Seven female beagle dogs (12 months of age) were used in the current study. The study was approved by the Animal Experiments Ethics Committee of Shanghai Jiao Tong University School of Medicine (Shanghai, China). All of the experimental dogs had optimal periodontal health prior to the study.

PDLCs were isolated from the anterior teeth of each dog via enzyme digestion. The isolated PDLCs were cultured in expansion medium composed of α-minimum essential medium (α-MEM) supplemented with 10% fetal bovine serum (FBS) (both from Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA), ascorbic acid phosphate ester and 1% antibiotic-antimycotic solution (Gibco; Thermo Fisher Scientific, Inc.) at 37°C in a humidified atmosphere of 95% air and 5% CO₂.

HBD-3 DNA was obtained from our previous study (17). The HBD-3 gene and the green fluorescent protein (eGFP) gene were cloned into the pLV.Des3d. P/puro vector [all from Cyagen Biosciences (Guangzhou), Inc., Guangzhou, China]. E. coli Stbl3 was used as the host. For amplification, the primer sequences were as follows: attB1-Kozak-Humacalx-F 5′-GGGGACAAGTTTGTACAAAAAAGCAGGCTGCCACCATGGGCTTCCAAAAGTTCTCCC-3′; attB2-Humacalx-R, 5′-GGGGACCACTTTGTACAAGAAAGCTGGGTTTAGTTGGCATTCTGGGGCATG-3′; attB1-Kozak-PTH-F, 5′-GGGGACAAGTTTGTACAAAAAAGCAGGCTGCCACCATGATACCTGCAAAAGACATGGC-3′; attB2-PTH-R, 5′-GGGGACCACTTTGTACAAGAAGCTGGGTTCACTGGGATTTAGCTTTAGTTAATAC-3′ (Sangon Biotech Co., Ltd., Shanghai, China). The polymerase chain reaction (PCR) amplification system included 5X primer STAR™ buffer (Mg²⁺ Plus, 10 µl), dNTP mixture (4 µl) (both from Takara Bio, Inc., Otsu, Japan), forward primer 1 µl, reverse primer 1 µl, DNA 1 µl, primer STAR™ HS DNA polymerase (0.5 µl; Takara Bio, Inc.) and ddH₂O. The PCR amplification conditions were as follows: 98°C for 3 min, then 30 cycles of 98°C for 30 sec, 60°C for 30 sec and 72°C for 1 min, and a final extension of 72°C for 5 min.

The recombinant plasmid pLV.EX3d. P/puro-EF1A-Humacalx-IRES/eGFP was constructed using Gateway Technology (Invitrogen; Thermo Fisher Scientific, Inc.). The recombinant plasmid pLV.EX3d. P/puro-EF1A-PTH-IRES/eGFP was constructed without HBD-3. The constructed expression plasmids were amplified in the E. coli Stbl3 strain.

The 293T cells [American Type Culture Collection (ATCC), Manassas, VA, USA] used in transfection experiments were cultured and passaged in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% FBS (both from Gibco, Thermo Fisher Scientific, Inc.), 100 U/ml penicillin and 100 µg/ml streptomycin at 37°C and 5% CO₂. The lentiviral vector containing HBD-3 was first transfected into 293T packaging cells. Following 8 h, the medium was replaced with fresh medium supplemented with 10% FBS. The viral particles were collected 48 h following plasmid transfection and were concentrated at 1,000 × g and 4°C for 15 min, with filtration at 0.45 µm to remove cell debris in the culture supernatant. The viral particles were then further concentrated by centrifugation at 5,000 × g for 90 min at room temperature. The obtained virus titer was approximately 6×10⁹ TU/ml (cells transfected with HBD-3 and GFP) and 2×10⁹ TU/ml (cells transfected with GFP only). The transfected PDLCs were cultured in DMEM and incubated with cells in 5% CO₂ at 37°C. The cells were then incubated for an additional 72 h. The transfection efficiency was calculated using a fluorescence microscope.

The GFP-HBD-3-PDLCs and GFP-PDLCs were cultured in 60 mm dishes until 80–90% confluency was reached. They were then collected with radioimmunoprecipitation lysis buffer (Beyotime Institute of Biotechnology, Haimen, China) and quantified using a bicinchoninic acid assay (Thermo Fisher Scientific, Inc.). Total proteins (20 µg/lane) were separated by 5–15% SDS-PAGE and transferred to PVDF membranes. Membranes were blocked with 5% nonfat milk for 1 h at room temperature and incubated overnight at 4°C with the corresponding primary antibodies: Anti-β-actin (cat no. ab1801; 1:1,000) and anti-β-defensin-3 antibody (cat no. ab19270; 1:1,000) (both from Abcam, Cambridge, UK). They were then incubated with a goat anti-rabbit IgG horseradish peroxidase-conjugated secondary antibody (cat no. KC-RB-035; 1:5,000; Shanghai Kangchen Biotechnology Co., Ltd., Shanghai, China) for 1 h at room temperature. After washing in PBST (PBS containing 0.1% Tween-20), the membranes were developed using an EZ-ECL detection kit following the manufacturer's instructions (Biological Industries Israel Beit Haemek Ltd., Beit Haemek, Israel) and were then imaged using a UVitec gel documentation system (UVitec Limited, Cambridge, UK).

GFP-HBD-3-PDLCs and GFP-PDLCs were plated into temperature-responsive culture dishes and cultured at 37°C in α-MEM supplemented with 10% FBS, ascorbic acid phosphate ester and 1% antibiotic-antimycotic solution. When the cells reached 100% confluence, the dishes were transferred into an incubator set at 20°C to induce the formation of cell sheets. The resultant cell sheets were harvested and fixed with 10% neutral formalin solution.

The culture supernatants were collected, and the amount of secreted HBD-3 was detected using an HBD-3 ELISA kit (cat no. JL19214; Shanghai Jiang Lai Biotechnology Co., Ltd., Shanghai, China) according to the manufacturer's instructions. Absorbance at 490 nm was determined with a microplate reader (BioTek ELx800; BioTek Instruments, Inc., Winooski, VT, USA). Each sample was analyzed in triplicate.

Actinomyces viscosus (ATCC 19246); Candida albicans (ATCC 10231); Porphyromonas gingivalis (ATCC 33277) and Streptococcus mutans (UA 159) were used to test the antimicrobial activity of the secreted HBD-3. All the strains were gifts from the Shanghai Research Institute of Stomatology. The strains were subcultured in brain heart infusion (BHI; BBL™ CHROMagar™; BD Biosciences, Franklin Lakes, NJ, USA) agar plates at 37°C in an aerobic (5% CO₂ and 95% air) or anaerobic (80% N₂, 10% H₂ and 10% CO₂) system for 24 h.

Cell-free culture supernatants from the transfected PDLC sheets were collected 48 h after transfection and were serially diluted. Subsequently, 100 µl suspension of the four microbial strains (10⁶ CFU/ml) were inoculated in BHI broth supplemented with 100 µl diluted supernatants at 37°C for 48 h. Microbial growth was assessed by measuring the optical density (OD) at 590 nm with a microplate reader. All experiments were performed in triplicate. Medium alone and untreated cells were used in the control groups.

The four microbial strains in cell-free culture supernatants from 35 mm dish were diluted 10⁶, and 100 µl of the solution were spread on BHI agar plates at 37°C. The number of colony-forming units (CFU) was counted after 48 h incubation. All experiments were performed in triplicate. Medium alone and untreated cells were used in the control groups.

The experimental dogs were fed a high sugar, high starch diet (daily diet of dog food mixed with 100 g sugar and 20 g mashed potatoes). In addition, elastic ligatures were subgingivally placed around the necks of all 3rd and 4th premolars (P3, P4) and all 1st molars (M1) to induce experimental periodontitis. The development and progression of periodontitis was monitored for 1 month.

Following intraperitoneal anesthesia and elevation of the buccal mucoperiosteal flap, the alveolar bone of the dogs was removed to create experimental periodontal bone defects. The defect was 5 mm in width, 45 mm in length, and 3 mm in depth, prepared from P3 to mesial root of M1.

The four sites of experimental periodontitis in each dog were randomly assigned into the following three experimental groups: Test group, transplanted with GFP-HBD-3-PDLC sheets (Fig. 1); control group, transplanted with GFP-PDLC sheets (Fig. 1); and untreated group, which received no treatment. The grouping results were as follows: 28 sites were involved in the study and there were 9, 12 and 7 sites assigned into the test group, the control group and the untreated group, respectively. To ensure that a uniform thickness was maintained between the different constructs, the same layers of cell sheets were used, and all procedures were completed by the same person. Following the surgery, all dogs used in this experiment received 4×10⁶ units amoxicillin intramuscularly daily for 7 days.

Inflammation during the healing period was observed 2 weeks following transplantation in the diseased regions of the untreated and control groups. Assessments were made by eye using clinical observations including, gingival swelling, periodontal overflow pus and loose teeth. At 8 weeks after the transplantation with cell sheets, the periodontitis model dogs were sacrificed and the alveolar bones of experimental sites were harvested. After excision, fixation in 4% paraformaldehyde for 24 h at room temperature and demineralization in 10% EDTA for 10 weeks at room temperature, buccal-lingual histologic specimens from P3, P4 and M1, each containing a tooth and its associated periodontium, were embedded in paraffin wax. Serial tissue sections (3-µm-thick) from the paraffin blocks were stained with hematoxylin and eosin for 5 min, respectively, at room temperature for cell visualization.

Following deparaffinization, antigen retrieval was performed with 1 mM Tris-EDTA (pH 8.0) in a 100°C water bath for 20 min and endogenous peroxidase activity was blocked with 3% hydrogen peroxide for 10 min at room temperature. All of the sections were then incubated with primary antibodies against GFP (cat no. ab183735; 1:200), tumor necrosis factor-α (TNF-α; cat no. ab199013; 1:200), interleukin-1β (IL-1β; cat no. ab34837; 1:200) and osteocalcin (OCN; cat no. ab13418; 10 µg/ml) (all from Abcam) for 1 h at room temperature. An EnVision™ Detection Systems kit (Peroxidase/DAB; cat no. K5007; Dako; Agilent Technologies, Inc., Santa Clara, CA, USA) was applied until the desired staining intensity was achieved, which was observed under a light microscope (Axio Scope A1; Carl Zeiss AG, Oberkochen, Germany). To ensure that pathological diagnosis was standardized, the cellular localization patterns, intensities, and numbers of cells with positive staining were assessed by two experienced oral pathologists in a double-blind manner. A consensus decision was reached in cases with discrepancies. Staining intensity was defined as 0, 1, 2 and 3 for no staining, light yellow staining, yellow-brown staining and brown staining, respectively. Expression levels were semi-quantitatively evaluated by multiplying the mean percentage of positive cells by the intensity score. The mean percentage of positive cells was determined in 4 random fields (magnification, ×400).

SPSS 19.0 software package (IBM Corp., Armonk, NY, USA) was used for all statistical analyses. Significant differences were calculated using one-way analysis of variance followed by the least significant difference or Tamhane test. The ELISA and antimicrobial activity assay data are expressed as the mean ± standard error. The IHC are expressed as the mean ± standard deviation. P<0.05 was considered to indicate a statistically significant difference.

Transfection efficiency was calculated by counting the cells that fluoresced green. The results demonstrated that the rate of PDLC transfection with the GFP + HBD-3 vector was 78.98% (Fig. 2A and B). The rate of PDLCs transfection with the GFP only vector was 74.96% (Fig. 2C and D).

To examine the protein expression level of HBD-3 in the transfected PDLCs, western blot analysis was performed. Distinct positive bands were observed corresponding to the PDLCs transfected with the HBD-3 + GFP vector. However, no positive bands were observed in the untransfected groups and GFP only vector (Fig. 3).

ELISA was performed to quantify the levels of HBD-3 in the PDLC sheets transfected with the HBD-3 and GFP genes. The untreated cell sheets were used as a control. The results showed that the concentrations of secreted HBD-3 were 279.22±2.17 µg/ml in the supernatants of PDLC sheets transfected with HBD-3 (Fig. 4), which was significantly higher than that of the corresponding cell sheets transfected with the control vector and that of the untreated control group (P<0.001).

As presented in Fig. 5A, Actinomyces viscosus, Candida albicans, Porphyromonas gingivalis and Streptococcus mutans were cultured with supernatants from PDLC sheets transfected with the HBD-3 gene expression vector at various dilutions, and exhibited lower OD values compared with those cultured with supernatant from control vector and untransfected cells.

A CFU assay was also performed to evaluate the effects of HBD-3 on the antimicrobial activity of the supernatant of cell sheets against the microbial strains. Supernatant from the PDLC sheets transfected with the HBD-3 vector demonstrated significantly lower colony counts compared with those of the control cell sheets (P<0.001; Fig. 5B).

During the course of the experiment, no apparent adverse reactions, such as severe infection and suppuration, were detected around the periodontal tissue. Although root exposure was observed in most dogs of all the groups, and gingival recession was marginally more severe in the control group and the untreated group than in the test group (Fig. 6).

Different degrees of epithelium attachment to the cementum-enamel junction were observed among the three experimental groups. Less inflammatory infiltration was observed in the epithelium, and gingival and periodontal ligament fibers were arranged in a more orderly and neat fashion in the test group compared with those in the untreated group. In the test group (Fig. 7A) and the control group (Fig. 7B), a number of osteoblasts and osteocytes, as well as newly formed alveolar bone with trabeculae, were observed. By contrast, osteoclasts, visible erosion of gingival epithelia and obvious absorption of the alveolar crest were observed in the untreated group (Fig. 7C).

In the test and the control group, a vast amount of GFP-positive cells could be observed in the gingiva of the experimental area (Fig. 8A) and scattered small amounts in the alveolar bone (Fig. 8B).

TNF-α-positive cells, predominantly macrophages, were distributed throughout the gingival epithelium, periodontal membrane matrix and osteoclasts. In the test group, only weak positive staining was observed (Fig. 9A), whereas in the control group (Fig. 9B.) and the untreated group (Fig. 9C.), TNF-α expression was significantly elevated compared with the test group (P<0.05; Fig. 10). There was no significant difference between the untreated group and the control group (P>0.05; Fig. 10).

Similar to the TNF-α-positive staining patterns described above, IL-1β-positive cells were primarily observed in the gingival epithelium, periodontal membrane matrix, osteoblasts and osteoclasts (Fig. 11). In the test group, only weak IL-1β expression was observed, and this expression was significantly lower than that in the control group and the untreated group (Fig. 10; P<0.05). No differences in IL-1β expression between the untreated group and the control group were detected (P>0.05; Fig. 10).

A marker for bone tissue, OCN immunoreactivity was observed in cells embedded in the alveolar bone and surrounding tissues. In the test group (Fig. 12A) and the control group (Fig. 12B), the bone and the extracellular matrix of the periodontal ligament showed moderate to strong OCN-positive staining. In the untreated group, the extracellular matrix and some cells in the alveolar bone exhibited weak staining for OCN (Fig. 12C). OCN staining was significantly higher in the test group compared to the untreated group (P<0.05; Fig. 10). There was no significant difference in OCN staining between the test group and the control group (P>0.05; Fig. 10).