The S-Layer Homology Domain-Containing Protein SlhA from Paenibacillus alvei CCM 2051T Is Important for Swarming and Biofilm Formation

Background Swarming and biofilm formation have been studied for a variety of bacteria. While this is well investigated for Gram-negative bacteria, less is known about Gram-positive bacteria, including Paenibacillus alvei, a secondary invader of diseased honeybee colonies infected with Melissococcus pluton , the causative agent of European foulbrood (EFB). Methodology Paenibacillus alvei CCM 2051T is a Gram-positive bacterium which was recently shown to employ S-layer homology (SLH) domains as cell wall targeting modules to display proteins on its cell surface. This study deals with the newly identified 1335-amino acid protein SlhA from P. alvei which carries at the C‑terminus three consecutive SLH-motifs containing the predicted binding sequences SRGE, VRQD, and LRGD instead of the common TRAE motif. Based on the proof of cell surface location of SlhA by fluorescence microscopy using a SlhA-GFP chimera, the binding mechanism was investigated in an in vitro assay. To unravel a putative function of the SlhA protein, a knockout mutant was constructed. Experimental data indicated that one SLH domain is sufficient for anchoring of SlhA to the cell surface, and the SLH domains of SlhA recognize both the peptidoglycan and the secondary cell wall polymer in vitro. This is in agreement with previous data from the S-layer protein SpaA, pinpointing a wider utilization of that mechanism for cell surface display of proteins in P. alvei. Compared to the wild-type bacterium ΔslhA revealed changed colony morphology, loss of swarming motility and impaired biofilm formation. The phenotype was similar to that of the flagella knockout Δhag, possibly due to reduced EPS production influencing the functionality of the flagella of ΔslhA. Conclusion This study demonstrates the involvement of the SLH domain-containing protein SlhA in swarming and biofilm formation of P. alvei CCM 2051T.


Introduction
The constitution of the cell surface of bacteria strongly influences the physicochemical properties of bacterial cells, the bacterial life-style (planktonic versus biofilm) and the potential for survival in a competitive habitat. Consequently, investigating cell surface compounds of bacteria, their display mechanism and their functional influence on the bacterium can add to our knowledge of strategies for interfering with bacterial colonization as relevant, for instance, in the context of combating bacterial infections.
S-layer homology (SLH) domains are cell wall-targeting modules employed by various Gram-positive as well as Gramnegative bacteria to display extracellular proteins such as enzymes, outer membrane proteins, and surface (S-) layer (glyco) proteins on the bacterial cell surface [1,2]. S-layers are 2D crystalline arrays that completely cover bacterial cells [3] that are in many bacteria of the Bacillaceae family, noncovalently attached to the bacterial cell envelope via their SLHdomains [4,5]. While lectin type-like binding of SLH-domains to a peptidoglycan (PG)-associated, non-classical, pyruvylated secondary cell wall polymer (SCWP) has been known for some time [2,6], we have shown recently for the S-layer protein SpaA of the Gram-positive bacterium Paenibacillus alvei CCM 2051 T that its SLH-domains have dual recognition function [2]: The SLH-domains recognize a SCWP with the structure [(Pyr4,6)-β-D-ManpNAc-(1→4)-β-D-GlcpNAc-(1→3)] n~11 -(Pyr4,6)-β-D-ManpNAc-(1→4)-α-D-GlcpNAc-(1→ that is linked via a phosphate-containing bridge to muramic acid residues of the PG backbone [7], and PG itself [2]. Furthermore, two out of three functional SLH-domains were found to be sufficient for cell wall binding of SpaA, regardless of the location of the SLHdomains [2]. In addition to the S-layer protein, the P. alvei CCM 2051 T genome reveals a suite of at least 17 more open reading frames encoding SLH-domains. Among these presumably surface located proteins is a protein named SlhA (S-layer homology domain protein A) that is the focus of the current study. SlhA is a 1335-amino acid protein showing only 17% of overall homology to SpaA. It comprises a typical Gram-positive N-terminal signal peptide (residues 1-31 of the pre-protein) followed by a galactose-binding domain (CBM6, residues 91-200) typical of proteins binding to specific ligands, such as cell-surface-attached carbohydrates, and three C-terminal SLH-domains containing the predicted modified binding motifs SRGE in SLH-domain 1 (residues 1125-1169), VRQD in SLHdomain 2 (residues 1198-1242), and LRGD in SLH-domain 3 (residues 1267-1319) ( Figure 1A). Thus, the question arose, if the cell surface display mechanism established for the SpaA S-layer protein [2] would also be valid for SlhA. It is important to note that while the highly conserved four amino acid motif TRAE present in many SLH-domains [8] plays a key role for the binding function to SCWP [8,9], functional variations in that motif have been reported for P. alvei CCM 2051 T where the motifs TVEE and TRAQ are present [2] as well as for Thermoanaerobacterium thermosulfurigenes EM1 [8].
The slhA gene of P. alvei CCM 2051 T is located on the bacterial chromosome between the predicted SCWP biosynthesis locus (downstream) and the S-layer gene spaA (upstream) where it is transcribed under its own promoter ( Figure 1B) [10]. Since this, together with its prominent location on the bacterial cell surface, might be indicative of the involvement of the SlhA protein in proper cell envelope formation and/or mediation of cell surface phenomena in P. alvei CCM 2051 T as potentially important to the life-style of the bacterium, we were interested in further characterizing this protein.
Regarding its life-style, P. alvei CCM 2051 T is a mesophilic, endospore-forming bacterium that has been reported to swarm vigorously on solidified standard culture medium [11,12]. Swarming is a flagella-driven strategy for motility utilized by a wide range of bacterial species, including paenibacilli [13,14]. While swarming has been previously regarded as a preliminary step in biofilm formation [15], it is now evident that it can be antagonistically regulated [16][17][18]. For SadB (surface attachment defective B) from Pseudomonas aeruginosa, for instance, it has been shown that swarming and biofilm formation are inversely regulated via the modulation of flagellar reversals and by influencing the production of exopolysaccharides [16][17][18]. In a bacterial biofilm, cells are tightly associated with each other and/or attached to a surface and embedded in a self-produced matrix of exopolymeric substances (EPS) [15,[19][20][21]. This matrix contains a variety of (glyco) proteins, (glyco) lipids, exopolysaccharides and, in several cases, extracellular DNA [22]. It is crucial for preventing the washout of enzymes, keeping them close to cells and allowing for effective degradation of polymeric and particulate material. Since the biofilm confers several advantages to the bacteria, such as increased resistance against toxic conditions or increased ability to escape the immune response of the host [23,24], it is the preferred life-style of many, especially pathogenic, bacteria. Among Paenibacillus species, Paenibacillus polymyxa, a growth-promoting rhizobacterium, is one of the best studied organisms regarding biofilm formation [25]. While the possibility to knock-out slhA demonstrated that the SlhA protein is not necessary for viability of P. alvei CCM 2051 T cells (this study), we speculated that it might affect bacterial motility.
In this study, we have investigated the predicted cell surface protein SlhA from P. alvei CCM 2051 T . This included (i) demonstration of its co-display with the S-layer protein SpaA on the bacterial cell surface in vivo by homologous coexpression of a chimera made of SlhA-enhanced green fluorescent protein (EGFP) and His 6 -tagged SpaA, (ii) assessment of the relevance of the SLH-domains of SlhA for cell surface anchoring in an in vitro binding assay using His 6 -tagged SlhA variants and native PG-containing cell wall sacculi that either contained SCWP or were deprived of it, and (iii) first functional insights into SlhA. In accordance with previous data from the S-layer protein SpaA our experiments indicated a dual recognition function of SlhA for both PG and SCWP with the innermost SLH-domain being sufficient for cell surface anchoring. This pinpoints a wider utilization of this mechanism for cell surface display of proteins in P. alvei CCM 2051 T . The slhA knockout mutant displayed changed colony morphology, lost its ability to swarm and showed impaired biofilm formation, possibly due to changes in the EPS. In contrast, a mutation that disrupted flagella synthesis did not show altered EPS production or colony morphology.

Materials and Methods
Bacterial strains and growth conditions P. alvei CCM 2051 T was obtained from the Czech Collection of Microorganisms (CCM; Brno, Czech Republic) and was grown at 37°C and 200 rpm in Luria-Bertani (LB) broth or on LB agar plates supplemented with 10 µg/ml chloramphenicol (Cm), when appropriate. Escherichia coli DH5α cells (Invitrogen) were cultivated in selective Luria Bertani (LB) medium (agar and broth) supplemented with 30 μg/ml chloramphenicol (Cm). All strains used in this study are listed in Table 1.

General molecular methods
All enzymes were purchased from Fermentas. Genomic DNA of P. alvei CCM 2051 T was isolated by using a Genomic Tip 100 kit (Qiagen) according to the manufacturer's instructions, except that cells were broken by repeated freezing and thawing cycles (10 times) [26]. The GeneJET™ Gel Extraction Kit (Fermentas) was used to purify DNA fragments from agarose gels and to purify digested plasmids and oligonucleotides. Plasmid DNA from transformed cells was isolated with the GeneJET™ Plasmid Miniprep kit (Fermentas). Agarose gel electrophoresis was performed as described elsewhere [27]. Primers for PCR and DNA sequencing were purchased from Invitrogen (Table 2). PCR was performed using the Phusion ® High-Fidelity DNA Polymerase (Fermentas) and the thermal cycler My Cycler TM (Bio-Rad). Transformation of chemically competent E. coli DH5α cells was done according to the manufacturer's protocol (Invitrogen). Transformation of P. alvei CCM 2051 T wild-type or P. alvei CCM 2051 T ΔslhA cells, with the latter showing a better transformation efficiency, is described elsewhere [26]. Transformants were screened by PCR using RedTaq ReadyMix PCR mix (Sigma-Aldrich), and recombinant clones were analyzed by restriction mapping and confirmed by sequencing (LGC).

Gene knockouts
Disruption of the slhA gene located in the SCWP biosynthesis locus of P. alvei CCM 2051 T and the flagella gene (hag, PAV_2c01710) was performed as described previously [26]. The Ll.LtrB targetron of pTT_wsfA243 was retargeted prior to transformation into P. alvei CCM 2051 T . Identification of potential insertion sites and design of PCR primers for the modification of the intron RNA was accomplished by a computer algorithm (www.Sigma-Aldrich.com/ Targetronaccess). The retargeted Ll.LtrB targetron was subsequently digested with HindIII and BsrGI and ligated into pTT_wsfA243 [28] digested with the same restriction enzymes, thereby replacing the wsfA targetron.
To complement the knockout, slhA under its native promoter was amplified from P. alvei CCM 2051 T genomic DNA using the primers P(SlhA) _HindIII_for/ SlhA_KpnI_rev. The amplification product was SphI/KpnI-digested and inserted into the linearized vector pEXALV. The resulting plasmid was named pEXALV_P(SlhA) _SlhA (Table 1) and was transformed into P. alvei CCM 2051 T ΔslhA cells [26].

Plasmid construction for in vivo studies
For analyzing co-display of SlhA and SpaA on P. alvei CCM 2051 T ΔslhA cells in vivo, the vector pSURF was designed based on the pEXALV_EGFP vector [2]. For the construction of a chimera of SlhA and enhanced green fluorescent protein (EGFP) including a GAGGAGGT linker, the slhA gene was PCR-amplified from genomic DNA of P. alvei CCM 2051 T using the primer pair SlhA_SphI_for/ SlhA_w/o Stop_Linker_XbaI_rev. The amplification product was SphI/ XbaI-digested and inserted into linearized pEXALV_EGFP. The resulting plasmid was named pEXALV_SlhA_Linker_EGFP (Table 1). Subsequently, the sgsE S-layer gene promoter of G. stearothermophilus NRS 2004/3a, P(SgsE), and the S-layer gene spaA of P. alvei CCM 2051 T were amplified from pEXALV_SP_SpaA_6HIS as a template using the primers P(SgsE) _SacI_for/ SpaA_6His_SacI_rev and ligated into pEXALV_SlhA_Linker_EGFP. The resulting plasmid was named pSURF (Table 1).

Recombinant production of His-tagged SlhA variants for in vitro assays
All SlhA variants used for in vitro assays were produced as His 6 -tagged constructs for detection purposes. The C-terminal His 6 -tag was fused to the sequence of slhA by PCR, using the primers SlhA_ohne SP_NcoI_for/ SlhA_ganz_XhoI_rev for slhA lacking the sequence coding for the signal peptide, SlhA_ohne SP_NcoI_for/ SlhA_His_SLH12_XhoI_rev for slhA lacking the sequences coding for the signal peptide and the third SLH domain, SlhA_ohne SP_NcoI_for/ SlhA_His_SLH1_XhoI_rev for slhA lacking the sequences coding for the signal peptide, the second and third SLH domain, and SlhA_ohne SP_NcoI_for/ SlhA_His_ohne SLH_XhoI_rev for slhA lacking the sequences coding for the signal peptide and all three SLH domains.
All amplifications were done using P. alvei CCM 2051 T genomic DNA as a template. The different His 6 -tagged slhA amplification products were digested with NcoI/XhoI and cloned  Figure 1A) and transformed into E. coli BL21 (DE3) cells. Freshly transformed cells were grown in LB medium [27] to the mid exponential growth phase (OD 600 ~0.6), protein expression was induced with a final concentration of 0.5 mM isopropyl-β-D-thiogalactopyranosid (IPTG) and cultures were grown for additional 4 h at 37°C and 200 rpm. Cells were harvested by centrifugation (4,500 x g, 30 min, 4°C).

Preparation of native peptidoglycan-containing cell wall sacculi
For the preparation of protein-free peptidoglycan (PG)containing cell wall sacculi including native SCWP of P. alvei CCM 2051 T (PG(+)), cells were grown until the late exponential phase and broken by ten freezing and thawing cycles [26]. After centrifugation (35,000 x g, 30 min, 4°C), the pellet was  [29].

Interaction studies of SlhA variants and PG-containing cell wall sacculi
The different His 6 -SlhA variants were tested for their ability to interact with PG(+) and PG(-) following a recently published protocol [2,8]. Briefly, ~1 µg of SlhA protein (crude cell extract) was incubated at 37°C for 1 h with 0.2 mg of lyophilized PG(+) in 25 mM Tris-HCl, pH 8.0, containing 0.2% (v/v) Tween 20 (incubation buffer), in a total volume of 125 µl. The mixture was centrifuged (16,100 x g, 20 min, 4°C) yielding a supernatant containing unbound protein and a pellet of the insoluble cell wall components including attached protein. The pellet was washed twice, suspended in incubation buffer, and fractions were investigated by SDS-PAGE and by Westernimmunoblotting using anti-His 6 mouse antibody (Roche) (see below). Mixtures without PG were used as a control. The pellet fraction and the supernatant fraction of the reaction as well as the control (without PG) were quantified using the Li-Cor Odyssey Infrared Imaging System (Li-Cor; see below) and the value of precipitated protein in the control was subtracted from that in the reaction pellet. For comparability of datasets, the sum of the pellet fraction and the supernatant fraction of each reaction was set to 100%. Experiments were performed in duplicate and standard deviations were calculated.

Fluorescence microscopy
The surface accessibility of displayed SlhA_EGFP and SpaA_His 6 chimeras was analyzed directly by fluorescence microscopy and after immunofluorescence staining, respectively. P. alvei CCM 2051 T ∆slhA cells expressing both proteins from pSURF (see Table 1) were harvested at an optical density of OD 600 ~0.6 and washed three times with phosphate-buffered saline (PBS). After resuspension in 100 µl of PBS, 5 µl of a penta-His Alexa Fluor 532 conjugate (Qiagen) was used and incubation was done for 3 h at 25°C on a horizontal shaker. After washing for three times, the cells were resuspended in 100 µl of PBS analyzed by fluorescence microscopy. Three μl of that suspension were mixed with 3 μl of a solution containing 0.9% molten agarose, 40 mM Tris-HCl (pH 8.0), 20 mM acetic acid, 1 mM EDTA, pH 8.0, and preheated to 55°C. The samples were covered immediately with glass to immobilize the cells in the solidified agarose. The samples were monitored using a Nikon Eclipse TE2000-S fluorescence microscope at a magnification of 100 x using an oil immersion objective. In the case of the Alexa Fluor 532labelled SpaA-His 6 , the TRITC filter block and for SlhA-EGFP, the GFP filter block were used. Pictures were taken with a Nikon Digital Sight DS-Qi1Mc camera and the NIS-Elements 3.22 imaging software.

General and analytical methods
SDS-PAGE was carried out according to a standard protocol [30] using a Protean II electrophoresis apparatus (Bio-Rad). Protein bands were visualized with Coomassie Brilliant Blue G 250 staining reagent.
For Western-immunoblotting of proteins onto a polyvinylidene difluoride membrane (Bio-Rad) a Mini Trans-Blot Cell (Bio-Rad) was used. Detection of the His 6 -tag fused to SlhA-variants was done with the Li-Cor Odyssey Infrared Imaging System using anti-His 6 mouse antibody (Roche) in combination with goat anti-mouse IgGIR Dye 800CW conjugate (Li-Cor). The integrated intensity of the detected bands was determined using the Li-Cor Odyssey Application Software 3.0.21 applying automatic background subtraction. PGcontaining cell wall sacculi (PG (+) and PG(-)) were analyzed for their content of aminosugars and diaminopimelic acid (DAP) using a Biochrom 30 amino acid analyzer as described elsewhere [2].

Colony morphology, swarming and adhesion test
For imaging of the colony morphology of single cells of P. alvei CCM 2051 T wild-type, P. alvei ΔslhA, complemented mutant (P. alvei ΔslhA comp ) and P. alvei Δhag, pictures were taken using a SteREO Discovery. V12 (Zeiss) microscope and an AxioCam MRc5 camera (Zeiss) after growth for 48 h at 37°C on 2% LB-plates.
To compare the adhesion of P. alvei CCM 2051 T wild-type, P. alvei ΔslhA, P. alvei ΔslhA comp and P. alvei Δhag cells to nutrient plates, an assay based on the adhesion and agar invasion assay for yeast was used [31]. Cells were grown to OD 600 ~1.0 and 5 µl of the culture were spotted on a 2% LBagar plate (pH 7.0). Plates were incubated overnight at 37°C. The adhesion of the cells to the agar plates was tested by intense rinsing of the plates with distilled water using a wash bottle. Pictures were taken with a SPImager (S&P Robotics).
To test the different P. alvei variants for their ability to swarm on agar plates, cells were grown to the exponential growth phase (OD 600 ~1.0) and 5 µl of each culture were spotted on 0.4% (soft agar), 1% (semi-solid agar) and 1.5% (hard agar) LB-agar plates (pH 6.0). P. alvei cells without the pEXALV plasmid were cultivated at 37°C for 24 h, while strains containing pEXALV were incubated at 37°C for 48° h. Pictures were taken using a SPImager (S&P Robotics).

Congo red assay
To quantify EPS production in P. alvei CCM 2051 T wild-type and mutants, cells from 1 ml of an overnight-culture grown in LB medium were pelleted and washed once with 1 ml of LB broth. The pellet was resuspended in 1 ml of broth containing 40 µg/ml Congo red and incubated for 2 h at 37°C with shaking (250 rpm). Congo red bound to P. alvei cells was removed by centrifugation and unbound Congo red was determined by measuring the absorbance of the supernatant at 490 nm [32]. Data represent mean values + SD of at least four independent experiments and were analyzed by the unpaired Student's T Test. Asterisks indicate significant differences (*, P < 0.05; **, P < 0.01; ***, P < 0.001).
To detect EPS production on plates, an overnight-culture was diluted to OD 600 ~1.0 and 5 µl of that culture were spotted on 2% LB plates containing 40 µg/ml Congo red. Plates were incubated at 37°C for 24 h and 48 h at 25°C, red spots indicated binding of Congo red [32].

Transmission electron microscopy (TEM)
To identify potential differences in flagellation of P. alvei CCM 2051 T wild-type, P. alvei ΔslhA and P. alvei Δhag, cells were grown over-night in LB medium. 30 µl of these cell suspensions were applied to Formvar-and carbon-coated 300mesh copper grids (Agar Scientific) that were rendered hydrophilic upon glow discharge using a Pelco easiGlow apparatus (Ted Pella) for TEM. The grids were incubated for 10 min face down on the cell suspensions. Samples were fixed with 2.5% glutaraldehyde, washed three times with MilliQwater, and stained with 1% uranyl acetate solution, pH 4.2, for 40 s [33]. Samples were investigated using a Tecnai G 2 20 Twin transmission electron microscope (TEM; FEI), operating at 80 keV. Pictures were taken with an FEI Eagle 4 k CCD camera (4096×4096 pixels).

Microtitre plate assay
The ability of the P. alvei CCM 2051 T variants to form biofilm on an abiotic surface was evaluated according to Yegorenkova et al. [21]. Cultures were grown overnight in LB medium, diluted to an OD 600 ~0.025 with sterile nutrient medium, and 1ml aliquots were pipetted into 24-well polystyrene plates (STARLAB). The control received only the sterile nutrient medium. Cultures were incubated at 37°C for 72 h without shaking. Subsequently, planktonic cells were gently removed with a pipette and the plates were washed twice with MilliQwater. The OD 600 of planktonic cells was determined. Each well received 1 ml of 1% (w/v) Crystal violet and the plates were allowed to stand for 20 min at 25°C. Crystal violet solution was withdrawn with a pipette and the plates were carefully washed with MilliQ-water. Bound dye was released by adding 2.0 ml of acetone-ethanol (1:4) to the wells. Biofilm forming ability was evaluated by measuring the absorbance at 590 nm (A 590 ) of the solution using a Hitachi U-2001 spectrophotometer. Crystal violet values were normalized to the corresponding absorbance of the planktonic cells (A 590 /A 600 ). Data represent mean values + SD of at least four independent experiments with each four replicates and were analyzed by the unpaired Student's T Test. Asterisks indicate significant differences (*, P < 0.05; **, P < 0.01; ***, P < 0.001).

Confocal laser scanning microscopy (CLSM)
P. alvei CCM 2051 T variants were grown for 72 h in uncoated plastic dishes (µ-Dishes, 35 mm high; Ibidi). Prior to CLMS, the biofilm was washed with 1 ml of PBS buffer to remove planktonic cells and 1 ml of PBS buffer was added.
To visualize the cells of the biofilm, 1 ml of a solution of Hoechst 33258 (Sigma Aldrich) (10 ng/ml) was added to the biofilm and incubated in the dark at room temperature for 25 minutes. Images were taken on an inverted TCS-SP5 confocal microscope (Leica) at excitation of 356 nm and emission of 465 nm [34].

Results
To investigate the role of the C-terminal SLH-domains of SlhA for cell surface attachment, different C-terminal truncations of SlhA devoid of the SP, i.e., SlhA recombinant, SlhA 12 (lacking the SP and SLH domain three), SlhA 1 (lacking SLH domains two and three), and SlhA w/o SLH (lacking all three SLH domains) ( Figure 1A), were tested for their ability to bind to native PG-containing cell wall sacculi (PG(+)) of P. alvei CCM 2051 T ; which mimicked the native cell envelope.
The results from the in vitro binding assay as evidenced by densitometric quantification of Western-immunoblots indicated that one SLH domain of SlhA (as evidenced for the innermost SLH-domain) is sufficient for binding to native PG(+), with 88% retention of binding observed for SlhA-SLH 12 and decreased binding for SlhA-SLH 1 (~44%) in comparison to the recombinant full-length protein (SlhA recombinant) (compare with Table 3). The truncated S-layer protein devoid of all three SLH domains (SlhA-w/o SLH) was non-reactive with that material confirming the binding hypothesis (Figure 2A).
To test, if the binding of SlhA was specific for the SCWP associated with the native PG of P. alvei CCM 2051 T , the cell walls were treated with HF to release the covalently bound SCWP, yielding PG(-) cell envelope material [29]. Amino acid analysis confirmed that this material was completely SCWPfree [2]. Using that PG(-) material in the in vitro binding assay clearly showed that recombinant full-length SlhA and also SlhA_SLH 12 still showed ~25.0% and ~7.5% binding capability, respectively, to PG(-), whereas SlhA_SLH 1 showed only residual binding of ~1.8% and the truncated S-layer protein devoid of all three SLH domains was non-reactive with PG(-) ( Figure 2B, Table 3).
Summarizing, the in vitro experiments suggest that the innermost SLH-domain (SLH1) is sufficient for anchoring of the SlhA protein (corresponding to SlhA  ) to the native PG layer. As has already previously been reported for the S-layer glycoprotein SpaA, the SLH-domains of SlhA have dual recognition function, both for the SCWP and for the PG [2].

Localization of SlhA to the cell surface of P. alvei ∆slhA
To localize the SlhA protein on the cell surface of P. alvei ∆slhA in relation to the S-layer protein SpaA, which is known to completely cover the bacteria as a 2D crystalline array with oblique symmetry (10), a cell surface co-display system was constructed based on a chimera made of SlhA-EGFP and SpaA-His 6 .
P. alvei ∆slhA cells simultaneously expressing the two tagged-proteins were analyzed by fluorescence microscopy. The appearance of a continuous fluorescent halo around the cells indicated that SlhA is located on the outermost surface of P. alvei ∆slhA cells. SpaA-His 6 was detected by probing the cells with penta-His Alexa Fluor 532 conjugate (Figure 3

Knockout of slhA causes changes in colony morphology and bacterial adhesion to agar plates
To determine a putative function of the SlhA surface protein, an slhA knockout was created based on bacterial mobile group II intron-mediated gene disruption [26]. Reconstitution of P. alvei CCM 2051 T slhA::Ll.LtrB (P. alvei ΔslhA) was done by plasmid-based SlhA expression. The ΔslhA mutant clearly differed from the wild-type bacterium in colony morphology when grown on 2% LB-agar plates ( Figure 4A). The colonies of the SlhA knockout were non-adhesive, flat and frayed, while those of the wild-type, the complemented bacterium and a flagella knockout (P. alvei Δhag) were sticky, convex and circular ( Figure 4A). P. alvei Δhag was included in this study for comparison based on our assumption that SlhA might be involved in bacterial motility.
An adhesion test, in which P. alvei CCM 2051 T wild-type, P. alvei ΔslhA, P. alvei ΔslhA comp and P. alvei Δhag cultures were spotted onto agar plates, incubated over night at 37°C and washed thoroughly, clearly showed an increased adhesion capability of wild-type, P. alvei ΔslhA comp and P. alvei Δhag cells compared to P. alvei ΔslhA cells ( Figure 4B). Pictures taken immediately before the washing step showed identical spot sizes for all variants, with P. alvei ΔslhA being more transparent, possibly due to different colony morphology ( Figure 4B, upper panel). After the washing step, the P. alvei wild-type, P. alvei ΔslhA comp and P. alvei Δhag spots remained visible, while the P. alvei ΔslhA colony has been completely  Table 3. The figure represents one of at least two independent repeats of the experiment. removed ( Figure 4B, lower panel). This clearly indicated differences in the adhesion capability to 2% LB-agar plates of P. alvei CCM 2051 T wild-type, P. alvei ΔslhA comp and P. alvei Δhag in comparison to P. alvei ΔslhA.

Knock-out of slhA does not affect flagellation of P. alvei
Following up the effect of SlhA on bacterial motility, it was investigated by TEM, if P. alvei CCM 2051 T wild-type, P. alvei ΔslhA and P. alvei Δhag cells grown in LB broth showed differences in flagellation. To avoid shear forces, cell suspensions of the overnight cultures were directly used for TEM analysis. According to the TEM evidence on both P. alvei CCM 2051 T wild-type and P. alvei ΔslhA cells flagella are clearly visible, but are absent in P. alvei Δhag cells. Thus, there is no evidence that SlhA would be directly involved in the production or formation of flagella ( Figure 5).

P. alvei CCM 2051 T ΔslhA cells lose the ability to swarm on LB-agar plates
Since P. alvei cells are known to vigorously swarm on agar plates [13], we were interested to know, if swarming was changed in the ΔslhA mutant. Swarming on 0.4%, 1% and 1.5% LB-agar plates was completely abolished for P. alvei ΔslhA, while the complemented strain P. alvei ΔslhA comp was able to swarm as effectively as the wild-type bacterium ( Figure  6). Considering our observation of the complemented bacterium requiring 48 h instead of 24 h incubation time to swarm as efficiently as the P. alvei wild-type, it was tested, if the pEXALV expression vector had an influence on bacterial swarming. In fact, P. alvei carrying empty pEXALV vector needed longer (48 h) to swarm to the level of the vector-free wild-type bacterium.
Thus, to investigate, if P. alvei cells carrying pEXALV have a slower growth phenotype (extended lag-phase) compared to the wild-type, growth curves were recorded for the P. alvei variants. Comparison of the generation times showed almost identical values for all tested P. alvei variants (data not shown).
Since the expression vector pEXALV affects the lag-phase and, consequently, the swarming ability of P. alvei CCM 2051 T on agar plates (48 h versus 24 h), but not the generation time during exponential growth, loss of swarming motility can be interpreted as a direct consequence of knocking out slhA. Since also, as demonstrated for P. alvei ΔslhA, P. alvei Δhag was not able to swarm on 0.4%, 1% or 1.5% LB agar plates we assume that the flagella might be non-functional in the slhA knockout background under the used conditions ( Figure 6).
Thus, knocking out of slhA caused a significant decrease in biofilm formation, being 3.1-fold less for cells without pEXALV and 3.2-fold less for cells carrying pEXALV in comparison to the wild-type bacterium. Plasmid-based expression of SlhA complemented biofilm formation to wild-type levels. Knocking out of hag resulted in 12.5-fold less biofilm formation compared to the wild-type level according to the microtiter assay.
SEM and CLSM confirmed the results of the microtiter assay. For each technique, three biological replicates were analyzed. The SEM pictures taken after 72 h of biofilm development on glass slides revealed a biofilm with a higher cell density for the wild-type and P. alvei ΔslhA comp compared to P. alvei ΔslhA and P. alvei Δhag ( Figure 7B). The P. alvei ΔslhA biofilm showed a reduced content of slimy matrix components, probably due to reduced EPS formation.
Summarizing, the biofilm of the wild-type is ~6-fold thicker than that of ΔslhA and ~3-fold thicker than that of Δhag, according to CLSM analysis. CLSM analysis of ΔslhA comp revealed again a slightly increased biofilm formation compared to the microtiter assay. In conclusion, the wild-type and the complemented strain showed by far the best ability for biofilm formation according to three different techniques (microtiter assay, CLSM; SEM). The ΔslhA and Δhag mutant strains both showed significantly reduced biofilm formation.

Knock-out of slhA decreases Congo red-staining of P. alvei CCM 2051 T cells
Given that colonies of the P. alvei CCM 2051 T wild-type were found to be sticky compared to P. alvei ΔslhA, the composition of the extracellular matrix could have been altered in the slhA mutant.
P. alvei CCM 2051 T wild-type, P. alvei ΔslhA, P. alvei ΔslhA comp and P. alvei Δhag were tested for their EPS production by spotting the cells on Congo red (CR)/LB agar plates. CR accumulates on the colony surface over time only when EPS is produced [36]. Upon incubation of the plates for 24 h at 37°C followed by incubation for 48 h at 25°C, colonies of the wild-type, ΔslhA comp , and Δhag strain had a reddish appearance, indicative of the presence of CR-reactive EPS on either variant. In comparison, spotted ΔslhA cells showed decreased adsorption of CR ( Figure 8A). Quantification of EPS production of planktonic cells showed that the wild-type bound 9.2 +2.3 µg CR/OD 600 , while ΔslhA bound 4.5 +1.7 µg CR/ OD 600 and ΔslhA comp bound 6.9 +1.7 µg CR/OD 600 ( Figure 6B). CR binding ability of Δhag of 9.0 +2.8 µg compared well to that of the wild-type ( Figure 8B). The value of 49% of CR binding of ΔslhA, in relation to 97% CR binding of Δhag compared to the wild-type bacterium, might be indicative of altered EPS production in the ΔslhA mutant.

Discussion
Gram-positive bacteria synthesize a thick cell wall, which serves as an assembly scaffold for the surface display of polypeptides, capsular polymers, and wall teichoic acids [37]. These cell surface molecules strongly influence the physicochemical and functional properties of bacterial cells and represent an important interaction zone with the environment. For cell surface display, many proteins are endowed with terminal SLH-domains that mediate non-covalent surface attachment via interaction with pyruvylated moieties of a SCWP [4,8,9,[38][39][40][41][42][43]. For this interaction, especially the TRAE motif that is located at the beginning of the α-helix within these domains has been reported to be critical [1,2,8]. Functional variations of that four amino acid motif have been shown for T. thermosulfurigenes EM1 [8] and P. alvei CCM 2051 T , where the motifs TVEE and TRAQ are involved in anchoring of the S-layer protein SpaA [2], and the motifs SRGE, VRQD, and LRGD are likely to be critical for anchoring of SlhA (this study). In contrast to the requirement of three functional SLH-domains as SCWP targeting modules in the cell envelope of T. thermosulfurigenes EM1 [8] or Bacillus anthracis [9], and four in Lysinibacillus sphaericus CCM 2177 [44], P. alvei CCM 2051 T has obviously elaborated a variation of that surface display mechanism. We have shown recently that for surface display of the S-layer protein of that bacterium, two functional SLH domains are sufficient [2]; for the SlhA protein characterized in the course of this study, even one SLHdomain (as tested for the innermost SLH-domain, SLH1) can mediate attachment in vitro, with a binding capacity to PG(+) of still 38.8 ±13.8%. This reduced requirement of SLH-domain/ SCWP interactions in P. alvei CCM 2051 T might be compensated by the simultaneously occurring PG/SCWP interaction which could be clearly demonstrated for the two investigated P. alvei CCM 2051 T proteins SpaA [2] and SlhA (this study). Comparison of the data from in vitro binding assays for the said P. alvei proteins indicated that binding to PG(-) was still detected for SlhA lacking the third SLH domain (SLH 12 ; 7.5 ±7.3%), and even for the SlhA variant additionally lacking the second SLH-domain weak binding was possible (SLH 1 ; 1.8 ±2.5%). In contrast, for the S-layer protein SpaA no binding to PG(-) could be detected for neither the single, nor the double or the triple SLH mutant [2]. Thus, it seemed as if SLH-domain 3 of SlhA played a minor role in the binding to PG. This might, on the other hand, enable the tri-amino acid motif RGD present in that SLH-domain to be involved in host-cell integrin binding [45].
Considering the full coverage of the P. alvei CCM 2051 T cell surface with a closed 2D crystalline SpaA S-layer lattice with oblique symmetry [26], it was important to verify the predicted cell surface localization of SlhA in vivo. This was approached by constructing a cell surface co-display system comprised of an SlhA-EGFP chimera and the His 6 -tagged S-layer protein SpaA. This strategy was based on the previous demonstration that His 6 -tagged S-layer protein gets integrated into the native SpaA-glycoprotein lattice in vivo and, thus, surface exposed [10]. The fluorescence signals obtained directly from SlhA_EGFP as well as from SpaA_His 6 after immunofluorescence staining clearly demonstrated surface codisplay of the proteins on P. alvei CCM 2051 T cells (compare with Figure 3). For this experiment, P. alvei CCM 2051 T ΔslhA cells were used because of improved transformation efficiency in comparison to the wild-type cells. This is a reliable approach, since it is known that deletion of slhA does not affect localization of SpaA (B. Janesch, unpublished data).  While it has been assumed previously that in S-layercovered bacteria the S-layer would be the exclusive (glyco) proteinaceous layer on the cell surface, these data corroborate recent findings with Bacillus anthracis [46] and Bacillus cereus G9241 [47], where in addition to the S-layer, S-layerassociated proteins were found (BSLs; Bacillus S-layer associated proteins). While both S-layer proteins of B. anthracis and BSLs bind the same SCWP, their deposition on the cell surface was found to be clearly confined. For example, the murein hydrolase BslO is targeted to septal PG zones, where it catalyzes the separation of daughter cells, while Sap S-layer patches were not observed at the septa [46]. Regarding the co-display of SlhA and SpaA on P. alvei CCM 2051 T cells, the distribution of SlhA over the cell surface has not been investigated in detail. However, due to the uniform fluorescence signal of the SlhA_EGFP chimera, it is conceivable that it would be regularly integrated in or associated with the S-layer lattice (Figure 3).
To unravel a putative function of the novel SlhA surface protein of P. alvei CCM 2051 T , a ΔslhA mutant was created and its phenotype as well as distinct cell surface mediated phenomena, such as swarming motility and biofilm formation, were compared to the wild-type bacterium. According to our data SlhA is not essential for the bacterium but affects its lifestyle. This mutant allowed, for the first time, investigating the functional relevance of a predicted cell surface protein in that bacterium. In contrast, it has been impossible to inactivate spaA (B. Janesch, unpublished data) implying that the S-layer is essential for cell envelope integrity of P. alvei CCM 2051 T . Since flagella are considered a prerequisite for bacterial swarming, a flagella mutant (Δhag) was created for phenotype comparison [18,48]. The first obvious difference between the wild-type and the ΔslhA mutant was a significant change in colony morphology from a sticky, convex and circular type into a non-adhesive, flat and frayed type (compare with Figure 4A), while colonies of Δhag showed wild-type morphology. Loss of stickiness of ΔslhA could be confirmed in an agar adhesion assay ( Figure 4B). Flagellation was visualized by TEM using negatively-stained, intact wild-type, ΔslhA and Δhag cells that were treated very carefully to avoid damage of flagella. Peritrichous flagella could be detected for wild-type and ΔslhA cells ( Figure 5). As expected, for Δhag, only cells without flagellation were found, however on some cells, remnants of a hook were visible ( Figure 5, arrow). In addition, the swarming motility on 0.4%, 1% and 1.5% LB agar plates as typical of P. alvei cells [11,12] was abolished in either mutant ( Figure 6). A lack of surfactant for reducing the surface tension between the substrate and the bacterial cell, thereby permitting spreading over surfaces, comparable to e.g., rhamnolipid of Pseudomonas or surfactin of Bacillus species, might be an explanation for abolished swarming but not for a lack in swimming ability [16,49]. Based on the findings of Cohen et al. [13], swimming should be possible at agar concentrations below 0.5%. Since we could not detect any movement for ΔslhA and Δhag, neither swimming nor swarming, at the tested agar concentration, it is conceivable that non-functional flagella and not a missing surfactant cause the lack in motility. It is important to note that, while no surfactant has been reported for P. alvei CCM 2051 T so far, surfactants have been found in other P. alvei strains. P. alvei LMD 50.12 is known to produce a layer of lubricating fluid for movement on hard surfaces [13] and in P. alvei ARN63, a biosurfactant, characterized as lipopeptide derivative, lowers the surface tension of media [50]. The presence of an ortholog of the sfp gene of Bacillus subtilis required for production of the lipoheptapeptide surfactin, on the genome of P. alvei CCM 2051 T (identity: 42%) [51], might be a further indication of a surfactant in this bacterium.
Knocking-out of slhA and hag translated into P. alvei CCM 2051 T cells with an impaired biofilm life-style according to Crystal violet assay [15,17,18]. While the ΔslhA mutant was 3.1-fold less efficient in biofilm formation compared to the wildtype, biofilm formation could be fully complemented by plasmid-encoded SlhA expression (Figure 7), indicating that SlhA directly correlated with biofilm formation without potential polar effects of the mutant additionally being involved. The Δhag mutant showed 12.5-fold less efficiency to form biofilm compared to the wild-type indicating the importance of functional flagella for biofilm formation in P. alvei CCM 2051 T . Crystal violet data were strongly supported by SEM and CLSM analyses. SEM images of wild-type and ΔslhA comp showed a thick biofilm with cells embedded in a slimy matrix. In contrast, P. alvei ΔslhA and Δhag showed a flat and thin biofilm, with the cell-covering material missing in ΔslhA, but not in Δhag. CLSM confirmed reduced biofilm formation for ΔslhA and Δhag compared to the wild-type, with wild-type and ΔslhA comp forming tower-like structures, while the ΔslhA and Δhag formed thin carpet-like biofilms.
Since the EPS matrix is generally an important factor in biofilm formation, P. alvei CCM 2051 T wild-type, ΔslhA, Δhag, and ΔslhA comp cells were analyzed for their EPS production upon spotting on LB plates containing CR in comparison to the respective planktonic cells ( Figure 8A). The adsorption of CR has previously been shown to have a positive correlation with the presence of EPS in Pseudomonas aeruginosa [52,53], Salmonella enterica serovar Typhimurium, and E. coli [54]. The ΔslhA mutant showed only half (49%) of EPS production compared to the wild-type ( Figure 8B). This might also affect the loss of stickiness of the ΔslhA colonies. Lauriano et al. showed that exopolysaccharide expression in Vibrio cholerae is regulated along a flagellum-dependent pathway [55]. They identified a sodium-driven flagellar motor as an essential component of the signaling pathway, because mutations in the motor abolished EPS production, biofilm formation and vps gene transcription. Cairns et al. recently showed that the level of DegU~P gets increased via the sensor kinase DegS due to an inhibition of flagellar rotation, which is an essential regulatory mechanism required for biofilm formation in Bacillus subtilis thereby supporting the importance of the flagellum for signal transduction purposes [56]. We hypothesize that SlhA might be involved in signal transduction based on its demonstrated effect on the motility, EPS production, and biofilm formation of P. alvei CCM 2051 T cells (this study) in conjunction with a missing link in the requirement of functional flagella for signal transduction and biofilm formation as proposed by others (see above). While there have so far been no reports on SLH domain-containing proteins (such as SlhA) functioning as receptors or their involvement in signaling cascades within the bacterial cell that could support this hypothesis, the question of how a missing or decreased flagellar rotation is sensed by the cell is still open. Considering that SlhA contains, in addition to the C-terminal SLH-domains, a galactose-binding module (CBM6; compare with Figure 1) typical of proteins binding to specific ligands, the option of SlhA functioning as a receptor remains possible. Since P. alvei Δhag showed abolished biofilm formation, despite production of EPS at wild-type level, we concluded that functional flagella are important for P. alvei CCM 2051 T biofilm formation. One might speculate that SlhA is involved in the signal transduction that leads to the activation of flagellation. A defect in flagellation might then lead to decreased EPS expression and reduced biofilm formation. Thus, in the future, it will be interesting to investigate the detailed composition of the EPS matrix and, especially, if SlhA is part of the biofilm matrix. There are several examples of EPS matrices known that harbor adhesive proteins. For instance, the Staphylococcus aureus matrix harbors biofilm-associated proteins that are anchored to the cell wall and serve to hold cells together, Bacillus subtilis synthesizes the EPS protein TasA and E. coli synthesizes the curli protein, with both proteins being critical for biofilm formation. Further, in Pseudomonas aeruginosa several surface proteins contribute to biofilm formation. Additionally lectin-binding proteins that facilitate cell-matrix or cell-cell interactions within the biofilms have been found [19].
Besides the effects of P. alvei CCM 2051 T SlhA on colony morphology, bacterial swarming, and biofilm life-style, several other functions remain open. One obvious function would be that of a surface-displayed enzyme; examples are known for EPS biofilm matrices [57]. Thus, based on some homologies of SlhA to putative amylo-pullulanases, α-amylases or pullulanases according to a BLAST search, we tested recombinant SlhA for the ability to hydrolyze starch and pullulan [58], however, without success. To extend the substrate range, P. alvei wild-type and P. alvei ΔslhA were tested for differences in their carbohydrate metabolism using an API ® 50 CH test (bioMérieux), but no differences were detected either. Thus, the question, if SlhA exerts also an enzymatic activity still remains open.
Summarizing, we showed in this study that SlhA is a novel, SLH-domain containing cell surface protein of P. alvei CCM 2051 T that is co-displayed with the S-layer lattice. For cell-envelope anchoring of SlhA in vitro, the innermost SLHdomain is sufficient. It is conceivable that the motifs SRGE, VRQD, and LRGD located in the SLH-domains 1 to 3 are critical for anchoring. The SLH-domains recognize both a pyruvylated SCWP and the PG. Knocking-out of SlhA causes drastic phenotypic changes concerning colony morphology, swarming motility on agar plates, and biofilm formation. In contrast, a mutation that disrupted flagella synthesis did neither show altered EPS production nor colony morphology but impaired biofilm formation. We conclude that SlhA possibly plays an important, still unknown role in signal transduction that leads to EPS expression. Further, a complex relationship between motility, EPS production, and biofilm formation seems to exist.