Microbiological Reduction of Molybdenum to Molybdenum Blue as a Sustainable Remediation Tool for Molybdenum: A Comprehensive Review

Molybdenum (Mo) microbial bioreduction is a phenomenon that is beginning to be recognized globally as a tool for the remediation of molybdenum toxicity. Molybdenum toxicity continues to be demonstrated in many animal models of spermatogenesis and oogenesis, particularly those of ruminants. The phenomenon has been reported for more than 100 years without a clear understanding of the reduction mechanism, indicating a clear gap in the scientific knowledge. This knowledge is not just fundamentally important—it is specifically important in applications for bioremediation measures and the sustainable recovery of metal from industrial or mine effluent. To date, about 52 molybdenum-reducing bacteria have been isolated globally. An increasing number of reports have also been published regarding the assimilation of other xenobiotics. This phenomenon is likely to be observed in current and future events in which the remediation of xenobiotics requires microorganisms capable of degrading or transforming multi-xenobiotics. This review aimed to comprehensively catalogue all of the characterizations of molybdenum-reducing microorganisms to date and identify future opportunities and improvements.


Introduction
Human activities are endangering the environment. Due to the growth of the population, intensive industrialization, in addition to urbanization and agriculture, has caused significant damage to the environment. The overexploitation of natural resources and ignorance of the laws of nature have exacerbated these problems. Pollution caused by hydrocarbons and metal ions has increased in recent years [1,2], representing a significant global problem.
Toxicity cases ranging from acute to chronic, within occupational and environmental high-exposure settings, have been identified to be caused by toxic agents from metals and their compounds. Typically, heavy metals naturally exist in the environment. Since the pre-industrial period, the level of heavy metals has significantly increased due to anthropogenic activities [3][4][5]. A considerable and indiscriminate release of pollutants into the environment has occurred in parallel with the growing population and intensity of industrialization [6,7]. Harmful effects on human health and biota can be exerted when heavy metals reach levels above the critical load [8][9][10][11]. In their elemental forms and different combinations, metals such arsenic, cadmium, chromium, cobalt, copper, lead,

Molybdenum Toxicity
Acute or chronic exposure can be used to evaluate the effects caused by molybdenum compounds. Data obtained on human toxicity as a result of chronic exposure are currently limited, and no relevant data are currently available on the acute toxicity of molybdenum compounds. Thus, the toxicity of these compounds needs to be assessed for different species of organisms because most of the chemicals are specified to their target and the effects may differ. The existence of numerous chemical forms of the molybdenum compounds should also be considered. The toxicity of molybdenum compounds in different animal models has been reported in several studies. Hence, it is important to identify the toxicity to spermatogenesis and oogenesis, particularly those of ruminants.

Toxicity to Spermatogenesis and Oogenesis
In one study, the effects of dietary molybdenum on pup growth during lactation and the reproductive ability of Long-Evans rats fed for 13 weeks with diets containing a graded dose of molybdenum (0.1, 2, 8, or 14 mg/kg/day) and 5-20 mg/kg copper showed noticeable infertility among males due to the degeneration of the seminiferous tubules. Furthermore, in the groups that received the two highest doses, more pronounced effects were seen. Lactating mothers exposed to the two highest doses and their pups experienced lower weight-gain during lactation compared with those in the lower-dose groups [38]. The oral gavage of sodium molybdate with a dose greater than 30 mg/kg body weight [39] has been reported to cause a dose-dependent degeneration in testicular morphology and its function. This was observed from the decline in sperm motility, concentration, normal morphology and epididymides of the tested rats. Therefore, molybdenum can affect sperm quality by modulating complex testicular oxidative stress processes. For rats undergoing treatment with 12 mg/kg/day of tetrathiomolybdate for sixty days, a significant decrease in their epididymal weight, sperm motility, count, and morphologic abnormalities, in addition to histopathological alterations in the epididymis and testes, were observed [40]. Another report revealed similar results in which a decrease in the number of germ cells and mature spermatocytes was found in the testes of rabbits fed with carrots containing 39 mg Mo/kg dry weight, in addition to the appearance of degenerated cells and a large number of syncytial giant cells among the spermatogenic cells in the seminiferous tubules relative to the control [41]. The usage of doses of molybdenum greater than 100 mg/L resulted in a negative effect on mice testes (complete lack of libido and sterility), accompanied by alterations in the levels of malondialdehyde (MDA), superoxide dismutase (SOD), and glutathione peroxidase (GPx) [18].

Mechanism of Molybdenum Toxicity in Ruminants
Molybdenum endorses cellular function by catalyzing a variety of hydroxylation and redox transfer reactions [18,19] that are essential within animal and plant physiology [20]. Molybdenum excretion for ruminants such as cattle, sheep, and horses is generally shared between faces and urine because the absorption from the gastrointestinal tract is incomplete [27,34,37]. In contrast, the intake and excretion of molybdenum for most nonruminants, including humans, are usually balanced [36]. Copper-deficient animals are likely to be more susceptible to molybdenum toxicity than those on copper-adequate diets, as seen in the interaction between molybdenum, copper, and sulfate. Dietary sulfate can be administered for non-ruminants to avoid molybdenum poisoning, whereas dietary sulfate administered to copper-deficient animals could worsen the situation [27,34,37]. For ruminants, a molybdenum concentration of 10 mg/kg of body weight will result in tissue copper depletion, an effect that is enhanced by dietary sulfate [27,42].
One of the most important biological interrelationships between animal nutrition and medicine can be seen in the complex interaction between molybdenum, copper, and sulfate metabolisms regarding the kinetics, bioavailability, and utilization of copper. This interrelationship, however, is not yet fully understood. Among ruminants, different molybdenum compounds react with sulfides to form thiomolybdate compounds that then further react with copper to form an insoluble complex that is poorly absorbed. A decreased copper uptake impairs its utilization and results in the synthesis of several copper-dependent proteins [26,33,37,43]. Thus, a secondary copper deficiency is triggered by a reduced copper bioavailability. Copper excretion increases with the excessive intake of molybdenum. The administration of tetrathiomolybdate may aid in the treatment of chronic copper poisoning. The tolerance of monogastric animals to molybdenum toxicity is usually due to the inadequate production of thiomolybdates by the intestinal flora [34,37]. In addition, the functions of several enzymes involved in collagen and elastin stability, as well as their maintenance, can be impaired due to chronic exposure to molybdenum [44]. Therefore, this may be associated with the risks of cardiovascular disorders. Molybdenum exposure may also reduce phospholipid synthesis in nerve tissues within a similar vein, resulting in demyelination and neurologic abnormalities [27,37].
A primary manifestation of molybdenum toxicity can be presented in impaired copper utilization and metabolism resulting from a secondary copper deficiency (molybdenosis) [26,45,46]. This usually can be seen in at the scale of the herd, with a morbidity of up to 80% in cattle. Most cattle present with symptoms of persistent diarrhea associated with green liquid feces, which is often referred to as teart scours. In addition, the depigmentation of the coat is typically noticed among black animals as a consequence of impaired tyrosinase activity and reduced melanin synthesis [44]. A decline in bone mineralization due to the competition of molybdenum with phosphorous utilization causes microcytic hypochromic anemia, joint pain with associated lameness, and osteoporosis. Regarding heifers, decreased fertility, milk production, and weight loss at puberty are typically seen, whereas among sheep (particularly lambs less than 30 days old), enzootic ataxia or sway back syndrome present as stiffness of the legs and back, respectively, and difficulty standing is seen [47]. Within one-to-two weeks of chronic exposure to molybdenum, the affected animals usually exhibit the abnormal development of connective tissue and growth. Because acute toxicity may occasionally be encountered in cattle or sheep, the symptoms of anorexia and lethargy may be witnessed within three days, and deaths can commence within the first week [27,45].

Bioremediation
In recent years, the bacterial-based remediation of environmental pollution has drawn significant interest as a promising innovative technology [48][49][50]. In order to abolish, attenuate, or transform pollutants into less hazardous products in combination with carbon dioxide, water, inorganic salts, and microbial biomass production, the unique catabolic property of microorganisms has been employed [51]. However, inorganic toxicants, e.g., heavy metals, are not degraded by microorganisms. Therefore, biologically encoded changes in the redox state and active metabolic capacity of the microbe are needed. In general, heavy metals are essentially needed by most microorganisms for life processes to occur. However, the adsorption capacity depends on both total microbial biomass and the system's geochemistry. A number of metals' oxyanions do not interact with microorganisms via assimilation, and so their bioremoval is focused on enzyme-catalyzed redox conversion to their precipitable forms [7].

Molybdenum Pollution
The diversity of molybdenum, which is used in superalloys, nickel-base alloys, lubricants, additives, glassworks, paint, pigments, electronics, and many other uses, has been proven to be valuable. Molybdenum can be present in the discharged effluent associated with these products [52]. Due to decades of heavy industrialization, heavy metal pollution in Tokyo Bay and the Black Sea, including that of molybdenum, has reached concerning levels [52]. In Tyrol, Austria, molybdenum in the soil used to feed ruminants can exceed 200 ppm, which resulted from decades of industrial exhaust discharge and has led to deaths and scouring [53]. A light-bulb plant built in the early 1970s in China has resulted in soil molybdenum pollution of 0.25-252 mg/kg [21]. Unusually high concentrations of Mo and Nb were found in the atmosphere of Islamabad, Pakistan, and the elevated levels of these elements were suggested to have originated from anthropogenic activities [54]. In Jordan, areas exposed to extensive oil-shale exploration and intensive agricultural activities have been found to be subject to molybdenum pollution of 11.7 mg/kg of soil [55].
In addition to industrial pollution, mining sites are a major cause of molybdenum pollution. Molybdenum is the main dissolved metal in the aqueous effluent from many uranium flotation mills, probably due to the high solubility of the molybdenum metal in the form of molybdate ions at the high pH values that are used, whereas most other metals are precipitated as hydroxides. For instance, in New Mexico, molybdenum and other toxic metals released from the Molycorp molybdenum mine resulted in levels of these metals exceeding the New Mexico environmental standard in the Red River and caused the extinction of many of the river's floras and fauna [56]. A complex tailings dam was found to contain molybdenum in the local region of the Miduk Copper Complex copper mine in Iran, where molybdenum is a byproduct [57]. Geochemical approaches have been used to analyze the speciation and levels of Mo, in addition to Zn, Cu, and Fe, in Nver River sediments polluted by Mo mining in western Liaoning, Northeast China. Metal characterization studied using X-ray fluorescence and X-ray diffraction has shown that the quantity of Mo approaches 1127.9 times the concentration of Mo in the Earth's crust. This analysis revealed that Mo tailings ponds may have a close association with the high sediment amounts of these metals, and Mo in sediment may pose a high danger to the surrounding ecosystem [58].
In Armenia, nearly 300 square kilometers of land are polluted with heavy metals such as lead, copper, and molybdenum from the Alaverdi copper-molybdenum mine in the area. Land adjacent to the mining activities suffers from the worst pollution, with levels of these heavy metals at 20-40 times the limit [59]. The concentration of dissolved Mo has been estimated to be as high as 900 ppm in the U.S.A. from a uranium mill in southern Colorado. Dissolved Mo exists in amounts up to around 25 ppm in the aqueous discharge from major molybdenum mills also located in Colorado [60]. One of the highest documented levels of molybdenum pollution in soils or water bodies is approximately 20.8 mM, or 2000 ppm [61]. In 2007 and 2008, a series of mysterious cow deaths occurred across the dark brown soil zone of Saskatchewan during the calving season, causing the deaths of 46 cows for 11 cow producers. Molybdenum toxicity was identified to be the major reason for the loss [62]. Molybdenum can be presented in the form of molybdenite, a byproduct of gold and copper mining. In recent years, the contamination of several thousand acres of paddy land occurred due to episodic pollution cases involving leakage from mining sites [63]. Furthermore, the main hidden source of molybdenum pollution in Malaysia is via the indiscriminate release of lubricant oil wastes, which contain 1-5% Mo, with the highest molybdenum concentration of 17.86 mg per kg soil found at one location [64].

Molybdenum Bioremediation
Microorganisms utilize the mechanisms of bioreduction, bioprecipitation, bioaccumulation or sequestration, efflux pumping, and biosorption to detoxify and immobilize metal ions of molybdenum, chromium, copper, and mercury [12,13,[65][66][67][68]. During the past century, microbial molybdate (Mo 6+ ) reduction was first reported in Escherichia coli by Capaldi and Proskauer [69]. However, the details of the reduction phenomenon were only discovered during the last three decades by Campbell et al. in E. coli K12 [70]. The work of [71] on Thiobacillus ferrooxidans (now Acidithiobacillus ferrooxidans) was then continued on Enterobacter cloacae strain 48 or EC 48 [72] and other Mo-reducing bacteria (Table 1). Molybdenum can also be reduced to a lower oxidation state by the action of sulfide, produced by sulphate-reducing bacteria such as Desulfovibrio desulphuricans and Desulfovibrio vulgaris, where molybdenum is reduced to MoS 2 (Mo 4+ ). Molybdenum disulfide is insoluble and precipitates out of solutions, which is another avenue for molybdenum bioremediation [73].
The earliest documented report on molybdenum bioremediation was in Tyrol, Austria. A plant-microbial consortium was used to treat large pasture areas that had been contaminated with molybdenum from industrial discharge. Soluble toxic molybdenum was detoxified using microbes from sewage, which transformed soluble molybdenum into insoluble compounds [53]. The use of molybdenum-reducing microorganisms in this case, as well as in the future using currently isolated molybdenum-reducing bacteria [74][75][76][77][78][79][80][81][82], represents a sustainable means of remediating molybdenum pollution.

Mechanism of Molybdenum Reduction to Mo-Blue
The mechanism of microbial molybdate reduction to Mo-blue was first proposed for E. cloacae strain 48 [72,83]. Enzyme-catalyzed redox was involved in the change in the oxidation state of molybdenum (M 6+ to M 5+ ) before phosphate joined to form Mo-blue. Similar Mo-blue spectra were produced in both the bacterial molybdate reduction and ascorbate-reduce phosphate determination methods, indicating the possible formation of the phosphomolybdate intermediate [84]. At this stage, the involvement of the phosphomolybdate species was first suggested as occurring under static growth conditions and high cell concentrations in a shaking mode of growth, and semi-anaerobic conditions may occur. Hence, most heterotrophs undergo fermentation and release organic acids. Consequently, the pH of the medium is lowered and molybdate ions are converted to phosphomolybdate [85,86].
In an aqueous medium (pH 7), molybdenum exists as molybdate ions (MoO 2 4+ ). Molybdate ions, however, form a unique complex structure called polymolybdate anions in an acidic solution. This is a hallmark of molybdenum chemistry. The polymolybdate ions then further combine with other anions, such as sulfate, phosphate, or silicate, to form a mega-complex (Figure 1), the so-called heteropolymolybdate (phosphomolybdate, silicomolybdate, and sulfurmolybdate) [85,87]. This phenomenon is believed to be the first to occur during molybdate reduction to Mo-blue. Reducing agents, such as dithionite, ferrous ions, and stannous ions, can donate two electrons to the formed heteropolymolybdate (PMo 12 O 3+ : 12-molybdophosphate), hence converting it to molybdenum blue. Both of these donated electrons are spread across the entire polymetallate sphere by a thermallyactivated hopping process, as shown in electron spin resonance (ESR). The two reduced electrons are in constant motion, as shown by nuclear magnetic spectroscopy, and therefore account for the valency of all of the twelve molybdenum atoms. The movement of electrons across the Keggin complex causes a mixed valency of molybdenum oxidation states, is between 6+ and 5+ [88,89].
cell concentrations in a shaking mode of growth, and semi-anaerobic conditions may occur. Hence, most heterotrophs undergo fermentation and release organic acids. Consequently, the pH of the medium is lowered and molybdate ions are converted to phosphomolybdate [85,86].
In an aqueous medium (pH 7), molybdenum exists as molybdate ions (MoO2 4+ ). Molybdate ions, however, form a unique complex structure called polymolybdate anions in an acidic solution. This is a hallmark of molybdenum chemistry. The polymolybdate ions then further combine with other anions, such as sulfate, phosphate, or silicate, to form a mega-complex (Figure 1), the so-called heteropolymolybdate (phosphomolybdate, silicomolybdate, and sulfurmolybdate) [85,87]. This phenomenon is believed to be the first to occur during molybdate reduction to Mo-blue. Reducing agents, such as dithionite, ferrous ions, and stannous ions, can donate two electrons to the formed heteropolymolybdate (PMo12O 3+ : 12-molybdophosphate), hence converting it to molybdenum blue. Both of these donated electrons are spread across the entire polymetallate sphere by a thermallyactivated hopping process, as shown in electron spin resonance (ESR). The two reduced electrons are in constant motion, as shown by nuclear magnetic spectroscopy, and therefore account for the valency of all of the twelve molybdenum atoms. The movement of electrons across the Keggin complex causes a mixed valency of molybdenum oxidation states, is between 6+ and 5+ [88,89].
The spectral scanning of Mo-blue produced by all isolated molybdenum-reducing bacteria is remarkably similar to the ascorbate-reduced phosphate determination method. In the phosphate determination method, the Mo-blue spectrum produces a characteristic peak at 890 nm and a shoulder at 710 nm, whereas the bacterial spectrum generally shows a peak at 865 nm and a shoulder at 700 nm [86]. Cell contact is required in molybdate reduction to Mo-blue, as reported in [70], and phosphomolybdate has been suggested as a possible identity of Mo-blue. Because there are a number of lacunary species of phosphomolybdate that can change, as well as due to changes in pH, it is hard to differentiate between the species formed during bacterial molybdate reduction. Spectral absorbance scanning has therefore been suggested as an adequate method to differentiate between The spectral scanning of Mo-blue produced by all isolated molybdenum-reducing bacteria is remarkably similar to the ascorbate-reduced phosphate determination method. In the phosphate determination method, the Mo-blue spectrum produces a characteristic peak at 890 nm and a shoulder at 710 nm, whereas the bacterial spectrum generally shows a peak at 865 nm and a shoulder at 700 nm [86]. Cell contact is required in molybdate reduction to Mo-blue, as reported in [70], and phosphomolybdate has been suggested as a possible identity of Mo-blue. Because there are a number of lacunary species of phosphomolybdate that can change, as well as due to changes in pH, it is hard to differentiate between the species formed during bacterial molybdate reduction. Spectral absorbance scanning has therefore been suggested as an adequate method to differentiate between the major heteropolymolybdates, such as phosphomolybdate, sulphomolybdate, and silicomolybdate [87]. Based on the similar absorption spectrum to that of Mo-blue using the ascorbic acid method and a reduction in pH during fermentation of most Mo-reducing bacteria, Shukor et al. proposed [86] that phosphomolybdate is an intermediate during the bacterial reduction of molybdenum to Mo-blue and that this phosphomolybdate is chemically formed (Figure 2). Furthermore, entrapping Enterobacter cloacae 48 in a dialysis tube [90] provided evidence to support the earlier suggestion of Campbell that Mo-blue production in microorganisms requires cell contact [29,70]. ascorbic acid method and a reduction in pH during fermentation of most Mo-reducing bacteria, Shukor et al. proposed [86] that phosphomolybdate is an intermediate during the bacterial reduction of molybdenum to Mo-blue and that this phosphomolybdate is chemically formed (Figure 2). Furthermore, entrapping Enterobacter cloacae 48 in a dialysis tube [90] provided evidence to support the earlier suggestion of Campbell that Mo-blue production in microorganisms requires cell contact [29,70].

Characteristics of Previously Isolated Molybdenum-Reducing Bacteria
To date, the molybdenum-reducing bacteria are known to come from twelve different genera (Acinetobacter, Bacillus, Burkholderia, Enterobacter, Escherichia, Klebsiella, Pseudomonas, Pantoea, Morganella, Clostridium, Raoultella, and Serratia) isolated from different soils and polluted waters around the world. From the isolated molybdenum-reducing bacteria (52 bacteria), 45% are from Malaysia (23 bacteria), followed by Nigeria (eight bacteria), Egypt (seven bacteria), and Indonesia (five bacteria). The remainder are from Sudan (two bacteria) and Pakistan (two bacteria), as well as Iraq, China, USA, and Antarctica, at one bacterium each (Table 1). E. coli K12 was the only standard ATCC-type strain (strain ATCC 10798). A molybdenum-tolerant bacterium (Bacillus subtilis LM 4-2) from China was isolated [91] from a molybdenum mine but was not reported to reduce molybdenum to molybdenum blue. Though a complete genome sequence of this molybdenum-resistant bacterium has been reported, the characterization work was excluded. Therefore, a new research direction has emerged in the genome annotation of molybdenum-reducing bacteria, particularly for sequencing work of the isolate. Of the twelve genera that are currently known for molybdenum reduction, the genera Bacillus spp. and Staphylococcus spp. are gram-positive rod and cocci, respectively. The remaining genera are gram-negative rods, with the majority belonging to the Enterobacteriaceae family (Table 1). For 80% of the isolated reducers, the optimum temperature that supports Mo-blue production was found to be between 30 and 37 °C; for 10%, the temperature was between 25 and 30 °C [67,92]; for 5%, it was between 15 and 20 °C [93]; and for 5%, it was 40 °C [94]. Of these, one was a psychrotolerant bacterium isolated from Antarctica, and two were thermophilic bacteria isolated from Egypt (

Characteristics of Previously Isolated Molybdenum-Reducing Bacteria
To date, the molybdenum-reducing bacteria are known to come from twelve different genera (Acinetobacter, Bacillus, Burkholderia, Enterobacter, Escherichia, Klebsiella, Pseudomonas, Pantoea, Morganella, Clostridium, Raoultella, and Serratia) isolated from different soils and polluted waters around the world. From the isolated molybdenum-reducing bacteria (52 bacteria), 45% are from Malaysia (23 bacteria), followed by Nigeria (eight bacteria), Egypt (seven bacteria), and Indonesia (five bacteria). The remainder are from Sudan (two bacteria) and Pakistan (two bacteria), as well as Iraq, China, USA, and Antarctica, at one bacterium each (Table 1). E. coli K12 was the only standard ATCC-type strain (strain ATCC 10798). A molybdenum-tolerant bacterium (Bacillus subtilis LM 4-2) from China was isolated [91] from a molybdenum mine but was not reported to reduce molybdenum to molybdenum blue. Though a complete genome sequence of this molybdenum-resistant bacterium has been reported, the characterization work was excluded. Therefore, a new research direction has emerged in the genome annotation of molybdenum-reducing bacteria, particularly for sequencing work of the isolate. Of the twelve genera that are currently known for molybdenum reduction, the genera Bacillus spp. and Staphylococcus spp. are gram-positive rod and cocci, respectively. The remaining genera are gram-negative rods, with the majority belonging to the Enterobacteriaceae family (Table 1). For 80% of the isolated reducers, the optimum temperature that supports Mo-blue production was found to be between 30 and 37 • C; for 10%, the temperature was between 25 and 30 • C [67,92]; for 5%, it was between 15 and 20 • C [93]; and for 5%, it was 40 • C [94]. Of these, one was a psychrotolerant bacterium isolated from Antarctica, and two were thermophilic bacteria isolated from Egypt ( Table 1). The molybdate reduction was discovered to be the most favorable at pH values between 5 and 6 [70], 5.5 and 6.8 [30,66,95-99], 7 and 8 [67,72,92,95,[100][101][102][103], and 6.5 and 7.5 [93].
As a general observation, molybdate reduction to molybdenum blue is inhibited at phosphate concentrations above 5 mM, which might be due to the physical interaction with phosphomolybdate substrate rather than an enzymatic inhibition [29]. Nearly all Moreducers achieve the optimal Mo-blue production at phosphate concentrations of between 5 and 7.5 mM, whereas the optimum of Pseudomonas aeruginosa strain Amr-11 is between 2.5 and 7.5 mM (Table 1).
Several other molybdenum-reducing bacteria, e.g., Klebsiella oxytoca strain Hkeem and Escherichia coli K12, can also tolerate and support optimal Mo-blue production at molybdate concentrations as high as 80 mM. Generally, most Mo-reducing bacteria exhibit an optimum reduction from 25 to 55 mM sodium molybdate (Table 1). In contrast, Bacillus sp. strain khayat, Klebsiella oxytoca strain Saw-5, P. aeruginosa strain Amr-11, Klebsiella oxytoca strain Aft-7, Pseudomonas putida strain Amr-12, Pseudomonas sp. strain DRY2, Acinetobacter calcoaceticus strain Dr.Y12, Serratia marcescens strain Dr.Y9, Serratia marcescens strain DRY6, and Enterobacter cloacae strain 48 require lower molybdate concentrations of between 5 and 20 mM. The highest is for Pseudomonas sp. isolated from Nigerian soil, which has an optimum at 100 mM (Table 1). This is helpful for molybdenum reducers that can tolerate and reduce the concentrations that are higher than 20 mM, because molybdenum pollution in a previous report reached as high as 2000 ppm (20.8 mM molybdate) [61].
Therefore, almost all of the Mo-reducers are similarly inhibited by the respiratory inhibitors and other toxic metals, such as silver, arsenic, cadmium, chromium, copper, lead, mercury, and zinc [30,90,93,98].
It has been reported that nearly all of the molybdenum reducers are susceptible to similar toxic heavy metals reported in this work (Table 1). In the chromate-reducing bacteria Bacillus sp. [105] and Enterobacter cloacae strain H01 [106], toxic cationic metals, such as mercury and copper, strongly inhibit bacterial chromate reduction and pose a threat to the remediation of anionic metals such as molybdate, chromate, and vanadate. These toxic cationic metal ions bind to the sulfhydryl, carboxyl, amide, phosphoryl, and amine groups of many housekeeping enzymes, including metal reductases, ultimately halting remediation. A strategy to combat this is the addition of nontoxic cationic metals' chelating or reacting compounds, such as manganese oxide, phosphate, calcium carbonate, and magnesium hydroxide, which react and reduce the solubility of cationic metals, thus allowing for the commencement of bioremediation [107,108]. Another alternative, specifically for molybdenum, is the immobilization of the Mo-reducing bacterium in an entrapment agent such as dialysis tubing, which has been demonstrated to offer protection to the toxicity of the co-presence cationic metal ions [109].  Note: OFAT = one-factor-at-a-time; RSM = response surface method; CCD = central composite design.

Statistical-Based Optimization Compared to One-at-a-Time Approach
Statistical-based optimization, such as the response surface method (RSM), is a composition of statistical and mathematical techniques for empirically building a model. It is a well-known and current approach for constructing approximation models based on experimental observations [140][141][142]. A newer approach is the use of artificial neural networks (ANNs). Both the RSM and ANNs have individual advantages and disadvantages [143]. The RSM is, however, more commonly reported and hence allows for a good comparison with established RSM data. The RSM was developed by Box and Wilson (1951) and uses a meticulous arrangement of experiments with the ultimate objective of maximizing an answer (output variable). Multiple independent variables influence the performance (input variables). The use of the RSM based on a central composite design (CCD) to optimize molybdenum reduction has been carried out in bacteria such as Serratia sp. strain MIE2 [130], Serratia sp. strain HMY1 [121], Bacillus sonorensis strain Pharon3 (MK078035), and Bacillus tequilensis strain Pharon2 (MK078034) [82], and it has successfully improved the process compared to the one-factor-at-a-time (OFAT) method.

Mathematical Modelling of Molybdenum Reduction Profile and Kinetics
Bacterial growth-linked processes often demonstrate a unique phase. The specific growth rate commences at a value of zero, after which it accelerates to a maximal value (µ max ) in a certain period, thus producing a lag time (λ). Within the growth curves, a final phase is involved in which the rate diminishes and ultimately reaches zero. Therefore, an asymptote (A) is obtained. Changes in the growth rate frequently result in a sigmoidal curve with the characteristic lag phase just after t = 0. An exponential phase then follows, and it is succeeded by a stationary phase and, finally, the death phase [144]. The lag period seen in the sigmoid shape has been attributed to the bacterial cells gearing their growth mechanism to adjust to a new environment of a vegetative state, particularly during storage. This adjustment period is commonly described as the lag period and can be described as a transient period that connects two autonomous systems. This introduction of the lag time or parameter is significantly less problematic than a mechanistic interpretation [145]. At the initial inoculation, it is presumed that each bacterial cell has different rates of growth. A nonlinear distribution can be illustrated if the rates of growth have been measured, as suggested by several studies [145,146].
In addition to the asymptotic value and the lag period, another valuable parameter of a growth curve is the maximum specific growth rate (µ m ). This is commonly used in the development of secondary models, such as of the effects of substrate, product, pH, and temperature on the growth rate of the organism [144]. The logarithm of the activity or cellular microbial number, µ m , can be demonstrated by the slope of the line when the activity or organisms grow or expand exponentially [147]. The parameter is measured through manual estimation from the part of the curve that is approximately linear, and a linear regression is used to determine the slope of this curve. In practice, the linearization of the sigmoidal curve using natural logarithmic data transformation is one of the most frequently used methods, and it has been used to determine the specific reduction rate for molybdenum reduction in Bacillus sp. strain A.rzi [68]. In addition, the entire set of data can also be illustrated by a nonlinear regression growth model, which then estimates µ max , λ and A from the model itself [148].
The molybdenum reduction to molybdenum blue is a detoxification process, whereas the production of Mo-blue is a growth-associated process [66,67,90,[94][95][96]102,109,131,149,150]. Various approaches to reduction modeling have been previously used [151], including logistic [144,152], Gompertz [144,153], Richards [144,154], Schnute [144], , von Bertalanffy [156,157], Buchanan three-phase [158], and the Huang model, which was used most recently [159] (Table 1) to model Mo-blue production from bacteria. It should be noted that the maximum specific growth rate or µ m was modified to q m and represents the maximum specific reduction rate. It appears that the modified Gompertz model is the best model to fit Mo-reduction in many cases (Table 2).
Buchanan Three-phase linear model 3 Note: A = Mo-blue lower asymptote; q m = maximum specific Mo-blue production rate; v = effects near which asymptote maximum Mo-blue production occurs; P = no of parameters; λ = lag time; y max = Mo-blue upper asymptote; e = exponent (2.718281828); t = sampling time; α, β, and k = curve fitting parameters; h 0 = a dimensionless parameter quantifying the initial physiological state of the reduction process. The lag time (h −1 ) can be calculated as equal to h 0 /q max .
Once the maximum specific reduction rate values are obtained from the primary modelling exercise, the values can be used to model the effect of the substrate (molybdate) on the rate of reduction. Several important parameters, such as the specific reduction rate, theoretical maximum reduction, and determination of the reduction at high molybdenum concentrations (which affects the lag period of reduction and the inhibitory property of the substrate, i.e., molybdenum to Mo-blue production), can be shown by the mathematical modeling of the reduction process [151]. In addition, the specific reduction rate can be used in secondary modeling via models such as those of Haldane, Monod, Teissier, Aiba, Yano, and Luong. Though the use of these approaches in the modeling of reduction kinetics involving metal-reducing microbial processes is seldom studied, a similar implementation has been undertaken in the research of organic compound biodegradation. Haldane-type inhibition has been reported in scientific studies of the modeling of the reduction kinetics of heavy metals, including mercury [160], arsenate [161], and chromate [162], whereas a Monod kinetic approach was reported [163] for uranium reduction. Beneficial kinetics information, which is vital in the modeling of the bioremediation of these heavy metals, was also obtained from these model kinetics [164]. Therefore, a secondary modeling activity can be used to determine whether the substrate inhibits the reduction rate (Monod) or vice versa (Haldane, Teissier, Aiba, Yano, and Luong). The best model can be assessed via a comparison using statistical tests such as root mean square error (RMSE), adjusted coefficient of determination (R 2 ), bias factor (BF), accuracy factor (AF), and corrected Akaike information criterion (AICc) [157]. It appears that the Luong model describes the effect of molybdate inhibition on Mo reduction in more bacteria than other models (Table 3). Despite the commonly reported Haldane model, the Luong and Teissier models allow for the determination of the critical concentration of substrate that can completely inhibit Mo-blue production.  [68,103,123] Note: q max = maximal reduction rate (h −1 ); K s = half saturation constant for maximal reduction (mM); S m = maximal concentration of substrate tolerated and (mM); m, n, and K = curve parameters; S = substrate concentration (mM); P = product concentration (mM).

Characteristic of the Partially Purified Molybdenum-Reducing Enzyme from other Bacterial Sources
The first attempt to purify the Mo-reducing enzyme was first carried out in E. cloacae strain 48 [83]. The purification began with ammonium sulfate fractionation followed by gel filtration using a Sephadex G-200, but a poor result was obtained. The partially purified enzyme was discovered to be heat-labile with a gradual loss of about half of the initial activity within 10 min at 40 • C and a total loss of activity at 100 • C within 1 min of incubation. The optimum temperature and pH for the enzymatic activity were 30 • C and 8.0, respectively. From the Lineweaver-Burk plot, the rate of molybdate (M +6 ) reduction was shown to increase with the concentration and halted as the concentration reached 100 mM. The K m (apparent) was 16.5 mM, and V max was 0.33 mU mg −1 . A second attempt at purifying the Mo-reducing enzyme [84] from E. cloacae strain 48 used phosphomolybdate (PM) as an electron acceptor using ammonium sulfate (40%-50%) fractionation, and ion exchange on DEAE-cellulose and gel filtration on Sephacryl S-200. An excellent ammonium sulfate fraction was then obtained with 6.5-fold enzyme purification and about 97% recovery of the initial value. A single peak was produced from ion exchange chromatography on DEAE-cellulose. Though the level of purification was outstanding, a significant decrease in the activity was recorded in both the gel filtration and ion exchange steps. About 1.6% of the final yield was found in the initial crude, indicating that a large degree of enzyme denaturation or loss occurred, and the final specific activity was 240 mU mg −1 protein. A 40-fold purification was achieved after the ion exchange and gel filtration chromatographic steps. From the SDS-PAGE of the concentrated gel filtration step, three protein bands with estimated molecular weights of 95, 100, and 105 kDa were obtained, indicating that the purification of the enzyme was not successful [84].
The first purification of the molybdenum-reducing enzyme was reported from the bacterium Serratia sp. Dr.Y5 [137]. The enzyme activity was eluted from a Mono Q anion exchange column at 330 mM of NaCl with a gradient from 0 to 500 mM NaCl. The final step involved gel filtration on Zorbax GF-250. The molecular weight (apparent) was estimated to be 100 kDa, and it was found to be a monomeric protein. Similar to the previous results, a significant reduction in enzyme activity after ion exchange chromatography was observed [137].
Characterizing the purified fraction from strain Dr.Y5 showed that the optimum pH was 6 and that temperature stability occurred between 25 and 35 • C. At 15 mM 12phosphomolybdate, a plot of initial rates against concentrations of substrate showed an apparent V max for NADH of 12 mU mg −1 protein and an apparent K m of 0.79 mM, whereas at 5 mM NADH, the apparent V max and apparent K m values for 12-MoP (phosphomolybdate) were 12.05 mU mg −1 protein and 3.87 mM, respectively. A higher V max for NADH was reported in strain Dr.Y5 than EC 48 at 6.28 mU mg −1 protein [137]. However, the apparent K m for NADH was lower than that of EC 48 at 1.65 mM, thus indicating a better affinity to a Mo-reducing enzyme in strain Dr. Y5. Both strains produced similar apparent V max values for 12-MoP. A higher apparent K m for 12-MoP in Dr.Y5, relative to 0.32 mM for EC 48, was obtained. which might suggest that the enzyme from EC 48 has a higher affinity for 12-MoP than strain Dr. Y5. In addition, the catalytic efficiency (k cat /K m ) of the Mo-reducing enzyme from Dr.Y5 was found to be 5.47 M −1 s −1 , which is about 2 × 10 5 less efficient than chromate reductase from Thermus scotoductus, a related metal reductase [172]. To date, the molybdenum-reducing enzyme from EC 48 appears to be the most efficient, with the highest V max and comparatively very low K m for NADH and phosphomolybdate [95].
The molybdenum-reducing enzyme was also successfully purified from Serratia sp. strain MIE [172], with steps involving an ammonium sulphate fractionation (between 40% and 50%) and gel filtration (GF-250). A final purification of 20.8-fold was obtained. The enzyme was monomeric and had a molecular weight of 100 kDa, an optimum pH of 5.0, and an optimum temperature of 35 • C for maximum activity. The K m and V max for NADH and phosphomolybdate were 0.859 mM and 16.11 mU mg −1 and 6.02 mM and 6.89 mU mg −1 , respectively. The catalytic efficiency measured as the ratio of K cat /K m was 7.89 M −1 s −1 . The most recent purification of the Mo-reducing enzyme came from a cyanide-degrading bacterium Serratia sp. strain UPM-FR1 [123]. The purification steps involved a 50%-60% ammonium sulphate fraction and gel filtration (GF-250). A much lower final purification of 5.69-fold was obtained. Similar to other works, the enzyme was monomeric and had a molecular weight of 100 kDa, an optimum pH of 5.0, and a temperature of between 25 and 35 • C for maximum activity. The K m and V max for NADH and phosphomolybdate were 1.81 mM and 21.2 mU mg −1 and 4.53 mM and 21.66 mU mg −1 , respectively, and the catalytic efficiency (K cat /K m ) was 5.35 M −1 s −1 ( Table 4). The work of Opperman et al. [173] on the identification of the chromate reductase in Thermus scotoductus SA-01 showed that the enzyme is not novel; rather, it is an existing enzyme called the Old Yellow enzyme with a novel chromate reductase activity. This is the reason why chromate reductase does not have an enzyme commission (E.C.) number. The only established toxic metal reductases are mercury reductase [174] and selenate reductase [175]. Hence, the identification of the Mo-reducing enzyme via the sequencing of the purified enzyme is vital.

Use of Electron Transport Chain Inhibitors in Probing the Location of Molybdenum Reduction
The electron transport chain (ETC) is a series of electron transfer reactions taking place within the inner mitochondrial membrane. Several heme-proteins, soluble enzymes, coenzymes, Fe-S clusters, and a host of several metal cofactors are involved in the electron transfer process and subsequently form the components of the system. The components of the electron transport chain are then organized in a unique manner to form complexes (complexes I, II, III, and IV), with ATPase serving as the fifth complex (complex V) [176]. Examples of the most significant known inhibitors of respiratory enzymes include chemical agents such as amytal, rotenone, antimycin A, carbon monoxide, sodium azide, and cyanides. Rotenone inhibits NADH dehydrogenase, whereas sodium azide and cyanide are the inhibitors of cytochrome oxidase. Moreover, both antimycin A and hydroxyquinoline-Noxide are cytochrome b inhibitors, and all of these inhibitors have been tested for their capacity to inhibit Mo reduction, with cyanide found to be the sole inhibitor of molybdenum reduction [72]. The inhibition by cyanide of Mo reduction in E. cloacae 48 indicates that the location occurred downstream from cytochrome [72]. However, all of the respiratory inhibitors tested on the recently isolated Mo-reducing bacteria showed no inhibition of the Mo-reducing enzyme [66][67][68]93,94,96,101,102,129]. The use of cyanide has only recently been discovered to increase the pH of the reaction mixture because it transforms the reaction mixture into a highly alkaline solution. Because phosphomolybdate is highly unstable under alkaline conditions, no reduction was observed when the effect of cyanide on E. cloacae strain 48 was revisited by adjusting the reaction mixture after addition of cyanide to a neutral pH [176]. Therefore, the electron transport chain is unlikely to be the location of molybdate reduction.

Identification of the Mo-Reducing Enzyme
The optimization of Mo-blue reduction for the purpose of bioremediation offers a method for the removal of molybdenum pollutants in addition to physical and chemical methods. Presently, molybdenum pollutants are chemically treated as a remediation technique [52]. Microbial-based molybdenum bioremediation offers several advantages over chemical-based remediation. First, Mo-blue is colloidal and forms a precipitate with bacterial biomasses [177]. This is suitable for bioremediation via entrapment with membranes, in which cells are either enclosed or immobilized. The main challenge in the elucidation of the mechanism of reduction is the identification of the enzyme. Though the enzyme has been purified and characterized in one bacterium [137], the sequencing of the enzyme was not successful.

Conclusions
This review aimed to provide an update on the current knowledge regarding the molybdenum-reducing bacteria that have been isolated to date. This review also provided an update on the ability of some molybdenum-reducers to degrade other xenobiotics, which is a feature that needs to be studied and optimized in the future. A number of these xenobiotics, such as amides, can be used as electron donors for molybdenum reduction, and the ability of other xenobiotics to act as electron donors should be studied in the future. The inhibitory effect on the molybdenum reduction of cationic heavy metals, such as mercury, copper, and silver, is also seen in other anionic reductions, such as chromate and arsenate microbial reductions. Thus, means to combat this issue need to be studied for real remediation applications in the future. The true identity of the molybdenum-reducing enzyme is currently unknown, and the future sequencing of the purified enzyme could reveal the underlying mechanism behind the reduction process. Inhibition kinetics studies have shown that molybdate is toxic at high concentrations. Thus, the limits of the capability of bacteria to remediate sites highly contaminated by molybdenum, in addition to molybdenum-rich effluent from mine tailings, should be studied. The activation energy for molybdenum reduction also needs to be explored in future studies to distinguish molybdenum from other anionic metal reductions and other xenobiotic degradation processes.