https://orcid.org/0000-0002-4919-1448
Figures
Abstract
Cytochrome P450 enzymes (P450s), particularly those of microbial origin, are highly versatile biocatalysts capable of catalyzing a broad range of regio- and stere-oselective reactions. P450s derived from extremophiles are of particular interest due to their potential tolerance to high temperature, salinity, and acidity. This study aimed to identify and classify novel microbial P450 enzymes from extreme environments across Türkiye, including hydrothermal springs, hypersaline lakes, and an acid-mine drainage site. The focus of this study was on classifying the sequence diversity of P450 enzymes in these sites. To that end, shotgun metagenomic analysis of six sites, using de novo binning, phylogenetic analysis, and functional gene annotation, was used to discover 311 putative P450 sequences, assigned to 87 families and 158 subfamilies, including 8 novel families and 49 new subfamilies. Of these, 237 were in 138 metagenomic bins, including 45 high-quality metagenome-assembled genomes. The distribution of P450 families varied across sites, reflecting distinct environmental conditions and microbial community compositions. These findings highlight the untapped potential of Türkiye’s extreme habitats as a source of novel biocatalysts. Beyond their industrial relevance, extremophile-derived P450s may also play key roles in enabling microbial adaptation to harsh environmental conditions, through their involvement in stress-responsive metabolic pathways and structurally resilient enzyme forms. This work provides a foundation for future studies into both their biotechnological applications and ecological functions.
Citation: Mumcu H, Zaugg J, Keles I, Kayrav A, Balci N, Nelson DR, et al. (2025) Exploring extreme environments in Türkiye for novel P450s through metagenomic analysis. PLoS One 20(9): e0330523. https://doi.org/10.1371/journal.pone.0330523
Editor: Preenan Pillay, University of South Africa—Florida Campus: University of South Africa—Science Campus, SOUTH AFRICA
Received: August 12, 2024; Accepted: August 1, 2025; Published: September 8, 2025
Copyright: © 2025 Mumcu et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The shotgun sequencing read data for the samples described in this study, as well as the 171 higher quality MAGs obtained from them, have been deposited in the NCBI Sequence Read Archive (SRA) (Accessions: SRR27869035–SRR27869040) and Genbank, respectively, under the bioproject accession PRJNA979897 (https://www.ncbi.nlm.nih.gov/bioproject/PRJNA979897/).
Funding: 1- NGK, Grant No. 1059B192100859, The Scientific and Technological Research Council of Turkey (TUBITAK), https://tubitak.gov.tr/en 2- NGK, Grant Nos. 42997 and 42953, ITU Scientific Research Projects Division, https://bap.itu.edu.tr/ The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Metagenomics is a culture-independent approach for studying microbial communities by extracting and sequencing genetic material directly from environmental samples (eDNA). Unlike traditional microbiology, which relies on cultivating microbes in the laboratory, metagenomics provides a much less biased view of microbial communities, including previously unculturable species [1]. This approach offers unprecedented insights into microbial diversity, metabolic functions, and ecological interactions, enabling researchers to study microorganisms in their natural habitats without the need for isolation [2].
Among metagenomic techniques, shotgun metagenomics has emerged as a powerful tool for exploring the functional capacity of microbial communities. By randomly sequencing genetic material within a sample, this approach enables the identification of novel genes, biosynthetic gene clusters, and entire metabolic pathways [3]. Through the use of the numerous computational tools that have been developed to process such data, it is now possible to reconstruct the genomes of novel microorganisms and functionally annotate their genes, providing researchers with insight into their ecological roles [4]. This approach is instrumental in identifying new enzymes and biomolecules with potential biotechnological applications, including cytochrome P450 enzymes, which play a crucial role in oxidative metabolism across various biological systems [5].
Cytochrome P450 heme-thiolate proteins (EC 1.14.14.1) are a superfamily of enzymes usually acting as monooxygenases. The majority of these enzymes catalyze the insertion of one oxygen atom from molecular oxygen into the substrate, with reduction of the other atom to water, a process facilitated by the presence of one or more redox partners that catalyze electron transfer from the reducing cofactor, NADPH. P450s bind molecular oxygen through their heme prosthetic group that is coordinated to the apoprotein through a conserved axial cysteine residue [6]. Although P450s catalyze different types of reactions, they have a common catalytic cycle consisting of nine steps [6] that involves the transfer of two electrons. The electrons are usually transferred to the heme center through redox protein partners such as ferredoxins/ferredoxin reductases or diflavin reductases in a multi-component electron transfer chain. However, some P450s are present as genetic fusions with one or more redox partners and are therefore considered self-sufficient [7].
To date, many bacterial and archaeal cytochromes P450 have been identified and classified [8–12]. Characterized P450s play roles in many catabolic and anabolic pathways such as fatty acid, steroid, and xenobiotic degradation, and the biosynthesis of primary and secondary metabolites [13,14]. Within those pathways, they act on diverse simple and complex molecules such as fatty acids, alkanes, terpenes, eicosanoids, vitamins, steroids, antibiotics, and a variety of drugs and other xenobiotics [15]. In addition to their wide substrate and reaction diversity, the most important feature of microbial P450s is that they can be regio- and stereo-specific [16]. Consequently, they are useful in synthesizing new drugs, fine and bulk chemicals, and agrochemicals in the pharmaceutical, flavour/fragrance, and agricultural sectors, as well as for pollutant removal [17]. The extensive intrinsic sequence diversity in microbial P450s and their potential to be used in many industrial processes make them attractive biocatalysts, and the identification of novel P450s is an area of intense interest [18].
Extreme environments, including hydrothermal vents, polar deserts, hypersaline lakes, acidic mines, and deep-sea sediments, host diverse microbial communities, collectively known as extremophiles. These microbes have evolved unique adaptive strategies to survive the harsh conditions characteristic of such environments, e.g., high temperatures, salinity, pH, and concentrations of heavy metals. Extremozymes—enzymes found in extremophiles—enable survival under these conditions and exhibit remarkable stability and activity, making them highly valuable for biotechnological applications [19]. P450 extremozymes, in particular, have garnered significant attention due to their diverse catalytic capabilities, but relatively few have been identified to date. Extremophilic P450s characterized to date include members of the self-sufficient CYP116 family [20], as well as the CYP119, CYP154, CYP174, CYP175, and CYP231 families [21]. Jiang et al. identified three moderately halophilic P450 fatty acid decarboxylases—CYP152L1_ortholog, CYP152L7, and CYP152L8—belonging to the CYP152 family [22]. Moreover, Nguyen and colleagues identified 36 potentially thermostable P450s from water samples collected at Binh Chau hot spring in Vung Tau, Vietnam, through metagenome shotgun sequencing [23]. They also discovered a novel moderately alkali-thermophilic P450 from the CYP203 subfamily, which exhibits optimal activity at 50 °C and pH 8.0 [24].
The climatic conditions at various locations across the Anatolian geography allow different species of living organisms to occupy unique habitats and ecological niches. Türkiye, one of the richest countries in Europe in terms of biodiversity, is home to many endemic species not commonly found elsewhere. The aim of the present study was to characterize the prokaryotic community and P450 diversity of six previously uncharacterized sites in Türkiye with extreme environmental conditions through de novo binning, phylogenetic analysis, and functional gene annotation of metagenomic data. This study identified and classified a total of 311 microbial cytochromes P450 across 87 families and 158 subfamilies, including 8 new families and 49 new subfamilies. The findings underscore the value of investigating extreme environments as a rich source of novel and functionally diverse enzymes.
Materials and methods
Sampling
Samples were collected from six sites in Türkiye characterized by extreme environmental conditions, with three samples collected from each site (Fig 1) (USGS National Map Viewer): Lake Acıgöl (37.8299 N, 29.8931 E; April 2024 spring) [25], Gömeç (Balıkesir; 39.386373 N, 26.835452 E; July 2019 summer), Hisaralan (Balıkesir; 39.287251 N, 28.341724 E; December 2021 winter), Armutlu (Yalova; 40.520437 N, 28.815628 E; July 2017 summer) [26], Balya (Balıkesir) acid mine drainage (39.749294 N, 27.578101 E; August 2010 summer) [27] and Tuz Gölü (38.818571 N, 33.347851 E; March 2022 spring). Lake Acıgöl, Tuz Gölü, and Gömeç are hyper-saline environments [25,28]. Located in hydrothermal regions, Hisaralan and Armutlu have average water temperatures of 98 °C and 74 °C [26], respectively. Balya acid mine drainage has a pH lower than four and contains high concentrations of sulfur and heavy metals such as Pb, Zn, and Cu [27].
Sediment samples were collected from Lake Acıgöl (the upper 10 cm of the lake bed sediments), Gömeç (the upper 10 cm of the lake bed sediments), Armutlu (at a depth of 10–20 cm of the pool) and Balya (at a depth of 10 cm of the acidic pools); a two-liter water sample was collected from Hisaralan (at a depth of 10–20 cm of the pool); and an approximately 110 g sample of salt crystals was collected from Tuz Gölü. The salt crystals precipitated from the water columns (< 20 cm) were collected from the lakebed. All collections were done in accordance with permits obtained from the Republic of Türkiye Ministry of Environment Urbanization and Climate Change explicitly for the field studies described here.
Environmental DNA extraction and shotgun metagenomic sequencing
Environmental DNA (e-DNA) was isolated from 0.5–1 g of each sediment sample using the Qiagen DNeasy PowerSoil Pro Kit. The hot spring and saltwater samples were dissolved slowly in 2 L phosphate buffer saline (PBS), filtered through a 0.22 µm sterile syringe filter with the help of vacuum, and then the e-DNA was isolated using a Qiagen DNeasy PowerWater Kit. DNA purity and quality were assessed using Qubit 2.0 DNA HS Assay (Life Technologies). Shotgun sequencing libraries were prepared using KAPA HyperPrep Kit (Roche) and library concentration and quality control were evaluated using Qubit 2.0 DNA HS Assay (Life Technologies) and Tapestation High Sensitivity D1000 Assay (Agilent Technologies). The 150 bp paired-end sequencing of prepared libraries was performed on an Illumina NextSeq 550 system. An overview of the experimental and computational (see below) methods used to process the samples is provided in Fig 2.
Metagenomic assembly and de-novo binning
Low quality reads were identified and removed with Trimmomatic (ver. 0.39, ILLUMINACLIP: NexteraPE-PE:2:30:10, SLIDINGWINDOW:4:15, MINLEN:50) [29]. Quality controlled reads were then assembled using metaSPAdes (ver. 3.15.4) [30] with default parameters. Quality controlled reads for each sample were mapped onto their respective scaffolds with minimap2 (ver. 2.17) [31] using the ‘make’ mode in the DNA read coverage calculator CoverM (ver. 0.6.1) [32]. Low quality read mappings were removed with the CoverM ‘filter’ mode (minimum identity 95% and minimum aligned length of 75%), and the number of remaining reads was used to calculate the fraction of the DNA mapping to the assembled scaffolds.
The assembly for each sample was binned using the metagenomic binning pipeline Aviary (ver. 0.5.6) [33]. Briefly, Aviary first maps reads from all samples to each individual assembly with minimap2 (ver. 2.17) as part of CoverM (ver. 0.6.1) to obtain differential coverage information for each assembly. Using this coverage information, metagenome contigs were then binned using the Maxbin (ver. 2.2.7) [34], MetaBAT (ver. 0.32.5) [35], MetaBAT2 (ver. 2.15) [36], CONCOCT (ver. 1.1.0) [37], Vamb (ver. 3.0.2) [38], Semibin (ver. 1.1.1) [39] and Rosella (ver. 0.4.2) [40] binning methods with a minimum contig length of 1,500 bp and minimum bin size of 200,000 bp. For each sample, an optimal, non-redundant set of bins produced from the various binning tools were selected by DAS Tool (ver. 1.1.2) [41]. The completeness and contamination of all 1,138 non-redundant bins were calculated by CheckM (ver. 1.1.3) [42]. Taxonomy was assigned to each bin using the Genome Taxonomy Database Toolkit (GTDB-Tk; ver. 2.3.0; with reference to GTDB R08-RS214) [43,44]. The non-redundant bins from across all samples were then clustered and dereplicated using CoverM ‘cluster’ (precluster-method = dashing) with an ANI threshold of 97% and accounting for bin quality (checkm-tab-table). Dereplication yielded 1,135 bins, 171 of which were higher quality with a quality value ≥ 50 (calculated as the completeness – (3 × contamination).
Metagenome community profiling
The relative abundance of the dereplicated bins was calculated by first mapping the reads from each sample to each using CoverM ‘make’ and removing low quality mappings with CoverM ‘filter’ (minimum identity 95% and minimum aligned percent of 75%). The mean coverage of each bin was then calculated with CoverM and the relative abundance of each, among those obtained, was calculated as its coverage divided by the total summed coverage of all bins (S1 Table).
To obtain a broader assessment of the community composition of each sample, the microbial community profiler SingleM (ver. 0.16.0) was used [45]. Taxonomic profiling tools typically rely on databases derived from reference genomes [46–50], limiting abundance calculations to known species while missing novel taxa [45]. In contrast, SingleM can identify lineages where no genome exists. Briefly, it achieves this by a) analyzing only those reads which cover highly conserved regions of single copy marker genes, b) clustering these reads de novo into operational taxonomic units (OTUs), independent of existing taxonomies, c) taxonomically classifying OTUs against the Genome Taxonomy Database (GTDB) [51,52], d) per marker gene, estimating the relative abundance of each taxon based on OTU classifications, and e) calculating a trimmed mean abundance taken across all the marker genes [45]. The bacterial and archaeal community composition of each sample was therefore determined by classifying those raw reads corresponding to 59 single-copy genes using the ‘pipe’ tool from SingleM, based on taxonomies derived from the GTDB R08-RS214. SingleM ‘condense’ was used to produce a single OTU table containing the trimmed mean coverage across each lineage, calculated across all genes. The relative abundance of each lineage was then calculated as its respective coverage divided by the total summed coverage for each sample. Shannon diversity was calculated for each sample from genus level mean coverage values from SingleM using phyloseq (ver. 1.50.0) [53]. Finally, Nonpareil was run on the quality-controlled reads using the k-mer alignment method to assess the fraction of the microbial community sampled by sequencing [54,55]. Community abundance stacked bar charts were created using the R package ggplot (ver. 3.4.4) [56], and heatmaps with Complex Heatmap (ver. 2.16.0) [57].
Gene extraction, and identification and classification of P450s
Protein-coding sequences (CDS) in the assembled scaffolds and bins were first predicted using Pyrodigal (ver. 2.0.2) [58], a Python library binding to Prodigal [59], in metagenomic mode. Sequences with start and stop codons, i.e., theoretically complete open reading frames, were extracted using mfqe [60] (ver. 0.5.0). Complete protein sequences (1,966,993) were clustered at 100% protein identity using CD-HIT (ver. 4.8.1) [61], with all members of each cluster required to have at least 80% of their sequence overlapping with the longest (seed) sequence. Protein sequences containing the cytochrome P450 domain (PF00067) were identified using HMMER hmmscan (ver. 3.3.2; -E 1e-5) [62] and by aligning the protein sequences against the CYPED database [63] with DIAMOND blastp (--evalue 0.00001, --query-cover 50, --subject-cover 50, --id 15) [64]. Of the 4,064 putative P450 sequences identified (2,730 BLAST, 1,334 HMMER), 311 were identified as complete P450s after manual inspection. The selected sequences were aligned using MAFFT (--localpair, ver. 7.455) [65], and the resulting alignment trimmed using trimAl (-automated1, ver. 1.4.1) [66]. A phylogenetic tree was then constructed using IQ-Tree (model LG + R7, ver. 2.1.2) [67] with 1,000 bootstraps and visualized using tvBOT [68]. Approximately 117 of the identified P450 sequences were either not found in a genome bin or were found in a bin with a poorly resolved taxonomic classification, i.e., the bin could not be taxonomically classified below the class level. For these sequences, similar sequences were searched for among the representative genomes from the GTDB (R08-R214) using MMseqs (ver. 13.45111; --min-seq-id 0.7 -c 0.7) [69], and the best hit was used to annotate the corresponding host-lineage in the phylogenetic tree.
Proteins within the P450 superfamily are classified in accordance with the guidelines set by the International P450 Nomenclature Committee [6,70]. Specifically, proteins sharing more than 40% sequence similarity were placed within the same family, while those with over 55% sequence similarity were categorized within the same subfamily [71]. Any proteins having less than 40% sequence similarity to known P450s were assigned to a novel P450 family.
Results
Taxonomic profiling of the extreme sites
Shotgun sequencing produced 18–24 Gbp of read data for each sample, except for the Armutlu hot spring, where 2.4 Gbp was obtained. Estimated coverage of the microbial communities ranged from 35–96% (66–96% excluding Armutlu; S1 Fig and S2 Table), suggesting that a substantial portion of the community was sampled. Dominant phyla (>10% relative abundance in at least one sample) included archaeal lineages from Halobacteriota (Tuz Gölü, Gömeç), Nanohaloarchaeota (Tuz Gölü) and Thermoproteota (Armutlu), and bacterial lineages Actinomycetota (Hisaralan), Aquificota (Hisaralan), Bacillota (Hisaralan), Bacteroidota (Lake Acıgöl), Bipolaricaulota (Hisaralan), Chloroflexota (Armutlu) and Pseudomonadota (Lake Acıgöl, Armutlu, Balya, Gömeç) (Fig 3, S1–S3 Tables). Notably, Halobacteriota are extremely halophilic archaea [72], Nanohaloarchaeota are exclusively derived from hypersaline habitats [73], and Thermoproteota are methanogenic and hyperthermophilic archaea (Fig 3). Of the bacteria, members of the Bipolaricaulota (15.7%) are known to fix carbon and dominate in some geothermal regions [74]; the family Thiomicrospiraceae (15.1% in Balya), belonging to the phylum Pseudomonadota, has an important role in sulfur oxidation pathways [75]; and Bacteroidota (26.4% in Acigol) is essential for the nitrogen cycle in hypersaline environments and significantly contributes to the elimination of greenhouse gases [76]. The taxonomic composition of these extreme environments reveals a diverse range of archaeal and bacterial lineages, with site-specific differences that may be shaped by distinct selective pressures.
Only the top five families/genera per sample are shown, with all other taxa grouped under ‘Other’.
Metagenome assembled genomes (MAGs)
A total of 1,138 metagenomic bins were obtained, 171 of which were deemed high quality (>50 combined completeness/contamination metric). These 171 MAGs were estimated to represent between 0.6–65% of the microbial communities from which they were derived (S4 Table). They belong to four archaeal phyla: (Halobacteriota, n = 21; Nanoarchaeota, n = 1; Nanohaloarchaeota, n = 7; and Thermoproteota, n = 5), and 28 bacterial phyla (Acidobacteriota, n = 4; Actinomycetota, n = 4; Aquificota, n = 2; Armatimonadota, n = 2; Bacillota, n = 10; Bacillota_A, n = 3; Bacillota_C, n = 1; Bacillota_F, n = 3; Bacteroidota, n = 32; Bipolaricaulota, n = 1; Campylobacterota, n = 1; Chloroflexota, n = 9; CSP1–3, n = 1; Cyanobacteriota, n = 3; Deinococcota, n = 3; Desulfobacterota, n = 6; Desulfobacterota_D, n = 1; Desulfobacterota_F, n = 1; DRYD01, n = 2; Fibrobacterota, n = 1; Gemmatimonadota, n = 1; Marinisomatota, n = 1; Nitrospirota, n = 1; Patescibacteria, n = 5; Planctomycetota, n = 3; Pseudomonadota, n = 32; Spirochaetota, n = 2; Thermotogota, n = 1; and, notably, one novel phylum) (Fig 4, S5 and S6 Tables). At lower taxonomic levels, many of the MAGs appear to represent novel lineages: 118 were unclassified at the species level, 22 at the genus, 4 at the family and 2 at or above the order level. A summary of all 1,138 bins is provided in the supplementary material (S5 Table). The recovered MAGs expand our understanding of the genomic diversity in these extreme environments, revealing several novel lineages that warrant further characterization.
Abundance values have been scaled by the fraction of the DNA that mapped to the MAGs. The number of MAGs per lineage is provided in the row labels by ‘n = #’.
Identification of P450s in metagenomes
Across the six samples, 544,659 proteins were predicted from Armutlu, 3,097,365 from Balya, 5,495,569 from Gömeç, 1,042,240 from Hisaralan, 5,046,949 from Acıgöl and 1,881,283 from Tuz Gölü (Fig 5; total of 1,958,703 proteins after clustering at 100% identity) representing a substantial reservoir of potentially useful extremophilic biomolecules. An initial screening of the full protein dataset, conducted using a combination of HMM profile searches and alignment with reference sequences from the CYPED database, identified hundreds of putative cytochrome P450 enzymes. The distribution across the six sites was as follows: 55 from Armutlu, 614 from Balya, 801 from Gömeç, 434 from Hisaralan, 579 from Acıgöl, and 541 from Tuz Gölü. Before classifying these putative P450s, amino acid sequences were filtered to those with complete sequences (including both start and stop codons) and were searched against the CYPED database to eliminate non-microbial sequences. Those with at least a 20% match to microbial P450s were then examined for the integrity of their heme-binding domains using the NCBI CDD (Conserved Domain Database), and those that did not contain the consensus heme binding motif F(x)nG/A(x)mCxG were removed (where: x is any amino acid; n is typically 2 but up to 5 in some families, e.g., CYP152; and m is typically 3 but up to 6 in some families [77]). After filtering, a total of 311 sequences remained: 52 thermophilic (Armutlu n = 3, Hisaralan n = 49), 92 acidophilic (Balya), and 167 halophilic (Gömeç n = 57, Lake Acıgöl n = 31, and Tuz Gölü = 79) (Fig 5). Among these sequences, 241 were found across 138 of the bins (104 bins with a taxonomic classification at the class level or below), including 45 of the higher-quality MAGs (S7 Table).
We did not observe a clear correlation between microbial diversity and either the number of P450s or the number of P450 families present in the samples (S2 Fig). P450s from Balya, which had a relatively high microbial diversity (Shannon diversity of ~5.5), were only encoded by members of the phylum Pseudomonadota. Notably, ten of the higher-quality Balya MAGs encoded multiple P450s (from different P450 families; S7 Table, S3 Fig). This included five members of the genus Novosphingobium that each encoded 4–9 P450s, and a Blastomonas fulva that encoded 7. At the hypersaline sites, Tuz Gölü and Gömeç (diversities of 3.7 and 6.5, respectively), members of the phylum Halobacteriota, specifically the families Haloarculaceae, Halobacteriaceae, Haloferacaceae, and Salinarchaeaceae, were the primary encoders with 1–5 P450s each. At the other hypersaline site, Lake Acıgöl (diversity of 5.8), P450s were primarily encoded by members of the phyla Bacteroidota, Halobacteriota, and Pseudomonadota (1–2 P450s). Among the hydrothermal sites Hisaralan and Armutlu (diversities of 4.0 and 5.4, respectively) bins were only obtained from Hisaralan, and the top encoders with 2–4 P450s included members of the phyla Actinomycetota, Bacillota, Chloroflexota, and Desulfobacterota_B.
Classification of P450s
The 311 P450s were named according to P450 nomenclature criteria [71], with those having less than 40% amino acid identity designated as a new family, and those with more than 40% but less than 55% identity assigned to a new subfamily (Fig 6). The site with the highest number of identified P450s was Balya acid mine drainage (n = 92), followed by Tuz Gölü (79), Gömeç (57), Hisaralan (49), and Lake Acıgöl (31). Only three P450s were identified in the Armutlu hot water sample, possibly due to low DNA read depth obtained from sequencing, however all three belonged to different families, one of which represented a novel subfamily. Aside from Armutlu, samples with the highest P450 family diversity were Balya (n = 37), Gömeç (27), Hisaralan (23), and Lake Acıgöl (16). While Tuz Gölü harbored the second highest number of P450s of the six samples, it had the lowest diversity (13) (Table 1).
The inner ring indicates the corresponding P450 subfamily (text) and family (highlight colour). The taxonomic classification of the host bin at the phylum and class level is shown in the second and third rings. For those proteins that were not found in one of the metagenomic bins or were in a bin with a poorly resolved taxonomic classification (not below the class level; see “Bin with class”), the closest P450 match in the GTDB reference genomes was used as a proxy where possible.
The family, subfamily, and potential functional characteristics of the identified P450s for each site are presented in S8 Table. In total, 311 putative microbial cytochrome P450 enzymes were identified and classified into 87 families and 158 subfamilies including 8 new families from all sites except Acıgöl and 49 new subfamilies except Armutlu (Table 2). Notably, Gömeç and Hisaralan exhibited the highest number of newly identified families and subfamilies. Three self-sufficient P450s (CYP116B304 from Hisaralan, CYP116B171 and CYP116B21 from Balya) [78,79] and seven P450s from the CYP152 family typically associated with peroxygenase activity [80] were identified across Hisaralan, Gömeç, Acıgöl, and Balya. Notably, 54% of the P450 families identified in this study have not been previously identified.
The CYP107 family was present in all sample sites except Armutlu, while there were the following numbers of site-specific P450 families: 6 for Tuz Gölü, 7 for Acıgöl, 14 for Gömeç, 12 for Hisaralan, 28 for Balya, and 1 for Armutlu. Dominant P450 families across the samples were CYP174 in Tuz Gölü (n = 28) and Gömeç (n = 11), CYP1103 in Acıgöl (n = 6), CYP107 and CYP197 in Hisaralan (n = 7), and CYP108 in Balya (n = 12) (Fig 7).
The sequences of CYP107, the common family in all extreme sites, were found in nine of the bins with taxonomic classifications at the class level or below (three higher-quality MAGs), from the following phyla: Actinomycetota (genera JAHWLC01 and Blastococcus and order Nitriliruptorales); CSP1–3 (genus HRBIN32); Deinococcota (genus JAABTL01); Bacillota (genera YIM-78166 and Ectobacillus); and Chloroflexota (genus Roseiflexus). A total of nine, four of which are novel, were identified from hypersaline environments (CYP107PH1, CYP107PH2, CYP107PH3, CYP107PJ1, CYP107PJ2, CYP107PK1, CYP107PL1, CYP107PM1, and CYP107PM2), seven from the hydrothermal environments (CYP107AQ30, CYP107AQ31, CYP107AQ32, CYP107AZ2, CYP107H11, CYP107JF5, and CYP107PN1), and one from the acidic environment (CYP107DG12).
The CYP174 and CYP109 families were commonly found in the hypersaline habitats: CYP174B72 and CYP109G34 from Lake Acıgöl; CYP174A, CYP174B, CYP174C, CYP174E, CYP109BL1, CYP109G35 and CYP109G36 from Tuz Gölü; and CYP174A, CYP174B, CYP174E and CYP174H, CYP109G37 from Gömeç. All identified hosts of the CYP174 sequences belonged to the phylum Halobacteriota (28 bins, 10 higher quality MAGs), across the genera Haloarchaeobius, Haloarcula, Halobacterium, Halobaculum, Halolamina, Halomicrobium, Halonotius, Haloplanus, Halorientalis, Halorubrum, Halosimplex, Natronomonas, QS-5-70-15 (family Haloarculaceae), Salinibaculum and Salinigranum (S7 Table). The previously identified CYP174s, the CYP174A and CYP174B subfamilies have been ascribed to archaea [12], as found here. The CYP109 sequences from hypersaline environments were found in archaeal bins from the phylum Halobacteriota, including the genera Halorientalis and Salinarchaeum (S7 Table).
In the hydrothermal site, Hisaralan, CYP197 (n = 7) and CYP125 (n = 5) were also common (Fig 7). Novel CYP197 subfamilies identified in Hisaralan included: CYP197AZ, CYP197BA, CYP197BB, CYP197BC, CYP197BD and CYP197Y. Across all samples, CYP197 sequences were present in eight bins with taxonomic classifications at the class level or below (including seven higher quality MAGs) from archaeal phyla (Halobacteriota: Halolamina and the Salinigranum genera, and the Halobacteriaceae family) and bacterial phyla (Bacillota, RAOX-1 (family); and Chloroflexota, CP2-2F and the JANWYT01 genus). CYP125 sequences were found across five taxonomically classified bins (two higher quality MAGs) from the phyla Actinomycetota (Blastococcus genus), Chloroflexota (HRBIN24 genus), Desulfobacterota_B (HRBIN30 genus), and Pseudomonadota (Rhizorhabdus, genus).
Balya acid mine drainage was one of the sites with the highest P450 diversity, with families CYP108 (subfamilies CYP108A, CYP108D, CYP108G and CYP108H) and CYP153 (subfamilies CYP153A and CYP153D) the most abundant in this area. Among the eight classified bins (six high quality MAGs) encoding CYP108 sequences, all were members of the phylum Pseudomonadota (genera Erythrobacter, Sphingobium, Blastomonas, Novosphingobium, and Hydrogenophaga). On the other hand, CYP153 sequences were found across seven bins (six higher quality MAGs) belonging to the phyla Actinomycetota (Blastococcus genus) and Pseudomonadota (Blastomonas and Novosphingobium genera).
Discussion
Extreme environments can support diverse, extremophilic microbial communities that have developed unique adaptive strategies to survive, and that can encode enzymes with novel structural and functional properties. Among such enzymes are cytochromes P450, a highly diverse superfamily of enzymes capable of catalyzing a broad range of reactions, including aromatic and aliphatic hydroxylation, heteroatom oxidation, epoxidation, and dealkylation at N-, O- and S-centers [81]. The ability of P450s to transform structurally diverse compounds such as fatty acids, steroids, terpenes, and aromatic hydrocarbons makes them key players in microbial metabolism. The products of these reactions are of value to industry, including in the pharmaceutical, bioremediation, and fine chemical sectors [82].
Although no clear correlation was observed between the microbial diversity (Shannon index) and the number of P450s (and P450 families) across the samples, microbial composition did have a strong influence on both the number and diversity of P450 enzymes. Specifically, members of Pseudomonadota and Actinomycetota contributed a wide variety of P450 families, including CYP108, CYP109, and CYP153, and often encoded multiple P450s in their respective genomes. In the hypersaline environments, Halobacteriota species often encoded CYP174s, while Bacillota and Chloroflexota were linked to CYP125 and CYP197 diversity in hydrothermal habitats. These results suggest that specific microbial groups may shape the diversity of P450s in extreme environments more than overall microbial diversity.
CYP107 sequences were found in the samples from all extreme conditions. Among the most extensively studied CYP107s are those involved in antibiotic biosynthesis. CYP107A1 (P450eryF) from Saccharopolyspora erythraea contributes to erythromycin biosynthesis [83]. CYP107L1 from Streptomyces venezuelae is integral in the production of pikromycin, neomethymycin, novamethymycin, neopikromycin, and novapikromycin [84]. Micromonospora griseorubida CYP107E1 (MycG) is associated with mycinamicin biosynthesis [85], Streptomyces himastatinicus CYP107B (HmtT) with himastatin [86,87], and Streptomyces thermotolerans CYP107C1 (orfA) with carbomycin [88]. Streptomyces avermitilis CYP107W1 [89] and Streptomyces sp. 307-9 CYP107FH5 (CYP TamI) [90] are involved in oligomycin and triandamycin biosynthesis, respectively. Additionally, other CYP107 forms are involved in the biosynthesis of other natural products of use in medicine: CYP107Z14 from Sebekia benihana contributes to the synthesis of the immunosuppressant cyclosporin A [91]; Streptomyces hygroscopicus CYP107G1 plays a role in the biosynthesis of the antifungal and antitumor agent rapamycin [92]; and Streptomyces sp. SN-593 CYP107E6 is associated with the biosynthesis of reveromycin T [93], used in osteoporosis treatment. CYP107H1 (P450Biol) from Bacillus subtilis plays a pivotal role in the synthesis of pimelic acid [94], a component involved in biotin synthesis, while CYP107BR1 (P450vdh) from Pseudonocardia autotrophica is engaged in vitamin D biosynthesis [95] and Streptomyces avermitilis CYP107X1 operates in the progesterone biosynthetic pathway [96]. CYP107 forms are also potentially useful for the detergent industry due to roles in glycocholic acid biosynthesis, as exemplified by Streptomyces coelicolor CYP107U1 [97]. While information on reactions, substrates, and products is available for the CYP107H subfamily (CYP107H1), the reactions catalyzed by the rest of the CYP107s identified in this study and their biotechnological significance are unknown. However, the predominance of CYP107 forms in the biosynthesis of complex secondary metabolites such as antibiotics suggests that the novel forms identified here may be useful in the search for new or modified antimicrobial agents or other natural products that may have useful properties.
The CYP174 and CYP109 families were widespread in hypersaline sites in this study. From a single previous study, CYP174 has been associated with terpene metabolism [98] but it is unclear whether this activity is common to other CYP174 family members. By contrast, there are many characterized CYP109s. A study of 128 Bacillus species identified the CYP109 family as the third most abundant P450 family [99], predicted to be involved in the synthesis of a wide range of secondary metabolites important to the physiology of Bacillus species. Among the characterized CYP109s, CYP109B1 from Bacillus subtilis strain 168 was found to be responsible for the hydroxylation of saturated fatty acids (C10-C18), methyl esters of saturated fatty acids (C12-C16), ethyl esters of saturated fatty acids (C12-C14) and unsaturated fatty acids (C14-C16). In addition to fatty acids, CYP109s can carry out the hydroxylation of primary n-alcohols (1-decanol, 1-dodecanol, and 1-tetradecanol) and the oxidation of the terpenes, α-ionone, β-ionone and (+)-valencene, which have an important place in the perfume, cosmetics, pharmaceutical, and other fine chemical industries [100,101]. Studies of CYP109E1 from Bacillus megaterium DSM319 have demonstrated that this enzyme can hydroxylate testosterone and vitamin D3 to synthesize industrially valuable products [102,103]. In addition, the CYP109E1 enzyme is capable of the hydroxylation of statins (compactin, lovastatin, and simvastatin) to synthesize drug metabolites and the hydroxylation of terpenes (α-ionone, β-ionone, nootkatone, isolongifolen-9-one, α-damascone, β-damascone, and β-damascenone) to synthesize valuable terpene derivatives with high regioselectivity [104]. Together with CYP109E1, CYP109A2—another CYP109 from B. megaterium DSM319—was found to hydroxylate vitamin D3 with high regioselectivity [105]. In addition to CYP109s from Bacillus species, studies with Sorangium cellulosum So ce56 showed that the organism has three CYP109s: CYP109C1, CYP109C2, and CYP109D1. CYP109D1 and CYP109C2 are responsible for the hydroxylation of lauric acid (C12), tridecanoic acid (C13), myristic acid (C14), and palmitic acid (C16), whereas CYP109D1 can also hydroxylate capric acid (C10) [106,107]. These studies suggest that the CYP109 family can catalyze many different reactions and substrates. Notably, the CYP109 and CYP174 members identified in this study are from subfamilies that have not been characterized in any detail biochemically. While the known substrate profiles of these families do not directly indicate roles in salt adaptation, their prevalence across the hypersaline microbiomes suggests they may possess structural features enabling function under high-salinity conditions. These observations highlight the need for future functional and structural studies to explore their potential halotolerance and biotechnological relevance.
The CYP125 and CYP197 families were also common in the hydrothermal site, Hisaralan. Previously characterized members of the CYP125 family are CYP125A6 and CYP125A7, which play a role in steroid hydroxylation pathways and in cholesterol catabolism in mycobacterial species [108]. These enzymes may also be linked to membrane lipid composition and ordering in thermophiles, which can be influenced by cholesterol across a wide temperature range [109,110]. Based on this information, it can be hypothesized that members of the CYP125 family, including the newly identified CYP125N, CYP125P, and CYP125AF subfamilies, may contribute to pathways that facilitate microbial adaptation to high temperatures in hydrothermal environments. Members of the CYP197 family have been found across various bacterial phyla and are frequently encoded within biosynthetic gene clusters associated with secondary metabolism [11,99]. While their specific enzymatic functions and underlying catalytic mechanisms remain uncharacterized, their presence in both hydrothermal sites suggests a role in the biosynthesis of heat-stable or stress-responsive metabolites. Functional characterization of these enzymes may uncover novel biocatalysts with potential applications in biotechnology and natural product discovery.
CYP108 was one of the two dominant P450 families at the acid mine drainage site, Balya. While there is limited research on CYP108, members of this family are known to catalyze the oxidation of α-terpineol [111]. For example, the CYP108D1 enzyme exhibits hydroxylase activity on aromatic hydrocarbons, including phenyl cyclohexane and p-cymene [112]. CYP153 was also common to the Balya site; this family has been associated with alkane degradation in diverse bacterial species, including members of the phyla Actinobacteria (now Actinomycetota) and Proteobacteria (now Pseudomonadota) [113–116]. To date, the best characterized members of the CYP153 family include CYP153A6 from Mycobacterium sp. HXN-1500, which hydroxylates medium-chain-length alkanes (C6 to C11) to 1-alkanols [117], and CYP153A13 from Alkanivorax borkumensis SK2. CYP153A13 has diverse catalytic capability, being able to hydroxylate not just the terminal end of short alkyl groups attached to aromatic rings but also the p-position of phenolic compounds substituted with a halogen or an acetyl group. Additionally, CYP153A13a demonstrated the ability to demethylate aromatic compounds containing methyl ether groups [118]. Organic compounds, including aromatic hydrocarbons and alkanes, are present in acid-mine drainage sites like Balya. Therefore, microorganisms from such sites, and the enzymes encoded in their genomes, may be useful for degrading these hydrocarbons [119]. Considering all the information known about the CYP108 and CYP153 families, undertaking further in-depth studies on the P450s from the Balya site to elucidate the hydrocarbon groups they degrade holds promise for advancing bioremediation initiatives.
Finally, rare exceptions to the F(x)nG(x)mCxG motif used for filtering sequences have been described previously [77], where one or more of the specified residues is conservatively substituted. However, the Cys residue is almost universally conserved and generally considered to be required for generating the highly reactive oxidizing species, compound I, involved in monooxygenase activity. Notably, among the sequences excluded based on the heme-binding motif was a CYP102A178 sequence that appeared to encode a plausible P450 sequence, with the exception that the conserved Cys was replaced by a Tyr residue. Further work is underway to characterise both the putative Tyr and Cys forms of this enzyme.
Conclusion
Metagenomics is a powerful tool for discovering novel biocatalysts from uncultured microorganisms. Through shotgun metagenomics and computational analyses, this study has identified 311 P450 sequences, including 8 novel families and 49 subfamilies, from diverse extreme environments across Türkiye. Of these sequences, 237 were associated with 138 metagenomic bins or metagenome assembled genomes (MAGs) of prokaryotic extremophiles, many representing taxonomically novel lineages. These findings underscore the untapped microbial diversity in Türkiye’s extreme environments and their potential as rich reservoirs for novel biocatalysts with applications in industrial and environmental biotechnology.
The taxonomic and P450 diversity uncovered in this study contributes to the growing catalogue of reference data for extremophilic microorganisms and their enzymes. These data can support the development of environment-specific microbial or enzymatic markers, aiding the identification of samples from similar geochemical conditions. Previous studies have shown that metagenomic data carry distinctive environmental signatures; for example, they have been used to infer the geographic origin of ancient samples [120], map the spatial distribution of antimicrobial resistance [121], and classify environments using machine learning models [122,123]. By contributing new reference data and uncovering novel P450 lineages, this study provides a valuable resource for future research into the ecological roles and biotechnological potential of extremophile-derived enzymes.
Supporting information
S2 Fig. Shannon diversity versus number of P450s and P450 families.
https://doi.org/10.1371/journal.pone.0330523.s002
(TIF)
S3 Fig. Number of P450s and P450 families per bin.
https://doi.org/10.1371/journal.pone.0330523.s003
(TIF)
S8 Table. P450s that were obtained from the six extremophile sample sites and the functions, where known, of previously characterized members of the same P450 family.
https://doi.org/10.1371/journal.pone.0330523.s011
(DOCX)
References
- 1.
Yadav BS, Yadav AK, Singh S, Singh NK, Mani A. Methods in metagenomics and environmental biotechnology. In: Environmental Biotechnology. Springer International Publishing; 2019. p. 85–113.
- 2. Garlapati D, Charankumar B, Ramu K, Madeswaran P, Ramana Murthy MV. A review on the applications and recent advances in environmental DNA (eDNA) metagenomics. Rev Environ Sci Biotechnol. 2019;18(3):389–411.
- 3. Tringe SG, Rubin EM. Metagenomics: DNA sequencing of environmental samples. Nat Rev Genet. 2005;6(11):805–14.
- 4. Quince C, Walker AW, Simpson JT, Loman NJ, Segata N. Shotgun metagenomics, from sampling to analysis. Nat Biotechnol. 2017;35(9):833–44. pmid:28898207
- 5. Prayogo FA, Budiharjo A, Kusumaningrum HP, Wijanarka W, Suprihadi A, Nurhayati N. Metagenomic applications in exploration and development of novel enzymes from nature: a review. J Genet Eng Biotechnol. 2020;18(1):39. pmid:32749574
- 6. Jeffreys LN, Girvan HM, McLean KJ, Munro AW. Characterization of Cytochrome P450 Enzymes and Their Applications in Synthetic Biology. Methods Enzymol. 2018;608:189–261. pmid:30173763
- 7. Hannemann F, Bichet A, Ewen KM, Bernhardt R. Cytochrome P450 systems--biological variations of electron transport chains. Biochim Biophys Acta. 2007;1770(3):330–44. pmid:16978787
- 8.
McLean KJ, Leys D, Munro AW. Microbial Cytochromes P450. In: Cytochrome P450. 2015. p. 261–407.
- 9. Nzuza N, Padayachee T, Syed PR, Kryś JD, Chen W, Gront D, et al. Ancient Bacterial Class Alphaproteobacteria Cytochrome P450 Monooxygenases Can Be Found in Other Bacterial Species. Int J Mol Sci. 2021;22(11):5542. pmid:34073951
- 10. Msweli S, Chonco A, Msweli L, Syed PR, Karpoormath R, Chen W, et al. Lifestyles Shape the Cytochrome P450 Repertoire of the Bacterial Phylum Proteobacteria. Int J Mol Sci. 2022;23(10):5821. pmid:35628630
- 11. Padayachee T, Nzuza N, Chen W, Nelson DR, Syed K. Impact of lifestyle on cytochrome P450 monooxygenase repertoire is clearly evident in the bacterial phylum Firmicutes. Sci Rep. 2020;10(1):13982. pmid:32814804
- 12. Ngcobo PE, Nkosi BVZ, Chen W, Nelson DR, Syed K. Evolution of Cytochrome P450 Enzymes and Their Redox Partners in Archaea. Int J Mol Sci. 2023;24(4):4161. pmid:36835573
- 13. Li S, Du L, Bernhardt R. Redox Partners: Function Modulators of Bacterial P450 Enzymes. Trends Microbiol. 2020;28(6):445–54. pmid:32396826
- 14. Dauda WP, Abraham P, Glen E, Adetunji CO, Ghazanfar S, Ali S, et al. Robust Profiling of Cytochrome P450s (P450ome) in Notable Aspergillus spp. Life (Basel). 2022;12(3):451. pmid:35330202
- 15. Girvan HM, Munro AW. Applications of microbial cytochrome P450 enzymes in biotechnology and synthetic biology. Curr Opin Chem Biol. 2016;31:136–45. pmid:27015292
- 16. Zhang X, Peng Y, Zhao J, Li Q, Yu X, Acevedo-Rocha CG, et al. Bacterial cytochrome P450-catalyzed regio- and stereoselective steroid hydroxylation enabled by directed evolution and rational design. Bioresour Bioprocess. 2020;7(1).
- 17. Kumar S. Engineering cytochrome P450 biocatalysts for biotechnology, medicine and bioremediation. Expert Opin Drug Metab Toxicol. 2010;6(2):115–31. pmid:20064075
- 18. Msomi NN, Padayachee T, Nzuza N, Syed PR, Kryś JD, Chen W, et al. In Silico Analysis of P450s and Their Role in Secondary Metabolism in the Bacterial Class Gammaproteobacteria. Molecules. 2021;26(6):1538. pmid:33799696
- 19. Elleuche S, Schröder C, Sahm K, Antranikian G. Extremozymes--biocatalysts with unique properties from extremophilic microorganisms. Curr Opin Biotechnol. 2014;29:116–23. pmid:24780224
- 20. Tavanti M, Porter JL, Sabatini S, Turner NJ, Flitsch SL. Panel of New Thermostable CYP116B Self‐Sufficient Cytochrome P450 Monooxygenases that Catalyze C−H Activation with a Diverse Substrate Scope. ChemCatChem. 2018;10(5):1042–51.
- 21. Harris KL, Thomson RES, Strohmaier SJ, Gumulya Y, Gillam EMJ. Determinants of thermostability in the cytochrome P450 fold. Biochim Biophys Acta Proteins Proteom. 2018;1866(1):97–115. pmid:28822812
- 22. Jiang Y, Li Z, Wang C, Zhou YJ, Xu H, Li S. Biochemical characterization of three new α-olefin-producing P450 fatty acid decarboxylases with a halophilic property. Biotechnol Biofuels. 2019;12:79. pmid:30996734
- 23. Tung NV, Hoang NH, Thoa NK. Mining cytochrome p450 genes through next generation sequencing and metagenomic analysis from Binh Chau hot spring. Tap Chi Sinh Hoc. 2019;41(3).
- 24. Nguyen K-T, Nguyen N-L, Milhim M, Nguyen V-T, Lai T-H-N, Nguyen H-H, et al. Characterization of a thermophilic cytochrome P450 of the CYP203A subfamily from Binh Chau hot spring in Vietnam. FEBS Open Bio. 2021;11(1):124–32. pmid:33176055
- 25. Kılıc M, Balci N, Gul Karaguler N, Stewart FJ. Draft Genome Sequence of Virgibacillus sp. Strain AGTR, Isolated from Hypersaline Lake Acıgöl in Turkey. Microbiol Resour Announc. 2022;11(10):e0055522. pmid:36043865
- 26. Oztug M, Cebeci A, Mumcu H, Akgoz M, Karaguler NG. Whole-Genome Sequence of Geobacillus thermoleovorans ARTRW1, Isolated from Armutlu Geothermal Spring, Turkey. Microbiol Resour Announc. 2020;9(24):e00269-20. pmid:32527772
- 27. Balci NÇ, Gül S, Kiliç MM, Karagüler NG, Sari E, Sönmez MS. Biogeochemistry of Balikesir Balya Pb-Zn mine tailings site and its effect on generation of acid mine drainage. Turk Jeol Bult. 2014;57(3):1–24. pmid:WOS:000443708500001
- 28. Akpolat C, Fernández AB, Caglayan P, Calli B, Birbir M, Ventosa A. Prokaryotic Communities in the Thalassohaline Tuz Lake, Deep Zone, and Kayacik, Kaldirim and Yavsan Salterns (Turkey) Assessed by 16S rRNA Amplicon Sequencing. Microorganisms. 2021;9(7):1525. pmid:34361960
- 29. Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics. 2014;30(15):2114–20. pmid:24695404
- 30. Nurk S, Meleshko D, Korobeynikov A, Pevzner PA. metaSPAdes: a new versatile metagenomic assembler. Genome Res. 2017;27(5):824–34. pmid:28298430
- 31. Li H. Minimap2: pairwise alignment for nucleotide sequences. Bioinformatics. 2018;34(18):3094–100. pmid:29750242
- 32. Aroney STN, Newell RJP, Nissen JN, Camargo AP, Tyson GW, Woodcroft BJ. CoverM: read alignment statistics for metagenomics. Bioinformatics. 2025;41(4):btaf147. pmid:40193404
- 33.
Newell R. Aviary. 2022. Available from: https://github.com/rhysnewell/aviary
- 34. Wu Y-W, Simmons BA, Singer SW. MaxBin 2.0: an automated binning algorithm to recover genomes from multiple metagenomic datasets. Bioinformatics. 2016;32(4):605–7. pmid:26515820
- 35. Kang DD, Froula J, Egan R, Wang Z. MetaBAT, an efficient tool for accurately reconstructing single genomes from complex microbial communities. PeerJ. 2015;3:e1165. pmid:26336640
- 36. Kang DD, Li F, Kirton E, Thomas A, Egan R, An H, et al. MetaBAT 2: an adaptive binning algorithm for robust and efficient genome reconstruction from metagenome assemblies. PeerJ. 2019;7:e7359. pmid:31388474
- 37. Alneberg J, Bjarnason BS, de Bruijn I, Schirmer M, Quick J, Ijaz UZ, et al. Binning metagenomic contigs by coverage and composition. Nat Methods. 2014;11(11):1144–6. pmid:25218180
- 38. Nissen JN, Johansen J, Allesøe RL, Sønderby CK, Armenteros JJA, Grønbech CH, et al. Improved metagenome binning and assembly using deep variational autoencoders. Nat Biotechnol. 2021;39(5):555–60. pmid:33398153
- 39. Pan S, Zhu C, Zhao X-M, Coelho LP. A deep siamese neural network improves metagenome-assembled genomes in microbiome datasets across different environments. Nat Commun. 2022;13(1):2326. pmid:35484115
- 40.
Newell R. Rosella 2022. Available from: https://github.com/rhysnewell/rosella
- 41. Sieber CMK, Probst AJ, Sharrar A, Thomas BC, Hess M, Tringe SG, et al. Recovery of genomes from metagenomes via a dereplication, aggregation and scoring strategy. Nat Microbiol. 2018;3(7):836–43. pmid:29807988
- 42. Parks DH, Imelfort M, Skennerton CT, Hugenholtz P, Tyson GW. CheckM: assessing the quality of microbial genomes recovered from isolates, single cells, and metagenomes. Genome Res. 2015;25(7):1043–55. pmid:25977477
- 43. Chaumeil PA, Mussig AJ, Hugenholtz P, Parks DH. GTDB-Tk: a toolkit to classify genomes with the Genome Taxonomy Database. Bioinformatics. 2019;36(6):1925–7. pmid:31730192
- 44. Chaumeil P-A, Mussig AJ, Hugenholtz P, Parks DH. GTDB-Tk v2: memory friendly classification with the genome taxonomy database. Bioinformatics. 2022;38(23):5315–6. pmid:36218463
- 45. Woodcroft BJ, Aroney STN, Zhao R, Cunningham M, Mitchell JAM, Blackall L, et al. SingleM and Sandpiper: Robust microbial taxonomic profiles from metagenomic data. bioRxiv. 2024:2024.01.30.578060.
- 46. Blanco-Míguez A, Beghini F, Cumbo F, McIver LJ, Thompson KN, Zolfo M, et al. Extending and improving metagenomic taxonomic profiling with uncharacterized species using MetaPhlAn 4. Nat Biotechnol. 2023;41(11):1633–44. pmid:36823356
- 47. Milanese A, Mende DR, Paoli L, Salazar G, Ruscheweyh H-J, Cuenca M, et al. Microbial abundance, activity and population genomic profiling with mOTUs2. Nat Commun. 2019;10(1):1014. pmid:30833550
- 48. Wood DE, Lu J, Langmead B. Improved metagenomic analysis with Kraken 2. Genome Biol. 2019;20(1):257. pmid:31779668
- 49. Sun Z, Liu J, Zhang M, Wang T, Huang S, Weiss ST, et al. Removal of false positives in metagenomics-based taxonomy profiling via targeting Type IIB restriction sites. Nat Commun. 2023;14(1):5321. pmid:37658057
- 50. Ounit R, Wanamaker S, Close TJ, Lonardi S. CLARK: fast and accurate classification of metagenomic and genomic sequences using discriminative k-mers. BMC Genomics. 2015;16(1):236. pmid:25879410
- 51. Parks DH, Chuvochina M, Waite DW, Rinke C, Skarshewski A, Chaumeil P-A, et al. A standardized bacterial taxonomy based on genome phylogeny substantially revises the tree of life. Nat Biotechnol. 2018;36(10):996–1004. pmid:30148503
- 52. Parks DH, Chuvochina M, Rinke C, Mussig AJ, Chaumeil PA, Hugenholtz P. GTDB: an ongoing census of bacterial and archaeal diversity through a phylogenetically consistent, rank normalized and complete genome-based taxonomy. Nucleic Acids Res. 2022;50(D1):D785–94. pmid:34520557
- 53. McMurdie PJ, Holmes S. phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data. PLoS One. 2013;8(4):e61217. pmid:23630581
- 54. Rodriguez-R LM, Gunturu S, Tiedje JM, Cole JR, Konstantinidis KT. Nonpareil 3: Fast Estimation of Metagenomic Coverage and Sequence Diversity. mSystems. 2018;3(3):e00039-18. pmid:29657970
- 55. Rodriguez-R LM, Konstantinidis KT. Nonpareil: a redundancy-based approach to assess the level of coverage in metagenomic datasets. Bioinformatics. 2014;30(5):629–35. pmid:24123672
- 56.
Wichkam H. ggplot2: Elegant graphics for data analysis. Springer; 2016.
- 57. Gu Z, Eils R, Schlesner M. Complex heatmaps reveal patterns and correlations in multidimensional genomic data. Bioinformatics. 2016;32(18):2847–9. pmid:27207943
- 58. Larralde M. Pyrodigal: Python bindings and interface to Prodigal an efficient method for gene prediction in prokaryotes. J Open Source Softw. 2022;7(72):4296.
- 59. Hyatt D, Chen G-L, Locascio PF, Land ML, Larimer FW, Hauser LJ. Prodigal: prokaryotic gene recognition and translation initiation site identification. BMC Bioinformatics. 2010;11:119. pmid:20211023
- 60.
Woodcroft BJ. mfqe. 2019. Available from: https://github.com/wwood/mfqe
- 61. Fu L, Niu B, Zhu Z, Wu S, Li W. CD-HIT: accelerated for clustering the next-generation sequencing data. Bioinformatics. 2012;28(23):3150–2. pmid:23060610
- 62. Eddy SR. Accelerated Profile HMM Searches. PLoS Comput Biol. 2011;7(10):e1002195. pmid:22039361
- 63. Fischer M, Knoll M, Sirim D, Wagner F, Funke S, Pleiss J. The Cytochrome P450 Engineering Database: a navigation and prediction tool for the cytochrome P450 protein family. Bioinformatics. 2007;23(15):2015–7. pmid:17510166
- 64. Buchfink B, Xie C, Huson DH. Fast and sensitive protein alignment using DIAMOND. Nat Methods. 2015;12(1):59–60. pmid:25402007
- 65. Katoh K, Standley DM. MAFFT multiple sequence alignment software version 7: improvements in performance and usability. Mol Biol Evol. 2013;30(4):772–80. pmid:23329690
- 66. Capella-Gutiérrez S, Silla-Martínez JM, Gabaldón T. trimAl: a tool for automated alignment trimming in large-scale phylogenetic analyses. Bioinformatics. 2009;25(15):1972–3. pmid:19505945
- 67. Nguyen L-T, Schmidt HA, von Haeseler A, Minh BQ. IQ-TREE: a fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies. Mol Biol Evol. 2015;32(1):268–74. pmid:25371430
- 68. Xie J, Chen Y, Cai G, Cai R, Hu Z, Wang H. Tree Visualization By One Table (tvBOT): a web application for visualizing, modifying and annotating phylogenetic trees. Nucleic Acids Res. 2023;51(W1):W587–92. pmid:37144476
- 69. Steinegger M, Söding J. MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets. Nat Biotechnol. 2017;35(11):1026–8. pmid:29035372
- 70. Sim SC, Ingelman-Sundberg M. The human cytochrome P450 allele nomenclature committee web site: Submission criteria, procedures, and objectives. In: Cytochrome P450 Protocols. 2005. p. 183–92.
- 71. Nelson DR. Cytochrome P450 nomenclature, 2004. Methods Mol Biol. 2006;320:1–10. pmid:16719369
- 72. Wang Z, Xu J-Q, Xu W-M, Li Y, Zhou Y, Lü Z-Z, et al. Salinigranum salinum sp. nov., isolated from a marine solar saltern. Int J Syst Evol Microbiol. 2016;66(8):3017–21. pmid:27151192
- 73. Xie Y-G, Luo Z-H, Fang B-Z, Jiao J-Y, Xie Q-J, Cao X-R, et al. Functional differentiation determines the molecular basis of the symbiotic lifestyle of Ca. Nanohaloarchaeota. Microbiome. 2022;10(1):172. pmid:36242054
- 74. Coskun ÖK, Gomez-Saez GV, Beren M, Ozcan D, Hosgormez H, Einsiedl F, et al. Carbon metabolism and biogeography of candidate phylum “Candidatus Bipolaricaulota” in geothermal environments of Biga Peninsula, Turkey. Front Microbiol. 2023;14:1063139. pmid:36910224
- 75. Wang Y, Bi H-Y, Chen H-G, Zheng P-F, Zhou Y-L, Li J-T. Metagenomics Reveals Dominant Unusual Sulfur Oxidizers Inhabiting Active Hydrothermal Chimneys From the Southwest Indian Ridge. Front Microbiol. 2022;13:861795. pmid:35694283
- 76. Lu H, Gao P, Phurbu D, Wu QL, Xing P. Salegentibacter lacus sp. nov. and Salegentibacter tibetensis sp. nov., isolated from hypersaline lakes on the Tibetan Plateau. Int J Syst Evol Microbiol. 2022;72(1):10.1099/ijsem.0.005202. pmid:35076362
- 77. Sezutsu H, Le Goff G, Feyereisen R. Origins of P450 diversity. Philos Trans R Soc Lond B Biol Sci. 2013;368(1612):20120428. pmid:23297351
- 78. Fulco AJ. P450BM-3 and other inducible bacterial P450 cytochromes: biochemistry and regulation. Annu Rev Pharmacol Toxicol. 1991;31:177–203. pmid:2064373
- 79. Correddu D, Di Nardo G, Gilardi G. Self-Sufficient Class VII Cytochromes P450: From Full-Length Structure to Synthetic Biology Applications. Trends Biotechnol. 2021;39(11):1184–207. pmid:33610332
- 80. Shoji O, Watanabe Y. Peroxygenase reactions catalyzed by cytochromes P450. J Biol Inorg Chem. 2014;19(4–5):529–39. pmid:24500242
- 81. Iizaka Y, Sherman DH, Anzai Y. An overview of the cytochrome P450 enzymes that catalyze the same-site multistep oxidation reactions in biotechnologically relevant selected actinomycete strains. Appl Microbiol Biotechnol. 2021;105(7):2647–61. pmid:33710358
- 82. Bernhardt R. Cytochromes P450 as versatile biocatalysts. J Biotechnol. 2006;124(1):128–45. pmid:16516322
- 83. Shafiee A, Hutchinson CR. Macrolide antibiotic biosynthesis: isolation and properties of two forms of 6-deoxyerythronolide B hydroxylase from Saccharopolyspora erythraea (Streptomyces erythreus). Biochemistry. 1987;26(19):6204–10. pmid:2446657
- 84. Cho M-A, Han S, Lim Y-R, Kim V, Kim H, Kim D. Streptomyces Cytochrome P450 Enzymes and Their Roles in the Biosynthesis of Macrolide Therapeutic Agents. Biomol Ther (Seoul). 2019;27(2):127–33. pmid:30562877
- 85. Li S, Tietz DR, Rutaganira FU, Kells PM, Anzai Y, Kato F, et al. Substrate recognition by the multifunctional cytochrome P450 MycG in mycinamicin hydroxylation and epoxidation reactions. J Biol Chem. 2012;287(45):37880–90. pmid:22952225
- 86. Zhang H, Chen J, Wang H, Xie Y, Ju J, Yan Y, et al. Structural analysis of HmtT and HmtN involved in the tailoring steps of himastatin biosynthesis. FEBS Lett. 2013;587(11):1675–80. pmid:23611984
- 87. Ma J, Wang Z, Huang H, Luo M, Zuo D, Wang B, et al. Biosynthesis of himastatin: assembly line and characterization of three cytochrome P450 enzymes involved in the post-tailoring oxidative steps. Angew Chem Int Ed Engl. 2011;50(34):7797–802. pmid:21726028
- 88. Arisawa A, Tsunekawa H, Okamura K, Okamoto R. Nucleotide sequence analysis of the carbomycin biosynthetic genes including the 3-O-acyltransferase gene from Streptomyces thermotolerans. Biosci Biotechnol Biochem. 1995;59(4):582–8. pmid:7772821
- 89. Han S, Pham T-V, Kim J-H, Lim Y-R, Park H-G, Cha G-S, et al. Functional characterization of CYP107W1 from Streptomyces avermitilis and biosynthesis of macrolide oligomycin A. Arch Biochem Biophys. 2015;575:1–7. pmid:25849761
- 90. Carlson JC, Li S, Gunatilleke SS, Anzai Y, Burr DA, Podust LM, et al. Tirandamycin biosynthesis is mediated by co-dependent oxidative enzymes. Nat Chem. 2011;3(8):628–33. pmid:21778983
- 91. Li F, Ma L, Zhang X, Chen J, Qi F, Huang Y, et al. Structure-guided manipulation of the regioselectivity of the cyclosporine A hydroxylase CYP-sb21 from Sebekia benihana. Synth Syst Biotechnol. 2020;5(3):236–43. pmid:32775708
- 92. Molnár I, Aparicio JF, Haydock SF, Khaw LE, Schwecke T, König A, et al. Organisation of the biosynthetic gene cluster for rapamycin in Streptomyces hygroscopicus: analysis of genes flanking the polyketide synthase. Gene. 1996;169(1):1–7. pmid:8635730
- 93. Takahashi S. Studies on Streptomyces sp. SN-593: reveromycin biosynthesis, β-carboline biomediator activating LuxR family regulator, and construction of terpenoid biosynthetic platform. J Antibiot (Tokyo). 2022;75(8):432–44. pmid:35778609
- 94. Cryle MJ, Matovic NJ, De Voss JJ. Products of cytochrome P450(BioI) (CYP107H1)-catalyzed oxidation of fatty acids. Org Lett. 2003;5(18):3341–4. pmid:12943422
- 95. Yasutake Y, Nishioka T, Imoto N, Tamura T. A single mutation at the ferredoxin binding site of P450 Vdh enables efficient biocatalytic production of 25-hydroxyvitamin D(3). Chembiochem. 2013;14(17):2284–91. pmid:24115473
- 96. Lin S, Ma B, Gao Q, Yang J, Lai G, Lin R, et al. The 16α-Hydroxylation of Progesterone by Cytochrome P450 107X1 from Streptomyces avermitilis. Chem Biodivers. 2022;19(5):e202200177. pmid:35426465
- 97. Tian Z, Cheng Q, Yoshimoto FK, Lei L, Lamb DC, Guengerich FP. Cytochrome P450 107U1 is required for sporulation and antibiotic production in Streptomyces coelicolor. Arch Biochem Biophys. 2013;530(2):101–7. pmid:23357279
- 98.
Hilberath T, Urlacher VB, Pohl M. Identification and Characterization of Novel Cytochromes P450 from Actinomycetes: Universitäts- und Landesbibliothek der Heinrich-Heine-Universität Düsseldorf. 2021.
- 99. Mthethwa BC, Chen W, Ngwenya ML, Kappo AP, Syed PR, Karpoormath R, et al. Comparative Analyses of Cytochrome P450s and Those Associated with Secondary Metabolism in Bacillus Species. Int J Mol Sci. 2018;19(11):3623. pmid:30453558
- 100. Girhard M, Klaus T, Khatri Y, Bernhardt R, Urlacher VB. Characterization of the versatile monooxygenase CYP109B1 from Bacillus subtilis. Appl Microbiol Biotechnol. 2010;87(2):595–607. pmid:20186410
- 101. Girhard M, Machida K, Itoh M, Schmid RD, Arisawa A, Urlacher VB. Regioselective biooxidation of (+)-valencene by recombinant E. coli expressing CYP109B1 from Bacillus subtilis in a two-liquid-phase system. Microb Cell Fact. 2009;8:36. pmid:19591681
- 102. Jóźwik IK, Kiss FM, Gricman Ł, Abdulmughni A, Brill E, Zapp J, et al. Structural basis of steroid binding and oxidation by the cytochrome P450 CYP109E1 from Bacillus megaterium. FEBS J. 2016;283(22):4128–48. pmid:27686671
- 103. Abdulmughni A, Jóźwik IK, Brill E, Hannemann F, Thunnissen A-MWH, Bernhardt R. Biochemical and structural characterization of CYP109A2, a vitamin D3 25-hydroxylase from Bacillus megaterium. FEBS J. 2017;284(22):3881–94. pmid:28940959
- 104. Putkaradze N, Litzenburger M, Abdulmughni A, Milhim M, Brill E, Hannemann F, et al. CYP109E1 is a novel versatile statin and terpene oxidase from Bacillus megaterium. Appl Microbiol Biotechnol. 2017;101(23–24):8379–93. pmid:29018905
- 105. Abdulmughni A, Jóźwik IK, Putkaradze N, Brill E, Zapp J, Thunnissen A-MWH, et al. Characterization of cytochrome P450 CYP109E1 from Bacillus megaterium as a novel vitamin D3 hydroxylase. J Biotechnol. 2017;243:38–47. pmid:28043840
- 106. Khatri Y, Hannemann F, Ewen KM, Pistorius D, Perlova O, Kagawa N, et al. The CYPome of Sorangium cellulosum So ce56 and identification of CYP109D1 as a new fatty acid hydroxylase. Chem Biol. 2010;17(12):1295–305. pmid:21168765
- 107. Khatri Y, Hannemann F, Girhard M, Kappl R, Même A, Ringle M, et al. Novel family members of CYP109 from Sorangium cellulosum So ce56 exhibit characteristic biochemical and biophysical properties. Biotechnol Appl Biochem. 2013;60(1):18–29. pmid:23586989
- 108. Ghith A, Bell SG. The oxidation of steroid derivatives by the CYP125A6 and CYP125A7 enzymes from Mycobacterium marinum. J Steroid Biochem Mol Biol. 2023;235:106406. pmid:37793577
- 109. Sterner R, Liebl W. Thermophilic adaptation of proteins. Crit Rev Biochem Mol Biol. 2001;36(1):39–106. pmid:11256505
- 110. Caron B, Mark AE, Poger D. Some Like It Hot: The Effect of Sterols and Hopanoids on Lipid Ordering at High Temperature. J Phys Chem Lett. 2014;5(22):3953–7. pmid:26276476
- 111. Wong NR, Liu X, Lloyd H, Colthart AM, Ferrazzoli AE, Cooper DL, et al. A new approach to understanding structure-function relationships in cytochromes P450 by targeting terpene metabolism in the wild. J Inorg Biochem. 2018;188:96–101. pmid:30170307
- 112. Bell SG, Yang W, Yorke JA, Zhou W, Wang H, Harmer J, et al. Structure and function of CYP108D1 from Novosphingobium aromaticivorans DSM12444: an aromatic hydrocarbon-binding P450 enzyme. Acta Crystallogr D Biol Crystallogr. 2012;68(Pt 3):277–91. pmid:22349230
- 113. He Z, Zhang K, Wang H, Lv Z. Trehalose promotes Rhodococcus sp. strain YYL colonization in activated sludge under tetrahydrofuran (THF) stress. Front Microbiol. 2015;6:438. pmid:26029182
- 114. Wang L, Wang W, Lai Q, Shao Z. Gene diversity of CYP153A and AlkB alkane hydroxylases in oil-degrading bacteria isolated from the Atlantic Ocean. Environ Microbiol. 2010;12(5):1230–42. pmid:20148932
- 115. Rojo F. Degradation of alkanes by bacteria. Environ Microbiol. 2009;11(10):2477–90. pmid:19807712
- 116. Alonso-Gutiérrez J, Teramoto M, Yamazoe A, Harayama S, Figueras A, Novoa B. Alkane-degrading properties of Dietzia sp. H0B, a key player in the Prestige oil spill biodegradation (NW Spain). J Appl Microbiol. 2011;111(4):800–10. pmid:21767337
- 117. Funhoff EG, Bauer U, García-Rubio I, Witholt B, van Beilen JB. CYP153A6, a soluble P450 oxygenase catalyzing terminal-alkane hydroxylation. J Bacteriol. 2006;188(14):5220–7. pmid:16816194
- 118. Otomatsu T, Bai L, Fujita N, Shindo K, Shimizu K, Misawa N. Bioconversion of aromatic compounds by Escherichia coli that expresses cytochrome P450 CYP153A13a gene isolated from an alkane-assimilating marine bacterium Alcanivorax borkumensis. J Mol Catalysis B Enzymatic. 2010;66(1–2):234–40.
- 119. Rambabu K, Banat F, Pham QM, Ho S-H, Ren N-Q, Show PL. Biological remediation of acid mine drainage: Review of past trends and current outlook. Environ Sci Ecotechnol. 2020;2:100024. pmid:36160925
- 120. Bozzi D, Neuenschwander S, Cruz Dávalos DI, Sousa da Mota B, Schroeder H, Moreno-Mayar JV, et al. Towards predicting the geographical origin of ancient samples with metagenomic data. Sci Rep. 2024;14(1):21794. pmid:39294129
- 121. Zhelyazkova M, Yordanova R, Mihaylov I, Kirov S, Tsonev S, Danko D, et al. Origin Sample Prediction and Spatial Modeling of Antimicrobial Resistance in Metagenomic Sequencing Data. Front Genet. 2021;12:642991. pmid:33763122
- 122. Kawulok J, Kawulok M, Deorowicz S. Environmental metagenome classification for constructing a microbiome fingerprint. Biol Direct. 2019;14(1):20. pmid:31722729
- 123. Anyaso-Samuel S, Sachdeva A, Guha S, Datta S. Metagenomic Geolocation Prediction Using an Adaptive Ensemble Classifier. Front Genet. 2021;12:642282. pmid:33959149