There is a variety of melanin sources; several common fruits and vegetables, such as apples, bananas, garlic, persimmons, and potatoes, can produce melanin (Lefevre and Perrett, 2015; Hagiwara et al., 2016; Qi et al., 2020). Melanin can also be obtained from plants, such as Mucuna monosperma (Wight) callus (Inamdar et al., 2014). Commercial melanin is prepared from sepia extract or by synthetic means (Prados-Rosales et al., 2015; Schroeder et al., 2015; Srisuk et al., 2016).

Nevertheless, these methods have the disadvantage of high production costs, low maneuverability, and environmental pollution risk. Therefore, the bioproduction of melanin by microorganisms such as fungi and bacteria as alternative melanin sources has attracted great attention. As they grow fast relatively and can be applied to the scale-up process for mass production. Besides, several attempts have been made to isolate melanin-producing strains from various environments to enhance melanin production through reactions and host cell engineering.

To date, several melanin-producing fungal strains have been reported (Rosas et al., 2000; Gomez et al., 2001; Nosanchuk et al., 2002; Morris-Jones et al., 2003; da Silva et al., 2006; Franzen et al., 2006; Nosanchuk et al., 2007; Walker et al., 2010; Nosanchuk et al., 2015). While fungal strains produce different types of melanin, the predominant melanin type is nitrogen-deficient allomelanin (Varga et al., 2016). The key enzymes responsible for melanin synthesis are tyrosinases, which are copper-dependent biocatalysts involved in ortho-specific hydroxylation and subsequent oxidation of monophenols like tyrosine (Kuzumaki et al., 1993). Laccase is another enzyme, which can catalyze the oxidation of a broad range of substrates like tyrosinase, including dihydroxyphenols and quinones (Nagai et al., 2003). Both enzymes are commonly abundant in plants and fungi rather than in bacteria. Therefore, fungal strains were potential candidate for melanin production. Moreover, the complex and dynamic membrane structure of fungi supplies a more suitable environment for melanin synthesis and deposition. For example, Cryptococcus neoformans melanin was reported to be located within the cell walls of branched polysaccharides and protein constructs (Nosanchuk and Casadevall, 2003). In addition, the presence of other cellular organizations, such as fungal vesicles, melanosomes, and anchoring structures, have been reported to assist in the efficient production and localization of fungal melanin (Nosanchuk and Casadevall, 2003; Nosanchuk et al., 2015; Camacho et al., 2019).

It should be noted, however, that the production of melanin-consuming fungus requires a relatively long incubation time due to the low cell growth rate of the fungus; for example, Auricularia auricula or Gliocephalotrichum simplex produced 2.97 g/L and 6.6 g/L of melanin in 8 and 6 days, respectively (Jalmi et al., 2012; Sun et al., 2016). Interestingly, Ribera et al. reported that 161 days of Armillaria cepistipes culture could produce 27.98 g/L of eumelanin, which was the highest as far as our understanding, in a 3% (w/v) tyrosine-supplemented medium (Ribera et al., 2019). However, it took a long time of 161 days to achieve this production titer.

It is possible to produce eumelanin from L-tyrosine and allomelanin via the polyketide pathway (Varga et al., 2016). However, limitations, such as low growth rate, sporulation, low extraction efficiency, and potential pathogenicity of fungal strains, need to be overcome to obtain desirable production titers. Recently, along with the development of genetic manipulation and sequencing technology, it has become possible to increase the productivity of various biochemicals with recombinant fungi through genetic engineering. In line with this, it could be possible to increase fungal melanin production through the expression of an external enzyme.

Several microbial melanins have also been reported (Cubo et al., 1988; Jalmi et al., 2012; Ganesh Kumar et al., 2013; Surwase et al., 2013; Guo et al., 2014a; Madhusudhan et al., 2014; Tarangini and Mishra, 2014; Perez-Cuesta et al., 2020). Also, it was reported the production of melanin by using wild-type bacteria of Klebsiella sp., Pseudomonas, Streptomyces, Bacillus, Amorphotheca, and Vibrio, or by the expressing tyrosinase in E. coli as summarized in Table 1 (Tarangini and Mishra, 2014; Oh et al., 2020; Wang et al., 2020; Ahn et al., 2021). Organisms with melanogenic capabilities have also been employed to develop production processes, which included production optimization of melanin by utilizing various carbon sources and culture variables; In particular, tyrosine, peptone, soy peptone, starch, and yeast extract were used as carbon sources or mixtures. This resulted in the biosynthesis of tyrosine-based eumelanin. The Klebsiella sp. GSK46 strain, which was isolated from crop field soil, was able to produce approximately 0.13 g/L of eumelanin when fed with 1 g/L of tyrosine (Sajjan et al., 2010).

However, melanin could be produced even in the absence of tyrosine. For example, marine Pseudomonas stutzeri, isolated from seaweed, was found to produce significant amounts of melanin, which was 6.7 g/L within 10 h of incubation in sea water production medium without tyrosine supplementation (Ganesh Kumar et al., 2013). As fruit waste extract provides good nutrition for biochemical production, it has been utilized for melanin production. Tarangini and Mishra reported that Bacillus safensis, isolated from garden soil, could produce 6.96 g/L of melanin within 10 h of incubation (Tarangini and Mishra, 2014; Valdez-Calderón et al., 2020).

Amino acids also have been utilized for melanin production through whole cell biotransformation, in addition to the use of sugar-based fermentation including glucose, starch, and molasses (Ghadge et al., 2020; Oh et al., 2020; Eskandari and Etemadifar, 2021). For example, Eskandari and Etemadifar reported cost effective melanin production using Pseudomonas koreensis UIS19 in a molasses medium with tyrosine supplementation (Mustafa et al., 2020; Eskandari and Etemadifar, 2021). A total of 32 g/L of sugar was consumed to obtain 5.4 g/L of dry cell mass and 0.44 g dry melanin/g weight of yield could be achieved from supplemented tyrosine. In addition, several amino acid-based mediums, such as peptone and yeast extract, were utilized for melanin production using Streptomcyes kathirae, Streptomyces glaucescens, Streptomyces sp. ZL-24, and Amorphoteca resinae, which resulted in several g/L of melanin (Guo J. et al., 2014; El-Naggar and El-Ewasy, 2017; Wang et al., 2019; Eskandari and Etemadifar, 2021). In particular, melanin production by S. kathirae could reach up to 13.7 g/L, but 128 h of incubation was required for the highest titer (Guo J. et al., 2014).

It is worth noting that metal ions are critical for eumelanin production. For example, ferrous and nickel ion supplementation has been reported to drive melanin production by improving tyrosinase activity or by inducing the synthesis of tyrosinase (Wang et al., 2019). According to optimization results, 1.33 g/L FeSO₄ and 3.05 mM NiCl₂ could produce approximately 189.9 mg/L of insoluble melanin and 4.24 g/L of soluble pure melanin. The supplementation of metal ions seemed to have a positive effect on the activation of melanin production; however, the produced melanin was also reported to be able to chelate or absorb metal ions, such as Cu(II) and Zn(II), which would result in a metal-melanin complex and affect its characteristic features (Caldas et al., 2020; He et al., 2020).

Another interesting type of melanin produced by bacteria is pyomelanin. Pyomelanin utilizes different synthetic pathways compared to bacterial eumelanin, even though they both are originated from L-tyrosine. The key enzyme in pyomelanin synthesis is the 4-hydroxyphenylpyruvate dioxygenase (4-HPPD) enzyme, which converts 4-hydroxyphenylpyruvate, a transaminated form of tyrosine, into homogentisate, a key intermediate in pyomelanin synthesis (Figure 1B). Recently, Ralstonia picketti was isolated and identified as capable of generating pyomelanin in the presence of tyrosine (Seo and Choi, 2020a). In the presence of 4 mM of tyrosine, R. picketti could produce about 0.09 g/L of pyomelanin within 62 h of incubation. To verify 4-HPPD-dependent pyomelanin synthesis, the encoding gene was isolated and cloned into E. coli BL21 (DE3). And the 4-HPPD overexpressing cells could produce 0.213 g/L of pyomelanin from 1 mM tyrosine within 24 h of incubation, suggesting that recombinant strain development could greatly enhance the production rate and titer (Seo and Choi, 2020a). Similarly, Bolognese et al. isolated the 4-HPPD enzyme and constructed a pyomelanin-producing recombinant E. coli strain that could produce 0.213 g/L of pyomelanin (Bolognese et al., 2019). In addition to bacterial strains, the yeast strain Yarrowia lypolytica W29 was isolated and verified to be capable of producing 0.5 g/L of pyomelanin by 1 g/L of tyrosine feeding (Ben Tahar et al., 2020). However, similar to fungal melanin production, a 72 h of incubation period was required to achieve the highest titer.

The development of a recombinant strain to produce allomelanin has also been extensively studied. For example, caffeic acid-based allomelanin production was investigated by our group. Jang et al. first reported the co-expression of feruloyl-CoA synthetase (FCS) and enoyl-CoA hydratase/aldolase (ECH) in an E. coli strain that drives allomelanin production in the presence of caffeic acids (Jang et al., 2018). These enzymes have been previously utilized in vanillin synthesis from ferulic acid (Gallage et al., 2014). As caffeic acid has a catechol moiety in its core structure, contrary to ferulic acid of which one hydroxyl group was blocked by the methoxyl group, the enzymatic modification of the other carboxylic moiety could readily lead to the formation of allomelanin. The FCS/ECH overexpressing recombinant strain could produce 0.2 g/L of allomelanin within a 12 h reaction (∼40.9 mg/L/h) (Jang et al., 2018). Ahn et al. also used the same strain to produce caffeic acid-based allomelanin and compared its chemical composition with that of other natural and synthetic melanin (Ahn et al., 2019). Interestingly, the caffeic acid-derived allomelanin showed substantial dyeing of the HEMA (hydroxyethyl methacrylate) polymer, which is generally used for soft contact lenses, suggesting the potential application of melanin as a UV-blocking contact lens.

Recombinant strains for melanin synthesis have several advantages, not only in terms of production rates and titers but also regarding extraction and purity. This was evident in studies conducted on the production of eumelanin and pyomelanin (Jang et al., 2018; Ahn et al., 2019; Bolognese et al., 2019; Park H. et al., 2020; Seo and Choi, 2020a). Another advantage is the possibility of an additional supply of melanin building blocks to control melanin chemical structure, this allows for the engineering of functionalities depending on the purpose of use. For example, we reported eumelanin engineering by co-expressing bacterial tyrosinase (MelC) with cytochrome P450 monooxygenase (CYP102G4), which is capable of catalyzing indole C2 hydroxylation (Park H. et al., 2020). The additionally supplied 2-hydroxyindole functioned as a new building block in melanin polymerization and could obtain different physical and electrical characteristics. However, several issues regarding the use of recombinant strains for melanin production should be addressed. For example, there is an issue regarding the safety of genetic engineering for use in cosmetics and pharmaceuticals. In addition, the dependency of the macroscopic structures and physical properties on the producing host should be considered.

To understand the reaction time and titer correlation in melanin production, a space-time yield analysis was conducted. Space-time yield analysis of the summarized microbial melanin production in g/L (closed circle, •) revealed that this biotransformation exhibited not distinct but observable two classes, namely those with a production rate range of less than 0.05 g/L/h and those with a range over 0.1 g/L/h (Figure 2). The first group includes most eumelanin- and pyomelanin-producing bacteria with less than 100 h of reaction time. The second group includes fungi and some Streptomyces species with more than 100 h of reaction time (Guo J. et al., 2014; Ribera et al., 2019). In particular, A. auricula showed the longest reaction time of 8 days with a moderate production titer (2.97 g/L) (Sun et al., 2016).

The second group had a relatively higher production rate. This group included several bacteria, such as P. stutzeri, P. koreensis, and B. safensis, which could produce more than 5 g/L of melanin within 24 h (Ganesh Kumar et al., 2013; Tarangini and Mishra, 2014; Eskandari and Etemadifar, 2021). Interestingly, melanin production by S. kathirae, with a 13.7 g/L titer within 128 h also belongs to this group, as it showed a production rate of more than 0.1 g/L/h (Guo J. et al., 2014). Compared to the fungal host system, the bacterial host system for whole-cell melanin production appears to be advantageous in terms of timescale, depending on the type of target melanin. However, several hurdles must be overcome to utilize bacterial hosts for industrial-scale production. One of the most limiting factors is the necessity of isolation and purification steps in circumstances where the synthesized melanin is not secreted. In addition, an adequate growth medium needs to be optimized to obtain a desirable cell mass.

In general, the pH for fungal melanin production was less than 6, whereas it was approximately neutral in bacterial cases. Although a pH of less than 7 was adopted for bacterial melanin production regarding P. stutzeri, S. kathirae, and B. safensis, which showed more than 5 g/L of production titer, the optimal pH for melanin production varied depending on the host cells (Ganesh Kumar et al., 2013; Guo J. et al., 2014; Eskandari and Etemadifar, 2021). The temperature for melanin synthesis is approximately 28°C for fungi and 30–37°C for bacterial systems. However, there seems to be no significant correlation between melanin production and temperature; rather, it seems more important to secure the maximal cell mass for melanin production under optimized conditions. Therefore, melanin production should be focused on the optimization of production parameters, such as growth medium composition, pH, temperature, extraction parameters, in addition to the design of response surface methodology in order to obtain a higher production titer and rate.

The basic melanin production process includes host selection, fermentation or biotransformation, followed by securing crude melanin through extraction and purification processes to obtain pure melanin (Figure 3). The method of extracting melanin differs depending on the host cell that is producing melanin, the intracellular localization of melanin, the structural properties of melanin, and the melanin crystal structure. As melanin pigment can easily be found in nature, research on extracting melanin was conducted early on (Aneesh et al., 2020). In particular, methods for extracting melanin pigment from melanocytes and melanin organs, which are generally extracted by dissolving in an alkali or strong acid solution and heating (Young, 1921; Voss, 1954). For example, crude melanin was obtained by simple alkali extraction; however, the yield was as low as 2.59% (Ma et al., 2018). The extraction and purification process of melanin affects the purity of melanin, depending on the extraction method, the number of repeat cycles, and the form of melanin, namely crude or pure melanin.

The detailed extraction process for microbial melanin production is presented in Table 2. Depending on the melanin source, it is divided into extracellular and intracellular melanin. Extracellular melanin extraction methods employ acid precipitation, whereas additional alkali extraction is necessary for intracellular melanin production. To assist alkali extraction, ultrasonic- or microwave-assisted methods were used (450 W for 50 min, or 70 W for 3 min periods with 30 cycles) (Sajjan et al., 2010; Jalmi et al., 2012; Lu et al., 2014; Hou et al., 2019; Liu et al., 2019). According to Hou et al., the ultrasonic-assisted extraction method yielded 37.33% pure melanin, whereas 24.24% was obtained without this step (Hou et al., 2019). Similarly, Lu et al. reported that a purification yield of 11.08%% could be achieved through a microwave-assisted extraction method, which was 40.43% higher than that obtained by alkali extraction and acid precipitation. In addition, an additional step of boiling at 80°C for 2 h was employed to increase the extraction yields (Oh et al., 2020).

Before acid precipitation, filtering through various materials, such as 0.45 μm glass fiber, 0.45 μm nitrocellulose membrane, 0.22 μm membrane filter, and Millipore 0.2 μm polyether sulfone membrane, were employed to remove cell debris and byproducts (Jalmi et al., 2012; Ribera et al., 2019; Oh et al., 2020; Wang et al., 2020). To assist precipitation, a boiling or incubation step for several hours may be added (Sajjan et al., 2010; Sun et al., 2016; Liu et al., 2019). After precipitation, a washing step with deionized water was conducted. Crude melanin could be prepared using these filtration-precipitation-centrifugation-washing procedures (Figure 3).

To increase the purity of the isolated crude melanin, several redissolution, precipitation, boiling, and washing steps were employed. In brief, crude melanin was dissolved in NaOH and collected by centrifugation. Thereafter, the pH of the collected sample was adjusted to approximately pH 2 with HCl, followed by incubation. The resuspended melanin was collected by centrifugation and washed several times with deionized water. Finally, the collected melanin was washed with CHCl₃, DCM, EA, and pure EtOH, followed by lyophilization. Depending on the melanin type and condition, extra boiling, acid-hydrolysis, and repetitive washing steps can be added to pure melanin (Sun et al., 2016; Liu et al., 2019; Ribera et al., 2019; Oh et al., 2020).

Several simplified extraction methods have been proposed, however, acid precipitation-pH adjustment-washing-resuspension steps are commonly used (Sajjan et al., 2010; Jalmi et al., 2012; Guo J. et al., 2014; Tarangini and Mishra, 2014; Ghadge et al., 2020; Wang et al., 2020). In addition, other useful technologies have been applied to melanin extraction. For example, enzymatic disruption of the cell membrane using protease or hydrolase enzymes, instead of alkali extraction, has been utilized. Additionally, a variety of organic solvents has been utilized for melanin extraction with excellent yields. With respect to bioprocess, an issue regarding environmental concerns should be considered as such use of organic solvent and effluent disposal for melanin isolation process. As described above, several extraction methods could be applied to different types of melanin and different sources, suggesting that no optimal method can be applied consistently. Depending on the chemical structure, type, solubility, and purpose of use, it seems appropriate to use optimized methods specific to each process.