Gözeneklilik ve Yoğunluk Değişimi

Experimental arthritis exacerbates Aggregatibacter actinomycetemcomitans- induced periodontitis in mice

Celso Martins Queiroz-Junior*†‡, Mila Fernandes Moreira Madeira†‡, Fernanda Matos Coelho‡, Camila Ribeiro de Oliveira*‡, Luíza Castro Menezes Cândido ‡, Gustavo Pompermaier Garlet§, Mauro Martins Teixeira‡, Daniele da Glória de Souza†, Tarcília Aparecida da Silva*‡

*Department of Oral Surgery and Pathology, Faculdade de Odontologia, Universidade Federal de Minas Gerais, Minas Gerais, Brazil;

†

Department of Microbiology, Instituto de Ciências Biológicas, Universidade Federal de Minas Gerais, Minas Gerais, Brazil;

‡

Department of Biochemistry and Immunology, Instituto de Ciências Biológicas, Universidade Federal de Minas Gerais, Minas Gerais, Brazil;

§

Department of Biological Sciences, School of Dentistry of Bauru, Universidade de São Paulo, São Paulo, Brazil.

Artigo científico submetido ao periódico Journal of Clinical Periodontology e avaliado pelos editores e revisores da revista. O trabalho está em revisão para responder as questões apontadas pelos revisores.

Artigo Científico 2

39 Experimental arthritis exacerbates Aggregatibacter actinomycetemcomitans-

induced periodontitis in mice

Celso Martins Queiroz-Junior*†‡, Mila Fernandes Moreira Madeira†‡, Fernanda Matos Coelho‡, Camila Ribeiro de Oliveira*‡, Luíza Castro Menezes Cândido ‡, Gustavo Pompermaier Garlet§, Mauro Martins Teixeira‡, Daniele da Glória de Souza†, Tarcília Aparecida da Silva*‡1

*Department of Oral Surgery and Pathology, Faculdade de Odontologia, Universidade Federal de Minas Gerais, Minas Gerais, Brazil;

†

Department of Microbiology, Instituto de Ciências Biológicas, Universidade Federal de Minas Gerais, Minas Gerais, Brazil;

‡

Department of Biochemistry and Immunology, Instituto de Ciências Biológicas, Universidade Federal de Minas Gerais, Minas Gerais, Brazil;

§

Department of Biological Sciences, School of Dentistry of Bauru, Universidade de São Paulo, São Paulo, Brazil.

Running title: Arthritis aggravates periodontitis

Keywords: Arthritis, Periodontal Disease, TNF-α, Bone Loss, Experimental Model

Conflict of interest and source of funding statement: The authors declare that they have no conflict of interests. This work was supported by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq, Brazil) and Fundação do Amparo a Pesquisas do Estado de Minas Gerais (FAPEMIG, Brazil).

1

Corresponding author: Tarcília Aparecida da Silva. Mailing address: Departamento de Clínica, Patologia e Cirurgia Odontológicas, Faculdade de Odontologia, Universidade Federal de Minas Gerais, Av. Presidente Antônio Carlos 6627, CEP 31.270-901, Belo Horizonte, Minas Gerais, Brazil. Phone: 55 31 3409-2478 (voice); 55 31 3409-2430 (Fax). E-mail: [email protected].

Artigo Científico 2

40 Abstract

Aim: This study aimed to investigate the influence of chronic antigen-induced arthritis (AIA) on infection-induced periodontitis (PD) in mice. The contribution of TNF-α was also evaluated.

Materials and methods: PD was induced in C57BL/6 mice by oral infection with Aggregatibacter actinomycetemcomitans. AIA was induced 30 d after infection. Anti- TNF-α and chlorhexidine treatments were used to investigate the role of TNF-α and oral infection on PD and AIA interaction. Maxillae, knee joints, lymph nodes and serum samples were used for histomorphometric, immunoenzimatic and/or real time-PCR analyses.

Results: AIA exacerbated alveolar bone loss triggered by PD infection. In contrast, PD did not influence AIA in the evaluated time point. PD exacerbation was associated with enhanced production of IFN-γ in maxillae and expression of Th1 transcription factor in submandibular lymph nodes. Increased serum levels of IL-6 and C-reactive protein were also detected. Anti-TNF-α and antiseptic therapies prevented the development and exacerbation of infectious-PD by reducing the expression of IFN-γ, TNF-α and IL-17 in maxillae.

Conclusions: Altogether, the current results indicate that exacerbation of PD by AIA is prevented by anti-TNF-α therapy and is associated with an alteration of the Th polarization pattern and increased systemic reactivity.

Artigo Científico 2

41 Clinical Relevance

Scientific rationale for the study: Clinical evidence suggests that PD is more severe in patients with RA. The current study investigated the interaction between PD and RA by assessing immune imbalances triggered by antigen-induced arthritis (AIA) in mice with PD.

Principal findings: AIA aggravated alveolar bone loss and inflammation in the maxillae of mice with PD. This exacerbation was associated with functional immune interferences, systemic reactivity to auto-antigens and amelioration by anti-TNF-α therapy.

Practical implications: Understanding the interaction between periodontitis and arthritis in a mouse model provides insights for therapeutic strategies to improve the course of periodontal breakdown in RA patients.

Artigo Científico 2

42 Introduction

Periodontal disease (PD) and rheumatoid arthritis (RA) are some of the most prevalent chronic inflammatory diseases that affect soft and bone structures in humans (Albandar, 2005, Firestein, 2005). Although PD is primarily triggered by infection and RA is an auto-immune disorder, the examination of both conditions indicates a remarkable similarity of their pathogenic features, such as bone resorption, cell infiltration and release of inflammatory mediators, e.g. tumor necrosis factor (TNF)-α, in affected sites (Pablo et al. 2009). These similarities have been suggested to account for the development and exacerbation of RA signs by PD (Ribeiro et al. 2005, Bartold et al. 2010, Cantley et al. 2011) and vice versa, increased PD in patients with RA (Tolo & Jorkjend, 1990, Mercado et al. 2000, Mercado et al. 2001, Pablo et al. 2009).

Recently, experimental studies from Bartold et al. (2010) and Cantley et al. (2011) showed that Porphyromonas gingivalis-induced inflammation can promote systemic alterations, which lead to the development and enhancement of arthritis signs in rodents. Inversely, our group (Queiroz-Junior et al. 2011) and others (Tatakis & Guglielmoni 2000, Ramamurthy et al. 2005, Trombone et al. 2010, Park et al. 2011) indicated that arthritis can induce alveolar bone loss in the presence of a non-manipulated oral microbiota. In a model of antigen-induced arthritis (AIA) in mice, we showed that AIA spontaneously triggered signs of PD through induction of T-helper (Th) 17 polarization and systemic hyper-reactivity to auto-antigens, in a TNF-α-dependent manner (Queiroz- Junior et al. 2011). Moreover, collagen-induced arthritis in mice triggered PD by increasing osteoclastic activity (Park et al. 2011). Altogether, these results support a causal hypothesis – RA triggering PD, and vice versa, through systemic immune activation (Pablo et al. 2009) – but the influence of RA on the severity of pre-existent PD has not been addressed yet. Besides an increase in prevalence, several clinical studies suggest greater severity of PD among individuals with RA (Tolo & Jorkjend, 1990, Mercado et al. 2000, Mercado et al. 2001, Havemose-Poulsen et al. 2006, Pischon et al. 2008). Accordingly, in a model of experimental arthritis induced by systemic challenge, it was observed exacerbation of infection-induced PD in mice with maximum acute inflammatory reactivity (Trombone et al. 2010). Nevertheless, the mechanisms by which RA exacerbates PD are not well understood.

In the current study, we investigated the influence of chronic antigen-induced arthritis (AIA) on Aggregatibacter actinomycetemcomitans-induced PD in mice. We also

Artigo Científico 2

43 evaluated the contribution of anti-TNF-α therapy, oral microbiota and systemic imbalances triggered by AIA on this condition.

Artigo Científico 2

44 Material and Methods

Mice

Experimental groups consisted of 6-week-old male C57BL/6 mice housed under standard conditions in the animal facilities of the Department of Microbiology, Instituto de Ciências Biológicas, Universidade Federal de Minas Gerais, Brazil. Mice had free access to chow and water. All experiments were performed according to a protocol approved by the local Institutional Committee for Animal Care and Use (protocol number: 165/2009).

Experimental periodontitis

Infectious periodontal disease (PD) was induced in two groups of mice – with or without experimental arthritis – as described previously (Garlet et al. 2006). Periodontal infection was achieved by oral delivery of 1 x 109 CFU of the periodontopathogen Aggregatibacter actinomycetemcomitans strain FDC Y4 (from the collection of Laboratório de Microbiologia Oral e Anaeróbios [Department of Microbiology, UFMG]) in 100 µL phosphate buffered saline (PBS) with 1.5% carboxymethylcellulose (CMC), on day 0, for 3 alternated days. Oral infection was induced 30 days before arthritis re-challenge. Mice were euthanized 45 days post-infection and maxillae, knee joints, spleen, serum, inguinal and submandibular lymph nodes were collected for analysis. Negative controls included naïve and sham-infected mice.

Chronic antigen-induced arthritis

Chronic antigen-induced arthritis (AIA) was performed as described earlier (Wengner et al. 2007, Queiroz-Junior et al. 2011). Mice were immunized with subcutaneous 100 µg mBSA (Methylated Bovine Serum Albumin, Sigma-Aldrich, Saint Louis, MO, USA) emulsified in Freund’s complete adjuvant (CFA; Sigma-Aldrich) and further reinforced with mBSA in Freund’s incomplete adjuvant (IFA; Sigma-Aldrich). In parallel, 200 ng of Bordetella pertussis toxin (Calbiochem, La Jolla, CA, USA) was injected (i.p.). These immunizations do not induce any signs of arthritis in mice (Wengner et al. 2007, Queiroz-Junior et al. 2011). Fourteen days after immunization, mice were orally inoculated for PD (day 0), as described above, and first joint-challenged. Subsequently, 30 d later, mice were re-challenged by intra-articular joint injection with mBSA to induce chronic AIA and were euthanized 15 days later (corresponding to 45 days of oral

Artigo Científico 2

45 infection). In the group PD, without AIA, mice were immunized but joint challenge with the antigen mBSA was replaced by vehicle. Non-immunized mice, inoculated for PD, were also evaluated but presented no difference from immunized animals.

Anti-TNF-α therapy

To evaluate the role of TNF-α on PD and AIA interaction, mice were treated with infliximab (Remicade®, Shering-Plough, Kenilwoth, NJ, USA, 10 mg/kg, i.p.), a chimeric monoclonal anti-TNF-α antibody, each 2 days, until the 45th day post-infection (Sachs et al. 2011).

Oral anti-microbial therapy

To investigate the effects of oral infection on PD and AIA association, two additional groups – PD+CLX and PD+AIA+CLX – were created by topical delivery of 50 µL chlorhexidine (1-chlorhexidine gluconate 0.12%) plus 2% CMC, each 2 days, until the 45th day post-infection (Trombone et al. 2010). Control group received 50 µL aqueous 2% CMC.

Morphometric evaluation of maxillae

At 45 days of A. actinomycetemcomitans infection, maxillae of 5 mice/group were collected and analyzed as described previously (Garlet et al. 2006, Queiroz-Junior et al. 2011). The palatal faces of maxillae samples were photographed using a stereomicroscope and a digital camera (Kodak EasyShare C743, Kodak, Manaus, Brazil). Quantitative analyzes comprised the measurement of the area between the cemento-enamel junction (CEJ) and the alveolar bone crest (ABC) in the 1st molar using Image J software (NIH, National Institute of Health, Bethesda, MD, USA).

In further experiments, the left hemi-maxillae were collected for immunoenzimatic or Real-Time PCR assays. The right hemi-maxillae were histologically processed and three distinct sections of each sample were stained for tartarate resistant acid phosphatase (TRAP, Sigma-Aldrich) for blinded osteoclast counting in the distal alveolar bone adjacent to the 1st molar in 5 consecutive microscopic fields (400x) per section.

Artigo Científico 2

46 Knee joints of mice euthanized for maxillae evaluation were also analyzed. Knee cavity was washed with PBS (2 x 5 µL) and total number of leukocytes was counted in a Neubauer chamber. Periarticular tissues were collected for immunoenzimatic assays. Knee joints were also collected for histological evaluation. Samples were processed and stained for H&E or toluidine blue (TB). Two sections per knee joint were scored in a blinded manner for different parameters (severity of synovial hyperplasia, intensity of inflammatory infiltrate and bone erosion) to obtain an arthritis index (range: 0-8) (Williams et al. 2007). TB stained slides were used to estimate joint proteoglycan content as described previously (Urech et al. 2010).

Quantification of neutrophil and macrophage enzymes

Quantification of the neutrophil enzyme marker myeloperoxidase (MPO) and the macrophage enzyme marker N-acetylglucosaminidase (NAG) was performed as described earlier (Barcelos et al. 2005). MPO and NAG activities in maxillary and knee joint tissues were evaluated by enzymatic reaction, at the absorbance of 450 nm and 405 nm, respectively.

Real-time PCR

The extraction of total RNA from maxillae, inguinal and submandibular lymph nodes was performed with Trizol reagent (Invitrogen, Rockville, MD, USA) and the cDNA synthesis was accomplished as described earlier (Garlet et al. 2006). Extraction of A. actinomycetemcomitans DNA was performed from maxillary tissues with DNA Purification System (Promega Biosciences Inc., San Luis Obispo, CA, USA). Real-time PCR quantitative mRNA or DNA analyzes were performed in a MiniOpticon system (Bio-Rad, Hercules, CA, USA), using Sybr Green PCR Master Mix (Invitrogen), 100 nM specific primers and 2.5 ng cDNA or 5 ng DNA in each reaction. Primer sequences and reaction properties are depicted in Table 1. For mRNA analysis, the relative level of gene expression was calculated with reference to β-actin using the cycle threshold (Ct) method. DNA levels of A. actinomycetemcomitans were determined using the Ct method and normalized by the tissue weight.

Detection of serum anti-collagen I antibodies

Estimation of anti-collagen I IgG in serum of mice was determined as described earlier (Van de Velde et al. 2010, Queiroz-Junior et al. 2011). Ninety-six-well microplates

Artigo Científico 2

47 were incubated with 20 µg/mL solution of murine collagen I (kind gift of Dr. G. T. Kitten, UFMG). Serum samples (1:2-1:100) were incubated, followed by addition of biotinylated goat anti-mouse IgG (Southern Biotechnology, Birmingham, AL, USA) and streptavidin-peroxidase. OPD (o-phenylenediamine dihydrochloride; Sigma- Aldrich) was the substrate buffer and 1 M H2SO4 the stop buffer. Results are expressed as optical density (O.D.) at the absorbance of 492 nm.

Spleen cell cultures

The protocol for stimulation of cell cultures with collagen was adapted from Berg et al. (2000). Whole spleen cells from individual mice were harvested 45 days after PD induction and cultured in triplicate in 96-well microplates at 106 cells/mL in the presence of collagen I 15 µg/well, concavalin (Con A) 2 µg/well or culture medium (RPMI, Flow Laboratories, Irvine, Scotland). Culture supernatants were collected 48 h later and TNF-α levels determined by ELISA.

ELISA

The concentration of interleukin (IL)-1β, IL-6, IL-17, IL-10, interferon (IFN)-γ and tumor necrosis factor (TNF)-α was measured in maxillae, serum and/or splenocytes culture supernatants, using commercially available kits, according to manufacturer instructions (R&D Systems, Minneapolis, MN). The results were expressed as picograms of cytokines (± S.E.M.) normalized for 100 mg tissue or 1 mL of sample.

C-Reactive Protein

Quantification of C reactive protein (CRP) levels was determined in serum samples of mice using a commercially available agglutination kit according to manufacturer instructions (Labtest Diagnóstica, São Paulo, Brazil).

Statistical analysis

Data are presented as mean ± SEM and the statistical significance among control, PD, PD+AIA and treated groups was detected by analyses of variance (ANOVA), followed by Newman-Keuls post hoc analysis. Tests were performed with GraphPad Prism 4.0 software (GraphPad Software Inc., San Diego, CA, USA). Results with P<0.05 were considered statistically significant.

Artigo Científico 2

48 Results

Influence of chronic experimental arthritis on A. actinomycemcomitans-induced periodontal disease

Infection with Aggregatibacter actinomycetemcomitans, referred to here as PD, induced alveolar bone loss (Fig. 1a and 1c), increased number of osteoclasts (Fig. 1b and 1d) and enhanced levels of MPO (Fig. 1e) in maxillae 45 days post-infection. AIA alone also triggered significant signs of PD, i.e. alveolar bone loss (0.78 ± 0.02 mm2; p<0.05), increased number of osteoclasts (1.25 ± 0.09 cells/field; p<0.05) and MPO levels (0.33 ± 0.07 relative units; p<0.05), which were similar to the induced by infection. Concomitant induction of PD+AIA significantly exacerbated alveolar bone loss and increased number of osteoclasts when compared with the PD group (Fig. 1a-d), although maxillary levels of MPO were not influenced (Fig. 1e).

Infectious PD also triggered the expression of cytokines in maxillae, including TNF-α, IFN-γ, IL-17 and IL-10 (Fig. 2a-d). This was similarly observed in mice with AIA alone, except for IL-10 whose expression was decreased (data not shown). Concomitant induction of AIA in mice with PD enhanced levels of IFN-γ (Fig. 2b) in comparison with mice subjected to PD only but expression of TNF-α, IL-17 and IL-10 (Fig. 2a-d) was not affected.

As TNF-α is a key mediator in RA and PD, the role of TNF-α in mediating exacerbated bone loss in animals subjected to PD and AIA was evaluated by using infliximab, a clinically prescribed anti-TNF antibody. Infliximab induced major effects in alveolar bone loss, number of osteoclasts and MPO levels in maxillae (Fig. 1c-e). Moreover, it reduced expression of TNF-α, IFN-γ and IL-17 in maxillae in PD and PD+AIA groups (Fig. 2a-c).

Aggregatibacter actinomycetemcomitans-induced PD does not interfere with AIA In order to evaluate the influence of PD infection on AIA signs, animals infected or not with A. actinomycetemcomitans were subjected to AIA and indices of joint inflammation were assessed. AIA triggered synovial hyperplasia and infiltration of mononuclear cells in the synovium, as scored in the arthritis index (Fig. 3a), proteoglycan loss in joint cartilage (Fig. 3b), increased leukocyte number in synovial fluid (Fig. 3c) and NAG levels in periarticular tissues (Fig. 3d). These signs were similarly increased when comparing groups without and with PD (AIA and PD+AIA,

Artigo Científico 2

49 respectively). A. actinomycemcomitans infection did not induce signs of joint lesion by itself (Fig. 3). Also, no signs of inflammation were observed in the contra-lateral unchallenged joint of mice subjected to AIA or PD+AIA.

In addition, the anti-TNF-α therapy decreased joint inflammation triggered by AIA (i.e., it reduced the arthritis index, proteoglycan loss and cell infiltrate) (Fig. 3).

Dependence of PD and AIA interaction on oral infection

Despite the greater alveolar bone loss (Fig. 1c) and expression of IFN-γ (Fig. 2b) in the group PD+AIA than in the group PD, A. actinomycetemcomitans load in maxillae was similarly increased in both groups (Table 2). Also, A. actinomycetemcomitans load was not affected by the anti-TNF-α therapy (Table 2). Although bacterial load was not influenced by AIA, it was necessary for PD development. Topical oral application of chlorhexidine (CLX) abrogated alveolar bone loss in the groups PD and PD+AIA (PD: 0.84 ± 0.02 mm2 versus PD+CLX: 0.63 ± 0.03 mm2, P<0.05; PD+AIA: 1.01 ± 0.08 mm2 versus PD+AIA+CLX: 0.54 ± 0.04 mm2, P<0.05; mean ± SEM), although it did not influence AIA parameters (data not shown).

T-helper transcription factors in submandibular lymph nodes

In view of the results indicating that AIA exacerbates PD, but not the opposite, a scenario ameliorated by anti-TNF-α therapy, we next investigated the expression of Th transcription factors in PD and AIA interaction. In submandibular lymph nodes, an increased expression of tBET, GATA3, RORγ and Foxp3 mRNA was associated with PD (Fig. 4). Nevertheless, in the group PD+AIA, the expression of tBET was exacerbated, while GATA3 and Foxp3 mRNA levels were reduced in relation to group PD (Fig. 4a- d). Infliximab decreased the expression of tBET (Fig. 4a) in PD+AIA mice, but it did not affect the other transcription factors (Fig. 4b-d). Regarding inguinal lymph nodes, which drain joint region, PD did not affect the expression of any of the evaluated transcription factors, while in the groups AIA and PD+AIA tBET, GATA3, RORγ and Foxp3 mRNA levels were indistinctly increased. Infliximab therapy reversed this polarization pattern of Th transcription factors, triggered by AIA in inguinal lymph nodes (data not shown).

Artigo Científico 2

50 We further investigated whether any soluble factor in the serum could account for the exacerbation of PD by AIA. No significant changes in levels of cytokines, including TNF-α, IL-1β, IL-17 and IL-10, were observed in serum of mice with or without AIA, 45 days post-PD-infection (data not shown). Nevertheless, serum levels of IL-6 were markedly increased in the group PD+AIA and this expression was ablated by infliximab (Fig. 5a). Increased serum C-reactive protein levels were detected in PD group (C: 3.14 ± 0.37 versus PD: 6.82 ± 0.13 mg/l; mean ± SEM, P < 0.05), while in the group PD+AIA these levels were even higher (PD+AIA: 11.61 ± 1.31 mg/l, P<0.05). In addition, AIA induced production of anti-collagen I IgG detected in serum and partially reduced by anti-TNF-α treatment (Fig. 5b). Stimulation of whole spleen cell cultures with collagen I enhanced the expression of TNF-α by cells from group PD+AIA, but not by cells from groups C and PD (Fig. 5c). Without stimulus or with ConA, cell cultures responded similarly (data not shown).

Artigo Científico 2

51 Discussion

Periodontal disease (PD) and rheumatoid arthritis (RA) have remarkably similar pathogenesis and, despite some controversies, emerging evidence suggests that RA might have a direct role in sustaining and aggravating the immune-mediated inflammatory response in established PD and vice versa (Golub et al. 2006, Pablo et al. 2009). Previously, we demonstrated that AIA induces signs of PD through increased systemic reactivity to auto-antigens (Queiroz-Junior et al. 2011). In the current study, we showed that chronic joint AIA exacerbated infection-induced PD in mice, a scenario that was ameliorated by anti-TNF-α therapy, although PD did not influence AIA. This interaction was associated with an altered polarization pattern of Th transcription factors and increased systemic reactivity.

The use of microorganisms to study PD in mice is useful and well established in the literature (Garlet et al. 2006, Garlet et al. 2010, Trombone et al. 2010, Cantley et al 2011). Indeed, in the present study, A. actinomycetemcomitans infection triggered signs of PD, including alveolar bone loss, neutrophil infiltration (as indicated by MPO levels) and release of inflammatory cytokines in maxillae. In view of these results, we