Materials and Methods

Chemicals and materials

All chemicals and materials used in the present study are listed in table 1.

Table 1: Chemicals and materials used in the present study

The plasmids, bacterial strains and primers used in this study are listed in tables 2, 3, 4 respectively.

Table 2: Plasmids used in the present study

^a The reference is omitted in case the plasmids were prepared for this study.

Table 3: Bacterial strains used in the present study

Table 4: Primers used in this study

FT, DNAse I footprinting; PE, primer extension; Pbmy, promoter of the bmy operon; PyczE, promoter of yczE; n.r. EMSA, non-radioactive EMSA; the enzyme recognition site within each primer is underlined.

2.3 Media and supplements

All media used in this work were prepared and sterilized according to [198, 199]. Supplements with different antibiotics and compounds are listed in table 5. For antibiotic production and mass spectrometry measurements, bacteria were grown either in Landy medium [200] or sucrose-ammonium citrate medium (ACS) [201].

Fungi were grown on “potato agar” at room temperature. When fungi and bacteria had to be simultaneously grown on plates, then Waksman agar was used and the microorganisms were let to grow at 27°C.

Table 5: Supplements

Standard molecular biology methods

DNA manipulation, such as digestion with restriction endonucleases and ligation, was performed according to the instructions supplied by the manufacturer. Agarose-gel-electrophoresis, fluorescent visualization of DNA with ethidium bromide, spectrophotometric quantitation of DNA as well as preparation of CaCl₂-competent E. coli cells followed by transformation of plasmid DNA were carried out with standard procedures described by [198]. Bacterial chromosomal DNA from Bacilli was prepared as described by [202]. Polymerase chain reaction (PCR) was done using the GeneAmp PCR system 2700 (Applied Biosciences) according to [203], under the appropriate conditions in each case. Ligation of PCR products to pGEM-T vector was carried out following the instructions of the manufacturer (Promega). Plasmid DNA isolation and recovery of DNA from agarose gels were performed with QIAprep Spin mini prep kit and QIAEX II gel extraction kit, respectively.

Competent cells of Bacillus subtilis were prepared according to the protocol published by [204]. Cells were grown overnight in 10 ml KM1 buffer at 32°C on a rotary shaker (150 rpm) and the next morning were 1:10 diluted in 50 ml KM1 buffer. Cells were further grown at 37°C under vigorous shaking (175 rpm). Every 30 minutes, samples were collected and the optical density at 600 nm was determined. At the beginning of stationary phase the culture was diluted 1:10 in 100 ml KM2 buffer and was further incubated at 37°C for 75 minutes (75 rpm). Subsequently the cells were harvested by a 10 minute centrifugation at 5000 rpm (room temperature) and the pellet was resuspended in 2 ml of the supernatant. Aliquots of 0,5 ml competent cells with 10% glycerol were stored at -80°C.

For the transformation, one aliquot was unfrozen by short incubation at 37°C. 1 μg of the desired DNA (chromosomal or linearized/circular plasmid DNA) was added and cells were incubated at 37°C for 30 minutes (50 rpm). Subsequently, 0,5 ml of LB medium, containing inducing concentration (0,1 μg/ml) of the appropriate antibiotic was added to the cells and they were further grown at 37°C for 75 minutes (200 rpm). Aliquots of the culture were plated on selective agar plates.

Buffers

Competent cells of Bacillus amyloliquefaciens were obtained by modifying the two-step protocol published by [205]. Cells were grown overnight in LB medium at 28°C (170 rpm). The next day, they were diluted in glucose-casein hydrolysate-potassium phosphate (GCHE) buffer to an OD₆₀₀ of 0,3. The cell culture was then incubated at 37°C under vigorous shaking (200 rpm) until the middle of exponential growth (OD₆₀₀ ~1,4). Dilution with an equal volume of GC medium followed and the cells were further incubated under the same conditions for 1 hour. Further on, the culture was divided in 2 ml Eppendorf tubes and cells were harvested by centrifugation at 6000 rpm for 5 minutes (room temperature). The pellets were resuspended in 200 μl of the supernatant and the desired DNA (1 μg) with 2 ml transformation buffer was added to them. After incubation at 37°C under shaking at 75 rpm for 20 minutes, 1 ml LB medium containing sublethal concentration (0,1 μg/ml) of the appropriate antibiotic was added. The cells were grown under vigorous shaking for 90 minutes and platted on selective agar plates.

Buffers

Suppression Subtractive Hybridization (SSH) is applied to two strains of the same species or genus and aims to find major sequence differences between them. SSH identifies unique DNA sequences of target strain (tester) that are absent from the reference strain (driver). The method was performed according to the protocol published by [206] and [207].

In principal, genomic DNA from two strains was digested separately with RsaI, yielding fragments of 100 to 1000 bp. The tester DNA was subdivided into two portions, each of which was ligated with a different adaptor (1 and 2R, see their sequence at the end of this section). The ends of the adaptors are unphosphorylated and thus only one strand of each adaptor attaches to the 5' end of the DNAs. At first ligation of the adaptor to the fragments of the tester strain was performed for 16 hours at room temperature. The mixture was then heated at 72°C for 5 min in order to inactivate the ligase.

Subsequently two hybridizations were performed. In the first one, excess of driver DNA was added to each adaptor-ligated lot separately. After denaturation of the two mixtures at 98°C for 2 minutes, the samples were allowed to anneal at 63°C for 90 minutes thus generating type a, b, c, d molecules (Fig. 10). During the second hybridization, the two primary hybridization samples are mixed together without denaturation in order to assure that the remaining single-stranded tester-specific DNAs can form the new type e molecules. The e molecules are double-stranded tester-specific DNAs with different ends that have resulted from ligation with different adaptors. Fresh denaturated driver is added to the mix to further enrich fraction e for tester-specific sequences. The samples were allowed to anneal at 63°C for 16 hours.

The entire population of molecules was then subjected to PCR. First the reaction mix was incubated in the thermal cycler at 72°C for 2 min in order for the adaptors to be extended (their recessed 3' ends were filled in during this step). PCR was performed using primers ssh1/ssh2 (Table 4), that annealed on the adaptors 1 and 2R respectively [T_den=94°C (30 sec), T_anneal=66°C (30 sec), T_ext=72°C (30 sec) for 35 cycles]. During the PCR, molecules a and d were missing the primer-annealing sites while type b molecules formed a panhandle-like structure. As a result, these three types of molecules could not be amplified. Type c molecules had only one primer annealing site and were thus amplified linearly. Due to the suppression PCR effect only type e molecules that had two different adaptors and contained tester-specific sequences could be exponentially amplified. The substracted DNAs were cloned into pGEM-T vector and sequenced.

The adaptors’ sequences are given below.

Adaptor 1

5'-CTAATACGACTCACTATAGGGCTCGAGCGGCCGCCCGGGCAGGT-3'

3'-GGCCCGTCCA-5'

Adaptor 2R

5'-CTAATACGACTCACTATAGGGCAGCGTGGTCGCGGCCGAGGT-3'

3'-GCCGGCTCCA-5'

•Hybridization buffer

	Figure 10: Schematic diagram of Suppression Subtractive Hybridization.
	Tester DNA fragments that are ligated with adaptors 1 and 2R separately are further hybridized separately with excess of driver DNA. The samples are mixed together without denaturation and are hybridized in the presence of fresh denaturated driver. After the second hybridization, the PCR mixture is incubated at 72°C for 2 minutes in order for the recessed 3' ends of the adaptors to be filled in. Type a, b and d molecules cannot be amplified, due to lack of primer annealing site (a and d) and to formation of a panhandle-like structure (b).Type c molecules have only one primer annealing site and thus are amplified linearly. Type e molecules are amplified exponential only if the sequence is present in the tester strain but absent from the driver strain. Solid lines stand for RsaI digested DNAs. Filled boxes represent the outer identical parts of adaptors 1 and 2R. Clear and shaded boxes indicate the inner parts of adaptors 1 and 2R, respectively and correspond to the sequence of primers ssh1, ssh2. The figure is reproduced from [207].

Pulsed Field Gel Electrophoresis allows size separation of DNA fragments ranging from a few kilobase pairs to 10 megabase pairs. It operates by applying electric fields from different angles, thus making even very large DNA fragments to move through the gel and be efficiently separated. DNA is embedded in agarose in order to prevent shearing during purification.

Mid-exponential-phase Bacillus cells were used to prepare DNA for the PFGE [208]. After centrifugation for 10 minutes at 4°C and 4000 rpm, the pellet was resuspended in wash buffer. Plugs of cell suspensions prepared by mixing with 1% low-melting point agarose were first incubated overnight with lysis buffer at 56°C and were subsequently incubated overnight with digestion buffer at 50°C. After incubation with 1 x TE buffer containing 100 μM PMSF for 1 hour at 37°C, the plugs were stored at 4°C in 1 x TE buffer. Digestion was performed overnight using SfiI according to the instructions of the manufacturer and was stopped by addition of stop buffer. The plugs were loaded on a 1,2 % agarose gel in TBE buffer and PFGE was performed at 10°C using the Gene navigator electrophoresis unit of Pharmacia Biotech. Direction of the applied electric fields (160V) changed every 2 seconds for the first 2 hours of the run, every 10 seconds for the next 8 hours, every 25 seconds for the next 8,5 hours and every 40 seconds for the last 6 hours of the run. Visualization of the DNA was performed with ethidium bromide.

Buffers

Southern blot is a way of permanently immobilizing DNA (that has been separated by agarose gel electrophoresis) to a solid support. It is designed to locate a particular sequence of DNA within a complex mixture, such as an entire genome. Hybridization and detection occurs by “anealling” with a complementary labelled DNA probe.

Synthesis of DIG-labelled probe

For each southern hybridization, an appropriate probe was labelled with Digoxigenin-11-dUTP (DIG-dUTP), according to the Ready-to-Go kit from Roche. The desired DNA region was amplified by PCR and purified, prior to labelling. 100 ng of the PCR fragment were denaturated by heating at 100°C for 10 minutes and then mixed with 5 μl dCTP (10 mM), 2,5 μl DIG-dUTP (1mM) to a final volume of 50 μl. The mixture was incubated at 37°C for 1,5 hours and was stored at -20°C until use.

1-2 μg of the chromosomal DNA in question were digested overnight with a suitable restriction endonuclease. Samples were initially separated on a 0,8 % agarose gel in 1 x TAE buffer at 70 Volt. The gel was washed twice for 20 minutes, initially with denaturation buffer and subsequently with neutralization buffer. Transfer on a nylon membrane was performed using the Biorad vacuum blotter (model 785). The DNA was fixed permanently on the membrane by cross-linking using UV radiation.

Buffers

The membrane was initially incubated for 1 hour at 65°C with 40 ml hybridization buffer and was hybridized overnight at 55°C with 5-10 ml hybridization buffer containing 5-25 ng/ml of denaturated DIG-labelled probe. The membrane was washed twice for 15 minutes, first with 2 x SSC/0,1 % SDS at room temperature and then with 0,5 x SSC/0,1 % SDS at 55°C.

Detection was achieved by a colorimetric approach. The membrane was first equilibrated with P1-DIG buffer and was then incubated for 30 minutes with P1-DIG buffer containing 3,75 units of the antibody Anti-Digoxigenin-Alkaline-Phosphatase. Unbound antibody was removed after a fifteen minute washing step. Addition of 10 ml Ap buffer containing 2,25 mg nitroblue tetrazolium salt (NBT) and 1,75 mg 5-bromo-4-chloro-3 –indolyl phosphate (BCIP) to the membrane and incubation in the dark allowed visualization of the hybridized DNA with our labelled probe.

Buffers

Samples from primer extension, DNAse I footprinting and sequencing reactions were analysed on denaturating sequencing gels. High concentrations of urea in the gel secured that the DNA was completely denaturated and thus could be better separated.

The gel was let to prerun before loading the samples for 1 hour at 60 Watt in 1x TBE buffer, using the SequinGen Sequencing Cell of Biorad. After loading the samples, DNA separation was allowed for approximately 2 hours more using the same running conditions. The gel was dried at 80°C for 1 hour using the vacuum SlaB Gel Dryer Model SE1160. An IP screen was put on the top of the dried gel and visualization was achieved using the Molecular Imager FX scanner (Biorad) or the phosphoimager 445SI (Molecular Dynamics).

Oligonucleotides can be radio-labelled at their 5'-OH end by the T4 Polynucleotide kinase (T4 PNK) that catalyses the transfer of the γ-phosphate from ³²P- ATP.

Therefore, 40 pmol of primer were mixed with 4 μl [γ-³²P]ATP (10μCi/ml) and phosphorylation took place by incubation of the mixture with T4-Kinase at 37°C for 30 minutes. The reaction was stopped by heat inactivation at 70°C for 10 minutes.

Sequencing reactions were carried out using the Thermo Sequenase cycle Sequencing kit (USB) according to the manufacturer’s instructions. 300 ng of plasmid DNA containing the desired fragment and 1 pmol of the radioactive primer were included in the reaction. Amplification was performed using a 23 cycle PCR program [T_den=94°C (30 sec), T_anneal=58°C (sec), T_ext=72°C (30 sec)]

Stationary-phase cells of Bacillus amyloliquefaciens were harvested for preparation of total RNA. 20 ml of the culture was mixed with 10 ml “killing” buffer (stops mRNA production) and centrifuged for 10 minutes at 4°C and 12000 rpm. The pellet was washed once more with 1 ml “killing” buffer and was then stored at -80°C.

Isolation of RNA was performed using the Nucleo Spin RNA L (Macherey Nagel). In order to remove possible DNA contaminations, the isolated RNA was additionally extracted with an acidic Phenol:Chloform:Isoamylalcohol (25:24:1) mixture and then chloroform. Ethanol precipitation followed and the pellet was resuspended in 20 μl DEPC-H₂O. The concentration of total RNA was spectrophotometrically determined, according to [198] whereas its quality was checked on a 1,5% RNA agarose gel under denaturating conditions (1xMEN, 16% formaldehyde). The samples were mixed with 1,6 volume loading buffer and were incubated at 65°C for 5 minutes prior to loading on the gel. The gel was run in 1 x MEN buffer at 60 Volt.

Buffers

Primer extension was used to map the 5' termini of mRNAs. 40 μg of total RNA was mixed with 0,15 μM radioactively (³²P) labelled primer at 70°C for 5 minutes. Then 4 μl 5 x reverse transcriptase buffer, 2 μl dNTPS (10 mM each) and 1 μl Ribonuclease inhibitor (40 units) were added to a final volume of 19 μl. After incubation at 37°C for 5 minutes, 1 μl reverse transcriptase (200 units) was added to the mixture and further incubation was allowed for 1 hour at 42°C. The primers used for identifying the transcriptional start(s) of bmy operon and yczE can be seen in table 4.

Electrophoretic Mobility Shift Assay (EMSA) is a technique used for determining protein-DNA interactions. It is based on the observation that DNA-protein complexes migrate slower through a non-denaturating polyacrylamide gel than free DNA fragments. Therefore, EMSA is a useful tool to determine if a protein binds directly to a DNA fragment or not.

In our case, the desired DNA fragment of the bmyD promoter region was amplified by PCR using primers bmyD1 and rev1 (Table 4), one of which was previously labelled at its 5'-end with [γ-³²P]ATP. The radio-labelled product (450bp) was purified with the Qiagen PCR purification kit. After dilution of the labelled DNA fragment to attain final activity of 10.000 cpm, the DNA was incubated at 37°C for 20 minutes with increasing concentrations of DegU protein in the 1xbinding buffer. The reaction mixtures were separated on 8% polyacrylamide gels, under non-denaturating conditions, in 1 x TBE buffer at 60 V. The gels were visualized using the Biorad Molecular Imager FX scanner.

Non radioactive EMSA experiments were performed in a similar manner, but visualization was done fluorescently using ethidium bromide. In particular, two smaller DNA fragments were amplified using primers bmyD1 / bmyD6 and bmyD2 / rev1. The obtained fragments, D1 (217 bp) and D2 (233 bp) respectively, result together in the whole 450 bp fragment used in the radioactive EMSA.

DNase I footprinting is a method of studying protein-DNA interactions and identifying the DNA region to which a protein binds. These experiments were done as described by [209]. A DNA fragment (450 bp) obtained by PCR, using primers bmyD1 and rev1, or plasmid DNA carrying the same fragment (pAK16), were incubated in binding buffer with 0, 0,8 and 1,6 μM DegU protein for 20 min at 37°C. Complexes were then treated with DNase I (0,6 μg/ml) for 20 seconds and the reaction was stopped by addition of 10 μl stop buffer containing 1,75 ng/μl non-specific DNA (salmon sperm) and rapid chilling on ice. Primer extension followed with ³²P-labelled primers bmyD1 and rev1, for the template strand and for the non template strand, respectively [T_den=94°C (30 sec), T_anneal=58°C (30 sec), T_ext=72°C (30 sec) for 23 cycles].

•Stop buffer (10 μl)

For the antifungal tests, B. amyloliquefaciens FZB42 and its derivatives were grown in Landy medium at 37°C for 24 hours. The cultures were centrifuged and 2 μl of the supernant were spotted on Waksman agar together with regularly arranged growing fungi. The plates were incubated at 27°C.

For the antibacterial tests, B. amyloliquefaciens FZB42 and its derivatives were grown treated in the same manner. The indicator strain was grown overnight at 37°C under vigorous shaking. 300 μl of the culture was mixed with 3 ml soft agar and poured on LB dishes. Supernatants obtained from the B. amyloliquefaciens FZB42 strains, grown in Landy medium for 24 hours, were applied on the plates and were incubated at 37°C.

MS analysis

B. amyloliquefaciens FZB42 was grown overnight on agar plates of Landy medium at 37°C. To record mass spectra, cell material was picked from the plate, spotted onto the target and covered with matrix medium, i.e. a saturated solution of α-cyanocinnamic acid in 40% acetonitrile-0,1% trifluoroacetic acid. It was air dried and then analyzed by matrix-assisted laser desorption/ionization-time of flight (MALDI-TOF) mass spectrometry (MS), as described in [210].

Alternatively, culture filtrate extracts were prepared by lyophilisation of the supernatants that resulted from cultures grown for 12, 24 or 48 hours at 37°C in Landy or ACS medium. A small sample of the culture filtrate was extracted with 70% acetonitrile-0,1% trifluoroacetic acid and then mixed with an equal volume of matrix medium. 1 μl aliquots were spotted on the target and were air dried prior to MS measurement [211]. Postsource decay (PSD) mass spectra were obtained with the same samples. Monoisotopic mass numbers were recorded.

Specific β-galactosidase activity was determined from growing liquid cultures in Difco medium, according to [199]. At different times of the growth curve the optical density of the culture at 600nm was determined and cells were harvested. Their pellets were frozen in order to be further used in the β-galactosidase assay. Pellets were resuspended in 640 μl Z-buffer and mixed with 160 μl lysozyme-buffer. After short vortexing, they were incubated at 37°C for 10 min. Further on, 8 μl of a 10% Triton-X solution was added to the samples, followed by ten-minute incubation on ice. The reaction began by addition of 200 μl ortho-nitrophenyl-β-D-galactopyranoside (ONPG) 4 mg/ml at 30°C and was stopped by addition of 400 μl 1M Na₂CO₃ when their colour changed to yellow. The samples were then centrifuged for 5 min and the supernatant’s absorbance was measured at 420 and 550 nm. Specific β-galactosidase activity was calculated in Miller units (MU) [212], according to the formula.

Miller units (MU) = 1000 x (OD ₄₂₀ – 1,75 x OD ₅₅₀ ) / (t x V x OD ₆₀₀ )

OD₄₂₀, OD₅₅₀, OD₆₀₀=optical density at 420, 550, 600 nm

T=reaction time (min)

V=volume of the sample of bacterial cells used for the reaction (ml)

Buffers

Sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis was performed according to [198]. The proteins were separated by 4% upper and 10% lower gels, using the “Mini-Protean II” apparatus of Biorad. Gels were run at 200 Volt in 1 x running buffer and were stained with Coomasie Brilliant Blue.

•1 x running buffer

The gels were electroblotted onto nitrocellulose membranes (0,45 μm pore size) using the “Trans-Blot SD Semi-Dry transfer cell” (Biorad). After incubation for 15 min with TBST buffer containing 10% powdered milk, the membranes were incubated overnight at 4°C with TBST buffer containing 1% powdered milk and 20 μl antibody “His-probe H15 sc-803 rabbit polyclonal IgG” (200 mg/ml; Santa Cruz). The membranes were washed 3 times with TBST buffer for 10 minutes and were then incubated with TBST buffer containing 2% powdered milk and 6 μl of the “Anti-rabbit IgG alkaline phoshatase conjugate” antibody (dilution 1:10.000) for 1 hour. Detection was achieved colorimetricly by addition of 10 ml Ap buffer containing 2,25 mg nitroblue tetrazolium salt (NBT) and 1,75 mg 5-bromo-4-chloro-3 –indolyl phosphate (BCIP).

•Transfer buffer

A DNA fragment containing the whole degU gene of B. amyloliquefaciens FZB42 was amplified using primers degU60a / degU60b (table 4), that contain restriction sites for the endonucleases NcoI and BglII respectively. DegU was cloned in the expression plasmid pQE60 (Qiagen) as a C-terminal His₆-tag fusion under regulation of an isopropyl β-D-thiogalactoside (IPTG) –inducible promoter. The resulting plasmid, pAK54, was used to overexpress DegU-His₆ fusion protein in E. coli. For this scope, DH5α strain was simultaneously transformed with pREP4 (a repressor plasmid carrying the lacI repressor) and pAK54 resulting in strain AK38.

Strain AK38 was grown overnight at 37°C in LB medium containing 100 μg/ml ampicillin and 20 μg/ml kanamycin. The culture was diluted in 500 ml LB-Ap/Km to an OD₆₀₀ of 0,03 and was further grown at 30°C under vigorous shaking. When the cells grew to an OD₆₀₀ of 1, ethanol was added to a final concentration of 3% to induce chaperone synthesis and minimize formation of inclusion bodies. After 15 minutes, IPTG was added to the cultures at 1 mM final concentration. The cultures were grown for 2,5 hours and were then centrifuged at 6000 rpm for 20 minutes. The pellets were stored at -80°C. The protein was then purified with the Protino^® Ni-1000 kit according to the manufacture’s instructions (Macherey Nagel) and was subsequently dialysed overnight against storage buffer at 4°C.

•Storage buffer

The genome of B. amyloliquefaciens FZB42 was sequenced in collaboration with the GenoMik Network in Göttingen using the random shotgun approach. Total genomic DNA was shared randomly or partially digested with Sau3AI, and DNA fragments 1 to 3 kb in size were cloned into pTZ19R or pCR2.2 TOPO (Invitrogen) to establish a shotgun library. The inserts of the recombinant plasmids were sequenced from both ends using the MegaBACE DNA Sequencing Systems 1000 and 4000 (Amersham-Biosciences) and ABI Prism 377 sequencers (Applied Biosystems) with dye terminator chemistry. Fosmid library and combinatorial multiplex PCR were performed in order to determine the RNA sequences present in the genome and their location within it.

Approximately 44068 sequences were processed with PHRED, assembled into contigs by using the PHRAP assembling tool [213] and edited with GAP4, which is a part of the STADEN package software [214]. The resulting contigs of B. amyloliquefaciens FZB42 were sorted using the genome of B. subtilis 168 as scaffold [7]. PCR-based techniques and primer walking on recombinant plasmids were applied in order to close remaining sequence gaps.

Identification of ORFs in all six different frames was performed using the Glimmer2 program. Annotation of the genome was done using the GeneSOAP program provided by Rainer Cramm. The full length of each ORF was determined according to the presence of putative ribosome-binding sites and putative start codons as well as by comparison to the ortholog ORFs in B. subtilis 168 [7] and in B. licheniformis DSM13 [11]. In addition, using GeneSOAP it was possible to look for conserved protein domains within the ORFs of B. amyloliquefaciens FZB42 by comparison to PFAM [215].