Many essential biological processes involve metal ions, and in particular, redox active transition metals like manganese, iron, cobalt, copper and magnesium. For instance, divalent ions such as Mn²⁺ and Fe²⁺, are indispensable for oxidative stress resistance or as cofactors for the respiratory chain in bacterial core metabolism (Palmer and Skaar, 2016). Hence, microorganisms have evolved pathways of metal sequestration to transport ions and to regulate ion availability. In the simplest case, the sequestration of metal ions occurs by adsorption. In Figure 2, metal ions are depicted as positively charged entities interacting with the bacterial membrane and other biomolecules. During adsorption, deprotonated functional groups (carboxyl, phosphonate, amine, and hydroxyl groups) on the cell surface result in a net negative charge which attracts metal cations and leads to non-specific binding of the metal to the cell surface (Saravanan et al., 2021). This process is independent of the cell metabolism, and it is mainly influenced by factors such as temperature, pH, ionic strength, concentration and by the complex composition of the microorganism’s surface. As an example, Gram-positive bacteria expose a large amounts of carboxyl groups in their peptidoglycan-rich cell walls, making them efficient metal chelators (Pham et al., 2022). As a consequence of this sequestration, these metals are enriched on the cell surface and can be taken up by specific transport processes (discussed below).

However, the extracellular sequestration of heavy metals not only allows to accumulate relevant metabolic cofactors, it is also a defense mechanism against toxic metals, by reducing the availability of unwanted metal ions. Polymers secreted by bacteria such as the exopolysaccharide (EPS) coatings of some species, have the ability to adsorb and bind metals extracellularly, thus acting as a protective layer that prevents the uptake of toxic metals (Bruins et al., 2000). Bacteria are likely to sequester non-essential metal ions, such as Au and Pd, by adsorption mechanisms. Compared to other microorganisms, Gram-negative bacteria such as Acinetobacter calcoaceticus, Erwinia herbicola, Pseudomonas maltophilia and Pseudomonas aeruginosa have the highest ability to accumulate metals such as gold (Tsuruta, 2004). This is a result of the surface charge present on the Gram-negative bacterial membrane, which promotes the gold adsorption from aqueous solutions. In the same way, the biosorption of Pd is presumably a result of the chemical affinity of metal complexes such as [PdCl₄]^2- for protonated groups on the cell surface at an acidic pH (Deplanche et al., 2010).

Free metal ions passively or actively transported from outside of the membrane into the cytoplasm can pose a danger to the cell and must be immobilized or transformed. Bacteria can produce metabolites, like metallothioneins, phytochelatins and glutathione with high affinity for metals for the intracellular sequestration of metal ions. Metallothioneins are small, cysteine-rich proteins that are directly involved in the homeostasis of different metal ions, including Cu⁺ and Zn²⁺. These proteins have been reported in bacteria such as Bacillus altitudinis MT422188 and Mycobacterium tuberculosis where they are expressed under Zn²⁺ and Cu⁺ induced stress, respectively (Johnstone and Nolan, 2015; Khan et al., 2022). The immobilization of the ions decreases the toxicity; however, the metal sequestration mechanisms are limited by the saturation of the binding sites in the extracellular and intracellular matrix (Bruins et al., 2000).

The metal ions in proximity to the extracellular layers can be imported by passive or active influx through the different layers of the cell wall, using non-specific transporters, metal-specific channels or passive diffusion (Figure 2) (Bruins et al., 2000). The indispensable uptake of essential metals is highly regulated and typically mediated by specific, energy-dependent uptake systems. In E. coli, Alcaligenes eutrophus and many other enterobacteria, essential trace elements such as Ni²⁺, Co²⁺, and Zn²⁺ ions are transported into the cell by constitutive Mg²⁺ transport systems (Nies, 1992; Komeda et al., 1997). These relatively unspecific uptake systems allow the “cost-effective” accumulation of macronutrients such as Mg and of trace metal ions in environments with habitual metal concentrations. In extreme conditions, with deficits in trace metals, bacteria can upregulate genes related to specific, ATP-dependent uptake systems for these elements in order to transport the ions against the concentration gradient. The TonB-dependent transporter, OprC, is repressed in P. aeruginosa by high exogenous Cu ion concentrations and expressed in anaerobic conditions to bind and transport Cu^+/2+ (Bhamidimarri et al., 2021). In environments with low availability of essential metals, many bacteria also secrete metal-sequestering proteins or other chelators to bind the biologically relevant ions and to specifically import them (Palmer and Skaar, 2016). An example of this mechanism are iron siderophores, which are low molecular weight compounds known for their high affinity for Fe ions. Siderophores are frequently used by pathogenic bacteria to chelate Fe²⁺ and Fe³⁺ in host environments that are otherwise deprived of free Fe ions (Johnstone and Nolan, 2015). The siderophores with bound ions are then recognized by specific outer membrane receptors that guide the chelated ions through energy-dependent ATP-binding cassette (ABC) transporters (Krewulak and Vogel, 2008; Ma et al., 2009). Because of the different chemical properties of Fe³⁺ and Fe²⁺, bacteria utilize different specific transport systems (and siderophores) for each ion (Lau et al., 2016; Crespo et al., 2017). Examples of different transport systems involved in bacterial Fe uptake, include MntH, ZupT, YfeABCD, FutABC, EfeUOB and Feo (Lau et al., 2016).

The uptake of precious metals like Au, Ag, Pt, and Pd ions into bacteria is not well understood, as cells do not possess inherent active, specific transport mechanisms to regulate non-essential ion uptake. The absorption of toxic heavy metals is presumably a process carried out by passive mechanisms such as diffusion. Therefore, at elevated concentrations, non-essential metals can interact with the cell, cross into the cytoplasm and accumulate there (Lusa et al., 2016). However, some authors suggest that heavy metals like Pd may also accidentally be absorbed and imported into living cells through systems specific for essential metallic metabolites, such as Fe, Ni, Cu or Zn (Omajali et al., 2015). Metal ions with comparable properties and ion radius, due to their similar conformations, can out-compete essential ions in binding to proteins. As an example, the high affinity of Hg and Cd ions to cysteine-rich sites in Zn-binding proteins results in translocation of these toxic metals. A similar effect is found in Cu-binding proteins with Au and Ag as competing metal ions (Waldron and Robinson, 2009). A special case of non-essential metal uptake is the homeostasis mechanism of specialized metal-resistant bacterial strains for some toxic heavy metal ions. In the case of Cupriavidus metallidurans, genes in the gig (gold-induced genes) operon are strongly upregulated after sequestration of Au ions in the membrane. The proteins encoded by this operon are suggested to be directly involved in the import of Au ions into the cytoplasm to prevent the harmful action of Au³⁺ in the periplasm. The reduced complexes there are later removed from the cytoplasm (Wiesemann et al., 2013; Zammit and Reith, 2013).

If not regulated, high concentrations of metals would have adverse effects in the metabolic processes of the cell. In metals such as Cu, their redox activity favors the formation of active oxygen species, which explains their toxicity. Excess of essential metals or toxic, non-essential heavy metals also leads to mismetallation of proteins which results in their inactivation (Andrei et al., 2020). The exposure to high concentrations of heavy metals activates a mechanism of defense called metal exclusion, in which the permeability of the cell membrane, the envelope, or the surface layer is modified to prevent or reduce the further influx of metal ions (Figure 2). Bacteria can alter the composition of the cell wall in an attempt to form a barrier for ions. A well-known example of this mechanism is the reduction of the expression of Porin channels in heavy-metal resistant mutants of E. coli. In this case, pore proteins such as OmpF are expressed at a reduced rate, leading to increased Cu and Ag resistance (Lutkenhaus, 1977; Rouch et al., 1995; Li et al., 1997). In a complementary mechanism, the active efflux of metal ions involves the use of highly specific membrane transport systems. This process to decrease the concentration of harmful metal ions inside the cell is wide-spread among bacteria. The transport of cations and anions from the cytoplasm against a concentration gradient requires energy. The involved efflux pumps can be broadly classified by their energy source into ATP-dependent and proton-gradient-based systems. Further classifications are based on sequence similarity, transport function and substrate specificity (Nies and Silver, 1995; Hynninen, 2010; Delmar et al., 2014).

Some examples of well-studied heavy metal efflux pumps are the P-type ATPase transport protein CopA, known to be involved in copper resistance (Rensing et al., 2000) and the resistance-nodulation-cell division (RND) family proteins, CzcA and CusA, which bind and export Cu⁺ and Ag⁺ in E. coli (Kim et al., 2011). In partly silver-resistant E. coli bacteria, the detoxification mechanism for silver is based on the expression of the CusCFBA copper/silver efflux systems that actively export the silver ions from the cytoplasm and bind it in the periplasm (Lin et al., 2014). In some bacteria, metal ions transported by efflux pumps may be retained in the periplasm even after the efflux process, along with the metal ions sequestered by EPSs and by other metal-binding proteins (Blindauer, 2011). The ions bioaccumulated in the cytoplasm, periplasm and in the extracellular matrix can later become sites for nucleation and formation of nanoparticles (Figure 2).

Some microorganisms have a remarkable ability to transform heavy metal ions in their environment and to produce MNPs. Even when sequestered, the high reactivity of transition metal ions could disrupt metabolic processes or damage DNA molecules. Various metal ions have a high binding affinity for biological macromolecules such as nucleic acids, or thiol-containing proteins (Delmar et al., 2014; Boedicker et al., 2021), and therefore, they need to be transformed into inert or less toxic substances. This is mediated by redox-active metabolic pathways in order to change solubility and toxicity, and often leads to complexation or precipitation of metals extracellularly (Silver, 1996; Bruins et al., 2000; Hood and Skaar, 2012). Alternatively, biomethylation is a metabolic process in which methyl groups are transferred to the toxic metal ions or metalloids to create less toxic, sometimes volatile compounds. Hg, Sn, As, Se, Te, Au, Tl, and Pb have been reported as methyl group acceptors in primary and secondary metabolic processes (Kosolapov et al., 2004). The resulting products are more toxic than the inorganic compounds in many cases, so it is debatable whether methylation is, in fact, a detoxification mechanism. However, it is important to note that the volatility has an important effect on reducing the concentration of the metals in the surrounding environment. Extended descriptions of the general pathways for biomethylation can be found elsewhere (Ridley et al., 1977a; Ridley et al., 1977b; Fatoki, 1997).

Some bacteria can decrease the mobility of metal ions through enzymatically promoted redox reactions that reduce the metals irreversibly, leading to the assembly into metal clusters. An illustration of the general process of reduction of the metal ions to neutral ions by redox proteins intra- and extracellularly can be found in Figure 2. The electron-accepting potential of different metal ions known to be reduced by bacteria are shown in Table 1. The reduction potential is expressed as the E ₀ value, ranging from very negative E ₀ values for compounds that are easily oxidized, to large positive E ₀ values for compounds that readily accept electrons (are easily reduced). “Precious” metals such as Au, Pd or Ag have cations species with very positive potentials and are frequently reduced into nanoparticles by bacteria.

The cornerstone of every living organism is energy conservation where chemical or light energy is ultimately converted into adenosine triphosphate (ATP). ATP production is facilitated by substrate-level phosphorylation, oxidative phosphorylation, or photophosphorylation. In order to achieve this, electrons need to be transferred in an oxidation-reduction coupled reaction. This electron movement within the cell is facilitated by electron carriers, such as nicotinamide adenine dinucleotide (NAD⁺/NADH), which promotes a diversity of reduction-oxidation reactions as they allow a variety of electron partners to interact with each other (Muller, 2003; Herrmann et al., 2008; Schuchmann and Müller, 2012). It is these metabolic reactions and electron carriers that are diverted and used for metal ion reduction.

On an extracellular level, bacterial growth is heavily influenced by the oxidation-reduction (redox) potential of the environment, and in turn, bacteria can modify the redox potential of their environment. Surprisingly, this ability is highly species-specific and can be exploited to identify different species by using redox electrodes (Reichart et al., 2007). A change in redox potential indicates the availability of electron acceptors and donors. Since bacteria can inhabit every niche that can support life, their diversity is also manifested in the type of electron acceptors that they can utilize intracellularly. Under aerobic respiration conditions, oxygen is the final electron acceptor while under anaerobic respiration conditions, a variety of electron acceptors can be utilized, such as nitrate (NO^3-), trimethylamine oxide/dimethyl sulfoxide (TMAO/DMSO), carbon dioxide (CO₂) or organic electron acceptors such as fumarate, depending on the bacterial species studied (Kröger et al., 1992; Vincent et al., 2021). Metal-based respiration is characterized by the utilization of manganic manganese (Mn⁴⁺) or ferric iron (Fe³⁺) as electron acceptors. Bacteria can reduce these metal compounds through mechanisms that involve redox-active proteins like reductases, cytochromes, and metallothioneins (Bloch et al., 2021).

Sulfate- and metal-reducing bacteria are frequently employed for the synthesis of metal nanoparticles, as these organisms are rich in membrane redox-active proteins and therefore easily enzymatically reduce different metal ions. These ‘reducing’ microorganisms can use the metal cations as terminal electron acceptors while performing anaerobic respiration, reducing Fe³⁺ to Fe²⁺ and Mn⁴⁺ to Mn²⁺ for energy conservation (Lloyd, 2003). These proteins however, are not necessarily highly specific and multiple enzymes are capable of reducing several redox-active metals ions (Barton and Fauque, 2009). The interaction of some bacteria with non-essential, toxic metal ions such as Au³⁺, Ag⁺, Pd²⁺ and Pt²⁺ also upregulates the expression of multiple redox-active enzymes located on the cell walls, in periplasmic space and intracellular contents, such as hydrogenases and NADH reductases, which can reduce ions to insoluble complexes that are subsequently transformed into MNPs (Omajali et al., 2015; Chen et al., 2019). One example is nitrate reductase known to be involved in the bioreduction of the bound Ag⁺ to Ag⁰ in Bacillus licheniformis as defense from heavy metals exposure (Kalimuthu et al., 2008; Rajora et al., 2016). As suggested by their negative reduction potential (Table 1), the NAD(P)H-dependent reduction of ions such as Cd²⁺, Zn²⁺, Co²⁺ or Ni²⁺ to metallic form may not be energetically favored, so the reduction of these ions is uncommon in bacteria and the particular cases of reduction must involve different mechanisms (Nies, 1992). In Shewanella strains, EPS-induced reduction and indirect electron transfer using electron shuttles such as flavins, L-cysteine, and quinones are possible alternative pathways that have been thoroughly described elsewhere (Yang et al., 2023). Metal NPs reduced and precipitated in the extracellular and intracellular space may possess different characteristics than their chemically produced counterparts. The biomolecules associated with them create a biological cap that stops nanoparticle growth, minimizes cytotoxicity, and may affect their chemical and physical properties (Bulgarini et al., 2021). The reduced metal particles accumulate without much effect on the cell population. Therefore, the formation of MNPs can be regarded as a by-product of microbial defense mechanisms, of bacterial metabolism, or as a biomineralization process intended to control, immobilize, or decrease the heavy metal bioavailability.

Even though naturally occurring nanoparticles are widely found in living organisms, the intracellular accumulation of solid metals is usually not a favorable condition for the cells and can lead to cell death. A complete metal particle detoxification would require energy and elaborated mechanisms of particle excretion (Griffin et al., 2017). This is probably the reason why only few bacterial species produce metal-based nanoparticles as part of their normal metabolic processes. The most remarkable example are magnetotactic bacteria, an extraordinary group of Gram-negative organisms that rely on biosynthetic iron-based crystals to orient and migrate along the geomagnetic field lines in a behavior known as “magnetotaxis” (Yan et al., 2012). The highly specialized organelles made of nanometer-sized metal crystals wrapped in a membrane are called magnetosomes. They are composed of magnetic iron oxide magnetite (Fe₃O₄) or iron sulfide greigite (Fe₃S₄) of different shapes and sizes depending on the exact bacterial species. The magnetosomes are frequently aligned in chains across the motility axis of the cell. This imparts the bacteria with a permanent magnetic dipole moment, causing them to align passively in a parallel orientation with external magnetic fields (Monteil and Lefevre, 2020). The cells swim along the geomagnetic field lines of the Earth to the oxic-anoxic interface of the aquatic habitat where are the most favorable conditions for growth (Le Nagard et al., 2018).

Despite the abundance of magnetotactic bacteria in the environment, only a few strains have been isolated into pure culture due to their particular growth requirements. Strains from the proteobacterial genus Magnetospirillum are the most common isolates. However, recent publications report up to 16 different lineages that include Proteobacteria, Nitrospirota, and Omnitrophota (Gareev et al., 2021; Goswami et al., 2022). The strains Magnetospirillum gryphiswaldense MSR-1 [first isolated in 1991 by Schleifer et al. (Schleifer et al., 1991)], Magnetospirillum magnetotacticum strain MS‐1 and Magnetospirillum magneticum AMB-1, are currently the most studied bacteria for the understanding of magnetosome formation and its complex genetic regulation (Nudelman and Zarivach, 2014). In M. gryphiswaldense MSR-1, the magnetosome island (MAI) comprises most of the genes that control the magnetosome synthesis. These genes encode all the essential processes in the majors steps of the organelle formation and determine the morphology and chemical composition of the particles (Uzun et al., 2020; Dziuba et al., 2021). In the past, different models of magnetosome biomineralization have been proposed (Schuler, 2002; Yan et al., 2012; Nudelman and Zarivach, 2014). Currently, four major steps are considered necessary to achieve the biomineralization. The first step is i) the cytoplasmic membrane invagination into vesicles, followed by ii) the arrangement of specific magnetosome proteins into the organelle membrane. Then, iii) the internalization and accumulation of iron into the vesicles by energy-dependent transport proteins takes place. Here, the cell rapidly chemically transforms the Fe ions and combines them into metal crystals. Specific proteins guide this process in order to avoid the toxic effect of intracellular Fe. Finally, iv) the magnetosomes are assembled into a linear chain along cytoskeletal filaments (Dziuba et al., 2021).

The bacteria tightly controls the size and morphology of the magnetosomes. Magnetosome crystals typically fall in a size distribution of about 30–120 nm. However, some species can form particles up to 250 nm in length (Yan et al., 2012). The spirillum-shaped bacterium M. magneticum AMB-1, originally collected from fresh water sediments of a natural spring in Tokyo, presents magnetite magnetosomes with an average diameter of 50 nm, aligned in chains of around 15 particles per cell (Matsunaga et al., 1991). In contrast, the M. gryphiswaldense MSR-1 bacteria were found to produce chains of up to 60 homogeneous magnetosomes with particle sizes of around 42 nm (Schuler, 2002). Both strains, like other bacteria from the Magnetospirillum genus, produce cubo-octahedral crystalline magnetosomes. In general, three different morphologies of magnetosomes can be found in nature: i) cubic or cubo-octahedral, ii) elongated hexa- or octahedral and iii) even more elongated crystals with large, anisotropic faces, showing tooth, arrowhead, or bullet shapes (Cypriano et al., 2022). These crystal structures are consistently replicated in the magnetosomes of bacteria of the same strain and have a direct influence on the final magnetic behavior of the particles.

The magnetosomes, due to their narrow size distribution, low aggregation, and strong ferromagnetism, have many potential technical applications. The low toxicity of bacterial magnetosomes in comparison with synthetic magnetic nanoparticles makes them especially attractive for biomedical applications. Cytotoxicity of bacterial magnetosomes varies depending on various factors, including cell type, incubation time, and concentration. While many types of magnetite nanoparticles [such as Superparamagnetic iron oxide nanoparticles (SPION)] are considered cytotoxic at exposure levels above 70–100 μg/ml (Alphandéry, 2014; Singh et al., 2010; Toropova et al., 2017), a study by Alphandéry et al. in MDA-MB-231 cells suggested low cytotoxicity due to magnetosome exposure at concentrations below 1,000 μg/ml which can be further improved by removing bacterial endotoxins and using biocoatings (Alphandéry et al., 2011; Gareev et al., 2021). In a recent study, Gwisai et al. evaluated the use of the ferromagnetic properties of AMB-1 bacterial magnetosomes for targeted cancer therapy. Thanks to their susceptibility to external magnetic fields, magnetotactic bacterial cells were magnetically manipulated and studied as a vehicle for tumor infiltration and drug delivery (Gwisai et al., 2022). In a similar approach, Xing et al. successfully demonstrated the colonization and ablation of tumors in mice by means of in vivo magnetically manipulated M. magneticum AMB-1 bacteria, showing their enormous potential for an efficient cancer treatment (Xing et al., 2021). Other potential application of magnetosomes include their use as magnetic resonance contrast agents. In vivo experiments were conducted on mice to examine how magnetosomes were distributed and eliminated after injection as a contrast agent in magnetic resonance imaging (MRI). The researchers concluded that their use in MRI was adequate in terms of spatial resolution and sensitivity (Nan et al., 2021).

Microorganisms play an important role in the process of biomineralization and metal geochemistry. The metals transformation into nanoparticles is also a natural process observed in some astounding microorganisms known as dissimilatory metal-reducing bacteria. In anoxic environments with abundant metal species such as Fe³⁺, Mn^3+/4+, U⁶⁺, Cr⁶⁺, Co³⁺ and Tc⁷⁺, these metal cations are reduced into nanoparticles by the bacteria (Lovley, 1993; Shi et al., 2016). All these elements have in common that they are good electron acceptors due to their high reduction potential. Then, during anaerobic respiration, the bacteria can replace oxygen as the terminal electron acceptor with the available cations. Fe³⁺ is the most abundant possible electron acceptor in anoxic environments. Due to its comparatively high reactivity with oxygen, elemental Fe is not frequently found in nature, but instead, is commonly found in the form of iron oxides. In metal- and sulfur-reducing microorganisms like, Geobacter sulfurreducens, Geobacter metallireducens GS15 and Shewanella oneidensis MR1, specialized membrane proteins are suggested to be involved in the Fe³⁺ reduction. Their outer membranes are rich in c-type cytochromes that transfer electrons coupled to the oxidation of lactate and other carbon sources to soluble Fe³⁺ oxides resulting in iron oxide NPs (Lovley et al., 1993; Mehta et al., 2005; Shi et al., 2012). Many Fe-based NPs naturally synthesized by bacteria consist of diverse compositions of iron oxides. The metal-based anaerobic mechanism of respiration is considered part of the natural cycle of precipitation of many heavy metals into anoxic sediments (Lovley, 1993). Multiple processes of bioremediation of heavy metals from the environment and contaminated water streams have been engineered based on the reducing ability of these bacteria (Iravani and Varma, 2020). Other transition metals such as gold, silver, and copper have high positive reduction potentials and are exchanged with iron. Bacteria readily precipitate them during remediation, which can then be applied to the biogenic production of these metal nanoparticles.

Bio-inspired NPs of different metals have been successfully synthesized using a large variety of bacterial strains. Under optimal conditions in temperature, pH and redox state, some bacterial species can reduce metal ions and form nanoparticles. Among the metals successfully recovered by bacteria in form of nanoparticles are iron, manganese, chromium, cobalt, palladium, gold, silver, arsenic, selenium, uranium, and polonium (Boedicker et al., 2021). Their resulting size-distribution, morphology and functional properties are highly dependent on the protocols used for the biosynthesis as well as on the type of bacteria and metal involved. In Table 2, details of recent, published protocols for metal nanoparticles fabrication using different bacterial strains are listed.