Gene expression analysis of monospecies Salmonella Typhimurium biofilms using Differential Fluorescence Induction

Journal of Microbiological Methods 84 (2011) 467–478

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

Gene expression analysis of monospecies Salmonella Typhimurium biofilms using Differential Fluorescence Induction Kim Hermans, T.L. Anh Nguyen, Stefanie Roberfroid, Geert Schoofs, Tine Verhoeven, David De Coster, Jos Vanderleyden ⁎, Sigrid C.J. De Keersmaecker ⁎⁎ Centre of Microbial and Plant Genetics, Department of Microbial and Molecular Systems, Katholieke Universiteit Leuven, Kasteelpark Arenberg 20, 3001 Leuven, Belgium

a r t i c l e

i n f o

Article history: Received 28 September 2010 Received in revised form 10 January 2011 Accepted 14 January 2011 Available online 21 January 2011 Keywords: Salmonella Biofilm DFI

a b s t r a c t Bacterial biofilm formation is an important cause of environmental persistence of food-borne pathogens, such as Salmonella Typhimurium. As the ensemble of bacterial cells within a biofilm represents different physiological states, even for monospecies biofilms, gene expression patterns in these multicellular assemblages show a high degree of heterogeneity. This heterogeneity might mask differential gene expression that occurs only in subpopulations of the entire biofilm population when using methods that average expression output. In an attempt to address this problem and to refine expression analysis in biofilm studies, we used the Differential Fluorescence Induction (DFI) technique to gain more insight in S. Typhimurium biofilm gene expression. Using this single cell approach, we were able to identify 26 genetic loci showing biofilm specific increased expression. For a selected number of identified genes, we confirmed the DFI results by the construction of defined promoter fusions, measurement of relative gene expression levels and construction of mutants. Overall, we have shown for the first time that the DFI technique can be used in biofilm research. The fact that this analysis revealed genes that have not been linked with Salmonella biofilm formation in previous studies using different approaches illustrates that no single technique, in casu biofilm formation, is able to identify all genes related to a given phenotype. © 2011 Elsevier B.V. All rights reserved.

1. Introduction During the last decades, it has become increasingly clear that bacteria, including pathogens such as Salmonella enterica serovar Typhimurium (S. Typhimurium), grow predominantly as biofilms in most of their natural habitats, rather than in planktonic mode (HallStoodley et al., 2004). S. Typhimurium is a Gram-negative, enteropathogenic bacterium that causes host-specific diseases ranging from self-limiting food-borne gastroenteritis to life-threatening systemic infections. Salmonella is capable of forming microcolonies and even mature biofilms on a wide diversity of surfaces, ranging from abiotic (Kusumaningrum et al., 2003; Latasa et al., 2005; Romling et al., 1998) to biotic (Barak et al., 2008; Boddicker et al., 2002; Brandl and Mandrell, 2002; Prouty et al., 2002) ones. Bacterial biofilms can be defined as structured communities of bacterial cells enclosed in a selfproduced matrix, adhering to inert or living surfaces (Costerton et al., 1999). Biofilm formation has been stated as a potential cause of the emerging (multi)drug resistance (Lewis, 2008), because of the protecting action of the self-produced matrix (Fux et al., 2005) and adaptation mechanisms of the bacteria residing in these multicellular

⁎ Corresponding authors: Tel.: +32 16321631; fax: +32 16321966. E-mail address: [email protected] (J. Vanderleyden). 0167-7012/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.mimet.2011.01.012

structures. Furthermore, Salmonella biofilm formation is an important survival strategy in non-host environments, which are fundamentally different from typical host environments (Mouslim et al., 2002; Romling et al., 2007), a strategy to induce chronic infections (Costerton et al., 1995; Davey and O'Toole, 2000) and even a possible way to colonize host organisms (Boddicker, et al., 2002; Prouty and Gunn, 2003). Taken together, Salmonella biofilm formation can be seen as an essential and integral part of the pathogen's life cycle and a source of reappearing infections by this pathogen (Rasschaert et al., 2007). Bacterial cells residing in biofilms are not only physiologically distinct from planktonic cells (with different gene expression patterns), but also vary from each other spatially, temporally and genetically as the biofilm formation proceeds (Sauer et al., 2002; Stewart and Franklin, 2008; Stoodley et al., 2002). High-throughput DNA microarray studies have been conducted to study biofilm formation in many model organisms and have identified a large number of genes showing differential expression under biofilm conditions (e.g. (Beenken et al., 2004; Beloin et al., 2004; Hamilton et al., 2009; Ren et al., 2004; Schembri et al., 2003; Shemesh et al., 2008; Whiteley et al., 2001)). This transcriptional profiling technique, however, generates a global value for the whole biofilm population (An and Parsek, 2007; Lazazzera, 2005; Stewart and Franklin, 2008) and as such, differences in gene expression patterns of subpopulations within biofilms are not taken into account.

468

K. Hermans et al. / Journal of Microbiological Methods 84 (2011) 467–478

In an attempt to address this problem and to refine expression analysis in biofilm studies, we decided to use the Differential Fluorescence Induction (DFI) single cell approach to study Salmonella biofilm formation. This technique was introduced by the Falkow group to sort S. Typhimurium clones differentially expressing GFP in low pH environments and within host cells (Valdivia and Falkow, 1996, 1997). DFI basically is an enrichment strategy using fragments of bacterial genomic DNA, cloned upstream of a promoterless gfp, and a flow cytometer with sorting module to monitor promoter activity. In this study, the genetic enrichment was performed by alternating rounds of positive (biofilm-inducing) and negative (planktonic conditions) selection of the bacterial population. This resulted progressively in the generation of bacterial subpopulations showing enrichment in biofilm upregulated inserts. Subsequent sequence determination of the genomic inserts in these enriched pools led to identification of DNA sequences that caused increased expression of the promoterless gfp gene in biofilm conditions in non-host environments. Using this enrichment technique in a biofilm context, we identified 26 genetic loci showing Salmonella biofilm specific induction of which 17 coincided with promoter regions of already annotated genes. For a selected number of DFI identified genes, we confirmed the results of the screening by the construction of defined promoter fusions. We also measured relative gene expression levels with qRT-PCR for a selection of DFI-retrieved genes to compare differential expression directly at the RNA level. Finally, for some of the identified genes, we investigated their impact on Salmonella biofilm formation by constructing and analyzing corresponding mutants. 2. Materials and methods 2.1. Bacterial strains, plasmids and media All strains and plasmids used in this study are listed in Table 1. In the planktonic state, S. Typhimurium strains were grown with

aeration at 37 °C in Luria-Bertani (LB) broth (Sambrook and Russel, 2001) or on LB plates containing 15 g/l agar (Invitrogen). If appropriate, antibiotics were added at the following concentrations: ampicillin (Ap), 100 μg/ml; chloramphenicol (Cm), 25 μg/ml and streptomycin (Sm), 25 μg/ml. Tryptic soy broth (BD Biosciences, 30 g/l) diluted 1/20 (TSB 1/20) was used for biofilm formation assays (De Keersmaecker et al., 2005). Standard protocols were used for molecular cloning (Sambrook and Russel, 2001). Restriction enzymes were purchased from New England Biolabs and used according to the manufacturer's instructions. Cloning steps were performed using E. coli DH5α and E. coli TOP10F′ (Sambrook and Russel, 2001) and the final, constructed plasmids were electroporated to the S. Typhimurium SL1344 strains using a Bio-Rad gene pulser. All primers used, as well as their purposes, are listed in Table 2. The sequences used for primer construction were obtained from the complete genome sequence of S. Typhimurium SL1344 (Hoiseth and Stocker, 1981), as available via the website of the Sanger Institute (U.K.), (http://www.sanger.ac.uk/ Projects/Salmonella). Reporter plasmids pCMPG5521, pCMPG5532, pCMPG5533 and pCMPG5539 were constructed by cloning the PCRamplified csgD, potF, STM1851 and csgB promoter regions, respectively, as a BamHI (for pCMPG5521 and pCMPG5539) or XbaI/BamHI (for pCMPG5532 and pCMPG5533) fragment into pFPV25. Complementation plasmids pCMPG5522, pCMPG5531 and pCMPG5538 were constructed by cloning the PCR-amplified potF, potFGHI and sitABCD coding sequences, as EcoRI (pCMPG5522) and XbaI/BamHI (pCMPG5531, pCMPG5538) fragments downstream of the constitutive nptII promoter into the RK2 based plasmid pFAJ1708 (Dombrecht et al., 2001). As such, complementation experiments were performed using this nptII promoter to drive expression of the mentioned genes. Correct orientation of all fragments was checked by PCR and restriction analysis. S. Typhimurium SL1344 mutants were constructed using the one-step chromosomal inactivation protocol, as previously described by Datsenko and Wanner (Datsenko and

Table 1 Bacterial strains and plasmids. Name Strains E. coli DH5α E. coli TOP10F′

S. Typhimurium SL1344 CMPG 5521 CMPG 5522 CMPG 5537 CMPG5579 CMPG5584 CMPG5589 CMPG10301 CMPG10305 CMPG10309 Plasmids pCMPG5521 pCMPG5522 pCMPG5531 pCMPG5532 pCMPG5533 pCMPG5538 pCMPG5539 pCP20 pFPV25

pFPV25.1 pKD3 pKD46 pFAJ1708

Description

Reference

F− endA1 glnV44 thi-1 recA1 relA1 gyrA96 deoR nupG − + ϕ80dlacZΔM15 Δ(lacZYA-argF)U169, hsdR17(r− K mK ), λ F′ {lacIq Tn10(TetR)} mcrA Δ(mrr-hsdRMS-mcrBC) ϕ80lacZΔM15 ΔlacX74 deoR recA1 araD139 Δ(ara-leu)7697 galU galK rpsL (StrR) endA1 nupG Parent strain, xyl hisG rpsL; virulent; SmR S. Typhimurium SL1344, potF::Cm S. Typhimurium SL1344, ΔpotF S. Typhimurium SL1344, ΔSTM1851 S. Typhimurium SL1344, ΔcsgD S. Typhimurium SL1344, sitA::Cm S. Typhimurium SL1344, ΔcpxP S. Typhimurium SL1344, ΔsanA S. Typhimurium SL1344, fhuA::Cm S. Typhimurium SL1344, ΔcsgB

Gibco BRL

csgD promoter cloned BamHI in pFPV25 potF cloned EcoRI in pFAJ1708 potF cloned XbaI/BamHI in pFAJ1708 potF promoter cloned XbaI/BamHI in pFPV25 STM1851 promoter cloned XbaI/BamHI in pFPV25 sitABCD cloned XbaI/BamHI in pFAJ1708 csgB promoter cloned BamHI in pFPV25 flp, ts-rep-[ciI857](λ) ts, ApR, CmR Promoter-trap vector constructed by inserting an EcoRI-HindIII fragment containing a promoterless gfpmut3 (Cormack, et al., 1996) into plasmid pED350 (colE1, bla, mob) (Derbyshire and Willetts, 1987); ApR 0.6 kb Sau3AI fragment inserted in the BamHI site of pFPV25, containing the promoter region of S. Typhimurium rpsM encoding for the ribosomal protein S13 (constitutive promoter); ApR Plasmid used as template for construction of Salmonella mutants; ApR, CmR Lambda Red helper plasmid, ApR Derivative of RK-2; ApR; TcR; contains nptII promoter of pUC18-2

Invitrogen

(Hoiseth and Stocker, 1981) This work This work This work This work This work This work This work This work This work This work This work This work This work This work This work This work (Datsenko and Wanner, 2000) (Valdivia and Falkow, 1996)

(Valdivia and Falkow, 1996) (Datsenko and Wanner, 2000) (Datsenko and Wanner, 2000) (Dombrecht et al., 2001)

K. Hermans et al. / Journal of Microbiological Methods 84 (2011) 467–478

469

Table 2 Primers used in this study. Primer

Sequence (from 5′ to 3′)

Purposea

PRO 4 PRO 0406 PRO0404 PRO0560 PRO 0703 PRO 0704 PRO 0752 PRO 0753 PRO 0987 PRO 0988 PRO 1382 PRO 1383 PRO 1814 PRO 1815 PRO1900 PRO1901 PRO2356 PRO2357 PRO2358 PRO2359 PRO2361 PRO2794 PRO 2365 PRO 2366 PRO 2367 PRO 2368 PRO2622 PRO2623 PRO2624 PRO2625 PRO3366 PRO3367 PRO3380 PRO3381 PRO3890 PRO3891 PRO4349 PRO4350 PRO4353 PRO4354 PRO4355 PRO4356

GTGCCACCTGACGTCTAAGAAACC CATATGTATATCTCCTTCTTAAATCTAG GGAGGATCCAGATCATAATTTGTCG GGAGGATCCAGCTTCTTATCCGCTTCCATCATATCC TCCCTTCCGATAATCAGGCAGTCG CAAACGGTGGAATGCCCGATAGC CATGTTTTAAACCACGCCTAATGGGTTCATTTGTTAACGGATTTCAGAAGGAAAGCGATGGTGTAGGCTGGAGCTGCTTC AACGATACAAACGGTGGAATGCCCGATAGCGCACGCTTATCAGGCCATACCGTTAATAAACATATGAATATCCTCCTTA AGGGGATCAAGATCTGATCAAGAGACAGG ATCGGGGGATCAAGCCCGCCTAATGAGCG TGGACGGCTAAACTGGTCGTACCACAATTAGCGTATTAACGGAGAGCACGGTGTAGGCTGGAGCTGCTTC ATGAGAAAAGGAAGAATAAATAACCCGCCTGGCGACGGGTTCTTTTTGAGCATATGAATATCCTCCTTA TGAAAAGGTATGCACATTTCAGGCATGTTTTAA AAGCTTAATGATGCCGCGACCCGCGCGGCAACGCAT CGCATTGGCGCTTTTGA GCTCCCCCGGCGATT GCAAAGGCTATATTCGATGATTAATTAACCACATTGTTGCGAGGGATACTGTGTAGGCTGGAGCTGCTTC ACCGTTGCGATACGTCACCGTGACTTGATCAACGGTAATCGCAGATTGACTCATATGAATATCCTCCTTA GTAAACTGTCTCTCGTTGAATCGCGACAGAAAAGATTTTGGGAGCAAGCGGTGTAGGCTGGAGCTGCTTC TCATGTGGGGGAAGACAGGGATGGTGTCTATGGCAAGGAAAACAGGGTTGTCATATGAATATCCTCCTTA AGCAAAGGCTATATTCGATGATTAA AGGAAATCCTTCCAGCGCCATAAGACTACTCTATATTATGA ACGCGGAAGAAGACGCGCAAAAAAC CCATCAGAGCACCCGCAACCAGACC ACCCGCGTGAACTGGTACGTGATGG TCGGCATCGTCGATTTCAAAAGCGC TCGGGTACAGTAGCCTGATCGTACTTTCATCCTGGTATCGAGTTTCCTGCGTGTAGGCTGGAGCTGCTTC CGGGAGTAGCAGAAAGGCTAATATGACAAATATCGTCTGTACATCCATGATCATATGAATATCCTCCTTA AATAATAATTATCGTTTACGTTATCATTCACTTTCATCAGAGATATACCAGTGTAGGCTGGAGCTGCTTC CGGGCACGGCAAGCCGTGCCCAAAAAGAAATTAGAAACGGAAGGTTGCCGTCATATGAATATCCTCCTTA ACGCTACTGAAGACCAGGAACAC GCATTCGCCACGCAGAATA GTCAGCACGTTCCGGTATTTG GCGAATTCACCCAGTTTGTGA CGGCTGACGCCAAAAATAA TCAGGGCGCAGCAGATAAT TCGACCAGGCAGGGAATTATA CTGGCATCGTTGGCATTG CGACGCTCCATACGACCAAT CGGCGAAGCGTTAGAAAAAC CCGACCGAACTGTTTGATGA CATATTCGTTACCCTGCTTGCA

FW, sequencing pFPV25 RV, sequencing pFPV25 FW, amplification csgD and csgB promoter region RV, amplification csgD and csgB promoter region FW, amplification of potF RV, amplification of potF FW, D&W construction CMPG5521 and 5522 RV, D&W construction CMPG5521 and 5522 FW, sequencing pFAJ1708 RV, sequencing pFAJ1708 FW, D&W construction CMPG5537 RV, D&W construction CMPG5537 FW, amplification of potFGHI RV, amplification of potFGHI FW, qRT-PCR STM1851 RV, qRT-PCR STM1851 FW, D&W construction CMPG5584 and 5588 RV, D&W construction CMPG5584 and 5588 FW, D&W construction CMPG5589 RV, D&W construction CMPG5589 FW, amplification of sitABCD RV, amplification of sitABCD FW, amplification potF promoter region RV, amplification potF promoter region FW, amplification STM1851 promoter region RV, amplification STM1851 promoter region FW, D&W construction CMPG10301 RV, D&W construction CMPG10301 FW, D&W construction CMPG10305 and 10306 RV, D&W construction CMPG10305 and 10306 FW, qRT-PCR csgD RV, qRT-PCR csgD FW, qRT-PCR rpsS RV, qRT-PCR rpsS FW, qRT-PCR potF RV, qRT-PCR potF FW, qRT-PCR csgB RV, qRT-PCR csgB FW, qRT-PCR gyrB RV, qRT-PCR gyrB FW, qRT-PCR purA RV, qRT-PCR purA

a

FW: forward primer; RV: reverse primer, D&W: Datsenko and Wanner protocol (Datsenko and Wanner, 2000).

Wanner, 2000) starting from plasmid pKD4 and using plasmids pKD46 and pCP20. These mutants were created by specifically deleting the entire coding sequences of the respective genes. To minimize polar effects the antibiotic resistance cassettes were removed using the pCP20 helper plasmid encoding an FLP recombinase. Operon disruptions were obtained without removing the antibiotic resistance cassettes, containing a transcription termination signal, from the first gene of the respective operons. All strains and constructs were verified by PCR and sequencing analysis. 2.2. Biofilm assays Two different biofilm assays were used: biofilms were formed using the peg system and at the bottom of petri dishes. Biofilm pegassay experiments were performed as previously described by De Keersmaecker, et al. (2005). Briefly, the device used for biofilm formation is a platform carrying 96 polystyrene pegs (Nunc no. 445497) that fits as a microtiter plate lid with a peg hanging into each microtiter plate well (Nunc no. 269787). For biofilm formation, an overnight culture of S. Typhimurium wild-type or mutant was diluted 1:100 into TSB 1/20 broth, and 200 μl (approximately 107 cells/ml) was added to each well of the microtiter plate. The pegged lid was placed onto the microtiter plate and the plate was incubated for 48 h at 16 °C or 25 °C without shaking. The biofilms developed on the

surface of the pegs, and after 24 h the lid was transferred into a new plate with a fresh medium. The optical density at 595 nm (OD595) was measured for the planktonic cells in the first plate using a Synergy MX microtiter plate reader (Biotek Instruments, Inc) to get an idea of the growth characteristics of the tested strains/mutants. For quantification of the amount of biofilm formed, the pegs were washed once in 200 μl phosphate buffered saline solution (PBS). The remaining attached bacteria were stained for 30 min with 200 μl ‘crystal violet’ staining solution (0.1% (wt/vol) in an isopropanol–methanol–PBS solution (1:1:18 [vol/vol])). Excess stain was rinsed off by placing the pegs in a 96-well plate filled with 200 μl distilled water per well. After the pegs were air dried (30 min), the dye bound to the adherent cells was extracted with 30% glacial acetic acid (200 μl). The OD570 of each well was measured using the Synergy MX microtiter plate reader and the results are an indirect measure for the biofilm forming capacity of the tested strains/mutants. Biofilm formation on the bottom of small polystyrene petri dishes (60 mm diameter, Greiner Bio-One) was performed by adding 10 ml of a 1:100 dilution of the particular S. Typhimurium strain into TSB 1/20 broth, with addition of Ap when appropriate. After 48 h stationary incubation at 25 °C, the bacteria formed a biofilm layer at the bottom. This layer was scraped off, passed through a 25 gauge syringe and vortexed to break up bacterial clumps when used in the DFI workflow because cell aggregates could cause false results (Rieseberg et al., 2001).

470

K. Hermans et al. / Journal of Microbiological Methods 84 (2011) 467–478

2.3. S. Typhimurium SL1344 promoter-probe library construction As previously reported by Valdivia and Falkow (1996), S. Typhimurium SL1344 genomic DNA was partially digested with Sau3AI and size-fractionated by agarose gel electrophoresis. DNA fragments of 0.4–1.6 kb were recovered and inserted into the BamHI site upstream of the promoterless gfpmut3 gene in pFPV25. gfpmut3 encodes a highly stable GFP variant (up to 24 h (Cormack et al., 1996)). The plasmid library was electroporated into S. Typhimurium SL1344, resulting in a pooled library of around 20,500 clones, subdivided into random pools of approximately 500 clones each. 2.4. Flow-cytometric analysis Bacterial cell suspensions were analyzed using a FACSCalibur (Becton Dickinson). Instrument settings were empirically optimized to our needs, whereby bacteria harboring pFPV25 and pFPV25.1 were used as negative and positive control, respectively. Fluorescence, side and forward scatter data were collected for 104 cells using logarithmic amplifiers. Bacteria were distinguished from other particles (cell fragments and culture debris) by applying forward scatter (FSC) as primary threshold (value 254) and fluorescence (FL1) as secondary threshold. Prior to each analysis, the light scatter and fluorescence parameters were calibrated with CaliBRITE 3 (Becton Dickinson), according to the manufacturer's recommendations. 2.5. DFI enrichment of biofilm-specific promoters To identify biofilm-induced S. Typhimurium SL1344 promoters, constructed library pools (approximately 500 clones per sorting round) were cultured overnight and subjected to biofilm-inducing growth conditions using small polystyrene petri dishes at 25 °C. After 48 h of induction, the culture was analyzed by fluorescence-activated cell sorting (FACS) and the bacterial population that exhibited fluorescence intensity exceeding that of an identically treated SL1344/pFPV25 (negative control) culture was sorted (at single cell mode (Bumann, 2002)), by gating the corresponding population at the computer terminal. 104 clones were sorted onto a 0.22 μm membrane filter (Millipore) and grown overnight in fresh LB broth, supplemented with Ap and Sm. This first-passage sublibrary was 1:100 diluted in 5 ml TSB 1/20 medium, supplemented with Ap. After overnight growth (TSB 1/20, 25 °C, aeration), planktonic, nonfluorescent bacteria were sorted (104 clones) and grown as described above. This non-fluorescent subpopulation was again subjected to biofilm-inducing conditions and fluorescent bacteria were collected as described above. This protocol was repeated until a total of 5 sorts was obtained (3 positive and 2 negative). After the last positive sort, the enriched pool was cultured overnight and plated out on LB agar plates with Ap and Sm to obtain single colonies. These single colonies were transferred to 96-well plates, cultured overnight and individually profiled as further described. 2.6. Individual profiling and sequence determination of potentially biofilm-influenced promoter fusions Biofilms of the picked clones were formed using the peg-system at 25 °C and fluorescence induction was assessed using FACS analysis, after sonicating the pegs in 200 μl FACSFlow™ Sheath Fluid (Becton Dickinson) for 20 min. Fluorescence profiles were also obtained for the 96 clones grown under planktonic conditions (TSB 1/20 broth, 25 °C, aeration) after similar sonication. Graphical overlays were made of biofilm-inducing versus planktonic conditions to determine the fluorescence distribution of the clones. The instrument settings as applied in the sorting steps were maintained. Quantitative measurements and overlays were made with the CELLQuest™ Pro software program (version 4.0.2.; Becton Dickinson). Subsequently, the DNA

sequences of interesting promoter-trap library clones (i.e., the ones that gave differential expression) were determined under BigDye terminator cycling conditions (Applied Biosystems) using primers PRO-4 and PRO-0406, on a 3730xl DNA analyzer (Applied Biosystems). Sequencing reactions were performed by Macrogen (Korea). These sequences were compared with the available, complete genome sequence of S. Typhimurium LT2 (McClelland et al., 2001), showing high similarity with the S. Typhimurium SL1344 sequence, by using the BLASTn algorithm (Altschul et al., 1990). 2.7. Investigation of bacterial growth Growth curves of wild-type and mutant strains were recorded using a Bioscreen C system (Oy Growth Curves Ab Ltd). Overnight cultures of the strains were 1:1000 diluted into LB and TSB 1/20 broth into three separate wells of a 100 well honeycomb plate (three biological repeats) and grown at 25 °C for 48 h. The experiment was performed under continuous shaking conditions and the optical density (OD595), reflecting the bacterial growth, was measured every 15 min. 2.8. qRT-PCR analysis For RNA isolation, the strains were grown in planktonic conditions for 24 h in TSB 1/20 with aeration. Approximately 5 x 108 cells/ml were added to a 1/5 volume ice-cold phenol:ethanol mixture (5:95) and transferred to a microcentrifuge tube which was immediately frozen in liquid nitrogen and stored at − 80 °C. For RNA isolation of biofilm cells, bacteria were grown in biofilms at the bottom of petri dishes for 24 h as described above. After the medium was removed, cells from the biofilm were scraped from the plate in a mixture of 1 ml TSB 1/20 and 200 μl ice-cold phenol:ethanol (5:95), transferred to a microcentrifuge tube, quick-frozen in liquid nitrogen and stored at −80 °C. Total RNA was isolated from the cells using the RNeasy Mini Kit (Qiagen), according to the manufacturer's instructions, and subsequent DNase treatment was performed using the TURBO DNA-free Kit (Ambion) according to the manufacturer's instructions. DNA contamination of the RNA samples was checked by PCR. 100 ng of Dnasetreated RNA was reverse transcribed using the RevertAid H Minus First strand cDNA Synthesis Kit (Fermentas), according to manufacturer's recommendations. After dilution of cDNA, 5 μl of cDNA (2 ng/μl), 0.9 μl of each specific primer (20 μM) and 3.2 μl of RT-qPCR grade water (Ambion) were mixed with 10 μl of Power SYBR Green PCR Master Mix (Applied Biosystems) and qRT-PCR reactions were performed on a StepOne-Plus System (Applied Biosystems). In order to confirm that there was no background contamination, a negative control was included in each run. For each target gene, PCR efficiency was determined. Melt curve analysis was performed to ensure that a single gene product was amplified. RT-PCR primers, listed in Table 2, were designed using Primer Express 3.0 (Applied Biosystems) and purchased from Integrated DNA Technologies, Inc (Leuven, Belgium). gyrB, purA and rpsS showed an almost invariant expression between the tested conditions and were used as endogeneous controls to normalize the target gene's expression. Each reaction was at least performed in triplicate and data were analyzed using the StepOne™ (version 2.1) and DataAssist™ (version 2.0) software from Applied Biosystems. 3. Results 3.1. CsgD is important in S. Typhimurium SL1344 biofilm formation In order to validate the DFI technique for its use under biofilm conditions, we needed a ‘control’ gene which is known to be involved in Salmonella biofilm formation and which could be used in a spiking experiment (Section 3.2.). As already shown in the enterobacteria

K. Hermans et al. / Journal of Microbiological Methods 84 (2011) 467–478

E. coli (Prigent-Combaret et al., 2001) and S. Typhimurium ATCC14028 (Romling, 2005), a csgD mutant is impaired in multicellular behaviour. CsgD can be seen as a master regulator that triggers a whole cascade of gene expressions related to biofilm formation (Grantcharova et al., 2010; Romling, 2005; Romling, et al., 1998). We first investigated whether csgD is also involved in S. Typhimurium SL1344 biofilm formation. SL1344 is the routinely used Salmonella strain in our laboratory. This strain has not yet been investigated extensively for its biofilm formation capacity, although there are some specific differences considering biofilm formation between this

471

S. Typhimurium strain and the S. Typhimurium ATCC14028 strain (Garcia et al., 2004). We constructed a SL1344 ΔcsgD mutant (CMPG5579) and tested its biofilm forming capacity. Site-specific deletion of csgD resulted in an 80–90% reduction of biofilm formation capacity. To test the expression of csgD in SL1344 under biofilm conditions using the DFI technique, the entire promoter region of csgD (i.e., total intergenic region between csgB and csgD (Gerstel et al., 2003; Gerstel and Romling, 2003)) was cloned in front of the promoterless gfpmut3 to yield pCMPG5521. S. Typhimurium SL1344 harboring this plasmid showed fluorescence upregulation under biofilm-inducing

Fig. 1. Flow-cytometric analysis of the biofilm induction of selected clones and constructs. Histograms (counts vs. fluorescence (FL1-H)) showing visual overlays of planktonic FACS profiles (grey shaded) with biofilm-induced FACS profiles (black lines). The curves represent populations of 104 cells. (A) SL1344 containing pCMPG5521 (csgD promoter), (B) Clone 0–2_10 (potF promoter), (C) Clone 2–2.5_47 (STM1851 promoter), (D) Skewed profile of clone 0–0.5_42 (sequence in kefC), (E) Skewed profile of clone 17–17.5_3 (region in STM4012 and STM4013), (F) SL1344 containing pCMPG5532 (potF promoter), (G) SL1344 containing pCMPG5533 (STM1851 promoter), and (H) SL1344 containing pCMPG5539 (csgB promoter).

472

K. Hermans et al. / Journal of Microbiological Methods 84 (2011) 467–478

conditions (Fig. 1A). Hence, the SL1344 csgD promoter can be seen as an example of a SL1344 biofilm-induced promoter. The profile of csgD expression illustrates what a FACS profile of a clear biofilm-induced gene looks like: csgD is already to some extent expressed in planktonic conditions (as compared to the negative control), but there is a drastic upregulation of GFPmut3 expression under control of the csgD promoter in biofilm-inducing conditions.

3.2. Proof of concept: an optimized DFI workflow is able to identify S. Typhimurium SL1344 biofilm upregulated genes The DFI technique was applied to perform a genome-wide screening to identify S. Typhimurium SL1344 promoters specifically induced under biofilm conditions by using the SL1344 promoter-trap library, as specified in Materials and methods. To prove that this strategy indeed succeeded in isolating Salmonella biofilm-induced promoters, we first performed a ‘spiking experiment’ with the promoter fusion construct of csgD (pCMPG5521), a gene shown to be important in S. Typhimurium SL1344 biofilm formation (Section 3.1.). SL1344 containing pCMPG5521 was spiked into a random library pool of the promoter trap library. This pool was grown under biofilm conditions and the most fluorescent bacteria of the pool's total population (0.3 to 5%) were isolated by cell sorting. This sorted sublibrary was expected to contain specifically biofilm-induced promoters as well as constitutive promoter fusions. Most of the constitutive promoters were discarded by FACS after a planktonic, ‘negative’ selection round, where the least fluorescent bacteria in this sublibrary, represented by approximately the lowest 10% of the total fluorescent population, were collected and cultured. In the next biofilm-inducing, positive sort, the pool of biofilm-induced promoter fusions was enriched. Again, the most fluorescent bacteria were sorted. Subsequent negative and positive sorts allowed further enrichment, as shown in Fig. 2A. This enrichment was further confirmed using DNA restriction analysis. Plasmid DNA isolated before the first and after the last sort was digested with appropriate enzymes to cut out the DNA inserts. As the sorting progressed, the complexity of DNA inserts drastically decreases, as visualized on agarose gel (Fig. 2B). Valdivia and Falkow previously showed that DNA inserts with a stimulus-dependent promoter activity represented a third to half of the total population of DNA inserts present in the DFI-enriched pools (Valdivia and Falkow, 1996). The remaining stimulus-independent fluorescent clones probably represented constitutive promoter fusions that could not be separated from stimulus-

inducible fusions with the FACS collection parameters used. These promoters, however, were discarded through downstream analysis. Indeed, after the final enrichment round, single colonies were collected and individually profiled for their biofilm-inducible GFP expression. To this end, each isolated colony was grown under planktonic and biofilm conditions, after which the corresponding GFP expression profiles were compared. According to these profiles (48 individually profiled clones out of an initial pool of 500), there was a good degree of profile-reciprocity, meaning that most profiles were retrieved more than once after the enrichment (data not shown). This shows that the chosen number of profiled clones for each enriched pool was a good ‘coverage estimation’ to get a clear view of the clones in the enriched sublibrary. Of clones containing promoter fusions showing differential expression between both conditions, plasmid DNA was isolated and the determined sequences of the corresponding inserted DNA fragments were analyzed with the BLASTn algorithm (Altschul et al., 1990). We were able to isolate the specific csgD FACS profile (Fig. 1A), as well as its sequence out of this spiked pool. This illustrates that the optimized DFI workflow was well designed to identify biofilm-induced genes in subsequent experiments.

3.3. Isolation of biofilm-induced promoters by DFI All random pools of the SL1344 promoter-probe library were grown under biofilm-inducing conditions and subjected to the optimized DFI parameters to identify specific biofilm upregulated promoter fragments, as elaborated in Section 3.2. Subsequent analysis, as outlined in Materials and methods Section 2.6., resulted in a set of biofilm upregulated promoters (and their corresponding genes) (Table 3). Inductions of fluorescence under biofilm conditions ranged from 1.3 to 13.1, with one clone (14–14.5_43) showing no real induction of fluorescence as measured by dividing the mean fluorescence value of the respective clone under biofilm-induced conditions by its corresponding value under planktonic conditions. This clone, however, showed a biofilm profile with a sharper peak in the fluorescence-count histogram with just a little variation around the mean, while under planktonic conditions there was a more flattened peak with more variation accounting for a lower than one value in Table 3. Similar phenomena have been encountered in other studies using DFI as well (Barker et al., 1998; Marra et al., 2002). Additionally, some profiles did not show a fold-induction of GFPmut3 expression in the main population. However, under biofilm-inducing conditions, only part of the population had an increased GFPmut3

Fig. 2. Visualization of the DFI enrichment. (A). FACS profile (counts vs. fluorescence intensity (FL-1 H)) of the enrichment of biofilm-induced clones by DFI: a, first biofilm-inducing sort (grey line); b, third biofilm-inducing sort (black line); c, negative control (SL1344/pFPV25, promoterless gfpmut3) (dotted line); and d, positive control (SL1344/pFPV25.1, constitutive promoter upstream of gfpmut3) (grey line). (B). DNA-agarose gel showing the DFI enrichment towards biofilm-induced promoters as the screening progressed: lane 1, Smartladder (Eurogentec) with 400 bp and 1600 bp indications; lane 2, EcoRI/XbaI digest of pool 0.5–1, prior to enrichment; lane 3, EcoRI/XbaI digest of pool 0.5–1, after the total enrichment process (3 positive alternated by 2 negative sortings); lanes 4 and 5, same as lanes 2 and 4, respectively, but for the pool 1.5–2. The band around 4800 bp represents the EcoRI/XbaI digested form of the pFPV25 plasmid. The arrows clearly indicate the enriched, biofilm upregulated fragments.

K. Hermans et al. / Journal of Microbiological Methods 84 (2011) 467–478

473

Table 3 S. Typhimurium promoters showing upregulation in biofilm conditions. Clonea

Sequence reading into gfpmut3b

Functionc

Profiled

0–0.5_41

sequence in kefC

skewed biofilm expression

4.5–5_48

fhuA promoter

12–16_44 4–4.5_45

STM0275 promoter fes promoter

1–1.5_47 0–2_10

ybeL promoter potF promoter

8–12_4 12–16_11 0.5–1_3

1.5–2_47

csgB promoter STM1255 promoter intergenic sequence between STM1606 and STM1607 in orientation of STM1607 sequence in sohB intergenic sequence between tdk and hns in orientation of tdk narK promoter intergenic sequence between STM1839 and yobF in orientation of yobF STM1851 promoter intergenic sequence between phsA and sopA in orientation of sopA sanA promoter

15.5–16 _9

sequence in STM2532

KefC (STM0086): glutathione-regulated potassium-efflux system protein FhuA (STM0191): outer membrane protein receptor/transporter for ferrichrome, colicin M, and phages T1, T5, and φ80 STM0275: putative cytoplasmic protein Fes (STM0586): enterobactin/ferric enterobactin esterase YbeL (STM0653): hypothetical protein PotF (STM0877): putrescine transporter subunit, periplasmic-binding component of ABC superfamily CsgB (STM1143): curlin minor subunit STM1255: putative ABC transporter periplasmic binding protein STM1606: putative benzoate membrane transport protein, STM1607: putative outer membrane lipoprotein SohB (STM1716): putative periplasmic protease H-NS (STM1751): global DNA-binding transcriptional dual regulator H-NS, Tdk (STM1750): thymidine kinase NarK (STM1765): nitrite extrusion protein YobF (STM1838): putative cytoplasmic protein, STM1839: hypothetical protein STM1851: putative cytoplasmic protein PhsA (STM2065): thiosulfate reductase precursor, SopA (STM2066): secreted effector protein SanA (STM2184): hypothetical protein (similar to E. coli vancomycin sensitivity gene) STM2532: putative inner membrane lipoprotein

3–3.5_45 0–4_3 0–0.5 _47 4–8_6 4–8_18 20–20.5_47 17–17.5_3

iroN promoter nrdH promoter sitA promoter STM3071 promoter nirB promoter cpxP promoter sequence in STM4012 and 4013 in orientation of STM4013

0.5–1_26

sequence in yjdB

15–15.5_45

STM4423 promoter

14–14.5_43 6.5–7_42 0–4_36 15.5–16_27 2–2.5_47 5.5–6_48

IroN (STM2777): TonB-dependent siderophore receptor protein NrdH (STM2805): glutaredoxin-like protein SitA (STM2861): Fur regulated Salmonella iron transporter STM3071: putative DNA-binding protein NirB (STM3474): nitrite reductase large subunit CpxP (STM4060): periplasmic repressor STM4012: putative coproporphyrinogen III oxidase and related FeS oxidoreductase, STM4013: putative membrane-associated metal-dependent hydrolase YjdB (STM4293): putative cell division protein (hypothetical 61.6 kDa protein in basS/pmrA-adiY intergenic region) STM4423: putative DNA-binding protein

2.1-fold induction

2.5-fold induction 1.4-fold induction 1.4-fold induction 1.4-fold induction 13.1-fold fold induction 2.8-fold induction skewed biofilm expression b 1-fold induction 1.7-fold induction 2.8-fold induction 2.7-fold induction 7.9-fold induction skewed biofilm expression skewed biofilm expression 3.5-fold induction with skewed biofilm expression skewed biofilm expression 1.6-fold induction 6.1-fold induction 3.3-fold induction 2.7-fold induction skewed biofilm expression skewed biofilm expression

skewed biofilm expression 1.3-fold induction

a

Numbers refer to the position of the respective clone in microplates as retrieved during the experimental workflow. b ‘Intergenic region’ refers to the intergenic region in between the relevant genes in the fully annotated S. Typhiumurium LT2 genome. c The function of the genes was derived from GenBank (Benson et al., 2002). d Skewed biofilm expression reflects the situation in which a part of the clonal population showed an upregulated GFPmut3 expression under biofilm conditions as compared to planktonic conditions, often observed as a slight upregulation in the right tail of the histogram (counts vs. FL1-H) under biofilm conditions. Fold induction represents the mean fluorescence value of biofilm-induced bacteria divided by the mean fluorescence value of the same clone under planktonic conditions. Values below 1 and values just exceeding 1 represent profiles of which the biofilm-induced form is not really shifted to a more fluorescent value, but of which a higher number of cells expresses the same fluorescence level as compared to the planktonic form.

expression and exhibited a skewed expression profile (tail) (Fig. 1D–E). This could point to a certain degree of bistable expression of the corresponding genes under these conditions. Of all 26 retrieved clones, 17 coincided with promoter regions of annotated genes (McClelland et al., 2001), whereas 9 mapped to regions with putative unknown regulatory elements in intergenic regions as well as inside annotated genes (Table 3). Five of the 17 clones with fragments mapping into promoter regions of already annotated genes, corresponded to promoters of putative genes with unknown functions (STM0275, STM1255, STM1851, STM3071 and STM4423), 2 to hypothetical proteins (ybeL (STM0653), sanA (STM2184)) and 10 to well-annotated genes being fhuA (STM0191), fes (STM0586), potF (STM0877), csgB (STM1143), narK (STM1765), iroN (STM2777), nrdH (STM2805), sitA (STM2861), nirB (STM3474) and cpxP (STM4060). According to the fully annotated genome (McClelland et al., 2001), these genes are involved in the following processes: polyamine transport (potF), iron (sitA) and siderophore transport (iroN, fhuA), nitrogen metabolism (narK, nirB), curli biosynthesis (csgB), oxidative stress responses (nrdH), enterochelin biosynthesis processes (fes) and the Cpx envelope stress response pathway (cpxP).

3.4. Confirmation of DFI screen results Because the results of our screen are based on a randomly constructed DNA library, we subsequently decided to make defined promoter fusions of some of the obtained results. The rationale behind this is that, although sequence analysis pointed at the fragments listed in Table 3 as being responsible for driving GFPmut3 expression, it cannot be excluded that the fluorescence expression could be the result of unwanted DNA scrambling occurring during the promoterprobe library construction. In addition to csgD confirmation (Section 3.1., Fig. 1A), three other genes were chosen for these initial confirmation studies: csgB, as already being documented to be important in Salmonella biofilms (Barnhart and Chapman, 2006; Romling, 2005); potF, as a link between polyamine metabolism and biofilm formation has been shown in related organisms (Shah and Swiatlo, 2008) and STM1851, as not yet been functionally identified. In clone 0–2_10 (Fig. 1B), GFPmut3 expression is driven by a DNA fragment containing the potF promoter (Table 3). PotF is the periplasmic binding protein of an ABC transporter encoded by the potFGHI operon, responsible for the uptake of the polyamine putrescine (Igarashi and Kashiwagi, 1999; Pistocchi et al., 1993).

474

K. Hermans et al. / Journal of Microbiological Methods 84 (2011) 467–478

Polyamines have been shown to have pleiotropic roles in bacterial pathogens, as reviewed by Shah and Swiatlo (2008), with direct and indirect effects on biofilm formation in for example Vibrio cholerae (Karatan et al., 2005), Yersinia pestis (Patel et al., 2006) and Pseudomonas putida (Sauer and Camper, 2001). Plasmid pCMPG5532, containing the potF promoter region in front of the gfpmut3 of pFPV25, was constructed. The expression profiles clearly showed the upregulation of the defined potF promoter fusion under biofilminducing conditions, thereby confirming our DFI results (Fig. 1F). Clone 2–2.5_47 showed an approximately 8-fold upregulation of fluorescence in biofilm growth conditions (Fig. 1C). The sequence driving GFPmut3 expression corresponds to the STM1851 promoter (Table 3). STM1851 encodes a small putative cytoplasmic protein of 79 amino acid residues with 88% sequence identity to the E. coli hypothetical protein YebV. The plasmid pCMPG5533, carrying the STM1851 promoter region upstream of the gfpmut3 of pFPV25, gave a similar expression pattern as the one from the random clone 2– 2.5_47, as can be seen from Fig. 1G, confirming our DFI results and verifying it is indeed the STM1851 promoter region responsible for the fluorescence expression. Similar phenomena were observed for clone 8–12_4 and the defined csgB promoter construct (pCMPG5539). Indeed, clone 8–12_4, in which the csgB promoter region drives the GFPmut3 expression, showed an approximately 13-fold induction of fluorescence under biofilm growth conditions (Table 3). CsgB functions as a nucleator in the assembly of curli (coiled surface structures) on the cell surface. Plasmid pCMPG5539 (Fig. 1H), carrying the defined csgB promoter region upstream of the gfpmut3 of pFPV25, gave similar expression patterns as the ones from the random clone 8–12_4, confirming the involvement of this gene during biofilm formation. 3.5. qRT-PCR analysis of a selection of DFI-retrieved genes qRT-PCR was used to investigate whether some selected biofilm upregulated genomic inserts in the DFI screen showed also differential expression at the RNA level and hence would have been isolated using traditional transcriptomic approaches. For this analysis, RNA was isolated from S. Typhimurium SL1344 planktonic and biofilm cells respectively, as described in Materials and methods Section 2.8., and the results are shown in Fig. 3. Expression levels were checked for the same loci as in the initial confirmation studies: csgD, csgB, potF and STM1851. In all cases, except for STM1851, there was a higher level of

Fig. 3. qRT-PCR analysis of a selection of DFI-retrieved genes. The transcription of a selection of DFI-retrieved genes was examined by qRT-PCR analysis of RNA extracted from planktonic and biofilm cells, as elaborated in Experimental procedures. Values represent mean fold expression normalized to the endogeneous control genes gyrB, purA and rpsS. These genes were shown to be expressed at comparable levels between the tested conditions for S. Typhimurium SL1344. The normalized level of expression of the genes of interest in the planktonic samples was set at 1 and is shown in grey, the respective biofilm samples are shown in black. Error bars represent the standard deviation of at least three independent reactions.

expression in the biofilm sample as compared to the planktonic sample (which was chosen as the reference sample). As such, this analysis shows that the DFI results and the qRT-PCR results in terms of differential expression are similar for csgD, csgB and potF, but different for STM1851. This implicates that csgD, csgB and potF by using the appropriate cut-off value could have been picked up using a traditional transcriptomic approach. STM1851, on the other hand would not have been picked up for the selected timepoint. This illustrates the fact that differences in terms of expression levels for a given gene not necessarily match in differential screens and hence that differential screens can be seen as complementary approaches. 3.6. Mutational analysis of DFI identified genes As this is the first use of the DFI technique in relation to biofilms, some DFI isolated genes were further chosen for an initial follow-up mutational study based on diverse criteria. The curli structural subunit encoded csgB gene was chosen because it has been documented, as part of the curli operon, to be important in Salmonella biofilms (Barnhart and Chapman, 2006; Romling, 2005). Most of the work on Salmonella curli, however, was done with strains different from strain SL1344 used in this study. Recently Hamilton and colleagues also showed upregulation of this gene in mature SL1344 biofilms, but mutational analysis in SL1344 has not yet been reported (Hamilton, et al., 2009). The polyamine uptake system encoded potFGHI (Shah and Swiatlo, 2008) as well as genes involved in iron metabolism (iron (sitABCD) and siderophore (fhuACDB) transport) (White et al., 2010; Yang et al., 2007) and the Cpx regulon gene cpxP (Beloin, et al., 2004) were chosen because these processes have been linked to biofilm formation in related organisms. An additional inclusion criterium for having selected cpxP is the fact that its promoter showed a skewed expression profile and not a fold induction profile as is the case for potFGHI, sitABCD and fhuACDB in the DFI isolation. STM1851 was chosen because it showed a very high level of upregulation under biofilm growth, as can be seen from Table 3 and Fig. 1C and G, and because it is a gene without known function so far. sanA, finally, was chosen because nothing is known about the function of this hypothetical gene in Salmonella and because it showed a skewed expression profile. Firstly, S. Typhimurium SL1344 potF and potFGHI mutants were constructed to investigate the possible roles of the gene and/or operon in the biofilm process (Fig. 4). Deletion of potF in SL1344 (CMPG5522) background resulted only in a small but reproducible reduction in

Fig. 4. Biofilm formation by potF mutants. The level of biofilm formation is expressed as a percentage of wild-type SL1344 biofilm formation. The data are representative for three independent repetitions, and the error bars show standard deviations of at least three measurements. CMPG5522: ΔpotF mutant; pCMPG5522/CMPG5522: complemented ΔpotF mutant; CMPG5521: potF::Cm mutant; pCMPG5531/CMPG5521: complemented potF::Cm mutant; pCMPG5531/SL1344: SL1344 wild-type strain harboring pCMPG5531.

K. Hermans et al. / Journal of Microbiological Methods 84 (2011) 467–478

biofilm formation at 16 °C (around 20%), while no significant reduction could be noticed at 25 °C (data not shown). This reduction could be complemented by electroporation of pCMPG5522, containing the potF gene driven by a constitutive promoter. CMPG5521, in which the whole potFGHI operon is disrupted, showed a further reduction in biofilm formation (up to 40%) (Fig. 4), which could almost be fully complemented by introducing the complete operon on a plasmid (pCMPG5531). The same effect was seen for biofilm formation at 25 °C (data not shown). For none of these mutants any significant effect on planktonic growth was observed (data not shown), indicating that inactivation of the potFGHI locus affects a property required for Salmonella to fully develop a biofilm. More importantly, the mutational analysis of the potFGHI locus allowed a direct phenotypic validation of results obtained with the DFI screening. To make sure the complementation plasmid backbone, the RK2 based pFAJ1708, did not itself impact on biofilm formation, as was shown for some IncP plasmids (Ghigo, 2001), we showed in a control experiment that an SL1344 wild-type strain harboring pFAJ1708 does not show any significant changes in biofilm formation capacity. Secondly, we constructed a disruption mutant (CMPG5584) of the sitABCD encoding iron transport system, another identified ABC transporter (Table 3). This mutant showed a 90% reduced biofilm formation capacity, and this could be complemented by introducing the complete operon on a plasmid (pCMPG5538) (Fig. 5). Thirdly, an fhuACDB disruption mutant (CMPG10305) also showed a drastic reduction in biofilm growth (Fig. 5). No significant growth defects were encountered for these iron metabolism related mutants (CMPG5584 and CMPG10305) as shown by Bioscreen analysis (data not shown). Links between iron metabolism and biofilm formation have been documented in other bacteria (e.g. (Banin et al., 2005; Karatan and Watnick, 2009)), showing that this is an important environmental factor governing biofilm behaviour in bacteria. Fourthly, deletion of csgB (CMPG10309) showed the anticipated large reduction (up to 95%) in biofilm formation. Fifthly, deletion of STM1851 (CMPG5537) did not show a clear reduction in the amount of biofilm formed by this mutant as compared to the wild-type S. Typhimurium SL1344 strain (Fig. 5). However, because the defined STM1851 promoter plasmid pCMPG5533 showed a clear upregulation under biofilm growth conditions, it is possible that this gene is not strictly necessary for biofilm formation but is being upregulated as a consequence of adaptation to the biofilm life style and

Fig. 5. Biofilm formation by some mutants in identified genes. The level of biofilm formation is expressed as a percentage of wild-type SL1344 biofilm formation. The data are representative for three independent repetitions, and the error bars show standard deviations of at least three measurements. CMPG5537: ΔSTM1851 mutant; CMPG5584: sitA::Cm mutant; pCMPG5538/CMPG5584: complemented sitA::Cm mutant; pCMPG5538/SL1344: SL1344 wild-type strain harboring pCMPG5538; CMPG5589: ΔcpxP mutant; CMPG10301: ΔsanA mutant; CMPG10305: fhuA::Cm mutant; CMPG10309: ΔcsgB mutant.

475

growth. Similar effects were obtained for deletions of cpxP (CMPG5589) and sanA (CMPG10301), as can be seen on Fig. 5. In summary, we can group the tested DFI genes in three categories: genes shown to be upregulated under biofilm growth in our DFI screen and of which the corresponding mutants are hindered in biofilm formation in the used biofilm test system (potFGHI, sitABCD and fhuACDB, csgB); genes shown to be upregulated under biofilm growth in our DFI screen but of which the corresponding mutant did not show a reduction in biofilm formation in our test system (exemplified by mutant STM1851); and genes shown to have a skewed expression profile under biofilm growth in our DFI screen that did not show a reduction in biofilm formation in our test system (cpxP and sanA). As not all the DFI identified genes have been fully tested yet phenotypically by mutational analysis, it is likely that more categories of ‘biofilm’ genes will be recognized in the future. 4. Discussion To our knowledge, this is the first reported study exploiting the high-throughput DFI promoter-capture approach (Valdivia and Falkow, 1996) to isolate a series of promoter constructs upregulated under bacterial biofilm growth conditions. Using this approach we were able to identify promoters (and their corresponding genes) that show increased expression under S. Typhimurium biofilm conditions. Because we have been working specifically under non-host conditions, we propose the identified genetic elements to be important during biofilm growth in the transmission phase of the pathogen; more specifically, under similar conditions as the ones we applied through our screen, which are for instance expected to be encountered in some parts of food processing industries (Leriche and Carpentier, 2000). Moreover, because of our sampling point (48 h), the identified genetic determinants are thought to be predominantly involved in the maturation stage of Salmonella biofilm formation. In contrast to the screening of mutant libraries, DFI does not exclude the identification of genes that are essential for bacterial survival. Further on, it allows the identification of promoters that are upregulated during biofilm growth and that drive the expression of genes that may be important for aspects of biofilm physiology, but not necessary for biofilm formation (as shown in the Results section). Although some of the identified genes have already been documented for mechanisms known to be important in biofilm formation, such as polyamine (potF) (Shah and Swiatlo, 2008) and iron metabolism associated genes (sitA, iroN, fhuA and fes) (Yang et al., 2007), we report here for the first time their importance in the particular case of S. Typhimurium SL1344 biofilm formation. A recent metabolomic S. Typhimurium ATCC14028 biofilm study, highlighting the importance of the metabolic responses to extracellular matrix production, however, identified fhuA as biofilm-induced under the experimental conditions tested using lux promoter fusions (White, et al., 2010). Further on, only one of the identified genes, csgB, as part of the curli biosynthesis operon csgDEFG-csgBAC, has already been conclusively shown to be important during Salmonella biofilm growth (Barnhart and Chapman, 2006; Hamilton et al., 2009; Romling, 2005; White et al., 2010). Except from csgB, only the hypothetical protein encoded by ybeL, showing similarities to an E. coli putative alpha helical protein without known function, was also found as being upregulated in a recently performed S. Typhimurium SL144 biofilm microarray/proteomics study (Hamilton et al., 2009). It cannot be excluded that differences in the used biofilm formation conditions between both studies might be responsible for the limited overlap of identified genes in both studies. However, as the microarray approach and the here applied single cell DFI approach are intrinsically different, we favour the idea that such approaches have a clear added value in identifying genes that are important for a given process and therefore will help in better understanding of it. Next to the above mentioned hypothetical gene ybeL, we also found some

476

K. Hermans et al. / Journal of Microbiological Methods 84 (2011) 467–478

other genes with unknown function (FUN genes) that are upregulated during SL1344 biofilm growth (STM0275, STM1255, STM1851, STM3071 and STM4423) and even another hypothetical gene, sanA. Given the significant number of FUN genes in the S. Typhimurium genome, and the increasing number of biofilm related functional genomics studies, it is to be expected that previously uncharacterized Salmonella genes will be found to be related to social behaviour processes, as also recently illustrated by Barak et al. (2009). Remarkably, STM1255 encodes a putative ABC transporter periplasmic binding protein. Together with potF, sitA and fhuA, this makes 4 out of 17 (24%) annotated, DFI-retrieved genes that are involved in surface and transport properties. Similar importance of diverse nutrient uptake systems under Salmonella biofilm growth has been notified by White et al. (2010), further elaborating the link between biofilm formation, matrix production and metabolic changes such as starvation. Therefore, some of our current research is focusing on the relationship between these transport systems and the homeostasis of the biofilm matrix. Comprehensive validation of the generated DFI results would require phenotypic confirmation (mutant analysis, qRT-PCR analysis and/or microscopic analysis) of all identified genetic loci, but was beyond the scope of our study. Using defined promoter constructs for a set of identified promoters we confirmed unambiguously their role in SL1344 biofilm formation, thereby validating our DFI screen (Fig. 1). qRT-PCR analysis of the same panel of identified promoters further confirmed these results but also showed the complementarity between different approaches because STM1851, for example, would not have been picked up using a traditional transcriptomic approach (Fig. 3). Initial follow-up mutational analysis of some of the identified genes substantiated their role in SL1344 biofilm growth (Figs. 4 and 5). Some identified promoters, however, showed an upregulation under biofilm growth conditions, as revealed by the DFI results (Table 3), but did not show a clear reduction in the amount of biofilm formed in the corresponding mutational analysis. It is likely that these genes are important for adaptation to the biofilm physiology, rather than for the biofilm formation itself. Apart from the isolated annotated promoters, we also isolated 9 clones of which the genomic inserts are within an annotated gene and/or of which the predicted transcription of all genetic elements within this region are in the opposite direction from that of the reporter gfpmut3 gene. This reflects a major advantage of the DFI strategy to study gene expression. Indeed, it is not biased by prior knowledge of gene annotation because of the use of a random promoter-probe library and as such it holds great promise for the identification of new regulatory elements. Similar observations were made in a recent RIVET screening study to identify genetic loci of Enterococcus faecalis involved in this opportunistic pathogen's biofilm formation (Ballering et al., 2009). It is possible that some of these 9 identified clones represent artefacts inherent to the use of promoterprobe libraries to perform a genome wide screening (Rediers et al., 2005) or might encode biofilm upregulated small non-coding RNAs (sRNAs) or peptides. However, none of the identified genetic loci were previously predicted or identified as being sRNAs (Sittka et al., 2008). Alternatively, these sequences might arise from antisense transcription. Recent work by Dornenburg and colleagues showed widespread transcription of antisense RNAs (approximately 1000 antisense RNAs on 4000 mRNAs) in E. coli (Dornenburg et al., 2010). Massive antisense transcription was also found in the ε-proteobacterium Heliobacter pylori (Sharma et al., 2010). Moreover, and further stressing the importance of our results, Koide and co-workers were recently able to unambiguously show the prevalence of transcriptional promoters within operons and coding sequences, as well as environmental modulation of operon architectures, using an archaeal model system (Koide et al., 2009). Because of the relatively large number of non-annotated, intergenic or in-gene hits found in our screen, we are currently considering a further, alternative screening

strategy combining DFI enrichment with tiling microarrays. Extracted biofilm enriched (or repressed) sublibrary pools directly hybridized to S. Typhimurium tiling arrays (Santiviago et al., 2009) should be able to give a better understanding and delineation of the exact genomic regions driving GFP expression under the used environmental conditions. Furthermore, and in line with the findings by Koide et al. (Koide et al., 2009), we are presently constructing GFP transcriptional fusions with some of the here identified in-gene sequences to check their corresponding transcriptional activity. This could not only lead to a better understanding of our results, but also could improve the S. Typhimurium gene annotation. Our DFI screening did, except for the earlier mentioned csgB (and to a lesser extent fhuA), not sort out well-known genes implicated in Salmonella biofilm formation, such as the genes involved in curli, cellulose and LPS production or genes required for motility (Gerstel and Romling, 2003; Kim and Wei, 2009; Lapidot and Yaron, 2009; Prouty and Gunn, 2003; Prouty et al., 2002; Solano et al., 2002). One can think of different reasons to explain this discrepancy. An obvious explanation for this could be the fact that these genes are also expressed to a certain basal threshold level in planktonic conditions and therefore were not enriched during the subsequent selection rounds. The differences of Salmonella strains used and applied biofilm formation conditions in the different studies could also explain the different outputs. It has been shown that strain background as well as biofilm formation conditions have a large impact on the importance and consequently the expression patterns of the genes involved (Garcia et al., 2004; Prouty and Gunn, 2003; Solano et al., 2002). In general, it is important to state that no single technique is able to identify all factors important for a given process (Mahan et al., 2000; Rediers et al., 2005), but rather complementary techniques have to be used. In this sense, the DFI technique can be added to those already widely used in biofilm research such as high-throughput microarrays (Beloin et al., 2004; Hamilton et al., 2009; Schembri et al., 2003; Whiteley et al., 2001), mutagenesis (Prigent-Combaret et al., 1999; Weiss-Muszkat et al., 2010), proteomics (Hamilton et al., 2009; Sauer et al., 2002) and metabolomics (Gjersing et al., 2007; White et al., 2010) to generate a more exhaustive list of biofilm determinants. Acknowledgements KH is a research assistant of the FWO-Vlaanderen. At the time of experiments, SDK was a post-doctoral researcher of the FWOVlaanderen. This work was also partially financially supported by the Industrial Research Fund of the Katholieke Universiteit Leuven (KP/06/014) and the Centre of Excellence SymBioSys (Research Council K.U.Leuven EF/05/007). We thank Prof. S. Falkow for kindly providing the pFPV25 and pFPV25.1 plasmids and Prof. C. Detweiler for a skillful training in the DFI technology. Appendix A. Supplementary data Supplementary data to this article can be found online at doi:10.1016/j.mimet.2011.01.012. References Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J., 1990. Basic local alignment search tool. Journal of Molecular Biology 215, 403–410. An, D., Parsek, M.R., 2007. The promise and peril of transcriptional profiling in biofilm communities. Current Opinion in Microbiology 10, 292–296. Ballering, K.S., Kristich, C.J., Grindle, S.M., Oromendia, A., Beattie, D.T., Dunny, G.M., 2009. Functional genomics of Enterococcus faecalis: multiple novel genetic determinants for biofilm formation in the core genome. Journal of Bacteriology 191, 2806–2814. Banin, E., Vasil, M.L., Greenberg, E.P., 2005. Iron and Pseudomonas aeruginosa biofilm formation. Proceedings of the National Academy of Sciences of the United States of America 102, 11076–11081.

K. Hermans et al. / Journal of Microbiological Methods 84 (2011) 467–478 Barak, J.D., Liang, A., Narm, K.E., 2008. Differential attachment to and subsequent contamination of agricultural crops by Salmonella enterica. Applied and Environmental Microbiology 74, 5568–5570. Barak, J.D., Gorski, L., Liang, A.S., Narm, K.E., 2009. Previously uncharacterized Salmonella enterica genes required for swarming play a role in seedling colonization. Microbiology 155, 3701–3709. Barker, L.P., Brooks, D.M., Small, P.L., 1998. The identification of Mycobacterium marinum genes differentially expressed in macrophage phagosomes using promoter fusions to green fluorescent protein. Molecular Microbiology 29, 1167–1177. Barnhart, M.M., Chapman, M.R., 2006. Curli biogenesis and function. Annual Review of Microbiology 60, 131–147. Beenken, K.E., Dunman, P.M., McAleese, F., Macapagal, D., Murphy, E., Projan, S.J., et al., 2004. Global gene expression in Staphylococcus aureus biofilms. Journal of Bacteriology 186, 4665–4684. Beloin, C., Valle, J., Latour-Lambert, P., Faure, P., Kzreminski, M., Balestrino, D., et al., 2004. Global impact of mature biofilm lifestyle on Escherichia coli K-12 gene expression. Molecular Microbiology 51, 659–674. Benson, D.A., Karsch-Mizrachi, I., Lipman, D.J., Ostell, J., Rapp, B.A., Wheeler, D.L., 2002. GenBank. Nucleic acids research 30, 17–20. Boddicker, J.D., Ledeboer, N.A., Jagnow, J., Jones, B.D., Clegg, S., 2002. Differential binding to and biofilm formation on, HEp-2 cells by Salmonella enterica serovar Typhimurium is dependent upon allelic variation in the fimH gene of the fim gene cluster. Molecular Microbiology 45, 1255–1265. Brandl, M.T., Mandrell, R.E., 2002. Fitness of Salmonella enterica serovar Thompson in the cilantro phyllosphere. Applied and Environmental Microbiology 68, 3614–3621. Bumann, D., 2002. Examination of Salmonella gene expression in an infected mammalian host using the green fluorescent protein and two-colour flow cytometry. Molecular Microbiology 43, 1269–1283. Cormack, B.P., Valdivia, R.H., Falkow, S., 1996. FACS-optimized mutants of the green fluorescent protein (GFP). Gene 173, 33–38. Costerton, J.W., Lewandowski, Z., Caldwell, D.E., Korber, D.R., Lappin-Scott, H.M., 1995. Microbial biofilms. Annual Review of Microbiology 49, 711–745. Costerton, J.W., Stewart, P.S., Greenberg, E.P., 1999. Bacterial biofilms: a common cause of persistent infections. Science 284, 1318–1322 (New York, N.Y.). Datsenko, K.A., Wanner, B.L., 2000. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proceedings of the National Academy of Sciences of the United States of America 97, 6640–6645. Davey, M.E., O'Toole, G.A., 2000. Microbial biofilms: from ecology to molecular genetics. Microbiology and Molecular Biology Reviews 64, 847–867. De Keersmaecker, S.C., Varszegi, C., van Boxel, N., Habel, L.W., Metzger, K., Daniels, R., et al., 2005. Chemical synthesis of (S)-4, 5-dihydroxy-2, 3-pentanedione, a bacterial signal molecule precursor, and validation of its activity in Salmonella typhimurium. The Journal of Biological Chemistry 280, 19563–19568. Derbyshire, K.M., Barken, K.B., 1987. Mobilization of the non-conjugative plasmid RSF1010: a genetic analysis of its origin of transfer. Mol Gen Genet. 206, 154–160. Dombrecht, B., Vanderleyden, J., Michiels, J., 2001. Stable RK2-derived cloning vectors for the analysis of gene expression and gene function in gram-negative bacteria. Molecular Plant-Microbe Interactions 14, 426–430. Dornenburg, J.E., DeVita, A.M., Palumbo, M.J., Wade, J.T., 2010. Widespread antisense transcription in Escherichia coli. mBio 1 (1), e00024–00010. doi:10.1128/ mBio.0024-0010. Fux, C.A., Costerton, J.W., Stewart, P.S., Stoodley, P., 2005. Survival strategies of infectious biofilms. Trends in Microbiology 13, 34–40. Garcia, B., Latasa, C., Solano, C., Garcia-del Portillo, F., Gamazo, C., Lasa, I., 2004. Role of the GGDEF protein family in Salmonella cellulose biosynthesis and biofilm formation. Molecular Microbiology 54, 264–277. Gerstel, U., Romling, U., 2003. The csgD promoter, a control unit for biofilm formation in Salmonella typhimurium. Research in Microbiology 154, 659–667. Gerstel, U., Park, C., Romling, U., 2003. Complex regulation of csgD promoter activity by global regulatory proteins. Molecular Microbiology 49, 639–654. Ghigo, J.M., 2001. Natural conjugative plasmids induce bacterial biofilm development. Nature 412, 442–445. Gjersing, E.L., Herberg, J.L., Horn, J., Schaldach, C.M., Maxwell, R.S., 2007. NMR metabolomics of planktonic and biofilm modes of growth in Pseudomonas aeruginosa. Analytical Chemistry 79, 8037–8045. Grantcharova, N., Peters, V., Monteiro, C., Zakikhany, K., Romling, U., 2010. Bistable expression of CsgD in biofilm development of Salmonella enterica serovar Typhimurium. Journal of Bacteriology 192, 456–466. Hall-Stoodley, L., Costerton, J.W., Stoodley, P., 2004. Bacterial biofilms: from the natural environment to infectious diseases. Nature Reviews 2, 95–108. Hamilton, S., Bongaerts, R.J., Mulholland, F., Cochrane, B., Porter, J., Lucchini, S., et al., 2009. The transcriptional programme of Salmonella enterica serovar Typhimurium reveals a key role for tryptophan metabolism in biofilms. BMC Genomics 10, 599. Hoiseth, S.K., Stocker, B.A., 1981. Aromatic-dependent Salmonella typhimurium are nonvirulent and effective as live vaccines. Nature 291, 238–239. Igarashi, K., Kashiwagi, K., 1999. Polyamine transport in bacteria and yeast. The Biochemical Journal 344 (Pt 3), 633–642. Karatan, E., Watnick, P., 2009. Signals, regulatory networks, and materials that build and break bacterial biofilms. Microbiology and Molecular Biology Reviews 73, 310–347. Karatan, E., Duncan, T.R., Watnick, P.I., 2005. NspS, a predicted polyamine sensor, mediates activation of Vibrio cholerae biofilm formation by norspermidine. Journal of Bacteriology 187, 7434–7443. Kim, S.H., Wei, C.I., 2009. Molecular characterization of biofilm formation and attachment of Salmonella enterica serovar Typhimurium DT104 on food contact surfaces. Journal of Food Protection 72, 1841–1847.

477

Koide, T., Reiss, D.J., Bare, J.C., Pang, W.L., Facciotti, M.T., Schmid, A.K., et al., 2009. Prevalence of transcription promoters within archaeal operons and coding sequences. Molecular Systems Biology 5, 285. Kusumaningrum, H.D., Riboldi, G., Hazeleger, W.C., Beumer, R.R., 2003. Survival of foodborne pathogens on stainless steel surfaces and cross-contamination to foods. International Journal of Food Microbiology 85, 227–236. Lapidot, A., Yaron, S., 2009. Transfer of Salmonella enterica serovar Typhimurium from contaminated irrigation water to parsley is dependent on curli and cellulose, the biofilm matrix components. Journal of Food Protection 72, 618–623. Latasa, C., Roux, A., Toledo-Arana, A., Ghigo, J.M., Gamazo, C., Penades, J.R., et al., 2005. BapA, a large secreted protein required for biofilm formation and host colonization of Salmonella enterica serovar Enteritidis. Molecular Microbiology 58, 1322–1339. Lazazzera, B.A., 2005. Lessons from DNA microarray analysis: the gene expression profile of biofilms. Current Opinion in Microbiology 8, 222–227. Leriche, V., Carpentier, B., 2000. Limitation of adhesion and growth of Listeria monocytogenes on stainless steel surfaces by Staphylococcus sciuri biofilms. Journal of Applied Microbiology 88, 594–605. Lewis, K., 2008. Multidrug tolerance of biofilms and persister cells. Current Topics in Microbiology and Immunology 322, 107–131. Mahan, M.J., Heithoff, D.M., Sinsheimer, R.L., Low, D.A., 2000. Assessment of bacterial pathogenesis by analysis of gene expression in the host. Annual Review of Genetics 34, 139–164. Marra, A., Asundi, J., Bartilson, M., Lawson, S., Fang, F., Christine, J., et al., 2002. Differential fluorescence induction analysis of Streptococcus pneumoniae identifies genes involved in pathogenesis. Infection and Immunity 70, 1422–1433. McClelland, M., Sanderson, K.E., Spieth, J., Clifton, S.W., Latreille, P., Courtney, L., et al., 2001. Complete genome sequence of Salmonella enterica serovar Typhimurium LT2. Nature 413, 852–856. Mouslim, C., Hilbert, F., Huang, H., Groisman, E.A., 2002. Conflicting needs for a Salmonella hypervirulence gene in host and non-host environments. Molecular Microbiology 45, 1019–1027. Patel, C.N., Wortham, B.W., Lines, J.L., Fetherston, J.D., Perry, R.D., Oliveira, M.A., 2006. Polyamines are essential for the formation of plague biofilm. Journal of Bacteriology 188, 2355–2363. Pistocchi, R., Kashiwagi, K., Miyamoto, S., Nukui, E., Sadakata, Y., Kobayashi, H., et al., 1993. Characteristics of the operon for a putrescine transport system that maps at 19 minutes on the Escherichia coli chromosome. The Journal of Biological Chemistry 268, 146–152. Prigent-Combaret, C., Vidal, O., Dorel, C., Lejeune, P., 1999. Abiotic surface sensing and biofilm-dependent regulation of gene expression in Escherichia coli. Journal of Bacteriology 181, 5993–6002. Prigent-Combaret, C., Brombacher, E., Vidal, O., Ambert, A., Lejeune, P., Landini, P., et al., 2001. Complex regulatory network controls initial adhesion and biofilm formation in Escherichia coli via regulation of the csgD gene. Journal of Bacteriology 183, 7213–7223. Prouty, A.M., Gunn, J.S., 2003. Comparative analysis of Salmonella enterica serovar Typhimurium biofilm formation on gallstones and on glass. Infection and Immunity 71, 7154–7158. Prouty, A.M., Schwesinger, W.H., Gunn, J.S., 2002. Biofilm formation and interaction with the surfaces of gallstones by Salmonella spp. Infection and Immunity 70, 2640–2649. Rasschaert, G., Houf, K., De Zutter, L., 2007. Impact of the slaughter line contamination on the presence of Salmonella on broiler carcasses. Journal of Applied Microbiology 103, 333–341. Rediers, H., Rainey, P.B., Vanderleyden, J., De Mot, R., 2005. Unraveling the secret lives of bacteria: use of in vivo expression technology and differential fluorescence induction promoter traps as tools for exploring niche-specific gene expression. Microbiology and Molecular Biology Reviews 69, 217–261. Ren, D., Bedzyk, L.A., Setlow, P., Thomas, S.M., Ye, R.W., Wood, T.K., 2004. Gene expression in Bacillus subtilis surface biofilms with and without sporulation and the importance of yveR for biofilm maintenance. Biotechnology and Bioengineering 86, 344–364. Rieseberg, M., Kasper, C., Reardon, K.F., Scheper, T., 2001. Flow cytometry in biotechnology. Applied Microbiology and Biotechnology 56, 350–360. Romling, U., 2005. Characterization of the rdar morphotype, a multicellular behaviour in Enterobacteriaceae. Cellular and Molecular Life Sciences 62, 1234–1246. Romling, U., Sierralta, W.D., Eriksson, K., Normark, S., 1998. Multicellular and aggregative behaviour of Salmonella typhimurium strains is controlled by mutations in the agfD promoter. Molecular Microbiology 28, 249–264. Romling, U., Pesen, D., Yaron, S., 2007. Biofilms of Salmonella enterica. In: Rhen, M., Maskell, D., Mastroeni, P., Threlfall, E.J. (Eds.), Salmonella Molecular Biology and Pathogenesis. Horizon Press, Norfolk, UK, pp. 127–145. Sambrook, J., Russel, D.W., 2001. Molecular Cloning: A Laboratory Manual 3rd Edition. Cold Spring Harbor Laboratory Press, Cold Spring Harbor. Santiviago, C.A., Reynolds, M.M., Porwollik, S., Choi, S.H., Long, F., Andrews-Polymenis, H.L., et al., 2009. Analysis of pools of targeted Salmonella deletion mutants identifies novel genes affecting fitness during competitive infection in mice. PLoS Pathogens 5, e1000477. Sauer, K., Camper, A.K., 2001. Characterization of phenotypic changes in Pseudomonas putida in response to surface-associated growth. Journal of Bacteriology 183, 6579–6589. Sauer, K., Camper, A.K., Ehrlich, G.D., Costerton, J.W., Davies, D.G., 2002. Pseudomonas aeruginosa displays multiple phenotypes during development as a biofilm. Journal of Bacteriology 184, 1140–1154. Schembri, M.A., Kjaergaard, K., Klemm, P., 2003. Global gene expression in Escherichia coli biofilms. Molecular Microbiology 48, 253–267. Shah, P., Swiatlo, E., 2008. A multifaceted role for polyamines in bacterial pathogens. Molecular Microbiology 68, 4–16.

478

K. Hermans et al. / Journal of Microbiological Methods 84 (2011) 467–478

Sharma, C.M., Hoffmann, S., Darfeuille, F., Reignier, J., Findeiss, S., Sittka, A., et al., 2010. The primary transcriptome of the major human pathogen Helicobacter pylori. Nature 464, 250–255. Shemesh, M., Tam, A., Kott-Gutkowski, M., Feldman, M., Steinberg, D., 2008. DNAmicroarrays identification of Streptococcus mutans genes associated with biofilm thickness. BMC Microbiology 8, 236. Sittka, A., Lucchini, S., Papenfort, K., Sharma, C.M., Rolle, K., Binnewies, T.T., et al., 2008. Deep sequencing analysis of small noncoding RNA and mRNA targets of the global post-transcriptional regulator, Hfq. PLoS Genetics 4, e1000163. Solano, C., Garcia, B., Valle, J., Berasain, C., Ghigo, J.M., Gamazo, C., et al., 2002. Genetic analysis of Salmonella enteritidis biofilm formation: critical role of cellulose. Molecular Microbiology 43, 793–808. Stewart, P.S., Franklin, M.J., 2008. Physiological heterogeneity in biofilms. Nature Reviews 6, 199–210. Stoodley, P., Sauer, K., Davies, D.G., Costerton, J.W., 2002. Biofilms as complex differentiated communities. Annual Review of Microbiology 56, 187–209.

Valdivia, R.H., Falkow, S., 1996. Bacterial genetics by flow cytometry: rapid isolation of Salmonella typhimurium acid-inducible promoters by differential fluorescence induction. Molecular Microbiology 22, 367–378. Valdivia, R.H., Falkow, S., 1997. Fluorescence-based isolation of bacterial genes expressed within host cells. Science 277, 2007–2011. Weiss-Muszkat, M., Shakh, D., Zhou, Y., Pinto, R., Belausov, E., Chapman, M.R., et al., 2010. Biofilm formation by and multicellular behavior of Escherichia coli O55:H7, an atypical enteropathogenic strain. Applied and Environmental Microbiology 76, 1545–1554. White, A.P., Weljie, A.M., Apel, D., Zhang, P., Shaykhutdinov, R., Vogel, H.J., et al., 2010. A global metabolic shift is linked to Salmonella multicellular development. PLoS ONE 5, e11814. Whiteley, M., Bangera, M.G., Bumgarner, R.E., Parsek, M.R., Teitzel, G.M., Lory, S., et al., 2001. Gene expression in Pseudomonas aeruginosa biofilms. Nature 413, 860–864. Yang, L., Barken, K.B., Skindersoe, M.E., Christensen, A.B., Givskov, M., Tolker-Nielsen, T., 2007. Effects of iron on DNA release and biofilm development by Pseudomonas aeruginosa. Microbiology 153, 1318–1328.