Introduction

Periodontitis is a chronic and inflammatory disease of bacterial etiology that results in loss of periodontal attachment [1]. In periodontitis, the gums detach from the teeth, and the supporting tissues are destroyed in the most advanced and severe stages, thereby leading to bone and tooth loss [2]. Moreover, this disease has been associated with several other important conditions, such as diabetes, cardiovascular diseases, and Alzheimer’s disease [3]. Periodontitis affects approximately 50% of the adult population worldwide [4] with varying degrees of severity and approximately 11% of this population has severe periodontitis [5]. In Colombia, 61.8% of the population has periodontitis, and 10.6% of cases are classified as severe periodontitis [6].

Microbiological and molecular studies were used to distinguish the microbiota associated with healthy periodontium from those associated with diseased periodontium [7]. For example, gram-positive and some gram-negative bacteria have been associated with a healthy state. Actinomyces and Streptococcus species are the first colonizers of pristine tooth surfaces and co-aggregate to form early dental biofilm. Notably, bacteria identified in healthy periodontal tissue samples include Peptostreptococcus, Gemella, Veillonella, Capnocytophaga, Neisseria, Rothia, and Corynebacterium [7,8,9]. Bacteria such as Porphyromonas gingivalis, Tannerella forsythia, and Treponema denticola have been identified in patients with periodontal disease. In addition, these patients had bacteria characterized by the presence of virulence factors associated with alteration in the tissue immune response (e.g., activation of destructive proinflammatory patterns or evasion of the immune response) and destruction of periodontal tissues (e.g., by lytic enzymes), promoting the onset and development of severe diseases [9, 10].

Other microbiological features of periodontitis are the presence of a greater diversity of bacteria and significant differences between the functional profiles of these microorganisms and those of healthy periodontium. In addition, a decrease, but not disappearance, in the proportion of some bacterial genera and species associated with healthy periodontium has been observed [9, 11]. These properties reflect the approach to disease onset: instead of a few pathogens, a dysbiotic polymicrobial community is considered to generate harmful inflammatory responses that perpetuate the imbalance or dysbiosis of the microbiota and its environment [11].

In addition to the accumulation of a diverse bacterial population in the sulcus of the gingiva, other risk factors are known to be involved in the occurrence of dysbiosis in systemically healthy individuals. These include unhealthy diet, smoking, and other habits that alter the environment of the tissues, i.e., the availability of oxygen and nutrients [12]. This promotes the emergence of a range of microorganisms, mainly bacteria, which act synergistically to enhance their metabolic activities, causing changes in the composition of the microbiota and thereby damaging periodontal tissues [7, 11, 13, 14].

Among the novel treatments being investigated for controlling periodontitis and maintaining a healthy periodontium, probiotics have been proposed as adjuvants to traditional treatments. Probiotics include live beneficial microorganisms that benefit the host when administered in sufficient quantity because they have properties that inhibit pathogenic microorganisms through the production of antimicrobial components or competition for host nutrients or tissue-binding sites. Furthermore, they may have immunomodulation properties, further strengthening the immune response [15].

The bacterial genera Lactobacillus and Bifidobacterium are most widely studied as probiotics to prevent periodontitis or as adjuvants during its treatment. Notably, these bacteria can inhibit periodontal pathogens and are safe for human consumption. However, even after examining the same probiotic strains, not all studies have discovered beneficial or consistent outcomes in clinical indices and/or in controlling the pathogenic microbiota [16,17,18]. This can be explained by various factors, such as the different administration routes, doses, and quantities; combinations with other probiotics; administration times; and/or sample sizes of clinical studies [16,17,18]. In addition, these bacteria are not normally abundant in periodontal tissues [8, 11], and some bacteria are of intestinal origin, which may affect their ability to colonize the oral cavity [19]. This further limits their effectiveness as probiotics; thus, further studies are needed to identify or investigate probiotics suitable for this tissue and periodontal health.

Bacteria that are more frequently in the microbiota of the healthy periodontium and have probiotic characteristics could be analyzed as possible probiotic candidates [20, 21]. Therefore, this review aimed to identify bacteria with probiotic characteristics in the microbiota of the healthy periodontium for treating periodontitis.

Material and methods

Type of study and search strategy

We conducted a systematic review of the literature as per the updated quality criteria of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement. The PubMed, ScienceDirect, Scopus, and Ovid databases were used to search for articles from July 2012 to April 2022 using both Medical Subject Headings (MeSH) and non-MeSH terms (Table 1); in addition, the associated filters for year and type of study were used. Furthermore, we conducted a manual search of information.

Table 1 Keywords and combinations according to the database
Full size table

The PICO format, Population (healthy population), Intervention (metagenomic analysis), Comparison (patients with periodontitis), Outcome (group of potential probiotics for the treatment of periodontitis), was used to formulate our clinical question: Which bacteria with probiotic characteristics for the treatment of periodontitis predominate in the microbiota of the periodontium of healthy patients? Inclusion criterion was original studies published in English that identified microorganisms present in subgingival tissue samples of healthy individuals using the high-throughput sequencing technique of the 16S ribosomal gene. Studies that did not identify bacteria in samples from healthy adults and patients with periodontitis in the same analysis and articles that did not report the nationality and/or sex of the patients were excluded. In addition, we selected studies that excluded patients with systemic diseases, those treated with antibiotics or other types of drugs, or those who were receiving periodontal treatment for at least 3 months before collecting the subgingival tissue samples.

Selection of the studies and collection of data

First, studies with unique titles were identified, their abstracts were read, and, eventually, those related to the purpose of the present study were selected. The selected studies were then thoroughly read to verify compliance with the selection criteria, leaving the final articles for review. This process was carried out by two researchers (MPAD and DM).

Finally, the studies were synthesized using the defined variables, including authors, origin, and type of the study; clinical parameters of the patients; selection criteria; analyzed samples; amplified sequence; and periodontal tissue bacteria identified in healthy individuals.

Level of evidence and risk of bias

The quality of the articles included in this review was assessed by two researchers (MPAD and CF) using the Joanna Briggs Institute (JBI) scale for case–control and cross-sectional studies. This scale enabled the evaluation of reliability, relevance, and results of the selected articles.

Results

Selection of studies and characteristics of the studies

Overall, 659 unique articles were found based on the information search; of these, 23 were selected for full reading, and 12 that met the selection criteria were eventually selected for review (Fig. 1). Most investigations were conducted in the United States (n = 4), followed by China (n = 2) and Spain (n = 2). The remaining four studies were conducted in Chile, Germany, Japan, and the Czech Republic (Table 2). The most common study types were case–control (58.3%, n = 7) and cross-sectional (41.6%, n = 5) studies (Fig. 1). Neither of the selected studies was a randomized clinical trial (Table 2).

Fig. 1
figure 1

Information flowchart of the different stages of a systematic review

Full size image
Table 2 Characteristics of the included studies
Full size table

Regarding patient selection, in addition to the exclusion criteria established in this study, some authors excluded pregnant and lactating patients, patients who had recently undergone prophylactic procedures, patients receiving oral antiseptics, smokers (only one study), patients with a certain number of natural teeth, etc. (Table 2).

Clinical criteria for the selection of healthy individuals were as follows: probing depth (PD) of < 2 to < 4 mm, clinical attachment level/loss (CAL) of 0 to < 4 mm, no bleeding on probing (BOP) or BOP in at least 10–20% of measured sites, no radiographic evidence of alveolar bone loss, and/or no evidence of inflammation. In most studies, samples were collected using paper points (66.7%, n = 8), whereas only two studies used curettes for sample collection (Table 2).

Regarding the most used amplified regions to identify bacteria, V1–V3 were predominantly used, followed by V4 (Table 2). Other regions used were V1–V2, V3, V3–V4, V4–V5, and V7–V9.

Bacteria of periodontal healthy tissue

Bacteria that were frequently identified by their presence and/or abundance in healthy periodontal tissue in the different reviewed studies included those from the genera Acinetobacter, Actinomyces, Aggregatibacter, Capnocytophaga, Corynebacterium, Gemella, Granulicatella, Haemophilus, Lautropia, Leptotrichia, Neisseria, Porphyromonas, Prevotella, Rothia, Streptococcus, and Veillonella. In particular, the species identified in each of these genera were Acinetobacter junii, A. antiviralis, A. baumannii, A. calcoaceticus, A. rhizosphaerae, Actinomyces naeslundii, A. viscosus, A. massiliensis, A. meyeri, Aggregatibacter aphrophilus, Capnocytophaga gingivalis, Corynebacterium matruchotti, Gemella morbillorum, Gemella haemolysans, Granulicatella adiacens, Haemophillus parainfluenzae, H. haemolyticus, H. parahaemolyticus, H. sputorum, Lautropia mirabilis, Neisseria lactamica, N. flava, N. subflava, N. oralis, Porphyromonas catoniae, Prevotella histicola, P. nigrescens, Rothia aeria, R. dentocariosa, Streptococcus sanguinis, S. mitis, S. intermedius, S. oralis, S. tigurinus, S. infantis, S. gordonii, Veillonella atypica, and V. dispar/V. parvula (Table 2, Fig. 2).

Fig. 2
figure 2

Summary of bacteria of healthy periodontal tissues as candidates of probiotics

Full size image

Bacteria like A. viscosus, A. naeslundii, G. haemolysans, H. parainfluenzae, P. nigrescens, R. dentocariosa, S. sanguinis, S. mitis, S. oralis, S. gordonii, and S. intermedius have characteristics like probiotics because they permit modulate the periodontal pathogenic microbiota. However, some strains have been associated with oral or systemic infections, particularly in patients with immune deficiencies. Also, these strains have shown antibiotic resistance (Table 3, Fig. 2).

Table 3 Bacteria that were frequently identified by their presence and/or abundance in healthy periodontal tissue and probiotics properties
Full size table

Level of evidence and risk of bias

Based on the JBI tool assessment, most studies met most quality criteria. Only two case–control studies and one cross-sectional study did not meet the identification criteria for confounding factors. Furthermore, the findings of one case–control and two cross-sectional studies were unclear in identifying the confounding factors (Appendix A1 and Appendix A2).

Discussion

Bacteria must meet certain requirements to be candidates for probiotics, including the ability to colonize the tissues in which they will act [76]; therefore, it is important to identify commensal bacteria that are common or abundant in healthy periodontal tissues [77]. In the present review, we found that both gram-positive and gram-negative bacteria were associated with the subgingival biofilm of healthy periodontium in the reviewed studies. These bacteria mainly comprised facultative anaerobes, including those from the genera Acinetobacter, Actinomyces, Aggregatibacter, Capnocytophaga, Corynebacterium, Gemella, Granulicatella, Haemophilus, Lautropia, Leptotrichia, Neisseria, Porphyromonas, Prevotella, Rothia, Streptococcus, and Veillonella.

Furthermore, bacteria of genera, such as Gemella (G. haemolysans), Granulicatella (G. adiacens), Haemophilus (H. parainfluenzae), Lautropia (L. mirabilis), Neisseria (N. flava, N. subflava, and N. oralis), Rothia (R. aeria and Rothia dentocariosa), Streptococcus (S. sanguinis, S. mitis, S. oralis, S. gordonii, and S. intermedius), Porphyromonas (P. catoniae), and Prevotella (P. nigrescens), were also reported as bacteria specific to patients with healthy periodontium in investigations excluded from this review based on the defined selection criteria [7, 10, 78,79,80,81,82,83]. In contrast, other genera, such as Acinetobacter (A. junii, A. antiviralis, A. baumannii, A. calcoaceticus, and A. rhizosphaerae), have been relatively frequently associated with disease conditions [84, 85].

Another frequently reported genus is Actinomyces spp., with the species having favorable properties for periodontal health, e.g., A. viscosus and A. naeslundii, and the species associated with periodontal disease, e.g., A. massiliensis, A. odontolyticus/meyeri, and, again, A. naeslundii [86].

Other bacteria that lead to discrepancies since they are associated with health or disease conditions are Aggregatibacter (A. aphrophilus), Capnocytophaga (C. gingivalis), Corynebacterium (C. matruchotti), and Veillonella (V. atypica, V. dispar/V. parvula) [7, 87]. These discrepancies may be attributed to the fact that the microbiome of healthy periodontal tissue demonstrates greater interpersonal variability than that of patients with periodontitis or gingivitis. This may indicate that the microbiome's composition can be influenced by environmental factors. Other factors that may affect the composition are as follows: diagnosis by selecting healthy controls, which may have caused an imbalance in the microbiota, as such individuals have no clinical symptoms; differences in criteria or clinical measures used to define healthy individuals, as noted in this review; differences in the individual experiences of the examiners; and differences in the sequenced region [8].

Another characteristic of probiotics is the ability to inhibit pathogens by producing antimicrobial agents or competing for nutrients, space, or environmental conditions [88]. Notably, the metatranscriptome analysis of subgingival plaque collected from healthy individuals revealed the expression of genes associated with competition, bacteriocin production (mutation), and quorum sensing [89]. Of the bacteria identified in the present review, A. viscosus, A. naeslundii, G. haemolysans, H. parainfluenzae, P. nigrescens, R. dentocariosa, S. sanguinis, S. mitis, S. oralis, S. gordonii, and S. intermedius have been reported as those capable of inhibiting the growth of periodontopathogens [50, 68, 90, 91]. For example, G. haemolysans produces an inhibitor of P. gingivalis; however, it does not inhibit other periodontal pathogens, such as F. nucleatum and T. denticola [45]. Notably, H. parainfluenzae has been reported to have the ability to inhibit the adhesion of periodontopathogenic bacteria, particularly P. gingivalis [48]. Streptococcus strains secrete substances that cause the lysis of P. gingivalis or eliminate its ability to generate a biofilm by affecting the expression of fimA and mfa-1, which are the components of the fimbria of P. gingivalis [66]. Regarding P. nigrescens, antimicrobial activity has been demonstrated against Actinomyces spp., P. gingivalis, P. intermedia, and T. forsythensis by a bacteriocin called nigrescin [68].

Bacteria must also meet other requirements to be candidates for probiotics; these include being harmless, having no pathogenic genes, and showing no characteristics of acquired antibiotic resistance [92, 93]. Some strains of the bacteria that were identified as possible candidates in the present review were found to cause endocarditis and opportunistic infections, as observed with G. haemolysans and H. parainfluenzae [46, 49]. In addition, P. nigrescens belongs to a genus of bacteria associated with pulp infections, periodontitis, and abscesses of dental and periodontal origin, and it is resistant to antibiotics [94]. In a study comparing different species of Prevotella (P. intermedia, P. nigrescens, and P. melaninogenica), P. nigrescens showed the greatest ability to resist amoxicillin/clavulanic acid, amoxicillin, and clindamycin, and it had more resistant genes to beta-lactams, tetracyclines, and lincosamides [94].

The genus Streptococcus includes a wide variety of species previously established as probiotics for human consumption owing to their proven safety [95, 96]. These bacteria are easy to culture under in vitro conditions. For example, the bacterium S. oralis, specifically the KJ3 strain, is used in a commercial product with proven safety for human consumption [63]. In addition, S. mitis strain YIT 12322 has been studied and found safe for humans [97]. Therefore, Streptococcus is a genus with strains such as those identified in the actual review (S. oralis, S. mitis, S. sanguinis, and S. gordonii) that could be used in future clinical trials. The strain KJ3 of S. oralis, along with S. uberis strain KJ2 and S. rattus strain JH145, constitutes a commercial product named ProBiora®. This product has shown to reduce pathogenic bacteria like S. mutans, Campylobacter rectus, and P. gingivalis in saliva and dental biofilm from young individuals [98]. Another strain of S. oralis is the sub-species dentisani strain 7746, which can adhere to the tooth to stabilize the pH through arginolytic activities and the production of bacteriocin which inhibits the growth of cariogenic bacteria and periodontopathogens [99]. Through a clinical study, strain 7746 was applied to a dental alginate ferrule every 48 h for one month in patients between 18 and 65 years old. The authors found improved clinical parameters like dental biofilm index, gingival index, and salivary flux [100].

Other bacteria suggested as probiotic candidates are S. mitis and S. sanguinis, which are primary colonizers of the oral cavity; these bacteria are characterized by inhibiting the adhesion of P. gingivalis, A. actinomycetemcomitans, P. gingivalis y P. intermedia to producing H2O2 [101]. These bacteria were used in replacement therapy for periodontitis in an animal model (dog), where the authors found, after a treatment of 12 weeks, a significant increase in the bone level in the treated periodontal pocket [102].

Finally, another probiotic candidate is S. gordonii which promotes the homeostasis of the biofilm through the inhibition of periodontopathogens like T. denticola, and P. gingivalis [103]. Also, S. gordonii causes a beneficial effect on the oral epithelium, stimulating the immune system (activating antigen-presenting cells) [104]. S. gordonii can modulate the expression of genes of gingival epithelial cells and decrease the expression of proinflammatory cytokines, indicating an anti-inflammatory potential [105].

However, some strains of the oral streptococci have been isolated from patients with infective endocarditis and other opportunistic infections [106, 107], underscoring the importance of evaluating the safety of bacterial strains as candidate probiotics before conducting clinical trials.

For the prevention or adjuvant treatment of periodontal disease through probiotics, the candidates must be able to maintain or restore the periodontal microecological balance. For this reason, it is necessary to combine several bacterial strains, which, in addition to possessing properties for the control of periodontopathogenic bacteria and regulation of the immune response of the host, comply with the corresponding safety for its use and ease of production at the industrial level.

This study has some limitations. First, not all studies included some factors affecting the microbiota, and therefore, they did not exclude patients who were smokers or pregnant and/or breastfeeding. Second, not all studies used the same number of healthy teeth, similar durations of treatment with antibiotics or other drugs, or similar clinical measures that distinguished between healthy individuals and patients with periodontitis. Finally, although species of bacteria with probiotic properties were identified, not all strains of the same species have the same properties. Nevertheless, this information allows us to focus our studies on the species identified, to isolate and study the most appropriate strains as probiotics.

Conclusion

There are several important variations in the microbiota associated with healthy periodontal tissue—an area where several of these bacteria have also been reported in disease conditions. This is not only related to the methodological differences between the studies but also to the characteristics of the patients and the nature of their native bacterial strains as well as their virulence factors and the metabolic interactions triggered. Therefore, it is important to reconsider the probiotic strategy because these agents favor the presence of beneficial native bacteria in these tissues.

This study identified bacteria that could be key candidates for maintaining the oral microbiota of periodontal tissues owing to their association with health conditions and their antimicrobial properties. These bacteria include those of the genus Streptococcus, such as S. sanguinis, S. oralis, and S. mitis, which have been studied as probiotics for oral health, but these bacterial strains must be selected based on their safety for human consumption.