Periodontal disease severity has traditionally been assessed using clinical parameters

Periodontal disease severity has traditionally been assessed using clinical parameters like pocket probing depth, clinical attachment loss, bleeding on probing, and radiographic determination of alveolar bone loss. Most of these techniques were established more than five decades ago, and lack the capacity to identify highly susceptible patients at risk for disease progression [28,57]. Periodontitis is a highly complex disease, which hampers the development of rapid, accurate, diagnostic and prognostic tests. Nevertheless, the development of innovative diagnostic tests for periodontal disease remains a high priority.
A small number of miRNA studies related to periodontal disease have been performed (Table 2). Xie et al. [58] used microarray analysis to examine miRNA expression, and transcript levels of selected inflammatory-related miRNAs were confirmed by quantitative reverse transcription polymerase chain reaction (qRT-PCR). The study used gingival tissues from ten healthy subjects and ten patients with periodontitis. Levels of some miRNAs were more than five-fold higher in tissues from periodontitis patients than in control tissues. In addition, the possible regulation of Toll-lik receptors (TLRs) in periodontal inflammation by miRNA pathways was also proposed.
A pilot investigation was conducted to determine whether miRNA tropisetron was altered by obesity or periodontal disease, and whether there were any potential interactions between obesity and periodontitis that could involve miRNA modulation [59]. In this study, gingival biopsy samples were obtained from 20 patients, ten of who were non-obese (BMI<30kg/m2) and ten of who were obese (BMI>30kg/m2). Each group of ten patients contained five patients with chronic periodontitis and five periodontally healthy patients. This was the first trial to assess the mechanisms underlying the pathogenesis of periodontitis and a common chronic condition (obesity), as well as the interaction between the two diseases [60].
Stoecklin-Wasmer et al. [61] examined the occurrence of miRNAs in healthy and diseased gingival tissues and validated the in silico-predicted targets through mRNA profiling using whole-genome microarray analysis of the same specimens. Four miRNAs were significantly overexpressed, and seven significantly underexpressed, in gingival tissues compared to controls. Gene Set Enrichment Analysis (GSEA) identified 60 enriched miRNA gene sets with target genes involved in immune/inflammatory responses and tissue homeostasis. This was the first study to examine concurrent mRNA and miRNA expression in the same gingival tissues.
Only a handful of studies investigating miRNAs in gingival tissues have been reported to date. Studies of miRNAs and their relationships to periodontal disease will be improved in the near future by the use of diverse samples, including saliva, and this will allow the inter-relationships of periodontal disease with other systemic diseases to be ascertained. In addition, some studies using animal periodontitis models and dental stem cells have been conducted [62,63], and these will help determine the mechanisms underlying the modulation of specific candidate miRNAs.

Limitations of miRNAs as biomarkers
First, differentially expressed candidate miRNAs identified through pilot studies need to be validated. Most previous studies using tissues were performed with samples from small numbers of individuals without matching for potential confounding factors known to influence periodontitis susceptibility such as age and gender. Further validation studies using large well-characterised cohorts are required [60].
Second, although recent advances in molecular biology and high-throughput screening techniques have enabled researchers to characterise miRNA patterns in body fluids such as serum, plasma, and saliva on a large scale, this is limited by the lack of suitable endogenous controls for normalisation of salivary miRNAs. Recent research attempted to identify endogenous control miRNAs displaying minimal expression variability between samples [38], but endogenous salivary miRNA controls are required for future exploitation of miRNA datasets. In addition, saliva samples collected from the same individual can display considerable heterogeneity according to the collection method used, and standardised methods for sample collection should therefore be considered.