Sampling

1. Measure the temperature of the sampling site using a thermometer.

2. For hot spring sediment: Collect the sediment with a sampling shovel, then use a small shovel to transfer the sample into an anaerobic tube. Ensure the tube is fully filled to displace oxygen; after filling, seal the tube tightly with a rubber stopper and store it in a sampling box. For water samples: When collecting water, wear heat-insulating gloves to submerge a sterilized anaerobic tube into the water. Fill the tube completely, seal it tightly with a rubber stopper, cover it with a cap, and place it in the sampling box. Alternatively, collect deep hot spring water using a long-handled sampling spoon.

3. Sample Processing: Suspend 2 g of the sample in an Erlenmeyer flask containing 18 mL of sterile water or 1/10-strength R2A medium (supplemented with glass beads), and shake the flask to achieve full suspension of the sample.

4. Serial Dilution: Transfer 1 mL of the suspension from the Erlenmeyer flask to a 15 mL centrifuge tube containing 9 mL of sterile physiological saline. Mix the liquid in the tube using a pipette, then transfer 1 mL of this suspension to another 15 mL centrifuge tube and add 9 mL of sterile water. Repeat the above steps to dilute the sample to 10⁻¹, 10⁻², 10⁻³, and so on.

5. Bacterial Culture: Transfer 100 μL of the suspension to the surface of a solid medium and spread it evenly. Plating is completed when no flowing droplets are observed after tilting the medium. Record and label information including sample type, temperature, dilution factor, and inoculation date. Place the plate face-up on a clean workbench for 2 hours. Afterwards, invert the Petri dish and place it in an incubator for cultivation at 45℃, 55℃, and 65℃ respectively. Observe the growth status after 3, 7, and 14 days of cultivation.

Metagenomics Experiment

1. Sample DNA extraction

The total genomic DNA of the microbial community was extracted according to the E.Z.N.A. ® soil DNA kit ( Omega Bio-tek, Norcross, GA, U.S. ) instruction. After DNA extraction, DNA concentration and purity were detected, and DNA integrity was detected by 1 % agarose gel electrophoresis. The DNA was fragmented by Covaris M220 ( Gene, China ), and a fragment of about 350 bp was screened for the construction of a PE library.

2. Construction of PE library

The library was constructed using NEXTFLEX Rapid DNA-Seq ( Bioo Scientific, USA ). The specific process is as follows :

1) Connector link ;

2) Magnetic bead screening was used to remove the self-connected fragments of the connector ;

3) Enrichment of library templates by PCR amplification ;

4) The PCR products were recovered by magnetic beads to obtain the final library.

3. Bridge PCR and sequencing

Metagenomic sequencing was performed using the Illumina NovaSeq TM X Plus ( Illumina, USA ) sequencing platform. The specific process is as follows :

1) One end of the library molecule is complementary to the primer base, and after a round of amplification, the template information is fixed on the chip ;

2) The other end of the molecule fixed on the chip is randomly complementary to another nearby primer and is also fixed to form a ' bridge ' ;

3) PCR amplification to produce DNA clusters ;

4) DNA amplicons are linearized into single strands.

5) The modified DNA polymerase and dNTP with four kinds of fluorescent markers were added, and one base was synthesized in each cycle.

6) Using laser scanning reaction plate surface, read each template sequence of the first round of reaction polymerization of nucleotide species ;

7) The ' fluorophore ' and ' termination group ' were chemically cleaved to restore the 3 ' end adhesion and continue to polymerize the second nucleotide ;

8) The sequence of the template DNA fragment was obtained by counting the fluorescence signal results collected in each round.

4. Data quality control

1) Fastp ( https://github.com/OpenGene/fastp, version 0.20.0 ) was used to cut the adapter sequences at the 3 ' and 5 ' ends of reads.

2) Fastp ( https://github.com/OpenGene/fastp, version 0.20.0 ) was used to remove reads with a length of less than 50 bp and an average base mass value of less than 20, and high-quality sequences were retained.

3) The reads were aligned with the host DNA sequence by software BWA ( http://bio-bwa.sourceforge.net, version 0.7.17 ), and the contaminated reads with high alignment similarity were removed.

5. Assembly and gene prediction

The optimized sequences were assembled using software MEGAHIT ( https://github.com/voutcn/megahit, version 1.1.2 ). Contigs ≥ 300 bp were selected as the final assembly results. Prodigal ( https://github.com/hyattpd/Prodigal, version 2.6.3 ) was used to predict the ORFs of contigs in the splicing results. Genes with nucleic acid length greater than or equal to 100 bp were selected and translated into amino acid sequences.

6. Construction of non-redundant gene set

CD-HIT ( http://weizhongli-lab.org/cd-hit/, version 4.7 ) was used to cluster the gene sequences predicted by all samples ( parameters : 90 % identity, 90 % coverage ). The longest gene in each class was taken as the representative sequence to construct a non-redundant gene set.

7. Gene abundance calculation

Using SOAPaligner software ( https://github.com/ShujiaHuang/SOAPaligner, version soap2.21 release ), the high-quality reads of each sample were compared with the non-redundant gene set ( 95 % identity ), and the abundance information of the gene in the corresponding sample was counted.

8. Bioinformatics analysis

The data analysis starts from the original sequence of the machine. Firstly, the original sequence is optimized by splitting, mass shearing and removing pollution. Then, the optimized sequences were used for assembly and gene prediction, and the obtained genes were annotated and classified in species and function, including NR, EggNOG, KEGG, etc. On the basis of the above analysis, we can carry out statistical analysis and exploration in multiple directions, such as similar clustering, grouping and sorting, difference comparison, etc., and visualize the results, mine effective information in the data, expose the hidden rules, verify the experimental hypothesis and discover new problems.

After obtaining the annotations of each database, we focused on NR ( species database ) and KEGG ( functional database ) for analysis, focusing on the results of polyamine metabolism-related functions in high-temperature specific species and high-temperature sample groups, and conducted detailed analysis.

Diamond ( https://github.com/bbuchfink/diamond, version 2.0.13 ) was used to align the amino acid sequence of the non-redundant gene set with the NR database ( BLASTP alignment parameter setting expectation value e-value is 1e-5 ), and the species annotation was obtained through the taxonomic information database corresponding to the NR database, and then the abundance of the species was calculated using the sum of the gene abundance corresponding to the species.

The amino acid sequence of the non-redundant gene set was compared with the KEGG database using Diamond ( https://github.com/bbuchfink/diamond, version 2.0.13 ) ( BLASTP alignment parameter setting expected value e-value is 1e-5 ). The corresponding KEGG function of the gene was obtained. The sum of gene abundance corresponding to KO, Pathway, EC and Module was used to calculate the abundance of corresponding functional categories.

Mining of key enzymes in polyamine synthesis pathway

According to the KEGG annotation results of metagenomics ( BLASTP alignment with the KEGG database ), the KO number of each key enzyme in the polyamine synthesis pathway was searched, and according to its KEGG annotation number, code 1 was used to map to the corresponding NCBI species number, and then according to code 2, the sequence of the polyamine synthesis pathway enzyme in each species was downloaded from NCBI ( FASTA format ), and the multi-sequence alignment and phylogenetic tree were constructed based on the type of enzyme : First, the sequence was fully selected in the MEGA software, and the target multi-sequence was aligned using the Align tool. The algorithm is ClustalW. All the full space positions were deleted, and the file was exported to the.meg format, and then dragged into the MEGA software. The Phylogeny tool was selected to construct the phylogenetic tree, and the file was exported to the.nwk format and uploaded to the ITOL website ( https://itol.embl.de/), ) for each species classification annotation, and finally exported to the PDF format.

The protein files were downloaded to the PDB ( https://www.rcsb.org/) and UnitProt ( https://www.uniprot.org/) ) websites in pdb format. The structure in the PDB database was the crystal structure obtained from the experiment, while the protein structure in the UniProt database contained prediction and crystal structure. The crystal structure of the corresponding enzyme verified in the literature was compared with the structure of the enzyme we excavated in Alpha Fold 3.0 ( AlphaFold Server ), and the predicted structure was obtained by inputting the amino acid sequence. Then it is imported into PyMol software for visualization, and the coverage of the enzymes in the three-dimensional structure is compared. The similarity between the enzyme structure we excavated and the known enzyme structure can be obtained, which is also a functional verification method for the excavated enzyme.

Strain isolation and purification

1. Silt sample treatment

In the anaerobic glove box, about 1 g of each original sample was weighed in turn, placed in a 50 mL sterile centrifuge tube, and the weight was recorded and weighed. The sterile protective agent was added at 1 : 10 w / v for vortex oscillation treatment for 2 min, and the residue was filtered with a filter and sub-packed into a sterile nut tube, 1 mL / branch ; the sample processing for aerobic separation is performed in an ultra-clean bench, and the other operations are the same as above.

2. Water sample processing

In the anaerobic glove box, the water sample was enriched on the filter by vacuum filtration device, and then the filter membrane was divided into two parts and transferred to two centrifuge tubes. 10 mL good / anaerobic sterile protective agent ( the amount of addition was not more than all the membranes ) was added for vortex shock treatment for 2 min, and the liquid in the suction tube was divided into sterile nut tubes, 1 mL / tube.

3. Sample coating separation

The treated samples were diluted to 10-5 by 10-fold gradient, and 10-2 ~ 10-5 were coated with aerobic / anaerobic R2A ( Hibo : HB0167 ), aerobic / anaerobic ZPM10 medium ( self-made ), aerobic / anaerobic ZPM11 medium ( self-made ), aerobic / anaerobic ZPM12 medium ( self-made ) and aerobic / anaerobic ZPM13 medium ( self-made ). In a good / anaerobic environment, the cells were cultured for 5-6 days according to the original sample temperature.

The above five liquid media were used to enrich the treated samples at a volume ratio of 1 : 10. After 6 days of culture, the corresponding solid medium was coated, and then the samples were cultured for 6-7 days according to the original sample temperature.

Single colonies were selected and cultured in aerobic TSB medium / anaerobic ZPM02 medium for 3 ~ 4 d.

The aerobic group was identified by MALDI-TOF-MS.After excluding non-target strains ( duplicate strains ) according to the MS results, the target strains were transferred to 96-well plates ( at this time, the bacterial liquid may be in a non-pure bacterial state ), and then the bacterial liquid in the 96-well plate was picked up and purified on the aerobic TSA medium. The anaerobic group was preliminarily identified by PCR after synchronous culture. After the non-target strains were removed according to the PCR results, the target strains were selected and purified or expanded on the anaerobic R2A medium. The original sample was cultured at aerobic / anaerobic temperature for 3 ~ 4 days. After growing a single colony, the third region single colony was selected according to the colony morphology grown on the plate to the aerobic TSB medium / anaerobic ZPM02 liquid medium. The original sample was cultured at temperature for 3 ~ 4 days, and finally identified by PCR.

( Note : During the purification process, the single-hole bacterial solution may grow in a variety of forms on the plate, or may grow aseptically on the plate. Therefore, the number of subsequent PCR entry and MALDI-TOF-MS identification may be inconsistent. )

4. Experimental instruments and materials

1) No.1 medium

2) No.2 medium

Trace elements solution (mg/L):

3) No.3 medium

Main component solution :

Take 8 mL of the main component solution and add the following components:

4) No.4 medium

5) No.5 medium

Multi-omics experiment

1. Transcriptome

1.1. RNA extraction

Total RNA was extracted from the tissue using CTAB method and genomic DNA was removed. Only high-quality RNA sample was used to construct sequencing library.

1.2. Library Construction and Sequencing

Ribosomal RNA (rRNA) depletion instead of poly(A) purification is performed by RiboCop rRNA Depletion Kit for Mixed Bacterial Samples（lexogen，USA）and then all mRNAs were broken into short (300nt) fragments by adding fragmentation buffer firstly. Secondly double-stranded cDNA was synthesized with random hexamer primers (Illumina). When the second strand cDNA was synthesized, dUTP was incorporated in place of dTTP. Then the synthesized cDNA was subjected to end-repair, phosphorylation and 'A' base addition according to Illumina's library construction protocol. RNA-seq transcriptome library was prepared following Illumina® Stranded mRNA Prep, Ligation (San Diego, CA) using of total RNA. paired-end RNA-seq library was sequenced with the Illumina NovaSeq X Plus (or other new sequenator) (Illumina Inc., San Diego, CA, USA) .The processing of original images to sequences, base-calling, and quality value calculations. The clean reads by removing low-quality sequences, reads with more than 10% of N bases (unknown bases) and reads containing adaptor sequences.

1.3. Bioinformatics Analysis

The data generated from Illumina platform were used for bioinformatics analysis. All of the analyses were performed using the online platform of Majorbio Cloud Platform (https://cloud.majorbio.com/) from Shanghai Majorbio Bio-pharm Technology Co.,Ltd. (Shanghai, China) .

2. Metabolome

2.1. Metabolite Extraction

Solid sample：

100 mg solid sample was added to a 2 mL centrifuge tube and a 6 mm diameter grinding bead was added. 800 μL of extraction solution (methanol: water = 4:1 (v:v) containing four internal standards ( 0.02 mg/mL L-2-chlorophenylalanine, etc.) were used for metabolite extraction. Samples were ground by the Wonbio-96c ( Shanghai wanbo biotechnology co., LTD) frozen tissue grinder for 6 min (-10°C, 50 Hz), followed by low-temperature ultrasonic extraction for 30 min (5°C, 40 kHz). The samples were left at -20°C for 30 min, centrifuged for 15 min (4°C, 13000 g), and the supernatant was transferred to the injection vial for LC-MS/MS analysis.

Liquid sample：

100 μL liquid sample was added to a 1.5 mL centrifuge tube with 800 μL solution (acetonitrile: methanol = 1:1(v:v)) containing four internal standards (0.02 mg/mL L-2-chlorophenylalanine, etc.) to extract metabolites. The samples were mixed by vortex for 30 s and low-temperature sonicated for 30 min (5°C, 40 KHz)。The samples were placed at -20°C for 30 min to precipitate the proteins. Then the samples were centrifuged for 15 min (4°C, 13000 g). The supernatant was removed and blown dry under nitrogen. The sample was then re-solubilized with 100 µL solution (acetonitrile: water = 1:1) and extracted by low-temperature ultrasonication for 5 min (5°C, 40 KHz), followed by centrifugation at 13000 g and 4°C for 10 min.The supernatant was transferred to sample vials for LC-MS/MS analysis.

2.2. UHPLC-MS/MS analysis

The LC-MS/MS analysis of sample was conducted on a UHPLC-Orbitrap Exploris 240 system equipped with an ACQUITY HSS T3 column (100 mm × 2.1 mm i.d., 1.8 μm; Waters, USA) at Majorbio Bio-Pharm Technology Co. Ltd. (Shanghai, China). The mobile phases consisted of 0.1% formic acid in water:acetonitrile (2:98, v/v) (solvent A) and 0.1% formic acid in acetonitrile (solvent B). The flow rate was 0.40 mL/min and the column temperature was 40℃. The injection volume was 5 μL. MS conditions:

The UPLC system was coupled to a UHPLC-Orbitrap Exploris 240 system Mass Spectrometer equipped with an electrospray ionization (ESI) source operating in positive mode and negative mode. The optimal conditions were set as followed: source temperature at 400℃ ; sheath gas flow rate at 40 arb; Aux gas flow rate at 10 arb; ion-spray voltage floating (ISVF) at -2800V in negative mode and 3500V in positive mode, respectively; Normalized collision energy , 20-40-60V rolling for MS/MS. Data acquisition was performed with the Data Dependent Acquisition (DDA) mode. The detection was carried out over a mass range of 70-1050 m/z.

2.3. Data preprocessing and annotation

The pretreatment of LC/MS raw data was performed by Progenesis QI (Waters Corporation，Milford, USA) software, and a three-dimensional data matrix in CSV format was exported. The information in this three-dimensional matrix included: sample information, metabolite name and mass spectral response intensity. Internal standard peaks, as well as any known false positive peaks (including noise, column bleed, and derivatized reagent peaks), were removed from the data matrix, deredundant and peak pooled. At the same time, the metabolites were identified by searching database, and the main databases were the HMDB (http://www.hmdb.ca/), Metlin ( https://metlin.scripps.edu/) and the self-compiled Majorbio Database (MJDB) of Majorbio Biotechnology Co., Ltd. (Shanghai, China)

The data matrix obtained by searching database was uploaded to the Majorbio cloud platform (https://cloud.majorbio.com) for data analysis. Fistly, the data matrix was pre-processed, as follows: At least 80% of the metabolic features detected in any set of samples were retained. After filtering, the minimum value in the data matrix was selected to fill the missing value and each metabolic signature was normalized to the sum. To reduce the errors caused by sample preparation and instrument instability, the response intensities of the sample mass spectrometry peaks were normalized using the sum normalization method, to obtain the normalized data matrix. Meanwhile, the variables of QC samples with relative standard deviation (RSD) > 30% were excluded and log10 logarithmicized, to obtain the final data matrix for subsequent analysis.Perform variance analysis on the matrix file after data preprocessing.

The R package "ropls"(Version 1.6.2) was used to perform principal component analysis (PCA) and orthogonal least partial squares discriminant analysis (OPLS-DA), and 7-cycle interactive validation evaluating the stability of the model. The metabolites with VIP>1, p<0.05 were determined as significantly different metabolites based on the Variable importance in the projeciton (VIP) obtained by the OPLS-DA model and the p-value generated by student's t test.

Differential metabolites among two groups were mapped into their biochemical pathways through metabolic enrichment and pathway analysis based on KEGG database (http://www.genome.jp/kegg/). These metabolites could be classified according to the pathways they involved or the functions they performed. Enrichment analysis was used to analyze a group of metabolites in a function node whether appears or not. The principle was that the annotation analysis of a single metabolite develops into an annotation analysis of a group of metabolites. Python packages "scipy.stats" (https://docs.scipy.org/doc/scipy/ ) was used to perform enrichment analysis to obtain the most relevant biological pathways for experimental treatments.

3. Proteome

3.1. Total protein extraction

Take out the samples in the frozen state and put it on ice. The samples were suspended in protein lysis buffer (8M urea,1% SDS) which included appropriate protease inhibitor to inhibit protease activity and the mixture were treated by high-flux tissue grinding machine for 3 times,180s each. Then, the non-contact cryogenic sonication was performed for 30 min. After centrifugation at 16000g at 8°C for 30min, the concentration of protein from the supernatant collected was determined by Bicinchoninic acid (BCA) method by BCA Protein Assay Kit（Thermo Scientific). Protein quantification was performed according to the kit protocol. After protein quantification, SDS-PAGE electrophoresis was performed.

3.2. Protein digestion

100 μg protein re-suspended with Triethylammonium bicarbonate buffer(TEAB) which with the final concentration of 100mM. The mixture was reduced with Tris(2-carboxyethyl)phosphine (TCEP) which with the final concentration of 10mM at 37 °C for 60min and alkylated with iodoacetamide (IAM) which with the final concentration of 40mM at room temperature for 40min in darkness. After centrifugation at 10000g at 4°C for 20min, the pellet was collected, which re-suspended with 100 μL Triethylammonium bicarbonate buffer(TEAB) which with the final concentration of 100mM. Trypsin was added at 1:50 trypsin-to-protein mass ratio and incubated at 37 °C overnight.

3.3. Peptide desalting and quantification

After trypsin digestion, the peptides were dried by vacuum pump. Then, the enzymatically drained peptides were re-solubilized with 0.1% trifluoroacetic acid (TFA), and the peptides were desalted with HLB and dried by vacuum concentrator. Finally, the peptides were quantified using the NANO DROP ONE(Thermo Scientific) by UV absorption value .

3.4. DIA mass detection

Based on peptide quantification results, the peptides were analyzed by an VanquishNeo coupled with an Orbitrap Astral mass spectrometer (Thermo, USA) at Majorbio Bio-Pharm Technology Co. Ltd. (Shanghai, China). Briefly, the uPAC High Throughptu column (75 μm×5.5 cm, Thermo, USA) was used with solvent A (water with 2% ACN and 0.1% formic acid) and solvent B ( water with 80% ACN and 0.1% formic acid). The chromatography run time was set to 8 minutes. Data-independent acquisition (DIA) data were acquired using an Orbitrap Astral mass spectrometer operated in DIA mode. The mass spectrometry scanning range was 100-1700 m/z.

3.5. Protein identification

Spectronaut software (Version 19) was used to search the DIA raw data. The parameters are as follows up : The peptide length range was set to 7-52；Enzyme cutting site was trypsin/P；The maximum missed cleavage site was 2；Carbamidomethylation of cysteines as fixed modification, and oxidation of methionines and protein N-terminal acetylation as variable modifications；Protein FDR≤0.01，Peptide FDR≤0.01，Peptide Confidence ≥99%，XIC width≤75ppm. The protein quantification method was MaxLFQ.

3.6. Statistical analyses

Bioinformatic analysis of proteomic data was performed with the Majorbio Cloud platform (https://cloud.majorbio.com). P-values and Fold change (FC) for the proteins between the two groups were calculated using R package "t-test". The thresholds of fold change (>1.2 or <0.83) and P-value <0.05 were used to identify differentially expressed proteins (DEPs). Functional annotation of all identified proteins was performed using GO (http://geneontology.org/) and KEGG pathway (http://www.genome.jp/kegg/). DEPs were further used to for GO and KEGG enrichment analysis. Protein-protein interaction analysis was performed using the String v11.5 .

Morphological experiment

4.1. Transmission Electron Microscopy (TEM)

4.1.1. Sample Preparation: Bacterial Cell Collection and Pre-fixation

Take 1 mL of bacterial suspension, centrifuge to collect the suspended bacteria. Add a mixed fixative solution containing 4% paraformaldehyde and 2.5% glutaraldehyde, let it stand for suspension fixation for 5 minutes. Then, centrifuge at a high speed of 12,000 rpm for 10 minutes, and discard the supernatant. After discarding the supernatant, the volume of the precipitate should be about the size of half a millet grain. Slowly add the TEM fixative solution along the tube wall, and do not disperse the bacterial cells.

4.1.2 Post-fixation

Wash the bacilli with ultrapure water 3 times, 10 minutes each time. Fix the bacilli with 1% osmium tetroxide for 1–2 hours, then wash them again with ultrapure water 3 times, 10 minutes each time.

4.1.3 Dehydration

Perform gradient dehydration using ethanol with the concentration gradient of 30% → 50% → 70% → 90% → 100% (replace with fresh 100% ethanol 3 times). Dehydrate for 15 minutes at each concentration.

4.1.4 Conductive Treatment

Use a pipette to drop the bacilli onto a silicon wafer, adhere the wafer to the sample stage with conductive adhesive, and place the stage into an ion sputter coater for gold sputtering.

4.1.5 Image Acquisition

Image acquisition of the samples was performed using a JSM-IT700HR scanning electron microscope (SEM) manufactured by JEOL (Japan Electron Optics Laboratory). For each sample, first observe the overall view of the sample under low magnification, then select the area to be observed for image acquisition to examine the specific morphology.

4.2. Scanning Electron Microscopy (SEM)

4.2.1 Sample Preparation

Transfer 1 mL of bacterial suspension into a 1.5 mL centrifuge tube, centrifuge at 3,000 rpm for 5 minutes, and retain the bacterial precipitate (which should be about the size of half a millet grain or a dense white block-like precipitate). Add 3% glutaraldehyde along the tube wall for suspension fixation.

4.2.2 Fixation

The samples are pre-fixed with 2.5% glutaraldehyde (fixed by the client), followed by re-fixation with 1% osmium tetroxide.

4.2.3 Dehydration

Perform gradient dehydration using acetone with the concentration gradient of 30% → 50% → 70% → 80% → 90% → 95% → 100% (replace with fresh 100% acetone 3 times).

4.2.4 Infiltration and Embedding

Infiltrate the samples sequentially with a mixture of dehydrating agent (acetone) and Epon-812 embedding agent at volume ratios of 3:1, 1:1, and 1:3 respectively. Embedding: Embed the samples using pure Epon-812 embedding agent.

4.2.5 Ultrathin Sectioning

Use an ultramicrotome to prepare 60–90 nm ultrathin sections, then fish out the sections onto copper grids.

4.2.6 Staining

First stain the sections with uranyl acetate for 10–15 minutes, then stain with lead citrate for 1–2 minutes. All staining steps are performed at room temperature.

4.2.7 Image Acquisition

Image acquisition of the copper grids was conducted using a JEM-1400FLASH transmission electron microscope (TEM) manufactured by JEOL (Japan Electron Optics Laboratory). For each copper grid, first observe all bacteria under low magnification, then select the target areas for image acquisition to examine specific pathological changes.

Fermentation Experiment

1. Activation: Take out the glycerol stock of strain SY-0058-A3 from the -80℃ refrigerator, thaw it on ice, streak and inoculate it onto LB medium, and culture overnight at 37℃.

2. Seed Solution Preparation: Inoculate the strain from the plate into 6 tubes containing 3ml LB medium each, and culture overnight at 37℃ with 250rpm shaking.

3. Inoculation: Measure the OD600 of the seed solution. Inoculate into 30ml medium at an OD amount of 0.1, and culture at 37/45/55℃ with 250rpm shaking.

4. Measurement: Measure OD600 every hour until the biomass stabilizes.

5. Growth Curve Plotting: Plot one growth curve for each temperature (37, 45, and 55℃) and calculate the specific growth rate for each.

6. Sampling: Based on the growth curves and specific growth rates under different temperature conditions, sample when the strain reaches the 4th passage (OD600=1.6).

7. Sample Processing: For each omics analysis of each sample, take 10mL bacterial solution into a centrifuge tube, centrifuge at 3000g for 5min at 4℃, and discard the supernatant. Add 1ml pre-cooled PBS, transfer to a 1.5ml centrifuge tube, rinse, then centrifuge at 3000g for 5min at 4℃ and remove the supernatant. Quick-freeze with liquid nitrogen for 0.5 hours and store in a -80℃ refrigerator.

1. Sampling

The samples were collected from four hot springs in Kangding City, Ganzi Tibetan Autonomous Prefecture, Sichuan Province, including Jinjiaheba (54℃), Zhonggu (60℃), Zheduotang (45℃/55℃) and Guanding (60℃/75℃). The carrying tools were sterile vacuum bags, sterile centrifuge tubes, sterile gloves, disinfectants (70% ethanol), marker pens, thermometers, samplers ( sterile spoons, syringes ). The samples were mainly composed of silt at the bottom of the hot spring, hot spring water and soil around the spring mouth. After arriving in the laboratory, the samples were stored in the chromatography cabinet at 4℃ for a long time.

Table.1 Statistics of hot spring sampling results in western Sichuan

2. Metagenomics Analysis

2.1 Data Quality Analysis

The quality test analysis of the metagenomic data showed that the median quality of the four samples was in the high-quality interval, indicating that the quality of the metagenomic data was excellent and could be used for subsequent analysis.

2.2 Analysis of overall community characteristics

Combined with the comprehensive analysis of various indexes of Alpha diversity, temperature will have a direct impact on species. High temperature will lead to the death of heat-sensitive species ( Sobs, Chao decline ), and thermotolerant bacteria will dominate through adaptive advantages ( Simpson rise, Shannon decline ). At the same time, high temperature will eliminate microenvironment differences and lead to community structure convergence. Because the order of temperature sensitivity is evenness > species abundance > comprehensive diversity, it shows that high temperature first destroys the balance of species distribution, then reduces the number of species, and finally reduces the overall diversity.

The hierarchical clustering of species showed that samples in similar temperature ranges showed high similarity in species composition, and the clustering branches were close, indicating that temperature was the core driving factor of community assembly. When the temperature difference exceeded 15℃, the branch distance between samples increased significantly, and the cross-temperature color blocks in the heat map were distinct (45℃ vs. 75℃), indicating that the temperature gradient would directly lead to the differentiation of community structure.

The analysis of the species PCoA map based on the NR database found that the high-temperature community was significantly differentiated, and the highest temperature group (75℃) and the sub-high temperature group (60℃) were significantly separated on the PCo1 axis, indicating that the increase in temperature led to a dramatic reconstruction of species composition. Temperature gradient dominated the difference, and PCo1 explained 64.4 %, which was much higher than PCo2 ( 20.96 % ), indicating that temperature was the core driving factor of community structure differentiation.

Analysis of the functional PCoA map based on the KEGG database found that the functional composition of high-temperature samples was highly specific. The high-temperature group (75℃) was far away from other groups on the PCo1 axis, and the function was temperature-sensitive. The interpretation degree of PCo1 was 79.24 %, indicating that the response of functional metabolism to temperature was more sensitive than species composition ( 64.4 % compared to species PCo1 ).

Based on this, the analysis of the overall community characteristics is summarized:

a) Environmental screening effect: High temperature can lead to a decrease in Alpha diversity ( Shannon index ) and an increase in Beta diversity ( PCoA separation ), which together confirm that temperature is the main factor of community assembly.

b)The sensitivity of functional metabolism to temperature ( PCo1 79.24 % ) was higher than that of species composition ( PCo1 64.40 % ), indicating that functional conservation was the core strategy of high temperature adaptation.

c)Transformation of community assembly mechanism: After the threshold was crossed, the community changed from the balanced maintenance state of the diversity community to the conservative mode of the core module of the dominant community, and the thermoduric bacteria and key pathways dominated the niche.

2.3 Species and functional composition analysis

According to the species percentage histogram, the species with survival advantages at the phylum / class / order / family / genus level and their functions are summarized as follows:

Table 2 Distribution table of dominant species at all levels

According to the abundance heat map analysis at the phylum level and the genus level, it was found that temperature would drive the abundance level, and 75 °C was the enrichment peak of sulfur metabolism genus. The abundance of sulfur metabolism genus reached the highest at 75 °C, and high temperature significantly screened sulfur metabolism functional groups. The order of abundance is Sulfurihydrogenibium > Thermus > Thermodesulfovibrio, indicating that the sulfur oxidation pathway dominates the high temperature sulfur cycle.

In the analysis according to the KO functional pathway, the KO pathway related to polyamine synthesis was searched. In the top 100 abundance pathways, the abundance of spermidine synthase (K00797) was found to be significantly higher at 75℃ than in the low temperature group (45℃/55℃/60℃), indicating that the polyamine synthesis process is more frequent in the high temperature environment.

The KO pathway of polyamine synthesis-related enzymes was queried, and the high-temperature group and the medium-low temperature group were compared in pairs. It was found that spermidine synthase ( K00797 ), S-adenosylmethionine decarboxylase ( K01611 ), aspartate kinase ( K00928 ), and aspartate semialdehyde dehydrogenase ( K00133 ) were highly abundant enzymes, and the expression of these pathways in the high-temperature group (75℃) was mostly higher than that in the medium-low temperature group (45℃/55℃/60℃), which further verified the importance of polyamine metabolism for the survival of microorganisms in high-temperature environments and its protective effect.

2.4 Polyamine synthase gene mining

The KO numbers corresponding to the following enzymes were searched for the metagenomic analysis proteins annotated in the KEGG database : arginine decarboxylase ( K01586 ), ornithine decarboxylase ( K01581 ), S-adenosylmethionine decarboxylase ( K01611 ), spermidine synthase ( K00797 ), aspartic kinase ( K00928 ), aspartic semialdehyde dehydrogenase ( K00133 ), and then mapped to the NCBI gene number and downloaded from NCBI to construct a phylogenetic tree.

Table 3 Statistical table of polyamine synthase mining results

The structure of the enzymes of the molecular mining results was compared with the following reference substances and their sequences.

1)Arginine decarboxylase_7P9B_ pdb_00007p9b(Providencia stuartii);

2)Ornithine decarboxylase_1QU4_ pdb_00001qu4(Trypanosoma brucei);

3)SAM decarboxylase_1TLU_ pdb_00001tlu(Thermotoga maritima);

4)Spermidine synthase_7XIF_ pdb_00007xif(Pyrobaculum calidifontis);

5)Aspartate Kinase_3L76_ pdb_00003l76(Synechocystis sp. PCC 6803);

6)Aspartate Semialdehyde Dehydrogenase_8JUO_pdb_00008juo(Porphyromonas gingivalis).

Table 4 Statistical table of comparative results of polyamine synthase structure

3. Multi-omics Analysis

3.1 Transcriptomics Analysis

To understand how thermotolerant Bacillus responds to elevated temperature at the molecular level, transcriptomic analysis was performed under different temperature conditions. This analysis aimed to identify key regulatory pathways and genes involved in heat stress adaptation, focusing on changes in metabolism, protein homeostasis, and signal regulation.

Differential gene expression profiling revealed that temperature increase caused broad transcriptional reprogramming. Genes associated with core metabolic processes, protein quality control, and signal transduction showed significant expression shifts, indicating a coordinated cellular response to heat stress. Functional enrichment and KEGG pathway analysis were used to classify the affected pathways and visualize systemic adaptations.

Heat stress triggers systemic responses across three dimensions: metabolism, protein homeostasis, and signal regulation.（总结）

The temperature increase significantly impacted:

a) Metabolic reprogramming: Genes related to carbohydrate, amino acid, nucleotide, and lipid metabolism were markedly upregulated, suggesting enhanced energy turnover and biosynthetic activity to maintain cellular functions under stress.

b) Protein homeostasis: Pathways involved in translation, protein folding, and degradation (including chaperones and proteases) were strongly activated, reflecting the need to counteract protein denaturation and aggregation caused by high temperature.

c) Membrane integrity and signaling regulation: Genes involved in membrane transport, signal transduction, and stress-response regulation were significantly upregulated, indicating enhanced communication and defense mechanisms to maintain cellular stability.

3.2 Proteomics Analysis

To complement the transcriptomic findings and verify cellular adaptation at the protein level, proteomic analysis was performed under different temperature conditions. The goal was to reveal how heat stress reshapes the proteome and identify key adaptive strategies related to metabolism, protein stability, and signaling regulation.

Quantitative proteomic profiling revealed extensive changes in protein abundance and functional categories in response to rising temperature. Differentially expressed proteins were classified into major biological processes and KEGG pathways to illustrate global trends in thermal adaptation.

Proteomic differences highlighted three core adaptive axes: metabolic reprogramming, maintenance of protein quality, and regulation of stress signaling.

a) Cellular Processes: Proteins associated with collective behavior and motility in prokaryotes were notably affected. High temperatures may impair or remodel motility-related systems such as flagella, reflecting altered group adaptability and surface attachment mechanisms under stress.

b) Environmental Information Processing: Pathways related to membrane transport and signal transduction were significantly upregulated. This suggests that Bacillus adapts to thermal stress by enhancing transmembrane transport efficiency and improving environmental signal recognition, enabling rapid perception and response to temperature fluctuations.

c) Metabolic Processes: Carbohydrate metabolism was the most enriched pathway, indicating major adjustments in energy acquisition and carbon utilization under high-temperature conditions. Other highly represented pathways included nucleotide, lipid, and energy metabolism, demonstrating comprehensive remodeling of cellular bioenergetics.

3.3 Metabolomics Analysis

To explore how metabolic processes respond to heat stress and sustain cellular adaptation, metabolomic profiling was conducted under different temperature conditions. This analysis aimed to reveal how high temperature influences energy metabolism, redox balance, and stress defense mechanisms at the metabolite level.

Comparative metabolomic analysis identified significant differences in metabolite abundance and pathway enrichment between high- and low-temperature groups. KEGG pathway mapping and compound classification revealed distinct metabolic signatures associated with thermal adaptation and stress resilience.

Heat stress reprograms cellular metabolism across multiple layers, promoting energy redistribution and metabolic flexibility to ensure survival.

a) Energy and Nucleic Acid Metabolism: In the high-temperature control group, nucleic acid–related metabolites were the most abundant, indicating that high temperatures primarily affect energy turnover, nucleotide synthesis and degradation, and amino acid cycling. This reflects the increased energy demand and biosynthetic activity required for cellular maintenance under stress.

b) Redox Balance and Cofactor Regulation: The significant enrichment of vitamins and cofactors suggests active regulation of redox reactions, electron transport chains, and antioxidant systems under high-temperature stress, contributing to the stabilization of intracellular oxidative balance.

c) Stress Response and Defense Mechanisms: Smaller molecules and antibiotic-related metabolites were relatively scarce, which may indicate that the cell reallocates resources from secondary metabolite production to essential defense and repair processes.

4. Morphological Analysis (SEM+TEM)

To visually examine the structural and morphological adaptations of Bacillus cells under different temperature conditions, scanning electron microscopy (SEM) and transmission electron microscopy (TEM) were employed. The aim was to reveal how rising temperature affects cell morphology, integrity, and ultrastructure, providing direct evidence of cellular adaptation and heat stress damage.

Cells cultivated at 37℃, 45℃, and 55℃ were observed under SEM and TEM. SEM provided surface morphology and cell shape information, while TEM allowed visualization of internal ultrastructural changes. Together, these observations illustrated how cells remodel their morphology to withstand increasing thermal stress.

With rising temperature, Bacillus cells underwent clear morphological transitions, indicating both adaptive changes and stress-induced damage:

a) 37℃: Cells appeared as short rods with smooth surfaces, uniform morphology, and intact internal structures, representing normal growth and stable cellular function.

b) 45℃: Cells became elongated and more dispersed, showing noticeable morphological variation. This elongation and increased spacing indicate enhanced metabolic activity and adaptive remodeling under moderate heat stress.

c) 55℃: Cells exhibited pronounced elongation or distortion, with localized vacuoles and partial structural damage. These alterations are likely caused by increased membrane fluidity, protein denaturation, and nucleic acid instability at high temperature. Despite these stresses, most cells retained overall surface integrity, demonstrating strong thermotolerance.

Microscopic observations confirm that high temperature induces both morphological adaptation and physiological stress. Moderate heat promotes cell elongation and metabolic activation as adaptive strategies, while extreme heat leads to structural deformation and partial damage. These findings visually support the multi-omics results, showing that thermophilic Bacillus achieves heat resistance through coordinated metabolic, structural, and molecular adjustments.

5. The Role of Polyamines

Genes associated with polyamine synthesis were all upregulated, though to a limited extent, suggesting that polyamines can exert stress-resistant functions even at low concentrations within organisms.

Genes associated with polyamine biosynthesis were consistently upregulated under heat stress, although the extent of upregulation was moderate. This suggests that even at low intracellular concentrations, polyamines can effectively contribute to stress resistance by stabilizing cellular components and maintaining physiological balance.

Enzymes involved in the synthesis of polyamine precursors—particularly ornithine—also showed significant transcriptional activation. The conversion of arginine to ornithine was strongly upregulated, highlighting the increased supply of substrates essential for polyamine synthesis. Both the aminopropyl donor and acceptor pathways were enhanced, reinforcing the idea that polyamine metabolism is tightly integrated into the cellular stress response network.

These findings indicate that heat stress not only triggers the upregulation of polyamine synthases but also stimulates precursor pathways, ensuring sustained polyamine production. Together, these regulatory changes underscore the vital role of polyamines in enhancing cellular stability, protecting macromolecules, and promoting adaptation under high-temperature conditions.