Periodontitis promotes gingival accumulation of cells with MDSC phenotype

Abstract

Background

Periodontitis is one of the most common inflammatory diseases in humans, mostly caused by bacterial infection and with diverse populations of immune cells involved. Myeloid-derived suppressor cells (MDSCs), a heterogeneous group of immature myeloid cells derived from hematopoietic precursor cells, have exhibited immunomodulatory functions by production of different molecules such as inducible NO synthase (iNOS) and it is thought to be involved in periodontitis. However, reports of characterization of cells with MDSC phenotypes in gingival tissues are very scarce. This study aimed to characterize gingival cells with MDSC phenotypes in healthy gingiva and periodontitis tissues.

Methods and Results

Human healthy gingival tissues and those with periodontitis were included to analyze cells with MDSC phenotypes by flow cytometry. Additionally, a mouse model of experimental periodontitis was used to identify cells with MDSC phenotypes and production of iNOS. Results showed an increased accumulation of CD45⁺HLA-DR^neg/lowCD11b⁺CD33⁺ cells in human gingival tissues with periodontitis. Experimental periodontitis promotes accumulation of CD45⁺CD11b⁺Gr-1⁺ and CD45⁺CD11b⁺Ly6G⁺ cells in gingival tissues. Experimental periodontitis did not promote accumulation of these subpopulations in other tissues as spleen. Additionally, gingival CD45⁺Gr-1⁺ iNOS⁺ cells were identified.

Conclusions

cells with MDSC phenotypes are resident in healthy gingival tissues and their accumulation is locally triggered by periodontitis. Cells with capacity of iNOS production could be implicated in generation of reactive nitrogen species, suggesting immunomodulatory properties.

Introduction

Periodontitis is an inflammatory disease characterized by soft tissue damage and, progressive destruction of the periodontal ligament and alveolar bone, which can finally lead to tooth loss, altering the quality of life of patients. ¹ In its severe form, this disease is the sixth most prevalent condition in the world, affecting around 10% of the adult population. ² Damage of the periodontal tissues is associated to inflammatory and immune responses promoted by accumulation of oral biofilms and alterations in the composition of oral microbial communities. An intense accumulation of immune cells such as neutrophils, macrophages, lymphocytes, plasma cells, and mast cells is generated in periodontitis.^1,3 In the absence of an effective and appropriate resolution of these processes and persistence of the stimulus, the immune response can promote activity of osteoclasts resulting in alveolar bone resorption.^1,3

Myeloid-derived suppressor cells (MDSCs) represent a highly heterogeneous population of myeloid progenitor cells and immature myeloid cells originated from hematopoietic precursors with immunoregulatory properties, classically with immunosuppressive effects on effector T cells. ⁴ Typically, mouse MDSCs have been characterized with the markers CD11b and Gr-1, and can be divided into two subpopulations: monocytic (M)-MDSCs (defined as CD11b⁺Ly6C⁺Ly6G^–cells), and granulocytic (G)-MDSCs defined as (CD11b⁺Ly6C^−/lowLy6G⁺ cells).^4,5 MDSC phenotypes in humans are far more diverse. Human MDSCs have been characterized by the expression of the common myeloid markers CD33 or CD11b, as well as the lack of mature myeloid cell markers such as HLA-DR.^4,5 Immunoregulatory activity of MDSCs involves processes such as direct ligand-receptor engagement, production of reactive oxygen species (ROS) and reactive nitrogen species (RNS). ⁴ For example, MDSC can produce inducible NO synthase (iNOS) which participates in L-arginine depletion (amino acids essential for T cell) by catalyzing the conversion of L-arginine to NO (nitric oxide) and L-citrulline.^6,7 Increased accumulation of MDSCs have been reported in diverse chronic inflammatory conditions including infectious diseases, dysbiosis, autoimmune disorders, or cancer. ⁴ Earlier, it was proposed that MDSCs could be involved in periodontal diseases. ⁸ Recently, MDSCs were identified in gingival tissues with periodontitis, ⁹ however subpopulations of cells with MDSC phenotypes in gingival tissues are widely unknown. The aim of this work is to identify cell subpopulations with MDSC phenotypes in healthy gingival tissues and with periodontitis.

Materials and methods

Ethics statement

The ethical approval to perform the experimental processes involving patients and mouse model was obtained from the Ethics Committee of Facultad de Medicina Mexicali (CEI-FMM) Universidad Autónoma de Baja California (No. FMM/CEI-FMM/014/2023-01). Written informed consent of all participating subjects was obtained prior to inclusion in the study. All subjects gave written informed consent in accordance with the Declaration of Helsinki. All experiments were conducted in accordance with the principles and procedures outlined in the Guideline of Laboratory Animals of the Institution. The ethical standards of experiments were in accordance with the guidelines provided by the World Medical Association Declaration of Helsinki on Ethical Principles for Medical Research Involving Experimental Animals.

Subjects

The clinical characteristics of patients are shown in Table 1. A total of eight patients were recruited in this study after being diagnosed by a periodontist. The subjects had an average age of 42 years old (62.5% were female and 37.5% were male patients).

Table 1.

Characteristics of patients and subjects.

Group	Patient	Gender	Age	BOP (%)	PD (mm)	CAL (mm)	Diagnosis (AAP 2018)	Sample site
Control	1	F	35	8	5 (gingival pocket)	No	Altered Passive Eruption (APE) 1A	Upper left central incisor
Control	2	F	40	6	2	No	Gingival health	Upper right First molar
Periodontitis	3	F	43	40	7	6	Periodontitis Stage 3, Grade B General	Lower left canine
Periodontitis	4	M	38	56	7	5	Periodontitis Stage 3, Grade B General	Lower first right molar
Periodontitis	5	F	55	28	5	4	Periodontitis Stage 2, Grade A Local	Upper left First molar
Control	6	M	27	7	6 (gingival pocket)	No	APE 1A	Upper right central incisor
Periodontitis	7	M	46	23	4	4	Periodontitis Stage 2, Grade A Local	Upper left second premolar
Periodontitis	8	F	53	48	6	5	Periodontitis Stage 3, Grade B General	Lower right second molar

BOP: bleeding on probing; PD: probing depth; CAL: clinical attachment loss; F, female; M, male.

Periodontal tissue specimens

Periodontal tissues were obtained from five patients with periodontitis and three subjects with clinically healthy gingiva, sample location site is indicated in Table 1. These specimens were collected from patients referred to the Periodontal Clinic, Facultad de Odontología Mexicali, Universidad Autónoma de Baja California. All participants were clinically examined for periodontal disease, and those with a tooth site demonstrating bleeding on probing, clinical attachment level ≥4 mm (Table 1) and radiographic assessment of bone loss were included in the periodontitis affected group; diseased gingival tissues were obtained at the time of periodontal surgery (debridement flap surgery) or extraction of severely involved teeth. Healthy gingival tissues were collected during crown-lengthening procedure showing characteristics such as no bleeding on probing (Table 1). Excised tissues were washed with phosphate-buffered saline (PBS) and immediately placed in an Eppendorf tube at 4°C. Samples were processed for flow cytometry analyses.

Experimental model of ligature-induced periodontitis

Pathogen-free male C57BL/6 mice at the age of 10–11 weeks, with no clinical signs of active periodontal disease, were used for this study. The mice were housed under ad libitum light/dark cycles in the temporary animal facility. The mice were anesthetized via intraperitoneal administration of a mixture of ketamine (100 mg/kg) and xylazine (10 mg/kg) and a sterile 6-0 silk suture was carefully placed around the maxillary second molar under a stereomicroscope.^10,11 The contralateral molar tooth in each mouse was left unligated to serve as control. The mice were monitored until full recovery from anesthesia and inspected daily for any signs of distress or complications. Mice were euthanized at 11–13 days post-ligation. Only mice that maintained the ligature in place throughout the experimental period were included. Gingival tissues were recollected from second molars and processed for flow cytometry analyses. Maxillary jaws were dissected to assess bone loss. Spleens of mice with ligature were collected and processed for flow cytometry analyses. Additionally, spleens from mice that did not receive any treatment were processed and included in the flow cytometry analysis as controls.

Cell suspensions

Single-cell suspensions of spleen were obtained after mechanical disaggregation using the Medimachine system (BD Biosciences, CA, USA), according to the manufacturer's instructions. Erythrocytes from spleen samples were lysed using an ACK (ammonium-chloride-potassium) buffer. ¹² Gingival tissues were minced and incubated with collagenase (2 mg/ml) and EDTA (0.5%) for 50 min at 37°C in continuous agitation as previously reported. ¹³ After washing, all samples were filtered through a 70 μm cell strainer and pelleted by centrifugation. Cell counting and viability were assessed using Trypan Blue. Finally, the cells were suspended, stained and analyzed using flow cytometry.

Flow cytometry

Cell suspensions were counted and resuspended in PBS 1x containing BSA (0.5%). Cells were incubated with 10% FBS for 10 min, then samples were stained by incubating the cell suspensions with the appropriate antibodies for 30 min, then cells were washed with PBS 1x containing BSA (0.5%). The following antibodies were used: anti-human CD45-FITC (J33, Beckman Coulter), anti-human HLA-DR-ECD (Immu-357, Beckman Coulter), anti-human CD11b-PC7 (Bear1, Beckman Coulter), anti-human CD33-PC5.5 (D3HL60.251, Beckman Coulter), anti-mouse CD45-PE-Cyanine 7 (30F11, eBioscience), anti-mouse CD11b-PE-Cyanine 5.5 (M1/70, eBioscience), anti-mouse Ly6G/Ly6C-PE (RB6-8C5, eBioscience), anti-mouse Ly-6G-PE (1A8-Ly6g, eBioscience) and anti-mouse iNOS-Alexa Fluor 488 (CXNFT eBioscience). The respective FMO controls were included. After staining of cell surface markers, intracellular iNOS staining was performed with Intracellular Fixation and Permeabilization Buffer Set (eBioscience), according to the manufacturer's instructions. The data were collected with Cytoflex flow cytometer (Beckman Coulter) and analyzed with CytExpert software (Beckman Coulter). Compensation was performed using VersaComp Antibody capture kit (Beckman Coulter). To analyze human gingival tissue cells, a minimum of 150,000 events were acquired per sample. At least 30,000 or 5000 events were acquired per sample of mouse spleen or gingival tissues, respectively. Initial gating included FSC-A/SSC-A and doublet exclusion using both FSC-A vs. FSC-H and FSC-A vs. FSC-W. Then, immune cells were identified according to CD45 expression. To analyze human gingival tissue cells, following gating on CD45⁺, the analyses include HLA-DR^−/low cells. Next, the cells of interest were analyzed as CD11b⁺ and CD33⁺. On the other hand, to analyze cells of mouse splenic and gingival tissues, following gating on CD45⁺, the analyses include Gr-1⁺ and CD11b⁺ cells or Ly6G⁺ and CD11b⁺ cells or iNOS⁺ and Gr-1⁺ cells.

Determination of bone loss induced by ligation

Soft excessive tissues were carefully removed from maxillary jaws using mechanical means. Hemi jaws were immersed in Hydrogen Peroxide (30%) for 16 h for complete removal of soft tissues. ¹⁰ Samples were dried and stained with methylene blue. After the samples were washed and dried, images of hemi jaws were obtained using a millimeter scale ruler under a stereomicroscope StemiDV4 (Zeiss) with a camera AxioCamERc5s (Zeiss) and the Blue Zen software (Zeiss). Image analysis was performed using ImageJ for quantification of bone loss. Bone loss was measured by assessing the linear distance between the cemento-enamel junction (CEJ) and the alveolar bone crest (ABC) at three points on each molar buccally.

Statistical analysis

GraphPad Prism Software (GraphPad Software, La Jolla, California, USA) was used for statistical analysis. The normality of the data distribution was analyzed via the Shapiro-Wilk test. The Student´s t-test was used to compare variables with normal distributions; the test applied is indicated in figure captions. Data are presented as mean ± standard deviation (SD) and the number of independent experiments is indicated in the figure captions. Results were considered significant at p ≤ 0.05.

Results

Gingival infiltration of cells with MDSC phenotype were increased in human periodontitis

Bioinformatic analysis and overexpression of MDSC markers have suggested that MDSCs are increased in gingival tissues of patients with chronic periodontitis^14,15 and cells with MDSC phenotype have not been analyzed by flow cytometry in human gingival tissues. Patients with periodontitis stage 2 and 3, and grade A and B (Table 1) showed increased accumulation of CD45⁺HLA-DR^neg/lowCD11b⁺CD33⁺ cells in gingival tissues (p = 0.01) compared with healthy gingival tissues (Figure 1 and Supplementary Figure 1A); studies analyzing this cellular phenotype in gingival tissues have not been previously reported. Thus, increased infiltration of cells with MDSC phenotype in gingival tissues is promoted by periodontitis. Clinical attachment loss (CAL) may indicate loss of periodontal tissue support, one of the main characteristics of periodontitis. ¹⁶ Thus, a correlation analysis was performed (Supplementary Figure 1B) to examine a possible relation of CAL and the levels of CD45⁺HLA-DR^neg/lowCD11b⁺CD33⁺ cells in gingival tissues of patients with periodontitis. The results did not show a significant correlation between frequency of these cells and CAL (p = 0.4).

Figure 1.

Increased accumulation of cells with MDSC phenotype in gingival tissue of patients with periodontitis. CD45⁺HLA-DR^−/lowCD11b⁺CD33⁺ cells were analyzed in healthy gingival tissue (control) and samples with periodontitis. Gating strategy is indicated and representative percentage of HLA-DR^−/lowCD11b⁺CD33⁺ cells of CD45⁺ cells is shown in plots [patient 2 (control) and patient 7 (periodontitis)]. Graphic shows mean and SD of % HLA-DR^−/lowCD11b⁺CD33⁺ cells of CD45⁺ cells in control (n = 3) and periodontitis groups (n = 5). Student´s t-test was used for statistical analysis. Results were considered significant at *p ≤ 0.05.

Experimental periodontitis induced accumulation of cells with MDSC phenotype

To strengthen these observations, a mouse model of experimental periodontitis induced by ligature was implemented. As expected, increased bone loss (p = 0.001) was detected in the ligated side of maxillae (Figure 2(a)) in contrast to the corresponding contralateral unligated side (Figure 2(a)). Analyses of CD45⁺CD11b⁺Gr-1⁺ cells by flow cytometry in gingival tissues with periodontitis showed increased accumulation of these cells (p = 0.03) as compared with healthy gingival tissues (Figure 2(b)). To characterize more these cells localized in periodontal tissues, the production of the iNOS enzyme, which is related to immunosuppressive potential of MDSCs, was analyzed. Results showed that CD45⁺Gr-1⁺ cells that are localized in healthy gingival tissues and are accumulated in tissues with periodontitis can produce iNOS (Figure 2(c)). Periodontitis slightly promoted accumulation of CD45⁺Gr-1⁺iNOS⁺ cells in gingival tissues; however, this was not significative (p = 0.2) as compared with healthy tissues (Figure 2(c)). Additionally, analyses of a subpopulation with phenotype CD45⁺CD11b⁺Ly6G⁺ showed increased accumulation in samples with periodontitis (p = 0.01) in comparison with healthy gingival samples (Figure 2(d)). Together, these results indicate that cells with MDSC phenotypes are residents in healthy gingival tissues and periodontitis induces their accumulation.

Figure 2.

Augmented accumulation of cells with MDSC phenotype in gingival tissue of an experimental periodontitis model induced by ligature. (a) Increased bone loss induced by ligation. Representative images of maxillae presenting bone loss of the ligated side (Periodontitis, n = 3) and of the corresponding contralateral unligated side (Control, n = 3) in a model of experimental periodontitis. Graphic represents the quantitative analyses of bone loss. CEJ, cementoenamel junction. ABC, alveolar bone crest. Scale bar indicates 1 mm. Student´s t-test was used for statistical analysis. Results were considered significant at *p ≤ 0.05. (b) CD45⁺CD11b⁺Gr-1⁺ cells were analyzed in healthy gingival tissue (control) and samples with experimental periodontitis in mouse model. Gating strategy is indicated and representative percentage of CD11b⁺Gr-1⁺ cells of CD45⁺ cells is shown in plots. Graphic shows mean and SD of % CD11b⁺Gr-1⁺ cells of CD45⁺ cells in control (n = 4) and experimental periodontitis groups (n = 4). Student´s t-test was used for statistical analysis. Results were considered significant at *p ≤ 0.05. (c) A proportion of periodontal CD45⁺Gr-1⁺ cells accumulated in gingival tissues of mice with experimental periodontitis showed production of iNOS. Representative percentage of Gr-1⁺iNOS⁺ cells of CD45⁺ cells is shown in plots. Graphic shows mean and SD of % Gr-1⁺iNOS⁺ cells of CD45⁺ cells in control (n = 2) and experimental periodontitis groups (n = 2). Student´s t-test was used for statistical analysis. (d) CD45⁺CD11b⁺Ly6G⁺ cells were analyzed in healthy gingival tissue (control) and samples with experimental periodontitis in a mouse model. Representative percent of CD11b⁺Ly6G⁺ cells of CD45⁺ cells are shown in plots. Graphic shows mean and SD of % CD11b⁺Ly6G⁺ cells of CD45⁺ cells in control (n = 4) and experimental periodontitis groups (n = 4). Student´s t-test was used for statistical analysis. Results were considered significant at *p ≤ 0.05.

Experimental periodontitis did not promote splenic expansion of cells with MDSC phenotype

To explore the capacity of the experimental periodontitis to induce expansion of cells with MDSC phenotype in other tissues, CD45⁺CD11b⁺ Gr-1⁺ cells were analyzed in spleen of mice with periodontitis. Analysis by flow cytometry showed that splenic CD45⁺CD11b⁺Gr-1⁺ cells were not increased in experimental periodontitis (p = 0.9) as compared to control mice (Figure 3(a)). Additionally, the splenic cellular subpopulation with phenotype CD45⁺CD11b⁺Ly6G⁺ were not augmented (p = 0.9) in mice with periodontitis (Figure 3(b)).

Figure 3.

The accumulation of splenic CD45⁺CD11b⁺Gr-1⁺ cells and splenic CD45⁺CD11b⁺Ly6G⁺ cells was not increased in the experimental periodontitis model induced by ligature. (a) CD45⁺CD11b⁺Gr-1⁺ cells were analyzed in spleen of control mice and with experimental periodontitis. Gating strategy is indicated and representative percent of CD11b⁺Gr-1⁺ cells of CD45⁺ cells is shown in plots. Graphic shows mean and SD of % CD11b⁺Gr-1⁺ cells of CD45⁺ cells in control (n = 3) and experimental periodontitis groups (n = 6). Student´s t-test was used for statistical analysis. (b) CD45⁺CD11b⁺Ly6G⁺ cells were analyzed in spleen of control mice and with experimental periodontitis. Representative percent of CD11b⁺Ly6G⁺ cells of CD45⁺ cells is shown in plots. Graphic shows mean and SD of % CD11b⁺Ly6G⁺ cells of CD45⁺ cells in control (n = 3) and experimental periodontitis groups (n = 6). Student´s t-test was used for statistical analysis. Results were considered significant at *p ≤ 0.05.

Discussion

Two reports performing in silico analysis suggested that MDSCs can be increased in diseased gingival tissues.^14,15 Recently, it was indicated by immunofluorescence that cells with MDSC phenotype CD11b⁺CD14⁺HLA-DR^low were increased in gingival tissue of periodontitis patients, ⁹ however information about other phenotypes is absent. At the best of our knowledge, this is the first report showing increased accumulation of CD45⁺HLA-DR^neg/lowCD11b⁺CD33⁺ cells in the gingival tissues of patients with periodontitis in different stages and grades. In accordance, different reports have determined that the presence of chronic inflammatory processes induce accumulation of MDSC cells with these phenotypes.^5,17–19 Previous publications have not addressed or discussed the relationship between levels of MDSCs and clinical characteristics of patients with periodontitis, in this sense, CD45⁺HLA-DR^neg/lowCD11b⁺CD33⁺ cell frequency and CAL severity did not show significant correlation, however it is important consider the reduced number of samples in the present study. Correlation analysis should be considered in future studies to connect immune phenotype with clinical phenotype and to determine possible translational relevance.

The results of this work showed that experimental periodontitis in a mouse model promoted accumulation of CD45⁺CD11b⁺Gr-1⁺ cells in diseased gingival tissues. In similar way, these cells were very recently found in mouse gingival tissues with increased accumulation promoted by experimental periodontitis, and it was indicated that this population presented immunosuppressive capacity. ⁹ Additionally, the cited report also indicated that subpopulation of cells with phenotype CD11b⁺Gr-1⁺ Ly6G⁻Ly6C^high, indicated as M-MDSCs, were increased in gingival tissues of mice with experimental periodontitis generating immunosuppression. ⁹ In the present work, the model of experimental periodontitis showed an increased accumulation of CD45⁺CD11b⁺Ly6G⁺ cells. Reports have related this phenotype to subpopulation of G-MDSCs. ²⁰ MDSC produces different enzymes to exert their immunomodulatory effects. ⁴ MDSCs can produce iNOS resulting in the production of reactive nitrogen species (primarily NO) that can react with superoxide to produce peroxynitrites which could promote apoptosis of T cells or nitration of the T-cell receptor (TCR) resulting in inhibition of T cell activation.^6,7 Interestingly, NO generates varied effects on the immune response depending on NO concentration and time of exposure. ²¹ The results showed a subpopulation of CD45⁺Gr-1⁺iNOS⁺ cells which had not been previously analyzed in gingival tissue. The production of iNOS by this myeloid subpopulation could have immunomodulatory effects in gingival tissues, however this remains to be determined.

This work and another published report ⁹ showed that different subpopulations with phenotypes of MDSCs were found in very low quantities in mice and healthy human gingival tissue, suggesting that they are gingival resident cells. Presence of a gingival inflammatory process as periodontitis triggers local accumulation of these diverse subpopulations with MDSC phenotype. Interestingly, experimental periodontitis by ligature did not promote expansion of these cell populations in spleen. In concordance, it was reported that subpopulations of MDSCs were not systemically expanded (blood and spleen) in mice with periodontitis induced via oral infection with P. gingivalis. ²² However, periodontitis associated with other inflammatory conditions [obesity, rheumatoid arthritis (RA) or cancer] can promote the expansion of MDSC subpopulations and their recruitment to tissues in mouse models,^22–24 suggesting that periodontitis and other inflammatory diseases can synergize to stimulate systemic MDSC expansion. In this sense, the systemic expansion of MDSCs by periodontitis in association with obesity was linked to enhanced loss of alveolar bone, augmented osteoclastogenesis, and severe periodontal inflammation.^3,23 Similarly, the increased accumulation of MDSCs by the simultaneous presence of periodontitis and RA was associated with more inflammation and exacerbated periodontal bone loss, as well as more severe arthritis symptoms.^3,22 Additionally, periodontal disease induced in a cancer mouse model promoted increased accumulation of MDSCs in cervical lymph nodes and head and neck tissues and this was associated with more metastasis in these tissues.^3,24 Thus, MDSCs could participate in a possible bidirectional mechanism connecting periodontitis with other inflammatory conditions. ⁴

It is important to note that one of the main limitations of this work is the small sample size. Consequently, caution should be exercised when generalizing these findings to a broader population. Thus, analyses of cell populations with MDSC phenotype should be performed in a more diverse and robust sample to strengthen the observations of the present study. Moreover, the small sample size may have resulted in low statistical power to detect subtle differences between groups and more susceptibility to the influence of outliers. It is possible that these effects could affect our correlation analysis and iNOS examination, which did not show significant differences. While the sample size is small, it provides valuable data for future large-scale investigations analyzing the possible relationship between frequency of these cells and important clinical parameters such as CAL; as well as determining the molecules generated by these cells in periodontitis to establish potential functional effects. Additionally, a more detailed characterization of MDSC subpopulations in gingival tissues of patients with periodontitis is necessary, considering that different subpopulations may perform specific functions. In this regard, further studies addressing an in-depth characterization of the diverse molecules with immunoregulatory properties differentially produced by MDSC subpopulations in gingival tissue could provide a broader picture of the roles of these cells in the pathogenesis of periodontitis and the molecular mechanisms involved. The presented results showed new populations of cells with MDSC phenotypes in gingival tissues, with potential immunomodulatory capacity since a population showed production of iNOS. Additionally, periodontitis promoted accumulation of cell subpopulations with MDSC phenotypes. These observations are important since it was reported that depletion of gingival Gr-1⁺ MDSC in an experimental periodontitis model resulted in reduced inflammation levels, decreased presence of osteoclasts and increased bone regeneration. ⁹ Thus, modulation of the presence or activity of gingival subpopulations with MDSC phenotypes could have potential therapeutics in gingival inflammation, however more studies are needed.

Conclusions

This study reveals new insights into cell subpopulations with MDSC phenotype presents in gingival tissues. Periodontitis promotes increased accumulation of cells CD45⁺HLA-DR^neg/lowCD11b⁺CD33⁺ in human gingival tissues. In experimental model, periodontitis stimulates the infiltration of CD45⁺CD11b⁺Gr-1⁺ cells and CD45⁺CD11b⁺Ly6G⁺ cells in gingival tissues. In addition to these cell populations, the gingival CD45⁺Gr-1⁺iNOS⁺ cells could be implicated in production of reactive nitrogen species, suggesting immunomodulatory properties.