Arsenotrophic Achromobacter aegrifaciens strains isolated from arsenic contaminated tubewell water and soil sources shared similar genomic potentials Authors: M. Nazmul Hoque, Ayman Bin Abdul Mannan, Anamica Hossian, Golam Mahbub Faisal, M. Anwar Hossain, Munawar Sultana Published: 2024 ABSTRACT BackgroundArsenic (As), found in diverse ecosystems, poses major public health risks in various parts of the world. Arsenotrophic bacteria in contaminated environments help reduce toxicity by converting arsenite (AsIII) to less harmful arsenate (AsV). We assumed that Achromobacter aegrifaciens strains from As-contaminated tubewell water and soil would share similar genomic characteristics associated with arsenic detoxification and bioremediation. To investigate this, we employed both culture-dependent and culture-independent viz. whole genome sequencing (WGS) methods to thoroughly elucidate the phenotypic and genotypic features of two A. aegrifaciens strains isolated from As-contaminated tubewell water (BAW48) and soil (BAS32) samples collected in the Bogura district of Bangladesh.ResultsBoth BAW48 and BAS32 isolates demonstrated As(III) oxidation in the KMNO4 test, which was corroborated by molecular analysis confirming the presence of aioA and arsB genes in both strains. These strains were found to be phylogenetically related to many strains of Achromobacter spp., isolated from biological inorganic reactors, environmental soils, sediments and human clinical samples across diverse geographical regions. Moreover, both strains possessed distinct heavy metal resistance genes conferring resistance to Co, Zn, Cu, Cd, Hg, As, and Cr. Three As gene clusters such as As(III) oxidizing aioBA, As(III) reducing arsRCDAB and the MMA(III) oxidizing ars resistance gene (arsHCsO) cluster were predicted in both genomes of A. aegrifaciens. Further genomic analyses revealed similar profiles in both strains, with mobile genetic elements, antimicrobials and heavy metal resistance genes, virulence genes, and metabolic features. Pangenome and synteny analysis showed that the two genomes are evolutionary distinct from other strains, but closely related to one another.ConclusionThe genomic data confirmed that A. aegrifaciens strains can oxidize As(III) and detoxify heavy metals like As, suggesting their potential for As detoxification and bioremediation. These findings align with our assumption and provide a basis for developing sustainable solutions for bioremediation efforts in As-contaminated environments.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12866-024-03676-9. Background Arsenic (As) is a hazardous pollutant identified in aquifers and soil, contributing to significant global health concerns [1]. In Bangladesh, approximately 75 million people, out of a total population of around 170 million, are consistently exposed to high levels of As (> 50 µg/L) in their drinking water [2], which is much higher than the World Health Organization (WHO) recommended limit of 10 µg/L [1]. Amid the numerous environmental challenges and often neglected issues in today’s world, substantial research is focused on understanding the toxicities of heavy metals like arsenic (As), lead (Pb), chromium (Cr), cadmium (Cd), and others [3]. According to the World Health Organization (WHO), Bangladesh is undeniably among the countries facing critical levels of heavy metal pollution in its ecosystems [4]. Unfortunately, underscoring the severity of the issue, groundwater in 62 out of 64 districts in Bangladesh is contaminated with As [5]. The problem is primarily aggravated by the widespread dependence on As-contaminated naturally occurring groundwater [6]. Arsenic-contaminated groundwater causes cancers, cardiovascular diseases, neurological impairments, reproductive issues, diabetes, weakened immunity, and widespread public health risks [7]. Arsenic exists in the environment in various forms, including inorganic arsenite (AsIII) and arsenate (AsV), organic compounds, and arsenic-rich minerals, contaminating soil, water, and air [6, 8]. Among these, As(III) is approximately 100 times more toxic than As(V) and poses greater challenges for mitigation [2, 8]. Due to the widespread presence of heavy metals in the environment, microorganisms have evolved mechanisms to withstand the harmful effects of these elements. A diverse groups of bacteria, including those from the genera Acidithiobacillus, Achromobacter, Bacillus, Deinococcus, Desulfitobacterium, Stenotrophomonas and Pseudomonas, have been documented for their resistance to As [9, 10]. Previously, the potential of various Achromobacter species for As remediation has been extensively studied [11, 12]. For example, Achromobacter sp. strain N2, isolated from pyrite- and cinder-contaminated soil, has the ability to transform As. It exhibits resistance to both As(III) and As(V), and due to the presence of the As(III) oxidase aioA gene, it can oxidize As(III) to As(V) [11]. Similarly, the detoxifying ability of A. arsenitoxydans SY8, isolated from As-contaminated soil, is enhanced by the presence of an As(III) oxidation operon and As resistance operons [12]. Additionally, A. xylosoxidans BHW-15, an As(III)-oxidizing bacterium isolated from contaminated tubewell water, plays a significant role in As detoxification, accumulation, and immobilization [13]. These studies significantly contributed to the understanding of how bacteria like Achromobacter can detoxify As and other harmful compounds while potentially playing a role in bioremediation and environmental cleanup efforts. The presence of these bacteria highlights their potential role in As-contaminated groundwater and soils due to their remarkable adaptability to extreme environments [2, 14, 15]. They can stabilize water and soil functions in affected areas. Microbial transformations of As influence its speciation, methylation, ecotoxicity, and environmental fate. Changes in the oxidation state of As affect its biogeochemical cycling [14]. Therefore, microorganisms that can tolerate and convert As(III) to As(V) offer promising bioremediation strategies, often acquiring traits through mutations, genetic material acquisition, and gene expression changes. Arsenic-related genes are commonly categorized into two groups: resistance and metabolism [16]. The ars operon (e.g., arsB, acr3) encodes As resistance or detoxification, safeguarding the cell against As without directly detoxifying the environmental As itself [14]. Arsenic metabolism involves several genes, including those responsible for methylation (arsM), oxidation (aioAB, arxAB), and dissimilatory reduction (arrAB) [16]. While the genetic elements responsible for As detoxification and metabolism are well-understood, the complete extent of gene distribution, diversity, and interspecies transfer in As detoxification and metabolism remains unknown. As(III)-oxidizing bacteria (AOB) harboring As(III) oxidase gene (aioA) has been reported to detoxify As(III) to As(V) [17]. While culture-based methods have long been considered the gold standard for characterizing microbiota, the past decade has seen a significant rise in the use of high-throughput next-generation sequencing and bioinformatics pipelines [18–20]. Previous genomic and metagenomic studies reported As-oxidizing bacterial genera, such as Pseudomonas, Alcaligenes, Thermus, Agrobacterium, Herminiimon, Thiomonas, and Achromobacter in the groundwater of Bangladesh [21, 22]. Achromobacter is a common and dominant inhabitant in the groundwater worldwide [23, 24]. Different strains of Achromobacter are known to be prevalent and abundant in As-contaminated waters and soils in the studied regions [13, 21]. While these organisms may not be the most extensively investigated, preliminary surveys and existing literature suggest they play a critical role in As bioremediation [13, 25]. Their genomic capabilities for arsenic detoxification and bioremediation make them particularly relevant for understanding microbial interactions with arsenic in contaminated ecosystems. Although various strains of A. aegrifaciens [26, 27] and A. xylosoxidans [13, 21] have been identified in the soil and groundwater of Bangladesh, their genomic potential as arsenite [As(III)] oxidizers remains poorly understood. Given the limited research on the As bioremediation potential of A. aegrifaciens, particularly in strains isolated from As-contaminated environments, it is crucial to investigate their complete genomes for a deeper understanding of their As detoxification and bioremediation capabilities. This gap raises the question of whether strains isolated from As-contaminated tubewell water and soil share common genomic features that facilitate As detoxification. To address this gap, this study hypothesizes that A. aegrifaciens strains from As-contaminated tubewell water and soil share genomic traits related to As detoxification. We conducted a comprehensive phenotypic and genotypic analysis of two A. aegrifaciens strains, BAW48 (isolated from As-contaminated tubewell water) and BAS32 (isolated from As-contaminated soil), using both culture-dependent and high-throughput genome sequencing. The results will enhance our understanding of As bioremediation mechanisms and the potential of these strains for environmental detoxification. Methods Sample collection and bacterial isolation We collected two samples: one from As-polluted tubewell water and the other from As-contaminated soil, sourced from Bogura district (24.85° N, 89.37° E), one of the most As-prone areas of Bangladesh. The collection of sample was conducted using sterile techniques, as previously described [28, 29]. According to the Department of Public Health Engineering, Government of Bangladesh report, tubewells in Bogura contained As levels ranging from 100 to 500 ppb in drinking water [2]. Tubewell water As concentrations were measured on-site using a low-range As field kit (Hach, USA) capable of detecting 0 to 500 ppb As. Collected water sample (250 ml) underwent filtration using a sterile 0.22 μm cellulose-nitrate filter (Osmonics, USA). Bacteria retained on the surface of the filter was resuspended in phosphate-buffered saline, and then enriched in 60 ml of a modified minimal salt medium supplemented with 2 mM NaAsO2 [2% (w/v)] (Merck, Germany), specifically designed for enrichment of autotrophic and heterotrophic As(III) oxidizing bacteria at 30 °C for 7 days [29]. Simultaneously, 2 gm of collected soil sample (approx. 10 gm) was dissolved in 20 ml 0.9% NaCl and shaken for 3 min. Subsequently, 2 ml of resulting suspension was spread onto heterotrophic agar medium for incubation at 30 °C on rotary shaker at 120 rpm for 7 days [28, 30]. Different single colonies were selected, picked up and inoculated in modified minimal salt both supplemented with 2 mM NaAsO2 for incubation at 30 °C with a rotary shaker set at 120 rpm for two weeks [2]. The enrichment broth from tubewell water and soil isolates were subjected to serial three-fold dilution, and the diluted samples were then plated onto agar media supplemented with As(III). In brief, evenly spread plates were sealed with parafilm, incubated at 30 °C in the shaking incubator set at 120 rpm, and observed growth after two weeks incubation period [2]. Finaly, two to five single colonies were isolated from each plate, subsequently purified, selected, and stored at -80 °C for further investigation. To assess the minimum inhibitory concentration (MIC) of As(III) for the BAS32 and BAW48 isolates, a 96-well microtiter plate was used. Each well was filled with 130 µL of sterile heterotrophic broth medium, supplemented with varying concentrations of As(III) as NaAsO2 (0, 2, 4, 6, 8, 10, 12, 14, 15, 16, 18, 20, and 22 mM). The BAS32 and BAW48 isolates were cultured in 5 mL of heterotrophic broth [29] without As until the optical density (OD) at 600 nm reached 0.1 on a rotary shaker at 120 rpm. Then, 5 µL of bacterial inoculum (OD590 = 0.1) was added to each corresponding well. A negative control was included in each row. Initial cell density and bacterial growth after 7 days were measured using a spectrophotometer at 600 nm, and the microtiter plate was incubated at 25˚C [2]. Screening for arsenite (AsIII) transformation Two isolates of A. aegrifaciens including one from tubewell water (A. aegrifaciens BAW48) and another from soil sample (A. aegrifaciens BAS32) were revived in heterotrophic As-enriched selective medium at 25° C for seven days [2]. The As(III) transformation test was conducted on pure cultures isolated from single colonies, utilizing colorimetric methods (KMnO4 and AgNO3 assays) as previously developed [13, 29]. Genomic DNA from arsenotrophic isolates (BAW48 and BAS32) was extracted using the Qiagen DNA Mini-Prep kit (Qiagen, USA) according to the manufacturer’s instructions. The isolates were confirmed as A. aegrifaciens through ribosomal gene (16 S rRNA) sequencing using 27 F (5′- AGAGTTTGATCCTGGCTCAG-3′) and 1492R (5′- GGTTACCTTGTTACGACTT-3′) primers [31]. The 16 S rRNA sequences of A. aegrifaciens BAW48 and BAS32 are in available in the GenBank under accession numbers PQ482439 and PQ488467, respectively. We used specific primers to perform PCR targeting arsenotrophic functional genes (Table 1). The presence of an As(III) oxidizing gene (aioA) and As(III) resistance gene (arsB) was confirmed by PCR [2, 13]. Table 1Primer sequences used for detection of bacterial 16 S rRNA gene, as(III) resistance gene, oxidizing gene, and corresponding annealing temperature used for PCRTargetgenePrimers Sequences (5ʹ→ 3ʹ)Ampliconsize (bp)Annealing T (0C)Reference16 S rRNA27 F5′- AGAGTTTGATCCTGGCTCAG-3′~ 1400-145055[31]1492R5′- GGTTACCTTGTTACGACTT-3′ aioA aoxBM1-2 F5’CACTTCTGCATCGTGGGNTGYGGNTA-3′110052[69]aoxBM3-2R5′- TGTCGTTGCCCCAGATGADNCCYTTYTC-3′ arsB darsB1F5′- GGTGTGGAACATCGTCTGGAAYGCNAC-3′75055[70] Table 1Primer sequences used for detection of bacterial 16 S rRNA gene, as(III) resistance gene, oxidizing gene, and corresponding annealing temperature used for PCRTargetgenePrimers Sequences (5ʹ→ 3ʹ)Ampliconsize (bp)Annealing T (0C)Reference16 S rRNA27 F5′- AGAGTTTGATCCTGGCTCAG-3′~ 1400-145055[31]1492R5′- GGTTACCTTGTTACGACTT-3′ aioA aoxBM1-2 F5’CACTTCTGCATCGTGGGNTGYGGNTA-3′110052[69]aoxBM3-2R5′- TGTCGTTGCCCCAGATGADNCCYTTYTC-3′ arsB darsB1F5′- GGTGTGGAACATCGTCTGGAAYGCNAC-3′75055[70] Primer sequences used for detection of bacterial 16 S rRNA gene, as(III) resistance gene, oxidizing gene, and corresponding annealing temperature used for PCR ~ 1400- 1450 Genome sequencing, assembly and annotation The DNA was extracted from the pure culture of both arsenotrophic isolates (BAS32 and BAW48) using FavorPrepTM Tissue Genomic DNA Extraction Mini Kit (Biosettia Inc., San Diego California USA) according to the manufacturer’s instructions. The purity and concentration of the extracted DNA were assessed using the NanoDrop 2000 UV-Vis Spectrophotometer (Thermo Fisher, Waltham, MA, USA). Libraries were generated from 1 ng DNA using the Nextera™ DNA Flex Library Prep Kit (Illumina, San Diego, USA), and WGS of the prepared libraries was carried out on the Illumina NovaSeq PE150 sequencer (Illumina, San Diego, CA, USA) with a 2 × 250-bp protocol. Generated raw reads (BAS32 = 328,86,904 bp and BAW48 = 204,90,222 bp) were trimmed using Trimmomatic v0.39 (with parameters leading:20, slidingwindow:4:20:20, trailing:20, and minlen = 36) [32] to remove Illumina adapters, known Illumina artifacts and phiX, and quality checked using FastQC v0.11.7 [33]. SPAdes v2.3 [34] was employed to assemble high-quality reads (BAS32 = 327,72,308 bp and BAW = 204,35,320 bp) with phred scores > 20, and quality of the assembled genome was checked using QUAST v.5.0.2 [35]. The assembled genomes of A. aegrifaciens BAW48 and BAS32 are available at GenBank under accession numbers JAVCPK000000000, and JAVCWD000000000, respectively. Genome annotation was performed using the NCBI Prokaryotic Genome Annotation Pipeline v6.5 [36]. The completeness of the genome was checked through CheckM v1.2.2 using Achromobacter CheckM marker set [37]. To identify the bacteria at the species level, assembled contigs were evaluated using PUBMLST [38], BLASTn and the k-mer method in the KmerFinder 3.2 [39]. Based on the PUBMLST, BLASTn and KmerFinder results, the study genomes (BAS32 and BAW48) and 24 reference genomes of Achromobacter spp. from different sources (Table S1) were used in phylogenetic analysis. Genomic alignment was conducted using MUSCLE v5.0 [40], and a maximum-likelihood phylogenetic tree was constructed with PhyML v3.0 [41], and resulting tree was visualized using iTOL (interactive tree of life) v3.5.4 [42]. Pairwise average nucleotide identity (ANI) values (%) were calculated for the study genomes (BAS32 and BAW48) and 24 closely related genomes of Achromobacter spp. (used in phylogenetic analysis), utilizing FastANI through the GTDB-tk v2 database [43]. Furthermore, BLAST Ring Image Generator (BRIG) v0.95 was employed for aligning and visualizing multiple genomes (n = 6), with A. xylosoxidans (GenBank Accession No.: AP028040) as the reference. The genome comparisons were executed using BRIG v0.95 with an e-value cutoff of 1e− 5 [44]. For pangenome analysis, Roary v3.11.2 [45], a rapid and high-speed stand-alone pipeline, was utilized to identify core and accessory genes. This analysis predicted four gene classes: ‘core’ (99% ≤ strains ≤ 100%), ‘soft core’ (95% ≤ strains < 99%), ‘shell’ (15% ≤ strains < 95%), and ‘cloud’ (0% ≤ strains < 15%). For the prediction of antibiotic resistance genes (ARGs), virulence factor genes (VFGs), mobile genetic elements (MGEs) and metabolic functions in the draft genomes, ResFinder 4.0 [46], VFDB [47], Proksee [48] and RAST server [49] were used, respectively. The predicted genes were also annotated using RAST server. SimpleSynteny was used to analyze the synteny of the As(III) oxidizing aioBA, As(III) reducing arsRCDAB and the MMA(III) oxidizing ars resistance (arsHCsO) gene clusters in BAW48 and BAS32 genomes with the closest strain [50]. PlasmidFinder v2.1.1 [51] and PHASTER [52] databases were used for prediction of plasmid and phage regions, respectively. Default parameters were used for all software unless otherwise specified. Results Phenotypic and molecular characteristics of A. aegrifaciens strains The bacterial strains BAW48 and BAS32 displayed features commonly associated with Gram-negative bacilli. These strains were aerobic, motile, and, when cultured on As-supplemented heterotrophic media, formed small, whitish, circular opaque colonies, demonstrating their resilience to As. Notably, both isolates displayed a pink color in the KMNO4 test, indicating that they had oxidized As(III) present in the solution, rendering the chemical ineffective as an oxidizing agent. This contrasted with the control, where the absence of bacterial transformation resulted in the chemical retaining its color, signifying its role as an oxidizing agent (Fig. S1a – b). Molecular analysis further supported these findings, revealing the presence of the aioA gene (1100 bp amplicon) (Fig. S1c) and the arsB gene (750 bp amplicon) (Fig. S1d) in both isolates. The MIC levels of As(III) for both BAW48 and BAS32 were determined by monitoring bacterial growth through OD measurements at 600 nm at the start of the experiment and after 7 days of incubation at 25 °C. The microtiter plate assay revealed that the MIC of As(III) was 20 mM for BAW48 and 30 mM for BAS32. The absence of growth in the negative control wells validated the reliability of the assay. These results indicate varying levels of As resistance between the two isolates, with BAS32 exhibiting significantly greater tolerance to As(III) compared to BAW48. Collectively, these phenotypic and molecular data confirmed the isolates as arsenotrophic candidates. Genomic features of A. aegrifaciens strains BAW48 and BAS32 To our understanding, this is the first-ever comparative genomic study focusing on arsenotrophic A. aegrifaciens strains isolated from As-contaminated samples from two different ecological origin. We sequenced the genome of the two arsenotropic A. aegrifaciens isolates (BAW48 and BAS32) having different As-tolerance properties. The genomes BAW48 and BAS32 were sequenced through WGS. The comparative genomic characteristics of these two strains are outlined in Table 2. CheckM assessment reveals 98.11% completeness of the both genomes. The size of the BAW48 and BAS32 draft genomes are about 6.9 Mbp, with GC content of 65.9%, and genome coverage of 45x and 43x, respectively. The final assembly of BAW48 comprised 52 contigs, with the largest contig of 1,001,563 bp. The N50 value for contigs larger than 1000 bp was 767,704 bp. Similarly, BAS32 comprised of 33 contigs, where the largest contig was 1,042,123 bp, and the N50 was 374,113 bp. The draft genome BAW48 harbored 6,350 coding sequences (CDS), 57 tRNAs, and one rRNA whereas BAS32 genome contained 6,331 CDS, 55 tRNAs, and four rRNAs (Table 2). According to PHASTER analysis, both genomes had two phage regions, each encoding five genes (Fig. S2). These genes elements included one responsible for phage packaging machinery (TrmL), three associated with the phage lysis module (genLysins, Holin, Rz/Rz1), and one dedicated to phage DNA synthesis (GtR). However, neither of the genomes contains any plasmid sequences. Table 2General genomic features of the A. aegrifaciens strains isolated from As-contaminated groundwater and soilFeatures (s)A. aegrifaciens strainsBAW48BAS32Genome size (bp)6,877,6536,857,615Genome coverage (x)4543GC content (%)65.965.9Total contigs5233Longest contig (bp)1,001,5631,042,123Shortest contig (bp)2232464Contig N50 (bp)767,704374,113L5045Total genes6,4236,394Coding sequences (CDSs)6,3506,331Protein coding genes6,3076,287RNA genes7363tRNA genes5755rRNAs14ncRNAs44Pseudo genes4344Genes with function prediction6,4976,467Features in subsystems362361Prophages22Antimicrobial resistance genes (ARGs)66Virulence factor gene (VFG)11 Table 2General genomic features of the A. aegrifaciens strains isolated from As-contaminated groundwater and soilFeatures (s)A. aegrifaciens strainsBAW48BAS32Genome size (bp)6,877,6536,857,615Genome coverage (x)4543GC content (%)65.965.9Total contigs5233Longest contig (bp)1,001,5631,042,123Shortest contig (bp)2232464Contig N50 (bp)767,704374,113L5045Total genes6,4236,394Coding sequences (CDSs)6,3506,331Protein coding genes6,3076,287RNA genes7363tRNA genes5755rRNAs14ncRNAs44Pseudo genes4344Genes with function prediction6,4976,467Features in subsystems362361Prophages22Antimicrobial resistance genes (ARGs)66Virulence factor gene (VFG)11 General genomic features of the A. aegrifaciens strains isolated from As-contaminated groundwater and soil Phylogenetic relatedness between BAW48 and BAS32 strains To elucidate the evolutionary relationship between the studied genomes (A. aegrifaciens strains BAW48 and BAS32) and 24 reference genomes of different strains of Achromobacter spp., a maximum-likelihood phylogenetic tree was constructed (Fig. 1, Table S1). The phylogenetic tree showed six major clades of 26 isolates from six continents where BAW48 and BAS32 clustered separately. However, A. aegrifaciens BAS32 strain is found to be a sub-lineage of the A. aegrifaciens BAW48. Furthermore, the phylogenetic analysis revealed a closer alignment of A. aegrifaciens BAW48 (JAVCPK000000000_Water_BD) with several strains of A. xylosoxidans previously isolated from biological inorganic reactor in Japan (AP028040_Reactor_JPN), human clinical samples from USA (CP053617_Human_USA, CP026973_Human_USA), environmental soil sample from China (CP045222_Soil_CHN), and one A. mucicolens strain isolated from environmental sediment of China (CP116006_Sediment_CHN) (Fig. 1). The phylogenetic results were further supported by the findings from the ANI analysis (Fig. 2). Phylogeny based on ANI revealed that both BAW48 and BAS32 genomes clustered closely with almost 100.0% identical nucleotides in their genomes. Though all strains shared > 95% ANI, six distinct clades could be observed supporting the results of maximum-likelihood phylogenetic tree (Fig. 2). The genomic map generated through BRIG comparison revealed minimal large-scale variation across the selected strains (Fig. 3). However, few non-homologous regions were identified around the reference genome, displaying over 95% identity. The BAW48 and BAS32 strains were found to share common regions of genetic variation with the reference strain (A. xylosoxidans H1_3_1 DNA; NCBI accession: AP028040) and several environmental strains from sediment (A. mucicolens DMF-4; NCBI accession: CP116006), soil (A. xylosoxidans DN002; NCBI accession: CP045222) and one human clinical isolate (A. xylosoxidans GN050; NCBI accession: CP053617) (Fig. 3). Most of these non-homologous regions might be linked to transposable elements. We further performed pangenome analysis to better elucidate the diversity and differences in the study genomes and phylogenetically close eight reference strains. The pangenome matrix based on the presence and absence of genes in ten valid reported strain of the Achromobacter species (Fig. 4a) showed that A. aegrifaciens strain LMG 26,852 is the closest strain of BAW48 and BAS32. The tree structure revealed that BAS32 and BAW48 branch off together from a common node, indicating that they shared a more recent common ancestor with each other than with the other strains. The pangenome curve showed a continuous increase in the number of total genes as more strains are added and conserved gene were almost in a conserved line (Fig. 4b). This suggested an open pangenome, indicating high genetic similarity within the species. In this analysis, a total of 10,509 genes were predicted in the pangenomes, including 6,284 genes in BAW48 and 6,264 genes in BAS32 genome. Among these, 4391 core genes, 2465 shell genes, and 3653 cloud genes were predicted, which were present in > 99% sequences of the valid strains (Fig. 4c). However, no soft-core gene was found in the pangenome analysis. Core genes were essential and shared by all strains, shell genes were present in some but not all strains and contributed to adaptability, while cloud genes were unique to a few strains, often providing specific advantages. Fig. 1The evolutionary relationships between A. aegrifaciens BAW48 and BAS32 and other Achromobacter spp. genomes sequenced from six continents worldwide. Whole genome sequences of twenty-four strains of human, plant and environmental origin retrieved from NCBI were used for phylogenetic analysis. The mid-point rooted tree was constructed using the NCBI Tree Viewer (https://www.ncbi.nlm.nih.gov/tools/treeviewer/), and visualized with iTOL (interactive tree of life). The evolutionary relationship was inferred using the maximum-likelihood method. Different colors (e.g., pinkish red for Asia, greenish cyan for Europe, yellowish green for North America, bluish cyan for South America, red for Australia and orange for Antarctica) are assigned according to the close evolutionary relatedness (clade) of the genomes. The scale bar is in the unit of the number of substitutions per site. The values on the branches are bootstrap support values based on 1000 replications. All the sequences were indicated by their accession numbers followed by the source and country code. The country codes according to the standard abbreviation are: United States of America (USA), Japan (JPN), China (CHN), Antarctica (ATA), Brazil (BRA), Germany (GER), Thailand (THA), Russia (RUS), India (IND), Czech Republic (CZE), Chile (CHL), Australia (AUS), Portugal (PRT) and Bangladesh (BD). The genomes of the A. aegrifaciens BAW48 (JAVCPK000000000_Water_BD) and A. aegrifaciens BAS32 (JAVCWD000000000_Soil_BD) are highlighted on green background The evolutionary relationships between A. aegrifaciens BAW48 and BAS32 and other Achromobacter spp. genomes sequenced from six continents worldwide. Whole genome sequences of twenty-four strains of human, plant and environmental origin retrieved from NCBI were used for phylogenetic analysis. The mid-point rooted tree was constructed using the NCBI Tree Viewer (https://www.ncbi.nlm.nih.gov/tools/treeviewer/), and visualized with iTOL (interactive tree of life). The evolutionary relationship was inferred using the maximum-likelihood method. Different colors (e.g., pinkish red for Asia, greenish cyan for Europe, yellowish green for North America, bluish cyan for South America, red for Australia and orange for Antarctica) are assigned according to the close evolutionary relatedness (clade) of the genomes. The scale bar is in the unit of the number of substitutions per site. The values on the branches are bootstrap support values based on 1000 replications. All the sequences were indicated by their accession numbers followed by the source and country code. The country codes according to the standard abbreviation are: United States of America (USA), Japan (JPN), China (CHN), Antarctica (ATA), Brazil (BRA), Germany (GER), Thailand (THA), Russia (RUS), India (IND), Czech Republic (CZE), Chile (CHL), Australia (AUS), Portugal (PRT) and Bangladesh (BD). The genomes of the A. aegrifaciens BAW48 (JAVCPK000000000_Water_BD) and A. aegrifaciens BAS32 (JAVCWD000000000_Soil_BD) are highlighted on green background Fig. 2Clustered heatmap of average nucleotide identity (ANI) values among the study genomes (BAW48 and BAS32) and 23 reference strains of the Achromobacter spp. (previously used in phylogenetic analysis; Table S2). The numbers represent the ANI values (%) between two genome sequences. An ANI above 95% between two genomes is an indication that they belong to the same species. Study genomes are highlighted in red box Clustered heatmap of average nucleotide identity (ANI) values among the study genomes (BAW48 and BAS32) and 23 reference strains of the Achromobacter spp. (previously used in phylogenetic analysis; Table S2). The numbers represent the ANI values (%) between two genome sequences. An ANI above 95% between two genomes is an indication that they belong to the same species. Study genomes are highlighted in red box Fig. 3Circular representation of the Achromobacter genomes. Circles (from inside to outside) 1 and 2 (GC content; black line and GC skew; purple and deep green lines), circle 3 (reference A. xylosoxidans strain H1_3_1 DNA; yellow color); circle 4 (mapped A. aegrifaciens BAW48 genome; red color); circle 5 (mapped A. aegrifaciens BAS32 genome; pinkish red color); circle 6 (mapped A. mucicolens strain DMF-4; dark tan color); circle 7 (mapped A. xylosoxidans strain DN002 genome; purple color), and circle 8 (mapped A. xylosoxidans strain GN050 genome; green color). Mapping of the genomes was done using BLASTn with an e-value cut-off 1e-5 using BRIG 0.95 Circular representation of the Achromobacter genomes. Circles (from inside to outside) 1 and 2 (GC content; black line and GC skew; purple and deep green lines), circle 3 (reference A. xylosoxidans strain H1_3_1 DNA; yellow color); circle 4 (mapped A. aegrifaciens BAW48 genome; red color); circle 5 (mapped A. aegrifaciens BAS32 genome; pinkish red color); circle 6 (mapped A. mucicolens strain DMF-4; dark tan color); circle 7 (mapped A. xylosoxidans strain DN002 genome; purple color), and circle 8 (mapped A. xylosoxidans strain GN050 genome; green color). Mapping of the genomes was done using BLASTn with an e-value cut-off 1e-5 using BRIG 0.95 Fig. 4Pangenome analysis of A. aegrifaciens strains. (a) Pangenome based (gene presence and absence) gene clustering matrix of A. aegrifaciens BAW48, A. aegrifaciens BAS32, and eight different strains of A. aegrifaciens. (b) Pangenomic curve (c) Breakdown of genes in BAW48 and BAS32 genomes Pangenome analysis of A. aegrifaciens strains. (a) Pangenome based (gene presence and absence) gene clustering matrix of A. aegrifaciens BAW48, A. aegrifaciens BAS32, and eight different strains of A. aegrifaciens. (b) Pangenomic curve (c) Breakdown of genes in BAW48 and BAS32 genomes Metal resistance genes and operon clusters in BAW48 and BAS32 strains Both of the strains contained resistance genes for a variety of heavy metals, including Cu, Zn, Co, Cd, Hg, Cr, As and others. These strains contain 23 genes involved in Cu homeostasis and tolerance, 10 genes involved in Co-Zn-Cd resistance, 4 genes involved in Hg resistance, and 11 genes involved in Cr resistance. Table 3 lists the metal resistance genes. They featured oxidase genes (aioA,aioB) that functioned in the detoxification of As compounds. Synteny analysis showed that the arrangement of arsenic resistance genes in these two genomes is nearly identical, with the exception of the arsH gene (Fig. 5). The As resistance genes identified in both of the genomes include As metallochaperone arsD which transfers the trivalent metalloid to arsAB pump, As(III)/antimonite: H + antiporter arsB, As(III)/antimonite pump driving ATPase arsA, As resistance protein arsH, As transcriptional regulator arsR, flavin dependent monooxygenase arsO which is associated with As resistance, glutathione and thioredoxin coupled As(V) reductase (arsC, arsCs), As(III) oxidase large subunit aioA, and As(III) oxidase small subunit aioB (Fig. 5; Table 3). However, we observed a relocation and inversion of the arsH gene in the BAW48 and BAS32 genomes (Fig. 5), suggesting significant genomic rearrangement that could impact gene expression, and enhance adaptability to As stress. Overall, there were three As gene clusters such as As(III) oxidizing aioBA, As(III) reducing arsRCDAB and the MMA(III) oxidizing ars resistance gene (arsHCsO) cluster in the both genomes of A. aegrifaciens strains/isolates (Fig. S3). In aioBA gene cluster (Fig. S3a), aioA,aioB are preceded by the ABC transporter substrate binding gene (aioX), and a two-component histidine kinase gene (aioS). Moreover, there is a Hg reductase gene cluster (mer) following aioBA gene cluster. In the arsRCDAB gene cluster (Fig. S3b), the As(V) reductase thioredoxin coupled gene (arsC) is preceded by phosphate ABC transporter ATP binding gene (pstB), arsD,arsA and arsB genes. In arsHCsO gene cluster (Fig. S3c), the arsH gene (As resistance gene) followed by gluteredoxin coupled arsenate reductase (arsCs), flavin dependent monooxygenase arsO, phosphate ABC transporter substrate binding genes (pstS,pstA,pstC,pstB) were present. The heatmap (Fig. 6) revealed that certain protein families, such as arsR,arsC,arsH are consistently present across all strains, indicating a potential core functional role. In contrast, protein families like aioA and aioB were found in only a few genomes, suggesting a more specialized or niche function. Synteny analysis with the reported genomes of different A. aegrifaciens strains (Fig. 7a) revealed that all strains possess at least one arsB gene, indicating similar genomic organization. However, only three strains exhibited an extra arsB gene in an operon system containing arsC, arsD, arsA, and arsB. The overall gene arrangement differs significantly among the three strains, with CTOTU4834 showing distinct clustering of resistance-related genes compared to the more linear arrangements seen in BAW48 and BAS32 (Fig. 7b). Importantly, both A. aegrifaciens strains BAW48 and BAS32 possessed the aioA (large subunit) and aioB (small subunit) genes for As(III) oxidation, while A. aegrifaciens CTOTU4834 lacks these genes. This difference indicates that BAW48 and BAS32 have the potential for As(III) oxidation, which is absent in CTOTU4834. Additionally, the arsA and arsD genes were found in both BAW48 and BAS32 but were absent in CTOTU4834 (Fig. 7b). Table 3Metal resistance genes detected in A. aegrifaciens strains BAW48 and BAS32MetalsResistance genesCu homeostatisciA,clfA,mO,copG,CRB,cusS,cusRCo-Zn-Cd resistanceczrR,hmhKZn resistance zraR Hg resistancemerR,merT,merA,merE,merPCu tolerancecutE,corC,cutAAs resistanceaioA,aioB,arsA,arsB,arsH,arsD,arsR,arsOCr resistancechrB,chrA,chrF Table 3Metal resistance genes detected in A. aegrifaciens strains BAW48 and BAS32MetalsResistance genesCu homeostatisciA,clfA,mO,copG,CRB,cusS,cusRCo-Zn-Cd resistanceczrR,hmhKZn resistance zraR Hg resistancemerR,merT,merA,merE,merPCu tolerancecutE,corC,cutAAs resistanceaioA,aioB,arsA,arsB,arsH,arsD,arsR,arsOCr resistancechrB,chrA,chrF Metal resistance genes detected in A. aegrifaciens strains BAW48 and BAS32 Fig. 5Synteny analysis of A. aegrifaciens As(III) oxidizing aioBA, As(III) reducing arsRCDAB and the MMA(III) oxidizing ars resistance (arsHCsO) gene clusters. From top to the bottom of the figure the genome of strains BAW48 and BAS32 are shown. Genes are represented by arrows of different color. Conserved genes are connected by lines Synteny analysis of A. aegrifaciens As(III) oxidizing aioBA, As(III) reducing arsRCDAB and the MMA(III) oxidizing ars resistance (arsHCsO) gene clusters. From top to the bottom of the figure the genome of strains BAW48 and BAS32 are shown. Genes are represented by arrows of different color. Conserved genes are connected by lines Fig. 6Heatmap depicting the distribution of arsenic resistance-related protein families across the ten strains examined (for pangenome analysis) Heatmap depicting the distribution of arsenic resistance-related protein families across the ten strains examined (for pangenome analysis) Fig. 7Synteny analysis of arsenic reducing and oxidizing gene clusters. (a) A. aegrifaciens As(III) reducing arsB gene clusters and (b) As(III) oxidizing aioBA gene clusters Synteny analysis of arsenic reducing and oxidizing gene clusters. (a) A. aegrifaciens As(III) reducing arsB gene clusters and (b) As(III) oxidizing aioBA gene clusters Genomic functional potentials of the BAW48 and BAS32 strains To elucidate the potential genomic features in the BAW48 and BAS32 strains of A. aegrifaciens, we predicted mobile genetic elements (MGEs), antimicrobial resistance genes (ARGs), virulence factor genes (VFGs), and metabolic functional genes/pathways. The overall genomic potentials representing MGEs, ARGs and VFGs were almost similar in both of the genomes. Different MGEs were predicted in the genomes of BAW48 (Fig. 8a) and BAS32 (Fig. 8b). Besides, both of the strains were found to harbor six ARGs such as oprZ,axyY,axyX,tetC,adeF and qacG (Fig. 8). There was a substantial overlap in the predicted metabolic features in both of the genomes. Figure 9 illustrates the comprehensive metabolic functional profile of the genomes. In BAW48 and BAS32 genomes, we identified 362 and 361 metabolic features under different SEED subsystem categories, respectively, with a coverage of 26%. The majority of these subsystems were primarily characterized by genes/pathways associated with metabolism amino acid and derivatives (22.2%), carbohydrate metabolism (13.18%), cofactors, vitamins, prosthetic groups and pigments (8.08%), protein (7.98%), and fatty acids, lipids and isoprenoids (7.38%) (Fig. 9). Furthermore, 222 genes (4.27%) were annotated for stress response, 174 genes (3.35%) for membrane transport, and 141 genes (2.71%) for virulence, disease and defense. Importantly, both of the genomes harbored brkB, a virulence factor that is essential for resistance to complement-dependent killing by serum. Fig. 8Circular genome representation of A. aegrifaciens strain (a) BAW48 and (b) BAS32 showing different antibiotic resistance genes (ARGs) and mobile genetic elements (MGEs). Different colors (e.g., blue for replication/recombination/repair, green for phage, black for transfer, pink for stability/transfer/defense, cyan for integration/excision and red for ARGs) are assigned for MGEs and ARGs of both genomes Circular genome representation of A. aegrifaciens strain (a) BAW48 and (b) BAS32 showing different antibiotic resistance genes (ARGs) and mobile genetic elements (MGEs). Different colors (e.g., blue for replication/recombination/repair, green for phage, black for transfer, pink for stability/transfer/defense, cyan for integration/excision and red for ARGs) are assigned for MGEs and ARGs of both genomes Fig. 9An overview of the metabolic functions predicted in A. aegrifaciens strains BAS32 and BAW48. Bar plots showing metabolic functional pathways (Y-axis) and the assigned number of genes (X-axis). The BAW48 and BAS32 genomes are represented in red and blue colors, respectively An overview of the metabolic functions predicted in A. aegrifaciens strains BAS32 and BAW48. Bar plots showing metabolic functional pathways (Y-axis) and the assigned number of genes (X-axis). The BAW48 and BAS32 genomes are represented in red and blue colors, respectively Discussion High As concentrations in groundwater and soil in Bangladesh present significant health risks. This study explores the As detoxification potential of A. aegrifaciens strains BAW48 and BAS32, which demonstrate the ability to oxidize As(III) and contain key arsenic-related genes, aioA and arsB. Genomic analysis confirms their role in As bioremediation and reveals their resistance to various heavy metals, crucial for survival in contaminated environments [2, 53]. Isolated from As-contaminated tubewell water and surrounding soil, these strains were shown to convert toxic As(III) to the less harmful As(V). High-throughput WGS uncovered arsenotrophic gene clusters and genes conferring resistance to both metals and antibiotics. Despite differences in MIC values between the two strains, their genomic similarity suggests related As resistance mechanisms influenced by genetic, metabolic, and environmental factors. Our findings are consistent with previous research by Istiaq et al. [21], who reported the ability of A. xylosoxidans strain BHW-15 to oxidize As(III). High-throughput WGS analysis revealed a range of genomic features that collectively define these strains as arsenotrophic. The genomic analysis of these bacterial isolates provides valuable insights into the coexistence of genes that confer resistance to both heavy metals and antibiotics [18, 54]. A key finding of this study is the presence of aioA and arsB genes in both Achromobacter aegrifaciens strains, which play vital roles in arsenic detoxification. The aioA gene encodes the As(III) oxidase enzyme, facilitating the conversion of toxic As(III) to less harmful forms [24], while arsB encodes a membrane protein that helps resist As(III) by exporting it from the bacterial cell, reducing its intracellular accumulation [21, 55]. Genome analysis of the BAW48 and BAS32 strains reveals promising features, including a substantial number of protein-coding genes and other genomic elements, such as prophage regions and antibiotic resistance genes. Phylogenetic analysis shows that these strains are closely related to other Achromobacter species, particularly A. xylosoxidans, which are isolated from diverse environments globally, including soil, water, and clinical samples [13, 21, 24, 56]. This suggests that A. aegrifaciens strains, like other arsenotrophic bacteria, are widely distributed across various ecological niches, emphasizing the global prevalence of As contamination and the role of arsenotrophic bacteria in mitigating its toxicity [29]. Microorganisms from environments assumed to be As-free still carry arsenotrophic genes, suggesting influences beyond local As concentrations. This highlights microbial adaptability and the pressing need for research on the ecological and health implications of As distribution, which can guide strategies to manage As contamination. Pangenome tree analysis of ten strains shows that A. aegrifaciens strain LMG 26,852 [56] is the closest relative to BAW48 and BAS32, with these two strains sharing a more recent common ancestor with each other than with other strains, underscoring their high genetic similarity. However, they also exhibit notable differences in gene presence and absence when compared to other A. aegrifaciens strains. The preliminary pangenomic curve suggests that A. aegrifaciens has an open pangenome, indicating significant genetic diversity. This diversity points to frequent gene acquisition, likely through horizontal gene transfer, which enhances the bacteria’s ability to adapt [57]. This study reveals that both A. aegrifaciens strains carry metal resistance genes, including for Cu, Zn, Co, Cd, Hg, Cr, and As. Their genomes show diverse defense systems, reflecting adaptability to various ecological niches. The strains also possess As-metabolic strategies, such as As(III) oxidation to the less toxic As(V), highlighting the role of environmental organisms in harboring resistance genes against both metals and toxic compounds [2, 31, 54]. In this study, we identified three As-related gene clusters in the A. aegrifaciens BAW48 and BAS32 genomes such as aioBA, arsRCDAB, and arsHCsO. These clusters, with distinct gene organization patterns, are commonly found in arsenotrophic bacterial genomes [58]. Furthermore, these clusters were previously found in highly efficient As(III)-oxidizing bacteria, including A. arsenitoxydans SY8 [11] and A. xylosoxidans BHW-15 [21]. The arsRDABC operon, which includes genes like arsA, arsB, and arsC, facilitates As resistance through an efflux pump [58, 59], likely evolved for better As tolerance. Our findings align with previous studies on adrenotropic bacteria including Pseudomonas putida KT2440 [60] and Pantoea sp. IMH [61]. The presence of multiple operons in BAW48 and BAS32 enhances their tolerance to high As concentrations, suggesting their role in As transformation and groundwater cycling. Despite the presence of multiple operons, the mechanisms of As resistance in these bacteria are complex and still underexplored. Further investigation into the regulation and contributions of these operons is needed to better understand As speciation. The synteny analysis across different strains of the species reveals a conserved genomic organization surrounding the arsB gene. This conservation suggests a fundamental role for the arsB gene in the As resistance mechanisms of these bacteria [62]. Three strains harbor an additional arsB gene within an operon system that includes arsC, arsD, and arsA genes. These strains, which possess the extra arsB gene operon, also contain the aioA and aioB gene operon, organized similarly. The comparable organization of these genes indicates a well-conserved, possibly co-regulated As resistance mechanism, encompassing both reduction (through ars genes) and oxidation (through aio genes). This coordination suggests enhanced resistance and potential adaptation to As-rich environments in these isolates [63]. Both A. aegrifaciens strains share genomic traits, including genes for heavy metal resistance, MGEs, ARGs, and VFGs. These similarities suggest a high probability of horizontal transfer of ARGs among arsenotrophic A. aegrifaciens strains [64]. Notably, both strains exhibit resistance to antibiotics and heavy metals via efflux pumps, supporting the idea of co-selection [65]. They also contain genes for stress response, secondary metabolites, and the brkB gene, which aids in resistance to serum killing [66–68]. These findings highlight their shared survival strategies in arsenic-contaminated environments. Conclusion In this study, two A. aegrifaciens strains, BAW48 and BAS32, were isolated from arsenic-contaminated tubewell water and soil samples, respectively. Both strains carried genes for As(III) oxidase (aioA; 1100 bp) and As(III) reduction (arsB; 750 bp), indicating their arsenotrophic genotypes. The MIC for As(III) was 20 mM for BAW48 and 30 mM for BAS32. WGS of BAW48 and BAS32 revealed 6.9 Mbp genomes with 98.11% completeness and a high GC content of 65.9%. Both genomes contained two phage regions encoding genes for phage packaging (TrmL), lysis (genLysins, Holin, Rz/Rz1), and DNA synthesis (GtR), but lacked plasmids. Phylogenetic analysis showed that A. aegrifaciens BAS32 is a sub-lineage of A. aegrifaciens BAW48, with an ANI value > 95%. These strains were closely related to various A. xylosoxidans strains isolated from biological reactors, soil, and human clinical samples from diverse regions. Pangenome analysis predicted 10,509 genes, including 6,284 in BAW48 and 6,264 in BAS32, with over 99% of these genes present in valid strain sequences. The BAW48 and BAS32 strains carried genes for resistance to heavy metals (Co, Zn, Cu, Cd, Hg, As, Cr) and three As resistance clusters (aioBA, arsRCDAB, arsHCsO). They also contain multiple MGEs and six ARGs namely oprZ, axyY, axyX, tetC, adeF, and qacG. Additionally, 362 and 361 metabolic features were identified in BAW48 and BAS32, respectively, with key functions in amino acid metabolism (22.2%), carbohydrate metabolism (13.18%), and stress response (4.27%). Both genomes harbored the brkB virulence factor, crucial for serum resistance. These findings suggest the emergence of novel variants of A. aegrifaciens with potential arsenotrophic and heavy metal resistance properties. However, further investigations are needed to understand the various virulence strategies employed by these strains in As polluted environment across different demographics. These findings have implications for the development of new strategies for As bioremediation. Electronic supplementary material Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1 Supplementary Material 2 Supplementary Material 2 Supplementary Material 3 Supplementary Material 3 Supplementary Material 4 Supplementary Material 4