Cheese yield, casein fractions and major components of milk of Saanen and Anglo-Nubian dairy goats

Rendimento de produção de queijos, frações de caseína e principais componentes do leite de cabras Saanen e Anglo-Nubiana

goats; milk; casein; cheese yield

cabras; leite; caseína; queijo

COMMUNICATION COMUNICAÇÃO

Cheese yield, casein fractions and major components of milk of Saanen and Anglo-Nubian dairy goats

Rendimento de produção de queijos, frações de caseína e principais componentes do leite de cabras Saanen e Anglo-Nubiana

J.P. Damián^{I, II}; I. Sacchi^I; S. Reginensi^II; D. De Lima^I; J. Bermúdez^II

^IFacultad de Veterinaria - UDELAR, Lasplaces, 1550 - CP 11600, Montevideo - Uruguay

^IIFacultad de Agronomía - UDELAR - Montevideo - Uruguay

Keywords: goats, milk, casein, cheese yield

RESUMO

Avaliaram-se os componentes, as frações de caseína e o rendimento na produção de queijo do leite de cabras das raças Saanen e Anglo-Nubiana, principais raças criadas no Uruguai. O estudo foi realizado em uma fazenda com sistema de criação semi-intensiva. O leite das cabras Anglo-Nubianas apresentou teores mais elevados de gordura (4,65 vs 3,59%), proteína total (3,48 vs 2,84%), caseína total (2,82 vs 2,23%), e maior rendimento na produção de queijos (22,00 vs 15,03kg/100l) que o leite das cabras Saanen. As fraçoes de αs1-caseína (6,99 vs 2,37g/l), β-caseína (13,95 vs 12,75g/l) e κ-caseína (4,24 vs 3,64g/l) também foram mais elevadas no leite das cabras Anglo-Nubianas, porém no teor de αs2-caseína (3,02 vs 3,60g/l) não se observaram diferenças. O rendimento na produção de queijos foi significativamente correlacionado com os teores de gordura, proteína, caseína total e com as frações de caseína. A α_s1-caseína representou a fração com maior correlação com os teores de proteína total, caseína, gordura e produção de queijo. Demonstrou-se, neste estudo, que o leite de cabras Anglo-Nubianas é mais indicado para a produção de queijo e o de cabras Saanen para utilização como leite fluido.

Palavras-chave: cabras, leite, caseína, queijo

Over the last two decades, goat production had significant increments around the world, particularly in developing countries (Boyazoglu and Morand-Fehr, 2001). Most of the goat milk production is used for small-scale artisan cheesemaking. Cheese properties, quality, and yield depend on milk composition, especially protein, casein, and fat content (Storry et al., 1983; Ambrosoli et al., 1988). Milk composition is also influenced by breed, age, stage of lactation, lactation number, month of sampling, nutrition, and environmental and genetic factors (Storry et al., 1983). Breed has been frequently reported in the literature as one of the main variables affecting goat milk composition (Sung et al., 1999; Clark and Sherbon, 2000), including casein fractions (Clark and Sherbon, 2000; Moatsou et al., 2004). The proportion of the different casein fractions, particularly the αs1-CN content, have been shown to influence coagulation properties, cheese yield, and protein content (Ambrosoli et al., 1988; Grosclaude et al., 1994; Clark and Sherbon, 2000).

Although there are studies that report the effects of breed on goat milk composition, researches in Latin America goat population are scarce and reports of breed effect (Saanen vs Anglo-Nubian) on casein fractions, milk composition, and cheese yield from goat milk under a semi-intensive production system were not found. Saanen and Anglo Nubian breeds are among the most common goats raised in Uruguay. Two reports compared milk composition of Saanen and Anglo-Nubian breeds (Clark and Sherbon, 2000; Sung et al., 1999) and only the former studied the αs1-CN fraction in the same breed. Clark and Sherbon (2000) did not found a breed effect (Saanen vs. Anglo-Nubian) on αs1-CN and total casein content, neither in coagulation properties. That study did not determine cheese yield or casein fractions. Thus, the aim of this study was to evaluate the individual milk components (fat, total protein, casein, and lactose), casein fractions (α_s1-CN, α_s2-CN, β-CN, and κ-CN) of two pure goat milk breeds, frequently raised in commercial Uruguayan dairy farms, and estimate the relationship between goat milk composition and individual laboratory cheese yield (ILCY).

Second lactation pure Saanen (n=11) and Anglo-Nubian (n=11) healthy goats, and raised under the same semi-intensive system and conditions, were selected from a commercial dairy farm in Canelones, Uruguay. All animals were fed pastures composed of lotus, clover, and rye grass. They also received a concetrate with 16% crude protein during milking time. Individual milk samples (100ml) from cows at mid-lactation (November) (120±10d) were collected on two consecutive days (morning and evening milkings) before milking. Samples collected from each goat were carefully stirred and pooled as a proportion of milk production in each milking. Half of the samples was preserved in 0.05% potassium dichromate and kept at 4ºC for casein fractions and chemical composition analysis. The other half (aseptically collected) was used for chesse making and bacteriological analysis (Moroni et al., 2005). For the later, milk samples were plated on blood agar and incubated for 24-48h at 37ºC. Pathogens microrganisms were not found. No clinical signs of intramammary infections were observed for any of the goats throughout the study, hence, no mastitis therapy was administered.

Total nitrogen (TN) and pH 4.6 soluble nitrogen (SN) were analyzed by the Kjeldahl method (Nx6.38). Casein nitrogen (CN) was calculated as: CN=TN-SN (Official..., 1991). Fat and lactose contents were obtained by using a Milkoscan^® apparatus, according to the IDF (Whole milk..., 1996). Whole milk from each animal was skimmed by centrifugation at 2500xg for 30min at 4ºC. Casein was obtained by isoelectric precipitation (3.3M Na-acetate buffer, pH 4.6) from skimmed milk. The precipitate was centrifuged at 500xg for 20min and further resuspended in distilled water for washing. The casein was precipitated once again by centrifugation, and washed with water three times. Casein samples of individual goat milk were obtained by dissolving one part (weight) of the casein precipitate with four parts (volume) of buffer (17.5mM 1,3-bis(tris)(hydroxymethyl-methylamino-propano) at pH 7.4 with 7M urea and 0.3% 2-mercaptoethanol. After standing at room temperature for 1 h, each sample was filtered through a 0.2μm pore size filter (cellulose acetate¹) and 40μλ were injected for analysis by reverse phase high performance liquid chromatography (RP-HPLC), as described by Trujillo et al. (2000), which allowed to separate and quantify the relative amounts of the four major casein fractions (α_s1-CN, α_s2-CN, β-CN, and κ-CN).

Casein standards² were used for identifying casein fractions by RP-HPLC. Goat casein, 90% purity (aprox.), determined by the method of biuret, and purified α-, β, and κ-CN fractions² were used. The fractions α_s1-CN and α_s2-CN were previously purificated ð as described by Igarashi (1999). Casein chromatography was performed on an Apex WP ODS reverse-phase column³ (7μm, 4.6x250mm) at 46ºC. Two solvents were used for the mobile phase as follows: A - acetonitrile-water-trifluoroacetic acid (100-900-1 v/v/v) and B - the same mixture with the proportions 900-100-0.7 (v/v/v). Proteins were eluted with a series of linear gradients: from 28.7 to 42.5% of solvent B in solvent A for 15min, 42.5 to 48.8% solvent B for 15min, 48.8 to 28.7% solvent B for 15min, finishing with 28.7% of solvent B for 15min in order to column reequilibration. Flow rate was 1ml/min and the eluted peaks were detected by UV-absorbance at 214nm. Quantification of the casein fractions was made by measurement of the peak area for each fraction as a proportion of the whole casein peak area. The concentration of casein fractions was expressed in g/l or its relative percentage of total casein (%).

The ILCY was determined in duplicate for each individual milk sample as described by Othmane et al. (2002). The pooled samples were previously equilibrated at room temperature for 30min before rennet addition. Ten ml of each preheated and homogenized sample plus 2¼l of liquid bovine rennet⁴ were curdled for 1h at 37ºC. The rennet was diluted at 1:20 with double distilled water before addition to the milk samples. The coagulum was cut (in way of cross with a spatula) and centrifuged for 25min at 2500rpm. The whey was decanted by inclining the test tube and the residue was weighed. ILCY was defined as the weight of the centrifuge residue expressed in kg/100 liters of milk. Milk composition, casein fractions, and ILCY were analyzed in a completely randomized design using the General Linear Model (GLM) procedure of the statistical software SAS (User's..., 2000). The statistical model included the effect of breed. Pearson´s correlation coefficients were determined using the Corr Procedure.

Major milk components and casein fractions of Anglo-Nubian and Saanen dairy goats are shown in Table 1. Contents of fat, total protein and casein were significantly higher in Anglo-Nubian than Saanen milk (P<0.01). No difference was found for lactose content among the breeds. The contents of fat and protein confirm previous reports on these breeds studied under other systems of production, however, contents of lactose and total casein were different (Sung et al., 1999; Clark and Sherbon, 2000). Sung et al. (1999) worked with second or third lactation Anglo-Nubian (n=9) and Saanen (n=12) goats, reared under intensive husbandry conditions, and found greater levels of protein and fat and lower lactose content in Anglo-Nubian goat milk. Clark and Sherbon (2000) did not find differences in total casein content between breeds (Saanen vs. Anglo-Nubian), but found differences in protein and fat contents. In addition, coagulation properties: coagulation time, coagulation rate, and curd firmness were not significantly different between the breeds. On the other hand, cheese yield was not affected. However, the comparison between the present results and those reported by Clark and Sherbon (2000) is difficult because those authors did not inform the type of management or the lactation number. The present study was performed using a small number of animals (Nubian n=11 and Saanen n=11), but all of them were in the second lactation, kept under the same condition, and on the same farm.

Figure 1 shows chromatographic pattern of αs1-CN and whole casein of Saanen and Anglo-Nubian milk by RP-HPLC. Table 2 shows the coefficients of correlation between milk components, casein fractions, and individual laboratory cheese yield. The αs1-CN content was the best correlated fraction with contents of total protein, total casein, fat, and ILCY in both expressions (g/l and %). ILCY was significantly (P<0.002) correlated with contents of fat (r=0.62), total protein (r=0.80), and casein (r=0.83) in goat milk. These results show a strong positive correlation between ILCY and contents of fat, total protein, and particularly, casein; which, in turn, affect cheese yield during cheesemaking. Othmane et al. (2002) worked with dairy sheep and found a high correlation between ILCY and contents of total protein (r=0.91) and fat (r=0.76). Results obtained for ILCY with a small number of goat milk samples (n=11) for breed have shown that this method could be sufficiently suitable to be applied in selection programs in order to optimize goat milk cheese yield.

Figure 2 shows the values of casein fractions (g/l) and ILCY in both breeds. ILCY for Anglo-Nubian milk was higher than for Saanen milk, as expected from the milk composition data, particularly the differences in casein content between the breeds. Cheese yield was 48% higher for Anglo-Nubian milk using the ILCY test. The higher content of αs1-CN in Anglo-Nubian milk was directly related to a higher yield of cheese when compared with Saanen milk. The αs1-CN fraction showed the highest differences between both breeds with nearly a three fold increase of its content in Anglo-Nubian milk. The β-CN and κ-CN fractions were also higher in Anglo-Nubian than Saanen milk. However, no significant difference was found in the content of αs2-CN between breeds. Expressed in relative percentages in referring to total casein (Table 1), Anglo-Nubian showed higher levels of αs1-CN, but less β-CN and αs2-CN than Saanen milk; moreover, no difference was found between the breeds for κ-CN. The lower percentage of αs1-CN content and the higher percentage of β-CN in Saanen milk agree with the findings of Moatsou et al. (2004) and Tomokate et al. (2006). Also, the αs1-CN shows the highest variability among the casein fractions of both breeds. This paper shows the content of casein fractions expressed in two different ways for more complete information, as some papers show levels of casein fractions in g/l (Mora-Gutierrez et al., 1991; Clark and Sherbon, 2000) or in relative percentage (Moatsou et al., 2004) between breeds, making it difficult to compare different studies.

The fraction of casein that showed the same difference between breeds in both expressions (g/l and %) was αs1-CN, while αs2-CN, β-CN, and κ-CN varied depending on their expression. The concentration of αs1-CN (g/l) or its percentage of total casein (%) was higher for Anglo-Nubian milk. This result was different from that reported by Clark and Sherbon (2000), who did not find differences in αs1-CN content between the same breeds, although Anglo-Nubian have shown differences in αs1-CN content when compared with other Swiss goat breeds (Toggenburg and Alpine). Besides the relative frequency of the alleles for αs1-CN, this casein fraction showed a strong difference between the breeds (Grosclaude et al., 1994; Clark and Sherbon, 2000; Moatsou et al., 2004). The variants of αs1-CN were associated with four levels of expression defining four quantitative classes named: "high", "medium", "low", and "null" alleles related to different amounts of synthesis of αs1-CN (Grosclaude et al., 1994). Differences found in this work in αs1-CN could be attributed to the predominance of "low" and "medium" alleles in Saanen goats, whereas Anglo-Nubian goats may have the predominance of "high" alleles for this casein fraction in the goat population.

Milk components (fat, total protein, and casein) and ILCY were positive and significantly correlated with αs1-CN content (g/l and %), and the relationships were higher than those from the other casein fractions. These findings agree with those reported by Ambrosoli et al. (1988) and Clark and Sherbon (2000), who found that a higher content of αs1-CN in goat milk is directly related to the contents of total solids, total protein, casein, and fat. The correlations among milk components and ILCY with αs2-CN, β-CN, and κ-CN were also variable depending on their expression (g/l or %). Bevilacqua et al. (2001) reported that goat milk with low αs1-CN content was less allergenic than goat milk with higher content of this fraction.

The high variability found between goat breeds and also within the same breed could be used in future selection plans targeting different uses of milk and value-added-products. Roncada et al. (2002) suggested that animals with "medium" alleles can be used to produce milk for allergic people while other genetic variants could be used to produce milk for cheesemaking dairy plants. In this study, Anglo-Nubian goats probably have "high" alleles for αs1-CN. Furthermore, the milk composition of this breed and the results of the ILCY study indicate advantages of this milk as a raw material for cheesemaking. On the other hand, higher milk production of Saanen goats could compensate the lower total solids production (Potter, 1996). In addition, less αs1-CN content suggests advantages of this breed for fluid milk sale, which may be employed for allergic subjects (Roncada et al., 2002; Tomotake et al., 2006).

In conclusion, casein fractions analysis indicated a wide variability of αs1-CN with a nearly three-fold increase in Anglo-Nubian milk compared to Saanen milk, which could explain the higher cheese yield and its better performance for cheesemaking. Saanen milk had a lower αs1-CN content, which suggests advantages of this breed for fluid milk sale.

ACKNOWLEDGEMENTS

This work was supported by the Comisión Sectorial de Investigación Científica (UDELAR, Uruguay). We thank José Mosquera from the farm "Los Tajamares", Canelones-Uruguay. We also thank Dr. Pedro Diaz and Mariana Sotelo for RP-HPLC assistance (Departmento de Bioquímica), Facultad de Agronomia, (UDELAR, Uruguay), and Dr. Paul Savello for critical review.

REFERENCES

AMBROSOLI, R.; Di STASIO, L.; MAZZOCO, P. Content of α_s1-casein and coagulation properties in goat milk. J. Dairy Sci., v.71, p.24-28, 1988.

BEVILACQUA, C.; MARTIN, P.; CANDAHL, C. et al. Goat´s milk of defective α_s1-casein genotype increases intestinal and systematic sensitization to β-Lactoglobulin in guinea pigs. J. Dairy Res., v.68, p.217-227, 2001.

BOYAZOGLU, J.; MORAND-FEHR, P. Mediterranean dairy sheep and goat products and their quality A critical review. Small Rum. Res., v.40, p.1-11, 2001.

CLARK, S.; SHERBON, J.W. Genetic variants of alpha_s1-CN in goat milk: breed distribution and associations with milk composition and coagulation properties. Small Rum. Res., v.38, p.135-143, 2000.

GROSCLAUDE, F.; RICORDEAU, G.; MARTIN, P. et al. Du gène au fromage: le polymorphisme de la caséine α_s1 caprine, ses effects, son évolution. (From the gene to the cheese: goat α_s1-casein polymorphism, its effects, its evolution). INRA Prod. Anim., v.7, p.3-19, 1994.

IGARASHI, Y. Separation of caseins by chemical procedures. Int. Dairy J., v.9, p.377-378, 1999.

MOATSOU, G.; SAMOLADA, M.; PANAGIOTOU, P. et al. Casein fraction of bulk milks from different caprine breeds. Food Chem., v.87, p.75-81, 2004.

MORA-GUTIERREZ, A.; KUMOSINSKI, T.F.; FARELL, H.M. Quantification of α_s1-casein in goat milk from French-Alpine and Anglo-Nubian breeds using reversed-phase high performance liquid chromatography. J. Dairy Sci., v.74, p.3303-3307, 1991.

MORONI, P.; PISONI, G.; Ruffo, G. et al. Risk factors for intramammary infections and relationship with somatic cell counts in Italian dairy goats. Prev. Vet. Med., v.69, p.163-173, 2005.

OFFICIAL methods of analysis, 15.ed. Wasghington, DC: AOAC, 1991.

OTHMANE, H.; DE LA FUENTE, L.F.; CARRIEDO, J.A. et al. Heritability and genetic correlations of test day milk yield and composition, individual laboratory cheese yield, and somatic cell count for dairy ewes. J. Dairy Sci., v.85, p.2692-2698, 2002.

POTTER, V. Goats of the world. 1.ed. Ipswich, UK: Farming Press, 1996.

RONCADA, P.; GAVIRAGHI, A.; LIBERATORI, S. et al. Identification of caseins in goat milk. Proteomics, v.2, p.723-726, 2002.

STORRY, J.E.; GRANDISON, A.S.; MILLARD, D. et al. Chemical composition and coagulating properties of renneted milks from different breeds and species of ruminants. J. Dairy Res., v.50, p.215-229, 1983.

SUNG, Y.Y.; WU, T.I.; WANG, P.H. Evaluation of Milk quality of Alpine, Nubian, Saanen and Toggenburg breeds in Taiwan. Small Ruminant. Res., v.33, p.17-23, 1999. TOMOKATE, H.; OKUYAMA, R.; KATAGIRI, M. et al. Comparison between Holstein cow´s milk and Japanese-Saanen goat´s milk in fatty acid composition, lipid digestibility and protein profile. Biosci. Biotechnol. Biochem., v.70, p.2771-2774, 2006

TRUJILLO, A.J.; CASSALS, I.; GUAMIS, B. Analysis of mayor caprine milk proteins by reverse-phase high-performance liquid chromatography and electrospray ionization-mass spectrometry. J. Dairy Sci., v.83, p.11-19, 2000.

USER´S guide: statistics. Cary, NC: SAS Institute, 2000.

WHOLE MILK: Determination of milk fat, protein and lactose contents. Guide for the operation of mid-infra-red instruments. Brussels: International Dairy Federation, 1996. (Doc. 141B).

Recebido em 29 de julho de 2008

Aceito em 25 de novembro de 2008

E-mail: [e-mail]

1

Sartorius AG - Germany.

2

Sigma Chemical Co. - St. Louis, USA.

3

Jones Chromatography Ltd. - Mid-Glamorgan, UK.

4

Pronacu Ltda - Montevideo, Uruguay.

Publication Dates

• Publication in this collection
  09 Feb 2009
• Date of issue
  Dec 2008

History

• Received
  29 July 2008
• Accepted
  25 Nov 2008