Introduction

The world’s population is predicted to reach 9.7 billion by 2050. Increasing demand for food and energy contributes to over-exploitation of natural resources and environmental degradation. Particularly, release of pollutants into the environment from manufacturing and resource industries is a major concern for ecosystem health. Pollutants, either organic (e.g. polycyclic aromatic hydrocarbons) or inorganic (e.g. heavy metals), when they interact with their physical environment, not only affect the quality of the environment but also have cascading effects on the health and wellbeing of organisms across all domains of life [1, 2]. Unlike organic pollutants, heavy metal(loid)s or trace metals (e.g. chromium-Cr, mercury-Hg, selenium-Se) are non-biodegradable, persist longer in the environment and bioaccumulate through the food web. Even at low concentrations, they are likely to have a long-term impact on ecosystem health [3, 4]. Naturally occurring metals are integral to the evolution of living organisms and are critical for metabolic activities e.g. co-factors for enzymes [5]. Despite their biological importance, large amounts of these metals can result in cellular and tissue damage, i.e. cytotoxicity in animals and growth inhibition in microbes; poor growth, low yields and nutrient inbalances in plants and metabolic interferences and mutangenesis in all types of organisms [6, 7]. Detailed reviews on the occurrence of heavy metals in the environment, industrial production and usage, potential for human exposure, and their molecular mechanisms of toxicity can be found in [8,9,10,11]. In this review, we also highlight the impact of tannery industries (i.e. Cr pollution – Table 1), selenium-polluting industries including mining (Table 2), and artisinal gold mining (i.e. Hg pollution – Table 3) as case studies.

Table 1 Chromium pollution from tannery industries—a case study
Table 2 Selenium an essential element with toxicity problems in the mining industry and beyond
Table 3 Mercury pollution from artisanal/small scale gold mining

Current remediation strategies of contaminated sites include chemical extraction (with acids or chelating agents), immobilisation, encapsulation and electrolysis. Although these strategies have been useful, they come with several limitations including significant alterations of the physicochemical properties, low efficiency and high cost of operation [38]. Phyto/bioremediation has been suggested as an alternative eco-friendly approach to detoxify metals from contaminated sites [39]. This approach leverages intrinsic biological mechanisms of plants and microorganisms to transform and/or bioaccumulate metals from the environment [1]. In particular, microorganisms possess remarkable abilities to bioaccumulate, retain and transform heavy metal ions [40,41,42] by taking advantage of reduction/oxidation (redox) and other processes e.g. modulating solubility of metals without changing the oxidation state of the metal [43, 44].

Metal(loids) exhibit different physical and chemical forms (i.e. differences in speciation) in the environment. Electronic configuration, oxidation state and ionic radius all define the chemical speciation of a particular metal and its fractionation (i.e. whether it is labile/inert, ligand complexed, precipitated or existing as a free ion). Consequently, the chemical form of a metal strongly influences its reactivity, toxicity, mobility and interaction with microorganisms in the environment [44, 45]. For instance, copper (Cu) becomes potentially toxic when it transitions between Cu(II) and Cu(I), soluble and toxic chromium (Cr(VI)) are less toxic when reduced to Cr(III), mercury (Hg(II)) becomes neurotoxic when its methylated (CH3Hg) and As(III) is more toxic than As(V) [46,47,48].

Recent research has highlighted the ability of aerobic methanotrophs, a specialised group of bacteria that can use methane (CH4) as a sole carbon and energy source, to transform metals (and also metalloids) such as Cu, Cr, Se and Hg [41, 49, 50]. Methanotrophs belong to the phyla Proteobacteria (classes Alphaproteobacteria and Gammaproteobacteria), Candidate division NC10 and Verrucomicrobia. Till date, more than 29 methanotroph genera and 8 families (i.e. Methylococcaceae, Methylothermaceae, Crenotrichaceae, Methylocystaceae, Beijerinkiaceae, and Methylacidiphilaceae and two currently unclassified families) have been identified within these phyla. Complete genome sequences for representatives of >23 genera are available in public repositories [51, 52]. Proteobacterial methanotrophs are active primarily in methane-oxygen counter gradients of oxic-anoxic interfaces [and in upland soils (high affinity atmospheric methane oxidisers)], while methanotrophs found in extremely acidic geothermal sites belong to the phylum Verrucomicrobia. Anaerobic microbial methane oxidation has been recently discovered that use reverse methanogenesis process to convert CH4 into CO2 [53, 54]. In contrast, the members of the candidate phylum NC10, such as Candidatus Methylomirabilis oxyfera, use all aerobic methane oxidated pathway-specific proteins, while it acquires oxygen through reduction of nitrite to the oxidation of methane via a unique oxygen-producing pathway [55]. The activity and diversity of methanotrophs and their impact on methane fluxes in different environments (e.g. landfills, rice paddies, natural gas seeps, hypogenic caves, saline lakes) have been studied extensively using both cultivation-dependent and molecular ecology tools [56,57,58,59]. In this review, we focus on recent developments in the physiology of aerobic methanotrophs with emphasis on their role in transformation and speciation of metals and metalloids such as Cu, Cr, Se and Hg. Detailed descriptions of the biochemistry and physiology of methanotrophs/methylotrophs are beyond the scope of this review and the reader is referred to [5, 60].

Aerobic methane oxidation and metalloenzymes

In aerobic methanotrophs, four major steps are involved in the enzymatic conversion of CH4 into biomass/CO2, in which the availabilities of different metal ions play a critical role (Fig. 1). The biocatalytic oxidation of CH4 to methanol (CH3OH) is modulated by the enzyme methane monooxygenase (MMO). Two forms of MMO exist: membrane-bound particulate MMO (known as pMMO and its divergent form pXMO [61]) and cytoplasmic soluble MMO (sMMO).

  1. (a)

    pMMO composed (i) β-subunit, PmoA (26 kDa); (ii) α-subunit, PmoB (45 kDa); and (iii) γ-subunit, PmoC (23 kDa) with an (αβγ)3 structure. Their genes are typically arranged in pmo-operons as pmoCAB. Th general hypothesis is that the enzyme pMMO obtains electrons from the cytochrome bc1 complex, which contains heme groups and Fe2S2 clusters [62, 63]. It has been proposed that Fe is required for pMMO activity [63], although a recent study shows only the presence of two mono-copper sites [64]. However, the alphaproteobacterial pMMO acquires electrons from ubiquinone pool through NADH oxidation [65, 66], while the pMMO activity in gammaproteobacterial methanotrophs is reported to be coupled to the oxidation of methanol to formaldehyde [67].

  2. (b)

    sMMO is a well-characterised three-component enzyme: (i) hydroxylase contains 3 sub-units (α:β:γ – 54: 42: 22 kDa, respectively) with an (αβγ)2 structure; (ii) reductase (38–40 kDa) that supplies electrons from NADH to the hydroxylase; and (iii) component B as a regulatory protein (15–17 kDa) [68,69,70]. The sMMO is encoded by mmoXYBZDC. It contains a di-iron active site cluster in the hydroxylase component as well as an Fe2S2 cluster and flavin adenine dinucleotide (FAD) moiety in its reductase component [52].

Fig. 1
figure 1

Methane oxidation by aerobic methanotrophs and metal co-factors of the enzymes. pMMO = particulate methane monooxygenase. sMMO = soluble methane monooxygenase. Xox-MeDH = XoxF-methanol dehydrogenase. Mxa-MeDH = MxaFI-methanol dehydrogenase. Fae = formaldehyde activating enzyme. FaDH = formate dehydrogenase. CBB = Calvin Benson Bassham Cycle. RuMP = Ribulose MonoPhosphate cycle. TCA – The Citric Acid cycle. PHB – Polyhydroxybutyrate cycle. Enzymes modulating the reaction are represented in red font, metals in blue & yellow fonts. Small vertical light blue arrows next to each metal ion indicate their effect on the expression and/or activity of enzymes

The two types of MMO differ completely in their metal ion requirements, substrate specificity and CH4 oxidation kinetics [52, 64, 70,71,72]. Specifically, Cu(II) to biomass ratio determines the expression and activity of MMOs in methanotrophs that can express either form of the enzyme and is often referred to as the ‘copper switch’ [52, 61, 72,73,74]. Most aerobic methanotrophs possess pMMO or pXMO (with exceptions of alphaproteobacterial members within the genera Methylocella, Methyloceanibacter and Methyloferula), while few possess both pMMO and sMMO (e.g. Methylosinus spp., Methylocystis spp., Methylobacter marinus, Methylocaldum marinum, Methylococcus capsulatus, Methylomagnum ishizawai 175, Methylomicrobium buryatense 5G, Methylovulum miyakonense HT12, Methylomonas spp. [52, 75,76,77];). A very few methanotrophs e.g. Methylocella spp., Methyloferulla spp., Methyloceanibacter spp., (Alphaproteobacteria) and Methyloterricola spp. (Gammaproteobacteria) contain only sMMO (Vekeman et al., 2016; Semrau et al., 2018). The methane turnover number per active site cells producing pMMOs (0.5–2.5 s-1) are comparatively lower than sMMO expressing cells (>3.5 s-1) [70].

In the second step, methanol is converted into formaldehyde by the enzyme methanol dehydrogenase (MeDH), which contains a pyrroloquinoline quinone (PQQ) cofactor. There are two distinct types of MeDH, (a) calcium dependent MeDH (i.e. MxaF-MeDH) or (b) the recently discovered XoxF, which is an MeDH dependent on rare earth elements (REEs) e.g. cerium or lanthanum [5, 78,79,80,81,82].

  1. (a)

    MxaF-MeDH is tetrameric (α2 – 66 kDa; β2 – 8.5 kDa), with greatest activity at pH~9 and requiring activation (e.g. by ammonia [83, 84];). The α sub-unit contains a single Ca2+ ion, the coordination sphere of which contains one nitrogen and two oxygen ligands from PQQ and a further four oxygen ligands from three aminoacyl sidechains (Glu 205, Asn 289 and Asp 331) [85].

  2. (b)

    XoxF-MeDH is a α2 homodimer, located in the periplasm and active at neutral pH (optimum ~7). It coordinates the REE at the active site in a 9-coordinate fashion, via the same three ligands from PQQ and six oxygen ligands derived from four amino acyl residues (Glu 197, Asn 285, Asp 327 and Asp 329) of the protein [86]. It has been reported that the activity of purified XoxF was higher with light REEs (atomic numbers 57–63; La-Eu) in comparison to heavy REEs (atomic numbers 64–71; Gb-Lu) [87].

It should be noted that XoxF-MeDHs are also present in yeast, moulds, fungi and non-methylotrophic bacteria [88]. In methanotrophs such as Methylomagnum ishizawai 175, Methylomicrobium kenyense AMO1, Methyloterricola oryzae 73a, Methylocystis sp., Mit Z-2018, Methylosinus spp., R-45379, Methyloacidiphilum fumariolicum SolV and Verrucomicrobium spp., the only form of MeDH that have been identified is the XoxF-MeDH. The physiology and genetic basis of the ‘REE-switch’ that controls expression of the alternate forms of MeDH is reviewed extensively in [5, 52]. XoxF-MeDH is reported to have higher affinity for methanol and faster conversion rate [89]. While there was an assumption that XoxF-MeDH oxidises methanol to formate by dual activity [90, 91], a recent study by Good and Colleagues [92], using XoxF purified from Methylobacterium extorquens AM1, confirmed that formaldehyde is the final product. There are 5 families of XoxF-MeDH that might display different catalytic properties [93], while all need to be validated to conclude whether the dual activity is relevant in vivo or not.

In subsequent steps, the formaldehyde derived from the oxidation of methanol is assimilated as biomass either via the ribulose monophosphate pathway (RuMP) in most of Gammaproteobacteria or the serine pathway in Alphaproteobacteria. There are multiple pathways for oxidation of formaldehyde to CO2 i.e. in conjugation with tetrahydromethanopterin (H4MPT) or tetrahydrofolate (H4F) [94].

  1. (a)

    The multi-step H4MPT- and H4F-dependent pathways, which also exist in non-methanotrophic methylotrophs such as Methylobacterium extorquens, operate by conjugating the formaldehyde and subsequent intermediates with the respective coenzymes. In model methanotrophs such as Mc. capsulatus (Bath) and Ms. trichosporium, OB3b studies have shown the use of both H4MPT- and H4F-dependent pathways [95,96,97,98].

  2. (b)

    DL-FalDH is homotetramer with a sub-unit mass of ~ 49 kDa, while the sub-unit contains a PQQ as red-ox co-factor at 1:1 ratio. It utilises cytochrome b559/569 complex as a electron acceptor [99].

  3. (c)

    N-FalDH was reported as an NAD(P)+-linked dehydrogenase from Mc. capsulatus (Bath) where the ability to oxidise formaldehye depended on a low molecular-mass heat-stable component [100, 101]. Later attempts to replicate these results yielded preparations that were active but found to be mixtures of other enzymes and cofactors involved elsewhere in the methane oxidation pathway [102]. Hence, whether an enzyme that directly catalyses NAD(P)+-linked oxidation of formaldhyde is a substantial contributor to the aerobic methane oxidation pathway remains unclear.

However, cytochrome bc1 (with Fe) is required for FalDH as an electron acceptor. Also, other enzymes involved in the conversion of methanol into formaldehyde, formate and finally into carbon dioxide primarily rely on Fe, with minor requirements of Cu, Ca, Mo and Zinc (Zn) ions. In particular, the membrane-associated formate dehydrogenase (FaDH) enzyme, which is 2 αβγδ protomers (Mw ~400 kDa). The holoenzyme contains flavin, iron, inorganic sulphide and molybdenum [103]. The FaDH is involved in the conversion of formate to CO2, contains four FexSx clusters and molybdenum (Mo) as cofactors [62]. Recently, the presence of two clusters encoding NAD+-dependent FaDH that require either tungsten (W) or Mo was reported [90]. Moreover, it was also observed that in Methylomicrobium alcaliphilum 20ZR formate production was considerably reduced during the presence of W [90]. Further work is required to develop a better understanding of whether a ‘W-Mo switch’ regulates expression of the two types of FaDH.

Metal uptake mechanisms in aerobic methanotrophs

Metal ions are integral to metabolic activities (Table 4) in methanotrophs. It has been estimated that about one quarter to one third of any bacterial cell proteins required metal ions to support their functions [104]. However, cells can restrict the number of metal ions to be transported into their cytoplasm, thereby creating a competition between different proteins requiring the same metal ions and thus influencing different enzyme activities in vivo. Specific energised pumps are used to facilitate metal ion transport either into the cell (importers) or out of it (exporters). Such transporters are generally located in the cell membrane and may use selective metal binding proteins or small molecules such as siderophores [105]. In methanotrophs, metals are transported into the cell by (a) passive diffusion e.g. direct transport through porins in surface membrane and/or (b) active transport e.g. through TonB-dependent metal transport or other metal transport proteins [106, 107].

Table 4 Redox states of metals with metallo-enzymes and their specific catalytic functions

Passive diffusion of metals

Passive transport of metals and organic solutes usually occurs via porins. The pore size and amino-acid compositions in the channel of porins determine their specificity towards different solutes and transport. In bacteria, the performance of porins (both non-specific and specific diffusion channels) is usually regulated by the availability of specific nutrients and downregulated by the presence of toxins or harmful solutes [108]. Based on their structure and diffusion characteristics, porins are classified as (a) specific monomeric, (b) specific trimeric, (c) non-specific monomeric and (d) non-specific trimeric diffusion channels [109], while little is known about the porin structures in methanotrophs. Reseachers have characterised a small number of outer membrane proteins (e.g. MopE and CorA) that can perform the function of copper transport in methanotrophs [106]. While MopE (some methanotrophs from Gammaproteobacteria express it) has been reported to bind copper with high affinity (< 1020 M-1) and to have a binding site composed of two imidazoles and a kynurenine (modified tyrptophan side-chain) group [110], other outer membrane proteins, as wells as their arrangements are yet to be characterised in detail. Unchelated metal ions are also reported to be transported via porins e.g. Cu and Hg transport in M. album BG8 [50].

Active transport of metals

It has been reported that about 10−6 M of intracellular ion concentration is required for any cellular activity and at lower levels, metal binding proteins or small molecules (e.g. siderophores) are synthesised by bacteria (including methanotrophs) to scavenge metals from the environment. For example, methanotrophs produce methanobactin under Cu limited conditions [111, 112]. Methanobactin, a chalkophore (chalk – copper in Greek), is a ribosomally produced and post-translationally modified peptide [72, 111, 113]. It plays a key role in active transport of metals and perhaps also enables ecological succession of different methanotrophs under external metal toxicities [72]. Specifically, the methanotrophs can export up to 3–50 methanobactin molecules per cell per second dependent on the Cu concentrations in the external solute [72]. Only very few methanotrophic species from the Alphaproteobacteria family are reported to biosynthesise methanobactin and only 10% of sequenced methanotrophs contain methanobactin biosynthesis genes [72, 74, 114]. Methanobactins produced by different methanotrophs are structurally distinct (i.e. till date only 7 methanobactins are characterised from methanotrophs), though with similar metal (i.e. Cu) binding sites, and are categorised as Group-I and II (Table 5). Ms. trichosporium OB3b was reported to produce the highest concentrations of Group-I methanobactin (35–60 mg l−1) followed by the Mc. capsulatus (Bath) (18–24 mg l−1) [115]. Interestingly, Mc. capsulatus (Bath) does not have the mbn cluster encoding methanobactin production [116, 117] indicating a different type of copper-binding molecule (including but possibly not restricted to MopE [118] used in high-affinity copper acquisition). Group-II methanobactin is mainly produced by Methylocystis sp. strain SB2, while recent bioinformatic analysis has shown that there are few methanotrophs that can make both the forms of methanobactins e.g. Methylocystis parvus OBBP, Methylocystis sp. LW5, Methylosinus sp. LW3, Methylosinus sp. R-45379 and Methylosinus sav2 [112].

Table 5 Difference between two well characterised known groups of methanobactins

As discussed earlier, methanobactin is reported to show high affinity towards Cu (~1011 to 1034 M−1) followed by Ag (~107 M−1) as summarised in Fig. 2. Based on the metal binding properties of methanobactin, metals are grouped under two categories: Group A metals—Ag(I), Au(III), Hg(II), Cu(II) and Pb(II) and Group B metals—Cd(II), Co(II), Fe(III), Mn(II), Ni(II) and Zn(II) [118]. Group A metals bind at both oxazolone rings and are reduced upon binding, while Group B metals bind at just one oxazolone ring and are not reduced upon binding [118, 119]. Based on the methanobactic metal selectivity, the metals are grouped as (a) Group-1 (high—100%): Cu (I) and Ag (I); (b) Group 2 (moderate—85–96%): Zn (II), Ni (II) and Co (II); and (c) Group 3 (low—<50%): Mn (II), Pb (II) and Fe (II). Studies have also shown that the methanobactin can also bind, transform and detoxify Hg [50, 69, 120] and U (VI) [121]. In a recent study, a methanobactin characterised from Methylosinus sporium was found to have one oxazolone and one imidazolone ring with disulphide bond between two Cys residues as in Group 1 [116]. The structural difference may affect the metal binding and affinity, which is not completely understood yet. So the classification, metal affinity and grouping of metals described above may applicable only to the methanobactins from Methylosinus trichosporium OB3b and Methylocystis sp. strain SB2, while it may vary for other methanobactin structures and requires further investigation. This knowledge is critical in determining how to exploit methanotroph strains and/or methanobactins to sequester the metal ions from the environment.

Fig. 2
figure 2

Grouping of metals based on their affinity and selectivity with methanobactin

Copper accumulation by methanotrophs

Copper is central to methanotrophic activity, while 40% of methanotrophs are reported to have Cu storage proteins and only 10% reported to make methanobactins for Cu uptake, while the remainder may use passive transport of metals using other porin related proteins [122]. It has been estimated that methanotrophs have at least a ten-fold greater Cu requirement compared with other bacteria [123]. While Cu is an essential metal ion for pMMO activity (i.e. CH4 to methanol conversion), it has also been shown to control expression of other enzymes (e.g. formaldehyde dehydrogenases, hemerythrin) and outer membrane proteins involved in Cu assimilation, regulation and transport. Moreover, it can influence the membrane structure or formation in methanotrophs e.g. M. capsulatus (Bath), M. album BG8 and Ms. trichosporium OB3b [123, 124]. The Cu-containing protein pMMO, which constitutes up to 20% of the total cellular proteins in methanotrophs [125] shows high-affinity towards CH4. The pMMO is expressed at high Cu concentrations, and at low levels, sMMO is expressed [126]. At high Cu concentrations, the Cu will bind the active site of sMMO and inhibit the electron transfer between the falvin adenine di-nucleotide and MMOH [127]. High copper-to-biomass ratio leads to pMMO expression through the canocial ‘copper switch’, but the exact mechanisms by which Cu activates pMMO expression is not yet clear. However, the rate of copper uptake by passive diffusion or active transport may differ between methanotrophs, mainly in Alphaproteobacteria and Gammaproteobacteria. During active transport, divalent Cu is reduced to monovalent Cu, which forms a more stable complex with methanobactin that will not be disassociated by simple dissolution mechanisms and transport into the cell. The methanobactin-Cu (I) complexes (at 1:1 stochiometry [119];) are reported to be too large to be transported via porins, and their transport is mainly mediated by outer membrane TonB-dependent transporters (TBDTs [106, 107];). However, it should be noted that the mechanism of Cu reduction by methanobactin is still unknown [111], while Cu concentrations and pH are known to influence the metal-methanobactin interactions [115, 128]. The methanobactin-Cu complex is also reported to regulate reductase-dependent oxidase activity, dismutation of O2 to H2O2, and the reductant-dependent reduction of H2O2 to H2O [129]. Recently, researchers have characterised the novel Csp-proteins i.e. Csp1 and Csp2 that were reported to be a major reservoir for Cu (up to 13 Cu(I) in one Csp protein molecule; affinity ~ 1017 M-1) and proposed to be exported from the cytosol to supply copper to pMMOs [130]. The related protein, Csp3, can hold upto 80 Cu(I) ions. Csp3 is speculated to sequester excess copper in the cytosol and rescue cellular activity from metal toxicity. Csp3, which also occurs in non-methanotrophs including Bacillus subtilis, is the only bacterial system known to store Cu in the cytoplasm [122]. The mechanism of Cu(I) release from the metallo-proteins methanobactin and the Csp is still not clear. Considering their copper requirements and ability to produce Cu-binding proteins and peptides, methanotrophs could be potentially exploited for copper extraction and recovery from Cu ore, minerals and tailings or remediation of Cu contaminated sites using either whole cell or immobilised protein based approaches [131]. Existing bioleaching methods work effectively under acidic pH (~2.0), whereas methanotroph-based bioleaching could work effectively under neutral pH and thus be considered as more eco-friendly.

Interaction of methanothrophs with chromium (VI)

Chromium is found in the environment mainly in its two most stable oxidation states, the highly soluble, bioavailable and oxidising hexavalent form and the less toxic, less soluble and less bioavailable trivalent form [132]. Despite world-wide regulation of the use of hexavalent chromium in metal plating, to inhibit corrosion and as a wood preservative, among other applications, hexavalent Cr (VI) continues to be a substantial environmental problem. Chromium is heavily used in a number of industries, especially in the production of chromium-iron alloys such as stainless steel and the use of Cr (III) salts in leather manufacture (see Table 1 for case study on the extent of Cr pollution from tannery industries). In 2017, total production of chromium-iron alloys was 31 million tons. In which, 70–80% comes from South Africa (15 MT), Kazakhstan (5.4 MT), India (3.2 MT) and Turkey (2.8 MT), while China and the USA are the top consumers of Cr alloys. Anthropogenic Cr contamination contains a significant amount of Cr (VI) which may leach into the aqueous environment [12, 133].

A wide range of bacteria including methanotrophs have been found able to bioremediate Cr(VI) by reducing it to the less harmful trivalent form [134]. Among the well characterised methanotrophs, Methylococcus capsulatus (Bath) is able to reduce Cr(VI) over a wide range of concentrations (tested across 1.4 to 1000 mg L-1), which offers the attractive possibility of using cheaply available methane to drive bioremediation of Cr(VI) and to sequester the Cr(III) product in the insoluble fraction, associated with the bacterial biomass (Fig. 3a). X-ray spectroscopy has confirmed that the Cr is in the +3 oxidation state and appears to have oxygen and phosphorous ligation [41]. Cell fractionation, together with in situ analysis of chromium distribution via X-ray photoelectron spectroscopy (XPS) and transmission electron microscopy coupled to energy-dispersive X-ray spectroscopy (TEM-EDX) showed that the chromium is predominantly intracellular [136]. Cells of the alphaproteobacterial methanotroph Ms. trichosporium did not reduce chromium (VI), although another gammaproteobacterial methanotroph Methylomonas koyamae SHU1, has been identified as able to reduce chromium (VI) [137]. Bioinformatic analysis of the genome of Mc. capsulatus (Bath) revealed five candidate reductases homologous to enzymes from other microorganisms known to reduce Cr(VI) [41]. Analysis of representatives of all methanotroph genera for which genome sequences is available indicates the presence of multiple potential chromium (VI) reductases in all of them. Comparative genomics of the three species for which the chromium (VI) reduction phenotype is known shows that only one homologue, a putative Na+-translocating NADH-quinone reductase subunit F (locus tag MCA2384 in Mc. capsulatus (Bath)), correlates with the ability to reduce chromium (VI) and so is a candidate for the chromium (VI)-reducing activity. This homologue is not found in any currently available alphaproteobacterial genomes and is present in 29 out of 44 gammaproteobacterial methanotroph genomes (Fig. 3b).

Fig. 3
figure 3

a Pathways of methane-driven metal biotransformation by obligate aerobic methanotrophs. b Genomic distribution of potential biomarker genes involved in metal transformation in methanotrophs. Presence/absence of biomarker genes are mapped to a phylogenomic tree constructed using 74 single-copy marker genes specific to Bacteria via the GtoTree (v1.5.22) pipeline (as described in [135]). Protein sequences were retrieved using HMMER3 tool and multiple alignments were produced using MUSCLE (v.3.8.31, default settings). Conserved alignment blocks were identified using trimal (v1.4; -automated1 option) and subsequently used for tree construction using the IQTREE2 (v2.0.3) using default setting and 1000 boostraps

A mixed culture in a membrane biofilm reactor system was found able to reduce chromium during continuous operation (feeding at 1–3 mg L-1 of Cr (VI)) with a microbial consortium that contained microorganisms of the genera Meiothermus and Methylosinus. It was concluded that the reduction of Cr (VI) was performed primarily by the Meiothermus utilising multicarbon nutrients released by the Methylosinus growing on methane [20]. This is consistent with observations that pure cultures of the well characterised alphaproteobacterial methanotroph Ms. trichosporium OB3b did not reduce Cr (VI) [41], while a mixed culture of Ms. trichosporium OB3b and a Cr (VI)-reducing strain of Escherichia coli was able to reduce Cr (VI) using methane as the only externally supplied carbon source (A. Al Hasin, T.J. Smith and P.H.E. Gardiner, unpublished observations). Collectively, these data indicate that methanotrophs, whether in pure culture or mixed microbial communities, have the capacity to bioremediate Cr (VI) contamination over a wide range of concentrations using cheaply available methane as the feedstock.

Interaction of methanothrophs with selenium

Selenium is an essential micronutrient across all domains of life (including prokaryotes and humans), principally because of its role in selenocysteine in certain enzymes such as glycine reductase, formate dehydrogenase, glutathione peroxidase, iodothyronine deiodinase and thioredoxin reductase. Selenium is also a substantial environmental problem, where the toxic and water-soluble oxyanions selenite (SeO32-) and selenate (SeO42-) may be present from natural and anthropogenic sources and are a risk to humans, animals and other forms of life [138] (see Table 2 for case study on problems of selenium associated with the mining industry and beyond). Some bacteria are able to respire using selenate (SeO42-) as their terminal electron acceptor, while others are able to reduce selenium species to elemental selenium and perform methylation reactions [40, 139]. The primary methylated forms of selenium produced by microorganisms are dimethyl selenide (CH3-Se-CH3) and dimethyl diselenide (CH3-Se-Se-CH3), although others including dimethyl selenone [(CH3)2SeO2], dimethyl triselenide (CH3-Se-Se-Se-CH3), methyl selenol (CH3-Se-H) and mixed selenium/sulphur-methylated species such as dimethyl selenyl sulphide (CH3-Se-S-CH3) and dimethyl selenyl disulphide (CH3-Se-S-S-CH3) have also been observed [40, 49]. Elemental selenium is insoluble and so is generally considered the most benign form of the element with lowest bioavailability. When ingested by mice, Se as nanoparticulate Se(0) had 7-fold lower LD50 compared with Se as selenite [140]. In contrast, ingestion of Se(0) nanoparticles by fish has about 5-fold lower LD50 compared with selenite. A study using the estuarine invertebrate Potamocorbula amurensis suggested that Se(0)-rich particles produced by environmental consortia were more bioavailable than nanoparticulate Se(0) produced chemically or by pure bacterial cultures [141]. The diversity of selenium nanoparticles that can be produced by microorganisms may find applications in electronics and other industries [142,143,144].

Pure cultures of the Mc. capsulatus (Bath) or Ms. trichosporium OB3b do not detectably transform selenate, although both are able to remove selenite. The principal product is elemental selenium in the form of extracellular nanoparticles (Fig. 3a), as well as a small proportion that is converted to methylated Se species. The removal of selenite occurs more rapidly in Mc. capsulatus (Bath) compared with Ms. trichosporium OB3b (2-fold difference in rate observed in comparable tests in the laboratory), at the respective optimum temperatures (45 and 30°C) of the two strains. Cultures of Mc. capsulatus (Bath) completely removed selenite from a starting concentration of 40 mg L-1 within 50 h, with 75% conversion to elemental Se and the production of detectable methylated species (Fig. 3a [49];).

While it is not clear whether all volatile selenium species are produced via elemental Se reduction, Mc. capsulatus (Bath) and Ms. trichosporium OB3b are each able to transform selenium nanoparticles into methylated species. Various mixtures of Se volatiles were detected, depending on the methanotroph strain used and the type of Se supplied (selenite, biogenic selenium nanoparticles produced by the methanotroph, or chemically produced selenium). Mc. capsulatus supplied with selenite produced the largest number of detectable Se volatiles: dimethyl selenide (CH3-Se-CH3), dimethyl diselenide (CH3-Se-Se-CH3), dimethyl selenyl sulphide (CH3-Se-S-CH3), methyl selenol (CH3-Se-H) and methyl selenoacetate (CH3-Se-CH2-COOH [49];).

A study by Lai and colleagues indicated that methane-driven conversion of selenate to elemental Se is possible [142]. A mixed community of microorganisms operating under anoxic conditions was able to perform methane-dependent reduction of selenate to elemental selenium. This consortium contained a substantial population of genera associated with aerobic methane oxidation, particularly Methylomonas. A more recent study of a methane-driven selenate-reducing community in a membrane biofilm reactor, in which the predominant methanotrophs were Methylocystis, showed that the rate of selenate reduction peaked at an intermediate rate of oxygen supply. These results are consistent with a role of methane monooxygenase in the methanotrophs in oxygen-dependent conversion of methane to methanol as the principal carbon and energy source of the community. Cross-feeding from methanotrophs to non-methanotophs then enables reduction of selenate, via a reaction that is suppressed in the presence of excess oxygen. Consistent with this explanation, as the oxygen delivery rate was decreased to the intermediate level at which the rate of methane-driven selenate reduction was maximised, expression of pMMO genes (pmoA) decreased only 5.4-fold. At the same time, expression of nitrate reductase genes (narG, which may be involved in selenate reduction in non-methanotrophs including Variovorax and Arthrobacter) increased 50-fold [145].

The reduction and methylation of Se species by methanotrophs or consortia offer the possibility for methane-driven remediation and concentration of Se species, as well as producing Se as nanoparticles with novel properties that may be useful in electronics and other technologies, and for other uses such as addition as a micronutrient to selenium-poor foodstuffs.

The reduction of selenite to elemental selenium has been attributed to a range of reductase enzymes and also to non-enzymatic reactions of selenite with thiol-containing molecules such as glutathione [40]. In Mc. capsulatus (Bath), analysis of the low molecular weight selenium-containing compounds during transformation of selenite and Se(0) suggested an extracellular mechanism of selenite reduction with methyl selenol as an intermediate [146]. Glutathione reductase, which has been implicated in the reduction of selenite in microorganisms including Pseudomonas stutzeri [147], appears to be absent from the genome of Mc. capsulatus (Bath), though is distributed among alpha- and gamma-proteobacterial methanotroph genomes (Fig. 3b), being present in at least 49 out of 66 such genomes including that of Ms. trichosporium OB3b, the alphaproteobacterial methanotroph known to reduce selenite to Se (0).

Mercury detoxification by methanotrophs

Mercury (II) or Hg (II), a priority pollutant is released into environment to the tune of 4500–7500 tonnes per year. Around 55% (~1500 tonnes per annum) of global industrial mercury emissions arise from China, India and the USA [28]. Coal-fired industries, especially power generation, have been identified as the major source of mercury pollution alongside gold mining industries (Fig. 4; see Table 3 for case study on the extent of Hg pollution from artisanal/small scale gold mining from Guyana) [29, 148]. In the environment, elemental mercury (Hg(0)) is oxidised to inorganic Hg (II), which can then react with various organic compounds in water and soil sediment by biotic reactions facilitated by bacteria, and abiotic reactions mediated by sunlight photolysis, resulting in conversion into organic mercury such as methylmercury [148]. Methyl mercury (CH3Hg+), which is a more toxic form than Hg (II) or Hg (0), can easily be absorbed by organisms from the lower levels of the food chain and accumulated in higher trophic organisms e.g. through fishes to humans. Methyl mercury can affect the human body negatively, especially the nervous system and is particularly dangerous for pregnant women as the foetus can be affected by mercury passing through the placenta. Damage done to the brains of babies leads to symptoms such as deafness, blindness, microcephally, cerebral palsy and problems with swallowing [28]. Methanotrophs are able to reduce mercury and detoxify methyl mercury [149, 150]. Recently Shi and colleagues [150] reported the presence of all genes required for the reduction of Hg (II) in a metagenome-assembled genome of the alphaproteobacterial methanotroph Methylocystis. Our analysis of representative genomes (Fig. 3b) detected merA genes in most of the gammaproteobacterial

Fig. 4
figure 4

A typical artisanal/small scale gold mining operation in Guyana

methanotrophs (13 out of 44), all of verrucomicrobial methanotrophs (6 out of 6) and in a very few alphaproteobacterial methanotrophs (4 out of 22). We also detected two copies of the merA gene in few methanotrophs such as Methylacidiphilum fumariolicum, Methylacidiphilum kamchatkense and Methylocystis sp. MitZ-2018. However, in other methanotroph genomes, we detected copies of merA homologues i.e. dihydrolipoamide dehydrogenase (DLD) and/or NADPH-glutathione reductase (gorA) genes. GorA, DLD and MerA belong to the family of flavin-dependent disulphide oxidoreductases that are mainly involved in redox reactions [151]. It has been reported that Hg (II) can bind strongly with sulfohydryl groups of proteins [152,153,154,155]. Previously characterised GorA and DLD enzymes are inhibited by Hg (II) upon binding at their sulfohydral binding sites, while MerA is catalytically active in the presence of Hg (II) [156]. This difference in the inhibitory effects of Hg (II) could be mainly due to the structural and redox cycling differences between the GorA or DLD and MerA [156, 157]. The possible role of DLD and GorA homologous proteins on Hg (II) or CH3Hg+ reduction and or detoxification in methanotrophs needs to be investigated.

It has been reported that there are two different mechanisms through which the specific enzymes are deactivated by binding of Hg (II). First mechanism: Hg (II) usually binds with proteins containing thiols (e.g. cysteine) and thioethers (e.g. methionine) causing protein unfolding, aggregation and precipitations [158]. Specifically, the N-termal cysteine residues are critical for Hg(II) binding and transport [156]. Second mechanism: In the absence of cysteine side chains, histidine binds Hg (II) at the constant of log Kf-7.4 [159]. Moreover, Stratton and colleagues [158] reported that the S-ligands offer higher binding constants with Hg (II) (i.e. log Kf -14.4–52.7) in comparison to N-ligands (i.e. log Kf 3.5–8.8).

Methanobactin can bind Hg (II) similar to Cu (II) or Cu (I) (binding with both oxazolone rings at the rate > 2000 s-1) [52]. As discussed above, methanobactins contain cys-thiols groups that readily bind Hg (II). However, the binding ratios differ for methanobactins from Ms. trichospoium OB3b and Methylocystis sp. strain SB2, which might be linked to the presence of different heterocyclic rings and associated enithiols [119]. It has been reported that the binding of Hg:methanobactin ratio were different for Ms. trichosporium OB3b-methanobactin (i.e. 0.1) and Methylocystis sp. strain SB2-methanobactin (i.e. 0.5–6), while the Hg replaced the Cu from Methylocystis sp. strain SB2-methanobactin when added together. Moreover, the binding ratios of methanobactin with different Hg species (e.g. Hg(II), CH3Hg+ or Hg(CN)2) are significantly different for the two methanobactins due to their structural disparity.

Baral and colleagues [119] observed that Hg (II) is reduced (presumably to Hg (0) as inferred from the appearance of a grey colour) by methanobactin from Ms. trichosporium OB3b, but not from Methylocystis sp. strain SB2. Similarly, Hg (0) produced from Hg (II) by Ms. trichosporium OB3b producing methanobactin is not volatile but associated with the biomass at a protein:mercury mass ratio of approximately 2:1 [50]. Methanobactin has been shown to bind with CH3Hg+, particularly with oxazolone rings and not with other heterocyclic ring [119]. Methanotrophs lacking the ability to produce methanobactin are able to take up CH3Hg+ [42, 149], but their accumulation capacity is comparatively lower. It is hypothesised that the methanobactin facilitated uptake of CH3Hg+ (e.g. 100 nM to up to 500 μM) can result in conversion into inorganic Hg by an non-conventional ‘oxidative demethylation’ (i.e. not through the conventional organomercurial lyase encoded by merB) [52, 160]. In contrast to the CH3Hg+ degradation system of methanotrophs, MerB enzymes usually show very poor affinity for CH3Hg+ (Km around 500 μM) and are expressed only under certain environmental conditions such as concentrations of Hg > 1 μM and pH > 7.0 (optimum pH~ 10.2) [42]. Not all methanotrophs degrade CH3Hg+; e.g. Ms. trichosporium OB3b can degrade CH3Hg+, while M. album BG8 cannot. It has been speculated that the CH3Hg+ initially binds with methanobactin and is internalised for MeDH to cleave the C-Hg+ bonds in Ms. trichosporium OB3b, as shown by the inhibition of demethylation of CH3Hg+ when methanol is added. Overall, mechanistic understanding of demethylation characteristics in methanotrophs will allow researchers to develop remediation strategies for contaminated environments. While there is no evidence of Hg (II) methylation by methanotrophs, the role of methanobactin on Hg (II) methylation in the presence of Geobacter sulfurreducens PCA and Desulfovibrio desulfuricans ND132 was recently reported [120]. Since there are no studies reporting the specific binding proteins from methanotrophs other than methanobactin regulating Hg (II) and CH3Hg+ toxicity, there is a need to characterise methanotroph cell surface proteins with sulfohydral groups to understand their possible role in Hg (II) and CH3Hg+ binding and transport [155].

Factors affecting metal transformation by methanotrophs

A number of abiotic factors can affect the passive or active metal uptake systems and transformation by methanotrophs, among which pH, temperature, available oxygen concentrations, carbon source (e.g. CH4) and other metal(loid)s coordination play a critical role. Metal affinity of methanobactin is influenced by pH (Fig. 5 [119];) along with metal to methanotroph biomass (and methanobactin) ratio. The binding rate and affinity can alter with the molar ratio of metal to methanobactin (i.e. Cu and methanobactin). If the ratio is 0.5 or above, methanobactin binds as a monomer and at lower ratios as tetramer or oligomer. In the presence of other metals, Cu will be the preferred metal ion that will be readily taken up by the methanotrophs, while least preferred is iron. In presence of Ag (I) or Au (II), the copper uptake will be limited. Our understanding of the physiology of methanotrophs, particularly the regulation of genes (e.g. TonB, mbnABCM, arsRBC, merABCD, etc.) for metal uptake and/or transformations in relation to concentrations of Cu and other metals is limited. Uptake of Zn can be affected by Hg (II) or Cd, which are thiophilic in nature and bind to the cysteine in methanobactin. Temperature influences the solubility of methane in solution and thereby influences methanotrophic growth, which in turn may significantly influence metal ion uptake. However, there is no detailed understanding on how metal affinity for methanobactin and other methanotroph metal uptake systems varies at different temperatures. While methanotrophs with no capacity to make methanobactin can also uptake and accumulate metals, studies on these methanotrophs are also very limited. Pure cultures and mixed methanotrophs cultures with or without heterotrophs (that mimic natural conditions) may also behave differently in metal uptake and transformations and future research is required to understand metal transformations in near in situ conditions. Lai and colleagues [161] found that the available nitrate had significant impact on bromate reduction, while they also correlated it with polyhydroxyalkanoate (PHA) accumulation capacity. However, not all the strains can make PHA as storage material.

Fig. 5
figure 5

Order of metal affinity for methanobactin under different pH conditions (red arrow indicate the order of change in metal afficinity with respect to pH change)

A range of biotic interactions (e.g. methanotroph—heterotroph interactions; Fig. 6) can regulate methanotroph distribution, activity, metal uptake and transformation [71, 162, 163]. Both synergistic and antagonistic biotic interactions are known to impact methanotroph functional diversity and have been extensively reviewed in [162]. It has been well-established that exchange of metabolites between methanotrophs and heterotrophs improve methanotroph growth and activity [163, 164]. Methanotroph-heterotroph interactions are also constrainted by various abiotic factors. For instance, the ratio of CH4 to O2 altered the methanotroph-heterotroph community structure in the enrichment. In particular, PHB accumulating alphaproteobacterial methanotrophs dominanted with increasing CH4 content [165, 166]. Moreover, differences in Cu to Fe ratio were also found to impact community composition [167]. Our understanding on the role of biotic interactions on metal transformation by methanotrophs is currently limited. Given the potential role of methanotrophs in bioremediation strategies, there is an immediate need to explore how community level biotic interactions impact metal transformations and uptake.

Fig. 6
figure 6

Schematic representation of potential biotic and abiotic interactions that constrains methanotroph-dependent metal(loid) transformation in the environment. N – nitrogen, P – phosphorus, K – potassium, C – carbon, Cu – copper, Fe – iron, Ni – nickel, Cr – chromium, Se – selenium, Hg – mercury, Zn – zinc, Co – Cobalt, Na – sodium, Cl – chlorine and F – flourine

Conclusions and future considerations

Manufacturing and resources industries are key drivers for economic growth, yet this comes at a huge environmental cost affecting not only ecosystem services but also the livelihood of local communities. Aerobic methanotrophs are metabolically versatile and are able to detoxify toxic heavy metals such as chromium and mercury while growing on a cheap feedstock i.e. methane. In order to fully exploit these traits for mitigation of polluted sites, future research is required to better understand (i) physiological and genetic basis of Cr(VI) reduction and demethylation of mercury, (ii) mechanism of Cu reduction by methanobactin, (iii) role of porins in passive uptake of metals in methanotrophs, and (iv) role of molybdenum and tungsten in formate dehydrogenase activity. More importantly, our knowledge of metal transformations by methanotrophs is based to a large extent on laboratory strains. Further research is required to understand metal uptake/transformation mechanisms in the environment, particularly in polluted sites with elevated and/or multiple metal concentrations.