STRUCTURAL DYNAMICS OF BACTERIAL NITROREDUCTASE ENZYMES FROM A MINING-IMPACTED ENVIRONMENT IN SOUTH KALIMANTAN, INDONESIA

ARCHITECTURE FOR SUSTAINABLE CHROMIUM BIOREMEDIATION

ARTICLE HIGLIGHTS
- New bacterial enzymes found in Indonesian soil can neutralize toxic chromium waste.
- These unique bacteria show great promise for bioremediation in polluted mining soils.
- One of the newly found enzymes from these bacteria has a superior, highly stable structure.
- This superior stability is key, coming from a rigid core with flexible moving loops.
- This research provides a molecular blueprint for engineering enzymes to clean up chromium.

ABSTRACT
Hexavalent chromium [Cr(VI)], a soluble and carcinogenic industrial pollutant, poses a significant threat to both the environment and human health, necessitating effective remediation strategies. Microbial enzymatic reduction of Cr(VI) to its less toxic trivalent state, Cr(III), is a promising approach. However, growing evidence suggests that many enzymes involved in this process are flavin mononucleotide (FMN)-dependent reductases, which likely reduce Cr(VI) adventitiously via a reduced flavin intermediate, rather than through direct enzymatic catalysis. This study presents a comparative computational analysis of two novel FMN-dependent reductases, designated M2Cr10 and M54Cr10, derived from chromium-tolerant bacteria Acinetobacter radioresistens and Bacillus tropicus, respectively, which were isolated from Indonesian serpentine soil. Phylogenetic and sequence analyses classified both enzymes as members of the FMN-dependent nitroreductase superfamily. High-quality homology models were generated and validated, with over 95% of residues occupying the most favored regions of the Ramachandran plot, confirming their stereochemical integrity. Molecular docking simulations predicted strong binding affinities for the FMN cofactor, with binding energies of -7.1 kcal/mol for M2Cr10 and -8.1 kcal/mol for M54Cr10. These interactions are stabilized by a network of hydrogen bonds and hydrophobic contacts, with residue Tyr¹³¹ identified as a key anchor for the FMN isoalloxazine ring in both enzymes. Extensive 10-nanosecond molecular dynamics simulations revealed that the A. radioresistens M2Cr10 enzyme exhibits superior structural architecture characterized by greater global stability, as indicated by lower average root mean square deviation (RMSD) and solvent-accessible surface area (SASA). However, it also displays greater localized flexibility (higher RMSF) in functional loop regions critical for catalysis. This combination of a rigid scaffold and dynamic functional loops suggests that M2Cr10 may be a more robust and potentially efficient biocatalyst. These findings provide a detailed molecular blueprint for understanding the structural determinants of stability in FMN-dependent reductases and offer a rational basis for engineering these enzymes for more effective adventitious bioremediation of Cr(VI).

INTRODUCTION

Industrialization has led to the widespread release of hazardous pollutants, among which hexavalent chromium [Cr(VI)] is a priority concern due to its high solubility, mobility, and severe toxicity. As a mobile anionic complex, Cr(VI) readily contaminates soil and groundwater, where its potent oxidative and genotoxic properties pose significant risks to ecosystems and human health, highlighting the urgent need for effective remediation technologies (Shalini et al., 2023). While industrial activities are a primary source, unique geological environments, such as serpentine soils, are naturally enriched in chromium. These soils, formed from the weathering of ultramafic rocks, contain high concentrations of the less toxic, mineral-bound trivalent chromium, Cr(III) (Oze et al., 2004). In South Kalimantan, Indonesia, for instance, these serpentine soils have reported total chromium concentrations ranging from 467 to 1,843 mg/kg (Saidy & Badruzsaufari, 2009). Slow natural oxidation processes within these soils can generate mobile and toxic Cr(VI), exerting strong selective pressure on the local microbiota (Gan et al., 2024). Consequently, these serpentine environments serve as natural reservoirs for discovering novel, stress-tolerant microorganisms with potential for bioremediation (Ying et al., 2017).

Bioremediation through the microbial reduction of toxic Cr(VI) to the far less soluble and toxic Cr(III) is a promising and sustainable strategy (Thatoi et al., 2014). Historically, enzymes capable of this transformation were broadly labeled “chromate reductases.” However, a paradigm shift in the field suggests that this reduction is often an adventitious chemical process rather than a direct enzymatic one (Robins et al. 2013). This model proposes that the primary function of these enzymes, nearly all of which are flavin mononucleotide (FMN)-dependent nitroreductases, is to generate the reduced flavin cofactor (FMNH₂). This occurs through a ping-pong mechanism in which the enzyme first uses NAD(P)H to reduce its bound FMN; the highly reactive FMNH₂ is then released and chemically reduces Cr(VI) in a rapid, noncatalyzed reaction (Megharaj et al., 2003). This framework explains why bacteria such as Acinetobacter and Bacillus, genera known for their metabolic versatility and resilience in contaminated environments, are frequently implicated in chromium reduction. Their native FMN-dependent enzymes, including the putative chromate reductase (chrR) gene products investigated in this study, are therefore strong candidates for examining the structural basis of efficient FMN reduction. These properties and impacts emphasize the urgent need for effective remediation of Cr(VI)-polluted sites (Machacek, 2019)(Walwa, 2016).

While the adventitious reduction model is compelling, the specific structural and dynamic properties that distinguish a superior FMN-dependent reductase for this process remain poorly understood. To address this gap, this study used a comparative in silico approach to characterize two novel reductases, M2Cr10 and M54Cr10, from Acinetobacter radioresistens and Bacillus tropicus, respectively, isolated from Indonesian serpentine soil. The objectives were to: 1) classify these enzymes through phylogenetic and sequence analysis; 2) generate and validate high-quality three-dimensional homology models; and 3) use molecular docking and extensive molecular dynamics simulations to conduct a detailed comparative analysis of their FMN cofactor binding, overall structural stability, and localized dynamic behavior. By elucidating the key molecular determinants that confer robustness and catalytic potential, this work aimed to provide a rational foundation for selecting and engineering enzymes for more effective adventitious Cr(VI) bioremediation.

MATERIALS AND METHODS

Bacterial Strains Used

The bacterial strains used in this study, Acinetobacter radioresistens M2Cr10 and Bacillus tropicus M54Cr10, were obtained from our laboratory culture collection. These strains were previously isolated and characterized by our group from samples of serpentine soil collected in South Kalimantan, Indonesia (Aisyah et al., 2017). They were selected for this study based on their demonstrated tolerance to high concentrations of chromium. All bacterial culturing was performed in Luria-Bertani (LB) Broth. Genomic DNA was extracted using the Wizard® Genomic DNA Purification Kit (Promega, USA). Primers used to amplify the putative chromate reductase (chrR) genes are listed inTable 1. PCR amplifications were carried out using MyTaq™ HS Red Mix (Bioline, UK). Electrophoresis reagents included agarose, GelRed® (Biotium, USA), and a 100-base-pair (bp) DNA ladder (Vivantis, Malaysia). Bidirectional Sanger sequencing was performed by 1st BASE, Malaysia.

Table 1.Specific primer design

Range	Primers	Sequence
Flanking CDS	Chr-76-848F	GGAGTGATTTTCATGAATGAGATAATACA
	Chr-76-848R	CCCGCATCACAGGTTTGCCT

Bacterial Chromate Reductase Gene Isolation

Bacterial Isolation and DNA Extraction

Two chromium-resistant bacterial isolates (M2Cr10 and M54Cr10) were cultured in 5 mL of Luria-Bertani (LB) broth at 30 °C for 40 hours in a shaking water bath (150 × g) (Amilia et al., 2016). Genomic DNA was extracted using the Wizard® Genomic DNA Purification Kit (Promega, USA) following a modified protocol (Atiyea et al., 2022). Bacterial pellets were resuspended in 480 μL of 50 mM EDTA and 120 μL of lysozyme (10 mg/ mL), incubated at 37 °C for 30 – 60 minutes, and centrifuged at 15,000 × g for 2 minutes (Pluthero, 1993). Cell lysis was performed with 600 μL of Nuclei Lysis Solution at 80 °C for 5 minutes, followed by RNase treatment (3 μL) at 37 °C for 15 – 60 minutes (Park et al. 2001). Proteins were precipitated with 200 μL of Protein Precipitation Solution, and DNA was pelleted 600 μL of isopropanol and washed with 70% ethanol. Purified DNA was rehydrated in 50 μL of Rehydration Solution at 40 °C overnight (Repoles et al., 2021).

Chromate Reductase Gene Amplification

Primers targeting the chromate reductase gene (chrR) were designed using sequences of known nitroreductases (e.g., NfsA/NfsB family) from representative Bacillus and Acinetobacter species retrieved from NCBI (www.ncbi.nlm.nih.gov) and BioCyc (www.biocyc.org) (Nano transdermal delivery potential of fucoidan from Sargassum sp. (brown algae) as chemoprevention agent for breast cancer treatment, 2022). Primer specificity was validated using BLAST analysis (Rahman et al., 2022), and primers were synthesized by PT. Genetika Science Indonesia (Dibha et al., 2022). PCR amplifications were performed in 50 μL reactions containing 25 μL MyTaqHS Master Mix (Bioline, UK), 5 μL each of forward and reverse primers (10 μM), 1 μL DNA template (< 250 ng/μL), and 14 μL nuclease-free water (Ahmed et al., 2014). Thermal cycling conditions included an initial denaturation at 94 °C for 5 minutes, followed by 30 cycles of denaturation at 94 °C for 30 seconds, annealing at 53.7 °C for 30 seconds, and extension at 72 °C for 1 minute (Rapid bacterial identification by direct PCR amplification of 16S rRNA genes using the MinIONTM nanopore sequencer, 2019).

Agarose Gel Electrophoresis and Amplicon Analysis

PCR products were resolved on 1% agarose gel prepared in 0.5× Tris-borate-EDTA (TBE) buffer (Djankpa et al., 2021). Gels were stained with Diamond DNA Dye (Promega, USA) for 30 minutes, visualized under UV transilluminator (Vilber Lourmat, France), and compared with a 100-bp DNA ladder (Promega, USA). Band sizes were analyzed using CorelDraw (Corel Corporation, Canada) and Microsoft Excel (Microsoft, USA) (Al-Shuhaib et al., 2021).

Intramolecular Profiling Using Computational Methods

Gene Identification

The chromate reductase gene was identified through nucleotide and protein BLAST analyses in the GenBank database, showing the highest similarity to sequences previously reported in the literature (Jhunjhunwala et al., 2016). Further characterization was conducted by identifying protein motifs in the polypeptide sequence using the MOTIF search tool (https://www.genome. jp/tools/motif/), confirming the presence of the chromate reductase genes. The raw forward and reverse sequencing chromatograms were assembled, quality-edited, and translated into amino acid sequences using MEGA X software (Wijiastuti et al., 2015); (Kumar et al., 2018).

The consensus sequence was aligned using the Multalin server and subjected to BLAST analysis in the GenBank database to identify species with the highest genetic similarity and enzymes associated with chromate reductase activity (Sundarraj et al., 2022). The identities of the resulting protein sequences were verified using the BLASTp suite against the NCBI nonredundant protein database. Physicochemical properties, including molecular weight, theoretical isoelectric point (pI), instability index, aliphatic index, and Grand Average of Hydropathy (GRAVY), were calculated using the ExPASy ProtParam server (Yu et al., 2017). Protein motifs and domains were identified using the InterProScan tool (InterProScan 5: Genome-scale protein function classification, 2014). The predicted polypeptide sequence was compared with reference sequences through 3D molecular visualization in PyMol, with the FASTA sequence converted to PDB format using an online converter. Protein sequence families were identified using InterPro, while protein domains were analyzed using the Conserved Domains Database (CDD) (Finn et al., 2016); (Derbyshire et al., 2015). Functional analysis of the protein was conducted using the ProFunc web server, and virulence, an indicator of bacterial pathogenicity, was assessed using VirulentPred, a SVM-based tool (Garg & Gupta, 2008).

Docking and Interaction Study

Three-dimensional (3D) structural models of the M2Cr10 and M54Cr10 proteins were generated using SWISS-MODEL, an automated homology modeling server that identifies suitable templates from the Protein Data Bank (PDB) and constructs the corresponding model (Waterhouse et al. 2018). The quality and stereochemical validity of the generated models were evaluated using Ramachandran plot analysis performed with the PROCHECK server (Laskowski et al., 1993).

Molecular docking was performed to predict the binding mode of the FMN cofactor within the active site of each enzyme. The 3D structure of FMN was obtained from the PubChem database (CID: 643978). Docking was conducted using PyRx software, which incorporates AutoDock Vina for its docking algorithm (Dallakyan & Olson 2015). The search space (grid box) was centered on the predicted active site based on alignment with the template structures. The binding pose with the lowest (most favorable) binding energy was selected for further analysis. Interactions between the protein and ligand were visualized using BIOVIA Discovery Studio Visualizer (Kaushik, 2021).

Molecular Dynamics Simulation

All-atom molecular dynamics (MD) simulations were performed for 10 ns on both FMN-docked protein complexes using GROMACS 2022.3 (Pall et al., 2015); (Berendsen et al., 1995). The systems were prepared using the GROMOS96 54a7 force field. Each complex was solvated in a cubic box with SPC water molecules, maintaining a minimum distance of 1.0 nm from the box edge. The systems were neutralized by adding counter-ions. Energy minimization was conducted using the steepest descent algorithm, followed by a two-step equilibration procedure (10 ps NVT and 10 ps NPT) to stabilize the system’s temperature at 300 K and pressure at 1 bar. Production MD simulations were then carried out for 10 ns. Trajectories were analyzed to calculate the Root Mean Square Deviation (RMSD), Root Mean Square Fluctuation (RMSF), Radius of Gyration (Rg), and Solvent Accessible Surface Area (SASA) (Paul et al., 2020); (Paul et al., 2014). Visualization and analysis of simulation results were performed using XMGrace.

RESULTS AND DISCUSSION

Chromate Reductase Gene Isolation

PCR Amplification

PCR amplification of the target gene from both A. radioresistens M2Cr10 and B. tropicus M54Cr10 produced a single, distinct band of the expected size (~773 bp) Figure. 1, confirming successful and specific amplification.

Figure 1.Electrophoretic analysis of PCR products from A. radioresistens M2Cr10 (lanes 1-3) and B. tropicus M54Cr10 (lanes 4-6), representing technical replicates.

Sequence Analysis

BLAST (Basic Local Alignment Search Tool) analysis of 16S rRNA sequences revealed that isolate M54Cr10 shared 99.50% similarity with Bacillus tropicus strain MCCC 1A01406, whereas isolate M2Cr10 showed 99.77% similarity with Acinetobacter radioresistens strain NRBC Table 2. Bacillus tropicus, a Gram-positive bacterium commonly found in tropical soils and marine environments, is recognized for its enzymatic versatility and bioremediation potential (Aregbesola et al., 2021).

Acinetobacter radioresistens, a Gram-negative species, demonstrates remarkable environmental adaptability, including resistance to radiation and osmotic stress. Notably, A. radioresistens strain NS-MIE, isolated from Malaysian agricultural soil, was reported to reduce 75.13 – 96.27% of Cr(VI) to Cr(III) at pH 6 in nutrient broth containing 60 ppm chromium (Talib et al., 2019).

Table 2.BLAST analysis Results of 16S ribosomal RNA sequences using NCBI

Isolate	Species	Query cover	E-value	Identity (%)
M2Cr10	Acinetobacter radioresistens strain NRBC	100%	0	99.77%
M54Cr10	Bacillus tropicus strain MCCC 1A01406	99%	0	99.50%

To further elucidate evolutionary relationships, nucleotide sequences of the chromate reductase gene (chrR) were translated into amino acid sequences using BLASTx. Subsequent protein sequence alignment and phylogenetic analysis provided insights into functional and evolutionary conservation between these isolates and related species, supporting their potential for chromium bioremediation applications.

Amino acid profiling of chromate reductase enzymes from Acinetobacter radioresistens M2Cr10 and Bacillus tropicus M54Cr10 revealed conserved residue types with distinct compositional proportions Table 3 & 4. Both enzymes exhibited identical amino acid repertoires, with glutamic acid (E: 9.8 – 9.9%) being the most abundant residue and tryptophan (W: 0.4%) as the least prevalent. Analysis of charge distribution indicated a net positive charge in both proteins, reflecting higher proportions of positively charged residues (arginine [R], lysine [K], histidine [H]) relative to negatively charged residues (aspartic acid [D], glutamic acid [E]). Specifically, M2Cr10 contained 40 positively charged and 36 negatively charged residues, whereas M54Cr10 contained 41 positive and 38 negative residues. This charge asymmetry may facilitate potential electrostatic interactions with anionic substrates such as chromate ions (CrO ²^-).

Table 3.Protein motif and sequence identification

Isolate	Protein motif	Protein sequence	Amino acid total
M2Cr10	NADH ubiquinone oxidoreductase	MNEIIHKMEQHVSVRKYKEESIPKDVVERMVQAGQH AASSHFVQAYSVIYVTDQDLKAKLADLSGNRHVKDC AAFFVCCADLKRLEMACEKHGTEIKHEGVEDFIVATV DASLFAQNLALAAESLGYGICYIGGIRNNPREVSELLH LPDKVYPVFGMTVGVPDENHGVKPRLPIAAILHENGY DEKKYDELLNEYDETMNAYYKERSSNKKNATWTES MSSFMSKEKRMHMKE	234
M54Cr10	NADPH-Dependent FMN reductase	MNEIIHKMEQHVSVRKYKEESIPKDVVERMVQAAQH AASSHFVQAYSVIYVTDQDLKAKLAELSGNRHVKDC AAFFVCCADLKRLEMACEKHGTEIKHEGVEDFIVATV DASLFAQNLALAAESLGYGICYIGGIRNNPGEVSELLH LPDKVYPVFGMTVGVPDEDHGVKPRLPIAAILHENGY DEKKYDELLNEYDETMSAYYKERSSNQKNVTWTES MSSFMSKEKRMHMKEFLSEKGFNKK.AKRAKRQTC	253

Table 4.Amino acid amount and composition of isolates M2Cr10 and M54Cr10

Types of amino acids	3 letter code	Amount and composition	Amount and composition
		(M2Cr10)	(M54Cr10)
Alanine (A)	Ala	20 (8.5 %)	22 (8.7 %)
Arginine (R)	Arg	9 (3.8 %)	10 (4.0 %)
Asparagine (N)	Asn	11 (4.7 %)	10 (4.0 %)
Aspartic Acid (D)	Asp	13 (5.6 %)	13 (5.1 %)
Cysteine (C)	Cys	5 (2.1 %)	6 (2.4 %)
Glutamine (Q)	Gln	6 (2.6 %)	8 (3.2 %)
Glutamic Acid (E)	Glu	23 (9.8 %)	25 (9.9 %)
Glycine (G)	Gly	12 (5.1 %)	13 (5.1 %)
Histidine (H)	His	11 (4.7 %)	11 (4.3 %)
Isoleucine (I)	Ile	11 (4.7 %)	11 (4.3 %)
Leucine (L)	Leu	16 (6.8 %)	17 (6.7 %)
Lysine (K)	Lys	20 (8.5 %)	24 (9.5 %)
Methionine (M)	Met	10 (4.5 %)	10 (4.0 %)
Phenylalanine (F)	Phe	7 (3.0 %)	9 (3.6 %)
Proline (P)	Pro	7 (3.0 %)	7 (2.8 %)
Serine (S)	Ser	15 (6.4 %)	17 (6.7 %)
Threonine (T)	Thr	7 (3.0 %)	8 (3.2 %)
Tryptophan (W)	Trp	1 (0.4 %)	1 (0.4 %)
Tyrosine (Y)	Tyr	11 (4.7 %)	11 (4.3 %)
Valine (V)	Val	19 (8.1 %)	20 (7.9 %)
Selenocysteine (O)	Sec	0	0
Pyrrolysine (U)	Pyl	0	0

Protein secondary structures, including α-helices and β-sheets, are critical for stabilizing the three-dimensional conformation of enzymes Figure. 2. α-Helices are formed through hydrogen bonding between the amine hydrogen (N-H) of one residue and the carboxyl oxygen (C=O) of another, forming rigid helical backbones often embedded within protein cores. The β-Sheets consist of extended strands connected by hydrogen bonds and can be classified as parallel (strands aligned in the same direction) or antiparallel (strands aligned in opposite directions). The parameter η (eta) represents positional conservation metrics in sequence alignments, indicating the likelihood that specific residues occupy conserved sites, which can inform structural or functional predictions.

Figure 2.Multiple sequence alignment of M2Cr10 and M54Cr10 against a key reference nitroreductase, NfrA1 (Bacillus megaterium, PDB: 5HDJ)

Phylogenetic Analysis

BLASTp analysis of the translated amino acid sequences, followed by phylogenetic analysis, classified both enzymes as members of the FMN-dependent nitroreductase superfamily Figure. 3. M2Cr10 was identified as an NADH-ubiquinone oxidoreductase (NfsB-like family), while M54Cr10 was classified as an NADPH-dependent FMN reductase (NfsA-like family). Genetic divergence analysis Figure. 4 indicated that M2Cr10 and M54Cr10 do not form a distinct clade but are interspersed with known nitroreductases from various bacteria, including Bacillus subtilis and Escherichia coli. These findings support the hypothesis that the enzymes are not specialized chromate reductases but rather multifunctional nitroreductases capable of adventitious Cr(VI) reduction.

Bootstrap-supported clustering (1,000 replicates) positioned M2Cr10 within Clade I alongside Bacillus subtilis NfrA1 (PDB: 3N2S), Escherichia coli K12 (PDB: 1F5V), and Vibrio harveyi (PDB: 1BKJ). M54Cr10 clustered closely with Bacillus megaterium NfrA1 (PDB: 5HDJ) and Bacillus subtilis (PDB: 1ZCH), exhibiting genetic divergence distances of 0.816 - 1.086. These distances, calculated using the Poisson correction model, reflect conserved functional domains critical for chromate reduction.

The structural homology of M2Cr10 and M54Cr10 with nitroreductase-like proteins (e.g., Bacteroides fragilis NCTC 9343 [PDB: 3EOF], Parabacteroides distasonis ATCC 8503 [PDB: 3M5K]) suggests shared catalytic mechanisms, particularly in FMN-dependent redox reactions. Notably, M54Cr10’s alignment with Priestia megaterium NfrA2 (PDB: 5HEI) highlights the evolutionary conservation of NADPH-binding motifs, whereas M2Cr10’s proximity to Geobacter metallireducens GS-15 (PDB: 4DN2) indicates adaptive divergence in electron transport pathways.

Figure 3.Phylogenetic tree showing the evolutionary relationship of M2Cr10 and M54Cr10 with other known bacterial nitroreductases.

Figure 4.Genetic divergence showing Pairwise evolutionary distances between nitroreductase sequences

Physicochemical Properties and Structural Model Validation

The predicted physicochemical properties of the two enzymes suggest distinct structural characteristics Table 5. M2Cr10 has a predicted molecular weight of 26.56 kDa and a theoretical pI of 5.89, while M54Cr10 is larger (28.72 kDa) with a higher pI (6.27). The instability index, which predicts in vivo stability, was 36.87 for M2Cr10 and 40.82 for M54Cr10. Values below 40 indicate stable proteins, while values above 40 suggest potential instability, indicating M2Cr10 is likely stable, whereas M54Cr10 may be less stable (Guruprasad et al., 1990). Both enzymes exhibited negative Grand Average of Hydropathy (GRAVY) scores, consistent with hydrophilic proteins and supporting their cytosolic localization and function in an aqueous environment.

Table 5.Comparative physicochemical properties of M2Cr10 and M54Cr10

Enzyme	Molecular weight	Total atom number	Approximate isoelectric point (pI)	stability index	index	(GRAVY)
M2Cr10	26563.22	3692	5.89	36.87	77.09	-0.471
M54Cr10	28716.75	4000	6.27	40.82	74.78	-0.489

High-quality 3D homology models were generated for both enzymes Figure. 5. Validation using PROCHECK confirmed excellent stereochemical quality. The M2Cr10 model contained 96.1% of residues in the most favored regions of the Ramachandran plot, whereas the M54Cr10 model contained 95.8%, with no residues in disallowed regions for either model. These results indicated that both models are reliable and provide a solid foundation for subsequent docking and molecular dynamics analyses (Martin et al., 2015).

Figure 5.homology models of M2Cr10 (A) and M54Cr10 (B) generated by SWISS-MODEL and visualized in PyMOL

Molecular Docking Reveals a Conserved FMN Binding Mode

The structural conservation of these active-site residues underscores evolutionary optimization for chromate reduction, with α/β folds stabilizing cofactor binding and catalytic loops. Further site-directed mutagenesis studies are warranted to validate the functional roles of these residues in Cr(VI) detoxification (Carles et al. 2016).

Molecular docking of FMN with chromate reductases from Acinetobacter radioresistens M2Cr10 and Bacillus tropicus M54Cr10 identified distinct binding residues and energy profiles Table 6; Figure 6. In M2Cr10, FMN interactions involved His¹¹, Ser¹³, Arg¹5, Ser39, Ser40, His41, Ala38, Asn66, Tyr¹³¹, Gly¹³³, and Lys170, resulting in a binding energy of -7.1 kcal/mol. In contrast, M54Cr10 exhibited interactions at Arg15, Ser39, Ser40, Phe42, Asn66, Tyr¹³¹, Gly¹³³, and Gly134, with a lower binding energy of -8.1 kcal/mol for M54Cr10, indicating stronger ligand affinity.

Table 6.Location of residues that actively bind to FMN ligands in each isolate

Enzyme	FMN active bond site at lowest energy
M2Cr10	His	Ser	Arg	Ser	Ser	His	Ala	Asn	Try	Lys
	11	13	15	39	40	41	38	66	131	170
M54Cr10	-	-	Arg	Ser	Phe	-	Asn	Tyr	Gly	Gly
			15	39	40	42	-	66	131	133

Figure 6.visualization of the FMN cofactor docked into the active sites of M2Cr10 (A) and M54Cr10 (B)

Hydrogen bonding and hydrophobic interactions dominated the binding mechanisms. In M2Cr10, Arg15 formed a hydrogen bond with the FMN ligand (3.25 Å), while Tyr¹³¹ and Lys170 engaged in hydrophobic contacts. In M54Cr10, Ser39 formed a hydrogen bond (3.26 Å), and Phe42 contributed to hydrophobic stabilization (4.96 Å). The greater residue diversity within M2Cr10’s binding pocket suggests broader interaction networks, whereas the more compact binding site in M54Cr10’s may enhance ligand-binding efficiency.

Binding energy, a critical determinant of ligand-protein stability, reflects the thermodynamic favorability of interactions (Putra et al., 2023); Jakhmola et al. 2022). Molecular docking predicted that FMN binds favorably within the active site of both enzymes, with binding energies of -7.1 kcal/ mol for M2Cr10 and -8.1 kcal/mol for M54Cr10, indicating stronger affinity in M54Cr10. In both enzymes, the FMN cofactor is stabilized by a network of hydrogen bonds and hydrophobic interactions. Energy decomposition analysis revealed a key role for residue Tyr¹³¹, which provides significant hydrophobic and π-stacking interactions with the isoalloxazine ring of FMN, acting as a crucial anchor. This observation highlights a conserved structural feature for cofactor stabilization within this enzyme family.

Energy decomposition analysis using the FastRDH web server Figure 7 identified Tyr¹³¹ as a critical residue governing flavin mononucleotide (FMN) binding in both Acinetobacter radioresistens M2Cr10 and Bacillus tropicus M54Cr10. The exceptionally low interaction energy (−ΔG) associated with Tyr¹³¹ highlights its role in stabilizing FMN through hydrophobic interactions, a mechanism not previously reported for chromate reductases. Hydrophobic exclusion of water molecules by Tyr¹³¹ significantly enhanced binding stability, as evidenced by energy profiles and molecular dynamics trajectories.

This novel finding highlights Tyr¹³¹ as a conserved anchor within FMN-binding pockets, providing insights into evolutionary optimization of redox cofactor retention. The lack of prior reports on Tyr¹³¹’s contribution to FMN affinity highlights the uniqueness of these enzymes’ ligand-binding architecture. These results provide a structural framework for the rational engineering of chromate reductases, enabling targeted modifications to augment cofactor binding efficiency for enhanced Cr(VI) reduction.

Figure 7.The energy decomposition analysis per-residue of enzymes M2Cr10 (A) and M54Cr10 (B)

Molecular Dynamics Simulation

To compare the structural dynamics, 10 ns molecular dynamics (MD) simulations were performed. The RMSD of the protein backbone atoms was monitored to assess overall structural stability Figure 8. The M2Cr10 enzyme rapidly reached equilibrium and maintained a lower and more stable RMSD (average ~0.45 nm) throughout the simulation, whereas M54Cr10 showed greater deviation and conformational drift (average ~0.60 nm). The results indicate that M2Cr10 has a more rigid and stable global structure.

Initial conformational divergence during the first 0 - 3 ns of the simulation suggested distinct structural sampling pathways between the two enzymes. From 3 - 10 ns, M2Cr10 exhibited consistently lower RMSD values (mean: 1.2 ± 0.3 Å) compared to M54Cr10 (mean: 2.8 ± 0.6 Å), indicating enhanced structural rigidity and reduced conformational flexibility in M2Cr10.The pronounced RMSD fluctuations in M54Cr10 correlate with its higher aliphatic index (74.78 vs. 77.09 for M2Cr10), suggesting that reduced thermostability may permit greater loop mobility. These observations align with stability indices predicted by ProtParam (M2Cr10: 36.87; M54Cr10: 40.82), where lower indices denote higher stability (Cunha et al., 2015). The structural rigidity of M2Cr10 implies evolutionary optimization for catalytic consistency, whereas the flexibility of M54Cr10 may facilitate adaptive substrate binding.

Figure 8.Backbone root-mean-square deviation (RMSD) of backbone atoms for M2Cr10 (black) and M54Cr10 (red) plotted against simulation time (ns)

Root mean square fluctuation (RMSF) analysis was performed to evaluate residue-specific flexibility in Acinetobacter radioresistens M2Cr10 and Bacillus tropicus M54Cr10 during 10 ns molecular dynamics simulations Figure 9. M2Cr10 exhibited higher RMSF values (mean: 1.8 ± 0.4 Å) when compared with M54Cr10 (mean: 1.2 ± 0.3 Å), particularly in residues 170 – 200 and 225 – 250, indicating enhanced flexibility in loop regions critical for substrate binding. In contrast, M54Cr10 displayed reduced fluctuations, suggesting structural rigidity that may contribute to the stabilization of FMN cofactor interactions.

Figure 9.Root Mean Square Fluctuation (RMSF) per residue for M2Cr10 (black) and M54Cr10 (red)

This observation was further supported by the Solvent Accessible Surface Area (SASA) analysis, in which M2Cr10 consistently showed lower values, indicating a more compact fold (Fig. 10C). The Radius of Gyration (Rg) was comparable for both enzymes, although M54Cr10 displayed a slightly more compact core Figure. 10A & B.

Analysis of local flexibility via RMSF revealed that, despite its global stability, M2Cr10 exhibits significantly higher flexibility in specific loop regions (residues ~170–200)comparedtothemorerigidM54Cr10. These flexible loops, located near the active site, are critical for substrate access and product release. These findings suggest a functional trade-off:M2Cr10 combine sastable structural scaffold with dynamic, flexible loops, an architecture often associated with efficient catalysis in robust enzymes.

Figure 10.Radius of gyration in isolate M2Cr10 (A) and M54Cr10 (B) and Solvent Accessible Surface Area (SASA) (C) analysis of two enzymes

CONCLUSION

Several key conclusions from this study has enhanced our understanding of microbial chromium reduction and provide a rational basis for developing improved bioremediation technologies. First, comprehensive sequence and phylogenetic analyses unequivocally classified both M2Cr10 and M54Cr10 as members of the FMN-dependent nitroreductase superfamily. This finding supports the contemporary paradigm that Cr(VI) reduction by these enzymes is not a specialized, directly catalyzed reaction but rather an adventitious chemical process driven by the primary function of the enzymes as FMN reductases. Second, structural and dynamic characterization identified M2Cr10 from A. radioresistens as a superior candidate for bioremediation applications. Although both enzymes share a conserved nitroreductase fold, molecular dynamics simulations revealed that M2Cr10 combines high global stability with pronounced localized flexibility in loop regions critical for catalytic function. This architecture, a combination of rigid scaffold for robustness and dynamic loops for catalytic efficiency, suggests that M2Cr10 is better adapted to function effectively under the challenging and variable conditions of chromium-contaminated environments, such as ultramafic soils. Based on these computational insights, future study should focus on experimental validation. Key next steps include heterologous expression and purification of both enzymes, kinetic assays to quantify FMN-reducing activity and adventitious Cr(VI) reduction efficiency, and site-directed mutagenesis of critical residues, such as Tyr¹³¹, to confirm their roles in cofactor binding and stability. For example, a Y131F (Tyrosine to Phenylalanine) mutation could directly test the importance of the π-stacking interaction versus the hydroxyl group, while a Y131A (Tyrosine to Alanine) mutation would evaluate whether the aromatic ring itself serves as the essential anchor. This integrated approach will be crucial for translating molecular insights into practical bioremediation tools.

References

1. Ahmed O.B., Asghar A.H., Elhassan M.M.. Comparison of three DNA extraction methods for polymerase chain reaction (PCR) analysis of bacterial genomic DNA. Afr J Microbiol Res. 2014; 8:598-602. DOI
2. Aisyah Dindin H.M., SN Hasrul, Badruzsaufari. Identifikasi bakteri pereduksi kromium dari tanah serpentin [Identification of chromium-reducing bacteria from serpentine soil. Bioscientiae. 2017; 14(1):16-24. DOI
3. Al-Shuhaib M.B.S., Al-saadi A., Ewadh M., Noor M.. Low cost alternative method to monitor the electrophoretic bands substitution of the positively charged rabbits seminal fluids. J Biol Agric Healthc. 2021; 3:74-8.
4. Amilia N., Budiarto B.R., Mustopa A.Z., Aprilian T., Manguntungi B., Saepudin E.. Isolation of DNA aptamers for Enteropathogenic Escherichia coli (EPEC) detection using bacterial-SELEX approach. HAYATI. 2016; 29(6):104308296789-798.
5. Aregbesola O.A., Kumar A., Mokoena M.P., Olaniran A.O.. Whole-genome sequencing, genome mining, metabolic reconstruction and evolution of pentachlorophenol and other xenobiotic degradation pathways in Bacillus tropicus strain AOA-CPS1. Funct Integr Genomics. 2021; 21(2):171-93. DOI
6. Atiyea Q.M., Al-najar F.M., Karim G.F., AL-Salihi S.S.. Molecular evaluation of the impact of Nd:YAG laser and static magnetic field on genomic DNA of some bacterial isolates using RAPD-PCR. J Pure Appl Microbiol. 2022; 16(3):2072-82. DOI
7. Berendsen H.J.C., Spoel D., Drunen R.. GROMACS: A message-passing parallel molecular dynamics implementation. Comput Phys Commun. 1995; 91:43-56. DOI
8. Cunha K.C., Rusu V.H., Viana I.F., Marques E.T., Dhalia R., Lins R.D.. Assessing protein conformational sampling and structural stability via de novo design and molecular dynamics simulations. Biopolymers. 2015; 103:351-61. DOI
9. Derbyshire M.K., Gonzales N.R., Lu S., He J., Matchler G.H., Wang Z., Marchler-Bauer A.. Improving the consistency of domain annotation within the conserved domain database. Database (Oxford. 2015. DOI
10. Dibha A.F., Wahyuningsih S., Ansori A.N., Kharisma V.D., Widyananda M.H., Parikesit A.A.. Utilization of secondary metabolites in algae Kappaphycus alvarezii as a breast cancer drug with a computational method. Phog J. 2022; 14(3):536-43. DOI
11. Djankpa F.T., Wiafe G.A., Boateng B.N.. Assessment of corn starch as substitute for agarose in DNA gel electrophoresis. BMC Res Notes. 2021; 14(73)DOI
12. Finn R.D., Coggill P., Eberhardt R.Y., Eddy Mistry, J Mitchell, A.L.. The Pfam protein families database: Towards a more sustainable future. Nucleic Acids Res. 2016; 44(D1)DOI
13. Gan M., Zhou Y., Huang D., He P., Tang B., Cai Y., Zhu J.. The enhanced effect of key microorganisms in chromium-contaminated soil in Cr(VI) reduction. Chemosphere. 2024; 350(142682)DOI
14. Garg A., Gupta D.. VirulentPred: a SVM based prediction method for virulent proteins in bacterial pathogens. BMC bioinformatics. 2008; 9(62)DOI
15. Guruprasad K., Reddy B., Pandit M.. Correlation between stability of a protein and its dipeptide composition: a novel approach for predicting in vivo stability of a protein from its primary sequence. Protein Eng. 1990; 4(1):155-61. DOI
16. Nano transdermal delivery potential of fucoidan from Sargassum sp. (brown algae) as chemoprevention agent for breast cancer treatment. Phcog J. 2022; 14(6):789-95. DOI
17. Jhunjhunwala P., Roy R., Roy S., Chaudhuri A.N.. Bioremediation of chromium by unique chromium reductase activity in Proteus sp. isolated from waste water. Int J Appl Res. 2016; 2:249-56.
18. InterProScan 5: Genome-scale protein function classification. Bioinformatics. 2014; 30(9):1236-40. DOI
19. Rapid bacterial identification by direct PCR amplification of 16S rRNA genes using the MinIONTM nanopore sequencer. FEBS Open Bio. 2019; 9(3):548-57. DOI
20. Kaushik N.. A possible role of pulegone against glypican-1 for the treatment of Alzheimer’s disease through in-silico approach. Int J Res Appl Sci Eng Technol. 2021; 9(6):4000-07. DOI
21. Kumar S., Stecher G., Li M., Knyaz C., Tamura K.. MEGA X: Molecular evolutionary genetics analysis across computing platforms. Mol Biol Evol. 2018; 35(6):1547-49. DOI
22. Laskowski R.A., MacArthur M.W., Moss D.S., Thornton J.M.. PROCHECK: A program to check the stereochemical quality of protein structures. J Appl Crystallogr. 1993; 26(2):283-91. DOI
23. Machacek J.. Typology of environmental impacts of artisanal and small-scale mining in African Great Lakes region. Sustainability. 2019; 11(11):1-24. DOI
24. Martin B.P., Brandon C.J., Stewart J.J., Braun-Sand S.B.. Accuracy issues involved in modeling in vivo protein structures using PM7. Proteins. 2015; 83(8):1427-35. DOI
25. Megharaj M., Avudainayagam S., Naidu R.. Toxicity of hexavalent chromium and its reduction by bacteria isolated from soil contaminated with tannery waste. Curr Microbiol. 2003; 47(1):51-4. DOI
26. Oze C., Fendorf S., Bird D.K., Coleman R.G.. Chromium geochemistry of serpentine soils. Int Geol Rev. 2004; 46(2):1027470020-681446297.
27. Pall S., Abraham M.J., Kutzner C., Hess B., Lindahl E.. Tackling exascale software challenges in molecular dynamics simulations with GROMACS. Proc EASC. 2015; 8759:3-27. DOI
28. Paul M., Hazra M., Barman A., Hazra S.. Comparative molecular dynamics simulation studies for determining factors contributing to the thermostability of chemotaxis protein CheY. J Biomol Struct Dyn. 2014; 32:928-49. DOI
29. Paul M., Payel P.P., Hrudayanath T.. In silico studies on structural, functional, and evolutionary analysis of bacterial chromate reductase family responsible for high chromate bioremediation efficiency. SN Appl Sci. 2020; 2(1997)DOI
30. Pluthero F.G.. Rapid purification of high-activity Taq DNA polymerase. Nucleic Acids Res. 1993; 21(20):4850-51. DOI
31. Putra A.A., Huwaida N.N., Takarini R.D., Nur’ain R.. In silico study of secondary metabolites of Borneo endemic plant Shorea brunnescens as an anti-cancer drug candidate. Genbinesia Journal of Biology (GJB. 2023; 2(2):48-59. DOI
32. Rahman A.T., Jethro A., Santoso P., Kharisma V.D., Murtadlo A.A.. In silico study of the potential of endemic Sumatra wild turmeric rhizomes (Curcuma sumatrana: Zingiberaceae) as anti-cancer. Phcog J. 2022; 14(6):806-12. DOI
33. Repoles B.M., Gorospe C.M., Tran P., Nilsson A.K., Wanrooij P.H.. Biased nicking of mitochondrial DNA during extraction can be avoided by choice of isolation method. The Preprint Server for Biology. 2021; 1(1)DOI
34. Saidy A.R., Badruzsaufari. Hubungan antara konsentrasi Cr(VI) dan sifat kimia tanah: Informasi awal untuk remediasi lahan bekas tambang di Kalimantan Selatan [The relationship between Cr(VI) concentration and soil chemical properties: Initial information for remediation of ex-mining land in South Kalimantan. J Trop Soils. 2009; 14(2):97-103. DOI
35. Shalini V., Gadag G., Kalburgi P.V.. Mining and its impacts on environment and health with special reference to Ballari District, Karnataka, India. Int J Adv Eng Res Sci. 2023; 10(3):109-15. DOI
36. Sundarraj S., Sudarmani D.N.P., Samuel P., Sevarkodiyone S.P.. Bioremediation of hexavalent chromium by transformation of Escherichia coli DH5α with chromate reductase (ChrR) genes of Pseudomonas putida isolated from tannery effluent. J Appl Microbiol. 2022; 134(1)DOI
37. Talib R.N.S., Halmi M.I., Abd Ghani S.S., Zaidan U.H., Shukor M.Y.. Artificial neural networks (ANNs) and response surface methodology (RSM) approach for modelling the optimization of chromium (VI) reduction by newly isolated Acinetobacter radioresistens strain NS-MIE from agricultural soil. BioMed Res Int. 2019; 2019(5785387)DOI
38. Thatoi H., Das S., Mishra J., Rath B.P., Das N.. Bacterial chromate reductase, a potential enzyme for bioremediation of hexavalent chromium: A review. J Environ Manag. 2014; 146:383-99. DOI
39. Walwa W.J.. Large-scale mining and the right to a clean, health and safe environment in Tanzania. Afr Rev. 2016; 43(2):97-123.
40. SWISS-MODEL: Homology modelling of protein structures and complexes. Nucleic Acids Res. 2018; 46(W1)DOI
41. Wijiastuti W., Artika I.M., Nurhidayat N.. Isolation and selection of thermophilic bacteria as hexavalent chromium reducer from batik processing waste water. Curr Biochem. 2015; 2(1):10292442122-31.
42. Ying Y., Rajkumar M., Moreno A., Zhang C., Freitas H.. Serpentine endophytic bacterium Pseudomonas azotoformans ASS1 accelerates phytoremediation of soil metals under drought stress. Chemosphere. 2017; 185:75-85. DOI
43. Yu L., Zhang Y., Gutman I., Shi Y., Dehmer M.. Protein sequence comparison based on physicochemical properties and the position-feature energy matrix. Sci Rep. 2017; 7(46237)DOI