You are on page 1of 11

Use of Different Dietary Protein Sources for Lactating Goats: Milk Production and Composition as Functions of Protein Degradability

and Amino Acid Composition 1
´ n Experimental del Zaidı *Estacio ´n, Consejo Superior de Investigaciones Cientı ´ficas, ´ n Animal, Departamento de Nutricio Profesor Albareda, 1.18008 Granada, Spain †Facultad de Farmacia, Universidad Complutense, Departamento de Bromatologı ´a, 28040 Madrid, Spain ‡Abbott Laboratories, 18004 Granada, Spain

ABSTRACT To establish the effect of the nature of four different protein sources [fababeans, 27.8% crude protein (CP); sunflower meal, 41.7% CP; corn gluten feed, 18.8% CP; and cottonseed, 18.3% CP] on milk protein production by goats, the ruminal degradation of these feeds was studied as was the amino acid ( A A ) composition of the original material and that of the undegradable fractions of the protein sources. Four diets were designed; 20% of their protein was supplied by each of the different sources. Four groups of 5 Granadina goats were used to study the utilization of these diets for milk production. No significant differences were observed in dry matter intake or milk production. The milk produced by goats fed the diet containing sunflower meal had the lowest protein concentration; the highest milk protein concentration was observed for goats fed the diet containing corn gluten feed. From a multivariate analysis, it was deduced that the quickly degradable protein fraction in the rumen and the ruminally undegradable protein fraction were the components of the protein sources most directly related to the milk protein produced. Given the similar AA profiles of the undegradable fractions of the different protein sources, the possible supplementation achieved from these ruminally undegradable fractions must be established by the amount of protein supplied regardless of AA composition. ( Key words: protein source, degradability, milk production, lactating goats)

Abbreviation key: CGF = corn gluten feed, CS = cottonseed, EAA = essential AA, FB = fababeans, SFM = sunflower meal. INTRODUCTION The systems currently in use to estimate the protein requirements of ruminant animals distinguish between the dietary protein fraction, which after ruminal degradation yields microbial protein, and the fraction that escapes ruminal fermentation and reaches the small intestine (1, 20). Since the introduction of these new systems, there has been much research (5, 27) to determine the effect of protein degradability on milk production and composition. All of these studies have attempted to optimize milk protein production, but the results have not always been predictable. Different researchers (5, 15, 26, 27, 31) have pointed out the inadequacy of using only the intake of RDP and RUP to formulate diets for lactating ruminants. In their opinions, it would be necessary to take into account the kinetics of ruminal degradation of the protein supplied and the amino acid profile of the RUP fraction of the protein supplied. The limited amount of research to date carried out in goats has only analyzed the effect of using less degradable protein sources instead of soya protein. Very few and even contradictory results have been obtained (11, 19). Morand-Fehr et al. ( 1 9 ) have stressed that in the majority of cases in which isonitrogenous and isoenergetic diets were used, the total protein and casein contents of goat milk did not appear to be very sensitive to changes in dietary protein source. Because of the scarce information available and because practically all of the goat milk produced in Spain is destined for cheese production, we analyzed the effect of using different protein

Received March 6, 1998. Accepted September 14, 1998. 1This study was supported financially by the Interministerial Commision of Science and Technology, Madrid, Spain (Project AGF93-0096 and ALI96-1024-C02-01). 1999 J Dairy Sci 82:555–565


Measurements and Analyses Every day of the principal trial period. the goats were weighed on the 1st. 40 CGF (g/kg) 360 200 .. Degradability and AA Composition of the Different Protein Sources The ruminal degradability of the N in each protein supplement was estimated using the nylon bag technique according to an international standard procedure (16).5 × 10 cm. the goats were hand-milked.12 g of Cu. The diet fed to the cannulated wethers was that proposed by Ørskov and McDonald ( 2 1 ) and Isac et al. 15th. which were previously milled through a 2. 3. and the last 4 d constituted the principal trial period.556 SANZ SAMPELAYO ET AL. One of the assays analyzed the characteristics of the ruminal degradation of each protein source. 520 IU of vitamin E. Treatments consisted of four diets. sunflower meal ( SFM) . corn gluten feed (CGF). 48. 0. The diet consisted of a good quality alfalfa hay with a mineral and vitamin supplement. 0. 0.08 million IU of vitamin D3. The amount of sample per bag was 2 to 3 g.. 40 CS 360 200 . Then. 2. The first 15 d of the experimental period were for adaptation. the goats consumed the experimental diets for 1 mo.. No. 2Contained per kilogram: 2. 1.. the bags were washed in a washing machine and were treated in a stomacher for 5 min to remove microbial contamination (16). Two wethers with permanent ruminal cannulas were used for incubations of 0. the daily rations were distributed. 200 . Before the experiment. sunflower meal (SFM). The four diets were similar in terms of N and gross energy contents. 0.92 g of Fe. once the orts from the ration that was offered the previous day had been collected. The ingredient composition of the concentrates and the chemical composition of diets are shown in Tables 1 and 2. 10.... 24.. and 19th d of the experimental period. MATERIALS AND METHODS Experimental Design Two types of assays were performed. .20 g of Mg...32 g of nicotinic acid. Milk Production and Composition Twenty Granadina goats that were midway through their second lactation were divided into four Journal of Dairy Science Vol. 82. and the pore size was 46 × 46 mm.. respectively. and cottonseed (CS). 6. .. After incubation. Subsequently. B6. 400 . The goats ( 1 4 ) were selected to have the same protein polymorphisms. two bags were used per animal.0 kg of concentrate. Every goat received a daily ration consisting of 1... The dimensions of the bags were 7.. 4. and B12.32 g of Ca.02 g of Co. . and cottonseed ( CS) on the production and composition of the milk of Granadina goats.84 g of P.0 kg of alfalfa hay and 1.. and 72 h. The 72-h incubation was repeated several times. The objectives of this study were to establish the overall changes in AA composition of the protein sources used that were due to ruminal incubation and to establish the pattern of relationship between those variables that defined the nature of the protein sources used and those variables that defined the amount and composition of the milk protein produced. Feed intake and milk production were monitored daily. samples of the forage and concentrate offered and samples of orts were collected to determine the composition of the diet fed and consumed. SFM. . corn gluten feed ( CGF) .33 million IU of vitamin A. and 20% of the total protein was supplied by the four experimental proteins: FB. Similarly.48 g of Mn. and 0. equal groups based on BW and milk production.0-mm screen. 12.. (13). ..60 g of Zn.. TABLE 1. Water was available at all times. Ingredient composition of the concentrates. 1. The other set of assays used groups of Granadina goats to determine the effect on milk production and composition of diets in which part of the protein content was provided by the alternative protein sources.. the goats were housed individually in crates for 19 more d. 400 40 sources with different ruminal degradation characteristics: fababeans ( FB) ... and CS.. After milking. For each incubation time. Diet1 Ingredient Oats Corn FB SFM CGF CS Mineral and vitamin supplement2 FB 360 360 240 . the specific N and energy requirements of this species and breed were considered in the dietary formulation ( 2 ) . . At 0900 h every day. samples of milk .0 g of NaCl.10 g of vitamins B1. and the residues were pooled to obtain sufficient material for AA analysis. 40 SFM 360 400 .. CGF. 0. examining at the same time the AA composition of each source before and after ruminal fermentation. 0. 1999 1Twenty percent of protein was provided by fababeans (FB).

The SDS-PAGE was performed on 20% homogeneous precast PhastGels™ (Pharmacia) in accordance with the instructions of the manufacturer (file no. 111) (24). All other analyses were performed on dried samples.00 2. Densitometric peak areas from different caseins and from different whey protein fractions were converted to a percentage of the total casein peak area or the total whey protein peak area. which involves precolumn derivatization with phenylisothiocyanate. When the effect of a covariate factor was not significant ( P > 0.65 38. which were obtained by oxidation with performic acid before 6 M HCl hydrolysis ( 7 ) . and acid detergent lignin contents were determined using the method of Goering and Van Soest (10).63 2. Diet1 FB DM. 220) (23). sunflower meal (SFM). 40 to 60°C).39 18. 2Acid detergent lignin. c = rate of degradation.49 88. The DM and N contents of the dietary component samples and orts. Tryptophan was not determined.4%/h was used for the latter parameter (13). Milk lactose was calculated as the difference between the amount of OM and protein plus fat. and k = rate of passage of the ingesta. 1999 . phosphorylase B (94 kDa). orts. Protein nitrogen values were converted to protein by multiplying by a factor of 6. 3. Uppsala. trypsin inhibitor (20. Cysteine and methionine were determined as cysteic acid and methionine sulfone.78 88. The N content was measured using the Kjeldahl method ( 3 ) .43 2.25. The AA composition of the original protein sources and of the residues in the bags after ruminal incubation was determined by HPLC using the Waters PicoTag method. MA) according to the Whole Band Analysis Program (18).14 22. The gels were scanned using a Bioimate analyzer (3CX ′ Bioimage and visage. milk.73 ( % of DM) 89. The gels were stained automatically in the development unit of PhastSystem™ following fast Coomassie blue staining (file no. carbonicanhydrase (30 kDa). the least squares means were calculated from the model after this term was omitted (32). % OM CP Fat NDF ADF ADL2 Gross energy.05).38.80 36. No. These values were converted to CP by multiplying by a factor of 6.. protein.24 4.26 18.1 17. and milk fat were analyzed from fresh samples. Bedford.76 19. In goats. total N was determined from whole milk samples.08 17. Band densities on the SDS gels were quantitated.21 27. corn gluten feed (CGF). Journal of Dairy Science Vol. evacuated tubes at 110°C for 24 h. and energy concentration in milk) was tested as a covariate in the model. b = slowly degradable fraction. Results were subjected to an ANOVA in accordance with the general linear models procedure of SAS (29). and the effect of DMI on milk production and composition (DM.45 21. regardless of diet type (12. fat.55 18.39 90.40 90. with no added preservatives were stored at –30°C until analysis. and cottonseed (CS). Milk protein fractions were analyzed by means of SDS-PAGE (PhastSystem™ Pharmacia. Millipore Corp. Once it was determined that DMI was not affected by diet. The effective degradability was calculated as a + ( b × c/c + k ) where a = quickly degradable fraction. The ash content of the feedstuffs. lactose.9 88. 82. Sweden). ovoalbumin (43 kDa). and milk was determined by incineration in an electric muffle furnace at 550°C. Standards used were molecular markers (Pharmacia). The model accounted for variation caused by the protein source in the diet.02 17.47 35. The NDF.9 percent of protein was provided by fababeans (FB).4 kDa). The fat contents were measured by extraction with petroleum ether (boiling point.PROTEIN SOURCE AND MILK PRODUCTION OF GOATS TABLE 2. The DM analysis of milk was carried out by lyophilization.96 4. 28). DMI is the factor that most determines milk production and composition. and NPN was determined from a filtrate of whole milk after precipitation with 12% (wt/vol) trichloroacetic acid (17).50 7. The N contents were measured using the Kjeldahl method ( 3 ) . The fat content was measured by the Gerber method (22). The DM of the feedstuffs was determined by oven-drying at 100 ± 2°C for 24 h.58 4.60 5. DMI was considered to be independent of the diet (32). Chemical composition and gross energy content of diets. A value of 2. albumin (64 kDa).49 21. ADF. MJ/kg of DM 1Twenty 557 SFM CGF CS 89.1 88. and the energy content was determined by adiabatic bomb calorimetry. Statistical Analysis The degradation kinetic parameters of each protein source were estimated using the model of Ørskov and McDonald (21). Protein nitrogen content of the milk samples was calculated as differences between total N and NPN. Protein hydrolysis was performed in 6 M HCl in sealed.15 18.1 kDa). respectively.70 41. and lactalbumin (14.

In general. Two different factors were derived. The results of this multivariate analysis are shown in Table 5 and Figure 1. The second factor was mainly determined by the concentration of His and Thr with positive loading values and by the concentration of Met and Val with negative loading values.94 5. multivariate factorial analyses were performed using the factor procedure of SAS (29).00 2.2 82. SFM had a higher ( P < 0. 3.50 19.00 15.35 4.42 32. respectively.05) of the quickly degradable fraction of CP than did the other three protein sources.65 SD 11. The EAA content of the protein sources was similar before and after ruminal incubation.0 c ED 1a = Quickly degradable fraction.2% for CGF).558 SANZ SAMPELAYO ET AL. The number of factors derived in each case.50 SD 1.05) than for the quickly degradable fractions.54 8. Organization of the feeds in the order of the intensity of change yielded the following: FB.8 and 26. 82. b = slowly degradable fraction.13 SD 3.64 b (%) Fababeans Sunflower meal Corn gluten feed Cottonseed X 54. No.21 84. To compare the overall changes in the AA composition of each of the protein sources that were caused by the action of the rumen. and Phe with positive loading values and by the concentration of nonessential AA and Thr with negative loading values. the algorithm used was PROC FACTOR.50 3.0% of the total variance. each goat was considered to be an experimental unit.30 42.86 6. and ED = effective degradability.0% for FB to 42.74 3.32 43.85 56. The lowest values for effective degradability were observed for CS. The methods for the factors extraction and rotation were principal component analysis and varimax.0 88. respectively (29). In Figure 1. The different protein sources had similar EAA contents (ranging from 51. the positions of the different protein sources before and after ruminal degradation and the situations of the different AA are represented in relation to these factors. Journal of Dairy Science Vol. and CGF. the first factor was mainly determined by the concentration of Arg.6%). TABLE 3. the goal was to establish the overall change in the AA composition of protein sources after ruminal incubation. c = rate of degradation. the maximum change was observed for FB (51. The second multivariate analysis was performed to establish the pattern of relationships between those variables that defined the nature of the protein sources used and those that defined the amount and composition of the milk protein produced. The experimental units were the protein sources both before and after ruminal fermentation.79 X 3. The maximum values for the fractional rates of degradation were recorded. Ruminal degradation parameters1 of the protein sources.4 64. The FB had more ( P < 0. Lys. 37. In accordance with the factor loadings.0 vs. the first of the multivariate analyses that were described previously was carried out. Two types of change in the AA composition of the protein sources were observed: one in the negative direction of axis 1 (FB and CS) and the other in the positive direction of axis 2 (CGF and SFM).05) rate than did the other protein sources. RESULTS Degradability of the Protein Sources The degradation characteristics of the CP of the protein sources in the rumen are shown in Table 3.92 79.08 5.31 2. AA Composition of Protein Sources and of RUP Fractions of Protein Sources Table 4 shows the AA composition of the protein sources and residues in the bags after 72 h of incubation. the concentrations of slowly degradable fractions of CP were greater ( P < 0. 1999 .97 4. To determine the patterns of relationships between specific variables. In the first analyses. CS. depended on the fraction of the total variance that each of the factors explained as well as on the extent to which each variable defined them.47 7. The correlation matrix was selected. the first and second accounting for 35. Two multivariate analyses were performed in this way. SFM. a X 53. The variables considered were the concentrations of each of the essential amino acids ( EAA) as well as the total nonessential AA. in addition to dependence on the essential objective of the analysis.

7 5.05) for goats fed the diets containing FB and SFM than for goats fed the diet containing CS.6 7.1 1.4 0.7 1.2 6. no change greater than ±1 kg was observed throughout the experimental period.6 7.0 4.7 5.7 0. milk production (grams per day). DMI.5 9.05) for goats fed the diet containing CGF than for goats fed the diet containing SFM. Thus.6 2.3 2.6 4.05).0 8.9 1.05) for goats fed diets containing FB and SFM than for goats fed diets containing CGF and CS. lactose.4 4. Amino acid composition of the protein sources1 and residues in the bag after 72 h of ruminal incubation.7 4.1 18. the Journal of Dairy Science Vol.9 2.5 5.8 4.1 kg. Live Weights of Goats.8 53.05).05) for goats fed diets containing CGF and CS than for goats fed the diet containing FB.5 8.0 49.PROTEIN SOURCE AND MILK PRODUCTION OF GOATS TABLE 4.5 39. Finally. 1999 .3 4. The b-CN percentage was higher ( P ≤ 0. Type of diet induced changes in milk DM and protein content ( P ≤ 0. 3Nonessential AA.3 1.6 18.0 7.1 0.6 5. The as-CN percentage was higher ( P ≤ 0.3 4.2 11.9 4.8 9.8 5. CGF = corn gluten feed.3 0.3 1. Table 6 shows the mean DMI (grams per day.3 9.7 41. slowly degradable protein in the rumen (grams per kilogram of BW0.3 1.05) for goats fed the diets containing CGF and FB than for goats fed the diet containing SFM.3 7.1 1.7 2. In the same way. No.0 4.7 5.1 24.7 0. the protein converted into quickly degradable protein in the rumen (grams per kilogram of BW0.2 6.75 per day).75 per day).0 44.8 0.75 per day). fat.9 5.9 6.6 62.7 7.5 9.5 6. The b-LG percentage was higher ( P ≤ 0.1 6.9 9.5 7.9 3. 3.6 7. and effectively degradable protein in the rumen (grams per kilogram of BW0.2 3.9 5.6 37.3 5.6 7.4 5.7 6.0 58.1 4.0 13.4 7.1 3.2 10.9 2.0 13. the a-LA percentage was higher ( P ≤ 0.7 1.0 = Fababeans.6 2. and Milk Production and Composition The mean live weight of the goats was 43 ± 3. Relationships Between the Amount and Composition of the Milk Protein Produced and the Nature of the Protein Sources Used For each goat. and energy contents and the mean values for the different protein fractions in the milk produced (percentages).5 3.4 2.05) for goats fed the diet containing SFM than for goats fed the diet containing CGF.1 55.3 3.0 7.6 5.3 4. protein.75 per day).2 3.4 1.7 3.6 15.4 51. and CS = cottonseed.2 2.7 4. Only the amounts of lactose were independent of DMI ( P > 0.6 9.1 4.5 4.75 per day) that was provided by each protein source ingested was calculated as a function of the degradation characteristics estimated for each source.1 4. The serum albumin percentage was higher ( P ≤ 0.1 7.1 5.8 60.5 1.1 9.6 9.6 5. SFM = sunflower meal. and RUP (milligrams per kilogram of BW0.2 4.3 58. The type of diet did not affect DMI ( P > 0.5 2.7 5. and milk composition (percentage and megajoules per kilogram) in terms of DM.1 3. FB Feed EAA2 Lys His Thr Arg Val Met Ile Leu Phe NEAA3 Tyr Cys Asp Glu Ser Gly Ala Pro Total EAA Total NEAA 1FB 2Essential 559 SFM Feed Residue Feed CGF Residue Feed CS Residue Residue (g/100 g of AA) 7.5 8.4 3.2 4.05).9 4.4 9.1 5.5 6. Milk protein content was greater ( P ≤ 0.2 3. 82.4 2.0 42.0 3.2 10.05) for goats fed the diet containing FB than for goats fed the diet containing SFM.6 1.3 23.1 16.9 7.7 7.8 4.0 3. The k-CN percentage was greater ( P ≤ 0.1 11.2 6.8 3.6 7.9 7. AA.9 1. milk DM content was higher ( P ≤ 0.7 5. and grams per kilogram of BW0.6 10.7 4.8 1.0 42.8 1.3 0.7 10.1 6.3 6.2 57.05) for goats fed the diet containing CGF than for goats fed the diet containing SFM.0 4.5 13.8 4.4 6.1 46.

Figure 2 shows the position of the different variables as well as the area of dispersion for the different protein sources with respect to factors 1 and 2. The first of the derived factors.1 Factor matrix Variable Arg Lys Phe Ile NEAA4 His Thr Leu Val Met Factor 12 0.8030 0. the different protein sources appeared in different positions. Effects of ruminal degradation on the factor pattern of the AA profile of the protein sources. FB were chosen because goats show a distinct preference for this feed. Finally.3880 –0. Factor pattern for amino acids from protein sources and residues in the bag after 72 h of ruminal incubation. TABLE 5.7906 0.75 per day) and the corresponding amounts of the milk protein fractions (grams per kilogram of BW0.0763 –0.0848 0. factor 4 was defined by the different milk whey fractions.1937 –0. DISCUSSION Factor 1 2 Variance explained 35. Results obtained from were derived from multivariate analysis.1763 –0. those defining the nature of the protein sources used and those defining the amount and composition of the milk protein produced. The other three sources were chosen because of their supposed lower degradability. was specially defined by the RUP and by its EAA content.1819 0.5993 1Results 2Expresses Factor 23 –0.75 per day). CGF = corn gluten feed.7134 0.0 Cumulative variance explained (%) 35.75 per day) that were provided by the corresponding RUP fraction were calculated from the amino acid composition of residues in the bag. Four different factors were derived. in accordance with the corresponding factor loadings. 3. No other variables determined this factor to have a high loading value. No.8 Characteristics of the Ruminal Degradation of the Protein Sources Of the protein sources used. 4Nonessential AA. CS = Cottonseed. and NEAA = nonessential AA.6223 0. Factor 3 was primarily determined by the slowly Journal of Dairy Science Vol. These variables appeared with only low factor loadings in the other factors. respectively.8. SFM = sunflower meal.5779 2. Point at end of arrow represents the factor pattern of the AA profile of residue remaining in the bag. third. Boza and Ferrando ( 4 ) reported that SFM and CGF are good protein concentrates. the first. With the values of all of these variables. amounts of each EAA (milligrams per kilogram of BW0.3204 –0. and with the goal of exploring the relationships among them.4438 –0.2114 0. 18. 1999 Figure 1. and Phe content in the protein sources and residues in the bag.6332 –0. and 10. CS is a good source of bypass protein as well as an excellent energy source.9.3% of the total variance. and the results are shown in Table 8. FB = fababeans.8 61. and fourth accounting for 50.2171 0. 82.7340 –0. a multivariate factorial analysis was carried out.560 SANZ SAMPELAYO ET AL. second. 11. 3Expresses mainly the His and Thr content in the protein sources and residues in the bag. The amount of milk protein produced and its different casein fractions as well as the quickly degradable protein in the rumen were the variables that most clearly defined the second factor. With respect to factor 1. Lys.9636 –0. mainly the Arg.8 26.3591 Final statistics Eigen value 3. All of these values are shown in Table 7 together with the total milk protein produced (grams per kilogram of BW0.5. .8219 degradable protein in the rumen and by the effectively degradable protein in the rumen. the different protein sources occupied more similar positions with respect to factor 2.8257 0.

61 5.1 70.71 4. and cottonseed (CS). 15).5b 29.3ab 35.52ab 2. the amounts of Ile and Phe might have increased. its concentration decreased ( 8 ) .4a 56.4 1098 16..70 0. In contrast.93 6. the results showed clearly Rulquin and Verite that the overall change of the AA profile depended on the feeds (Figure 1). Phe and Lys decreased in the protein sources.1 74.PROTEIN SOURCE AND MILK PRODUCTION OF GOATS TABLE 6.9 70..29 7.16 7. % Fat. % of Whey protein a3-CN.2b 25.42 1. except in CGF and CGF and CS.25a 6.3a 14. % Lactose. percent of protein was provided by fababeans (FB). g/d DMI.. No.40a 3. 15) reported data from which it can be deduced that the use of the original AA profile of one protein source to predict EAA available for absorption is not accurate because the change that is undergone differs according to the protein source.. 3P > 0. sunflower meal (SFM).3 1071 15. Milk production and composition by goats fed diets differing in protein source. Romero et al.. when a multivariate analysis was carried out that was similar to that by ´ (26).4a 56.32 5. As a result.1b 66.3a 29.7ab 5.50a 6.75 per d Milk production.74 15.30 3. The degree of modification brought about by ruminal fermentation of the AA profile of each protein source is given by the magnitude of the vector that joins the position of each one before and after ruminal fermentation...0 80.5a 14.10 0. At the same time.0b 8. % of Casein a.71 0. this study for the characteristics of ruminal degradation showed that FB.9ab 23. g/d Milk composition DM. 4Serum albumin. . *** P ≤ 0..18ab 6. g/kg of BW0.2 971 16. Diet1 FB DMI.2a 26.11 2..0b 37.21b 3..9b 1157 69.9a 19.31 0.. However..001. In the present study.7b 29. % Energy.6ab 7. . Rulquin and Verite cluded that these differences were small in comparison with the differences that continue to exist among feeds..17 269. .66 Covariate . % of Casein k-CN.2ab 28. the AA compositions of their corresponding RUP fractions were much closer.70 5.12 3.05.8a 11.5 1153 15. corn gluten feed (CGF). a marked decrease was noted in the concentration of Arg in the four protein sources.63 4. 3. The AA profiles of FB and SFM were very different before ruminal fermentation. Thr increased in all four of the protein sources.97 3.. *** *** ** *** NS ** . 2Residual standard deviation.8a 56. 1999 .1a 7. % of protein SA.77 4.9b 64. such as was inferred from this study.bMeans 1Twenty 561 P CGF 1064 65. % Protein. ( 2 5 ) worked with goats fed diets containing soybean meal or CGF as the protein source.8ab CS 1155 68. Journal of Dairy Science Vol.. and the amount of Lys might have decreased (9.67 0. 82.6a 27.05.33a 3. SFM. Corn protein has a lower degradability than does soybean meal protein.4 % of Whey protein a-LA. . Those researchers ( 2 5 ) attributed this result to the higher content of RUP that reached the duodenum.01.87b 6. . * P ≤ 0.1ab 58.0b 26.. Something similar occurred with the CS and CGF. respectively. After examination of the effect of ruminal incubation on the AA profile by ´ ( 2 6 ) conmultivariate analysis.8ab 62. ** P ≤ 0. In the present study.7a within a row with unlike superscripts differ ( P ≤ 0. those researchers noted a higher milk production by goats fed the CGF than that by goats fed soybean meal.1b 11. The AA profiles of the residues of feeds subjected to ruminal fermentation shows that Arg is one of the amino acids that is most sensitive to such fermentation.57 5. % of Whey protein b-LG.. Diet NS3 NS NS * ** NS NS NS * * * * * ** * * SFM 1064 61. % of Casein b-CN. . and CGF had similar effective degradation..00 5. . other researchers (8.8b 63. .6ab RSD2 254. MJ/kg Protein fractions Casein... 9.05). % of protein Whey protein. AA Composition of the Protein Sources Used and of Their RUP Fractions ´ ( 2 6 ) opinion In spite of the Rulquin and Verite that feed AA profiles can be taken as a first guide to estimate the AA profiles of RUP.

With this in mind and with the goal of determining the effect of the different types of diet on milk composition.12 0.21 0.91 2.75 QRDP.40 1.28 1. No additional differences were detected as a result of the type of diet.89 1.36 4.69 2.09 0. 7Serum albumin.26 CGF 0.04 2.77 97.5 0.20 1. energy intake is the factor that is the most significant determinate of milk composition (28).26 8.83.80 6.46 2.11 0. sunflower meal (SFM). studies carried out on goats to investigate this effect (11.22 14.56 1.19 23.40 0. considering rates of DMI as a covariance factor. In goats.93 134. The milk production by goats fed diets with practically equal energy and N contents is known to be dependent on the intake of the diet.03 0. However.05 0. less degradable proteins are used to increase the milk protein content as long as the animal has the potential to increase milk protein ( 5 ) .47 0.33 4.26 69.47 3.07 0.13 0.61 24. Journal of Dairy Science Vol. g/kg of BW0.43 2.97 1.01 0.96 0.07 3.562 Milk Production and Composition SANZ SAMPELAYO ET AL.2 g SRDP. In the lactating ruminant.33 2. the effect of a change in the protein source of the diet on the protein content of caprine milk has always been analyzed by comparing alternative protein sources with soybean protein.21 1.41 1.75 MP as-CN b-CN k-CN SA7 a-LA b-LG 1.88 1.01 0.59 1. corn gluten feed (CGF). 3Slowly degradable protein in the rumen.07 0. However.13 0. In the present study. Morand-Fehr et al.28 SEM 0. 4Effectively degradable protein in the rumen.22 12.87 1.90 6.72 0.98 1.59 0.56 3.35 4. 5Essential AA.4 g RUP.00 23.96 0. yielded results that indicated that production depended on intake. Diet1 FB Daily intake per kg of BW0.0 5. Here the best results were obtained when goats were fed the diet contain- TABLE 7.08 0.82 0. Results could vary with the nature of the protein that the less degradable protein replaces.8 16.29 X 0.92 0. the DMI in each case was used as the covariance factor. The correlation between the energy intake and milk production halfway through lactation is 0.13 0.65 5.10 0.09 1. and cottonseed (CS).25 CS 0. the values of different parameters that are indicative of composition were analyzed statisti- cally.36 8.10 1.21 2.63 1. mg EAA5 in RUP.87 25.27 1. 6Milk protein. 19) have given different and.38 1.25 0. 1999 .90 0.23 2.70 13.25 0. mg Lys His Thr Arg Val Met Ile Leu Phe MP6 and MP Fractions.96 10.73 1.06 0.8 6.86 0.80 4.32 SFM 0.3 g ERDP.02 0.14 0.1 2.75 1. No.90 0.44 2.40 1.49 2.14 0.01 0.49 7.94 3.01 0.28 4.56 0.51 0.79 1. 3. at times.49 0.04 1. contradictory results. 82.10 0. Diet and overall means of the variables that define the nature of the protein sources and the amount and composition of the milk protein produced.43 0. ( 1 9 ) have reported that in the majority of cases in which isoenergetic diets with the same N content are used.06 0.74 10.44 0. The same result was found to be true for milk fat content.14 0. 2Quickly degradable protein in the rumen.18 8. Hadjipanayiotou and Morand-Fehr ( 1 2 ) reported that feed intake is the chief factor that determines milk production by goats.55 2.14 35.02 1Twenty percent of protein was provided by fababeans (FB).05 0.79 0. analyses of milk production.16 0.52 347.92 1. the protein and casein contents of the goat milk do not appear to be particularly sensitive to changes in the protein source of the diet.

Factor matrix Variable Ile Leu RUP Thr Met Arg Lys Phe His Val k-CN b-CN MP QRDP as-CN SRDP ERDP a-LA b-LG SA Factor 11 0.0945 0.0266 0.8560 0.0165 0. The milk production by goats fed the diets containing SFM had the lowest protein content.9087 0.0160 Factor 22 0.0764 0.0341 0. the milk casein percentage was higher than that produced by goats fed the other diets. No. When the effects that the kinetics of ruminal degradation of a protein can have on the use of the protein supplied for milk production are considered.0588 0.1163 0.0178 –0. These results suggest the necessity of taking into account the process of milk protein production from the protein consumed by the animal. 31) indicate that it is necessary to take into account first.9912 0. 82.5 Factor Eigen value Variance explained 1Expresses the RUP and its essential AA content.0255 0.0927 –0. b-LG. these results were due to the effect of the nature of the protein source used on the amount and composition of the milk protein produced.0289 0.1413 0.0486 0.5 10.1787 Factor 44 0.6337 Cumulative variance explained (%) 1 2 3 4 10. 20) suggest that animals need to absorb a certain amount of protein to ensure that there is a certain supply of RDP and RUP.0341 –0.2 91.1659 –0.0821 0.0315 0.0102 –0. and serum albumin (SA). Nature of the Protein Sources and Amount and Composition of the Milk Protein The new systems used to evaluate the protein in the feeds of lactating ruminants (1. 2Expresses the quickly degradable protein in the rumen (QRDP).0276 –0.1136 0.0742 0.8 18. 4Expresses the different whey protein fractions.0542 0.3247 –0.1353 0.9864 –0.7736 2.7322 0.PROTEIN SOURCE AND MILK PRODUCTION OF GOATS 563 ing CGF.0061 0. However.0651 0.3413 –0.0493 –0.0829 0.3372 0. 26.0505 –0.3727 0.0922 0.0063 0.0319 0. 3Expresses the slowly degradable protein in the rumen (SRDP) and effectively degradable protein in the rumen (ERDP).0643 0.0116 0.0533 0.0273 –0. 27. although not significantly higher than that produced by goats fed the diet containing FB. opinions and experimental results (5.4961 0.9963 0.9514 0. weakly associated with milk protein ( M P ) and with its as-CN and b-CN fractions. 1999 . In addition to the higher milk protein production by goats fed this diet.0327 0.9 11.0690 50. As is discussed later.9977 0. the AA profile of the RUP fractions of the protein supplied.9926 0.7 81.0795 0.2993 0.9946 0.0745 0. 3.2240 0.3072 0.0522 0. it is necessary to examine the values estimated here for the different parameters considered in the model of ruminal degradation developed by Ørskov and McDonald ( 2 1 ) as a function of the protein source TABLE 8.0983 –0.9972 0.0048 0. a-LA.9897 0. Journal of Dairy Science Vol.0198 Final statistics Factor 33 0.9940 0.3899 0. 15.2784 0. weakly associated with as-CN and a-LA.0439 0. Variables that define the nature of the protein sources and the amount and composition of the milk protein produced.0411 0. Results were derived from multivariate analysis.8123 0.0494 0.0386 0. This increase was closely linked to the b-CN percentage. the kinetics of the ruminal degradation of the protein supplied and second.8735 0.8 69.3 50.0109 0.6414 0.2704 –0.2953 2.9488 0. with greater frequency.1565 3.9895 0. mainly associated with MP and with its different casein fractions. which is a multivariate process.8938 0.

the RUP. it was possible to distinguish among the protein sources. 2. QRDP = Quickly degradable protein in the rumen. In our case. If the AA composition of the RUP fractions of the protein sources (Table 4 ) is considered. sunflower meal (horizontal lines). regardless of its EAA composition. according to the results obtained. in particular to its casein fractions. multivariate statistical analysis is an appropriate methology to identify biological similarities among them. EAA = essential AA. The significance of this additional N for the lactating ruminant appears to depend on both the amount of N supplied and the AA profile (5. it can be seen that the proportions of some AA were similar (His. second. the RUP fraction. and CGF were very similar despite the fact that the values for the quickly degradable fraction in the rumen. With regard to a determined physiological process. especially with respect to the first factor. No. SFM. Dispersion area of protein sources: fababeans (vertical lines). and third. the quickly degradable protein in the rumen. The animal would thus acquire an additional supply of N. the slowly degradable fraction in the rumen. 1999 available. In our . ERDP = effectively degradable protein in the rumen. the identification of groups of variables that behaved differently and independently was possible.564 SANZ SAMPELAYO ET AL. Val. and Leu). First and second factors in the multivariate analyses were used to plot the variables. it is worthwhile to determine which aspects of the composition or nature of the protein sources used most affect the amount and even the composition of the milk protein that is produced. SFM. 15. respectively. and the rate of degradation were quite different. it may be possible to establish a hierarchy among the four protein sources: CS. With regard to the significance of the quickly degradable protein in the rumen as deduced here. and Phe) and those for others were different (Thr. the effective degradabilities estimated for the protein from FB. Journal of Dairy Science Vol. 82. examination of the pattern of relationships that existed among the considered variables and. or 3. 30) have indicated how the results obtained in this way acquire statistical as well as biological significance. 26. To a much lesser extent. the factors represent new variables that are not correlated with each other ( 6 ) . CGF. Met. the N that is supplied to the animal in the form of microbial protein). These variables intervened only in the definition of factors 1. in this type of analysis. 31). Variables that define the nature of the protein sources and the amount and composition of milk protein ( M P ) as related to source of protein used. it is worth bearing in mind the observations of Chandler ( 5 ) . and SA = serum albumin. the results highlighted the different behaviors of first. and cottonseed (crosshatched pattern).. Various researchers (6. also was related to milk protein with regard to as-CN and b-CN fractions. Because.e. Chandler ( 5 ) also pointed out that the supply of this microbial protein could be balanced using protein sources that are resistant to ruminal degradation. With regard to the variables that define the nature of the protein sources. In this way. Figure 2 separates and classifies the feeds. SRDP = slowly degradable protein in the rumen. the determination of whether. two different goals could be realized with this type of multivariate analysis: first. No protein supplied fraction was related to any whey protein fraction. used (Table 3). corn gluten feed (dotted pattern). a multivariate factorial analysis was carried out with the values of all derived variables: those defining the nature of the protein sources used and those defining the amount and composition of the milk protein produced. As Chatfield and Collins ( 6 ) reported. 3. the slowly degradable protein in the rumen and the effective protein degradable in the rumen. This examination reveals that it is possible to derive similar values for effective degradability from quite different parameters. Because of these results. it is necessary to identify those protein sources that will be quickly degraded first (i. 27. when the values of a series of relevant variables are Figure 2. resulting in nearly identical total proportions of EAA at the same time for the different protein sources. The quickly degradable protein in the rumen was the fraction most closely related to the milk protein produced. who reported that to satisfy the protein requirements of lactating ruminants best. In fact. So the different whey protein fractions were the most independent of the nature of the protein source. and FB. second.

Version 3. C. Effect of protein supplementation on milk yield of goats grazing a semiarid temperature rangeland.PROTEIN SOURCE AND MILK PRODUCTION OF GOATS 565 study. P. The Netherlands. P. England. Situacio aprovechamiento de los subproductos en Espan ˜ a. E. M. M.. McDonald. K. Agric. kinetics of digestion and passage and rumen fermentation pattern in goats and sheep offered medium quality forages at the maintenance level of feeding. the concentrations of these two AA were similar in the RUP fractions from each source. M.. ed. Cary. P. M. 8 Cozzi.D. 1989. and A. Suministros Gra 5 Chandler. Torrie. NY. 77:541–551. Blanchart. 32 Steel. Morand-Fehr. 220. ´ . 14 Law. Ruminant Nitrogen Usage. and J. England. Dairy Sci. C. Pudoc. J. 1990. 1987. the possible supplementary N achieved from the RUP fraction was established by the amount supplied. 6 Chatfield. Pages 209–224 in Goat Nutrition. J. and I. M. New York. P. Nutr. as can be observed from Figure 1. Handbook No. Amino acid profile and intestinal digestibility in dairy cows of rumen-undegradable protein from various feedstuffs. 18 Millipore. CGF. Comparison and Pattern Matching. 1994. 21 Ørskov. and G. Garcı ´a. 23 Pharmacia AB. Hadjipanayiotou. and P. J. Nottingham. P. M.. ( 3 1 ) who observed that the EAA with the most limiting effect on milk production are Lys and Met. Tarvin. A. E. 13: 21–25. J. 28 Sanz Sampelayo. 40:237–248.. 1976. Bozak. M. Laboratory Techniques in Food Analysis. Botha. 1986. SFM. L. Acad.. Cole. Principles and Procedures of Statistics: A Biometrical Approach. Millipore Corp. 17 Martı ´n-Herna ´ n de la leche de cabra de Martı ´n-Alvarez.S. Daccord.. J. Aguilar.) 12 Hadjipanayiotou. Jua ´ rez. Importances des acides amine ´ res. With regard to the milk protein produced. The estimation of protein degradability in the rumen from incubation measurements weighted according to rate of passage. Sci. all of the undegradable fractions were more similar in amino acid composition than were protein sources prior to ruminant degradation. and V. E. Amigo. Prieto. J. REFERENCES 1 Agricultural Research Council. (Paris) 36:326. 3 Association of Official Analytical Chemists. Bedford. C. Ann. 39:201–212. S. ´ s dans la 15 Le Henaff. Anim. The Nutrient Requirements of Ruminant Livestock. 1975. and E. S. Pages 55–77 in Recent Advances in Animal Nutrition. Shimada. Influence of feeding on goat milk composition and technological characteristics. No. and P. 1999 . Results of an european ringtest. L. M. In this sense. Procedures. and C. P. The quickly degradable protein in the rumen was most closely related to the milk protein produced. Remeuf. J. G. the RUP fraction was also related to milk protein production. Morand-Fehr. C. Huber.. Santiago. Composicio razas Murciana y Granadina. 1993. In situ ruminal disappearance of essential amino acids in protein feedstuffs. Nutr. F. Bio Image. Quantitation. To a much lesser extent and regardless of the EAA composition. 1994. and R.2. England. ed.. 9 Erasmus. the protein sources used consisted of different and independent entities.. Prediction of in situ protein degradability in the rumen. Cruywagen. Pe Forage of different physical forms in the diets of lactating Granadina goats: nutrient digestibility and milk production and composition. Washington. van Soest. A. SAS Inst. 20 National Research Council. 1980. 1988. MA. Dairy Sci. Whitehouse. Bas. Pudoc. F. J. Effect of protein source on performance in lactation of Chios ewes and Damascus goats. Hvelplund.. Spain. G. M. Washington. 1994. NC. J. DC. Gihad.. Livest. Sweden. 10 Goering. 1970. 2-Extent of lysine limitation. Uppsala. McGraw-Hill Book Co. and A. 1991. Ann. Garnsworthy and J.. S. I. 1987. Sci. Anal.. The Netherlands. Intensive feeding of dairy goats. Dairy Sci. 1979. and F. Amino acid limitation and flow to the duodenum at four stages of lactation. L. 1992..A. and D.. Collins. ´ ndez. D. Bromatol.A.. A. R. Zootech. and T. 1996. 12th ed. Zootech. Prod. R. 1995. L. A comparative study of nutrient digestibility. J. Ferrando. Boza. Arch. Washington. Quality of bypass proteins fed to high-producing cow is important. Slough. 46:37–50. ed. SDS-PAGE in homogeneous media. W. 29 SAS® User’s Guide: Statistics. and E. 78:161–171. one should consider the comments of Schwab et al. Dairy Sci. 1992. Andrighetto. H. Journal of Dairy Science Vol. GigerReverdin. 22 Pearson. Press. C. Quantitative fractionation of caprine casein by cation-exchange FPLC. Introduction to Multivariate Analysis. R. Morand-Fehr.. Sensitive silver staining. PhastSystem Separation Technique File No. Sevilla. Br. 1991. Pages 197–208 in Goat Nutrition. P. and J. A. MorandFehr. J. A Manual of Advanced Techniques for Amino Acid Analysis. ARS-USDA. AOAC. Aguilera. Wageningen.J. 2nd. Butterworths.. M.R.2. Forage Fiber Analyses (Apparatus. Ramos. and L. and Some Applications). Amino acid nutrition of dairy cows advances but still has ways to go. 75:3503–3518. Thesis. (Camb. and P. A. L. J. J. A. France.G. 15. (Paris) 27:75–81. 111.. 27 Santos. 1991. J. Georghiades. Version 6 Edition. Univ.. 1985. H.. 1994. T. Reagents. M.. Agric. D.D. Official Methods of Analysis. H. D. CAB. Pages 9–70 in ´ n Animal III. nutrition des vaches laitie Rennes. London. MA. J. Nuevas Fuentes de Alimentos para la Produccio ´ ficos. and P. 63:165–175. C. Wageningen. Natl. Tziboula. 3. 81:492–498. England. 379. Molina Alcaide. and degradability of protected and unprotected soybean meal in the rumen of goats. M.. Bedford. Rennes. Mowlen. ´ n actual en el estudio y 4 Boza. G. and J. Milchwissenschaft 47:558–562. J. Whole Band Software. Millipore Corp.. N. 19 Morand-Fehr. Analyse des variations individuelles en nutrition animale: application de l’analyse en ` l’e ´ tude de la secre ´ tion lipideque du composantes principales a ` vre. K. Sauvant. In the present study. and CS. 1993. Koumas. 1978. 24 Pharmacia AB. 25 Romero. Small Ruminant Res. Fonolla energy metabolism of lactating Granadina goats. A. and P. Sizing. and T. F. Furthermore.O. 16 Madsen. ´ . PhastSystem Development Technique File No.(Abstr. 7 Cohen. DC. Meys.) 92: 499–503. F. Protein and 2 Aguilera. The Pico-Tag Method. 1980. 11 Hadjipanayiotou.. Chapman and Hall Ltd. P. Sweden. Univ. P. ed. 1989. 1980.. E. 13 Isac. Amino acid nutrition of dairy 26 Rulquin.. Feedstuffs 68(34): 12–13. Press. Feedstuffs 67(33):11–14. Berzaghi. CONCLUSIONS The composition of the milk produced by Granadina goats appears to be sensitive to diets with 20% of the protein provided by FB. A. DC. 1998. Uppsala. M.. 1987. R. lait de che 31 Schwab. 1994. London. Verite cows: productive effects and animal requirements. regardless of its EAA composition. Inc. ´ rez. 82.. 30 Sauvant. Olson.