Original papers
Purification and Characterization of Bovine β-Lactoglobulin Variants A and B (Characterization of Bovine β-Lactoglobulin Variants)
2020 Volume 26 Issue 3 Pages 399-409
Details
2020 Volume 26 Issue 3 Pages 399-409
β-Lactoglobulin is a major allergen in whey, which have more than 9 variants. Among them, variants A and B are the most predominant forms in bovine milk. In this study, variants A and B were purified by combining gel filtration chromatography with anion-exchange chromatography for the first time. The purity of variants A and B were 96.2% and 94.3%, and the variants yield was 35.56% from the whey. The structure characterization has been monitored by the Circular dichroism spectrometry, Fourier transform infrared spectroscopy and Fluorescence spectroscopy. IgG binding of variants was evaluated by Western blotting which proved both keep well and β-Lactoglobulin B had a stronger ability. The substitution Val of Ala in β-Lactoglobulin B decreased its hydrophobic packing. In short, β-Lactoglobulin B had a stronger antigenicity but smaller hydrophobic packing than that of A.
Whey has become a new source of nutritional and functional ingredient, which plays an essential role in the formulation of balanced amino acid in food production chain (Etzel, 2004). β-Lactoglobulin accounts for 50% of whey protein (Santos et al., 2012), and whose eleven genetic variants have been identified. Among variants, variant A (18.363 kDa, pI=5.26) and B (18.276 kDa, pI=5.34) are predominant in the milk of various breeds of cows (Naqvi et al., 2013). There are only two differences in the amino acid residues at positions 64 (Asp→Gly) and 118 (Val→Ala), which locates on a turn and a β-sheet structure (Dong et al., 1996). The positions 64 fully exposes to the solvent, and positions 118 immerses in the hydrophobic core of the protein (Bello et al., 2010). Small structural differences in the vicinity of the two mutations were reported by Oliveira et al. (2001) and Qin et al. (1998). For example, replacement of Val to Ala at position 118 causes a void space in the core of the variant B and conformations of the loop C-D arises from variations in the local electrostatic potential induced by the change Asp/Gly at position 64. The difference in amino acid sequence also significantly demonstrates unique properties of the two variants (Rosa et al., 2006). For example, variant B is more favorable for cheese making because it more readily contributes to the coagulation of milk than variant A.
In general, β-Lactoglobulin has been separated by membranes, reverse-phase high-performance liquid chromatography (RP-HPLC) (Santos et al., 2012) and chromatographic technologies according to its properties. Among them, Size-exclusion chromatography and Ion-exchange chromatography are more efficient and ease of operation while preserving proteins' native state. Size-exclusion chromatography based on the molecular size has been used for macromolecule isolation since 1959. Ion-exchange chromatography also widely employs in the separation of proteins where the target proteins will not be retained at their pIs, only be retained by anion resins at pH above their pIs or by cation resins at pH below their pIs. β-Lactoglobulin variants were purified by ultrafiltration and Ion-exchange chromatography, which was reported by Kristiansen et al. (1998). The protein content was between 96.3% and 97.2% in dry matter, however, little fat presented in the preparations. Anion exchange chromatography (Mono Q column) coupled to a Fast Protein Liquid Chromatography system for the fractionation of major whey proteins and β-Lactoglobulin variants were proposed by Santos et al. (2012). Though a 60.5% (w/w) recovery of the two β-Lactoglobulin variants was obtained, the purity of the variants was not good enough. Up to now, there is little information on purification of β-Lactoglobulin A and β-Lactoglobulin B by Size-exclusion chromatography and Ion-exchange chromatography, which may be a novel and relatively easy way.
In our work, β-Lactoglobulin variants were purified by Size-exclusion chromatography with Sephadex G-75 gel followed by Ion-exchange chromatography with DEAE-sepharose Fast Flow gel. Identification and purity analysis of variants were determined by mass spectrometry and RP-HPLC. After obtaining high-purify variants, IgG binding was evaluated by western blotting and their structure were characterized by Circular dichroism spectroscopy, Fourier transform infrared spectroscopy and Fluorescence spectroscopy.
Chemicals and reagents Fresh raw milk was collected from Jiangxi Sunshine Dairy Group Co., Ltd, China. Standard β-Lactoglobulin (from bovine milk, 90% a mixture of β-Lactoglobulin A and β-Lactoglobulin B), β-Lactoglobulin A (from bovine milk, 90%), β-Lactoglobulin B (from bovine milk, 90%) and sheep anti-rabbit Ig/Horseradish Peroxidase were all purchased from Sigma (USA). Antiserum against a mixture of β-lactoglobulin A and B from rabbits were obtained by our lab in previous work. Polyvinyl difluoride membranes were obtained from Millipore (Bedford, MA, USA). DEAE-Sepharose Fast Flow gels and Sephadex G-75 gel were from General Electric Company, USA. All solvents and reagents were analytical grade.
Preparation of whey protein Fresh raw milk was filtrated with quadrilateral gauze for removing some impurities. Then milk was centrifuged at 3 000 rpm for 30 min at 4 °C, and the fat layer in the top was removed. The skimmed milk was adjusted to pH 4.6, which made casein precipitated. After that the supernatants were collected and concentrated by hollow-fiber membrane with 10 kg·mol−1 cut-off. The proteins were stored at −20 °C until used.
Purification by two steps of chromatography Step 1: Size-exclusion chromatography. Prior to loading sample, the column of Size-exclusion chromatography was equilibrated in 20 mM sodium phosphate buffer (pH 6.8) containing 150 mM Sodium chloride and the buffer should be degassed before using (Li et al., 2006). Then, whey protein (containing 36.0 mg β-Lactoglobulin) was carried out by Sephadex G-75 (3.8 cm×100 cm) column with a flow of 2 mL· min−1. Each fraction was collected with a volume of 10 mL. Then the absorbance of fractions in the elution profile was monitored at 280 nm using an Ultraviolet detector (Lambda 25, PerkinElmer) and the protein concentration was quantified by using Coomassie Brilliant Blue G 250 Staining (Bradford, 1976). The collected fractions were identified by 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) using pre-stained molecular weight marker (ranging from 10 to 170 kg·mol−1) according to Laemmli (1970).
Step 2: anion-exchange chromatography. In the second step of purification, anion–exchange chromatography was performed on DEAE-Sepharose gel packed in a column (2.6 cm×60 cm). Similarly, the column was equilibrated with 20 mM phosphate buffer (pH 6.8) after loading the sample. Then the bound proteins were eluted at a linear gradient by increasing the ionic strength of the elution buffer (0 to 0.6 M Sodium chloride) at a flow rate of 5 mL·min−1. Contents of collected protein was determined as in step 1. The purified β-Lactoglobulin variants was checked by native polyacrylamide gel electrophoresis(Native-PAGE) with standard bovine β-Lactoglobulin A and β-Lactoglobulin B.
Native-PAGE The gel of Native-PAGE was prepared with a modification according to Wittig et al (Wittig and Braun, 2006), and stacking gel were performed at current of 60 V for 30 min and 80 V for 120 min, respectively. Loading samples were mixed well prior to electrophoresis. 12 µL purified protein solutions and 4 µL prestained marker protein were transferred to each well. The gels were stained with Coomassie Brilliant Blue R-250 at least 15 min, rinsed several times until protein bands showed visibly, pictured by GS-800 (Bio-Rad, USA) biological imaging.
Mass spectrometry assay Purified β-Lactoglobulin variants were analyzed by ESI-Q-q-TOF MS appliance from Bruker Daltonics (Billerica, MA, USA). 5 mg of standard β-Lactoglobulin, isolated β-Lactoglobulin A and β-LactoglobulinB, were dissolved in the solutions with acetonitrile to water 1:1 including 0.1% formic acid, then 5 µL aliquots of the sample solution was injected into the Q-q-TOF MS systems. Mass spectra was scanned with pressure of nebulizer at 0.4 Bar, the flow rate of 0.3 mL·min−1 and the dry heater temperature of 180 °C. ESI source in positive mode was used for the ionization of samples.
Identification purity by RP-HPLC The purity of isolated variants was determined by RP-HPLC (LC-20A equipment, Shimadzu, Japan) with Inertsil WP300 C8 column (250 mm×4.6 mm×5 µm) (GL Sciences, China) at 30 °C. The column was equilibrated with solvent A (Milli-Q water containing 0.1% tallow fatty acid) and solvent B (HPLC grade-acetonitrile containing 0.1% tallow fatty acid). All samples were filtered through a 0.22 µm membrane before loading on RP-HPLC. 10 µL of 1 mg· mL−1 protein was injected in the column and ultraviolet absorption was measured at 280 nm. A gradient elution started from 20% to 50% of solvent B in 0–10 min, then 50% to 80% in 11–20 min and 80% to 20% for 2 min.
Western blotting SDS-PAGE was performed as the previous step. The amount and volume of the purified protein were about 5~20 µg and 12 µL, respectively. The gel containing isolated β-Lactoglobulin A and β-Lactoglobulin B was transferred to polyvinyl difluoride membranes with a current of 50 mA for 1 h. Then polyvinyl difluoride membranes were blocked with 1% porcine skin gelatin in TBS (Tris-buffered saline; pH 7.5) for 1 h at 37 °C followed by washing three times for 5 min with TBS-T (TBS containing 1% Tween 20). In further, the membrane was incubated with antiserum from rabbit with dilution of 1: 10 000 in TBS overnight at 4 °C. After washing in the next day, the membrane was incubated with sheep anti-rabbit IgG /Horseradish Peroxidase diluted to 1: 5 000 in TBS at 37 °C for 1 h. Finally, the membrane was colored with substrate for 20 min until bands were visible, then washed three times again in water and pictured.
Circular dichroism spectroscopy The secondary structure of isolated β-Lactoglobulin variants was performed by Far-UV CD spectroscopy with a JASCO J810 spectropolarimeter (Jasco, Tokyo, Japan)) at room temperature. The spectra ranged from 190 to 250 nm, the speed of scanning was 60 nm· min−1, the response time was 1–4 s and the bandwidth was 1 nm. The concentration of protein was 0.1 mg· mL−1 dissolved in 20 mM phosphate buffer (pH 6.8). CD spectra were measured for three times. Finally, the contents of different secondary structures were calculated with online Circular dichroism website software (http://dichroweb.cryst.bbk.ac.uk).
Fourier transform infrared spectroscopy Fourier Transform infrared spectroscopy (FT-IR) measurements were performed on a Nicolet-5700 FT-IR spectrometer (Thermo Scientific, USA), equipped with a germanium attenuated total reflection (ATR) accessory, a DTGS KBr detector, and a KBr beam splitter. All spectra were recorded via the ATR method with resolution of 4 cm−1 and 60 scans. Freeze-dried β-Lactoglobulin variants (2 mg) were ground into a fine powder with KBr (200 mg) and pressed into discs. Infrared spectra were carried out in the range of 1 600 and 1 700 cm−1 that display the characterization of secondary structure at room temperature. The white noise spectra were regarded as substrate from the final protein spectrum. The inverted second-derivative spectra were obtained and then curve fitted of the amide I band were analyzed.
Fluorescence spectroscopy The fluorescence was measured with a spectrophotofluorometer F-4500 (Hitachi, Japan) equipped with a 150 W Xenon lamp and a thermostat bath. For intrinsic fluorescence measurements, the excitation wavelength was 280 nm and emission wavelengths were among 300 and 450 nm. While during 8-Anilino-1-naphthalenesulfonic acid (ANS) fluorescence measurements, the excitation wavelength was 375 nm and the emission wavelength were from 400 to 650 nm. 30 µL of ANS solution (5 mM in 0.01 M sodium phosphate buffer, pH 7.4) was mixed with 3.0 mL of protein sample with a concentration of 0.1 mg· mL−1 and then fully blended by vortex. The widths of both the excitation slit and emission slit were 2.0 nm. The response time was 0.2–0.5 s and the scan speed was 1 200 nm· min−1. Each spectrum was measured for three times.
Size-exclusion chromatography Size-exclusion chromatography based on the molecular size has been used for macromolecule isolation. According to the molecular weight of whey protein, we chose Sephadex G-75 gel as separated medium. It has reported (Hobman, 1984) that major components in whey proteins were β-Lactoglobulin and α-lactalbumin(α-LA) while minor ones were bovine serum albumin, immunoglobulin, lactoperoxidase and lactoferrin (Table 1). Four major fractions separated by Size-exclusion chromatography profile were shown in Fig. 1A, and detected by SDS-PAGE shown in Fig 1B. We can find that large molecules in first band were bovine serum albumin together with some casein in peak a, while β-Lactoglobulin and α-LA appeared in peak b and c, respectively. However, peak b and c cannot be separated well as shown in Fig. 1A. It might be reasonable the main protein was α-LA in peak c, and it also contained a few amounts of β-Lactoglobulin. Given the lower content of β-Lactoglobulin, peak c was deserted for further fractionation process, which may bring about lower production but higher purity of β-Lactoglobulin. However, no protein could be detected in peak d which might be solvent elution peak shown in Fig. 1B. As for Peak b, the main protein was β-Lactoglobulin. But the purity of this peak was not good enough as shown by SDS-PAGE pattern in Fig. 1B. The different fractions of fraction B was detected by SDS-PAGE in Fig. 1C. Apparently, β-Lactoglobulin was dominant in the middle fractions from lane 3–7, while the first three fractions contained some large molecular weight of proteins and some of α-LA also appeared in lane 7 and 8. The collected crude β-Lactoglobulin was 31.71 mg from 72.54 mg whey protein and the total β-Lactoglobulin recovery rate was 43.71% (data not shown). More importantly, variants A and B could not be distinguished in the gel because the molecular weights of two variants are almost the similar. Therefore, it is necessary to purify the protein in peak b on Ion-exchange chromatography in further to get higher purity of β-Lactoglobulin variants.
| Protein | Molecular weight (kDa) | Isoelectric Point (pI) | Percentage of the whey protein fraction |
|---|---|---|---|
| Immunoglobulin | 150.0–900.0 | 5.8–7.3 | 12.5 |
| Lactoperoxidase | 78 | 9.6 | <1 |
| Lactoferrin | 78 | 8 | <0.9 |
| Bovine serum albumin | 66 | 4.9–5.1 | 6.3 |
| β-Lactoglobulin | 18.3 | 5.2–5.4 | 56.3 |
| α-lactoglobulin | 14.2 | 4.7–5.1 | 25 |
Purification of whey protein by Sephadex G-75 gel filtration.
A, Chromatogram of the whey protein isolated from size-exclusion chromatography. B, Electrophoretic patterns of whey protein isolated from size-exclusion chromatography; lane M, marker; lane 1–4, peak a–d. C, SDS-PAGE pattern of different fractions of peak b; lane M, maker; lane 1, the fraction of peak b in the beginning; lane 2–3, the fraction of peak b at the ascending phase; lane 4–5, the fraction of peak b at the summit; lane 6–8, the fraction of peak b at the descending phase.
Anion-exchange chromatography Ion-exchange chromatography also widely employs in the separation of proteins where the target proteins will not be retained at their pIs, only be retained by anion resins at pH above their pIs or by cation resins at pH below their pIs. There were two marked peaks in Fig. 2A, and they were identified as β-Lactoglobulin B and β-Lactoglobulin A by Native-PAGE with standard bovine β-Lactoglobulin variants in Fig. 2B. Since pH of variant B (pI=5.34) was higher than that of variant A (pI=5.26), variant B was eluted firstly at a 0.28 M salt followed by variant A when the concentration of salt reached to 0.32 M. This result was in good agreement with Santos et al. (2012) whose results also indicated that β-Lactoglobulin B was eluted firstly. The slight difference in the amino acid composition of the two variants (Oliveira et al., 2001) caused β-Lactoglobulin A competed for the adsorption sites and bonded more strongly to the resin, thus β-Lactoglobulin B was the variant being eluted firstly. Compared with the areas of two peaks, we can find that β-Lactoglobulin A accounted for a higher proportion of the total β-Lactoglobulin as shown in Figure 2A. Ganai once reported variant A had a higher β-Lactoglobulin protein concentration than variant B (Ganai et al., 2009). It was likely caused by the different levels of expression of the corresponding A and B alleles of the β-Lactoglobulin gene (Hill, 1993). It is reported that the amount of variant A was 30–50% higher than the amount of variant B (Graml et al., 1989). In addition, the purities of isolated variants were higher compared with standard variants as shown in Fig. 2B, and their purities were above 90%. The content of variants A and B were 18.8 mg in 40 mL and 7.0 mg in 20 mL, respectively. The two variants recovery rate was 81.36% from crude β-Lactoglobulin loaded from Size-exclusion chromatography. The electrostatic attraction between protein and resin will rely on the net charge and the distribution of that charge of the protein. It should be possible to separate variants A and B, due to their different net charge and molecular configurations (Boardman and Partridge, 1955). The practical range of ionic strengths used in Ion-exchange chromatography was considered to be between 50 mM and 500 mM, since molecules cannot move closer to each other, which may be in favor of protein separation. Even if concentration of sodium chloride changed from 0 to 0.5 M (0–50% of salt) in 500 mL, the chromatogram of variants A and B were overlapped. Instead, 600 mL of eluting solution was applied which can separate the two variants well in Fig. 2A. The lower ionic strength in 600 mL eluting solution could reduce the electrical double layer thickness, which may separate protein easily. In addition, it may also be related to the surface area and the surface structure of DEAE-Sepharose resin. Generally speaking, the elution volume of proteins is proportional to the distribution coefficient between the stationary and mobile phases. Other authors also proved that the linear gradient was the optimal elution mode for purifying proteins (Ishihara and Yamamoto, 2000). Besides, the flow rate of the eluent was 5 mL· min−1 which was quite fast compared the normal one. Luckily, such resin at high flow rates provided powerful ability for the purification of proteins. High purity and yield of variants proved the separated method was successful and easy to operate in laboratory study. Furthermore, the method can be enlarged for pilot production according to the parameters we established. For example, keeping ratio of sample volume to volume of medium and the same elution profile; increasing whey protein concentration and the column volume in the current work (Ng and Snyder, 2013).
Anion-exchange chromatography on DEAE–Sepharose Fast Flow of peak b from Size-exclusion chromatography. A. Elution profile of β-Lactoglobulin variants A and B; B. Native-PAGE pattern. Lane 1: standard β-Lactoglobulin A; Lane 2: standard β-Lactoglobulin B; lane 3: Peak 1; Lane 4: Peak 2
ESI-MS Mass spectrometry analysis after the process of deconvolution were shown in Fig. 3. In Fig. 3A, mass spectra of purified β-Lactoglobulin A, there were many peaks which indicated the target protein had different hydrogen ion charges. The result showed the purified protein was β-Lactoglobulin A and its relative molecular weight was 18.363 kg·mol−1 which was consistent with the known literature (Qin et al., 1999). In Fig. 3B, the relative molecular mass of the main protein was 18.277 kg·mol−1, which was β-Lactoglobulin B. However, some details from mass spectrum deconvolution report (not shown) revealed the purified sample contained other variants, whose peaks were not marked in Fig. 3B, possibly because of either the low concentrations in the sample or failure to be ionized. In general, the purified variants were β-Lactoglobulin A and β-Lactoglobulin B exactly, which were confirmed by MS again in addition to the Native-PAGE.
Mass spectra of purified β-Lactoglobulin variants obtained by ESI-MS.
A. fraction 2 (Elution profile of β-Lactoglobulin variants A by Anion-exchange chromatography on DEAE–Sepharose Fast Flow of peak b); B. fraction1 (Elution profile of β-Lactoglobulin variants B by Anion-exchange chromatography on DEAE-Sepharose Fast Flow of peak b)
RP-HPLC A unique RP-HPLC peak for standard β-Lactoglobulin A and purified β-Lactoglobulin A appeared at a retention time (RT) of 22.55 min and 22.54 min in Fig. 4 A and B. Similarly, standard β-Lactoglobulin B and purified β-Lactoglobulin B showed similar RP-HPLC profile, with the single strong peak at the RT of 22.61 min and 22.63 min (Fig. 4C and D). In general, the ratio of individual peak area to the total peak area can be employed to the purity analysis with UV absorption wavelength at 280 nm. After calculating, the purity of isolated β-Lactoglobulin A was as high as 96.2% and β-Lactoglobulin B was 94.3%. Though the high purify of variants, there were some minor peaks in both variants. It may contribute to the solvent elution peak and machine instability, which can be called as signal to noise ratio (SNR). Additional, the retention time of variant B was longer than that A, which demonstrated the polarity of variant B was weaker than variant A based on the principle of RP-HPLC. Clearly, there were no palpable detectable contamination of purified variants and the high purity of variants proved again the purification process was effective.
RP-HPLC profiles of the standard and purified β-Lactoglobulin.
A. standard β-Lactoglobulin A; B. purified β-Lactoglobulin A; C. standard β-Lactoglobulin B; D. purified β-Lactoglobulin B
Western blotting Fig. 5 showed the electrophoresis pattern of purified variants and specific IgG-antigen complexes. Obviously, two clear bands were detected at a molecular weight around 18 kg·mol−1 position in Fig. 5A, implicating these were β-Lactoglobulin A and β-Lactoglobulin B. The positive western blotting strips in Fig. 5B manifested the presence of specific IgG-antigen complexes, and the color of the strip B was deeper than that of A. The higher response was related to amino acid residues found only in the variant B, lying in surface regions of the molecule which may be caused by the different acids at position 64 and 118. (Efstratiadis et al., 1982). In general, we can fully prove that the purified β-Lactoglobulin variants maintained a well-preserved IgG binding ability.
Detection of purified β-Lactoglobulin A and B by Western-blotting
A. SDS-PAGE pattern; B.Immunoblotting with anti-β-Lactoglobulin rabbit serum. Lane M: Prestained Marker; Lane 1: purified β-Lactoglobulin A; Lane 2: purified β-Lactoglobulin B
Circular dichroism spectroscopy Secondary structure of purified variants was monitored by Far-UV Circular dichroism spectrum at 190–240 nm as shown in Fig. 6. The shape of two curves were similar, showing a positive peak at 196 nm, two well defined negative peak at 207 and 221 nm and a crossover point with X-axis at 203 nm. Clearly, the differences of the two variants peaks were particular at the 190, 207 and 221 peak. The peak intensity of B was greater at 190 nm, however, A was more significant at 207 and 221 nm. Previously reported (Dong et al., 1990) the spectrum at 190 nm was dominated by the strong α-helix and a negative one peaking at 205 to 220 nm by β-sheet. In short, a higher content of α-helix while a lower content of β-sheet was at variant B compared with A. The concrete content of secondary structure of variants can be calculated by JASCO secondary structure software. As shown in Fig. 6B, the two variants contained a large amount of β-sheets and the content of β-sheets in variant A was higher than that of B, which agreed well with the results of Fig. 6A. This would reflect diversities in the packing of the aromatic side chains of the two variants. The earlier results also reported that β-sheets were predominant, in both monomeric and dimeric β-Lactoglobulin variants A and B. Besides, there was no clear difference in the content of β-turn and random coil that accounted for a large proportion in both variants. This behavior confirmed the earlier analysis (Townend et al., 1969) of the β-Lactoglobulin as containing a small amount of α-helix, almost 50% of unordered structure, and the remainder in an antiparallel β conformation. In short, the CD data presented here underlined the major similarities between variants rather than differences.
Secondary structure of purified β-Lactoglobulin variants A and B by Far-UV CD spectroscopy.
A. Far-UV CD spectra; B. Secondary structure contents of two variants
Fourier transform infrared spectroscopy Secondary structure of purified variants had been analyzed by FT-IR based on the amide I band (Vidal et al., 2014), which attributed primarily to the C=O stretching vibrations of the protein backbone in the wavenumbers between 1 600 and 1 700 cm−1. In Fig. 7, an overall similarity between the peak shape and peak intensity of variants A and B in the upper FT–IR spectra was observed and both variants exhibit absorbance maxima for their amide I bands at 1 631 cm−1, indicating a predominant β-sheet structure. Besides, the small differences were at the band ascribed to β-sheet near 1 654 cm−1 and the band assigned to turns at 1 663–1 664 cm−1. The results also reflected the structural differences at amino acid residues 118 and 64, which located at β-sheet and β-turn structure (Monaco et al., 1987). In order to quantitative the areas of the overlapping band, Fourier self-deconvolution, second-derivative and curve-fitting analysis were employed. It revealed a total of nine band components in both variants. The percentage of β-sheet was highest, which represented as 39.96% and 38.91%. This agreed well with CD spectroscopy results. In brief, there was small but readily detectable differences between two variants resulted from substitutions at positions 64 and 118, which caused β-Lactoglobulin A had greater conformational mobility.
FT-IR of purified variants A and B.
Upper curve: original spectrum. Lower curve: curve-fitted Fourier self-deconvoluted and inverted second-derivative spectra of β-Lactoglobulin variants.
A. purified β-Lactoglobulin A; B. purified β-Lactoglobulin B.
Peak 1–9 were sub-peaks of Gaussian curve fitting.
Peaks 1–4 (1 661–1 700 cm−1), β-sheet; Peak 5 (1 650–1 660 cm−1), α-helix; Peak 6 (1 640–1 649 cm−1), random coil; Peak 7 (1 635–1 640 cm−1), no specific secondary structure; Peak 8 (1 623–1 632 cm−1), β-turn; Peak 9 (1 600–1 612 cm−1), side chain of proteins.
Intrinsic fluorescence and ANS fluorescence The tryptophan and tyrosine micro environment (Ioniţă et al., 2014) in β-Lactoglobulin variants can be monitored by intrinsic fluorescence measurement. In Fig. 8A, the two spectrum shapes were very similar and λ max of the variant A and B spectra appeared at 335.0 nm and 335.2 nm. Moreover, the fluorescence intensity of variant A was higher than that of variant B. The surface hydrophobicity of variants was determined by ANS, which can bind to exposed hydrophobic residues of protein and measure the compactness of protein conformation. In Fig. 8B, variant A showed characteristic λ max at 487.2 nm and variant B at 487.6 nm. Besides, variant A showed greater fluorescence intensity (FI) than that of B. The substitution Val of Ala with smaller side chain in variant B resulted in the loss of two methyl groups, promoting formation of a cavity in the hydrophobic core of the protein, which caused significant decrease in the hydrophobic packing of variant B. In addition, FI of ANS binding of both variants decreased greatly comparing with the intrinsic fluorescence, which may contribute to the combining ability of ANS with protein.
Fluorescence Spectroscopy of purified variants A and B.
A. Intrinsic fluorescence spectra of variants; B. ANS fluorescence spectra of variants
It is the first time to purify β-Lactoglobulin variants A and B efficiently by Size-exclusion chromatography with Sephadex G-75 gel followed by Ion-exchange chromatography with DEAE-sepharose Fast Flow gel. The purity of two variants were higher than 90% and β-Lactoglobulin A had a higher concentration than β-Lactoglobulin B in native β-Lactoglobulin. The CD and FT-IR all indicated β-sheet was dominant in both variants. Overall, the obvious decrease in the hydrophobic packing was found in variant B. Meanwhile, variant B showed smaller conformational mobility than variant A, which made β-Lactoglobulin B had a stronger antigenicity than that of A.
Acknowledgments The work was supported by National High Technology Research and Development Program of China (863 Program, No. 2013AA102205), International Science & Technology Cooperation Program of China (No.2013DFG31380), National Natural Science Foundation of China (No.31171716, 31260204 and 31301522), the Research Program of State Key Laboratory of Food Science and Technology(SKLF-ZZB-201510), Jiangxi Province funding program for outstanding youth (20162BCB23016) and Special project of Technology Research from Guangdong Province in 2013 (No.201 090600060).
