Food Chemistry 185 (2015) 362–370

Contents lists available at ScienceDirect

Food Chemistry journal homepage: www.elsevier.com/locate/foodchem

Composition and microstructure of colostrum and mature bovine milk fat globule membrane Xiaoqiang Zou a,b, Zheng Guo b, Qingzhe Jin a, Jianhua Huang a, Lingzhi Cheong b, Xuebing Xu b,⇑, Xingguo Wang a,⇑ a State Key Laboratory of Food Science and Technology, Synergetic Innovation Center of Food Safety and Nutrition, School of Food Science and Technology, Jiangnan University, 1800 Lihu Road, Wuxi 214122, Jiangsu, PR China b Department of Engineering, Aarhus University, Gustav Wieds Vej 10, 8000 Aarhus C, Denmark

a r t i c l e

i n f o

Article history: Received 22 May 2014 Received in revised form 29 March 2015 Accepted 31 March 2015 Available online 4 April 2015 Keywords: Milk fat globule membrane Microstructure variations Phospholipids Sphingomyelin Confocal laser scanning microscopy

a b s t r a c t The microstructures of colostrum and mature bovine milk fat globule membrane (MFGM) were investigated using confocal laser scanning microscopy (CLSM) at different temperatures, and the relationships between microstructure variations and the chemical compositions of the MFGM were also examined. Using a fluorophore-labeled phospholipid probe, we found that non-fluorescent domains on the MFGM were positively correlated with the amount of sphingomyelin at both room (20 °C) and physiological (37 °C) temperatures. However, at the storage temperature (4 °C), there were more non-fluorescent domains on the MFGM. These results indicate that the heterogeneities in the MFGM are most likely to be the result of the lateral segregation of sphingomyelin at the room and physiological temperatures, and at the storage temperature, phospholipids with saturated fatty acids affect the formation of these domains. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction Milk is an oil-in-water emulsion that contains approximately 3.5–5% lipids in an aqueous environment (Jensen, Ferris, & Lammi-Keefe, 1991). Lipids in milk are present as milk fat globules (MFG) which are surrounded by a biological membrane (MFGM). The size of MFG naturally ranges in diameter from 0.2 to 20 lm and the size of the distribution profile has been shown to vary with lactation stages (Briard, Leconte, Michel, & Michalski, 2003; Patton & Keenan, 1975). These globules are important delivery vehicles of triacylglycerols (TAGs), fat-soluble nutrients and bioactive molecules for neonates (German & Dillard, 2006). Before 2010, the most widely used model of the MFGM was based on a trilayer that consisted of an inner monolayer composed of proteins and phospholipids derived from the endoplasmic reticulum and a bilayer from the apical plasma membrane of the epithelial cells. The structure of the bilayer membrane can be described by the fluid mosaic model, which suggests that the ⇑ Corresponding authors. Tel.: +45 89425089; fax: +45 86123178 (X. Xu). Tel./fax: +86 510 85876799 (X. Wang). E-mail addresses: [email protected] (X. Zou), [email protected] (Z. Guo), [email protected] (Q. Jin), [email protected] (J. Huang), [email protected] (L. Cheong), [email protected] (X. Xu), [email protected] (X. Wang). http://dx.doi.org/10.1016/j.foodchem.2015.03.145 0308-8146/Ó 2015 Elsevier Ltd. All rights reserved.

phospholipids are the backbone of the membrane and exist in a fluid state, while proteins are globular molecules that are partially embedded in or protruding from the membrane (Dewettinck et al., 2008). In 2011, Lopez’s group proposed a new model for the MFGM, which has rigid lipid rafts surrounded by more fluid lipids composed of glycerophospholipids and glycoproteins (Lopez, 2011; Lopez, Madec, & Jimenez-Flores, 2010). The MFGM is composed of polar lipids, cholesterol, proteins, glycoproteins, gangliosides and enzymes (Dewettinck et al., 2008; Lopez, 2011). Analysis of the proteins in bovine MFGM by SDS–PAGE has demonstrated that the MFGM contains a number of polypeptides. Many of the proteins present in the MFGM are reported to have physiological functions, such as butyrophilin (which suppresses multiple sclerosis), xanthine oxidase (a bactericidal agent), fatty acid binding protein (a cell growth inhibitor), and betaglucuronidase inhibitor (which inhibits colon cancer) (Ito, Kamata, Hayashi, & Ushiyama, 1993; Martin, Hancock, Salisbury, & Harrison, 2004; Mañá et al., 2004; Spitsberg, Matitashvili, & Gorewit, 1995). The lipids in the MFGM are mainly polar lipids, including phosphatidylinositol (PI), phosphatidylcholine (PC), phosphatidylethanolamine (PE), phosphatidylserine (PS), and sphingomyelin (SM) (Rombaut, Camp, & Dewettinck, 2005). The glycophospholipids with high saturated fatty acid contents play a major role in the fluidity of the MFGM (Lopez et al., 2010). SM is characterized by its ceramide core, which contains

X. Zou et al. / Food Chemistry 185 (2015) 362–370

sphingosine bonded with a molecule of fatty acid via an amide link, esterified with one polar head group that is either phosphocholine or phosphoethanolamine. SM mainly contains saturated fatty acids, which contribute to its high melting temperature (Fong, Norris, & MacGibbon, 2007). The distribution of the MFGM polar lipids is asymmetric, with PC and SM largely located in the outer layer of the membrane and PE, PI and PS concentrated in the inner surface (Deeth, 1997). The MFGM structure has been investigated with microscopy techniques. Wooding (1971) used electron microscopy to observe a dense-staining layer of 10–20 nm between the fat globule and the plasma membrane when the fat globule approached the plasma membrane. Horisberger, Rosset, and Vonlanthen (1977) observed that glycoproteins of bovine and human milk fat globules were clustered and uniformly distributed over the external membrane surface. Confocal laser scanning microscopy (CLSM) has been used to observe the heterogeneous organization of the MFGM using lipophilic probes and lectin (Evers et al., 2008) and to reveal the distribution of polar lipids, membrane proteins, glycolipids and glycoproteins in the bovine and human MFGM (Lopez & Ménard, 2011; Lopez et al., 2010, 2011). The heterogeneities in the MFGM are due to the lateral segregation of SM, cholesterol and highly saturated phospholipids in rigid liquid-ordered domains surrounded by a fluid matrix of liquiddisordered glycophospholipids (Lopez et al., 2010; Zheng, Jiménez-Flores, Gragson, & Everett, 2014). Lopez et al. (2011) illustrated that these heterogeneities were exclusive of proteins. The reason for the observation of phase separation under CLSM is that the liquid-ordered domains are so tightly packed that the fluorescent probes cannot insert into this area, whereas they insert easily into the liquid-disordered domains, which have high fluidity and a disordered molecular arrangement (Lopez et al., 2010). The major component of the MFGM is polar lipids, whose relative proportions and fatty acid compositions vary with factors such as the lactation stage, diet, genetics and MFG sizes. Whether and how the composition of the polar lipids affects the microstructure of the MFGM remains unknown. Lopez et al. (2011) hypothesized that the non-fluorescent domains observed under CLSM were mainly composed of SM, on the basis of its special properties. Gallier, Gragson, Jiménez-Flores, and Everett (2010) used fluorophore-labeled SM to stain the SM-rich domains, but the images obtained under CLSM were similar to other phospholipid-based fluorescent probes. Furthermore, studies on the microstructure of the MFGM have been conducted mainly at the room temperature (20 °C), and the microstructure changes under the storage (4 °C) and physiological temperatures (37 °C) have rarely been reported, except that Lopez and Ménard (2011) investigated the human MFGM at the physiological temperature. Based on the abovementioned considerations, colostrum and mature bovine milk, which have significant differences in polar lipid proportions and their fatty acid compositions, were used to investigate the relationship between the lipid composition of the MFGM and its microstructure at different temperatures. 2. Materials and methods 2.1. Samples and reagents Bovine colostrum (1–5 days, n = 10) and mature (after 16 days, n = 10) milk samples were obtained from a local Danish dairy farm. Lipid-soluble Nile Red fluorescent dye (9-diethylamino-5H-benzoalpha-phenoxazine-5-one), bought from Sigma–Aldrich Corp. (St Louis, MO, USA), was prepared at 1 mg/mL in acetone and used to stain the triacylglycerol core of the bovine MFG. The fluorescent dye N-(lissamine rhodamine B sulfonyl) di-oleoyl-

363

phosphatidylethanolamine (Rd-DOPE, 1 mg/mL in chloroform) was used to label the phospholipid membrane, purchased from Avanti Polar Lipids, Inc. (Alabaster, AL, USA). Phospholipid standards were supplied by Sigma–Aldrich Corp.: PE (L-a-phosphatidylethanolamine, dioleoyl, purity 99%), PI (L-a-phosphatidylinositol ammonium salt from soybean; purity 98%), PS (1,2-diacylsn-glycero-3-phospho-L-serine from bovine brain; purity 97%), PC (1,2-dipalmitoyl-sn-glycero-3-phosphocholine; purity 99%) and SM (SM from bovine brain; purity 97%). Silicic acid 60G TLC plates and BF3 methanol solution (10% w/v) were purchased from Sigma– Aldrich Corp. Methanol, chloroform, hexane, diethyl ether and heptanes were all of high-performance liquid chromatography purity. 2.2. Size distribution The MFG size distributions were determined with integrated light scattering using a Mastersizer 2000 (Malvern Instruments Ltd., Malvern WR14 1XZ, UK). The refractive index for bovine milk fat is 1.460 and 1.458 at 466 and 633 nm, respectively (Michalski, Michel, Sainmont, & Briard, 2002). The casein micelles were dissociated by diluting the milk in 35 mM EDTA buffer (pH 7) and the sample was diluted in the measurement cell to reach 10% obscuration. The size distribution was evaluated by the P 3 P 2 volume-surface average diameter (d32, defined as ni di = ni di ) and the volume-weighted average diameter (d43, defined as P 4 P 3 ni di = ni di ), where ni is the number of globules in a size class of diameter di. 2.3. Lipid analysis 2.3.1. Extraction of total lipids Total lipids were extracted from the freeze-dried samples by homogenization with chloroform/methanol (2:1, v/v), as described by Folch, Lees, and Sloane-Stanley (1957). The extract was shaken and equilibrated with one-fourth volume of a saline solution (NaCl 0.86%, w/w). The solvent phase was filtered and evaporated under vacuum and the obtained total lipids were stored at 20 °C for further chemical analysis. 2.3.2. Analysis of the fatty acid composition of polar lipids The polar lipids were separated from the total lipids with silica gel G TLC plates using a developing solvent system of hexane/diethyl ether/acetic acid (80:20:1, v/v/v). The polar lipids were scraped off the baseline and extracted with 3 mL chloroform: methanol:water (5:5:1, v/v/v) mixture (Benoit et al., 2010). After centrifugation at 4000 rpm for 10 min, the organic phase was collected. The remaining water phase was extracted twice with the same method and the organic solvent was pooled and evaporated. Three hundred microlitres of BF3 methanol solution was added for methylation and the screw-capped tubes were kept at 100 °C for 90 min. Next, 600 lL of heptane and 500 lL of saturated NaCl solution were added. The mixture was centrifuged at 4000 rpm for 10 min at 20 °C. The solvent phase was collected and dried with anhydrous sodium sulfate. After centrifugation, the upper layer was injected into a gas chromatograph for fatty acid analysis. Fatty acid methyl esters were analyzed on a gas chromatograph (Thermo Trace GC Ultra, USA) equipped with an autosampler, a flame ionization detector and an ionic liquid capillary column (SLB-IL 100, 60 m  0.25 mm  0.2 lm). Helium was used as the carrier gas at a flow rate of 1 mL/min. The column oven temperature was kept at 170 °C and the running time for each sample was 60 min. The injection port and detector temperatures were both set at 250 °C. The fatty acid methyl esters were identified

364

X. Zou et al. / Food Chemistry 185 (2015) 362–370

by comparing the retention time with the standards, and the relative contents expressed as mol% were then calculated. 2.3.3. Analysis of polar lipids Polar lipid separation was conducted on a high-performance liquid chromatograph that was equipped with an evaporative light scattering detector, as described by Rombaut et al. (2005) with slight modification. Nitrogen was used as the nebulizing gas at a flow rate of 1 L/min and the evaporating temperature was set at 85 °C. A silica column with a 5 lm particle diameter and a precolumn with the same packing and internal diameter were used. The elution was performed under isocratic conditions with 87.5:12:0.5 (v/v/v) chloroform/methanol/triethylamine buffer (pH 3, 1 M formic acid) from 0 to 10 min and then with a linear gradient from 87.5:12:0.5 (v/v/v) at t = 11 min to 28:60:12 (v/v/v) at t = 45 min. The mobile phase was brought back to the initial conditions at t = 47 min, and the column was allowed to equilibrate until the next injection at t = 55 min. The flow rate was maintained at 0.5 mL/min, the injection volume was 10 lL and the samples and column were equilibrated at 40 °C. The identification of the polar lipids was conducted by performing a comparison with the retention times of the pure standards. 2.4. Microstructural analysis The phospholipids in the MFGM were labeled by Rd-DOPE with 5 lL solution in 0.5 mL milk. The sample was mixed by gentle swirling and inversion of the vial until the fluorescent probe was evenly dispersed. The mixed samples were maintained at the room temperature for 20 min before microstructural analysis. Labeled milk (100 lL) was mixed with 100 lL agarose (10 g/L in deionized water, kept at 45 °C). An aliquot (25 lL) was placed on a slide, and a coverslip was quickly applied without excessive pressure. The prepared slides were kept at the room (20 °C), storage (4 °C), or body temperature (37 °C) for 30 min to allow the redistribution of components in the MFGM to reach a new equilibration at that temperature (Zou et al., 2012). The slides were then placed in a thermostat with a transparent window for microstructure analysis at the room, storage, or body temperature. The microstructure of the MFG was analyzed with a Zeiss LSM 510 Meta confocal microscope. A 63  1.4 oil immersion objective was used throughout. Nile red and Rh-DOPE were both excited with the 543 nm line of the He–Ne laser, and the emission was captured with 560–620 nm band pass filters. Excitation of Fast green was achieved using the 488 nm line from the Ar laser, with the band-pass filter set at 505–530 nm. 2.5. Statistical analysis Analysis of variance tests were performed using the Statistical Analysis System software (SAS, Cary, NC). The significance level was set at a = 0.05 and differences were considered to be significant at P < 0.05. 3. Results and discussion 3.1. Fatty acid composition of phospholipids in the colostrum and mature bovine MFGM To investigate the effect of the fatty acid compositions of the phospholipids on the microstructure variation, the SM fatty acids were excluded using a relatively mild methylation method. Table 1 lists the fatty acid composition of the phospholipids (PE, PI, PC and PS) in the colostrum and mature bovine milk. As shown in Table 1, the content of saturated fatty acids in the mature bovine

Table 1 Fatty acid composition of phospholipids from colostrum and mature bovine MFGM (Mean ± SD). Fatty acids

Colostrums (n = 10)

Mature milk (n = 10)

Stata

C4:0 C6:0 C8:0 C10:0 C12:0 C14:0 C14:1 x-5 C15:0 C15:1 C16:0 C16:1 x-7 C17:0 C17:1 C18:0 C18:1 x-9 tans C18:1 x-9 C18:2 x-6 tans C18:2 x-6 C18:3 x-3 C20:0 C20:1 x-9 C21:0 C20:2 x-6 C20:3 x-6 C22:0 C20:5 x-3 C23:0 C22:2 x-6 C24:0 C24:1 x-9 C22:5 x-6 C22:5 x-3 C22:6 x-3 SFA MUFA PUFA

0.26 ± 0.04 0.38 ± 0.03 0.57 ± 0.03 1.21 ± 0.06 2.51 ± 0.20 3.97 ± 0.23 0.34 ± 0.06 0.61 ± 0.07 0.16 ± 0.04 15.30 ± 0.72 1.57 ± 0.08 0.98 ± 0.09 0.36 ± 0.10 13.29 ± 0.91 0.76 ± 0.13 39.28 ± 1.03 0.28 ± 0.04 6.53 ± 0.39 0.61 ± 0.04 0.30 ± 0.05 1.02 ± 0.14 0.18 ± 0.02 0.24 ± 0.04 0.17 ± 0.04 2.04 ± 0.15 0.86 ± 0.09 2.23 ± 0.15 0.45 ± 0.09 1.88 ± 0.17 0.53 ± 0.14 0.28 ± 0.01 0.56 ± 0.02 0.30 ± 0.06 45.71 ± 1.10 44.01 ± 0.79 10.28 ± 0.66

0.31 ± 0.06 0.45 ± 0.06 0.69 ± 0.12 1.52 ± 0.11 2.73 ± 0.11 5.22 ± 0.33 0.75 ± 0.13 0.77 ± 0.11 0.21 ± 0.04 17.24 ± 0.80 1.73 ± 0.11 0.71 ± 0.06 0.33 ± 0.07 16.24 ± 1.17 0.59 ± 0.06 31.98 ± 0.81 0.31 ± 0.08 7.06 ± 0.44 0.43 ± 0.05 0.27 ± 0.03 0.75 ± 0.10 0.12 ± 0.07 0.20 ± 0.06 0.14 ± 0.03 2.31 ± 0.21 0.23 ± 0.03 2.79 ± 0.30 0.25 ± 0.04 2.49 ± 0.25 0.35 ± 0.06 0.28 ± 0.04 0.32 ± 0.03 0.23 ± 0.04 53.86 ± 1.18 36.69 ± 0.74 9.45 ± 0.64

NSb NS

Probability of F-test: ⁄P < 0.05, ⁄⁄P < 0.01, a Results of the analysis of variance. b NS: no significant difference.

⁄⁄⁄

⁄

NS NS ⁄⁄ ⁄⁄

NS NS ⁄

NS ⁄

NS ⁄⁄⁄ ⁄ ⁄

NS NS ⁄⁄

NS ⁄⁄ ⁄

NS NS NS ⁄⁄ ⁄⁄ ⁄⁄ ⁄⁄ ⁄⁄

NS ⁄⁄

NS ⁄⁄ ⁄⁄⁄ ⁄⁄

P < 0.001.

milk was significantly higher than that in the colostrum, which indicates that the mature MFGM had more possibility to have phospholipids with two saturated fatty acids which could segregate from other phospholipids with unsaturated fatty acids at room or lower temperatures due to their relatively higher melting temperature. Some studies have reported the fatty acid composition of different types of phospholipids in the MFGM, showing that the saturated fatty acid composition of PI is 36%, PS 40%, PE 25% and PC 50% (Benoit et al., 2010). PC has the highest saturated fatty acid composition, which could play a major role in the segregation of phospholipids with SM and cholesterol, whereas PE has the lowest saturated fatty acid content, which could greatly contribute to the fluidity of the membrane. 3.2. Polar lipid composition of the colostrum and mature bovine MFGM Polar lipids are the backbone of the MFGM. Because different polar lipids have different properties, the polar lipid composition could greatly influence its microstructure. The compositions and relative proportions of the polar lipids are given in Table 2. As shown in Table 2, the total amount of polar lipids in the colostrum was significantly lower than that in the mature bovine milk. The analysis of the size distributions showed that the mean volumeweighted diameter of the colostrum (5.15 ± 0.09 lm) was higher than that of the mature MFG (4.26 ± 0.11 lm). Due to larger globule size of the mature bovine milk than that of the colostrum, their amount of polar lipids on a specific surface area basis is not

365

X. Zou et al. / Food Chemistry 185 (2015) 362–370 Table 2 The amounts of polar lipids of colostrum and mature bovine MFGM on a total lipid weight and specific surface area basis and their relative proportions (Mean ± SD). Polar lipidsa

Concentration of polar lipids (mg polar lipids per g of total lipids) Colostrum (n = 10)

PE PI PS PC SM Total a b

a

1.45 ± 0.19 0.47 ± 0.02a 0.35 ± 0.03a 1.20 ± 0.04a 1.31 ± 0.11a 4.78 ± 0.22a

b

Amount of polar lipids on a specific surface area basis (mg/m2)

Relative proportion of polar lipids (% of polar lipids)

Mature milk (n = 10)

Colostrum (n = 10)

Mature milk (n = 10)

Colostrum (n = 10)

Mature milk (n = 10)

1.92 ± 0.25b 0.72 ± 0.03b 0.51 ± 0.02b 1.22 ± 0.06a 0.84 ± 0.08b 5.21 ± 0.27b

0.77 ± 0.10a 0.25 ± 0.01a 0.19 ± 0.02a 0.64 ± 0.02a 0.70 ± 0.06a 2.54 ± 0.12a

0.92 ± 0.12b 0.35 ± 0.01b 0.25 ± 0.01b 0.59 ± 0.03b 0.40 ± 0.04b 2.50 ± 0.13a

30.23 ± 2.69a 9.89 ± 0.87a 7.32 ± 0.99a 25.20 ± 1.88a 27.36 ± 1.07a

36.72 ± 2.92b 13.85 ± 1.21b 9.83 ± 0.32b 23.49 ± 2.28a 16.11 ± 0.79b

PE: phosphatidylethanolamine; PI: phosphatidylinositol; PS: phosphatidylserine; PC: phosphatidylcholine; SM: sphingomyelin. Means ± SD with the same letter are not significantly different at the 0.05 probability level.

significant even though the mature bovine milk has more content of polar lipids. As for the relative proportion of the polar lipids, the amounts of PE were the highest, which were 30% and 36.7% for the colostrum and mature bovine milk, respectively. The amounts of PC in the MFGM of both milk types were more than 20%, but were not significantly different. The amounts of SM in the colostrum and mature bovine milk were 27% and 16%, respectively, which was significantly different (P < 0.001). 3.3. Microstructure investigation of the colostrum and mature bovine MFGM using CLSM Two-dimensional images of the colostrum and mature bovine MFGM at the room temperature are shown in Fig. 1A1 and B1. Non-fluorescent domains were observed on the surface of the globules in both milk types, which indicates the heterogeneous distribution of the polar lipids in both MFGM. However, some

individual globules had non-fluorescent domains on their equatorial section, and others had an entire equatorial circle. The globules without non-fluorescent domains on their equatorial sections did not necessarily lack domains of this type on their surfaces. However, according to the possibility distribution of the non-fluorescent domains, the higher the number of domains that are observed on the equatorial section, the greater is the likelihood that the globules have non-fluorescent domains on their surfaces. Comparing the non-fluorescent domains in the 2D images of the colostrum and mature bovine MFGM did not reveal which one had more non-fluorescent domains, and thus investigation of their 3D images was required. 3D images of both the colostrum and mature bovine MFGM are shown in Fig. 1A2 and B2. Most of the globules in the colostrum had domains on their surfaces, whereas in mature milk, the domains were not seen on many of the globules. According to their chemical composition, the colostrum had more content of SM in the

Fig. 1. 2D and 3D images of bovine MFGM labeled by the Rh-DOPE fluorescent probe observed under CLSM at 20 °C. Arrows indicate the non-fluorescent domains. Colostrum (A1 and A2) and mature milk (B1 and B2).

366

X. Zou et al. / Food Chemistry 185 (2015) 362–370

membrane than that of the mature bovine milk, whereas the mature bovine milk had more content of saturated fatty acids in phospholipids than that of the colostrum. The results indicated that the composition of the polar lipids had more influence on the microstructure of the MFGM than the fatty acid composition of the phospholipids at room temperature. The reason for the insignificant influence of saturated fatty acids was largely due to their distribution in the phospholipids. Although the mature bovine milk had a higher saturated fatty acid content, they were distributed across different types of phospholipids, and not concentrated in a certain type of phospholipid. Some phospholipids with two molecules of saturated fatty acids, which have high melting temperature and can segregate from other phospholipids with SM to form liquid-order domains, may evenly distribute into different layers of membrane, which further decreases the influence of saturated fatty acids on the microstructure. Through an extensive study of the microstructure of the MFGM, we found that the number of non-fluorescent domains on the surface of the membrane was positively proportional to the size of the globule, which was in accordance with the results of Gallier et al. (2010). The smaller their size, the smaller was the number of domains observed on the globules, and vice versa. Non-fluorescent domains were rarely observed on some small globules under CLSM, which was also shown by Lopez et al. (2011). Gallier et al. (2010) statistically calculated the number of domains on different sizes of globules: the small globules with diameters of under 2 lm had approximately three to four domains on average, the globules with diameters between 2 and 5 lm had an average of five to six domains, and the number of domains on the surfaces of the large globules varied greatly. Then, they used different types of fluorophore-labeled phospholipid probes, including PE, PC and SM probes, to label the membranes of the globules to prove the composition of the domains. The results showed that the morphology of the MFGM

with the fluorophore-labeled SM probe was similar to those of the other phospholipid probes, which was probably due to the bulky fluorophore in the fatty acids of the SM; this circumstance limited its ability to pack together and enter the liquid-order phase. Comparing the microstructures of the colostrum and mature bovine MFGM, which had significantly different SM contents, showed that the number of non-fluorescent domains on the surface of the globules was associated with the amounts of SM, which corroborated the hypothesis proposed by Gallier et al. (2010), Lopez et al. (2011) and Lopez and Ménard (2011). However, further verification of the theory should be conducted using artificial membranes with different polar lipid compositions. The microstructure analysis of the bovine MFGM by observation under CLSM showed that the MFG with the same diameter had different surface morphologies. Fig. 2A shows two globules of the same diameter, one has non-fluorescent domains on the surface and the other has no such domains. Fig. 2B shows two globules of the same diameter both have domains on their surfaces, but the sizes of the domains differ. One globule has tiny domains distributed on its surface, and the other has larger ones. Fig. 2C shows globules have considerably larger domains (approximately 1 lm in diameter) than those observed on other globules. Fig. 2D shows a globule has both small and large domains. The differences in the number and size of the domains among the globules with the same diameter or different diameter classes were probably due to the differences in the chemical compositions of the MFGM and their microenvironments. Based on the mechanism of MFGM formation, the bilayer of the membrane is derived from the apical plasma membrane of the mammary epithelial cells when the milk is secreted. On the surface of the mammary epithelial cell membrane, there are lots of lipid rafts in which SM is concentrated. When the globules are secreted, these lipid rafts are taken away with the plasma membrane enveloping the surface of the lipid droplets.

Fig. 2. 3D images of bovine MFGM with the non-fluorescent domains labeled by the Rh-DOPE fluorescent probe observed under CLSM at 20 °C. Arrows indicate the nonfluorescent domains, and circles highlight the merging domains.

X. Zou et al. / Food Chemistry 185 (2015) 362–370

367

Fig. 3. 3D images of bovine MFGM labeled by the Rh-DOPE fluorescent probe observed under CLSM at 4 °C. Arrows indicate the non-fluorescent domains. Colostrum (A1, A2 and A3), mature milk (B1, B2 and B3) and MFGM on the top (C1) and bottom (C2).

368

X. Zou et al. / Food Chemistry 185 (2015) 362–370

The smaller the globules are, the fewer domains they receive, and vice versa. However, some special situations could be encountered as described above, which are probably due to the difference in composition of the mammary epithelial cell plasma membrane, which leads to differences in the chemical composition of the MFGM at the same diameter class. The shape of the domains on the globule membrane depends on the line tension between the liquid-ordered and liquid-disordered phases and the dipole–dipole interactions between molecules at the interface of the two phases (Lopez & Ménard, 2011). To minimize the domain boundaries and edge energy, circular domains form at room temperature. Due to the high fluidity of the membrane, liquid-ordered SM-based domains can move freely in the liquid-disordered phospholipidbased environment. Therefore, there is a strong likelihood that two domains will collide with each other and fuse to form larger domains. The areas highlighted by white circles in Fig. 2E and F are the fused domains that were observed under CLSM. This phenomenon was also observed by Evers et al. (2008). Therefore, it can be speculated that the large domains on the surface of the MFGM were probably formed by the fusion of small domains as they moved in the phospholipid-based environment. Phospholipids, composed of unsaturated fatty acids, have low melting temperatures and thus have little influence on the liquid-ordered domains on the surface of MFG at room temperature. However, when the temperature is decreased sufficiently, the influence of these phospholipids on the microstructure may differ. Therefore, the milk samples were kept at the storage temperature (4 °C) to investigate how the microstructure of the MFGM changed. The overview and zoomed-in images of the bovine MFGM observed under CLSM at the storage temperature (4 °C) are presented in Fig. 3. As shown in the overview images of the bovine colostrum (Fig. 3A1) and mature (Fig. 3B1) MFGM, the number and

size of the non-fluorescent domains are apparently increased compared with these in room temperature (Fig. 1A2, B2, A3 and B3). The zoomed-in images show that the shapes of the domains of both MFGM become irregular. The changes in the appearance of the domains on the surfaces of the globules from circular at room temperature to angular and elongated at storage temperature indicate that the substances on the surface of the membrane are rearranged. The phospholipids with saturated fatty acids might have great influence on this organization of the MFGM, and the formation of TAG crystals at this temperature might also lead to these changes. At the room temperature, the non-fluorescent domains on the MFGM can be formed by the segregation of SM and cholesterol, and phospholipids with relatively high melting temperatures may also form these domains with SM and cholesterol or alone. The function of highly saturated phospholipids in the formation of non-fluorescent domains on the MFGM has been described by Zheng, Jiménez-Flores, Gragson et al. (2014), who proved that the presence of SM was not a necessary requirement for the formation of liquid-ordered regions by the construction of giant unilamellar vesicles to mimic the morphology of the MFGM using bovine milk-derived phospholipids, and by applying washing processes using aqueous solutions with different degrees of stringency, Zheng, Jiménez-Flores, and Everett (2014) proved that PC was mainly located on outer layer of the MFGM and phospholipids that induced liquid-ordered regions was most likely to be PC due to its high contents of saturated fatty acids. As the temperature decreased, the phospholipids with one molecule of saturated fatty acid (relatively higher melting temperature than other phospholipids with two molecules of unsaturated fatty acids) separated and gathered around the formed domains. The fluorescent probes, due to their high fluidity (low melting temperature), were elbowed out. Through this course, the domains became larger, which could

Fig. 4. 3D images of bovine MFGM labeled by the Rh-DOPE fluorescent probe observed under CLSM at 37 °C. Arrows indicate the non-fluorescent domains. Colostrum (A1 and A2) and mature milk (B1 and B2).

X. Zou et al. / Food Chemistry 185 (2015) 362–370

explain the increase in the sizes of the non-fluorescent domains. However, when the milk was maintained at the storage temperature, the shapes became irregular, indicating that the domains, at this moment, probably corresponded to the gel phase in which the lipids were semi-crystalized (Evers et al., 2008), not to the liquid-ordered phase. The partial crystallization of the lipids in the membrane was due to the low storage temperature. We can speculate that if a suitable decrease rate of temperature is used, the domains that are both large size and circular could be obtained as well. As shown in Fig. 4, the microstructure images of the colostrum and mature bovine MFGM were similar at the storage temperature, even though they had significantly different saturated fatty acid compositions. As for the influence of TAG cores, their crystallization at the storage temperature might alter the internal environment of the MFGM, and thus might also have influence on the organization changes. The microstructure of the bovine MFGM which was at the storage temperature for a short period is shown in Figs. 3C1 (bottom) and C2 (top). As obviously seen in these images, the morphologies of the domains on the surface of the MFGM were obviously different. Some of the domains were still circular, some started to become irregular, and some connected to one another and formed larger domains. The morphology differences in the different domains on the MFGM may reveal how enlarged and angular domains are formed at low temperature. The microstructures of the colostrum and mature bovine MFGM at 37 °C are shown in Figs. 4A and B, respectively. As shown in Fig. 4, even when the MFG were kept at physiological temperature, non-fluorescent domains could also be observed, and similar results were also reported by Lopez and Ménard (2011). However, due to the difference in the polar lipid composition of the MFGM, the number of domains in the globule membrane of colostrum was more than that in mature milk. The saturated fatty acid content of the phospholipids in the mature MFGM was significantly greater than that in the colostrum, but the saturated fatty acids were distributed in different species of phospholipids, which decreased their influence on the microstructure of the MFGM. Compared with the domains in room temperature, the size of these non-fluorescent domains in the physiological temperature was obviously smaller, which was largely due to the transition of some phospholipids in these domains from liquid-ordered phase to liquid-disordered phase or a lowering of the amount of the gel phase phospholipids. The melting temperature of SM is 35 °C (Malmsten, Bergenståhl, Nyberg, & Odham, 1994). However, the non-fluorescent domains could still be observed at the physiological temperature, indicating that the major substance, that is, SM, in the domains was still packed together and the formation of lipid domains is therefore largely the result of phase separation, which is in accordance with a recent study of Guyomarch et al. (2014), who reported that phase separation was clearly visible at temperatures below 35 °C by the observation of supported lipid bilayers composed of milk SM and dioleoylphosphatidylcholine 50/ 50 mol% using atomic force microscopy height imaging. During the course of observation, a few large domains could also be found in some globules, as shown in Fig. 4A2, which were probably the result of the fusion of small domains due to the increased fluidity of the membrane at this temperature.

4. Conclusion In conclusion, the investigation of the bovine MFGM using CLSM and Rh-DOPE indicated that there were heterogeneous domains in the MFGM. By observation of the images of the colostrum and mature bovine MFGM at the room and physiological temperature, the number of domains was increased with the

369

increase of the amounts of SM. At the storage temperature, the shapes of the domains became irregular, the number of domains was significantly increased and no difference in the morphology between the colostrum and mature bovine MFGM was observed. These results indicates that the heterogeneities in the MFGM are largely the result of the lateral segregation of SM at the room and physiological temperature, and at low temperature, the phospholipids with saturated fatty acids had influence on the formation of these domains.

Acknowledgements This work was supported by the National Science & Technology Pillar Program during the Twelfth Five-Year Plan Period (No. 2011BAD02B04) and Fundamental Research Funds for the Central Universities (JUSRP11439).

References Benoit, B., Fauquant, C., Daira, P., Peretti, N., Guichardant, M., & Michalski, M.-C. (2010). Phospholipid species and minor sterols in French human milks. Food Chemistry, 120(3), 684–691. Briard, V., Leconte, N., Michel, F., & Michalski, M. C. (2003). The fatty acid composition of small and large naturally occurring milk fat globules. European Journal of Lipid Science and Technology, 105(11), 677–682. Deeth, H. (1997). The role of phospholipids in the stability of milk fat globules. Australian Journal of Dairy Technology, 52(1), 44–46. Dewettinck, K., Rombaut, R., Thienpont, N., Le, T. T., Messens, K., & Van Camp, J. (2008). Nutritional and technological aspects of milk fat globule membrane material. International Dairy Journal, 18(5), 436–457. Evers, J. M., Haverkamp, R. G., Holroyd, S. E., Jameson, G. B., Mackenzie, D. D., & McCarthy, O. J. (2008). Heterogeneity of milk fat globule membrane structure and composition as observed using fluorescence microscopy techniques. International Dairy Journal, 18(12), 1081–1089. Folch, J., Lees, M., & Sloane-Stanley, G. (1957). A simple method for the isolation and purification of total lipids from animal tissues. The Journal of Biological Chemistry, 226(1), 497–509. Fong, B. Y., Norris, C. S., & MacGibbon, A. K. (2007). Protein and lipid composition of bovine milk-fat-globule membrane. International Dairy Journal, 17(4), 275–288. Gallier, S., Gragson, D., Jiménez-Flores, R., & Everett, D. (2010). Using confocal laser scanning microscopy to probe the milk fat globule membrane and associated proteins. Journal of Agricultural and Food Chemistry, 58(7), 4250–4257. German, J. B., & Dillard, C. J. (2006). Composition, structure and absorption of milk lipids: A source of energy, fat-soluble nutrients and bioactive molecules. Critical Reviews in Food Science and Nutrition, 46(1), 57–92. Guyomarch, F., Zou, S., Chen, M., Milhiet, P., Godefroy, C., Vie, V., et al. (2014). Milk sphingomyelin domains in biomimetic membranes and the role of cholesterol: morphology and nanomechanical properties investigated using AFM and force spectroscopy. Langmuir, 30(22), 6516–6524. Horisberger, M., Rosset, J., & Vonlanthen, M. (1977). Location of glycoproteins on milk fat globule membrane by scanning and transmission electron microscopy, using lectin-labelled gold granules. Experimental Cell Research, 109(2), 361–369. Ito, O., Kamata, S., Hayashi, M., & Ushiyama, K. (1993). Milk fat globule membrane substances inhibit mouse intestinal b-glucuronidase. Journal of Food Science, 58(4), 753–755. Jensen, R. G., Ferris, A. M., & Lammi-Keefe, C. J. (1991). The composition of milk fat. Journal of Dairy Science, 74(9), 3228–3243. Lopez, C. (2011). Milk fat globules enveloped by their biological membrane: Unique colloidal assemblies with a specific composition and structure. Current Opinion in Colloid & Interface Science, 16(5), 391–404. Lopez, C., Briard-Bion, V., Ménard, O., Beaucher, E., Rousseau, F., Fauquant, J., et al. (2011). Fat globules selected from whole milk according to their size: Different compositions and structure of the biomembrane, revealing sphingomyelin-rich domains. Food Chemistry, 125(2), 355–368. Lopez, C., Madec, M.-N., & Jimenez-Flores, R. (2010). Lipid rafts in the bovine milk fat globule membrane revealed by the lateral segregation of phospholipids and heterogeneous distribution of glycoproteins. Food Chemistry, 120(1), 22–33. Lopez, C., & Ménard, O. (2011). Human milk fat globules: Polar lipid composition and in situ structural investigations revealing the heterogeneous distribution of proteins and the lateral segregation of sphingomyelin in the biological membrane. Colloids and Surfaces B: Biointerfaces, 83(1), 29–41. Malmsten, M., Bergenståhl, B., Nyberg, L., & Odham, G. (1994). Sphingomyelin from milk-characterization of liquid crystalline, liposome and emulsion properties. Journal of the American Oil Chemists’ Society, 71(9), 1021–1026. Mañá, P., Goodyear, M., Bernard, C., Tomioka, R., Freire-Garabal, M., & Liñares, D. (2004). Tolerance induction by molecular mimicry: prevention and suppression of experimental autoimmune encephalomyelitis with the milk protein butyrophilin. International Immunology, 16(3), 489–499.

370

X. Zou et al. / Food Chemistry 185 (2015) 362–370

Martin, H. M., Hancock, J. T., Salisbury, V., & Harrison, R. (2004). Role of xanthine oxidoreductase as an antimicrobial agent. Infection and Immunity, 72(9), 4933–4939. Michalski, M.-C., Michel, F., Sainmont, D., & Briard, V. (2002). Apparent f-potential as a tool to assess mechanical damages to the milk fat globule membrane. Colloids and Surfaces B: Biointerfaces, 23(1), 23–30. Patton, S., & Keenan, T. (1975). The milk fat globule membrane. Biochimica et Biophysica Acta (BBA) – Reviews on Biomembranes, 415(3), 273–309. Rombaut, R., Camp, J., & Dewettinck, K. (2005). Analysis of phospho-and sphingolipids in dairy products by a new HPLC method. Journal of Dairy Science, 88(2), 482–488. Spitsberg, V. L., Matitashvili, E., & Gorewit, R. C. (1995). Association and coexpression of fatty-acid-binding protein and glycoprotein CD36 in the bovine mammary gland. European Journal of Biochemistry, 230(3), 872–878.

Wooding, F. (1971). The mechanism of secretion of the milk fat globule. Journal of Cell Science, 9(3), 805–821. Zheng, H., Jiménez-Flores, R., & Everett, D. W. (2014a). Lateral lipid organization of the bovine milk fat globule membrane is revealed by washing processes. Journal of Dairy Science, 97(10), 5964–5974. Zheng, H., Jiménez-Flores, R., Gragson, D., & Everett, D. W. (2014b). Phospholipid architecture of the bovine milk fat globule membrane using giant unilamellar vesicles as a model. Journal of Agricultural and Food Chemistry, 62(14), 3236–3243. Zou, X.-Q., Guo, Z., Huang, J.-H., Jin, Q.-Z., Cheong, L.-Z., Wang, X.-G., et al. (2012). Human milk fat globules from different stages of lactation: a lipid composition analysis and microstructure characterization. Journal of Agricultural and Food Chemistry, 60(29), 7158–7167.