Cultivated and not-yet-cultivated bacteria in oral biofilms

Abstract

This review gives an overview of the bacterial diversity of cultivated and not-yet-cultivated bacterial species in oral biofilms. Examples are given from the healthy oral cavity of youngsters, adults, and the elderly; caries in primary and permanent teeth; root caries in the elderly; subgingival plaque; aggressive periodontitis; chronic periodontitis; necrotizing ulcerative periodontitis; halitosis; noma; endodontic infections; and spreading odontogenic infections. Transfer of biofilm bacteria to the blood is also discussed. Techniques used for identifying these organisms are mainly PCR, cloning, and 16S rRNA gene sequencing, as well as microarrays. As much as 50% or more of the microbiota in oral biofilms cannot yet be cultured. This may have significant implications for our knowledge of the pathogens in major biofilm infections in the oral cavity such as caries, periodontitis, peri-implantitis, and mucositis. Furthermore, several bacterial species not traditionally believed to be oral pathogens have also been shown to be associated with disease.

• oral biofilm
• PCR
• cloning
• 16S rRNA gene sequencing
• not-yet-cultivated bacteria

Introduction

Traditionally, the oral microbiota has been studied using culture techniques. The initial idea of culturing was that all major components of the microbiota would be identified. This was considered particularly important for oral infections involving ‘unusual’ pathogens. Culturing is also a prerequisite for assessing antibacterial resistance. However, it is expensive, technique-sensitive, and requires personnel with considerable knowledge in microbiology [1]. This particularly applies to fastidious microorganisms of the oral cavity, most of which are strictly anaerobic. There is also a problem with choice of media and atmosphere for culturing, and optimal collection and transport of samples containing anaerobic bacteria to the laboratory are not always carried out [2]. Following the development of anaerobic techniques for culturing of oral bacteria, over 250 species were isolated, characterized, and named [3]. Direct microscopic examination such as dark-field or phase-contrast microscopy has clearly shown that there are several microorganisms present in dental plaque that cannot be cultured. In 1963, Socransky et al. [4] suggested that only 50% of the microbiota in the oral cavity had been cultured. For example, it was clear that the medium- and large-sized spirochetes [5], [6] had not yet been cultivated.

A breakthrough in the detection of known oral bacterial species, even fastidious ones, was the application of the checkerboard technique developed by Socransky et al. in 1994 [7], by which large numbers of plaque samples could be handled and a panel of 40 bacterial species identified simultaneously. These authors found that the pathogenicity of the oral microbiota could be better understood by focusing on bacterial consortia rather than on a single bacterial species such as Porphyromonas gingivalis, Tannerella forsythia or Aggregatibacter actinomycetemcomitans (specific plaque theory). They suggested that the most pathogenic bacteria in periodontitis were those comprising the so-called red complex, namely P. gingivalis, T. forsythia, and Treponema denticola. These species depend on earlier colonization of the periodontal pocket by pathogenic organisms called the orange complex. These and other complexes were identified by Socransky et al. [8] after analyzing over 13 000 dental plaque samples from 185 patients using whole DNA genomic probes in checkerboard assays. However, these assessments were based on the use of probes targeting microorganisms that had been previously cultured.

16S rRNA gene sequencing for identification of cultivable and not-yet-cultivated species

It is noteworthy that 50% of the oral microbiota cannot be cultured because the major pathogens in oral diseases have been proposed based on knowledge from culture studies. Also therapeutic recommendations in terms of antibiotics are based on cultivated species alone. Fortunately, knowledge on the oral microbiota is much better than that from other environments, e.g. soil and sediments, where only 1% of the species have been cultivated so far [9].

By using the polymerase chain reaction (PCR) and cloning strategies that target bacterial 16S rRNA genes, it is possible to determine the bacterial composition of any ecological site. Briefly, DNA is amplified by PCR using conserved primers for the 16S rRNA genes. The 16S rRNA gene amplicons are then cloned into Escherichia coli and the cloned inserts are sequenced to determine species identity [3]. Species, or more precisely phylotypes, are then identified with >98.5% identity. Strains or clone sequences that have <98.5% identity to previously defined species or clones are regarded as representatives of new species. This approach has been used in clinical studies to define the association of single or sets of bacterial species with health and disease in the oral cavity.

Microarray analysis for assessment of oral microbiota

Based on 16S rRNA gene sequencing, a microarray system (HOMIM) was recently developed for detection of about 300 of the most prevalent oral bacterial species, including those that cannot yet be grown in vitro (Figure 1) (http://mim.forsyth.org/). The system can be used to compare bacterial associations in health vs disease, monitor effects of therapy on the oral ecology, and perform microbial perturbation studies.

Figure 1.  Microarray (HOMIM) illustrating the bacterial profile of a subgingival plaque sample.

Oral biofilm

Biofilm (plaque) in the oral cavity is located both on the soft and hard tissues. Even the healthy mucosa is colonized with a biofilm. It is often associated with oral diseases such as dental caries, marginal and apical periodontitis, peri-implantitis, and mucositis. Biofilms are complex structures consisting of pure or mixed microcolonies surrounded by a glycocalyx [10]. The latter is primarily composed of exopolysaccharides produced by resident bacteria. There are water channels inside the biofilm that permit passage of nutrients and waste to the overlaying bathing fluid. Inside the biofilm, cell-to-cell communication and transfer of genetic information are common and enable changes in the biofilm bacteria in response to their environments. Furthermore, bacteria in biofilms are protected against phagocytosis and other host mechanisms. It is important to recognize that bacteria in biofilms behave differently from planktonic microorganisms, which are not organized, but rather free floating. This overview deals with the microbiota of biofilms in different subject groups and habitats with different health status encompassing both cultivable and not-yet-cultivated microorganisms.

Healthy oral cavity

Youngsters and adults

Most studies on the oral microbiota have been devoted to examining the cause of disease. However, before we understand the role of pathogenic bacteria, we should have a better understanding of those microorganisms associated with health. In a recent study [11], nine oral sites (tongue dorsum, lateral sides of the tongue, buccal epithelium, hard palate, soft palate, supragingival plaque of tooth surfaces, subgingival plaque, maxillary anterior vestibule, and tonsils) were sampled in five patients (23–55 years old) with no signs of mucosal disease. 16S rRNA genes were amplified, cloned and transformed into E. coli. Inserts of 16S rRNA genes were sequenced to determine species identity or closest relatives. Among the 2589 clones analyzed, 141 species were detected. Of these, over 60% had not been previously cultivated. Thirteen new phylotypes were identified. Overall, genera such as Gemella, Granulicatella, Streptococcus, and Veillonella were commonly detected. Some species were subject-specific and were detected in most sites, others were site-specific. There were 20–30 different predominant species in most sites, and the number of predominant species from all 9 sites varied from 34 to 72.

Elderly people

Recently, the bacterial diversity of the oral cavity in the elderly (73–93 years old) was investigated [12]. Samples were collected from the following sites: tongue dorsum, mucosa of the buccal fold, hard palate, supragingival plaque from sound root surfaces, and subgingival plaque from the same roots. A 16S rRNA gene-based bacterial microarray, the Human Oral Microbe Identification Microarray (HOMIM) was used for the simultaneous detection of approximately 300 bacterial species, including not-yet-cultivated bacterial species. Overall, 175 species and clusters were detected that represented eight phyla. Species belonging to the genera Streptococcus, Veillonella, and Fusobacterium were common to all sites. The number of species detected per subject varied from 51 to 81. About 40 species showed significant association with at least one of the sites. The cheek and palate regions demonstrated the highest bacterial diversity. The findings revealed a highly diverse oral bacterial flora of the elderly that was to a large extent site-specific rather than subject-specific.

Caries in primary and permanent teeth

In a reverse-capture checkerboard assay (Figure 2), 243 samples were first analyzed for 110 prevalent bacterial species [13]. By sequencing of 1285 16S rRNA clones from this material, 197 bacterial species/phylotypes were detected. Half of these had not been cultivated. Twenty-two new phylotypes were identified. There were both health- and disease-associated species. It was noteworthy that in subjects with Streptococcus mutans, additional species such as species of the genera Atopobium, Propionibacterium, and Lactobacillus were significantly more abundant than S. mutans. In subjects with no detectable S. mutans, Lactobacillus spp., Bifidobacterium dentium, and low pH non-S. mutans streptococci were found. Munson et al. [14] also detected a diverse bacterial community, including S. mutans, Lactobacillus spp., Rothia dentocariosa, and Propionibacterium spp., in the middle and advancing front of dental caries in adults and found that numerous new taxa were present in carious lesions. Chour et al. [15] used similar molecular methods to Munson et al. [14] to assess advanced caries in adults. They found an abundance of species in the genera Lactobacillus, Prevotella, Selenomonas, Dialister, Fusobacterium, Eubacterium, Olsenella, Bifidobacterium, Propionibacterium, and Pseudoramibacter. Notably, S. mutans was not a common finding. It is clear that, in addition to S. mutans, bacteria such as Veillonella, Lactobacillus, Bifidobacterium, and Propionibacterium, low pH non-S. mutans streptococci, Actinomyces, and Atopobium spp. are likely to play a role in the caries process.

Figure 2.  Reverse capture checkerboard analysis of subjects with rampant caries and those with no caries. Plaque samples from healthy subjects (h) were compared to samples from intact enamel (a), whitespot lesions (b), cavitated lesions (c) and deep dentinal caries (d) from subjects with severe decay of the young permanent dentition. Mean bacterial signal for 29 of 110 selected bacterial species significantly associated with health and caries [13]. Bacteria associated with health (green) and caries (red). Bacterial signal was measured by intensity of spot on checkerboard.

Corby et al. [16] studied the bacteria associated with dental caries and health in 204 twins aged 1.5–7 years. Associated with disease was a strain of an Actinomyces sp., S. mutans, and Lactobacillus spp. Streptococcus parasanguinis, Abiotrophia defectiva, Streptococcus mitis, Streptococcus oralis, and Streptococcus sanguinis dominated in caries-free subjects, which was confirmed in another study [13].

Root caries in the elderly

The root caries microbiota was studied in a cohort of elderly subjects (70–101 years old) with culture-independent methods [17], [18]. Supragingival plaque was collected from sound root surfaces in subjects with and without root caries, as well as from carious root surfaces. In addition, the underlying dentine from the same carious roots was sampled. In the first study, a 16S rRNA gene sequencing technique was used to evaluate the bacterial diversity of healthy and carious root surfaces [17]. A total of 3544 clones were analyzed and 245 predominant species or phylotypes were detected, representing 8 bacterial phyla. The majority (54%) of the species had not previously been cultivated. Species of Selenomonas and Veillonella were common in all samples. Fusobacterium nucleatum subsp. polymorphum, Leptotrichia spp., Selenomonas noxia, Streptococcus cristatus, and Kingella oralis were characteristic for the sound root surfaces. Lactobacilli were absent, S. mutans was present in one, and Actinomyces spp. were found in half of the controls. Conversely, Actinomyces spp., Lactobacillus spp., S. mutans, Enterococcus faecalis, Selenomonas sp. clone CS002, Atopobium and Olsenella spp., Prevotella multisaccharivorax, Pseudoramibacter alactolyticus, and Propionibacterium sp. strain FMA5 dominated the bacterial profile of root caries. There was a considerable subject-to-subject variation in the root caries microbiota. The data indicated that putative microbial agents of root caries include not only S. mutans, lactobacilli, and Actinomyces, but also species of Atopobium, Olsenella, Selenomonas, Pseudoramibacter, and Propionibacterium.

A second study with a similar study design but a higher number of subjects employed HOMIM to better define the bacterial profiles of healthy and carious root surfaces in the elderly [18]. Altogether, 179 bacterial species and species groups were detected. Control subjects had a more diverse microbiota than patients with root surface caries. Lactobacillus spp. and P. alactolyticus dominated in most carious root samples. S. mutans was detected more frequently in infected dentine than in other samples, but the difference was not significant. Actinomyces spp. were found more often in control subjects. Based on the findings of these two studies [17], [18], it was concluded that bacteria other than those traditionally considered as pathogens, i.e. Actinomyces and S. mutans, may play a role in the root caries process.

Subgingival plaque

Subgingival plaque was analyzed from healthy subjects and subjects with refractory periodontitis, adult periodontitis, HIV periodontitis, and acute necrotizing ulcerative gingivitis [19]. A total of 2522 clones were examined and approximately 60% of them fell among 132 known species. Seventy of these were identified from multiple subjects. About 40% (215) of the clones were novel phylotypes, of which 75 were identified from multiple subjects. P. gingivalis, T. forsythia, and T. denticola were recovered from multiple subjects, but as a minor component of the microbiota of plaque. Notably, several phylotypes fell into two phyla associated with extreme natural environments with no representative cultivable species. Many species or phylotypes were detected only in subjects with disease, and a few were found exclusively in healthy subjects. The predominant subgingival microbial community comprised 347 species of phylotypes that fell into 9 bacterial phyla. It was estimated that there are 68 additional unseen species, thus producing a total of 415 species in subgingival plaque.

In another study, PCR and oligonucleotide probes were used to evaluate the subgingival and tongue microbiota during early periodontitis in 141 healthy and early periodontitis adults [20]. Most species differed in associations with sampled sites. Subgingival species were usually associated with subgingival samples. Few species were found more frequently in early periodontitis by DNA probes. Direct PCR associated P. gingivalis and T. forsythia with early periodontitis.

Aggressive periodontitis

Samples from 10 patients with generalized aggressive periodontitis were selected, DNA was extracted and the 16S rRNA gene was amplified [21]. Of the 1007 clones sequenced, 110 species were identified from the 10 subjects. Seventy of these species were most prevalent, with 57% representing uncultivated taxa. In all subjects, several species of Selenomonas and Streptococcus were detected. Selenomonas sputigena was most commonly found (9 of 10 subjects). Selenomonas noxia, Selenomonas sp. EW084, Selenomonas sp. EW076, Selenomonas FT050, Selenomonas sp. P2PA_80, and Selenomonas sp. strain GAA14 were also present at high levels. The classic putative periodontal pathogens, such as Aggregatibacter actinomycetemcomitans, were below the limit of detection. It was suggested that species other than the traditional periodontal pathogens, such as species of Selenomonas, may be involved in aggressive periodontitis.

Chronic periodontitis

Kumar et al. [22] observed associations with chronic periodontitis for several new species and phylotypes such as uncultivated clones D084 and BHO17 from the Deferribacterium phylum, AU126 from the Bacteroidetes phylum, Megasphaera clone BB166, clone X112 from the OP11 phylum, and clone 1025 from the TM7 phylum. Also the named species Eubacterium saphenum, Porphyromonas endodontalis, Prevotella denticola, and Cryptobacterium curtum were seemingly involved. In periodontal health, two uncultivated phylotypes, Deferribacteres clone W090 and Bacteroidetes clone BU063 were prevalent, as were Atopobium rimae and Atopobium parvulum.

Necrotizing ulcerative periodontitis (NUP)

Paster et al. [23] compared the microbial profiles of subgingival plaque in patients with NUP and chronic periodontitis and in periodontally healthy subjects using a battery of over 200 oligonucleotide probes in a PCR-based reverse capture, checkerboard DNA–DNA hybridization assay. Sequence analysis of over 400 clones revealed 108 species. At that time, 26 of these were novel to NUP subjects. Most commonly detected were Bulleidia extructa, Dialister, Fusobacterium, Selenomonas, Peptostreptococcus, Veillonella, and members of the phylum TM7. Microbial plaques from periodontally healthy subjects were different and less complex than the profiles of both the disease groups.

Another study on the predominant bacterial and fungal species associated with gingivitis, periodontitis, and linear gingival erythema (LGE) in HIV-positive subjects (n=14) with differing immune status [24] detected 119 bacterial species. Almost half of the species had not been cultivated. Classic periodontal pathogens such as P. gingivalis, T. denticola, and T. forsythia were not detected. On the other hand, species of Gemella, Dialister, Streptococcus, and Veillonella were dominant. Gemella morbillorum constituted 84% of the clones in one HIV-positive subject with periodontitis and low viral load. Saccharomyces cerevisiae was the only fungal species detected in an LGE subject and in periodontitis subjects with high viral loads. In periodontitis patients with low viral loads, Candida albicans dominated while S. cerevisiae was only a minor component. Classic periodontal pathogens did not seem to be involved in periodontitis of subjects with HIV. Periodontitis in HIV subjects is likely an opportunistic infection in a severely compromised host.

Halitosis

Kazor et al. [25] analyzed scrapings from the tongue dorsa of healthy subjects with no halitosis and from subjects with halitosis. The latter was defined as having an organoleptic score of 2 or more and volatile sulfur levels >200 ppb. Typically, 50–100 clones were analyzed from each subject. In addition, 51 strains isolated from the tongue dorsa of healthy subjects were examined. For novel species or phylotypes, nearly complete sequences of about 1500 bases were obtained. Of 750 clones analyzed, 92 different bacterial species were detected. About half of the clones had not yet been cultivated. Twenty-nine of these were new to the tongue microbiota. Fifty-one of the 92 species or phylotypes were detected in more than one subject. Streptococcus salivarius, Rothia mucilaginosa, and Eubacterium strain FTB41 were most associated with healthy subjects. Streptococcus salivarius was the predominant species in healthy subjects, representing 12–40% of the total clones. The microbiota of the healthy tongue was clearly different from that of the tongue dorsa with halitosis. The species most associated with halitosis were A. parvulum, Dialister clone BS095 (now Dialister invisus), Eubacterium sulci, TM7 clone DR034, Solobacterium moorei, and Streptococcus clone BW009. Recently, the dorsal surface of the tongue has been suggested to serve as a potential reservoir for bacterial species involved in ventilator-associated pneumonia [26].

Noma

A total of 67 bacterial species or phylotypes were detected from advanced noma lesions in four Nigerian children; 25 of them had not been grown in vitro[27]. Nineteen of the species or phylotypes such as Propionibacterium acnes, Staphylococcus spp., and the opportunistic pathogens, Stenotrophomonas maltophilia and Ochrobactrum anthropi, were found in more than one subject. Other known species detected included Achromobacter spp., Afipia spp., Brevundimonas diminuta, Capnocytophaga spp., Cardiobacterium sp., Eikenella corrodens, Fusobacterium spp., Gemella haemolysans, and Neisseria spp. Other bacteria in noma were found among Eubacterium, Flavobacterium, Kocuria, Microbacterium, Porphyromonas, Sphingomonas,Treponema, and S. salivarius.

Endodontic infections

Endodontic infections are frequently caused by oral biofilm bacteria entering the pulp or root canal. Samples were taken from the root canals of 21 untreated teeth and 22 root-filled teeth [28]. All of them had radiographic evidence of periradicular bone destruction. 16S rRNA-nested or hemi-nested PCR detected 13 species or phylotypes of bacteria. All species or phylotypes were found in at least one case of primary infection. Most prevalent in primary infections were D. invisus (81%), Synergistes oral clone BA121 (33%), and Olsenella uli (33%). Only these three species were detected in persistent infections. Like marginal periodontitis, apical periodontitis may be caused by hitherto unknown bacterial species.

In molecular analysis of the root canal microbiota associated with endodontic failures, Sakamoto et al. [29] detected bacteria in all cases. Seventy-four bacterial taxa belonging to six phyla were found in the nine cases investigated. Of these, 55% were as-yet-uncultivated phylotypes. Altogether, 25 new phylotypes were identified. A mean of 10 taxa was detected in each case. There was high inter-individual variability in the microbiota. It seems that endodontic treatment failures are associated with a number of not-yet-cultivated bacteria species and taxa other than Enterococcus faecalis.

Similar findings were obtained in a study that examined 20 surgically removed apical lesions from therapy-resistant teeth (T. Handal, D.A. Caugant, I. Olsen, P.T. Sunde, unpublished observations). A total of 236 clones were examined. Bacterial DNA was detected in 17 of the 20 samples. A total of 79 different taxa were identified and 35% of these species had not previously been cultivated. Commonly detected species were Fusobacterium spp., Prevotella spp., T. forsythia, P. endodontalis, T. denticola, Bacteroides-like spp., and Peptostreptococcus and Streptococcus spp. Altogether, 15 samples (88%) contained not-yet-cultivated species.

Spreading odontogenic infections

In a study by Riggio et al. [30], pus samples from four cases of spreading odontogenic infections were examined. Molecular detection methods identified a far more diverse microbiota than did culturing. Prevotella was the predominant genus, representing 102 of 203 clones. The most abundant species was Prevotella oris, which constituted 45 of the 203 clones analyzed. Twelve clones represented not-yet-cultivable species: Prevotella PUS9.180, uncultured Peptostreptococcus species, and an uncultured species belonging to the Bacteroidetes phylum.

Biofilm bacteria in blood

In a study by Bahrani-Mougeot et al. [31] on oral bacterial species in blood following dental procedures, sequence analysis of 16S rRNA genes identified 98 different bacterial species cultured from 151 bacteremic subjects. Forty-eight of the isolates represented 19 novel species of Prevotella, Fusobacterium, Streptococcus, Actinomyces, Capnocytophaga, Selenomonas, and Veillonella. Such organisms may play a key role in contributing to cardiovascular diseases through cumulative daily exposure of the vascular system inducing inflammation.

Concluding remarks

About half of the microbiota in oral biofilms consists of not-yet-cultivated bacterial species. Several examples of such biofilms and their microbiota have been given in the present review. Since most infections in the oral cavity are caused by biofilm bacteria, this may have large implications for our understanding of the pathogenesis of major oral diseases such as caries, periodontitis, peri-implantitis, and mucositis. Actually, the knowledge may initiate new concepts related to bacterial pathogenesis where hitherto unknown species will be ascribed a significant clinical role. There is no reason to believe that the not-yet-cultivated species should be less pathogenic than cultured ones. Another aspect of this new knowledge is its possible influence on treatment planning. How are we going to direct our treatment in a reasonable way when we do not know which bacteria are causing the disease? What effect do antimicrobial agents have on these organisms? How can we know if they are not resistant to our treatment? Another important question is the relationship between not-yet-cultured organisms and the host. How can we know anything about this relationship if the organism cannot be cultured? At least, now we can monitor these not-yet-cultivated species to begin to assess their importance in health and disease.

Acknowledgements

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.