The Gram-positive endospore-forming bacterium Bacillus licheniformis can be found widely in nature and it is exploited in industrial processes for the manufacturing of antibiotics, specialty chemicals, and enzymes. Both in its varied natural habitats and in industrial settings, B. licheniformis cells will be exposed to increases in the external osmolarity, conditions that trigger water efflux, impair turgor, cause the cessation of growth, and negatively affect the productivity of cell factories in biotechnological processes. We have taken here both systems-wide and targeted physiological approaches to unravel the core of the osmostress responses of B. licheniformis. Cells were suddenly subjected to an osmotic upshift of considerable magnitude (with 1 M NaCl), and their transcriptional profile was then recorded in a time-resolved fashion on a genome-wide scale. A bioinformatics cluster analysis was used to group the osmotically up-regulated genes into categories that are functionally associated with the synthesis and import of osmostress-relieving compounds (compatible solutes), the SigB-controlled general stress response, and genes whose functional annotation suggests that salt stress triggers secondary oxidative stress responses in B. licheniformis. The data set focusing on the transcriptional profile of B. licheniformis was enriched by proteomics aimed at identifying those proteins that were accumulated by the cells through increased biosynthesis in response to osmotic stress. Furthermore, these global approaches were augmented by a set of experiments that addressed the synthesis of the compatible solutes proline and glycine betaine and assessed the growth-enhancing effects of various osmoprotectants. Combined, our data provide a blueprint of the cellular adjustment processes of B. licheniformis to both sudden and sustained osmotic stress.
Funding: This study was financially supported by grants from the German Ministry of Education and Research via the Bacell-SysMo2 consortium (to ML, MH and EB) and the competence network “Genome Research in Bacteria” (to ML, MH, KHM and TS). Additional Funds were provided through the LOEWE program of the State of Hessen (via the Centre for Synthetic Microbiology; SynMicro, Marburg) (to EB), the Fonds der Chemischen Industrie (to EB) and the French National Centre for Scientific Research (UPR 9073), Université Paris VII-Denis Diderot (to HP). The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Copyright: © 2013 Bremer et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
While an effective water management is certainly the cornerstone for the cellular response to high salinity by most microorganisms [ 32 , 47 ], the overall adjustment process to this environmental challenge is a rather complex process. This is evident from genome-wide transcriptomic and proteomic assessments of the responses of B. subtilis and of the food pathogen B. cereus to acute and sustained high salinity environments [ 39 , 41 , 65 , 66 ]. Here we have combined physiological approaches, multi-omics techniques and genome mining to derive a comprehensive picture of the molecular and cellular events that allow the adaptation of B. licheniformis to high osmolarity surroundings.
The cellular stress response to high osmolarity is typically a multi-phasic process [ 32 ]. It initially entails in many microorganisms the rapid uptake of K + ions as an emergency reaction [ 53 ] and the subsequent replacement of this ion with a class of organic osmolytes that are highly compliant with cellular physiology, the compatible solutes [ 48 ]. Synthesis and uptake of the compatible solutes glycine betaine and proline play a key role in the defense of B. subtilis against the insults of high salinity [ 53 - 64 ]. A qualitative assessment by natural abundance 13 C-NMR spectroscopy has previously revealed that B. licheniformis belongs to the group of Bacilli that synthesize large amounts of proline when they are continuously exposed to high salinity surroundings [ 33 , 34 ].
Despite the existence of water-conduction channels, the aquaporins, in a considerable number of microorganisms [ 46 ], it is worth recalling that no bacterial cell can actively pump water across the semi-permeable cytoplasmic membrane to compensate for the water influxes or effluxes instigated by changes in the external osmolarity [ 32 , 47 ]. As a consequence, microorganisms have to balance the vital osmotic gradient across their cytoplasmic membrane indirectly by influencing the osmotic potential of the cytoplasm to direct the flux of water in or out of the cell [ 48 , 49 ]. They accumulate water-attracting ions and organic osmolytes (compatible solutes) when they face hyperosmotic conditions to prevent cellular dehydration [ 32 , 48 ], and they rapidly expel these compounds through the transient opening of mechanosensitive channels to avert cell rupture when the osmolarity suddenly drops [ 50 - 52 ].
As its cousins Bacillus subtilis [ 35 , 36 ] and Bacillus cereus [ 37 ], B. licheniformis possesses a general stress response system that is under the control of the alternative transcription factor SigB [ 37 , 38 ]. Members of the SigB-controlled regulon provide pre-emptive stress resistance to a multitude of environmental insults, various cellular constraints, and nutritional limitations [ 35 - 37 ]. Detailed studies in B. subtilis have shown that high salinity is one of the most-strongest inducers of the general stress response [ 39 , 40 ], and many members of its SigB regulon contribute to survival when the cells are exposed to severe and growth limiting osmotic up-shocks [ 41 , 42 ]. However, due to the transient nature of the induction of the SigB-regulon in response to acute salt stress [ 40 , 43 , 44 ], this general stress response system is not crucial for the ability of B. subtilis to strive under sustained high-salinity growth conditions [ 43 ]. Under these circumstances, an effective cellular water management is key to ascertaining a physiologically adequate level of hydration of the cytoplasm and magnitude of turgor in order to sustain growth [ 32 , 45 ].
An important challenge with which B. licheniformis has to cope, both in its varied natural habitats [ 1 - 5 ] and in industrial settings [ 30 ], are fluctuations in the external osmolarity. In its soil ecosystem, B. licheniformis will be frequently exposed on one hand to high osmolarity micro-niches that are caused by desiccation processes and increases in salinity; on the other hand, rainfall and flooding of the soil habitat will confront the cell with rapid osmotic down-shifts. During biotechnological applications, high-level excretion of metabolites and substrate feed will increase the osmolarity of the growth medium [ 27 , 30 , 31 ], a process leading to water efflux from microbial cells that causes a reduction in turgor and eventually leads to cessation of growth [ 32 ]. Hence, the overall productivity of B. licheniformis cell factories will be negatively affected by high-osmolarity growth conditions. However, the cellular adjustment process to either suddenly imposed or sustained osmotic stress is not well studied in this Bacillus species [ 33 , 34 ].
The sequencing of the genomes of two closely related B. licheniformis strains, B. licheniformis DSM 13 T [ 13 ] and B. licheniformis ATTC 14580 [ 14 ], has provided a blueprint for further in-depth studies of the physiology of this industrial workhorse and incentives for the rational design of industrial relevant strains with improved production capabilities [ 15 - 18 ] and enhanced biosafety [ 19 ]. Genome-wide transcriptomic and proteomic investigations of stressed B. licheniformis cells have allowed detailed insights into its genetic regulatory circuits, metabolic networks, biosynthetic capabilities, and cellular stress adaptation responses [ 10 , 20 - 26 ]. These studies have provided valuable knowledge when one considers that large-scale and high-density fermentation processes impose considerable constraints on microbial cells and can impair their fitness and capacity to produce biotechnologically valuable products efficiently [ 27 - 29 ].
Bacillus licheniformis is a Gram-positive endospore-forming microorganism that is widely distributed in nature and can readily be isolated from soils and animal and plant material [ 1 - 5 ]. It is extensively exploited in industrial processes [ 6 - 8 ]. In particular, the excellent protein secretion capacities of B. licheniformis [ 9 , 10 ] have made it an attractive host for the large-scale production of commercially employed enzymes (e.g., amylases, phytases, proteases). It is generally regarded as safe, and since some strains of this species are considered as probiotic, B. licheniformis is also used in the food and feed industry [ 11 , 12 ], but it can also be considered as a food spoilage bacterium [ 5 ].
To assess the resistance of B. licheniformis DSM 13T against the growth-inhibiting effects of high salinity, we grew cells in a chemically defined medium (SMM) with glucose as the carbon source and different salinities in shake-flask experiments at 37° C for 14 h and then determined the growth yield of the cultures by measuring their optical densities (OD578). As shown in Figure 1A, B. licheniformis DSM 13T can readily withstand salt concentrations up to 1 M NaCl but a further increase in the external salinity rapidly leads to a strong decline in growth yield; the presence of 1.3 M NaCl in the minimal medium resulted to a complete inhibition of growth. This osmotic stress resistance profile of B. licheniformis DSM 13T is similar, but not identical, to that of B. subtilis . B. licheniformis DSM 13T is thus a representative of the group of Bacilli exhibiting an intermediate degree of osmotic stress resistance, and most of these species synthesize proline as their dominant osmoprotectant [33,34]. Bacilli that exhibit a considerably higher degree of salt tolerance (e.g., Virgibacillus salexigens) than B. licheniformis DSM 13T or B. subtilis typically synthesize the compatible solute ectoine, whereas those Bacilli that synthesize only glutamate as their osmoprotectant (e.g., B. cereus) are rather salt-sensitive species [33,34,66].
Figure 1.Growth yields, proline production and osmoprotection of B. licheniformis DSM 13T by compatible solutes.
(A) Cultures of B. licheniformis DSM13T were grown at 37° C in SMM with glucose as the carbon source in the presence of the indicated NaCl concentrations. Growth yields of the cultures (as assessed by measuring the OD578) were determined after 14 h of incubation. (B) Proline content of osmotically stressed B. licheniformis DSM 13T cells. Cultures were grown in SMM with the indicated salinities either in the absence (red symbol) or in the presence (black symbol) of 1 mM of the osmoprotectant glycine betaine to an optical density (OD578) of approximately 2. The proline content of the cells was determined by HPLC analysis. The data shown represent one typical experiment. (C) Salt-stress protection of B. licheniformis DSM 13T by exogenously provided compatible solutes. Cultures of B. licheniformis DSM 13T were grown in SMM either in the absence (hatched bar) or in the presence of 1.3 M NaCl (black bars) in the absence (-) or in the presence of various compatible solutes. GB: glycine betaine; Cho: choline; Pro: proline; PB: proline betaine; Car: carnitine; COS: choline-O-sulfate; DMSP: dimethylsulfoniopropionate; Ect: ectoine; OHEct: hydroxyectoine. The compatible solutes were added to the growth medium at a final concentration of 1 mM. Growth yields of the cultures were measured after 14 h of incubation at 37 °C in a shaking water bath. The data shown were derived from two independently grown cultures.
Using natural abundance 13C-NMR spectroscopy, it has previously been found that B. licheniformis DSM 13T produces increased amounts of proline when cells are grown in SMM containing 1 M NaCl . To investigate this salt-stress responsive de novo proline biosynthesis in greater detail, we grew cultures of B. licheniformis DSM 13T to the same optical density (OD578 = 2) in SMM with different salinities and then measured the proline content of the cells by HPLC analysis (Figure 1B). An increase of the external salinity up to 0.2 M NaCl had little effect on the proline content of the cells, but further increases led to a graded rise in the cellular proline pool in a fashion that was directly proportional to the degree of the imposed osmotic stress (Figure 1B). Hence, it can be inferred from this experiment that B. licheniformis DSM 13T must have the ability to detect incremental increases in the external salinity and that the cell is then able to genetically convert this information into the build-up of a situation-conform pool of an osmostress-relieving compound. The size of the proline pool at a given external salinity was strongly influenced by the presence of 1 mM glycine betaine in the growth medium. Externally provided glycine betaine is an excellent osmoprotectant for B. licheniformis DSM 13T (Figure 1C) and its presence in the growth medium repressed the build-up of the osmostress responsive proline pool entirely (Figure 1B). Hence, osmotically challenged B. licheniformis DSM 13T cells preferred the import of a preformed osmoprotectant such as glycine betaine over the synthesis of proline, the only compatible solute that they can produce de novo [33,34]. The same phenomenon has also been observed in B. subtilis [54,57] and it might be connected with the cell’s attempt to cope with bioenergetic constraints under high-salinity growth conditions  or to optimize the solvent properties of the cytoplasm .
We used a combined transcriptomic and proteomic global gene expression analysis in order to identify relevant genes and gene clusters whose pattern of transcription is altered osmotic stress conditions in B. licheniformis. Although the two-dimensional PAGE-based separation of L-[35S]-methionine-labeled proteins allows for a good visualization of new highly expressed marker proteins, it shows only a very restricted window of the induced proteins under these stress conditions. Furthermore, as revealed in many other gene expression profiling studies, there is no direct quantitative correlation between the transcript and the protein levels. In most cases induction ratios at the transcriptomic level are significantly higher than at the proteomic level. This was also observed in our study. One typical example is the gene encoding the organic hydroperoxide resistance protein OhrB (YkzA) (see Table S2 and S3), which revealed a remarkable induction ration in osmotic stress that exceeded 500-fold at the transcript level but led to only a maximal 10-fold induction at the proteomic level. The overall expression pattern of the detectable proteins was in most cases consistent with their mRNA levels. There are very few exceptions; as an example, hag in this study, we detected a clear down-regulation of the hag gene at the transcriptomic level but observed a significantly increased accumulation at the proteomic level.
Our study offers an in-depth analysis of the salt-stress response of B. licheniformis at the transcriptome and proteome level, and we have combined this system-wide assessment with physiological approaches that address the synthesis and uptake of osmoprotectants. Combined with genome mining of systems known to be important for the management of osmotic stress in B. subtilis [32,41,45,52,54,69,71], the data presented here provide a detailed view of the B. licheniformis cell’s attempt to cope with and ameliorate the negative effects of high salinity on its physiology. A considerable overlap with the salt stress response of B. subtilis was evident but also stress reactions that are specific to B. licheniformis were found.
It well established that large-scale growth conditions of microbial cells influence the outcome of industrial size bioprocesses [96,97]. High-level excretion of the desired product from the microbial producer cell into the culture broth will successively increase the osmolarity of the medium [30,31] and such an increase can limit cell density and volumetric productivity. The feeding of osmoprotective compounds such as glycine betaine, choline, carnitine, proline or proline-containing peptides to osmotically stressed cells [58,60,71] will likely ameliorate such negative effects (Figure 1C) .
Microorganisms used in large-scale reactor environments are continuously exposed to various types of gradients [96,98]. This comprises also the development of osmotic gradients as a result of the feeding of high concentrated nutrient solutions and insufficient mixing of the culture broth. Since such osmotic gradients will certainly induce stress response in the microbial cell factory [28,31], the data presented here will help to understand such processes in B. licheniformis on a much more solid footing. Furthermore, the ability to distinguish between essential osmotic-stress-relieving pathways and dispensable regulons, which are not required or are even perturbing under industrial scale process conditions, could help to design more robust and efficient production strains of B. licheniformis.
We surmise that the B. licheniformis DSM 13T-derived proH-proI-proAA operon for the osmostress-adaptive proline biosynthesis (Figure 4) might be exploited in the context of synthetic microbiology as a bio-brick to engineer salt stress resistance in salt-susceptible microorganisms. Furthermore, the salient features of the osmotically inducible promoter driving the expression of this gene cluster (Figures 7 and 8), and the regulatory elements of other osmotically controlled B. licheniformis genes identified in our study (Figure 2) (Table S1), might turn out to be useful tools in developing novel types of environmentally responsive expression systems for Bacilli.