Cassamo U.
Mussagy
*ab,
Valéria C.
Santos-Ebinuma
b,
Rondinelli D.
Herculano
bc,
João A. P.
Coutinho
d,
Jorge F. B.
Pereira
*e and
Adalberto
Pessoa
Jr.
a
aDepartment of Pharmaceutical-Biochemical Technology, School of Pharmaceutical Sciences, University of São Paulo, São Paulo, Brazil. E-mail: cassamo.mussagy@unesp.br; jfbpereira@eq.uc.pt
bDepartment of Engineering of Bioprocesses and Biotechnology, School of Pharmaceutical Sciences, São Paulo State University (UNESP), Araraquara 14800-903, São Paulo, Brazil
cTerasaki Institute for Biomedical Innovation (TIBI), 11570 West Olympic Boulevard, Los Angeles, CA 90064, USA
dCICECO-Aveiro Institute of Materials, Department of Chemistry, University of Aveiro, 3810-193, Aveiro, Portugal
eUniv. Coimbra, CIEPQPF, Department of Chemical Engineering, Rua Sílvio Lima, Pólo II – Pinhal de Marrocos, 3030-790, Coimbra, Portugal
First published on 22nd November 2021
In order to replace conventional organic solvents with more benign equivalents such as ionic liquids or eutectic solvents, in this work we attempt to develop a simple and ecofriendly process using these alternative solvents for the extraction of astaxanthin and β-carotene from Phaffia rhodozyma biomass, which could simultaneously be used as a colorant and plasticizer agent for the preparation of bioactive starch-based biodegradable films without further purification. The use of cholinium-based eutectic solvents appears to be a promising solution envisioning the development of functionalized carotenoid-rich biofilms.
Fig. 1 Chemical structures of carotenoids and precursors for the synthesis of ionic liquids and eutectic solvents. |
The development of sustainable, efficient, and eco-friendly extraction platforms has become a critical challenge for the scientific community and the industrial sector. As a consequence, over the last two decades, several studies have been proposing the use of ionic liquids (ILs) and eutectic solvents (ESs) as “ideal solvents” for the extraction of compounds from natural matrices, mainly due to their outstanding solvation properties.17–19 Interestingly, the successful use of biocompatible ILs and ESs containing cation/anion or hydrogen bond donor/acceptor (HBA/HBD), respectively, fully (or partially) derived from low-cost and eco-friendly renewable natural sources has been growing.20 Good examples of more biocompatible and less toxic alternative solvents include the cholinium-based ILs/ESs, whose cation or HBA is derived from the same precursor, i.e., choline, a compound that can be found in nature.21,22 Despite advances in using ILs and ESs as prominent “greener” solvents for the recovery of biomolecules from complex matrices,17,23 there is no clear evidence as to which of these two classes are in fact the “best” extraction solvents. Considering the specificity of each matrix, for efficient solvent selection, fundamental key parameters, i.e., the type of microorganism, biomass type, target compound characteristic and the technical-economic analysis, must also be considered.
In the search for designing alternative, greener and efficient extraction platforms, as well as finding the most efficient ones between ILs and ESs, this work provides a comprehensive study regarding the extraction abilities of cholinium-based ILs and equivalent ES counterparts for the selective recovery of astaxanthin and β-carotene from P. rhodozyma biomass and their application in the production of starch-based films. Therefore, we first evaluated the extraction performance of ILs and ESs synthesized with the same precursors (choline chloride/choline bicarbonate, lactic and butyric acid, cf. respective chemical structures in Fig. 1), comparing respective recovery yields with that of DMSO as a control and those of the precursors. Considering the highest recovery yield of ES composed of cholinium chloride:butyric acid ([Ch]Cl:But), subsequently, the extraction performance was maximized as a function of the HBA:HBD molar ratio, water content and solid–liquid ratio (i.e., solvent-to-biomass ratio or wet cells concentration). Finally, ES extracts containing carotenoids were used as plasticizer agents for the formulation of cornstarch-based biodegradable films, demonstrating the industrial potential of using ESs for creating new functionalized bio-based materials for food packaging and other commercial applications.
The performance of aqueous solutions of two ILs, i.e., choline lactate ([Ch][Lac]) and choline butanoate ([Ch][But]), and two ESs, i.e., choline chloride + lactic acid ([Ch]Cl:Lac) and choline chloride + butyric acid ([Ch]Cl:But) (in molar ratios of 1:1 and 1:2), in the simultaneous extraction of intracellular astaxanthin and β-carotene from wet biomass of P. rhodozyma (at a concentration of 0.2 g mL−1) was screened after 1 h of stirring at 300 rpm and 65 °C (Fig. 2). All experimental protocols and detailed values are given in the Solid–liquid extraction of carotenoids using ILs and ESs section and Table S3 in the ESI,† respectively. Note that all the results in Fig. 2 and Table S3† are expressed as the extraction recovery yield (% w/w) relative to the initial content of carotenoids in the P. rhodozyma wet biomass, i.e., astaxanthin = 7.9 μg mLwet biomass−1 and β-carotene = 25.3 μg mLwet biomass−1.
The results (p ≤ 0.05) depicted in Fig. 2 demonstrate that although the control (DMSO) and aqueous solutions (at 80% w/w) of ILs and ESs allow the recovery, at 65 °C, of at least 10% (w/w) of astaxanthin (red bars) and β-carotene (yellow bars), except for the [Ch]Cl:But (1:1) solution, both carotenoid recovery yields were lower than the control (DMSO). Comparing the performance of ILs and ESs, significant differences (p ≤ 0.05) in the recovery of astaxanthin and β-carotene were found. The highest recovery yields of astaxanthin [47.9 ± 0.8 (w/w)] and β-carotene [46.0 ± 1.0 (w/w)] were achieved using an ES solution of ChCl:But (1:2) (at 80 w/w), with increases of ∼10% (w/w) compared to the control.
Despite the lower recovery yields of other IL/ES aqueous solutions, there seems to be a relationship between the nature of each solvent and its intrinsic capacity to extract each pigment. Specifically, the extraction of astaxanthin (red bars) using ESs was favoured by [Ch]Cl:But (1:2), while [Ch]Cl:Lac (1:1/1:2) always exhibited high selectivity for β-carotene (yellow bars). Both ILs also revealed high selectivity for β-carotene recovery. The change in the carotenoid selectivity using an ES solution of 80% (w/w) of [Ch]Cl:But seems to be a result of the increase in the relative hydrophobicity of HBD (logKow of But = 0.78).24 In this case, since the hydroxyl and keto groups present in the astaxanthin molecule give it a complex polar-nonpolar-polar nature,25 the most suitable solvent for the recovery of astaxanthin from wet cells of P. rhodozyma is a mixture of polar (i.e., [Ch]Cl and water) and non-polar (i.e., butyric acid) components, that is, solvents with an “amphiphilic nature” seem to be effective solubilizing agents of wet biomass, as previously demonstrated by our research group.8
Previously some researchers reported that the pH and viscosity of solutions are crucial for the recovery of carotenoids.17,26 To infer these effects, the pH values and viscosities of each IL and ES solution were determined. As depicted in Fig. 2, there is no significant influence (p ≤ 0.05) of the pH parameter in the recovery of astaxanthin and β-carotene. The highest recoveries of both carotenoids were achieved using [Ch]Cl:But (1:2), a solution with a pH value of 1.68, pH values similar to or lower than those of IL/ES solutions with a weaker extraction performance (i.e., pH values similar to [Ch]Cl:Lac at 1:1 and 1:2 and lower than those of [Ch][Lac] and [Ch][But] solutions). On the other hand, regarding the viscosity, it is widely recognized that the high viscosity of solutions limits the mass-transfer process, hindering the effect of solvents on the membrane permeation and consequently reducing the recovery of carotenoids from the intracellular environment.17 The results presented in Fig. 2 confirm a viscosity dependence, since the lowest recovery yields were obtained with more viscous ES solutions (i.e., [Ch]Cl:Lac), while the highest recovery was achieved with [Ch]Cl:But, which has a viscosity of 9.99 mPa s at 65 °C (about half the viscosity of [Ch]But). However, despite some influence of viscosity on recovery abilities, it is important to note that the highest recovery yields are mainly a result of the high relative hydrophobicity of the solvent, i.e., the more hydrophobic ES solutions ([Ch]Cl:But) (as well as DMSO) have the greatest capacity to recover astaxanthin and β-carotene from P. rhodozyma wet cells (Fig. 2).
The importance of HBD as a key parameter to control carotenoid recovery yields was revealed. Thus, to gain a deeper understanding of the relative influence of the HBA:HBD molar ratio, the extraction of astaxanthin and β-carotene using concentrated (80% w/w) solutions of different [Ch]Cl:But ESs (molar ratios of 1:2, 1:3, 1:4, 1:5, 1:6, 1:7 and 1:8) was performed. This experimental step was carried out during 1 h of stirring at 300 rpm and 65 °C, using a concentration of 0.2 g mL−1of P. rhodozyma wet biomass (yeast cells). The respective recovery yields (% w/w) of astaxanthin and β-carotene are shown in Fig. 3A and detailed in Table S4 in the ESI.†
As shown in Fig. 3A, the effect of the HBA:HBD molar ratio of the concentrated [Ch]Cl:But solution was analysed, specifically by evaluating how the increase in HBD alters the selective recovery of carotenoids. Interestingly, there is a clear effect of the HBA:HBD molar ratio not only on increasing the extraction performance, but also on the selectivity of carotenoids. Regarding the extraction performance, the recovery yield was maximized by increasing the HBA:HBD molar ratio up to 1:5, achieving values of 82.2% ± 0.4 and 81.8% ± 0.3 (w/w) for β-carotene and astaxanthin, respectively. Subsequent increases of HBD concentration (i.e., 1:6, 1:7 and 1:8) were not favourable, particularly for astaxanthin recovery. Despite similar selectivity performance from 1:2 to 1:4 molar ratio (recovery of β-carotene > astaxanthin), Fig. 3-A demonstrates a change in the carotenoid extraction profile at a molar ratio of 1:5 (recovery of β-carotene = astaxanthin, p ≤ 0.05), from which an increase of HBD (from 1:6 to 1:8 molar ratio) turns the ES solution more selective for recovery of β-carotene. Together, these results confirm that the increase in the relative hydrophobicity of ESs favours the extraction of the more hydrophobic carotenoid (β-carotene), while an amphiphilic character of ES is desirable to maximize the recovery yields of both intracellular carotenoids from wet yeast biomass.
Concerning water influence (only evaluated for [Ch]Cl:But 1:5), as shown in Fig. 3B (detailed values in Table S5 in the ESI†), the recovery yields linearly decreased with the increase of water concentration in the [Ch]Cl:But solution, achieving decreases of 50% in the extraction performance with a solution with 50% of water. As expected, the treatment with concentrated [Ch]Cl:But confers a more hydrophobic character to the solvent enhancing the solubilization of the carotenoids and the weakening of the cell structure. At this point, it is important to emphasize the importance of balancing diffusion and solubility, i.e., although more diluted solutions are less viscous benefiting the diffusion of the carotenoids, the negative influence on increasing the solvent polarity confirms the importance of ensuring an adequate hydrophobicity of the solvent for the solubilization of the two carotenoids.
From an industrial perspective, the solid–liquid ratio (SLR) is also a crucial parameter for designing cell disrupting procedures to recover intracellular molecules from microbial biomass. Therefore, to identify the optimal SLR (i.e., P. rhodozyma wet biomass/ES solution volume), the ES solution with higher extraction abilities ([Ch]Cl:But at 1:5 molar ratio) at five different concentrations (0.05, 0.10, 0.20, 0.50 and 1.00) were tested to maximize the recovery of astaxanthin and β-carotene. As shown Fig. 3C (detailed values are listed in Table S6 in the ESI†), the concentration of both carotenoids (astaxanthin and β-carotene) increased up to a concentration of 0.2 g mL−1 of P. rhodozyma wet cells, from which the extraction abilities decreased due to the solubility limits of carotenoids in the ES solution.
A general analysis of Fig. 3 confirms the potential of ILs and ESs (mainly But-based ones) as “greener” solvents for the recovery of intracellular carotenoids from microbial matrices, with clear indications that the use of eutectic mixtures is preferable. In sequence, to obtain a comprehensive understanding of the extraction performance of the [Ch]Cl:But ES solution, equivalent IL ([Ch][But]) and respective ES precursor ([Ch]Cl and butyric acid), aqueous solutions of these components were used for the selective recovery of astaxanthin and β-carotene from P. rhodozyma biomass. Note that in this set of experiments, all solutions were prepared at 50% (w/w) due to the solubility limits of [Ch]Cl, while all other conditions were similar to those of previous experiments (a wet cell concentration of 0.2 g mL−1, in 1 h of stirring at 300 rpm, and 65 °C). As shown in Fig. 4 (detailed values are provided in Table S7 in the ESI†), the aqueous solution of [Ch]Cl, the most hydrophilic, has the lowest extraction ability (5.0% ± 0.3 and 6.4% ± 0.5 of astaxanthin and β-carotene, respectively), with intermediate yields (about 15% for both carotenoids) being obtained for butyric acid and [Ch][But] solutions, and the highest astaxanthin and β-carotene recovery yields achieved with the [Ch]Cl:But (1:5) aqueous solution. These results confirm that more than the influences of pH and viscosity (no relationship of these parameters with recovery yields) is the nature of the [Ch]Cl:But ES, which is the key for maximizing the recovery of astaxanthin and β-carotene. In fact, the increase in the extraction abilities using ESs in aqueous media seems to be a result of a hydrotropic mechanism (ability to boost the aqueous solubility of hydrophobic solutes23) of [Ch]Cl:But (1:5). Recent experimental works have confirmed that hydrotropy is a key mechanism behind the outstanding ability of ESs to solubilize/extract hydrophobic molecules.27,28 Furthermore, as microbial biomass is “wet”, the presence of a minimal amount of water is crucial to aid in the extraction of intracellular molecules. Recently, we8,15 demonstrated that the presence of water in solvent mixtures not only facilitates the miscibility of solvents with the wet biomass but also helps in permeabilization/disintegration of the yeast cell wall.
As noted above, the results depicted in Fig. 4 reinforce the notion that pH and even viscosity does not have (or have a low) influence (p ≤ 0.05) on the recovery of astaxanthin and β-carotene. For example, aqueous solutions of [Ch]Cl (pH = 4.06 and viscosity = 0.84 mPa s at 65 °C) and But (pH = 2.88 and viscosity = 1.07 mPa s at 65 °C) do not extract the same carotenoids as [Ch]Cl:But (1:5) (pH = 1.61 and viscosity = 1.30 mPa s at 65 °C) even with similar acidity values and lower viscosity.
To implement a circular and sustainable industrial ES-based extraction platform, the recovery of astaxanthin and β-carotene from [Ch]Cl:But (1:5) and solvent recycling arise as the most problematic/challenging issues, mainly due to the low vapour pressure of [Ch]Cl and But.29 Fortunately, there are successful studies for the polishing of carotenoids and recovery of the ES and IL solutions, including, for example, three-phase partition systems induced by precipitation (with antisolvents or strong salting-out salts)17 or using adsorptive membranes.30 However, in this particular case, due to the naturally biocompatible character of ES and excellent biological activities of carotenoids, we propose to combine both as green alternatives to obtain new commercial bioproducts/biomaterials. Therefore, due to the acidic character of the [Ch]Cl:But solution rich in carotenoids, it can be used directly (“one-pot”) as a green plasticizer solution for biopolymers, e.g. chitosan or starch, and can produce environment friendly biodegradable active packaging films with high antioxidant activities.31 The process integration of the proposed technology for the recovery of astaxanthin and β-carotene from yeast biomass using [Ch]Cl:But (1:5) with a subsequent formulation of corn-starch biofilms (CS) was here attempted. The [Ch]Cl:But (1:5) extract rich in astaxanthin and β-carotene was mixed with corn-starch according to the methodology described in the Preparation of biofilms section in the ESI.†
As depicted in Fig. 5A and B, a CS-based thermoplastic formulation was efficiently produced using [Ch]Cl:But-rich carotenoids as green plasticizers. The FTIR spectra (Fig. S2 in the ESI†) obtained from the CS/[Ch]Cl:But and CS/[Ch]Cl:But-rich carotenoids showed almost the same pattern. The strong peak (broad band at 3304 cm−1) observed in the biofilm spectrum corresponds to the hydrogen bonds formed by the hydroxyl groups of starch and ES. The tensile strength, elasticity modulus (Young's modulus) and elongation at break of the CS-based biofilms were determined as the main mechanical properties and to estimate the resistance to tensile ruptures. The values of the elasticity modulus (2.6 ± 0.1 MPa), tensile strength (1.0 ± 0.2 MPa) and elongation at break (70.0 ± 3.1%) show the material to be less rigid for the CS:ES ratio tested, supporting the plasticizer character of the solution. These values are in line with those obtained by Ibrahim et al.32 using fructose, sorbitol and urea as green plasticizers for the formulation of CS-based films.
Finally, in order to evaluate if the proposed integrated approach to produce CS-based biofilms is safe and does not affect the biological activity of astaxanthin and β-carotene, the antioxidant activity of the formulated biofilms was evaluated. For that purpose, the DPPH˙ radical scavenging method was performed to estimate the antioxidant activity of biofilms using electron paramagnetic resonance (EPR) spectroscopy, which directly measures the concentration of the DPPH˙ radical. As observed in Fig. 5C, the DPPH˙ solution exhibited a decrease in the EPR signal intensity after the addition of biofilms containing two-different concentrations of carotenoid-rich extracts, confirming that the biofilms preserve the original antioxidant properties of astaxanthin and β-carotene. Biofilms containing carotenoids revealed encouraging results in the DPPH˙ radical scavenging assay, with a percentage inhibition increased with the increase of the concentration of the extracts in biofilms, in which the CS-based biofilm containing 4 μg mL−1 of astaxanthin and 15 μg mL−1 of β-carotene [biofilm + extract 2 (Fig. 5C)] allowed the highest antioxidant activity (99%).
These results confirm that carotenoid-rich biofilms exhibit powerful antioxidant activities, which can be further used in the formulation of biodegradable active packaging films for food, pharmaceutical, nutraceutical and cosmeceutical purposes.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: 10.1039/d1gc03521e |
This journal is © The Royal Society of Chemistry 2022 |