Production and Analytical Aspects of Natural Pigments to Enhance Alternative Meat Product Color

Color is a major feature that strongly influences the consumer’s perception, selection, and acceptance of various foods. An improved understanding regarding bio-safety protocols, health welfare, and the nutritional importance of food colorants has shifted the attention of the scientific community toward natural pigments to replace their toxic synthetic counterparts. However, owing to safety and toxicity concerns, incorporating natural colorants directly from viable sources into plant-based meat (PBM) has many limitations. Nonetheless, over time, safe and cheap extraction techniques have been developed to extract the purified form of coloring agents from raw materials to be incorporated into PBM products. Subsequently, extracted anthocyanin has displayed compounds like Delphinidin-3-mono glucoside (D3G) at 3.1 min and Petunidin-3-mono glucoside (P3G) at 5.1 277, 515, and 546 nm at chromatographic lambda. Fe-pheophytin was successfully generated from chlorophyll through the ion exchange method. Likewise, the optical density (OD) of synthesized leghemoglobin (LegH) indicated that pBHA bacteria grow more rigorously containing ampicillin with a dilution factor of 10 after 1 h of inoculation. The potential LegH sequence was identified at 2500 bp through gel electrophoresis. The color coordinates and absorbance level of natural pigments showed significant differences (p < 0.05) with the control. The development of coloring agents originating from natural sources for PBM can be considered advantageous compared to animal myoglobin in terms of health and functionality. Therefore, the purpose of this study was to produce natural coloring agents for PBM by extracting and developing chlorophyll from spinach, extracting anthocyanins from black beans, and inserting recombinant plasmids into microorganisms to produce LegH.


Introduction
PBM are food products generated from plant-based ingredients and designed to resemble conventional meat in terms of taste, texture and apperance, without the integration of animal components [1,2]. The goal of PBM is to provide a meat-like experience for those who want to reduce their consumption of animal products for ethical, health, or environmental reasons [3]. Some popular examples of PBM include burgers, sausages, and chicken nuggets produced from ingredients such as soy, pea protein, and mushrooms [4,5]. Nevertheless, the widening gap between the existing supply of meat and its imminent demand has increased the need to produce PBM analogs [6,7]. Therefore, demand for PBM products to replace conventional meat is increasing due to issues linked with extensive agriculture including livestock production, animal welfare, and potential climate change have made it unsustainable to increase meat production to meet future demand. Moreover, previous reports suggest that replacing animal protein with plant-based protein in the daily The existing literature regarding the production, extraction and analytical aspects of natural pigments developed for PBM is very limited; however, some corresponding studies have been reported [13,22,29,[32][33][34]. Therefore, researchers are taking a profound interest in utilizing natural pigments as replacements for their synthetic counterparts, particularly by developing sustainable extraction techniques for the efficient retrieval of bioactive compounds for use in PBM products and nutraceuticals. To the best of our knowledge, this is the first report describing the extraction and analytical aspects of natural pigments to be incorporated into PBM products. The chemicals used in this study were anthocyanins, ethanol, extraction solvent (Bioextrax AB, Lund, Sweden), distilled water, Milli-Q water, chlorophyll, acetic acid, petroleum ether (PE), diethyl ether, silica gel, ferric chloride (FeCl 2 ), sodium hydroxide (NaOH), sodium acetate, propanol, agarose, Tris-acetate-EDTA (TAE), methanol, acetonitrile, formic acid, and sodium dodecyl sulfate-polyacrylamide (SDS-PA).

Instruments
The instruments used in this study were a biosafety level 2 cabinet (BSC), CO 2 incubator, culture dishes, flasks and plates, colony counter, Buchner funnel with No. 1 Whatman filter paper, rotary evaporator, column chromatography assembly, agar plates, spectrophotometer, Milli-Q water purification apparatus, electrophoresis equipment, colorimeter, high-performance liquid chromatography (HPLC), and a 0.45 µm filter. Figure 1 shows the chemical structure of anthocyanins. Black beans were purchased from Morning Crops Co. (Gimpo, Republic of Korea), and anthocyanins were extracted from 500 g of black beans using an extraction solvent (Bioextrax AB, Lund, Sweden). The anthocyanins were extracted at 25 • C with acetic acid which was dispensed to a hydroalcoholic solution (60% ethanol) used for extraction to reach a 0.1% acetic acid concentration in the solvent extraction mixture. The resulting mixture was stirred at 250 rpm for 4 h, followed by sedimentation for an additional 1 h. Using vacuum filtration, the supernatant was recovered using Whatman paper No. 1. To eliminate the ethanol, the resultant extract was concentrated using a rotary evaporator by setting the water bath temperature to 50 • C. Subsequently, the concentrated extract was lyophilized, and the resulting lyophilized powder was stored at −80 • C [35].

Chemical Structure and Extraction of Chlorophyll
The chemical structure of chlorophyll is presented in Figure 2. Spinach (Spinacia oleracea) was purchased from Gomgom (Seoul, Republic of Korea). Under controlled conditions, a sample of 500 g was dried at 50-60 °C for 24 h to eliminate moisture. The dehydrated sample, weighing 37.86 g, was extracted with ethanol at a mass: solvent ratio of 1:10 w/v at 60 °C for 30 min. The crude chlorophyll filtrate was separated using a Buchner funnel with No. 1 Whatman filter paper. A rotary evaporator was used to concentrate the extract. Following concentration, petroleum ether (PE) was added to sonicate and dissolve the extracts. Thereafter, the extracts were stored at −20 °C with no influence of light [36].

Chemical Structure and Extraction of Chlorophyll
The chemical structure of chlorophyll is presented in Figure 2. Spinach (Spinacia oleracea) was purchased from Gomgom (Seoul, Republic of Korea). Under controlled conditions, a sample of 500 g was dried at 50-60 • C for 24 h to eliminate moisture. The dehydrated sample, weighing 37.86 g, was extracted with ethanol at a mass: solvent ratio of 1:10 w/v at 60 • C for 30 min. The crude chlorophyll filtrate was separated using a Buchner funnel with No. 1 Whatman filter paper. A rotary evaporator was used to concentrate the extract. Following concentration, petroleum ether (PE) was added to sonicate and dissolve the extracts. Thereafter, the extracts were stored at −20 • C with no influence of light [36]. drated sample, weighing 37.86 g, was extracted with ethanol at a mass: solvent ratio of 1:10 w/v at 60 °C for 30 min. The crude chlorophyll filtrate was separated using a Buchner funnel with No. 1 Whatman filter paper. A rotary evaporator was used to concentrate the extract. Following concentration, petroleum ether (PE) was added to sonicate and dissolve the extracts. Thereafter, the extracts were stored at −20 °C with no influence of light [36].

Synthesis of Fe-Pheophytin by Ion Exchange Chromatography
The process of synthesis of FePhe derivatives has been adopted from Nelson and Ferruzzi [29], and the detection of various FePhe derivatives samples extracted from spinach was separated by ion exchange chromatography through an established procedure of Moustafa and Morsi [37]. Consequently, through the ion exchange method, the extracted chlorophyll sample was added with 1 N NaOH to adjust the pH to 8.5, followed by heating, stirring, and dissolving in distilled water. The sample dissolved in distilled water was adjusted to pH 3 using 3-acetic acids and heated at 60-80 • C for 30 min. A solution of 1.3 M FeCl 2 and 0.25 M sodium acetate was prepared and dissolved in acetic acid. The prepared acetic acid solution was added five times to the sample, mixed with the heated sample, and further heated at 60-80 • C for 30 min. The pH of the sample was titrated to 8.5 by adding NaOH and distilled water, and then heating and stirring the extract. The sample extracted in distilled water was freeze-dried to form a powder and stored at −80 • C.
Subsequently, a column was prepared by mixing silica gel with PE. After filling with silica gel, spinach extract and sand were added to protect the sample layer. PE was added until color separation into green (chlorophyll), yellow (pheophytin), and orange (Fe-pheophytin); when this separation was confirmed, the yellow and orange sample layers were extracted with the addition of PE alongside 10% diethyl ether. The green sample layer was extracted using PE with 5% 2-propanol until the sample was not dissolved in the extraction solvent. After extraction, the sample was stored at −20 • C and protected from light ( Figure 3).

Plasmid
The sequence for producing LegH was custom-made by Bioneer Co. (Daejeon, Republic of Korea). This sequence was inserted into the plasmid BHA (pBHA) at a size of 1987 bp [38]. In the case of pBHA cloning, only pBHA recombinant microorganisms can be obtained from an ampicillin-containing medium, as this plasmid confers an antibiotic and ampicillin-resistant sequence. The sequence that produces LegH is 528 bp in size (Bioneer Co.; Daejeon, Republic of Korea) ( Table 1). extracted with the addition of PE alongside 10% diethyl ether. The extracted using PE with 5% 2-propanol until the sample was not tion solvent. After extraction, the sample was stored at −20 °C an ( Figure 3).     CTTTTAAGGCTAACATTCCACAATACTCTGTTGTTTTCTATACTTCTATTTTGGA  GAAGGCTCCTGCTGCTAAAGATTTGTTTTCTTTCTTGTCTAACGGTGTTGATCCATCTAATCCTAAGTTGACTGGTCATGCTGAAAAATTG  TTTGGTTTGGTTAGAGATTCTGCTGGTCAATTGAAGGCTAATGGTACTGTTGTT  GCTGATGCTGCTTTGGGTTCTATTCACGCTCAAAAGGCTATTACTGATCCACAATTCGTTGTTGTTAAGGAGGCTTTGTTGAAAACTATTA  AGGAGGCTGTTGGAGATAAATGGTCTGATGAATTGTCCTCTGCTTGGGAAGTTGCTTAC  GATGAATTGGCTGCTGCTATTAAGAAGGCTTTTCACCACCACCACCACCACGCTGA  GCATCACCTGTAAGTCGGACGAATTCGGCGCTCTTCCGCTTCCTCGCTCAC  TGACTCGCTGCGCTCGGTCGTTCGGCTGCGGCGAGCGGTATCAGCTCACTCAAAGGCGGTAATACGGTTATCCACAGAATCAGGGGATA  ACGCAGGAAAGAACATGTGAGCAAAAGGCCAGCAAAAGGCCAGGAACCG  TAAAAAGGCCGCGTTGCTGGCGTTTTTCCATAGGCTCCGCCCCCCTGACGAGCATCACAAAAATCGACGCTCAAGTCAGAGGTGGCGAA  ACCCGACAGGACTATAAAGATAC  CAGGCGTCTCCCTCGTGCGCTCTCCTGTTCCGACCCTGCCGCTTACCGGATACCTGTCCGCCTTTCTCCCTTCGGGAAGCGTGGCGCTTTC  TCATAGCTCACGCTGTAGGTATCTCAGTTCGGTGTAGGTCGTTCGCTCCAA  GCTGGGCTGTGTGCACGAACCCCCCGTTCAGCCCGACCGCTGCGCCTTATCCGGTAACTATCGTCTTGAGTCCAACCCGGTAAGACACGA  CTTATCGCCACTGGCAGCAGCCACTGGTAACAGGATTAG  CAGAGCGAGGTATGTAGGCGGTGCTACAGAGTTCTTGAAGTGGTGGCCTAACTACGGCTACACTAGAAGAACAGTATTTGGTATCTGCG  CTCTGCTGAAGCCAGTTACCTTCGGAAAAAGAGTTGGTAGCTCTTGATCCGGCAAACAAAC  CACCGCTGGTAGCGGTGGTTTTTTTGTTTGCAAGCAGCAGATTACGCGCAGAAAAAAAGGATCTCAAGAAGATCCTTTGATCTTTTCTAC  GGGGTCTGACGCTCAGTGGAACGAAAACTCACGTTAAGGGATTTTGGTCATGAGAT  TATCAAAAAGGATCTTCACCTAGATCCTTTTAAATTAAAAATGAAGTTTTAAATCAATCTAAAGTATATATGAGTAAACTTGGTCTGACA GTTACCAATGCTTAATCAGTGAGGCACCTATCTCAGCGATCTGTCTATTTCGTTCATCCA TAGTTGCCTGACTCCCCGTCGTGTAGATAACTACGATACGGGAGGGCTTACCATC The extraction technology for LegH Prep was adopted from Impossible™ Foods (Redwood City, CA, USA, Private limited) P. pastoris production strain, MXY0291 (GRAS notification for soy leghemoglobin protein preparation derived from Pichia pastoris) (patent approved, 2017) [39]. This LegH Prep does not contain the production organism or antibiotic resistance genes. Impossible™ Foods has conducted mass spectrometry to identify the P. pastoris protein that is present in LegH Prep at >1% of the total protein fraction. The sequence of each protein was analyzed to ensure that the P. pastoris proteins present in the LegH Prep did not contain significantly similar homology to known allergens [40].
The production of LegH from legumes using microorganisms was adopted from a previously established procedure described by Jin et al. [41]. The microbial culture was produced using LB broth (Sigma-Aldrich, St. Louis, MO, USA) and LB agar plates (Sigma-Aldrich, St. Louis, MO, USA) as the bacterial culture medium. Ampicillin (Roche Holding AG, Basel, Switzerland) was procured from Sigma-Aldrich (St. Louis, MO, USA). The Escherichia coli (DH5a) with or without pBHA (with LegH) were incubated with LB broth in an incubator at 37 C for 24 h. The cultured bacteria were diluted from 10 1 to 10 9 and the optical density (OD) was measured at 600 nm every hour for 24 h using a Multiskan SkyHigh Microplate Spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). E. coli were then separated into media with and without ampicillin, added to an LB plate, and spread to check for growth. Subsequently, pJAN, with an ampicillin resistance sequence, was inserted into P. pastoris. Similarly, P. pastoris was cultured in yeast extract peptone dextrose (YPD) medium containing 0.1 g/L of ampicillin. The culture process was conducted for up to 48 h at 30 C. Through this methodological process, the generated LegH sequence extracted from pBHA was inserted into pJAN to grow P. pastoris ( Figure 4). Moreover, pBHA contains a promoter that drives gene expression, a selection marker for identifying cells that have taken up the plasmid, and a multiple cloning site for inserting the gene of interest. The pBHA vector plasmid is named after its components, which include the promoter from the human beta-actin gene (B), the hygromycin selection marker (H), and the polyadenylation signal from the bovine growth hormone gene (A). Similarly, pJAN had a size of 4329 base pairs (bp). The pJAN is a custom-designed or proprietary plasmid developed by a specific research group.

Electrophoresis
The procedure was conducted by mixing 1% agarose gel with 1× TAE buffer (Bioneer, Daejeon, Republic of Korea), and 1× TAE buffer solution. The 50× TAE buffer (Bioneer, Republic of Korea) was diluted 1× with water produced using a Milli-Q water purification system (Millipore, Billerica, MA, USA). A 1 kb ladder from GeneDireX (Taiwan, China)

Absorbance Measurement
The absorbance of the anthocyanin, Fe-chlorophyll, LegH, and myoglobin samples was set to a concentration of 0.7% at 535 nm absorbance. Absorbance was measured in the 400-700 nm range using a Multiskan SkyHigh Microplate Spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) [45].

High-Performance Liquid Chromatography (HPLC) Analysis
Based on the determination of the standard, the experiment was performed using HPLC under the same conditions. Methanol, water, and acetonitrile (Sigma-Aldrich, Steinheim, Germany) were used as the HPLC-grade products. Samples were prepared by filtration using a 0.45 µm filter (GVS, Bologna, Italy). Agilent 1100 HPLC (Agilent, Santa Clara, CA, USA) and an Agilent C18 column were used for HPLC. For the mobile phase, solvent A was water: ACN: formic acid (45:45:10, vol/vol), and solvent B was 10% formic acid water. The flow rate used was 1 mL/min, the analysis time was 60 min, and the injection volume was 20 µL. The following wavelengths were detected: 277, 515, and 546 nm [46].

Statistical Analysis
Shapiro-Wilk and Levene's tests were performed to measure the normality and homoscedasticity assumptions. One-way analysis of variance (ANOVA) was used to analyze the data using R Studio (Ver. 4.0.2, Software & Tech Services, Boston, MA, USA). Duncan's multiple range tests were performed to assess the differences between means. Data are expressed as mean ± standard deviation. Significant differences among the groups were determined using a significance level of p < 0.05. All experiments were conducted in triplicate.

High-Performance Liquid Chromatography
Recently, a sustainable methodology based on green extraction technologies has been adopted for the extraction of bioactive compounds from plant sources. Based on the current green extraction technologies, anthocyanin is a plant-specialized water-soluble compound that confers red, violet, and blue colors to various food products [47]. Anthocyanin samples were analyzed by HPLC at 277 and 546 nm using a c18 column. The standard peaks of Deldelphinidin-3-mono glucoside (D3G) and Petunidin-3-mono glucoside (P3G) were observed at 3.1 and 5.1 min, respectively (S1 and S2). This phenomenon has been documented previously that characterized anthocyanin profiles of red table grape cultivars [48].
Previously, Lianza et al. [49] determined that the efficiency of anthocyanin extraction from untreated black beans was better than that of anthocyanin extraction experiments performed after peeling and drying black beans. The possible reason for this observation is that anthocyanin is water soluble; therefore, rehydration may lead to pigment deprivation during the extraction process. Moreover, it has been proven that anthocyanin extraction is positively affected by the incorporation of hydrochloric acid, rather than acetic acid. Additionally, Teixeira et al. [50] have extracted anthocyanin from broken black bean hulls using optimum conditions with a 30:70 (v/v) ratio of ethanol and citric acid solution 0.1 mol L −1 , at a flow rate of 4 mL min −1 at 60 • C. This study further indicated that Delphinidin-3-glucoside was the primary anthocyanin that has been linked to numerous health benefits, such as antidiabetic and antiradical activities. Furthermore, Pomar, Novo and Masa [48] estimated a typical HPLC chromatogram of anthocyanin at 546 nm. The authors calculated the chemical structure of anthocyanins by hydrolysis and chromatographic analysis. Various peaks were numbered which correspond to mono glucosides of five anthocyanins found in grapes.
In addition, the amount of anthocyanin extracted varied according to the concentration of acid added to the extraction solvent. When hydrochloric, acetic, formic, phosphoric, and citric acids were compared, hydrochloric acid showed the highest yield. The concentrationdependent extraction (0.1, 0.2, and 0.3%) revealed that 0.3% concentrations of these acids resulted in a higher yield of anthocyanin; however, higher or lower concentrations than 0.3% resulted in lower yields [51]. Similarly, based on the current observations, Pappas, Athanasiadis, Palaiogiannis, Poulianiti, Bozinou, Lalas and Makris [34] conducted a proportional evaluation of diverse novel technologies for the extraction of total anthocyanins from freeze-dried saffron tepals using aqueous solutions of citric acid and the lactic acid at various concentrations. Contrastingly, these results showed that no specific configuration was observed for the type of acid or concentration. The authors further clarified that the best performance for anthocyanin extraction and antioxidant capacity was obtained using a stirred-tank extraction with 1% (w/v) lactic acid solution; this yielded 3.25 g of cyanidin-3-Oglucoside equivalents (CGE)/kg in the dry weight of tepals. For anthocyanins, quantitative analysis has been widely conducted in terms of the total anthocyanin content. In recent years, however, an increasing number of studies report individual anthocyanin concentrations, and influencing factors have also been investigated. The major contribution to the extraction of anthocyanins from natural sources for food health applications is described in detail by previous authors [16,52,53].
However, because the aim of extracting anthocyanin as a pigment is for the manufacture of PBM as a food product, it is necessary to determine the residual amount of acid that is harmful to the human body and conduct a corresponding safety test. Furthermore, additional experiments on the anti-inflammatory and antioxidant effects of anthocyanins are required. Nonetheless, regarding the toxicity of anthocyanins, previous studies have indicated that this compound is widely used as a natural colorant in food and beverages in Europe, Japan, and the United States [22]. Additionally, JECFA concluded that anthocyanin-containing extracts had a very low or negligible level of toxicity upon consumption [24,54]. Figure 3 describes the conversion of chlorophyll into its derivatives pheophytin and Fe-pheophytin. Heme protein is considered the preferred protein for consumption [29]. However, the consumption of heme protein has several limitations, including cultural, socioeconomic, and environmental issues [55]. To counter these societal complications, an alternative strategy has been adopted by replacing the heme protein with a structurally similar tetrapyrrole. Chlorophyll molecules can be modified by replacing centrally chelated Mg with other metals, such as Fe, to generate different metalloporphyrins [56,57]. Furthermore, lipophilic and hydrophilic derivatives have been produced and applied in the food and pharmaceutical industries; these derivatives include Zn-and Cu-pheophytins, pyropheophytins, and Cu-and Fe-chlorophyllin as food colorants and nutritional supplements [58]. Inconsistent with the current study FURUYA et al. [59] synthesized and purified the Fepheophytin from crude spinach extract successfully. Similarly, the degradation of natural chlorophylls to pheophytins occurs instantly in natural plant extracts [60] and thermally treated vegetables [61]. Moreover, the extraction and incorporation of chlorophyll and its derivatives have been demonstrated in detail [62,63]. In the supporting literature, chlorophyll derivatives, pheophytins, and Zn-pheophytins from chlorophylls were extracted from spinach, characterized, and evaluated for their antioxidant and anti-inflammatory activities [63,64].

Chlorophyll and Fe-Pheophytin
Additionally, Nelson and Ferruzzi [29] utilized gastric and small intestine digestion simulation trials with Fe-chlorophyll; overall, this study indicated that the recovery rate of Fe-chlorophyll was 52.3%, the recovery rate of Fe-chlorophyll derivative was 58.7%, and digestion results were stable. Therefore, the development of chlorophyll-derived heme mimetics offers opportunities to expand the current Fe fortification strategies. Previous studies have indicated that the chlorophyll molecule can be modified by replacing centrally chelated Mg with other metals, such as Cu, Zn, and Fe, to form new metalloporphyrin complexes [56][57][58]. A successful trial was conducted for assessing the quantitative yield of chlorophylls into pheophytins and Fepheophytins, supporting these findings, and FePhe derivatives were synthesized at 78.5 ± 0.1% efficiency [29]. Due to the limited available literature, further discussion on this area of research is difficult.

Microbial Culture
In the current study, the extracted LegH underwent a series of experiments to determine the safety level regarding toxic and allergic ingredients. E. coli culture confirmed that pBHA-containing bacteria proliferated in LB medium containing ampicillin ( Figure 5A). From the results of culturing P. pastoris with and without LegH sequences in YPD medium, a color difference could be observed when observed with the naked eye.
Furthermore, based on the OD results, bacteria containing pBHA were grown in a medium containing ampicillin. Bacteria with a dilution factor of 10 were confirmed to grow 1 h after inoculation, and observable growth of bacteria with a dilution factor of 10 9 started after 18 h. Thus, it was confirmed that the growth rate was slow in samples with a high dilution factor because of the low number of initial bacteria. P. pastoris containing pJAN was able to grow in YPD medium containing ampicillin because of ampicillin resistance conferred by the pJAN plasmid ( Figure 5B).
As shown in Table 1, the LegH sequence used did not match the sequences of allergens and toxins that cause allergies. Additionally, 11 P. pastoris proteins with allergen and 13 proteins with toxic protein homology showed similarity to the protein of Saccharomyces cerevisiae, a microorganism previously used for food. In P. pastoris, LegH digestion occurs by the proteolytic enzyme pepsin, at pH 2, similar to that of the stomach [41]. Heme has a long history of safe consumption, while soy LegH is a novel food ingredient. Therefore, soy LegH was subjected to rigorous safety testing, including tests for allergenicity, mutagenicity, chromosome damage, and toxicity [32]. In addition to being rapidly digested by pepsin, soy LegH does not share any meaningful similarity to known allergens or toxins [41,65]. Furthermore, LegH produced by P. pastoris and a 28-day trial on animal subjects indicated that no mortality was associated with LegH administration. This further clarified that there was no LegH-associated adverse clinical evidence, including metabolic, ophthalmological, or histopathological alterations [32]. Therefore, the consumption of LegH using P. pastoris as a food ingredient has no adverse effects, particularly allergenic and toxic reactions. Furthermore, based on the OD results, bacteria containing pBHA were grown in a medium containing ampicillin. Bacteria with a dilution factor of 10 were confirmed to grow 1 h after inoculation, and observable growth of bacteria with a dilution factor of 10 9 Additionally, Impossible™ Foods incorporated soy LegH protein using the yeast P. pastoris. LegH generated from the nodules of soy plants acts as a plant-based heme molecule. Moreover, LegH has tremendous Fe-binding capabilities, with regulatory functions in oxygen tension and diffusion. The broad-range resemblances of LegH and hemoglobin are red coloration, Fe-binding capability, and flavor [32]. Impossible™ Foods transferred the soy LegH gene into yeast to produce large quantities of LegH protein as sustainably as possible. Production of this ingredient by yeast fermentation has a smaller environmental footprint than digging up soybean root nodules and extracting the protein; however, it is identical to the LegH protein found in such root nodules [41] In addition, the detection and purification of modified LegH were extracted from soybean root nodules. Nodule extracts were first chromatographed on hydroxylapatite at pH 6.8. Over 98% of the peptidases bound to the column and the LegH were recovered in the wash fraction, and subsequent ion-exchange chromatography of the wash fraction at pH 8.0 yielded several fractions [66]. The previous supporting literature indicated a successful extraction of LegH from soy roots [67][68][69].

SDS-PAGE
The results for the extracted pBHA were confirmed by electrophoresis. The basic pBHA sequence size was 1987 bp, the LegH-producing sequence was 528 bp, and when the LegH sequence was inserted into pBHA, the size was 2515 bp. The corresponding size was inferred using the 1 kb reference ladder, the sample could be identified at approximately 2500 bp ( Figure 6).

REVIEW
13 of 18

Color Measurements Pigments
The color coordinates of the extracted pigments for the PBM are listed in Table 2 and pigment weight and concentration are demonstrated in Table 3. Myoglobin, as a control, revealed color coordinates with the L* (6.84), a* (10.95), and b* (6.23) values. Similarly, the Fe-chlorophyll sample showed the highest L* value (22.59), whereas a* exhibited 10.60 and b* value exhibited 26.29; these values were all significantly different (p < 0.05) from the control. Based on the color indices, the L* value of anthocyanin was 3.90 and the a*

Color Measurements Pigments
The color coordinates of the extracted pigments for the PBM are listed in Table 2 and pigment weight and concentration are demonstrated in Table 3. Myoglobin, as a control, revealed color coordinates with the L* (6.84), a* (10.95), and b* (6.23) values. Similarly, the Fe-chlorophyll sample showed the highest L* value (22.59), whereas a* exhibited 10.60 and b* value exhibited 26.29; these values were all significantly different (p < 0.05) from the control. Based on the color indices, the L* value of anthocyanin was 3.90 and the a* value was 5.85, which was the fourth highest value, except for the control. However, LegH showed no significant difference with the control.  The extraction and generation of natural pigments for PBM is a challenging task, despite the availability of different renewable sources and various approaches [70]. A possible reason for this difficulty could be that the optimal pH of the colorant is dissimilar. This limitation of natural colorants is more likely to be overcome by adding acidulants, such as acetic acid, citric acid, or lactic acid [13].
In the current study, myoglobin, Fe-chlorophyll, and anthocyanin extracts were optimized to the level used in PBM products (Figure 7). For instance, some companies, such as Beyond Meat and Light life, use beet juice or powder to imitate a "bleeding" state for their products, whereas Impossible™ Foods uses soy LegH to confer a red meat-like appearance in its burger products [3]. Various food colorants from natural sources have been incorporated into various foods, such as anthocyanins [13,[20][21][22], chlorophyll derivatives [29,30,62,63], and LegH [32,41]. The incorporation of natural colorants into patties were recently reported: the authors synergistically integrated beetroot pigments, laccase, and pectin to enhance the color coordinates of PBM patties [6]. Similarly, decolonization and detoxication of plant-based proteins were reported [11].  Previously, our team optimized and extracted two natural colorants (red yeast rice and lactoferrin), and thereafter synergistically incorporated these colorants into PBM analog patties; this ultimately resulted in the production of PBM analog patties that resembled real meat in many aspects [19].

Absorbance Level
The concentrations of anthocyanin, chlorophyll and LegH extracted were determined at an absorbance of 0.7 at 535 nm. The absorbance at 0.7 was measured at 0.54 g of antho-

Myoglobin
Fe-pheophytin Anthocyanin Leghemoglobin Previously, our team optimized and extracted two natural colorants (red yeast rice and lactoferrin), and thereafter synergistically incorporated these colorants into PBM analog patties; this ultimately resulted in the production of PBM analog patties that resembled real meat in many aspects [19].

Absorbance Level
The concentrations of anthocyanin, chlorophyll and LegH extracted were determined at an absorbance of 0.7 at 535 nm. The absorbance at 0.7 was measured at 0.54 g of anthocyanin, 1.00 g of Fe-chlorophyll, and 1.76 g of LegH. The absorbance measurement spectrum and pigment weight with the concentration per colorant are shown in Figure 8. In addition, ideal extraction conditions were determined for each particular dye by determining the maximum absorbance level or optical density value at a precise absorbance wavelength using a UV-visible absorbance spectrophotometer [33]. By measuring the absorbance of the samples, it was confirmed that the absorbance increased from 400 nm to 535 nm and then decreased after the maximum value of 535 nm. Anthocyanins and Fe-chlorophyll showed a decreasing pattern after a 400 nm wavelength. These results correspond to those previously described by Nelson and Ferruzzi [29], and Furuya, Inoue, and Shirai [59]: both studies confirmed the effective synthesis of Fe-pheophytin derivatives in crude spinach extracts. Moreover, He and Giusti [22] describe in detail the incorporation of anthocyanins as natural colorants into food additives, and their absorbance and toxicity levels. The absorbance of the level of various natural pigments has been reported previously by various authors [18,71,72].

Conclusions
Natural pigments are exceptional sources of bioactive combinations that are consid ered harmless for human consumption. In this study, anthocyanins extracted from blac beans, Fe-chlorophyll derived from spinach, and LegH were produced from microbi culture after the recombination and insertion of a modified plasmid. The effectiveness anthocyanin extraction is the result of a combination of the solvent used and the applie technique. The chromatographic results reveal successful indications of anthocyanin-r lated compounds in the extracted sample from beans. Through microbial culture, th leghemoglobin was successfully synthesized by plasmid insertion and the sequence ind cated negligible homology to toxic agents. The chlorophylls extracted from spinach an their synthesized derivatives including Fe-pheophytins were characterized by the ion e change method. Furthermore, extraction technologies should be optimized to provide e vironmentally feasible and low-cost colorants from traditional and novel sources. The u of natural colorants in food systems is still limited because of technological issues. Ther fore, further investigations are needed to determine how to stabilize these colorants in diverse range of pH and temperatures for use in future food systems.

Conclusions
Natural pigments are exceptional sources of bioactive combinations that are considered harmless for human consumption. In this study, anthocyanins extracted from black beans, Fe-chlorophyll derived from spinach, and LegH were produced from microbial culture after the recombination and insertion of a modified plasmid. The effectiveness of anthocyanin extraction is the result of a combination of the solvent used and the applied technique. The chromatographic results reveal successful indications of anthocyanin-related compounds in the extracted sample from beans. Through microbial culture, the leghemoglobin was successfully synthesized by plasmid insertion and the sequence indicated negligible homology to toxic agents. The chlorophylls extracted from spinach and their synthesized derivatives including Fe-pheophytins were characterized by the ion exchange method. Furthermore, extraction technologies should be optimized to provide environmentally feasible and low-cost colorants from traditional and novel sources. The use of natural colorants in food systems is still limited because of technological issues. Therefore, further investigations are needed to determine how to stabilize these colorants in a diverse range of pH and temperatures for use in future food systems.