Lactation Responses to Sulfur-Containing Amino Acids from Feather Meal or Rumen-Protected Methionine

Transcription

1 Lactation Responses to Sulfur-Containing Amino Acids from Feather Meal or Rumen-Protected Methionine S. Pruekvimolphan and R. R. Grummer University of Wisconsin, Madison ABSTRACT The objectives of this experiment were to determine: 1) if Cys in hydrolyzed feather meal (HFM) can contribute to the supply of sulfur amino acids for meeting requirements of dairy cows; 2) if the feeding value of meat and bone meal (MBM) can be enhanced by HFM or ruminally protected Met (rpmet); and 3) the value of HFM sulfur amino acids relative to rpmet. Fifteen multiparous Holstein cows were used in a replicated 5 5 Latin square design with 21-d periods. The control (CTRL) diet was designed to contain feeds that were low in RUP and Met and consisted of 50% alfalfa silage and 50% corn-based concentrate. Additional treatments were modifications of CTRL in which MBM (4% of DM), MBM + rpmet (Smartamine M) (4 and 0.08% of DM), MBM + HFM (4 and 2% of DM), and MBM + rpmet + HFM (4, 0.04, and 1% of DM) replaced corn grain. Feeding MBM depressed milk and 3.5% fatcorrected milk (FCM) yields. Adding rpmet to MBM enhanced dry matter intake, milk and 3.5% FCM yields, and milk crude protein percentage. Milk fat percentage and 3.5% FCM yield were increased when HFM and rpmet were added to MBM. Supplementing HFM to a diet containing MBM could not duplicate the response of adding rpmet to MBM. Results of this study indicate that feeding HFM may not alleviate Met deficiency in lactating dairy cows. (Key words: meat and bone meal, methionine, cysteine, hydrolyzed feather meal) Abbreviation key: CTRL = control, DMETMP = digestible methionine, as a percentage of metabolizable protein, HFM = hydrolyzed feather meal, MBM = meat and bone meal, MP = metabolizable protein, rpmet = ruminally protected Met, SBM = soybean meal. INTRODUCTION The U. S. meatpacking and rendering industry produces byproducts that are high in protein, greater than Received April 4, Accepted July 2, Corresponding author: R. R. Grummer; rgrummer@ facstaff.wisc.edu. 50% of DM. Hydrolyzed feather meal (HFM) and meat and bone meal (MBM) are examples of animal byproducts used in diets fed to livestock. Research suggests that Met or Lys are the first or second limiting amino acids in most dairy diets (Schwab et al., 1992; Armentano and Bertics, 1993; Rulquin and Delaby, 1994). In the future, protein feeding strategies for dairy cattle may be similar to that of poultry and swine, i.e., feeding a low-cost protein source and supplementing diets with a specific limiting amino acid. Consequently, it is important to determine the feeding value of animal proteins that may be low in one or both of these amino acids. For example, MBM is rich in undegradable protein. It is an excellent source of Lys but a poor source of Met. Therefore, supplementing Met that escapes degradation in the rumen may enhance the nutritive value of MBM. Hydrolyzed feather meal is rich in sulfur amino acids, mainly Cys. In nonruminant diets, Cys can spare Met and satisfy 50% of the requirement for sulfur containing amino acids (Met + Cys) (Chung and Baker, 1992). It is not known whether Cys can spare the Met requirement in diets fed to lactating cows. If Met is the first limiting amino acid in diets containing MBM, then feeding HFM with MBM may resolve a Met deficiency and increase the likelihood of obtaining a production response when feeding MBM. The objectives of the present study were to determine 1) if Cys in HFM can contribute to the sulfur amino acid requirement of dairy cows; 2) if the feeding value of MBM can be enhanced by HFM or ruminally protected Met (rpmet); and 3) the value of HFM sulfur amino acids relative to rpmet. MATERIALS AND METHODS Fifteen multiparous Holstein cows averaging 640 kg were used in a replicated 5 5 Latin square design. Five of 15 animals were fitted with rumen cannulas and made up one square. The remaining cows were blocked by square according to days postpartum. Cows averaged 61 ± 14 (SD) DIM at the start of the trial. Experimental periods were 21 d each, with a 16-d adaptation period followed by a 5-d collection period.

2 Table 1. Ingredient and nutrient composition of diets and estimated amino acid flows to the intestine. Treatment 1 MBM + MBM + MBM + HFM + CTRL MBM rpmet HFM rpmet (% of DM) Ingredient Alfalfa silage Corn grain Blood meal MBM, porcine HFM rplys rpmet Urea Tallow Vitamins and minerals Nutrient composition CP, % CP, % RUP, % RDP, % NDF, % ADF, % Ca, % P, % Ash, % NE L, Mcal/kg of DM Predicted Lys flow, 5 g/d Predicted Met flow, 5 g/d Predicted RUP Cys, 6 g/d Predicted Met plus Cys flow greater than control, 7 g/d CTRL = Control (methionine-deficient diet); MBM = control diet supplemented with meat and bone meal; MBM + rpmet = Control diet supplemented with meat and bone meal plus rumen-protected methionine (Smartamine M, Antony, France); MBM + HFM = control diet supplemented with meat and bone meal plus hydrolyzed feather meal; MBM + HFM + rpmet = control diet supplemented with meat and bone meal plus hydrolyzed feather meal plus rumen-protected methionine. 2 Analysis of concentrate mixes and alfalfa silage. 3 Calculated with measured CP for alfalfa, MBM and HFM, and NRC (2001) values for other feeds. 4 Predicted (NRC, 2001) using measured CP for alfalfa, MBM, and HFM and NRC (2001) values for other feeds. Assumes 640-kg cow, 25 kg of DMI/d and 60 DIM. 5 Predicted amino acid flow to the duodenum (NRC, 2001). Assumes 640-kg cow, 25 kg of DMI/d and 60 DIM. Assumes rpmet is 100% protected. 6 The NRC (2001) does not predict total Cys flow to the duodenum. This value is what NRC (2001) would predict for Cys flow to the duodenum from RUP assuming the amino acid profile of the RUP is the same as CP in feeds. 7 A prediction of the increase in total sulfur amino acid flow to the duodenum assuming that the contribution of Cys from microbial and endogenous protein is constant across treatments. Cows were fed control (CTRL) diet or one of the following treatments: MBM (4% of DM); MBM + rpmet (Smartamine M [75.5% methionine] Rhone-Poulenc Animal Nutrition, Antony, France) (4 and 0.08% of DM); MBM + HFM (4 and 2% of DM); and MBM + rpmet + HFM (4, 0.04, and 1% of DM). Ingredient and nutrient compositions of the experimental diets are in Table 1. Feeds used to formulate CTRL were selected to increase the likelihood that cows would be deficient in Met. Rumen-protected Lys (36% Lys, Archer Daniels Midland, Decatur, IL) and spray-dried blood meal were added to each diet to minimize the likelihood that Lys would be the first limiting amino acid. Previous research demonstrated greater milk and casein yield when this protected Lys product was added to Lysdeficient diets (Armentano, 1996). The MBM diet was formulated to provide more RUP than the CTRL diet, but still be deficient in Met (negative control). The MBM + rpmet was formulated to determine whether the feeding value of MBM could be enhanced by supplementation with Met. We anticipated that this treatment would serve as a positive control for evaluating HFM. The MBM + HFM diet was formulated to test whether a protein rich in undegradable Cys as a sulfur

3 amino acid source could replace Met. In the event that Cys can only partially spare Met, the MBM + HFM + rpmet treatment was included to test the response when approximately one-half of the supplemental sulfur amino acids is contributed by HFM and one-half by rpmet. Predicted amino acid flows were calculated using the NRC model (NRC, 2001; Table 1). Cows were fed individually a TMR twice daily, and the amount of feed offered was adjusted daily to allow a 10% feed refusal. The 50:50 forage-to-concentrate ratio in the TMR was maintained by determining DM content of feeds weekly. Alfalfa silage was sampled weekly. Concentrate samples were obtained each time a new batch was mixed. Weights of feed offered and refused were recorded daily. Orts were sampled on d 17, 19, and 21 of each period, then samples were dried for 48 h in a 60 C forced-air oven and ground to pass a 2-mm Wiley mill screen (Arthur H. Thomas, Philadelphia, PA). Orts were composited on a proportional basis according to amounts of feed DM refused each day. Ort and feed samples were analyzed for DM, OM, and CP (AOAC, 1990), ADF (Goering and Van Soest, 1970), and NDF (Van Soest, 1990). Hydrolyzed feather meal and MBM were analyzed for estimated rumen degradability and intestinal digestion (Calsamiglia and Stern, 1995). Cows were milked twice daily. Milk yields were recorded daily and average production was calculated from the last 5dofeach period. Milk samples were collected on 10 consecutive milkings during the last 5 d of each period and were analyzed for fat and protein content by near-infrared analysis (AgSrouce Cooperative Services, Menomonie, WI). Blood samples were collected in heparinized Vacutainers at 3 h postfeeding on d 19, 6 h postfeeding on d 20, and 9 h postfeeding on d 21 of each period and centrifuged at 2700 rpm for 10 min at 4 C. Plasma was removed and frozen at 20 C until analyzed for urea nitrogen (Raabo and Terkildsen, 1960) and glucose concentration (Weichselbaum et al., 1969) with Sigma Diagnostics kits (no. 640 and no. 510, respectively, St. Louis, MO). Plasma NEFA was determined according to Johnson and Peter (1993; Waco NEFA-C kit, Wako Chemical Industries USA, Inc., Richmond, VA). Plasma amino acids were measured on a composite blood sample representing all time points for each cow within period. Plasma was deproteinized using 4 vol of plasma:1 vol of 15% (wt/vol) 5-sulfosalicylic acid and analyzed for individual free amino acids with a Beckman 6300 amino acid analyzer (Spinco Division, Beckman Instruments, Palo Alto, CA). Only Met and Cys concentration in plasma is reported. Rumen fluid samples were collected from cannulated cows at 0, 3, 6, and 9 h postfeeding on d 21 of each period for analysis of ph, NH 3 -N (Chaney and Marbach, 1962) and VFA concentration (Erwin et al., 1961; Technical bulletin #856A; Supelco, Inc., Bellfonte, PA). Data were examined by analysis of variance using PROC MIXED of SAS (1998), with the following statistical model: where Y ijkl = u + S i + C j (S)i + P k + T l + E ijkl, Y = the dependent variable; u = the overall mean of the population; S i = the average effect for square i; C j (S)i = the average effect of cow j within square i; P k = the average effect of period k; T l = the average effect of treatment l; and E ijkl = the unexplained residual error assumed to be normally and independently distributed. Blood (glucose, NEFA, and urea nitrogen) and rumen fluid (ph, NH 3 N, and VFA concentration) data were analyzed as repeated measurements over time. Therefore, autoregressive covariance (AR 1) was added into the model to account for correlated error. Before establishing the final model, the interactions square treatment (S T) il, square period (S P) ik, and period treatment (P T) kl were tested in a fixed model, but they were not significant sources of variation (P > 0.10) and were dropped from the final model. Nonorthogonal contrasts were MBM versus CTRL, MBM versus MBM + rpmet, MBM versus MBM + HFM, and MBM versus MBM + HFM + rpmet. All data are expressed as least squares means. Differences were not considered significant if P > Data from all periods for two cows that experienced mastitis were dropped from analysis. During period 2, a cow was treated for ketosis, and her data from that period were excluded from statistical analysis. None of the data that were excluded from analysis were from cannulated cows. RESULTS AND DISCUSSION The most recent edition of the NRC (2001) was used to evaluate the five treatments. To evaluate the diets for RUP, RDP, and metabolizable protein (MP) using the new NRC model, it would have been preferable to have extensive analysis of the nitrogen fractions for all feeds. We only conducted CP analysis on MBM, HFM, and alfalfa silage, and they were substituted for values in the NRC (2001) feed library. The feed library

4 was used for other parameters required to estimate duodenal flow of amino acids (NRC, 2001). The RUP values we determined (66.3 and 86.6% of CP for MBM and HFM) were from in situ analysis and based on CP disappearance after 16 h of incubation. In contrast, the NRC (2001) requires estimates of three nitrogen fractions and a disappearance rate of the potentially degradable fraction for the determination of RUP. Therefore, we could not use our estimates of RUP for the diet evaluation and instead relied on NRC (2001) estimates. Intestinal digestibility of RUP was only measured for MBM and HFM, and was determined on residue recovered following a 16-h ruminal incubation period (Calsamiglia and Stern, 1995). Our values were more conservative (lower) than estimates in the NRC (2001) (54.4 and 41.7% vs. 60 and 65% for MBM and HFM), therefore we substituted them for those in the NRC (2001) feed library. Hence, RUP and RDP of diets as well as the second row of CP values (Table 1) were calculated using our laboratory measurements of CP for alfalfa, MBM, and HFM and intestinal digestion coefficients of MBM and HFM; NRC (2001) values were used for all other parameters and feeds. The first row of CP values (Table 1) was obtained from wet lab analysis of the concentrate mixes and alfalfa silage. The CTRL diet provided 16.5% CP, with 4.2% RUP, and 12.3% RDP, respectively (Table 1). The NRC (2001) recommendation for a mature 640-kg Holstein cow fed the CTRL diet and producing 39 kg/d of milk containing 3.8% fat and 2.9% CP is 14.9% CP, 5.0% RUP, and 9.9% RDP (DM basis). Although the CTRL diet was beyond recommendations for CP (NRC, 2001), RUP and MP balance were predicted (NRC, 2001) to be negative ( 209 and 163 g/d, respectively). Consequently, cows on this trial would be predicted to respond to supplemental RUP, if the RUP was not limiting in amino acids. For cows fed CTRL diet, the predicted Met flow to the duodenum was 57 g/d (NRC, 2001). Digestible Met, as a percentage of MP (DMETMP) was estimated to be 2.0% (NRC, 2001). This is considerably less than 2.4%, which was determined to maximize milk protein yield and percentage (NRC, 2001). The prediction of negative MP balance and low value of DMETMP suggests that cows fed the CTRL diet were deficient in Met. Digestible Lys, as a percentage of MP was approximately 7.7%; beyond the amount recommended to maximize milk protein yield and percentage (NRC, 2001). Therefore, Lys should not have been the limiting amino acid in the CTRL diet. Meat and bone meal is a relatively poor source of Met. Supplementing MBM at 4% of DM, primarily at the expense of corn grain, had little effect on predicted Met flow to the duodenum (Table 1). Assuming the Met provided by Smartamine M has 100% ruminal protection and 100% digestibility, adding rpmet to the MBM diet could have increased Met flow 16 g/d and DMETMP to 2.4%. Compared with feeding MBM, feeding MBM plus HFM should have had little impact on Met flow to the duodenum, but could have increased Cys flow by about 13 g/d (assuming Cys flow from microbial and endogenous protein remained constant). Compared with MBM, feeding MBM + HFM + rpmet may have increased total sulfur amino acid flow to the duodenum by about 14 g/d, approximately one-half being contributed by Met and one-half being contributed by Cys. Feeding MBM or HFM did not significantly reduce DMI. Santos et al. (1998) summarized 15 trials to evaluate the effects of replacing soybean meal (SBM) with supplements designed to increase RUP content. These supplements included animal proteins such as blood meal, fishmeal, HFM, and meat meal. These supplements provided an average of 65% of the supplemental protein (the remainder was from SBM). There were no statistically significant differences in DMI when SBM was replaced by the alternative RUP sources in any of the trials. However, values for DMI were consistently greater for cows fed SBM than for cows fed animal proteins (12 comparisons: 19.4 vs kg/ d per cow; Santos et al., 1998). Dry matter intake (Table 2) was significantly (P < 0.05) greater when rpmet was added to a diet containing MBM. This may have been the result of alleviating an amino acid deficiency and increasing milk production. Supplementing HFM or rpmet + HFM to the MBM diet did not increase DMI. Milk yield and 3.5% FCM production (Table 2) were significantly decreased (P < 0.01) by MBM supplementation alone although the cause is not obvious. The addition of rpmet to diets containing MBM increased (P < 0.1) both milk yield and 3.5% FCM, and suggested that the MBM diet was deficient in Met. The milk response we observed was similar to those observed in other studies when rpmet was supplemented (Guinard and Rulquin, 1995; Wu et al., 1997; Overton et al., 1998). Supplementation of HFM alone or HFM + rpmet to the diet containing MBM could not duplicate the milk production response seen with supplementation of rpmet alone. This suggests that feeding Cys as HFM is not an appropriate feeding strategy to alleviate a Met deficiency. Harris et al. (1992) supplemented 14 or 18% CP diets with 0, 3, or 6% HFM and observed a quadratic milk response to HFM when feeding the low protein diets but not the high protein diets. Milk yield was increased 3.7 kg/d when feeding 3% HFM in the 14% CP diet but not the 18% CP diet. However, the experiment was not designed to determine whether responses were due to the provision of a specific limiting amino acid. In another study, replacing 50 or

5 Table 2. Feed intake, lactation performance, and milk composition. Treatment 1 MBM + HFM CTRL MBM MBM + rpmet MBM + HFM + rpmet LSM 2 SE LSM SE LSM SE LSM SE LSM SE Contrasts 2 DMI, kg/d B** Milk yield, kg/d A***, B* 3.5% FCM, kg/d A***, B* Milk protein % B**, C** kg/d A***, B*** Milk fat % kg/d A***, D* 1 CTRL = Control (methionine-deficient diet); MBM = control diet supplemented with meat and bone meal; MBM + rpmet = control diet supplemented with MBM plus rumen-protected methionine (Smartamine M, Antony, France); MBM + HFM = control diet supplemented with MBM plus hydrolyzed feather meal; MBM + HFM + rpmet = control diet supplemented with MBM plus HFM plus rpmet. 2 A = MBM vs. CTRL; B = MBM vs. MBM + rpmet; C = MBM vs. MBM + HFM; D = MBM vs. MBM + HFM + rpmet. 3 Least square means. *P < **P < ***P < % of the MBM in diets with HFM did not affect milk production (Kellems et al., 1989). Milk protein percentage and yield were significantly higher (P < 0.05) when rpmet was supplemented to diets containing MBM. Meat and bone meal is considered low in Met compared with the amino acid profile of ruminal bacteria and milk (Met = 3.7, 5.2, and 5.1% of total essential AA for MBM, ruminal bacteria, and milk, respectively; Schwab, 1994). Therefore, the increase in milk protein may have been due to a more favorable amino acid profile being absorbed from the duodenum when rpmet was added. Several others (Robinson et al., 1995; Armentano et al., 1997; Rulquin and Delaby, 1997; Sniffen et al., 1999; Wu et al.,1999) have observed a similar response when supplementing rpmet to diets fed to lactating cows. Milk protein production is typically more responsive than milk yield when supplementing limiting amino acids (NRC, 2001). The absence of a response in milk protein yield when feeding HFM also suggests that supplementing HFM is not a suitable feeding strategy to alleviate a Met deficiency. Kellems et al. (1989) also did not observe a change in milk protein percentage or yield when substituting HFM for a portion of MBM. Replacing HFM for SBM in isonitrogenous diets causes a decrease in milk protein percentage (Harris et al., 1992), suggesting a less desirable amino acid profile in HFM compared with SBM. Milk fat percentage (Table 2) was not affected by treatments. Milk fat yield results were similar to those for milk yield (Table 2); yield from cows fed the CTRL diet was significantly greater (P < 0.01) than those fed the MBM diet. Cows fed the MBM + HFM + rpmet diet had greater (P < 0.1) milk fat yield than cows fed the MBM diet. Others have reported no differences in milk fat percentage or milk fat yield when substituting HFM for MBM or SBM (Kellems et al. 1989; Harris et al., 1992). Plasma glucose concentration (Table 3) of cows fed the MBM + HFM diet was significantly (P < 0.05) higher than that of cows fed the MBM diet (52.3 vs mg/dl, respectively). The cause for this is not known. Nonesterified fatty acid concentration (Table 3) was significantly (P < 0.05) higher for cows receiving MBM compared with cows receiving CTRL, MBM + rpmet, and MBM + HFM, (213 vs. 174, 163, and 187 µm, respectively). This may have been related to the reduced DMI and plasma glucose for cows fed MBM. Low DMI and low concentration of plasma glucose would lead to low plasma insulin. Insulin acts to suppress lipolysis and reduce plasma NEFA (Sutton et al., 1988). Additionally, rpmet may play a role in lowering NEFA. Blum et al. (1999) conducted an experiment to examine the bioavailability of rpmet in dairy cows and showed that plasma NEFA concentrations were decreased (P < 0.05) after rpmet feeding. Pisulewski et al. (1996) conducted a study to measure the effects of postruminal infusion of Met on lactation performance and plasma metabolites. The treatments were duodenal infusions of 10 g/d of Lys (as control) plus supple-

6 Table 3. Plasma metabolite concentrations. Treatment 1 MBM + HFM CTRL MBM MBM + rpmet MBM + HFM + rpmet LSM 3 SE LSM SE LSM SE LSM SE LSM SE Contrasts 2 Glucose, mg/dl C** NEFA, µm A**, B**, D* Urea N, mg/dl A** Plasma amino acids, nmol/ml Met B***, D** Cys Met + Cys B*, D** 1 CTRL = Control (methionine deficient diet); MBM = meat and bone meal; MBM + rpmet = MBM plus rumen-protected methionine (Rhone- Poulenc, Antony, France); MBM + HFM = MBM plus hydrolyzed feather meal; MBM + HFM + rpmet = MBM plus HFM plus rumen-protected methionine. 2 A = MBM vs. CTRL; B = MBM vs. MBM + rpmet; C = MBM vs. MBM + HFM; D = MBM vs. MBM + HFM + rpmet. 3 Least square means. *P < **P < ***P < mentation of 6, 12, 18, and 24 g/d of Met. Duodenal infusion of Met caused a linear decrease (P < 0.01) in plasma NEFA concentration. They suggested a decrease in plasma NEFA concentrations could be explained by reduced mobilization of body fat reserves or by improved use of NEFA for synthesis of very lowdensity lipoproteins in the liver. Plasma urea nitrogen (Table 3) was significantly higher (P < 0.05) for cows fed MBM compared with cows fed CTRL diet, which had the lowest concentration of plasma urea nitrogen. In general, all other treatments containing animal by- products or rpmet resulted in higher plasma urea nitrogen compared with the CTRL diet. The differences were extremely small; they probably lack biological significance, and can probably be explained by differences in amino acid intake. Cows fed MBM + rpmet had higher plasma concentrations of Met (P < 0.01) and Met + Cys (P < 0.4) than cows fed MBM (Table 3). Cows fed MBM + HFM + rpmet also had higher plasma concentration of Met (P < 0.05) than cows fed MBM. However, Cys concentration was not increased when HFM was supplemented. Table 4. Ruminal fluid composition and ph. Treatment 1 MBM + HFM CTRL MBM MBM + rpmet MBM + HFM + rpmet LSM 3 SE LSM SE LSM SE LSM SE LSM SE Contrasts 2 ph NH 3 N, mg/dl Total VFA 4,mM VFA, mm VFA, mol/100 mol Acetate Propionate Isobutyrate B* Butyrate Isovalerate B* Valerate CTRL = Control (methionine deficient diet); MBM = meat and bone meal; MBM + rpmet = MBM plus rumen-protected methionine (Rhone- Poulenc, Antony, France); MBM + HFM = MBM plus hydrolyzed feather meal; MBM + HFM + rpmet = MBM plus HFM plus rpmet. 2 B = MBM vs. MBM + rpmet. 3 Least squares means. 4 Total VFA = Acetate + propionate + isobutyrate + butyrate + isovalerate + valerate. *P < 0.10.

7 Low digestibility of HFM in the small intestine may have decreased the likelihood of seeing an increase in plasma Cys concentration. To our knowledge, the proportion of absorbed Cys that is catabolized by the ruminant liver is unknown. Ruminal ph and ammonia and total VFA concentrations (Table 4) were similar across all treatments. Molar proportions of acetate, propionate, and butyrate were also not affected by treatment. These results would be expected because diets did not differ greatly in nonfiber carbohydrate content. There were some small but statistically significant decreases in isobutyrate and isovalerate when rpmet was added to the MBM diet. However, the causes of these changes were not obvious and were probably of little biological significance. CONCLUSIONS Supplementing rpmet can enhance the feeding value of MBM for lactating dairy cows. Supplementing Cys as HFM could not duplicate this response. This suggests that feeding HFM may not be a viable strategy for eliminating a Met deficiency in lactating dairy cattle. The absence of an increase in plasma Cys when supplementing HFM precluded evaluation of Cys relative to Met as a sulfur amino acid source for supporting milk production. We do not know why there was not an increase in plasma Cys concentration or a lactation response when supplementing HFM. Possible explanations include, but are not limited to, low intestinal digestibility of RUP or extensive hepatic metabolism of absorbed Cys. ACKNOWLEDGMENTS The authors acknowledge the Fats and Proteins Research Foundation for partial financial support and Rhone-Poulenc (Antony, France) for kindly providing the rpmet used in this study. Great appreciation is also extended to the barn crew at the UW-Madison Dairy Cattle Research Center for animal care and P. M. Crump for assisting with the statistical analyses. The skilled laboratory assistance of Sandra J. Bertics is greatly appreciated. REFERENCES Armentano, L. E Addition of lysine improves lactation performance in cows fed high levels of distiller s dried grains. Pages in Distillers Feed Conference Proceedings. Vol. 51. Orlando, FL. Distillers Feed Research Council, Inc., Ft. Mitchel, KY. Armentano, L. E., and S. J. Bertics Lactation response to rumen protected methionine, or methionine with lysine, in diets based on alfalfa haylage. J. Dairy Sci. 76(Suppl. 1):202. (Abstr.) Armentano, L. E., S. J. Bertics, and G. A. Ducharme Response of lactating cows to methionine or methionine plus lysine added to high protein diets based on alfalfa and heated soybeans. J. Dairy Sci. 80: Association of Official Analytical Chemists Official Method of Analysis. 13th ed. Assoc. Offic. Anal. Chem., Washington, DC. Blum, J. M., R. M. Bruckmaier, and F. Jans Rumen-protected methionine fed to dairy cows: Bioavailability and effects on plasma amino acid pattern and plasma metabolite and insulin concentrations. J. Dairy Sci. 82: Calsamiglia, S., and M. D. Stern A three-step in vitro procedure for estimating intestinal digestion of protein in ruminants. J. Anim. Sci. 73: Chaney, A. L., and E. P. Marbach Modified reagents for determination of urea and ammonia. Clin. Chem. 8: Chung, T. K., and D. H. Baker Maximal portion of the young pig s sulfur amino acid requirement that can be furnished by cystine. J. Anim. Sci. 70: Erwin, E. S., G. J. Marco, and E. M. Emery Volatile fatty acid analyses of blood and rumen fluid by gas chromatography. J. Dairy Sci. 44: Goering, H. K., and P. J. Van Soest Forage Fiber Analysis (Apparatus, Reagents, Procedures, and Application). Agric. Handbook No ARS-USDA, Washington, DC. Guinard, J., and H. Rulquin Effects of graded amounts of duodenal infusions of methionine on the mammary uptake of major milk precursors in dairy cows. J. Dairy Sci. 78: Harris, B., D. E. Dorminey, W. A. Smith, H. H. Van Horn, and C. J. Wilcox Effects of feather meal at two protein concentrations and yeast culture on production parameters in lactating dairy cows. J. Dairy Sci. 75: Johnson, M. M., and J. P. Peters Technical note: An improved method to quantify nonesterified fatty acid in bovine plasma. J. Anim. Sci. 71: Kellems, R. O., K. L. Powell, G. A. Romo, M. V. Wallentine, D. Andrus, and R. Jones Effects of replacing 50% and 100% of the supplemental crude protein derived from meat and bone meal with feather meal on performance of high producing lactating Holstein cows. J. Dairy Sci. 72(Suppl. 1):531. (Abstr.) National Research Council Nutrient Requirements of Dairy Cattle. 7th rev. ed. National Academy Press, Washington, DC. Overton, T. R., L. S. Emmert, and J. H. Clark Effects of source of carbohydrate and protein and rumen-protected methionine on performance of cows. J. Dairy Sci. 81: Pisulewski, P. M., H. Rulquin, J. L. Peyraud, and R. Verite Lactational and systemic responses of dairy cows to postruminal infusions of increasing amounts of methionine. J. Dairy Sci. 79: Raabo, E., and T. C. Terkildsen On the enzymatic determination of blood glucose. Scand. J. Clin. Lab. Invest. 12: Robinson. P. H., A. H. Fredeen, W. Chalupa, W. E. Julien, H. Sato, T. Fujieda, and H. Suzuki Ruminally protected lysine and methionine for lactating dairy cows fed a diet designed to meet requirements for microbial and postruminal protein. J. Dairy Sci. 78: Rulquin, H., and L. Delaby Effects of energy status on lactational responses of dairy cows to rumen-protected methionine. J. Dairy Sci. 77(Suppl. 1):91. (Abstr.) Rulquin, H., and L. Delaby Effects of the energy balance of dairy cows on lactational responses to rumen-protected methionine. J. Dairy Sci. 80: Santos, F. A. P., J. E. P. Santos, C. B. Theurer, and J. T. Huber Effect of rumen-undegradable protein on dairy cow performance: A 12-year literature review. J. Dairy Sci. 81: SAS User s Guide: Statistics, 7th edition SAS Inst., Cary, NC. Schwab, C. G Optimizing amino acid nutrition for optimum yields for milk and milk protein. Pages in Proc. Southwest Nutr. Manage. Conf., Phoenix, AZ. Dept. Anim. Sci., Univ. Arizona, Tucson.

8 Schwab, C. G., C. K. Bozak, N. L. Whitehouse, and V. M. Olsen Amino acid limitation and flow to duodenum at four stages of lactation. 2. Extent of lysine limitation. J. Dairy Sci. 75: Sniffen, C. J., D. S. Tsang, C. S. Ballard, W. Chalupa, W. E. Julien, T. Fujieda, T. Ueda, H. Sato, and H. Suzuki Effect of rumen-protected lysine and methionine supplementation with different sources of metabolizable protein on milk production of high production dairy cows. J. Dairy Sci. 82(Suppl. 1):94. (Abstr.) Sutton, J. D., I. C. Hart, S. V. Morant, E. Schuller, and A. D. Simmonds Feeding frequency for lactating cows: Diurnal patterns of hormones and metabolites in peripheral blood in relation to milk-fat concentration. Br. J. Nutr. 60: Van Soest, P. J Methods for dietary fiber, NDF, and nonstarch polysaccharides. J. Dairy Sci. 73(Suppl. 1):225. (Abstr.) Wu, Z., R. J. Fisher, C. E. Polan, and C. G. Schwab Lactational performance of cows fed low or high ruminally undegradable protein prepartum and supplemental methionine and lysine postpartum. J. Dairy Sci. 80: Weichselbaum, T. E., J. C. Hagerty, and H. B. Mark, Jr A reaction rate method for ammonia and blood urea nitrogen utilizing pentacyanonitrosyloferrate catalyzed Berthelot reaction. Anal. Chem. 41: Wu, Z., C. Le Guilloux, and L. D. Satter Supplementing rumen protected methionine to lactating cows fed different amounts of protein. J. Dairy Sci. 82(Suppl. 1):65. (Abstr.)