Advertisement

Compositional and Functional Properties of Buttermilk: A Comparison Between Sweet, Sour, and Whey Buttermilk1

      Abstract

      Buttermilk is a dairy ingredient widely used in the food industry because of its emulsifying capacity and its positive impact on flavor. Commercial buttermilk is sweet buttermilk, a by-product from churning sweet cream into butter. However, other sources of buttermilk exist, including cultured and whey buttermilk obtained from churning of cultured cream and whey cream, respectively. The compositional and functional properties (protein solubility, viscosity, emulsifying and foaming properties) of sweet, sour, and whey buttermilk were determined at different pH levels and compared with those of skim milk and whey. Composition of sweet and cultured buttermilk was similar to skim milk, and composition of whey buttermilk was similar to whey, with the exception of fat content, which was higher in buttermilk than in skim milk or whey (6 to 20% vs. 0.3 to 0.4%). Functional properties of whey buttermilk were independent of pH, whereas sweet and cultured buttermilk exhibited lower protein solubility and emulsifying properties as well as a higher viscosity at low pH (pH ≤ 5). Sweet, sour, and whey buttermilks showed higher emulsifying properties and lower foaming capacity than milk and whey because of the presence of milk fat globule membrane components. Furthermore, among the various buttermilks, whey buttermilk was the one showing the highest emulsifying properties and the lowest foaming capacity. This could be due to a higher ratio of phospholipids to protein in whey buttermilk compared with cultured or sweet buttermilk. Whey buttermilk appears to be a promising and unique ingredient in the formulation of low pH foods.

      Key words

      Introduction

      Buttermilk is the aqueous phase released during the churning of cream in butter manufacture. It contains all the water-soluble components of cream such as milk protein, lactose, and minerals. It also encloses material derived from milk fat globule membrane (MFGM), which is disrupted during the churning and mostly migrates to the buttermilk fraction (
      • Corredig M.
      • Dalgleish D.G.
      Isolates from industrial buttermilk: emulsifying properties of materials derived from the milk fat globule membrane.
      ). Buttermilk contains more phospholipids than milk because of its high content in MFGM material, which is rich in phospholipids that constitute about one-third of the MFGM DM (
      • Mulder H.
      • Walstra P.
      The milk fat globule.
      ). For instance,
      • Elling J.L.
      • Duncan S.E.
      • Keenan T.W.
      • Eigel W.N.
      • Boling J.
      Composition and microscopy of reformulated creams from reduced-cholesterol butteroil.
      reported 7 times more phospholipids in buttermilk than in whole milk, with concentrations equal to 0.89 mg/g and 0.12 mg/g, respectively.
      • Christie W.W.
      • Noble R.C.
      • Davies G.
      Phospholipids in milk and dairy products.
      determined a 4-fold increase of phospholipids in buttermilk compared with whole milk, with a phospholipid content of 0.72 mg/mL and 0.15 mg/mL, respectively. The high content of phospholipids in buttermilk makes this dairy ingredient interesting for use as a functional ingredient because of the emulsifying properties of phospholipids (
      • Elling J.L.
      • Duncan S.E.
      • Keenan T.W.
      • Eigel W.N.
      • Boling J.
      Composition and microscopy of reformulated creams from reduced-cholesterol butteroil.
      ;
      • Corredig M.
      • Dalgleish D.G.
      Buttermilk properties in emulsions with soybean oil as affected by fat globule.
      ;
      • Wong P.Y.Y.
      • Kitts D.D.
      A comparison of the butter milk solids functional properties to nonfat dried milk, soy protein isolate, dried egg white, and egg yolk powders.
      ). In addition, phospholipids have been shown to possess biological activity. Some studies have demonstrated the anticarcinogenic potential of phospholipids, especially against colon cancer (
      • Dillehay D.L.
      • Webb S.J.
      • Schmelz E.M.
      • Merill A.H.
      Dietary sphingomyelin inhibits 1,2-dimethylhydrazine-induced colon cancer in CF1 mice.
      ;
      • Schmelz E.M.
      • Dillehay D.L.
      • Webb S.K.
      • Reiter A.
      • Adams J.
      • Merill A.H.
      Sphingomyelin consumption supresses aberrant colonic crypt foci and increases the proportion of adenomas versus adenocarcinomas in CF1 mice treated with 1,2-di-methylhydrazine: Implications for dietary sphingolipids and colon carcinogenis.
      , 
      • Schmelz E.M.
      • Dombrink-Kurtzman M.A.
      • Roberts P.C.
      • Kozutsumi Y.
      • Kawasaki T.
      • Merill A.H.
      Induction of apoptosis by fumosinin B1 in HT-29 cells is mediated by the accumulation of endogenous free sphingoid bases.
      ), as well as their protective effect against bacterial toxins and infection (
      • Rueda R.
      • Sabatel J.L.
      • Maldonado J.
      • Molina-Font J.S.
      • Gil A.
      Addition of gangliosides to an adapted milk formula modifies levels of fecal Escherichia coli in preterm newborn infants.
      ;
      • Sprong R.C.
      • Hulstein M.F.E.
      • van der Meer R.
      Bovine milk fat components inhibit food-borne pathogens.
      ).
      The overall production of liquid buttermilk is 4.1 million tons worldwide and 0.6 million tons in the United States, based on the annual production of butter from 40% cream (
      International Dairy Federation
      World Dairy Situation.
      ). In the United States, the commercial use of buttermilk is mainly for the baking industry (39%), prepared dry mixes (33%), and for the dairy industry (23%) (
      International Dairy Foods Association
      Milk facts.
      ). In the baking industry, buttermilk is used to improve flavor and texture of bakery products (
      • Vetter J. L
      Dairy products for the cereal processing industry.
      ). Other industrial uses of buttermilk are to prepare functional mixes for various foods, such as sauces, chips, and chocolate products (
      • Chandan R.
      Dairy-based ingredients.
      ). In the dairy industry, buttermilk is used in cheese making (
      • Joshi N.S.
      • Thakar P.N.
      • Jana A.H.
      Utilization of buttermilk in cheese making: A review.
      ), in formulation of ice cream (
      • Chandan R.
      Dairy-based ingredients.
      ) or yogurt (
      • Trachoo N.
      • Mistry V.V.
      Application of ultrafiltered sweet buttermilk and sweet buttermilk powder in the manufacture of nonfat and low fat yogurts.
      ), or in the manufacture of recombined milks (
      • Singh H.
      • Tokley R.P.
      Effects of preheat treatments and buttermilk addition on the seasonal variations in the heat stability of recombined evaporated milk and reconstituted concentrated milk.
      ). A unique functionality reported in buttermilk is its ability to increase the heat stability of recombined milks (
      • Singh H.
      • Tokley R.P.
      Effects of preheat treatments and buttermilk addition on the seasonal variations in the heat stability of recombined evaporated milk and reconstituted concentrated milk.
      ), mainly due to phospholipid-protein interactions preventing protein coagulation during sterilization (
      • McCrae C.H.
      Heat stability of milk emulsions: Phospholipid-protein interactions.
      ). It has also been demonstrated that addition of buttermilk in the manufacture of low-fat Cheddar can improve the texture of the cheese because of the high water-holding capacity of phospholipids (
      • Raval D.M.
      • Mistry V.V.
      Application of ultrafiltered sweet buttermilk in the manufacture of reduced fat process cheese.
      ;
      • Turcot S.
      • Turgeon S.L.
      • St-Gelais D.
      Effect of buttermilk phospholipid concentrations in cheese milk on production and composition of low fat Cheddar cheese.
      ). For all the applications, an important functional property of buttermilk is its emulsifying property (
      • Elling J.L.
      • Duncan S.E.
      • Keenan T.W.
      • Eigel W.N.
      • Boling J.
      Composition and microscopy of reformulated creams from reduced-cholesterol butteroil.
      ;
      • Corredig M.
      • Dalgleish D.G.
      Buttermilk properties in emulsions with soybean oil as affected by fat globule.
      ;
      • Wong P.Y.Y.
      • Kitts D.D.
      A comparison of the butter milk solids functional properties to nonfat dried milk, soy protein isolate, dried egg white, and egg yolk powders.
      ). On the other hand, foaming capacity is significantly lower for buttermilk than for skim milk (
      • Wong P.Y.Y.
      • Kitts D.D.
      A comparison of the butter milk solids functional properties to nonfat dried milk, soy protein isolate, dried egg white, and egg yolk powders.
      ), probably because of the antifoaming properties of the phospholipids when combined with proteins (
      • Coke M.
      • Wilde P.J.
      • Russel E.J.
      • Clarck D.C.
      The influence of surface composition and molecular diffusion on the stability of foams formed from protein/surfactants mixtures.
      ;
      • Vaghela M.N.
      • Kilara A.
      Foaming and emulsifying properties of whey protein concentrates as affected by lipid composition.
      ).
      Studies have been undertaken to assess and enhance the functional properties of buttermilk. Heat treatment applied to cream in butter manufacture has been shown to be detrimental to the functionality of buttermilk, because of denaturation of whey proteins and their interaction with MFGM (
      • Corredig M.
      • Dalgleish D.G.
      Isolates from industrial buttermilk: emulsifying properties of materials derived from the milk fat globule membrane.
      ,
      • Corredig M.
      • Dalgleish D.G.
      Effect of heating of cream on the properties of milk fat globule membrane isolates.
      ). Several physicochemical treatments have been applied to increase the concentration of phospholipids in buttermilk, such as microfiltration (
      • Sachdeva S.
      • Buchheim W.
      Recovery of phospholipids from buttermilk using membrane processing.
      ;
      • Astaire J.C.
      • Ward R.
      • German J.B.
      • Jimenez-Flores R.
      Concentration of polar MFGM lipids from buttermilk by microfiltration and supercritical fluid extraction.
      ;
      • Corredig M.
      • Roesch R.R.
      • Dalgleish D.G.
      Production of a novel ingredient from buttermilk.
      ;
      • Morin P.
      • Jiménez-Flores R.
      • Pouliot Y.
      Effect of temperature and pore size on the fractionation of fresh and reconstituted buttermilk by microfiltration.
      ), ultrafiltration (
      • Sachdeva S.
      • Buchheim W.
      Recovery of phospholipids from buttermilk using membrane processing.
      ), super-critical fluid extraction (
      • Astaire J.C.
      • Ward R.
      • German J.B.
      • Jimenez-Flores R.
      Concentration of polar MFGM lipids from buttermilk by microfiltration and supercritical fluid extraction.
      ), eventually preceded by a preliminary treatment to remove the casein fraction (
      • Sachdeva S.
      • Buchheim W.
      Recovery of phospholipids from buttermilk using membrane processing.
      ;
      • Corredig M.
      • Roesch R.R.
      • Dalgleish D.G.
      Production of a novel ingredient from buttermilk.
      ).
      Most of the work on the functionality of buttermilk as a food ingredient has been done with sweet buttermilk, which is the major source of commercial buttermilk. No study has yet been developed to diversify the sources of buttermilk and to evaluate the effect of this diversification on its functionality. However, other types of buttermilk can be produced from milk fat, as cultured buttermilk obtained from churning of cultured cream, in the manufacture of European-style butter; or whey buttermilk, from churning of whey cream, in manufacture of whey butter. These ingredients are produced in much smaller quantities compared with the volume of sweet buttermilk; however, their potential market should not been neglected. The potential market for whey buttermilk is significant based on the large volume of whey produced each year. Considering an average fat content of 0.13% in whey, 57,000 tons of liquid whey buttermilk could be produced each year in the United States, which would constitute 10% of the current volume of sweet buttermilk (
      National Agricultural Statistics Service
      Dairy products, 2003 Summary.
      ).
      The main objectives of this work were to assess the compositional and functional properties (solubility, viscosity, emulsifying and foaming properties) of sweet, sour, and whey buttermilk, and to evaluate the potential of nonconventional buttermilks (cultured and whey buttermilks) as food ingredients. In addition, the functional properties of the different buttermilks were compared with those of milk and whey.

      Materials and Methods

       Samples

      Seven different dairy ingredients were studied, including 1 commercial medium-heat skim milk powder (SMP), 1 commercial whey powder (WP), and 5 buttermilk powders (BM). Two buttermilk powders (BM1 and BM2) were commercial sweet buttermilks kindly provided by 2 Californian dairy companies; the 2 other buttermilk powders (BM3, BM4, and BM5) were manufactured on site at the plant of the Dairy Products Technology Center (DPTC), California Polytechnic State University (San Luis Obispo) from sweet cream, cultured cream, and whey cream, respectively. For each trial, 100 to 200 L of cream were churned. Sweet cream and whey cream were provided by Foster Farms (Modesto, CA) and Hilmar Cheese Company (Hilmar, CA), respectively. They were churned after a waiting period of 16 h at 4°C. Cultured cream was produced at DPTC from sweet cream (Foster cream, Foster Farms). The cream was preheated at 20°C, inoculated with 0.07% (wt/wt) commercial mesophilic culture Flora Danica (Chr. Hansen, Milwaukee, WI), then incubated at the same temperature for 16 h, and finally churned. The final pH of cultured cream was 4.90 ± 0.05. The 3 creams (sweet, sour, and whey cream) were churned to butter using a continuous pilot-scale butter churn (Egli AG, Bütschwil, Switzerland). Buttermilk was recovered in a milk can after butter fines were removed by filtration through cheesecloth, then spray dried using a Niro Filterlab Spray Drier (Hudson, WI). The buttermilk production on site at DPTC was repeated twice, with 2 different lots of the 3 creams. For commercial samples, 2 different lots of powder were used.

       Compositional Analyses

       Nitrogen and Protein Determination

      The levels of total nitrogen (TN), nitrogen soluble at pH 4.6 (SN), and NPN were determined via the Kjeldahl method (
      Association of Official Analytical Chemists
      Official Methods of Analysis.
      ). All measurements were carried out in duplicate. The total protein was calculated as TN × 6.38. The NPN, expressed in protein equivalent, was calculated as NPN × 6.38. The soluble protein at pH 4.6 was calculated as (SN − NPN) × 6.38. The insoluble protein at pH 4.6 was calculated as (TN − SN) × 6.38. The protein profile was established by SDS-PAGE using precast gels with 4 to 20% gradient (Gradipore, Frenchs Forest, Australia). All samples were diluted to 3.3 mg/mL of total protein with deionized water. One volume of sample was added to 1 vol.me of reducing buffer containing 59 mM Tris-HCl pH 6.8, 24% glycerol, 2% SDS, 5% 2-β-mercaptoethanol, and 0.01% bromophenol blue. A sample of 10 μL of the mixture was loaded after boiling. Gels were stained with Coomassie Blue R-250 (BioRad, Hercules, CA). Proteins were identified according their molecular weight by comparison to protein standard (BioRad).

       Determination of Moisture, Fat Content, Phospholipids, Ash, and Lactose

      Moisture was determined by drying each sample for 5 h in a vacuum oven at 100°C (
      American Dairy Products Institute
      Standards for grades of dry milks.
      ). Fat content was determined by the Mojonnier ether extraction method as described by
      • Marshall R.T
      Standards methods for the examination of dairy products.
      . Extracted lipids were then diluted to 10 mg of lipids per mL with 1:2 chloroform methanol and kept in a freezer (−20°C) until analysis. Phospholipids were analyzed by HPLC with an electro-evaporative light scattering detector as described in
      • Morin P.
      • Jiménez-Flores R.
      • Pouliot Y.
      Effect of temperature and pore size on the fractionation of fresh and reconstituted buttermilk by microfiltration.
      . All reagents were electrophoresis or HPLC grade. Ash content was determined by ignition for 16 h at 550°C in an electric muffle furnace (
      Association of Official Analytical Chemists
      Official Methods of Analysis.
      ). All measurements were carried out in triplicate. Content of lactose + lactic acid was calculated by difference [total solid − (total protein + fat + ash)] as proposed by
      • Guzman-Gonzalez M.
      • Morais F.
      • Ramos M.
      • Amigo L.
      Influence of skimmed milk concentrate replacement by dry dairy products in a low fat set-type yoghurt model system. I: Use of whey protein concentrates, milk protein concentrates and skimmed milk powder.
      .

       Determination of pH

      The pH of 4% protein (wt/wt) reconstituted powder was determined using a pH meter Φ34 (Beckman, Fullerton, CA). Measurements were done in quadruplicate.

       Functional Properties

      The functional properties were determined at 4 pH levels (initial normal pH, pH 6, pH 5, and pH 4), and at a defined and normalized protein concentration (2 to 5% wt/wt), depending on the property tested, as recommended by
      • Hall G. M
      Methods of testing protein functionality.
      .

       Preparation of Samples

      The dairy ingredients were dissolved in deionized water at room temperature for 1 h. The pH of the solution was eventually adjusted to pH 6, pH 5, or pH 4, using 1 M and 0.1 M HCl or NaOH and waiting one additional hour for pH stability.

       Solubility

      Protein solubility was determined as described by
      • Wong P.Y.Y.
      • Kitts D.D.
      A comparison of the butter milk solids functional properties to nonfat dried milk, soy protein isolate, dried egg white, and egg yolk powders.
      . Protein solutions (5% TN, wt/wt) at different pH values were centrifuged at 12,000 × g for 15 min at 25°C. Supernatant liquid was analyzed for TN by the Kjeldahl method. Measures were done in triplicate.

       Viscosity

      The viscosity of 4% (wt/wt) protein solutions at 4 pH levels was determined at 20°C with a control-stress rheometer (model SR5000; Rheometric Scientific Inc., Piscataway, NJ). The rheometer was equipped with a concentric cylinder device consisting of a cup (32-mm diameter) and a bob (29.5-mm diameter, 44.25-mm length). About 17 mL of protein solution was transferred into the cup of the rheometer and the bob was lowered until the whole bob surface was covered. Five minutes were allotted for the sample temperature to equilibrate to 20°C before each analysis. Steady stress sweep in the range 0.1 to 1 Pa was applied to the sample to obtain a flow curve in the shear rate range 0.1 to 100 s−1. Apparent viscosity (in Pa·s) was determined at 50 s−1. Three replicates were performed and fresh sample was used for each replicate.

       Emulsifying Properties

      Emulsions were prepared with 40 mL of a 2% (wt/wt) protein solution combined with 10 mL of corn oil (Mazola, Bestfoods, Englewood Cliffs, NJ) to form a 20% (vol/vol) emulsion at 4 pH values. The emulsion was created using an Ultra-Turrax mixer (model T18, IKA Works, Wilmington, NC) for 2 min at 22,000 rpm at room temperature (
      • Flanagan J.
      • Fitzgerald R.J.
      Functionality of Bacillus proteinase hydrolysates of sodium caseinate.
      ;
      • Raymundo A.
      • Franco J.M.
      • Empis J.
      • Sousa I.
      Optimization of the composition of low-fat oil-in-water emulsions stabilized by white lupin protein.
      ). Then, volume particle diameter size distributions were determined using a laser diffraction particle size analyzer (model LS230, Beckman Coulter, Miami, FL). The arithmetic mean of the diameter size was calculated for each distribution. The optical parameters (adsorption coefficient, refractive index) of the corn oil droplets for the measurements were determined by spectrophotometry (Spectra Max Plus spectrophotometer, Molecular Devices Corporation, Sunnyvale, CA), and refractometry (Abbe refractometer, Fisher Scientific, Pittsburgh, PA) as recommended by
      • Michalski M.C.
      • Briard V.
      • Michel F.
      Optical parameters of milk fat globules for laser light scattering measurements.
      . The adsorption coefficient, Ka, was calculated as 2.6 × 10−7. The refractive index measured was 1.473. These values were used for calculation based on the Mie theory for the laser light-scattering techniques. Each measurement was performed in triplicate.

       Foaming Properties

      Fifty milliliters (VS) of a 4% (wt/wt) protein solution at 4 pH values was blended for 3 min using a Ultra-Turrax mixer (IKA Works) at 18,000 rpm and poured into a graduated cylinder. The initial volume of foam (VF) was recorded and the foam was then left undisturbed for 10 min. The volume of liquid drained beneath the foam (VD) was measured. Foam capacity (FC), expressed in milliliters of foam generated divided by milliliter of solution (VI), and foam stability (FS), expressed in milliliters of liquid remaining foamed after 10 min divided by milliliter of solution, were calculated using the following equations:
      FC=VFVIandFS=(VIVDVI


      Each measurement was performed in triplicate.

       Statistical Analyses

      Results were evaluated statistically using Minitab 13.1 Software (Minitab Inc., State College, PA). A 2-factor ANOVA with interaction was performed to determine the effects of both sample and pH on functional properties of dairy ingredient. Multiple comparison of means was performed using Tukey's pairwise comparison at an α-level of 5%.

      Results and Discussion

       Composition

       Gross Composition

      Table 1 shows the gross composition on a DM basis and the pH of the different powders.
      Table 1Gross composition (%) on a DM basis, and natural pH of the samples in this work
      Samples: One commercial skim milk powder, one commercial whey powder, and 5 buttermilk (BM) powders. Three buttermilks were obtained from sweet cream (BM1, BM2, and BM3), one buttermilk sample was obtained from cultured cream (BM4), and one was obtained from whey cream (BM5). Two buttermilks were commercial (BM1 and BM2), and 3 were manufactured at a pilot-scale (BM3, BM4, and BM5).
      SamplesTotal NFatsPhospholipidsAshLactosepH
      pH was determined in a 4% protein solution.
      Skim milk powder35.9
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      0.3
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      ND
      ND = Not determined.
      8.0
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      55.8
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      6.39
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      Whey powder13.3
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      0.4
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      ND7.5
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      78.8
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      5.58
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      Buttermilk powders
       BM132.9
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      5.7
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      1.29
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      7.6
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      53.8
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      6.46
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
       BM233.1
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      7.2
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      1.34
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      7.3
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      52.4
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      6.59
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
       BM331.5
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      13.1
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      1.27
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      6.7
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      48.7
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      6.61
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
       BM427.8
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      22.3
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      1.15
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      6.2
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      43.7
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      5.39
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
       BM514.1
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      15.5
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      1.87
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      7.0
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      63.4
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      5.98
      –eMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      a –eMeans within a column not sharing a common superscript differ significantly (P ≤ 0.05).
      1 Samples: One commercial skim milk powder, one commercial whey powder, and 5 buttermilk (BM) powders. Three buttermilks were obtained from sweet cream (BM1, BM2, and BM3), one buttermilk sample was obtained from cultured cream (BM4), and one was obtained from whey cream (BM5). Two buttermilks were commercial (BM1 and BM2), and 3 were manufactured at a pilot-scale (BM3, BM4, and BM5).
      2 pH was determined in a 4% protein solution.
      3 ND = Not determined.
      The protein content was slightly lower for buttermilk powders obtained from sweet cream (BM1, BM2, BM3) and cultured cream (BM4), than for skim milk powder (27.8 to 33.1% compared with 35.9%). This has been reported in other studies (
      • Surel O.
      • Famelart M.H.
      Ability of ceramic membranes to reject lipids of dairy products.
      ;
      • Elling J.L.
      • Duncan S.E.
      • Keenan T.W.
      • Eigel W.N.
      • Boling J.
      Composition and microscopy of reformulated creams from reduced-cholesterol butteroil.
      ;
      • O’Connell J.E.
      • Fox P.F.
      Heat stability of buttermilk.
      ;
      • Turcot S.
      • Turgeon S.L.
      • St-Gelais D.
      Effect of buttermilk phospholipid concentrations in cheese milk on production and composition of low fat Cheddar cheese.
      ;
      • Scott L.L.
      • Duncan S.E.
      • Sumner S.S.
      • Waterman K.M.
      • Kaylegian K.E.
      Influence of emulsifying component composition on creams formulated with fractionated milkfat.
      ). A small part of the protein fraction of the sweet cream remains in the butter after churning. On the other hand, the buttermilk obtained from whey cream (BM5) contained a much smaller amount of protein, 14%, so less than half the protein content of the buttermilk was obtained from sweet cream. This was due to the lack of casein in the whey cream as compared with cream, yielding values of protein of 0.89 and 1.94%, respectively (

      Morin, P., Y. Pouliot and R. Jiménez-Flores. A comparative study of the fractionation of regular buttermilk and whey buttermilk by microfiltration. J. Food Eng. doi:10.1016/j.jfoodeng.2005.06.065

      ). It should be noted that the level of protein in whey buttermilk was very similar to that observed in whey (13%), which is consistent with the expectations of whey buttermilk resembling whey composition as opposed to regular buttermilk, which resembles skim milk in composition.
      Differences were observed for the fat content among buttermilks. Commercial buttermilks BM1 and BM2 contained 6 to 7% fat, whereas the buttermilks produced at DPTC had higher lipid content (13 to 22% fat). The same observation was reported by
      • Elling J.L.
      • Duncan S.E.
      • Keenan T.W.
      • Eigel W.N.
      • Boling J.
      Composition and microscopy of reformulated creams from reduced-cholesterol butteroil.
      . This was due to the lack of fat removal process during the manufacture of buttermilk on site, which did not include a centrifugation step to remove the excess lipid, as used for commercial buttermilk. The centrifugation of the buttermilk was not performed on site because the quantity of buttermilk produced (50 to 100 L) was too low to allow the use of the pilot-plant centrifuge. On the other hand, skim milk powder and whey powders contained very little fat (<0.5%) compared with commercial buttermilks (>5%). This low fat content has been reported by others (
      • Surel O.
      • Famelart M.H.
      Ability of ceramic membranes to reject lipids of dairy products.
      ;
      • Elling J.L.
      • Duncan S.E.
      • Keenan T.W.
      • Eigel W.N.
      • Boling J.
      Composition and microscopy of reformulated creams from reduced-cholesterol butteroil.
      ;
      • Turcot S.
      • Turgeon S.L.
      • St-Gelais D.
      Effect of buttermilk phospholipid concentrations in cheese milk on production and composition of low fat Cheddar cheese.
      ) and is due to the presence of MFGM fractions, small milk fat globules, and free lipids not extractable by centrifugation (
      • Corredig M.
      • Dalgleish D.G.
      Isolates from industrial buttermilk: emulsifying properties of materials derived from the milk fat globule membrane.
      ).
      Phospholipids content of the buttermilks showed similar composition according to their origin and place of manufacturing, except for whey buttermilk BM5. Although the phospholipids content is significantly higher in BM5, the fact that the ratio of phospholipids to protein is almost 3.5 times higher in this sample is even more striking.
      Ash content was between 6 and 8%, and lactose between 43 and 79%. Differences observed in lactose content were strongly related to the difference in protein content. The ingredients with the low protein levels, WP and whey buttermilk (BM5), were the ones with the highest level of lactose. The pH of the sweet buttermilks was close to the pH of skim milk (pH 6.5 to 6.6 compared with pH 6.4), whereas the pH of cultured buttermilk, whey buttermilk, and WP was lower (pH < 6) because of the acidification occurring during the ripening of the cream (for cultured buttermilk) or during the ripening of milk in cheese making (for whey buttermilk and WP). The pH of the cultured buttermilk was higher than the pH of ripened cream (pH 5.4 as compared with pH 4.9), because the measurement of the pH was done in a 4% protein solution, which had higher buffering capacity compared with a 2% protein concentration as in the case of cream.

       Nitrogen Composition

      Table 2 shows the content of the nitrogen fraction for the different powders. Results are expressed as a percentage of the total protein content. Figure 1 illustrates the protein composition of the various powders as obtained by SDS-PAGE.
      Table 2Content of the nitrogen fraction of a commercial skim milk powder, a commercial whey powder, and 5 buttermilk powders
      Three buttermilks were obtained from sweet cream (BM1, BM2, and BM3), one was obtained from cultured cream (BM4), and one was obtained from whey cream (BM5). Two buttermilks were commercial (BM1 and BM2), and 3 were manufactured on a pilot-scale (BM3, BM4, and BM5).
      ,
      Nonprotein nitrogen, soluble protein at pH 4.6, and insoluble protein at pH 4.6 are expressed in % (wt/wt) of total nitrogen.
      SamplesNPNpH 4.6 soluble proteinpH 4.6 insoluble protein
      Skim milk powder6.2
      –dMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      6.8
      –dMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      87.0
      –dMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      Whey powder28.3
      –dMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      38.7
      –dMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      33.0
      –dMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      Buttermilk powders
       BM17.6
      –dMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      6.9
      –dMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      85.5
      –dMeans within a column not sharing a common superscript differ significantly (P≤0.05).
       BM26.3
      –dMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      9.3
      –dMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      84.4
      –dMeans within a column not sharing a common superscript differ significantly (P≤0.05).
       BM36.7
      –dMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      14.9
      –dMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      78.4
      –dMeans within a column not sharing a common superscript differ significantly (P≤0.05).
       BM48.8
      –dMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      13.9
      –dMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      77.2
      –dMeans within a column not sharing a common superscript differ significantly (P≤0.05).
       BM527.1
      –dMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      32.9
      –dMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      40.0
      –dMeans within a column not sharing a common superscript differ significantly (P≤0.05).
      a –dMeans within a column not sharing a common superscript differ significantly (P ≤ 0.05).
      1 Three buttermilks were obtained from sweet cream (BM1, BM2, and BM3), one was obtained from cultured cream (BM4), and one was obtained from whey cream (BM5). Two buttermilks were commercial (BM1 and BM2), and 3 were manufactured on a pilot-scale (BM3, BM4, and BM5).
      2 Nonprotein nitrogen, soluble protein at pH 4.6, and insoluble protein at pH 4.6 are expressed in % (wt/wt) of total nitrogen.
      Figure thumbnail gr1
      Figure 1Sodium dodecyl sulfate-PAGE of the various dairy powders. Lanes: S = molecular weight standard; SMP = commercial skim milk powder; WP = commercial whey powder; BM1 and BM2 = commercial sweet buttermilk powders; BM3 = pilot-scale sweet buttermilk powder; BM4 = pilot-scale cultured buttermilk powder; and BM5 = pilot-scale whey buttermilk.
      The NPN fraction (Table 2) represented 6 to 28% of the total protein content of the powders. It was low in skim milk powder and sweet buttermilk powder (6 to 7%), and higher in cultured buttermilk (9%) and whey buttermilk (27 to 28%), because of the proteolytic activity of the starter used for cream ripening or cheese making, respectively. The percentage of NPN was especially high in whey buttermilk because of the low TN content (only 14% on a DM basis compared with 28% for the cultured buttermilk).
      The protein profile as obtained by SDS-PAGE (Figure 1) reveals the lack of caseins in WP and whey buttermilk powder, as expected. The band of whey proteins appears a little fuzzy and with a higher molecular weight in whey powder compared with other powders. This can be attributed to lactosylation of the whey proteins, which occurs during heat treatment of liquid whey and storage of the whey powder and increases the molecular weight of the whey protein (
      • Leonil J.
      • Molle D.
      • Fauquant J.
      • Maubois J.L.
      • Pearce R.J.
      • Bouhallab S.
      Characterization by ionization mass spectrometry of lactosyl β-lactoglobulin conjugates formed during heat treatment of milk and whey and identification of one lactose-binding site.
      ). Milk fat globule membrane proteins were observed in the sweet, cultured, and whey buttermilks, whereas they were not detected in whey or milk, indicating the presence of MFGM fragment in buttermilks, as reported in previous studies (
      • Corredig M.
      • Dalgleish D.G.
      Buttermilk properties in emulsions with soybean oil as affected by fat globule.
      ;
      • O’Connell J.E.
      • Fox P.F.
      Heat stability of buttermilk.
      ;
      • Scott L.L.
      • Duncan S.E.
      • Sumner S.S.
      • Waterman K.M.
      • Kaylegian K.E.
      Influence of emulsifying component composition on creams formulated with fractionated milkfat.
      ).
      The protein fraction soluble at pH 4.6 (Table 2) included native whey protein, and some of the MFGM proteins, as the glycoprotein B (also called PAS 6/7;
      • O’Connell J.E.
      • Fox P.F.
      Heat stability of buttermilk.
      ), which does not precipitate at acidic pH. This fraction was high in whey powder and whey buttermilk (39 and 33%, respectively) because whey proteins were the major part of the protein in these powders. It constituted a smaller part in skim milk and in sweet and cultured buttermilks, where the major part of the protein was casein, which precipitates at pH 4.6. However, some differences were noticeable among the 4 buttermilks obtained from milk cream. The 2 buttermilks BM3 and BM4 produced on a pilot scale contained more soluble protein (around 14%) at pH 4.6 than did the industrial buttermilks, BM1 and BM2 (8 to 9%). This difference is probably due to the nature of the heat treatment applied to buttermilk before drying. In industry, buttermilk is exposed to severe heat treatment before drying to initiate Maillard reactions and generate browning and flavoring compounds to enhance the sensory properties (
      • Walstra P.
      • Geurts T.J.
      • Noomen A.
      • Jellema A.
      • van Boekel M.A.J.S.
      Dairy Technology: Principles of milk properties and processes.
      ). This severe heat treatment could partly denature the whey proteins. When they are denatured, the whey proteins are insoluble at pH 4.6. For buttermilk manufactured at the DPTC pilot plant, no heat treatment was applied before drying, so no significant denaturation of the whey protein occurred.
      The insoluble protein at pH 4.6 (Table 2) represents the casein, some of the MFGM proteins, and the denatured whey protein, which precipitate at an acidic pH. This fraction was high (>77%) for skim milk powder and buttermilks from sweet or cultured cream, BM1 to BM4, which contained casein. Surprisingly, although this fraction was only half the amount in the regular and cultured buttermilks, the total insoluble protein was not negligible for whey powder and whey buttermilk (33 to 40%), which did not contain casein according to results of SDS-PAGE (see Figure 1). One possible explanation is that part of the whey protein of whey and whey buttermilk was denatured during processing. Manufacture of sweet whey powder can involve various processes, including heat treatments (multiple pasteurizations, hot well holding, evaporation, spray drier chamber;
      • Banavara D.S.
      • Anupama D.
      • Rankin S.A.
      Studies on physicochemical and functional properties of commercial sweet whey powders.
      ) that denature part of the whey protein. This seems to be the case with our commercial sweet whey powder, because already there is some apparent lactosylation of the whey proteins (see Figure 1), which would prove that the whey has been subjected to high heat treatments. Considering the commercial processing conditions, the whey cream is subjected to a high heat treatment (82°C for 45 s) after centrifugation, which may have partly denatured the whey proteins. In addition, whey buttermilk contained a non-negligible fraction of MFGM proteins (see Figure 1), which are large proteins insoluble at acidic pH (
      • O’Connell J.E.
      • Fox P.F.
      Heat stability of buttermilk.
      ).

       Functional Properties

       Protein Solubility

      Protein solubility is an important functionality for protein powders, which governs many other functional properties (
      • Kinsella J.E.
      Functional properties of proteins in foods: A survey.
      ). Protein solubility of the different powders at various pH levels is reported in Figure 2.
      Figure thumbnail gr2
      Figure 2Protein solubility determined in a 5% protein solution of various dairy powders: a commercial skim milk powder (SMP), a commercial whey powder (WP), and 5 buttermilk powders (BM). Three buttermilks were obtained from sweet cream (BM1, BM2, and BM3), one was obtained from cultured cream (BM4), and one was obtained from whey cream (BM5). Two buttermilks were of commercial origin (BM1 and BM2), and 3 were manufactured on a pilot scale (BM3, BM4, and BM5). Measurements were performed at 4 pH values: initial pH (black bar), pH 6 (white bar), pH 5 (diagonal hatch), and pH 4 (shaded). Error bars represent the standard deviations.
      Protein solubility varied from 12 to 92% according to the type of dairy powder and the pH of the solution. The effect on protein solubility of the powder and pH were both highly significant, as well as the interaction (P < 0.001). The protein solubility of the powder was strongly dependent on pH, with lower solubility when pH was lower than 5. However, the effect of pH was much less important for whey powder and whey buttermilk powder. For instance, protein solubility at initial pH and pH 4 was equal to 92 and 80%, respectively, for whey powder, and to 77 and 73%, respectively, for whey buttermilk, whereas it decreased from 60 to 79% to 12 to 23%, respectively, for the other buttermilks and for the skim milk powder. These differences in protein solubility as a function of pH are due to differences in casein and whey protein contents. Caseins have been shown to be highly insoluble at acidic pH (where they precipitate), whereas whey protein remains mainly soluble at low pH (
      • Chobert J.M.
      • Bertrand-Harb C.
      • Nicolas M.G.
      Solubility and emulsifying properties of caseins and whey proteins modified enzymatically by trypsin.
      ). Whey powder and whey buttermilks contained no casein, whereas casein represented 77 to 87% of total protein in sweet and cultured buttermilks as well as skim milk powder (see Table 2).

       Viscosity

      The flow behavior of the different dairy powders in 4% protein solution has been studied; the viscosity results determined at 50 s−1 at 4 pH levels are illustrated in Figure 3.
      Figure thumbnail gr3
      Figure 3Apparent viscosity determined at a shear rate of 50 s−1 in a 4% protein solution of various dairy powders: a commercial skim milk powder (SMP), a commercial whey powder (WP), and 5 buttermilk powders (BM). Three buttermilks were obtained from sweet cream (BM1, BM2, and BM3), one was obtained from cultured cream (BM4), and one was obtained from whey cream (BM5). Two buttermilks were of commercial origin (BM1 and BM2), and 3 were manufactured on a pilot scale (BM3, BM4, and BM5). Measurements were performed at 4 pH values: initial pH (black bar), pH 6 (white bar), pH 5 (diagonal hatch), and pH 4 (shaded). Error bars represent the standard deviations.
      The effects of the dairy powder and pH as well as their interaction were highly significant (P < 0.001). The protein solutions containing casein (solution of skim milk powder and sweet and cultured buttermilks) exhibited a drastic increase in viscosity (3 to 4 times increase) at pH 4 compared with initial pH. This change in viscosity was due to the precipitation of the pH 4.6 insoluble protein fraction at low pH, which represented about 80% of the total protein fraction. The precipitation of this fraction created a thickening of the solution. On the other hand, the viscosity of the protein solution of whey powder and whey buttermilk, devoid of casein, remained remarkably stable in the pH range studied. This may be due to the lower content of the insoluble protein fraction in these powders at low pH, where it is only half of the amount of the buttermilk powders (e.g., BM5 has 40% insoluble protein compared with BM1, which has 85.5%). When compared at initial pH, the viscosity range of the 5 buttermilks was in the order: BM1 = BM2 < BM3 < BM4 = BM5. These differences are due to various factors; 1) the fat content, higher for the buttermilks produced on site (BM3, BM4, BM5) than commercial buttermilks (BM1 and BM2); 2) the DM content, higher for the 4% protein solution of whey buttermilk BM5 than for other buttermilks, respectively 35% compared with 15 to 18%, because the protein content of whey buttermilk BM5 was the lowest (see Table 1); and 3) the presence of cells and microbial exopolysaccharide in the solution of buttermilk BM4, ripened with a culture of Flora Danica.

       Emulsifying Properties

      The volume particle diameter size distribution in 20% oil in water emulsion has been determined with 2% protein solution at various pH levels for the 7 dairy powders. The arithmetic mean for the size of the fat globules is reported in Figure 4.
      Figure thumbnail gr4
      Figure 4Average particle size of 20% (vol/vol) oil in water emulsions stabilized by 2% protein solution of various dairy powders: a commercial skim milk powder (SMP), a commercial whey powder (WP), and 5 buttermilk powders (BM). Three buttermilks were obtained from sweet cream (BM1, BM2, and BM3), one was obtained from cultured cream (BM4), and one was obtained from whey cream (BM5). Two buttermilks were of commercial origin (BM1 and BM2), and 3 were manufactured on a pilot scale (BM3, BM4, and BM5). Measurements were performed at 4 pH values: initial pH (black bar), pH 6 (white bar), pH 5 (diagonal hatch), and pH 4 (shaded). Error bars represent the standard deviations.
      There was a significant effect (P < 0.001) of the type of dairy powder and the pH of the solution, as well as the interaction, on the size of the globules. For skim milk powder and sweet or cultured buttermilk powders (BM1, BM2, BM3, BM4), the pH had a strong effect on the average globule size, which is 50 to 300% higher when the pH is lower than 5. On the other hand, there was no effect of pH on globule size when protein solution is obtained from whey powder or whey buttermilk. This lack of effect of pH on the emulsifying capacity of the whey protein has been reported in the work of
      • Chobert J.M.
      • Bertrand-Harb C.
      • Nicolas M.G.
      Solubility and emulsifying properties of caseins and whey proteins modified enzymatically by trypsin.
      . It can be attributed again to the low level of insoluble protein at pH 4.6 in whey protein powder. In our case, the insoluble protein fraction represented less than 40% for whey powder and whey buttermilk, and represented about 80% for skim milk and sweet and cultured buttermilks.
      Furthermore, the size of the fat droplet was significantly different (P < 0.05) between SMP and sweet or cultured buttermilk (BM1, BM2, BM3, BM4), and between WP and whey buttermilk (BM5). Figure 5 illustrates the distribution of the fat globules at pH 6 for skim milk, whey, buttermilk (sweet and sour), and whey buttermilk. The globule size was 7 to 8 μm with sweet and cultured buttermilks compared with 10 μm with skim milk, and 5.5 μm for whey buttermilk compared with 9.7 μm with whey. One of the major differences in composition for sweet and cultured buttermilk vs. skim milk, or for whey buttermilk vs. whey, was the presence of MFGM fractions, as evidenced by the presence of MFGM proteins in the SDS-PAGE of the powders (See Figure 1). Milk fat globule membrane fractions have been shown to have a strong emulsifying capacity (
      • Kanno C.
      Emulsifying properties of bovine milk fat globule membrane in milk fat emulsion: Condition for the reconstitution of fat globules.
      ). This can probably explain the better emulsifying capacity of whey buttermilk compared with whey, and sweet or cultured buttermilk compared with skim milk. Furthermore, at pH > 5, the size of the fat globules was significantly different (P < 0.05) between the various buttermilks. The range was BM5 < BM4 < BM1, BM2, and BM3. The whey buttermilk BM5 had lower protein content compared with sweet (BM1, BM2, BM3) or cultured buttermilk (BM4) (14 vs. ∼30%; Table 1). However, the phospholipids content was the same in sweet and whey buttermilks, about 1.2% (Table 1). Consequently, the proportion of phospholipids to protein content was higher for whey buttermilk compared with sweet buttermilk, around 13 and 4%, respectively. In protein solutions used to prepare the emulsions, the phospholipid concentration was higher for whey buttermilk solution than for sweet buttermilk, which could enhance its emulsion capacity. These results are similar to those obtained by
      • Roesch R.R.
      • Rincon A.
      • Corredig M.
      Emulsifying properties of fractions prepared from commercial buttermilk by microfiltration.
      with buttermilk concentrates and MFGM isolates. Those authors investigated the emulsifying properties of fractions prepared from commercial buttermilks by microfiltration with or without citrate. When microfiltration was carried out with citrate, casein was dissociated and retentates contained a high amount of MFGM fractions. The resulting MFGM isolates are devoid of casein, but rich in β-lactoglobulin because of the heat induced protein-protein interactions occurring during the processing of butter. In these fractions, the original ratio of protein in buttermilk (casein, whey protein, and MFGM proteins) was modified. On the other hand, when microfiltration was operated without citrate, this ratio was maintained and the retentate constituted buttermilk concentrate. The study demonstrated better emulsion properties for MFGM isolates than for buttermilk concentrates, and this result was attributed to the higher ratio of MFGM fraction to protein in MFGM isolates as compared with buttermilk concentrates. In our study, the same tendency was observed in a different system, composed of whey buttermilk and regular buttermilk, the former devoid of casein and with a higher ratio of MFGM fraction to protein compared with the latter. Finally, the bacteria and fermentation by-products contained in cultured buttermilk may play a positive role in emulsification and explain the smaller size of the fat globules in emulsion with cultured buttermilk BM4, compared with the one with sweet buttermilk BM1, BM2, and BM3.
      Figure thumbnail gr5
      Figure 5Particle size distribution of 20% (vol/vol) oil in water emulsions stabilized by a 2% protein solution at pH 6 of various dairy powders: a commercial skim milk powder (milk), a commercial whey powder (whey), and 3 buttermilk powders. Buttermilks were respectively obtained from sweet cream (sweet buttermilk), cultured cream (sour buttermilk), and whey cream (whey buttermilk), and manufactured at a pilot scale level.
      Thus, for emulsifying properties, the size of the fat globule in oil in water emulsion was smaller when buttermilk was used (sour, sweet, or whey buttermilk) compared with skim milk or whey. Furthermore, because of the lack of sensitivity of the whey protein to pH, there was no instability of the emulsion at acidic pH when whey buttermilk was used, whereas sweet or cultured buttermilk lost their emulsifying capacity at pH 5 or pH 4. Consequently, whey buttermilk offers an interesting alternative to create an emulsion for the formulation of low pH foods.

       Foaming Properties

      The foaming capacity of the various dairy powders has been determined in 4% protein solution and is reported in Figure 6.
      Figure thumbnail gr6
      Figure 6Foaming capacity determined in a 2% protein solution of various dairy powders: a commercial skim milk powder (SMP), a commercial whey powder (WP), and 5 buttermilk powders (BM). Three buttermilks were obtained from sweet cream (BM1, BM2, and BM3), one was obtained from cultured cream (BM4), and one was obtained from whey cream (BM5). Two buttermilks were of commercial origin (BM1 and BM2), and 3 were manufactured on a pilot scale (BM3, BM4, and BM5). Measurements were performed at 4 pH values: initial pH (black bar), pH 6 (white bar), pH 5 (diagonal hatch), and pH 4 (shaded). Error bars represent the standard deviations.
      The effect of the type of dairy powder, as well as the effect of the pH and the interaction, were highly significant (P < 0.001). The range of foaming capacity was BM5 ≤ BM4 ≤ BM3 < BM1 = BM2 < SMP < WP. The data were similar for the foaming stability and are not reported in this manuscript. Buttermilks in general had much lower foaming capacity than skim milk or whey powder, with 1.5 to 2.5 mL of foam per mL of liquid, compared with 2 to 4 mL/mL for skim milk and whey, as reported in the study of
      • Wong P.Y.Y.
      • Kitts D.D.
      A comparison of the butter milk solids functional properties to nonfat dried milk, soy protein isolate, dried egg white, and egg yolk powders.
      . This finding can be attributed to the increase of phospholipid concentration in buttermilk by a concentration factor of 4 to 7 (
      • Christie W.W.
      • Noble R.C.
      • Davies G.
      Phospholipids in milk and dairy products.
      ;
      • Elling J.L.
      • Duncan S.E.
      • Keenan T.W.
      • Eigel W.N.
      • Boling J.
      Composition and microscopy of reformulated creams from reduced-cholesterol butteroil.
      ), which directly affects the foaming capacity by competitive displacement of the protein from the surface by the surfactant (
      • Coke M.
      • Wilde P.J.
      • Russel E.J.
      • Clarck D.C.
      The influence of surface composition and molecular diffusion on the stability of foams formed from protein/surfactants mixtures.
      ;
      • Vaghela M.N.
      • Kilara A.
      Foaming and emulsifying properties of whey protein concentrates as affected by lipid composition.
      ). Among the various buttermilks, pilot-scale-produced buttermilks (BM3, BM4, and BM5) exhibited lower foaming capacity than did commercial buttermilks (BM1 and BM2). This could be attributed to their higher fat content (13 to 22% vs. 6 to 7%, respectively; Table 1). Fats are known to decrease the foaming ability of the protein solution, because of their amphiphilic nature and their ability to displace the protein from the surface (
      • Vaghela M.N.
      • Kilara A.
      Foaming and emulsifying properties of whey protein concentrates as affected by lipid composition.
      ). Finally, among the pilot-scale-produced buttermilks, the whey buttermilk (BM5) is the one that exhibited the lowest foaming capacity. It could be due to its higher phospholipids to protein ratio compared with sweet buttermilk (Table 1), as already discussed in the previous section. This lack of foaming capacity in whey buttermilk makes this new ingredient particularly attractive for some industrial applications, in which foaming is a critical issue.

      Conclusions

      Whey buttermilk showed significant differences in composition and functional properties compared with sweet or cultured buttermilks. The composition of sweet or cultured buttermilk was comparable with that of skim milk, whereas the composition of whey buttermilk was close to that of whey, except for the fat content, which was always higher for buttermilk. This finding is related to the fact that, by definition, buttermilk is the plasma in which the fat globules are dispersed, which is skim milk in case of sweet or cultured cream, and whey in case of whey cream, with the additional MFGM fractions. The functional properties of whey buttermilk were different from sweet and cultured buttermilk. Whey buttermilk exhibited higher emulsification properties and lower foaming ability compared with sweet or cultured buttermilk, possibly because of a higher ratio of phospholipids to protein. On the other hand, whey buttermilk showed stable levels of protein solubility, emulsifying capacity, and viscosity over a pH range of 4 to 6, whereas sweet or cultured buttermilks, rich in casein, had lower solubility and emulsifying capacity, and a higher viscosity at acidic pH (pH < 5).
      Our findings suggest that whey buttermilk could be an interesting and novel dairy ingredient, especially in formulation of low-pH food. However, the sensory properties of whey buttermilk need to be evaluated and compared with commercial buttermilks to ensure its suitability in food formulation. This is the focus of an ongoing study by the authors of this work.

      Acknowledgments

      This work has been funded in part by Dairy Management Inc. , the CSU Agricultural Research Initiative, California Dairy Research Foundation, Natural Science and Engineering Research Council of Canada, and Hilmar Cheese Company. The authors thank Lorna Lassonde, Jessica Lee, and Jerry Mattas for their technical assistance.

      References

        • American Dairy Products Institute
        Standards for grades of dry milks.
        ADPI, Chicago, IL1990 (bulletin 916)
        • Association of Official Analytical Chemists
        Official Methods of Analysis.
        16th ed. AOAC, Gaithersburg, MD1995
        • Astaire J.C.
        • Ward R.
        • German J.B.
        • Jimenez-Flores R.
        Concentration of polar MFGM lipids from buttermilk by microfiltration and supercritical fluid extraction.
        J. Dairy Sci. 2003; 86: 2297-2307
        • Banavara D.S.
        • Anupama D.
        • Rankin S.A.
        Studies on physicochemical and functional properties of commercial sweet whey powders.
        J. Dairy Sci. 2003; 86: 3866-3875
        • Chandan R.
        Dairy-based ingredients.
        Eagan Press, St. Paul, MN1997
        • Chobert J.M.
        • Bertrand-Harb C.
        • Nicolas M.G.
        Solubility and emulsifying properties of caseins and whey proteins modified enzymatically by trypsin.
        J. Agric. Food Chem. 1988; 36: 883-892
        • Christie W.W.
        • Noble R.C.
        • Davies G.
        Phospholipids in milk and dairy products.
        J. Soc. Dairy Technol. 1987; 40: 10-12
        • Coke M.
        • Wilde P.J.
        • Russel E.J.
        • Clarck D.C.
        The influence of surface composition and molecular diffusion on the stability of foams formed from protein/surfactants mixtures.
        J. Colloid. Interface Sci. 1990; 138: 489-503
        • Corredig M.
        • Dalgleish D.G.
        Isolates from industrial buttermilk: emulsifying properties of materials derived from the milk fat globule membrane.
        J. Agric. Food Chem. 1997; 45: 4595-4600
        • Corredig M.
        • Dalgleish D.G.
        Buttermilk properties in emulsions with soybean oil as affected by fat globule.
        J. Food Sci. 1998; 63: 476-480
        • Corredig M.
        • Dalgleish D.G.
        Effect of heating of cream on the properties of milk fat globule membrane isolates.
        J. Agric. Food Chem. 1998; 46: 2533-2540
        • Corredig M.
        • Roesch R.R.
        • Dalgleish D.G.
        Production of a novel ingredient from buttermilk.
        J. Dairy Sci. 2003; 86: 2744-2750
        • Dillehay D.L.
        • Webb S.J.
        • Schmelz E.M.
        • Merill A.H.
        Dietary sphingomyelin inhibits 1,2-dimethylhydrazine-induced colon cancer in CF1 mice.
        J. Nutr. 1994; 124: 615-620
        • Elling J.L.
        • Duncan S.E.
        • Keenan T.W.
        • Eigel W.N.
        • Boling J.
        Composition and microscopy of reformulated creams from reduced-cholesterol butteroil.
        J. Food Sci. 1996; 61: 48-53
        • Flanagan J.
        • Fitzgerald R.J.
        Functionality of Bacillus proteinase hydrolysates of sodium caseinate.
        Int. Dairy J. 2002; 12: 737-748
        • Guzman-Gonzalez M.
        • Morais F.
        • Ramos M.
        • Amigo L.
        Influence of skimmed milk concentrate replacement by dry dairy products in a low fat set-type yoghurt model system. I: Use of whey protein concentrates, milk protein concentrates and skimmed milk powder.
        J. Sci. Food Agric. 1999; 79: 1117-1122
        • Hall G. M
        Methods of testing protein functionality.
        Blackie Academic and Professional, London, UK1996
        • International Dairy Federation
        World Dairy Situation.
        FIL-IDF, Brussels, Belgium2002
        • International Dairy Foods Association
        Milk facts.
        IDFA, Washington, DC2003
        • Joshi N.S.
        • Thakar P.N.
        • Jana A.H.
        Utilization of buttermilk in cheese making: A review.
        Indian Food Packer. 1994; 48: 59-65
        • Kanno C.
        Emulsifying properties of bovine milk fat globule membrane in milk fat emulsion: Condition for the reconstitution of fat globules.
        J. Food Sci. 1989; 54: 1534-1539
        • Kinsella J.E.
        Functional properties of proteins in foods: A survey.
        CRC Crit. Rev. Food Sci. Nutr. 1976; 7: 219-280
        • Leonil J.
        • Molle D.
        • Fauquant J.
        • Maubois J.L.
        • Pearce R.J.
        • Bouhallab S.
        Characterization by ionization mass spectrometry of lactosyl β-lactoglobulin conjugates formed during heat treatment of milk and whey and identification of one lactose-binding site.
        J. Dairy Sci. 1997; 80: 2270-2281
        • Marshall R.T
        Standards methods for the examination of dairy products.
        16th Ed. The American Public Health Association Inc., Washington, DC1992
        • McCrae C.H.
        Heat stability of milk emulsions: Phospholipid-protein interactions.
        Int. Dairy J. 1999; 9: 227-231
        • Michalski M.C.
        • Briard V.
        • Michel F.
        Optical parameters of milk fat globules for laser light scattering measurements.
        Lait. 2001; 81: 787-796
        • Morin P.
        • Jiménez-Flores R.
        • Pouliot Y.
        Effect of temperature and pore size on the fractionation of fresh and reconstituted buttermilk by microfiltration.
        J. Dairy Sci. 2004; 87: 267-273
      1. Morin, P., Y. Pouliot and R. Jiménez-Flores. A comparative study of the fractionation of regular buttermilk and whey buttermilk by microfiltration. J. Food Eng. doi:10.1016/j.jfoodeng.2005.06.065

        • Mulder H.
        • Walstra P.
        The milk fat globule.
        Centre for Agricultural Publishing and Documentation, Wageningen, The Netherlands1974
        • National Agricultural Statistics Service
        Dairy products, 2003 Summary.
        USDA, Washington, DC2004
        • O’Connell J.E.
        • Fox P.F.
        Heat stability of buttermilk.
        J. Dairy Sci. 2000; 83: 1728-1732
        • Raval D.M.
        • Mistry V.V.
        Application of ultrafiltered sweet buttermilk in the manufacture of reduced fat process cheese.
        J. Dairy Sci. 1999; 82: 2334-2343
        • Raymundo A.
        • Franco J.M.
        • Empis J.
        • Sousa I.
        Optimization of the composition of low-fat oil-in-water emulsions stabilized by white lupin protein.
        J. Am. Oil Chem. Soc. 2002; 79: 783-790
        • Roesch R.R.
        • Rincon A.
        • Corredig M.
        Emulsifying properties of fractions prepared from commercial buttermilk by microfiltration.
        J. Dairy Sci. 2004; 87: 4080-4087
        • Rueda R.
        • Sabatel J.L.
        • Maldonado J.
        • Molina-Font J.S.
        • Gil A.
        Addition of gangliosides to an adapted milk formula modifies levels of fecal Escherichia coli in preterm newborn infants.
        J. Pediatr. 1998; 133: 90-94
        • Sachdeva S.
        • Buchheim W.
        Recovery of phospholipids from buttermilk using membrane processing.
        Kieler Milchw. Forsch. 1997; 49: 47-68
        • Schmelz E.M.
        • Dillehay D.L.
        • Webb S.K.
        • Reiter A.
        • Adams J.
        • Merill A.H.
        Sphingomyelin consumption supresses aberrant colonic crypt foci and increases the proportion of adenomas versus adenocarcinomas in CF1 mice treated with 1,2-di-methylhydrazine: Implications for dietary sphingolipids and colon carcinogenis.
        Cancer Res. 1996; 56: 4936-4941
        • Schmelz E.M.
        • Dombrink-Kurtzman M.A.
        • Roberts P.C.
        • Kozutsumi Y.
        • Kawasaki T.
        • Merill A.H.
        Induction of apoptosis by fumosinin B1 in HT-29 cells is mediated by the accumulation of endogenous free sphingoid bases.
        Toxicol. Appl. Pharmacol. 1998; 148: 252-260
        • Scott L.L.
        • Duncan S.E.
        • Sumner S.S.
        • Waterman K.M.
        • Kaylegian K.E.
        Influence of emulsifying component composition on creams formulated with fractionated milkfat.
        J. Agric. Food Chem. 2003; 51: 5933-5940
        • Singh H.
        • Tokley R.P.
        Effects of preheat treatments and buttermilk addition on the seasonal variations in the heat stability of recombined evaporated milk and reconstituted concentrated milk.
        Aust. J. Dairy Technol. 1990; 45: 10-16
        • Sprong R.C.
        • Hulstein M.F.E.
        • van der Meer R.
        Bovine milk fat components inhibit food-borne pathogens.
        Int. Dairy J. 2002; 12: 209-215
        • Surel O.
        • Famelart M.H.
        Ability of ceramic membranes to reject lipids of dairy products.
        Aust. J. Dairy Technol. 1995; 50: 36-40
        • Trachoo N.
        • Mistry V.V.
        Application of ultrafiltered sweet buttermilk and sweet buttermilk powder in the manufacture of nonfat and low fat yogurts.
        J. Dairy Sci. 1998; 81: 3163-3171
        • Turcot S.
        • Turgeon S.L.
        • St-Gelais D.
        Effect of buttermilk phospholipid concentrations in cheese milk on production and composition of low fat Cheddar cheese.
        Lait. 2001; 81: 429-442
        • Vaghela M.N.
        • Kilara A.
        Foaming and emulsifying properties of whey protein concentrates as affected by lipid composition.
        J. Food Sci. 1996; 61: 275-280
        • Vetter J. L
        Dairy products for the cereal processing industry.
        American Association of Cereal Chemists Inc., St. Paul, MN1984
        • Walstra P.
        • Geurts T.J.
        • Noomen A.
        • Jellema A.
        • van Boekel M.A.J.S.
        Dairy Technology: Principles of milk properties and processes.
        Marcel Dekker, New York, NY1999
        • Wong P.Y.Y.
        • Kitts D.D.
        A comparison of the butter milk solids functional properties to nonfat dried milk, soy protein isolate, dried egg white, and egg yolk powders.
        J. Dairy Sci. 2003; 86: 746-754