Abstract
L-asparaginase has been used in the remission of malignant neoplasms such as acute lymphoblastic leukemia. The search for new sources of this enzyme has become attractive for therapeutics. Traditional methods for biomolecule purification involve several steps. A two-phase system may be a good strategy to anticipate one of these stages. This study aimed to produce and purify a fungal L-asparaginase through an aqueous two-phase micellar system (ATPMS) using Triton X-114. The fungus Penicillium sp.–encoded 2DSST1 was isolated from Cerrado soil. Plackett-Burman design followed by a 24 full factorial design was used to determine the best conditions to produce L-asparaginase. The evaluated variables were L-asparagine, L-proline, wheat bran, potato dextrose broth, ammonium sulfate, yeast extract, sucrose and glucose concentrations, incubation temperature, incubation period, and initial pH of the culture medium. L-asparaginase quantification was valued by the formation of β-aspartyl hydroxamate. The significant positive variables, L-asparagine, L-proline, potato dextrose broth, and sucrose concentrations, were evaluated at 2 levels (+ 1 and − 1) with triplicate of the central point. After 34 runs, maximum activity (2.33 IU/mL) was achieved at the factorial design central point. A central composite design was performed in ATPMS at two levels (+ 1 and − 1) varying Triton X-114 concentration (w/v), separation phase temperature, and crude extract concentration (w/v). The L-asparaginase partition coefficient (K) was considered the experimental design response. Out of the 16 systems that were examined, the most promising presented a purification factor of 1.4 and a yield of 100%.
Keywords: L-Asparaginase, ATPMS, Plackett-Burman design, Purification, Agro-industrial residue
Introduction
L-asparaginase (ASNase), L-asparagine amidohydrolase, EC 3.5.1.1, is an enzyme that selectively catalyzes the hydrolysis of L-asparagine in ammonia and L-aspartic acid. The clinical use of this enzyme is based on the fact that a metabolic difference exists between normal and cancer cells. Normal cells do not depend on extracellular L-asparagine since they can produce this amino acid by the activity of L-asparagine synthetase, which does not occur in certain types of lymphomas [1]. In 1962, ASNase was included in the treatment of acute lymphoblastic leukemia (ALL).
ALL is a malignant lymphocytic neoplasm characterized by the accumulation of immature T and B cells in the bone marrow [2]. This disease has a peak incidence in children between the ages of 2 and 5 years old. However, it accounts for only about 20% of acute leukemias diagnosed in adults [2, 3].
Escherichia coli–derived L-asparaginase is administered at doses of 6000 to 10,000 IU every 3 days for 3 to 4 weeks via intramuscular or intravenous route. PEG-ASNase, Pegaspargase, is an enzyme conjugated to polyethylene glycol polymers. It can be produced on an industrial scale, and is administered in doses of less than 5000 IU, generally every 14 days [4]. The combination of L-asparaginase, corticosteroids, and chemotherapeutics agents has been shown to be effective for remission of the disease [5].
There are three forms of treatment for ALL using L-asparaginase in clinical practice: the native enzyme derived from Escherichia coli, the pegylated enzyme, and an isolated ASNase from Erwinia chrysanthemi [6–8]. Treatment with bacterial L-asparaginase has produced some important side effects, such as neurological crisis, anaphylaxis, coagulation disorders, leukopenia, thrombosis, pancreatitis, hyperglycemia, and hepatotoxicity [2]. Many of these effects could be attributed to bacterial structures and to a partial glutaminase activity presented by some L-asparaginases that depletes glutamine plasma concentration causing a toxic environment for healthy cells. In this way, it is important to find an L-asparaginase with low glutaminase activity [9, 10].
Traditional methods for biomolecule purification involve several steps such as chromatography, dialysis methods, ultrafiltration, or even non-chromatographic methods. However, the use of a two-phase system may be a good strategy to anticipate one stage of this process [11]. Actually, the production of L-asparaginase of eukaryotic origin is of great importance, due to the possibility decreased adverse effects in patients when compared with the drug from prokaryotes. The search for new sources of L-asparaginase for the treatment of ALL has become increasingly interesting. The use of alternative methods of cultive and purification aimed at reducing the production costs of this enzyme is also a relevant point to be observed.
Aqueous two-phase system (ATPS), a liquid-liquid fractionation method, has been recognized as a versatile and well-established process for separation. Aqueous two-phase micellar system (ATPMS) utilizes a binary phase micellar system with the help of surfactants at low concentrations in aqueous solutions to induce micelles. Once the temperature of the surfactant solution reaches a cloudy point temperature, two phases are formed in ATPMS due to the difference between the physicochemical environments. When the temperature is above the cloudy point, the surfactant molecules self-aggregate to maintain an aqueous (surfactant-depleted) phase and result into two-phase formation, i.e., in a bottom, micelle-rich phase and a top, micelle-poor phase. [12]. The partitioning of proteins (enzymes) are governed primarily by steric, repulsive, and excluded-volume interactions between the globular hydrophilic proteins and the non-charged cylindrical micelles, which drive desired biomolecules preferential into the bottom micelle-rich phase while simultaneously eliminating most of the contaminants (proteins) to the micellar-poor, top phase, due to excluded-volume interactions between the contaminates and micelles. The formation of micelles represents a delicate balance of several intermolecular forces including those of the hydrophobic, van der Waals, steric, and electrostatic types. The resulting micelles are labile microstructures formed by the non-covalent binding of individual surfactant molecules, contrary to polymers, which are formed by the covalent binding of individual monomers [13]. In addition, micellar size and shape, as well as the associated micellar solution properties, can be tuned in situ by varying solution conditions, such as temperature, surfactant concentration, salt type, and concentration. [13]. The enzyme’s behavior in an aqueous two-phase micellar system can be defined by the partition coefficient (K), which is determined by the ratio of the enzyme activity found in the micelle-poor phase and the micelle-rich phase. The partition of a biomolecule may also be more selective when a mixed micelle system, consisting of charged surfactants or surfactant with affinity ligands groups, is incorporated to nonionic surfactants [11, 14, 15]. According to Liu, when comparing works using ATPMS and ATPS, some advantages of the micellar system are observed in front of the polymeric systems: (1) The micelle characteristics such as size, shapes, and ionic charge aid in optimizing purification; (2) micelles provide hydrophilic and hydrophobic conditions to the system allowing partition selectivity; (3) it is possible to use mixed micelles and surfactant-type affinity ligand to aim a specific target. These advantages do not prevent the use of ATPS for purification of proteins [16]. ATMPS has been already described in purification and concentration of molecules such as albumin [17], antibiotics [18, 19], enzymes [20, 21], and organic compounds [22].
The present study aims the production and partial purification of a fungal L-asparaginase using wheat bran as agro-industrial residue and ATPMS as one step for L-asparaginase purification.
Material and methods
Reagents
The culture media and enzymatic assay reagents such as L-asparagine, L-proline, yeast extract, and peptone were purchased from Sigma-Aldrich ®. The salts were purchased from Dinâmica®. The residue, wheat bran, was obtained after the processing of wheat in a stage in the west lake region in Brasília, Federal District, Brazil.
Microorganism culture
The filamentous fungus strain used in this study was isolated from Brazilian Cerrado soil and is being kept by the Laboratory of Enzymology of the University of Brasilia, Brazil. The strain is deposited in Embrapa Genetic Resources and Biotechnology Collection under the accession number 2DSST1. The isolation procedure of the filamentous fungus was performed using serial dilution plating method in Petri dishes. The fungus used in this study was maintained on plates containing potato dextrose agar medium at 28 °C.
Preparation of the wheat bran residue
Wheat bran, an agro-industrial residue, was used as the substrate for the Penicillium sp. fermentation. It was autoclaved for 2 h at 121 °C and 1 atm, as this process is important to break lignocellulosic structures. After that, it was washed successively with tap water and then dried at 60 °C for 48 h.
Screening of variables using Plackett-Burman design
Submerged fermentation was carried out in 250-mL Erlenmeyer flasks containing 50 mL of a Czapek-Dox medium (0.15% K2HPO4, 0.052% KCl, 0.052% MgSO4, 0.001% FeSO4, 0.001% ZnSO4, 0.001% CuSO4) (w/v) and different concentrations of L-asparagine (X1), L-proline (X2), wheat bran residue (X3), potato dextrose broth (X4), ammonium sulfate (X5), yeast extract (X9), sucrose (X10), and glucose (X11). The incubation temperature (X6), incubation time (X7), and initial pH (X8) were also evaluated. All the cultures were inoculated with 3.7 × 107 spores. Plackett-Burman design (PBD) was proposed to identify the most important variables in L-asparaginase production in this model. Variables were evaluated in two levels with a triplicate at the central point, as shown in Table 1. After autoclave, all experiments were carried out in a temperature-controlled orbital shaker at 120 rpm. The PBD was predicted using software Design Expert® 7.0.
Table 1.
Experimental design variables and levels
| Independent variables | Symbol | Levels and values | ||
|---|---|---|---|---|
| − 1 | 0 | + 1 | ||
| L-asparagine (%) | X1 | 0.40 | 0.90 | 1.40 |
| L-proline (%) | X2 | 0.50 | 1.75 | 3.00 |
| Residue (%) | X3 | 1.00 | 3.00 | 5.00 |
| Potato dextrose broth (%) | X4 | 1.00 | 3.00 | 5.00 |
| Ammonium sulfate (%) | X5 | 0.50 | 1.50 | 2.50 |
| Incubation temperature (°C) | X6 | 25 | 30 | 35 |
| Incubation period (h) | X7 | 48 | 72 | 96 |
| Initial pH | X8 | 5.0 | 7.0 | 9.0 |
| Yeast extract (%) | X9 | 0.50 | 1.00 | 1.50 |
| Sucrose (%) | X10 | 0.20 | 0.70 | 1.20 |
| Glucose (%) | X11 | 0.20 | 0.70 | 1.20 |
After the incubation period, culture medium was filtered and the fungal biomass was kept at − 80 °C until analysis, when it was grounded using a mortar and pestle. The macerated biomass was mixed with buffer solution (50 mM Tris-HCl, pH 8.6) and centrifuged and the supernatant was used as crude enzyme extract.
Response surface methodology
After the PBD, a 24 factorial design was used to optimize the production process. After PDB evaluation, variables with positive significance (L-asparagine, L-proline, potato dextrose broth, and sucrose) were maintained in the optimization process and the negative significative variables (yeast extract and wheat bran residue) were kept at their lowest level. Variables with no significance were excluded from the design. The 24 factorial design was predicted using software Design Expert® 7.0.
L-asparaginase assay
The L-asparaginase assay was carried out as described by Drainas et al. [23]. A total of 500 μL of crude extract containing L-asparaginase were incubated for 30 min with 200 μL of 0.1 M L-asparagine, 200 μL of 1 M hydroxylamine hydrochloride, and 1.1 mL of 50 mM Tris-HCl buffer, pH 8.6. The reaction was stopped with the addition of 500 μL of a solution containing 10% trichloroacetic acid, 0.66 M HCl, and ferric chloride 10% (w/v). Afterwards, the samples were centrifuged at 5000 rpm for 5 min at 4 °C. The absorbance was measured in the supernatant at 500 nm. Sample blank was performed without the addition of L-asparagine and hydroxylamine hydrochloride solutions.
Protein assay
The concentration of total proteins in the samples was quantified using Pierce™ BCA Protein Assay Kit. A sample of 50 μL was incubated with 200 μL of BCA Protein Assay Kit mixture for 30 min at 37 °C; after this, the absorbance was measured at 420 nm. A standard curve of bovine serum albumin was performed in varying concentrations of 2 to 2000 μg/mL.
Aqueous two-phase micellar system
The ATPMS was constructed through different concentrations of Triton X-114 and crude extract. The total weight of the systems was 3 g. A central composite design (CCD) was proposed in two factorial levels (+ 1 and − 1) and two axial levels (− 1.68 and + 1.68) with a triplicate at the central point to vary the concentrations of surfactant (w/v), crude extract (w/v), and temperature (°C) to analyze the best conditions for L-asparaginase purification using this system as shown in Table 2.
Table 2.
Experimental central composite design variables and levels for micellar aqueous two-phase system
| Independent variables | Symbol | Levels and values | ||||
|---|---|---|---|---|---|---|
| − 1.68 | − 1 | 0 | 1 | 1.68 | ||
| Temperature (°C) | Y1 | 28 | 29 | 0.90 | 31 | 32 |
| Triton X-114 (%) | Y2 | 0.63 | 2 | 4 | 6 | 7.4 |
| Crude extract (%) | Y3 | 6.3 | 20 | 40 | 60 | 73.6 |
A Triton X-114/Tris-HCl buffer coexistence curve was established according to the mist point [24]. Systems were subjected to established temperatures in a water bath for 1 h, and then remained at room temperature for 30 min until complete separation of phases. Phases were then separated and submitted to enzymatic activity and total protein assays.
Calculation of the partition coefficient, mass balance, yield, purification factor, and selectivity
The partition coefficient (K) was determined through the enzymatic activities defined for each phase of the system:
Ac = L-asparaginase activity in rich micelles phase
Ad = L-asparaginase activity in poor micelles phase
Mass balances (MB) were calculated to identify possible losses of the enzyme during the process. The equation below determines how the mass balances were calculated:
where A0 refers to the activities of L-asparaginase in the fermented broth initially added to the system, Ac and Ad refer to the activities in the phases with high concentration of micelles and low concentration of micelles, respectively. Vc, Vd, and V0 are the volumes of the concentrate phase (c), dilute phase (d), and the stock solution (0) initially added to the system, respectively.
The yield (η) was determined from the equation below. The yield resulting from the liquid-liquid partition of L-asparaginase is critical for further industrial application.
Vt = volume of the L-asparaginase phase was extracted in the highest proportion; Art = enzymatic activity in the phase where there is a higher proportion of extraction. V0 = volume of broth subjected to extraction; A0 = enzymatic activity of the original sample.
The purification factor (PF) was calculated by the ratio of the specific activity (IU/mg protein) in the diluted phase and in the fermented broth.
Aed = specific activity (IU/mg protein) in the concentrated phase.
Ae0 = specific activity (IU/mg protein) in the fermented broth.
Statistical analysis
All statistical analyses were performed using the software Design Expert ® 7.0. Enzymatic assays were carried out in triplicates and the average of the samples was used to calculate the amount of L-asparaginase or protein concentration.
Results
Submerged fermentation and optimization
The PBD carried out 15 runs to determine the significant variables. The highest activity was found on run 1 (Table 3). The ANOVA (Table 4) revealed the significance (p < 0.05) with positive effect to L-asparagine, L-proline, potato dextrose broth, and sucrose. Significant variables with negative effect (residue, yeast extract, and glucose) were added in the factorial design at its lowest value.
Table 3.
Plackett-Burman design for L-asparaginase production in submerged culture
| Run | X1 | X2 | X3 | X4 | X5 | X6 | X7 | X8 | X9 | X10 | X11 | L-asparaginase activity (UI/mL) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 1.4 | 3.00 | 1.0 | 5.0 | 2.5 | 35 | 48 | 5.0 | 0.5 | 1.2 | 0.2 | 1.27 |
| 2 | 0.4 | 3.00 | 5.0 | 1.0 | 2.5 | 35 | 96 | 5.0 | 0.5 | 0.2 | 1.2 | 0.15 |
| 3 | 1.4 | 0.50 | 5.0 | 5.0 | 0.5 | 35 | 96 | 9.0 | 0.5 | 0.2 | 0.2 | 0.33 |
| 4 | 0.4 | 3.00 | 1.0 | 5.0 | 2.5 | 25 | 96 | 9.0 | 1.5 | 0.2 | 0.2 | 0.49 |
| 5 | 0.4 | 0.50 | 5.0 | 1.0 | 2.5 | 35 | 48 | 9.0 | 1.5 | 1.2 | 0.2 | 0 |
| 6 | 0.4 | 0.50 | 1.0 | 5.0 | 0.5 | 35 | 96 | 5.0 | 1.5 | 1.2 | 1.2 | 0.14 |
| 7 | 1.4 | 0.50 | 1.0 | 1.0 | 2.5 | 25 | 96 | 9.0 | 0.5 | 1.2 | 1.2 | 0.59 |
| 8 | 1.4 | 3.00 | 1.0 | 1.0 | 0.5 | 35 | 48 | 9.0 | 1.5 | 0.2 | 1.2 | 0.81 |
| 9 | 1.4 | 3.00 | 5.0 | 1.0 | 0.5 | 25 | 96 | 5.0 | 1.5 | 1.2 | 0.2 | 0.43 |
| 10 | 0.4 | 3.00 | 5.0 | 5.0 | 0.5 | 25 | 48 | 9.0 | 0.5 | 1.2 | 1.2 | 0.41 |
| 11 | 1.4 | 0.50 | 5.0 | 5.0 | 2.5 | 25 | 48 | 5.0 | 1.5 | 0.2 | 1.2 | 0.09 |
| 12 | 0.4 | 0.50 | 1.0 | 1.0 | 0.5 | 25 | 48 | 5.0 | 0.5 | 0.2 | 0.2 | 0.06 |
| 13 | 0.9 | 1.75 | 3.0 | 3.0 | 1.5 | 30 | 72 | 7.0 | 1.0 | 0.7 | 0.7 | 0.29 |
| 14 | 0.9 | 1.75 | 3.0 | 3.0 | 1.5 | 30 | 72 | 7.0 | 1.0 | 0.7 | 0.7 | 0.36 |
| 15 | 0.9 | 1.75 | 3.0 | 3.0 | 1.5 | 30 | 72 | 7.0 | 1.0 | 0.7 | 0.7 | 0.37 |
Table 4.
Analysis of variance (ANOVA) and effect for PBD for L-asparaginase production in wheat bran
| Source | p value | Effect |
|---|---|---|
| Model | 0.0079 | - |
| X1: L-asparagine | 0.0021 | 0.38 |
| X2: L-proline | 0.0018 | 0.39 |
| X3: residue | 0.0033 | − 0.32 |
| X4: potato dextrose broth | 0.0539 | 0.04 |
| X5: ammonium sulfate | 0.1768 | 0.066 |
| X6: temperature | 0.0698 | 0.10 |
| X7: time | 0.1101 | − 0.084 |
| X8: pH | 0.1176 | 0.081 |
| X9: yeast extract | 0.0322 | − 0.14 |
| X10: sucrose | 0.0273 | 0.15 |
| X11: glucose | 0.1850 | − 0.064 |
| Lack of fit | 0.1642 | - |
The variables L-asparagine, L-proline, potato dextrose broth, and sucrose presented p value < 0.05 and positive effect (Table 4). They were selected for the continuation of the optimization process. The variables that did not express significance (p > 0.05) were excluded from the experimental design. However, for the significant variables that had a negative effect, when the proposed values decrease, there is a tendency to increase the response. They were kept at the minimum level until the end of the optimization.
There were 19 runs (Table 5) with 24 factorial design (16 factor points, 3 central points). The best response was found at the central point of the optimization design (run 17, 2.33 IU/mL). The L-asparaginase activity found in factorial design was approximately 1.8 times higher when compared with the greater activity found in the Plackett-Burman design (Table 3).
Table 5.
24 factorial planning matrix for L-asparaginase production
| Run |
X1 (% p/v) |
X2 (% p/v) |
X4 (% p/v) |
X10 (% p/v) |
Experimental L-asparaginase activity | Predicted L-asparaginase activity |
|---|---|---|---|---|---|---|
| 1 | 1.4 | 3.0 | 0.5 | 0.0 | 0.06 | 0.20 |
| 2 | 4.2 | 3.0 | 0.5 | 0.0 | 0.70 | 0.72 |
| 3 | 1.4 | 5.0 | 0.5 | 0.0 | 0.91 | 0.80 |
| 4 | 4.2 | 5.0 | 0.5 | 0.0 | 1.05 | 0.98 |
| 5 | 1.4 | 3.0 | 1.0 | 0.0 | 0.75 | 0.77 |
| 6 | 4.2 | 3.0 | 1.0 | 0.0 | 1.48 | 1.28 |
| 7 | 1.4 | 5.0 | 1.0 | 0.0 | 1.18 | 1.11 |
| 8 | 4.2 | 5.0 | 1.0 | 0.0 | 1.04 | 1.27 |
| 9 | 1.4 | 3.0 | 0.5 | 0.2 | 0.91 | 0.89 |
| 10 | 4.2 | 3.0 | 0.5 | 0.2 | 1.87 | 1.72 |
| 11 | 1.4 | 5.0 | 0.5 | 0.2 | 1.36 | 1.33 |
| 12 | 4.2 | 5.0 | 0.5 | 0.2 | 1.64 | 1.83 |
| 13 | 1.4 | 3.0 | 1.0 | 0.2 | 1.17 | 1.02 |
| 14 | 4.2 | 3.0 | 1.0 | 0.2 | 1.52 | 1.84 |
| 15 | 1.4 | 5.0 | 1.0 | 0.2 | 1.01 | 1.20 |
| 16 | 4.2 | 5.0 | 1.0 | 0.2 | 2.05 | 1.68 |
| 17 | 2.8 | 4.0 | 0.75 | 0.1 | 2.33 | 2.07 |
| 18 | 2.8 | 4.0 | 0.75 | 0.1 | 2.05 | 2.07 |
| 19 | 2.8 | 4.0 | 0.75 | 0.1 | 1.83 | 2.07 |
After 19 runs, the maximum activity of L-asparaginase was determined under the proposed conditions. L-asparaginase activity was calculated by the model equation proposed:
Prediction L-asparaginase activity = 2.07 + 0.25X1 + 0.11X2 + 0.11X4 + 0.27X10–0.085X1X2–0.0025X1X4 + 0.079X1X10–0.066X2X4–0.037X2X10–0.11X4X10–0.90X12. R2 = 0.9915. Level of significance of 95%.
The surface response graph (Fig. 1) as a function of the significant variables, L-asparagine and sucrose, presents an optimal region for L-asparaginase activity.
Fig. 1.
Surface response graph for L-asparaginase activity as a function of the variables X1 (L-asparagine) and X10 (sucrose) using wheat bran
Aqueous micellar two-phase system
The partition coefficient (K) of L-asparaginase was considered the dependent variable (response) of the experimental design (Table 6). Also, the purification factor and yield were evaluated to choose the best conditions. Values obtained in the experimental design for enzyme partition coefficient (K), mass balance (MB), yield (η), and purification factor (Pf) are shown in Table 7.
Table 6.
Analysis of variance (ANOVA) for factorial design 24 for L-asparaginase production of L-asparaginase in wheat bran
| Source | p value |
|---|---|
| Model | 0.0191 |
| X1: L-asparagine | 0.0312 |
| X2: L-proline | 0.1160 |
| X4: potato dextrose broth | 0.1253 |
| X10: sucrose | 0.0067 |
| X1X2 | 0.5464 |
| X1X4 | 0.7136 |
| X1X10 | 0.2113 |
| X2X4 | 0.2509 |
| X2X10 | 0.4182 |
| X4X10 | 0.1173 |
| X12 | 0.0024 |
| Lack of fit | 0.3707 |
Table 7.
Central composite design of liquid-liquid L-asparaginase extraction
| Systems | Real values | ||||||
|---|---|---|---|---|---|---|---|
| Y1 | Y2 | Y3 | K | MB | η | Pf | |
| 1 | 29 | 2 | 20 | 0.31 | 78.9 | 60 | 0.2 |
| 2 | 31 | 2 | 20 | 0.33 | 75.4 | 57 | 0.3 |
| 3 | 29 | 6 | 20 | 0.21 | 0.4 | 38 | 0.6 |
| 4 | 31 | 6 | 20 | 0.32 | 56.8 | 43 | 1.0 |
| 5 | 29 | 2 | 60 | 0.37 | 88.2 | 64 | 0.7 |
| 6 | 31 | 2 | 60 | 0.21 | 103.6 | 85 | 0.7 |
| 7 | 29 | 6 | 60 | 0.09 | 70.2 | 64 | 0.8 |
| 8 | 31 | 6 | 60 | 0.07 | 100 | 100 | 1.4 |
| 9 | 28* | 4 | 40 | 0.57 | 2.9 | 1.8 | 0.0 |
| 10 | 32* | 4 | 40 | 0.60 | 1.2 | 0.7 | 0.0 |
| 11 | 30 | 7.4* | 40 | 0.19 | 1.7 | 1.5 | 0.0 |
| 12 | 30 | 4 | 6.3* | - | - | 0 | - |
| 13 | 30 | 4 | 73.6* | 0.09 | 52.74 | 48.4 | 0.7 |
| 14 | 30 | 4 | 40 | 0.06 | 10.0 | 9.5 | 0.1 |
| 15 | 30 | 4 | 40 | 0.05 | 10.46 | 9.9 | 0.1 |
| 16 | 30 | 4 | 40 | 0.05 | 9.6 | 9.1 | 0.0 |
Y1, temperature; Y2, concentration of Triton X-114; Y3, concentration of the crude extract; K, enzyme partition coefficient; MB, mass balance; η, yield; Pf, purification factor
*Axial points
In the individual analysis of each system, system 8 presented a purification factor (Pf) of 1.4, in addition to a low partition coefficient, demonstrating that L-asparaginase migrated preferentially to the micelle-poor phase. In addition, it was demonstrated there was no enzymatic loss.
The ANOVA for the response (partition coefficient, K) revealed significance for the model at the 95% confidence level (p ≤ 0.05). Table 8 presents the calculated p values for each variable. ANOVA did not observe a significant effect on the interaction between the variables; However, the quadratic temperature evaluation was significant (p = 0.0004).
Table 8.
Analysis of variance (ANOVA) for central composite design of the partition coefficient (K)
| Source | SS | DF | MS | F-value | p value |
|---|---|---|---|---|---|
| Model | 0.46 | 9 | 0.051 | 8.29 | 0.0091 |
| Y1: temperature | 1.01*10−4 | 1 | 1.01*10−4 | 0.016 | 0.9021 |
| Y2: Triton X-114 | 0.023 | 1 | 0.023 | 3.79 | 0.0994 |
| Y3: crude extract | 5.55*10−3 | 1 | 5.55*10−3 | 0.90 | 0.3795 |
| Y1Y2 | 5.51*10−3 | 1 | 5.51*10−3 | 0.89 | 0.3812 |
| Y1Y3 | 0.012 | 1 | 0.012 | 1.95 | 0.2125 |
| Y2Y3 | 0.011 | 1 | 0.011 | 1.70 | 0.2398 |
| Y12 | 0.32 | 1 | 0.32 | 51.57 | 0.0004 |
| Y22 | 0.017 | 1 | 0.017 | 2.74 | 0.1492 |
| Y32 | 3.97*10−3 | 1 | 3.97*10−3 | 0.64 | 0.4531 |
| Lack of fit | 0.037 | 4 | 9.26*10−3 | 3968.75 | 0.0003 |
| Pure error | 4.66*10−6 | 2 | 2.33*10−3 |
SS, sum of squares; DF, degrees of freedom; MS, mean squares
No evaluated variable showed significance for the proposed model. Likewise, ANOVA did not show a significant effect (Table 8) on the interaction between the variables; however, the quadratic temperature evaluation was significant (p = 0.0004), which implies that the model should be readjusted for larger variations of the temperature factor.
The partition coefficient of L-asparaginase in the system can be predicted by the equation below, in a confidence level of 95% and R2 = 0.98;
The response surface graph (Fig. 2) showed an optimum region to find a good partition coefficient for L-asparaginase purification. When it is low in the levels of the response K, the better will be the enzyme migration to the micelle-poor phase.
Fig. 2.
Response surface graph as a function of Triton X-114 and temperature for response K
As shown in Fig. 2, interactions through variables did not directly influence the separation profile, as expected by ANOVA results. As said above, temperature quadratic values are the only variables associated to a significant change. This is expected, since too low or too high temperatures can change dramatically the coexistence curve, because the partitioning of system is temperature dependent [21].
Discussion
Several reports of ASNase production have been found in the literature. Badoei-Dalfard purified, optimized, and characterized a bacterial L-asparaginase from a Pseudomonas aeruginosa strain. After cloning and purification, the maximum activity found using the method proposed by Imada et al. [25] for the quantification of free ammonia identified 785 U/mL [26].
Another study reported the production of the enzyme in Pseudomonas fluorescens as homotetramero and total molecular weight of 140 KDa and after purified this ASNase had an activity of 26 U/mg protein and half-life higher than commercially available L-asparaginase from E. coli. [27].
Souza et al. identified 19 studies related to the production of L-asparaginase from fungi. Of the 19 studies selected, only 6 showed total enzyme purification. These data demonstrate the need for production of this enzyme by eukaryotic microorganisms. The asparaginase purification studies selected in this systematic review showed enzymes with molecular weights ranging from 35 to 133 kDa [28].
Baskar and Renganathan determined an activity of 40.86 IU/mL in a strain of Aspergillus terreus through a delineation of the central composite where it identified the positive relation of the concentration of L-asparagine and L-proline, resembling our results [29].
Another species of the genus Aspergillus was also studied by Dias and Sato [30]. The study presented a Plackett-Burman design for selection of significant variables with variation of temperature, agitation, inoculum concentration, initial pH, L-proline concentration, L-asparagine concentration, glucose concentration, and concentration of yeast extract and concentration of oats. Variables temperature, inoculum concentration, and pH of the culture medium presented a significant and positive effect for the proposed model. Following these results, they performed a factorial design with central composite, using as variables just pH and inoculum concentration. According to the results presented in the experimental design, there is a tendency of improvement as the cultivation temperature increases. As an increase in temperature of cultivation could modify the fungal biological characteristics, it was decided to keep the temperature at 35 °C. This temperature has already been demonstrated to be adequate in other studies. [31–33].
After all the steps of the optimization, it is possible to observe that the use of wheat bran can be a viable alternative to produce yeast L-asparaginase. Our PBD evaluated 7 reagent variables, 1 agro-industrial residue, and 3 external factors that can be easily controlled, aiming the reduction of production costs of this enzyme.
Duarte et al. found that the use of an ATPMS was more efficient than an ATPS for the purification of a lipase produced by Leucosporidium scottii L117. They had a yield of 93.85% and a partition coefficient of 7.76 was observed, indicating that the enzyme was mostly in the micelle-rich portion of the system, because lipases have many hydrophobic residues in their structure. The coexistence curve used with McIlvaine buffer has a temperature lower than that used in the present work. In this case, the partition coefficient greater than 1 indicates migration of the enzymes to the micelle-rich phase. The ATPMS constructed through different concentrations of Triton X-114 and buffer was able to exhibit a purification factor of 1.7 and 73% yield [34]. Liu et al. used ATPMS with Triton X-114 as a surfactant to purify 3′,5′-bisphosphate nucleotidase from yeast. In the bottom phase of the system, a purification factor of 1.4 and an enzymatic recovery of 63.58% were found [15]. A purification of bromelain using a micellar system composed of McIlvaine and Triton X-114 buffer presented by Spir et al. demonstrated a purification factor of 0.2, in addition to a partition coefficient > 1, indicating that migration to the micelle-rich phase of the system [35].
In the present work, it had been found a 1.4 purification factor and 100% of enzyme recovery for L-asparaginase, which indicates a promising result for one of the enzymatic purification steps mainly when it is been using a residue agro-industrial as carbon and nitrogen source. The ATPMS was able to extract pigments for the micelle-rich phase and the target biomolecule for the opposite phase. ATPMS is a method used as a pre-purification method and for the reduction of contaminants and to protein extraction is an alternative to the salting out or organic solvents in purification steps of macromolecules.
Jaramillo et al. applied an ATPMS Triton X-114/sodium phosphate buffer for the purification of a pectinase produced by Aspergillus oryzae. A factorial design 23 was applied to obtain a system with a yield of 375% at 30 °C, 6% Triton X-114, and 15% crude extract. The author still defines that the partition coefficients were smaller than 1, which indicates that the pectinase migrated preferentially to the micelles-poor phase [36]. The ATPMS was able to extract pigments for the micelle-rich phase and the target biomolecule for the opposite phase. The temperature used and the concentration of surfactant in the work described by Jaramillo et al. resemble that used for the purification of L-asparaginase in system 8, in which the best yield was obtained.
To the best of the authors’ knowledge, there is no previously published work using ATPMS for extraction of a fungal L-asparaginase. An optimized culture medium was presented in this work for the Penicillium sp. L-asparaginase production containing 2.8% L-asparagine, 4.0% L-proline, 0.75% potato dextrose broth, 0.1% sucrose, and 0.5% wheat bran. The best response was found at the central point of the optimization design (run 17, 2.33 IU/mL). The L-asparaginase activity found in factorial design was approximately 1.8 times higher when compared with the greater activity found in the Plackett-Burman design.
It is important to highlight that this work had used the Drainas methodology for quantified L-asparaginase activity. However, as shown by Freitas et al., L-asparaginase activity most of the time was assayed from crude protein extracts using the Nessler method, which is an indirect measurement of asparaginase activity that determines the concentration of ammonia generated after the action of the enzyme on the substrate, L-asparagine. Nevertheless, ammonia is also generated throughout microbial fermentations and this endogenous ammonia will reduce the Nessler reagent if crude microbial extracts are used to determine total L-asparaginase activity. Thus, many L-asparaginase activities reported in the literature may be overestimated when Nessler reagent is used, which can justify the low levels we found in this work [37, 38].
Considering these results and the low cost of the purification process using ATPMS, we consider it relevant for the study of the partial purification of the enzyme L-asparaginase through non-chromatographic techniques as part of the process, thus reducing its industrial production costs.
Acknowledgments
This work was carried out with the financial support of the State Funding Agency of Distrito Federal (FAPDF), National Council for Scientific and Technological Development (CNPq), Coordination of Superior Level Staff Improvement (CAPES), and University of Brasilia.
Funding information
This work was supported by the State Funding Agency of Distrito Federal (FAPDF) – process number 193.001.661/2017 and the São Paulo Research Fundation (FAPESP) – process number: 13/08617-7.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Contributor Information
Samuel L. Cardoso, Email: samuel.cardoso@aluno.unb.br
Marcela M. de Freitas, Email: marcelafreitas@aluno.unb.br
Paula M. de Souza, Email: paulasouza@unb.br
Mauricio Homem-de-Mello, Email: mauriciohmello@unb.br.
Dâmaris Silveira, Email: damaris@unb.br.
Yris Maria Fonseca-Bazzo, Email: yrisfonseca@unb.br.
Edivaldo X. Filho, Email: eximenes@unb.br
Adalberto P. Junior, Email: pessoajr@usp.br
Pérola O. Magalhães, Email: perolamagalhaes@unb.br
References
- 1.Lopes AM, de Oliveira-Nascimento L, Ribeiro A, Tairum CA, Jr, Breyer CA, de Oliveira MA, Monteiro G, de Souza-Motta CM, de Oliveira Magalhães P, Avendaño JGF. Therapeutic l-asparaginase: upstream, downstream and beyond. Crit Rev Biotechnol. 2017;37(1):82–99. doi: 10.3109/07388551.2015.1120705. [DOI] [PubMed] [Google Scholar]
- 2.Kebriaei P, Anastasi J, Larson RA. Acute lymphoblastic leukaemia: diagnosis and classification. Best Pract Res Clin Haematol. 2002;15(4):597–621. doi: 10.1053/beha.2002.0224. [DOI] [PubMed] [Google Scholar]
- 3.Onciu M. Acute lymphoblastic leukemia. Hematol Oncol Clin North Am. 2009;23(4):655–674. doi: 10.1016/j.hoc.2009.04.009. [DOI] [PubMed] [Google Scholar]
- 4.Moghrabi A, Levy DE, Asselin B, Barr R, Clavell L, Hurwitz C, Samson Y, Schorin M, Dalton VK, Lipshultz SE. Results of the Dana-Farber Cancer Institute ALL Consortium Protocol 95-01 for children with acute lymphoblastic leukemia. Blood. 2007;109(3):896–904. doi: 10.1182/blood-2006-06-027714. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Quintanilla-Flores DL, Flores-Caballero MÁ, Rodríguez-Gutiérrez R, Tamez-Pérez HE, González-González JG (2014) Acute pancreatitis and diabetic ketoacidosis following L-asparaginase/prednisone therapy in acute lymphoblastic leukemia. Case Rep Oncol Med 2014:139169. 10.1155/2014/139169 [DOI] [PMC free article] [PubMed]
- 6.Egler RA, Ahuja SP, Matloub Y. L-asparaginase in the treatment of patients with acute lymphoblastic leukemia. J Pharmacol Pharmacother. 2016;7(2):62–71. doi: 10.4103/0976-500X.184769. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Pieters R, Hunger SP, Boos J, Rizzari C, Silverman L, Baruchel A, Goekbuget N, Schrappe M, Pui CH. L-asparaginase treatment in acute lymphoblastic leukemia: a focus on Erwinia asparaginase. Cancer. 2011;117(2):238–249. doi: 10.1002/cncr.25489. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Torres-Obreque K, Meneguetti GP, Custódio D, Monteiro G, Pessoa-Junior A, Rangel-Yagui CO (2018) Production of a novel N-terminal PEGylated crisantaspase. Biotechnol Appl Biochem 66(3):281–289. 10.1002/bab.1723 [DOI] [PubMed]
- 9.Chan WK, Lorenzi PL, Anishkin A, Purwaha P, Rogers DM, Sukharev S, Rempe SB, Weinstein JN. The glutaminase activity of L-asparaginase is not required for anticancer activity against ASNS-negative cells. Blood. 2014;123(23):3596–3606. doi: 10.1182/blood-2013-10-535112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Warrell RP, Chou T-C, Gordon C, Tan C, Roberts J, Sternberg SS, Philips FS, Young CW. Phase I evaluation of succinylated Acinetobacter glutaminase-asparaginase in adults. Cancer Res. 1980;40(12):4546–4551. [PubMed] [Google Scholar]
- 11.Mazzola PG, Lopes AM, Hasmann FA, Jozala AF, Penna TC, Magalhaes PO, Rangel-Yagui CO, Pessoa A., Jr Liquid–liquid extraction of biomolecules: an overview and update of the main techniques. J Chem Technol Biotechnol. 2008;83(2):143–157. [Google Scholar]
- 12.Nadar SS, Pawar RG, Rathod VK. Recent advances in enzyme extraction strategies: a comprehensive review. Int J Biol Macromol. 2017;101:931–957. doi: 10.1016/j.ijbiomac.2017.03.055. [DOI] [PubMed] [Google Scholar]
- 13.Malpiedi LP, Nerli BB, Abdala DS, de Alcântara Pessôa-Filho P, Pessoa A., Jr Aqueous micellar systems containing Triton X-114 and Pichia pastoris fermentation supernatant: a novel alternative for single chain-antibody fragment purification. Sep Purif Technol. 2014;132:295–301. [Google Scholar]
- 14.Chuo SC, Mohd-Setapar SH, Mohamad-Aziz SN, Starov VM. A new method of extraction of amoxicillin using mixed reverse micelles. Colloids Surf A Physicochem Eng Asp. 2014;460:137–144. [Google Scholar]
- 15.Liu Y, Sun M-H, Shao S-K, Deng G. An affinity-based aqueous two-phase mixed micellar system and its purification of Yeast 3′, 5′-bisphosphate nucleotidase. J Chromatogr B. 2017;1060:215–220. doi: 10.1016/j.jchromb.2017.05.032. [DOI] [PubMed] [Google Scholar]
- 16.Liu CL, Nikas Y, Blankschtein D. Novel bioseparations using two-phase aqueous micellar systems. Biotechnol Bioeng. 1996;52(2):185–192. doi: 10.1002/(SICI)1097-0290(19961020)52:2<185::AID-BIT1>3.0.CO;2-M. [DOI] [PubMed] [Google Scholar]
- 17.Shiomori K, Ebuchi N, Kawano Y, Kuboi R, Komasawa I. Extraction characteristic of bovine serum albumin using sodium bis (2-ethylhexyl) sulfosuccinate reverse micelles. J Ferment Bioeng. 1998;86(6):581–587. [Google Scholar]
- 18.Rundlett KL, Armstrong DW. Effect of micelles and mixed micelles on efficiency and selectivity of antibiotic-based capillary electrophoretic enantioseparations. Anal Chem. 1995;67(13):2088–2095. doi: 10.1021/ac00109a030. [DOI] [PubMed] [Google Scholar]
- 19.Soto A, Arce A, Khoshkbarchi MK. Partitioning of antibiotics in a two-liquid phase system formed by water and a room temperature ionic liquid. Sep Purif Technol. 2005;44(3):242–246. [Google Scholar]
- 20.Rangel-Yagui C, Pessoa-Jr A, Blankschtein D. Two-phase aqueous micellar systems: an alternative method for protein purification. Braz J Chem Eng. 2004;21(4):531–544. [Google Scholar]
- 21.Rangel-Yagui CO, Lam H, Kamei DT, Wang DI, Pessoa A, Jr, Blankschtein D. Glucose-6-phosphate dehydrogenase partitioning in two-phase aqueous mixed (nonionic/cationic) micellar systems. Biotechnol Bioeng. 2003;82(4):445–456. doi: 10.1002/bit.10586. [DOI] [PubMed] [Google Scholar]
- 22.Saitoh T, Hinze WL. Concentration of hydrophobic organic compounds and extraction of protein using alkylammoniosulfate zwitterionic surfactant mediated phase separations (cloud point extractions) Anal Chem. 1991;63(21):2520–2525. doi: 10.1021/ac00021a023. [DOI] [PubMed] [Google Scholar]
- 23.Drainas C, Kinghorn J, Pateman J. Aspartic hydroxamate resistance and asparaginase regulation in the fungus Aspergillus nidulans. Microbiology. 1977;98(2):493–501. [Google Scholar]
- 24.Van Der Meeren P, Cocquyt J, Flores S, Demeyere H, Declercq M. Quantifying wetting and wicking phenomena in cotton terry as affected by fabric conditioner treatment. Text Res J. 2002;72(5):423–428. [Google Scholar]
- 25.Imada A, Igarasi S, Nakahama K, Isono M. Asparaginase and glutaminase activities of micro-organisms. Microbiology. 1973;76(1):85–99. doi: 10.1099/00221287-76-1-85. [DOI] [PubMed] [Google Scholar]
- 26.Badoei-Dalfard A. Purification and characterization of l-asparaginase from Pseudomonas aeruginosa strain SN004: production optimization by statistical methods. Biocatalysis Agric Biotechnol. 2015;4(3):388–397. [Google Scholar]
- 27.Sindhu R, Manonmani H. Expression and characterization of recombinant l-asparaginase from Pseudomonas fluorescens. Protein Expr Purif. 2018;143:83–91. doi: 10.1016/j.pep.2017.09.009. [DOI] [PubMed] [Google Scholar]
- 28.Souza PM, de Freitas MM, Cardoso SL, Pessoa A, Guerra ENS, Magalhaes PO. Optimization and purification of L-asparaginase from fungi: a systematic review. Crit Rev Oncol Hematol. 2017;120:194–202. doi: 10.1016/j.critrevonc.2017.11.006. [DOI] [PubMed] [Google Scholar]
- 29.Baskar G, Renganathan S. Optimization of L-asparaginase production by Aspergillus terreus MTCC 1782 using response surface methodology and artificial neural network-linked genetic algorithm. Asia Pac J Chem Eng. 2012;7(2):212–220. [Google Scholar]
- 30.Dias FF, Sato HH. Sequential optimization strategy for maximum l-asparaginase production from Aspergillus oryzae CCT 3940. Biocatalysis Agric Biotechnol. 2016;6:33–39. [Google Scholar]
- 31.Farag AM, Hassan SW, Beltagy EA, El-Shenawy MA. Optimization of production of anti-tumor l-asparaginase by free and immobilized marine Aspergillus terreus. Egypt J Aquat Res. 2015;41(4):295–302. [Google Scholar]
- 32.Gurunathan B, Sahadevan R. Optimization of culture conditions and bench-scale production of L-asparaginase by submerged fermentation of Aspergillus terreus MTCC 1782. J Microbiol Biotechnol. 2012;22(7):923–929. doi: 10.4014/jmb.1112.12002. [DOI] [PubMed] [Google Scholar]
- 33.Huang L, Liu Y, Sun Y, Yan Q, Jiang Z. Biochemical characterization of a novel L-Asparaginase with low glutaminase activity from Rhizomucor miehei and its application in food safety and leukemia treatment. Appl Environ Microbiol. 2014;80(5):1561–1569. doi: 10.1128/AEM.03523-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Duarte AWF, Lopes AM, Molino JVD, Pessoa A, Sette LD. Liquid–liquid extraction of lipase produced by psychrotrophic yeast Leucosporidium scottii L117 using aqueous two-phase systems. Sep Purif Technol. 2015;156:215–225. [Google Scholar]
- 35.Spir LG, Ataide JA, De Lencastre Novaes LC, De Borba GD, Moriel P, Silveira E, Pessoa A, Jr, Tambourgi EB, Mazzola PG. Application of an aqueous two-phase micellar system to extract bromelain from pineapple (Ananas comosus) peel waste and analysis of bromelain stability in cosmetic formulations. Biotechnol Prog. 2015;31(4):937–945. doi: 10.1002/btpr.2098. [DOI] [PubMed] [Google Scholar]
- 36.Jaramillo PMD, Gomes HAR, de Siqueira FG, Homem-de-Mello M, Ferreira Filho EX, Magalhães PO. Liquid–liquid extraction of pectinase produced by Aspergillus oryzae using aqueous two-phase micellar system. Sep Purif Technol. 2013;120:452–457. [Google Scholar]
- 37.de Freitas MM, Souza PM, Cruvinel K, Barros T, Santos SN, Long PF, Pessoa A, Magalhães PO. Interferences that impact measuring optimal l-asparaginase activity and consequent errors interpreting these data. Appl Microbiol Biotechnol. 2019;103(13):5161–5166. doi: 10.1007/s00253-019-09890-0. [DOI] [PubMed] [Google Scholar]
- 38.Drainas D, Drainas C. A conductimetric method for assaying asparaginase activity in Aspergillus nidulans. Eur J Biochem. 1985;151(3):591–593. doi: 10.1111/j.1432-1033.1985.tb09144.x. [DOI] [PubMed] [Google Scholar]