Introduction
Periodontitis is inflammatory diseases induced by complex periodontic pathogens which results in irreversible alveolar bone destruction by imbalance of bone resorbing and bone formation. Fusobacterium nucleatum is a Gram-negative obligate anaerobe bacterium, commonly cultivated from the subgingival plaque of inflamed gingivitis and induced synergistic alveolar bone loss with other microbial pathogens, such as Tannerella forsythia and Porphyromonas gingivalis [1,2].
Osteoclasts are multinucleated cells and play crucial roles in bone resorption. Osteoclast formation requires the presence of osteoblast or stromal cells which express a receptor activator of nuclear factor-κB (RANK) ligand (RANKL). Many hormones and cytokines are able to promote bone resorption by regulating RANKL expression from osteoblast or stromal cells [3,4].
Mammalian γ-glutamyltranspeptidase (GGT) was identified as a novel bone-resorbing factor [5], which catalyzes the hydrolysis of γ-glutamyl compounds such as glutathione and the transfer of γ-glutamyl moieties to amino acids and peptides [6]. In animal study, the administration of recombinant human GGT to the gingival sulcus induces an increase in the proteins were properly folded number of osteoclasts at the surface of the alveolar bone [7]. GGT is widely distributed in living organism including bacteria. Interestingly, from gram-positive bacteria such as Bacillus species secrete GGT as an extracellular enzyme to gram-negative bacteria such as Escherichia coli and Helicobacter pylori secrete GGT as periplasmic protein were reported [8-10]. Recently, GGT has been shown to be a virulence factor in microbes like H. pylori and B. subtilis GGT of B. subtilis act as a virulence factor in pathogenesis of bone resorption similar to mammalian GGT in vitro. In addition, GGT from H. pylori assisted in the inhibition of T-cell proliferation resulting in persistent colonization and infection [11,12].
The fact that F. nucleatum has GGT was reported in 1997 and the GGT of F. nucleatum was purified by chromatographic techniques and showed enzymatic activity [13]. However, it has not been shown whether the GGT of F. nucleatum has bone-resorbing activity. Thus, we examined whether the GGT of F. nucleatum induces osteoclastogenesis in vitro.
Materials and Methods
1. Alignment of amino acid sequences
The full-length protein sequences of GGT were aligned using the Align X program of Vector NTI version 9.1 (Invitrogen, Carlsbad, CA, USA) and Sequencher 5.1 (Gene Codes Corp., Ann Arbor, MI, USA). Amino acid sequences of GGT in this study were based on the genome sequences of Homo sapiens (human), B. subtilis 168 and F. nucleatum ATCC 25586 (GenBank accession numbers AAA52546, NP_389723, and AAL95137, respectively).
2. Bacterial strains and culture media
The bacterial strain and plasmids for this present study described in Table 1. The strain was cultivated overnight at 37℃ in brain heart infusion (BHI) agar plate. A single colony was inoculated in BHI culture media at 37℃ for 24 hours. The cultured bacteria were centrifuged at 10,000 × g for 10 minutes.
3. Cloning of ggt gene
As shown Fig. 1, a chromosomal DNA of F. nucleatum was extracted from F. nucleatum wild-type strain. The ggt gene was amplified by polymerase chain reaction (PCR) from the DNA template using specific primers listed in Table 2. The restriction enzyme sites of EcoRI and XhoI were inserted into forward and reverse primers, respectively. The PCR parameters were as follows: 94℃ for 40 seconds, followed by 30 cycles 94℃ for 40 seconds, 55℃ for 40 seconds, 72℃ for 40 seconds, and a final extension at 72℃ for 10 minutes. Amplified F. nucleatumggt gene was ligated into pET21b (Novagen, Madison, WI, USA) at the EcoRI and XhoI restriction enzyme recognition sites, resulting in pET21b-Fn ggt . The pET21b vector was used for expression of recombinant GGT proteins of F. nucleatum (Fn rGGT). The pET21b-Fn ggt was transformed into E. coli BL21 (DE3), which is listed in Table 1.
4. Expression and purification of Fn rGGT protein
The pET21b-Bs ggt plasmid has been used in previous study and description can be found Kim et al. [12]. For this study, the pET21b-Bs ggt plasmid was utilized as a positive control. The Fn rGGT expressed in E. coli BL21 (DE3) which was grown at 37℃ in Luria-Bertani (LB) broth containing selection marker ampicillin (100 μg/mL) until approximately 0.6 optical density (OD) at 570 nm. In order to express Fn rGGT, isopropyl β-D- 1-thiogalactopyranoside (IPTG, 0.025 mM) was added to the bacteria culture [14]. IPTG induction was performed at 20℃ for indicated times. It was centrifuged the bacterial cells at 10,000 × g for 15 minutes. The recombinant proteins with C-terminal His-tag were purified by using Ni-NTA spin kit (Qiagen, Hilden, Germany). Binding buffer is composed of 5 mM imidazole, 500 mM NaCl, and 20 mM Tris-HCl (pH 7.9), which was resuspended with cell pellets for purification of this recombinant protein. The resuspended cells were sonicated to obtain cell lysate (bursts with 30 seconds for 5 minutes). The soluble rGGT in supernatant of cell lysate was purified using a nickel-nitrilotriacetic acid (Ni2+-NTA) resin in a column. The column was washed washing buffer with Tris (pH 7.9), 50 mM NaCl and 60 mM imidazole. To elute recombinant proteins, bound protein was eluted with the binding buffer containing 1 M imidazole. Bradford protein assay (Bio-Rad, Hercules, CA, USA) was performed to measure amount of purified recombinant proteins.
5. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis
Samples were heated in sample loading buffer (containing 2% SDS and 5% 2-mercaptoethanol) at 100℃ for 5 minutes. According to the manufacturer’s instruction, protein amount was measured with protein assay kit (Bio-Rad). SDS-PAGE gel electrophoresis was performed using 12% acrylamide gels.
6. Co-culture of osteoblast and bone marrow-derived macrophage
To simulate osteoclastogenesis in presence of GGT of F. nucleatum, osteoblast and bone marrow-derived macrophage were co-cultured. Primary osteoblasts were isolated from the calvaria of the 1- to 2-day old ddy mice (SLC Inc., Shizuoka, Japan) and bone marrow cells were isolated from tibiae and femurs of 5- to 8-week-old ddy mice as described previously [15]. Osteoblasts (106 cells) were seeded in a 10 cm culture dish and grown to confluence. Using Gibco trypsin- EDTA (Gibco, Grand Island, NY, USA) cells were detached from the cultured dishes. Subsequently, the osteoblasts (8 × 103 cells/well) were co-cultured with bone marrow cells (8 × 104 cells/well) derived macrophage in alpha-minimum essential medium (α-MEM) media containing 10% fetal bovine serum (FBS). Fifty μL of each purified Bs rGGT and Fn rGGT at concentrations of 1 nM to 20 nM was added to each co-culture after the medium was exchanged on day 3. Fifty μL of serumfree α-MEM medium was used in the presence of 1α, 25-dihydroxyvitamin D3 (10 –8 M) as positive control, respectively. Every 3 days, culture media were exchanged by replacing with fresh media. For 7 days, the co-culture was maintained.
7. Tartrate resistant acid phosphatase (TRAP) staining
Before TRAP staining, adherent cells were fixed with 10% formaldehyde in phosphate- buffered saline (PBS), and treated with ethanol-acetone (50:50). More than three nuclei TRAPpositive multinucleated cells were considered as osteoclast. Stained cells were photographed with CKX41 (Olympus, Tokyo, Japan).
Results
1. Homology of GGT protein among F. nucleatum , human and B. subtilis
Amino acid sequences were compared among representative GGTs of human, B. subtilis, and F. nucleatum. GGT proteins of human, B. subtilis and F. nucleatum showed sizes approximately 580 amino acids (570, 580, and 588 residues, respectively). GGT proteins between human and B. subtilis shared 28.1% identity (Fig. 2A). Two proteins between human and F. nucleatum shared 25.8% identity (Fig. 2B). When it compared between two bacteria, B. subtilis and F. nucleatum, showed 37.7% identity in the amino acid sequences (Fig. 2C). Therefore, the bacterial GGTs of the B. subtilis and F. nucleatum showed higher similarity than that of the comparison with human.
2. Purification of F. nucleatum recombinant GGT
To express Fn rGGT protein in E. coli , ggt gene of F. nucleatum was cloned into a pET21b. As shown in Fig. 3A, Fn rGGT expression was observed at 12, 24, 48, and 65 hours after IPTG induction (lane 2–5). Autocatalysis of the precursor to heavy and light subunits occurred in the Fn rGGT proteins. The hydrolysis activity of F. nucleatum is increased by timedependent manner. From 48 hours to 65 hours induction time, it was detected that most of GGT precursor was cleaved to heavy and light subunits. After 48 hours induction, therefore, Fn rGGT were purified using His tag and autocatalysis of the precursor to light and heavy subunits occurred in the Fn rGGT similar with Bs rGGT (Fig. 3B). In addition, GGT of F. nucleatum was observed similar molecular sizes to Bs rGGT (light subunit; 21–25 kDa, and heavy subunit; 40–50 kDa).
3. Effect of rGGT proteins on osteoclastogenesis
To prove that GGT of F. nucleatum is sufficient to induce osteoclastogenesis, a co-culture system of osteoblast and bone marrow-derived macrophage was treated with the purified Fn rGGT. TRAP-positive osteoclast cells were photographed and presented in Fig. 4. The absence or presence of 1α, 25-dihydroxyvitamin D3 (10 –8 M) was used for negative and positive controls, respectively (Fig. 4, upper lane) and Bs rGGT was used as a control of the rGGT (Fig. 4, middle lane). This result showed that Fn rGGT induced the osteoclast formation in a dose-dependent manner (Fig. 4, lower lane), similar to that of Bs rGGT.
Discussion
Bacterial GGT was previously identified as a novel osteoclastogenic factor [12]. This study is to prove that the GGT of F. nucleatum act as a bone resorbing factor based on osteoclastforming activity using in vitro co-culture system of osteoblast and bone marrow-derived macrophage.
In a periodontal disease process, F. nucleatum coaggregates with early colonizers, and permit late colonizers including oral pathogens to bind and eventually form a biofilm [16,17]. The prevalence of F. nucleatum increases with the severity of disease, progression of inflammation and pocket depth [18]. Animal studies supported a causative role of F. nucleatum in periodontal infections. Oral inoculation of F. nucleatum in mice caused inflammation of the gingival tissue and activated osteoclasts via CD4+RANKL+cells, resulting in alveolar bone resorption [19] and its effect on alveolar bone loss synergized co-infected with other oral species [1,2]. Here, we showed GGT of F. nucleatum known as periodontopathic bacteria induced osteoclast formation. This induction was increased by a dose-dependent manner and showed similar levels with treatment of Bs rGGT known to stimulate the osteoclastic bone resorption.
Similar with mammalian GGTs which exists as a heterodimer of a light and a heavy subunit, B. subtilis GGT is composed of light (21-kDa) and heavy (45-kDa) subunits processed from a single molecule of precursor by an autocatalytic cleavage [20]. Our previous data proved that the heavy subunit of Bs GGT showed osteoclastogenesis activity and the induction of osteoclastogenesis activity was independent of enzymatic activity and it was related to upregulation of RANKL, an osteoclast differentiation factor [12]. In this study, GGT of F. nucleatum has shown that the intact polypeptide sequences reveal 37.7% identical positions along with GGT of B. subtilis. In addition, there was the predicted region of heavy subunit [21] which shown to be conserved the N-terminal domain (residues 1 to 400) between Bs GGT and Fn GGTs (similarity; about 53.9%).
In summary, our findings suggest that F. nucleatum GGT can serve as a bone resorbing factor. Therefore, GGT of F. nucleatum may play an important role as a virulence factor in alveolar bone destruction. Further study should investigate that GGTs of periodontopathic pathogens other than F. nucleatum involve to active the bone resorption.