The human oral cavity harbors a highly diverse and unique microbiome, which exists in a continuously changing environment where pH, organic carbon and oxygen levels fluctuate on a hundred-fold or even a thousand-fold scale within minutes [1, 2]. From the first discovery that dental plaque pH will decrease after sucrose consumption and then return to baseline values [3, 4], the oral microbial community has been a highly studied system not only because of its health-related significance but also for demonstrating interactions between species and functional analysis of multi-species communities in general [5–7]. Difficulties in studying this complex and structurally heterogeneous environment are multifaceted, and include problems related to the species variability of human subjects, continuous access to samples over time, small sample sizes and complicated ethical issues. This has led to the development of both synthetic consortia biofilm model systems and in vitro microcosm plaque from the human natural oral flora by using growth systems, including chemostats , the constant-depth film fermentor (CDFF) , saliva-conditioned flow cells [10, 11] and artificial mouths . Results from these studies show that the multispecies biofilms, containing a hand full of bacterial species, are functionally reproducible with heterogeneous structures and pH behaviors consistent with those of natural plaque [13, 14]. As investigators started to understand that the microbial diversity within the oral microbiome is highly important in both health and disease they began to explore multiple species interactions by using mixed-species models consisting of up to 10 defined species. This led to the synthesis of the 'ecological plaque hypothesis,’ which proposes that selection of cariogenic bacteria is directly coupled to alterations in the environment that shifts the balance of the community . According to this hypothesis, if the pH remains below the critical pH (value of 5.5) for demineralization for extended time periods after a carbohydrate pulse, a shift in the bacterial populations to more cariogenic organisms that are acid-producing (acidogenic) and acid-tolerant (aciduric) occurs [16, 17]. Another important aspect of this hypothesis is that any species with relevant traits can contribute to the disease process [15, 18]. This was also supported by multiple findings that bacterial species, other than well-known pathogens (for example, Streptococcus mutans) are present in caries-active sites [19, 20]. Also, a recent study shows that one can detect several low-pH active species present in a healthy plaque, which may be responsible for the onset of caries disease .
In order to fill in the knowledge gaps in species diversity for these complex communities, the Human Microbiome Project (HMP) [22, 23], the Human Oral Microbiome Database (HOMD)  and the CORE database  have been established. The major goal of the HMP was to characterize bacterial communities that are associated with several different body sites as well as generate a catalog of reference genomes from species derived from human hosts . The HOMD specifically contains information on the prokaryotic species present in the human oral cavity and has the capacity to link sequence data with phenotypic, phylogenetic, clinical and bibliographic information. The curated version of HOMD, published by Dewhirst and colleagues in 2010 contains approximately 619 validated taxa with 1,178 total taxa identified, of which 24% are named, 8% are cultivated but unnamed and 68% represented uncultivated phylotypes . Human oral taxa are defined as sharing 98.5% similarity in 16S rRNA (henceforth abbreviated as 16S) gene sequences . A major hurdle to understanding the oral microbiome is the unknown contribution of the very large uncultivated fraction of the existing bacterial diversity. In fact, the greatest number of the 'most wanted taxa’ (that is, those that have been seen by 16S sequencing but remain uncultivated) targeted for whole genome sequencing in the human body reside in the oral cavity [28, 29]. The identity of these oral phylotypes can only be linked to possible functions by using techniques such as nucleic acid-based stable isotope probing (SIP)  or single-cell genomics based sequencing approaches [30, 31]. Moreover, due to the high taxonomic variability of the oral microbiome between study subjects it is extremely difficult to track species and strains temporally and spatially . Also, small sample sizes that are dictated by availability of volunteers and costs, limit the statistical power needed to detect small, but important differences among communities. Hence, to gain a deeper ecological understanding of the processes that are involved in the gradual succession of healthy oral microbiomes to disease-associated microbiomes, it is important to continue the development of oral microbial model systems where experiments can be conducted in a controlled environment. The advantages with such systems are many as they provide novel opportunities to study microbial community ecology with systems biology perspectives by using global omics experimental tools (metagenomics, metatranscriptomics, metabolomics). A model system also allows for generating biological replicates and contributes to the analyses of large samples that are needed to obtain reliable spatial and temporal dynamics data of bacterial populations within a community.
In this study, our aim was to develop a mixed-community biofilm model system comprising the highest possible cultivable bacterial diversity representative of the resident saliva-derived microbiome responsible for plaque formation in the human oral cavity. We used a recently developed growth medium (SHI medium) that previously was shown to support growth of a highly diverse microbiome and that also had a high coverage of species found in the original inoculum saliva samples obtained from healthy adults . To evaluate the reproducibility of this in vitro model system and to address its overall metabolic potential we applied both conventional community fingerprinting, denaturing gradient gel electrophoresis (DGGE), and next-generation sequencing. The broader taxonomic reproducibility of biofilms growing in two different research laboratories was tested by comparing 16S gene profiles from DGGE. To obtain a broader understanding of the whole biofilm community diversity, deep sequencing of 16S genes via 454-pyrosequencing and whole genome shotgun (WGS) sequencing on the Illumina HiSeq platform was performed. To our knowledge, the model we have developed represents the highest oral bacterial diversity that has been reported for an in vitro system thus far. This model can be used to gain a deeper understanding of molecular mechanisms that underlie the evolution of healthy oral microbiomes into disease-associated microbiomes where cariogenic groups such as mutans streptococci and Lactobaccilli become more abundant [7, 19, 33, 34]. This model will also aid in the understanding of oral microbial communities by facilitating discovery and functional characterization of known, as well as uncultivated bacteria, within a mixed-species system that is approaching the diversity of in vivo conditions. Importantly, it will allow systematic investigation of species, specific genes/domains, gene products and metabolic pathways that define the synergistic and competitive contributions to health and disease in the complex oral microbiome.