| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Agriculture and Agri-Food Canada, Research Centre, Lethbridge, AB, T1J 4B1, Canada
1 Corresponding author: beauchemink{at}agr.gc.ca
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Key Words: physically effective neutral detergent fiber • chewing • digestion • dairy cow
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
It is unclear which measure of peNDF and what peNDF contents provide the most accurate estimation of chewing activity and ruminal pH in dairy cows, although a number of studies have been conducted in recent years (Yang et al., 2001; Krause et al., 2002b; Beauchemin et al., 2003; Kononoff et al., 2003b; Yansari et al., 2004; Beauchemin and Yang, 2005). These different methods of peNDF estimation result in very different values (Yang et al., 2001; Beauchemin et al., 2003; Einarson et al., 2004). Yansari et al. (2004) showed that chewing activity per unit intake of peNDF was consistent across diets varying in particle length when estimated using systems that incorporated a 1.18-mm sieve (i.e., peNDF1.18 or peNDFps-3s). A consistent ratio of chewing activity to peNDF is desirable in terms of predicting chewing time based on peNDF intake. However, in some studies, peNDFps-3s did not differentiate diets that differed in forage particle length (Kononoff and Heinrichs, 2003b).
Increased chewing activity as a result of increased peNDF intake can increase ruminal pH, and help minimize ruminal acidosis in dairy cows (Krause et al., 2002b; Yansari et al., 2004). However, this is not always the case; chewing activity can also increase with increasing intake of peNDF without affecting ruminal pH, particularly when diets contain highly fermentable carbohydrate sources (Kononoff et al., 2003b; Kononoff and Heinrichs, 2003a; Beauchemin and Yang, 2005). Thus, it is not clear whether systems of measuring peNDF are useful for predicting both chewing and ruminal pH.
A number of studies on the effects of peNDF on feed intake, digestibility, and milk production and composition have also been conducted (Krause et al., 2002a; Kononoff et al., 2003a; Plaizier, 2004; Yansari et al., 2004; Yang and Beauchemin, 2005). However, the results have been inconclusive due to differences in measuring peNDF.
The objectives of the present study were to determine the effect of increasing the peNDF content of a diet containing CS on feed intake, chewing activity, ruminal pH and fermentation, microbial protein synthesis, digestibility, and milk production in lactating dairy cows. The second objective was to determine the most appropriate method of determining peNDF for use in diet formulation to promote chewing and prevent ruminal acidosis. Thus, the peNDF content of the diets was determined using the PSPS with 2 and 3 sieves, and by expressing the values as a proportion of DM and NDF, and using a dry sieving technique as particles retained on a 1.18-mm sieve (Mertens, 1997).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
. These analyses were conducted commercially by Cumberland Valley Analytical Service, Inc. (Maugansville, MD).
|
Cows were offered 1 of 3 diets, which were chemically identical with approximately 54% of concentrate and 46% of corn silage (CS; Table 2), but differed in peNDF level: high, medium, and low. The 3 dietary peNDF levels were obtained using CS differing in particle length: long silage (high), medium silage (medium), and short silage (low) (Table 3). The diets were formulated using the Cornell-Penn-Miner System (CPMDairy, Version 3.0.4a; Cornell University, Ithaca, NY; University of Pennsylvania, Kennett Square, PA; and William H. Miner Agricultural Research Institute, Chazy, NY) to supply adequate ME and MP for a 650-kg cow producing 35 kg/d of milk containing 3.5% fat and 3.2% protein. Each period consisted of 11 d for adaptation to diets and 10 d for experimental measurements.
|
|
Meal Duration and Feeding Behavior
Feeding behavior was monitored for 48 h on d 13 to 15 of the period. Mangers attached to load cells (Omega Engineering Inc., Stamford, CT) that were connected to a computer were used to measure the feed weight over time. An average weight was obtained every 11 s, and stored using Collect software (Labtronics, Inc., Guelph, ON, Canada). During feeding activity, the weight of the manger decreased. Feeding activity was separated into meals using a meal criterion of 27 min as outlined by DeVries et al. (2003). A meal criterion is the minimum time interval between 2 meals and is used to determine meal frequency and meal duration. A meal was defined as eating activity greater than 30 s and more than 300 g of feed being removed from the feeder. Rate of DMI was calculated as the ratio of DM ingested and duration of the meal.
Chewing Activity
Chewing activity was measured concurrently with feeding behavior. Cows were fitted for 48 h with leather halters that measured jaw movements. Each halter contained a piezo disk (Edmund Scientific Company, Barrington, NJ), which was inserted within the halter, and positioned under the jaw. Chewing action places stress on the disk generating an electrical signal, which is then processed and counted as a single jaw movement. A datalogger (Campbell Scientific, Inc., Logan, UT) was used to receive the output signal from each cow. The number of jaw movements was summed each minute and stored for subsequent analysis. The jaw movements were then designated as eating or ruminating chews based on several criteria. Jaw movements were considered as eating chews if they occurred during a meal (i.e., duration of the eating activity was greater than 30 s and more than 300 g of feed was removed from the feeder during the meal). Jaw movements were designated as ruminating chews if they did not occur during a meal and rate of chewing exceeded 30 chews/min. All other jaw movements were considered associated with licking, drinking, and grooming and were not included in the eating or ruminating categories. Total number of chews was calculated as the sum of eating and ruminating chews. Total time spent eating, ruminating, and chewing (eating + ruminating) was based on the duration of chewing activity.
Ruminal pH and Fermentation
Ruminal pH was monitored for 48 h concurrent with measurements of feeding behavior and chewing activity. The pH was measured using an industrial electrode (model S650-CDHF, Sensorex, Garden Grove, CA) linked to a pH controller (model PHCN-37, Omega Engineering Inc.) as described in detail by Penner et al. (2006). The electrodes were suspended approximately 60 cm in the rumen using a cable that was anchored to the ruminal cannula plug. The electrode was weighted down to ensure submersion within the rumen contents. This indwelling continuous ruminal pH measurement device results in values that are about 0.05 units lower than manual measurements of pH taken from the same location within the rumen (Penner et al., 2006). The electrodes were covered with a perforated guard to prevent them from coming into direct contact with the rumen wall. The pH electrodes were removed from the rumen for 20 min daily between 1400 and 1430 h for calibration using pH 4.0 and 7.0 standards. Continuous measurements from the indwelling probe were sent to a datalogger (Campbell Scientific, Inc.) every 5 s and averaged every 5 min. Ruminal pH data were summarized daily for each cow as mean pH, minimum pH, area between the observed pH and a line drawn at pH 5.8 or 5.5, and time (h) under pH 5.8 or 5.5. The minimum pH value for each day and each cow was obtained from the raw input data using PROC MEANS (SAS Institute, 1996). The area was calculated by adding the absolute value of negative deviations in pH from 5.5 or 5.8 for each 5-min interval. A ruminal pH of 5.8 was chosen as benchmark because this threshold value is often used to indicate subclinical ruminal acidosis. The pH of 5.5 was used to further categorize subclinical ruminal acidosis in terms of severity.
Ruminal fluid was collected on d 13 at 0900, 1300, and 1600 h from multiple sites in the rumen for VFA and NH3 N determination. Samples were immediately squeezed through 4 layers of cheesecloth with a mesh size of 250 µm. Five milliliters of filtrate was preserved by adding 1 mL of 25% HPO3 and used to determine VFA, and 5 mL of filtrate was preserved by adding 1 mL of 1% H2SO4 and used to determine NH3 N. The samples were subsequently stored frozen at –20°C until analyzed.
Apparent Digestion in the Total Tract
Apparent total tract digestion of nutrients was measured using YbCl3 (Rhône-Poulenc Inc., Shelton, CT). Ytterbium product was mixed into the concentrate portion of the diets at a rate of 400 g of YbCl3/t of concentrate to achieve intake of approximately 1 g of Yb/d per cow. Fecal samples (approximately 200 g, wet weight) were collected for each cow from the rectum twice daily at various times during the last 6 d of the period. Samples were immediately subsampled (about 50 g), composited across sampling times for each cow and each period, dried at 55°C, ground to pass a 1-mm sieve (standard model 4, Arthur Thomas Co.), and stored for chemical analysis. Apparent nutrient digestibility in the total tract was calculated from concentrations of Yb and nutrients in the diet, orts, and feces using the following equation: apparent digestibility = 100 – [100 x (Ybd/Ybf) x (Nf/Nd)], where Ybd = Yb concentration in the diet actually consumed (i.e., after adjustment for Yb concentration of orts), Ybf = Yb concentration in the feces, Nf = concentration of the nutrient in the feces, and Nd = concentration of the nutrient in the diet actually consumed.
Microbial Protein Synthesis
A mixed ruminal liquid and solid sample (about 0.8 L) from various locations of the rumen was obtained daily from each cow during the last 5 d of the period. Samples were squeezed immediately through 4 layers of cheesecloth. Ruminal particles obtained by squeezing were blended (400 g of particles plus 400 mL of 0.9% NaCl) in a Waring blender (Waring Products Division, New Hartford, CT) for 1 min, and then squeezed through 4 layers of cheesecloth. Filtrates from squeezed and strained homogenate were mixed, centrifuged (800 x g for 10 min at 4°C) to remove protozoa and feed particles, and the supernatant was centrifuged (27,000 x g for 30 min at 4°C) to obtain a mixed ruminal bacteria pellet. Bacterial isolates were accumulated by period for each cow, freeze-dried, and ground using a ball mill (Mixer Mill MM2000; Retsch, Haan, Germany) to a fine powder for chemical determination.
Total urine was collected from each cow on d 16 to 18 using indwelling Foley catheters (26 French, 75-cc ballon; C. R. Bard, Inc., Covington, GA), which were inserted on d 16 of each experimental period. Fresh containers with 600 mL of 4 N H2SO4 were attached to catheters at 1100 and 2300 h to obtain daily samples (final pH <3). Urine output was measured every 12 h for 2 d, and then 2 sets of 3-mL aliquots were taken, diluted to 15 mL with distilled water, and stored at –20°C for analysis of allantoin and uric acid.
Total purine derivatives (PD) excreted (mmol/d) were estimated as the sum of uric acid and allantoin. Excretion of the endogenous PD was assumed constant at 0.385 mmol/kg of BW0.75 for individual cows (Chen and Gomes, 1992). Purine absorption of microbial origin was calculated as: (total PD excreted – endogenous PD)/ 0.85 (Chen and Gomes, 1992). Synthesis of microbial N within the rumen was calculated as: (purine absorption x 70)/(purine N:total N in mixed rumen bacteria x 0.83 x 1,000) (Chen and Gomes, 1992). The average purine N:total N in mixed rumen bacteria measured in this study was 0.129.
Chemical Analyses
Dry matter content of the ingredients (silage, grain, concentrate mix) and TMR was determined by oven drying at 55°C for 48 h. Analytical DM content of oven-dried samples was determined by drying at 135°C for 3 h (AOAC, 1990). The OM content was calculated as the difference between DM and ash contents, with ash determined by combustion at 550°C overnight. The NDF and ADF contents were determined using the methods described by Van Soest et al. (1991) with amylase and sodium sulfite used in the NDF procedure. Starch was determined by enzymatic hydrolysis of 
-linked glucose polymers as described by Rode et al. (1999). Content of Yb in the TMR, orts, and fecal samples was determined using inductively coupled plasma optical emission spectroscopy according to the AOAC method (1990) modified such that no KCl was used during sample digestion. Content of N in the samples was determined according to the method of Smith and Tabatabai (2004) by flash combustion, chromatographic separation, and thermal conductivity detection (Carlo Erba Instruments, Milan, Italy). Ruminal VFA were separated and quantified by gas chromatography (Varian 3700; Varian Specialties Ltd., Brockville, ON) using a 15-m (0.53-mm i.d.) fused silica column (DB-FFAP column; J and W Scientific, Folsom, CA). Ammonia content of ruminal samples was determined using the method described by Weatherburn (1967) modified to use a plate reader. Allantoin in urine was determined by autoanalyzer using the procedure of Pentz (1969) with modifications by Lindberg and Jansson (1989). Uric acid was determined using a commercial kit (no. 292; Sigma Chemical Co., St. Louis, MO). Purine content in the bacterial pellet was determined using the procedure of Zinn and Owens (1986).
The pef of CS and TMR were determined using dry sieving techniques. Firstly, pef was determined using the PSPS as the proportion of DM retained on 2 sieves: 19 and 8-mm (pefps-2s; Lammers et al., 1996) or on 3 sieves: 19, 8, and 1.18-mm (pefps-3s; Kononoff et al., 2003a). The NDF content of the original sample and of all materials retained on the sieves was measured. The pef was also determined using a vertical oscillating sieve shaker (Analysette 3; Fritsch, Oberstein, Germany) equipped with a stack of sieves (W. S. Tyler, Inc., Mentor, OH) arranged in descending mesh size. Sieve mesh sizes for the shaker were 9.5, 6.7, 3.35, 1.18, 0.6, and 0.15 mm. Approximately 15 g of feed was placed on the top sieve, and the stack of sieves was shaken until the distribution of material did not change (approximately 10 min). In this case, the pef was determined based on the proportion of DM retained on the 1.18-mm sieve (pef1.18), according to Mertens (1997).
The peNDFps-2s, peNDFps-3s, and peNDF1.18 were calculated by multiplying NDF content of the feed by the pefps-2s, pefps-3s, and pef1.18, respectively. The peNDF was also calculated (peNDFps-2sndf and peNDFps-3sndf) as the proportion of NDF retained on 2 and 3 sieves of the PSPS.
Statistical Analyses
The data were analyzed using the mixed model procedure of SAS (SAS Institute, 1996) to account for effects of square, period within square, cow within square, and treatment. The treatment was considered a fixed effect; square, period within square, and cow within square were considered random effects. Estimation method was restrictive maximum likelihood (REML) and the degrees of freedom method was Kenward-Rogers. Data for feeding behavior, chewing activity, and ruminal pH were summarized by day and then analyzed using the same mixed model but with day included as a repeated measure using compound symmetry. Similarly, data for VFA and ammonia were analyzed by sampling time using repeated measures techniques. Data for particle distribution, pef, and peNDF of forages and diets were averaged by period and analyzed by including particle length as a fixed effect and period as a random effect. Linear and quadratic orthogonal contrasts were tested using the CONTRAST statement of SAS. Pearson correlation coefficients were estimated using the CORR procedure of SAS. Effects of the factors were declared significant at P < 0.05 unless otherwise noted; trends were discussed at P < 0.15.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
). Overall, the silages were very similar in nutrient profiles, even though there were some small statistically significant differences detected. Specifically, contents of DM, NDF, and ADF of CS were quadratically affected by TCL with no effect on starch and CP contents. Higher DM content of medium-cut CS compared with long- and short-cut CS was unexpected because it was harvested first. Reduced TCL of CS from long to medium tended to decrease NDF (P < 0.06) and ADF (P < 0.11) contents, with no further reduction in NDF and a small increase in ADF content as TCL was reduced from medium to short. Decreased NDF content by decreasing TCL of CS was reported in several studies (Bal et al., 2000; Schwab et al., 2002). However, those studies also reported a concomitant change in starch concentration that was not observed in the present study. It is likely that the observed variations in nutrient content of CS were caused by sampling. In support of this theory, Bal et al. (2000) noted that sampling was more uniform for finely chopped CS than for longer chopped silage. Variations in chemical composition of the diets reflected those of the CS used because the diets differed in the TCL of CS (Table 2).
Particle Length and Physically Effective Fiber
As expected, the values for pef and peNDF obtained using the PSPS with 3 sieves (pefps-3s, peNDFps-3s, and peNDFps-3sndf) were greater than those obtained using 2 sieves (pefps-2s, peNDFps-2s, and peNDFps-2sndf), because of the materials retained on the 1.18-mm sieve (Table 3). In general, particle size distribution, pef, and peNDF of CS measured using the PSPS with either 2 or 3 sieves decreased linearly or quadratically with decreased TCL. However, there was a much smaller range in pef of CS when 3 sieves, rather than 2, were used (range of 0.93 to 0.96 vs. 0.41 to 0.72). As a result, the range in peNDFps-2s and peNDFps-2sndf with decreasing TCL of CS was greater than that of peNDFps-3s and peNDFps-3sndf. Thus, using the PSPS with 3 sieves, expressed on the basis of proportion DM or NDF, did not greatly differentiate the peNDF content of diets varying in chop length of silage.
The present results are in agreement with other reports that used the new PSPS to measure particle distribution of CS. Soita et al. (2005) reported that almost all (99%) the sample was retained by the PSPS when the 1.18-mm sieve was used, and furthermore, there was no differentiation between long- (19.1 mm) and short- (9.5 mm) cut CS. Similarly, no difference in pef was obtained for long- (22.3 mm) and short- (4.8 mm) cut CS in the study by Kononoff and Heinrichs (2003b). In contrast, distinct differences in pef were observed for alfalfa hay measured using the new PSPS for forages with a TCL of 19, 10, and 2 mm (Yansari et al., 2004). It appears that the new PSPS may not be suitable for determining the pef of some forages, including CS.
The higher values for peNDF when expressed on the basis of fractional NDF (peNDFps-2sndf and peNDFps-3sndf) than on the basis of total NDF (peNDFps-2s or peNDFps-3s) were expected due to the higher NDF content of the material on the 19- and 8-mm sieves relative to the total NDF content of CS (Kononoff and Heinrichs, 2003b).
Quadratic effects of TCL on pef and peNDF indicated that the increase in the proportion retained on the 19-and 8-mm sieves was substantial from short (4.8 mm) to medium (15.9 mm) cut, but there was virtually no further increase for further increasing TCL from medium to long (28.6 mm) cut. In fact, Kononoff and Heinrichs (2003b) reported an 18% increase in pefps-2s (from 0.72 to 0.85) for CS when TCL increased from 4.8 to 22.3 mm, whereas there was no increase in the proportion of long particles (retained by an 8.98-mm sieve) when the TCL of CS increased from 19 to 32 mm (Onetti et al., 2003).
The PSPS pef1.18 measured using dry sieving with a vertical oscillating sieve shaker, was also quadratically increased (P < 0.06) with increasing TCL of CS, but pef1.18 was lower than pefps-3s, especially for short-cut CS. Yansari et al. (2004) also found that estimates of pef1.18 and pefps-3s were different for CS (pef1.18 vs. pefps-3s; 0.83 vs. 0.92), even though estimates were similar for alfalfa hay. Although both these systems of measuring pef incorporate a sieve with mesh size of 1.18 mm, the PSPS sieve consists of nonadjacent round openings, whereas the sieve used with the oscillating shaker consists of adjacent square holes that allow particles to tip through on the diagonal. Thus, longer particles are expected to pass through the 1.18-mm sieve of the oscillating shaker compared with the 1.18-mm sieve of the PSPS, accounting for the differences in pef values obtained using these 2 techniques. Consequently, the pef1.18 differentiated the CS of varying TCL to a greater extent that did the PSPS with 3 sieves.
For the diets offered, particle size distribution, pef, and peNDF reflected the variations in TCL of the CS (Table 4). As observed with CS, the proportion of the TMR retained on the 19- and 8-mm sieves, and pefps-2s was substantially increased from the low to the medium peNDF diets and then moderately increased from the medium to the high peNDF diets. In contrast, the increase of pefps-3s with increasing dietary peNDF was small with only a difference (P < 0.06) detected between the medium and high peNDF diets. These data confirm the study of Kononoff and Heinrichs (2003b) in which dietary peNDFps-3s contents were similar even though the proportion of CS in the diets varied from 45 to 57%. In contrast, with diets containing alfalfa hay, dietary peNDFps-3s contents were increased from 17.2 to 22.6 to 25.1% for diets containing fine, medium, and long hay, respectively (Yansari et al., 2004). It appears that the new PSPS may not adequately measure the pef of CS-based diets, although it may distinguish the pef of hay-based diets. In our study, values for dietary pef1.18 and peNDF1.18 contents were intermediate of the 3 systems used, but the values did not differentiate among diets formulated using CS varying in TCL. Thus, peNDF values obtained using a 1.18-mm sieve, either as part of the PSPS or alone, do not adequately reflect differences in particle length of diets containing CS of varying TCL.
|
Intake, Digestibility, and Microbial Production
Intake of DM (kg/d or % of BW) was not affected by diet (Table 5). Several studies (Bal et al., 2000; Soita et al., 2005; Yang and Beauchemin, 2005) have indicated a lack of effect of chop length on DMI of dairy cows consuming CS with TCL ranging from 9 to 19 mm. However, other studies have reported a decrease in DMI of dairy cows when TCL of CS was increased which can be explained by increased rumen retention time of feed resulting in increased rumen fill and reduced intake. However, increased rumen retention time is not always the case. For example, Schwab et al. (2002) did not observe an increase of rumen retention time of feed although DMI was reduced by 0.8 kg/d when the TCL of CS was increased from 19 to 32 mm. The DMI was also reduced when the TCL of CS was increased from 13 to 19 mm in that study. Furthermore, De Boever et al. (1993) reported a 1.4 kg/d reduction in DMI when TCL of CS was increased from 4 to 8 mm. It appears that the effect of the TCL of CS on DMI is not consistent, which may relate to the inconsistent effect of increased TCL of forages on the peNDF content of the diets. In the present study, increasing the TCL of CS from 15.9 to 28.6 mm did not greatly increase the proportion of coarse forage particles or the peNDF content of the diets.
|
Apparent digestibilities of DM, OM, and starch in the total tract were unaffected by dietary peNDF, whereas increasing dietary peNDF linearly increased digestibility of CP and tended to increase digestibilities of NDF (P < 0.15) and ADF (P < 0.11; Table 5). These results are in agreement with some reports (Bal et al., 2000; Schwab et al., 2002; Yang and Beauchemin, 2005) but are in contrast to others (Kononoff and Heinrichs, 2003b; Fernandez et al., 2004) for cows fed CS-based diets. The discrepancy in effects of peNDF on nutrient digestibility in the total tract across the studies is attributed to various dietary factors. Chopping forage finely increases surface area available for microbial attack thereby accelerating digestion, but fine particles often have faster passage rate from the rumen such that digestibility is reduced. In addition, long particles are reduced in size during ingestive mastication, thereby eliminating the effects of particle length. Fernandez and Michalet-Doreau (2002) reported that mean particle length markedly decreased after ingestive mastication, and the decrease was greater for coarsely chopped than for finely chopped CS. Fermentability of concentrate could be another factor influencing digestion. In our previous study (Yang and Beauchemin, 2005) using a barley grain-based diet, total digestibility was increased with increasing dietary peNDF. However, when using a corn grain-based diet in the present study, the increase of digestibility in the total tract was minimal with increasing TCL of CS. Dietary particle length may enhance ruminal fermentation to a greater extent for barley-based diets than for corn-based diets because barley is rapidly degradable in the rumen following ingestion. Similarly, an increase of fiber digestion in the total tract with increasing particle length of CS was reported when processed CS (Bal et al., 2000; Schwab et al., 2002; Yang and Beauchemin, 2005), rather than unprocessed CS (Kononoff and Heinrichs, 2003b; Fernandez et al., 2004), was fed because processing increases ruminal starch digestibility (Schwab et al., 2002).
Daily urinary excretion of uric acid and allantoin was higher with the medium dietary peNDF than with the high or low dietary peNDF, so that urinary excretion of total PD was higher with the medium peNDF diet (Table 6). Consequently, calculated intestinal flow of microbial N (g/d) and microbial efficiency (g/kg of digestible OM intake) were quadratically increased with increasing dietary peNDF. The microbial N supply estimated in the present study was similar to that obtained by Krause and Combs (2003), who estimated microbial N supply using the same procedure from diets based on alfalfa silage, corn silage, and corn grain, and found an average microbial N supply of 297.7 g/d. Krause et al. (2002a) reported that ruminally fermentable carbohydrate is the key dietary factor affecting microbial protein synthesis, whereas dietary particle size indirectly influences microbial protein synthesis by its effects on feed intake. In the present study, lower microbial N supply for the high peNDF than for the medium peNDF diet could be due to lower starch digestion in the rumen resulting from a shift in starch digestion to the intestine with high peNDF (Yang and Beauchemin, 2005). Lower microbial N supply with low dietary peNDF is in agreement with our previous study (Yang and Beauchemin, 2005) in which microbial protein synthesis was decreased with decreasing dietary peNDF due to lower nutrient digestion in the rumen. Digestibilities of protein and fiber (P < 0.15) in the total tract were linearly lowered by reducing the dietary peNDF in the present study (Table 5).
|
). The lower milk yield of the cows fed the low peNDF diet was consistent with the lower digestibilities of protein and fiber in the total tract and the lower microbial protein synthesis. In animal digestion studies, the milk yield response to dietary peNDF is usually minimal (Zebeli et al., 2006) because most studies are conducted using few cows with short experimental periods (e.g., 21 d). Furthermore, the cows used in the present study, were quite variable in stage of lactation, ranging from 60 to 180 DIM, which could account for some of the variability in response to the treatments.
|
). Lack of effect of dietary peNDF on milk fat content suggests that the diets contained adequate fiber to maintain milk fat percentage above 3.5%. The contents of dietary NDF (31.1%) and NDF from forage (21.9%) in the current study were above the minimum recommendation by NRC (2001). Beauchemin and Rode (1997) reported a depression of milk fat when NDF was below minimum requirements. Using the peNDF system, Kononoff and Heinrichs (2003b) reported a tendency (P < 0.10) for lower milk fat percentage with peNDFps-2s and peNDFps-3s decreasing from 18.4 to 15.6% and from 30.6 to 32.4%, respectively. However, Yansari et al. (2004) reported a 10% reduction in milk fat percentage when dietary peNDFps-2s decreased from 11.2 to 2.0% and peNDFps-3s decreased from 25.1 to 17.2%. Hence, results from the literature and the present study support the recommendation of 19.7% peNDF1.18 by Mertens (1997) to maintain a milk fat of 3.4% as well as the suggestion of Beauchemin and Yang (2005) that a level of dietary peNDFps-2s above 10% is required to avoid a reduction of chewing activity.
Chewing Activity
Increasing peNDF content of the diet tended to decrease the number of eating chews per unit of DMI (P < 0.10) or NDF intake (P < 0.08), but did not affect the numbers of ruminating chews (Table 8). Cows spent similar time eating the high, medium, and low peNDF diets. However, ruminating time, as well as total chewing time, was linearly increased with increasing dietary peNDF. A tendency to reduce the number of eating chews with no difference in eating time for increasing dietary peNDF was not expected and contrasts with the literature (Fischer et al., 1994; Beauchemin and Yang, 2005). In those studies, although eating time was similar, number of chews per kilogram of DMI during eating was linearly increased (P < 0.10) with increasing dietary peNDF. The increase in total chewing time due to increased dietary peNDF resulted from increased ruminating time rather than eating time, which is consistent with some reports (Kononoff and Heinrichs, 2003b; Kononoff et al., 2003b), but different from others (Soita et al., 2000; Kononoff and Heinrichs, 2003a; Beauchemin and Yang, 2005) in which both eating and ruminating times were increased. Thus, increasing peNDF typically increases chewing time, mainly because of increased rumination time, and in some cases, increased eating time as well.
|
). The values are consistent with those of Kononoff and Heinrichs (2003b), but different from other reports (Kononoff et al., 2003b; Beauchemin and Yang, 2005) for CS-based diets. Mertens (1997) suggested that the ratio of total chewing activity to DM is an attribute of the feed, for cows of the same genotype, size, and level of intake. Zebeli et al. (2006) reviewed 99 published treatment means and found that cows spent 17.9 to 47.1 min/kg of DM and 54.3 to 160.1 min/ kg of NDF in total chewing, indicating that intake of DM or NDF is not a reliable variable to predict chewing duration. Furthermore, the chewing index based on intake of peNDF (min/kg of peNDF) was constant among diets with peNDF1.18 (118 to 122), slightly more variable with peNDFps-3s (98 to 113; P < 0.10), and particularly variable with peNDFps-2s (173 to 287). The implication is that systems of measuring peNDF that incorporate a 1.18-mm sieve (peNDF1.18 or peNDFps-3s) provide a more consistent prediction of the chewing index of the feed than the PSPS with 2 sieves (peNDFps-2s). The present results are consistent with the findings of Yansari et al. (2004) who reported for high, medium, and low dietary peNDF, a more consistent chewing index for peNDF1.18 (120, 112, and 118, respectively) and for peNDFps-3s (112, 105, and 102, respectively) than for peNDFps-2s (250, 409, and 926, respectively). Values for chewing index (min/kg of peNDFps-3s) of CS-based diets reported by Kononoff and Heinrichs (2003b) ranged 82 to 93 and from 71 to 81 in the study by Kononoff et al. (2003b). However, a considerable variation in chewing index based on peNDF1.18, ranging from 71 to 644, was reported for 87 published treatment means (Zebeli et al., 2006). Therefore, none of the current systems of evaluating peNDF provides a consistent measure of chewing activity, although chewing time is generally increased with increasing dietary peNDF content. Because other factors such as animal, level of intake, and type of feed also affect chewing activity, dietary peNDF level may have variable effects on promoting chewing among cows and studies. Furthermore, particle distribution of the feed actually consumed by the cow can differ substantially from that offered due to sorting in favor or against long particles, and this can drastically affect the actual intake of peNDF (Yang and Beauchemin, 2006).
Number of meals per day tended (P < 0.06) to be linearly decreased, but amount of DM and NDF per meal was linearly increased, with increasing dietary peNDF (Table 8). Intake rate (g of DM/min) was similar among the treatments. Linearly increased meal size due to increased dietary peNDF is consistent with the observation of Beauchemin and Yang (2005) who reported that amount of NDF per meal tended (P < 0.13) to be higher with higher dietary peNDF. The linearly increased meal size was associated with fewer meals rather than a faster intake rate.
Ruminal pH and Fermentation
Ruminal pH parameters including mean pH, area under pH 5.8 or 5.5, duration pH < 5.8, and the lowest pH were not affected by dietary peNDF except duration of pH < 5.5, which tended (P < 0.13) to be linearly reduced with increasing dietary peNDF (Table 9). Reduced duration of pH < 5.5 in the rumen indicated a small improvement of ruminal pH status (1 h reduction during a 24-h period). Lack of effect of CS particle size on mean ruminal pH in the present study is consistent with a number of studies (Kononoff and Heinrichs, 2003b; Plaizier, 2004; Beauchemin and Yang, 2005), although it contrasts with the commonly accepted principle that increasing particle length promotes chewing activity, and thereby increases buffering capacity within the rumen due to increased saliva secretion. Zebeli et al. (2006) conducted a regression analysis by pooling 100 treatment means and showed that mean ruminal pH responded in a quadratic fashion either to peNDF1.18 or to peNDFps-2s. The authors further determined that the plateau for ruminal pH in response to peNDF1.18 occurred at 22.6% dietary peNDF1.18. The average dietary peNDF1.18 of 24.4% in the present study was above this threshold, which may explain the lack of effect of peNDF on ruminal pH.
|
). The similar ruminal VFA concentration across the treatments was consistent with the ruminal pH data. However, there was a linear reduction in ruminal ammonia concentration with increasing dietary peNDF.
Correlations
Particles retained on the 19- or 8-mm sieves of the PSPS and the dietary peNDF content, regardless of methods used to determine it, were not correlated to the number of chews (Table 10). However, chewing activity (min/d) was correlated with particles on the 8-mm sieve (r = 0.52), peNDFps-2s (r = 0.52), peNDFps-2sndf (r = 0.49), peNDFps-3s (r = 0.45, P < 0.10), and peNDFps-3sndf (r = 0.47). The positive correlation of dietary peNDF with chewing time is consistent with the linear increase in chewing time with increasing dietary peNDF levels (Table 8). Lack of correlation between peNDF1.18 and chewing activity was due to similar dietary peNDF1.18 content across the treatments (Table 4). The correlation between particles >8 mm and chewing time is in agreement with our previous report (Beauchemin and Yang, 2005), but contrasts with the finding of Kononoff and Heinrichs (2003a), who suggested that the proportion of particles >19 mm may be a primary factor affecting chewing activity. It appears that the effect of the individual particle fractions of the PSPS on chewing activity depends on the relative contribution of each particle fraction to the pef (Beauchemin and Yang, 2005).
|
). Thus, mean ruminal pH was correlated with dietary peNDF measured using 2 sieves, despite the lack of statistically significant effects of treatment peNDF levels on ruminal pH (Table 9). This apparent contradiction can be explained by the fact that within treatment peNDF level (high, medium, and low), actual peNDF content of the diets consumed by individual cows was variable depending upon the degree of selectivity for and against long particles. The implication of these results is that even though a diet may be formulated to provide specific peNDF content, whether the diet promotes adequate chewing to maintain high rumen pH will depend on the individual cows. Thus, from a ration formulation approach, a margin of safety needs to be incorporated into recommendations for minimum peNDF. Correlation of ruminal pH with dietary peNDF and with chewing activity coincides with the premise that increasing particle length increases chewing activity and salivary buffer production. However, this premise is not always consistent in studies, because the correlation between ruminal pH and chewing activity in the literature is usually poor (Krause et al., 2002b), especially for diets containing highly fermentable ingredients such as barley grain (Beauchemin and Yang, 2005; Yang and Beauchemin, 2006). The relationship between peNDF and rumen pH is inconsistent because the concept of peNDF does not account for differences in ruminal fermentability of feeds, which can have a major impact on ruminal pH (Yang et al., 2001; Krause et al., 2002b). Furthermore, there is tremendous variability in ruminal pH among cattle, even when fed the same diet (Bevans et al., 2005).
Choice of peNDF System
The ability to differentiate the physical effectiveness of diets is essential in terms of predicting the biological response of the cow. In this study, dietary peNDF was positively associated with fiber digestibility, as well as chewing and ruminating time. However, there was no relationship between peNDF and milk production because the cows were in mid to late lactation. The contents of pef and peNDF of forage and TMR estimated using the new PSPS with 3 sieves or a 1.18-mm screen were similar, but higher than those obtained using the original PSPS with 2 sieves. However, determining peNDF using the systems that incorporate a 1.18-mm sieve did not differentiate between diets differing in particle size distribution due to varying TCL of CS. Thus, peNDFps-3s and peNDF1.18 did not appropriately determine dietary pef and peNDF of CS-based diets. Calculating the peNDF content of diets based on the fractional NDF content retained by the individual sieves, rather than the total NDF content of the sample, increased the values but did not change the classification of diets, particularly because the diets did not vary in proportion of concentrate. The disadvantage of a system with fractional NDF is that it requires more laboratory analyses (i.e., NDF determination), which greatly restricts its application, especially on farm. However, the fractional NDF system would be advantageous for differentiating peNDF of diets that vary in concentrate, particularly when the concentrates are of variable size, which would affect the proportions retained on the sieves.
The peNDF1.18 was not correlated with chewing activity. In contrast, peNDF measured by both the original and new PSPS was positively correlated to chewing activity. Thus, the PSPS is a useful and practical device to determine the physical effectiveness of fiber, which is a good indication of the rumination potential of the feed. The consistency of chewing index, the ratio of total chewing activity to peNDF, across diets varying in chop length of CS when using the new PSPS would seem to indicate that it may be the best of the systems evaluated for measuring effects of diet on chewing time. However, the values for peNDF using the new PSPS were not different among diets that varied considerably in chop length of CS. This lack of variation makes this system difficult to use in ration formulation to account for effects of dietary differences in particle length. Thus, although each system evaluated in this study has its advantageous and disadvantages, it would appear that the PSPS with 2 sieves is the most useful of the systems evaluated for meeting the requirements of dairy cows for peNDF. The PSPS with 2 sieves differentiated diets based on particle size of forage and was correlated with chewing and ruminal pH.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Received for publication November 10, 2005. Accepted for publication February 2, 2006.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. K. Miller-Cushon and T. J. DeVries Effect of dietary dry matter concentration on the sorting behavior of lactating dairy cows fed a total mixed ration J Dairy Sci, July 1, 2009; 92(7): 3292 - 3298. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Adin, R. Solomon, M. Nikbachat, A. Zenou, E. Yosef, A. Brosh, A. Shabtay, S. J. Mabjeesh, I. Halachmi, and J. Miron Effect of feeding cows in early lactation with diets differing in roughage-neutral detergent fiber content on intake behavior, rumination, and milk production J Dairy Sci, July 1, 2009; 92(7): 3364 - 3373. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Z. Yang and K. A. Beauchemin Increasing physically effective fiber content of dairy cow diets through forage proportion versus forage chop length: Chewing and ruminal pH J Dairy Sci, April 1, 2009; 92(4): 1603 - 1615. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Dubuc, D. DuTremblay, M. Brodeur, T. Duffield, R. Bagg, J. Baril, and L. DesCoteaux A randomized herd-level field study of dietary interactions with monensin on milk fat percentage in dairy cows J Dairy Sci, February 1, 2009; 92(2): 777 - 781. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Q. Zebeli, M. Tafaj, B. Junck, V. Olschlager, B. N. Ametaj, and W. Drochner Evaluation of the Response of Ruminal Fermentation and Activities of Nonstarch Polysaccharide-Degrading Enzymes to Particle Length of Corn Silage in Dairy Cows J Dairy Sci, June 1, 2008; 91(6): 2388 - 2398. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Q. Zebeli, J. Dijkstra, M. Tafaj, H. Steingass, B. N. Ametaj, and W. Drochner Modeling the Adequacy of Dietary Fiber in Dairy Cows Based on the Responses of Ruminal pH and Milk Fat Production to Composition of the Diet J Dairy Sci, May 1, 2008; 91(5): 2046 - 2066. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Z. Yang and K. A. Beauchemin Altering Physically Effective Fiber Intake Through Forage Proportion and Particle Length: Digestion and Milk Production J Dairy Sci, July 1, 2007; 90(7): 3410 - 3421. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Z. Yang and K. A. Beauchemin Altering Physically Effective Fiber Intake Through Forage Proportion and Particle Length: Chewing and Ruminal pH J Dairy Sci, June 1, 2007; 90(6): 2826 - 2838. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
