Live Porphyromonas gingivalis and Candida albicans Synergistically Induce RANKL in Osteoblast-like PDLFs But Not in Undifferentiated PDLFs

Abstract

Background/Aim

Human periodontal ligament fibroblasts (PDLFs) can acquire osteoblast-like characteristics within periodontal lesions and contribute to osteoclastogenesis through the expression of receptor activator of nuclear factor (NF) κB ligand (RANKL). However, the microbial and cellular conditions required for RANKL induction remain unclear. This study compared the microbial responsiveness of undifferentiated and osteoblast-like PDLFs to clarify the mechanisms regulating RANKL expression.

Materials and Methods

PDLFs were cultured in standard or osteogenic medium and stimulated with TLR2/TLR4 ligands, a Dectin-1 ligand, or infected with live or heat-killed Porphyromonas gingivalis (P.g.) and Candida albicans (C.a.). Gene expression was quantified by qRT-PCR, and the RANKL/osteoprotegerin (OPG) ratio was calculated. Cell viability was assessed using metabolic assays.

Results

Undifferentiated PDLFs failed to upregulate RANKL in response to TLR2/TLR4 ligands, a Dectin-1 ligand, or stimulation with live or heat-killed P.g. or C.a., even though TLR2 stimulation increased IL-17A expression. In contrast, osteoblast-like PDLFs exhibited increased cell viability and robust RANKL induction following stimulation with live C.a., and these responses were further enhanced by co-stimulation with live P.g. and C.a. Co-stimulation synergistically increased the RANKL/OPG ratio and IL-17A expression, whereas heat-killed microbes had no effect.

Conclusion

RANKL induction in PDLFs requires osteoblast-like differentiation and multifactorial stimulation provided by live microbes, including synergistic interactions between bacteria and fungi. These findings offer new insights into the mechanisms driving periodontal bone destruction.

Introduction

Periodontitis is characterized by chronic inflammation of periodontal tissues, leading to progressive alveolar bone destruction. A central molecular mechanism underlying this process is the promotion of osteoclast differentiation by the receptor activator of NF-κB ligand (RANKL) (1, 2). RANKL is primarily produced by T cells, osteoblasts, and fibroblasts, and the balance between RANKL and its decoy receptor osteoprotegerin (OPG) determines the extent of bone resorption (1, 2). An increased RANKL/OPG ratio has been reported in periodontal lesions and is associated with enhanced osteoclastogenesis (1).

Recently, periodontitis has been reconceptualized from a simple bacterial infection to a dysbiosis-driven disease involving complex interactions among bacteria and fungi (3). Notably, the coexistence of Porphyromonas gingivalis (P.g.), a keystone pathogen in periodontitis, and the oral commensal fungus Candida albicans (C.a.) has been reported to contribute to the severity of periodontal inflammation (4).

Human periodontal ligament fibroblasts (PDLFs), which are abundant in periodontal tissues, are considered a source of RANKL (5). However, the mechanisms that regulate RANKL expression in PDLFs remain unclear. P.g. lipopolysaccharide (P.g. LPS) has been reported to induce the expression of RANKL and OPG in PDLFs, thereby influencing the RANKL/OPG ratio (6). In contrast, we previously reported that P.g. LPS does not affect RANKL expression (7). Moreover, transcriptome analyses of PDLFs stimulated with P.g. LPS have shown that, although numerous inflammation-related genes are induced, changes in RANKL expression were not identified as a prominent pattern (8). Differences in stimulation conditions, cell lines or donors, and detection methods for membrane-bound versus soluble RANKL have been proposed as explanations for these discrepancies (1, 5). In addition, inflammatory cytokines, including IL-1β and TNF-α, enhance RANKL expression in PDLFs (9), suggesting that cooperation between microbial stimuli and inflammatory signals is important for RANKL induction. The Th17 cytokine IL-17A is known to strongly induce RANKL expression in osteoblasts and fibroblasts and is considered a key mediator of bone destruction in periodontitis (10). In contrast, IL-17F, another Th17-related cytokine, exhibits weaker biological activity and limited capacity to induce RANKL (11).

In addition, PDLFs possess the ability to differentiate into an osteoblast-like phenotype, and culture in an osteogenic medium induces characteristics similar to osteoblasts (12). Although changes in differentiation status are thought to influence RANKL expression, differences in microbial responsiveness between undifferentiated and osteoblast-like PDLFs have not been fully elucidated.

Based on these considerations, the present study aimed to investigate how microbial components and live and heat-killed bacteria and fungi affect RANKL expression in undifferentiated PDLFs and PDLFs differentiated into an osteoblast-like phenotype.

Materials and Methods

Reagents. Zymosan-depleted and ultrapure LPS from Porphyromonas gingivalis (P.g.) was purchased from InvivoGen (San Diego, CA, USA).

Cell culture. PDLFs (ScienCell, Tokyo, Japan) were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% heat-inactivated FBS and 1% penicillin-streptomycin (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) (13). Cells were maintained in 10-cm dishes at 37°C in a humidified 5% CO₂ incubator. PDLFs were cultured in α-MEM supplemented with Osteoblast-Inducer Reagent (Takara Bio, Kusatsu, Shiga, Japan) containing 1% (v/v) ascorbic acid and 2% (v/v) β-glycerophosphate for three days to evaluate the early changes associated with their transition toward an osteoblast-like phenotype while retaining their fibroblastic characteristics.

Strain and culture conditions for P.g.and C.a.. Strains P. g. ATCC 33277 and C.a. SC5314 were used. Culture and heat-inactivation of P. g. and C.a. were performed as described previously (13, 14). PDLFs were seeded in 12-well plates at a density of 1 × 10⁵ cells per well. Once the cells reached the desired confluence, they were infected with P.g. and C.a. at the defined MOI.

RNA isolation and quantitative reverse transcription PCR. Total RNA was extracted from the cells using NucleoSpin RNA (Takara Bio). Total RNA (500 ng) was then reverse-transcribed using ReverTra Ace reverse transcriptase (Toyobo, Osaka, Japan) with oligo(dT)₁₈ and random hexamer primers. Quantitative reverse transcription PCR (qRT-PCR) was performed using TB Green Premix Ex Taq II (Tli RNase H Plus; Takara Bio) on a LightCycler 480 (Roche Diagnostics, Basel, Switzerland). Pre-designed primer sets were obtained from Takara Bio; therefore, the primer sequences are not shown. The reaction conditions were as follows: pre-denaturation at 95°C for 30 s, followed by 40 cycles of denaturation at 95°C for 5 s and annealing at 60°C for 20 s. Gene expression levels were assessed using the ΔΔCt method. Results are presented as relative expression levels normalized to those of ACTB. The RANKL/OPG ratio was calculated as the ratio of the relative RANKL expression to that of OPG in the same sample.

Statistics. Statistical analyses were performed using GraphPad Prism 8 (GraphPad Software, San Diego, CA, USA). Comparisons between two groups were conducted using Student’s t-test. Comparisons between three or more groups were performed using Dunnett’s test. Statistical significance was set at p<0.05.

Results

TLR4 and TLR2 ligands do not induce RANKL expression in undifferentiated PDLFs. We first examined the effects of the TLR4 ligand P.g. LPS and the TLR2 ligand Pam3CSK4 on PDLFs. MTT [3-(4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide) assay showed that neither ligand significantly affected cell viability (Figure 1A). Under both unstimulated conditions and upon stimulation with P.g. LPS or Pam3CSK4, undifferentiated PDLFs exhibited extremely low RANKL expression (Figure 1B). These findings indicate that undifferentiated PDLFs do not induce RANKL expression in response to TLR2/TLR4 stimulation.

Figure 1.

Effects of TLR4 and TLR2 ligands on RANKL expression in undifferentiated PDLFs. (A) TLR4 and TLR2 ligands do not affect cell viability. Human periodontal ligament fibroblasts (PDLFs) were stimulated with Porphyromonas gingivalis (P.g.) lipopolysaccharide (P. g. LPS; TLR4 ligand; 0.1, 1, and 10 μg/ml) or Pam3CSK4 (TLR2 ligand; 0.1, 1, and 10 μg/ml) for 24 h. Cell viability was assessed using the MTT assay. (B) TLR4 and TLR2 ligands do not affect RANKL expression. PDLFs were stimulated with P. g. LPS (1 μg/ml) or Pam3CSK4 (1 μg/ml) for 24 h. RANKL mRNA levels were determined using quantitative RT-PCR (B). Mean±standard deviation values for one (triplicate) of the three independent experiments are shown. n=3, Dunnett’s test. n.d.: Not detected.

Undifferentiated PDLFs produce IL-17A upon TLR2 stimulation, but this does not lead to RANKL expression. We evaluated the expression of IL-17A, a cytokine involved in RANKL induction (10). P.g. LPS did not induce IL-17A expression (Figure 2A), whereas Pam3CSK4 significantly increased IL-17A levels (Figure 2B). However, this induction did not induce RANKL expression (Figure 1B).

Figure 2.

Effects of TLR4 and TLR2 ligands on IL-17A expression in undifferentiated PDLFs. (A, B) TLR2 ligand, but not TLR4, induces IL-17A expression. PDLFs were stimulated with Porphyromonas gingivalis (P. g.) LPS (1 μg/ml) or Pam3CSK4 (1 μg/ml) for 24 h. IL‑17A mRNA expression was quantified using RT-PCR. Mean±standard deviation values for one (triplicate) of the three independent experiments are shown. *Significantly different at p<0.05, Student’s t-test.

Live or heat-killed P.g. and C.a. do not induce RANKL in undifferentiated PDLFs. To evaluate more physiologically relevant stimuli, we examined RANKL expression following exposure to live or heat-killed P.g. and C.a. Heat-killed microbes were included because disruption of the cell wall is thought to enhance stimulatory activity (13, 15). Neither live nor heat-killed P.g. or C.a. affected cell viability (data not shown). None of these stimuli induced RANKL expression (data not shown).

Microbial ligands do not induce RANKL in osteoblast-like PDLFs differentiated in osteogenic medium. Because PDLFs in periodontal lesions have been reported to acquire osteoblast-like properties and express RANKL (16), we cultured PDLFs in osteogenic medium to induce an osteoblast-like phenotype and assessed their responsiveness to microbial ligands. To evaluate β-glucan-mediated Dectin-1 signaling by the cell wall components of C.a. without activating TLR2, we used purified β-glucan preparation that lacks TLR2-stimulating components (zymosan-depleted). Ligands derived from P.g. or C.a., or Pam3CSK4 did not affect cell viability (Figure 3A). In osteoblast-like PDLFs, low- RANKL expression was detected under basal conditions (Figure 3B); however, microbial ligands did not induce RANKL expression (Figure 3B).

Figure 3.

Effects of TLR4, TLR2, Dectin-1 ligands on RANKL expression in osteoblast‑like PDLFs. (A) TLR4, TLR2, Dectin-1 ligands do not affect cell viability. PDLFs differentiated in osteogenic medium were stimulated with microbial ligands derived from Porphyromonas gingivalis (P. g.) LPS (0.1, 1, and 10 μg/ml), Pam3CSK4 (0.1, 1, and 10 μg/ml), or zymosan (0.1, 1, and 10 μg/ml) for 24 h. Cell viability was assessed using the MTT assay. (B) TLR4, TLR2, Dectin-1 ligands do not affect RANKL expression. PDLFs differentiated in osteogenic medium were stimulated with P. g. LPS (1 μg/ml), Pam3CSK4 (1 μg/ml), or zymosan (1 μg/ml) for 24 h. RANKL mRNA expression was quantified using RT-PCR. Mean±standard deviation values for one (triplicate) of the three independent experiments are shown. n.d.: Not detected.

Live microbial stimulation induces RANKL expression in osteoblast-like PDLFs. When osteoblast-like PDLFs were stimulated with live or heat-killed P.g. or C.a., live C.a. alone and co-stimulation with live P.g. and live C.a. significantly increased cell viability (Figure 4A). Live P.g. alone did not alter cell viability (Figure 4A). In osteoblast-like PDLFs, live C.a. and co-stimulation with live P.g. and live C.a. significantly induced RANKL expression (Figure 4B). OPG expression was also increased by live P.g., live C.a., and their combination (Figure 4C). The RANKL/OPG ratio was significantly elevated upon co-stimulation with live P.g. and live C.a. (Figure 4D). In contrast, heat-killed microbes did not affect cell viability or RANKL or OPG expression (Figure 4A-C). Co-stimulation with live P.g. and live C.a. significantly increased IL-17A expression (Figure 4D).

Figure 4.

Effects of live or heat‑killed Porphyromonas gingivalis (P.g.) and Candida albicans (C.a.) on RANKL expression in differentiated PDLFs. (A) Live C.a. and co-stimulation with live P.g. and live C.a. increased cell viability. Differentiated PDLFs were stimulated with live or heat-killed P. g. (MOI: 1, 10, 100) or C. a. (MOI: 1, 10, 100), alone or in combination, for 24 h. Cell viability was measured using the MTT assay. (B-E) Co-stimulation with live P.g. and live C.a. significantly increased RANKL/OPG ratio and IL‑17A expression. Differentiated PDLFs were stimulated with live or heat-killed P. g. (MOI: 10) or C. a. (MOI: 10), alone or in combination, for 24 h. RANKL (B), OPG (C), and IL‑17A (E) mRNA levels were quantified using RT-PCR. The RANKL/OPG ratio (D) was calculated using RANKL and OPG mRNA expression levels. Mean±standard deviation values for one (triplicate) of the three independent experiments are shown. *Significantly different at p<0.05 versus control by Dunnett’s test.

Discussion

We investigated the mechanisms underlying RANKL induction by comparing microbial responsiveness of undifferentiated PDLFs and PDLFs that were differentiated into an osteoblast-like phenotype. Undifferentiated PDLFs did not induce RANKL expression in response to TLR2/TLR4 ligands, a Dectin-1 ligand, or stimulation with live or heat-killed P. g. or C.a. (Figure 1B). Although TLR2 stimulation induced IL-17A expression in undifferentiated PDLFs (Figure 2), RANKL expression remained unchanged (Figure 1B). In contrast, osteoblast-like PDLFs produced both IL-17A and RANKL following co-stimulation with live P.g. and live C.a. (Figure 4B and E). These findings indicate that IL-17A alone is insufficient for RANKL induction, and that both the differentiation state of the cells and multifactorial microbial stimulation are required.

A key finding was that in osteoblast-like PDLFs, heat-killed microbes did not affect cell viability or RANKL expression, whereas live C.a. alone or co-stimulation with live P.g. and live C.a. increased cell viability and induced RANKL expression (Figure 4A and B). Notably, the RANKL/OPG ratio was higher under co-stimulation than under stimulation with live C.a. alone (Figure 4D), supporting an emerging dysbiosis model of periodontitis (3). The limited stimulatory capacity of heat-killed microbes, despite exposure to pathogen-associated molecular patterns such as LPS and β-glucan, is likely due to activation of only single pathways such as TLRs or Dectin-1. Consistent with this hypothesis, isolated microbial components (LPS or β-glucan) also failed to induce RANKL (Figure 3). In contrast, live microbes release diverse factors including metabolic products, outer membrane vesicles, extracellular vesicles, proteases, and short-chain fatty acids, which can simultaneously activate multiple receptors (17). C.a. has been reported to enhance the activity of P. g. virulence factors and to interact with P.g. in biofilm formation (18). Therefore, the coexistence of P.g. and C.a. may synergistically amplify RANKL-inducing signals. Moreover, P.g. has been shown to alter host gene expression in distant tissues such as the hippocampus (19), supporting the idea that live microbial activity exerts broader biological effects than isolated components. Taken together, these findings suggest that, in vivo, RANKL induction requires not only the differentiation state of PDLFs, but also bacteria-fungus stimulation.

Study limitations. First, we did not assess the protein levels of membrane-bound or soluble RANKL; thus, the extent to which mRNA changes reflect functional RANKL remains unclear. Second, the specific microbial factors of live microbes responsible for RANKL induction in live organisms were not identified. Third, we used a monolayer culture model that does not fully recapitulate the complex microenvironment of periodontal lesions, including immune cells, vasculature, and mechanical forces. Despite these limitations, our findings demonstrate that RANKL expression and the RANKL/OPG ratio in PDLFs are regulated by three major factors: the cellular differentiation state, multifactorial stimulation unique to live microbes, and bacterial-fungal interactions. In particular, the observation that live C.a. enhances both cell viability and RANKL expression and that co-stimulation with live P.g. and live C.a. further increases the RANKL/OPG ratio provides new insights into the mechanisms of bone destruction in periodontitis.

Conclusion

Undifferentiated PDLFs did not produce RANKL in response to microbial ligands or stimulation with live or heat-killed microbes. In contrast, osteoblast-like PDLFs responded specifically to live P.g. and live C.a. Live C.a. induced RANKL expression, and co-stimulation with live P.g. and live C.a. synergistically enhanced the RANKL/OPG ratio and IL-17A expression. These findings indicate that RANKL induction in periodontal ligament cells requires osteoblast-like differentiation, multifactorial stimulation by live microbes, and bacterial-fungal interactions.

Conflicts of Interest

The Authors declare there are no competing interests.

Authors’ Contributions

Experimental work (cell culture, infection assays, viability assays): MA, KK, SO. Data processing and statistical analysis: MA, KK, SO, MI. Figure preparation and data visualization: MA, KK, SO, MI. Manuscript drafting: MA, MI. Critical revision of the manuscript: JH, HS. Overall supervision and project administration: MA, JH, HS, MI. All Authors read and approved the final manuscript.

Acknowledgements

The Authors would like to thank Editage for editing and reviewing this article for English language.

Funding

This work was supported by Grants-in-Aid for Scientific Research (KAKENHI) from the Japan Society for the Promotion of Science to M.I. (24K13267) and M.A. (23K09466).

Artificial Intelligence (AI) Disclosure

During the preparation of this manuscript, large language model (ChatGPT, OpenAI) was used solely for language editing and stylistic improvements in select paragraphs. No sections involving the generation, analysis, or interpretation of research data were produced by generative AI. All scientific content was created and verified by the authors. Furthermore, no figures or visual data were generated or modified using generative AI or machine learning-based image enhancement tools.