Effects of growth conditions on biofilm formation by Actinobacillus pleuropneumoniae.

Biofilm formation is an important virulence trait of many bacterial pathogens. It has been reported in the literature that only two of the reference strains of the swine pathogen Actinobacillus pleuropneumoniae, representing serotypes 5b and 11, were able to form biofilm in vitro. In this study, we compared biofilm formation by the serotype 1 reference strain S4074 of A. pleuropneumoniae grown in five different culture media. We observed that strain S4074 of A. pleuropneumoniae is able to form biofilms after growth in one of the culture conditions tested brain heart infusion (BHI medium, supplier B). Confocal laser scanning microscopy using a fluorescent probe specific to the poly-N-acetylglucosamine (PGA) polysaccharide further confirmed biofilm formation. In accordance, biofilm formation was susceptible to dispersin B, a PGA hydrolase. Transcriptional profiles of A. pleuropneumoniae S4074 following growth in BHI-B, which allowed a robust biofilm formation, and in BHI-A, in which only a slight biofilm formation was observed, were compared. Genes such as tadC, tadD, genes with homology to autotransporter adhesins as well as genes pgaABC involved in PGA biosynthesis and genes involved in zinc transport were up-regulated after growth in BHI-B. Interestingly, biofilm formation was inhibited by zinc, which was found to be more present in BHI-A (no or slight biofilm) than in BHI-B. We also observed biofilm formation in reference strains representing serotypes 3, 4, 5a, 12 and 14 as well as in 20 of the 37 fresh field isolates tested. Our data indicate that A. pleuropneumoniae has the ability to form biofilms under appropriate growth conditions and transition from a biofilm-positive to a biofilm-negative phenotype was reversible.

INTRODUCTION

Actinobacillus pleuropneumoniae, a member of the Pasteurellaceae, is an important swine pathogen responsible for economic losses in the swine industry. To date, 15 serotypes of A. pleuropneumoniae have been described based on capsular antigens [3, 10]. The virulence of the bacteria is mediated by the coordinated action of several virulence factors, namely the capsule, lipopolysaccharides (LPS), Apx toxins and outer membrane proteins involved in iron uptake [4, 11, 14, 18, 19, 28, 29].

It is widely accepted that the majority of bacteria in virtually all ecosystems (natural, engineered and pathogenic ecosystems) grow in matrix-enclosed biofilms [7]. The matrix provides biofilm cells with a protected microenvironment containing nutrients, secreted enzymes and DNA. The matrix also contributes to the increased resistance to antibiotics and host defenses exhibited by biofilm cells [15]. All members of the Pasteurellaceae are inhabitants of mucosal surfaces of mammals and therefore formation of a biofilm may be crucial to their persistence in vivo. However, biofilms have only been investigated in a few species of the Pasteurellaceae family [16]. In A. pleuropneumoniae, the formation of biofilms on polystyrene microtiter plate is dependent on the production of poly-N-acetylglucosamine (PGA) a linear polymer of N-acetylglucosamine residues in β(1,6) linkage [17, 20]. The production of PGA is encoded by the genes pgaABCD [20]. A novel insertion element, ISApl1, was recently identified in an A/T rich region of the pgaC gene of the biofilm-negative A. pleuropneumoniae strain HB04 [25]. PGA is a substrate for dispersin B (DspB), a biofilm-releasing glycosyl hydrolase produced by Aggregatibacter (Actinobacillus) actinomycetemcomitans and A. pleuropneumoniae [20, 22]. It has also been reported that only 2 of the 15 A. pleuropneumoniae reference strains, representing serotypes 5b and 11, were able to form a biofilm in vitro and that the transition from a biofilm-positive to biofilm-negative phenotype was irreversible [21]. However, Li et al. [24] recently observed slight biomass of biofilm when the A. pleuropneumoniae serotype 1 reference strain S4074 was grown in serum-free TSB but not in serum-containing TSB. In addition, an enhanced biofilm formation was observed in luxS [24] and hns [8] mutants of A. pleuropneumoniae strain S4074.

The aims of the present study were: (i) to re-evaluate biofilm formation by A. pleuropneumoniae reference strain S4074 (serotype 1) under different growth conditions using a standard microtiter plate and crystal violet staining protocol; (ii) to evaluate the ability of 16 reference strains and 37 fresh field isolates to form biofilm in the growth condition shown to allow the best biofilm formation and (iii) to determine the transcriptomic profile of A. pleuropneumoniae strain S4074 when grown in that culture condition.

MATERIALS AND METHODS

Bacterial strains and growth conditions

Bacterial strains used in the present study are listed in Table I. Bacteria were grown on brain heart infusion agar plates (BHI; Difco Laboratories, Detroit, MI, USA) supplemented with 15 μg/mL nicotinamide adenine dinucleotide (NAD). A colony was transferred into 5 mL of Luria-Bertani broth (LB; Difco), tryptic soy broth (TSB; Difco), Mueller Hinton broth (MH; Difco) or BHI (BHI-A; Difco or BHI-B; Oxoid Ltd, Basingstoke, Hampshire, UK) with 5 μg/mL NAD and incubated at 37 °C overnight with agitation. This culture was used for the biofilm assays.

Table I.

A. pleuropneumoniae strains used in the present study.

Strains	Relevant traits	Source
Reference strains
S4074	Serotype 1	K.R. Mittal1
4226	Serotype 2	K.R. Mittal1
1421	Serotype 3	K.R. Mittal1
1462	Serotype 4	K.R. Mittal1
K17	Serotype 5a	K.R. Mittal1
L20	Serotype 5b	K.R. Mittal1
FEMO	Serotype 6	K.R. Mittal1
WF.83	Serotype 7	K.R. Mittal1
405	Serotype 8	K.R. Mittal1
13261	Serotype 9	K.R. Mittal1
13039	Serotype 10	K.R. Mittal1
56153	Serotype 11	K.R. Mittal1
832985	Serotype 12	K.R. Mittal1
N2734	Serotype 13	M. Gottschalk1
39064	Serotype 14	M. Gottschalk1
HS143	Serotype 15	M. Gottschalk1
Field strains
05-7430, 05-7431	Serotype 1	M. Ngeleka2
111A, 719, 2398, 2521	Serotype 1	D. Slavic3
05-4817, 05-C996, 06-996	Serotype 5a	S. Messier1
04-37943, 04-3128, 05-508	Serotype 5a	M. Ngeleka2
05-6501, 06-4091	Serotype 5b	S. Messier1
03-14796, 03-22382, 03-22383, 05-4832	Serotype 5b	M. Ngeleka2
366A, 400, 564D, 888	Serotype 5b	D. Slavic3
05-3695, 06-3008, 06-3060, 06-4108	Serotype 7	S. Messier1
04-37257, 05-14401	Serotype 7	M. Ngeleka2
881, 986, 1951, 4648	Serotype 7	D. Slavic3
05-13146, 05-14657, 05-20080, 05-20081, 05-2983	Serotype 15	M. Ngeleka2

Faculté de médecine vétérinaire, Université de Montréal, St-Hyacinthe, QC, Canada.

Prairie Diagnostic Services, University of Saskatchewan, Saskatoon, SK, Canada.

Ontario Veterinary College, University of Guelph, Guelph, ON, Canada.

These strains are NAD-independent and belong to biotype II.

Biofilm assay in microtiter plates

The microtiter plate biofilm assay is a static assay particularly useful for examining early events in biofilm formation [27]. The wells of a sterile 96-well microtiter plate (Costar® 3599, Corning, NY, USA) were filled in triplicate with a dilution (1/100) of an overnight bacterial culture. Following an incubation of 6 or 24 h at 37 °C, the wells were washed by immersion in water and excess water was removed by inverting plates onto a paper towel. The wells were then filled with 100 μL of crystal violet (0.1%) and the plate was incubated for 2 min at room temperature. After removal of the crystal violet solution, the plate was washed and dried in a 37 °C incubator for 30 min and 100 μL of ethanol (70%) were added to the wells. Absorbance was measured at 590 nm using a spectrophotometer (Powerwave, BioTek Instruments, Winooski, VT, USA).

Scanning laser confocal microscopy

The same biofilm assay protocol was used as described previously. After the 6 or 24 h incubation, the wells were filled with 100 μL of Wheat Germ Agglutinin (WGA)–Oregon Green 488 (Molecular Probes, Eugene, OR, USA) diluted 1/100 in PBS and the plate was incubated for 30 min at room temperature in the dark. The plate was then washed with water and filled with PBS. The plate was observed with a confocal microscope (Olympus FV1000 IX81). WGA was excited at 488 nm and detected using 520 nm filters. The images were processed using Fluoview software (Olympus).

Transcriptomic microarray experiments

RNA extractions

For the microarray experiments, BHI-A or BHI-B broths were inoculated with 500 μL of an overnight culture of A. pleuropneumoniae serotype 1 strain S4074 and grown at 37 °C in an orbital shaker until an optical density of 0.6 was reached. Ice-cold RNA degradation stop solution (95% ethanol, 5% buffer-saturated phenol), shown to effectively prevent RNA degradation and therefore preserve the integrity of the transcriptome [2], was added to the bacterial culture at a ratio of 1:10 (vol/vol). The sample was mixed by inversion, incubated on ice for 5 min, and then spun at 5 000 g for 10 min to pellet the cells. Bacterial RNA isolation was then carried out using the QIAGEN RNeasy MiniKit (QIAGEN, Mississauga, ON, Canada), as prescribed by the manufacturer. During the extraction, samples were subjected to an on-column DNase treatment, as suggested by the manufacturer and then treated with Turbo DNase (Ambion, Austin, TX, USA) to ensure that all DNA contaminants were eliminated. The RNA concentration, quality and integrity were assessed spectrophotometrically and on gel.

Microarray construction and design

For the construction of AppChip2, 2033 ORFs from the complete genome sequence of A. pleuropneumoniae serotype 5b strain L20, representing more than 95% of all ORFs with a length greater than 160 nt, were amplified and spotted in duplicate on the chip. Spotted sheared genomic DNA from A. pleuropneumoniae L20 and porcine DNA are used as controls (GEO: GPL6658). Additional information concerning chip production is described by Gouré et al. [13].

Microarray hybridizations

cDNA synthesis and microarray hybridizations were performed as described [6]. Briefly, equal amounts (15 μg) of test RNA and control RNA were used to set up a standard reverse transcription reaction using random octamers (BioCorp, Montreal, QC, Canada), SuperScript II (Invitrogen, Carlsbad, CA, USA) and aminoallyl-dUTP (Sigma, St. Louis, MO, USA), and the resulting cDNA was indirectly labelled using a monofunctional NHS-ester Cy3 or Cy5 dye (Amersham, Buckinghamshire, UK). The labelling efficiency was assessed spectrophotometrically. Labelled samples were then combined and added to the AppChip2 for overnight hybridization. Five hybridizations were performed for the serotype 1 strain S4074 BHI-A versus BHI-B experiments. All slides were scanned using a Perkin-Elmer ScanArray Express scanner.

Microarray analysis and bioinformatics

Microarray data analysis was conducted with the TM4 Suite of software from the J. Craig Venture Institute [30] as described by Deslandes et al. [9]. Briefly, raw data was first generated using SpotFinder v.3.1.1. Locally weighted linear regression (lowess) was then performed in the Microarray Data Analysis System (MIDAS) in order to normalize the data. The Significance Analysis of Microarray (SAM) algorithm [33], which is implemented in TIGR Microarray Expression Viewer (TMEV), was used to generate a list of differentially expressed genes. During SAM analysis, a false discovery rate (FDR) of 0% was estimated for the serotype 1 strain S4074 BHI-A versus BHI-B experiments.

Effects of DspB and zinc on biofilm formation

Biofilms were grown for 6 or 24 h in BHI-B as described above. The wells were washed with water and then filled with 100 μL of PBS containing 0.2, 2.0 or 20 μg/mL of DspB (Kane Biotech Inc, Winnipeg, MB, Canada) as described by Izano et al. [17]. After incubation at 37 °C for 5 min, the wells were rinsed with water and stained with crystal violet. To monitor the effect of zinc on biofilm formation, bacteria were grown for 6 or 24 h in BHI-B supplemented with 50–250 μg/mL of ZnCl2.

Statistical analysis

The statistical significance (p value) of differences in biofilm phenotypes (mean optical density values) was determined by a paired, one-tailed t-test using GraphPad Prism version 4.0 (GraphPad Software, San Diego, CA, USA).

RESULTS

Biofilm formation and growth conditions

The ability of the A. pleuropneumoniae serotype 1 reference strain S4074 to form biofilms was evaluated using different growth media (Fig. 1). No biofilm was present in the wells containing bacterial cells grown in LB broth while only a slight biofilm was observed in wells containing cells grown in TSB, MH or BHI-A broths after 24 h of incubation. However a pronounced biofilm (p < 0.01) was formed when strain S4074 was grown in BHI-B broth. This was not due to an increased growth in BHI-B compared to BHI-A as similar growth curves were observed in both media.

Figure 1.

Biofilm formation by A. pleuropneumoniae serotype 1 reference strain S4074 grown in different culture media using the crystal violet staining protocol described in Materials and methods. LB: Luria-Bertani; TSB: tryptic soy broth; M-H: Mueller Hinton; BHI: brain heart infusion.

We then evaluated biofilm formation by all the reference strains of A. pleuropneumoniae after growth for 6 or 24 h in BHI-B. Similarly to what was observed with the serotype 1, we found that growth in BHI-B, but not BHI-A, allows biofilm formation in reference strains representing serotypes 4, 5a and 14. In addition to the already reported biofilm formation in serotypes 5b and 11, we also observed biofilms for serotype 3 and 12 reference strains. Moreover, biofilm formation (OD590nm > 0.1) was observed in 20 (54%) of the 37 fresh field isolates of serotypes 1, 5, 7 and 15 that were tested (Fig. 2). In general, serotypes 5a, 5b and 7 field isolates tend to form more biofilms (mean OD of 1.15, 1.47 and 1.47 after 24 h) than isolates from serotypes 1 and 15 (mean OD of 0.36 and 0.80 after 24 h).

Figure 2.

Thirty-seven independent fresh field isolates of A. pleuropneumoniae (representing serotypes 1, 5, 7 and 15) were tested for their ability to form biofilms when grown for 6 h (A) and 24 h (B) in BHI-B using the microtiter plate assay.

When A. pleuropneumoniae strain S4074 grown in BHI-A (no or slight biofilm) was transferred to BHI-B we observed the formation of a pronounced biofilm (p < 0.05). When these cells were then transferred back to BHI-A, the phenotype returned to a slight biofilm (p < 0.05). This was also observed with field isolates representing different serotypes (data not shown).

We observed that for many reference strains, including strain S4074, and field isolates, pronounced biofilms were present after a short incubation period of only 6 h (Fig. 2). The biofilm was visualized by confocal laser scanning microscopy using a fluorescent probe (WGA-Oregon Green) specific to the PGA matrix polysaccharide (Fig. 3). It is evident from these micrographs that A. pleuropneumoniae strain S4074 does not form biofilm when grown in BHI-A while a thick PGA matrix is formed by A. pleuropneumoniae serotype 5b strain L20 grown in the same condition. However, both strains showed a pronounced biofilm when grown in BHI-B. In the case of strain S4074, the biofilm is even more important after 6 h than 24 h of incubation (Fig. 3). Because scanning laser confocal microscopy allows optical sectioning of the biofilm either in the horizontal or the vertical dimension it is possible to evaluate the thickness of the biofilm. We evaluated the thickness of A. pleuropneumoniae strain S4074 biofilm to be of ~ 25 μm after growth in BHI-B for 6 h (Fig. 3C) and even greater (~ 65 μm) for A. pleuropneumoniae strain L20.

Figure 3.

Confocal scanning laser microscopic images of A. pleuropneumoniae serotype 1 strain S4074 (A and C) and serotype 5b strain L20 (B) biofilms stained with WGA-Oregon Green 488. (C) Stack of sections through the X–Z plane of a biofilm formed after 6 h in BHI-B. Bars = 50 μm.

Transcriptomic profiling under different growth conditions

To assess the transcriptional response of A. pleuropneumoniae S4074 after growth in BHI-B compared to BHI-A, transcript profiling experiments using DNA microarrays were performed. Overall, 232 genes were significantly differentially expressed during growth in BHI-B; 152 being up-regulated and 80 being down-regulated (Tab. II). The genes that showed the highest level of up-regulation after growth in BHI-B belonged to the “amino acid biosynthesis”, “energy metabolism”, “transport and binding proteins”, “cell envelope” and “hypothetical/unknown/unclassified” functional classes (Fig. 4). Genes such as tadC and tadD (tight adherence proteins C and D), genes with homology to autotransporter adhesins (APL_0443 and APL_0104) as well as genes pgaABC involved in PGA biosynthesis were up-regulated after growth in BHI-B. A cluster of genes involved in dipeptide transport (dppABCDF) and genes involved in the synthesis of an urease (ureAEFG) were also up-regulated. Down-regulated genes after growth in BHI-B mostly belonged to the “transport and binding proteins”, “cell envelope”, “protein synthesis” and “hypothetical/unknown/unclassified” functional classes. Most notably, cys genes involved in sulphate transport systems were down-regulated, as well as a gene (APL_1096) sharing 59% identity with the DspB gene of A. actinomycetemcomitans.

Table II.

A. pleuropneumoniae strain S4074 genes that are up- or down-regulated after growth in BHI-B compared to growth in BHI-A.

Locus tag	Gene	Description	Fold change
Amino acid biosynthesis
APL_0728	ilvH	Acetolactate synthase small subunit	5.707
APL_0662	aspC	Putative aspartate aminotransferase	5.324
APL_0427	gdhA	NADP-specific glutamate dehydrogenase	4.943
APL_0727	ilvI	Acetolactate synthase large subunit	4.204
APL_0099	ilvG	Acetolactate synthase isozyme II large subunit (AHAS-II)	3.915
APL_1499	thrC	Threonine synthase	3.198
APL_0097	ilvD	Dihydroxy-acid dehydratase	3.142
APL_0393	leuA	2-isopropylmalate synthase	3.000
APL_0098	ilvM	Acetolactate synthase isozyme II small subunit (AHAS-II)	2.934
APL_2027	hisF	Imidazole glycerol phosphate synthase subunit hisF	2.833
APL_0702	serC	Phosphoserine aminotransferase	2.788
APL_0432	leuB	3-isopropylmalate dehydrogenase	2.643
APL_0899	dapA	Dihydrodipicolinate synthase	2.401
APL_0211	glyA	Glycine/serine hydroxymethyltransferase	2.398
APL_0133	cysB	HTH-type transcriptional regulator CysB	2.340
APL_1853	ilvC	Ketol-acid reductoisomerase	2.313
APL_0072	ilvE	Branched-chain-amino-acid aminotransferase	2.001
APL_0859	trpCF	Tryptophan biosynthesis protein trpCF	1.883
APL_2025	hisH	Imidazole glycerol phosphate synthase subunit hisH	1.777
APL_2026	hisA	Phosphoribosylformimino-5-aminoimidazole carboxamide ribotide isomerase	1.739
APL_1198	APL_1198	Putative NAD(P)H nitroreductase	1.708
APL_0139	leuC	3-isopropylmalate dehydratase large subunit 2	1.605
APL_1230	serB	Phosphoserine phosphatase	1.438
APL_0620	aroG	Phospho-2-dehydro-3-deoxyheptonate aldolase	1.428
APL_1873	dapE	Succinyl-diaminopimelate desuccinylase	1.380
Biosynthesis of cofactors, prosthetic groups, and carriers
APL_0207	Dxs	1-deoxy-D-xylulose-5-phosphate synthase (DXPS)	−1.555
APL_1461	menA	1,4-dihydroxy-2-naphthoateoctaprenyltransferase	−1.631
APL_0382	ribD	Riboflavin biosynthesis protein	−1.726
APL_1408	gshA	Glutathione biosynthesis bifunctional protein GshAB	−1.789
Cell envelope
APL_1494	ftpA	Fine tangled pili major subunit	5.705
APL_1921	pgaA	Biofilm PGA synthesis protein PgaA precursor	5.308
APL_0460	plpD	Lipoprotein Plp4	3.801
APL_1923	pgaC	Biofilm PGA synthesis N-glycosyltransferase PgaC	3.591
APL_1922	pgaB	Biofilm PGA synthesis lipoprotein PgaB precursor	3.093
APL_0006	ompP2A	Outer membrane protein P2	2.515
APL_0550	tadC	Tight adherence protein C	1.985
APL_0442	sanA	SanA protein	1.776
APL_0549	tadD	Tight adherence protein D	1.749
APL_0332	hlpB	Lipoprotein HlpB	1.627
APL_1364	gmhA	Putative phosphoheptose isomerase	1.386
APL_0873	rlpB	Putative rare lipoprotein B	−1.391
APL_1028	APL_1028	Possible lipooligosaccharide N-acetylglucosamine glycosyltransferase	−1.445
APL_0747	mepA	Penicillin-insensitive murein endopeptidase precursor	−1.446
APL_0436	mreC	Rod shape-determining protein MreC	−1.585
APL_1086	ompW	Outer membrane protein W precursor	−1.606
APL_1029	APL_1029	Hypothetical protein	−1.650
APL_1424	oxaA	Inner membrane protein OxaA	−1.772
APL_0933	ompP1	Putative outer membrane protein precursor	−2.808
Cellular processes
APL_1489	Tpx	Putative thiol peroxidase	2.252
APL_0988	hktE	Catalase	−1.461
APL_0669	APL_0669	Putative iron dependent peroxidase	−1.483
APL_1442	apxID	RTX-I toxin secretion component	−1.506
APL_1346	ftsY	Cell division protein FtsY-like protein	−1.530
Central intermediary metabolism
APL_1615	Gst	Putative glutathione S-transferase	3.269
APL_1614	ureE	Urease accessory protein UreE	2.601
APL_1613	ureF	Urease accessory protein UreF	2.478
APL_1612	ureG	Urease accessory protein UreG	2.165
APL_1618	ureA	Urease gamma subunit UreA	1.653
DNA metabolism
APL_1931	tagI	3-methyladenine-DNA glycosidase	−1.500
APL_1474	dnaG	DNA primase	−1.551
APL_1282	dnaQ	DNA polymerase III subunit	−1.579
APL_1255	parE	DNA topoisomerase IV subunit	−1.630
APL_1505	holC	DNA polymerase III subunit	−1.663
Energy metabolism
APL_1197	APL_1197	3-hydroxyacid dehydrogenase	3.100
APL_0841	pntB	NAD(P) transhydrogenase subunit beta	2.726
APL_1908	xylA	Xylose isomerase	2.243
APL_0894	fdxH	Formate dehydrogenase, iron-sulfur subunit	2.161
APL_1425	napC	Cytochrome c-type protein NapC	2.159
APL_1799	torC	Pentahemic c-type cytochrome	2.156
APL_0892	fdxG	Formate dehydrogenase, nitrate-inducible, major subunit	2.116
APL_1798	torA	Trimethylamine-N-oxide reductase precursor	1.977
APL_0381	glpC	Anaerobic glycerol-3-phosphate dehydrogenase subunit C	1.919
APL_0842	pntA	NAD(P) transhydrogenase subunit alpha	1.903
APL_0895	fdnI	Formate dehydrogenase, cytochrome b556 subunit	1.816
APL_1208	adhC	Putative alcohol dehydrogenase class 3	1.801
APL_0971	APL_0971	Putative acyl CoA thioester hydrolase	1.796
APL_0652	manB	Phosphomannomutase	1.677
APL_0483	APL_0483	Predicted nitroreductase	1.668
APL_0142	glxK	Glycerate kinase	1.564
APL_0452	sucC	Succinyl-CoA synthetase beta chain	1.515
APL_0461	APL_0461	Predicted hydrolases of the HAD superfamily	1.456
APL_0687	Dld	D-lactate dehydrogenase	1.439
APL_1510	gpsA	Glycerol-3-phosphate dehydrogenase (NAD(P)+)	1.414
APL_1427	napH	Ferredoxin-type protein NapH-like protein	1.360
APL_0789	APL_0789	Dioxygenase	1.253
APL_0983	tktA	Transketolase 2	1.233
APL_1036	pflB	Formate acetyltransferase	−1.653
APL_1498	mgsA	Methylglyoxal synthase	−1.790
APL_1840	ubiC	4-hydroxybenzoate synthetase (chorismate lyase)	−1.952
APL_0857	sdaA	L-serine dehydratase	−3.016
Fatty acid and phospholipid metabolism
APL_1407	Psd	Phosphatidylserine decarboxylase	−1.419
APL_1384	fabH	3-oxoacyl-[acyl-carrier-protein] synthase 3	−1.826
APL_1385	plsX	Fatty acid/phospholipid synthesis protein PlsX	−2.706
Mobile and extrachromosomal element functions
APL_1056	APL_1056	Transposase	1.560
APL_0985	APL_0985	Transposase	1.271
Protein fate
APL_0871	pepE	Peptidase E	2.551
APL_1101	pepA	Putative cytosol aminopeptidase	1.913
APL_0254	pepD	Aminoacyl-histidine dipeptidase	1.903
APL_1883	ptrA	Protease 3 precursor	1.680
APL_0928	hscB	Co-chaperone protein HscB-like protein	1.377
APL_1068	secF	Protein-export membrane protein SecF	−1.496
APL_0321	dsbB	Disulfide bond formation protein B	−1.557
APL_1035	pflA	Pyruvate formate-lyase 1-activating enzyme	−1.774
Protein synthesis
APL_1821	rpmE	50S ribosomal protein L31	2.211
APL_0484	rimK	Ribosomal protein S6 modification protein	1.533
APL_1781	rpsM	30S ribosomal protein S13	−1.401
APL_0205	APL_0205	Predicted rRNA methyltransferase	−1.538
APL_0399	ksgA	Dimethyladenosine transferase	−1.578
APL_0679	glnS	Glutaminyl-tRNA synthetase	−1.584
APL_0641	truB	tRNA pseudouridine synthase B	−1.742
APL_1383	trmB	tRNA (guanine-N(7)-)-methyltransferase	−1.756
APL_0574	APL_0574	tRNA-specific adenosine deaminase	−1.778
APL_0723	Tgt	Queuine tRNA-ribosyltransferase	−1.937
Purines, pyrimidines, nucleosides, and nucleotides
APL_0958	purH	Bifunctional purine biosynthesis protein PurH	1.856
APL_0593	guaB	Inosine-5′-monophosphate dehydrogenase	1.485
APL_1343	Cdd	Cytidine deaminase	1.278
APL_1014	deoD	Purine nucleoside phosphorylase DeoD-like protein	−1.430
APL_0351	Ndk	Nucleoside diphosphate kinase	−1.531
APL_1839	Udp	Uridine phosphorylase	−1.617
APL_1075	purA	Adenylosuccinate synthetase	−1.762
Regulatory functions
APL_0059	narP	Nitrate/nitrite response regulator protein	2.552
APL_0823	glpR	Glycerol-3-phosphate regulon repressor	1.908
APL_1295	argR	Arginine repressor	1.896
APL_0126	APL_0126	HIT-like protein	1.580
APL_0395	rseA	Putative sigma-E factor negative regulatory protein	1.524
APL_1668	rbsR	Ribose operon repressor	1.302
APL_1270	sprT	Putative SprT-like protein	−1.483
APL_1233	malT	HTH-type transcriptional regulator MalT	−1.484
APL_1540	tldD	TldD-like protein	−1.578
Transcription
APL_0560	rhlB	ATP-dependent RNA helicase RhlB	1.409
APL_0423	rnhA	Ribonuclease HI	1.345
APL_0201	nusB	Transcription antitermination protein NusB	−1.457
Transport and binding proteins
APL_0967	gltS	Sodium/glutamate symport carrier protein	4.155
APL_0377	glpT	Glycerol-3-phosphate transporter	3.247
APL_0064	dppA	Periplasmic dipeptide transport protein	3.168
APL_0869	abgB	Aminobenzoyl-glutamate utilization-like protein	3.004
APL_1857	merP	Copper chaperone MerP	2.911
APL_0068	dppF	Dipeptide transport ATP-binding protein DppF	2.860
APL_1665	gntP_1	Gluconate permease	2.723
APL_0066	dppC	Dipeptide transport system permease protein DppC	2.640
APL_1440	znuA	High-affinity zinc uptake system protein ZnuA precursor	2.600
APL_0065	dppB	Dipeptide transport system permease protein DppB	2.229
APL_0067	dppD	Dipeptide transport ATP-binding protein DppD	2.036
APL_1448	afuC	Ferric ABC transporter ATP-binding protein	1.855
APL_1319	ptsB	PTS system sucrose-specific EIIBC component	1.744
APL_1320	thiQ	Thiamine transport ATP-binding protein ThiQ	1.569
APL_1622	cbiM	Predicted ABC transport permease protein CbiM	1.433
APL_1620	cbiO	Predicted ABC transport ATP-binding protein CbiO	1.417
APL_1173	pnuC	Nicotinamide mononucleotide transporter	1.408
APL_0749	APL_0749	Potassium efflux system KefA	−1.436
APL_1212	tehA	Tellurite resistance protein TehA	−1.543
APL_0716	APL_0716	Iron(III) ABC transporter, permease protein	−1.547
APL_1253	APL_1253	Putative sodium/sulphate transporter	−1.598
APL_1846	cysT	Sulfate transport system permease protein cysT	−1.684
APL_0191	APL_0191	Predicted Na+-dependent transporter of the SNF family	−1.751
APL_1083	arcD	Putative arginine/ornithine antiporter	−1.786
APL_2016	fhuA	Ferrichrome-iron receptor FhuA	−2.031
APL_1847	cysW	Sulfate transport system permease protein cysW	−2.195
APL_1844	cysN	Sulphate adenylate transferase subunit 1	−2.375
APL_1848	cysA	Sulfate/thiosulfate import ATP-binding protein cysA	−2.401
APL_1843	cysJ	Sulfite reductase [NADPH] flavoprotein alpha-component	−2.757
APL_1127	APL_1127	Predicted Na+/alanine symporter	−3.402
Hypothetical/unknown/unclassified
APL_1100	APL_1100	Hypothetical protein	3.395
APL_0920	APL_0920	Hypothetical protein	2.835
APL_1882	APL_1882	Hypothetical protein	2.776
APL_1856	APL_1856	Hypothetical protein	2.775
APL_1855	APL_1855	Hypothetical protein	2.763
APL_0443	APL_0443	Autotransporter adhesin	2.762
APL_1252	APL_1252	Hypothetical protein	2.739
APL_0134	APL_0134	Hypothetical protein	2.681
APL_0836	APL_0836	Putative transcriptional regulator	2.661
APL_1588	APL_1588	Predicted TRAP transporter solute receptor	2.464
APL_1491	APL_1491	Hypothetical protein	2.282
APL_0104	APL_0104	Autotransporter adhesin	2.231
APL_1069	ftnA	Ferritin-like protein 1	2.194
APL_1059	APL_1059	Hypothetical transposase-like protein	2.172
APL_1690	APL_1690	Inner membrane protein	2.168
APL_0245	APL_0245	Transferrin binding protein-like solute binding protein	2.097
APL_1191	namA	NADPH dehydrogenase	2.078
APL_1948	APL_1948	Hypothetical protein	2.061
APL_0870	APL_0870	Putative C4-dicarboxylate transporter	2.034
APL_0643	APL_0643	Hypothetical protein	2.029
APL_1743	APL_1743	Ser/Thr protein phosphatase family protein	1.999
APL_0426	APL_0426	Hypothetical protein	1.994
APL_1791	APL_1791	Putative periplasmic iron/siderophore binding protein	1.944
APL_0970	APL_0970	Hypothetical protein	1.908
APL_1070	ftnB	Ferritin-like protein 2	1.907
APL_1894	APL_1894	Hypothetical protein	1.907
APL_1374	APL_1374	Hypothetical protein	1.803
APL_1206	APL_1206	Plasmid stability-like protein	1.794
APL_1881	APL_1881	Hypothetical protein	1.792
APL_0038	APL_0038	Hypothetical protein	1.730
APL_1355	APL_1355	Hypothetical protein	1.716
APL_0471	APL_0471	Hypothetical protein	1.707
APL_1438	APL_1438	Hypothetical protein	1.689
APL_1437	APL_1437	Hypothetical protein	1.643
APL_1423	APL_1423	Hypothetical protein	1.612
APL_0125	APL_0125	Hypothetical protein	1.608
APL_0096	APL_0096	Zinc transporter family protein ZIP	1.592
APL_0220	APL_0220	Putative lipoprotein	1.583
APL_1934	APL_1934	Hypothetical protein	1.570
APL_1574	APL_1574	Hypothetical protein	1.543
APL_0036	APL_0036	Hypothetical protein	1.533
APL_0222	APL_0222	Putative lipoprotein	1.518
APL_1088	APL_1088	Hypothetical protein	1.512
APL_1207	APL_1207	Hypothetical protein	1.510
APL_0463	APL_0463	Predicted sortase and related acyltransferases	1.448
APL_1859	APL_1859	Probable NADH-dependent butanol dehydrogenase 1	1.448
APL_1828	APL_1828	PilT protein-like protein	1.447
APL_0433	msrB	Methionine sulfoxide reductase B	1.415
APL_1189	APL_1189	Hypothetical protein	1.393
APL_0090	APL_0090	Hypothetical protein	1.360
APL_1709	APL_1709	Hypothetical protein	−1.307
APL_0357	APL_0357	Hypothetical protein	−1.328
APL_1380	APL_1380	Hypothetical protein	−1.394
APL_1729	APL_1729	Hypothetical protein	−1.401
APL_1062	APL_1062	Hypothetical protein	−1.468
APL_0179	APL_0179	Hypothetical protein	−1.481
APL_0940	APL_0940	Hypothetical protein	−1.482
APL_1273	APL_1273	Putative fimbrial biogenesis and twitching motility protein PilF-like protein	−1.488
APL_1131	APL_1131	Hypothetical protein	−1.540
APL_0583	APL_0583	Hypothetical protein	−1.585
APL_1096	APL_1096	Hypothetical protein (59% ID dispersine B)	−1.594
APL_0936	APL_0936	Hypothetical protein	−1.616
APL_1115	APL_1115	Hypothetical protein	−1.639
APL_0811	APL_0811	Hypothetical protein	−1.682
APL_1898	ap2029	Hypothetical protein	−1.798
APL_1654	gidB	Methyltransferase GidB	−1.816
APL_0340	APL_0340	Hypothetical protein	−1.893
APL_1381	APL_1381	Hypothetical protein	−1.926
APL_0053	typA	GTP-binding protein	−2.043
APL_1681	APL_1681	Hypothetical protein	−2.233

Figure 4.

Functional classification of the differentially expressed genes during growth of A. pleuropneumoniae S4074 in BHI-B according to TIGRFAM. AAB: amino acids biosynthesis; BCPC: biosynthesis of cofactors, prosthetic groups and carriers; CE: cell envelope; CP: cellular processes; CIM: central intermediary metabolism; DNA: DNA metabolism; EM: energy metabolism; FAPM: fatty acid and phospholipid metabolism; HUU: hypothetical proteins/unclassified/unknown; MEEF: mobile and extrachromosomal element functions; PF: protein fate; PS: protein synthesis; PPNN: purines, pyrimidines, nucleosides and nucleotides; RF: regulatory functions; TR: transcription and TBP: transport and binding proteins.

Effect of DspB on biofilm formation

Enzymatic treatment with DspB of biofilms of A. pleuropneumoniae strains S4074 and L20 grown for 6 or 24 h almost completely dispersed them (p < 0.05) confirming the presence of PGA in the biofilm matrix.

Effect of zinc on biofilm formation

Chemical analysis showed differences in some divalent cations concentration between BHI-A (Fe < 0.10 ppm, Zn 2.03 ppm) and BHI-B (Fe 0.10 ppm, Zn 1.75 ppm) while no differences were observed for others (Ca, Cu, Mg, Mn). We therefore hypothesized that the difference in biofilm formation observed after growth in BHI-B compared to BHI-A might be due to cations concentration. Since the concentration of zinc was found to be higher in BHI-A (no or slight biofilm) we tested a possible inhibitory effect of this cation on biofilm formation. The addition of ZnCl2 to BHI-B inhibited, in a dose-dependent manner, the formation of biofilms by A. pleuropneumoniae strains S4074 and L20 (Fig. 5). A complete inhibition (p < 0.01) was observed when 100 μg/mL of ZnCl2 was added to BHI-B, a concentration which did not affect growth after 24 h (data not shown). A similar inhibition was also observed with the addition of ZnSO4, ZnO, and Zn3(PO4)2 but not with MgCl2 or CaCl2 thus confirming that the inhibition was due to the addition of zinc. Biofilm formation in A. actinomycetemcomitans was also inhibited by zinc (data not shown). Interestingly, genes potentially involved in zinc transport (znuA and APL_0096) were up-regulated after growth in BHI-B (Tab. II).

Figure 5.

Effect of the addition of ZnCl2 on biofilm formation by A. pleuropneumoniae serotype 1 strain S4074 (App1) and serotype 5b strain L20 (App5b) grown for 6 h.

DISCUSSION

Biofilm formation is an important virulence trait of many bacterial pathogens including A. pleuropneumoniae. It has been previously reported that only 2 of the 15 A. pleuropneumoniae reference strains, representing serotypes 5b and 11, were able to form a biofilm in vitro [21]. We observed however an increased stickiness of colonies when strain A. pleuropneumoniae S4074 was grown on plates made of BHI from one of two different suppliers. In addition, Li et al. [24] recently observed slight biomass of biofilm when the A. pleuropneumoniae serotype 1 reference strain S4074 was grown in serum-free TSB and that an enhanced biofilm formation was observed in luxS [24] and hns [8] mutants of A. pleuropneumoniae S4074. These observations brought us to re-evaluate biofilm formation by strain A. pleuropneumoniae S4074 under different growth conditions using a standard microtiter plate and crystal violet staining protocol. Our data indicate that strain S4074 has the ability to form a pronounced biofilm when grown in the appropriate conditions, and that the biofilm was sensitive to DspB treatment and can be inhibited by zinc. Transition from a biofilm-positive to a biofilm-negative phenotype is not irreversible in contrast to what was reported by Kaplan and Mulks [21] under different conditions.

Transcript profiling experiments using DNA microarrays indicated that overall, 232 genes were significantly differentially expressed during growth in BHI-B. Genes such as tadC, tadD, genes with homology to autotransporter adhesins as well as genes pgaABC involved in PGA biosynthesis were up-regulated after growth in BHI-B. While we can hypothesize that these genes might be important for the formation of the biofilm itself, it is also interesting to note that many of the same genes (tadB, rcpA, gene APL_0443 with high homology to the Hsf autotransporter adhesin of Haemophilus influenzae as well as genes pgaBC involved in biofilm biosynthesis) were up-regulated, when the transcriptomic profile of A. pleuropneumoniae was determined after contact with porcine lung epithelial cells [1], thus emphasizing the possible importance of biofilm formation for the establishment of the infection.

Initial steps in biofilm development require the transcription, early on, of genes involved in reversible attachment and motility, before a subsequent switch towards the transcription of genes involved in the irreversible attachment of bacteria [35]. This second irreversible attachment might require the synthesis of adhesive organelles, such as the curli fibers (csg genes). Interestingly, gene APL_0220 is a putative lipoprotein of the CsgG family, responsible for the transport and assembly of curli fibers. The up-regulation of other genes possibly involved in adhesion processes (tadC, tadD, Hsf homolog APL_0443) might indicate that bacterial cells were entering or in the middle of this irreversible attachment phase. In A. actinomycetemcomitans, the Tad locus is essential for biofilm formation [32]. The fact that the transcription of a zinc-specific transporter (znuA) was increased, combined with the decrease in transcription of an hypothetical Zn-dependant protease (APL_1898) and lower concentration of this metal in BHI-B lead us to believe that Zn restriction might be a signal leading to increase biofilm formation.

It is tempting to speculate that growth in BHI-B affected the expression of regulators which in turn affected PGA expression and biofilm formation. Indeed, it has been recently shown that an enhanced biofilm formation was observed in a hns mutant of A. pleuropneumoniae strain S4074 [8] and that over-expression of RpoE in a rseA mutant is sufficient to alleviate repression of biofilm formation by H-NS1. However, other genes have been shown to affect biofilm formation in A. pleuropneumoniae. An enhanced biofilm formation was observed in a quorum sensing (luxS) mutant [24] while a mutant in the ArcAB two-component system facilitating metabolic adaptation to anaerobicity (arcA) [5] and an autotransporter serine protease (AasP) mutant were deficient in biofilm formation [31]. It is interesting to note that many genes involved in branched-chain amino acid biosynthesis (ilv genes) were up-regulated after growth in BHI-B. Limitation of branched-chain amino acids was shown to be a cue for expression of a subset of in vivo induced genes in A. pleuropneumoniae, including not only genes involved in the biosynthesis of branched-chain amino acids, but also other genes that are induced during infection of the natural host [34].

Our data indicate that many strains of A. pleuropneumoniae have the ability to form biofilms under appropriate growth conditions. This is an important observation considering that A. pleuropneumoniae biofilm cells exhibit increased resistance to antibiotics compared to planktonic cells [17] and may also exhibit increased resistance to biocides [12]. Biofilms are often associated with chronic infections but the fact that A. pleuropneumoniae can form an important biofilm after only 6 h of incubation suggests that biofilm formation might also play a role in acute infections.

We have undertaken the screen of a large library of mini-Tn10 isogenic mutants of A. pleuropneumoniae S4074 in order to identify other genes that are involved in biofilm formation and/or regulation. A better understanding of biofilm formation in A. pleuropneumoniae might lead to the development of molecules or strategies to interfere with biofilm formation and prevent infection in pigs. In that respect, we made an important, and unexpected, observation that zinc could completely inhibit biofilm formation in A. pleuropneumoniae and A. actinomycetemcomitans, which also synthesizes PGA [20]. We do not know at this time how zinc interferes with PGA biosynthesis and biofilm formation but some glycosyltransferases have been shown to be inhibited by zinc [23]. Hypozincemia which occurs during infection and inflammation [26] might therefore favour biofilm formation by A. pleuropneumoniae. Knowing that PGA functions as a biofilm matrix polysaccharide in phylogenetically diverse bacterial species such as Staphylococcus aureus, S. epidermidis, and Escherichia coli [20], it would be worth investigating whether zinc can also interfere with PGA biosynthesis in these other bacterial pathogens.