The genomic characteristics of ATCC 13939K were analyzed in comparison with the ATCC BAA-816 strain and three additional D. radiodurans R1 genome sequences available in the National Center for Biotechnology Information’s (NCBI) Genome database. These sequences, referred to as R1-2016 (Hua and Hua, 2016), ATCC 13939E (Repar et al., 2021), and ATCC 13939O (Eugénie et al., 2021), were included to provide a comprehensive comparison of genomic attributes across strains (Table 1). Each strain encompasses two circular chromosomes and two plasmids, pMP and pCP. The aggregate genome length of ATCC 13939K stands at 3,285,071 bp, accompanied by a G + C content of 66.65%. Bioinformatic predictions have identified a sum of 3,150 protein-coding sequences (CDSs), in addition to 50 tRNA genes and nine rRNA genes (with three copies each of 5S, 16S, and 23S) within its genome (Table 1). The genome magnitude of ATCC 13939K aligns closely with that of BAA-816 (3,284,156 bp) and is roughly six kbp in excess compared to ATCC 13939E and ATCC 13939O. This disparity is ascribed mainly to variations in the length of chromosome 1 among the strains (Table 1). Prominent insertions discerned in chromosome 2, as well as the two plasmids of R1-2016 (Hua and Hua, 2016), were absent in other R1 strains (Table 1). In pairwise alignments, ATCC 13939K exhibited a 99.98% nucleotide-level identity with BAA-816 and manifested considerable synteny conservation, underscoring its collinearity and the absence of pronounced genomic structural variations.

In the genomic comparison between BAA-816 and ATCC 13939K, a total of 436 DNA sequence differences were identified, including 100 single nucleotide variations (SNVs), 278 insertions with sizes ranging from 1 to 6 base pairs (bp), and 58 short deletions (Table 2 and Supplementary Table S1). Previous studies have shown that R1-2016 and ATCC 13939E exhibit 577 and 559 genetic variations compared to BAA-816 (Hua and Hua, 2016; Repar et al., 2021). The genomic alignment of these differences with the BAA-816 genome revealed that 107 are located in intergenic regions, while the majority, 329 (75.7%), occur within gene regions, including pseudogenes and the 23S rRNA genes (Table 2 and Supplementary Table S1). These variations are observed across 259 distinct genes, with several genes harboring multiple differences. For instance, gene DR_1922, encoding exonuclease SbcC, presented with 4 SNVs and 5 insertional events (Supplementary Table S1). Among the 76 SNVs detected in CDSs, 21 were synonymous substitutions, while 53 led to amino acid (aa) changes. Notably, 2 SNVs lead to the elimination of stop codons in genes DR_1333 and DR_2250 (Supplementary Table S1). Furthermore, 241 insertion or deletion (InDel) events were identified, resulting in frameshifts in 164 CDSs and alterations in the reading frames of 46 pseudogenes (Supplementary Table S1).

Predominantly, InDel events lead to modifications in the length of CDSs or cause the fusion of genes. In the latter case, two or three adjacent genes may merge to form a singular CDS. For a comparative analysis of alterations in the CDSs, we reconfigured the initiation positions of each chromosome and plasmid in ATCC 13939K to coincide with those of BAA-816. Additionally, the designation of the predicted CDSs prefixed with ‘KDR’ was synchronized with the inaugural annotation of the BAA-816 genes. Within the BAA-816 genome, the first gene of chromosome 1, designated DR_0001, is characterized to encode the DNA polymerase III β clamp (DnaN) consisting of 393 aa, as per the initial annotation provided by the authors (GenBank genome AE000513.1). However, this CDS is omitted in the annotation crafted by NCBI (NC_001263.1) (Supplementary Table S2). This absence suggests that the protein homology in particular segments might be insufficient to validate protein-coding designations robustly (Tatusova et al., 2016). In the ATCC 13939K genome, the dnaN gene is newly annotated and assigned the locus tag KDR_0001. This gene spans 1,086 nucleotides and encodes a β-clamp of 361 aa (Supplementary Table S2), a consequence of a 1-bp deletion corresponding to guanine (G) at the 1,037th position in DR_0001 (Supplementary Table S1). This observation is consistent with a previous study (Niiranen et al., 2015).

A comparative genomic analysis among Deinococcus species revealed several annotation inaccuracies in the foundational genomes of BAA-816 (AE000513.1, AE001825.1, AE001826.1, and AE00001827.1). Many of these misannotations were rectified in the reannotated genomes (NC_001263.1, NC_001264.1, NC_000958.1, and NC_000959.1) using the annotation protocols from NCBI. More information on this can be found at https://ncbi.nlm.nih.gov/refseq/about/prokaryotes/reannotation/. An illustrative example is DR_0003. When its orientation is reversed, this gene encodes the DNA damage response protein DdrC (de Groot et al., 2009). Within the reference sequence NC_001263.1, DdrC is appropriately annotated as DR_RS00015 and corresponds to KDR_0003r from ATCC 13939K (Supplementary Table S2). Consequently, the features of the predicted CDSs from ATCC 13939K were also juxtaposed with those in the NCBI-curated annotation (Supplementary Table S2).

Upon examining the CDS across the ATCC 13939 strains, we observed that 2,557 CDSs possess the same length. In addition, 2,629 ATCC 13939K CDSs align with those in ATCC 13939E, while 2,584 match ATCC 13939O (Supplementary Table S3). Within the ATCC 13939K genome, the second gene on chromosome 1, labeled KDR_0002, is annotated as encoding a DnaA protein with a length of 454 aa. In contrast, the corresponding DnaA proteins in ATCC 13939E and ATCC 13939O are annotated as having 466 aa (Supplementary Table S3). This discrepancy arises from different annotation procedures, as the gene exhibits no sequence differences (Supplementary Table S1). Considering the three CDSs, DR_RS00010, E5E91_RS00010, and DRO_RS00010, annotated by NCBI share consistency (Supplementary Tables S2, S3), there is a possibility that the length of KDR_0002 might undergo revision in future NCBI annotations. In addition, among the newly identified CDSs in ATCC 13939K, several with a length of less than 50 aa, such as KDR_0131n, 0333n, and 0347n, are absent in other 13,939 strains (Supplementary Table S3), indicating that NCBI may exclude these short CDSs in their annotation processes. To better understand the genetic variations across the sequenced R1 strains, we juxtaposed the ATCC 13939K genes that exhibit sequence differences (Supplementary Table S1) against their counterparts annotated in R1-2016, ATCC 13939E, and ATCC 13939O (Supplementary Table S4).

Gene gain and loss often happen through the addition and removal of various-sized genome segments, frequently involving mobile genetic elements (MGEs) (Iranzo et al., 2019). Intercellular MGEs like plasmids and phages facilitate DNA transfer between bacterial cells. On the other hand, DNA movement within cells is primarily mediated by specific MGEs such as transposons (Tns) and insertion sequences (ISs). IS elements represent the most basic type of Tn, containing only essential genes for their transposition process (Bennett, 2004). Notably, IS elements are more abundant in D. radiodurans than E. coli and B. subtilis, which serve as model organisms for GN and GP bacteria (Makarova et al., 2001).

In D. radiodurans, the ISDra2 element, originally termed IS8301 and belonging to the IS200/IS605 family, comprises two genes: tnpA and tnpB. These genes code for transposase and RNA-guided DNA nuclease (Islam et al., 2003; Karvelis et al., 2021). ISDra2’s transposition activity significantly increases under γ-radiation or UV radiation exposure (Mennecier et al., 2006). In the ATCC 13939 strain, ISDra2 exists as a single functional copy (DR_1652-DR_1651) and one inactive, degenerate copy (DR_0177-DR_0178), while BAA-816 contains seven complete and one incomplete ISDra2 copies (Pasternak et al., 2010). Comparative genomic analysis revealed the loss of six ISDra2 elements in ATCC 13939K (Figures 2A–F) and other examined 13,939 strains (Supplementary Table S3). This loss led to the reconstitution of genes disrupted by ISDra2 insertion. For instance, KDR_1930m and KDR_2322m, coding for α/β-hydrolase fold enzyme and serine protease were restored (Figures 2D,E). In the ATCC 13939K strain, the ISDra2 segment DR_0979-DR_0978 was replaced by five ORFs (Figure 2F). Among these ORFs, KDR_0977n2 is not defined in other 13939 strains (Supplementary Table S3). The enzyme acetoin dehydrogenase, produced by the genes KDR_0977n3 (acoA) and KDR_0977n4 (acoB), plays a critical role in the breakdown of acetoin (Xiao and Xu, 2007). Additionally, the genes KDR_0980n and KDR_0980 produce glutamate dehydrogenase (Gdh), which facilitates the reversible conversion of glutamate into α-ketoglutarate and ammonia (Miñambres et al., 2000). In a unique case of gene disruption by ISDra2, the segment KDR_1963n1-KDR_1963n2 is integrated into the pilT gene, which codes for the pilus retraction ATPase (Figure 2G). This disruption is exclusive to the ATCC 13939K strain and does not occur in other 13939 strains (Supplementary Table S3). Given that the DNA translocation system in D. radiodurans is associated with type IV pili, mutations in the pilT gene lead to a decrease in the efficiency of natural transformation (Ithurbide et al., 2020). The reduced transformation efficiency was observed in the ATCC 13939K strain (Supplementary Figure S6). This disruption in the 13939K strain is a notable genomic difference that affects the strain’s phenotype.

IS2621, spanning 1,322 bp and enclosed by 19-bp perfect inverted terminal repeats, represents another transpositionally active IS in D. radiodurans (Narumi et al., 1997; Mennecier et al., 2006). The BAA-816 strain originally contained five copies of IS2621, identified as DR_0870, DR_1,334, DR_1618, DR_2222, and DR_B0059 (Long et al., 2015). However, in the ATCC 13939K strain, three of these copies were lost (Figures 2H–J), aligning with findings from previous research (Long et al., 2015). Additionally, ATCC 13939K acquired three new IS2621 copies (Figures 2K–M). Among these, two copies, KDR_0854n and KDR_1841n, were exclusively found in ATCC 13939K, while the third copy, KDR_A0342n, was present in other 13939 strains (Supplementary Table S3). The occurrence of IS elements, specifically in ATCC 13939K, including KDR_0854n, KDR_1841n, and KDR_1963n1-1963n2, was confirmed through PCR assay (Supplementary Figure S7). Cyclic-di-GMP (c-di-GMP) acts as a crucial second messenger in the signal transduction pathways of many bacteria, regulating various cellular processes. The cellular levels of c-di-GMP are controlled by a balance between its production, catalyzed by diguanylate cyclases that contain GGDEF domains, and its breakdown, mediated by phosphodiesterases that possess either EAL or HD-GYP domains (Petchiappan et al., 2020). DR_A0342 is notable for harboring both HD-GYP and GGDEF domains within its C-terminal region, illustrating its role in c-di-GMP homeostasis (Galperin et al., 2001; Pei and Grishin, 2001). The insertion of IS2621 into DR_A0342 splits it into two parts, KDR_A0342a and KDR_A0342b (Figure 2M). Importantly, KDR_A0342b retains the intact domains (data not shown), suggesting its potential functionality in regulating c-di-GMP levels.

In the genome NC_001263.1, DR_RS16465 is identified as IS1, equivalent to KDR_0835r in the ATCC 13939K strain (Supplementary Table S2). Additionally, an extra copy of IS1 has been inserted in both ATCC 13939K (Figure 2N) and other 13939 strains (Supplementary Table S3). ISDra3, ISDra4, and ISDra6 have been classified as members of the IS630 family in the ISFinder database.¹ These elements are characterized by sequences containing stretches of A residues. For instance, ISDra3, which includes DR_0255, DR_B0139, and DR_C0004, features a run of nine As, and ISDra6, incorporating DR_1523, DR_B0056, and DR_B0113, is marked by a sequence of eight As (Baranov et al., 2005). In the BAA-816 strain, the 23S rRNA genes were not fully characterized: rrnaA23S (DR_r05) and rrnaB23S (DR_r09) had a length of only 876 nucleotides each, while rrnaC23S (DR_r02) extended to 1,943 nucleotides, a result of the insertion by ISDra3 element DR_0255 (Pei et al., 2009; Supplementary Table S2). In ATCC 13939K, re-annotation and the absence of DR_0255 allowed for the identification of KDR_r02, KDR_r05, and KDR_r09, each at 2,876 nucleotides, harmonizing the lengths of all three 23S rRNA genes (Figure 2O and Supplementary Table S2). Furthermore, KDR_1721n and KDR_0687n were found as additional insertions of ISDra3 and ISDra6, respectively, in the ATCC 13939K strain (Figures 2P,Q), consistent with the previous report (Long et al., 2015).

Gene gain or loss in bacteria can occur without MGEs due to improper recombination processes, such as homologous recombination or inaccurate non-homologous repair activities. Errors in DNA replication or the repair process, mainly when fixing aberrant replication forks, can result in the deletion or duplication of genes. Consequently, this leads to either the loss or gain of genetic material (Periwal and Scaria, 2015). The genomic comparison between 13939K and BAA-816 reveals that the genes DR_A0268 and DR_A0269, present in BAA-816, are absent in the 13939K strain (Figure 3A). Examination of a 1,697 nucleotide sequence that includes DR_A0268 and DR_A0269 shows that a portion of DR_A0268, between coordinates 287,484 and 288,083, is nearly identical to the C-terminal sequence of DR_A0270, found between coordinates 289,001 and 289,600, with only three bases differing (Figure 3B). Additionally, the sequence spanning 288,083–288,973 on chromosome 2 aligns perfectly with the segment of chromosome 1 between coordinates 2,463,286–2,464,176 (Figure 3B). This alignment suggests that recombination between homologous sequences of DR_A0270 and DR_A0268 could have resulted in the loss of the DR_A0269 region in strain ATCC 13939K. Furthermore, DR_A0268 has been updated to DR_RS17005 in annotations (Supplementary Table S2), and the C-terminal aa sequences of KDR_A0270 match those of DR_RS17005 (Supplementary Figure S8). In the ATCC 13939K strain, there was a notable presence of two newly inserted genes, KDR_1221n1 and KDR_1221n2 (Figure 3C). Proteins in the drug/metabolite transporter (DMT) superfamily are characterized by a unique structure that includes varying numbers of TM α-helices, commonly at counts of 4, 5, 9, or 10 per protein. The DMT proteins with 10 TM segments are formed by duplicating a fundamental five-segment precursor within the gene (Jack et al., 2001). Specifically, the DMT protein encoded by gene KDR_1221n2 possesses 10 helical segments (data not shown), implying that it may have evolved through a gene duplication event.

The D. radiodurans strains, ATCC BAA-816 and ATCC 13939, originated from the ATCC repository. Cultures were grown at 30°C in TGY broth (0.5% tryptone, 0.3% yeast extract, and 0.1% glucose) or on TGY plates supplemented with 1.5% Bacto-agar. The chemicals used throughout the study were purchased from Sigma-Aldrich (St. Louis, MO, USA).

From single colonies on TGY agar, the D. radiodurans ATCC 13939K strain was cultured in TGY broth, shaking at 200 rpm at 30°C for 18 h. Post-cultivation, the cells were centrifuged, and genomic DNA was extracted using the AccuPrep® Genomic DNA Extraction Kit (Bioneer, Korea) according to the manufacturer’s instructions. Specifically, 1 × 10⁹ cells were lysed using 180 μL of lysis buffer (20 mM Tris–HCl, 2 mM sodium EDTA, 1.2% Triton® X-100, pH 8.0), supplemented with 20 μL of lysozyme (100 mg/mL) and 10 μL of RNase A. The lysate was incubated at 37°C, followed by the addition of 20 μL of Proteinase K and 200 μL of GB Buffer, with subsequent incubation at 60°C. DNA purification was then performed using spin columns. For genome sequencing, a combination of Illumina NovaSeq 6000 and PacBio Sequel II platforms was used. DNA libraries were prepared using the TruSeq DNA Nano kit (Illumina, San Diego, CA, USA) and the SMRTbell Prep Kit 3.0 (PacBio, Menlo Park, CA, USA). The initial assembly was achieved with PacBio reads achieving 298.4× coverage, using the CANU V1.7 software (Koren et al., 2017), and was further refined using Illumina reads at 480.8× coverage through Pilon V1.21 software (Walker et al., 2014).

The genome sequence comprises four circular DNA structures totaling 3,285,071 bp. Analysis of GC content was conducted using the ANI Calculator (Richter et al., 2016). Prokka (v1.13) was employed to define the genomic profile, including CDS, tRNA, and rRNA genes (Seemann, 2014). Functional annotation of identified proteins was aligned with COG and KEGG databases, and sequence verification was performed against NCBI databases using BLASTP and BLASTX. The nucleotide sequence of ATCC 13939K was deposited in the NCBI Nucleotide database (GenBank number, CP150840 ~ CP150843).

Raw read data from Illumina sequencing was used for SNP detection. Following filtration, the clean read sequences were aligned to the reference sequence using the Burrows-Wheeler Aligner (BWA, version 0.7.17) (Li and Durbin, 2010). After mapping, duplicate reads were removed using Sambamba (version 0.6.8) (Tarasov et al., 2015), and SAMTools were used to identify variants (Danecek et al., 2021). Amino acid changes from these variants were obtained using SnpEff (version 4.3 t) (Cingolani, 2022).

For the preparation of competent cells, exponentially growing cells were collected via centrifugation and then resuspended in TGY medium at a concentration of 5 × 10⁸ cells/mL. The medium was supplemented with 30 mM CaCl₂ and 10% (v/v) glycerol, after which it was quickly frozen and stored at −80°C for preservation. Two plasmids were used for the transformation assays: pRadgro with a chloramphenicol resistance marker, which replicates in D. radiodurans, and pKatAPH3, carrying a kanamycin resistance gene, used for targeted gene disruption (Ohba et al., 2005; Misra et al., 2006). For the pKatAPH3-based assay, the upstream and downstream regions (approximately 1 kb) of bphP (DR_A0050) were PCR amplified using the primer pairs, Up-F (5′-tatctcgagtcgcgcggcctgtt-3′)/Up-R (5′-tatgatatcgatgctgggcctcttccggga-3′) and Down-F (5′-tatggatccggcaacgggtcccggctcat-3′)/Down-R (5′-tatctcgagcgacgttgcggttgtcttcgcc-3′). The PCR products were digested with XhoI/EcoRV and BamHI/PstI to produce the upstream and downstream fragments and ligated into the pKatAPH3 vector. The recombinant plasmid was named pKatAPH3-BphP. For transformation, aliquots of competent cells (100 μL each) were thawed on ice and mixed with an equal TGY medium containing 30 mM CaCl₂. Subsequently, 250 ng of either plasmid DNA was added. Following a 30 min incubation at 0°C, the mixture underwent heat treatment at 32°C for 90 min. After the heat treatment, 800 μL of TGY media were added to the mixture to aid in the recovery and expression of antibiotic resistance, followed by further incubation for 5 h. Diluted samples were then plated onto TGY plates containing the appropriate antibiotics for selection.

PCR assays were utilized to validate the integration of specific IS elements within the genome of the ATCC 13939K strain. The dedicated primer pairs were used for the detection of each IS element: 0854F (5′-gtgggctcgcctagcatgtt-3′)/0855R (5′-gaacttctacccccaggcgg-3′) for KDR_0854n, 1841F (5′-atgttcggcgccatctggta-3′)/1842R (5′-cctcaacctctacgccgagg-3′) for KDR_1841n, and 1963F (5′-taggcgtccatcgtgaccatgc-3′)/1963R (5′-gatgtactcgatgatgaacgagcc-3′) for KDR_1963n1 and KDR_1963n2. Amplification was carried out using PrimeSTAR DNA polymerase (TaKaRa Bio Inc., Japan) under the following conditions: an initial denaturation step at 94°C for 40 s, followed by 20 cycles of amplification consisting of 40 s at 94°C, 40 s at 57°C, and 3 min at 72°C. A final elongation step at 72°C for 5 min was performed. The PCR products were then subjected to agarose gel electrophoresis.

Overnight cultures in the stationary phase were inoculated into fresh TGY broth at a 1:100 dilution and grown to log phase (OD₆₀₀ ≈ 1.0). These cultures were then diluted to an OD₆₀₀ of approximately 0.1 in TGY broth and treated with varying concentrations of mitomycin C (MMC) and hydrogen peroxide (H₂O₂) for 1 h at 30°C. Post H₂O₂ treatment, cells were treated with catalase (Sigma-Aldrich, Saint Louis, CA, USA) to degrade any residual H₂O₂, followed by 10-fold serial dilutions in distilled water, and subsequent spotting onto TGY agar plates. At room temperature, γ-radiation exposure was carried out using a ⁶⁰Co gamma irradiator (AECL, IR-79; MDS Nordion International Co. Ltd., Ottawa, Canada) at the Advanced Radiation Technology Institute, Republic of Korea. Ultraviolet C (UV-C) exposure was performed using a UV crosslinker (CX-2000, UVP LLC, Upland, CA, USA), where plates with spotted cells were subjected to UV-C radiation. Post-exposure, the TGY plates were incubated at 30°C for 2–3 days, allowing colony-forming unit (CFU) enumeration.