Impaired innate immune signaling due to combined Toll-like receptor 2 and 4 deficiency affects both periodontitis and atherosclerosis in response to polybacterial infection

Abstract

Plasma membrane-associated Toll-like receptor (TLR2 and TLR4) signaling contributes to oral microbe infection-induced periodontitis and atherosclerosis. We recently reported that either TLR2 or TLR4 receptor deficiency alters recognition of a consortium of oral pathogens, modifying host responses in periodontitis and atherosclerosis. We evaluated the effects of combined TLR2^−/−TLR4^−/− double knockout mice on innate immune signaling and induction of periodontitis and atherosclerosis after polybacterial infection with Porphyromonas gingivalis, Treponema denticola, Tannerella forsythia and Fusobacterium nucleatum in a mouse model. Multispecies infections established gingival colonization in all TLR2^−/−TLR4^−/− mice and induced production of bacterial-specific IgG antibodies. In combined TLR2^−/−TLR4^−/− deficiency there was, however, reduced alveolar bone resorption and mild gingival inflammation with minimal migration of junctional epithelium and infiltration of inflammatory cells. This indicates a central role for TLR2 and TLR4 in periodontitis. Atherosclerotic plaque progression was markedly reduced in infected TLR2^−/−TLR4^−/− mice or in heterozygotes indicating a profound effect on plaque growth. However, bacterial genomic DNA was detected in multiple organs in TLR2^−/−TLR4^−/− mice indicating an intravascular dissemination from gingival tissue to heart, aorta, kidney and lungs. TRL2 and TLR4 were dispensable for systemic spread after polybacterial infections but TLR2 and 4 deficiency markedly reduces atherosclerosis induced by oral bacteria.

TLR2/TLR4 deficiency, periodontitis, atherosclerosis, polybacterial infection, oral bacteria, innate immunity

INTRODUCTION

Pattern recognition receptors (PRRs) such as Toll-like receptors (TLRs1–9) and nucleotide oligomerization domain (NOD)-like receptors (NLRs) are highly developed potent microbe-associated molecular pattern (MAMP) recognition receptors. These receptors initiate host innate immune responses to periodontopathic bacteria in gingivitis and periodontitis and are further involved in activating adaptive immunity. Recent studies have suggested a central role for plasma membrane-associated TLR2, TLR4 and intracellular innate sensor TLR9 in the initiation and progression of inflammatory responses in mild, moderate and severe periodontitis, in some cases due to overexpression in diseased tissues (Mori et al.2003; Chen et al.2014; D'Souza et al.2016; Song et al.2017). Expression of TLR2, 4 and 9 is significantly elevated in gingivitis and periodontal tissues such as pocket epithelium, spinous epithelial layer, gingival fibroblasts, periodontal ligament fibroblasts and connective tissues, suggesting that TLRs can induce early periodontitis and drive progression. Chronic periodontitis with a persistent inflammation of the gingival layers, has a strong association with several systemic diseases including atherosclerotic vascular disease (ASVD) (Lockhart et al.2012), adverse pregnancy outcomes (Madianos, Bobetsis and Offenbacher 2013), rheumatoid arthritis (Leech and Bartold 2015) and Alzheimer's disease (Cerajewska, Davies and West 2015). Similar to TLR2, 4 and 9 in periodontitis, there is an expanding body of evidence demonstrating intricate interactions between microbial infections and increased innate immune responses in the initiation and acceleration of chronic inflammatory ASVD (Libby 2002; Libby 2012). TLR1, 2 and 4 expression are markedly over expressed in endothelial cells on the surface of human atherosclerotic lesions (Edfeldt et al.2002), while TLR4 is expressed inside lipid-rich atherosclerotic plaques. Indeed, we have also demonstrated an elevated expression of TLR1 and TLR9 mRNA in aortic tissues of mice orally infected with a polybacterial inoculum (Velsko et al.2015b).

The study by Mullick et al. (2008) has shown increased expression of TLR2 on the surface of endothelial cells at sites prone to development of atherosclerotic plaque, specifically on the inner curvature of the aortic arch in LDLR^−/− mice. Similarly, Higashimori et al. (2011) observed that TLR2 signal deficiency decreased foam cell accumulation in the aorta of ApoE^−/− mice yielding additional support for a pathogenic role for TLR2 signaling in ASVD. An increase in TLR4 gene expression was observed in human atherosclerotic plaques (Xu et al.2001). Similar to TLR2, atherosclerosis-prone ApoE^−/− mice with deficiency of TLR4 have attenuated atherosclerosis progression through decreased macrophage recruitment (Michelsen et al.2004). In addition, TLR4 expression is associated with accelerated foam cell formation with significantly more effect than TLR2 (Higashimori et al.2011).

Chronic periodontitis is now known to be induced by a synergistic subgingival microbial dysbiosis with deregulated immune responses and diseased subgingival plaque comprised of a highly complex microbial consortium of bacteria that includes Porphyromonasgingivalis, Treponemadenticola, Tannerellaforsythia

and Fusobacteriumnucleatum. Approximately, 9–10 oral bacterial species have been detected in human atherosclerotic plaques signifying the potential for a causative role in ASVD (Haraszthy et al.2000; Fiehn et al.2005). Recently, we provided proof-of-concept evidence for a role for TLR2 and TLR4 in multi-species polybacterial periodontitis and ASVD in ApoE^−/− mice (Chukkapalli et al.2017b). Specifically, TLR mice deficient in either TLR2 or TLR4 are resistant to multi-species chronic infection resulting in reduced bone resorption, and suggesting that alveolar bone loss in periodontitis might additionally be regulated by TLR2 and TLR4. Whether plasma membrane-associated TLR2/4 dual genetic deficiency plays a significant role in protection from periodontitis and also ASVD in polybacterial infection is unknown.

In this study, we examined atherosclerosis after polybacterial infection induction of periodontitis (Kesavalu et al.2007; Rivera et al.2013; Chukkapalli et al.2015a; Velsko et al.2015b) using this mouse model to assess the combined roles of TLR2 and 4 microbial-induced signaling in the induction of periodontitis and ASVD using TLR2^−/−TLR4^−/− double knockout (DKO) mice. Our well-characterized chronic infection model of mice periodontitis mirrors, in part, human infection, which is characterized by multiple concurrent oral infections. Accordingly, the primary aim of the current study was to understand the mechanisms by which combined TLR2/4 deficiency contributes to periodontal inflammation, alveolar bone resorption (ABR), systemic inflammation and progression of periodontitis and ASVD. This is the first study to examine the functional role of combined TLR2/4 signaling in periodontitis and ASVD.

MATERIALS AND METHODS

Bacterial strains and inoculum

Porphyromonasgingivalis ATCC 53977, T. denticola ATCC 35404, T. forsythia ATCC 43037 and F. nucleatum ATCC 49256 were grown in a Coy anaerobic chamber at 37°C for 2–3 d, as previously described (Kesavalu et al.2007; Rivera et al.2013; Velsko et al.2015b; Chukkapalli et al.2015a, 2017b). Bacterial concentrations were determined by enumerating in a Petroff-Hausser bacterial counting chamber, and 2.5 × 10⁸P. gingivalis, 2.5 × 10⁸T. denticola, 2.5 × 10⁸T. forsythia, 2.5 × 10⁸F. nucleatum bacteria were mixed in equal concentrations to attain a final inoculum with a concentration of 10⁹ total bacteria per ml (Kesavalu et al.2007; Rivera et al.2013; Velsko et al.2015b; Chukkapalli et al.2015a, 2017b). Periodontal bacteria were suspended in equal volume of reduced transport fluid (RTF) and 4% carboxymethylcellulose (CMC). This bacterial mixture was used for gingival infection of TLR2^−/−TLR4^−/− (double homozygous), TLR2^+/−TLR4^−/− (TLR2 heterozygous, TLR4 homozygous) and TLR2^−/−TLR4^+/− (TLR2 homozygous, TLR4 heterozygous) mice (Table 1), as described previously (Velsko et al.2015b; Chukkapalli et al.2015a, 2017b) (Fig. 1a).

Figure 1.

(A) Schematic diagram of the experimental design illustrating polybacterial infections, gingival sampling, PCR and euthanasia. (B-G) Bacteria-specific IgG and IgM antibody analysis. (B) Serum IgG response in TLR2^−/−TLR4^−/−mice (N = 13). (D) Serum IgG response in TLR2^+/−TLR4^−/−mice (N = 5). (F) Serum IgG response in TLR2^−/−TLR4^+/− mice after 24 weeks of polybacterial infection with, P. gingivalis, T. denticola, T. forsythia and F. nucleatum (N = 8). The graph shows the response to gingival infection with periodontal bacteria expressed as fold increase over sham-infected control mice and hence no error bars depicted in the data (Velsko et al.2015b; Chukkapalli et al.2015a, 2017a, 2017b). Serum IgM levels specific to P. gingivalis, T. denticola, T. forsythia and F. nucleatum after 24 weeks of infection in TLR2^−/−TLR4^−/−mice (N = 13) (C), TLR2^+/−TLR4^−/− (N = 5) (E), TLR2^−/−TLR4^+/− (N = 8) (G) mice. The graph shows the response to gingival infection expressed as fold increase over sham-infected control mice.

Table 1.

Distribution of TLR2 and TLR4 double knockout mice and infection groups.

Mice Group	TLR 2/4 genotype	Bacteria	Total mice
I	TLR2−/−TLR4−/−	Pg + Td + Tf + Fn	13
II	TLR2−/−TLR4−/−	Sham infection	12
III	TLR2+/−TLR4−/−	Pg + Td + Tf + Fn	5
IV	TLR2+/−TLR4−/−	Sham infection	3
V	TLR2−/−TLR4+/−	Pg + Td + Tf + Fn	8
VI	TLR2−/−TLR4+/−	Sham infection	9

Pg—P. gingivalis, Td—T. denticola, Tf—T. forsythia and Fn—F. nucleatum. Group I and II are TLR2^−/−TLR4^−/− double knockout homozygous mice (N12–13), Group III and IV are TLR2 heterozygous and TLR4 homozygous (N = 3–5), and Group V and VI are TLR2 homozygous and TLR4 heterozygous mice (N = 8–9). All mice are same age groups. Our focused study was TLR2^−/−TLR4^−/− double knockout homozygous mice groups I and II. However, we utilized all available heterozygous mice (TLR2^+/−TLR4^−/−: Groups III, IV and TLR2^−/−TLR4^+/−: Groups V, and VI) instead of euthanizing. We have used both sexes in this study.

Generation of TLR2^−/−TLR4^−/− DKO mice

Breeding pairs of the TLR2^−/− (B6.129-Tlr2^tm1Kir/J) and TLR4^−/− (C.C3-Tlr4^Lps-d/J) mice were shipped from Jackson Laboratories (Bar Harbor, ME) to the University of Florida, Animal Care Services. All the animals were housed in Mice Breeding Facility and bred in a specific pathogen-free animal facility in micro isolator cages. Upon arrival, two female TLR2^−/− and one male TLR4^−/− breeding pairs were allowed to adapt to the new environment for at least 1 week prior to initiation of mouse breeding. Weaning of pups was done at 18–21 days after birth. Intercrosses between these F1 heterozygotes yielded mice of all three genotypes (+ /−,−/− and + / +) for initial studies. Tail snipping and subsequent TLR2^−/−TLR4^−/− genotyping was done for confirmation of DKO in mice at TransnetYX Inc (Cordova, TN, USA). Pups were ear notched for identification. Heterozygous TLR2^+/−TLR4^−/−, TLR2^−/−TLR4^+/− and homozygous TLR2^−/−TLR4^−/− knockout male and female mice appeared normal in size, general health and longevity. Disruption of the TLR2 and TLR4 genes did not appear to cause any features of embryonic lethality or affect general health and viability of the male and female mice. No gross microanatomical anomalies attributable to the TLR2^−/−TLR4^−/− knockout were seen in any of the mice examined.

Ethical approval

The University of Florida (UF) has an Assurance with Office of Laboratory Animal Welfare (OLAW) and follows United States Public Health Services (PHS) policy, the Animal Welfare Act and Animal Welfare Regulations, and the Guide for the Care and Use of Laboratory Animals. The UF is also Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC) accredited. Adequate measures were followed at all times to minimize pain and discomfort in all mice. All animal experimentation procedures of this study were approved by University of Florida, Institutional Animal Care and Use Committee (IACUC) protocol # 201304539.

TLR2^−/−TLR4^−/− mouse infection and oral plaque sampling

Mice were housed under specific-pathogen-free facility in University of Florida vivarium. Available TLR2^−/−TLR4^−/−, TLR2^+/−TLR4^−/−, TLR2^−/−TLR4^+/− mice aged 8 weeks were randomly assigned to three infection and three control sham-infection groups (Table 1). Mice were acclimated for 1 week and treated with antibiotics to reduce normal oral flora, as described previously (Velsko et al.2015b; Chukkapalli et al.2015a, 2017b). The infected mice groups (aged approximately10 weeks) were orally inoculated as gavage at the gingival margins of maxilla and mandible molars with 10⁹ total cells per mL in RTF-4% CMC as previously described (Velsko et al.2015b; Chukkapalli et al.2015a, 2017a, 2017b), while sham-infected control mice were infected with RTF-4% CMC vehicle only. The health and behavior of all the mice were monitored daily for 24 weeks and none of the mice died during the study duration. At 24 weeks of chronic infection, mice were euthanized and sera, jaws and internal organs were collected for analyses.

Mice euthanasia, terminal bleeding and tissue collection

Mice were euthanized by CO₂ inhalation followed by cervical dislocation. Blood samples from infected and sham-infected mice were obtained by cardiac puncture at necropsy. Whole blood was centrifuged at 3000 RPM for 5 min to obtain serum, and stored at −20°C. Maxillae (left and right) and mandibles (left and right) were carefully stripped of musculature, leaving an intact periosteum for assessment of horizontal ABR and intrabony defects. Maxillae and mandibles were also collected in 10% phosphate-buffered formalin (pH 7.4) for 48 h for assessment of gingival inflammation (histology) (Velsko et al.2015b; Chukkapalli et al.2015a, 2017a, 2017b). An aliquot of heart and aorta (aortic arch, thoracic, abdominal) were collected in 10% phosphate-buffered formalin (pH 7.4) for 48 h for assessment of atherosclerotic plaque (Velsko et al.2015b; Chukkapalli et al.2015a, 2017a, 2017b). An aliquot of heart, aorta, lung, liver and kidney were placed in liquid nitrogen and stored at −80°C for later detection of bacterial genomic DNA.

Detection of periodontal bacterial genomic DNA in gingival plaque and systemic tissues

Gingival plaque samples were taken 72 h after each infection cycle with a sterile veterinary cotton swab, stored in 150 μl TE (Tris-EDTA) buffer, and used directly to perform a colony PCR with bacterial species-specific primers to monitor gingival colonization as previously reported (Velsko et al.2015b; Chukkapalli et al.2015a, 2017a, 2017b). Mouse tissues (heart, aorta, liver, kidney, lung) (N = 6–12) for bacterial genomic DNA analysis were collected in liquid nitrogen at sacrifice and stored at −80°C. PCR was run by using a Bio-Rad thermal cycler with the following primers, which detect bacterial species-specific 16S rDNA: P. gingivalis forward 5΄- GGT AAG TCA GCG GTG AAA CC-3΄, reverse 5΄- ACG TCA TCC ACA CCT TCC TC-3΄, T. denticola forward 5΄-TAATACCGAATGTGCTCATTTACAT-3΄, reverse 5΄-CTGCCATATCTCTATGTCATTGCTCTT-3΄, T. forsythia forward 5΄-AAAACAGGGGTTCCGCATGG-3΄, reverse 5΄-TTCACCGCGGACTTAACAGC-3΄, F. nucleatum forward 5΄-TAAAGCGCGTCTAGGTGGTT-3΄, reverse 5΄-ACAGCTTTGCGACTCTCTGT-3΄. Bold letters were altered from the reference for a 100% match with F. nucleatum strain ATCC 49256. Bacterial species-specific primers were confirmed by NCBI PrimerBLAST to specifically amplify only the four bacterial species that they were designed to detect, despite the fact that either individual forward or reverse primer may individually detect numerous cultivable and uncultivable bacterial species. Genomic DNA extracted from four bacterial strains served as positive controls and PCR performed with no template DNA served as negative control. PCR products were separated by 1.5% agarose gel electrophoresis and the bands were observed using a BioRad Gel Doc XR (BioRad, CA, USA). Each PCR assay could detect at least 0.05 pg of DNA standard.

Bacterial-specific serum antibody analysis

TLR2/4 (TLR2^−/−TLR4^−/−, TLR2^+/−TLR4^−/−, TLR2^−/−TLR4^+/−) mice sera was collected on sacrifice and stored at −20°C for bacterial-specific antibody analyses. Porphyromonasgingivalis specific, T. denticola-specific, T. forsythia-specific and F. nucleatum-specific IgM and IgG antibody titers were determined by ELISA (Velsko et al.2015b; Chukkapalli et al.2015a, 2017a, 2017b) using whole cell P. gingivalis, T. denticola, T. forsythia or F. nucelatum as antigen. Mouse serum was diluted 1:100 while secondary goat anti-mouse conjugated to alkaline phosphatase (Bethyl Laboratories, Inc. Montgomery, TX) was used at a 1:5000 dilution. Absorbance of each well was read at OD₄₀₅ using a Bio-Rad Microplate Reader. Infected and sham-infected mouse serum antibody concentrations were determined by using a gravimetric standard curve that consisted of eight mouse IgG and IgM concentrations (Sigma). The fold change in bacteria-specific antibody titers between infected and control mice was determined by dividing the mean antibody titer of infected mice by the mean antibody titer of uninfected mice. The quotient, which represents the mean fold change in specific antibody titer due to infection, was graphed (Velsko et al.2015b; Chukkapalli et al.2015a, 2017a, 2017b).

Morphometric measurement of alveolar bone resorption (ABR)

Histomorphometric analysis of horizontal ABR of TLR2/4 (TLR2^−/−TLR4^−/−, TLR2^+/−TLR4^−/−, TLR2^−/−TLR4^+/−) mice maxillae and mandibles were performed as previously described (Velsko et al.2014, 2015b; Chukkapalli et al.2015a, 2017a, 2017b) with the exception that cementum was not stained with methylene blue prior to measuring ABR. Digital images of both buccal and lingual root surfaces of all maxillary and mandibular molar teeth were captured under a 10 × stereo dissecting microscope (SteReo Discovery V8; Carl Zeiss Microimaging, Inc, Thornwood, NY), after superimposition of buccal and lingual cusps to ensure reproducibility and consistency. The surface perimeters of cementoenamel junction (CEJ) i.e. the location where the enamel that covers the anatomical crown of teeth and the cementum that covers anatomical root of teeth meet and alveolar bone crest (ABC) were traced using the calibrated line tool (AxioVision LE 29A software version 4.6.3.). Examiners blinded to the study performed all morphometric measurements twice at separate times. The means of the measurements were obtained for each of the two quadrants. Total ABR was calculated by adding ABR calculated on maxilla palatal, maxilla buccal and mandible lingual alveolar bone.

Histology of gingival inflammation

Five right mandibles from TLR2^−/−TLR4^−/− polybacterial-infected and sham-infected control mice were decalcified in PBS containing 0.4 M EDTA and 2% formaldehyde, embedded in paraffin and sectioned (4 μm) along the mesio-distal plane. Sections were stained with hematoxylin and eosin (H&E) for histological analysis and scanned with a ScanScope CS system (Aperio, Vista, CA). The scanned slides were viewed at 200× magnification with ImageScope viewing software (Aperio). Evidence of gingival inflammation was determined as previously described (Bainbridge et al.2010; Velsko et al.2015b).

Morphometric analysis of aortic atherosclerosis

TLR2^−/−TLR4^−/−, TLR2^+/−TLR4^−/− and TLR2^−/−TLR4^+/− mice hearts (N = 6) and aortas (N = 6) were fixed in 10% neutral buffered formalin and embedded in paraffin. Heart and aortic specimens were then sectioned and assessed for atherosclerotic plaque with particular focus on the aortic root where there is redisposition to plaque growth as previously described (Velsko et al.2014, 2015b; Chukkapalli et al.2014, 2015a). Atherosclerotic plaque area, intimal thickness, medial thickness and calculated intimal/medial thickness ratios were measured by a reviewer blinded to the polybacterial infection using an Olympus DP71 microscope and ImagePro MC 6.0 software standardized to the microscopic objective.

Statistics

The minimum sample size needed for this experimental design is 10–12 mice to detect a difference in means of 0.5 units with a power above 80% and a type I error rate of 0.05. This calculation is based on the observed standard deviation of 9.0 units in our published PD and atherosclerosis study (Rivera et al.2013; Chukkapalli et al.2015a; Velsko et al.2015b). Unpaired, two-tailed Student's T test was used to assess for statistical significance of mice (infected and sham-infected) serum antibody levels, and horizontal ABR data, with GraphPad Prism 5.0 software. ELISA and horizontal ABR graphs have shown mean with standard deviation. ANOVA was used to determine significance for histology measurements using the Statview program and post hoc PLSD analysis; graphs are presented as mean ± standard error. P < 0.05 was considered statistically significant.

RESULTS

Effects of impaired TLR2/4 signaling in polybacterial infection and immune response

Combined TLR2^−/−TLR4^−/− gene deficient mice were infected with polybacterial inoculum. Gingival surface was swabbed after 4 days infection cycle (2^nd, 3^rd and 6^th infection cycles) and bacterial species-specific PCR was performed on gingival plaque samples to examine the presence of bacterial genomic DNA in order to monitor bacterial colonization and infection (Table 2). The majority of infected gingival surfaces in TLR2^−/−TLR4^−/− (Pg/Td/Tf/Fn 100%), TLR2^+/−TLR4^−/− (Pg/Td/Fn 100%; Tf 60%) and TLR2^−/−TLR4^+/− (Pg/Td/Fn 100%; Tf 87%) mice were colonized with all four bacterial species (Table 2 and Tables S1–3, Supportive Information). It is possible that the sampling technique was not sensitive enough to detect all four bacteria, which may explain why not all mice had consistently positive samples in infection cycles examined. None of the gingival surfaces in sham-infected control TLR2^−/−TLR4^−/−, TLR2^+/−TLR4^−/− and TLR2^−/−TLR4^+/− mice were positive for P. gingivalis, T. denticola, F. nucleatum and T. forsythia genomic DNA by PCR (Table 2 and Tables S1–3, Supportive Information). There were no gender differences observed in bacterial colonization and infection of gingival surfaces in either TLR2/4 DKO mice or single TLR2 or TLR4 knockout mice (data not shown).

Table 2.

Distribution of gingival plaque samples positive for bacterial genomic DNA by PCR.

	Positive gingival plaque samples
Group/genotype	Polybacterial infection	2nd Infection	3rd Infection	6th Infection
I TLR2−/−TLR4−/−	Pg/Td/Tf/Fn	a6/7/3/8 (n = 13)	5/6/4/8 (n = 13)	8/7/6/6 (n = 13)
II TLR2−/−TLR4−/−	Sham-infected miceb	0/0/0/0 (n = 12)	0/0/0/0 (n = 12)	0/0/0/0 (n = 12)
III TLR2+/−TLR4−/−	Pg/Td/Tf/Fn	1/2/0/1 (n = 5)	3/3/2/4 (n = 5)	2/2/1/2 (n = 5)
IV TLR2+/−TLR4−/−	Sham-infected mice	0/0/0/0 (n = 3)	0/0/0/0 (n = 3)	0/0/0/0 (n = 3)
V TLR2−/−TLR4+/−	Pg/Td/Tf/Fn	4/3/1/5 (n = 8)	6/5/3/5 (n = 8)	5/3/3/5 (n = 8)
VI TLR2−/−TLR4+/−	Sham-infected mice	0/0/0/0 (n = 9)	0/0/0/0 (n = 9)	0/0/0/0 (n = 9)

Gingival plaque samples collected at second, third and sixth infection cycles following polybacterial infection for determination of microbial colonization by species-specific PCR analysis. Gingival plaque samples not collected at first, fourth, fifth, seventh and eighth infection cycles to facilitate the bacterial adherence to the gingival surface without disruption, invade epithelial cells, multiply and dissemination. Pg—P. gingivalis, Td—T. denticola, Tf—T. forsythia and Fn—F. nucleatum.

a

The first value corresponds to the number of mice that tested positive for P. gingivalis genomic DNA, the second value to the number of mice that tested positive for T. denticola genomic DNA, the third value to the number of mice that tested positive for T. forsythia genomic DNA and the fourth value to the number of mice that tested positive for F. nucleatum genomic DNA at each time point.

P. gingivalis, T. denticola, T. forsythia and F. nucleatum were positive in 100% of mice in-group I (Table S1, Supporting Information for individual mice data). P. gingivalis, T. denticola, and F. nucleatum were positive in 100% of mice in-group III (Table S2, Supporting Information for individual mice data). T. forsythia was positive in 3 out of 5 mice in group III (Table S2, Supporting Information for individual mice data). P. gingivalis, T. denticola and F. nucleatum were positive in 100% of mice in group V (Please refer Table S3, Supporting Information for individual mice data). T. forsythia was positive in seven out of eight mice in group V (Please refer Table S3, Supporting Information for individual mice data).

b

Oral plaque samples were collected from sham-infected mice periodically and were examined for the presence of bacterial genomic DNA of P. gingivalis, T. denticola and T. forsythia and F. nucleatum. None of the sham-infected mice was genomic DNA positive for any of the bacteria (Please refer Tables S1–3, Supporting Information for individual mice data).

Highly significant serum IgG response (> 10–1000-fold) were mounted to P. gingivalis, T. denticola and F. nucleatum in combined TLR2^−/−TLR4^−/− deficient mice (P < 0.001, Fig. 1b) when compared to sham-infected control mice. Similarly, TLR2^+/−TLR4^−/− mice and TLR2^−/−TLR4^+/− mice mounted highly significant serum IgG responses (> 10–1000-fold) (P < 0.001, Fig. 1d, IF) when compared to sham-infected control mice. The serum IgM response of three mouse groups to all four bacterial species was not significant (Fig. 1c, e and g). These IgG and IgM data strongly suggest that T. forsythia poorly colonizes in the oral cavity of TLR2^−/−TLR4^−/− mice and in TLR2 or TLR4 deficient (single knockout) mice, and subsequently no humoral immune response is induced against the infections.

Deficiency of TLR2/4 signaling reduces gingival inflammation and alveolar bone resorption (ABR)

Histological examination of gingival tissue of infected TLR2^−/−TLR4^−/− KO mice demonstrated minimal apical migration of junctional epithelium (JE), gingival hyperplasia, mild inflammatory cellular infiltration in connective tissue and ABR in ABC-stained sections (Fig. 2di) when compared to sham-infected TLR2^−/−TLR4^−/− mice (Fig. 2dii). These findings, indicate that TLR2/4 play an important role in the inflammatory response against polybacterial infection. Accordingly, 24-week-polybacterial infection did not lead to a significantly enhanced ABR in the maxilla palatal, maxilla buccal and mandible lingual sides in combined TLR2^−/−TLR4^−/− KO mice (Fig. 2a, b and c) as well as both male and female mice (Fig. 3a-d). Similarly, TLR2^+/−TLR4^−/− mice (Fig. 4a and b), and TLR2^−/−TLR4^+/− mice (Fig. 4c and d) did not enhance the ABR to polybacterial infection relative to their sham-infected control mice. ABR is the outcome of strong gingival inflammatory processes, and mice lacking both TLR2/4 receptors have dampened periodontal inflammatory responses.

Figure 2.

(A) Total horizontal Alveolar bone resorption measurements in infected TLR2^−/−TLR4^−/−mice (Group I; N = 13) and sham-infected TLR2^−/−TLR4^−/−mice (Group II; N = 12). (B) Representative image of the mandible lingual surface depicting the horizontal ABR following polybacterial infection in TLR2^−/−TLR4^−/− mice. (C) Representative image of mandible lingual surface following sham-infection in TLR2^−/−TLR4^−/− mice. (D) Representative image of infected TLR2^−/−TLR4^−/− mice gingival tissue demonstrating minimal apical migration of JE and infiltration of inflammatory cells in connective tissue and bone resorption lacuna in ABC. (E) Representative image of sham-infected TLR2^−/−TLR4^−/− mice gingival tissue demonstrating no signs of infiltration of inflammatory cells in connective tissue and bone resorption lacuna in ABC. ABC-alveolar bone crest; CEJ-cementoenamel junction; CT-connective tissue, D-dentin, M1, first molar; M2, second molar; M3, third molar.

Figure 3.

Representative images of the mandible lingual surface depicting the horizontal ABR following (A) polybacterial infection in male TLR2^−/−TLR4^−/− mice and (B) polybacterial infection in female TLR2^−/−TLR4^−/− mice. Representative images of the mandible lingual surface depicting the horizontal ABR following (C) sham-infection in male TLR2^−/−TLR4^−/− mice and (D) sham-infection in female TLR2^−/−TLR4^−/− mice. M1, first molar; M2, second molar; M3, third molar.

Figure 4.

(A) Total horizontal Alveolar bone resorption measurements in infected TLR2^+/−TLR4^−/− mice (Group III; N = 5) and sham-infected control TLR2^+/−TLR4^−/− mice (Group IV; N = 3). (B) Representative image of the mandible lingual surface depicting the horizontal ABR following (i) polybacterial infection in TLR2^+/−TLR4^−/− mice and (ii) sham infection in TLR2^+/−TLR4^−/− mice. (C) Total horizontal Alveolar bone resorption measurements in infected TLR2^−/−TLR4^+/− mice (Group V; N = 8) and sham-infected control TLR2^−/−TLR4^+/− mice (Group VI; N = 9). (D) Representative image of the mandible lingual surface depicting the horizontal ABR following (i) polybacterial infection in TLR2^−/−TLR4^+/− mice and (ii) sham-infection in TLR2^−/−TLR4^+/− mice. M1, first molar; M2, second molar; M3, third molar; ABC- alveolar bone crest; CEJ- Cementoenamel junction.

Systemic dissemination of periodontal bacteria

To investigate whether oral bacteria colonized the gingival margins of molar teeth disseminated intravascular to multiple systemic organs, we extracted bacterial genomic DNA from heart, aorta, liver, kidney and lung and PCR was done to detect the presence of specific bacterial genomic DNA. Only P. gingivalis, T. denticola and F. nucleatum genomic DNA were positive in all organs indicating its hematogenous dissemination with highest genomic DNA presence in the hearts and aortas of both TLR2^−/−TLR4^−/− deficient mice (Table 3). Tannerellaforsythia genomic DNA was not detected in any of the systemic organs examined. Similarly, only P. gingivalis and T. denticola genomic DNA were detected in the heart, aorta, kidney and lung of TLR4 deficient mice (TLR2^+/−TLR4^−/−) (Table 3). Furthermore, P. gingivalis, T. denticola and F. nucleatum genomic DNA were positive in all organs except liver in TLR2 deficient mice (TLR2^−/−TLR4^+/−) (Table 3). Similar to gingival bacterial infection, T. forsythia genomic DNA was not detected in any of the organs of infected TLR2^−/−TLR4^−/−, TLR2^−/− and TLR4^−/− deficient mice, possibly due to its reduced adherence to gingiva and invasion of gingival junctional epithelium.

Table 3.

Distribution of tissue samples positive for P. gingivalis, T. denticola, T. forsythia and F. nucleatum genomic DNA by PCR.

	Positive systemic tissue samples
Group/genotype	Poly infection	Heart (n = 12)	Aorta (n = 12)	Liver (n = 6)	Kidney (n = 6)	Lung (n = 6)
I TLR2−/−TLR4−/−	Pg/Td/Tf/Fn	a7/5/0/2	6/3/0/5	0/2/0/0	4/3/0/1	3/1/0/1
II TLR2−/−TLR4−/−	Sham-infected mice	0/0/0/0	0/0/0/0	0/0/0/0	0/0/0/0	0/0/0/0
III TLR2+/−TLR4−/−	Pg/Td/Tf/Fn (n = 5)	2/1/0/0	1/1/0/0	0/0/0/0	1/0/0/0	1/1/0/0
IV TLR2+/−TLR4−/−	Sham-infected mice (n = 3)	0/0/0/0	0/0/0/0	0/0/0/0	0/0/0/0	0/0/0/0
V TLR2−/−TLR4+/−	Pg/Td/Tf/Fn (n = 8)	2/2/0/1	2/2/0/2	0/0/0/0	2/1/0/2	3/1/0/2
VI TLR2−/−TLR4+/−	Sham-infected mice (n = 9)	0/0/0/0	0/0/0/0	0/0/0/0	0/0/0/0	0/0/0/0

Pg—P. gingivalis, Td—T. denticola, Tf—T. forsythia, Fn—F. nucleatum. Tissue samples were analyzed using appropriate bacterial species-specific PCR primers with positive and negative controls.

a

The first value corresponds to the number of mice tissues that tested positive for P. gingivalis genomic DNA and the second value to the number of mice tissues that tested positive for T. denticola genomic DNA. The third value to the number of mice tissues that tested positive for T. forsythia genomic DNA and the fourth value to the number of mice tissues that tested positive for F. nucleatum genomic DNA.

Deficiency of TLR2^−/−TLR4^−/− signaling reduces progression of aortic atherosclerotic plaque

To investigate whether multiple oral bacteria acting as a synergistic polybacterial infection, can induce aortic atherosclerosis progression in TLR2^−/−TLR4^−/−, TLR2^+/−TLR4^−/− and TLR2^−/−TLR4^+/− deficient mice, we measured atherosclerotic plaque growth in the aorta at the level of the aortic valve. As expected, the TLR2^−/−TLR4^−/− DKO mice exhibited a trend toward increased plaque area in infected mice when compared to sham-infected control mice (Fig. 5a), on measurement of intimal thickness (Fig. 5b), medial thickness (Fig. 5c), or intimal-medial thickness ratio between infected and sham-infected control mice (Fig. 5d). However, in TLR2^−/−TLR4^−/− mice this increase in plaque size after polybacterial infections no longer achieved significance. The infected TLR2^−/−TLR4^−/− mice exhibited smaller plaque areas (Fig. 5e) similar to sham-infections (Fig. 5f and g) which also had minimal plaque (Fig. 5f and g). Of interest, mice with partial expression of TLR2 (TLR2^+/−) or TLR4 (TLR4^+/−) also had minimal plaque growth when compared to sham infections. These findings suggest that TLR2 and 4, whether completely absent in the DKO TLR2^−/−TLR4^−/− or partially absent in the TLR2 (TLR2^+/−TLR4^−/−) and TLR4 (TLR2^−/−TLR4^+/−) heterozygotes leads to a marked suppression of plaque growth after polybacterial infection. These data clearly demonstrate that in the absence of both TLR2 and TLR4 or partial absence of either TLR2 or TKR4 signaling prevents an increase in atherosclerotic plaque progression after polybacterial-infected when compared to sham-infected control mice.

Figure 5.

Chronic polybacterial (Pg + Td + Tf + Fn) infection induced alterations in the aorta of TLR2^−/−TLR4^−/− mice (N = 6). (A) Bar graphs of morphometric analysis of mean aortic plaque area in the polybacterial-infected and sham-infected control TLR2^−/−TLR4^−/− mice, (B) intimal layer thickness, (C) medial layer thickness and (D) intimal/medial layer thickness ratios. Atherosclerotic plaque in the H&E stained aortic leaf of polybacterial-infected and sham-infected control TLR2^−/−TLR4^−/− mice. Panel E is a representative image of plaque in infected TLR2^−/−TLR4^−/− mice at 24 weeks of infection at 40× magnification. Panel F is a representative image of plaque in sham-infected control TLR2^−/−TLR4^−/− mice at 24 weeks of infection at a lower magnification 20 × and panel G is a representative image of plaque in sham-infected TLR2^−/−TLR4^−/− mice at a higher magnification 40×. Panel H is a representative image of plaque in infected TLR2^+/−TLR4^−/− mice at 24 weeks of infection at 10x magnification. Panel I is a representative image of plaque in infected TLR2^+/−TLR4^−/− mice at 24 weeks of infection at a lower magnification 4×. Panel J is a representative image of plaque in infected TLR2^−/−TLR4^+/− mice at 24 weeks of infection at a lower magnification 4×. Black arrow indicate plaque margin. L-lumen; I-intimal layer; M-medial layer.

DISCUSSION

Atherosclerosis is a chronic inflammatory disease with plaque growth in the intimal layer of the arterial wall. TLR2 and TLR4 are microbial sensors and atherogenic promotors that have been implicated in atheroma development and progression, causing ASVD. The TLRs also mediate pro-inflammatory signals involved in the pathogenesis of ASVD and periodontitis as shown in in vivo animal studies (Velsko et al.2014; Chukkapalli et al.2014, 2015a). The cells of the periodontium (pocket epithelium, spinous epithelial layer, gingival fibroblasts and periodontal ligament fibroblasts) express several TLRs, and NLRs, and TLR sensing and signaling plays a major role in maintaining the periodontal health. These shared inflammatory processes may explain the links between ASVD and chronic periodontitis, at least in part. Roshan, Tambo and Pace (2016) demonstrated a reduced atherosclerosis-associated inflammatory response in TLR2^−/− and TLR4^−/− mice. However, the precise roles for combined TLR2^−/−TLR4^−/− DKO in inflammatory responses to polybacterial infections are unclear. It is known that both P. gingivalis and T. denticola subvert TLR signaling, which is thought to contribute to pathogenesis of periodontitis (Hajishengallis et al.2013).

We recently reported that periodontal bacteria are able to establish gingival colonization in TLR2^−/− and TLR4^−/− mice and induce a pathogen-specific IgG immune response, accompanied by reduced ABR, and indicating a major role in infection-induced periodontitis (Chukkapalli et al.2017b). While bacteria disseminate from gingival tissue to the heart and aorta, there was no increase in atherosclerotic lesions in the aortic arch. In addition, polybacterial infection does not alter levels of serum risk factors such as oxidized low-density lipoprotein, nitric oxide and lipid fractions in both mice indicating a strong correlation with diminished atherosclerosis and with that of TLR2 and TLR4 (Chukkapalli et al.2017b). Infected TLR2^−/− mice demonstrated significant levels (P < 0.05 to P < 0.01) of T helper type 2 [transforming growth factor-β₁, macrophage inflammatory protein-3α, interleukin-13 (IL-13)] and T helper type 17 (IL-17, IL-21, IL-22, IL-23) splenic T-cell cytokine responses (Chukkapalli et al.2017b).

Therefore, in this study, we focused on the evaluation of the role of combined deficiency of TLR2/4 on the progression of both periodontitis and atherosclerosis after a polybacterial oral infection. The majority of TLR2^−/−TLR4^−/− DKO mice, TLR2^+/−TLR4^−/− mice and TLR2^−/−TLR4^+/− mice gingival surfaces were effectively colonized with all four bacterial species, induced a pathogen-specific IgG immune response, but reduced ABR. There are no gender differences in bacterial colonization and infection in the gingival surfaces in either TLR2^−/−TLR4^−/− DKO mice or single TLR2 or TLR4 knockout mice. In addition, mice lacking both TLR2^−/−TLR4^−/− receptors have dampened periodontal inflammatory responses correlating with reduced ABR, and indicating a major role of TLR2/4s in periodontitis. Porphyromonasgingivalis, T. denticola and F. nucleatum genomic DNA were positive in the hearts and aortas indicating their hematogenous dissemination in TLR2^−/−TLR4^−/− deficient mice, corroborating polybacterial colonization and specific-immune responses. Furthermore, deletion of TLR2/4 gene expression was associated with reduced atherosclerotic plaque in the aortic arch suggesting that TLR2/4 play a major role in infection-induced atherosclerosis.

Several studies were reported using monoinfection with P. gingivalis alone either in TLR2 deficient mice or in TLR4 deficient mice (Hayashi et al.2010, 2012). We have conducted five previous studies with four significant oral bacteria as polybacterial infection in TLR2/4 expressing mice (Rivera et al.2013; Velsko et al.2015b; Chukkapalli et al.2015a, 2017a, 2017b). Similarly, we have conducted four studies as monoinfections in TLR2/4 expressing mice (Chukkapalli et al.2014, 2015b; Velsko et al.2014, 2015a). These studies have highlighted the significance of TLR2 and TLR4 in induction of PD as well as ASVD. One potential limitation in this study is lack of wild-type control TLR2/4 expression mice to compare with TLR2^−/−TLR4^−/− deficient mice. In general, wild-type mice have very little or minimal if any proatheroscerotic potential as has been published (Wang, Chang and Kuan 1965; Paigen et al.1985).

Our recent study found that mRNA expression of Tlr1 and Tlr9 were altered in aortic tissues in the presence of periodontitis, while Tlr2 and Tlr4 were not, suggesting that Tlr1 and Tlr9 signaling may be more important in periodontitis-influenced atherosclerosis (Velsko et al.2015b). Interaction of intracellular innate sensor TLR9 is intriguing as this receptor senses intracellular microbial DNA, as opposed to plasma membrane-associated sensors TLR2 and TLR4 that sense bacterial surface components. These data together suggest that intracellular bacterial invasion is a more potent driver of inflammation than bacterial-host cell surface interactions. Few other studies also reported the protective role of TLR3 and TLR7 in ASVD (Cole et al.2011; Salagianni et al.2012). These findings highlight the significance to look beyond plasma membrane-associated TLRs and potentiate the initiation of future studies to understand the intricate role that endosomal TLRs play in initiation of atherosclerosis and periodontitis. Bacterial–bacterial interactions may change gene expression of the interacting organisms (Aruni, Roy and Fletcher 2011; Tan et al.2014) and thereby modify protein expression and behavior of the bacteria. In a way, bacterial–host cell interactions in polybacterial infections may vary significantly from those observed in single-bacterial species infections (Shin et al.2013). As this is the first study on the role of multiple oral bacterial infections that examines plasma membrane-associated TLR2/4 signaling in periodontitis and atherosclerosis, we could not compare our findings with similar published studies. Hence, with the observations of the current study it is not emphatically possible to draw conclusions about the extent various TLR receptors and signaling pathways are altered in PD and their contribution to ASVD. Thus, further studies are warranted with other TLR signaling molecules to understand their mechanistic pathways of activation and their individual contribution to ASVD.

CONCLUSIONS

In summary, we demonstrate that combined genetic deficiency of TLR2^−/−TLR4^−/− significantly abrogates inflammation in the periodontium, reduces ABR (periodontitis) and aortic arch atherosclerotic plaque progression. Our findings demonstrate an important role of both TLR2 and TLR4 in the progression of polybacterial infection-induced periodontitis and atherosclerosis. Future studies are recommended to understand the significance of other plasma membrane-associated TLRs along with endosomal TLRs to understand the significance of individual TLRs in PD and ASVD.

Acknowledgements

We thank Dr. Irina Velsko for previewing this manuscript, critical comments and Dr. Aravindraja Chairmandurai for previewing and formating figures.

FUNDING

This work was supported by National Institute of Health/National Institute for Dental and Craniofacial Research [R01DE020820]. The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript.

Conflict of interest. None declared.