HGF is genetically heterogeneous and five subtypes are described corresponding to different candidate susceptibility loci, including gingival fibromatosis Types 1–5 (GINGF1 to GINGF5)[42–46]. Besides, the son of sevenless-1 gene (SOS1, GINGF1)[42] and the RE1-silencing transcription factor gene (REST, GINGF5)[46, 47] were clearly associated with the occurrence of HGF [48, 49]. Moreover, ALK, CD36, and ZNF862 (zinc finger protein 862) are new genes found recently, which are also associated with HGF [50, 51]. Histologically, HGF is characterized by proliferative fibrous overgrowth of the gingival tissues and causes enlargement of the attached gingival tissue[7]. This fibrosis is characterized by densely arranged collagen bundles, numerous fibroblasts, and connective tissue that is hyperkeratotic with elongated rete ridges and relatively avascular. Small calcified particles, ulceration of the overlying mucosa, islands of osseous metaplasia, and inflammation can be occasionally observed[49, 52]. Many studies have attempted to explore the pathogenesis of HGF currently, Gao Q et al. found that KCNQ1, a member of the voltage-gated potassium channel family, played a critical role in the pathogenesis of HGF, promoted fibrogenic activity and facilitated extracellular matrix (ECM) synthesis and accumulation, and was upregulated in gingival tissues derived from HGF patients compared with normal gingival tissues, which may be a novel target for HGF therapy [53]. Another possible pathogenesis is the inflammatory stimulus of teeth eruption that would induce the expression of transforming growth factor ß1(TGFß1), more sub-epithelial fibroblast proliferation, greater collagen, and fibronectin synthesis, reduction in the matrix metalloproteinases (MMPs), which could degrade collagen, increasing in gingival connective tissue synthesis coupled with inhibiting connective tissue decomposition[54–57].
Periodontitis already affects 1,087.37 million people worldwide according to the Global Burden of Disease Study (GBD) 2019, with a global prevalence of 50%[58]. The incidence of periodontitis is approximately 47% in adults aged 30–65 years[59]. Chronic periodontitis is considered a slowly infectious progressing disease resulting in inflammation within the supporting tissues of teeth, progressive attachment loss, and bone loss. This inflammatory disease progresses more rapidly with severe systemic conditions or environmental factors, such as diabetes and smoking[60]. While aggressive periodontitis is another genetically inherited disease, the localized form of aggressive periodontitis is characterized by incisor and first molar bone loss with interproximal attachment loss[21, 61], whereas periodontal tissue destruction involves other areas in the dentition in the generalized form of aggressive periodontitis, which occurred around puberty with a lack of local factors such as heavy amounts of plaque and calculus in patients with good oral hygiene[29]. There have been some cases reported regarding the co-existence of HGF with chronic and aggressive periodontitis, but the detailed pathogenesis mechanisms are still not known [62].
By bioinformatics analysis, we obtained differentially expressed genes and ceRNA associated with HGF and periodontitis. GSE10334, GSE16134, and GSE4250 datasets were used and IL6, FLG2, and KRT2 were included in ceRNA after calculation and analysis. Interleukin (IL) -6 is a significant proinflammatory cytokine modulating the development of various human diseases, including chronic periodontitis[63, 64]. A previous study reported that in the gingival tissue of chronic periodontitis, the presence of sIL-6R increases the binding of IL-6 to human gingival fibroblasts [65]. Additionally, IL-6 could accelerate gingival fibroblasts to produce collagenolytic enzymes, resulting in tissue destruction that exists in periodontitis. Interestingly, TGF-β1 and IL-6 are also produced in greater amounts by HGF fibroblasts, promoting an increase in type I collagen and a decrease in MMP-1 and MMP-2 expression in fibroblasts from HGF [56]. Besides, ECM components of HGF, such as type I collagen and fibronectin, are produced at higher levels in HGF fibroblasts and these cells exhibit a doubled proliferation rate when compared with control group cells[25]. Hence, there is more sub-epithelial fibroblast proliferation and greater collagen and fibronectin synthesis in HGF[66], however, IL-6 may affect the progression of both HGF and periodontitis, but needs to be further confirmed.
Although there has no literature reported that Filaggrin 2(FLG2) and Keratin 2(KRT2) have a relationship with HGF, these two key genes are involved in skin and epithelial keratosis, which reminds us they may be joined in the development of HGF. FLG2 was expressed throughout the cornified cell layers and colocalized with corneodesmosin that plays a crucial role in maintaining cell-cell adhesion in the epidermis and was required for proper cornification and to protect the barrier in the skin, while a recent study reported that the absence of FLG2 could lead to the peeling phenotype in the skin of patients[67]. Besides, FLG2 is related to a generalized ichthyotic form of peeling skin syndrome and has a significant impact on the mechanical strength and integrity of the stratum corneum of the epidermis[68]. KRT2, a member of the keratin gene family and a structural constituent of the cytoskeleton, is expressed largely in the upper spinous layer of epidermal keratinocytes and is associated with keratinocyte activation, proliferation, and keratinization[69, 70]. These two molecules have not been investigated whether they are involved in the pathogenesis of HGF and periodontitis, which may be a new target for further study.
MicroRNAs(miRNAs) are short noncoding RNAs, with a length ranging from 20 to 23 nucleotides, that act as posttranscriptional blockers by adhering to mRNAs and silencing certain target genes[71, 72]. Hsa-miR-149-5p, miR-760-3p, and miR-376c-3p were screened out as key miRNAs involved in the current ceRNA network. Hsa-miR-149-5p was identified as a crucial determinant of differential effects in periodontitis and periimplantitis by high-throughput sequencing[73, 74]; hsa-miR-149-5p also can regulate the IL-6 expression in chronic obstructive pulmonary disease and is associated with inflammation[75]; Another study also reported that hsa-miR-149-5p promotes IL6 transcription in macrophage to increase macrophage-mediated inflammation in the progression of abdominal aortic aneurysms[76]. Furthermore, miR-760-3p might be involved in inflammation in an oxygen and glucose deprivation/reoxygenation(OGD/R) cell model by regulating Map3k8 to activate the NF-κB pathway in cerebral ischemia[77]; has-miR-376c-3p, a key miRNA in our ceRNA network, is identified as one of the differentially expressed miRNAs and may become novel diagnostic biomarkers and therapeutic targets for coronary artery disease in a bioinformatics analysis[78].
LncRNAs, more than 200 nucleotides, could compete for miRNA-mRNA binding and thus affect the negative regulation of gene expression by sponging with miRNAs[79]. It has been linked to the pathophysiology of various disorders and can influence gene expression at the transcriptional and posttranscriptional stages[71, 80, 81]. Additionally, lncRNA KCNQ1OT1, nuclear enriched abundant transcript 1 (NEAT1), lncRNA LRRC75A-AS1, and HELLPAR are key long non-coding RNAs involved in our ceRNA network to bridge the connection of HGF and periodontitis.
KCNQ1OT1 (potassium voltage-gated channel subfamily Q member 1 overlapping transcript 1), a crucial lncRNA in our ceRNA network, is involved in regulating lipid metabolism, atherosclerosis, and inflammation[82]. LncRNA KCNQ1OT1 regulated high glucose-induced proliferation, oxidative stress, ECM accumulation, and inflammation in human glomerular mesangial (HGMC) cells via the miR-147a/SOX6 axis in diabetic nephropathy (DN)[83], LncRNA KCNQ1OT1 can also target miR-214, and regulate ALP, BMP2, Runx2, OPN, and OCN in osteogenesis and related disorders, which may include osteoporosis and periodontitis [84]. For instance, KCNQ1OT1 could positively regulate the osteogenic differentiation of bone marrow mesenchymal stem cells (BMSCs) by acting as a ceRNA to regulate BMP2 expression through sponging miR-214[85]; down-regulation of lncRNA KCNQ1OT1 could inhibit the proliferation and osteogenic differentiation of human periodontal ligament stem cells (hPDLSCs) by targeting the up-regulated expression of miR-24-3p[86].
NEAT1 is a nuclear-enriched lncRNA and an essential scaffolding factor of nuclear paraspeckles[87], which is also related to inflammatory reactions. NEAT1 promotes the production of inflammatory cytokines such as IL-6, IL-8, and matrix metalloprotease 13 (MMP13) in synoviocytes [88]. NEAT1 is a potential therapeutic target for smoking-related periodontitis because NEAT1 can decrease autophagy flux, which is pivotal for inflammation factor production in nicotine-treated PDLSCs [89]. It was worth noting that NEAT1 was significantly highly expressed in chronic periodontitis tissues and LPS-induced PDLSCs, silencing of NEAT1 could protect PDLSCs against LPS-induced inflammation and apoptosis by targeting the miR-200c-3p/TRAF6 axis, thereby contributing to alleviating the progression of chronic periodontitis and may act as a potential therapeutic target for chronic periodontitis therapy in clinical applications[90]. In addition, a low level of NEAT1 was observed in the peripheral blood samples, but with a high level in tissue samples obtained from patients with periodontitis when compared with controls, which confirmed its pivotal value during the periodontitis diagnosis and treatments [91].
Although there has no study to report lncRNA LRRC75A-AS1 involve in HGF and periodontitis, it plays an important role during some other inflammation response. LncRNA LRRC75A-AS1 may exert crucial roles in modulating the inflammation response during the initiation and progression of breast cancer[92]; LRRC75A-AS1 can regulate the expression of tight junction (TJ) proteins through LRRC75A, affecting the development of inflammation in Escherichia coli-introduced cell model of bovine mastitis [93].
The different mRNAs, miRNAs, and lncRNAs in our ceRNA networks point out there have a close relationship between HGF and periodontitis, and these RNAs could be used as new biomarkers for diagnosis, providing a new therapy method, and exploring a novel way to investigate the pathophysiology of HGF and periodontitis. However, there still have some limitations in our study, such as the small sample size of GSE4250 which only have 4 HGF patients due to its rare incidence; in addition, follow-up experimental confirmation is also needed to confirm our bioinformatics predictions. Finally, our study provided a different perspective to evaluate the mechanism of HGF and periodontitis, and these results will help us to better understand, diagnose, and treat these two diseases in near future.