L-alanine-induced germination in Bacillus licheniformis -the impact of native gerA sequences

Background L-alanine, acting through the GerA receptor, was recently found to be an efficient germinant in Bacillus licheniformis ATCC14580/DSM13. Results In this study, we show that several of 46 examined B. licheniformis strains germinate remarkably slower than the type strain when exposed to L-alanine. These strains are not necessarily closely related, as determined by MLST (multi-locus sequence typing). Three of the slow-germinating strains were further examined in order to see whether nucleotide substitutions in the gerA sequences were responsible for the slow L-alanine germination. This was performed by complementing the transformable type strain derivate MW3ΔgerAA with gerA variants from the three slow-germinating strains; NVH1032, NVH1112 and NVH800. Conclusions A wide selection of B. licheniformis strains was evaluated for L-alanine-induced germination efficiency. Our results show that gerA substitutions could only partially explain why spores of some B. licheniformis strains responded slower than others in the presence of L-alanine.


Background
Spores of Bacillus licheniformis and other Bacillus species are frequent contaminants in foods [1,2]. Exposure to nutrients triggers spores to leave dormancy in the process of germination [3][4][5]. This process involves several steps leading to rehydration of the spore core and loss of dormancy. Under favorable conditions, spores will grow out and multiply to numbers that can cause food spoilage and sometimes disease [6]. While dormant spores are extremely heat resistant, germinated spores can easily be killed by a mild heat treatment [7]. Therefore, a food preservation technique applied by food manufacturers to reduce spore numbers in food products is "induced germination". The consequence of induced germination is spores germinated into vegetative cells will be heat sensitive and can therefore be inactivated, by successive heating below temperatures that compromise food quality (modified Tyndallization) [8][9][10]. The effectiveness of such a strategy depends on the germination rate of the spore population. A slow and/or heterogeneous germination rate of a specific spore population will reduce the effectiveness of such treatments [11][12][13][14].
Nutrient germinant receptors (GRs), localized to the inner spore membrane [15][16][17], are involved in the spore's recognition of specific nutrients in its environment, which is the initial step in the spore's return to growth [18]. Binding of nutrient to the receptors is believed to trigger the release of the spore core's large depot of Ca-dipicolinic acid (CaDPA), followed by rehydration of the spore core and degradation of the spore cortex [3]. Current knowledge about GRs and their nutrient specificity is mainly achieved from Bacillus subtilis and Bacillus cereus. However, genes encoding GRs are widely distributed among Bacillus and Clostridium species [5,19], implicating an essential role in triggering of spore germination in most spore-forming bacteria. Interestingly, the nutrient specificity of the receptors and the interaction between them varies between and even within species, as has been shown for B. cereus-group members [20][21][22].
GRs are generally encoded by polycistronic operons that are expressed late in sporulation under the regulation of the forespore-specific transcription factor, sigma G (σ G ) [23,24]. These genes constitute a family (gerA family) of homologous genes that probably have evolved from the same ancestor [4,19]. Three putative gerA family operons, gerA (A, B, C), gerK (A, C, B) and ynd (D,E 3 E 2 , F 1, E 1 ) and the single gerAC homologue yndF2 have been identified within the B. licheniformis type strain ATCC14580/DSM13 genome [25][26][27]. Of these, only the gerA operon has been functionally characterized so far [28]. gerA was found to be essential for germination in presence of L-alanine. A similar role has been described for gerA in B. subtilis [18]. L-alanine is probably the most universal single nutrient germinant among spore formers [19].
The Bacillus GRs which have been described so far are usually composed of three subunits termed A, B and C. The A and B subunits are predicted to contain 5-6 (A) and 10-11 (B) membrane-spanning domains, respectively [5,29], while the C subunit is thought to be a membrane-anchored lipoprotein [30]. The tertiary structure of B. subtilis GerBC was determined a few years ago [31]. The B-subunit, whose amino acid sequence shows homology to proteins of the APC (amino acidpolyamine-organocation) superfamily, is proposed to be the most likely site of ligand binding, as mutations within this subunit alter ligand specificity [4,32]. However, since mutations in any of the three cistrons are shown to disturb receptor function, the exact site of nutrient binding is still unknown [5].
The genetic relationship of 53 strains of the foodspoilage agent B. licheniformis, a close relative of B. subtilis, was recently described by a novel MLST scheme [33]. One of these strains, NVH1032, was isolated after surviving an "induced germination"-regime (Tyndallization), applied by the food industry to eliminate spore contamination. Preliminary results in our lab suggested that NVH1032 and other B. licheniformis strains germinate considerably slower than the type strain when exposed to Lalanine. Such slow-germinating strains pose a challenge to food manufacturers that want to implement "induced germination" as a strategy to reduce/eliminate spores during processing.
In this study, 46 of the 53 genotyped strains were screened for efficiency of L-alanine-induced germination, and the correlation between the genotype and the induced germination was determined. Furthermore, it was investigated whether the slow germination of three particular B. licheniformis strains was due to sequence differences in the gerA operon.

Results and discussion
Screening of L-alanine-induced germination in B. licheniformis strains In order to evaluate the efficiency of L-alanine-induced germination of the 46 B. licheniformis strains, the level of germination was recorded after addition of L-alanine in a screening assay. The results showed that germination efficiency, determined by reduction of absorbance (A 600 ) varied from~1 to 60% between the tested strains 2 h after the addition of germinant (Additional file 1). A drop in A 600 of 60% was equivalent to > 95% germinated spores, as verified by phase contrast microscopy. About 30 of the strains germinated well with a reduction in absorbance of 40% or more, while six strains germinated poorly (10% or less in reduction of absorbance).
In general, differences in germination between strains may be due to differences in lag time (interval between addition of germinant and loss of refractivity) and differences in rate of germination (slope of the germination curve/ΔA 600 min −1 ). Several factors may account for these differences: (i) permeability of the outer spore layers, restricting access of germinant to the inner membrane [34], (ii) germinant specificity [20,22], (iii) GR (nutrient germinant receptors) level [35], (iv) dysfunctional GRs [36], (v) GR synergism/antagonism [37] and/or (vi) structure of the cortex [38]. Within single populations of B. subtilis, a reduced level of GRs has been suggested to be one of the main reasons for slow germination or "superdormancy" [35], probably by increasing the lag time until CaDPA is released [14]. In B. subtilis, GRs have been proposed to be present in a relatively low number (<40) in the spore's inner membrane where they form discrete clusters, so-called germinosomes [16,39], however, it has recently been reported that this number may be highly underestimated [40]. The number of germination receptors has been shown to be strongly dependent on the sporulation conditions [4,41,42]. In this study, sporulation and germination conditions (e.g. temperature, sporulation medium, pH, activation time/temperature, germinant concentration) were optimized with respect to the type strain ATCC14580/DSM13. However, these conditions may not be optimal for all strains.

Distribution and characterization of the gerA operon
The gerA locus was detected by PCR in all of the 53 genotyped B. licheniformis strains (GenBank: KF358523-KF358575). To investigate whether certain gerA sequence variants were associated with slow germination, partial gerA operon sequences of all strains were analysed, aligned and organized into clusters. The resulting neighbourjoining (NJ) tree is presented in Figure 1. With the exception of two strains (NVH1109/"1a" and NVH1077/"1b") the NJ-dendogram was congruent with the MLST tree generated from six house-keeping genes [33]. Thus, the gerA locus seemed to have evolved in parallel to the house-keeping genes. The ratio of non-synonymous versus synonymous base substitutions (dN/dS) was 0.0845 which is somewhat higher than the calculated values for the individual MLST loci (0.0000-0.0457) [33], but far below the limit of 1.0 that is often set for loci undergoing positive selection. Thus, the gerA locus, similar to the house-keeping genes, seems to be subject to purifying (stabilizing) selection [43,44].
A total of seven unique alleles were distributed into four main clusters, determined "1a", "1b", "1c" and "2" (Figure 1). Cluster "2" was represented by only three strains, NVH1032, NVH800 and NVH1112, that all showed a slower and less efficient germination response (Additional file 1) compared to the type strain, ATCC14580/DSM13 (cluster "1b"). However, slow-germinating strains were also found within each of the other clusters. Thus, this part of the gerA operon sequence (718 bp ranging from 3′ end of gerAB to 5′ end of gerAC) was not suitable in order to completely distinguish slow-germinating and fastgerminating strains.

Germination of gerA complementation strains
In order to further investigate the influence of gerA sequences on germination rate, MW3ΔgerAA was complemented with gerA operons originating from the type strain ATCC14580/DSM13 [28], and the three slowgerminating strains (Figure 2c,d). The gerA sequences of ATCC14580/DSM13 , NVH1032 and NVH800 nearly restored the phenotype of the sequence originating strains, while complementing MW3ΔgerAA with the gerA sequence from NVH112 increased the germination rate of the complemented strain compared to NVH1112 wild-type ( Figure 2a,c). Still, the order of the germination rate between the four strains was consistent between the two experiments (NVH1112/NVH1321 < NVH1032/NVH1309 < NVH800/NVH1322 < ATCC14580/NVH1311), substantiating that the phenotypes of the complemented MW3ΔgerAA (See figure on previous page.) Figure 1 Cluster analysis of partial gerA sequences from 53 B. licheniformis strains. Dendogram of partial gerA operon sequences (626 bp) in 53 B. licheniformis strains. The sequences cover parts of the last two genes (gerAB and gerAC) of the tricistronic gerA operon. The dendogram was calculated using the NJ-method with tree branch quality assessed using bootstrap values (500 replicates) as shown next to the branches. The evolutionary distances were computed using the Maximum Composite Likelihood method and are in the units of the number of base substitutions per site. MLST sequence type (ST) is indicated in brackets behind each strain and gerA cluster (1a, b, c and 2) is indicated with solid vertical lines to the right. Analyses were conducted in MEGA5.

Figure 2
Spore germination of slow-germinating strains and of gerAA disruption mutant complemented with gerA sequences from slow-germinating strains. ab: Germination of MW3ΔgerAA (x), the wild-type strains ATCC14580 (■), NVH 1032 (▲), NVH1112 (•) and NVH800 (♦) measured as reduction in absorbance (A 600 ) after addition of germinant (100 mM L-alanine). cd: Spore germination of the MW3ΔgerAA (x), and MW3ΔgerAA complemented with gerA from ATCC14580 (□ NVH1311), NVH1032 (Δ NVH1309), NVH1112 (○ NVH1321) and NVH800 (◊ NVH1322) measured as reduction in absorbance (A 600 ) after addition of germinant (100 mM L-alanine). The results represent the average (SD) of three independent spore batches. The type strain derivate MW3 (dotted line) has been included in Figure 3D for comparison. mutant to some extent restored the phenotypes of the gerA originating strains. Germination data of MW3 carrying pHT315 (MW3_pHT315) showed that carrying the empty vector, or the use of erythromycin in the cultures, hampered the germination rate of the MW3 strain (Additional file 2). However, we assume that comparing the effect of the complementing sequences is acceptable since they are all carried by the same vector.
An important observation was that, in contrast to Løvdal et al. 2012 [28], L-alanine-induced germination was not completely abolished in MW3ΔgerAA (NVH1307). This weak germination (~10% phase dark spores after 120 min) was not observed in absence of germinant, indicating that germination receptors other than GerA might be weakly activated by L-alanine. We also noted that spores of the slow-germinating strain NVH1112 hardly germinated at all, and to a lesser extent than MW3ΔgerAA (Figure 2a,b). When complementing MW3ΔgerAA with the gerA operon from NVH1112 (NVH1321) germination efficiency increased, indicating that the gerA operon of NVH1112 has some functionality in presence of L-alanine. A faster and more efficient germination of the complementation mutants compared to their respectively gerA originating strains was also observed for NVH1322 (gerA from NVH800) and NVH 1309 (gerA from NVH1032).
The imperfect complementation of the phenotypes may be due to several different factors. Firstly, a twoto seventeen-fold increase in expression level of gerAA was observed when MW3ΔgerAA was complemented with different gerA sequences and compared to the wildtype strains from where the gerA sequences originated (Figure 3). The increased gerAA expression level in the complementation mutants might be related to the copynumber of the plasmid pHT315 (15 copies per cell). Previous experiments have shown that a 2-200 fold overexpression of ger genes may increase germination rate [45,46]. Secondly, since the complementing gerA genes in this experiment were plasmid-born (pHT315 encoding erythromycin resistance), 1 μg ml −1 erythromycin was used in the sporulation medium to maintain the plasmid throughout the sporulation process and MW3 carrying the pHT315 empty vector germinated slower and with less efficiency than the wild-type MW3 strain (Additional file 2). Despite this observation, MW3ΔgerAA complemented with slow germinating gerA sequences germinated better than the strains from where the gerA sequences originated (Figure 2a-d).
Thirdly, the entire gerA operon and the 151 bp region upstream of the start codon of gerAA was cloned in the complementing vector pHT315. Alignments of the promoter sequence of strain NVH1032, NVH800, NVH1112 and ATCC14580/MW3 can be viewed in Additional file 3. No differences were observed between the type-strain and the slow germinating strains in the −10 and −35 promoter region of gerA. However, differences outside these regions may influence the transcriptional level. pHT315 [47] contains the inducible lac promoter, but transcription from this promoter cannot be excluded even without induction.
Despite the imperfect restoration of the wt phenotypes, the results of the germination assays in this study indicate that the gerA sequences have an impact on germination rate and efficiency. Differences in the GerA amino acid sequence may lead to altered protein 3-D structure, which again may cause impaired assembly and stability of the receptor complex in the inner membrane, lower or higher substrate affinity or influence the interactions with other membrane proteins.

Structural analysis of amino acid substitutions in the GerA receptor
Analyses of single amino acid substitutions have previously been conducted in B. subtilis GerAA [48], GerAB [49] and GerBC [50]. None of these positions were substituted in the GerA sequences examined in the present study. Alignments of the GerAA, GerAB and GerAC sequences of B. licheniformis NVH1032, NVH800, NVH1112 and ATCC14580/DSM13 are presented in Additional files 4, 5 and 6. Thus, on the basis of this knowledge and the lack of a 3D structure of any proteins in the GerAA and GerAB families of proteins, the relevance of the observed differences within these two subunits is difficult to determine. However, the crystal structure of B. subtilis GerBC has recently been determined [31]. Using this structure as a template for prediction of B. licheniformis GerAC structures, one of the perhaps most interesting substitutions is F342S (NVH1032 and NVH800) which lies in the so-called "region 2" of domain III [50] (Additional file 7). Region 2 is reported to be a well conserved region in GerBC among Bacillales and substitutions within this region were previously shown to affect receptor function in B. subtilis [50]. On the other hand, the F342S substitution was neither observed in the gerA sequences of the slowest germinating strain NVH1112 or the fastest germinating strain ATCC14580/DSM13 suggesting that the role of this site seems unclear. It should be mentioned that the aa sequence of the GerAC protein of NVH1112 is much closer to that of MW3 than the other two (Additional file 6), indicating that GerAC is not crucial to germination efficiency. Ultimately, the lack of information about the exact germinant binding site, as well as the fact that only the C subunit has been structurally characterized, makes it difficult to interpret the effect of single substitutions on the GerA receptor function.

Conclusions
This study shows that spores of 46 B. licheniformis strains are able to germinate in the presence of L-alanine, but that the germination rate and efficiency differ significantly between the strains. About 10% of the strains germinated poorly, even in presence of high (100 mM) concentrations of probably the most universal and potent germinant for Bacillus species in general, and B. licheniformis in particular. Germination rate of different bacterial strains are of importance to the food industry, using so-called "induced germination", eg Tyndallization, to decrease spore contamination in processed foods. Delayed germination may reduce the efficiency of Tyndallization by allowing ungerminated spores to survive. Our results demonstrate that nutrient-induced germination followed by inactivation can be challenging when dealing with specific B. licheniformis strains.
The germination phenotype was partly restored when complementing a gerAA disruption mutant with gerA operons from either slow-or fast-germinating B. licheniformis strains. This observation indicates that differences in gerA family operons are partly responsible for differences in germination efficiency of B. licheniformis in response to L-alanine.

Strains
Strains included in this work are listed in Table 1. The 53 strains were previously characterized and genotyped by a novel MLST scheme [33].
MW3 ΔgerAA (NVH1307) and the complementation mutant NVH1311 are described in Løvdal et al. 2012 [28]. The complementation mutants NVH1309, NVH1321 and NVH1322 were constructed in this work as described later on.

DNA extraction
Bacteria were grown on sheep blood agar at 30°C overnight. Single colony material was inoculated in 20 mL Luria broth (LB). The bacterial culture was grown overnight at 30°C and centrifuged at 3000 × g for 10 min. The supernatant was discarded and the pellet resuspended in 1 mL enzymatic lysis buffer (20 mM Tris · Cl, pH 8.0, 20 mM Tris · Cl, pH 8.0, 1.2% Triton® X-100, 20 mg mL −1 lysozyme (Sigma, Steinheim, Germany)). Further DNA extraction was performed according to the protocol provided by DNeasy Blood and Tissue Kit (Qiagen, USA).

PCR and sequencing of the gerA operon
Primer A7F and A7R (Table 2) were used to amplify a 718 bp region of the gerA operon, including 3′ end of gerAB and 5′ end of gerAC. Additionally, complete gerA operons from strain NVH800, NVH1032 and NVH1112 were amplified in smaller fragments for DNA sequencing using primers listed in Additional file 8. All amplification reactions were performed in 20 μL using 2 μL DNA (10 ng μL −1 ) as a template. PCR reactions were performed in a LightCycler® 480 System using LightCy-cler® 480 SYBR Green I Master (Roche Diagnostics GmbH, Germany) according to recommendations given by the manufacturer of the kit. The temperature program was as follows: 5 min initial denaturation at 95°C followed by 35 cycles of denaturation at 95°C for 10 s, annealing at 56°C for 10 s and extension at 72°C for 30 s. The amplifications were terminated after a final elongation step of 7 min at 72°C. The PCR fragments were verified by electrophoresis using Bioanalyzer (Agilent Technologies, USA). PCR products were purified and sequenced by Eurofins MWG Operon (Ebersberg, Germany) using the dideoxy chain termination method on an ABI 3730XL sequencing instrument (Applied Biosystems, USA).

Data analysis
The Staden Package [52] was used for alignment, editing and construction of consensus sequences based on the ABI sequence chromatograms. Consensus sequences (626 bp) were entered into the MEGA5 software [53] and aligned by CLUSTALW [54]. Dendograms were constructed in MEGA5 using the Neighbor-Joining method (NJ) [55] with branch lengths estimated by the Maximum Composite Likelihood method [56]. Branch quality was assessed by the bootstrap test using 500 replicates. Sequences were trimmed to be in frame, which means that eight bases in the transition between gerAB and gerAC were removed, before entering into S.T.A.R.T. 2 [57]. This program was used to calculate the dN/dS ratio (ratio of nonsynomous versus synonymous substitutions) [58].

Construction of B. licheniformis MW3ΔgerA complementation mutants
The entire gerA operons including the putative sigG promoter from B. licheniformis strain NVH1032, NVH800 and NVH1112 were cloned into the pHT315 [47] shuttle vector and introduced into the gerAA deletion mutant strain MW3ΔgerAA by electroporation as described previously [28]. Briefly, PCR, with primers ( Table 2) containing SalI and XbaI restriction sites, was used to amplify the gerA operon including 151 bp upstream of the gerAA start codon and 177 bp downstream of the gerAC STOP codon. The amplified fragments were cloned into the SalI/XbaI restriction site of pHT315, giving the complementation plasmids. For details regarding primers, PCR conditions, DNA isolation and electroporation see Løvdal et al. 2012 [28]. The strains created in this study were designated as follows: B. licheniformis NVH1309 (MW3ΔgerAA _NVH1032gerA); NVH1321 (MW3ΔgerAA _NVH1112gerA) and NVH1322 (MW3ΔgerAA _NVH800gerA). Correct construction of the complementation plasmids was confirmed by sequencing and the complementation mutants were verified by PCR analysis. Sequence editing and alignments were performed as already described in the Data analysis section.

Bacterial growth and sporulation
Sporulation was performed according to Løvdal et al. 2012 [28], with minor modifications. Bacteria were precultured overnight in LB-broth with agitation (230 rpm) at 37°C. Complementation mutants were grown in presence of 1 μg mL −1 erythromycin. 10 μL of preculture was transferred to 50 mL of the non-defined, rich sporulation medium [28] in 500 mL EM flasks. Incubation was performed with agitation (230 rpm) at 37°C for 3-7 days until ≥ 80% phase bright spores as judged by phase   Table 2. The amplification was performed using StepOne PCR software (Applied Biosystems) with thermal cycling conditions set at 10 min at 95°C, followed by 40 cycles of 15 s at 95°C and 1 min at 60°C. Fluorescence was monitored during each extension phase and a melting curve analysis was performed after each run to confirm the amplification of specific transcripts. Each qPCR of the RNA samples was performed in triplicate, no template was added in negative controls, and rpoB was used as internal control. The qPCR analysis was performed on three independent biological replicates. Slopes of the standard curves and PCR efficiency (E) for each primer pair were estimated by amplifying serial dilutions of the cDNA template. For quantification of mRNA transcript levels, Ct (threshold cycle) values of the target genes (gerAA) and the internal control gene (rpoB) derived from the same sample in each real-time PCR reaction were first transformed using the term E −Ct . The expression levels of target genes were then normalized by dividing their transformed Ct-values by the corresponding values obtained for internal control gene [64,65].

Germination assays
Storage water was decanted and the spores were resuspended in autoclaved Milli-Q water (20°C) immediately before heat activation at 65°C in a heating block (QBD2, Grant Instruments Ltd) for 20 min. The heat-activated spores were rapidly cooled down by centrifugation for 3 min 4500 × g at 4°C before resuspension in germination buffer (200 mM K-phosphate buffer pH 7.2). The A 600 of the buffered spore suspension was adjusted to~2.1 (Shimadzu UV-160A, Shimdazu Europe GMBH). L-Alanine (Sigma) was dissolved in Milli-Q water and filter sterilized prior to use through a 0.45 μm pore size filter. 100 μL of 0.05 -0.2 M L-Alanine germinant solution was added to 100 μL buffered spore suspension in a 96-well microplate (BD) giving an initial A 600 of~1. Germination was by monitored by reading the drop in absorbance (A 600 ) in a 96-well microplate reader (Tecan Infinite M200). Readings were performed at regular intervals (2 min) and the plate was shaken 10 s prior to each reading. Set point temperature during germination was 37°C (36.5 -37.5). The screening of 46 strains was performed in duplicate with a single spore preparation. All other experiments were performed with three independent spore preparations.