ABSTRACT _Porphyromonas gingivalis_ and _Tannerella forsythia_ have been thought to be associated with periodontitis; however comprehensive histopathological localization of bacteria in

affected human periodontal tissues is not well documented. In the present study, we examined formalin-fixed paraffin-embedded gingival and subgingival granulation tissues from 71 patients

with chronic periodontitis and 11 patients with aggressive periodontitis, using immunohistochemistry with novel monoclonal antibodies specific to _P. gingivalis_ or _T. forsythia_, together

with quantitative real-time polymerase chain reaction for each bacterial DNA. Immunohistochemisty revealed both bacterial species extracellularly, as aggregates or within bacterial plaque,

and tissue-invasiveness of these periodontal bacteria. Detection of these bacteria in the capillary endothelium in some samples suggested possible bacterial translocation into the systemic

circulation from inflamed gingival and subgingival granulation tissues. Immunohistochemistry with the novel antibodies showed high specificity and sensitivity, and can be used to locate

these periodontal bacteria in routinely-used formalin-fixed paraffin-embedded human tissue sections from systemic locations. SIMILAR CONTENT BEING VIEWED BY OTHERS DEFINING _PORPHYROMONAS

GINGIVALIS_ STRAINS ASSOCIATED WITH PERIODONTAL DISEASE Article Open access 14 March 2024 ORAL MICROBIAL EXTRACELLULAR DNA INITIATES PERIODONTITIS THROUGH GINGIVAL DEGRADATION BY

FIBROBLAST-DERIVED CATHEPSIN K IN MICE Article Open access 14 September 2022 COMPARISON OF THREE QPCR-BASED COMMERCIAL TESTS FOR DETECTION OF PERIODONTAL PATHOGENS Article Open access 17

March 2021 INTRODUCTION Periodontitis is a multifactorial inflammatory disease (one factor being bacteria biofilm residing on the teeth) causing destruction of surrounding tooth supporting

tissues and eventually leading to tooth loss1. At the same time periodontitis has been associated with systemic diseases due to increased risk for atherosclerotic cardiovascular diseases2,

adverse pregnancy outcomes3, rheumatoid arthritis4,5, aspiration pneumonia6,7, and cancer8,9. Initial inflammatory reaction occurs as a host response by the human body towards colonizing

periodontal bacteria. Among the six closely related groups of periodontal bacterial species recognised by Socransky and Haffajee10, _Porphyromonas gingivalis_ (PG), _Tannerella forsythia_

(TF), and _Treponema denticola_ are referred to as the “red complex” and are defined as being the species most strongly associated with periodontitis. Several molecular techniques have been

developed over the last few decades that have allowed us to study periodontal bacteria. These techniques include polymerase chain reaction (PCR)-based methods such as conventional

PCR11,12,13, multiplex PCR14,15, and real-time PCR16, as well as hybridization-based techniques such as oligonucleotide probes17,18,19, whole-genomic checkerboard DNA-DNA hybridization10 and

DNA microarray20. Also, next generation sequencing methods have revealed bacterial community composition in health and periodontitis21,22. Despite their accuracy, high sensitivity, and

detection specificity, these methods cannot visualise the location of periodontal bacteria in tissue specimens while preserving the tissue architecture for histopathologic examination. PG

and TF have been implicated as microorganisms that are associated with chronic periodontitis in the 1996 Consensus report on Periodontal Diseases23. The objectives of the present study were

to locate PG and TF using immunohistochemistry (IHC) with novel monoclonal antibodies in formalin-fixed, paraffin-embedded tissue sections of clinically-removed gingival and subgingival

tissues affected by chronic or aggressive periodontitis. Extracellular and intracellular localization of these periodontal bacteria in each of the tissue samples was analysed

histopathologically with regard to the tissue-invasiveness and cell-invasiveness of each bacterium. The bacterial localization detected by IHC was compared with the bacterial density

detected by real-time PCR, as well as with clinical profiles of patients from whom the tissue samples were obtained. RESULTS ANTIBODY SPECIFICITY The specificity of the novel monoclonal

antibodies is shown in Table 1. By IHC, the anti-PG and anti-TF antibodies gave positive reactions in the Kupffer cells in PG_-_ and TF-infected rat liver sections, respectively, and did not

cross-react with other bacterial species. The specificity results obtained by IHC were the same when the reactivity was checked using western blot analysis of whole cell bacterial lysates.

Western blot analysis with each of the anti-PG and anti-TF antibodies showed a ladder pattern of positive bands ranging from 31 to 76 kiloDaltons (kDa) and from 38 to 225 kDa, respectively.

There was no cross-reactivity and no background-reactivity with other examined bacterial species, or with the PBS control (Fig. 1). HISTOLOGIC LOCALIZATION OF THE BACTERIA IHC revealed both

bacterial species extracellularly as aggregates or within bacterial plaque and intracellularly in stromal inflammatory cells, squamous epithelium, and capillary endothelium of granulation

tissue (Figs 2 and 3). TF and PG cells, when detected together extracellularly, were intermixed within bacterial plaques, mostly with a predominant number of TF cells (Fig. 2a), and a

predominant distribution of TF cells in the central core and PG cells in the marginal area (Fig. 3h). Bacterial plaques comprising only TF cells were occasionally observed, but bacterial

plaques comprising only PG cells were seldom observed (Fig. 3g). Focal clusters of squamous epithelial cells with one or a few bacterial cells intracellularly or intercellularly between

desquamative epithelial cells were occasionally observed in the erosive epidermal layer adjacent to superficial bacterial aggregates (Fig. 3e), and sometimes found in deeper epidermal layers

(Fig. 2b). Immunofluorescence double-staining signals for PG or TF cells overlapped in the area of the epithelial cytoplasm, which was detected using the anti-cytokeratin antibody (Fig. 4).

PG and TF cells were occasionally clustered in the area of inflammatory cell infiltration and localized in some of the inflammatory cells, which morphologically appeared to be macrophages

(Figs 2d and 3f). Intracellular PG or TF cells were rarely detected in the capillary walls of granulation tissue (Fig. 2c). The intracellular bacterial localization in the capillary

endothelium was confirmed by IHC with anti-human CD31 antibody (Fig. 3a–c). DETECTION FREQUENCY OF THE BACTERIA IN EACH LOCATION The extracellular detection and intracellular detection

frequencies (in squamous epithelium, stromal inflammatory cell, or capillary endothelium) of PG and TF are shown in Table 2, including the detection frequency by PCR. The extracellular

detection frequency by IHC was most similar to the results by PCR. The detection frequency did not differ significantly between chronic and aggressive periodontitis for any location

examined. PG and TF were only detected intracellularly in samples that also had extracellular detection of the same species. The frequency of the extracellular and intracellular detection

was generally higher for TF than PG. Extracellular PG was only detected in the presence of extracellular TF in 55 (67%) samples, whereas extracellular TF was detected even in the absence of

extracellular PG in 15 (18%) samples. Similarly, intracellular PG was only detected in the presence of intracellular TF in 23 (28%) samples, whereas intracellular TF was detected even in the

absence of intracellular PG in 20 (24%) samples. Samples without either species detected extracellularly and intracellularly were found in 12 (15%) and 39 (48%) samples, respectively.

Intracellular TF and PG in the capillary endothelium was detected in 8 (19%) and 4 (17%) of the 43 and 23 samples with intracellular TF and PG detected, respectively. The endothelial

localization of TF was detected in all samples with endothelial PG. All samples with PG or TF detected in the capillary endothelium were accompanied by the same species detected in the

stromal inflammatory cells. ASSOCIATION BETWEEN LOCALIZATION AND DENSITY OF THE BACTERIA The number of genome copies for each bacterial species was compared between samples with positive and

negative IHC results (Fig. 5). Detection status of the bacterium by IHC and PCR was highly associated for both species (_P_ < 0.0001). Positive IHC results of PG and TF were obtained

from 52/59 (88%) and 69/75 (92%) of samples with more than one genome copy detected by PCR for the same species, respectively. PCR for TF and PG was negative for one sample and three samples

that were IHC-positive for the same species, respectively. In both species, the bacteria number (bacterial density) ranged from 0 to 100 (mean ± SD: PG, 6 ± 17 and TF: 14 ± 22) genome

copies in the IHC-negative samples and from 10 to 27,000 (mean ± SD: PG, 1577 ± 3428 and TF: 2997 ± 5618) genome copies in the IHC-positive samples. Results from IHC detection of

extracellular or intracellular localization in either stromal inflammatory cells, squamous epithelium, or capillary endothelium were analysed by comparing with bacterial density of the

samples (Table 3). TF density, when detected extracellularly by IHC, was approximately nine times higher in samples with extracellular PG than in those without extracellular PG (_P_ = 

0.0009). Intracellular detection of PG in stromal inflammatory cells was associated with a higher PG density (_P_ = 0.037). Intracellular detection of TF in epithelial cells and in stromal

inflammatory cells was associated with the higher TF density (_P_ = 0.0021 and 0.0010) followed by a higher PG density (_P_ = 0.0031 and 0.013), respectively. ASSOCIATION BETWEEN BACTERIAL

LOCALIZATION AND CLINICAL PROFILES The results obtained from IHC were analysed based on the clinical profiles of the patients (Table 3). Patients with both bacteria detected extracellularly

by IHC were older than patients without both bacteria (_P_ = 0.0095). The other bacterial localization patterns were not associated with probing pocket depth, age, or sex of the patients.

The IHC and PCR results did not differ significantly between chronic and aggressive periodontitis. DISCUSSION This study describes the first histologic localization of PG and TF in the

gingival and subgingival tissues affected by chronic or aggressive periodontitis. These periodontal pathogens were located extracellularly on the epithelial surface as aggregates or within

bacterial plaques, and intracellularly in the stromal inflammatory cells, squamous epithelium, and capillary endothelium of the granulation tissues. IHC with the novel anti-PG and anti-TF

antibodies showed that the antibodies were highly specific and sensitive for detecting the _in situ_ localization of these bacteria in formalin-fixed paraffin-embedded tissue sections, and

can therefore be generally used for histopathologic examination of the presence of these periodontal bacteria in other systemic organs. The production of species-specific monoclonal

antibodies to detect TF and PG cells in routinely-used formalin-fixed paraffin-embedded tissue sections was successful in the present study with the screening of hybridoma clones by IHC with

formalin-fixed paraffin-embedded tissue sections of the liver from rats infected with these target bacteria. The specificity of the antibodies was confirmed by IHC with liver tissue

sections of rats infected with other control bacteria as well as confirmed by western blot analysis with bacterial lysates. These species-specific antibodies did not cross-react with other

bacterial species within the same order of Bacteroidales (Supplementary Table S1) when used for IHC (Supplementary Fig. S1) and western blot. Species-specific reactivity of the anti-PG

antibody was also confirmed by demonstrating no cross-reactivity between PG and _Porphyromonas levii_ within the same genus of _Porphyromonas_ (Supplementary Fig. S1 and S2). A ladder

pattern of positive bands by each of the novel antibodies on western blots suggested that bacterial lipopolysaccharides (LPS) are target antigens of these species-specific antibodies.

Indeed, the anti-PG antibody reacted with commercially-available LPS purified from bacterial lysate of PG, showing a ladder pattern of positive bands (Supplementary Fig. S3). LPS, which

locates in the outer leaflet of the outer membrane of Gram-negative bacteria, is composed of three parts: lipid A, core sugars, and O-antigen. The O-antigen is a polysaccharide, and the

O-specific chain consists of up to 50 repeating oligosaccharide units formed of 2 to 8 monosaccharide components in a highly species-specific manner24. Further studies are needed to identify

the epitope recognized by each of the novel antibodies. We also attempted to produce an antibody specific to _T. denticola_ for use in this study, but our attempts were unsuccessful.

Comparison between the results from IHC and PCR supported the sensitivity of IHC with the anti-PG and anti-TF antibodies. Positive IHC results of PG and TF were obtained from 52/59 (88%) and

69/75 (92%) samples with more than one genome copy detected by PCR for the same species, respectively. The fact that no genome copy was detected by PCR in a few samples with positive IHC

results requires some discussion because PCR detection sensitivity is believed to be higher than that of IHC detection. Samples with a few signals (presumably less than 10 signals) detected

by IHC may be positive in one tissue section but negative in the other semi-serial sections by IHC due to an uneven distribution of the bacterium across the tissue samples. IHC of gingival

and subgingival tissue samples revealed extracellular PG only when TF was present, whereas TF was detected even in the absence of PG, which is consistent with the results of a previous study

using baseline subgingival plaque samples25. Moreover, the TF tissue density, when detected by IHC, was remarkably higher in samples with extracellular PG compared to those without

extracellular PG. These results suggest that increased levels of extracellular TF produces an environment that promotes the colonization and proliferation of PG, which is consistent with the

finding that TF cell extracts stimulate PG growth26. The lack of bacterial plaques comprising only PG cells, and the predominant distribution of TF cells in the central core and PG cells in

the marginal area of many bacterial plaques, also suggested a PG growth-supporting property of TF. Of particular note, however, even in many samples with extracellular co-localization of PG

and TF, intracellular PG was only found in the presence of intracellular TF. On the other hand, intracellular TF was found even in the absence of intracellular PG. This suggests that

cell-invasiveness (represented by invasion into squamous epithelium) or tissue-invasiveness (represented by tissue stromal inflammatory cells containing bacteria) of PG may also be supported

by the cell-invasiveness or tissue-invasiveness of TF. Involvement of TF is directly implicated in the development of periodontal disease27,28. Indeed, both extracellular and intracellular

detection by IHC was more frequent with TF than with PG, and the bacterial density of tissue sections, when detected by IHC, was higher for TF than PG. The concentration of pathogenic

bacteria within subgingival plaque may need to surpass a particular threshold in order to induce disease; identifying this threshold may allow the progression of periodontal disease to be

predicted29,30. In the present study, higher density of TF and PG was associated with detection of the same species in epithelial cells and stromal cells, suggesting that the bacterial

density is a major factor responsible for cell-invasiveness and tissue-invasiveness of these periodontal bacteria. Cell-invasiveness of these periodontal bacteria is associated with cell

molecules; such as major fimbriae protein (FimA)31 and gingipain32 of PG, and leucine-rich surface protein (BspA)27 and surface-layer proteins33 of TF. Gingipain also damages host cells and

connective tissues, which in turn induces apoptosis32 and further enhances the ongoing inflammatory reaction leading to periodontal tissue destruction and ulceration in the gingival

epithelium (tissue-invasiveness). Otherwise, PG suppresses apoptotic cell death, for intracellular survival, by modulating major apoptotic pathways in gingival epithelial cells34. Although

TF is clinically linked with disease, the virulence mechanisms of this organism remain unclear. One possible mechanism of virulence that has been suggested is that TF invades epithelial

cells35,36,37. The ability of TF to attach to and invade epithelial cells is promoted by PG and the outer membrane vesicles produced by PG27. PG is able to invade the deeper structures of

connective tissues via a paracellular pathway by degrading epithelial cell-cell junction complexes38. Indeed, we observed the colocalization of PG and TF in the intercellular spaces between

desquamative squamous epithelial cells. Consequently, the disrupted epithelial cell barrier caused by intracellular TF infection (cell-invasiveness), supported by intercellular

co-localization with PG and other unknown factors, permits extracellular TF and PG invasion into tissue stromal space (tissue-invasiveness). This results in tissue inflammation accompanied

by many stromal inflammatory cells that have ingested TF and PG. Since only one section for each sample was used for IHC detection analysis, the results may be biased especially for

intracellular positivity. Nevertheless, samples with any combination of intracellular bacterium detected in stromal inflammatory cells, squamous epithelium, or capillary endothelium of the

granulation tissue were always accompanied by extracellular detection of that bacterium. Among them, intracellular detection of bacteria in the capillary endothelium of granulation tissues

was always accompanied by intracellular detection of the same species in stromal inflammatory cells. This suggests that the tissue-invading bacteria, many of which are phagocytosed by

stromal inflammatory cells, can invade into capillary vessels. IHC-based detection of intracellular bacteria in the capillary endothelium suggested bacterial translocation via the

bloodstream. It is generally believed that gingival ulceration in periodontal pockets enables the translocation of bacteria into the systemic circulation, causing bacteremia in patients with

periodontitis, which may be an atherogenic stimulus39. Tissue-invading periodontal bacteria may translocate to lymph nodes via the lymphatic drainage of the oral cavity40, similarly to

_Helicobacter pylori_ translocating into gastric lymph nodes after invading the gastric mucosa41. Finally, IHC, with the novel antibodies prepared for the present study, allows

histopathologic examination to locate these periodontal bacteria in routinely-used formalin-fixed paraffin-embedded tissue sections from surgically-removed specimens to investigate possible

translocation of periodontal bacteria to other systemic organs. METHODS SAMPLING A total of 82 samples from 82 patients who attended the Periodontics Clinic, Dental Hospital, Faculty of

Dentistry, Tokyo Medical and Dental University, were enrolled in the study. The clinical profiles of the patients are shown in Supplementary Table S2 online, and the diagnosis of periodontal

disease was made according to the 1999 classification system for periodontal diseases produced by a workshop for clinical periodontitis42,43. The mean age was 58 years (SD, 11.9). All

patients underwent one round of initial periodontal therapy prior to the surgery. Written informed consent was obtained from all patients after explanation of the purpose of the study. The

Ethics Committee of Tokyo Medical and Dental University approved the study (reference 857). All methods were performed in accordance with the relevant guidelines and regulations. The samples

obtained included the deepest non-healing pocket of each tooth with periodontitis, including gingival tissues and subgingival granulation tissues that were curetted following internal bevel

incision, during open-flap debridement, EmdogainTM treatment, guided tissue regeneration therapy, apically positioned flap, crown lengthening, or connective tissue graft. Each tissue sample

was immediately fixed in 10% buffered formalin for 4 h and then processed using methods for preparing tissues for routine pathologic examination. BACTERIA CULTURE Four strains of PG (ATCC

53978, ATCC 33277, BAA-1703, and A7A128) and four strains of TF (ATCC 43037, and 3 clinical isolates) were cultured on Tripticase soy agar44,45 plates containing defibrinated horse blood,

hemin, and vitamin-K. For TF strain cultures, 15 μl of 20 mg/ml N-acetyl muramic acid solution containing 100 μm sterilised paper disc per Tripticase soy agar plate was added.

_Aggregatibacter actinomycetemcomitans_, _Fusobacterium nucleatum_, _Prevotella intermedia_, _Prevotella nigrescens_, _Bacteroides fragilis_, and _Bacteroides vulgatus_ were cultured on

Anaero Columbia Agar with Rabbit Blood (BD Diagnostic Systems). _Propionibacterium acnes_ was cultured on Gifu Anaerobic Medium plates (Nissui Pharm. Co., Ltd.). The above bacterial species

were cultured under anaerobic conditions at 35 °C. _Escherichia coli_, _Enterococcus faecalis_, _Streptococcus pneumoniae_, _Pseudomonas aeruginosa_, _Staphylococcus aureus_, and _Klebsiella

pneumoniae_ were cultured on sheep blood agar M58 (Eiken Chemical Co., Ltd.). _Streptococcus pyogenes_ was cultured on Todd-Hewitt agar plates (BD) supplemented with 0.2% yeast extract.

_Legionella pneumophila_ was cultured on buffered charcoal-yeast extract agar plates (BD). _Mycobacterium tuberculosis_ was cultured on Middlebrook 7H11 agar plates (BD). The above bacterial

species were cultured under aerobic conditions at 35 °C. _Haemophilus influenza_ was cultured on Chocolate II agar plates (BD) under 5% CO2 conditions. MONOCLONAL ANTIBODY PRODUCTION

Monoclonal antibodies were generated according to the previously described protocol46. Briefly, sonicated whole-cell bacterial lysates of PG or TF were used as an immunogen. Hybridoma cell

lines were checked with an enzyme-linked immunosorbent assay using bacterial antigens as an immunogen, and screened by IHC with formalin-fixed and paraffin-embedded tissue sections of PG_-_

or TF_-_infected rat liver (infected by intravenously injecting 30 mg of heat-killed bacteria into female Sprague–Dawley rats [CLEA Japan] 1 h before killing the rat). Similarly-prepared

liver tissue sections of rats injected by each strain of other control bacteria were also examined to confirm the specificity to PG or TF by enzyme IHC with culture supernatant from each

hybridoma clone, according to the methods described hereafter. After selecting the hybridomas that produced a strong reaction specific to each bacterium followed by two rounds of cloning

with limiting dilution, a single hybridoma clone was implanted in the intraperitoneal space of severe combined immunodeficiency mice, and ascites was collected and used without further

purification. The monoclonal antibodies against PG and TF were named anti-PG antibody and anti-TF antibody, respectively. WESTERN BLOT ANALYSIS The specificity of the anti-PG and anti-TF

antibodies was examined by western blot analysis with four strains of PG, four strains of TF, and many strains of other control bacteria (Table 1). Western blot analysis was performed with

whole-cell lysate and commercially-available LPS purified from PG (LPS-PG Ultrapure, InvivoGen, San Diego, CA). Whole-cell lysate of each bacteria were prepared by sonication on ice with an

ultrasonic homogenizer (VP-5S ULTRA S homogenizer; Taitec, Koshigaya, Japan) twice at level 6 for 1 min with 100 mg/ml of each bacterial species in phosphate-buffered saline (PBS). Each

sample was separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis using 12% polyacrylamide gel. All samples were transferred electrically to a polyvinylidene difluoride

membrane with Mini Trans-Blot cell (Bio-Rad Laboratories). Membranes were blocked overnight at 41 °C with Block Ace (DS Pharma Biomedical Co. Ltd.). The membranes were then incubated for 90 

min at room temperature with the appropriately diluted antibodies (1:1000 for both anti-TF and anti-PG antibodies). Membranes were incubated for 30 min with biotinylated rabbit anti-mouse

immunoglobulins (DAKO) at a 1:2000 dilution ratio. The membranes were then incubated with FluoroLinkTMCyTM3-labelled streptavidin (GE Healthcare) for 30 min in the dark with a 1:3000

dilution ratio. Before and after each of the last three steps, the membranes were washed in PBS containing 0.5% Tween-20. Finally, the membranes were washed in PBS before drying them in the

dark for 1 h at 37 °C. The dried membranes were viewed using a Bio-Rad BPC-500, Molecular Imager (Quantity One-Analysis Software Version 4.6.9, Bio-Rad). The specificity of both antibodies

was evaluated with and without the primary antibody to rule out cross-reactions of whole cell lysates of bacteria with the biotinylated rabbit anti-mouse immunoglobulins. ENZYME

IMMUNOHISTOCHEMISTRY The enzyme IHC method used with the anti-PG and anti-TF antibodies was the same as described previously46. Histologic sections (4 µm-thick) were cut from formalin-fixed

and paraffin-embedded rat liver tissues, and human gingival and subgingival tissue samples, and mounted on silane-coated slides (Muto Pure Chemicals). After the sections were de-paraffinized

and rehydrated, they were microwaved (Microwave Processor H2850; Energy Beam Sciences, East Granby, CT, USA) in 10 mmol/l citrate buffer (pH 6.0) for 40 min at 97 °C. The sections were then

treated with 3% hydrogen peroxide in methanol for 10 min. The sections were first incubated with normal horse serum (Vectastain Universal Elite ABC Kit; Vector Laboratories) and then

incubated overnight at room temperature with one of the appropriately diluted antibodies (anti-PG antibody 1:10,000 or anti-TF antibody 1:128,000) in a humidified chamber. The sections were

then incubated for 30 min with biotinylated secondary antibody, followed by 30-min incubation with streptavidin–peroxidase complex (Vectastain Universal Elite ABC Kit), both at room

temperature. Before and after each step, the sections were washed in PBS containing 0.5% Tween-20. The signal was developed as a brown reaction product using peroxidase substrate

diaminobenzidine (Histofine Simplestain DAB Solution; Nichirei Biosciences Inc.). The staining procedure for IHC with the anti-human CD31 monoclonal antibody (M0823, DAKO) for endothelial

cells diluted 1:20 was basically the same except that the microwaving was performed in 10 mmol/l citrate buffer (pH 6.0) for 5 min at 97 °C after deparaffinizing and rehydrating the

sections. All specimens were counterstained with Mayer’s hematoxylin. Adjacent histologic sections were also examined with hematoxylin and eosin staining. Since no background signals were

produced by the PBS control without primary antibodies in the gingival tissues (Supplementary Fig. S4), the IHC results were considered positive for a single tissue section from each sample

when any positive signals were identified in each tissue section or in each type of cell; including squamous epithelium, stromal inflammatory cells, and capillary endothelium. Stained

sections were observed and photographed using an OLYMPUS BX-51 research microscope (DP2-BSW application software; Shinjuku, Tokyo, Japan). HISTOLOGIC EVALUATION Bacterial aggregates are

recognised as clusters of individual bacterial cells on epithelial surfaces and bacterial plaque is recognised as a mass of dense bacterial clusters in free spaces beyond the epithelial

surface. PG and TF were considered to be extracellularly localized when at least one positive signal was detected by the anti-PG or anti-TF antibody in the bacterial aggregates or plaques,

and intracellularly localized when at least one positive signal was detected by the anti-PG or TF antibody in the cytoplasm of the squamous epithelium or submucosal stromal cells, using

serial hematoxylin and eosin-stained sections. We similarly evaluated the intracellular location of PG and TF in the capillary endothelium using serial sections immunostained with anti-human

CD31 monoclonal antibody. Three independent observers (MN, TI, and DK) evaluated the histopathologic findings and IHC results. In case of any disagreements between the three observers, the

observers re-evaluated the specimens and reached a consensus after discussion. IMMUNOENZYME DOUBLE-STAINING For simultaneous identification of PG and TF in the same tissue section,

immunoenzyme double-staining was performed. According to the method described previously47, sections were first reacted with the anti-PG antibody using the avidin-biotin-complex method with

the VECTASTAIN UNIVERSAL Elite ABC Kit (PK-7200, VECTOR) and the Histofine Simple stain DAB Solution (415174, Nichirei Biosciences Inc.), and microwaved again for 20 min. After these

procedures, the sections underwent the secondary reaction with the anti-TF antibody, followed by IHC using a polymer method with Histofine Simple stain AP (M) (414241, Nichirei Biosciences

Inc.) and Vector Blue Alkaline Phosphatase Substrate Kit III (SK-5300, Vector). IMMUNOFLUORESCENCE DOUBLE-STAINING Immunofluorescence double-staining was performed to confirm the

intracytoplasmic localization of PG or TF in epithelial cells. According to the method described previously41, tissue sections were first reacted with the anti-PG or anti-TF antibody using

the avidin-biotin-complex method with fluorescein isothiocyanate-conjugated streptavidin (DAKO), and microwaved again for 20 min. After these procedures, the sections underwent the secondary

reaction with anti-human cytokeratin AE1/AE3 antibody (clones:AE1/AE3; DAKO) diluted 1:50, followed by anti-mouse immunoglobulin conjugated with tetra-methylrhodamine isothiocyanate (DAKO).

Stained sections were observed and photographed using a FLUOVIEW FV1000-D confocal laser scanning microscope (FV10-ASW1.6 software version; Olympus Co.). Samples showing intracellular

positivity were used to acquire consecutive cross-sectional images (XYZ) using the line sequential scanning version of the same microscope. The acquired images were used to produce images in

the XY plane, and to confirm the intracellular localization of bacteria, orthogonal projections were produced in the XZ and YZ planes. REAL-TIME PCR Real-time PCR was performed according to

the methods described previously40. Three serial sections (3 μm-thick) adjacent to the sections used for IHC were sliced from formalin-fixed and paraffin-embedded tissue blocks and placed

in sterilised 1.5-ml centrifuge tubes. We used a different microtome blade for each sample to avoid sample-to-sample contamination. The DNA of each section was extracted with a QIAamp for

DNA Mini Kit (Qiagen) according to the tissue protocol of the manufacturer. Real-time PCR was performed to amplify fragments of 16S ribosomal RNA of PG and TF. The primers used were PG-F,

5′-GTTCAGCCTGCCGTTG AAAC-3′ and PG-R, 5′-ACCGCTACACCACGAATTCC-3′, and TF-F, 5′-CGACGGAGA GTGAGAGCTTTC T-3′ and TF-R, 5′-GCGCTCGTTATGGCACTTAAG-3′. TaqMan probes were PG-P, 5′-

CTTGAGTTCAGCGGCGGCAGG-3′ and TF-P, 5′-CGTCTATGTA GGTGCTGCATGGTTGTCG-3′. The probe was labelled with 6-carboxyfluorescein on the 5′-end and 6-carboxytetramethylrhodamine (TAMRA) on the

3′-end. Real-time PCR mixtures contained 5 µl of template DNA, 100 nM of each primer, 40 nM of the probe, and Absolute QPCR ROX (500 nM) Mix (ABgene) in a total volume of 50 µl.

Amplification and detection were performed using a detection system (ABI PRIZM Sequence Detection System; Applied Biosystems) under conditions of 95 °C for 5 min, followed by 50 cycles of 95

 °C for 15s and 60 °C for 1 min. The amount of bacterial DNA in each sample was estimated from internal standard samples of serially diluted bacterial DNA (10 ng, 1 ng, 100 pg, and 10 pg)

and expressed in terms of the number of bacterial genomes, with 2.5 × 109 Da per genome used in the conversion48. Based on the conversion, 10 ng of bacterial DNA equals 2,000,000 copies of

16S rRNA genes. Finally, the results of the quantitative PCR for each sample are expressed as the number of bacterial genomes in the sample. Negative controls without bacterial DNA were

included in every PCR; background values were always less than one genome copy. Samples with one or more bacterial genome copy were considered to have positive results. STATISTICAL ANALYSIS

All analyses were performed using GraphPAD PRISM ver. 6 (GraphPad Software, Inc.). The association of the IHC detection frequency of the bacterium between chronic and aggressive

periodontitis, and the association between the detection status of the bacterium by IHC and PCR were analysed using Fisher’s exact test. The Mann–Whitney U test was used to compare the

number of genome copies for each bacterial species between samples with both PG and TF, and only TF detected by IHC in each location. Associations of the number of genomic copies for each

bacterial species with clinical profiles such as patients’ sex or pocket-probing depth were analysed with the Mann-Whitney U test. In both tests, a _P_-value of less than 0.05 was considered

statistically significant. REFERENCES * Löe, H., Anerud, A., Boysen, H. & Morrison, E. Natural history of periodontal disease in man. Rapid, moderate and no loss of attachment in Sri

Lankan laborers 14 to 46 years of age. _J. Clin. Periodontol._ 13, 431–45 (1986). Article  PubMed  Google Scholar  * Haraszthy, V. I., Zambon, J. J., Trevisan, M., Zeid, M. & Genco, R.

J. Identification of Periodontal Pathogens in Atheromatous Plaques. _J. Periodontol._ 71, 1554–1560 (2000). Article  PubMed  CAS  Google Scholar  * Lin, D., Moss, K., Beck, J. D., Hefti, A.

& Offenbacher, S. Persistently high levels of periodontal pathogens associated with preterm pregnancy outcome. _J. Periodontol._ 78, 833–41 (2007). Article  PubMed  Google Scholar  *

Kaur, S., White, S. & Bartold, P. M. Periodontal Disease and Rheumatoid Arthritis. _J. Dent. Res._ 92, 399–408 (2013). Article  PubMed  CAS  Google Scholar  * Pischon, N. _et al_.

Association Among Rheumatoid Arthritis, Oral Hygiene, and Periodontitis. _J. Periodontol._ 79, 979–986 (2008). Article  PubMed  CAS  Google Scholar  * Tan, L., Wang, H., Li, C. & Pan, Y.

16S rDNA-based metagenomic analysis of dental plaque and lung bacteria in patients with severe acute exacerbations of chronic obstructive pulmonary disease. _J. Periodontal Res._ 49, 760–9

(2014). Article  PubMed  CAS  Google Scholar  * Fourrier, F., Duvivier, B., Boutigny, H., Roussel-Delvallez, M. & Chopin, C. Colonization of dental plaque: a source of nosocomial

infections in intensive care unit patients. _Crit. Care Med._ 26, 301–8 (1998). Article  PubMed  CAS  Google Scholar  * Abnet, C. C. _et al_. Tooth loss is associated with increased risk of

gastric non-cardia adenocarcinoma in a cohort of Finnish smokers. _Scand. J. Gastroenterol._ 40, 681–7 (2005). Article  PubMed  Google Scholar  * Hiraki, A., Matsuo, K., Suzuki, T., Kawase,

T. & Tajima, K. Teeth loss and risk of cancer at 14 common sites in Japanese. _Cancer Epidemiol. Biomarkers Prev._ 17, 1222–7 (2008). Article  PubMed  Google Scholar  * Socransky, S. S.,

Haffajee, A. D., Cugini, M. A., Smith, C. & Kent, R. L. Microbial complexes in subgingival plaque. _J. Clin. Periodontol._ 25, 134–44 (1998). Article  PubMed  CAS  Google Scholar  *

Ashimoto, A., Chen, C., Bakker, I. & Slots, J. Polymerase chain reaction detection of 8 putative periodontal pathogens in subgingival plaque of gingivitis and advanced periodontitis

lesions. _Oral Microbiol. Immunol._ 11, 266–73 (1996). Article  PubMed  CAS  Google Scholar  * Slots, J., Ashimoto, A., Flynn, M. J., Li, G. & Chen, C. Detection of putative periodontal

pathogens in subgingival specimens by 16S ribosomal DNA amplification with the polymerase chain reaction. _Clin. Infect. Dis._ 20(2), S304–7 (1995). Article  PubMed  CAS  Google Scholar  *

Umeda, M. _et al_. The distribution of periodontopathic bacteria among Japanese children and their parents. _J. Periodontal Res._ 39, 398–404 (2004). Article  PubMed  Google Scholar  * Seol,

J. H., Cho, B. H., Chung, C. P. & Bae, K. S. Multiplex polymerase chain reaction detection of black-pigmented bacteria in infections of endodontic origin. _J. Endod._ 32, 110–114

(2006). Article  PubMed  Google Scholar  * Tran, S. D. & Rudney, J. D. Multiplex PCR Using Conserved and Species-Specific 16S rRNA Gene Primers for Simultaneous Detection of

_Actinobacillus actinomycetemcomitans_ and _Porphyromonas gingivalis_. _J. Clin. Microbiol._ 34, 2674–2678 (1996). PubMed  PubMed Central  CAS  Google Scholar  * Suzuki, N., Yoshida, A.

& Nakano, Y. Quantitative analysis of multi-species oral biofilms by TaqMan Real-Time PCR. _Clin Med Res_ 3, 176–185 (2005). Article  PubMed  PubMed Central  CAS  Google Scholar  * Dix,

K. _et al_. Species-specific oligodeoxynucleotide probes for the identification of periodontal bacteria. _J. Clin. Microbiol._ 28, 319–23 (1990). PubMed  PubMed Central  CAS  Google Scholar

  * Tsai, C. Y., Wolff, L. F., Germaine, G. & Hodges, J. A rapid DNA probe test compared to culture methods for identification of subgingival plaque bacteria. _J. Clin. Periodontol._ 30,

57–62 (2003). Article  PubMed  CAS  Google Scholar  * Moncla, B. J., Braham, P., Dix, K., Watanabe, S. & Schwartz, D. Use of synthetic oligonucleotide DNA probes for the identification

of _Bacteroides gingivalis_. _J. Clin. Microbiol._ 28, 324–7 (1990). PubMed  PubMed Central  CAS  Google Scholar  * Paster, B. J., Olsen, I., Aas, J. A. & Dewhirst, F. E. The breadth of

bacterial diversity in the human periodontal pocket and other oral sites. _Periodontol. 2000_ 42, 80–87 (2006). Article  PubMed  Google Scholar  * Griffen, A. L. _et al_. Distinct and

complex bacterial profiles in human periodontitis and health revealed by 16S pyrosequencing. _ISME J._ 6, 1176–85 (2012). Article  PubMed  CAS  Google Scholar  * Abusleme, L. _et al_. The

subgingival microbiome in health and periodontitis and its relationship with community biomass and inflammation. _ISME J._ 7, 1016–25 (2013). Article  PubMed  PubMed Central  CAS  Google

Scholar  * Consensus report. Periodontal diseases: pathogenesis and microbial factors. _Ann. Periodontol._ 1, 926–932 (1996). Article  Google Scholar  * Matsuura, M. Structural Modifications

of Bacterial Lipopolysaccharide that Facilitate Gram-Negative Bacteria Evasion of Host Innate Immunity. _Front. Immunol._ 4, 109 (2013). Article  PubMed  PubMed Central  CAS  Google Scholar

  * Byrne, S. J. _et al_. Progression of chronic periodontitis can be predicted by the levels of _Porphyromonas gingivalis_ and _Treponema denticola_ in subgingival plaque. _Oral Microbiol.

Immunol._ 24, 469–477 (2009). Article  PubMed  CAS  Google Scholar  * Yoneda, M. _et al_. Stimulation of growth of _Porphyromonas gingivalis_ by cell extracts from _Tannerella forsythia_.

_J. Periodontal Res._ 40, 105–109 (2005). Article  PubMed  Google Scholar  * Inagaki, S., Onishi, S., Kuramitsu, H. K. & Sharma, A. _Porphyromonas gingivalis_ vesicles enhance

attachment, and the leucine-rich repeat BspA protein is required for invasion of epithelial cells by ‘_Tannerella forsythia_’. _Infect. Immun._ 74, 5023–5028 (2006). Article  PubMed  PubMed

Central  CAS  Google Scholar  * Tran, S. D., Rudney, J. D., Sparks, B. S. & Hodges, J. S. Persistent Presence of _Bacteroides forsythus_ as a Risk Factor for Attachment Loss in a

Population With Low Prevalence and Severity of Adult Periodontitis. _J. Periodontol._ 72, 1–10 (2001). Article  PubMed  CAS  Google Scholar  * Brown, L. F., Beck, J. D. & Rozier, R. G.

Incidence of attachment loss in community-dwelling older adults. _J. Periodontol._ 65, 316–323 (1994). Article  PubMed  CAS  Google Scholar  * Papapanou, P. N. _et al_. Subgingival

microbiota in adult Chinese: prevalence and relation to periodontal disease progression. _J.Periodontol._ 68, 651–666 (1997). Article  PubMed  CAS  Google Scholar  * Yilmaz, O., Watanabe, K.

& Lamont, R. J. Involvement of integrins in fimbriae-mediated binding and invasion by _Porphyromonas gingivalis_. _Cell. Microbiol._ 4, 305–14 (2002). Article  PubMed  CAS  Google

Scholar  * Sheets, S. M., Robles-Price, A. G., McKenzie, R. M. E., Casiano, C. A. & Fletcher, H. M. Gingipain-dependent interactions with the host are important for survival of

_Porphyromonas gingivalis_. _Front. Biosci._ 13, 3215–38 (2008). Article  PubMed  PubMed Central  CAS  Google Scholar  * Sakakibara, J. _et al_. Loss of adherence ability to human gingival

epithelial cells in S-layer protein-deficient mutants of _Tannerella forsythensis_. _Microbiology_ 153, 866–76 (2007). Article  PubMed  CAS  Google Scholar  * Yilmaz, O., Jungas, T.,

Verbeke, P. & Ojcius, D. M. Activation of the phosphatidylinositol 3-kinase/Akt pathway contributes to survival of primary epithelial cells infected with the periodontal pathogen

_Porphyromonas gingivalis_. _Infect. Immun._ 72, 3743–51 (2004). Article  PubMed  PubMed Central  CAS  Google Scholar  * Conrads, G., Gharbia, S. E., Gulabivala, K., Lampert, F. & Shah,

H. N. The use of a 16s rDNA directed PCR for the detection of endodontopathogenic bacteria. _J. Endod._ 23, 433–8 (1997). Article  PubMed  CAS  Google Scholar  * Khader, Y. S. & Ta’ani,

Q. Periodontal Diseases and the Risk of Preterm Birth and Low Birth Weight: A Meta-Analysis. _J. Periodontol._ 76, 161–165 (2005). Article  PubMed  Google Scholar  * Kinane, D. F. Causation

and pathogenesis of periodontal disease. _Periodontol. 2000_ 25, 8–20 (2001). Article  PubMed  CAS  Google Scholar  * Katz, J., Sambandam, V., Wu, J. H., Michalek, S. M. & Balkovetz, D.

F. Characterization of _Porphyromonas gingivalis_-induced degradation of epithelial cell junctional complexes. _Infect. Immun._ 68, 1441–9 (2000). Article  PubMed  PubMed Central  CAS 

Google Scholar  * Genco, R. J. & Van Dyke, T. E. Prevention: Reducing the risk of CVD in patients with periodontitis. _Nat. Rev. Cardiol._ 7, 479–80 (2010). Article  PubMed  Google

Scholar  * Amodini Rajakaruna, G. _et al_. Possible translocation of periodontal pathogens into the lymph nodes draining the oral cavity. _J. Microbiol._ 50, 827–836 (2012). Article  PubMed

  CAS  Google Scholar  * Ito, T. _et al_. _Helicobacter pylori_ invades the gastric mucosa and translocates to the gastric lymph nodes. _Lab. Invest._ 88, 664–81 (2008). Article  PubMed  CAS

  Google Scholar  * Lang, N. Consensus Report: Aggressive Periodontitis. _Ann. Periodontol._ 4, 53–53 (1999). Article  Google Scholar  * Lindhe, J. _et al_. Consensus Report: Chronic

Periodontitis. _Ann. Periodontol._ 4, 38–38 (1999). Article  Google Scholar  * Tanner, A. C., Maiden, M. F., Zambon, J. J., Thoren, G. S. & Kent, R. L. Rapid chair-side DNA probe assay

of _Bacteroides forsythus_ and _Porphyromonas gingivalis_. _J. Periodontal Res._ 33, 105–17 (1998). Article  PubMed  CAS  Google Scholar  * Wolff, L. F. _et al_. Dark-pigmented Bacteroides

species in subgingival plaque of adult patients on a rigorous recall program. _J. Periodontal Res._ 23, 170–4 (1988). Article  PubMed  CAS  Google Scholar  * Negi, M. _et al_. Localization

of _Propionibacterium acnes_ in granulomas supports a possible etiologic link between sarcoidosis and the bacterium. _Mod. Pathol._ 25, 1284–97 (2012). Article  PubMed  PubMed Central  CAS 

Google Scholar  * Bae, Y. _et al_. Intracellular _Propionibacterium acnes_ infection in glandular epithelium and stromal macrophages of the prostate with or without cancer. _PLoS One_ 9,

e90324 (2014). Article  ADS  PubMed  PubMed Central  CAS  Google Scholar  * Baess, I. Determination and re-examination of genome sizes and base ratios on deoxyribonucleic acid from

mycobacteria. _Acta Pathol. Microbiol. Immunol. Scand. B._ 92, 209–11 (1984). PubMed  CAS  Google Scholar  Download references ACKNOWLEDGEMENTS We would like to thank the clinical and

nursing staff of the Periodontics Clinic, Dental Hospital, Faculty of Dentistry, Tokyo Medical and Dental University, for their continuous kind support during sample collection. We would

like to thank Dr. Yasuo Takeuchi and Dr. Nay Aung of Department of Periodontology, Tokyo Medical and Dental University for supporting with bacterial culture and Ms. Yuuki Ishige of

Department of Human Pathology, Tokyo Medical and Dental University for the technical support. AUTHOR INFORMATION Author notes * G. Amodini Rajakaruna and Mariko Negi contributed equally to

this work. AUTHORS AND AFFILIATIONS * Department of Periodontology, Graduate School and Faculty of Dentistry, Tokyo Medical and Dental University, Tokyo, 113-8510, Japan G. Amodini

Rajakaruna, Walter Meinzer & Yuichi Izumi * Department of Human Pathology, Graduate School and Faculty of Medicine, Tokyo Medical and Dental University, Tokyo, 113-8510, Japan Mariko

Negi, Asuka Furukawa, Takashi Ito, Daisuke Kobayashi, Yoshimi Suzuki & Yoshinobu Eishi * Division of Surgical Pathology, Tokyo Medical and Dental University Hospital, Tokyo, 113-8510,

Japan Keisuke Uchida, Masaki Sekine, Takumi Akashi & Yoshinobu Eishi * Department of Periodontology, Osaka Dental University, Osaka, 540-0008, Japan Makoto Umeda * Global Center of

Excellence for Tooth and Bone Research, Tokyo Medical and Dental University, Tokyo, 113-8510, Japan G. Amodini Rajakaruna & Yuichi Izumi * Research Fellow, International Scientific

Exchange Fund Program, Japan Dental Association, Tokyo, Japan G. Amodini Rajakaruna Authors * G. Amodini Rajakaruna View author publications You can also search for this author inPubMed 

Google Scholar * Mariko Negi View author publications You can also search for this author inPubMed Google Scholar * Keisuke Uchida View author publications You can also search for this

author inPubMed Google Scholar * Masaki Sekine View author publications You can also search for this author inPubMed Google Scholar * Asuka Furukawa View author publications You can also

search for this author inPubMed Google Scholar * Takashi Ito View author publications You can also search for this author inPubMed Google Scholar * Daisuke Kobayashi View author publications

You can also search for this author inPubMed Google Scholar * Yoshimi Suzuki View author publications You can also search for this author inPubMed Google Scholar * Takumi Akashi View author

publications You can also search for this author inPubMed Google Scholar * Makoto Umeda View author publications You can also search for this author inPubMed Google Scholar * Walter Meinzer

View author publications You can also search for this author inPubMed Google Scholar * Yuichi Izumi View author publications You can also search for this author inPubMed Google Scholar *

Yoshinobu Eishi View author publications You can also search for this author inPubMed Google Scholar CONTRIBUTIONS G.A.R. and M.N. contributed equally to this paper. Y.E. conceived, designed

and coordinated the study. G.A.R., M.U. and Y.I. collected samples. G.A.R., M.N., K.U. and A.F. performed laboratory experiments. A.R., M.N., M.S., T.I. and Y.S. performed histological

experiments. M.N., K.U., D.K. and T.A. performed histological analysis. G.A.R., M.N., K.U., A.F., W.M. and Y.E. wrote the paper. All authors reviewed the manuscript. CORRESPONDING AUTHOR

Correspondence to Yoshinobu Eishi. ETHICS DECLARATIONS COMPETING INTERESTS The authors declare no competing interests. ADDITIONAL INFORMATION PUBLISHER'S NOTE: Springer Nature remains

neutral with regard to jurisdictional claims in published maps and institutional affiliations. ELECTRONIC SUPPLEMENTARY MATERIAL SUPPLEMENTARY INFORMATION RIGHTS AND PERMISSIONS OPEN ACCESS

This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as

long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third

party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the

article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright

holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Reprints and permissions ABOUT THIS ARTICLE CITE THIS ARTICLE Rajakaruna, G.A., Negi, M., Uchida,

K. _et al._ Localization and density of _Porphyromonas gingivalis_ and _Tannerella forsythia_ in gingival and subgingival granulation tissues affected by chronic or aggressive periodontitis.

_Sci Rep_ 8, 9507 (2018). https://doi.org/10.1038/s41598-018-27766-7 Download citation * Received: 23 October 2017 * Accepted: 08 June 2018 * Published: 22 June 2018 * DOI:

https://doi.org/10.1038/s41598-018-27766-7 SHARE THIS ARTICLE Anyone you share the following link with will be able to read this content: Get shareable link Sorry, a shareable link is not

currently available for this article. Copy to clipboard Provided by the Springer Nature SharedIt content-sharing initiative