Buffalo milk is the second most commonly produced milk in the world, with approximately 90 billion KG produced each year, accounting for 13% of all milk produced in the world (International Dairy Federation, 2010). It is higher in protein content, fat, lactose, total solids, vitamins and minerals than cow milk. These differences can influence the flavour and taste of buffalo milk, and it also makes buffalo milk a highly suitable ingredient for the manufacture of other milk products such as cheese and yogurt (Menard et al. Reference Menard, Ahmad, Rousseau, Briard-Bion, Gaucheron and Lopez2010; Hussain et al. Reference Hussain, Bell and Grandison2011).
China is one of the world's largest producers of buffalo milk, with more than 12 million buffalo raised in China. However, most Chinese buffalo are the native swamp breed with lower milk production than other breeds. Therefore, river buffalo breeds such as Murrah (M) and Nili-Ravi (N) have been introduced from India and Pakistan in an effort to improve milk production through crossbreeding with the native swamp buffalo.
Similar to other mammals, buffalo milk includes caseins and whey protein. The dissimilar distribution of αs1, αs2, β, and κ-casein in buffalo milk proteins compared to Holstein proteins has been reported (Bonfatti et al. Reference Bonfatti, Giantin, Rostellato, Dacasto and Carnier2013). The primary structure of buffalo and bovine αs1-casein is homologous and is made up of 199 amino acids with a theoretical MW of 22·80 kDa (dephosphorylated form). Buffalo β-casein consists of 209 amino acids with a theoretical MW of 24·04 kDa (Ferranti et al. Reference Ferranti, Scaloni, Caira, Chianese, Malorni and Addeo1998). There are several protein variants found in buffalo milk, such as κ-casein (X1: Ile135 and X2: Thr135) and αs1-casein (A: Leu178 and B: Ser178), while more than 10 variants were found in cattle protein (Caroli et al. Reference Caroli, Chessa and Erhardt2009; Bonfatti et al. Reference Bonfatti, Giantin, Rostellato, Dacasto and Carnier2013). The milk protein polymorphism also has some relationship with the breeds, and two β-lactoglobulin variants (A and B) have been detected in Indian and Egyptian buffalo; however, only one variant (B) has been detected in the Mediterranean water buffalo (Patel et al. Reference Patel, Chauhan, Singh and Soni2007; Feligini et al. Reference Feligini, Bonizzi, Buffoni, Cosenza and Ramunno2009).
Various buffalo milk proteins have been previously studied, and their post-translational modifications, genetic variants, native and non-native disulphide bonds, and non-disulphide cross-links have been identified using proteomic techniques (Chevalier & Kelly, Reference Chevalier and Kelly2010; Holland et al. Reference Holland, Gupta, Deeth and Alewood2011). Some studies have discussed the protein content from several varieties of River water buffalo, such as M and N that originated in India and Pakistan, and Mediterranean water buffalo (Aggarwal et al. Reference Aggarwal, Sodhi, Mukesh, Ahlawat, Kumar, Chakra-Vorty and Raina2007; Bonfatti et al. Reference Bonfatti, Gervaso, Coletta and Carnier2012). However, few studies have focused on the differences in milk protein composition between swamp water buffalo and their crossbreeds.
Notably, the resulting prolific crossbreed offspring yielded a great improvement in milk production (Nanda & Nakao, Reference Nanda and Nakao2003) and total milk protein content (Han et al. Reference Han, Meng, Li, Yang, Ren, Zeng and Nout2007). Microbiological evidence, compositional information, genetic polymorphisms of the milk proteins and breed detection have all been reported (Han et al. Reference Han, Meng, Li, Yang, Ren, Zeng and Nout2007; Cottenet et al. Reference Cottenet, Blancpain and Golay2011; Ren et al. Reference Ren, Miao, Chen, Zou, Liang and Liu2011), but there is limited information on the influence of crossbreeding on the content and distribution of milk protein components.
The objective of this study is, therefore, to investigate the effects of crossbreeding on the protein composition, content and distribution among the four types of buffalo milk. The results obtained in this study will be useful in future research that directs breeding to produce high-quality raw buffalo milk for specific purposes, such as cheese-making.
Material and methods
Materials
Milk samples were obtained from the Guangxi Buffalo Institute Experiment Farm in Nanning, Guangxi Province of China. A total of 156 individual milk samples (500 ml) were obtained from pure river buffalo (M, n = 41; and N, n = 34), crossbreeds between M and N (M-N, n = 36), and crossbreeds of river buffalo and local swamp buffalo (C, n = 45). All of the milk samples were taken from animals during mid-lactation that were fed the same diet (Table 1). Buffalos were housed in indoor tie stalls and fed individually with free access to feed and water. The diets were offered as roughage first then the concentrate was fed. The buffalos were milked twice a day. All of the buffalo were 493 ± 30 kg in weight, 2~5 parity, and 3~6 years old. All of the selected crossbreeds derived from a third-generation breeding or after, where the M-N crossbreed was obtained from an M father and an N mother and the C crossbreed was obtained from a local swamp buffalo father with either a 25 M or 20 N mothers. The fresh milk was collected, and the protein content was analysed in triplicate samples by infrared spectroscopy using a Milko Scan FT120 (Foss Electric, Denmark). The milk samples were stored at −20 °C until the protein composition was tested and two-dimensional gel electrophoresis was performed.
Table 1. Ingredients and chemical compositions of experimental diets (DM basis)
CP, Crude protein; NDF, Neutral detergent fibre; ADF, Acid detergent fibre; GE, Gross energy
† One kilogram of premix contained 3 000 IU VE, 150 000 IU VD, 500 000 IU VA, 1·3 g Cu, 4·0 g Fe, 3·0 g Mn, 80 mg I, 6·0 g Zn, 80 mg Co, 50 mg Se
Protein composition analysis
The milk samples were mixed 1:1 with a buffer solution (pH 7, containing 0·1 m BisTris buffer, 6 m guanidine hydrochloride, 5·37 mm sodium citrate, and 19·5 mm dl-dithiothreitol), shaken for 10 s, incubated in the 1·5 ml tube for 1 h at room temperature, centrifuged (16 000 g ) for 10 min and then diluted 1:3 in a solvent consisting of acetonitrile and water according to Bonfatti et al. (Reference Bonfatti, Grigoletto, Cecchinato, Gallo and Carnier2008). An Agilent 1100 Series (Agilent Technologies, Santa Clara, CA, USA) HPLC equipped with an analytical C8 column (Zorbax 300SB-C8 RP, 3·5 µm, 300 Å, 150 × 4·6 I.D., Agilent Technologies) preceded by a Security Guard Cartridge System (ZorbaxC4 4 × 3·0 mm2, Agilent Technologies) was used for separation. The gradient elution conditions according to Bonfatti et al. (Reference Bonfatti, Grigoletto, Cecchinato, Gallo and Carnier2008) included a flow rate of 0·5 ml/min. The total analysis time per sample was 45 min, with a column temperature maintained at 45 °C, a detection wavelength of 214 nm, and an injection volume of 10 µl. The protein composition was identified by comparing the elution times of the various chromatographic peaks with reported cow and buffalo proteins (Bobe et al. Reference Bobe, Beitz, Freeman and Lindberg1998; Bonfatti et al. Reference Bonfatti, Grigoletto, Cecchinato, Gallo and Carnier2008, Reference Bonfatti, Giantin, Rostellato, Dacasto and Carnier2013). The chemicals used in this experiment were purchased from Sigma-Aldrich (St. Louis, MO, USA).
Two-dimensional gel electrophoresis separation
Three random individual samples from each breed (a total of 12 milk samples) were used for two-dimensional gel electrophoresis (2-DE). The milk samples were centrifuged at 3000 g for 15 min at 4 °C, and the top fat layer and precipitate were removed (Yang et al. Reference Yang, Wang, Yuan, Bu, Yang, Sun and Zhou2013). The concentration of buffalo milk protein was determined by the Bradford protein assay kit (Bio-Rad Laboratories, Hercules, USA) with bovine serum albumin as standard according to the manufacturer's protocol. Buffalo milk protein samples (approximately 280 µg) were mixed with 350 µl of solubilisation buffer consisting of 8 m urea (GE HealthCare), 4% 3-[(3-cholamidopropyl)-dimethylammonio]-1-propane sulphonate (CHAPS) (GE HealthCare), 0·2% pH 4-7 carrier ampholytes (GE HealthCare), 6·5 mm dithiothreitol (Bio-Rad) and 0·001% bromophenol blue (Songon Biotech, Shanghai, China). The sample was placed on a 17 cm, pH 4-7, immobilised pH gradient (IPG) strip (Bio-Rad) for 12 h of passive rehydration at room temperature and subjected to isoelectric focusing using an Ettan IPG phor 3 (GE Healthcare). The programme including six steps was as following: 100 V, 1 h; 500 V, 1 h; 1000 V, 1 h; 4000 V, 1 h; 8000 V, 1 h; and 8000 V, 65 000 Vhr.
Focused strips were embedded with 0·5% agarose (GE HealthCare) on top of 12·5% polyacrylamide gels in an Ettan DALT II system (GE HealthCare). The gels were stained with Colloidal Coomassie Brilliant Blue G-250 (GE HealthCare) and destained in acetic acid. For the technical replication, three gels were run for an individual buffalo in each breed, and all sample gels were performed under identical conditions. Images were captured using an ImageScanner (GS800, Bio-Rad) and ImageMaster platinum 6·0 (GE HealthCare) to determine the protein spots of interest as reported by Goncalves Lda et al. (Reference Goncalves Lda, Soares, Nogueira, Garcia, Camisasca, Domont, Feitosa, Pereira Dde, Zingali and Alves2010). Briefly, the 2D gels of triplicate biological were compared each other to obtained a master gel. Every protein spot position, shape and optical density was averaged in each master gel. At least 5 well-defined landmarks were chose to match gels. The spots in other studied breeds exhibited a greater than two fold change compared to breed C which was considered as differentially expressed. Then the different expressed spots would be sent to a commercial company for MS analysis.
In-gel digestetion
The twenty-three spots were removed manually from the gel using pipette tips then placed into a 1·5 ml tube. After washed with Milli-Q water for 15 min, the gel fragments were destained with 100 mm ammonium bicarbonate (Sigma) in 30% acetonitrile (Merck, Darmstadt, Germany). Subsequently, the spots were digested with 5 µl of 10 ng/μl trypsin solution ((Promega, modified sequencing grade) and incubated overnight at 37 °C, and then stopped reaction with 100 µl of 60% acetonitrile in 0·1% trifluoroacetic acid (Merck). After sonication for 15 min, the supernatants were concentrated to near dryness.
Mass spectrometry analysis
Mass spectrometry (MS) experiments were carried out on a 4800 Plus MALDI TOF/TOF™ Analyser (Applied Biosystems, USA) and data acquisition in the positive ion mode with a selected mass range of 800 to 4000 Da. The resuspended sample in 20% acetonitrile was placed on the ABI 4800 target plate (384 Opti-TOF 123 × 81 mm2 ss, Applied Biosystems) and then 0·5 µl of the supersaturation CHCA matrix solution in 50% acetonitrile/0·1% trifluoroacetic acid was added. Eight of the most intense ions (a signal to noise ratio above 50) were performed for MS/MS analysis. Protein was identified of using the MASCOT software to search the NCBI database (http://www.ncbi.nlm.nih.gov, 2012-2-10). The following search parameters were used with trypsin as the enzyme, carbamidomethyl (C) as a fixed modification, mammalia as taxonomy, monoisotopic mass values, unrestricted protein mass, ±100 ppm peptide mass tolerance, ±0·4 Da fragment mass tolerance, and max missed cleavages 1.
Statistical analysis
The relative intensity of protein spots in the buffalo milk of different breeds were captured using the Quantity One software (v4·6·2, Bio-Rad). Statistical analysis was conducted by GLM procedure of SAS (SAS, V9·3) with Duncan's multiple range tests. All differences were considered statistically significant at P < 0·05.
Results and discussion
In this study, both reverse-phase high performance liquid chromatography (RP-HPLC) and a gel-based proteomic approach were used to investigate the content, composition and distribution of milk protein from the N, M, Nili-Ravi-Murrah crossbreeds (M-N) and crossbreeds of river buffalo with local swamp buffalo (C). A total of 23 proteins were identified. Some previous studies have reported milk proteomes similar to the buffalo in Holstein, Jersey and other bovine species (Hinz et al. Reference Hinz, O'Connor, Huppertz, Ross and Kelly2012; Yang et al. Reference Yang, Zheng, Zhao, Zhang, Han, Ma, Zhao, Li, Guo and Wang2015). In a previous work, a quantitative proteomic method performed on the Holstein, Jersey and buffalo milk fat globule membrane fractions revealed abundant κ-casein in the Jersey bovine milk and abundant β-casein and αs1-casein in the buffalo milk (Yang et al. Reference Yang, Zheng, Zhao, Zhang, Han, Ma, Zhao, Li, Guo and Wang2015). In another study, a higher abundance of κ-casein was found in the Bangladeshi buffalo than in cows’ milk (Islam et al. Reference Islam, Alam, Islam, Khan, Ekeberg, Rukke and Vegarud2014). In our study, high levels of κ-casein were discovered in buffalo breed C and confirmed by two analytical methods, indicating that through specific crossbreeding, buffalo breed C can yield milk with more κ-casein content that is high enough for improving cheese-making yield than other studied breeds. Thus, this finding suggests that this crossbreed may modify the pathway of milk protein synthesis in the buffalo mammary glands through activation of mammalian target of rapamycin (mTOR) signalling pathway and janus kinase 2 - signal transducer and activator of transcription 5 (Jak2-Stat5) signalling pathway (Bionaz & Loor, Reference Bionaz and Loor2011; Shimobayashi & Hall, Reference Shimobayashi and Hall2014). However, further studies are needed to explore the mechanism.
Nanda & Nakao (Reference Nanda and Nakao2003) reported that the milk nutrient content in crossbred buffalo were higher than the content measured in purebred river buffalo, and therefore, crossbreeding may improve milk production in addition to providing a higher nutrient content. In addition to the type of breeds, many other factors, such as forage, feeding systems, milking frequency and method, seasonal changes and lactation period, can affect the physicochemical parameters of buffalo milk (Khedkar et al. Reference Khedkar, Kalyankar, Deosarkar, Caballero, Finglas and Toldrá2016). In this study, all of the milk samples came from the same farm at the same time to avoid factors involving seasonality. However, future research into the differences in genetics and the translational efficiency of transcripts are still needed to provide methods for developing buffalo breeds with optimised milk protein production.
Protein composition
The protein composition of individual buffalo milk samples obtained by RP-HPLC is shown in Table 2 and Fig. 1. Significant differences (P < 0·05) in the total protein content were found among the four groups, with the crossbred group (C) having the highest protein percentage and N containing the lowest which was agreed with Ren et al. (Reference Ren, Zou, Lin, Chen, Liang and Liu2015). The total protein content did not have significant difference between C and M buffalo. The results in this study were consistent with the protein contents for purebred water buffalo, M (4·27%) and N (4·16%), and for the native swamp crossbred buffalo, C (4·75–5·23%), previously reported by Han et al. (Reference Han, Meng, Li, Yang, Ren, Zeng and Nout2007). Han et al. (Reference Han, Meng, Li, Yang, Ren, Zeng and Nout2007) showed that the protein level in the milk with each buffalo crossbreed and multi-generation crossbreed in their study were similar to the levels observed among the multi-generation crossbreeds in this study (4·75 ± 0·53% and 4·46 ± 0·48%, respectively), suggesting that crossbreeds have great potential to produce more milk protein than other buffalo.
Fig. 1. RP-HPLC chromatograms of individual samples of the four types of buffalo milk with different αS1- and κ-casein genetic variants (αS1-casein A or B, and κ-casein X1 or X2) obtained using the optimised elution conditions reported by Bonfatti et al. (Reference Bonfatti, Grigoletto, Cecchinato, Gallo and Carnier2008).
Table 2. Protein profiles of milk from four breeds of buffalo
Means with different superscripts in the same row differ significantly (P < 0·05)
M, Murrah; N, Nili-Ravi; M-N, Crossbreeds from Murrah and Nili-Ravi; C, Crossbreeds of river buffalo (Murrah and Nili-Ravi) with local swamp buffalo
† Total protein (%) was determined by infrared spectroscopy using a Milko Scan FT120, and protein composition (%) was determined by RP-HPLC
The differences (P < 0·05) were also observed in the α-lactalbumin and κ-casein content of the milk. The lowest α-lactalbumin content was found in the milk from the crossbred C buffalo (6·23%) and was significantly lower than the levels found in the milk from the N (7·60%) buffalo, while the α-lactalbumin content in M, N, M-N buffalo were similar. The amount of κ-casein in the milk was significantly higher in the crossbred buffalo C (11·14%) than in the other buffalo. The result that crossbred buffalo C had higher κ-casein content than M buffalo was consistent with Ren et al. (Reference Ren, Zou, Lin, Chen, Liang and Liu2015). However, Ren et al. (Reference Ren, Zou, Lin, Chen, Liang and Liu2015) found that there was no significant difference in κ-casein content between crossbred buffalo C and N buffalo which may be due to the more amount of buffalo milk used in this study. The content of κ-casein is an important factor in cheese making, as it can influence the milk coagulation ability, cheese yield and quality (Comin et al. Reference Comin, Cassandro, Chessa, Ojala, Dal Zotto, De Marchi, Carnier, Gallo, Pagnacco and Bittante2008; Ren et al. Reference Ren, Chen, Chen, Miao and Liu2013; Sanchez-Macias et al. Reference Sanchez-Macias, Morales-delaNuez, Torres, Hernandez-Castellano, Jimenez-Flores, Castro and Arguello2013). In the future, it may be possible to produce high quality cheese by specifically crossbreeding to yield milk with high κ-casein content.
As shown in Fig. 1, κ-casein consisted of several partially co-eluting peaks at 10 and 16 min that can be ascribed to the presence of different glycosylated and phosphorylation forms of this protein (Bonfatti et al. Reference Bonfatti, Giantin, Rostellato, Dacasto and Carnier2013). The separation of buffalo protein has been achieved by several studies, but only a few of them detected the buffalo κ-casein polymorphism (Chianese et al. Reference Chianese, Quarto, Pizzolongo, Calabrese, Caira, Mauriello, De Pascale and Addeo2009; Feligini et al. Reference Feligini, Bonizzi, Buffoni, Cosenza and Ramunno2009; Bonfatti et al. Reference Bonfatti, Giantin, Rostellato, Dacasto and Carnier2013). Similarly, in the case of αs1-casein, two alternative forms were detected, and all of the samples gave rise to a double peak, which can be ascribed to the varying degrees of post-translational phosphorylation (Addeo et al. Reference Addeo, Mercier and Ribadeau-Dumas1977).
The expression profile of buffalo milk protein
The protein expression profiles of the four types of milk are shown in Fig. 2, and the changes in the protein spot quantity values are summarised in Fig. 3. The identification of the proteins contained in the spots is listed in Table 3. As shown in Fig. 2, a total of 23 differentially expressed spots were performed for MS analysis.
Fig. 2. Two-dimensional electrophoresis of the four types of buffalo milk. C, crossbreeds of river buffalo (Murrah and Nili-Ravi) with local swamp buffalo; M-N ,crossbreeds from Murrah and Nili-Ravi; M, Murrah; N, Nili-Ravi.
Fig. 3. Densitometric values of protein spots (numbered in Fig. 2) detected in the milk of the four studied breeds of buffalo. The bars indicate the standard error mean (n = 3). Means with different letters differ significantly (P < 0·05). ND, Non-detectable; PIGR, polymeric immunoglobulin receptor; C, crossbreeds of river buffalo (Murrah and Nili-Ravi) with local swamp buffalo; M-N, crossbreeds from Murrah and Nili-Ravi; M, Murrah; N, Nili-Ravi.
Table 3. List of selected milk protein spots identified in Fig. 2 from two-dimensional electrophoresis gels and identified using MALDI-TOF-MS/MS
PIGR, Polymeric immunoglobulin receptor;
gi, GenInfo Identifier
Polymeric immunoglobulin receptor
Protein spot 1 is a polymeric immunoglobulin receptor (PIGR), with a molecular weight of 83·71 kDa that was found only in the C buffalo (Table 3; Fig. 3). PIGR is an important member of the major histocompatibility complex family and immunoglobulin superfamily (Kaetzel, Reference Kaetzel2005), which can be expressed in mammary gland cells (De Groot et al. Reference De Groot, Van Kuik-Romeijn, Lee and De Boer2000) and in human and macaque milk (Beck et al. Reference Beck, Weber, Phinney, Smilowitz, Hinde, Lonnerdal, Korf and Lemay2015). PIGR plays an important role in mucosal immunity by translocating IgA and IgM into external body fluids (Kaetzel, Reference Kaetzel2005). Furthermore, the amount of PIGR produced is a limiting factor in the transport of IgA into the milk under normal non-inflammatory circumstances (De Groot et al. Reference De Groot, Van Kuik-Romeijn, Lee and De Boer2000). So PIGR could play a part in the humoral immunity through the transport of IgA and IgM then in turn preventing microorganisms and foreign proteins from penetrating the mucosal surfaces (Monteiro & Van De Winkel, Reference Monteiro and Van De Winkel2003). Therefore, the relatively high abundance of PIGR in C buffalo milk may indicate that this could provide more immunological protection and even the development of immune function in humans.
κ-casein
The eight protein spots shown in Table 3 correspond to κ-casein, and five of these spots—2, 3 (which is in a different configuration with 4 and 5), 7, 8 and 23—were observed in high abundance in buffalo breed C and there were no significant difference between other buffalos. Although the κ-casein in spot 9 was upregulated in breed N compared with other breeds, this upregulation amount was not enough to affect the total content of κ-casein in the milk of breed C. This may be partly consistent with the high content of κ-casein in breed C (Table 2), suggesting that the crossbreed may modify the pathway of milk protein synthesis in the buffalo mammary glands. Many spots corresponding to the phosphorylated, dephosphorylated, phosphorylated-glycosylated, phosphorylated-diglycosylated and dephosphorylated-diglycosylated forms of κ-casein were found in the buffalo milk (D'Ambrosio et al. Reference D'Ambrosio, Arena, Salzano, Renzone, Ledda and Scaloni2008). Additionally, amino acid replacements may be another reason for the two variants of κ-casein that were found in the buffalo, as shown in Fig. 1. Similar results were observed in Mediterranean water buffalo, where κ-casein X1 and X2 were identified with the amino acid Ile135 changed to Thr135 in the mature polypeptide chain (Bonfatti et al. Reference Bonfatti, Giantin, Rostellato, Dacasto and Carnier2013). Casein composition and structure, especially κ-casein content and protein phosphorylation, are important to cheese making (Ageitos et al. Reference Ageitos, Vallejo, Poza and Villa2006; Ren et al. Reference Ren, Chen, Chen, Miao and Liu2013) and the antioxidant activity of yogurts (Perna et al. Reference Perna, Intaglietta, Simonetti and Gambacorta2013), primarily because of the role of κ-casein in milk aggregation and the relationship of Ca2+-sensitive proteins in the formation of casein micelles. These properties imply that modulating the casein content in varieties of milk will have an effect on any resulting dairy product, including cheese or yogurt.
β-casein
Additionally, eight spots related to β-casein were identified (spots 6, 10, 12, 13, 14, 19, 20 and 22) (Table 3; Fig. 3). The β-casein identified in spot 14 for buffalo breeds M-N, N and M, however, appeared in different configurations for breed C, corresponding to spots 12 and 13. This difference may be attributed to milk protein polymorphism. Bonfatti et al. (Reference Bonfatti, Giantin, Rostellato, Dacasto and Carnier2013) showed that analytical methods employing RP-HPLC could not separate the β-casein polymorphisms. The reason for this inability to separate the proteins may be due to their similar molecular weights and isoelectric points, which we also observed in the β-casein variants contained within spots 12 and 13. The different breeds and geographic locations may influence this as well. The expression of β-casein in spots 6 and 19 was most abundant in breed M-N, whereas the levels of β-casein expression in spots 10 and 14 were the most abundant in breed C buffalo than other buffaloes. This finding may explain the no effect of the breed on total β-casein (Table 2). The expression of β-casein may have a compensation effect between the variants.
Some researchers reported that more than one genetic variant of β-casein has been identified in Italian buffalo (Ferranti et al. Reference Ferranti, Scaloni, Caira, Chianese, Malorni and Addeo1998), but only one variant was found in Venezuelan and Mediterranean water buffalo (Ferranti et al. Reference Ferranti, Scaloni, Caira, Chianese, Malorni and Addeo1998; Bonfatti et al. Reference Bonfatti, Giantin, Rostellato, Dacasto and Carnier2013). Two-dimensional electrophoresis is a powerful tool that may aid the identification of β-casein polymorphisms, where four allele genes (A1, A2, B and I) and 10 variants of β-casein were found in Holstein cow milk by 2-DE (Jensen et al. Reference Jensen, Holland, Poulsen and Larsen2012). Compared to κ-casein, few spots corresponding to β-casein were identified during the course of this study. This difference may be attributed to the lack of a putative phosphorylation site in the β-casein sequence, which leads to a reduced degree of phosphorylation in buffalo β-casein compared with other ruminants (Mamone et al. Reference Mamone, Caira, Garro, Nicolai, Ferranti, Picariello, Malorni, Chianese and Addeo2003; Farrell et al. Reference Farrell, Jimenez-Flores, Bleck, Brown, Butler, Creamer, Hicks, Hollar, Ng-Kwai-Hang and Swaisgood2004), thereby resulting in fewer variants of the protein to detect.
αs1-casein
The four spots for αs1-casein are shown in Fig. 2. The spot containing αs1-casein in M-N (spot 18) corresponded to different spots (spot 16 and 17) in crossbreed C. The differences in the proteins were also evident from the RP-HPLC analysis, as shown in Fig. 1, and may also be due to protein polymorphism. These results are similar to those reported by Bonfatti et al. (Reference Bonfatti, Giantin, Rostellato, Dacasto and Carnier2013), who discovered two variants of αs1-casein. This polymorphism is the result of the substitution of amino acid Leu178 in variant A to Ser178 in variant B of the mature polypeptide chain. This has also been previously reported in Mediterranean water buffalo and river buffalo, which are similar to breeds M and N (Ferranti et al. Reference Ferranti, Scaloni, Caira, Chianese, Malorni and Addeo1998; Chianese et al. Reference Chianese, Quarto, Pizzolongo, Calabrese, Caira, Mauriello, De Pascale and Addeo2009). Compared to other ruminants, the αs1-casein protein in buffalo exhibits reduced phosphorylation activity because it lacks a series of phosphorylation sites that are present in bovine and ovine proteins (Farrell et al. Reference Farrell, Jimenez-Flores, Bleck, Brown, Butler, Creamer, Hicks, Hollar, Ng-Kwai-Hang and Swaisgood2004). In this study, αs1-casein of breed C in spot 15 has the lowest expression, while αs1-casein in spot 18 has the highest expression. No significant difference was found between breeds M, N and M-N buffalo. Likewise, no significant difference was found in the expression of total αs1-casein, which may be the result of a compensation effect between the variants (spot 15 and 18).
αs2-casein
Only two spots corresponding to αs2-casein (spot 11 and 21) were found, which are shown in Fig. 2, and both of them were identified as αs2-casein. No polymorphism has previously been described for αs2-casein in the Mediterranean water buffalo as analysed by RP-HPLC (Ferranti et al. Reference Ferranti, Scaloni, Caira, Chianese, Malorni and Addeo1998; Feligini et al. Reference Feligini, Bonizzi, Buffoni, Cosenza and Ramunno2009). The expression of αs2-casein in spot 11 was more abundant in breed M-N than M and C, which is consistent with the relatively high content of αs2-casein in M-N (Table 2). Little is known about the relatively high expression of αs2-casein in buffalo breed M-N, and further studies are needed to explore this.
β-lactoglobulin and α-lactalbumin
As for αs2-casein, unique peaks for β-lactoglobulin were expected (Fig. 1). Similar results have been shown in the analysis of milk from Mediterranean water buffalo; however, two variants of β-lactoglobulin have been detected in Indian and Egyptian buffalo populations (Patel et al. Reference Patel, Chauhan, Singh and Soni2007). Similar to previous research studies (Chianese et al. Reference Chianese, Caira, Lilla, Pizzolongo, Ferranti, Pugliano and Addeo2004), we also found two variants of α-lactalbumin (B: Asn45 changed to A: Asp45).
In conclusion, this study found that the C buffalo has higher milk protein content, κ-casein content, and lower α-lactalbumin content than the other three breeds studied. And the distribution differences of milk casein are attributed to milk protein polymorphism in the various crossbreeds. Therefore, understanding how crossbreeding can influence the protein content and composition will provide insight into breeding buffalo with higher total protein and κ-casein contents in their milk and potentially provide a method for controlling quality in cheese production applications.
This work was supported by the National Science Foundation of China (Grant No. 31260384 and 31501504), the Opening Project of Guangxi Key Laboratory of Buffalo Genetics, the Reproduction and Breeding (SNKF-2014-02) and the Fundamental Research Funds for the Central Universities (2016FZA6016). The authors thank Dr Moushumi Paul, the Dairy and Functional Foods Research Unit, the Eastern Regional Research Center, the ARS and the USDA for their useful advice and linguistic help.
