Journal of Applied Microbiology ISSN 1364-5072
ORIGINAL ARTICLE
Individual microflora beget unique oral microcosms R.G. Ledder1, P. Gilbert1, A. Pluen1, P.K. Sreenivasan2, W. De Vizio2 and A.J. McBain1 1 School of Pharmacy and Pharmaceutical Sciences, University of Manchester, Manchester, UK 2 Colgate-Palmolive Company, Piscataway, NJ, USA
Keywords denaturing gradient gel electrophoresis, dendrogram, dental plaque, multiple Sorbarod device, perfused microcosm. Correspondence A.J. McBain, School of Pharmacy and Pharmaceutical Sciences, University of Manchester, Manchester, M13 9PL, UK. E-mail:
[email protected]
2005/0703: received 20 June 2005, revised 19 September 2005 and accepted 10 October 2005 doi:10.1111/j.1365-2672.2006.02847.x
Abstract Aims: To examine the efficacy of the multiple Sorbarod device (MSD) for the reproduction of inter-individual variations in oral microbiotas. The MSD supports sessile growth on parallel cellulose filters, perfused with artificial saliva. This enables biofilms (BF) to be grown and sampled, together with released cells in eluted medium (perfusates, PAs). Methods and Results: Two sets of triplicate MSDs were established. One set was inoculated using fresh saliva from three separate volunteers; the second set was inoculated from one saliva donor. Both were incubated in an anaerobic cabinet. BF and PA were analysed at 24-h intervals by PCR-denaturing gradient gel electrophoresis (DGGE) of 16S rDNA. Hierarchical dendrograms were constructed in order to sort community fingerprints over time, based on community relatedness. The MSD supported complex oral communities, as evidenced by DGGE (>20 distinct DGGE bands) and confocal scanning laser microscopy. DGGE band sequencing revealed bacteriological diversity and a high incidence of anaerobic species, including Prevotella sp. Dendrograms demonstrated marked inter-individual variation in the relative species abundance within salivary inocula from different volunteers (DV) and each associated MSD (all >45%, majority c. 85% concordance). Less variation was shown between triplicate models established using saliva from a single volunteer (SV) (all >58%; majority c. 95% concordance). PAs clustered together with the associated biofilms and inocula in the majority of cases for the DV MSDs whilst SV MSD community profiles clustered between replicate MSDs. Conclusions: Data indicate that marked inter-individual variations in human salivary composition can be partially replicated in individualized MSD microcosms. Significance and Impact of the Study: This study demonstrates the in vitro reproduction of individual oral microbiotas and suggests that taking inter-individual variability into account will increase the relevance of microcosm studies.
Introduction In vitro models have proved useful as a substitute for human volunteers when studying the microbial physiology and ecology of the oral microbiota for giving insight into the effect of antimicrobial treatments upon acidogenicity and tooth demineralization, etc. Commonly used models include simple multi-well culture plates (Guggenheim et al. 2001), various continuously fed biofilm reac-
tors (Marsh et al. 1983; Bradshaw et al. 1996) including constant depth film fermenters (Pratten and Wilson 1999; Vroom et al. 1999; McBain et al. 2003a,b) flow cells (Foster and Kolenbrander 2004; Pratten et al. 2004), artificial mouths which involve drip feeds onto solid substrata (Sissons et al. 1991, 1992; Sissons 1997), drip-flow reactors (Adams et al. 2002) and chemostats (McKee et al. 1985). Hydroxyapatite discs can also be suspended within liquid cultures in order to broadly simulate a tooth
ª 2006 The Authors Journal compilation ª 2006 The Society for Applied Microbiology, Journal of Applied Microbiology 100 (2006) 1123–1131
1123
Individualized oral microcosms
R.G. Ledder et al.
surface (Bradshaw and Marsh 1998). Regardless of the apparatus used, microcosms, by definition utilize abstracted microbial material from in vivo as inocula. Microcosms thus enable natural mixed oral microbiota to develop and evolve in vitro (Wimpenny 1988). The human oral microbiota has received considerable research attention (Moore and Moore 1994; Paster et al. 2001). This environment is highly heterogeneous, with distinct sessile microbial assemblages, separately colonizing the hard surfaces of the tooth (supra- and sub-gingival plaque), as well as soft surfaces such as the tongue and buccal mucosa (Pratten and Wilson 1999; Guggenheim et al. 2001; Mager et al. 2003; McBain et al. 2003a,b). Many resident oral species have evaded culture or are difficult to cultivate in the laboratory. Those that are amenable to culture may be difficult to differentiate from their congeners using phenotypic criteria (Moore and Moore 1994; Wall-Manning et al. 2002). Cultureindependent approaches can therefore be used to complement culture, or in certain cases as a replacement, whilst rDNA sequence analysis has proved a useful means of identifying cultured taxa (McBain et al. 2003a). Just as the statistical power of any meaningful clinical trial of a drug requires multiple human volunteers to account for inter-individual variation, it is arguable that the accurate and realistic modelling of any host-associated microbial ecosystem requires the considerable inter-individual variation in microflora be taken into account and if possible, reproduced in vitro. Such variations have previously been noted and may be highly significant when investigating the effects of antimicrobials upon microbial communities (McBain et al. 2003b), or the metabolic capacity and degradative potential of natural consortia (Hopkins et al. 2003). We have previously developed a small-scale fermentation system that enables the maintenance of five independent, sessile oral (salivary) communities upon cellulose filters (Sorbarods) (McBain et al. 2005). The multiple Sorbarod device (MSD) uses a simple two-piece stainless steel housing, yields relatively large amounts of biomass, and enables continuous monitoring of population dynamics through the analysis of perfusates (PA, spent culture fluid). Being a continuous flow system, the mean growth rate can be regulated by altering the rate of perfusion of medium through the MSD. Denaturing gradient gel electrophoresis (DGGE) has been used to generate reproducible fingerprints of consortia associated with the oral cavity, oral microcosms, and the general environment (Fujimoto et al. 2003; McBain et al. 2003a,b; Zijnge et al. 2003). In this technique, extracted community DNA is amplified by PCR, utilizing primers specific for variable regions of rDNA. PCRs carried out with consortial DNA give a mixture of PCR products of 1124
similar length, but varying sequence. These products can then be separated electrophoretically on a denaturing gradient gel on the basis of PCR product sequence, and hence melting characteristics. The phylogenetic origin of selected bands can be identified by sequence analysis. A further enhancement of this technique is the application of image analysis to construct dendrograms by the unweighted pair group method, using arithmetic averages (UPGMA) based on lane matching profiles (Ibekwe et al. 2001; Yang et al. 2001; Boon et al. 2002). This allows the identification of band pattern motifs that are characteristic of particular states or conditions. Because MSDs require relatively small volumes of salivary inocula, pooling of saliva from individual or groups of volunteers is unnecessary, enabling the in vitro maintenance of oral microfloras abstracted from individual subjects. The present study examines the efficacy of the MSD with respect to inter-individual variations in the oral microflora. To remove culture bias, DGGE was used and measures of bacterial community composition (UPGMA dendrogram construction) made. Sequencing of DGGE bands was used to identify dominant PCR-amplified phylotypes. Materials and methods Multiple Sorbarod devices Multiple Sorbarod devices (MSDs) were used to support the growth of oral microbial consortia under environmental conditions similar to those occurring in the human mouth, including nutrient availability, surfaces for colonization and oxygen status. The device has been previously described (McBain et al. 2005) and enables replicate biofilms (n ¼ 5) to be established on Sorbarod filters (paper cylinders of 10 mm diameter; 20 mm length). Temperature (36 ± 0Æ5C) and anaerobiosis (atmosphere H2, 10%; CO2, 10%; N2, 80%) were maintained by housing the devices in an anaerobic chamber (Don Whitley Scientific, Shipley, UK). Microcosm growth conditions A modified artificial saliva was used as growth medium (Shah et al. 1976; Marsh et al. 1983). This contained (g l)1 in distilled water): mucin (type II, porcine gastric), 2Æ5; bacteriological peptone, 2Æ0; tryptone, 2Æ0; yeast extract, 1Æ0; NaCl, 0Æ35; KCl, 0Æ2; CaCl2, 0Æ2; cysteine hydrochloride, 0Æ1; haemin, 0Æ001; vitamin K1, 0Æ0002. Unless otherwise stated, chemicals were obtained from Sigma (Poole, Dorset, UK). Prior to inoculation with fresh human saliva, the filters were conditioned, in situ with the culture medium continu-
ª 2006 The Authors Journal compilation ª 2006 The Society for Applied Microbiology, Journal of Applied Microbiology 100 (2006) 1123–1131
R.G. Ledder et al.
ously supplied by peristaltic pump (Minipulse 3; Gilson, Villiers-Le-Bel, France) for c. 2 h and throughout the study (flow rate, 7Æ0 ± 0Æ1 ml h)1). In order to ensure that spent culture fluid did not accumulate in the models, a second peristaltic pump was used on the waste pipe, set to an identical flow rate to the medium supply pump. The pumps were switched off during primary inoculation. For each experiment, fermenters were inoculated on two occasions, 4 h apart, by removing the upper section of the MSDs and depositing 4 ml of freshly collected saliva from the same donor directly into the upper equilibration chamber. Fresh portions of these inocula were removed and archived at )60C for subsequent DGGE analyses. MSD consortial variation experiments In order to investigate reproducibility of the MSD systems, two triplicate sets of MSDs were simultaneously set up: the first set used a separate volunteer for each of three fermenters (2 M, 1 F), age range 26–53 (volunteers and their respective models were designated A, B and C). The second set of three fermenters utilized saliva obtained from a 26-year-old individual female volunteer (models were designated 1, 2 and 3). In all cases, saliva donors were healthy, had no history of extant periodontal disease and had not taken antibiotics for the previous three years. Analysis of biofilm (BF) and perfusate (PA) Once equilibrated (after c. 3 days), PA (spent culture fluid) samples (c. 5 ml) were collected daily by disconnecting the waste tube at the bottom of the MSD and placing a sterile plastic Universal bottle directly underneath. Once the sample had been harvested, the tube and lower portion of the MSD were disinfected by immersion in ethanol (80%, v/v) and then reconnected. For sampling of biofilm growth, MSDs were opened aseptically and a single Sorbarod filter was removed and immediately replaced with a new, sterile filter. Sorbarod filter BFs, together with PAs were archived at )60C for subsequent DGGE analysis. Confocal scanning laser microscopy (CSLM) image analysis Developed biofilms were exposed by cutting the Sorbarod filter sheaf longitudinally using a sterile scalpel and a portion from the upper section was transferred to a sterile plastic Petri dish where it was stained for 10 min in the dark at room temperature using 100 ll of BaclightTM LIVE/DEAD stain (Molecular Probes, Leiden, The Netherlands), prepared according to manufacturer’s instructions in phosphate-buffered saline (0Æ01 mol l)1, pH 7Æ0). Biofilms were then examined using a Zeiss combi
Individualized oral microcosms
LSM 510 META/Confocor II inverted microscope (Jena, Germany). To image live and dead bacteria, 488 and 633 nm excitation wavelengths were used respectively. Confocal images were obtained using 40 · 1Æ3NA DIC oil immersion objectives. Each biofilm was scanned at randomly selected positions. Z-series were generated by vertical optical sectioning at every position with the slices’ thickness set to 0Æ4 lm. Image acquisition and analysis was made with the software combi lsm (Zeiss, version 3.2). DNA extraction DNA was extracted from the archived periodontal samples using a DNeasy Tissue Kit (Qiagen Ltd, West Sussex, UK) in accordance to manufacturer’s instructions. The amounts and quality of DNA extracted were estimated by electrophoresis of 5 ll aliquots on a 0Æ8% agarose gel and comparison with a molecular weight standard (100 bp Low Ladder, Sigma) stained with ethidium bromide. DNA extracts were stored at )60C prior to analysis in nuclease free containers. PCR amplification and DGGE analysis The V2–V3 region of the 16S rRNA gene (corresponding to positions 339–539 of Escherichia coli) was amplified with the eubacterium-specific primers HDA1-GC (5¢-CGC CCG GGG CGC GCC CCG GGC GGG GCG GGG GCA CGG GGG GAC TCC TAC GGG AGG CAG CAG T-3¢) and HDA2 (5¢-GTA TTA CCG CGG CTG CTG GCA C-3¢) as previously described (Walter et al. 2000). The reactions were performed in 0Æ2-ml tubes with a DNA thermal cycler (model 480; Perkin-Elmer, Cambridge, UK). In all cases, reactions were carried out with Red Taq DNA polymerase ready mix (25 ll; Sigma), HDA primers (2 ll of each, 5 l mol l)1), nanopure water (16 ll), and extracted community DNA (5 ll). Optimization studies, as described by Muyzer and Smalla 1998 showed that in some cases, extracted community DNA required 1 : 10 dilution to ensure reliable PCRs. The thermal program was as follows: 94C (4 min), followed by 30 thermal cycles of 94C (30 s), 56C (30 s), and 68C (60 s). The final cycle incorporated a 7-min chain elongation step (68C). A D-Code universal mutation detection system (Bio-Rad, Hemel Hempstead, UK) with 16 · 16 cm (1 mm deep) polyacrylamide gels (10%), run with 1x TAE buffer diluted from 50x TAE buffer (40 mmol l)1 Tris base, 20 mmol l)1 glacial acetic acid, and 1 mmol l)1 EDTA) was used to analyse community DGGE amplicon mixtures. Gels were polymerized by adding tetramethylethylenediamine (TEMED) (50 ll) and ammonium persulphate (APS) (1%, w/v; 100 ll) immediately prior to pouring. Normal gel pouring temperatures ranged between 18 and 22C. Dena-
ª 2006 The Authors Journal compilation ª 2006 The Society for Applied Microbiology, Journal of Applied Microbiology 100 (2006) 1123–1131
1125
Individualized oral microcosms
R.G. Ledder et al.
Figure 1 A confocal scanning laser micrograph showing LIVE/DEAD-stained Sorbarod filter material. (a) Uninoculated and (b) after 14 h growth. Image shows a 58 · 58 lm area of filter fibres.
turing gradients for parallel DGGE analysis, ranged from 30% to 60% as previously validated (McBain et al. 2003a,b). Electrophoresis was carried out at 40 V at 60C for approximately 17 h. Gels were stained using SYBR Gold stain (diluted to 10)4 in 1x TAE) [Molecular Probes (Europe), Leiden, The Netherlands] for 30 min. Gels were viewed and TIFF images recorded using a BioDocitTM system (UVP, Upland, CA, USA). Construction of dendrograms All separate DGGE gel images were aligned into a single TIFF file using Adobe Photoshop 6.0 (Adobe Systems UK, Uxbridge, UK) prior to analysis with Phoretix 1D version 2003.02 (NonLinear Dynamics, Newcastle, UK). Accurate alignment of gels was achieved by comparing replicated samples across gels and by confirming the identities of key band positions on the constructed gel image. Manually detected bands indicative of unique positions in the gel were used to create a synthetic reference lane. Each lane on the gel was then compared with the reference lane, generating a matching profile and UPGMA dendrograms.
grams. Corresponding bands were excised from replicated gel lanes and from the polyacrylamide gels with a sterile scalpel under UV illumination and incubated at 4C for 20 h together with 50 ll of nanopure water in nuclease-free universal bottles. Portions (5 ll) were removed and used as a template for a PCR identical to that outlined for DGGE analysis. PCR products were purified with QIAquick PCR purification kits (Qiagen) and sequenced with the reverse (non-GC clamp) primer (HDA2). The sequencing reaction was as follows: 94C (4 min) followed by 25 cycles of 96C (30 s), 50C (15 s), and 60C (4 min). Once chain termination was complete, sequencing was carried out in a Perkin-Elmer ABI 377 sequencer. Unidirectional DNA sequences were checked and edited using chromas lite (Technelysium, Tewantin, Australia). For excised DGGE band PCRs, the presence of a GC clamp upon sequence analyses confirmed that the correct target had been reamplified, rather than a contaminant. Results
Sequencing of bacterial isolates and excised gel bands
CSLM image analysis
Major resolved DGGE amplicons were identified by examining the synthetic reference lanes on the dendro-
Figure 1 shows CSLM images of a LIVE/DEAD-stained (A) uninoculated Sorbarod filter and (B) filter removed
1126
ª 2006 The Authors Journal compilation ª 2006 The Society for Applied Microbiology, Journal of Applied Microbiology 100 (2006) 1123–1131
R.G. Ledder et al.
Table 1 Sequencing of PCR amplicons derived from DGGE gels
Individualized oral microcosms
Unique bands*
Sequence length
Ambiguity (%)
Closest relative (% sequence similarity)
b c(i) c(ii) d(i) d(ii) d(iii) d(iv) e f(i) f(ii) g(i) g(ii) h(i) i(i) i(ii) i(iii) i(iv) i(v) i(vi) i(vii) i(viii) i(ix) i(x) i(xi) j(xii) j(xv) k l m n(i) n(ii) o p q r(i) r(ii) r(iii) r(iv) r(v) r(vi) s t u(i) u(ii)
120 183 185 172 180 171 179 175 174 186 185 139 174 177 185 178 178 171 178 136 178 176 183 177 164 182 169 168 183 168 169 170 180 182 173 185 183 159 185 190 186 181 180 187
0 1 5 2Æ86 1Æ3 0 8Æ9 5Æ14 0 9Æ7 2Æ70 2Æ6 0 0 1 0Æ55 0Æ56 0Æ55 0 2Æ2 0 2 0Æ55 2Æ26 3Æ66 0Æ55 2Æ96 3 2 3Æ0 1Æ2 1Æ2 1 0Æ55 0 0Æ54 0Æ55 8Æ18 1Æ08 2Æ1 0 2Æ2 2 1
Pseudomonas sp. AF448043 (93) Neisseria subflava AF479578 (95) Neisseria dentiae AF487709 (91) Prevotella sp. AY349396 (91) Prevotella pallens Y13106 (83) Prevotella sp. AY349396 (98) No significant match Peptostreptococcus sp. AY665684 (97) Prevotella sp. AY349396 (82) Prevotella sp. AY349394 (96) Micromonospora sp. AF432238 (96) Cytophaga sp. AB015543 (92) Fusobacterium canifelinum AY162222 (95) Prevotella sp. AY349396 (83) Prevotella sp. AY349396 (89) Prevotella sp. AY349396 (87) Prevotella sp. AY349396 (96) Prevotella sp. AY349396 (88) Prevotella sp. AY349396 (98) Prevotella sp. AY349394 (93) Prevotella sp. AF385512 (99) Prevotella sp. AY349396 (84) Prevotella sp. AY349396 (95) Prevotella sp. AF481227 (95) Fusobacterium sp. AY827903 (89) Fusobacterium nucleatum subsp. animalis AF342835 (88) Uncultured Flavobacterium clone AY712104 (100) Helicobacter cholecystus AY686606 (100) Bacteroides salyersae AY608696 (93) Deferribacteres sp. oral clone AY349371 (96) Deferribacteres sp. oral clone AY349372 (92) Syntrophomonas sp. AB021306 (96) Haemophilus parainfluenzae AY365452 (84) Veillonella dispar AF439639 (95) Streptococcus cristatus AY281090 (98) Streptococcus agalactiae AF459432 (90) Streptococcus cristatus AY281090 (98) Streptococcus sp. AY005047 (89) Streptococcus iniae AY489404 (93) Uncultured bacterium clone AF087802 (89) Flexispira rappini AY192528 (94) Ruminococcus sp. RSP315982 (95) Prevotella albensis AF073773 (95) Prevotella melaninogenica AY323525 (90)
*Refers to key denaturing gradient gel electrophoresis (DGGE) bands that were sequenced, given in the order of gel-band position (see Fig. 4). Replicate, identically migrated bands sequenced from other gel lanes shown in parenthesis. Similarities are based on pair wise alignments with published sequences, according to BLAST searches and indicate similarity, not guaranteed identity.
from an MSD after 14 h incubation. The image (B) shows a large, multi-species aggregate and demonstrates a high degree of bacterial vitality (green cells) with lesser numbers of dead (red) cells. Considerable morphological diversity is apparent.
Characterization of consortia by DGGE and comparison by cluster analysis The DGGE fingerprints obtained from triplicate MSDs were established separately using (i) the saliva of three
ª 2006 The Authors Journal compilation ª 2006 The Society for Applied Microbiology, Journal of Applied Microbiology 100 (2006) 1123–1131
1127
Individualized oral microcosms
A saliva
A-BF
A-PA
R.G. Ledder et al.
B saliva
B-BF
B-PA
C saliva
C-BF
C-PA
b c d e f h g i j k l m n o p q s
r t u
Pseudomonas sp. Neisseria spp. Neisseria dentiae Peptostreptococcus sp. Prevotella spp. Micromonospora* Fusobacterium canifelium Prevotella spp. Fusobacterium sp. Flavobacterium sp. Helicobacter cholecystus Bacteroides salyersae Deferribacteres Syntrophomonas sp. Haemophilus parainfluenzae Veillonella dispar Streptococcus sp. Flexispira rappini Ruminococcus sp. Prevotella melaninogenica
Figure 2 A negative image constructed to show denaturing gradient gel electrophoresis profiles of selected, derived multiple Sorbarod device perfusate (PA) and biofilm (BF) sample profiles together with salivary inocula. The image includes a diagrammatic representation of the migration positions of sequenced bands their identities of closest relatives (based on sequencing). *Possible co-migration of phylotypes.
different volunteers (DV) and (ii) a single collection of fresh saliva taken from a single volunteer (SV). Analysis of salivary inocula, perfusates and the formed biofilms using PCR-DGGE, showed that PA and BF samples from DV MSDs formed distinct clusters in two of three donors (Fig. 3). The 5-day BF from volunteer A clustered tightly with the perfusates (90% concordance) and there was considerable similarity between the BF and PA at the 85% level for individual C. Clustering of PA and BF samples was considerably less prevalent in the same volunteer (SV) MSDs (Fig. 4) and PA and BF samples from separate fermenters (1, 2 and 3) in many cases co-clustered, in one case at the 100% level and in majority of cases at over 90% concordance. The complex communities that developed in the DV devices gave upwards of 20 distinct dominant PCR amplicons. Key DGGE bands, identified during dendrogram construction were excised and sequenced for identity (Table 1). The synthetic reference lane generated by image analysis was used to identify key bands for sequencing. This lane is represented in diagrammatic form in Fig. 2, together with representative gel lanes (PA and BF) from MSD established with three separate volunteers (corresponding to the dendrogram given in Fig. 3). Data in Table 1 show the closest phylogenetic relatives based on BLAST searches with DNA sequences obtained from these DGGE gel bands identified by PHORETIX 1D analysis, excised and sequenced from the gels. Table 1 presents band sequences in the order they occurred on the gels, top to bottom, unique bands being given assigned codes b–u and replicate bands from separate samples sequences being represented by Roman numerals. Overall, sequence data suggest that complex microbial consortia colonized 1128
the MSDs with a high prevalence of Gram-negative anaerobic species. Prevotella sp. was the predominant DGGE amplicon in the perfusates and biofilms of all MSDs whilst Veillonella sp., Neisseria dentiae, Fusobacterium sp. and various streptocococci also gave strong DGGE bands. Bacteria less commonly associated with the oral microcosms included organisms with sequence homology to Helicobacter cholecystus, and the related bacterium Flexispira rappini, together with Pseudomonas sp., the mycelial, sporulating bacterium micromonospora and Syntrophomonas sp. Whilst the community fingerprints were closely related to, and clearly derived from that of the donor saliva, distinct differences were noted between perfusates and biofilm. Specifically, Fusobacterium sp. was present in higher abundance in perfusates whilst Ruminococcus sp. was a major biofilm amplicon in volunteer C but was absent from PA samples (data not shown). Sequential filters taken from the MSDs at 1-day intervals showed that major amplicons were maintained and that variations apparent in Fig. 3 were mainly because of less dominant amplicons. Discussion The variation of community composition between individuals is an issue of concern when attempting to model any commensal microflora. Such variation applies not only to the species composition but also to baseline metabolic activities and to the consortial response to various growth substrates or antimicrobials. We have used the MSD, a simple small-scale biofilm device, to grow complex salivary microcosms using small volumes of saliva as inocula. DGGE sequencing data
ª 2006 The Authors Journal compilation ª 2006 The Society for Applied Microbiology, Journal of Applied Microbiology 100 (2006) 1123–1131
R.G. Ledder et al.
Individualized oral microcosms
C-PA 5 C-PA 4 C-PA 3 C-PA 2 C-PA 1 C-BF 5 C-BF 4 C-BF 3 C-BF 2 C-BF 1 C Saliva A-PA 5 A-PA 4 A-PA 3 A-PA 2 A-PA 1 A-BF 5 A Saliva A-BF 1 A-BF 2 A-BF 3 A-BF 4 B-BF 5 B-BF 4 B-PA 5 B-PA 4 B-PA 3 B-PA 2 B-BF 3 B-BF 2 B-BF 1 B-PA 1 B Saliva
50
60
70 80 Percent similarity
90
100
Figure 3 A UPGMA dendrogram showing percentage matching of multiple Sorbarod device (MSD) samples from three MSD models inoculated simultaneously with salivary inocula from three separate donors (A, B and C). BF, biofilm; PA, perfusate. Sample numbers (1–5) refer to consecutive days of sampling.
showed that the MSD microcosms became dominated by anaerobic species such as prevotella, fusobacteria and Peptostreptococcus sp. (Table 1). This observation is in agreement with previous studies where culture and checkerboard DNA–DNA hybridization identified a dominant population of Gram-negative anaerobes within MSD-mouths (McBain et al. 2005). The dendrogram clustering of perfusate (PA) and biofilm (BF) and associated inoculum samples demonstrates that the microcosm broadly reproduced the inoculum communities whilst diverging from inocula by up to 22% (Fig. 3). Failure of microcosm plaques to exactly match the inoculum is most likely due to the device not exactly reproducing the physico-chemical conditions extant in the mouth. The data are therefore consistent with the definition of a microcosm as ‘a laboratory model of the natural system from which it originates, but also from which it evolves’ (Wimpenny 1988). The clustering between microcosms established with inocula from different individuals gives evidence for a founder effect, i.e. that a daughter community cannot contain more than a fraction of the total genetic variability of the parent population. In this study, the parent population is present in the saliva of the donor and the daughter population is what becomes established within the MSD mouth. Saliva contains pooled bacteria from a variety of distinct oral niches including teeth, soft palate, the dorsum of the tongue and from the external environment and as such will contain considerable microbial diversity, whilst the MSD communities will most likely support reduced genetic diversity due to the selective pressures in the Sorbarod. Such environmental selection is a key issue in microbial ecology. No in vitro model of a human-associated microflora can precisely reproduce in vivo conditions. Pooling of inocula, apparent in many other studies might therefore increase the potential microbial diversity of the microcosm at the expense of relevance, possibly masking variation. We have previously reported that MSDs, inoculated with freshly collected saliva and perfused with artificial saliva rapidly achieved dynamic steady-states with respect to the numbers of bacteria associated with the filters and also with the spent medium (perfusate). Such steadystates could be maintained for at least 5 days post equilibration (McBain et al. 2005). The dendrograms shows that dynamic stability was maintained over 5 days in both BF and PA and that these samples differed in composition (Figs 3 and 4). MSD biofilm samples might be viewed as being analogous to plaque and perfusates as being analogous to saliva, but this is simplistic as salivary consortia are not simply a subset of those present on the teeth, but a complex mixture derived from a variety of oral tissues. Variations between BF and PA communities are due to: (i) the initial retent-
ª 2006 The Authors Journal compilation ª 2006 The Society for Applied Microbiology, Journal of Applied Microbiology 100 (2006) 1123–1131
1129
Individualized oral microcosms
R.G. Ledder et al.
3-PA 4 3-PA 3 3-BF 4 3-PA 2 3-BF 5 1, 2, 3 Saliva 3-BF 2 2-PA 5 3-BF 1 2-PA 4 2-PA 3 2-PA 2 3-BF 3 1-BF 3 3-PA 5 2-BF 4 2-BF 5 1-BF 2 1-BF 1 2-BF 2 2-BF 3 2-PA 5 1-BF 4 1-BF 5 1-PA 1 1-PA 5 1-PA 2 1-PA 3 1-PA 4 2-PA 4 2-PA 3 2-PA 2 2-PA 1 4*-BF 1 4*-BF 2 4*-BF 3 4*Saliva 5*-BF 2 5*-BF 1 5*Saliva 60
70
80 90 Percent similarity
100
Figure 4 A UPGMA dendrogram showing percentage matching of multiple Sorbarod device (MSD) samples from three MSD models inoculated simultaneously with salivary inocula from the same donor (donor A from Fig. 3) (1–3 in bold text). BF, biofilm; PA, perfusate. Sample numbers (1–5) refer to consecutive days of sampling. MSDs 3 and 4 were established using the same donor.
ion of some species of bacteria and the loss of others; (ii) variations in growth rate under conditions extant in the devices; and (iii) interspecies differences in bacterial retention within biofilms. Our data suggest that fusobac1130
teria, in particular, being more prevalent in perfusates than biofilm, exhibited high growth rates and low biofilm retention. This observation is in agreement with previous experiments where checkerboard DNA–DNA hybridization and differential culture gave similar results (McBain et al. 2005). Saliva samples taken from an individual over a course of 3 months had similarity indexes of c. 80% (Fig. 4). Such stability is somewhat surprising considering the open nature of the oral cavity and the considerable temporal variation in growth substrates which a normal diet will provide. A recent study by Rasiah et al. (2005) supports the observation of stability within oral microbiotas, suggesting the composition of oral consortia is host specific. These researchers used the multi-plaque artificial mouth model also support the observation that host-specific microbiotas can be maintained to a degree during the establishment of mature plaque microcosms (Rasiah et al. 2005). Microcosms normally rely on consistency in medium supply to achieve microbiological equilibrium, whereas the mechanisms for microbial equilibrium on a host-associated microflora remain poorly understood and possibly relate to immune factors that would be absent from a microcosm. In conclusion, we have shown that the MSD can be used to support stable, individualized oral microcosms and that when studying the micro-ecology of the oral cavity, such variation should be taken into account in order to maximize the relevance of the study. References Adams, H., Winston, M.T., Heersink, J., Buckingham-Meyer, K.A., Costerton, J.W. and Stoodley, P. (2002) Development of a laboratory model to assess the removal of biofilm from interproximal spaces by powered tooth brushing. Am J Dent 15, 12B–17B. Boon, N., De-Windt, W., Verstraete, W. and Top, E.M. (2002) Evaluation of nested PCR-DGGE (denaturing gel electrophoresis) with group-specific 16S rDNA primers for the analysis of bacterial communities from different waste water treatment plants. FEMS Microbiol Ecol 39, 101–112. Bradshaw, D.J. and Marsh, P.D. (1998) Analysis of pH-driven disruption of oral microbial communities in-vitro. Caries Res 32, 456–462. Bradshaw, D.J., Marsh, P.D., Allison, C. and Schilling, K.M. (1996) Effect of oxygen, inoculum composition and flow rate on development of mixed-culture oral biofilms. Microbiology 142, 623–639. Foster, J.S. and Kolenbrander, P.E. (2004) Development of a multispecies oral bacterial community in a saliva-conditioned flow cell. Appl Environ Microbiol 70, 4340–4348.
ª 2006 The Authors Journal compilation ª 2006 The Society for Applied Microbiology, Journal of Applied Microbiology 100 (2006) 1123–1131
R.G. Ledder et al.
Fujimoto, C., Maeda, H., Kokeguchi, S; Takashiba, S., Nishimura, F., Arai, H., Fukui, K. and Murayama, Y. (2003) Application of denaturing gradient gel electrophoresis (DGGE) to the analysis of microbial communities of subgingival plaque. J Periodontal Res 38, 440–445. Guggenheim, B., Giertsen, E., Schu¨pback, P. and Shapiro, S. (2001) Validation of an in-vitro biofilm model of supragingival plaque. J Dent Res 80, 363–370. Hopkins, M.J., Englyst, H.N., Macfarlane, S., Furrie, E., Macfarlane, G.T. and McBain, A.J. (2003) Degradation of cross-linked and non-cross-linked arabinoxylans by the intestinal microbiota in children. Appl Environ Microbiol 69, 6354–6360. Ibekwe, A.M., Papiernik, S.K., Gan, J., Yates, S.R., Yang, C.H. and Crowley, D.E. (2001) Impact of fumigants on soil microbial communities. Appl Environ Microbiol 67, 3245– 3257. Mager, D.L., Ximenez-Fyvie, L.A., Haffajee, A.D. and Socransky, S.S. (2003) Distribution of selected bacterial species on intra-oral surfaces. J Clin Periodontol 30, 644– 654. Marsh, P., Hunter, J.G., Bowden, I., Hamilton, A., McKee, A.S., Hardie, J. and Ellwood, D. (1983) The influence of growth rate and nutrient limitation on the microbial composition and biochemical properties of a mixed culture of oral bacteria grown in a chemostat. J Gen Microbiol 129, 755–770. McBain, A.J., Bartolo, R.G., Catrenich, C.E., Charbonneau, D., Ledder, R.G. and Gilbert, P. (2003a) Effects of triclosancontaining rinse on the dynamics and antimicrobial susceptibility of in vitro plaque ecosystems. Antimicrob Agents Chemother 47, 3531–3538. McBain, A.J., Bartolo, R.G., Catrenich, C.E., Charbonneau, D., Ledder, R.G. and Gilbert, P. (2003b) Effects of a chlorhexidine gluconate-containing mouthwash on the vitality and antimicrobial susceptibility of in-vitro oral bacterial ecosystems. Appl Environ Microbiol 69, 4770–4776. McBain, A.J., Sissons, C., Ledder, R.G., Sreenivasan, P.K., De Vizio, W. and Gilbert, P. (2005) Development and characterization of a simple perfused oral microcosm. J Appl Microbiol 98, 624–634. McKee, A.S., McDermid, A.S., Ellwood, D.C. and Marsh, P.D. (1985) The establishment of reproducible, complex communities of oral bacteria in the chemostat using defined inocula. J Appl Bacteriol 59, 263–275. Moore, W.E. and Moore, L.V. (1994) The bacteria of periodontal diseases. Periodontology 2000 5, 66–77. Muyzer, G. and Smalla, K. (1998) Application of denaturing gradient gel electrophoresis (DGGE) and temperature gradient gel electrophoresis (TGGE) in microbial ecology. Antonie Van Leeuwenhoek 73, 127–141. Paster, B.J., Boches, S.K., Galvin, J.L., Ericson, R.E., Lau, C.N., Levanos, V.A., Sahasrabudhe, A. and Dewhirst, F.E. (2001) Bacterial diversity in human subgingival plaque. J Bacteriol 183, 3770–3783.
Individualized oral microcosms
Pratten, J. and Wilson, M. (1999) Antimicrobial susceptibility and composition of microcosm dental plaques supplemented with sucrose. Antimicrob Agents Chemother 43, 1595– 1599. Pratten, J., Nazhat, S.N., Blaker, J.J. and Boccaccini, A.R. (2004) In vitro attachment of Staphylococcus epidermidis to surgical sutures with and without Ag-containing bioactive glass coating. J Biomater Appl 19, 47–57. Rasiah, I.A., Wong, L., Anderson, S.A. and Sissons, C.H. (2005) Variation in bacterial DGGE patterns from human saliva: over time, between individuals and in corresponding dental plaque microcosms. Arch Oral Biol 50, 779–787. Shah, H.N., Williams, R.A.D., Bowden, G.H. and Hardie, J.M. (1976) Comparison of the biochemical properties of Bacteroides melinogenicus from human dental plaque and other sites. J Appl Bacteriol 41, 473–492. Sissons, C.H. (1997) Artificial dental plaque biofilm model systems. Adv Dent Res 11, 110–126. Sissons, C.H., Cutress, T.W., Hoffman, M.P. and Wakefield, J.S. (1991) A multi-station dental plaque microcosm (Artificial Mouth) for the study of plaque growth, metabolism, pH and mineralization. J Dent Res 70, 1409– 1416. Sissons, C.H., Cutress, T.W., Faulds, G. and Wong, L. (1992) pH responses to sucrose and the formation of pH gradients in deep ‘‘artificial mouth’’ microcosm dental plaques. Arch Oral Biol 37, 913–922. Vroom, J.M., De Grauw, K.J., Gerritsen, H.C., Bradshaw, D.J., Marsh, P.D., Watson, G.K., Birmingham, J.J. and Allison, C. (1999) Depth penetration and detection of pH gradients in biofilms by two-photon excitation microscopy. Appl Environ Microbiol 65, 3502–3511. Wall-Manning, G.M., Sissons, C.H., Anderson, S.A. and Lee, M. (2002) Checkerboard DNA:DNA hybridisation technology focused on the analysis of gram positive cariogenic bacteria. J Microbiol Methods 51, 301–311. Walter, J., Tannock, G.W., Tilsala-Timisjarvi, A., Rodtong, S., Loach, D.M., Munro, K. and Alatossava, T. (2000) Detection and identification of gastrointestinal Lactobacillus species by using denaturing gradient gel electrophoresis and species-specific PCR primers. Appl Environ Microbiol 66, 297–303. Wimpenny, J.W.T. (1988) Introduction. In CRC Handbook of Laboratory Model Systems for Microbial Ecosystems ed. Wimpenny, J.W.T. pp. 1–17. Boca Raton, FL: CRC Press. Yang, C.H., Crowley, D.E., Borneman, J. and Keen, N.T. (2001) Microbial phyllosphere populations are more complex than previously realized. Proc Natl Acad Sci USA 27, 3889–3894. Zijnge, V., Harmsen, H.J., Kleinfelder, J.W., van der Rest, M.E., Degener, J.E. and Welling, G.W. (2003) Denaturing gradient gel electrophoresis analysis to study bacterial community structure in pockets of periodontitis patients. Oral Microbiol Immunol 18, 59–65.
ª 2006 The Authors Journal compilation ª 2006 The Society for Applied Microbiology, Journal of Applied Microbiology 100 (2006) 1123–1131
1131
