Upon starting the mixing sequence, the system collected WS with a belt conveyor from a forage wagon, conveyed it to the batch mixer, and weighed the allotted.
Effects of Feeding Alkaline Hydrogen Peroxide-Treated Wheat Straw-Based Diets on Digestion and Production by Dairy Cows M. G. CAMERON, G.
c. FAHEY, JR.,'
ratio. Cows fed the low and medium wheat straw diets had slightly lower DM intakes but production responses were similar to cows fed the control diet containing alfalfa haylage and corn silage as fiber sources. (Key words: treated wheat straw, ruminal fermentation, hydrogen peroxide)
ABSTRACT
Twelve Holstein cows, averaging 34 d postpartum, were used in three replicat i o n s of a 4 x 4 Latin -s design to determine the effects of feeding different levels of a h l i n e hydrogen peroxidetreated wheat straw on digestion and production responses in lactating dairy cows. Complete mixed diets consisted of 50% concentrate (DM basis) plus varying proportions of treated wheat straw, alfalfa haylage, and corn silage as the forage source. Treatment contained 0 (control). 12.5 (low), 25.0 (medium), or 37.5% (high) treated wheat smw in the diet. Dry matter intakes were 18.5, 17.2, 17.4, and 16.7 kgld for the four treatments, respectively. Apparent digestibilities of DM and OM were decreased (appmxjmately 4.4 percentage units), and NDF and ADF digestibilities were increased by 9.4 and 3.0 percentage units, respectively, with the high wheat straw diet. Yields of milk and 4% FCM, and SNF percentage did not differ among the treatment groups. Milk fat percentage increased (from 3.07 to 3.32%) and milk protein percentage decreased (from 2.61 to 2.56%) as the proportion of treated wheat straw increased in the diet. Cows fed the higher proportions of treated wheat straw had increased ruminal concentrations of total VFA and molar percentage acetate but a decreased molar percentage propionate, resulting in a greater acetate to propionate
INTRODUCTION
Reccivcd January 23, 1990. Accepted June 11, 1990. 'Address correspondence to G.C. Pahey. Jr., Dept of Animal Scienccs, University of Illinois, 126 Animal Sciences Laboratoxy, 1207 W Gregory, Urbana, IL 61801.
1990 J Dairy Sci 73:3544-3554
J. H. CLARK, N. R. MERCHEN, and L. L. BERGER Depattmnt of Animal Sciences University of Illinois U t b w 61801
Historically, altemative high fiber feedstuffs have not often been fed by the US dairy producer. However, as cereal grains increase in demand for direct human use, altemative energy feedstuffs will be used increasingly for livestock produdon, including dairy. Crop residues am high in structural carbohydrates (cellulose, hemicelluloses) and lignin and are major underutilized energy sources for the Nminant. Because corn stover, wheat straw, and soybean residue 8ccount for more than 80% of the 500 million tons of US farm residue (36), the potential for incorporating these residues into the diets of dairy cows is enormous. Improvement of the nutritive value of residues, such as cereal straws, has been the objective of much research. Chemical treatment has the potential to increase nutrient digestibdity and feed intake by the ruminant animal. Alkaline treatments disrupt the plant cell wall by dissolving some hemicelluloses and lignin and by swelling cellulose microfibrils (8). However, diets that contained NaOH-treated roughages fed to lactating dairy cows have resulted in varying effects on DM intake, milk yield, and milk composition when compared with control diets. Soper et al. (32) reported that NaOH-treated corn cobs decreased DM intake and milk yield, but in other studies (6, 17), DM intake and milk yield were increased when cows were fed NaOH-treated roughages. In contrast, other research has indicated no
3544
TREATED WHEAT STRAW FOR COWS
change in DM intake (26) or milk yield (26, 27). Gould (15, 16) demonstrated that a dilute alkaline solution containing H202 could partially delignify lignocellulosic roughages and make the roughages more susceptible to enzymatic and microbial attack, Kerley et al. (19) reported that this alkaline H202 (AHP) treatment of wheat straw (WS) increased degradation of plant cell walls by increasing bacterial colonization and adhesion to fiber particles. The AHP treatment appeared to solubilize a portion of the plant lignin, disrupt certain hemicellulose-lignin bonds, and d u m the ordered structure of the cellulose polymers. Cameron et al. (5) reported that growing wethers fed AHP-WS diets ingested and digested more DM than wethers fed NaOHtreated WS diets. Addition of H2Q to the NaOH resulted in a 15 to 17% increase in digestible nutrient intake by growing wethers compared with WS treated with only NaOH. Therefore, the research conducted with AHPWS (5,19,20,22) suggests that AHP treatment of fibrous by-products has the ability to provide adequate energy and could enhance feed intake and milk yield in lactating dairy cows. The objective of this experiment was to determine the effects of various levels of AHP-WS in the diet on feed intake, nutrient digestion, ruminal fermentation, and milk production responses in dairy cows during early lactation.
3545
three spray nozzles. Upon starting the mixing sequence, the system collected WS with a belt conveyor from a forage wagon, conveyed it to the batch mixer, and weighed the allotted amount of WS. Sodium hydroxide was added as 5.0% of the DM of the WS. Both the water and NaOH pumps were started, and the solution was sprayed onto the WS until the scale reached a set point. The pumps shut off automatically, and mixing continued for 3 min. Upon completion of the 3 min of mixing, H202 was added at 2.0% of the WS DM. Water and H2q2 were sprayed onto the WS until the scale reached the finaI set point, the pumps shut off, mixing continued for 3 min, and the end product was discharged into another forage wagon. The final moisture content of the AH€-WSwas 37.W0 and the product was stored in an oxygen-limiting silo during the experiment. Treatment of native W S with AHP resulted in an approximately 20 percentage unit decrease in NDF and a 12 percentage unit decrease in ADF. Experlmntal Animals and Dlets
Twelve Holstein cows were used in three replications of a 4 x 4 Latin square design. The replicated Latin squares were run simultaneously with the following groups (four cows per group): 1) nuninally cannulated multip m u s cows; 2) intact primiparous cows, and 3) intact multiparous cows. Each replicate contained cows of similar age and stage of lactaMATERIALS AND METHODS tion. Cows averaged 34 d postpartum at the beginning of the experiment and ranged from Treatment of Wheat Straw 16 to 53 d. Initial BW at the onset of the Approximately 113 kg of WS (ground experiment was: replicate 1 = 566 kg, replicate through a IO-mm screen) was treated in each 2 = 498 kg, and replicate 3 = 634 kg. During batch. The native WS contained 92.2% DM, the trial, cows were housed in conventional 91.2% OM, 84.8% NDF, 61.0% ADF, and stanchions, fed daily at 1100 and 2200 h, and 9.0% acid detergent lignin (ADL). The AHP- allowed to exercise from 0800 to lo00 h. The WS required for the experiment was prepared ruminally cannulated cows were not allowed to using a 3041 Marion Batch Mixer (Marion exercise during the collection phase (last 3 d) Mixers, Jnc., uarion, IA) in conjunction with a of each period. Toledo 8142 Digital Indicator (Toledo Scale, All cows were fed a complete mixed diet of Westerville, OH) and load cell system. The 50% forage and 50% concentrate (C) on a DM Toledo 8142 was programmed to start and stop basis, with varying levels of AHP-WS, alfalfa the appropriate motors and pumps that deli- haylage (AH), and corn silage (CS) as the vered the WS, water, and chemical reagents to sources of forage. The ratio of AH and CS in the batch mixer in weighed amounts for mix- each diet was maintained in approximately a 3: ing. Hydrogen peroxide, NaOH, and water were 1 ratio. The four diets were 1) 38.4% AH, pumped into the batch mixer and applied via 12.2%CS,0%AHP-WS,49.4%C (control); 2) J o d of Dairy Science Vol. 73, No. 12, 1990
3546
CAMWON ET AL.
TABLE 1. Ingredient d chemical composition of dicta as a percentage of
Ingredients Alfalfa haylagc corn silage! AHP-ws
38.4 12.2
the
DM.
28.6 9.3 125
25 .O
9.6 3.1 37.5
33.7 14.0
295 17.8
25.1 21.8
19.2 6.1
concentrate mix G m d ahelled Soybun meel
Corn
Sodium bicarbonate
--vimix2 Dicalcium phosphate LimCStone
Magnesium oxide Calcium sulfate Potassium chlonide chemical aaalysis
OM CFJ ADP
NDF
37.1 10.2 .a3 .15
.a3 .44 .15
.a2 .14 92.2 17.6 18.0 30.8
.15 .67 .74 .16 .13 .14 91.3 17.9 21.1 33.4
.15 .74 1.05 .17
.15 33 1.35
20
.I8 .24
.18
.42
90.6 17.8 24.0 36.1
89.3 17.6 26.7 38.5
lAlkaline hydrogen peroxidetreated wheat straw. %onrains I, .W% Fe,2,096; Zn, 3.096; Mn, 3.096; M g , 5.W; Cu, 5%;Co, .004% Se, .015% S, 10.046,K, 7.5% vitamin A, 2200 nr/g, vitamin Dj, 660 N/g; vitamin E, 7.7 W g .
a d CS contained 19.1 and 9.0% CP, respectively Vable 2). Experiment periods were 14 d. The first 3 d of each period were used to gradually change The complete mixed diet was available ad libi- the cows from one experimental diet to another. tum,allowing for at least 10% om. Water was On d 1 of each period, cows were fed 75% of available continually. Ingredient and chemical the previous period and 25% of the current composition of diets and individual feedstuffs period diet, on d 2 the diets were 5050, on d 3 fed to lactating cows is presented in Tabla 1 they were 2575, and by d 4, cows were fed and 2. Minerals were added to each diet to meet 100% of the diet for that period. The first 9 d or exceed NRC (2.5) recommendations. Sodium were used for adjustment to diets followed by 5 bicarbonate was not included in diets contain- d of sample and data collection. ing AHl'-WS as recommendations for Na were met due to the high level of Na pnsent in the Dry Matter Intake, Mllk Production, AHP-WS.Limestone and KCl were increased end Mllk Composltlon in the diet as AHP-WS in the diet was inDry matter intake was measured daily during creased because AHP-WS contains little Ca or K. Percentage CP varied slightly among diets. d 10 to 14. Sampks of forages and feed The variation in OM,NDF, and ADF contents refusals were obtained on d 10 to 14 and ~ 8 m among the dids was primarily due to the vary- plcs of concentrate were taken on d 12. Saming concentrations of AHP-WS m the diet. The ples were dried at 55'C, ground in a Wiley mill AHP-WS is higher in structural carbohydrates (2-mm scnen), and cornposited at the end of than is either AH or CS (Table 2). Forages fed each period. Feed and orts were analyzed for during the experiment were of high quality;AH DM and OM (1). Nitrogen was measured using 28.6% AH, 9.3% CS, 12.5% AHP-WS,49.6% C (low); 3) 19.2% AH, 6.1% CS,25.0% AHPWS, 49.7% C (medium); and 4) 9.6% AH, 3.1% CS, 37.5% AHP-WS,49.8% C (high).
Journal of Dairy Sciance VoL 73, No. 12, 1990
3547
TREATED WHEAT STRAW POR COWS
TABLE 2. chemical composition of feeds as a pacentage of DM. Ingredient1
DMZ
OM
CP
NDF
ADF
ADL3
Alfal€a hayIage
sa2
sY
492 67.9
89.1 w.7 87.4
19.1 9.0 3.2
46.7 445 63.4
37.8 265 53.9
85 6.6 7.9
86.6 87.6 87.4 87.6
93.8 93.1 922 90.6
19.1 22.4 25.6 28.9
13.2 13.4 12.8 125
4.1 4.5 4.6 4.7
AHP-W Concentrate mix con(r01 Low AHP-ws Medilrm AHP-WS High AHP-ws
IMean particle size for alfalfa haylage, 2735 ptq corn silage, 3390 ptq *DM on as-fed basis. 3Acid detergent lignin. 4Alkaline hydrogen peroxide-treated wheat straw.
the Kjeldahl method (1). Neutral detergent fiber was measured using the a-amylase procedure of Robertson and Van Soest (28). Acid b r gent fiber and ADL were measured according to Goering and Van Soest (13). Cows were milked at o600 and 1700 h. and milk weights were recosded daily. Milk samples were taken at each milking on d 11 and 12 of each period. Daily am. and p.m. milk samples were cornposited based on milk yield and analyzed for SIW (14). milk p t e i n , and milk fat (Tnfrared analysis. Dairy Lab Service,Inc., Dubuque, IA). Body weights were taken on d 1 and 14 of each perid Rumen Sarnpllng and Analysls
On d 12 of each collection period, approximately 2.3 kg of the feed offered to each ruminally cannulated cow was marked with 3 g of Yb (7.32 g of YbAcy4H20). The Yb was dissolved in 150 ml of &tilled water, and the resulting solution was s p y e d and mixed slowly with the 2.3-kg podon of the diet. Ruminal jmticulate umtents were taken from the rumen at 4,lO, 16,22,28,34,40, and 46 h following administration of the Yb marker to determine particulate (Yb) dilution rate. Particulate matter was dried at 55'C and ground through a 2-mm screen before Yb analysis. Ruminal particulate samples were prepared and analyzed for Yb using an atomic absorption spectrophotomekr (perkin-Elmer 2380, Norwalk, 0 as described by Ellis et aL (9).
.4 .5
.9 12
m - w s , 1018 pm.
particulate dilution rate was deterrnined as the slope of the regression of the natural logarithm of Yb concentration on time. On d 14 of each period, the ruminally-cannulated cows wete dosed inmmmmall ' y at 0800 h with 50 g O-EDTA dissolved in 500 ml of water as described by Uden et al. (35) to determine Nminal fluid dilution rate. The C o - m A solution was placed in several locations in the rumen via a plastic tube attached to a plastic funnel. Ruminal fluid samples (50 ml) were taktn hourly for 24 h beginning at 0800 h from several locations in the rumen at each sampling time via a suction pump. Ruminal pH measurements were taken immediately using a Beckman 31 pH meter (Beckman Instruments, Inc., Palo Alto, CA). Ruminal fluid samples were strained through eight layers of cheesecloth after sampling, centrifuged at 10,000 x g for 15 min, and the supernatant aciMied with 1 ml of 5096 sulfuric acid @Hd) and refrigerated until analysis of Co, WA, and ammonia. Ruminal fluid ammonia concentrations were determined according to the procedure of Chaney and Marbach (7).Samples for Co and VFA analyses were prepared by centrifuging the ruminal fluid sample at 18,000 x g for 20 min. W a l t content of the ruminal fluid was measured with an atomic absorption spectrophotometer. Fluid (Co) dilution rate was calculated as the slope of the regression of the natural logarithm of Co concentration on time. The VFA in ruminal fluid were prepared for Joamal of Dairy Science Vol. 73, No. 12, 1990
3548
CAMERON ET AL.
gas chromatographic analysis following the procedure of Erwin et al. (11). Samples were analyzed for VFA using an automated Hewlett Packard Model 5890 (Hewlett Packd, Inc., Palo Alto, CA) gas chromatograph equipped with a hydrogen flame ionization detector. A glass column (approximately 210 cm x 8 mm 0.d.) packed with Chmmosorb WAW (80 to 100 mesh) coated with 1% phosphoric acid was used. Total Tract Apparent Dlgestlbllltles
Cows (replicate 1) were dosed via the ruminal cannula with 10 g 0 2 0 3 twice daily at 12-h intervals on d 5 to 14 of each period to determine apparent digestibility of the diets. Fecal grab samples were obtained twice M y (d 10 to 14), and 100 g of each sample (wet basis) were composited for each cow during the collection phase and dried at 55'C. Dried fecal samples were ground in a Wiley mill (2-mm screen) and assayed for DM, OM, N, NDF, and ADF as described previously. Chromium concentration of feces was determined by atomic absorption spectrophotometry (37). Statlstlcal Analysls
Data were analyzed by ANOVA for a Latin square design according to the General Linear Models procedure of SAS (30). Model sums of squares for the Latin square were animal, period, and diet effects. Model sums of squares for the replicated Latin Square were separated into replication, cow within replication, period, and diet effects. Sums of squares for diet effects were sepatated further into orthogonal contrasts for linear and quadratic effects. One observation from the control diet (period 4) for the intact multiparous replication was lost for reasons unrelated to treatment. Least squares means were calculated for the replicated Latin square design for DM intake, change in BW, milk production, and milk composition. Least squares means were calculated for apparent nutrient digestibility and ruminal fermentation data for replicate 1 (nuninally cannulated cows) only. RESULTS AND DISCUSSION Feed lngredlent and Dlet Composltlon
Ingredient and chemid composition of diets and individual feedstuffs fed to cows in Journal of Dairy Science Vol. 73, No. 12. 1990
early lactation is presented in Tables 1 and 2. The control diet consisted of feedstuffs (AH, CS, gromd shelled corn, and soybean meal) that are typically fed and readily available to dajl producers in the Midwest. The AHP-WS padally replaced the AH and CS portions of the control diet at 12.5, 25.0, and 37.5%, respectively, in the other three diets. Because AHP-WS supplies mostly energy (it is nearly devoid of protein), its inclusion in the diet resulted in the use of varying amounts of ground shelled corn, soybean meal, and minerals to meet NRC (25) recommendations. Percentage CP varied little among diets, ranging from 17.6 to 17.9%, which supplied an adequate amount of d i e m CP for cows in early lactation. The concentrate mix fed to COWS eating the control, low, medium, and high AHPWS diets contained 19.1, 22.4, 25.6, and 28.9% CP respectively; thus, the complete mixed diets contained at least 17.0% CP. It is important to supplement AHP-WS diets with dietary CP sources that will establish an optimal N:energy ratio,which will meet the requirements of both the host animal and the ruminal microbes. Addition of AH and soybean meal to the AHPWScontaining diets will provide ruminal degradable and ruminal escape protein fractions that may increase ruminal digestion of AHPWS and yet allow for both bypass and microbial protein flows to the small intestine. Percentages of NDF and ADF increased with increased levels of AHP-WS in the diet. Neuasll detergent fiber increased from 30.8% for the control diet to 38.5% for cows fed the high AHP-WSdiet, an increase of 7.7 percentage units. The change in the percentage of ADF followed a similar pattern with an increase of 8.7 percentage units from the control to the high AHP-WS diet. It appeared that the AHPWScontaining diets contained sufficient fiber to ensure normal ruminal function and help prevent milk fat depression. Table 2 indicates that the percentages of NDF and ADF in AHPWS were noticeably higher than the percentages in either AH or CS. Dry Matter Intake, Mllk Yleld and Composltlon Dry matter intakes by cows used in the replicated Latin square design were 18.5, 17.2, 17.4, and 16.7 kg/d for the control, low, medi-
3549
TREATED WHEAT STRAW FOR COWS TABLE 3. Least .quam means for DM intake, change in BW. and lactation performsncc.1 ~
~~
Diet
Significance of effect (P.20.
um, and high AHP-WSdiets, respectively (Table 3). The DM intake decreased linearly (P= .13) when the amount of AH€-WSwas increased in the diet. Dry matter intake for the nuninally cannulated cows (Table 4) was not different (b.20)among the diets. Consumption of DM, when expressed as a percentage of BW, decreased linearly (P= .15)from 3.48% for the control to 3.13% for the high AHP-WS diet (Table 3). It appears that increasing AHP-WS in the diet of cows during early lactation slightly decreases DM intake. Body weight change was not altered (b.10) in by feeding increased amounts of AH€-WS the diet. However, most cows were losing BW and were apparently mobilizing body reserves to support milk production during early lactation. These data suggest that cows fed the control and low AHP-WS diets mobilized a smaller quantity of body reserves (-11.2 and -4.0 kg) than cows fed medium and high AH€WS diets (-17.7 and -16.9 kg). Phipps et al. (26) reported similar results in that cows fed grass and maize silage-based diets had higher DM intakes (13.4vs. 12.5 kg/d) and greater liveweight gains (.14vs. -.15 kgld) compared with cows fed grass silage and NaOH-treated barley straw diets. These researchers suggested
the grater loss of liveweight by cows fed the grass silage and NaOH-treated barley straw diets was because of excessive tissue mobiliation in an attempt to sustain milk production in early lactation. Other studies (6, 17) reported that NaOH treatment of low quality roughages resulted in an improved DM intake and milk yield when compared with control diets containing the untreated roughages. Feeding the higher levels of AHP-WSdid not affect (f5.10)yield of milk, 4% FCM, milk fat, or SNJ? (Table 3). Milk production ranged from 30.0 kg/d for cows fed the high AH€-WS diet to 31.2 kg/d for cows fed the control diet, a difference of 1.2 kgld. Milk fat percentage increased linearly (P= .14),milk protein percentage decreased (P= .1l), and milk protein yield decl.eased (P = .11) when increasing amounts of AHP-WS were fed to the cows. Milk fat percentage increased from 3.07% for cows fed the control diet to 3.32% for cows fed the high AHP-WSdiet. This is consistent with results of other studies (4,12, 17,26) in which inclusion of untreated or NaOH-treated roughage in the diet resulted in improved milk fat percentages. Milk pmtein percentage and yield decreased as more AHP-WSwas added to the diet. There is evidence that as dietary crude fiber or ADF in that
J o d of Dairy Science Vol. 73, No. 12. 1990
3550
CAMERON ET AL..
TABLE 4. Least squares
m e a ~ dfor
intake and .pparent digestibility of rm~& by lactating cows.’ S i g n i f i of ~
Diet Item
AHP-ws* Medium
&st
(P
