Sequence, Structure, and Function of DNA-Binding Protein in Deinococcus Wulumuqiensis R12

Deinococcus wulumuqiensis R12, which was isolated from arid irradiated soil in Xinjiang province of China, belongs to a genus Deinococcus that is well-known for its extreme resistance to ionizing radiation and oxidative stress. The DNA-binding protein Dps has been studied for its great contribution to oxidative resistance. To explore the role of Dps in D. wulumuqiensis R12, the Dps sequence and homologous structure were analyzed. In addition, the dps gene was knocked out and proteomics was used to verify the functions of Dps in D. wulumuqiensis R12. Docking data and DNA binding experiments in vitro showed that the R12 Dps has a better DNA binding ability with the N-terminal than the R1 Dps1. When the dps gene was deleted in D. wulumuqiensis R12, its resistance to H 2 O 2 and UV rays was greatly reduced, and the cell envelope was destroyed by H 2 O 2 treatment. Additionally, the qRT-PCR and proteomics data suggested that when the dps gene was deleted, the catalase gene was signicantly down-regulated in cells. And the proteomics data indicated the metabolism, transport and oxidation-reduction processes in D. wulumuqiensis R12 were down-regulated after the deletion of dps gene. Dps protein might play an important role in Deinococcus wulumuqiensis R12. in this study, we constructed a Δdps R12 mutant of strain R12 through homologous recombination and subjected it to UV and H 2 O 2 -induced oxidative stress to investigate the properties and protective function of Dps in D. Wulumuqiensis R12. Finally, proteomics analysis was utilized to examine the role of Dps on a wider, systemic level. The results showed that after the deletion of dps gene, the antioxidative capacity was reduced, and the envelope was more easily destroyed after H 2 O 2 treatment. Finally, the proteomic data revealed that the dps gene and Dps protein might affect metabolism, transport, and cell wall etc.. and comparative proteomics was used to investigate the functions of Dps in D. wulumuqiensis R12. The qRT-PCR and proteomics data suggested that when the dps gene was knocked out in D. wulumuqiensis R12, the catalase gene was down-regulated. The proteomics data also suggested that that the metabolism, transport and oxidation-reduction processes were down-regulated in D. wulumuqiensis R12 after the dps gene was knocked out.


Introduction
In 1956, a bacterium that had survived exposure to an extremely high dose of ionizing radiation (IR) was accidentally discovered as a contaminant in a can of supposedly to sterilized meat [1]. Now well-known as Deinococcus radiodurans, it is one of the most radiation-resistant organisms known to science [2][3]. It is not only tolerant to gamma radiation, but also to other DNA damage and oxidative stress-generating conditions such as UV, desiccation, or high temperature [4][5][6]. The radiation tolerance of D. radiodurans can reach to 15,000 Gy [7], which is 100-fold that of typical microorganisms, 250-fold that of Escherichia coli, and 3000-fold that of humans [8][9]. Therefore, D. radiodurans is an ideal model strain for studying the oxidative stress response and radiation resistance [10][11].
Several studies have investigated the remarkable oxidative resistance mechanisms of this bacterium, which can be divided into three categories: DNA self-repair [4]; e cient cell evolution mechanism [10]; and effective scavenging of reactive oxygen species (ROS) [11][12][13]. The DNA repair system of D. radiodurans has been identi ed as a major determinant of its IR resistance. Intermediate dose of IR can cause death to most cells due to numerous double strands broken [4]. However, studies have shown that the protein damage is just as important as the DNA damage following exposure to IR. Daly et al. also put forward the viewpoint of proteomes are important macromolecule that can be affected by IR [14]. The powerful antioxidative system is another determinant of stress resistance in D. radiodurans [4]. To adapt to the oxygen-enriched environment of irradiated soil and remove the resulting ROS, microorganisms have evolved various ROS scavenging systems [4,10]. Notably, D. radiodurans has several unprecedented antioxidative systems to protect itself from oxidative stress that are not found in other microorganisms [4]. Among them, DNA-binding protein (Dps), a conserved protein found in most bacterial species, has been devoted a great deal of attention, since it is a vital factor that protects DNA from various oxidative damage in stressed or starved cells [3,[15][16].
Recent studies have painted a clearer picture of the two mechanisms through which Dps exerts its protective in cells: (1) Dps can effectively bind DNA, thereby protecting DNA from the attack of oxygen free radicals [17]; (2) The ferroxidase activity is the key feature of Dps that prevents the formation of highly toxic ROS from the reaction of iron (II) with hydrogen peroxide or dioxygen [18]. Dps can also be oxidized to protect DNA from a distance by DNA charge transfer (CT), which may be another effective DNA protection mechanism [19]. The dps gene is also critical for cell survival under stress conditions [20]. When the dps gene was knocked out in Salmonella enterica, the mutant was more sensitive to antibiotics than the parental strain [21]. Similarly, the Δdps mutant of Riemerella anatipestifer, was more sensitive to H 2 O 2 under iron-rich conditions [22]. However, there are no studies have shown changes in the antioxidant capacity of Δdps mutant in genus Deinococcus.
The genus Deinococcus is interesting because many of its members exhibits extreme radioresistance [23]. D. radiodurans R1 was the rst strain in this family isolated from canned meat which shows extraordinary resistance to IR [24][25]. D. Wulumuqiensis R12 is a member of Deinococcus family which was previously isolated by our team from a radiation-contaminated area of Xinjiang Uigur Autonomous Region of northwest China. Strain D. Wulumuqiensis R12 is a Gram-positive, reddish orange, non-sporeforming coccus, and its gamma radiation resistance was more than 10 kGy and UV resistance was over 700 J m −2 [26]. The D. Wulumuqiensis R12 shows higher tolerance for gamma radiation and UV light than that of in D. radiodurans R1, and the genome of R12 was also sequenced in our previous work [26]. The R12 genome revealed a single dps gene of 645 bp size. Dps is a classic antioxidant protein in organisms. With the aim of shedding some light on the role of Dps in extremophilic genus Deinococcus (strain R12), in this study, we constructed a Δdps R12 mutant of strain R12 through homologous recombination and subjected it to UV and H 2 O 2 -induced oxidative stress to investigate the properties and protective function of Dps in D. Wulumuqiensis R12. Finally, proteomics analysis was utilized to examine the role of Dps on a wider, systemic level. The results showed that after the deletion of dps gene, the antioxidative capacity was reduced, and the envelope was more easily destroyed after H 2 O 2 treatment. Finally, the proteomic data revealed that the dps gene and Dps protein might affect metabolism, transport, and cell wall etc..
Model building of R1 / R12 Dps and protein/DNA docking The R12 Dps structure was modeled using RoseTTAFold, and the N-terminal of R1 Dps structure (PDB code: 2C2F) was repaired using the same method. The structures were presented and analyzed using PyMol.
The DNA model was downloaded from PDB. The protein/DNA docking was carried out using Autodock 4.2.6, and the complex with lowest energy was obtained.

Plasmid construction
The primers used in this study are listed in Table S1  The cells from 50 mL culture were harvested by 8,000 ×g centrifugation for 5 min, and the discarded the supernatants. Then, the cells were resuspended with 3 mL PBS buffer. The cells were disrupted by 300 W sonication for 15 min, and centrifuged at 10,000 ×g for 20 min at 4 °C. The supernatants were obtained and loaded onto 1 mL Ni-NTA resin to purify proteins that was pre-equilibrated with 5 mL buffer A (pH 8.0) that contains 20 mM imidazole and 300 mM NaCl. Then 5 mL buffer A was used to remove miscellaneous proteins. Finally, 3 mL buffer B (pH 8.0) contains 300 mM imidazole and 300 mM NaCl was used to elute the target proteins. The supernants and puri ed proteins were analyzed by SDS polyacrylamide gel electrophoresis (SDS-PAGE), and the protein concentration was measured ultramicro spectrophotometer (Colibri, Germany).

Construction and screening of the Δdps knockout strain
The positive recombinant colonies grown on NA plates containing kanamycin were transferred into TGY liquid medium without antibiotics, and cultured for 2-3 days at 31 °C. Then, 500 μL of the resulting culture were plated onto TGY solid medium containing 10% sucrose and cultured at 31 °C for 3-5 days. Single colonies without bacteriolysis on sucrose TGY solid medium were streaked onto TGY solid medium with and without kanamycin, respectively. Then, clones that grew on TGY solid medium without kanamycin and did not grow on TGY solid medium with kanamycin were transferred into TGY liquid medium without antibiotics and cultured for 2-3 days at 31 °C. Finally, the genomic DNA was extracted using a commercial kit (Takara, china), and the Δdps R12 mutant was veri ed by PCR ampli cation and sequencing.
The DNA protection of R12 Dps and R1 Dps1 protein The pET-22b-R12-Dps and pET-22b-R1-Dps1 recombinant plamids were constructed, the used primers were listed in Tablse S1. The 40 μM puri ed R12 Dps and R1 Dps1 protein in double distilled water and crosslinker uid (pH 8.0 20 mM phosphate buffer with 80 mM NaCl and 0.1% glutaraldehyde) incubated in room temperature for 30 min, respectively. Then, mix 10 μL pET-22b plasmids (60 ng/μL) and 10 μL 40 μM R12 Dps or R1 Dps1 protein were incubated in room temperature for 30-60 min. The 30 ng/μL plamisds were using as control. Finally, agarose gel electrophoresis was utilized to verify the protection of Dps proteins.

Validation of calatase expression by qRT-PCR
The WT R12 strain and Δdps R12mutant was grown in TGY medium at 31 °C for three days. The 4 mL cells were harvested by centrifugation at 12000 ×g for 3 min. The total RNA was used bacteria total RNA isolation kit (Sangon, Shanghai). The cDNA was obtained using one-strp gDNA removal and cDNA synthesis supermix (TransGen Biotech, Beijing). The real-time PCR was carried out in Roche LightCycler96 real-time uorescence quantitative PCR instrument. 16S rRNA was used as internal reference gene. The primers using in this test listed in Table S1. Excel was used to calculate student's ttest.
Growth curve analysis of WT R12 strain and Δdps R12 mutant The WT R12 strain and Δdps R12mutantatthe logarithmic growth stage were transferred into 50 mL of TGY liquid medium, an inoculation amount of 2%, after which the OD 600 value was measured every 2 h.
Each sample was measured in triplicate and the mean value was recorded.
Survival rate of WT R12 strain and Δdps R12 mutant under oxidative stress The Δdps R12mutant was cultured for 48 h in TGY liquid medium at 31 °C. The resulting seed culture was used to inoculate, fresh TGY medium to an initial OD 600 of 0.6-0.8. Next, the cells were treated with 80 mM H 2 O 2 for 0, 10, 20, 30, and 50 min at 31 °C and 800 rpm inheating block. After the stress treatment, 100 μL (10 9 CFU) aliquots of serial 10-fold dilutions (10 -1 -10 -5 ) of cells were plated onto TGY agar and grown at 31 °C for 3 days. The survival rate was calculated based on the number of colonies in the treated samples compared with the untreated sample (control group). The 10 -5 dilution was plated onto TGY agar to calculate the number of colony-forming units (CFU) in triplicate.
Survival rate of WT R12 strain and Δdps R12 mutant after exposed to UV irradiation The ΔdpsR12mutant was cultured to the stationary stage and diluted in a 10 6 -fold gradient. Then, 200 μL (10 6 CFU) of the diluted cell suspension were plated onto TGY agar, exposed to 0, 3, 6, 9 and 12 min 700 J m -2 UV irradiation, and cultured for 2-3 days at 31 °C. The survival rate was calculated after counting the the colonies. Each group included three independent repeats. The WT R12 strain was included under the same conditions as the control group.

Transmission electron microscopy (TEM)
The WT R12 strain and Δdps R12 mutant were grown in TGY liquid medium at 31 °C from starting OD 600 value reached to 0.6-0.8, at which point they were treated with 80 mM H 2 O 2 for 30 min 31 °C. Then, the cells were washed twice with PBS and collected by centrifugation at 12,000 ×g for 2 min. The cells were xed overnight at 4 °C with 2.5% glutaraldehyde, harvested by centrifugation at 4000 ×g for 5 min, and embedded in 2% agarose. The slices were stained with uranyl acetate for 15 min and observed under a Hitachi H-7650 transmission electron microscope.

Protein extraction and digestion
The 200 μg samples (which the strains grown to OD 600 0.8 were used as samples) were shock frozen -80°C . A proper of cells was weighed into a pre-cooled mortar, and liquid nitrogen was added. Then, 4 times the volume of lysis buffer (1% Triton X-100 and 1% protease inhibitor) was added to each sample. The cells were disrupted by sonication (240 W) for 15 min. After centrifugation at 12000 ×g for 10 min, the supernatant was transferred to a fresh centrifuge tube and the protein concentration was determined using a BCA assay kit (Sangon, Shanghai, China). An equal amount of each sample was subjected to enzymatic hydrolysis, and the volume was adjusted with lysis buffer. Then, 1 volume of pre-cooled acetone was added and vortexed, after which 4 times of pre-cooled acetone was added. The precipitation took place at -20 °C for 2 h. After centrifugation at 4,500 ×g for 5 min, the supernatant was discarded and the precipitate was washed twice with pre-cooled acetone. After drying and precipitation, TEAB with a nal concentration of 200 mM was added, followed by ultrasonic dispersion of the precipitate. Trypsin was added at a ratio of 1:50 (protease: protein, m/m) and enzymatic hydrolysis was conducted overnight. DTT was added to a nal concentration of 5 mM and incubated at 56 °C for 30 min. Then, IAA was added to a nal concentration of 11 mM and incubated at room temperature in the dark for 15 min. Finally, the peptides were desalted by C18 SPE column. Each strain was acquired for triplicate, and each sample was treated as described above.

Liquid Chromatography-Mass Spectrometry/Mass Spectrometry (LC-MS/MS) analysis
The peptide segments were separated on NanoElute in UPLC (Bruker Daltonics) mobile phase A and then separated using a NanoElute ultra-high performance liquid phase system. Mobile phase A was water with 0.1% formic acid and 2% acetonitrile. Mobile phase B was acetonitrile with 0.1% formic acid and 100%.
The mobile phase gradient settings were as follows: 0-70 min, 4-22% B; 70-84 min, 22-30% B; 84-87 min, 30-80% B; 87-90 min, 80% B. The ow rate maintained at 450 nL/min. The peptide segments were separated by UPLC and then ionized by injecting into the capillary ion source. The peptide segments were analyzed using a timsTOF Pro (Bruker Daltonics) mass spectrometry instrument. Mass spectrometry data was acquired by Bruker Compass HyStar (Version 5.1.8.1), and analyzed by MaxQuant 1.6.6.0. The voltage of the ion source was set at 1.75 kV, and the parent ion of the peptide segments and its secondary fragments were detected and analyzed using high-resolution TOF scanning. The scanning range of secondary mass spectrometry was set to 400-1500 m/z. The data acquisition was conducted in parallel cumulative serial fragmentation (PASEF) mode. After one set of rst-order mass spectrometry data was collected, secondary spectrographs with the charge of the parent ion in the range of 0-5 were collected in PASEF mode 10 times. The dynamic elimination time of tandem mass spectrometry scanning was set to 30 s to avoid repeated scanning of the parent ion.
In our results, at least one razor or unique peptide of a protein was to be considered as identi ed, the minimum score of peptides was set as 40, and false discovery rate (FDR) was set as 1% to ensure the identities are authentic. FDR is a measure of the incorrect peptide spectral matches (PSMs) among all accepted PSMs [27][28][29]. Proposed by Benjamini and Hochberg [30] as an alternate to the Bonferroni correction, it is de ned as the rate of false positives in accepted hits. FDR is a less stringent metric for global con dence assessment. In the context of proteomics, it is a global estimate of the false positives present in the results obtained by a database search algorithm [31].

Differential protein screening
Protein difference analysis rst picks out the samples to be compared, calculates the quantitative mean protein of the repeated samples, and nally the difference multiple of the comparison group was calculated. The calculation formula is showed in . In order to judge the signi cance of the difference, Ttest was carried out for the relative quantitative value of each protein in the two comparison samples, and the corresponding P value was calculated, which was taken as the signi cance index. The default P value was ≤ 0.05. To make the test data conform to the normal distribution of the T-test requirements. Before testing, the relative quantitative values of proteins need to undergo Log2 logarithmic conversion. The calculation formula is showed in β. Through the above difference analysis, when P value ≤ 0.05, the change of differential expression level over 1.5 was regarded as the change threshold of signi cantly upregulated, and less than 1/1.5 was regarded as the change threshold of signi cantly down-regulated. The summary data of all differentially expressed proteins in this project are shown in "MS_identi ed_information" (Supporting information). A and B were the sample, R represents the relative quantity of protein, i represents the sample, and k represents the protein.

GO and KEGG enrichment analyses
The GO and KEGG enrichment analyses were performed using Blast2go (version 5.2, Biobam, Valencia, https://www.blast2go.com/) and DAVID (version 6.8, https://david.ncifcrf.gov), respectively. GO terms and KEGG pathways with corrected P-values of less than 0.05 were considered to be signi cantly enriched for differentially expression proteins.

Protein-protein interaction prediction
The number of sequences of the differentially expressed-proteins screened according to the fold change over 1.5 in different comparison groups was compared with the protein network of the STRING database (v.11.0), and the interaction of the differentially expressed proteins were extracted according to the con dence score > 0.7 (high con dence). The "Cytoscape" software was used to visualize the interaction network of differentially expressed proteins.

Results And Discussion
Sequence and structure analyses of D. wulumuqiensis R12 Dps To explore the function of D. wulumuqiensis R12 Dps, the sequence and structure were analyzed. Through sequence alignment, Dps1 from Deinococcus radiodurans R1 (PDB code: 2C2F and the Nterminal was missing) with the highest identity (81.31%), and the alignment between R12 Dps and R1 Dps1 sequences was carried out. The alignment result showed that the N-terminal sequences of R1 Dps1 nad R12 Dps were unconservative, that might play an important role in function differences. The recent research articles have reported that Dps proteins from E. coli have identi ed the N-terminal lysine residues are involved in DNA binding [32].
In order to verify the function of R12 Dps protein N-terminal. The R12 Dps model was built through RoseTTAFold. The N-terminal of R1 Dps1 crystal structure was too exible that can not be analyzed, therefore, RoseTTAFold was used to repair the structure. Furthermore, the protein/DNA docking was carried out through AutoDock 4.2.6. The docking result showed that the DNA binding ability of R12 Dps was better than that of R1 Dps1. The Lys3, Ser8, Ser20, Lys20, Asp22, Ser129 and Ala132 residues bond to DNA by hydrogen bonds in R12 Dps/DNA docking result. And in R1 Dps1/DNA docking result, only Thr22 bonds to DNA.

Puri cation of Dps proteins, DNA binding and protection tests
The docking results showed that R12 Dps DNA binding ability was better than that of R1 Dps1. The DNA binding tests need to verify this computational result. The dps gene was ampli ed from D. wulumuqiensis R12 genome, and the recombinant plasmid pET-22b(+)-R12Dps was constructed and expressed in E.coli BL21 (DE3). The same method was utilized to obtain the pET-22b(+)-R1Dps1 plasmid which the template was D. radiodurans R1 genome. The SDS-PAGE result (Fig. 2C) showed that R1 Dps1 and R12 Dps proteins can be successfully expressed and puri ed in vitro. The molecule weight of R1 Dps1 was 24 kDa, and the R12 Dps was ~25 kDa, which were consistent with the theoretical molecular weights.
As shown in Fig. 2A, the plasmids (pET-22b) appeared superhelical in vitro. The superhelical level of plasmids was reduced with the addition of R1 Dps1 protein or R12 Dps proteins. Compare with the superhelical level between R1 Dps1 and R12 Dps proteins, the lower superhelical level the R12 Dps re ected, indicating the DNA binding ability of the R12 Dps was truly better than R1 Dps1 that con rmed to the docking data.
The Dps protein can combine Fe 2+ ions to prevent fenton reaction [4]. Therefore, 80 mM and 200 μM FeSO 4 were added to plasmids. As shown in Fig. 2B, the plasmids have been destroyed without the addition of Dps proteins, while the plasmids combined with the R1 Dps1 or R12 Dps proteins were remained few plasmids, indicating that the Dps protein can combine Fe 2+ that the fenton reaction can be prevented. Also, the bromophenol blue was oxidized to yellow under fenton reactions, and the solution which was added Dps proteins that was remain the blue color, demonstrating that the Dps protein can combine Fe 2+ .

Knockout of the dps gene in D. wulumuqiensisR12
In order to explore the speci c functions of Dps in D. wulumuqiensis R12, the dps gene was knocked out through homologous recombination. In this study, the suicide plasmid pK18mobSacB was utilized as the bifunctional screening vector with kanamycin resistance gene as a positive selection marker, and the sacB gene which encodes a secretory levansucrase as a negative selection marker. The recombinant plasmid was used to knock out the dps gene through homologous recombination in D. wulumuqiensis R12 as shown in Fig. 3A.
The initial screening was conducted on kanamycin plates to obtain the strains containing the recombinant plasmids (Fig. 3B). Then, the secondary screening was carried out, and the single colonies were selected from the sucrose NA plate and streaked onto TGY plates with or without kanamycin respectively to obtain the strains which has been cured of the recombinant plasmid without marker genes. Clones that grew on the NA plate without kanamycin but not on kanamycin were selected. The genomic DNA of the rescreened knockout and WT R12 strains as isolated, and homologous primers were used for PCR ampli cation. The genomes of wild-type R12 and Δdps R12 mutant were extracted and sequenced, the sequencing primers were homologous arm F1 and R2, the sequencing result showed that 784 bp which contain dps gene was knocked out in the Δdps R12 mutant genome (Fig. S1). Here, the target we designed to replace contains extra 139 bp on the both sides of the dps gene (645 bp) to ensure the target gene was successful knockout. Therefore, the nal fragments deleted in the genome contains totally 784 bp. Finally, the PCR products which used primers F1, and R2 (Table S1) to amplify were veri ed by sequencing, which demonstrated that the dps gene was indeed knocked out from R12 genome, resulting in the Δdps R12 mutant.
The D. wulumuqiensis R12 growth condition after deletion of dps gene Dps has been shown to play an important role during exponential phase and stationary phase growth in E. coli [3]. In this study, we want to investigate the effects ofthe D. wulumuqiensis R12 growth after the deletion of dps gene. Hence the WT R12 strain and Δdps R12 mutant growth curves were measured. As shown in Fig. 4A (Fig. 5C), and the cell envelope was seriously destroyed in Δdps R12 mutant (Fig. 5D). The unusually thick cell envelope is the rst barrier of D. wulumuqiensis R12 against external stresses when faced with a hostile environment. The antioxidant capacity of D. wulumuqiensis R12 was signi cantly reduced when the Dps proteins were not in D. wulumuqoensis R12. And previous research showed that Nterminus of DrDps2 in the D. radiodurans can interact with the membrane [33]. We hypothesised that without the protection of Dps protein, the antioxidant capacity of cell envelope would be reduced.

Difference of proteins expression between the WT R12 strain and the Δdps R12 mutant
In this study, both WT R12 strain and the Δdps R12 mutant were used in triplicate, and each sample was in logarithmic phase, and a total of 1,009 proteins were detected by spectrum search analysis. Among these 116 proteins were up, and 111 were down-regulated, with fold change > 1.5 and p value < 0.05 (Fig.  6). Thus, the numbers of up and down-regulated proteins following dps knockout were simliar. In this work, we mainly focused on a series of adverse effects caused by dps gene knockout in D. wulumuqiensis R12. Therefore, the down-regulated proteins were analyzed further.

GO analysis of differentially expressed proteins in the WT R12 strain and the Δdps R12 mutant
To further understand the functional characteristics of the differential expressed proteins, GO enrichment analyses of the categories Cellular Component, Molecular Function, and Biological Process were performed. The up-regulated proteins were enriched for the Cellular Component categories DNA-directed RNA polymerase complex, nucleoid, RNA polymerase. They were also enriched for the molecular functions 5'-3' RNA polymerase activity, ribonucleoside binding, nucleoside binding, RNA polymerase activity, DNA-directed 5'-3' RNA polymerase activity, nucleotidyltransferase activity, zinc ion binding as well as the Biological Process category hexose catabolic process, galactose metabolic process and monosaccharide catabolic process (Fig. 7A). The down-regulated proteins were enriched in the Cellular Component categories outer membrane-bound periplasmic space, cell envelope, envelope, periplasmic space. They were also enriched in the Molecular Function categories disul de oxidoreductase activity, cofactor binding, tetrapyrrole binding, heme binding, and zinc ion binding as well as the Biological Process category carboxylic acid catabolic process (Fig. 7B). Additionally, in down GO term, metabolic process, cellular process, growth, response to stimulus, interspecies interaction between organisms, localization, multi-organism process, biological regulation in Biological Process; cell, intracellular, proteincontaining complex in Cellular Component; catalytic activity, binding, antioxidant activity, and molecular carrier activity were in down-regulation (Fig. 7C). Also, in up GO term, the cellular process, metabolic process, response to stimulus, growth, biological regulation (Biological Process); cell, intracellular, proteincontaining complex (Cellular Component); and catalytic activity, binding, transporter activity, antioxidant activity, molecular function regulator, and transcription regulator activity (Molecular Function) were in upregulation (Fig. 7D). These data imply that Dps may in uence a large number of proteins with different functions.
The TEM results showed that the cell envelope of the Δdps R12 mutant was severely damaged by treatment with 80 mM H 2 O 2 . Cells of D. wulumuqiensis R12 can form tetrads, and possess a thick cell envelope (Fig. 5). The cell envelope of Deinococcus exhibits an unusual structure and composition [34], and some Deinococcus species have six layers in the cell envelope. The entire D. radiodurans cell is enveloped by a dense carbohydrate shell, a thick cellular structure that might contribute to its extreme stress resistance [34][35]. D. wulumuqiensis R12 shares similar cell-envelope characteristics, and its envelope is even thicker than that of in D. radiodurans [6]. The Δdps R12 mutant's proteins in the GO catagory cell envelope were signi cantly down-regulated, among which the ABC transporter substratebinding protein, phosphate-binding protein, and thiamine ABC transporter substrate-binding protein were in down-regulation, which might explain why the envelope was more sensitive to H 2 O 2 and damaged by the stress treatment.
To investigate the response to oxidative stress in D. wulumuqiensis R12, several crucial proteins were analyzed. In the Biological Process and Molecular Function catagories, "response to oxidative stress" GO term, such as the catalase (DVJ83_01425), and dihydrolipoyl dehydrogenase (lpdA) were signi cantly down-regulated. Catalase is a metalloenzyme that detoxi es H 2 O 2 into water and O 2 , thereby protecting organisms from oxidative damage caused by H 2 O 2 [36]. Three catalases (DVJ83_14715, DVJ83_03805 and DVJ83_01425) were identi ed in D. wulumuqiensis R12, the catalases encoded by the DVJ83_14715 and DVJ83_03805 genes were expressed under normal conditions, while catalase DVJ83_01425 was down-regulated. The proteomic data showed that the catalase DVJ83_01425 was down-regulated 2.07-fold in response to chemical, 1.91-fold in cofactor metabolic processes, 2.48-fold in response to oxidative stress, 4.14-fold in nucleoside bisphosphate metabolic process, 5.39-fold in heme binding, and 2.45-fold in cofactor binding. The catalase DVJ83_01425 was in down-regulated when the dps gene was knocked out, reducing the antioxidative capacity of D. wulumuqiensis R12. Also, we found two catalases (catalase1 and catalase2) in D. wulumuqiensis R12 genome, and the catalase2 was signi cantly down-regulated. These data indicated that the catalases would be affected with the deletion of dps gene. Whereas the H 2 O 2 tolerance was reduced not only the Dps was lost, but the catalases were down-regulated in D. wulumuqiensis R12.
The growth of the Δdps R12 mutant was signi cantly slower than that of the WT strains. ABC substratebinding transporter were signi cantly down-regulated. ABC transporters (ATP-binding cassette transporter) are one of the largest and oldest protein families, which plays a crucial role in the physiology of all organisms [37][38]. It uses the energy of ATP hydrolysis to transport a wide range of biomolecules across the membrane [38]. The substrate-binding proteins bind the substrate with high a nity and deliver it to the transporter [39]. ABC importers are major determinants of the acquisition of essential nutrients in bacteria [40][41]. The slow growth of D. wulumuqiensis R12 after dps gene knock out can be explained by the down-regulation of ABC substrate-binding transporter. In addition, in the down-regulated GO term, the "growth" in Biological Process GO term was down-regulated, among which "Fe-S cluster assembly protein SufB (sufB)", "Phosphate transport system permease protein PstA (DVJ83_13895)", "Pyridoxal 5'phosphate synthase subunit PdxS (pdxS)", "Phosphate transport system permease protein (pstC)", "Cysteine-tRNA ligase (cysS)", "Glycine dehydrogenase (gcvP)", and "Elongation factor G (fusA)". These proteins were down-regulated that might cause the lower growth of Δdps R12 mutant compared to WT R12 strain.
The Δdps R12 mutant grew slower than the WT R12 strain (Fig. 4A). ABC transporter are the important for the import of nutrients [38]. As shown in Fig. S3A, mineral and organic ion transporters (TbpA); oligosaccharide, polyol, and lipid transporters (ChiE, BmpA and NupB); as well as phosphate and amino acid transporters (PstS, PstA, PstC, PstB, livK and TcyA) were down-regulated. ATP-dependent proteins are valuable for the removal of oxidative damages and dysfunctional proteins [42]. These results indicated that Dps may also be important for the growth of D. wulumuqiensis.
Bacteria cells can communicate using a cell-density dependent regulatory system known as quorum sensing (QS) [43]. QS is associated with a number of cellular processes, such as motility, bio lm formation or antibiotic production [44][45][46]. However, the molecular mechanisms underlying QS circuits in many bacterial species remain unclear, while QS might play important roles in the response to environmental stresses in D. radiodurans [42]. On the basis of proteomic data (Fig. S3B), the QS KEGG pathway (map02024) was signi cantly down-regulated, and 9 proteins were mapped, including branchedchain amino acid ABC transporter substrate-binding protein (DVJ83_00750, DVJ83_02800, DVJ83_06100, and DVJ83_07370), a multifunctional fusion protein (secD), and ABC transporter substrate-binding protein (DVJ83_03180, DVJ83_03190, DVJ83_04825, and DVJ83_13605). The downregulation of these proteins can also help explain the reduced stress resistance of D. wulumuqiensis R12 following the knockout of the dps gene.

Protein-protein interaction network of differentially expressed proteins
Deletion of the dps gene in D. wulumuqiensis R12 induced a series of unfavourable phenotypes resulting from the interaction of multiple proteins. Cell growth and antioxidant defenses were decreased when the dps gene was deleted. A total of 48 up-and 34 down-regulated proteins were predicted to interact with each other. The protein-protein interaction (PPI) network was divided into 8 modules (A-I) as shown in Fig.  9. The proteins in module A are mainly involved in metabolism. Notably, glyceraldehyde-3-phosphate dehydrogenase (A0A345IFK5), which is a key enzyme in the glycolytic pathway [47], was strongly downregulated. Dihydrolipoyl dehydrogenase (A0A345II30) is a component of the pyruvate dehydrogenase complex that plays an important role in the decarboxylation of pyruvate to produce acetyl-CoA [48].
Module D contained proteins that are part of the phosphate transport system, and down-regulated proteins were predicted to interact in pairs. Module E was composed of ABC transporter, which was signi cantly down-regulated. Module I was mainly contained antioxidative enzymes, among which superoxide dismutase (A0A345IKM4) and catalase (A0A345IEC2) were down-regulated in the PPI. These data indicated that the metabolism, transport and oxidation-reduction processes might be affected in D. wulumuqiensis R12 when loss of Dps proteins.

Conclusions
In this study, the sequence, structure, and function of Dps in D. wulumuqiensis R12 were analyzed, the Nterminal of Dps protein might play an important role in DNA binding or protection. The docking data and DNA binding experiments in vitro showed that DNA binding data of R12 Dps is better than that of R1 Dps1. In addition, Δdps R12 mutant was constructed, and comparative proteomics was used to investigate the functions of Dps in D. wulumuqiensis R12. The qRT-PCR and proteomics data suggested that when the dps gene was knocked out in D. wulumuqiensis R12, the catalase gene was downregulated. The proteomics data also suggested that that the metabolism, transport and oxidationreduction processes were down-regulated in D. wulumuqiensis R12 after the dps gene was knocked out.        The circle color indicates the enrichment signi cance P-value, while the circle size is the number of different proteins in the functional class.