Porphyromonas gingivalis Promotes Oral Squamous Cell Carcinoma Progression in an Immune Microenvironment

Abstract

Increasing evidence has revealed a significant association between microorganisms and oral squamous cell carcinoma (OSCC). Porphyromonas gingivalis, the keystone pathogen in chronic periodontitis, is considered an important potential etiologic agent of OSCC, but the underlying immune mechanisms through which P. gingivalis mediates tumor progression of the oral cancer remain poorly understood. Our cohort study showed that the localization of P. gingivalis in tumor tissues was related to poor survival of patients with OSCC. Moreover, P. gingivalis infection increased oral lesion multiplicity and size and promoted tumor progression in a 4-nitroquinoline-1 oxide (4NQO)–induced carcinogenesis mouse model by invading the oral lesions. In addition, CD11b ⁺ myeloid cells and myeloid-derived suppressor cells (MDSCs) showed increased infiltration of oral lesions. Furthermore, in vitro observations showed that MDSCs accumulated when human-derived dysplastic oral keratinocytes (DOKs) were exposed to P. gingivalis, and CXCL2, CCL2, interleukin (IL)–6, and IL-8 may be potential candidate genes that facilitate the recruitment of MDSCs. Taken together, our findings suggest that P. gingivalis promotes tumor progression by generating a cancer-promoting microenvironment, indicating a close relationship among P. gingivalis, tumor progression of the oral cancer, and immune responses.

Full text

https://doi.org/10.1177/0022034520909312
Journal of Dental Research
1 –10
© International & American Associations
for Dental Research 2020
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/0022034520909312
journals.sagepub.com/home/jdr
Research Reports: Biological
Introduction
Oral squamous cell carcinoma (OSCC) is the most common
malignant cancer of the head and neck (Chi et al. 2015), which
accounted for approximately 180,000 deaths worldwide in
2018 (1.9% of total cancer cases), and is among the top 15
common cancers worldwide (Bray et al. 2018). OSCC is often
preceded by oral potentially malignant disorders (OPMDs),
but the mechanisms by which OPMDs progress into cancer
remains largely uncertain. Among the major risk factors for
oral cancers, alcohol consumption, areca nut (betel quid) chew-
ing, and human papillomavirus infection were extensively
studied (Mehanna et al. 2013; Winn et al. 2015; Mehrtash et al.
2017). However, around 15% cases of oral carcinogenesis
remain unexplained (Chocolatewala et al. 2010). Thus, it is
urgent to find new pathogenic factors so that we can prevent
the tumor progression of oral cancer.
An increasing and substantial number of microorganisms
have been confirmed to be strongly associated with human car-
cinogenesis, including Helicobacter pylori in gastric cancer
(Amieva and Peek 2016), Fusobacterium nucleatum in
colorectal cancer (Yu et al. 2017), and so on. Porphyromonas
gingivalis, a Gram-negative oral bacteria as “keystone patho-
gen” in periodontitis (Hajishengallis et al. 2012), shows strong
carcinogenic potential in gastrointestinal tract cancers, such as
colon cancer (Ahn et al. 2012), pancreatic cancer (Michaud
et al. 2013), esophageal cancer (Peters et al. 2017), and OSCC.
The relative abundance of P. gingivalis in gingival carcinoma
has been reported to be 33% higher than that in normal gingival
909312JDRXXX10.1177/0022034520909312Journal of Dental ResearchP. gingivalis Promotes OSCC Progression in an Immune Microenvironment
research-article2020
1Hospital of Stomatology, Guanghua School of Stomatology, Guangdong
Provincial Key Laboratory of Stomatology, Sun Yat-Sen University,
Guangzhou, Guangdong, China
2State Key Laboratory of Oncology in South China, MOE Key
Laboratory of Gene Function and Regulation, School of Life Sciences,
Sun Yat-sen University, Guangzhou, Guangdong, China
3Department of Microbiology, Zhongshan School of Medicine, Key
Laboratory for Tropical Diseases Control of the Ministry of Education,
Sun Yat-sen University, Guangzhou, Guangdong, China
4Discipline of Oral Bioscience, Sydney Dental School, Faculty of Medicine
and Health, The University of Sydney, Westmead, NSW, Australia
A supplemental appendix to this article is available online.
Corresponding Authors:
Z. Wang, Hospital of Stomatology, Guanghua School of Stomatology,
Guangdong Provincial Key Laboratory of Stomatology, Sun Yat-Sen
University, No. 56, Lingyuan West Road, Yuexiu District, Guangzhou,
Guangdong 510055, China.
Email: wangzh75@mail.sysu.edu.cn
B. Cheng, Hospital of Stomatology, Guanghua School of Stomatology,
Guangdong Provincial Key Laboratory of Stomatology, Sun Yat-Sen
University, No. 56, Lingyuan West Road, Yuexiu District, Guangzhou,
Guangdong 510055, China.
Email: chengbin@mail.sysu.edu.cn
Porphyromonas gingivalis Promotes Oral
Squamous Cell Carcinoma Progression
in an Immune Microenvironment
L. Wen1, W. Mu1, H. Lu1, X. Wang1, J. Fang1, Y. Jia1, Q. Li1, D. Wang1, S. Wen1,
J. Guo1, W. Dai1, X. Ren1, J. Cui2, G. Zeng3, J. Gao4, Z. Wang1, and B. Cheng1
Abstract
Increasing evidence has revealed a significant association between microorganisms and oral squamous cell carcinoma (OSCC).
Porphyromonas gingivalis, the keystone pathogen in chronic periodontitis, is considered an important potential etiologic agent of OSCC,
but the underlying immune mechanisms through which P. gingivalis mediates tumor progression of the oral cancer remain poorly
understood. Our cohort study showed that the localization of P. gingivalis in tumor tissues was related to poor survival of patients with
OSCC. Moreover, P. gingivalis infection increased oral lesion multiplicity and size and promoted tumor progression in a 4-nitroquinoline-1
oxide (4NQO)–induced carcinogenesis mouse model by invading the oral lesions. In addition, CD11b+ myeloid cells and myeloid-
derived suppressor cells (MDSCs) showed increased infiltration of oral lesions. Furthermore, in vitro observations showed that MDSCs
accumulated when human-derived dysplastic oral keratinocytes (DOKs) were exposed to P. gingivalis, and CXCL2, CCL2, interleukin
(IL)–6, and IL-8 may be potential candidate genes that facilitate the recruitment of MDSCs. Taken together, our findings suggest that
P. gingivalis promotes tumor progression by generating a cancer-promoting microenvironment, indicating a close relationship among
P. gingivalis, tumor progression of the oral cancer, and immune responses.
Keywords: microbiology, bacteria, 4-Nitroquinoline-1-oxide, precancerous conditions, mucosal immune response, retrospective
studies

2 Journal of Dental Research 00(0)
tissues, suggesting an association between P. gingivalis and
OSCC (Katz et al. 2011). More recently, a meta-analysis indi-
cated that the existence of P. gingivalis increased OSCC devel-
opment by 1.36 times (Sayehmiri et al. 2015). These findings
indicated that P. gingivalis might be a new candidate that pro-
motes the tumor progression of oral cancer. However, few
studies have explored the molecules and mechanisms underly-
ing these associations.
Importantly, while P. gingivalis infection is strongly corre-
lated with chronic periodontitis, the infection result in tissue
destruction attributes to host immune response but not bacteria
itself (Hajishengallis 2015). The interaction of the host immune
system with the oral bacteria in healthy states and in diseased
states has been widely described; nevertheless, the possible
interrelation of infection of P. gingivalis, immune microenvi-
ronment development, and oral cancer is rarely explored.
The current study investigated the mechanisms by which
P. gingivalis participates in tumor progression of oral cancer
and its association with the immune microenvironment.
Therefore, we first examined whether the existence of P. gingi-
valis was correlated with survival in patients with OSCC. To
delineate the role of P. gingivalis in oral carcinogenesis, we
determine whether P. gingivalis invades the oral precancerous
and cancer lesions of an established 4-nitroquinoline-1 oxide
(4NQO)–induced carcinogenesis mouse model. Furthermore,
our results show that myeloid cells and myeloid-derived sup-
pressor cells (MDSCs) were more infiltrated in the P. gingiva-
lis infection group, which might be associated with interleukin
(IL)–6, IL-8, CCL2, and CXCL2 expression. This study pro-
vides novel insights and experimental basis in vivo and in vitro
to support the role of P. gingivalis in the tumor progression of
oral cancer.
Materials and Methods
Collection of Clinical Samples
Fifty-six formalin-fixed, paraffin-embedded (FFPE) blocks of
OSCC tissues were collected from the Stomatological Hospital,
Sun Yat-Sen University (SYSU). The patients were pathologi-
cally with OSCC and over 18 y of age. Overall survival,
defined as the time from the first disease diagnosis to the date
of death or last contact, was designated as the end point.
Recurrence was monitored by clinical and pathological diag-
nosis. Informed consent was obtained from all patients. The
use of human samples for this study was approved by the ethics
committee of SYSU and was conducted according to the
STROBE (Strengthening the Reporting of Observational
Studies in Epidemiology) statement.
Bacterial Stains and Culturing
Strain P. gingivalis ATCC33277, P. gingivalis W83, and
Streptococcus mutans UA159 (S. mutans) were purchased
from ATCC. The bacterial were grown in Brain Heart Infusion
broth (BHI) supplemented with yeast extract at 5 mg/mL,
cysteine at 1 mg/mL, vitamin K1 at 0.5 µg/mL and hemin at 5
µg/mL under anaerobic conditions with oxygen concentration
<1% at 37°C (Bactrox-2; SHELLAB). Human-derived dys-
plastic oral keratinocytes (DOKs; Sigma) were grown and cul-
tured in high-glucose Dulbecco’s modified Eagle’s media
(DMEM; Gibco) supplemented with 10% fetal bovine serum
(FBS; Millipore Sigma) and 0.05% hydrocortisone (Millipore
Sigma) at 37°C in 5% CO2. DOK cells were starved in low-
glucose serum-free DMEM (Gibco) overnight prior to experi-
ments, followed by continued treatment in serum-free
low-glucose medium.
4NQO-Induced Oral Tumorigenesis
Model and Animal Experiments
Six-week-old wild-type C57BL/6 mice were purchased from
Guangzhou University of Chinese Medicine and were main-
tained in the SYSU in a specific pathogen-free facility at stan-
dard laboratory conditions. The chemical carcinogen 4NQO
(Sigma-Aldrich) was performed according to the published
methods (Chen et al. 2018). For the infected model, mice were
randomly divided into a control group (n = 6) and a P. gingiva-
lis group (n = 12). P. gingivalis ATCC33277 (1 × 108 colony-
forming units [CFU] per 100 µL) suspended in 2%
carboxymethylcellulose was orally inoculated to mice every
other day for a period of 10 wk. For the control group, the mice
were treated with BHI. Mice were euthanized at week 26, and
tongue, spleen, and draining lymph node were collected. The
gross examination of tongue lesions was performed according
to a previous study (Wu et al. 2018). All experimental handing
was conducted in compliance with the ARRIVE (Animal
Research: Reporting of In Vivo Experiments) guidelines. The
study procedure was approved by the Animal Care and Use
Committee of SYSU.
Detection of Bacteria by Immunohistochemistry
and Fluorescence In Situ Hybridization
Immunohistochemistry (IHC) and fluorescence in situ hybrid-
ization (FISH) were employed to detect the colonization of
P. gingivalis. Consecutive 4-µm-thick paraffin sections were
cut from each block, 2 for human samples and 3 for mice. The
first section was used as a negative control, and the second and
third sections were used for the detection of P. gingivalis.
Antigen retrieval was performed under high-pressure heating
with EDTA buffer (ZLI-9067, ZSGB-BIO, pH 8.0). A mono-
clonal mouse anti-RgpB antibody (gift from the laboratory of
Jinlong Gao, University of Sydney; 1:100) was employed as
primary antibody for IHC. FISH was carried out with appropri-
ate specific probes according to the manufacturer’s instruc-
tions. FFPE tissue sections were probed with 5 mg/mL
P. gingivalis 16S ribosomal RNA-specific oligonucleotide
POGI (CAATACTCGTATCGCCCGTTATTC), 5′-TGCAC
AAGGCACAACGCAACAGGGCA-3′ labeled with Cy3
(Takara, Japan) dye. FISH was performed according to the
published methods (Romero-Lastra et al. 2019).

P. gingivalis Promotes OSCC Progression in an Immune Microenvironment 3
Tissue Preparation, Histology,
and Immunostaining
Harvested oral lesions for hematoxylin and eosin (H&E) stain-
ing and IHC were stained as previously described (Wen et al.
2019). Samples for immunofluorescence (IF) were embedded
in OCT compound (Sakura Tissue-Tek) and sectioned into
8-µm sections. IHC was then carried out with anti-Ki67 rabbit
polyclonal antibody (ab15580, Abcam; 1:500) and anti-Gr-1 rab-
bit polyclonal antibody (GB11229, Servicebio; 1:500). IF was
performed with anti-CD11b rabbit monoclonal antibody
(ab133357, Abcam; 1:200).
Flow Cytometry
A single cell suspension was prepared from tongue, spleen, and
draining lymph nodes of the mice. Tongues were dissected,
minced, and resuspended in complete media (RPMI 1640 with
10% FBS) supplemented with Collagenase-IV (17104019;
Gibco) at 1 mg/mL and DNase I (1121MG010; BioFroxx) at
0.1 mg/mL and incubated at 37°C for 30 min with shaking to
form a single cell. Tissue were passed through a 70-µm strainer,
collected, and washed with phosphate-buffered saline (PBS)
with 1% FBS. Spleen and draining lymph nodes were dissected
and minced to form a single cell system. Cells were processed
to live/dead cell discrimination using Fixable Viability Dye
(423105; BioLegend) and were stained for 30 min at 4°C with
the following antibodies (all from BioLegend): CD45 (103112),
CD3 (100203), CD4 (100528), CD8 (100726), CD11b
(101210), and Gr-1 (108408). All flow cytometry data acquisi-
tion was done using CytExpert (2.1.0.92; Beckman Coulter)
and analyzed using FlowJo software version 10 (TreeStar).
RNA Extraction and Quantitative Real-Time
Polymerase Chain Reaction
Total RNA was isolated from cells with TRIzol reagent
(Invitrogen) in accordance with the manufacturer’s instruc-
tions, and 500 ng total RNA was reverse transcribed using the
PrimeScript RT Reagent Kit (Perfect Real Time; Takara) for
quantitative real-time polymerase chain reaction (qPCR).
qPCR was performed in triplicate with an Applied LightCycler
96 quantitative PCR system (Roche). The Ct values obtained
from different samples were compared using the 2–DDCt method.
β-Actin served as an internal reference gene. Primers used in
this study are shown in Appendix Table 1.
MDSC Isolation and Chemotaxis Assays
MDSCs were isolated from mice bone marrow. Bone marrow
was dissected, minced, and resuspended in complete media
(DMEM with 10% FBS). Then, the bone marrow cells were
passed through a 70-µm strainer, collected, and washed with
PBS with 1% FBS. After staining with CD11b (101210;
BioLegend) and Gr-1 (108408; BioLegend), MDSCs were iso-
lated by BD FACSAria Fusion.
Chemotaxis assays were performed using 24-well plates
with 5-µm pore size inserts (3421; Corning) according to the
manufacturer’s instructions. Then, 2.5 × 105 DOK cells with or
without coculture with P. gingivalis 33277 (multiplicity of
infection [MOI] = 1:100, 24 h) in complete medium (DMEM +
10% FBS) were added to the lower chamber. Also, a total of 1
× 106 MDSCs in serum-free medium were loaded into the
upper chamber. After 24 h of incubation, migrated cells were
counted as previously described (Zhang et al. 2016). These
experiments were repeated 3 times.
Statistical Analysis
Kaplan-Meier survival curves and log-rank tests were per-
formed for overall survival and recurrence-free survival.
Imaging was reviewed by 2 certified pathologists. The positive
cells were counted under 400× magnification using ImageJ
software (National Institutes of Health), and 5 randomly
selected independent microscopic fields were counted for each
sample to ensure that the data were representative and homo-
geneous. Differences in quantitative data between groups were
assessed using an unpaired Student’s t test. Comparisons of
means among multiple groups were performed by 1-way anal-
ysis of variance (ANOVA) tests. Measurements are expressed
as the mean ± standard deviation (SD). All statistical analyses
and graphs were generated with GraphPad Prism version 7.0
(GraphPad Software). P values less than 0.05 were considered
statistically significant (*P < 0.05, **P < 0.01, ***P < 0.001,
****P < 0.0001).
Results
Existence of P. gingivalis Is Associated with Oral
Squamous Cell Carcinoma Patient Mortality
To examine whether P. gingivalis is related to the prognosis of
patients with OSCC, we measured the relative abundance of
P. gingivalis in 56 OSCC FFPE samples using IHC staining.
According to the staining intensity, patients were classified as
P. gingivalis negative (0%, n = 22) or P. gingivalis positive
(>0%, n = 34) (Fig. 1A). Patients diagnosed with OSCC were
reviewed by 2 certified pathologists. Among the 56 OSCC
cases, 19 deaths occurred, including 15 deaths of P. gingivalis
positive patients, during a median patient follow-up period of
5.25 y (interquartile range [IQR], 1.16–5.71). We performed
univariate regression analyses and multivariate regression
analyses of baseline information in this cohort, including
P. gingivalis infection, age, location, T stage, lymphatic metas-
tasis, smoking, and alcoholic drinking (Appendix Tables 2 and
3). In our cohort study, colonization with P. gingivalis was sig-
nificantly associated with shorter overall survival (OS; hazard
ratio [HR] = 2.928; 95% confidence interval [CI], 1.183–7.247;
P < 0.05 by log-rank test; Fig. 1B, left). We also observed

4 Journal of Dental Research 00(0)
nonsignificantly higher abundance of P. gingivalis associated
with recurrence-free survival (RFS; HR = 1.492; 95% CI,
0.4775–4.695; P > 0.05 by log-rank test; Fig. 1B, right).
Consequently, our results showed that infection with P. gingi-
valis was associated with mortality in the patients with OSCC.
P. gingivalis Promotes Tumor Progression of
Oral Mucosa Epithelium in 4NQO-Induced Mice
P. gingivalis infection was related to a poor outcome in patients
with OSCC, which prompted us to investigate whether chronic
P. gingivalis infection could promote the progression of oral
squamous cell carcinoma. In our preliminary experiment, we
administered P.g ATCC33277 (P. gingivalis group),
Streptococcus mutans UA159 (S. mutans group), or BHI broth
(control group) to 4NQO-inducted mice by oral gavage (Appendix
Fig. 1). S. mutans was used as a negative control since
S. mutans has not been shown to have an obvious effect on oral
tumor progression (Gholizadeh et al. 2016; Orlandi et al. 2019).
A diagram of our experimental protocol is shown in Figure
2A. As shown in Figure 2B, 10 wk after P. gingivalis adminis-
tration, lesions in 4NQO-induced mice progressed to white
masses with a cauliflower-like or verrucous appearance, while
the control group still exhibited a wrinkled paper-like lesion
appearance. Notably, we found that the P. gingivalis group
showed a significant increase in lesion number (2.50 ± 1.22 vs.
1.17 ± 0.41, P < 0.05) and lesion diameter (2.61 ± 2.40 mm vs.
1.20 ± 0.29 mm, P < 0.05) (Fig. 2C). In addition, during the
course of administration, we observed that the P. gingivalis
group had significant weight loss compared to
the control group (Fig. 2D).
Then, H&E staining showed that mice with
or without introduction of P. gingivalis exhib-
ited different stages of oral carcinogenesis (Fig.
2E). A 2-category system was used to analyze
the oral cancer risk (normal/hyperplasia/mild
or moderate dysplasia: low risk; severe dyspla-
sia/invasive cancer: high risk) of mice accord-
ing to a previous study (Wu et al. 2018). In the
P. gingivalis group, 9 of 12 of mice showed a
high risk of oral carcinogenesis, compared to 1
of 6 in the control group (Appendix Table 4).
Moreover, P. gingivalis–treated mice showed
increased basal proliferation of hyperplastic
regions in the tongue (Fig. 2F). These data sug-
gest that P. gingivalis infection of oral mucosa
can promote oral carcinogenesis.
P. gingivalis Promotes Tumor
Progression by Colonizing Oral Tissue
A previous study reported that P. gingivalis is a
type of intracellular colonizing bacteria that
invades, replicates, and survives in human pri-
mary gingival epithelial cells. Therefore, we
hypothesized that P. gingivalis could accelerate
tumorigenesis by invading and colonizing oral tissues in mice.
To test our hypothesis, we detected the existence of P. gingiva-
lis by IHC and FISH. As shown in Figure 3A and 3B, P. gingi-
valis was always found in the P. gingivalis group (12/12) and
was not found in the control group (0/6) (P < 0.0001). In addi-
tion, we observed P. gingivalis was present not only in the epi-
thelial layer in mouse tissue (Fig. 3A, red arrows) but also in
the lamina propria (Fig. 3A, blue arrows), which was also
shown in human OSCC samples (Fig. 1A, red and blue arrows).
P. gingivalis Infection Expands CD11b+ Myeloid
Cells and MDSCs in Local Lesions
Immune cells and their effectors are key components of tumors
and promote neoplastic progression. To determine whether
P. gingivalis contributes to tumorigenesis by affecting intratu-
moral immune cells, we characterized infiltrating immune
cells from lesions, the spleen, and draining lymph nodes of the
P. gingivalis group and control group. Appendix Fig. 2A shows
the gating strategy for analysis of immune cells in oral lesions.
Interestingly, we observed an increase in infiltrating CD11b+
myeloid cells (P < 0.001) and MDSCs (P < 0.01) in the lesions
of the P. gingivalis group (Fig. 4A), while the populations of
CD3+, CD4+, and CD8+ T lymphocytes were not significantly
different (Fig. 4B). The populations of CD3+, CD4+, and CD8+
T lymphocytes; CD11b+ myeloid cells; and MDSCs in spleen
and draining lymph nodes showed no differences between the
2 groups (Appendix Fig. 2B, C). To further confirm the changes
in the immune microenvironment induced by P. gingivalis, we
Figure 1. Intratumoral Porphyromonas gingivalis infection is associated with
outcome in patients with oral squamous cell carcinoma (OSCC). (A) Representative
immunohistochemistry-stained paraffin sections of P. gingivalis–negative and P. gingivalis–
positive group using monoclonal mouse anti-RapB antibody in human OSCC samples (red
arrows, P. gingivalis infection at the epithelial layer; blue arrows, P. gingivalis infection at the
lamina propria). Scale bar = 20 μm. (B) Overall survival (left) and recurrence-free survival
(right) were compared between the patients with or without P. gingivalis infection.

P. gingivalis Promotes OSCC Progression in an Immune Microenvironment 5
Figure 2. Porphyromonas gingivalis infection promoted oral carcinogenesis in 4-nitroquinoline-1 oxide (4NQO)–induced mice. (A) Diagram of the
experimental protocol. (B) Representative images of tongue tissues of control group (n = 6) and P. gingivalis group (n = 12). (C) The average oral lesion
numbers and average lesion size per mice in the control group and P. gingivalis group. (D) The body weights of mice in the control and P. gingivalis
group. The data are represented as mean ± SEM. (E) Representative hematoxylin and eosin sections of pathology, including mild dysplasia, moderate
dysplasia, severe dysplasia, and invasive carcinoma. Scale bar = 50 μm. (F) Representative immunostaining for Ki67 in oral lesions. The percentage of
proliferating cells was evaluated in severe dysplasia from the control and P. gingivalis groups using a Ki67 rabbit polyclonal antibody. Scale bar = 20 μm.
*P < 0.05. **P < 0.01. ***P < 0.001.

6 Journal of Dental Research 00(0)
Figure 3. Porphyromonas gingivalis could accelerate tumorigenesis by invading and colonizing oral tissues in mice. (A) Representative
immunohistochemistry images and analysis of paraffin tissue sections in oral lesions from the control group and P. gingivalis group (yellow-brown,
P. gingivalis; red arrows, P. gingivalis infection in epithelial layer; blue arrows, infection in lamina propria). Scale bars = 20 μm. (B) Representative
fluorescence in situ hybridization images and analysis of paraffin tissue sections of 2 groups (red, P. gingivalis). Scale bars = 20 μm. ****P < 0.0001.

P. gingivalis Promotes OSCC Progression in an Immune Microenvironment 7
Figure 4. Porphyromonas gingivalis infection selectively recruits CD11b+ myeloid cells and myeloid-derived suppressor cells (MDSCs) in oral lesions of
4-nitroquinoline-1 oxide (4NQO)–induced mice. (A) The proportion of CD11b+ myeloid cells and MDSCs in oral lesions from the control and
P. gingivalis groups. (B) The proportion of CD3+, CD4+, and CD8+ T cells in oral lesions from the control and P. gingivalis groups. (C) Representative
immunofluorescence images and analysis of CD11b expression in oral lesions between the 2 groups. And representative immunohistochemistry
staining images and analysis of Gr-1 expression in oral lesions between the 2 groups. Scale bars = 20 μm. *P < 0.05. **P < 0.01. ***P < 0.001.

8 Journal of Dental Research 00(0)
subjected FFPE samples from the P. gingivalis group and con-
trol group to IF (with an anti-CD11b antibody) and IHC stain-
ing (with an anti-Gr-1 antibody). As shown in Figure 4C,
infiltration of CD11b+ myeloid cells and MDSCs was increased
in the P. gingivalis group, which was consistent with the pri-
mary results. We also found that in the clinical OSCC samples,
the presence of P. gingivalis was associated with more MDSC
influx than those tumors free of P. gingivalis (P < 0.05)
(Appendix Fig. 3).
P. gingivalis Infection Is Associated
with the Recruitment of MDSCs In Vitro
MDSCs are immature myeloid cells that have potent immuno-
suppressive activity. Given our findings of P. gingivalis–
induced MDSC expansion in mouse oral precancerous lesions,
we investigated the potential mechanisms. First, to explore the
relationship between P. gingivalis infection and MDSC recruit-
ment, we established a coculture system with Transwell mem-
branes (5 µm) in vitro. MDSCs were added to medium without
FBS in the upper chamber, and P. gingivalis, DOK cells, or
P. gingivalis–infected DOK cells (MOI = 1:100, 24 h) in 10%
FBS were added to the lower chamber (Fig. 5A). After 24 h of
incubation, the number of migrating MDSCs was significantly
elevated following coculture with P. gingivalis–infected DOK
cells compared to those cultured only with P. gingivalis or
DOK cells (Fig. 5B). In addition, we extracted RNA from
DOK cells infected with P. gingivalis for 24 h (MOI = 1:100)
and analyzed the expression levels of CCL2, CXCL2, CXCL5,
CCL5, IL-6, and IL-8, which have been described as potential
candidate genes for oral cancer in response to chronic infection
with P. gingivalis (Geng et al. 2019) and the recruitment and
expansion of MDSCs (Zhang et al. 2016). DOK cells without
P. gingivalis infection were cultured at the same time as con-
trol. DOK cells exposed to P. gingivalis showed significantly
increased expression of CCL2 (P < 0.05),
CXCL2 (P < 0.01), IL-6 (P < 0.01), and
IL-8 (P < 0.0001) compared to uninfected
DOK cells (Fig. 5C).
Discussion
P. gingivalis is regarded as a keystone
pathogen of periodontitis because of its
ability to disrupt the host immune
response, and it has been reported to be
associated with OSCC progression.
Therefore, we investigated whether it does
play an important role in OSCC progres-
sion. First, our cohort study showed that
the colonization of tumors by P. gingivalis
was negatively correlated with overall
survival in patients with OSCC. Recently,
many in vitro studies have provided
insights into the effects of P. gingivalis in
promoting oral cancer progression, including activation of cell
proliferation, inhibition of apoptosis, and promotion of cellular
invasion (Perera et al. 2016). However, the potential effect of P.
gingivalis on the immune microenvironment during tumor pro-
gression of oral cancer is poorly understood. In the present
study, we found that P. gingivalis invades oral precancerous
lesions and recruits the myeloid-derived suppressor cells by
expressing chemokines such as CCL2 and CXCL2 and cyto-
kines such as IL-6 and IL-8.
Entry of microbes and/or microbial metabolites into the
tumor microenvironment promotes neoplastic progression by
eliciting tumor-potentiating immune cell responses (Jobin
2012). Studies in mice using micro-osmotic pumps implanted
into the lumbodorsal region to deliver P. gingivalis showed that
the percentage of MDSCs was increased in the bone marrow
and spleen (Su et al. 2017). Interestingly, we observed a sig-
nificant increase in infiltrating CD11b+ myeloid cells and
MDSCs in local sites of oral lesions resulting from infection of
P. gingivalis in mice exposed to 4NQO. Moreover, P. gingiva-
lis presence was associated with more MDSC influx than those
tumors free of P. gingivalis in the OSCC samples. An in vitro
study showed that P. gingivalis can upregulate the MDSC
recruitment-related genes and recruit MDSCs. The main fea-
ture of MDSCs in the tumor environment is their potent immu-
nosuppressive activity, which is often related to poor patient
survival (Kumar et al. 2016). This indicates that during the
process of oral tumorigenesis, myeloid cells, especially
MDSCs, create an immunosuppressive microenvironment that
favors tumor progression.
It is well known that tumor-produced chemokines or cyto-
kines cause MDSCs to be trafficked through the circulatory
system or migrate into solid tumors. Among cytokines, IL-6
and IL-8 have been identified as major cytokines involved in
P. gingivalis infection (Liu et al. 2014; Yee et al. 2014). IL-6
has been reported as a promising predictor for the early diagno-
sis of tongue squamous cell carcinoma (Hussein et al. 2018). In
Figure 5. Porphyromonas gingivalis infection promoted myeloid-derived suppressor cell (MDSC)
migration in vitro, and it was associated with interleukin (IL)–6, IL-8, CCL2, and CXCL2
expression. (A) Schematic diagram of MDSC chemotaxis assays. MDSCs were seeded into the
upper chamber of a Transwell system, and P. gingivalis and dysplastic oral keratinocyte (DOK) cells
with or without P. gingivalis infection were added to the lower chamber. (B) The analysis of MDSC
chemotaxis assays. (C) The relative expression level of CCL2, CXCL2, CXCL5, CCL5, IL-6, and
IL-8 between uninfected DOK cells and DOK cells infected with P. gingivalis. *P < 0.05. **P < 0.01.
***P < 0.001. ****P < 0.0001.

P. gingivalis Promotes OSCC Progression in an Immune Microenvironment 9
addition, IL-6 has been reported as a candidate to illustrate the
role of P. gingivalis infection in promoting OSCC initiation
and progression (Geng et al. 2019). Furthermore, IL-6 and
IL-1β play an important role in driving both the accumulation
and suppressive potency of murine MDSCs (Bunt et al. 2007).
IL-8, which is frequently secreted in the tumor microenviron-
ment, promotes tumor progression through chemotaxis of
MDSCs (Alfaro et al. 2017). Whether P. gingivalis may have a
direct and an indirect impact on the production of IL-8 in our
study remains unknown. P. gingivalis expresses a variety of
virulence factors that play different roles in subverting the host
immune response (Zenobia and Hajishengallis 2015). Different
virulence factors of P. gingivalis may exert contrasting influ-
ence on the production of IL-8 through various mechanisms
(Isberg et al. 2013; Zhang and Li 2015), and our study shows
that P. gingivalis can promote IL-8 production of DOKs in
vitro. Therefore, whether P. gingivalis may have a direct and
indirect impact on the production of IL-8 may depend on which
virulence factor play a leading role in our model. Regarding
chemokines, CCL2 has been reported to promote colorectal
carcinogenesis by enhancing the population and function of
polymorphonuclear MDSCs (Chun et al. 2015). Another che-
mokine, CXCL2, promotes the generation of monocytic
MDSCs (Shi et al. 2018). Furthermore, CXCL5 and CCL5 are
also associated with tumor progression (Ban et al. 2017; Najjar
et al. 2017). Our results showed that CCL2, CXCL2, IL-6, and
IL-8 were significantly increased in DOK cells after exposure
to P. gingivalis in vitro, indicating that these chemokines and
cytokines are involved in the recruitment of MDSCs, which
contributes to tumor progression of oral cancer.
To our knowledge, this is the first report that illustrates that
P. gingivalis mediates oral carcinogenesis by recruiting
MDSCs, and we realize that our current study has some limita-
tions. The mechanism by which P. gingivalis affects the func-
tion of MDSCs in the microenvironment needs to be explored.
In conclusion, P. gingivalis promotes tumor progression by
recruiting MDSCs via increasing secretion of IL-6, IL-8,
CCL2, and CXCL2 from infected oral dysplastic keratino-
cytes. These findings suggest that P. gingivalis is another
potential target for the management of tumor progression of
oral cancer.
Author Contributions
L. Wen, W. Mu, H. Lu, contributed to design, data acquisition, and
analysis, drafted and critically revised the manuscript; X. Wang, J.
Fang, Y. Jia, contributed to data analysis and interpretation, criti-
cally revised the manuscript; Q. Li, D. Wang, S. Wen, J. Guo, W.
Dai, contributed to data acquisition and analysis, critically revised
the manuscript; X. Ren, J. Cui, G. Zeng, contributed to data inter-
pretation, critically revised the manuscript; J. Gao, contributed to
conception and design, critically revised the manuscript; Z. Wang,
B. Cheng, contributed to conception and design, drafted and criti-
cally revised the manuscript. All authors gave final approval and
agree to be accountable for all aspects of the work.
Acknowledgments
This project was supported by grants from the National Natural
Science Foundation of China (No. 81772896, 81630025, and
81972532), the Science and Technology Planning Project of
Guangzhou City of China (No. 2017004020102), the Science and
Technology Program of Guangzhou city of China (No. 20180
4010144), and the China Postdoctoral Science Foundation (No.
2019TQ0388). The authors declare no potential conflicts of inter-
est with respect to the authorship and/or publication of this
article.
References
Ahn J, Segers S, Hayes RB. 2012. Periodontal disease, Porphyromonas
gingivalis serum antibody levels and orodigestive cancer mortality.
Carcinogenesis. 33(5):1055–1058.
Alfaro C, Sanmamed MF, Rodríguez-Ruiz ME, Teijeira Á, Oñate C, González
Á, Ponz M, Schalper KA, Pérez-Gracia JL, Melero I. 2017. Interleukin-8 in
cancer pathogenesis, treatment and follow-up. Cancer Treat Rev. 60:24–31.
Amieva M, Peek RM Jr. 2016. Pathobiology of helicobacter pylori-induced
gastric cancer. Gastroenterology. 150(1):64–78.
Ban Y, Mai J, Li X, Mitchell-Flack M, Zhang T, Zhang L, Chouchane L, Ferrari
M, Shen H, Ma X. 2017. Targeting autocrine CCLl5-CCR5 axis reprograms
immunosuppressive myeloid cells and reinvigorates antitumor immunity.
Cancer Res. 77(11):2857–2868.
Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. 2018. Global
cancer statistics 2018: Globocan estimates of incidence and mortality world-
wide for 36 cancers in 185 countries. CA Cancer J Clin. 68(6):394–424.
Bunt SK, Yang L, Sinha P, Clements VK, Leips J, Ostrand-Rosenberg S. 2007.
Reduced inflammation in the tumor microenvironment delays the accumu-
lation of myeloid-derived suppressor cells and limits tumor progression.
Cancer Res. 67(20):10019–10026.
Chen Y, Li Q, Li X, Ma D, Fang J, Luo L, Liu X, Wang X, Lui VWY, Xia J,
et al. 2018. Blockade of PD-1 effectively inhibits in vivo malignant trans-
formation of oral mucosa. Oncoimmunology. 7(2):e1388484.
Chi AC, Day TA, Neville BW. 2015. Oral cavity and oropharyngeal squamous
cell carcinoma—an update. CA Cancer J Clin. 65(5):401–421.
Chocolatewala N, Chaturvedi P, Desale R. 2010. The role of bacteria in oral
cancer. Indian J Med Paediatr Oncol. 31(4):126–131.
Chun E, Lavoie S, Michaud M, Gallini CA, Kim J, Soucy G, Odze R, Glickman
JN, Garrett WS. 2015. CCL2 promotes colorectal carcinogenesis by
enhancing polymorphonuclear myeloid-derived suppressor cell population
and function. Cell Rep. 12(2):244–257.
Geng F, Wang Q, Li C, Liu J, Zhang D, Zhang S, Pan Y. 2019. Identification of
potential candidate genes of oral cancer in response to chronic infection with
Porphyromonas gingivalis using bioinformatical analyses. Front Oncol. 9:91.
Gholizadeh P, Eslami H, Yousefi M, Asgharzadeh M, Aghazadeh M, Kafil
HS. 2016. Role of oral microbiome on oral cancers, a review. Biomed
Pharmacother. 84:552–558.
Hajishengallis G. 2015. Periodontitis: from microbial immune subversion to
systemic inflammation. Nat Rev Immunol. 15(1):30–44.
Hajishengallis G, Darveau RP, Curtis MA. 2012. The keystone-pathogen
hypothesis. Nat Rev Microbiol. 10(10):717–725.
Hussein AA, Forouzanfar T, Bloemena E, de Visscher J, Brakenhoff RH,
Leemans CR, Helder MN. 2018. A review of the most promising biomark-
ers for early diagnosis and prognosis prediction of tongue squamous cell
carcinoma. Br J Cancer. 119(6):724–736.
Isberg RR, Takeuchi H, Hirano T, Whitmore SE, Morisaki I, Amano A, Lamont
RJ. 2013. The serine phosphatase SerB of Porphyromonas gingivalis sup-
presses IL-8 production by dephosphorylation of NF-κB RelA/p65. PLoS
Pathog. 9(4):e1003326.
Jobin C. 2012. Colorectal cancer: CRC—all about microbial products and bar-
rier function? Nat Rev Gastroenterol Hepatol. 9(12):694–696.
Katz J, Onate MD, Pauley KM, Bhattacharyya I, Cha S. 2011. Presence of
Porphyromonas gingivalis in gingival squamous cell carcinoma. Int J Oral
Sci. 3(4):209–215.
Kumar V, Patel S, Tcyganov E, Gabrilovich DI. 2016. The nature of myeloid-
derived suppressor cells in the tumor microenvironment. Trends Immunol.
37(3):208–220.

10 Journal of Dental Research 00(0)
Liu J, Wang Y, Ouyang X. 2014. Beyond Toll-like receptors: Porphyromonas
gingivalis induces IL-6, IL-8, and VCAM-1 expression through NOD-
mediated NF-κB and ERK signaling pathways in periodontal fibroblasts.
Inflammation. 37(2):522–533.
Mehanna H, Beech T, Nicholson T, El-Hariry I, McConkey C, Paleri V, Roberts
S. 2013. Prevalence of human papillomavirus in oropharyngeal and nonoro-
pharyngeal head and neck cancer—systematic review and meta-analysis of
trends by time and region. Head Neck. 35(5):747–755.
Mehrtash H, Duncan K, Parascandola M, David A, Gritz ER, Gupta PC,
Mehrotra R, Amer Nordin AS, Pearlman PC, Warnakulasuriya S, et al.
2017. Defining a global research and policy agenda for betel quid and areca
nut. Lancet Oncol. 18(12):e767–e775.
Michaud DS, Izard J, Wilhelm-Benartzi CS, You DH, Grote VA, Tjonneland A,
Dahm CC, Overvad K, Jenab M, Fedirko V, et al. 2013. Plasma antibodies
to oral bacteria and risk of pancreatic cancer in a large European prospec-
tive cohort study. Gut. 62(12):1764–1770.
Najjar YG, Rayman P, Jia X, Pavicic PG, Jr., Rini BI, Tannenbaum C, Ko J,
Haywood S, Cohen P, Hamilton T, et al. 2017. Myeloid-derived suppres-
sor cell subset accumulation in renal cell carcinoma parenchyma is associ-
ated with intratumoral expression of IL1β, IL8, CXCl5, and Mip-1α. Clin
Cancer Res. 23(9):2346–2355.
Orlandi E, Iacovelli NA, Tombolini V, Rancati T, Polimeni A, De Cecco L,
Valdagni R, De Felice F. 2019. Potential role of microbiome in oncogen-
esis, outcome prediction and therapeutic targeting for head and neck cancer.
Oral Oncol. 99:104453.
Perera M, Al-Hebshi NN, Speicher DJ, Perera I, Johnson NW. 2016. Emerging
role of bacteria in oral carcinogenesis: a review with special reference to
perio-pathogenic bacteria. J Oral Microbiol. 8:32762.
Peters BA, Wu J, Pei Z, Yang L, Purdue MP, Freedman ND, Jacobs EJ, Gapstur
SM, Hayes RB, Ahn J. 2017. Oral microbiome composition reflects pro-
spective risk for esophageal cancers. Cancer Res. 77(23):6777–6787.
Romero-Lastra P, Sanchez MC, Llama-Palacios A, Figuero E, Herrera D, Sanz
M. 2019. Gene expression of Porphyromonas gingivalis ATCC 33277 when
growing in an in vitro multispecies biofilm. PLoS One. 14(8):e0221234.
Sayehmiri F, Sayehmiri K, Asadollahi K, Soroush S, Bogdanovic L, Jalilian FA,
Emaneini M, Taherikalani M. 2015. The prevalence rate of Porphyromonas
gingivalis and its association with cancer: a systematic review and meta-
analysis. Int J Immunopathol Pharmacol. 28(2):160–167.
Shi H, Han X, Sun Y, Shang C, Wei M, Ba X, Zeng X. 2018. Chemokine
(C-X-C motif) ligand 1 and CXCL2 produced by tumor promote the
generation of monocytic myeloid-derived suppressor cells. Cancer Sci.
109(12):3826–3839.
Su L, Xu Q, Zhang P, Michalek SM, Katz J. 2017. Phenotype and function
of myeloid-derived suppressor cells induced by Porphyromonas gingivalis
infection. Infect Immun. 85(8). pii: e00213-17.
Wen L, Lu H, Li Q, Li Q, Wen S, Wang D, Wang X, Fang J, Cui J, Cheng B,
et al. 2019. Contributions of T cell dysfunction to the resistance against
anti-PD-1 therapy in oral carcinogenesis. J Exp Clin Cancer Res. 38(1):299.
Winn DM, Lee YC, Hashibe M, Boffetta P; INHANCE Consortium. 2015. The
INHANCE consortium: toward a better understanding of the causes and
mechanisms of head and neck cancer. Oral Dis. 21(6):685–693.
Wu JS, Zheng M, Zhang M, Pang X, Li L, Wang SS, Yang X, Wu JB, Tang
YJ, Tang YL, et al. 2018. Porphyromonas gingivalis promotes 4-nitroquin-
oline-1-oxide-induced oral carcinogenesis with an alteration of fatty acid
metabolism. Front Microbiol. 9:2081.
Yee M, Kim S, Sethi P, Duzgunes N, Konopka K. 2014. Porphyromonas gingi-
valis stimulates IL-6 and IL-8 secretion in GMSM-K, HSC-3 and H413 oral
epithelial cells. Anaerobe. 28:62–67.
Yu T, Guo F, Yu Y, Sun T, Ma D, Han J, Qian Y, Kryczek I, Sun D, Nagarsheth
N, et al. 2017. Fusobacterium nucleatum promotes chemoresistance to
colorectal cancer by modulating autophagy. Cell. 170(3):548–563.e16.
Zenobia C, Hajishengallis G. 2015. Porphyromonas gingivalis virulence factors
involved in subversion of leukocytes and microbial dysbiosis. Virulence.
6(3):236–243.
Zhang H, Ye YL, Li MX, Ye SB, Huang WR, Cai TT, He J, Peng JY, Duan TH,
Cui J, et al. 2016. CXCLl2/MIF-CXCR2 signaling promotes the recruit-
ment of myeloid-derived suppressor cells and is correlated with prognosis
in bladder cancer. Oncogene. 36(15):2095–2104.
Zhang Y, Li X. 2015. Lipopolysaccharide-regulated production of bone sia-
loprotein and interleukin-8 in human periodontal ligament fibroblasts: the
role of toll-like receptors 2 and 4 and the MAPK pathway. J Periodontal
Res. 50(2):141–151.