In vitro anaerobic biofilms of human colonic microbiota

Journal of Microbiological Methods 83 (2010) 296–301

Contents lists available at ScienceDirect

Journal of Microbiological Methods j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j m i c m e t h

In vitro anaerobic biofilms of human colonic microbiota K.M. Sproule-Willoughby a, M. Mark Stanton a, K.P. Rioux b, D.M. McKay c, A.G. Buret a,d, H. Ceri a,⁎ a

Biofilm Research Group, Department of Biological Sciences, University of Calgary, 2500 University Drive NW, Calgary, AB, Canada T2N 1N4 Department of Medicine, Division of Gastroenterology and Department of Microbiology and Infectious Disease, University of Calgary, Health Sciences Centre, 3330 Hospital Drive NW, Calgary, AB, Canada T2N 4N1 c Gastrointestinal Research Group, Department of Physiology and Pharmacology, the Calvin, Phoebe and Joan Snyder Institute for Infection, Immunity and Inflammation, University of Calgary, AB, Canada T2N 4N1 d Inflammation Research Network, University of Calgary, Faculty of Medicine, Health Sciences Centre, 3330 Hospital Drive NW, Calgary, AB, Canada T2N 4N1 b

a r t i c l e

i n f o

Article history: Received 11 June 2010 Received in revised form 22 September 2010 Accepted 24 September 2010 Available online 12 October 2010 Keywords: Anaerobic bacteria Biofilms Intestinal microbiota

a b s t r a c t The human gastrointestinal tract hosts a complex community of microorganisms that grow as biofilms on the intestinal mucosa. These bacterial communities are not well characterized, although they are known to play an important role in human health. This study aimed to develop a model for culturing biofilms (surfaceadherent communities) of intestinal microbiota. The model utilizes adherent mucosal bacteria recovered from colonic biopsies to create multi-species biofilms. Culture on selective media and confocal microscopy indicated the biofilms were composed of a diverse community of bacteria. Molecular analyses confirmed that several phyla were represented in the model, and demonstrated stability of the community over 96 h when cultured in the device. This model is novel in its use of a multi-species community of mucosal bacteria grown in a biofilm mode of growth. © 2010 Elsevier B.V. All rights reserved.

1. Introduction The human gastrointestinal tract harbors a complex symbiotic population of microbes which outnumbers the eukaryotic cells by tenfold (Tancrède, 1992). These organisms play an important role in maintaining our health; they help us digest our food, develop our immune system, and protect us from exogenous pathogens (Camp et al., 2009; Macfarlane and Macfarlane, 2006; Moreau and Corthier, 1988). Despite the importance of these organisms, relatively little is known about them. A major challenge in studying these communities is that approximately 80% of the species have never been cultured (Eckburg et al., 2005). Therefore, most studies cataloguing the organisms that inhabit our gastrointestinal tract are limited to DNA-based techniques,

Abbreviations: BBE, Bacteroides bile esculin agar; bp, base pair(s); CBA, Columbia blood agar; cfu, colony forming units; Ct, threshold cycle; dATP, deoxyadenosine triphosphate; dCTP, deoxycytodine triphosphate; ddH2O, double-distilled water; dGTP, deoxyguanosine triphosphate; DNA, deoxyribonucleic acid; dNTP, deoxynucleotide triphosphate; dTTP, deoxythymidine triphosphate; EDTA, (ethylenedinitrilo)-tetraacetic acid; FAA, Fastidious Anaerobe Agar; mEA, mEnterococcus agar; MRS, de Man, Rogosa and Sharpe; OPLS-DAPBS, orthogonal partial least squares discriminant analysis phosphate-buffered saline; PCR, polymerase chain reaction; PEA, phenylethanol agar; qPCR, quantitative polymerase chain reaction; sTSY, supplemented tryptic soy with yeast broth; TBE, Tris-borate-EDTA; T-RF, terminal restriction fragment; T-RFLP, terminal restriction fragment length polymorphism. ⁎ Corresponding author. Biofilm Research Group, Department of Biological Sciences, University of Calgary, 2500 University Drive NW, Calgary, AB, Canada T2N 1N4. Tel.: +1 403 220 6960; fax: +1 403 289 9311. E-mail address: [email protected] (H. Ceri). URL: http://homepages.ucalgary.ca/~ceri (H. Ceri). 0167-7012/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.mimet.2010.09.020

which do not give information about viability, metabolism, or other behaviors (such as interactions with other bacteria or host cells). In vitro models of intestinal bacterial populations are limited to those organisms that can be cultured using current techniques. Fermentation systems are able to model intestinal bacteria to an extent, but the seed communities for these experiments are generally fecal samples (Mäkivuokko and Nurminen, 2006). Fecal bacteria are not exact representatives of the mucosal bacterial populations that have the most interaction with human cells (Eckburg, et al., 2005; Probert and Gibson, 2002; Tannock, 2005). Another unique feature of these mucosal communities is they exist in a sessile, biofilm mode of growth, rather than a non-adherent planktonic state. Biofilms are known to behave differently than planktonic cells, and biofilms in the gastrointestinal tract have different metabolic activities than nonadherent populations (Macfarlane and Macfarlane, 2006). In vivo models of intestinal bacteria are also lacking. While animal models have been useful in assessing disease processes, the microbiota here do not closely represent human microbiota. Intestinal bacterial populations are relatively species specific (Henriksson, 2006) and it is normal for rodents to have commensal bacteria directly adherent to the intestinal epithelium, whereas in humans this is generally only the case in diseased individuals (Tannock, 2005). In order to gain a greater understanding of the complex ecosystem of our intestines, new methods for studying these bacterial communities are necessary. Descriptive studies identifying DNA profiles have provided the majority of the knowledge we have to date, but in order to truly understand these communities we must be able to grow these intestinal bacterial biofilms in vitro. A culture-based model is advantageous

K.M. Sproule-Willoughby et al. / Journal of Microbiological Methods 83 (2010) 296–301

because it can be easily manipulated and sampled, and the behaviors of the organisms within the communities can be evaluated. In this report, we describe a novel model for studying the colonic microbiota. The model is unique in that it uses mucosal, rather than luminal bacteria from the human colon, and allows these multispecies communities to grow in a surface-adherent biofilm state. 2. Methods and materials 2.1. Microbiological media Colonic microbiota were cultured on Columbia blood agar (CBA, Oxoid, Nepean, ON, Canada) or Fastidious Anaerobe Agar (FAA, Lab M, Bury, Lancashire, UK) with 5% defibrinated sheep blood (Dalynn Biologicals, Calgary, AB, Canada). Phenylethanol agar (PEA), Bacteroides bile esculin agar (BBE) and mEnterococcus agar (mEA) selective media were purchased from Dalynn Biologicals. de Man, Rogosa and Sharpe (MRS) agar was purchased from BD Biosciences (Mississauga, ON, Canada). Anaerobic bacteria were cultured in supplemented tryptic soy broth with yeast (sTSY). The base consisted of tryptic soy broth (EMD Chemicals, Gibbstown, NJ, USA) prepared to manufacturer's specifications supplemented with 4 g/l yeast extract (BD Biosciences). After autoclaving the following was added per 100 ml of base: 1 ml of 5% L-cysteine-HCl (Sigma Aldrich) solution in sterile, double distilled water (ddH2O) and 1 ml of hemin-menadione solution (50 mg bovine hemin (Sigma Aldrich), 1.74 g K2HPO4, 0.4 g NaOH, and 1 ml of a 1 mg/ml menadione (Sigma Aldrich) solution in 95% ethanol, all dissolved in 99 ml ddH2O).

297

ethics approval through the Intestinal Inflammation Tissue Bank at the Foothills Medical Centre in Calgary and met approval of the Conjoint Health Research Ethics Board of the University of Calgary and Alberta Health Services. Samples were collected in BBL Port-A-Cul tubes (BD Biosciences). Biopsies used for culture experiments were immediately placed in an anaerobic chamber (Bactron Anaerobic/Environmental Chamber; Sheldon Manufacturing Inc., Cornelius, OR, USA) with 90% nitrogen, 5% hydrogen and 5% carbon dioxide gases. Processing was carried out using a protocol adapted from Conte et al. (2006). Biopsies were incubated in 200 μL phosphate-buffered saline (PBS) with 0.016% dithiothreitol for 2 min and rinsed 3 times with gentle shaking in PBS. Biopsies to be used for direct DNA extraction were frozen in 80 μl PBS at −70 °C until extraction could be performed (Section 2.6). Biopsies to be used for biofilm formation were homogenized in 200 μL sTSY using a sterile mini-tissue homogenizer (Pellet Pestle® Microgrinder System; Kimble-Kontes, Vineland, NJ, USA). 2.3. Biofilm formation from colonic microbiota Biofilms were seeded immediately following biopsy collection using the Calgary Biofilm Device (Ceri et al., 1999) (commercially available as the MBEC™ P&G device, Innovotech Inc., Edmonton, AB, Canada). Biopsy homogenates were diluted 20- or 50-fold in sTSY, and 150 μl of the diluted homogenate were added to each well of the device. The devices were sealed in anaerobic bags (AnaeroGen™ Compact System, Oxoid, Nepean, ON, Canada), removed from the anaerobic hood, and placed on a gyrotary shaker at 150 rpm for 144 h. 2.4. Sampling of anaerobic microbiota biofilms

2.2. Biopsy collection and processing Mucosal biopsies of the ascending and descending colon were collected from five healthy volunteers undergoing colonoscopy for colon cancer screening (patients A, B, C, Y and Z) after standard colonoscopy preparation (Colyte™ polyethylene glycol–electrolyte solution). Patients who had taken antibiotics or probiotics in the four weeks prior to testing were excluded. Biopsies were obtained with

Bilofilm sampling was done inside an anaerobic chamber. Using sterile pliers, pegs were broken from the Calgary Biofilm Device. For viable cell counts, pegs were rinsed in 200 μl of 0.9% saline and each peg was sealed in a tube containing 200 μl saline. These tubes were removed from the anaerobic chamber, sonicated on high for 10 min, and returned to the anaerobic chamber. Samples were serially diluted in 0.9% saline and plated on FAA or CBA for viable cell counting.

Table 1 Primers used for PCR and qPCR studies. Target bacteria

Primer

Sequence (5′→3′)

Reference

All Eubacteria

Uni331F Uni797R g-BfraF g-BfraR Bac303F Bac708R BT-1 BT-2 BV-1 BV-2 BIA-1 BIA-2 BIL-1 BIL-2 CC-1 CC-2 g-Ccoc-F g-Ccoc-R FPR-1 FPR-2 LAA-1 LAA-2 PSP-1 PSP-2 g-Prevo-F g-Prevo-R

TCCTACGGGAGGCAGCAGT GGACTACCAGGGTATCTATCCTGTT CACRaGTAAACGATGGATGCC GGTCGGGTTGCAGACC GAAGGTCCCCCACATTG CAATCGGAGTTCTTCGTG GGCAGCATTTCAGTTTGCTTG GGTACATACAAAATTCCACACGT GCATCATGAGTCCGCATGTTC TCCATACCCGACTTTATTCCTT GGAAAGATTCTATCGGTATGG CTCCCAGTCAAAAGCGGTT GTTCCCGACGGTCGTAGAG GTGAGTTCCCGGCATAATCC CCGCATGGCAGTGTGTGAAA CTGCTGATAGAGCTTTACATA AAATGACGGTACCTGACTAA CTTTGAGTTTCATTCTTGCGAA AGATGGCCTCGCGTCCGA CCGAAGACCTTCTTCCTCC CATCCAGTGCAAACCTAAGAG GATCCGCTTGCCTTCGCA AACTCCGGTGGTATCAGATG GGGGCTTCTGAGTCAGGTA CACRaGTAAACGATGGATGCC GGTCGGGTTGCAGACC

Nadkarni et al. (2002)

Bacteroides fragilis group Bacteroides-Prevotella species Bacteroides thetaiotaomicron Bacteroides vulgatus Bifidobacterium adolescentis Bifidobacterium longum Clostridium clostridiiforme Clostridium coccoides group Faecalibacterium prausnitzii Lactobacillus acidophilus Peptostreptococcus productus Prevotella species a

R = purine.

Matsuki, et al. (2002) (Bartosch, et al., 2004; Bernhard and Field, 2000) Wang, et al. (1996) Wang, et al. (1996) Wang, et al. (1996) Wang, et al. (1996) Wang, et al. (1996) Matsuki, et al. (2002) Wang, et al. (1996) Wang, et al. (1996) Wang, et al. (1996) Matsuki, et al. (2002)

298

K.M. Sproule-Willoughby et al. / Journal of Microbiological Methods 83 (2010) 296–301

Colonies on the agar plates were counted after at least four days of anaerobic incubation at 37 °C. Samples to be used for DNA extraction were rinsed in 0.9% saline and frozen at −20 °C until extraction was performed (Section 2.6). 2.5. Confocal microscopy To visually confirm biofilm formation in the Calgary Biofilm Device, growth surfaces were stained without fixative using the Live/Dead® BacLight™ Kit (Molecular Probes, Burlington, ON, Canada) according to manufacturer's instructions. Biofilms were visualized using a Leica Microsystems (Bannockburn, IL, USA) DM IRE2 spectral confocal and multiphoton microscope with a Leica TCS SP2 acoustic optical beam splitter. Images were rendered using Imaris™ (Bitplane Scientific Software, Saint Paul, MN, USA). 2.6. DNA extraction DNA extraction was performed using the QIAamp® DNA Mini Kit (Qiagen, Mississauga, ON, Canada) and protocols from Conte et al. (2006) with the addition of a bead-beating step. Samples were thawed at room temperature and incubated with shaking at 37 °C for 30 min in 200 μl of lysozyme solution (20 mg/ml lysozyme (Sigma Aldrich) in 20 mM Tris–HCl, pH 8.0, 2 mM EDTA and 1.2% Triton X-100). Buffer ATL (200 μl; Qiagen) and Proteinase K (20 μl; Qiagen) were added to each sample, along with 0.5 g of sterile 0.1 mm glass beads (BioSpec Products Inc., Bartlesville, OK, USA). Samples underwent bead beating in a Precellys® 24 lysis homogenization system (Bertin Technologies, Montigny le Bretonneux, France) at 6500 rpm with three cycles of 30 s beating and 45 s rest. Samples were incubated at 56 °C for 30 min and 95 °C for 10 min. Two hundred microliters of Buffer AL (Qiagen) were added and samples were mixed by pulse vortexing for 15 s and incubated at 70 °C for 10 min. Two hundred microliters of absolute ethanol were added and the samples were mixed by pulse vortexing for 15 s. For biofilm samples analyzed by qPCR, DNA from 4 pegs was pooled. Samples were applied to QIAamp® Mini Spin Columns (Qiagen) and the remaining washing and elution steps were carried out according to manufacturer's instructions (DNA was eluted in 60 μl of Buffer AE). Samples were quantified by absorbance at 260 nm using an Ultrospec 200 UV/visible spectrophotometer (BioChrom, Cambridge, UK). DNA extractions for terminal restriction fragment length (T-RFLP) analysis were performed using the procedure described above without the bead-beating step. When DNA was extracted from colonies on agar plates a swab was used to transfer the bacteria into 2 mL screw-cap tubes. Biofilm samples were not pooled when DNA was extracted for T-RFLP analysis.

PCR products were separated on 0.7% to 1.0% agarose gels (made with Ultrapure Agarose; Invitrogen) in 1X Tris-borate-EDTA (TBE) buffer (10.9 g Tris, 5.5 g H3BO3 and 0.465 g EDTA per L). One microliter loading dye (4 g C12H22O11 and 25 mg bromophenol blue (Bio-Rad, Mississauga, ON, Canada) in 10 ml ddH2O) was added to 5 μl product and the samples were applied to the wells of an agarose gel, along with the appropriate ladder (1 Kb+ DNA ladder or 50 bp DNA ladder; Invitrogen). Gels were run in a Sub-Cell GT MINI gel electrophoresis unit (Bio-Rad) at 70–100 V for 30 to 50 min. Products were visualized with a UV transilluminator (Bio-Rad) after staining for 15–20 min with 0.5 μg/ml ethidium bromide in TBE buffer. Gels were photographed using a Gel Doc 2000 and visualized using Quantity One Software (Bio-Rad).

2.8. Terminal restriction fragment length polymorphism analysis Reaction mixtures for PCR (50 μl each) consisted of 1 μl template DNA (5 ng/μl), 5 μl 10× PCR reaction buffer (without MgCl2; Invitrogen), 1.5 μl MgCl2 (50 mM), 1 μl dNTP mix (10 mM), 1 μl forward primer (10 pmol/μl), 1 μl reverse primer (10 pmol/μl), 0.25 μl recombinant Taq polymerase and 39.25 μl ddH2O. VIC-labeled universal Eubacterial forward primer 8f (5′-VIC®-AGA GTT TGA TCC TGG CTC AG-3′) was purchased from Applied Biosystems (Streetsville, ON, Canada). Universal Eubacterial reverse primer 926r (5′-CCG TCA ATT CCT TTG AGT TT-3′) was purchased from Sigma Aldrich. The 8f/926r primer set flanks the V1–V3 variable regions of the bacterial 16S rRNA gene (Liu et al., 1997). Recombinant Taq polymerase and MgCl2 were acquired from Invitrogen and all other PCR reagents were purchased from Qiagen. PCR reactions were carried out in a GeneAmp 2400 PCR System thermocycler (Applied Biosystems). The reaction conditions were as follows: 94 °C for 2 min, 25 cycles of 94 °C, 56 °C and 72 °C for 1 min each, and a final extension at 72 °C for 10 min. PCR reactions were run in triplicate, pooled and purified using the QIAquick® PCR purification kit (Qiagen). Samples were eluted in 30 μl elution buffer. Purified PCR products were digested with HpaII (Invitrogen) overnight at 37 °C. Reaction mixtures (20 μl total) consisted of the following: 5 μl PCR product, 2 μl 10× REACT® 8 buffer (Invitrogen), 1 μl HpaII (20 units/μl; Invitrogen) and 12 μl ddH2O. Digests were ended by incubation at 65 °C for 15 min. Digested products were purified with the QIAquick® PCR purification kit (Qiagen) and eluted in 40 μl elution buffer. Five microliters of the purified product were submitted to the University of Calgary Core DNA Services (Calgary, AB, Canada) for fragment analysis using a 3730 × l (96 capillary) genetic analyzer (Applied Biosystems). The LIZ1200 size standard (Applied Biosystems) was used to size fragments along with the G5 filter set to detect the VIC®-labelled fragments. T-RFLP data was analyzed using GeneMapper 3.0 Software (Applied Biosystems). Binning was performed at 1 bp and only peaks with a minimum peak threshold of 50 fluorescence units were

2.7. Group-specific PCR DNA extracted from biopsies and biofilms was analyzed using PCR with group-specific primers. Table 1 summarizes the primers used for group-specific PCR as well as qPCR. Reactions mixtures (50 μl total) consisted of 0.5 μl DNA template, 5 μl 10× PCR reaction buffer (Invitrogen, Burlington, ON, Canada), 3 μl 50 mM MgCl2 (Invitrogen), 1 μl dNTP mix (10 mM each of dATP, dCTP, dGTP and dTTP); Qiagen), 1 μl forward primer (50 pmol/μl), 1 μl reverse primer (50 pmol/μl), 0.25 μl recombinant Taq polymerase (Invitrogen) and 38.25 μl ddH2O. PCR reaction conditions for Bacteroides fragilis and Prevotella species were: 94 °C for 7 min, 40 cycles of: 94 °C for 20 s, 55 °C for 20 s and 70 °C for 30 s, followed by incubation at 72 °C for 5 min (Matsuki et al., 2002). For all other primer sets the reaction conditions were: 95 °C for 7 min, 35 cycles of: 94 °C for 30 s, 55 °C for 10 s and 74 °C for 33 s, by incubation at 74 °C for 2 min and 45 °C for 2 s (Wang et al., 1996).

Fig. 1. Bacteria recovered from anaerobic biofilms of colon bacteria. Biopsies from the descending colon of patients Y (biopsies Y1 and Y2) and Z (biopsies Z1 and Z2) were processed and used to seed biofilms in the Calgary Biofilm Device, which were enumerated after 144 hours by plating on agar media. Four biofilms were enumerated for each biopsy.

K.M. Sproule-Willoughby et al. / Journal of Microbiological Methods 83 (2010) 296–301

299

Fig. 2. Confocal microscopy images of anaerobic biofilms of colon bacteria. Representative images of anaerobic biofilms formed in vitro from colonic bacteria (biofilms shown were formed from the colonic microbiota of patient B (A) and patient C (B). Biofilms were stained with a live/dead stain and imaged using confocal scanning laser microscopy. Images were rendered using Imaris™ (Bitplane Scientific Software).

included. Fragments with a relative abundance of ≥1% in at least one sample were included in statistical analyses. 2.9. Quantitative PCR Quantitative PCR (qPCR) was used to measure specific bacterial groups in biopsy and biofilm samples. The groups that were quantified were Bacteroides-Prevotella species and the Clostridium coccoides group (primers are listed in Table 1). qPCR reactions were carried out using the iQ5 Real-Time PCR Detection System and iQ5 software (Bio-Rad). Reaction mixtures (25 μl total) were made up of 4 μl template DNA (5 ng/μl), 1 μl each primer (50 pmol/μl), 12.5 μl 2X iQ SYBR Supermix (Bio-Rad) and 6.5 μl ddH2O. Reaction conditions were as follows: 95 °C for 3 min, 35 cycles (Bacteroides-Prevotella) or 40 cycles (C. coccoides) of: 95 °C for 30 s, 58 °C (Bacteroides-Prevotella) or 55 °C (C. coccoides) for 30 s and 72 °C for 30 s (Bartosch et al., 2004; Bernhard and Field, 2000; Wang, et al., 1996). All qPCR reactions were followed by melt curve analysis from 60 °C to 99 °C with a temperature increase of 1 °C per second. Standard curves (Ct verses log10 cfu equivalents) were created for each experiment using serial dilutions of a known quantity of DNA from a single species. Cfu equivalents per biofilm sample were calculated from ng of DNA using the method of Harrow et al. (2007). Average genome sizes were determined from a review by Fogel et al. (1999).

Table 2 Growth of biofilm bacteria on selective media. Patient Y Biofilm Y1

Biofilm Y2

Biofilm Z1

Biofilm Z2

MacConkey agara

+ (Pink and white) + + (Black) + + (Purple)

+ (Pink) + + (Black) + + (Purple)

−

+ (Pink) + + (Black) + + (Purple)

MRS agard mEAe

Group data are expressed as mean ± standard deviation. Figures and univariate analyses were compiled using Prism 5 Software (GraphPad Inc., La Jolla, CA, USA). Viable cell count and qPCR data were transformed by log base 10 and analyzed by Kruskal–Wallis test with Dunn's multiple comparison post-test. P values b 0.05 were considered statistically significant. T-RFLP data were analyzed using multivariate analysis by supervised orthogonal partial least squares discriminant analysis (OPLS-DA), which allowed for the direct comparison of the variance between origin of sample (biopsy or biofilm; y variable) and fragment relative abundances (x variable). OPLS-DA was performed using SIMCA-P software (Umetrics, Sweden). 3. Results and discussion Biofilms formed in the Calgary Biofilm Device by the mucosal microbiota from descending colon biopsies were assessed by viable cell counting. The anaerobic bacteria formed biofilms ranging from 2 × 103–2 × 105 cfu of culturable bacteria per peg (Fig. 1); this process only enumerated bacteria that were culturable as pure colonies on agar media. Plate counts likely underestimated the total size of the biofilms, since some bacteria may have grown in the complex community of the biofilm but did not grow on the agar media. Confocal microscopy confirmed the successful development of biofilms from colonic biopsy samples (Fig. 2). Biofilm thicknesses at the atmosphere-liquid interface ranged from 20–80 μm.

Patient Z

Medium

PEAb BBEc

2.10. Statistical analysis

+ + (Black) + + (Purple)

a Bacterial strains that ferment lactose grow as pink colonies (E. coli or other Enterobacteriaceae) on MacConkey agar. E. coli also produces a bile precipitate. Non-lactose fermenting Enterobacteriaceae grow as white colonies (Difco, 1985). b Phenylethanol agar, selective or staphylococci and streptococci (Difco, 1985). c Bacteroides bile esculin agar, members of the Bacteroides fragilis group turn BBE black (Livingston et al., 1978). d deMan, Rogosa and Sharpe agar, selective for lactobacilli (Difco, 1985). e mEnterococcus agar, selective for enterococci; E. faecalis grows as dark purple colonies (Difco, 1985).

Table 3 Presence of bacterial groups or species in colon biopsy samples and biofilms. Patient Y Target organism or group

Bacteroides fragilis group Bacteroides thetaiotaomicron Bacteroides vulgatus Bifidobacterium adolescentis Bifidobacterium longum Clostridium clostridiiforme Faecalibacterium prausnitzii Lactobacillus acidophilus Peptostreptococcus productus Prevotella species Bx = biopsy sample. Bio = biofilm sample.

Patient Z

Bx

Bx

Bio

Bio

Bx

Bx

Bio

Bio

Y3

Y4

Y1

Y2

Z3

Z4

Z1

Z2

+ + + + + + − + + +

+ + + − + + − + + +

+ + + − + − − + + −

+ + + − + − − + − −

+ + + − + + + + + −

+ + + − + + − + + −

+ + + − + + − + + −

+ + + + + + + + + −

300

K.M. Sproule-Willoughby et al. / Journal of Microbiological Methods 83 (2010) 296–301

Table 4 OPLS-DA analysis of T-RFLP data sets derived directly from biopsy and biofilm DNA or from cultured samples of the biopsies and biofilms. Sample Preparation

Fragment diversity captured (R2X)

Difference between biopsy and biofilm captured (R2Y)

Cross-validation strength of difference (Q2Y)

Direct DNA extraction Culture on agar media prior to extraction

0.96 0.40

0.99 0.44

0.94 0.003

Colonies from the viable cell count plates were cultured on differential media; the results are presented in Table 2. All of the biofilm samples formed colonies on media that selected for the following bacteria: Streptococcus and/or Staphylococcus species, the Bacteroides fragilis group, lactobacilli and enterococci. Enterobacteriaceae were also detected in 3 of the 4 samples. The results indicate the presence of bacteria from at least three different phyla, including both Gram positive and Gram negative species, demonstrating that this model system successfully establishes complex, multi-species biofilms. PCR with group-specific primers confirmed that both biopsies and biofilms contained a diverse community of organisms (Table 3). The biopsy samples contained representatives from prevalent phyla in the colon, such as Bacteroidetes (B. fragilis group), Firmicutes (C. coccoides group, C. leptum group and lactobacilli) and Actinobacteria (bifidobacteria). Importantly, not only were the biofilms diverse, but they included the majority of bacterial groups and species that were detected in the biopsies, suggesting that the biofilm model captures some of the diversity of the colonic microbiota. T-RFLP was used to determine the differences in community composition between colon biopsies and the biofilms formed in the Calgary Biofilm Device. OPLS-DA was able to strongly predict the differences between biopsy and biofilm composition for T-RFLP data gathered directly from samples, but not after culture on agar media (Table 4). While biopsy and biofilm samples were clearly separated in the OPLS-DA model, 73% (38/52) of the fragments did not differ significantly between biopsy and biofilm samples (Fig. 3). These results provide further evidence that a subset of the colonic microflora was successful cultured in the biofilm model. Sixty unique T-RFs were identified in the biopsy samples grown on agar plates; 85% (51 of 60) of these T-RFs were also present in the agar plate samples of at least one biofilm. Culturing the community on agar may have caused a similar shift in community structure in both biopsy and biofilm samples, and therefore decreased the differences seen between them.

In an attempt to further compare bacterial communities from the biopsies with those grown in the present model, qPCR was used to enumerate two important groups of the human gut microbiota: the Bacteroides–Prevotella group and the C. coccoides cluster. These groups were tested for in the biopsy samples and corresponding biofilms of patients A, B and C. Patient A biopsies had lower numbers of both Bacteroides-Prevotella and C. coccoides group bacteria than patient B and C biopsies (Table 5). Similarly, when grown in the biofilm model, bacterial communities from patient A contained fewer BacteroidesPrevotella and C. coccoides bacteria than those of patients B and C. This observation supports our hypothesis that the bacterial biofilms grown in this model reflect at least some of the differences between the microbial communities of each individual. The presence of these groups was maintained over a 96-hour period of biofilm growth, indicating that the diversity of the biofilm communities continued throughout the growth period of the biofilms in the device (Table 4).

3.1. Conclusions In summary, we have created a novel model for studying selected representatives from the human colonic microbiota. This model uses mucosal bacterial communities from the human colon, and allows them to grow in a surface-adherent mode of growth. The resulting biofilms were complex, multi-species communities and contained representatives from the most prolific bacterial groups in the human colon. In this model system, biofilms are stable in composition over an extended period, making them uniquely useful for investigating the effects of exogenous microbial, environmental, and pharmaceutical/ pharmabiotic influences on bacterial community structure and function in the intestine.

Acknowledgements This work was supported by grants from the Crohn's and Colitis Foundation of Canada to HC, DMM and AGB, and the Natural Science and Engineering Research Council of Canada (NSERC) to HC and AGB. KMS was supported by graduate student fellowships from NSERC and the Alberta Heritage Foundation for Medical Research. Special thanks to Ms. Ida Rabbani of the Intestinal Inflammation Tissue Bank at the University of Calgary for providing the biopsy samples. The authors are grateful to Dr. D. Morck, Dr. J. Kauffman, K. Cannon, and G. Duggan for sharing equipment and technical expertise.

Fig. 3. OPLS-DA analysis of T-RFLP data obtained directly from biopsy and biofilm samples. Biopsies and biofilms were separated on the basis of T-RFLP profiles by the OPLS-DA model (A). However, only 14 of the 52 fragments (white bars) were statistically different between the biopsy and biofilm communities (B). R2X = 0.225362.

K.M. Sproule-Willoughby et al. / Journal of Microbiological Methods 83 (2010) 296–301

301

Table 5 Enumeration of Bacteroides–Prevotella and C. coccoides group bacteria in descending colon biopsies. Bacteroides–Prevotella species Patient A B C

Biopsiesa 5.10 ± 0.21 5.89 ± 0.26 6.04 ± 0.30

Clostridium coccoides group

Biofilms (48 h)b c

5.00 ± 0.51 5.96 ± 0.55 6.17 ± 0.71

d

Biofilms (96 h)b 4.40 ± 0.38 6.09 ± 0.12 5.32 ± 0.56

d

Biopsiesa 2.72 ± 0.21 4.35 ± 0.31 4.02 ± 0.30

Biofilms (48 h)b c

1.98 ± 0.74 3.61 ± 0.66 3.34 ± 0.60

e

Biofilms (96 h)b 1.42 ± 0.13e 3.43 ± 0.75 3.46 ± 0.53

a At least five biopsies were tested from each patient; each biopsy was tested in duplicate. Results are displayed as log10 average cfu per 20 ng sample of extracted DNA ± standard deviation. b At least four biofilms were tested from each patient per timepoint; each biofilm was tested in duplicate. Results are displayed as log10 average cfu per 20 ng sample of extracted DNA ± standard deviation. c Biopsies: Bacteroides–Prevotella patient A vs. patients B (p b 0.001) and C (p b 0.001). C. coccoides group patient A vs. patient B (p b 0.01) and patient C (p b 0.001). d Bacteroides-Prevotella biofilms: at 48 h, patient A vs. patients B (p b 0.05) and C (p b 0.01); at 96 h, patient A vs. patient C (p b 0.001). e C. coccoides group biofilms: at 48 h, patient A vs. patients B and C (p b 0.01); at 96 h, patient A vs. patients B and C (p b 0.01).

References Bartosch, S., Fite, A., Macfarlane, G.T., McMurdo, M.E.T., 2004. Characterization of bacterial communities in feces from healthy elderly volunteers and hospitalized elderly patients by using real-time PCR and effects of antibiotic treatment on the fecal microbiota. Appl. Environ. Microbiol. 70, 3575–3581. Bernhard, A.E., Field, K.G., 2000. A PCR assay to discriminate human and ruminant feces on the basis of host differences in Bacteroides-Prevotella genes encoding 16S rRNA. Appl. Environ. Microbiol. 66, 4571–4574. Camp, J.G., Kanther, M., Semova, I., Rawls, J.F., 2009. Patterns and scales in gastrointestinal microbial ecology. Gastroenterology 136, 1989–2002. Ceri, H., Olson, M.E., Stremick, C., Read, R.R., Morck, D., Buret, A., 1999. The Calgary Biofilm Device: new technology for rapid determination of antibiotic susceptibilities of bacteria biofilms. J. Clin. Microbiol. 37, 1771–1776. Conte, M.P., Schippa, S., Zamboni, I., Penta, M., Chiarini, F., Seganti, L., Osborn, J., Falconieri, P., Borrelli, O., Cucchiara, S., 2006. Gut-associated bacterial microbiota in paediatric patients with inflammatory bowel disease. Gut 55, 1760–1767. Difco, 1985. Difco Manual. Difco Laboratories Incorporated, Detroit. Eckburg, P.B., Bik, E.M., Bernstein, C.N., Purdom, E., Dethlefsen, L., Sargent, M., Gill, S.R., Nelson, K.E., Relman, D.A., 2005. Diversity of the human intestinal microbial flora. Science 308, 1635–1638. Fogel, G.B., Collins, C.R., Li, J., Brunk, C.F., 1999. Prokaryotic genome size and SSU rDNA copy number: estimation of microbial relative abundance from a mixed population. Microb. Ecol. 38, 93–113. Harrow, S.A., Ravindran, V., Butler, R.C., Marshall, J.W., Tannock, G.W., 2007. Real-time quantitative PCR measurement of ileal Lactobacillus salivatius populations from broiler chickens to determine the influence of farming practices. Appl. Environ. Microbiol. 73, 7123–7127. Henriksson, A., 2006. Animal models for the human gastrointestintal tract. In: Ouwehand, A., Vaughan, E.E. (Eds.), Gastrointestinal Microbiology. Taylor and Francis Group, New York, pp. 253–272. Liu, W.T., Marsh, T.L., Cheng, H., Forney, L.J., 1997. Characterization of microbial diversity by determining terminal restriction fragment length polymorphisms of genes encoding 16S rRNA. Appl. Environ. Microbiol. 63, 4516–4522. Livingston, S.J., Kominos, S.D., Yee, R.B., 1978. New medium for selection and presumptive identification of Bacteroides fragilis group. J. Clin. Microbiol. 7, 448–453.

Macfarlane, S., Macfarlane, G.T., 2006. Composition and metabolic activities of bacterial biofilms colonizing food residues in the human gut. Appl. Environ. Microbiol. 72, 6204–6211. Mäkivuokko, H., Nurminen, P., 2006. In vitro methods to model the gastrointestinal tract. In: Ouwehand, A., Vaughan, E.E. (Eds.), Gastrointestinal Microbiology. Taylor and Francis Group, New York, pp. 237–252. Matsuki, T., Watanabe, K., Fujimoto, J., Miyamoto, Y., Takada, T., Matsumoto, K., Oyaizu, H., Tanaka, R., 2002. Development of 16 S rRNA-gene-targeted group-specific primers for the detection and identification of predominant bacteria in human feces. Appl. Environ. Microbiol. 68, 5445–5451. Moreau, M.C., Corthier, G., 1988. Effect of the gastrointestinal microflora on induction and maintenance of oral tolerance to ovalbumin in C3H–HrJ mice. Infect. Immun. 56, 2766–2768. Nadkarni, M.A., Martin, F.E., Jacques, N.A., Hunter, N., 2002. Determination of bacterial load by real-time PCR using a broad-range (universal) probe and primers set. Microbiology 148, 257–266. Probert, H.M., Gibson, G.R., 2002. Bacterial biofilms in the human gastrointestinal tract. Curr. Issues Intest. Microbiol. 3, 23–27. Tancrède, C., 1992. Role of the human microflora in health and disease. Eur. J. Clin. Microbiol. Infect. Dis. 11, 1012–1015. Tannock, G.W., 2005. Microbiota of mucosal surfaces in monogastric animals. In: Nataro, J.P., Cohen, P.S., Mobley, H.L.T., Weiser, J.N. (Eds.), Colonization of Mucosal Surfaces. ASM Press, Washington, pp. 163–178. Wang, R.F., Cao, W.W., Cerniglia, C.E., 1996. PCR detection and quantitation of predominant anaerobic bacteria in human and animal fecal samples. Appl. Environ. Microbiol. 62, 1242–1247.

Glossary Biofilm: a surface-adherent community of microorganisms, usually encased in a matrix of extracellular polysaccharide. Microorganisms growing in a biofilm are phenotypically distinct from free-swimming (planktonic) cells. Planktonic: a free-swimming or free-floating mode of growth.