Primary Recovery of Miraculin from Miracle Fruit, by AOT Reverse Micellar System

Zuxing He, Joo Shun Tan, Sahar Abbasiliasi, Oi Ming Lai, Yew Joon Tam, Murni Halim, Arbakariya B. Ariff

Miracle fruit, Synsepalum dulcificum, contains a glycoprotein known as miraculin. After consuming this glycoprotein, sour foods taste sweet and the effect lasts for up to 4 hours. With the increasing demand for natural and “low-calorie” sweeteners, the use of miraculin as an additive is increasing enormously in the food, medicine, and cosmetic industries.

Introduction

Synsepalum dulcificum, a shrub native to tropical West Africa, produces red berries that have the unusual ability to modify a sour taste into a sweet taste. The active ingredient in the berries, miraculin, is a taste-modifying protein that causes the sour taste components such as citric and ascorbic acids to be perceived as sweet after consumption in the mouth.

The mechanisms behind the sweet-inducing activity of miraculin have not yet been identified but histidine residues in miraculin have been linked to its taste-modifying activity. Twenty micrograms of chromatographically purified miraculin produce a marked increase in the sweetness of lemon and concomitantly a marked diminution of sourness.

However, the activity of miraculin is prone to be destroyed when the solution is boiled or exposed to a high concentration of organic solvents at room temperature. The activity was also decreased at high pH (pH > 12) and is greatly decreased (pH < 2.5). Although a lot of experiments have been carried out to explore the structure and mechanism of miraculin and to study the actual function of miraculin, the purification procedures for miraculin nowadays are thought to be labor-intensive, time-consuming and costly.

The main objective of this study was to investigate the feasibility of using the reverse micelle extraction method to extract miraculin from S. dulcificum. The effects of various factors that might influence performance were evaluated, such as crude pH, surfactant concentration during forwarding extraction and pH, isopropanol concentration, and salt concentration in the aqueous phase during backward stripping. The significant factors were also optimized to enhance extraction yield and product purity.

 Materials and methods

2.1. Miracle fruits

The skin and seeds of fresh miracle, S. dulcificum fruits were separated manually using a knife and the pulp was freeze-dried, then ground into a fine powder using a blender. The pulp powder was kept at 30 C prior to use in the extraction and purification procedures.

2.2. Chemicals

Sodium di (2-Ethylhexyl) sulfosuccinate (AOT) used without further purification, Isooctane, Bradford reagent, Sodium chloride, Isopropanol, Miraculin standard (~95% purity).

2.3. Preparation of miraculin extract

The extraction of miraculin was carried out with some modifications. In this method, 4 g of lyophilized pulp powder was suspended with 40 mL of water and homogenized for 2 min.  The homogenate was centrifuged at 12,000g for 30 min. After discarding the supernatant, the sediment was homogenized for 2 min in 30 mL of 0.5 M NaCl solution (pH 6.8). The homogenate was clarified by centrifugation at 12,000 g for 20 min and the colorless supernatant at pH 3 was stored at 30 C.

2.4 Forward extraction

We performed the forward extraction and backward stripping with some modifications. Briefly, the organic phase was prepared by dissolving various concentrations of AOT (0.03, 0.05, 0.1, and 0.2 mol/L) in isooctane. The pH of the crude was adjusted with either 1 mol/L NaOH or 1 mol/L HCl. The isoelectric point (pI) of miraculin is 9 and miraculin was reported to be stable during storage between pH 2.5 and 12. Based on this information, the pH of the crude in the aqueous phase was adjusted to various pH values ranging from 3 to 10 to avoid miraculin precipitation and loss of activity during the experiments. Equal volumes (0.5 mL each) of aqueous and organic solutions were mixed gently in a tube and the mixtures were then shaken mechanically for 10 min. The mixtures were then centrifuged at 4000 g for 5 min to reach a clear separation of the two phases.

2.5. Backward stripping

The reversed micellar solution with loaded protein was added to an equal volume of an aqueous solution which consisted of 0.02 mol/L phosphate buffer at the required pH (7, 8, 9, 10, and 11) in a tube. The required concentrations of NaCl (0, 0.5, 1, 1.5 and 2 mol/L) and isopropanol (0, 50, 100, 150 and 200 mL/L) were also added to the tube. The organic phase and fresh aqueous phase were shaken for 20 min. The mixtures were then centrifuged at 4000 g for 5 min to reach a clear separation of the two phases. The total protein in the stripping aqueous solution was determined.

2.6. Total protein assay

The total protein concentration in the crude sample was determined using the Bradford method using bovine serum albumin (BSA) as a standard. Ten mL of the sample was added to 200 ml of the diluted dye reagent (1 part dye reagent concentrate with four parts distilled, deionized water) in a microtiter plate and incubated at room temperature for at least 5 min. The absorbance was measured at 595 nm against a reagent blank.

2.7. Reverse-phase high-performance liquid chromatography analysis

The miraculin concentration in the sample was analyzed using reversed-phase high-performance liquid chromatography with some modifications. Sample (40 mL) was injected into the column and equilibrated with 1 mL/L trifluoroacetic acid (TFA) in water. The column was eluted using a linear gradient of acetonitrile with increasing concentration from 150 mL/L to 700 mL/L, and the flow rate was fixed at 1 mL/min. The absorbance of the sample was read at 280 nm. Miraculin standards were prepared at concentrations ranging from 100 to 1000 mg/L. The relationship between miraculin concentration (mg/L) and peak area (AU min) was observed as 0.00013 mg miraculin/L/peak area. The purity of the peak was analyzed based on the percentage of total peak area using Empower software (System Software, Waters Co.) for data acquisition and analysis.

2.8. SDS-PAGE and silver staining

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed in a Bio-Rad electrophoresis unit as described by Laemmli (1970). The acrylamide gel was prepared as a 120 mL/L resolving gel and a 45 mL/L stacking gel. Protein samples recovered from the top phase were concentrated and precipitated using 100 mL/L trichloroacetic acids (TCA) solution, which removed the salts that affect the electrophoresis process. The pellets were resuspended in denaturing buffers (0.1 mol/L Tris HCl pH 6.8, 40 g/L SDS, 100 mL/L 2-mercaptoethanol, 200 mL/L glycerol, and bromophenol blue). The electrophoresis was run at 110 V and 36 mA for 75 min. The gel was stained with a buffer solution consisting of 0.5 mL/L Coomassie Brilliant Blue G-250, 300 mL/L methanol, and 100 mL/L acetic acids. After destaining, protein bands were visualized using the same buffer solution in the absence of Coomassie Brilliant Blue. The gel was then stained with a PageSilver™ silver staining kit (Fermentas, St. Leon-Rot, Germany).

2.9. Miraculin sensory analysis

The taste-modifying activities of miraculin were evaluated by five subjects by tasting 0.2 mL of partially purified miraculin solution and held in mouth (tongue) for 5 min. Subsequently, each subject expectorated out the partially purified miraculin solution, washed the mouth with distilled water and sipped 5 mL of 0.02 mol/L citric acid and finally evaluated the presence of taste modifying activities in the purified miraculin.

2.10. Definition

Specific miraculin in crude ¼ Miraculin in crude extract Total protein in crude extract Specific miraculin in back extraction aqueous phase ¼ Miraculin in back extraction aqueous phase Total protein in back extraction aqueous phase.

Results and discussion

3.1. Effect of AOT concentration during forward extraction

The effect of AOT concentration on protein recovery by reverse micelles AOT based reverse micellar phases were selected in the present study as the results indicated that miraculin can be easily extracted into reverse micelle when AOT was used as a surfactant. This could be due to AOT which is an anionic surfactant, is suitable for purifying proteins with low molecular weight and high isoelectric point. On the other hand, cationic surfactants such as CTAB generally used to separate proteins with high molecular weights and low isoelectric points were not suitable in this study. There of, in the present study, AOT was used for the separation of miraculin which also has a low molecular weight (only 28 kDa) and a high isoelectric point value of 9.

A similar result used AOT to separate low molecular weight and high isoelectric point proteins such as resistance-like protein PeB and pentatricopeptide repeat-containing protein. Protein recovery was increased marginally from 37% to 50% as AOT concentration increased from 0.03 to 0.2 mol/L, while miraculin recovery was increased from 47% to 63% as AOT concentration increased from 0.03 to 0.1 mol/L, and remained constant (63%) at 0.2 mol/L. An increase in AOT concentration leads to an increase in the aggregation number of AOT and the size of reverse micelles.

Subsequently, the increase in size would lead to a decrease in the steric hindrance of reverse micelles. These effects contributed to the increasing amount of protein extracted. For AOT concentrations ranging from 0.1 to 0.2 mol/L, miraculin recovery remained constant, while protein recovery increased from 49% to 54%, indicating that the latter increase was caused by impurities. The highest purification factor (1.27) was obtained at 0.1 mol/L AOT, which was further confirmed by the HPLC result, giving a purity of 44%. Therefore, 0.1 mol/L AOT was used in subsequent experiments.

3.2. Effect of crude pH during forwarding extraction

The effect of crude pH on the partitioning behavior of reverse micelles as the pH of the crude increased from 3 to 6, protein recovery significantly decreased from 50% to 32%. Protein recovery was further decreased to 17% at pH 7 and declined slightly to 16% at pH 8. Protein recovery at crude pH 10 was only 1.5%. A similar trend was also observed with a change in pH from 3 to 6, where protein recovery decreased from 61% to 25%.

Miraculin recovery was stable at pH values ranging from 6 to 8 and miraculin was not recovered at pH 10. Because of the rapid decrease in miraculin recovery, the purification factor was lowest at pH 6 (0.778), while the highest purification factor was obtained at pH 8 (1.48). From HPLC analysis, the purity obtained at pH 6 was 48% and the purity was increased markedly to 89% and 90% at pH 7 and 8, respectively.

When the pH is lower than the protein's pI value, the protein will provide a positive charge at its surface, resulting in the development of electrostatic interactions between the charged amino acid residues on the protein surface and the electrical double-layer.

The results from this study also showed that the pI values of most protein contaminants were below 7 or 8, and the relative values of pI and pH determined the amount of protein recovery using AOT as a surfactant. In accordance with the general conclusions from the forward extraction studies, the condition of 0.1 mol/L AOT/isooctane mixed with crude at pH 8 was applied for subsequent studies on backward transfer.

3.3. Effect of isopropanol concentration during backward stripping

According to previous work, accomplishing backward extraction of protein from the solvent phase to the aqueous phase may be difficult. It was reported that the rate of stripping is about three orders of magnitudes slower than forwarding extraction. Considering the traditional method of backward extraction for nattokinase stripping found that backward extraction was rather difficult when changing the pH and salt concentration.

When different concentrations of isopropanol were added to the aqueous phase, 150 mL/L isopropanol strongly promoted the backward transfer of nattokinase (2004). A similar result was reported where the nearly complete backward transfer of porcine pepsin and 70% backward transfer of bovine chymosin were obtained after the addition of 100e150 mL/L isopropanol. When small amounts of alcohol are added, they may affect the micellar interactions in the reverse micelle system.

The hydrophobic hydrocarbon group suppresses the intermicellar attractive interaction in proportion to their carbon chain length, while the hydrophilic hydroxyl group enhances the interaction. The addition of isopropanol probably leads to an increase in the attractive interactions between reverse micelles and the arrangement of AOT molecules in isooctane, and this causes instability in the reverse micelles and also the exclusion of protein from reverse micelles. Although protein recovery increased slightly from 15% to 17% with increasing isopropanol concentrations from 0 to 200 mL/L, miraculin recovery (ranging from 23% to 24%) was not significantly improved.

In addition, the purity of miraculin (ranging from 44% to 45%) as shown by HPLC chromatography was not significantly improved. Results from this study showed that nearly all proteins and miraculin were stripped without the addition of isopropanol, or that isopropanol had little effect in this case.

3.4. Effect of pH during backward stripping

The mechanism by which pH affects extraction performance during backward stripping was similar to that observed for forwarding extraction. Briefly, the electrical interactions between the charged AOT head groups and the protein surface implied that the pH in the added aqueous phase should be higher than the pI of miraculin. The same negative charge on the protein surface and on the surfactant head groups would lead to the stripping of protein from the solvent phase to the aqueous phase.

No miraculin was observed through HPLC under pH 7 and 8, and little miraculin was stripped at pH 9 (miraculin recovery was 1.2%), with a purification factor of 0.1. Nevertheless, miraculin recoveries markedly increased to 22% and 23% when the pH values of the modified aqueous phase were adjusted to pH 10 and 11, respectively.

Protein recovery and the purification factor at pH 10 were close to that at pH 11. However, the purity indicated by HPLC at pH 11 was 94%, which was relatively higher than that obtained at pH 10 (90%). Thus, pH 11 was considered the optimal pH of the aqueous phase during backward extraction.

3.5. Effect of salt concentration during backward stripping

The effect of salt concentration during backward stripping on the performance of protein and miraculin extraction. The protein recovery increased from 13.5% to 16.5% with the addition of NaCl in the range 0.5e2 mol/L. Change in salt concentration affects both the size and hydrophobicity of reverse micelles and the hydrophobicity of proteins. The increase in ionic strength with increasing NaCl concentration led to a decrease in electrostatic interactions between AOT reverse micelles and protein, which promoted backward transfer. When increasing the NaCl concentration, the miraculin recovery did not follow the same trend as protein recovery but remained constant at about 22%. This strongly implies that miraculin had been completely stripped under this condition.

This result also led to a decrease in purification factor from 1.6 to 1.3, confirmed by purity analysis using HPLC (reduced from 95% to 89%). A turbid aqueous phase was observed during the stripping stage when NaCl was omitted (data not shown). This phenomenon has been observed by many researchers for soluble and insoluble proteins in aqueous buffers. However, this cloudy aqueous phase may affect protein release.

3.6. Determination of the taste-modifying activities and purity of miraculin after reverse micelle treatment

The taste-modifying activities of purified miraculin were determined by sensory analysis. The results showed that the purified miraculin successfully changed the sourness of citric acid into sweetness. The purity of miraculin obtained from reverse micelles performed under optimal conditions was 94.8%, as determined by HPLC. High purity miraculin was eluted at retention times of 6.4 min, with some minor peaks found at retention times of 5.8 min and 9.8 min. The realization of recovery of miraculin in the aqueous phase after reverse micelle treatment in a partially purified form was confirmed by SDS-PAGE with silver staining.

Interestingly, a small peak appeared at a retention time of 7 min, near the major peak. This result differs from those where only one peak was observed during HPLC. However, silver staining under reducing conditions proved that miraculin occupied the major band in the gel. The molecular weight of miraculin varies from 25 to 28 kDa under reducing conditions, indicating that miraculin was successfully purified in this study. This is probably because miraculin shows dimeric and tetrameric structures after reverse micelle treatment.

The reverse micelle extraction could be used as a partial purification step of the miraculin from miracle fruit. In this study, the functionality or structural state of the purified protein remains unknown as partially purified miraculin could not be accessed in functionality or structural study. This is a subject for future study after a further purification step such as size exclusion purification or gel-filtration is performed to obtain pure miraculin.

Conclusions

Reverse micelle extraction can be applied as a simple and convenient process for the purification of miraculin from miracle fruit, S. dulcificum. Optimization of purification parameters was required to improve total miraculin purified, purification factor, and purity. Miraculin can be extracted under a wide range of crude extract pH values but pH 8 gives the highest purification factor and purity of miraculin. It should be noted that the amount of miraculin extracted at crude pH 3 was nearly three times that at pH 8, with a decreasing purification factor and purity.

Optimizing pH had more effect than optimizing the salt concentration. During the backward stripping stage, it is not necessary to add isopropanol because similar miraculin recoveries were obtained using different isopropanol concentrations. Results of HPLC and SDS-PAGE silver staining analyses also indicated that high purity miraculin could be obtained by purification using reverse micelles.

Reference:

Full journal research copy available here.