Journal of Periodontology; Copyright 2016 DOI: 10.1902/jop.2016.

160081

Enhanced Activity of the Macrophage M1/M2 Phenotypes and

Phenotypic Switch to M1 in Periodontal Infection

Ting Yu*†, Li Zhao‡, Xin Huang*, Chanjuan Ma*, Yixiong Wang*, Jincai Zhang*§,
Dongying Xuan║

*Department of Periodontology, The Affiliated Hospital of Stomatology, Southern

Medical University, Guangzhou, China.

†Department of Periodontology, Hospital and School of Stomatology, Guangzhou

Medical University, Guangzhou, China.

‡Department of Prosthodontics, Guanghua School of Stomatology, Sun Yat-sen

University, Guangdong Provincial Key Laboratory of Stomatology, Guangzhou,
China.

§Savaid Medical School, University of Chinese Academy of Sciences, Hangzhou,

China.

║Hangzhou Dental Hospital, Savaid Medical School, University of Chinese Academy

of Sciences, Hangzhou, China.
Ting Yu and Li Zhao contribute equally to this work.

Dongying Xuan and Jincai Zhang are the co-corresponding authors.

Background: Macrophages are central players in the pathogenesis of periodontitis. However,

phenotypic switch of macrophage M1/M2 remains uncertain.

Methods: Adult male mice were divided into periodontitis (P) or control (C) groups. Bone
marrow-derived macrophages (BMMs) were stimulated with Porphyromonas gingivalis
lipopolysaccharide. In both the periodontium and serum, macrophage M1 and M2 phenotypes were
detected in vivo and in vitro, via immunofluorescence, immunohistochemistry,
electrochemiluminescence immunoassays, quantitative PCR assays and ELISAs. The M1-type markers
used included nitric oxide synthase-2 (NOS2), tumor necrosis factor alpha, interleukin-1 beta,
interleukin-6 and C-reactive protein, while the M2-type markers included arginase-1, CD206 and
interleukin-10.

Results: Compared with the C group, the P group had a 14-fold increase in F4/80+ NOS2+
cells and 4-fold more F4/80+ CD206+ cells with an enhanced NOS2/CD206 ratio in the periodontium
(P < 0.01). NOS2- CD206+ and dual NOS2+ CD206+ macrophages dominated in the C and P groups,
respectively. The P group had significantly increased M1- and M2-type cytokines in both the
periodontium and serum, and also had an enhanced interleukin-6/interleukin-10 ratio in the serum (P <
0.05). M1-type markers were significantly upregulated at the mRNA level, whereas M2-type markers
were downregulated at both the mRNA and protein levels in BMMs after lipopolysaccharide
stimulation (P < 0.01).

1
Journal of Periodontology; Copyright 2016 DOI: 10.1902/jop.2016.160081

Conclusion: Periodontal inflammation associates with an enhancement of both the M1 and

M2 phenotypes of macrophages, in which a phenotypic switch of M2 to M1 might be a critical
mechanism in mediating periodontal tissue damage including alveolar bone loss.

KEY WORDS
Periodontal disease; Macrophage; Phenotype; Nitric oxide synthase; Arginase;
Cytokine.

Periodontitis is characterized by progressive inflammation-induced alveolar bone loss

resulting from an altered host-biofilm interaction. 1 In the pathogenesis of
periodontitis, macrophages are central players with versatile functions that participate
in the initiation of inflammation, the resolution of inflammation and tissue repair, the
activation of lymphocyte-mediated adaptive immunity and the mediation of alveolar
bone resorption. 1, 2
The functional diversity of macrophages depends on the development of highly
plastic phenotypes in response to microenvironmental signals 1. For simplicity,
macrophages are commonly divided into two phenotypes/subsets, i.e., classically
(M1) and alternatively (M2) activated macrophages. 1, 3 M1 macrophages can be
primed by interferon-γ (IFN-γ) and produce high levels of the metabolic marker nitric
oxide synthase-2 (NOS2,) and the pro-inflammatory cytokines tumor necrosis factor
alpha (TNFα), interleukin-1 beta (IL1β) and interleukin-6 (IL6). 1, 3 In contrast, M2
macrophages can be primed by interleukin-4/13 (IL4/13) and expresse high levels of
the metabolic marker arginase-1 (Arg1), CD206 (also known as mannose recptor 1, a
cell surface marker) and the anti-inflammatory cytokine interleukin-10 (IL10).1, 3 M1-
like macrophages mediate bacterial killing and promote inflammation, whereas M2-
like macrophages are dominant in tissue homeostasis, the inhibition and resolution of
inflammation and tissue repair. 1, 3 An imbalanced M1/M2 ratio is a hallmark in the
pathogenesis of many inflammatory diseases, such as infection, obesity and cancer. 4
Despite the dichotomy of these phenotypes, macrophages in vivo that are shaped by
the tissue microenvironment have been suggested to acquire mixed activation states
along the entire monocytic spectrum. 1, 3
The M1/M2 paradigm is largely derived from the NOS2-Arg1 balance, in which
the two enzymes compete for arginine metabolism and thus regulate M1- and M2-
related functions 5. Nitric oxide, one of the NOS2 downstream products, is an
important mediator in initiating inflammation and killing bacteria, and polyamines
and proline, the downstream products of Arg1, are necessary substrates in tissue
proliferation and repair 5. CD206, a transmembrane C-type lectin, is a commonly-used
and specific cell surface marker that is preferentially expressed by M2-like
macrophages and functions in antigen internalization and presentation. 3, 6
Clinical and animal studies that utilized either metabolic markers, 7-11
inflammatory cytokines 11-13 or cell surface markers 11, 14 for macrophage
phenotyping, both in vivo and in vitro, have consistently suggested an enhanced
phenotype and increase in the number of M1-like macrophages in periodontitis.

2
Journal of Periodontology; Copyright 2016 DOI: 10.1902/jop.2016.160081

However, there are conflicting reports regarding changes in the M2 phenotype. 10, 12,
14, 15
Some studies found an enhancement of the M2 phenotype in periodontitis, 10, 12
whereas other studies found a decrease. 14, 15 In this context, phenotypic switch of
macrophage M1/M2 remains uncertain because of the lack of an in situ analysis of the
macrophage phenotypes using classical markers including NOS2 and CD206. 11, 12, 14
This study aimed to explore the phenotypic switch of macrophages in the context
of periodontitis by analyzing macrophage phenotypes in situ and cytokine profile in
vivo and in vitro. We hypothesized that there is an enhancement of both the M1 and
M2 phenotypes and phenotypic switch of macrophages is toward M1 in periodontitis.

MATERIALS AND METHODS

Ethics Statement
Animals were provided by and cared for at the Guangdong Medical Laboratory
Animal Center. The entire animal study was approved by the Animal Ethics
Committee of Guangdong Medical Laboratory Animal Center and in accordance with
the National Institutes of Health guide for the care and use of laboratory animals.

Experimental Periodontitis
Twenty C57 BL/6J mice (male, 6 weeks old) were randomly divided into
periodontitis (P) or control (C) groups (n = 10 per group). Porphyromonas gingivalis
(P.g) ¶ was cultured as previously described. 16 Under anesthesia by 4% chloral
hydrate (i.p.), the P group was ligated at the bilateral maxillary second molars with
P.g-adhered ligature (5-0 silk) for 10 days to induce periodontitis. As the control, the
C group was sham-ligated with sterile silk that was removed at once.
On day 10, all the mice were euthanized via cardiac puncture. Fasting serum was
separated. Gingival tissue was harvested from one side of the upper jaw for RNA
extraction, leaving the bare alveolar bone for micro-computed tomography (Micro-
CT) and morphometric analysis. The other side of the upper jaw was fixed in 4%
paraformaldehyde for histology.

Micro-CT and Morphometric Analysis of Alveolar Bone Loss

Alveolar bone was scanned using a Micro-CT system # at an 8-μm resolution
followed by reconstruction of the Micro-CT images. The volume of interest was
delineated on the axial planes between the mesial root surface of the first molar and
the distal root surface of the third molar. 17, 18 The contours were delineated
continuously every five data planes from the furcation roof to the root apex until a
three-dimensional volume of interest was generated. 17, 18 The total volume was
calculated by excluding the root volume from the volume of interest. 17, 18 The bone
volume fraction was calculated by dividing the remaining bone volume by the volume
of interest 17, 18. After Micro-CT analysis, the alveolar bone was stained with 1%
methylene blue as previously described. 19 The average distance from the

3
Journal of Periodontology; Copyright 2016 DOI: 10.1902/jop.2016.160081

cementoenamel junction to the alveolar bone crest at 18 sites on the three molars was
measured on photographs. 20

Leukocyte and Osteoclast Enumeration

See the supplementary appendix in online Journal of Periodontology. Osteoclasts
were stained with tartrate-resistant acid phosphatase (TRAP). ** 21

Immunohistochemistry
Immunohistochemistry was performed with two-step methods. Primary antibodies ††
against mouse CD68 (1:100), TNFα (1:100), IL1β (10 μg/mL), IL10 (1:100) and Arg1
(1:200) were used. For the blank control, primary antibodies were replaced with
phosphate buffered saline. For the color reaction, 3,3’-diaminobenzidine was used.
Immuno-stained sections were counterstained with hematoxylin. Positive cells that
expressed TNFα, IL1β or IL10 were counted in one field (400×) mesial and distal to
the second molars, and the number was divided by the number of total cells in the
same filed to determine the percentage of positive cells.

Immunofluorescence
Two-color immunofluorescence (anti- F4/80 and anti-NOS2 or anti-F4/80 and anti-
CD206) was performed with two-step methods. Primary antibodies ‡‡ against mouse
F4/80 (rat IgG, 1:20), NOS2 (rabbit IgG, 1:100) and CD206 (rabbit IgG, 1:200) were
utilized. For anti-F4/80 and anti-NOS2 double-staining, donkey anti-rat §§ and goat
anti-rabbit ǁǁ secondary antibodies were used. For anti-F4/80 and anti-CD206
double-staining, goat anti-rat ¶¶ and goat anti-rabbit secondary antibodies were used.
Cell nuclei were counterstained by 4′,6-diamidino-2-phenyl-indole (DAPI). Confocal
microscopy images mesial and distal to the second molars were captured with a laser
scanning microscope ##, and the corresponding software was used for the image
processing. Adjacent sections were used for the two-color staining of F4/80 and
NOS2 and F4/80 and CD206 so that the ratio of NOS2+ cells to CD206+ cells could
be calculated. The NOS2+ CD206- (strict M1), NOS2- CD206+ (strict M2) and NOS2+
CD206+ (mixed M1-M2) cell percentages in the F4/80+ macrophages were estimated
from the two-color staining data based on the set concept.

Cell Culture
The bone marrow cells of the femur and tibia from C57 BL/6J mice (male, 6 weeks
old) were isolated and cultured in Dulbecco’s modified Eagle’s medium containing
10% fetal bovine serum ***, 1% penicillin-streptomycin †††, and 10 ng/mL
macrophage colony-stimulating factor (M-CSF) ‡‡‡. On day 3, the cells were
supplemented with complete medium containing M-CSF. On day 7, the mature bone
marrow-derived macrophages (BMMs) were re-plated into 6-well plates (106
cells/well) and cultured for 1 day. Then, the BMMs were stimulated with medium
containing 10 ng/mL P.g lipopolysaccharide (LPS) §§§ or control medium for 2 or 4
hours followed by cell harvesting.

4
Journal of Periodontology; Copyright 2016 DOI: 10.1902/jop.2016.160081

Quantitative PCR
The total RNA of gingiva and BMMs was extracted, reverse transcribed, and
quantified via a fluorescence quantitative PCR assay ǁǁǁ using a real-time PCR
analyzer ¶¶¶ . For the gingiva, the mRNA levels of receptor activator of nuclear factor
kappa-B ligand (RANKL), CD68, TNFα, IL1β, IL6 and Arg1 were determined, with
glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as the endogenous control. For
the BMMs, the mRNA levels of TNFα, IL1β, IL6, NOS2, IL10 and Arg1 were
determined, with peptidylprolyl isomerase A (PPIA) as the endogenous control. 22
The primers used were referred to in a non-profit platform ### as follows (5’ to 3’,
forward/reverse): Gapdh AGGTCGGTGTGAACGGATTTG,
GGGGTCGTTGATGGCAACA; Ppia GAGCTGTTTGCAGACAAAGTTC,
CCCTGGCACATGAATCCTGG; Rankl CAGCATCGCTCTGTTCCTGTA,
CTGCGTTTTCATGGAGTCTCA; Cd68 TGTCTGATCTTGCTAGGACCG,
GAGAGTAACGGCCTTTTTGTGA; Tnfa CAGGCGGTGCCTATGTCTC,
CGATCACCCCGAAGTTCAGTAG; Il1b GAAATGCCACCTTTTGACAGTG,
TGGATGCTCTCATCAGGACAG; Il6 CTGCAAGAGACTTCCATCCAG,
AGTGGTATAGACAGGTCTGTTGG; Il10 CTTACTGACTGGCATGAGGATCA,
GCAGCTCTAGGAGCATGTGG; Nos2 GTTCTCAGCCCAACAATACAAGA,
GTGGACGGGTCGATGTCAC; and Arg1 CTCCAAGCCAAAGTCCTTAGAG,
GGAGCTGTCATTAGGGACATCA. The relative mRNA levels were calculated
using the 2-ΔΔCt method.

Cytokine Detection in Serum and BMM Supernatants

C-reactive protein (CRP) in serum and IL1β and IL10 in BMM supernatants were
detected with ELISA kits ****. Serum TNFα, IL1β, IL6 and IL10 were detected
using electrochemiluminescence immunoassays ††††.

Statistics
Statistical analysis was performed using commercially available software ‡‡‡‡. Two
independent samples were compared using Student’s t test. The normally distributed
data were presented as the means ± SD. For the genetic data, the ΔCt value was used
for comparisons, and the median was presented in the description. For the data
estimated based on the set concept, the non-parametric Mann-Whitney U test was
used for comparisons, and an estimated range was presented in the description.
Statistical significance was set at P < 0.05.

RESULTS

Establishment of the Periodontitis Model

The periodontitis model was confirmed by vertical bone loss, bone volume fraction,
leukocyte and osteoclast counts, and osteoclast activity-related genes. The vertical
bone loss in the P group was significantly increased compared with that in the C

5
Journal of Periodontology; Copyright 2016 DOI: 10.1902/jop.2016.160081

group (P < 0.001), as revealed by the morphometric analysis (Figure 1A, B, I).
Conversely, the bone volume fraction was significantly decreased in the P group
relative to that in the C group (P < 0.001), as revealed by the Micro-CT analysis
(Figure 1C, D, J). The numbers of leukocytes infiltrating into the periodontium and
osteoclasts on the alveolar bone surface were both significantly greater in the P group
than those in the C group (P < 0.005), as indicated by the hematoxylin and eosin
(H&E) (Figure 1E, F, K) and TRAP staining (Figure 1G, H, L), respectively.
Moreover, the mRNA level of pro-osteoclastic RANKL was significantly upregulated
in the P group relative to that in the C group (P < 0.05, Figure 1M). These data
suggested that the periodontitis model was successfully established.

M1 and M2 Localization in the Periodontium

To localize macrophages in the periodontium, macrophages were labeled with the
specific cell surface markers CD68 or F4/80. CD68+ cells were sharply increased in
the gingival epithelium and sub-epithelial connective tissue in the P group compared
with those in the C group, as shown by immunohistochemistry (Figure 2A, B), which
was consistent with the change in CD68 at the mRNA level (P < 0.05, Figure 2C).
There was a similar change for F4/80, as revealed by immunofluorescence (Figure
2D, G), in which F4/80+ cells accounted for 24.1% of the total cells in the P group vs.
only 4.7% of the total cells in the C group in the periodontium (P < 0.005, Table 1).
To differentiate between the macrophage subsets in the periodontium, M1 and M2
macrophages were double-labeled with antibodies to F4/80 and NOS2 (a cytoplasmic
marker for the M1 phenotype) or F4/80 and CD206 (a cell surface marker for M2
macrophages), respectively. Compared with the C group, the F4/80+ CD206+ cell
(M2) percentage of the total cells was increased by 4-fold in the P group (P < 0.005),
as indicated by immunofluorescence (Figure 2F, I and Table 1). In contrast, the
F4/80+ NOS2+ cell (M1) percentage of the total cells was increased by 14-fold in the
P group relative to that in the C group (P < 0.001, Figure 2E, H and Table 1).
Among the F4/80+ macrophages, the CD206+ cells accounted for nearly all of the
macrophages in both the groups with no significant difference between them (P:C,
97.4 ± 0.8 : 95.4 ± 1.9 (%)). In contrast, the NOS2+ cells accounted for nearly all of
the macrophages in the P group with a 2-fold enhancement of the percentage of
NOS2+ cells compared with that in the C group (P = 0.001). The ratio of NOS2+ to
CD206+ cells was significantly greater in the P group than in the C group (P < 0.01,
Figure 3 and Table 1).
Based on the data from the two-color staining of F4/80 and NOS2 and F4/80 and
CD206, an estimated range was calculated for the NOS2+ CD206-, NOS2- CD206+
and NOS2+ CD206+ cell percentage among the F4/80+ macrophages (Table 1). Based
on this estimation, the percentage of NOS2- CD206+ cells among the F4/80+
macrophages was considerably lower in the P group than in the C group (0-2.2 vs.
64.5-69.1 (%), P < 0.05). The NOS2+ CD206+ cells, a mixed subset, accounted for
nearly all of the macrophages in the P group, in which the percentage increased 2-fold
relative to that in the C group (95.2-97.4 vs. 26.3-30.9 (%), P < 0.05). The percentage
6
Journal of Periodontology; Copyright 2016 DOI: 10.1902/jop.2016.160081

of NOS2+ CD206- cells among the macrophages was 0.4%-2.6% and 0-4.6% in the P
and C groups, respectively, and the difference was difficult to compare because of the
overlapping percentage ranges.

M1- and M2-Type Inflammatory Cytokines in the Periodontium

To explore macrophage phenotype-related inflammatory functions, the inflammatory
cytokines TNFα, IL1β, IL6 and IL10, which are primarily expressed by macrophages,
1, 23
were localized in the periodontium via immunohistochemistry. Consistent with
the results from the anti-NOS2/anti-CD206 immunofluorescence assays, the
percentages of cells that positively expressed the M1-type cytokines TNFα or IL1β or
the M2-type cytokine IL10 were all significantly increased by 2- to 4-fold in the P
group relative to the percentages in the C group (P < 0.005, Figure 4B-D, G-I, K-M).
Additionally, the M1-type cytokine IL1β and the M2-type marker Arg1, an enzyme
that competes with NOS2 in arginine metabolism, were both significantly upregulated
at the mRNA level in the P group compared with the C group (Figure 4N, O). The
modest upregulation of Arg1 in the P group was focused in the gingival epithelium
and was also present in the deeper connective tissue with leukocyte involvement, as
indicated by immunohistochemistry (Fig. 4E, J).

M1- and M2- Type Inflammatory Cytokines in Serum

To explore the consistency of the systemic response regarding the phenotypes of
tissue macrophages, which are primarily recruited from the circulation under
inflammatory conditions, 1 M1- and M2-type inflammatory cytokines in the serum
were detected. Compared with the C group, the M1-type cytokine IL6 and the M2-
type cytokine IL10 in the serum were both significantly increased in the P group (P <
0.05), despite no significant changes in the M1-type cytokines TNFα, IL1β or CRP
(Figure 5A). Moreover, the M1/M2 ratio, calculated based on the serum IL6/IL10
ratio, was significantly greater in the P group than in the C group (P < 0.05) (Figure
5A).

M1- and M2-Type Markers in BMMs Stimulated by P.g LPS

Using primary BMMs, changes in the macrophage phenotypes in response to
periodontal pathogens were investigated in vitro. On a challenge with P.g LPS,
BMMs exhibited no significant changes in the M1- or M2-type genes until 4 hours
after the stimulation, when the M1-type cytokines IL1β and IL6 were upregulated by
~2-fold (P < 0.01), and the M2-type markers IL10 and Arg1 were downregulated by
~2-fold at the mRNA level (P < 0.01) (Figure 5B). In the BMM supernatants, the
protein level of IL1β was not significantly changed by either a 2-hour or 4-hour
stimulation, whereas that of IL10 was significantly downregulated by stimulation for
both 2 and 4 hours stimulation (P < 0.05, Figure 5C).

7
Journal of Periodontology; Copyright 2016 DOI: 10.1902/jop.2016.160081

DISCUSSION

This animal study investigated phenotypic changes of macrophages in periodontitis in

vivo (especially in situ) and in vitro using multiple types of phenotypic markers. It
was found that both the phenotypes M1 and M2 are enhanced in periodontitis, as
revealed by the increased percentage of NOS2+ and CD206+ cells in the periodontium
and enhanced levels of pro- and anti-inflammatory cytokines in both the periodontium
and serum. There is phenotypic switch of M2 to M1 (M1 polarization) in
periodontitis, as indicated by the enhanced ratios of NOS2/CD206 in the
periodontium and of IL6/IL10 in the serum as well as the increased ratio of M1-
type/M2-type markers in BMMs. The NOS2- CD206+ macrophages are dominant
under normal conditions, whereas a mixed subset of NOS2+ CD206+ macrophages is
dominant in periodontitis. These data suggest that both the M1 and M2 phenotypes of
macrophages are activated in periodontitis and the phenotypic switch seems toward
M1 macrophages.
Previous studies have consistently suggested an enhanced activity including the
number of M1-like macrophages. 7-14 However, there are conflicting results regarding
the M2 phenotype and phenotypic switch of macrophages is still undefined because of
the lack of in situ analyses. 10, 12, 14, 15 In this context, the results of the present study
are novel regarding the co-localization of F4/80 and NOS2 and F4/80 and CD206 in
the periodontium. In the present study, in addition to an enhanced number of M1-like
(NOS2+) macrophages in the inflamed periodontium, which is consistent with
previous studies, 8 the number of M2-like (CD206+) macrophages was also enhanced,
which has not been reported previously. Moreover, the present study found an
enhanced NOS2/CD206 (M1/M2) ratio in the inflamed periodontium, indicating
phenotypic switch of M2 to M1 (M1 polarization) in periodontitis. Another animal
study also found an increased M1/M2 ratio in inflamed gingiva using flow cytometry,
however, the number of M2-like macrophages was decreased possibly because of the
much longer infection duration (~2 months). 14
The enhanced activity of both the phenotypes M1 and M2 in periodontitis is
further supported by the enhanced levels of pro- and anti-inflammatory cytokines in
both the inflamed periodontium and serum, which has been previously well
documented. 13, 24, 25 Macrophages are the primary source for both the pro- and anti-
inflammatory cytokines, including TNFα, IL1β, IL6 and IL10, 1, 23 which function via
certain common pathways including metalloproteinases and osteoclastogenesis to
cause or to abate periodontal tissue damage. 24, 26 Under conditions of periodontitis,
various signals including the M1-priming (IFNγ) and M2-priming (IL-4) cytokines
upregulated in T helper cells and the pro- and anti-inflammatory cytokines
upregulated in macrophages themselves may act together to shape the macrophage
phenotype, 4, 12, 27 and activate both M1- and M2-like functions. Enhanced M1-like
functions are necessary to fight large amounts of bacteria, whereas enhanced M2-like
functions may be necessary to restrain the magnitude of inflammation. 1, 28 With

8
Journal of Periodontology; Copyright 2016 DOI: 10.1902/jop.2016.160081

phenotypic switch to M1 such as the enhanced NOS2/CD206 ratio in this study and
the enhanced ratios of pro-inflammatory to anti-inflammatory cytokines in previous
studies, 13, 25 the inflammatory pattern would be skewed from protective to chronic
and destructive. 2, 24, 28
Phenotypic switch to M1 also seems to occur in human gingival tissue with
experimental gingivitis or with drug-induced gingival hyperplasia as with severe
inflammation. 29-31 The similarity of the phenotypic switch to M1 among periodontal
diseases indicates that M1 polarization might be a critical mechanism in driving the
development of periodontal inflammation and the related consequences. Phenotypic
switch to M1 might be explained by a pro-inflammatory state in the circulation (such
as the enhanced IL6/IL10 ratio in the serum in the present study), which would recruit
more pro-inflammatory monocytes to the infection sites, or by a phenotypic switch of
the tissue macrophages from M2 to M1 at local sites. 1, 32
Another novel finding in the present study is that the number of CD206+
macrophages is not simply enhanced independent of the NOS2+ macrophages.
Instead, the CD206+ macrophages account for nearly all the macrophages both under
the normal condition and in periodontitis, and the differences between the two
conditions are primarily in terms of the NOS2+ and NOS2- cell percentages within the
CD206+ macrophages. Under normal conditions, the CD206+ NOS2- macrophages are
dominant, and the dual CD206+ NOS2+ macrophages play a small role (1/4-1/3). In
contrast, in periodontitis, the CD206+ NOS2+ cells are markedly increased and
account for nearly all the macrophages. These findings indicated that the development
of periodontitis appears to be associated with an enhanced number of a mixed subset
of CD206+ NOS2+ macrophages. An enhanced number of CD206+ NOS2+
macrophages has been related to the resolution of inflammation by expressing high
levels of IL10 and low levels of TNFα in experimental peritonitis. 33 Whereas, in the
present study, the dual CD206+ NOS2+ macrophages appear to express enhanced
levels of both TNFα and IL10, suggesting that the CD206+ NOS2+ macrophages in
the inflamed periodontium are more pro-inflammatory than pro-resolving. Under
normal conditions, there is a co-existence of the CD206+ NOS2- and CD206+ NOS2+
macrophages, which might be necessary for tissue homeostasis and immune defense
against the commensal oral flora. 1, 34 In contrast, there are nearly no NOS2+ CD206-
cells (the strict M1 phenotype) in both the conditions, possibly because an M2-related
inherent negative feedback mechanism exists in both conditions. 28 Regardless,
functionally distinct macrophage subsets appear to exist in periodontium and differ
between the two conditions, in which the respective roles of these cells, including the
CD206+ NOS2+ macrophages, are not yet clear, and require a functional analysis of
the macrophage expression patterns.
The enhanced levels of both NOS2 and Arg1 in the periodontium with
periodontitis in this study are consistent with the changes in gingival NOS2 and
salivary Arg1 in patients with chronic periodontitis. 7-9 In addition, the NOS2+ cells
were highly associated with F4/80+ macrophages, similar to the co-localization results
of NOS2 and CD68 by Hussain et al (2016). 8 However, these authors found that the
9
Journal of Periodontology; Copyright 2016 DOI: 10.1902/jop.2016.160081

NOS2+ cells are primarily localized in the gingival epithelium, 8 whereas those in the
present study appeared to be evenly distributed across the epithelium and sub-
epithelial connective tissue. The differences in the distribution patterns of NOS2
might indicate changes in the distribution of macrophages at various stages of
inflammation. Enhanced levels of NOS2 upregulate the level of nitric oxide which is
used to kill bacteria but can also activate the pathways of metalloproteinases and
osteoclasts and thereby cause periodontal tissue damage. 35 The present study found
for the first time that the enhanced expression of Arg1 in periodontitis was mainly
localized in the gingival epithelium but was also present in the connective tissue with
leukocyte involvement. The distribution pattern of Arg1 in periodontitis might be
associated with the reactive hyperplasia of the gingival epithelium, given that Arg1
has been found to participate in not only anti-inflammation but also tissue repair and
hyperplasia. 5, 36-38 Arg1 has also been associated with an impaired immune response
in certain chronic inflammatory conditions such as obesity. 39 For instance, obesity-
educated macrophages have an upregulated level of Arg1 but have blunted M1-like
functions, which could be recovered by inhibiting the expression of Arg1. 39 Thus,
NOS2 and Arg1 may both have two sides, and the inflammatory status should be
considered when interpreting changes of the NOS2-Arg1 balance and the potential
pathological significance of this balance.
This study was limited by using only a single time point for inducing periodontitis
because macrophages might differ in their functions at different stages of
periodontitis. 1 For example, M2-like macrophages have been suggested to be
dominant at the resolution stage of periodontal inflammation. 1 Therefore, an
exploration of the dynamics of macrophage phenotypes in periodontitis is required in
future studies. Additionally, the M1/M2 paradigm is largely based on animal and in
vitro studies, and the precise manifestation of macrophage phenotypes in humans is
expected to differ substantially from that in animals, 4, 40 therefore, more
investigations of the human version of periodontitis are required.

CONCLUSION

In summary, periodontal inflammation associates with an enhancement of both the

M1 and M2 phenotypes of macrophages, in which a phenotypic switch of M2 to M1
might be a critical mechanism in mediating periodontal tissue damage, including
alveolar bone loss.

CONFLICT OF INTEREST
No potential conflicts of interest are declared.

ACKNOWLEDGMENTS
This study was supported by the National Natural Science Foundation of China, Beijing, China (Nos.
81271160, 81371151 and 81470750).

10
Journal of Periodontology; Copyright 2016 DOI: 10.1902/jop.2016.160081

REFERENCES
1. Sima C, Glogauer M. Macrophage subsets and osteoimmunology: tuning of the immunological
recognition and effector systems that maintain alveolar bone. Periodontol 2000 2013;63:80-101.

2. Hasturk H, Kantarci A, Van Dyke TE. Oral inflammatory diseases and systemic inflammation:
role of the macrophage. Front Immunol 2012;3:118.

3. Morris DL, Singer K, Lumeng CN. Adipose tissue macrophages: phenotypic plasticity and
diversity in lean and obese states. Curr Opin Clin Nutr Metab Care 2011;14:341-346.

4. Wynn TA, Chawla A, Pollard JW. Macrophage biology in development, homeostasis and disease.
Nature 2013;496:445-455.

5. Rath M, Muller I, Kropf P, Closs EI, Munder M. Metabolism via arginase or nitric oxide synthase:
two competing arginine pathways in macrophages. Front Immunol 2014;5:532.

6. Taylor PR, Martinez-Pomares L, Stacey M, Lin HH, Brown GD, Gordon S. Macrophage receptors
and immune recognition. Annu Rev Immunol 2005;23:901-944.

7. Lappin DF, Kjeldsen M, Sander L, Kinane DF. Inducible nitric oxide synthase expression in
periodontitis. J Periodontal Res 2000;35:369-373.

8. Hussain QA, McKay IJ, Gonzales-Marin C, Allaker RP. Detection of adrenomedullin and nitric
oxide in different forms of periodontal disease. J Periodontal Res 2016;51:16-25.

9. Ozmeric N, Elgun S, Uraz A. Salivary arginase in patients with adult periodontitis. Clin Oral
Investig 2000;4:21-24.

10. Gheren LW, Cortelli JR, Rodrigues E, Holzhausen M, Saad WA. Periodontal therapy reduces
arginase activity in saliva of patients with chronic periodontitis. Clin Oral Investig 2008;12:67-72.

11. Holden JA, Attard TJ, Laughton KM, Mansell A, O'Brien-Simpson NM, Reynolds EC.
Porphyromonas gingivalis lipopolysaccharide weakly activates M1 and M2 polarized mouse
macrophages but induces inflammatory cytokines. Infect Immun 2014;82:4190-4203.

12. Navarrete M, Garcia J, Dutzan N, et al. Interferon-gamma, interleukins-6 and -4, and factor XIII-A
as indirect markers of the classical and alternative macrophage activation pathways in chronic
periodontitis. J Periodontol 2014;85:751-760.

13. Gorska R, Gregorek H, Kowalski J, Laskus-Perendyk A, Syczewska M, Madalinski K.

Relationship between clinical parameters and cytokine profiles in inflamed gingival tissue and
serum samples from patients with chronic periodontitis. J Clin Periodontol 2003;30:1046-1052.

14. Lam RS, O'Brien-Simpson NM, Lenzo JC, et al. Macrophage depletion abates Porphyromonas
gingivalis-induced alveolar bone resorption in mice. J Immunol 2014;193:2349-2362.

15. Gullu C, Ozmeric N, Tokman B, Elgun S, Balos K. Effectiveness of scaling and root planing
versus modified Widman flap on nitric oxide synthase and arginase activity in patients with
chronic periodontitis. J Periodontal Res 2005;40:168-175.

16. Kimura S, Nagai A, Onitsuka T, et al. Induction of experimental periodontitis in mice with
Porphyromonas gingivalis-adhered ligatures. J Periodontol 2000;71:1167-1173.

11
Journal of Periodontology; Copyright 2016 DOI: 10.1902/jop.2016.160081

17. Park CH, Abramson ZR, Taba MJ, et al. Three-dimensional micro-computed tomographic imaging
of alveolar bone in experimental bone loss or repair. J Periodontol 2007;78:273-281.

18. Zhang W, Ju J, Rigney T, Tribble G. Porphyromonas gingivalis infection increases osteoclastic

bone resorption and osteoblastic bone formation in a periodontitis mouse model. BMC Oral Health
2014;14:89.

19. Li CH, Amar S. Morphometric, histomorphometric, and microcomputed tomographic analysis of

periodontal inflammatory lesions in a murine model. J Periodontol 2007;78:1120-1128.

20. Abe T, Hajishengallis G. Optimization of the ligature-induced periodontitis model in mice. J

Immunol Methods 2013;394:49-54.

21. Saadi-Thiers K, Huck O, Simonis P, et al. Periodontal and systemic responses in various mice
models of experimental periodontitis: respective roles of inflammation duration and
Porphyromonas gingivalis infection. J Periodontol 2013;84:396-406.

22. Moreno-Navarrete JM, Escote X, Ortega F, et al. A role for adipocyte-derived lipopolysaccharide-
binding protein in inflammation- and obesity-associated adipose tissue dysfunction. Diabetologia
2013;56:2524-2537.

23. Arango DG, Descoteaux A. Macrophage cytokines: involvement in immunity and infectious
diseases. Front Immunol 2014;5:491.

24. Liu YC, Lerner UH, Teng YT. Cytokine responses against periodontal infection: protective and
destructive roles. Periodontol 2000 2010;52:163-206.

25. Passoja A, Puijola I, Knuuttila M, et al. Serum levels of interleukin-10 and tumour necrosis factor-
alpha in chronic periodontitis. J Clin Periodontol 2010;37:881-887.

26. Kayal RA. The role of osteoimmunology in periodontal disease. Biomed Res Int
2013;2013:639368.

27. Cassetta L, Cassol E, Poli G. Macrophage polarization in health and disease.

ScientificWorldJournal 2011;11:2391-2402.

28. Ivashkiv LB. Inflammatory signaling in macrophages: Transitions from acute to tolerant and
alternative activation states. Eur J Immunol 2011;41:2477-2481.

29. Zwadlo G, Voegeli R, Schulze OK, Sorg C. A monoclonal antibody to a novel differentiation
antigen on human macrophages associated with the down-regulatory phase of the inflammatory
process. Exp Cell Biol 1987;55:295-304.

30. Topoll HH, Zwadlo G, Lange DE, Sorg C. Phenotypic dynamics of macrophage subpopulations
during human experimental gingivitis. J Periodontal Res 1989;24:106-112.

31. Iacopino AM, Doxey D, Cutler CW, et al. Phenytoin and cyclosporine A specifically regulate
macrophage phenotype and expression of platelet-derived growth factor and interleukin-1 in vitro
and in vivo: possible molecular mechanism of drug-induced gingival hyperplasia. J Periodontol
1997;68:73-83.

32. Italiani P, Boraschi D. From monocytes to M1/M2 macrophages: phenotypical vs. functional
differentiation. Front Immunol 2014;5:514.

12
Journal of Periodontology; Copyright 2016 DOI: 10.1902/jop.2016.160081

33. Bystrom J, Evans I, Newson J, et al. Resolution-phase macrophages possess a unique

inflammatory phenotype that is controlled by cAMP. Blood 2008;112:4117-4127.

34. Irie K, Novince CM, Darveau RP. Impact of the oral commensal flora on alveolar bone
homeostasis. J Dent Res 2014;93:801-806.

35. Menaka KB, Ramesh A, Thomas B, Kumari NS. Estimation of nitric oxide as an inflammatory
marker in periodontitis. J Indian Soc Periodontol 2009;13:75-78.

36. Kampfer H, Pfeilschifter J, Frank S. Expression and activity of arginase isoenzymes during normal
and diabetes-impaired skin repair. J Invest Dermatol 2003;121:1544-1551.

37. Singh K, Coburn LA, Barry DP, Boucher JL, Chaturvedi R, Wilson KT. L-arginine uptake by
cationic amino acid transporter 2 is essential for colonic epithelial cell restitution. Am J Physiol
Gastrointest Liver Physiol 2012;302:G1061-G1073.

38. Kitowska K, Zakrzewicz D, Konigshoff M, et al. Functional role and species-specific contribution
of arginases in pulmonary fibrosis. Am J Physiol Lung Cell Mol Physiol 2008;294:L34-L45.

39. Richard G, Trivedi N, Belta C, Amar S. Partial restoration of macrophage alteration from diet-
induced obesity in response to Porphyromonas gingivalis infection. PLoS One 2013;8:e70320.

40. Dalmas E, Clement K, Guerre-Millo M. Defining macrophage phenotype and function in adipose
tissue. Trends Immunol 2011;32:307-314.

Contact information: Dongying Xuan, Hangzhou Dental Hospital, University of

Chinese Academy of Sciences, No.1 Pinghai Road, Hangzhou 310000, China. Fax:
+86 571 87283570. E-mail: xuanxuan187@126.com
Submitted February 07, 2016; accepted for publication April 07, 2016.

Figure 1

Establishment of the periodontitis model as confirmed by vertical bone loss, bone volume fraction,
leukocyte and osteoclast count and osteoclastic activity-related genes.

The periodontitis group (A, C, E, G) exhibits significantly increased vertical bone loss (A, B, I),
leukocyte infiltrate in the periodontium (E, F, K), osteoclastogenesis (G, H, L) and osteoclastic activity
(M) and a decreased bone volume fraction (C, D, J) compared with those of the control group (B, D,
F, H). Red lines indicate the cementoenamel junction-alveolar bone crest distance at 18 sites of three
molars. The data are presented as the means ± SD for CEJ-ABC (n = 8-9/group), bone volume fraction
(n = 9/group), and the numbers of leukocytes (n = 4/group) and TRAP+ cells (n = 5-8/group), and the
median is presented for Rankl (n = 5-9/group). P, periodontitis group; C, control group; ABC,
alveolar bone crest; Den, dentin; CEJ-ABC, cementoenamel junction-alveolar bone crest distance;
TRAP, tartrate-resistant acid phosphatase; rankl, receptor activator of nuclear factor kappa-B ligand;
white scale bar, 500 μm; black scale bar, 50 μm.

13
Journal of Periodontology; Copyright 2016 DOI: 10.1902/jop.2016.160081

Figure 2

Single-color immuno-staining for macrophage markers in the periodontium with and without
periodontitis.

The periodontitis group (A, D-F) has markedly increased expression of the macrophage markers CD68
(A-C) and F4/80 (D, G), the M1 marker NOS2 (E, H) and the M2 marker CD206 (F, I) compared with
those of the control group (B, G-I). For the gene Cd68, the data are presented as the median (n = 8-
10/group). Scale bar, 50 μm; P, periodontitis group; C, control group.

Figure 3

Two-color immunofluorescence for M1 and M2 macrophages in the periodontium with and without
periodontitis.

The F4/80+ macrophages express more NOS2 in the periodontitis group (A-C) than in the control
group (D-F). In contrast, nearly all the F4/80+ macrophages are CD206+ in both the periodontitis (G-
I) and control (J-L) groups. Short arrows indicate representative double-stained cells that co-express
F4/80 and NOS2 or F4/80 and CD206. Triangular arrowheads indicate single-stained cells that
express F4/80 but not NOS2 or F4/80 but not CD206. Scale bar, 20 μm.

Figure 4

Expression of M1- and M2-type inflammatory cytokines in the periodontium with and without
periodontitis.

The periodontitis group (B-D) has significantly increased expression of the M1-type cytokines TNFα
(B, G, K) and IL1β (C, H, L) and the M2-type cytokine IL10 (D, I, M) compared with those of the
control group (G-I), as revealed by immunohistochemistry. Additionally, the periodontitis group has
significantly increased expression of the M1-type cytokine IL1β (N) and the M2-type marker Arg1 (O)
at the mRNA level. Anti-Arg1 immunohistochemistry indicates that the modest upregulation of Arg1 in
the periodontitis group (E) relative to that in the control group (J) is focused in gingival epithelium
and is also present in deeper connective tissue with leukocytes involvement. The data are presented as
the means ± SD for TNFα (n = 5/group), IL1β (n = 5/group) and IL10 (n = 4/group) and as the median
for Il1b (n = 10/group) and Arg1 (n = 10/group). Rectangles in the H&E-stained sections (A, F)
indicate a representative area from which the partial enlarged views of the immuno-stained sections
(B-E, G-J) originate. Scale bar, 50 μm; P, periodontitis group; C, control group; TNFα, tumor
necrosis factor α; Il1b/IL1β, interleukin-1β; IL10, interleukin-10; Arg1, arginase-1.

Figure 5

M1- and M2-type inflammatory cytokines in the serum with and without periodontitis and the
expression of M1- and M2-type markers in BMMs stimulated with P.g LPS.

(A) Compared with the control group, the periodontitis group has increased levels of the M1-type
cytokine IL6 and M2-type cytokine IL10 and an enhanced IL6/IL10 ratio in the serum. The data are
presented as the means ± SD (n = 4-8/group). (B) In the P.g LPS-stimulated BMMs, the M1- and M2-
type genes, including inflammatory cytokines, are not significantly changed until 4 hours after
stimulation, when the M1-type genes Il1b and Il6 are upregulated by ~2-fold, whereas the M2-type

14
Journal of Periodontology; Copyright 2016 DOI: 10.1902/jop.2016.160081

genes Il10 and Arg1 are downregulated by ~2-fold compared with baseline levels. The data are
presented as the median (n = 6-10/group). (C) In the BMM supernatants, the M1-type cytokine IL1β is
not changed by either a 2-hour or 4-hour stimulation, whereas the M2-type cytokine IL10 is
downregulated by both 2-hour and 4-hour stimulations. The data are presented as the means ± SD (n =
3-5/group). BMMs, bone marrow-derived macrophages; LPS, lipopolysaccharide; P group,
periodontitis group; C group, control group; Tnfa/TNFα, tumor necrosis factor α; Il1b/IL1β,
interleukin-1β; Il6/IL6, interleukin-6; Il10/IL10, interleukin-10; Nos2, nitric oxide synthase-2; Arg1,
arginase-1; CRP, C-reactive protein.

Table 1

M1 and M2 percentage and M1/M2 ratio in the periodontium with and without periodontitis.
Ratio of positive cells Periodontitis group Control group P value
Mφ/total cells ( F4/80+/DAPI+) 24.1% ± 7.0% 4.7% ± 2.1% 0.002
+ + +
M1/total cells (F4/80 NOS2 /DAPI ) 20.9% ± 4.8% 1.4% ± 1.1% < 0.001
M2/total cells (F4/80+ CD206+/DAPI+) 23.5% ± 6.8% 4.6% ± 2.0% 0.002
+ + + 30.9% ±
M1/total macrophages (F4/80 NOS2 /F4/80 ) 97.8% ± 0.7% 0.001
11.7%
M2/total macrophages (F4/80+ CD206+/F4/80+) 97.4% ± 0.8% 95.4% ± 1.9% 0.117
M1/M2 ratio (F4/80+ NOS2+/F4/80+ CD206+) 1.00 ± 0.02 0.38 ± 0.09 0.006
Strict M1/total macrophages (F4/80+ NOS2+
0.4%-2.6% 0-4.6% -
CD206-/F4/80+)
Strict M2/total macrophages (F4/80+ NOS2-
0-2.2% 64.5%-69.1% 0.016
CD206+/F4/80+)
Mixed M1-M2/total macrophages (F4/80+
95.2%-97.4% 26.3%-30.9% 0.016
NOS2+ CD206+/F4/80+)

Total macrophages, M1, M2 and total cells were recognized as F4/80+, F4/80+ NOS2+, F4/80+ CD206+
and DAPI+, respectively. The data from the three-color staining regarding the ratios of strict M1, strict
M2 and mixed M1-M2 are estimated from the data from the two-color staining based on the set
concept. The data are presented as the means ± SD and as an estimated range for the two-color and
three-color staining, respectively (n = 4-5/group). “-” means the two groups are difficult to compare
because of the overlapping percentage ranges. Mφ, macrophage; DAPI, 4′,6-diamidino-2-phenyl-
indole.

¶ 33277; ATCC, Manassas, VA.

# Skyscan 1172, Bruker, Germany.

** Sigma-Aldrich, St. Louis, MO.

†† CD68, ab31630; TNFα, ab6671; IL1β, ab9722; IL10, ab9969; Arg1, ab60176; Abcam, Cambridge, MA.

‡‡ F4/80, ab16911; NOS2, ab3523; CD206, ab64693; Abcam, Cambridge, MA.

§§ Alexa Fluor 488-conjugated; Life Technologies, Beijing, China.

ǁǁ Alexa Fluor 594-conjugated; ZSGB-BIO, Beijing, China.

¶¶ 113633, FITC-conjugated ; Jackson ImmunoResearch, West Grove, PA.

## LSM 710NLO; Zeiss, Jena, Germany.

*** Merck-Millipore, Victoria, Australia.

††† Gibco, Pleasanton, CA.

15
Journal of Periodontology; Copyright 2016 DOI: 10.1902/jop.2016.160081

‡‡‡ R&D Systems, Minneapolis, MN.

§§§ InvivoGen, San Diego, CA.

ǁǁǁ RNAiso plus/PrimeScript RT reagent kit/SYBR Premix Ex Taq PCR kit; Takara Bio, Otsu, Japan.

¶¶¶ ViiA 7; Applied Biosystems, Waltham, MA.

### PrimerBank; The Harvard University, Cambridge, MA

**** R&D Systems, Minneapolis, MN.

†††† V-Plex; Meso Scale Discovery, Gaithersburg, MD.

‡‡‡‡ SPSS 17.0; IBM, Armonk, NY.

16
Journal of Periodontology; Copyright 2016 DOI: 10.1902/jop.2016.160081

17
Journal of Periodontology; Copyright 2016 DOI: 10.1902/jop.2016.160081

18
Journal of Periodontology; Copyright 2016 DOI: 10.1902/jop.2016.160081

19
Journal of Periodontology; Copyright 2016 DOI: 10.1902/jop.2016.160081

20
Journal of Periodontology; Copyright 2016 DOI: 10.1902/jop.2016.160081

21