Complement 3 mediates periodontal destruction in patients with type 2 diabetes by regulating macrophage polarization in periodontal tissues

Abstract Objectives Diabetes aggravates the risk and severity of periodontitis, but the specific mechanism remains confused. Complement 3 (C3) is closely related to complications of type 2 diabetes (T2DM). In the present study, we concentrated on whether C3 mediates the development of periodontitis in T2DM. Materials and Methods Levels of C3 in blood and gingival crevicular fluid (GCF) of patients were measured first. A C3‐knockout diabetic mouse model was established, real‐time PCR, Western blotting and histological investigation were performed to evaluate the progress of periodontitis. Microcomputed tomography (micro‐CT) and TRAP staining were performed to detect alveolar bone resorption. Immunofluorescence was performed to detect polarization of macrophages. Results Our data showed that C3 levels were elevated in the blood and GCF of T2DM patients compared with non‐diabetic individuals. Increased C3 was closely related to the upregulation of inflammatory cytokines including interleukin (IL)‐1, IL‐6 and tumour necrosis factor‐alpha (TNF‐α), as well as the decline of the bone volume density (BMD) and bone volume over total volume (BV/TV) of the alveolar bones in diabetic mice. The deletion of C3 inhibited inflammatory cytokines and rescued the decreased BMD and BV/TV of the alveolar bones. C3‐mediated polarization of macrophages was responsible for the damage. Conclusion T2DM‐related upregulation of C3 contributes to the development of periodontitis by promoting macrophages M1 polarization and inhibiting M2 polarization, triggering a pro‐inflammatory effect on periodontal tissues.

The complement system is closely related to the development of diabetes complications. 8 A large number of complement components, activated fragments and end product membrane attack complex (MAC) deposits are found in the glomerular basement membrane, retinal vascular layer and neural tissues of patients with chronic complications of diabetes. [9][10][11] Complement 3 (C3) is a key molecule of the complement system. 12,13 C3 activation was detected in T2DM. 14 Epidemiological and experimental studies have jointly suggested that C3 is closely related to vascular complications of diabetes and exacerbates diabetic nephropathy, retinopathy and neurological disease. [15][16][17] In addition, C3 was hyper-activated in periodontitis. Deletion of C3 has been shown to effectively inhibit the destruction of periodontal tissue and alveolar bone resorption in a periodontitis mouse model. 18,19 Thereafter, further investigation into whether C3 mediates the development of periodontitis in T2DM patients is necessary.
The purpose of this study was to clarify the role of C3 in the progression of T2DM-related periodontitis. In addition, preliminary mechanism was explored. In the present study, the levels of C3 in blood and gingival crevicular fluid (GCF) of patients were measured first, and a C3-knockout diabetic mouse model was established to study the roles of C3 in the progression of T2DM-related periodontitis. were enrolled from the outpatient clinic of the Stomatological Hospital (College) of Xi'an Jiaotong University.

| Case inclusion
The inclusion criteria of T2DM: patients diagnosed with T2DM for more than six months and in stable condition; good periodontal conditions without gingivitis or periodontitis; no antibiotic or immunosuppressant treatment in the past three months; not pregnant; and no acute infections or allergies. The inclusion criteria of the non-diabetic patients: individuals with normal blood glucose without type-1 diabetes or T2DM; who had good periodontal conditions without gingivitis or periodontitis; who had no systemic diseases; who did not consume antibiotics or immunosuppressants in the past three months; who were not pregnant; and who had no acute infection or allergy. The inclusion criteria of periodontitis: patients diagnosed with periodontitis without any periodontal treatments in the last 6 months; who with normal blood glucose without type-1 diabetes or T2DM; no antibiotic or immunosuppressant treatment in the past three months; not pregnant; and no acute infections or allergies. The inclusion criteria of diabetic periodontitis: patients diagnosed with T2DM for more than six months and in stable condition; patients diagnosed with periodontitis without any periodontal treatments in the last 6 months; no antibiotic or immunosuppressant treatment in the past three months; not pregnant; and no acute infections or allergies.
All included patients signed the written informed consent and agreed to the test of C3 content in blood and GCF and the rou-

| Sample collection and clinical evaluation
Six millilitre of peripheral blood was collected from all patients with empty stomach in the morning, serum was centrifuged at 4°C, 3000 r/ min, 15 minutes, and then stored at −70°C. Filter paper (Whatman) was prepared into 2 mm × 8 mm size. Bilateral maxillary and mandible molars (The third molars were not included) were tested at mesial, distal and buccal positions. For GCF collection, the filter paper was inserted into the sulcus for 30 seconds and washed by centrifugation with 1% PBS. Bleeding on Probing (BOP), gingival index (GI), probing depth (PD) and tooth mobility (Mob) were evaluated for each patient.

| Animals
C57BL/6 male mouse weighing ~23 g (n = 40) was used in the present study. All mice were housed under standard conditions with a 12 hour light/dark cycle in a specific-pathogen-free (SPF) facility.
Mice were randomly divided into four groups. The control group was fed with basic food for 4 weeks before 4 weeks of injection with normal saline (n = 10). The STZ group was fed with basic food for 4 weeks before 4 weeks of injection with streptozotocin (STZ, 60 mg/kg) (n = 10). Four weeks of high-sugar and high-fat food followed by 4 weeks of normal saline injection were performed as HF/HS diet group (n = 10). Four weeks of high-sugar and highfat food followed by 4 weeks of low dose (60 mg/kg) STZ intraperitoneal injection were performed to simulate T2DM (n = 10).

| Isolation of total RNA and RT-PCR
Gingiva was harvested from the maxillary second molars and stored at −70°C for real-time PCR. Total RNA was isolated using the Trizol reagent (Invitrogen) according to the manufacturer's instructions and diluted in 30 μL of RNase-free water. Equal amounts of total RNA (2 μg) were reverse transcribed with a First Strand cDNA Synthesis Kit (Fermentas). Gene expression was analysed using an iQ5 (Bio-Rad) with SYBR ® Premix Ex Taq™ II (TaKaRa). Sequences of the primers are provided in Table 1.

| Histological examinations
Maxillary bones were excised and immediately fixed in 4% paraformaldehyde neutral buffer solution for 48 hours. Then, the maxillary specimens were decalcified with 10% EDTA at room temperature for 7 days until the alveolar bone could be easily penetrated followed by conventional dehydration and paraffin embedding. Immunohistochemical detection of C3 and RANKL as well as HE staining was performed in each group of samples. Examination and analysis were performed in blind. For immunohistochemical detection, 5 μm sections were prepared. Deparaffinized sections were treated with methanol containing 3% hydrogen peroxide before conducting antigen retrieval using a microwave oven at 95°C for 5 minutes, and cooling at room temperature for 15 minutes for two times. After washing with phosphate-buffered saline (PBS), 5% bovine serum albumin was applied for 10 minutes. The sections were incubated with antibody 16-18 hours overnight at 4°C.

| Micro-computed (Micro-CT)
Bilateral maxillary was fixed in 4% paraformaldehyde. Tissues were scanned on a micro-CT system. The parameters are set as follows: 70 kV, 114 μA, 12 μm resolution; Root bifurcation area of second molar was defined as the volume of interest (VOI). Thirty slices prior to and after the identified furcation slice were added to generate a VOI. The distance between cemento-enamel junction and the alveolar bone crest (CEJ-ABC) was assessed. Bone volume density (BMD) and bone volume over total volume (BV/TV) were analysed.

| Tartrate-resistant acid phosphatase (TRAP)
TRAP staining was performed by using Acid Phosphatase, Leukocyte (TRAP) Kit (Sigma) according to the manufacturer's instructions.
Briefly, slices were deparaffinized and rinsed with deionized water.
TRAP staining was performed at 37°C in the dark for 1 hour. Nuclei were stained with haematoxylin for 2 minutes.

| Immunofluorescence
Detailed methods are described as previously. 22 Briefly, 5 μm sections were prepared. Slices were deparaffinized and incubated with 5% bovine serum albumin for 30 minutes. Afterwards, samples were incubated with primary antibody at 4°C overnight. After washing with PBS, the sections were incubated for with secondary antibody for 2 hours. Fields of connective tissue adjacent to the junctional epithelium as well as alveolar bone were selected from each sample to calculate the number of positive cells. Primary antibodies used were as follows: Rat Anti-CD68 antibody (1:100, Abcam, ab53444) and

| Statistical analysis
The quantitative data shown in the figures are presented as the mean ± SD. One-way analysis of variance (ANOVA) and Student's t test were performed to analyse differences among or between groups using SPSS 16.0 software (SPSS, Inc). In all analyses, a P value < .05 was taken as the level of significance.

| C3 levels were increased in the blood and GCF of T2DM patients
In this study, we compared the levels of C3 and its fragments in the blood of non-diabetic individuals, T2DM patients, periodontitis patients and diabetic periodontitis patients. Table 2 shows the age and sex of the participants of this study. As shown in Figure 1A-C, the blood C3, C3a and iC3b of the T2DM patients were elevated compared with non-diabetic individuals. What's more, the concentration of C3 and C3a was higher in diabetic periodontitis patients compared with T2DM patients. Further measurement of the C3 levels in the GCF of the four groups showed that T2DM patients had higher C3, C3a and iC3b levels than non-diabetic individuals. In addition, their levels in diabetic periodontitis patients were elevated compared with T2DM patients ( Figure 1D-F).

| T2DM patients had worse periodontal conditions
The weight of GCF was gained in the T2DM group compared with non-diabetic individuals, while the weight gained even more in the periodontitis patients and diabetic periodontitis patients ( Figure 2A). The level of IL-1β in the GCF, which was the key cytokine in the development of periodontal disease, also increased in the T2DM group, periodontitis group and diabetic periodontitis group ( Figure 2B). Further examination of the periodontal conditions in the four groups revealed that the BOP and GI of patients with T2DM were higher than that of non-diabetic individuals, while they were even higher in the periodontitis patients and diabetic periodontitis patients ( Figure 2C,D). No significant difference in the probing depth and tooth mobility was found between T2DM group and nondiabetic individuals, while they were increased in the periodontitis group and diabetic periodontitis group ( Figure 2E,F).

| Increased C3 was closely related to periodontal destruction in diabetic mice
According to Figures 1 and 2, we suspected that the increase in C3 The C3 levels in T2DM group were significantly higher than those of normal mice, while no significant difference was seen between normal mice and T1DM mice. We next observed inflammation of the gingival tissue in the diabetic mice ( Figure 3E-G). The expression of inflammatory cytokines including interleukin (IL)-1, IL-6 and tumour necrosis factor-alpha (TNF-α) in the gingiva of diabetic mice was significantly increased. While no significant difference was seen in the Diet group. Therefore, we may conclude that T2DM-related C3 upregulation was closely related to the development of periodontal inflammation.

| Deletion of C3 inhibited periodontal destruction in diabetic mice
To further prove that C3 is a key molecule for T2DM to promote periodontal disease, we established a C3 knockout mouse model and induced them with T2DM. The knockout effect of the mouse model was first verified in the gingival tissues by Western blotting and immunohistochemistry, showing that C3 expression was strongly inhibited (Figure 4A,B). Decreased inflammation of the gingival tissue was observed in the C3 KO mice ( Figure 4C). Further investigation of the diabetic mouse model showed that the deletion of C3 significantly downregulated the inflammatory cytokines IL-1, IL-6 and TNF-α compared with the non-knockout group ( Figure 4D).
Micro-CT showed that there was no difference in CEJ-ABC among the three groups, while the deletion of C3 rescued the decreased BMD and BV/TV of the alveolar bones ( Figure 4E). No significant difference was observed in the alveolar bone of C3 KO and WT mice ( Figure 4F).

| C3-mediated polarization of macrophages was responsible for the periodontal damage
In this study, we also assessed osteoclasts in the alveolar bone and showed that they increased significantly in the diabetic mice, while their number in alveolar bone decreased after the deletion of C3 ( Figure 5A). RANKL was an indicator of alveolar bone destruction in periodontitis. So we next evaluated RANKL expression in the alveolar bone. We found that the expression of RANKL was increased in the diabetic mice, while decreased after the deletion of C3 ( Figure 5B). The quantitative results are shown in Figure 5C. We suggest that although the increase in C3 due to T2DM did not cause evident bone destruction, the increased number of osteoclasts may be the reason for the promotion of alveolar bone resorption.
We next investigated the mechanism of C3 regulation in the

| D ISCUSS I ON
In this study, we found that the increase in C3 levels of T2DM patients may be the cause of periodontal destruction. Although it did not cause obvious bone destruction, the increase in the number of F I G U R E 1 C3 levels were increased in the blood and GCF of T2DM patients. A, The blood C3 level of the non-diabetic individuals, T2DM patients, patients with periodontitis and diabetic periodontitis patients. ***P < .001. B, The blood C3a levels in the four groups. ***P < .001. C, iC3b levels in the blood of four groups. ***P < .001. D, C3 levels in the GCF of non-diabetic individuals, T2DM patients, patients with periodontitis and diabetic periodontitis patients. ***P < .001. E, The GCF C3a level of the four groups. *P < .05, ***P < .001. F, The GCF iC3b level of the four groups. *P < .05, **P < .01, ***P < .001 osteoclasts may aggravate periodontal bone destruction. Increase in C3 caused macrophages in the periodontal tissues to polarize to M1 and inhibited the M2 polarization which may be a mechanism by which it promotes periodontitis.
Diabetes affected bacteria-host interactions to promote inflammation and periodontal disease. 23 The inflammatory response modified oral microbiota to render it more pathogenic. Complement system may also be involved in the process. Amplified immune response mediated by T2DM-related C3 upregulation may lead to further bacteria-host interactions which aggravate periodontal damage. We showed that C3 levels in the blood and GCF of T2DM patients were elevated, and so was the weight of the GCF. In healthy periodontal tissues, the secretion of GCF is limited. Increased GCF was resulted from the break of bacteria and their products from biofilm into gingival crevasse. Therefore, increased GCF is one of the main manifestations of early gingivitis and often precedes changes in clinical characteristics. 24 When gingivitis becomes obvious, the GCF increases substantially, and inflammatory cytokine concentrations also increase. 25 This study showed that the weight of GCF of T2DM patients was higher than that of non-diabetic individuals. In addition, the inflammatory cytokine concentrations of T2DM patients were also elevated, suggesting that T2DM patients had a tendency for early gingival inflammation compared with non-diabetic individuals, which may be related to the increase of C3 levels.
In the present study, we first fed mice with high fat and high sugar food for 4 weeks to induce insulin resistance, then low dose of STZ injection was followed for next 4 weeks to destruct islet tissues to establish T2DM model. Basic diet with STZ injection (T1DM) did not change the expression of C3, which was in line with previous studies. 26 We further showed that the significant increase in C3 levels in in the late stage. Therefore, we may suggest that by controlling C3 concentration at early stage of T2DM, we can prevent later periodontal destruction. C3 concentrations in GCF of diabetic patients may also be a potential indicator of T2DM-related periodontitis.

| CON CLUS ION
In summary, this study showed that C3 levels were significantly increased in the blood and GCF of T2DM patients. Increased C3 levels were accompanied by periodontal destruction. Deletion of C3 extenuated gingival inflammation and rescued potential bone destruction. C3 increased M1 macrophage polarization and decreased M2 polarization, suggesting that C3 may promote the development and progression of periodontal inflammation by regulating macrophages polarization. Xi'an Jiaotong University (xtr012019007).

CO N FLI C T O F I NTE R E S T
The authors declare no conflicts of interest to the authorship and/or publication of this article.

AUTH O R CO NTR I B UTI O N S
Ye Li, contributed to conception, design, data acquisition, analysis, and interpretation, drafted and critically revised the manuscript.
Xinxin Wang, contributed to conception, design, data analysis and interpretation, critically revised the manuscript. Saisai Wang, Chunhui Zhu and Jing Guo contributed to data acquisition, analysis, and interpretation, critically revised the manuscript. Ke Li, contributed to conception, data acquisition and analysis, critically revised the manuscript. Ang Li, contributed to conception, design, data analysis and interpretation, drafted and critically revised the manuscript.
All authors gave final approval and agreed to be accountable for all aspects of the present work.

All experiments were reviewed and approved by the Ethics
Committee of College of Stomatology, Xi'an Jiaotong University.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.