Thrombospondin-1 Production Is Enhanced by Porphyromonas gingivalis Lipopolysaccharide in THP-1 Cells

Periodontitis is a chronic inflammatory disease caused by gram-negative anaerobic bacteria. Monocytes and macrophages stimulated by periodontopathic bacteria induce inflammatory mediators that cause tooth-supporting structure destruction and alveolar bone resorption. In this study, using a DNA microarray, we identified the enhanced gene expression of thrombospondin-1 (TSP-1) in human monocytic cells stimulated by Porphyromonas gingivalis lipopolysaccharide (LPS). TSP-1 is a multifunctional extracellular matrix protein that is upregulated during the inflammatory process. Recent studies have suggested that TSP-1 is associated with rheumatoid arthritis, diabetes mellitus, and osteoclastogenesis. TSP-1 is secreted from neutrophils, monocytes, and macrophages, which mediate immune responses at inflammatory regions. However, TSP-1 expression in periodontitis and the mechanisms underlying TSP-1 expression in human monocytic cells remain unknown. Here using real-time RT-PCR, we demonstrated that TSP-1 mRNA expression level was significantly upregulated in inflamed periodontitis gingival tissues and in P. gingivalis LPS-stimulated human monocytic cell line THP-1 cells. TSP-1 was expressed via Toll-like receptor (TLR) 2 and TLR4 pathways. In P. gingivalis LPS stimulation, TSP-1 expression was dependent upon TLR2 through the activation of NF-κB signaling. Furthermore, IL-17F synergistically enhanced P. gingivalis LPS-induced TSP-1 production. These results suggest that modulation of TSP-1 expression by P. gingivalis plays an important role in the progression and chronicity of periodontitis. It may also contribute a new target molecule for periodontal therapy.


Introduction
Periodontitis is an inflammatory disease that is caused by gram-negative anaerobic periodontopathic bacteria, which colonize the tooth surface at marginal gingiva and the subgingival area [1]. Bacterial stimuli induce an immune response that produces a variety of pro-inflammatory cytokines and leads to the destruction of periodontal supportive tissue, resorption of the alveolar bone, and sometimes loss of the tooth [2]. Porphyromonas gingivalis is strongly associated with chronic periodontitis and is believed to play a crucial role in the disease process [3]. P. gingivalis possesses virulence factors such as fimbriae, gingipain, hemagglutinin, outer membrane proteins, lipopolysaccharide (LPS), and capsules [4,5]. Of these, LPS is an important pathogenic component in the initiation and development of periodontitis [6]. Bacterial LPS promotes gingival inflammation with the increased expression of inflammatory cytokines and osteoclastogenesis activation that results in alveolar bone resorption.
Toll-like receptors (TLRs) on dendritic cells, monocytes, macrophages, and polymorphonuclear cells recognize invading stimuli such as LPS and promote activation of the innate immune system in tissues affected by periodontitis. In periodontal pockets, neutrophils are the dominant immune cells against bacteria. However, if neutrophils do not provide sufficient clearance, then bacterial penetration results in the activation of the monocyte/lymphocyte axis. The severity of disease progression is largely due to intrinsic differences in the monocyte/lymphocyte response traits [7]. During periodontal inflammation, monocytes play a key role in this immune response [8]. Human monocytes are exquisitely sensitive to LPS and respond by expressing inflammatory mediators. P. gingivalis LPS is reported to activate human monocytes [9] to induce inflammatory chemokines and cytokines such as IL-1a, IL-1b, IL-6, IL-10, TNF-a, CXCL10, and IL-32 [10,11]. Thus, P. gingivalis LPS acts as an inflammatory signal through TLR pathways [12]. However, few studies have comprehensively examined the enhancement or suppression of the particular gene expression initiated by the host response against periodontal pathogens. An investigation to determine the numerous genes in the human monocytic cell line THP-1 cells stimulated by P. gingivalis LPS has not yet been conducted.
Thrombospondin-1 (TSP-1) is a 420-450-kDa homotrimeric multifunctional extracellular matrix protein, which was first isolated from human blood platelets as a thrombin-sensitive protein [13]. TSP-1 is secreted from endothelial cells, fibroblasts, neutrophils, monocytes, and macrophages [14]. Interacting with multiple different cell surface receptors, proteins, and proteoglycans of specific domains, TSP-1 has various but often converse biological effects [15]. TSP-1 is a regulator of TGF-b activation, which mediates wound healing, proliferation, cell differentiation, and cytokine responses [16]. TGF-b is a pleiotropic immunoregulatory cytokine found upregulated in periodontal gingival tissues [17]. TSP-1 stimulates macrophage migration, neutrophil phagocytosis, and monocyte chemotaxis to modulate the inflammatory response [18,19]. It has an antiinflammatory effect to inhibit NO activation and angiogenesis by binding CD36 [20], whereas it has been reported that CD47-TSP-1 interaction perpetuates the inflammation of rheumatoid synovitis [21]. TSP-1 promotes T-cell function by binding integrin or CD47 to regulate immune responses [21,22]. Activated T-cells induces inflammatory cytokine production and receptor activator of nuclear factor-kB ligand (RANKL) expression, which causes local inflammation and bone destruction. Previous studies have shown that TSP-1 is abundantly expressed in damaged and inflamed tissues such as rheumatoid synovium, atherosclerotic lesions, and diabetes mellitus [18,21]. TSP-1 deficiency decreases obesity-induced adipose tissue inflammation [19] and also attenuates experimental autoimmune encephalomyelitis [23]. Moreover, CD47-TSP-1 interaction is associated with RANKL-driven osteoclastogenesis and bone resorption [24]. These may be related to gingival inflammation and destruction of alveolar bone, which is characteristic of periodontitis. Although many reports have shown that TSP-1 is involved in the inflammation process, little is known about TSP-1 expression during periodontitis and the reaction of TSP-1 with P. gingivalis LPS.
The purpose of this study was to examine P. gingivalis LPS gene expression in human monocytic cells, to evaluate TSP-1 expression in periodontal gingival tissues, and to investigate the modulation of TSP-1 expression by periodontopathic bacteria. Here we show that P. gingivalis LPS induced TSP-1 expression in human monocytic cells. TSP-1 mRNA level was significantly increased in periodontitis gingival tissues. TSP-1 production was dependent upon TLR2/NF-kB signaling and was enhanced by T-cell cytokines. These results indicate that TSP-1 expression may play an important role in the progression of periodontitis.
THP-1 cells were differentiated with phorbol 12-myristate 13-acetate (PMA) (Wako) at 25 ng/ml for 72 h. We removed the PMA containing media and incubated PMA-treated cells in RPMI-1640 for 24 h. PMA-treated cells were incubated at 37˚C in a humidified atmosphere containing 5% CO 2 .

Human gingival tissues
Fifteen human inflammatory gingival tissues were obtained during flap operations from deep periodontal pocket gingiva [probing pocket depth (PPD) $4 mm, bleeding of probing was positive], 13 inflammatory gingival tissues were obtained during apically positioned flap surgery from shallow periodontal pocket gingiva (PPD ,4 mm, bleeding of probing was positive), and 7 healthy gingival grafts with no inflammation were obtained during free gingival graft operations from palatal gingival tissues. The samples of human gingival tissues were divided into 3 groups. Collected samples were immediately immersed in RNAlater (RNA Stabilization Reagent; QIAGEN, Hilden, Germany). Single tissue pieces were obtained from the tube and ground using FastPrep-24 and Lysing Matrix A (M.P. Biomedicals, Santa Ana, CA, USA). Samples were stored at 220˚C.
This study was performed with the approval of the Ethics Committee of Tokyo Medical and Dental University (#879). All sampling was performed after obtaining written informed consent from the patient.

DNA microarray
For gene expression study, mRNA was extracted from THP-1 cells treated with 1.0 mg/ml P. gingivalis LPS or untreated after 4 h [10]. Total RNA was electrophoresed using a 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA) and measured for quality and concentration using Nano Drop (Thermo SCIENTIFIC, Waltham, MA, USA). The concentrations of P. gingivalis LPS stimulating sample (Test) and control sample (Control) were 141.9 ng/ml and 254.0 ng/ml, respectively. The ratio of optical density (OD) at 260 nm to OD at 280 nm (OD 260/280 ) for both Test and Control was similar at 2.07. Each sample underwent RNA amplification using an Ambion Amino Allyl aRNA kit (#1753; Ambion, Carlsbad, CA, USA) using the steps of first and second strand cDNA synthesis, clean-up of double-strand cDNA, synthesis of aRNA, and clean-up of aRNA. The samples were checked by Nano Drop for aRNA amplification. Using 10 ml of the amplified RNA, the Test was labeled with Amersham Cyanine-5 Monoreactive Dye (Cy5; #PA25001; GE, Healthcare Life Sciences Pittsburgh, PA, USA) and the Control was labeled with Amersham Cyanine-3 Monoreactive Dye (Cy3; #PA23001; GE). Samples were purified (YM30/#42407; Microcon, Billerica, MA, USA), fragmented, and purified again (YM10/#42410, Microcon). To measure the Nano Drop labeling efficiency, each 1.0-ml sample was hybridized using Human Oligo chip 25K (TORAY, Tokyo, Japan) at 37˚C for 16 h using a hybridization chamber (Takara-bio, #TX711, Otsu, Japan) and a bio-shaker. Following hybridization and washing, the chip was scanned and quantified using a 3D-Gene Scanner 3000 (TORAY). The raw data sheets show the BG subtraction value, which was obtained by subtracting the background value from the raw data, and the global normalization value of the normalized data, which was obtained by adjusting the median of the Cy3/Cy5 ratio to 1. This DNA array analysis was all performed by 3D-Gene (TORAY).

RNA extraction and real time RT-PCR
Total RNA from cells and gingival tissues was extracted using an RNeasy Mini kit (QIAGEN) according to the manufacturer's instructions. RNA quantitation and purity were calculated by measuring the ratio of absorbance at 260 nm to the absorbance at 280 nm using NanoDrop Lite (Thermo Scientific). cDNA synthesis was conducted using a QuantiTect Reverse Transcription kit (QIAGEN). For each sample, 1.0 mg of total RNA was used for each 20-ml reverse transcription reaction system. The reaction mixture contained 12.5 ml of SYBR Premix Ex Taq II (Takara-bio), 0.8 ml of forward and reverse primers and 1.0 ml of sample cDNA. PCR reaction efficiency was optimized, and the final concentration of each primer was 0.4 mM. Real time RT-PCR was performed with a Thermal Cycler Dice Real Time System II (Takara-bio) with the following thermal profile: and sets of primers (Takara-bio): One ELISA THP-1 cells were stimulated with 0, 0.01, 0.1, and 1.0 mg/ml P. gingivalis LPS for 4 h. THP-1 cells were stimulated with 1.0 mg/ml P. gingivalis LPS for 0, 1, 4, 6, 12, 24, 48, and 72 h. THP-1 supernatants were analyzed to determine the amount of TSP-1 using ELISA kits (Quantikine TSP-1 ELISA kit; R&D SYSTEMS) and following the manufacturer's instructions. OD was measured at a wavelength of 450 nm.

Statistical analysis
All results are shown as means and standard deviations (SD). Data from triplicate samples were used for statistical analysis. The statistical significance of the results was calculated using the Tukey-Kramer multiple comparisons test by InStat (GraphPad Software, La Jolla, CA, USA). Differences were considered significant at P,0.05.

TSP-1 mRNA expression was significantly upregulated in periodontitis gingival tissues
The level of TSP-1 mRNA expression was significantly upregulated in the deep periodontal pocket gingiva group (PPD $4 mm) compared with that in the shallow periodontal pocket gingiva (PPD ,4 mm) and no inflammatory gingiva groups (Healthy; Fig. 1). PPD was 5.6¡1.5 mm (mean ¡ SD) in the Deep periodontal pocket group and 2.4¡0.5 mm in the Shallow periodontal pocket group. Between these two groups, a 2-tailed P value of ,0.0001 was considered extremely significant. There was a tendency for increased production of TSP-1 mRNA as the periodontal pocket became deeper. P. gingivalis LPS induced different types of gene expression in THP-1 cells 3D-Gene microarray-based gene expression profiling was conducted using a gene chip focused on major molecules. THP-1 cells were stimulated with 1.0 mg/ml P. gingivalis LPS for 4 h and 1.0 mg of total RNA was used for detection. The results of this analysis are listed in S1 Table. The total data sheet shows the raw data values; the BG subtraction value, which was obtained by subtracting the background value from the raw data, and the global normalization value of normalized data, which was obtained by adjusting the median of the Cy3/Cy5 ratio to 1. In addition to known chemokines and cytokines, expression of a wide variety of genes, including that encoding TSP-1, was observed. A total of 25,369 genes were detected, of which an increased expression of 2-fold or more was found for 1974 genes, whereas 1247 genes were similarly downregulated. We focused on 134 genes whose expression was increased more than 10-fold. Of particular interest was TSP-1, a protein encoded by the THBS1 gene; this was increased more than 14.5-fold.
TSP-1 was expressed in response to P. gingivalis LPS by THP-1 cells Fig. 2 demonstrates that TSP-1 expression was enhanced by P. gingivalis LPS in THP-1 cells. The highest expression was observed at a concentration of 1.0 mg/ml ( Fig. 2A). TSP-1 mRNA expression at concentrations of 1.0 mg/ml and 0.1 mg/ml P. gingivalis LPS was significantly different from that observed without stimulation (P,0.05 and P,0.001, respectively). The highest expression was observed at 4 h of stimulation with the P. gingivalis LPS preparation (Fig. 2B). THP-1 cells with P. gingivalis LPS preparations reached a peak after 4 h; this corresponds to the fact that TSP-1 is transiently released early during the acute inflammatory phase. The secretion of TSP-1 protein was increased in a time and dose-dependent manner (Fig. 2C, 2D). TSP-1 mRNA expression was also upregulated in PMA-treated THP-1 cells stimulated with P. gingivalis LPS (Fig. 2E).

Discussion
This study investigated the modulation of TSP-1 expression by periodontal pathogenic factor P. gingivalis LPS. Our study revealed that TSP-1 mRNA was significantly upregulated in severe periodontitis gingival tissues. P. gingivalis LPS induced TSP-1 expression in human monocytic cells through TLR2/NF-kB signaling; this TSP-1 production was also enhanced by inflammatory cytokines such as IL-17F.
Here we focused on TSP-1 from the results of a 3D-Gene microarray. Many well-known genes were identified in the array results; among them, we particularly focused on TSP-1, which has not been previously reported in periodontitis. In this study, high levels of TSP-1 mRNA expression were observed in human inflammatory gingival tissues; however, TSP-1 mRNA expression was hardly observed in healthy tissues in the oral cavity. Inflammatory cytokines such as IL-1b, TNF-a, IFN-c, and IL-2 were observed in inflamed gingival tissue, which had deep probing pocket depths [25]. TSP-1 mRNA expression was also significantly enhanced in deep periodontal pocket sites where gingival tissue destruction was severe. P. gingivalis has an ability to attach and invade periodontium cells, and significant correlation was observed between the periodontal pocket depth and the degree of P. gingivalis invasion [26]. In this study, P. gingivalis LPS acted in a dose-dependent manner on TSP-1 mRNA and protein expression in in vitro experiments, suggesting that high amounts of P. gingivalis have the potential to increase TSP-1 in inflammatory gingival tissues.
In some previous studies, large amounts of TSP-1 were expressed at the sites of damaged and inflamed tissues. TSP-1 expression was upregulated in rheumatoid synovial inflammatory tissues [21,27]. Increased expression of TSP-1 facilitated T-cell infiltration, which resulted in persistent tissue damage in inflamed joints [21]. TSP-1 mRNA expression was upregulated in myositis muscles to perpetuate the inflammatory response [28]. Periodontopathic bacterial infection is believed to activate RANKL expression by T-cells. TSP-1 produced by P. gingivalis LPS may be involved in periodontal inflammation and RANKL-driven bone resorption. Therefore, TSP-1 expression may be associated with the severity and chronicity of periodontitis.
However, other studies indicate different inflammatory effects of TSP-1. TSP-1deficient mice had persistent pneumonia because of a defect in the regulation of inflammatory cells [29]. TSP-1 peptide decreased the inflammation of experimental erosive arthritis by reducing granulocytosis [30]. In an experimental autoimmune uveoretinitis model, TSP-1 knockout mice developed severe diseases [23]. These findings suggest that TSP-1 has both anti-inflammatory and proinflammatory mechanisms. TSP-1 interacts with multiple receptors such as CD36, CD47, and integrins [20], suggesting that TSP-1 has contrasting functions. CD36 has been reported to be abundantly co-expressed with TSP-1 in rheumatoid synovium macrophages and endothelial cells [31]. TSP-1, CD36, and CD47 expression have been reported to be upregulated in inclusion body myositis and dermatomyositis muscle specimens [28]. TSP-1 has a strong anti-angiogenic function, which is believed to impact chronic inflammation. These results indicate that TSP-1 may have some role in periodontal inflammation and alveolar bone resorption.
This study was undertaken to investigate the modulation of TSP-1 expression in periodontitis through the stimulation of human monocytic cell line THP-1 cells by P. gingivalis LPS. The present study shows that not only P. gingivalis LPS but also Pam2CSK4 and E. coli LPS promote TSP-1 expression in THP-1 cells. Immunocompetent cells such as macrophages, monocytes, and dendritic cells and periodontal tissue component cells such as gingival epithelial cells and fibroblasts expressed TLR on their cell surfaces. Previous studies have shown that TLR2 and TLR4 are expressed on periodontium cells [32]; in particular, TLR2 and TLR4 are deeply involved in the immune response of periodontal tissues [33]. In a previous study, TLR2 and TLR4 mRNA were expressed at significant levels in THP-1 cells, and TLR2 expression appeared to be stronger than TLR4 expression [34]. In our study, TSP-1 was upregulated by stimulation with the P. gingivalis LPS preparation. Its expression was inhibited by anti-TLR2 but remained unchanged with anti-TLR4, although the LPS of gram-negative bacteria was normally recognized by TLR4 [35]. There is a report that E. coli LPS-stimulated bone marrow-derived macrophages enhance TSP-1 production via TLR4 [36]. On the other hand, P. gingivalis LPS is less dependent on TLR4 signaling than E. coli LPS [37]. P. gingivalis LPS can be utilized through both TLR2 and TLR4 with its unique lipid A structure [12]. In mouse models, TLR2-deficient mice resisted alveolar bone loss following oral infection with P. gingivalis [38], although another study showed that TLR4, not TLR2, was important in protecting against alveolar bone loss [39]. TLR2 plays a sentinel role in the defense against microbial pathogens in the innate immune response [40,41]. It is believed that TSP-1 is produced through TLR2 as well as TLR4 in THP-1 cells by P. gingivalis LPS, and TLR2 is a primary pathway for P. gingivalis LPS, which leads to TSP-1 expression in THP-1 cells. Moreover, its expression is dependent on the NF-kB pathway downstream of TLR2.
The present findings showed that TSP-1 production by P. gingivalis LPS was enhanced by inflammatory cytokines. This is consistent with previous findings where high plasma levels of TSP-1 were correlated with CCL4/MIP1b and TGF-b in patients with rheumatoid arthritis [42]. TNF-a upregulates TSP-1 production in human skeletal muscle-derived cell line myoblasts [28]. Furthermore, in periodontitis, there is a possibility that TSP-1 co-stimulatory with various cytokines activates T-cells or other immune cells. A previous study demonstrated that the addition of Th1 and Th2 cytokines reduced E. coli LPS-mediated TSP-1 production, whereas IL-17 had no effect on bone marrow-derived macrophages [36]. In our study, co-stimulation with P. gingivalis LPS and either IL-4, IFN-c, IL-17A, or particularly IL-17F upregulated TSP-1 expression in THP-1 cells. TSP-1 is closely associated with Th17 differentiation through its ability to activate latent TGF-b [23]. It is therefore suggested that THP-1 cells can continue to secrete TSP-1 in a Th17 environment. Innate immune cells such as neutrophils also produce IL-17 cytokine at an early phase of stimulation [43,44]. A large amount of IL-17 is observed in P. gingivalis oral injection model mice [44]. High amounts of IL-17 in periodontal regions have been reported, and P. gingivalis has been associated with IL-17 levels [45]. P. gingivalis and IL-17 work in tandem in TSP-1 production. However, TSP-1 expression was not increased by IL-17F alone. Although further studies are needed, IL-17F may exhibit potential interaction through the downstream signaling of TLR2. Through the interaction with various cytokines, TSP-1 may have a synergistic effect on aggravated inflammation. It may be useful for the development of treatment and new prevention. In some cases, the inflammatory response that should protect against infection promotes host's tissue destruction. It is necessary to note the disease treatment strategy.
Our study demonstrated that TSP-1 is elevated in periodontal gingival tissues and TSP-1 expression is induced by P. gingivalis LPS stimuli in human monocytic cells. TSP-1 production is upregulated through TLR2 signaling on the NF-kBdependent pathway. Inflammatory cytokines such as IL-17F are involved in TSP-1 induction. In conclusion, the specific role of TSP-1 in periodontal disease is unknown, but it may be a potential target during the progression of periodontitis. Future studies are needed to elucidate the function of TSP-1 in periodontal disease.
Supporting Information S1 Table. Gene expression in THP-1 cells stimulated with P. gingivalis LPS by DNA microarray. THP-1 cells were stimulated with 1.0 mg/ml of P. gingivalis LPS for 4 h. Total RNA from stimulated (Test) and non-stimulated (Control) cells were amplified. The P. gingivalis LPS-stimulated sample was labeled with Cy3 and the control sample with Cy5. The total data sheet shows the raw data, BG subtraction values, and global normalization values, including the Cy3/Cy5 ratio. In total, 25,369 genes were detected. Various genes were upregulated in THP-1 cells, of which TSP-1 expression was increased 14.5-fold. doi:10.1371/journal.pone.0115107.s001 (XLS)