre of the University of
cial effects on clinical and biochemical parameters of Amsterdam, Amsterdam, The Netherlands (approval
ACVD,12 we hypothesised that periodontal treatment number: MEC 07/264).38 All participants were
Moonen et al. 3

informed about the purpose of the study and signed an Sample collection
informed consent form. Researchers handling the
Plasma samples from patients and controls for both
plasma samples (C.G.J.M. and M.R.J.F.) could not
part 1 and part 2 of this study were available.
retrieve the identity of the donors.
Samples for the first part of the study were retrieved
from previous yet unpublished studies (E.A.N., M.G.
General characteristics B. and E.L.). For part 2 of this study, 99 patients com-
pleted the study, including all follow-up appointments
At the start of the study, demographic characteristics
up to 12 mo.38 However, for part 2 of this study,
such as age, sex, ethnicity and education were recorded
plasma samples of 91 patients were available and
by means of a questionnaire. A participant was consid-
used. In general, study participants were scheduled
ered a smoker if he or she was currently smoking or
for blood collection between 8:00 am and 12:00 noon.
had quit #6 mo before baseline examination. The body
Non-fasting blood was collected by venipuncture from
mass index (BMI) was calculated from height and
the antecubital fossa into EDTA tubes (VacuetteVR ;
weight.
Greiner Bio-One, Alphen a/d Rijn, the Netherlands)
which were put on ice until further handling. The
Periodontitis patient and control subject collected whole blood was processed (whole blood cen-
characteristics trifugation, 2000 g, 10 min, at 4% C) within 2 h, and
aliquots of all plasma samples were stored at & 80% C
Periodontitis cases were defined as consented in a
until further analysis.
European Workshop,39 specifically the presence
of proximal attachment loss of $ 5 mm in $ 30% of
teeth present. Alveolar bone loss was confirmed on
periapical radiographs. The healthy control subjects
did not have more than one pocket of 4–5 mm, in the
Circulatory PMN isolation for NET formation
absence of proximal bone loss (third molars excluded). To generate NETs for degradation experiments, PMNs
This was confirmed on bitewing radiographs not older were isolated from three healthy male donors (age
than 1 yr, displaying a distance between the alveolar 30–50 yr) which were not related to the aforementioned
bone and the cemento-enamel junction of # 3 mm. study population of either part 1 or part 2 of this
Periodontitis cases were excluded if they had received study. From the three donors, each on a separate d,
any form of periodontal therapy within the last 2 yr. venous blood was obtained in lithium heparin tubes
Further exclusion criteria for both periodontitis and (VacuetteVR ; Greiner Bio-One, VWR, Amsterdam, The
controls were pregnancy/lactation, the use of antimi- Netherlands). All three volunteers were informed about
crobials in the previous 6 mo, the presence of < 20 the purpose of the study and gave informed consent.
natural teeth, the use of omega-3/omega-6 fatty acids In total, 90 ml blood was taken per donor to obtain a
supplementation, systemic diseases (such as diabetes or sufficient number of PMNs to study NET degradation
RA) or the (regular) use of any medicine that could of all plasma samples of part 1 and part 2 on 1 d.
modulate the inflammatory response (such as statins Blood was diluted 1:1 in 1% PBS citrate (1.55 M
or non-steroidal anti-inflammatory drugs) in the previ- sodium citrate, 0.10 M citric acid in sterile water,
ous 2 wk. pH 7.4; both Merck Millipore, Darmstadt, Germany,
pH 7.4), carefully layered on Lymphoprep (Axisshield
Po CAS, Oslo, Norway), and centrifuged for 30 min at
Periodontal therapy 800 g without brake at room temperature (RT; 21% C).
All periodontitis patients in part 2 of the study received The supernatant was removed. Remaining erythrocytes
non-surgical periodontal treatment in three appoint- and PMNs were suspended in cold lysis buffer (1.5 M
ments within 1 wk. Of all patients participating in NH4Cl, 100 mM NaHCO3, 1 mM disodium EDTA; all
part 2 of the study, 42 (randomly determined) received Sigma–Aldrich, St Louis, MO, diluted 10' in sterile
systemic antibiotics treatment in conjunction with peri- Milli-Q water). After erythrocytes were lysed
odontal therapy.38 Systemic antibiotics consisted of (10 min), the samples were centrifuged (500 g, 10 min,
amoxicillin 375 mg and metronidazole 250 mg, both 4% C), and the supernatant was discarded. Finally,
three times daily for 7 d. After completion of the PMNs were washed in cold PBS (4% C) and recovered
active therapy with or without antibiotics, patients in phenol-red free culture medium (Roswell Park
were seen for maintenance treatment every 3 mo until Memorial Institute (RPMI) 1640; Gibco BRL,
the end of the follow-up (1 yr).38 Paisley, UK).
4 Innate Immunity 0(0)

NET formation The control condition contained unstimulated PMNs

5 where no NETs were observed after 3 h (Figure 1a).
For NET formation, PMNs (5 ' 10 in phenol red-free
Typical web-like structures of NETs were observed after
culture medium) were seeded onto a flat-bottom non-
3 h of stimulation of PMNs with PMA (Figure 1b). For
treated 96-well plate (Greiner Bio-One, Monroe, NC)
the visualisation of NET degradation, plasma samples
pre-blocked overnight with 1% BSA in PBS. After 30
from both a periodontitis patient and a healthy control
min baseline incubation at 37% C, selected wells were
were 10' diluted in PBS and added to NETs and incu-
stimulated with 75 nM PMA (Sigma–Aldrich) for 3 h
bated at 37% C for another 3 h. Degradation of NETs (pro-
at 37% C to induce NET formation. Culture medium
cedures described below) by plasma from a healthy donor
served as a negative control where no NETs formed.
and periodontitis patient is shown in Figure 1c and d.
In addition to quantification, NET formation
and degradation were microscopically validated with
visualisation assays which were performed in the same NET degradation quantification
manner as quantification assays. Briefly, PMNs First, aliquots of plasma samples of part 1 and part 2
(1 ' 106) suspended in phenol red-free culture medium were simultaneously defrosted on ice. Thereafter, 20 ml
were seeded onto a transparent 48-well plate (Greiner of the plasma samples was individually transferred in
Bio-One) pre-blocked with 1% BSA-PBS. After 30 min duplicate to 96-well plates, and 180 ml cold PBS was
baseline incubation, selected wells were stimulated with added. This procedure was performed in triplicate to
75 nM PMA. After 3 h of incubation at 37% C, 50 nM obtain a sufficient number of plates for each of the
SYTOXTM green was added to each well and visualised three experimental days. Thus, NET degradation by
with fluorescence microscopy (magnification 20' , Leica plasma samples of part 1 and part 2 was tested in dupli-
DFC320; Leica Microsystems, Wetzlar, Germany). cate on NETs which were generated from PMNs from

Figure 1. NET. (a) PMNs without any NETs visible. (b) PMA-induced NET formation. (c) NET degradation post incubation with 10%
plasma of a healthy subject (d) and a periodontitis patient. NETs were visualized using the extracellular DNA binding dye SYTOXTM
Green. Magnification 20' , scale bars represent 100 mm.
Moonen et al. 5

for Windows v25 (IBM Corp., Armonk, NY).

125 NS Differences between patients and healthy controls
(part 1) were tested with an unpaired parametric
100 t-test for age and BMI, and chi-square tests (Fisher’s
NET degradation (%)

exact tests where appropriate) for sex, ethnicity, educa-

75 tion and smoking. NET degradation results were
analysed using GraphPad Prism v6.07 (GraphPad
50 LLC, La Jolla, CA). The distribution of NET degra-
dation percentages for part 1 and part 2 was assessed
25 with D’Agostino–Pearson tests for normal distribution
and found to be not normally distributed. Therefore,
0 before proceeding with parametric statistics, log-
periodontitis Healthy controls transformed values were calculated. NET degradation
patients
results were presented in scatter plots with means
! SEM. For part 1, NET degradation differences
Figure 2. NET degradation by plasma from periodontitis between periodontitis patients and healthy subjects
patients and healthy controls. NET degradation levels of peri- were first tested with an unpaired t-test. Thereafter,
odontitis patients (n¼38) and healthy controls (n¼38). Each an analysis of covariance (ANCOVA), taking into
individual data point represents the mean of three independent
account group variabilities such as smoking, ethnicity,
experiments from one plasma sample. Horizontal lines represent
the overall mean percentages ! SEM. NS: not significant. age, BMI and education, was also performed using
SPSS. These statistical tests yielded adjusted P Values
three different blood donors. All plates containing all (Padj). For part 2 of this study, the effect of non-
the plasma samples from both part 1 and part 2 of the surgical periodontal treatment over time on NET
study were stored at –80% C until the analysis. degradation was tested with ANOVA followed by
To study in vitro NET degradation, the diluted Tukey multiple comparison tests for more than three
plasma samples on 96-well plates were defrosted on comparisons. Differences with P < 0.05 were consid-
ice and were carefully transferred onto the freshly gen- ered significant. Additionally, to explore the possible
erated NETs and incubated for 3 h at 37% C, as previ- adjunctive effect of systemic antibiotics and other
ously described.25 Next, 15 ml 14.3 IU/ml micrococcal potential confounding factors for changes in NET deg-
nuclease (MNase; Invitrogen by Thermo Fisher radation after non-surgical periodontal therapy,
Scientific, Eugene, OR) was added and incubated for we first calculated the individual differences of NET
15 min at 37% C to digest any PMN bound NET degradation between baseline and 3 mo post therapy,
DNA.25 Then, cells and debris were pelleted (10 min, and similarly for 6 and 12 mo. The obtained values
1800 g, RT; 21% C), and the supernatant was transferred were entered as a dependent variable in a multiple
linear regression analysis. First, antibiotic usage
to a black U-bottom 96-well plate (Greiner Bio-One)
during therapy for about half of the patients was
containing 50 nM SYTOXTM green (Invitrogen by
explored as a possible predictor. Similarly, baseline
Thermo Fisher Scientific). A fluorospectrophotometer
NET degradation levels were explored as possible pre-
(FLUOstar Galaxy, BMG; MTX Lab Systems,
dictors. Potential patient characteristics were also
Bradenton, FL) was used to quantify NETs between
tested as predictors. First, periodontitis severity (mean
485 and 525 nm at 37% C and analysed with Fluostar
probing pocket depth, PPD) was explored and thereafter
Galaxy software v4.2. On the same plate, background
age, sex, ethnicity, education, smoking and BMI were
absorption of plasma was quantified in a PMN-free
entered as predictors in a fully adjusted model. The
condition, and these fluorescence values were sub-
explorative modelling yielded adjusted P Values, and
tracted from the final NET quantification readings.
Padj <0.05 was considered significant.
The percentage of NET degradation for each plasma
sample was normalised to 100% and 0% degradation
values. The 100% degradation was based on 3 h of Results
NET digestion (instead of plasma) with 15 ml of
1 IU/ml MNase, and a similar condition without Characteristics of periodontitis patients and healthy
MNase represented the 0% degradation standard.25 subjects (part 1)
Baseline characteristics of the study population of part
Statistical analysis 1 are presented in Table 1. This study population con-
Background characteristics of participants for part 1 sisted of 38 periodontitis patients and 38 controls with
and part 2 were analysed with IBM SPSS Statistics mean ages of 49 and 43 yr, respectively. There were
6 Innate Immunity 0(0)

Table 1. Baseline characteristics of periodontitis patients and Table 2. Baseline characteristics of periodontitis patients par-
healthy controls participating in part 1 of this study. ticipating in part 2 of this study.

Periodontitis Controls Periodontitis

patients (n¼38) (n¼38) P Value patients (n ¼ 91)

Age (yr) 49.1 ! 11.6 43.4 ! 12.2 0.043 Age (yr) 48.5 ! 9.0
Sex Sex
Female 19 (50%) 16 (42.1%) 0.490 Female 37 (40.7%)
Ethnicity Ethnicity
Caucasian 26 (68.4%) 34 (89.5%) 0.047 Caucasian 80 (87.9%)
Education Education
$ High school 35 (92.1%) 38 (100%) 0.240 $ High school 74 (81.3%)
Smoking Smoking
Smokers 20 (52.6%) 7 (18.4%) 0.002 Smokers 62 (68.1%)
BMI (kg/m2) 27.0 ! 4.5 24.6 ! 2.9 0.006 BMI (kg/m2) 25.3 ! 3.7
Systemic antibiotics 42 (46.1%)
Values (age and BMI) represent means ! SD. Categorical data (sex, eth-
Number of teeth present 27.0 ! 2.8
nicity, education and smoking) are presented as absolute numbers (per-
centages) of subjects. Statistical differences (P Values) between patients
Probing pocket depth (mm) 3.9 ! 0.6
and controls are presented. P Values were calculated with unpaired Clinical attachment loss (mm) 4.2 ! 1.0
parametric t-tests, and categorical data were compared with chi-square Bleeding on probing (%) 66.3 ! 14.8
tests or Fisher’s exact tests where appropriate. Plaque (%) 62.8 ! 24.2
Sites with probing 34.0 ! 17.7
pocket depth $ 6 mm
more Caucasians among controls (89.5%) compared to Values represent means ! SD or absolute numbers (percentages).
patients (68.4%). The education level of high school or
above in the control group was 100% and in the patient at baseline (untreated) and at 3, 6 and 12 mo after
group it was 92%, but the difference was not statistically non-surgical periodontal therapy.38 Patient character-
significant. More than half of the patient population istics were determined at baseline and are presented in
smoked, while 81.6% did not smoke in the control Table 2. The study population comprised mainly
group (P ¼ 0.002). The mean BMI in the patient group Caucasians (87.9%) with a mean age of 48.5 ! 9.0 yr
was significantly higher than the control group (P ¼ 0.009). and a mean BMI of 25.3 ! 3.7 kg/m2. Of the 91
patients, 62 (68.1%) were smokers, and 42 (randomly
Similar NET degradation in healthy subjects and assigned) patients received systemic antibiotics in con-
periodontitis patients (part 1) junction with non-surgical periodontal therapy. The
NET degradation capacity of plasma from 38 peri- clinical results of non-surgical periodontal therapy
odontitis patients and 38 healthy controls was investi- have been recalculated for the current patient popula-
gated. NET degradation did not differ significantly tion and are presented in Supplementary Material
between periodontitis patients and healthy subjects Table S1.
(Figure 2; P ¼ 0.1316); 62.8 ! 4.5% and 68.7 ! 4.3%
NET degradation was shown in patients and controls NET degradation is significantly increased after
(mean percentages ! SEM), respectively. In an periodontal treatment
ANCOVA model, we considered differences in subject
In part 2 of this study, the plasma samples from the 91
characteristics between periodontitis patients and
periodontitis patients were individually added to fresh
healthy controls such as age, sex, ethnicity, education,
PMA-induced NETs to investigate NET degradation
smoking and BMI. However, in this model, subject-
at baseline, and at 3, 6 and 12 mo following non-
related background characteristics did not influence
surgical periodontal treatment (Figure 3). Altogether,
the NET degradation differences between periodontitis
patients and controls (Padj ¼ 0.307). Accordingly, none plasma-derived NET degradation was increased by
of the potential confounders appeared a significant approximately 10% after non-surgical periodontal
covariate. therapy (overall P<0.0001). Accordingly, NET degra-
dation was 65.6 ! 1.7% before periodontal treatment
(baseline) and amounted to 75.7 ! 1.2%, 75.3 ! 1.2%
Characteristics of periodontitis patients receiving
and 74.9 ! 1.3% at 3, 6 and 12 mo post treatment,
non-surgical periodontal therapy (part 2) respectively (means ! SEM). The NET degradation
The study population of part 2 of this investigation for the three time points post treatment was significant-
consisted of 91 periodontitis patients who were seen ly higher compared to baseline (P<0.0001 Tukey post
Moonen et al. 7

potential confounders were significant, meaning that

150 **** the NET degradation levels were maintained solely by
**** periodontal therapy and subsequent maintenance care.
125 ****
NET degradation (%)

100
Discussion
75
The first objective of this study was to compare the
50 NET degradation by plasma from periodontitis
patients and healthy controls. Since we hypothesised
25 that periodontal therapy might benefit NET degrada-
0
tion, the second objective was to quantify NET degra-
Baseline 3 months 6 months 12 months dation by plasma from periodontitis patients at
baseline and at 3, 6 and 12 mo after periodontal treat-
Figure 3. NET degradation by plasma from periodontitis
ment. Altogether, our results showed that plasma-
patients pre and post non-surgical periodontal treatment. NETs induced NET degradation did not differ significantly
were incubated with 10% plasma from periodontitis patients between periodontitis patients and healthy controls.
(n¼91) obtained before and 3, 6 and 12 mo after non-surgical Importantly, we found that non-surgical periodontal
periodontal therapy. NET degradation pretreatment (baseline) treatment increased the NET clearance capacity by
and 3, 6 and 12 mo after periodontal treatment. Statistical sig- 10% after 3 mo, being stably improved at 6 and 12 mo.
nificance was calculated using an ANOVA with repeated meas-
Although the authors of a previous investigation of
ures. Each individual data point represents the mean of three
independent experiments for one plasma sample and accordingly NET production by circulatory PMNs in periodontitis
representing one individual patient. Horizontal lines represent found no difference between periodontitis patients and
the overall mean percentages ! SEM. ****P<0.0001. controls,25 normal NET production may be accompa-
nied by impaired NET degradation. If so, this may
hoc test; Figure 3). To explore whether the use of potentially lead to NET accumulation, with negative
systemic antibiotics affected the positive effect of peri- consequences for tissue homeostasis in periodontitis
odontal treatment on NET degradation, we performed and its systemic comorbidities.36,40 Therefore, in this
a multiple linear regression analysis with antibiotic study, we only investigated NET degradation.
usage as a predictor. From this analysis, no effect White et al. compared NET degradation in peri-
of systemic antibiotics was found for changes in NET odontitis patients and matched controls, and showed
degradation from baseline to 3, 6 and 12 mo (P val- that NET degradation and DNaseI levels were lower in
ues ¼ 0.868, 0.526 and 0.907, respectively). periodontitis patients than in healthy controls.25 In the
In contrast, the baseline NET degradation level of current unmatched study population about twice the
size, we also found a slightly lower mean NET degra-
patients was a strong predictor for its increase after
dation in periodontitis patients compared to controls,
therapy (Padj<0.0001 for 3, 6 and 12 mo after treat-
but that did not differ significantly between patients
ment). With bivariate correlation analyses, we found
and controls, even when we adjusted for variability in
high negative correlations (q ¼ –0.699, –0.705 and
patient background characteristics. Periodontitis
–0.729 for 3, 6 and 12 mo, respectively), meaning that
patients represent a heterogeneous population, and in
lower NET degradation levels at baseline predicted
most cases, the progression of periodontitis is not a
higher increases of NET degradation at 3, 6 and linear process, but rather one consisting of periods of
12 mo post treatment. Furthermore, we investigated exacerbation and remission.41,42 Publications investi-
whether the severity of periodontitis at baseline gating plasma-induced NET degradation (in periodon-
would predict the improved NET degradation levels. titis and SLE) included age- ( ! 5 yr) and sex-matched
As such, the mean PPD at baseline was tested as a controls.25,32 Unlike in our study, lifestyle factors of the
predictor and found to be not significant (Padj ¼ subjects were matched in the aforementioned studies.
0.792). To investigate further whether other back- However, in our study, testing for differences in NET
ground characteristics of the patients were possibly degradation between periodontitis patients and con-
associated with the improvement of NET degradation trols with age- and sex-matched pairs resulted in non-
after therapy, we extended the model by also entering significant differences.
age, sex, ethnicity, education, smoking and BMI as In our study, plasma samples were obtained
predictors. From this fully adjusted model, it became from periodontitis patients and healthy controls.
apparent that a higher BMI at baseline was a signifi- Unlike White et al., we included smokers in our study
cant predictor in the improvement of NET degradation group; more than half of our patient population
at 3 mo (Padj ¼ 0.026). At 6 and 12 mo, none of the (part 1: 52.6%; part 2: 68.1%) consisted of smokers.
8 Innate Immunity 0(0)

While the detrimental effects of smoking on the onset degradation after treatment, while high NET degrada-
and progression of periodontitis and the association tion levels at baseline were correlated with limited
with a wide range of systemic diseases including increases after therapy.
ACVD is well known,43,44 there is limited knowledge Systemic diseases evidently contribute to the onset
on the effects of smoking on NET formation and deg- and/or progression of periodontitis. Vice versa, peri-
radation. However, a recent study showed the negative odontitis has been suggested to influence diabetes and
effects of smoking on PMN chemotaxis, reactive ACVD negatively. While the biological mechanisms
oxygen species formation and NET formation.45 of these associations are still under investigation, a
Thus, smoking may also negatively influence NET for- strong relationship between periodontitis, diabetes
mation and the associated degradation profiles of these and ACVD has been reported.6,46–49 The ACVD and
subjects. The higher numbers of smokers in our peri- metabolic syndrome–related parameters of our study
odontitis patient group may therefore have contributed group were previously investigated, and we reported
to the absence of a difference in NET degradation. that periodontal treatment improved periodontal
In general, differences in sociodemographic character- conditions and cardiovascular parameters such as sys-
istics, smoking habits and BMI may have influenced tolic blood pressure and triglycerides, and the number
the comparison of NET degradation in the first part of patients with metabolic syndrome also reduced.12
of our study and may have been why no significant Interestingly, a related and significant area of NET
difference was found between two groups. Despite research is the relationship between NETs and
this, ANCOVAs resulted in non-significant differences, ACVD. Moreover, NETs play an important protective
suggesting that this was not a result of a multifactorial role by preventing bacterial dissemination into the
cause–effect relationship. vasculature and possibly lymphatics. On the other
The objective of part 2 of this study was to investi- hand, the accumulation of NETs present at the injured
gate whether a substantial reduction of the inflamma- endothelium could activate venous endothelium where
tory state of periodontal tissues after non-surgical PMNs and platelets are then recruited.34,50,51 This acti-
periodontal treatment would benefit NET degradation. vated endothelium induces more NET formation,
We found that NET degradation increased significantly which creates a vicious cycle, resulting in increased
after non-surgical periodontal treatment. These results vessel-wall damage. Eventually, accumulation of acti-
are in line with previous observations.25 However, our vated platelets possibly leads to the formation of
population was almost five times larger than that in the microthrombi and potential obstruction of blood ves-
previous study, and we investigated two extra time sels. Increased NET production has been reported in
points (6 and 12 mo after treatment). Although part ACVD patients37 but not in periodontitis patients.25
of our study population (n ¼ 42) received antibiotic However, periodontal treatment has been proven to
treatment in addition to non-surgical periodontal treat- reduce endothelial dysfunction in patients with severe
ment, this did not influence NET degradation levels. periodontitis.8–11 Therefore, we investigated whether
Furthermore, severity of periodontitis per se (mean periodontal treatment would benefit NET degradation
PPD) was not a predictor for the increased NET deg- and thus could be a possible reason for improved
radation levels in our study. Thus, non-surgical peri- ACVD profiles. Using the same study population as
odontal treatment is beneficial not only for the clinical Bizzarro et al., our finding that NET degradation was
periodontal parameters38 of the periodontitis patient, significantly increased 3 mo after periodontal treatment
but also for NET degradation levels, regardless of their and remained at these levels after 6 and 12 mo is in line
PPD at baseline levels. Interestingly, patients with a with previously reported ACVD clinical parameters.12
higher BMI at baseline showed a greater improvement This suggests that periodontal treatment synergistically
in NET degradation at 3 mo. This can possibly be improves ACVD profiles, clinical periodontal parame-
explained by the fact that periodontal therapy also ters and NET degradation capacities.
had a positive effect on the symptoms of metabolic
syndrome.12 However, the confounding effect of BMI
on NET degradation was absent at 6 and 12 mo,
Conclusion
meaning that the NET degradation levels were solely We found that plasma-induced NET degradation did
maintained by periodontal therapy. From multiple not quantitatively differ between periodontitis patients
regression and bivariate correlation analyses, it and healthy controls. We demonstrated that non-
became apparent that NET degradation levels at base- surgical periodontal treatment increased NET degrada-
line were significant predictors for the improvement of tion capacity by 10% after 3 mo and that this remained
NET degradation at 3, 6 and 12 mo post periodontal stably elevated at 6 and 12 mo. Our findings are in line
treatment. In other words, lower degradation levels with previously reported improved ACVD clinical
at baseline predict a larger increase in NET parameters such as systolic blood pressure and
Moonen et al. 9

triglycerides.12,13 This suggests that non-surgical peri- 6. Schenkein HA and Loos BG. Inflammatory mechanisms
odontal treatment synergistically affects ACVD linking periodontal diseases to cardiovascular diseases.
profiles, periodontal inflammatory parameters and J Periodontol 2013; 84: S51–69.
NET degradation. Unraveling NET formation and 7. Aarabi G, Zeller T, Seedorf H, et al. Genetic susceptibil-
ity contributing to periodontal and cardiovascular
degradation mechanisms in periodontitis patients may
disease. J Dent Res 2017; 96: 610–617.
improve the understanding of the pathophysiology of
8. Teeuw WJ, Slot DE, Susanto H, et al. Treatment of
periodontitis and the aetiopathogenetic links between periodontitis improves the atherosclerotic profile: a sys-
periodontitis and its systemic co-morbidities such as tematic review and meta-analysis. J Clin Periodontol
ACVD, while also enabling the development of new 2014; 41: 70–79.
therapeutic approaches. 9. Orlandi M, Suvan J, Petrie A, et al. Association between
periodontal disease and its treatment, flow-mediated
Acknowledgements dilatation and carotid intima-media thickness: a
The author(s) are grateful to Dr J Hirschfeld (University of systematic review and meta-analysis. Atherosclerosis
Birmingham, UK) for her knowledge and advice on labora- 2014; 236: 39–46.
tory procedures. 10. Tonetti MS, D’Aiuto F, Nibali L, et al. Treatment of
periodontitis and endothelial function. N Engl J Med
Declaration of conflicting interests 2007; 356: 911–920.
11. Seinost G, Wimmer G, Skerget M, et al. Periodontal treat-
The author(s) declared no potential conflicts of interest with ment improves endothelial dysfunction in patients with
respect to the research, authorship and/or publication of this severe periodontitis. Am Heart J 2005; 149: 1050–1054.
article. 12. Bizzarro S, van der Velden U, Teeuw WJ, et al. Effect of
periodontal therapy with systemic antimicrobials on
Funding parameters of metabolic syndrome: a randomized clinical
trial. J Clin Periodontol 2017; 44: 833–841.
The authors received no financial support for the research,
13. Arvanitidis E, Bizzarro S, Alvarez Rodriguez E, et al.
authorship and/or publication of this article. Reduced platelet hyper-reactivity and platelet–leukocyte
aggregation after periodontal therapy. Thromb J 2017;
ORCID iD 15: 1–10.
Carolyn GJ Moonen https://orcid.org/0000-0003-1601- 14. Vidal F, Cordovil I, Figueredo CMS, et al. Non-surgical
4768 periodontal treatment reduces cardiovascular risk in
refractory hypertensive patients: a pilot study. J Clin
Supplemental material Periodontol 2013; 40: 681–687.
15. Matthews JB, Wright HJ, Roberts A, et al. Neutrophil
Supplemental material for this article is available online.
hyper-responsiveness in periodontitis. J Dent Res 2007;
86: 718–722.
References 16. Chapple ILC and Matthews JB. The role of reactive
1. Carrizales-Sep!ulveda EF, Ordaz-Far!ıas A, Vera-Pineda oxygen and antioxidant species in periodontal tissue
R, et al. Periodontal disease, systemic inflammation and destruction. Periodontol 2000 2007; 43: 160–232.
the risk of cardiovascular disease. Heart Lung Circ 2018; 17. Fine N, Hassanpour S, Borenstein A, et al. Distinct oral
27: 1327–1334. neutrophil subsets define health and periodontal disease
2. Lockhart PB, Bolger AF, Papapanou PN, et al. states. J Dent Res 2016; 95: 931–938.
Periodontal disease and atherosclerotic vascular disease: 18. Kumar V and Sharma A. Neutrophils: Cinderella of
does the evidence support an independent association? innate immune system. Int Immunopharmacol 2010; 10:
A scientific statement from the American Heart 1325–1334.
Association. Circulation 2012; 125: 2520–2544. 19. Kubes P. The enigmatic neutrophil: what we do not
3. Dietrich T, Sharma P, Walter C, et al. The epidemiolog- know. Cell Tissue Res 2018; 371: 399–406.
ical evidence behind the association between periodontitis 20. Brinkmann V. Neutrophil extracellular traps in the
and incident atherosclerotic cardiovascular disease. second decade. J Innate Immun 2018; 10: 414–421.
J Periodontol 2013; 84: S70–84. 21. Moonen CGJ, Hirschfeld J, Cheng L, et al. Oral neutro-
4. Beukers N, Van der Heijden G, Van Wijk AJ, et al. phils characterized: chemotactic, phagocytic, and neutro-
Periodontitis is an independent risk indicator for phil extracellular trap (NET) formation properties.
atherosclerotic cardiovascular diseases among 60 174 Front Immunol 2019; 10: 1–15.
participants in a large dental school in the Netherlands. 22. Brinkmann V, Reichard U, Goosmann C, et al.
J Epidemiol Commun Health 2017; 71: 37–42. Neutrophil extracellular traps kill bacteria. Science
5. Humphrey LL, Fu R, Buckley DI, et al. Periodontal dis- 2004; 303: 1532–1535.
ease and coronary heart disease incidence: a systematic 23. Deniset JF and Kubes P. Neutrophil heterogeneity:
review and meta-analysis. J Gen Intern Med 2008; 23: bona fide subsets or polarization states? J Leukoc Biol
2079–2086. 2018; 103: 829–838.
10 Innate Immunity 0(0)

24. Jorch SK and Kubes P. An emerging role for neutrophil periodontal therapy: a randomized controlled trial.
extracellular traps in noninfectious disease. Nat Med J Clin Periodontol 2016; 43: 778–788.
2017; 23: 279–287. 39. Tonetti MS and Claffey N. Advances in the progression
25. White P, Sakellari D, Roberts H, et al. Peripheral blood of periodontitis and proposal of definitions of a peri-
neutrophil extracellular trap production and degradation odontitis case and disease progression for use in risk
in chronic periodontitis. J Clin Periodontol 2016; 43: factor research: Group C Consensus report of the 5th
1041–1049. European Workshop in Periodontology. J Clin
26. Delgado-Rizo V, Mart!ınez-Guzmán MA, I~ niguez- Periodontol 2005; 32: 210–213.
Gutierrez L, et al. Neutrophil extracellular traps and its 40. Yang H, Biermann MH, Brauner JM, et al. New insights
implications in inflammation: an overview. Front into neutrophil extracellular traps: mechanisms of forma-
Immunol 2017; 8: 1–20. tion and role in inflammation. Front Immunol 2016; 7:
27. Munafo DB, Johnson JL, Brzezinska AA, et al. DNase I 1–8.
inhibits a late phase of reactive oxygen species production 41. Hajishengallis G. Periodontitis: from microbial immune
in neutrophils. J Innate Immun 2009; 1: 527–542. subversion to systemic inflammation. Nat Rev Immunol
28. Jiménez-Alcázar M, Rangaswamy C, Panda R, et al. 2015; 15: 30–44.
Host DNases prevent vascular occlusion by neutrophil 42. Loos BG, Papantonopoulos G, Jepsen S, et al. What is
extracellular traps. Science 2017; 358: 1202–1206. the contribution of genetics to periodontal risk? Dent Clin
29. Sohn DH. NETosis in autoimmune diseases. J Rheum North Am 2015; 59: 761–780.
Dis 2016; 23: 82–87. 43. Leite FRM, Nascimento GG, Scheutz F, et al. Effect of
30. Branzk N and Papayannopoulos V. Molecular mecha- smoking on periodontitis: a systematic review and meta-
nisms regulating NETosis in infection and disease. regression. Am J Prev Med 2018; 54: 831–841.
44. Ambrose JA and Barua RS. The pathophysiology of cig-
Semin Immunopathol 2013; 35: 513–530.
arette smoking and cardiovascular disease: an update.
31. Leffler J, Martin M, Gullstrand B, et al. Neutrophil
J Am Coll Cardiol 2004; 43: 1731–1737.
extracellular traps that are not degraded in systemic
45. White PC, Hirschfeld J, Milward MR, et al.
lupus erythematosus activate complement exacerbating
Cigarette smoke modifies neutrophil chemotaxis, neutro-
the disease. J Immunol 2012; 188: 3522–3531.
phil extracellular trap formation and inflammatory
32. Hakkim A, Furnrohr BG, Amann K, et al. Impairment
response-related gene expression. J Periodontal Res
of neutrophil extracellular trap degradation is associated
2018; 53: 525–535.
with lupus nephritis. Proc Natl Acad Sci USA 2010; 107:
46. Tonetti MS and Van Dyke TE. Periodontitis and athero-
9813–9818.
sclerotic cardiovascular disease: consensus report of the
33. Kahlenberg JM and Kaplan MJ. Mechanisms of Acute
Joint EFP/AAP Workshop on Periodontitis and
Inflammation and Vascular Injury in SLE. Duboi’s Systemic Diseases. J Clin Periodontol2013; 40: 24–29.
Lupus Erythematosus and related Syndromes By Wallace 47. Chistiakov DA, Orekhov AN and Bobryshev YV. Links
DJ and Hahn BH. (2013) pp 166–174 doi: 10.1016/B978-1- between atherosclerotic and periodontal disease.
4377-1893-5.00015-7 ISBN: 978-1-4377-1893-5: Elsevier Exp Mol Pathol 2016; 100: 220–235.
Saunders; Philadelphia, PA, USA. 48. Martinez-Herrera M, Lopez-Domenech S, Silvestre FJ,
34. Fuchs TA, Brill A, Duerschmied D, et al. Extracellular et al. Chronic periodontitis impairs polymorphonuclear
DNA traps promote thrombosis. Proc Natl Acad Sci leukocyte–endothelium cell interactions and oxidative
USA 2010; 107: 15880–15885. stress in humans. J Clin Periodontol 2018;1429–1439.
35. Knight JS, Luo W, O’Dell AA, et al. Peptidylarginine 49. Sanz M, Geriello A, Buysschaert M, et al. Scientific
deiminase inhibition reduces vascular damage and mod- evidence on the links between periodontal diseases and
ulates innate immune responses in murine models of ath- diabetes: consensus report and guidelines of the joint
erosclerosis. Circ Res 2014; 114: 947–956. workshop on periodontal diseases and diabetes by the
36. Sørensen OE and Borregaard N. Neutrophil extracellular International Diabetes Federation and the European
traps – the dark side of neutrophils. J Clin Invest 2016; Federation of Periodontology. Diabetes Res Clin Pract
126: 1612–1620. 2018; 137: 231–241.
37. Qi H, Yang S and Zhang L. Neutrophil extracellular 50. D€oring Y, Soehnlein O and Weber C. Neutrophil extra-
traps and endothelial dysfunction in atherosclerosis and cellular traps in atherosclerosis and atherothrombosis.
thrombosis. Front Immunol 2017; 8. Circ Res 2017; 120: 736–743.
38. Bizzarro S, Van der Velden U and Loos BG. Local dis- 51. Gaul DS, Stein S and Matter CM. Neutrophils in cardio-
infection with sodium hypochlorite as adjunct to basic vascular disease. Eur Heart J 2017; 38: 1702–1704.