Journal of Dairy Science
Volume 91, Issue 2 , Pages 490-499, February 2008

Enhanced Functionalities of Whey Proteins Treated with Supercritical Carbon Dioxide

Department of Food Science and Technology, University of Tennessee, Knoxville 37996

Received 3 September 2007; accepted 23 October 2007.

Article Outline

Abstract 

The functionality of whey proteins can be modified by many approaches; for example, via complexation with carbohydrates, enzymatic cross-linking, or hydrolysis, and the objective of this work was to research the effects of supercritical carbon dioxide (scCO2) treatments on the functionalities of commercial whey protein products including whey protein isolates (WPI) and whey protein concentrates (WPC). The WPI and WPC powders and a 10% (wt/vol) WPI solution were treated with scCO2. The WPI solution was treated at 40°C and 10MPa for 1h, whereas WPI and WPC powders were treated with scCO2 at 65°C and 10 or 30MPa for 1h. Dynamic rheological tests were used to characterize gelation properties before and after processing. Compared with the unprocessed samples and samples processed with N2 under similar conditions, scCO2-treated WPI, whether dispersed in water or in the powder form during treatments, formed a gel with increased strength. The improvement in gelling properties was more significant for the scCO2-treated WPC. In addition, the scCO2-processed WPI and WPC powders appeared to be fine and free-flowing, in contrast to the clumps in the unprocessed samples. Proximate compositional and surface hydrophobicity analyses indicated that both compositional and structural changes may have contributed to enhanced whey protein functionalities. The results suggest that functionalities of whey proteins can be improved by scCO2 treatment to produce novel ingredients.

Key words: whey protein, supercritical carbon dioxide, functionality, rheology

 

Back to Article Outline

Introduction 

Whey proteins have been studied extensively because of their widespread use as nutritious, highly functional but affordable food ingredients. Depending on the separation processes used to extract proteins, variations in product compositions can be obtained. Whey protein concentrates (WPC) are commercial products with a protein content of 25 to 80%, whereas whey protein isolates (WPI) usually have more than 90% protein (Foegeding et al., 2002). Variations in the composition and preparation methods significantly change protein functionality (Damodaran, 1997; van Vliet et al., 2002).

Protein functionality is a general term used to describe any function a protein provides: solubility, water-holding capacity, structure formation in dairy and meat products, the ability to form and stabilize oil-water (in emulsions) and air-water (in foams) interfaces, and the ability to remain in suspension during thermal processing among others (Foegeding et al., 2002). Functional properties of protein ingredients are affected by factors such as origin, preparation method (which leads to different compositions and structure changes), physiochemical properties of the systems in which proteins are incorporated, and storage conditions (van Vliet et al., 2002; Baier and McClements, 2005).

Approaches to modifying protein functionality include complexation with carbohydrates via physical (noncovalent) or chemical (covalent) bonds to enhance functionalities such as interfacial activity (Dickinson and Galazka, 1991; Dickinson and Semenova, 1992), cross-linking (via thermal aggregation or enzymatic reaction), and hydrolysis (Foegeding et al., 2002). When modifying protein functionality, however, attention must be paid to regulations that may limit the application of these modified products as ingredients in foods (McClements, 2005).

Modifying protein functionality with physical methods may encounter fewer regulatory hurdles. High hydrostatic or dynamic pressure has been investigated as a physical method to modify whey and soy protein functionality. The irreversible change in structure needed to alter protein functionality of α-LA, β-LG, and glycinin occurs above 400, 150, and 400MPa, respectively (Tanaka et al., 1996a, b; Tanaka and Kunugi, 1996; Molina et al., 2001).

Supercritical carbon dioxide (scCO2) may be a more practical and affordable approach to modifying protein functionalities, a fact that has been exemplified by the commercial production of decaffeinated coffee. Carbon dioxide has also been used as a weak acid to precipitate milk and soy proteins dispersed in aqueous solutions (Tomasula et al., 1997, 1998; Strange et al., 1998; Hofland et al., 1999; Thiering et al., 2001a,b; Tomasula and Yee, 2001). For the case of whey proteins, separation of WPC dispersions into α-albumin-rich or β-LG-rich fractions can be achieved by manipulating precipitation conditions (Tomasula et al., 1998; Tomasula and Yee, 2001). Compared with a reference WPC solution, the fractionated β-LG-rich fraction enhanced interfacial activity and thickening and gelling properties (Tomasula and Yee, 2001). Another technique of supercritical fluid extrusion has been developed to process foods with novel functionalities properties (Alavi et al., 1999; Gogoi et al., 2000).

Thermodynamically, scCO2 has a liquid-like density and gas-like diffusivity and viscosity (McHugh and Krukonis, 1994). Supercritical conditions of CO2 are relatively easy to achieve (critical temperature of 31.1°C and critical pressure of 7.38MPa). Further, CO2 is an environmentally friendly solvent that is ideal for food processing. It is inflammable, nontoxic, chemically inert, physiologically safe, and can be easily recycled and reused. Further, scCO2 has a surface tension of zero (McHugh and Krukonis, 1994), which leads to complete and rapid wetting, thus allowing penetration of complex structures (Zhang et al., 2006). The scCO2 can be used to extract lipids and other nonpolar and polar (with a cosolvent such as ethanol) compounds in commercial whey protein ingredients and cause direct changes in compositions. Protein structures may also change due to the depletion or redistribution of nonpolar compounds. These aspects may introduce novel protein functionalities.

The objective of this work was thus to evaluate the functionalities of whey proteins before and after scCO2 processing, with a focus on gelation properties. Gelation of whey proteins has been studied extensively at various solvent conditions. pH values of 3.35 and 7.0 were selected as representive conditions below and above the isoelectric point of whey proteins, respectively; both pH values have been used to evaluate gelation properties of whey protein ingredients (Ikeda et al., 1999; Resch et al., 2005). The outcome of this work may lead to specialized ingredients of novel applications.

Back to Article Outline

Materials and Methods 

Materials 

A WPC 80% sample and a WPI sample were kindly supplied by Davisco Foods International Inc. (Eden Prairie, MN). Samples were stored at −20°C upon arrival and used directly without further purification. All processed and unprocessed powder samples were analyzed for total protein (AOAC, 1990; method 984.13), lipids (AOAC, 1990; method 948.15), and moisture (AOAC, 1990; method 950.46) contents by ABC Research Corporation (Gainesville, FL). Duplicate tests were done for protein and moisture contents, but only one test was performed for the total lipids content because of the insufficiency of sample quantities from the same batch treatment.

Equipment and Processing Protocols 

The scCO2 system (model SAS50, Thar Technologies, Pittsburgh, PA) is illustrated schematically in Figure 1. Bone-dry grade CO2 (Airgas, Chicago, IL) from a cylinder was cooled to 3°C by a heat exchanger and delivered by a CO2 pump continuously at a controlled mass flow rate. The CO2 was heated to a set temperature by another heat exchanger before entering the pressure vessel (500-mL volume). The vessel pressure was regulated by an automatic pressure regulator downstream, and the vessel temperature was controlled by an outer heat jacket.

In a typical experiment, 25g of WPI or WPC powder was contained in a sample basket with a 5-μm frit at the bottom, which was then placed into the pressure vessel. The pressure vessel cap was then closed, and the valve to the CO2 cylinder was opened. All processing parameters were controlled by computer. A treatment time was defined as that after the system reached the set pressure and temperature (in about 5min). After 1h treatment with CO2 at a mass flow rate of 50 g/min, the vessel was depressurized in about 15min and the sample was collected and stored at −20°C before rheological tests. In another set of experiments, the sample basket was replaced by an open-top stainless steel container with a similar diameter, filled with 100mL of a 10% (wt/vol in deionized water without pH adjustment) WPI solution and processed with scCO2 as above. A control was processed similarly with compressed nitrogen (N2) for both WPI powders and a 10% (wt/vol) WPI solution.

Because BSA shows partial, reversible unfolding between 42 and 50°C (Lin and Koenig, 1976), a temperature of 40°C was chosen for the WPI solution treatments to single out the scCO2 effects from the thermal denaturation of proteins. On the other hand, because a higher temperature facilitates the extraction of lipids from powder samples (Froning et al., 1990; Manganiello et al., 2000; Wolf et al., 2003), a temperature of 65°C was selected, while keeping the temperature below the denaturation temperatures of α-LA (67.1°C) and β-LG (76.3°C; Ju et al., 1999), the 2 major whey proteins.

Rheological Tests 

Sample Preparation 

Five grams of WPC or WPI powder was hydrated in 50mL of deionized water overnight at room temperature. Then, the pH was adjusted to 3.35 or 7.0 with 1 N HCl or 1 N NaOH. The final powder concentration was 9.9% (wt/vol) at pH 7.0 and 9.4% (wt/vol) at pH 3.35. The pH 7.0 samples also contained 100mM NaCl. The scCO2-processed 10% WPI solution had a pH of 5.80 after opening the vessel (a time lag of approximately 30min after stopping the CO2 stream, much slower depressurization than processing powders to prevent extensive foaming) and some protein precipitated (due to acidification). The suspension was continuously stirred overnight at room temperature to deplete the dissolved CO2 and dissolve the precipitate. The stirred solution recorded no difference in pH (pH = 6.75) from the unprocessed sample before pH adjustment and was then adjusted to pH 3.35 with 1 N HCl. The solubility of N2 in water is low, and the WPI solution processed by N2 did not have the foaming issue and recorded no pH change. The pH adjustment to 3.35 was therefore done shortly after processing. The amount of 1 N HCl used to adjust pH to 3.35 was identical for the unprocessed WPI solution and the solutions processed by CO2 and N2.

Rheological Test Procedures 

Dynamic rheological tests were performed with an AR2000 rheometer (TA Instruments, New Castle, DE) using a Searle set up (bob, outside diameter = 28mm and cup, inside diameter = 30mm). After positioning the bob and removing the excess sample, a layer of mineral oil was applied on top of the sample, and a sealing cap was used to minimize moisture loss during measurements. A 1% strain amplitude (within the linear viscoelasticity regimen) was used for the following steps: 1) a linear heating step from 20 to 90°C at 1°C/min, 2) an isothermal step at 90°C for 3h, 3) a frequency sweep step from 0.01 to 10Hz at 90°C, 4) a linear cooling step from 90 to 20°C at 1°C/min, and 5) a frequency sweep step from 0.01 to 10Hz at 20°C. The oscillation frequency during steps 1, 2, and 4 was 1Hz. Triplicate tests were performed for each sample.

Surface Hydrophobicity of Proteins 

Salting-out properties of WPI were used as an indirect measurement of relative protein surface hydrophobicity (Lindahl et al., 1981). Different concentrations of ammonium sulfate, (NH4)2SO4 (0 to 3.465 M, pH 3.35) were prepared by mixing different proportions of a (NH4)2SO4 stock solution (3.465 M in 20mM sodium phosphate, adjusted to pH 3.35) and a 20mM sodium phosphate solution (pH 3.35). The (NH4)2SO4 solution (990μL) was mixed with 10μL of the WPI pH 3.35 samples (the unheated leftover from rheological tests), and the mixture was continuously agitated for 30min on an end-to-end shaker (Lab Industries Inc., Berkeley, CA) at room temperature. The suspension was then centrifuged at 14,500×g for 15min (MiniSpin Personal, Eppendorf, Westbury, NY). The supernatant was analyzed for the total protein concentration by the Bradford method (Bradford, 1976) using an assay kit containing Coomassie (Bradford) reagent and BSA standard from Pierce Biotechnology (Rockford, IL). All samples were measured in triplicate; results, in averages and 95% confidence intervals, were reported by normalizing to the corresponding sample protein concentration at 0 M (NH4)2SO4.

Scanning Electron Microscopy 

The scanning electron microscopy tests were performed with a LEO 1525 SEM microscope (LEO Electron Microscopy, Oberkochen, Germany) for WPI and WPC powder samples before and after processing. The powder sample was glued directly onto an adhesive tab mounted on the specimen stub and sputter-coated with a thin layer of gold (∼5nm).

Statistical Analysis 

An ANOVA was used to evaluate the compositional and rheological data; the Tukey test was performed with the assistance of JMP software (version 6.0, SAS Institute, Cary, NC). However, statistical analysis was not performed for the lipids content because only one test was done for each sample.

Back to Article Outline

Results and Discussion 

Sample Compositions 

Proximate composition analyses of WPI and WPC powders before and after processing are listed in Table 1. The measured compositions of the unprocessed WPI sample were similar to the product label that listed 92.7% protein, 0.3% fat, and 5.0% moisture. The moisture content of WPI powders decreased slightly after scCO2 processing, and the decrease was more significant for the 30MPa treatment than for the 10MPa treatment. Also decreased was the lipid content, which was expected from the well-known capability of scCO2 to extract lipids. The protein content per unit mass of powders was statistically greater for the scCO2-processed samples. In contrast, there was no significant difference in moisture and protein contents between the unprocessed sample and that treated with N2. These data indicate that scCO2, because of zero surface tension, had much better penetration capability than N2 at similar temperatures and pressures, which increased the rate of moisture and lipids transportation out of protein powder granules.

Table 1. Sample compositions of whey protein concentrate (WPC) and whey protein isolate (WPI) powders1 unprocessed and processed with supercritical CO2 (scCO2) or N2
SampleTemperature (°C)Pressure (MPa)Composition
Protein1 (%)Lipids2 (%)Moisture1 (%)
Unprocessed WPI——91.21±0.16b0.135.93±0.17a
scCO2-processed WPI651092.27±0.13a0.115.17±0.08ab
scCO2-processed WPI653092.71±0.01a0.054.41±0.29bc
N2-processed WPI651090.48±0.19b0.215.80±0.23a
Unprocessed WPC——75.46±0.21e7.164.44±0.46bc
scCO2-processed WPC651076.63±0.15d7.743.51±0.12c
scCO2-processed WPC653078.58±0.44c6.413.53±0.28c

a–eMeans with a different letter in each column are statistically different.

1Values are averages±standard deviations from 2 tests.

2Only 1 replicate was performed, as explained in the Materials and Methods section.

For WPC, the measured compositions of unprocessed WPC sample were also similar to the labeled 78.0% protein, 8.0% fat, and 4.5% moisture. In comparison, reduced moisture content was recorded for the scCO2-treated samples. The lipids content after the 10MPa treatment was similar to that in the unprocessed samples, possibly due to the inefficiency of the CO2 extraction at this pressure. Conversely, the sample after the 30MPa treatment had a lower lipid content. The protein content significantly increased after the scCO2 extraction.

Sample Appearance and Notable Sensory Changes 

Visual appearance of processed and unprocessed whey protein powder samples is presented in Figure 2. The unprocessed WPI (panel A) and that processed by N2 (panel D) had loose clumps and did not move freely, whereas those processed by scCO2 appeared as fine powders (panels B and C). Similarly, the unprocessed WPC had many clumps (panel E) that still existed after being processed by scCO2 at 65°C and 10MPa (panel F); but no more clumps were observed after processing with scCO2 at 65°C and 30MPa (panel G). The changes in lipid content (Table 1), although lacking statistical support, seemed to correlate with the sample appearance depicted in Figure 2: treatments that lowered the lipid content (panels B and C vs. panels A and D; panel G vs. panels E and F) may have also reduced the chance of sticking between powder granules. When observed with scanning electron microscopy, granules appeared irregular in shape and size, and no strong conclusions could be drawn for the samples before and after processing with scCO2 or N2 (Figure 3). Therefore, the advantages from scCO2-processed powders include better flowability and a better visual appearance that may attract ingredient buyers and consumers.

  • View full-size image.
  • Figure 2. 

    Appearance of powders (5g) of A) unprocessed whey protein isolate (WPI); B) WPI processed with CO2 at 65°C and 10MPa; C) WPI processed with CO2 at 65°C and 30MPa; D) WPI processed with N2 at 65°C and 10MPa; E) unprocessed whey protein concentrate (WPC); F) WPC processed with CO2 at 65°C and 10MPa; and G) WPC processed with CO2 at 65°C and 30MPa.

  • View full-size image.
  • Figure 3. 

    Scanning electron microscopy images for powders of A) unprocessed whey protein isolate (WPI); B) WPI processed with supercritical (sc)CO2 at 65°C and 10MPa; C) WPI processed with scCO2 at 65°C and 30MPa; D) WPI processed with N2 at 65°C and 10MPa; E) unprocessed whey protein concentrate (WPC); F) WPC processed with scCO2 at 65°C and 10MPa; and G) WPC processed with scCO2 at 65°C and 30MPa. Scale bar = 100μm.

In addition, WPC and WPI powders had a reduced creamy, sweet aroma when processed with scCO2. This may have been due to the well-established fact that scCO2 is capable of extracting small nonpolar molecules (Maheshwari et al., 1995). Sensory analyses and characterization of volatile compounds, however, are future research focuses. Nevertheless, whey proteins with a milder aroma and taste may be incorporated in products in which the aroma and flavor of protein ingredients are undesirable, while providing other functionalities of proteins.

Rheological Data of WPI Solutions Processed by scCO2 and N2 

For the WPI solution that interacted with scCO2, the sample after resuspending precipitates and adjusting pH to 3.35 showed a steady development in storage modulus upon heating (Figure 4A), further strengthened during cooling (Figure 4B). Storage modulus (G′) of the scCO2-processed WPI solution was greater than that of the unprocessed sample and that of the sample processed with N2 under the same conditions (Figure 4). This approach of modifying WPI suspended in solutions was not pursued further because similar results were obtained from scCO2-processed powder samples (Table 2), an easier operation that does not present the foaming issue and is easier to scale up. Solutions processed by scCO2 and N2 were not analyzed for composition with the assumption that no significant compositional changes would occur because proteins were dispersed in water. Water evaporation during processing was expected via the container opening due to the continuous gas stream inside the pressure vessel. However, it would be expected that the amount of water evaporated would be similar for scCO2 and N2 treatments. The apparent increase in gel strength for the scCO2-treated solution compared with the controls (unprocessed and that processed by N2) may have been caused by the dissolved scCO2 that interacted with proteins, causing irreversible structure changes after CO2 depletion.

  • View full-size image.
  • Figure 4. 

    Storage modulus of 9.4% (wt/vol) whey protein isolate (WPI) solutions (pH 3.35) processed with supercritical CO2 or N2 and unprocessed during (A) heating from 20 to 90°C at 1°C/min and holding at 90°C for 3h and (B) during cooling from 90 to 20°C at 1°C/min. Data are averages from 3 tests.

Table 2. Storage modulus (G′)1 of whey protein isolate (WPI) or whey protein concentrate (WPC) samples (9.9%, wt/vol, at pH 7.0; 9.4%, wt/vol, at pH 3.35) unprocessed and processed with supercritical CO2 (scCO2) or N2
SampleTemperature (°C)Pressure (MPa)G′ (Pa) after step (2) at 90°CG′ (Pa) after step (4) at 20°C
pH 3.35pH 7pH 3.35pH 7
Unprocessed WPI——449±19c6,198±66a1,279±31b19,197±138a
scCO2-processed WPI solution6510586±7a—1,518±15a—
N2-processed WPI solution6530421±14c—1,200±16b—
scCO2-processed WPI powder6510526±25a6,262±11a1,518±20a21,130±132a
scCO2-processed WPI powder6530508±13b6,599±47a1,519±30a20,590±355a
N2-processed WPI powder6510419±9c—1,229±8b—
Unprocessed WPC——55±3e334±17c190±9e2,025±42c
scCO2-processed WPC powder651075±7e397±7bc314±14d2,253±16bc
scCO2-processed WPC powder6530113±12d443±15b405±13c2,603±37b

a–eMeans with a different superscript letter within a column are statistically different.

1Results are averages±95% confidence intervals from 3 tests.

Phase-angle changes of WPI solutions during the second step (isothermal at 90°C) where storage modulus developed steadily are plotted in Figure 5. The tangential function of a phase angle is defined as a ratio of loss (G′) to storage (G′) modulus. A phase angle <45° indicates a storage modulus that dominates the loss modulus, and a system behaves more elastic- or solid-like. Therefore, the phase angle is a convenient way of monitoring the gelation process and a system with a phase angle <45° is commonly referred to as a gel (Ross-Murphy, 1994). With the assumption that all 3 samples had identical composition, the WPI processed with scCO2 reached a phase angle of 45° (i.e., gelation) faster than the unprocessed WPI and that processed with N2 (insert in Figure 5). The latter 2 samples had an almost identical gelation time. The differences in gelation time further indicated that the dissolved CO2 changed whey protein structures (CO2 has a much greater solubility in water than does N2), which enabled a faster gelation and a stronger gel. Both computer simulations (of lysozyme) and experiments (from films of albumin and lysozyme) have shown that scCO2 alternates the secondary structure (the relative amount of α-helix, β-sheet, and random coil) of proteins (Striolo et al., 2003; Liu et al., 2004). Further studies are needed to illustrate the exact mechanisms of whey protein molecular structure changes induced by CO2 and how these changes correlate with rheological properties.

  • View full-size image.
  • Figure 5. 

    Phase angle of 9.4% (wt/vol) whey protein isolate (WPI) solutions (pH 3.35) processed with supercritical CO2 or N2 and unprocessed during holding at 90°C for 3h after heating from 20 to 90°C at 1°C/min. Data are averages from 3 tests.

Frequency sweep tests during step 3 (at 90°C) and step 5 (at 20°C) showed that G′ was much greater than G″, exemplified in Figure 6 from step 5, indicating that all samples formed weak gels. Data suggested that the gel formed from the CO2-treated WPI was statistically stronger than the other 2 samples (P<0.05).

  • View full-size image.
  • Figure 6. 

    Mechanical spectra of 9.4% (wt/vol) whey protein isolate (WPI) solutions (pH 3.35) processed with supercritical CO2 or N2 and unprocessed after cooling to 20°C. Data are averages from 3 tests and error bars are 95% confidence intervals. G′ = storage modulus; G″ = loss modulus.

Rheological Data of WPI Powders Processed by scCO2 and N2 

Comparing thermal gelation properties, WPI powders processed by scCO2 formed a stronger gel (higher G′) than those unprocessed or processed by N2 (Figure 7). During the heating and holding steps, the sample extracted by scCO2 at 65°C and 10MPa formed the strongest gel, whereas that extracted by scCO2 at 65°C and 30MPa was slightly weaker (Figure 7A). Upon cooling (Figure 7B), the 2 scCO2-treated samples had an almost identical gel strength. Comparing the compositions of 2 scCO2-processed samples (Table 1), there was no statistical difference in protein content, which might have led to a similar gel strength after cooling.

  • View full-size image.
  • Figure 7. 

    Storage modulus (G′) of 9.4% (wt/vol) whey protein isolate (WPI) solutions (pH 3.35) constituted from powders processed with supercritical CO2 or N2 and that unprocessed during (A) heating from 20 to 90°C at 1°C/min and holding at 90°C for 3h and (B) cooling from 90 to 20°C at 1°C/min. Data are averages from 3 tests.

The sample processed by scCO2 at 30MPa showed a much faster gelation (a shorter time to reach a phase angle of 45°) than the other 3 samples (Figure 8). The gelation time for the unprocessed sample was similar to that of samples processed by scCO2 at 10MPa and shorter than for samples processed by N2. The gelation times correlated well with protein content (R2 = 0.95) and lipid content (R2 = 0.98) but not with moisture content (R2 = 0.65).

  • View full-size image.
  • Figure 8. 

    Phase angle of 9.4% (wt/vol) whey protein isolate (WPI) solutions (pH 3.35) constituted from powders processed with supercritical CO2 or N2 and that unprocessed during holding at 90°C for 3h after heating from 20 to 90°C at 1°C/min. Data are averages from 3 tests.

The time to reach a phase angle of 45° correlated with observations from the “salting out” of proteins (Figure 9). The WPI powder after scCO2 processing at 65°C and 30MPa had the greatest soluble protein content after precipitation by ammonium sulfate, indicating that the sample had the lowest surface hydrophocity. This lowered surface hydrophobicity, possibly due to removal of bound lipids (Table 1), may have induced the faster gelation. There was no difference in surface hydrophobicity for the unprocessed sample and that processed by scCO2 at 65°C and 10MPa (Figure 9), which correlated with no difference in the gelation time for these 2 samples (Figure 8). Conversely, the sample processed by N2 had the poorest protein solubility during precipitation (Figure 9), which is reflected by the longest gelation time of this sample (Figure 8). It seemed that surface hydrophobicity may have contributed to the observed differences in gelation kinetics. Eventually, if the sample is allowed enough time to form chemical and physical bonds, the total protein concentration (Table 1) may be a critical factor determing the final gel strength (Table 2, Figure 7), because heating denatures whey proteins before formation of the bonds.

  • View full-size image.
  • Figure 9. 

    Salting-out of whey protein isolate (WPI) samples at pH 3.35 by different concentrations of ammonium sulfate. The remaining soluble protein concentrations are normalized to that at 0 M ammonium sulfate.

Storage moduli of samples following rheological test steps 2 and 4 are compiled in Table 2. When correlating compositions in Table 1 and G′ data in Table 2, protein contents correlated much better with rheological data at pH 3.35 (R2 = 0.89 for G′ after step 2 and 0.94 for G′ after step 4) than lipid and moisture contents (all R2<0.77). The unprocessed sample and that processed by N2 showed no statistical difference in G′, whereas the scCO2-processed samples formed gels with similar strength. It is also interesting to note that there was no statistical difference in gel strength for the scCO2 treatment of WPI in either powder form or dispersed in water. However, to judge whether the same interactions between protein and scCO2 occurred for the solution and powder treatments, detailed information of composition, surface hydrophobicity, secondary (possibly tertiary) structures of proteins, and interactions between proteins and lipids is needed.

Storage moduli of samples at pH 7.0 and 0.1 M NaCl were much greater than the corresponding sample at pH 3.35 (Table 2). Enhanced gel strength was also observed for scCO2-processed WPI at pH 7.0, but the increase was not statistically significant. Ikeda et al. (1999) studied whey protein gelation with different NaCl concentrations at pH 7.0 and observed the strongest gel at 0.1 M NaCl. Therefore, the addition of 0.1 M NaCl to the WPI sample may have lessened the enhanced gel strength due to scCO2 processing.

Rheological Data of WPC Powders Processed by scCO2 

Compared with the unprocessed sample, the WPC powders processed by scCO2 formed a gel with a much greater storage modulus (Figure 10), with the 30MPa treatment being more significant than the 10MPa treatment. Table 1 shows statistical differences of protein content among all samples, but only the 30MPa treatment lowered the lipids content. The enhanced gelling properties may have been induced by compositional changes and the interaction between scCO2 and lipid-bound protein in WPC. Greater lipid content in WPC compared with WPI signified the enhancement in gelation properties by scCO2 treatments.

  • View full-size image.
  • Figure 10. 

    Storage modulus (G′) of 9.4% (wt/vol) whey protein concentrate (WPC) solutions (pH 3.35) constituted from powders processed with supercritical CO2 and that unprocessed during (A) heating from 20 to 90°C at 1°C/min and holding at 90°C for 3h and (B) cooling from 90 to 20°C at 1°C/min. Data are averages from 3 tests.

When comparing phase-angle changes of WPC samples during step 2, the gelation time followed a decreasing order of 30MPa treatment, 10MPa treatment, and untreated sample (Figure 11). Storage modulus data after test steps 2 and 4 are summarized in Table 2. At pH 3.35, G′ increased from 55 to 75Pa for the 10MPa treatment and to 113Pa for the 30MPa treatment after step 2; after step 4, G′ increased from 190 to 314Pa for the 10MPa treatment and to 405Pa for the 30MPa treatment. Similar to WPI, protein contents correlated much better with rheological data at pH 3.35 (R2 = 1.00 for G′ after step 2 and 0.95 for G′ after step 4) than did lipid and moisture contents (all R2<0.81). The improvement in the relative gel strength after scCO2 processing was less significant at pH 7.0: G′ increased from 334 to 397Pa for the 10MPa treatment and to 443Pa for the 30MPa treatment after step 2; and G′ increased from 2,025 to 2,253Pa for the 10MPa treatment and to 2,603Pa for the 30MPa treatment after step 4. Similar to pH 3.35, protein contents correlated much better with rheological data at pH 7.0 (R2 = 0.95 for G′ after step 2 and 1.00 for G′ after step 4) compared with lipid and moisture contents (all R2<0.81).

  • View full-size image.
  • Figure 11. 

    Phase angle of 9.4% (wt/vol) whey protein concentrate (WPC) solutions (pH 3.35) constituted from powders processed with supercritical CO2 and unprocessed during holding at 90°C for 3h after heating from 20 to 90°C at 1°C/min. Data are averages from 3 tests.

In summary, gelling properties of whey proteins, either dissolved in water or in powder form, can be improved after scCO2 processing. Compositional and surface hydrophobicity analyses indicated that rheological properties were affected not only by composition (particularly protein content), but also by structural changes due to the scCO2 treatment. Removal of lipids from WPC was incomplete and may be improved further under other conditions, making scCO2 a low-cost purification technique compared with those based on membrane filtrations or chromatography. We expect that compositional and structural changes will also significantly change interfacial properties of proteins; this aspect will be studied in the near future. We are currently extending extraction conditions and characterizing the corresponding functionality changes over a broader range of testing conditions and techniques, with a goal of establishing correlations between extraction, composition, and structure and functionalities of whey proteins.

Back to Article Outline

Acknowledgments 

This project was supported by the Tennessee Agricultural Experiment Station Hatch project and UTIA Idea Grant. Discussions with Allen Foegeding, Jochen Weiss, and Jack Davis are highly appreciated.

Back to Article Outline

Supplementary data 

Download file

download text

Interpretive summary.

Back to Article Outline

References 

  1. Alavi SH, Gogoi BK, Khan M, Bowman BJ, Rizvi SSH. Structural properties of protein-stabilized starch-based supercritical fluid extrudates. Food Res. Int. 1999;32:107–118
  2. AOAC. Official Methods of Analysis. 15th ed.. Arlington, VA: AOAC; 1990;
  3. Baier SK, McClements DJ. Influence of cosolvent systems on the gelation mechanism of globular protein: Thermodynamic, kinetic, and structural aspects of globular protein gelation Comp. Rev. Food Sci. Food Safety. 2005;4:43–54
  4. Bradford MM. A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 1976;72:248–254
  5. Damodaran S. Food proteins: An overview. In:  Damodaran S,  Paraf A editor. Food Proteins and Their Application. New York, NY: Marcel Dekker; 1997;p. 1–21
  6. Dickinson E, Galazka VB. Emulsion stabilization by ionic and covalent complexes of beta-lactoglobulin with polysaccharides. Food Hydrocoll. 1991;5:281–296
  7. Dickinson E, Semenova MG. Emulsifying properties of covalent protein dextran hybrids. Colloids Surf. 1992;64:299–310
  8. Foegeding EA, Davis JP, Doucet D, McGuffey MK. Advances in modifying and understanding whey protein functionality. Trends Food Sci. Technol. 2002;13:151–159
  9. Froning GW, Wehling RL, Cuppett SL, Pierce MM, Niemann L, Siekman DK. Extraction of cholesterol and other lipids from dried egg yolk using supercritical carbon dioxide. J. Food Sci. 1990;55:95–98
  10. Gogoi BK, Alavi SH, Rizvi SSH. Mechanical properties of protein-stabilized starch-based supercritical fluid extrudates. Int. J. Food Prop. 2000;3:37–58
  11. Hofland GW, van Es M, van der Wielen LAM, Witkamp GJ. Isoelectric precipitation of casein using high-pressure CO2. Ind. Eng. Chem. Res. 1999;38:4919–4927
  12. Ikeda S, Foegeding EA, Hagiwara T. Rheological study on the fractal nature of the protein gel structure. Langmuir. 1999;15:8584–8589
  13. Ju ZY, Hettiarachchy N, Kilara A. Thermal properties of whey protein aggregates. J. Dairy Sci. 1999;82:1882–1889
  14. Lin VJC, Koenig JL. Raman studies of bovine serum albumin. Biopolymers. 1976;15:203–218
  15. Lindahl M, Faris A, Wadstrom T, Hjerten S. A new test based on “salting out” to measure relative surface hydrophobicity of bacterial cells. Biochim. Biophys. Acta. 1981;677:471–476
  16. Liu HL, Hsieh WC, Liu HS. Molecular dynamics simulations to determine the effect of supercritical carbon dioxide on the structural integrity of hen egg white lysozyme. Biotechnol. Prog. 2004;20:930–938
  17. Maheshwari P, Ooi ET, Nikolov ZL. Off-flavor removal from soy-protein isolate by using liquid and supercritical carbon dioxide. J. Am. Oil Chem. Soc. 1995;72:1107–1115
  18. Manganiello L, Ríos A, Valcárcel V. Automatic micro-gravimetric determination of fats in milk products by use of supercritical fluid extraction with on-line piezoelectric detection. J. Chromatogr. A. 2000;874:265–274
  19. McClements DJ. Food Emulsions: Principles, Practices, and Techniques. 2nd ed.. Boca Raton, FL: CRC Press; 2005;
  20. McHugh MA, Krukonis VJ. Supercritical Fluid Extraction: Principles and Practice. 2nd ed.. Boston, MA: Butterworth-Heinemann; 1994;
  21. Molina E, Papadopoulou A, Ledward DA. Emulsifying properties of high pressure treated soy protein isolate and 7S and 11S globulins. Food Hydrocoll. 2001;15:263–269
  22. Resch JJ, Daubert CR, Foegeding EA. The effects of acidulant type on the rheological properties of beta-lactoglobulin gels and powders derived from these gels. Food Hydrocoll. 2005;19:851–860
  23. Ross-Murphy SB. Rheological Methods. In:  Ross-Murphy SB editors. Physical Techniques for the Study of Food Biopolymers. London, UK: Blackie Academic & Professional; 1994;p. 343
  24. Strange ED, Konstance RP, Tomasula PM, van Hekken DL, Smith PW, Boswell TR, et al. Functionality of casein precipitated by carbon dioxide. J. Dairy Sci. 1998;81:1517–1524
  25. Striolo A, Favaro A, Elvassore N, Bertucco A, Di Noto V. Evidence of conformational changes for protein films exposed to high-pressure CO2 by FT-IR spectorscopy. J. Supercrit. Fluid. 2003;27:283–295
  26. Tanaka N, Koyasu A, Kobayashi I, Kunugi S. Pressure-induced change in proteins studied through chemical modifications. Int. J. Biol. Macromol. 1996;18:275–280
  27. Tanaka N, Kunugi S. effect of pressure on the denaturation exchange reaction of alpha-lactalbumin and beta-lactoglobulin. Int. J. Biol. Macromol. 1996;18:33–39
  28. Tanaka N, Tsurui Y, Kobayashi I, Kunugi S. The pressure-induced structural change of bovine alpha-lactabumin as studied by fluorescence hydrophobic probe. Int. J. Pept. Protein Res. 1996;48:259–264
  29. Thiering R, Hofland G, Foster N, Witkamp GJ, van de Wielen L. Carbon dioxide induced soybean protein precipitation: Protein fractionation, particle aggregation, and continuous operation. Biotechnol. Prog. 2001;17:513–521
  30. Thiering R, Hofland G, Foster N, Witkamp GJ, van de Wielen L. Fractionation of soybean proteins with pressurized carbon dioxide as a volatile electrolyte. Biotechnol. Bioeng. 2001;73:1–11
  31. Tomasula PM, Craig JC, Boswell RT. A continuous process for casein production using high-pressure carbon dioxide. J. Food Eng. 1997;33:405–419
  32. Tomasula PM, Parris N, Boswell RT, Moten RO. Preparation of enriched fractions of alpha-lactalbumin and beta-lactoglobulin from cheese whey using carbon dioxide. J. Food Process. Pres. 1998;22:463–476
  33. Tomasula PM, Yee WCF. Enriched fractions of alpha-lactalbumin (alpha-LA) and beta-lactoglobulin (beta-LG) from whey protein concentrate using carbon dioxide. Functional properties in aqueous solution. J. Food Process. Pres. 2001;25:267–282
  34. van Vliet T, Martin AH, Bos MA. Gelation and interfacial behavior of vegetable proteins. Curr. Opin. Colloid Interf. Sci. 2002;7:462–468
  35. Verheul M, Roefs SPFM. Structure of particulate whey protein gels: Effect of NaCl concentration, pH, heating temperature, and protein composition J. Agric. Food Chem. 1998;46:4909–4916
  36. Wolf WR, LaCroix DE, Goel R, Kaur M. Total fat analysis in milk- and soy-based infant formula powder by supercritical fluid extraction. J. Am. Oil Chem. Soc. 2003;80:853–857
  37. Zhang J, Burrows S, Gleason C, Matthews MA, Drews MJ, LaBerge M, et al. Sterilization of Bacillus pumilus spores using supercritical carbon dioxide. J. Microbiol. Meth. 2006;66:479–485

PII: S0022-0302(08)71391-2

doi:10.3168/jds.2007-0663

Journal of Dairy Science
Volume 91, Issue 2 , Pages 490-499, February 2008

Article Tools

* Image Usage & Resolution