香京julia种子在线播放

Lirima is a geothermal site located in the Andean highlands of northern Chile (3.997 m.a.s.l., 19°51.118W, 68°54.402S). The Chilean Altiplano presents multiple co-occurring environmental stressors: the highest UV radiation ever reported on Earth (Cordero et al., 2016); variable daily temperatures (from below 0°C up to 30°C); high altitude (4000 m); and low relative humidity (Cordero et al., 2016). Microbial mats in different geothermal environments (aqueous and non-aqueous, temperature ranging from 30°C to over 90°C) were sampled during a field campaign in April 2011. The microbial mats were collected and stored using sterile material. After sampling, the sample was preserved at 4°C until enrichment. For aerobic enrichment, one gram of homogenized mat was inoculated in nutrient broth (NB) (Biolife, Italy), pH adjusted to 7.2, and incubated at room temperature for 7 days under aerobic conditions. The mismatch between in situ and cultivation temperature was used to favor the selection of dormant forms. 100 μL of the enrichment culture was plated on nutrient agar (NA) (Biolife, Italy) and incubated for 24 h at room temperature (RT) under aerobic conditions. Axenic cultures were isolated after repeated serial dilutions on NA medium at RT for 24 h. Strains were confirmed for purity by microscopic observation. Pure isolates were cryopreserved in 30% (v/v) of glycerol at -80°C. Isolates were screened for phase bright spore-like structures by microscopy after having been selected for dormancy by nutrient deprivation. The isolate Lr5/4 described here was isolated from a double layer microbial mat in a source pond at 54°C (Figure 1), alongside four other strains that were identified as belonging to the genus Bacillus.

Cell growth was monitored at different temperatures (4, 15, 25, 30, 37, 45, 55, and 60°C) for up to 72 h in nutrient broth medium (Biolife, Italy), measuring optical density at 600 nm (OD600) with a spectrophotometer Genesys 10S UV-Vis Thermo Scientific (Thermo Fisher Scientific Inc., United States). In addition, growth was measured at different pH (2 to 13) and at different NaCl concentrations (1, 2, 3, 3.5, 4, 5, 5.5, 6, 6.5, 8, 9, 10, 11, and 12% w/v) with a microplate reader at 590 nm (UVM 340 Microplate Reader, ASYS, Hiteck, United Kingdom) for up to 72 h. The need for oxygen during growth was verified using the method of thioglycollate medium (Brewer, 1940).

Phospholipids were extracted from the whole cells using a single-phase (methanol/dichloromethane/phosphate buffer; 2:1:0.8; v/v/v) extraction procedure (Guckert et al., 1985). Fatty acid methyl esters (FAMES) were prepared from the extracted acyl lipid mixture by acid-catalyzed methanolysis (Ichihara and Fukubayashi, 2010). The chemical characterization of the lipids was performed with the Agilent gas chromatograph 6890 coupled to the Agilent 5973 quadrupole mass selective detector (GC-MS; Palo Alto, CA, United States). For the analysis of FAME fractions, the system was equipped with the Agilent free fatty acids phase (FFAP) fused silica capillary column (50 m length, 0.20 mm i.d.) coated with nitroterephthalic acid modified polyethylene glycol stationary phase (film thickness 0.33 μm). A sample aliquot was injected split-less at a temperature of 200°C. Helium was used as carrier gas (1 mL/min flow rate). After an initial period of 2 min at 100°C, the column was heated to 240°C at 5°C/min followed by an isothermal period of 30 min. The MS was operated in the electron impact mode at 70 eV, source temperature of 250°C, emission current of 1 mA and multiple-ion detection with a mass range from 50 to 600 amu. Compound identifications were based on comparison of standards, GC retention time, and mass spectrometric fragmentation patterns.

For G+C content analysis and DNA-DNA hybridization, the services of DSMZ, Germany were used. Strain Lr5/4 was tested against S. marcescens (DSMZ 30121) and S. ureilytica (DSMZ 16952). MALDI-TOF mass spectra were acquired on a Bruker Microflex RLF mass spectrometer (Bruker Daltonics, Bremen, Germany). Spectral mass resolutions and signal-to-noise ratios were determined with the software Flex Analysis 3.3.65 (Bruker Daltonics).

For PCR amplifications, genomic DNA was extracted using the InnuPREP Bacteria DNA Kit (Analytik Jena, Germany), according to the manufacturer’s instructions. DNA was quantified fluorometrically with a Qubit^® dsDNA HS Assay Kit with a Qubit^® 2.0 Fluorometer (Invitrogen Ltd., Paisley, United Kingdom) according to the manufacturer’s instructions. PCR amplification of the 16S rRNA gene was performed using the primer set GM3F and GM4R (Muyzer et al., 1995). The PCR products were purified with a MultiScreen PCRμ96 Filter Plate (Millipore), according to the manufacturer’s instructions and sequenced using the services of GATC Biotech (Germany). To obtain the full-16S rRNA gene sequence, Lr5/4 PCR products were sequenced in addition with the primers 907r, 520r, and 926f primers (Muyzer et al., 1995; Forney et al., 1997). The 16S rRNA gene sequence was compared in GenBank by using the BLASTn tool (Altschul et al., 1997). After alignment, a series of sequences that had an identity match over 98% were selected, along with 16S rRNA gene sequences from cultured representative strains belonging to different species of the genus Serratia and a 16S rRNA gene sequence of a Geobacillus sp. (serving as an outgroup in order to root the tree). These sequences were used to build-up a cladogram using the online tools of phylogeny.fr website (Dereeper et al., 2008). Multiple sequence alignment has been performed using MUSCLE and the phylogenetic tree has been constructed using PhyML. To determine the confidence values for individual branches, 100 bootstrap replications were calculated for each generated tree.

Colonies of strain Lr5/4 were obtained and observed after overnight growth on NA. Gram staining was performed on an overnight solid culture using the Hucker staining method (Gerhardt, 1994). Spores were observed with a phase contrast microscope (Leica DM R, magnification 1000×). Vegetative cells and spores were also observed by Scanning and Transmission Electron Microscopy (SEM and TEM). Vegetative cells were observed from a fresh 24 h culture inoculated in NB. Cyst-like structures were collected from a 3-month old solid culture. Both preparations were fixed in 2.5% glutaraldehyde in a cacodylate buffer (0.1 M; pH7.4) for 2 h at room temperature and then overnight at 4°C. They were then washed by gentle immersion in cacodylate buffer (0.2M; pH7.4) post fixed with 1% OsO₄ in the same buffer, and carefully washed with the above buffer. For SEM, the samples were dehydrated in 15 to 100% of ethanol solution and finally fixed on Poly-L-Lysine slides and coated with a 23 nm gold layer in a BaltecSCD005 sputter apparatus. The samples were observed with a Philips XL30 SEM at acceleration voltages of 10–20 kV. For TEM, the samples were dehydrated in 15–100% acetone and embedded in Spurr’s resin. Serial sections were made with a Reichert Ultracut-S microtome, mounted on copper grids, double stained with uranyl acetate and lead citrate, and observed with a Philips CM 100 TEM at 60 or 80 kV.

Nutrient deprivation on solid media was one method for the production of light-refracting spherical cysts. Other environmental triggers were also put to test. Overnight liquid cultures in NB were subjected to temperature shocks. These shocks included two rounds of placing the liquid culture in an 80°C water bath for 20 min, then moved to 4°C for 20 min. The culture was then placed at room temperature overnight and observations for light refracting spheres was observed at 24 or 48 h. Overnight solid cultures on NA were subjected to desiccation and UV radiation, in order to verify whether these factors trigger the formation of cyst-like bodies. Overnight solid cultures were placed in a desiccator for 72 h until the solid medium dried to a 1 mm thin layer. Formation of light refracting spheres was examined at 24 and 48 h. Overnight solid cultures were also placed 60 cm from an UVc lamp (approximate wavelength 254 nm). The experiment was held under sterile conditions (BSL II hood), and the lids of the petri dishes were removed so that the bacterial colonies were directly exposed to the radiation. At t = 5, 10, 20, 30, 60, 120, 180, 240, and 720 min, three petri dishes were removed, closed sterilely and kept at room temperature. Observations for light refracting spheres was made at 24 or 48 h. Finally, overnight liquid cultures in NB were autoclaved and subjected to temperature extremes (no heat-shock). The extreme temperatures tested were -80, -20, 4, 60, 80, and 100°C. The overnight liquid cultures were incubated for 24 h at these temperatures, then observed for cyst formation. S. ureilytica str. Lr5/4 was tested along with Bacillus subtilis str. NEU1294, Anoxybacillus sp. str. Lr10/3 (isolated along Lr5/4 from Lirima), S. marcescens DSMZ 30121, S. ureilytica DSMZ 16952, and Escherichia coli str. NEU1006.

Spores produced by nutrient deprivation and heat-shocks were subjected to resistance tests. A heat resistance test for the spores of strain Lr5/4 was performed as previously described (Ajithkumar et al., 2003), along with S. marcescens (DSMZ 30121), S. ureilytica (DSMZ 16952), Bacillus subtilis (Neu1294), and Anoxybacillus sp. (Lr10/3). Suspensions of these cultures in tryptic soy broth (TSB) medium were heat-shocked at 70 and 75°C for 20 min and then re-cultured under optimal conditions. Growth was macroscopically verified after 24 and 48 h. Cyst-like or spore preparations, as well as vegetative cells from old S. marcescens str. DSMZ 30121, S. ureilytica str. DSMZ 16952, and E. coli str. NEU1006, were also autoclaved and exposed to UV radiation (radiation applied at 60 cm from a UV-C lamp, 200–280 nm) and to temperature extremes (as described above for cyst production). After exposure, biomass from the solid or liquid medium was stroked on solid media and cultured under optimal conditions in order to verify re-growth.

The presence of dipicolinic acid (DPA) in the spores was assessed on a 3-month old Lr5/4 preparation (20 mg wet weight), according to a previously published method (Brandes Ammann et al., 2011). This preparation contained almost exclusively refracting light structures. Fluorescence was measured within a Perkin-Elmer LS50B fluorometer. The excitation wavelength was set at 272 nm with a slit width of 2.5 nm. Emission was measured at 545 nm (slit width 2.5 nm). The device was set in the phosphorescence mode (equivalent to time-resolved fluorescence). The delay between emission and measurement was set at 50 μs. Measurements were performed every 20 ms. The integration of signal was performed over a duration of 1.2 ms. Values recovered for each measurement corresponded to the mean of the relative fluorescence unit (RFU) values given by the instrument within the 30 s following sample introduction in the device. Finally, to transform RFU units into DPA concentrations, a 10-point standard curve was established using increasing concentrations of DPA from 0.5 up to 10 μM.

Genomic DNA was extracted from an overnight culture using the Genomic-tip 20/G Kit (Qiagen GmbH, Germany). Sequencing was performed with PacBio RS II system based on single molecule, real-time (SMRT) technology (Pacific Biosciences, California). Genome assembly was performed using hgap 2.0 in the SmrtPortal. The full closed genome of S. ureilytica strain Lr5/4 presents a unique contig of 5,39 Mbp in size, and a G+C content of 59.2%. Genome annotation utilized an Ergatis based (Hemmerich et al., 2010) workflow with minor manual curation. The circular genome with the additional features was created using DNAplotter (Carver et al., 2009). The genome sequence was submitted to NCBI under Bioproject accession number PRJNA260750.

An extensive literature search was performed in order to compile all spore- and sporulation-specific genes reported for the known spore-forming phyla (Actinobacteria, Cyanobacteria, Firmicutes, and Proteobacteria). Whether a protein encoded by those genes was submitted to UniProt or NCBI was then verified. These proteins, along with their homologs from other spore-forming species were downloaded and multiple sequence alignments were performed using Clustal Omega, with default parameters (Sievers et al., 2011). For these proteins, profiles were built using HMMER version 3.1 with default parameters (Eddy, 2011). An additional extended literature search was performed in order to catalog all reported spore-forming and non-spore-forming species of the four phyla (Actinobacteria, Cyanobacteria, Firmicutes, and Proteobacteria). The genomes of the reported spore-formers were downloaded from NCBI, when available. These genomes were thus screened using the generated profiles and the latter were updated. The updated profiles were used to screen the genome of S. ureilytica Lr5/4.

The characterization of the strain Lr5/4 showed oval cells (0.5 × 1–3 μm) with a negative Gram-staining. Growth tests showed that Lr5/4 is a mesophile (temperature growth range between 10 and 37°C) with an optimum at 25°C. This range is not within the temperature measured at the sampling site (56°C), suggesting that the strain was present in the geothermal spring in a dormant state. Isolate Lr5/4 is also moderately halotolerant (growth up to 8% NaCl), able to develop over a vast pH range (from 3 to 11; optimal 5 to 6), and a facultative anaerobe, which are all characteristics that can be expected from a poly-extremophilic bacterium inhabiting the Chilean Altiplano. Based on the entire 16S rRNA gene sequence, isolate Lr5/4 belongs to the genus Serratia, with more than 97% identity to both Serratia marcescens and Serratia ureilytica. To establish the phylogenetic affiliation of the strain, a series of additional analysis were performed. While, MALDI-TOF characterization suggested that strain Lr5/4 belonged to S. marcescens (Supplementary Figure 1), PFLA analysis (Supplementary Table 1) showed that palmitic acid is the major fatty acid in the three strains. Based on the total fatty acid composition, Lr5/4 is more closely related to S. ureilytica, although the fatty acid profile of S. ureilytica type strain correlates better with S. marcescens than with Lr5/4. DNA/DNA hybridization analysis between Lr5/4 and its closest relatives (S. ureilytica DSM-16952 and S. marcescens DSM-30121) suggests that it is a member of S. ureilytica, with above 99% DNA/DNA relatedness to S. ureilytica DSM-16952. Nonetheless, DNA/DNA relatedness was also found to be above 70% with S. marcescens subsp. marcescens, which is the recommended threshold for delineating bacterial species (18). This observation is intriguing as it has been previously shown that DNA/DNA relatedness between the two reference species (S. ureilytica and S. marcescens subsp. marcescens) is only 43.7% (19). Finally, a phylogenetic tree constructed using all the available full genomes of Serratia isolates was constructed showing the relationship between the species that belong to this genus (Supplementary Figure 2). Considering all biochemical and molecular tests performed, we concluded that strain Lr5/4 represents a novel strain of Serratia ureilytica.

After nutrient deprivation, S. ureilytica Lr5/4 differentiated into elliptic-spherical cell forms that appeared phase-bright at the phase contrast microscope (Figure 2A). The morphological characterization of dormant cells in strain Lr5/4 was further conducted using electron microscopy. Using scanning electron microscopy, vegetative cells display a bacillus-like shape of approximately 2.0 μm × 0.5 μm in size, while dormant cells corresponded to round spheres of 0.5 μm in diameter (Figure 2B). Using TEM and osmium (OsO₄) fixation, the interior of vegetative cells appears stained while dormant cells present a clearer core surrounded by two thin outer layers (Figure 2C and inset). Although the exact composition of these two layers is not yet known, they prevent the penetration of the osmium tetraoxide (OsO₄), used for fixation. The latter penetrated easily the cell wall and membrane of the vegetative form, which became dark gray (Figure 2C). OsO₄ is used for fixation of spore structures of Firmicutes, Actinomycetes (Lechevalier et al., 1966), and Myxococcus (Müller et al., 2015). An increase in coat thickness hinders the penetration of OsO₄, resulting in poor resolution of the inner coat and the spore protoplasm (Stevenson et al., 1972). The same has also been shown during germination experiments. In the latter, the protoplasm of spores is brighter and stains darker during germination as the spore coat is remodeled during outgrow (Hoeniger and Headley, 1968). Finally, experiments on the formation of resting light-refracting structures previously performed (Muliukin et al., 1997), have shown that TEM pictures of cells that resume their growth cycle stain dark, as observed also for S. ureilytica str. Lr5/4 (Figure 2C). This suggests that in spite of the apparent simplicity of the outer layers in the dormant form of strain Lr5/4, they are nonetheless impermeable to OsO₄ and prevent staining. Overall the morphological characteristics of the dormancy cells in Serratia suggest that they differ in size and form to the vegetative cells.

Although many environmental factors can hinder bacterial survival, nutrient deprivation is a well-studied dormancy triggering factor (Barton, 2005). Accordingly, S. ureilytica Lr5/4 dormant cells developed in response to starvation (after 3 months of growth on agar medium). Other environmental factors may also influence spore-formation, as well as resting cell production in non-spore-formers (Suzina et al., 2006), however, these factors have not been thoroughly tested in previous studies. Therefore, for this study, the roles of other known triggers of dormancy that could be relevant in the geothermal spring from which strain Lr5/4 was isolated (UV radiation, thermal shock, temperature extremes and desiccation) were also measured. Dormant cells of Lr5/4 were produced in response to UV radiation and thermal shock (Supplementary Table 2). In contrast, cells of Lr5/4 were largely tolerant to desiccation. Extreme temperatures (down to -80°C and up to 100°C applied overnight), did not induce dormancy in S. ureilytica Lr5/4. B. subtilis str. NEU1294 and Anoxybacillus sp. str. Lr10/3 responded to desiccation and extreme temperatures by producing endospores. On the contrary, S. marcescens DSMZ 30121, S. ureilytica DSMZ 16952, and E. coli str. NEU1006 did not produce any cyst-like cells. Although cyst-like resting structures have been previously reported for E. coli (Suzina et al., 2004), it appears that the specific environmental triggers that are characteristic of the Atacama Desert did not trigger cyst-production in the E. coli strain tested here. No spore or cyst production was observed in response to autoclaving. Autoclaving killed all cells tested.

In order to evaluate the resistance of the dormant forms produced by S. ureilytica Lr5/4, cyst-like cells from 90-day old and heat-shocked cultures were challenged with multiple environmental stressors. Dormant forms of Lr5/4 were able to withstand and generate an actively growing culture after wet heat-shock at 75°C, 240 min UV radiation exposure, and incubation at extreme temperatures. This was comparable to the survival of endospores from the endospore-forming bacilli B. subtilis and Anoxybacillus sp. Lr10/3, although the latter resisted to 120 min UV radiation exposure (Supplementary Table 3). However, cells of S. marcescens DSMZ 30121 and S. ureilytica DSMZ 16952, as well as cells of E. coli did not survive these treatments. Autoclaving killed all spores and cells tested. When dormant forms of Lr5/4, B. subtilis and Anoxybacillus sp. Lr10/3 were subjected to multiple stressors applied simultaneously (starvation, desiccation, and UV radiation), only the endospores of the Firmicutes species could be revived, while Lr5/4 did not survive.

Although for most of the stressors studied, the precise molecular mechanism involved in resistance is unknown, in endospores of Firmicutes the protection of DNA in response to wet heat is conferred by the accumulation of dipicolinic acid (DPA) in the core (Setlow et al., 2006). The presence of DPA (Brandes Ammann et al., 2011) in the dormant cells of Lr5/4 was assessed on dilutions of a 3-month old culture (Supplementary Figure 3). The DPA content per dormant cell was calculated to be between 2.36 and 2.55 × 10^-7 μg/dormant cell, a value nearly 14 times lower than the DPA concentration in spores of B. subtilis. DPA was also reported as detectable in the case of S. marcescens subsp. sakuensis [10 mg (g bacteria)^-1] (Ajithkumar et al., 2003), but the values cannot be compared to those obtained here as the dormant to vegetative cell ratio was not provided in the case of S. marcescens subsp. sakuensis.

In order to investigate the genomic imprints of dormancy in strain Lr5/4, the complete genome of the strain was sequenced (5.39 Mb in size, 59.2% G+C) and annotated. We next compared the 5,056 annotated genes to genes reported to be involved in sporulation in other clades. The molecular pathways for bacterial cellular differentiation to the formation of a spore are well described in Actinobacteria, Cyanobacteria, Firmicutes, and Proteobacteria. Therefore, a database with all the known genes involved in dormancy in these taxa (Supplementary Table 4) was built and the genome of S. ureilytica Lr5/4 was screened for the presence of those genes. The detection of these genes using a BLAST search is indicative rather than demonstrative of the presence of this genetic determinants and further experiments, including gene expression and functional characterization during sporulation, need to be performed. Therefore, these results should be regarded with caution. Nonetheless, significant homology was detected between 12 Lr5/4 loci and myxospore-formation genes, nine in actinobacterial exospores, and 24 genes involved in spore formation in the Firmicutes (Supplementary Table 5). These genes participate in peptidoglycan synthesis, signaling, and transcription regulation. These genes are dispersed throughout the genome of Lr5/4 (Figure 3). The phylogenetic analysis of the genes that are potentially involved in dormancy show that most of them are likely of proteobacterial origin (Figure 4), suggesting they have not recently been acquired via horizontal gene transfer. A thorough phylogenomic and functional analysis of the dormancy-related genes showed that they contain conserved domains homologous to proteins that, in other species, participate mainly to general stages of cell division and sensing (Table 1), with homologs reported also in non-dormant bacteria.