| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,4
* Department of Animal Sciences, University of Illinois, Urbana 61801
Department of Dairy Science, University of Wisconsin, Madison 53706
Agricultural Research Service, USDA, US Dairy Forage Research Center, Madison 53706
2 Corresponding author: Ignacio.Ipharraguerre{at}lucta.es
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Key Words: dairy cow • nutrient flow • omasum • duodenum
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Surgical cannulation of the gastrointestinal tract is required in ruminant animals to assess the kinetics of nutrient digestion in the rumen and the passage of nutrients to the absorption sites in the intestines. Although the commonly used simple-T cannula offers fewer postsurgical complications than the reentrant cannula, digesta samples may not be representative of digesta leaving the rumen (Faichney, 1993; Harmon and Richards, 1997). Because digesta flow at the duodenum is tubular in nature, the use of closed-T duodenal cannulas with complete diversion of the flow and frequent sampling throughout the day may result in collection of representative samples (Firkins et al., 1998). However, this cannulation surgery is invasive, maintenance of cannulated animals is laborious, and the longevity of experimental animals often is compromised.
An alternative involves sampling digesta that flow out of the rumen into the omasal canal. The omasal sampling technique developed by Punia et al. (1989) and modified by Huhtanen et al. (1997) and Ahvenjärvi et al. (2000) can be used for such a purpose, using animals fitted with only ruminal cannulas. Because ruminal cannulation is less invasive and expensive than duodenal cannulation, more experimental animals can be used to increase the statistical power and detect differences among treatments that otherwise would be declared insignificant (i.e., reducing type II error). Compared with duodenal sampling, collection of omasal samples avoids limitations imposed by abomasal digestion and contamination of digesta with abomasal secretions. However, omasal sampling is more labor intensive and requires the use of more sophisticated equipment. More important, because samples taken from the omasal canal are not representative of true digesta, a multiple marker system must be used to mathematically reconstitute digesta samples, increasing the probability of compounding analytical errors.
Omasal flows of some N and fiber fractions have been shown to be representative of duodenal flows (Ahvenjärvi et al., 2000) under feeding conditions (high forage, small grain-based diets) that differ substantially from those usually found in the United States and other countries (corn-based diets) in terms of the physical characteristics of both the diets fed and the digesta leaving the rumen. Data from experiments comparing flow of nutrients determined at the omasal canal and the duodenum of dairy cows fed corn-based diets that contained different sources of supplemental CP are not available. Consequently, the objectives of the present trial were to identify, compare, and discuss major limitations associated with the use of the omasal canal and duodenal sampling techniques for measuring nutrient passage in dairy cows fed corn-based diets that contained different sources of CP, and to evaluate the impact of different digesta and microbial markers as well as microbial references on the estimation of nutrient flows at these sites of the digestive tract.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Cows were housed in tie stalls bedded with straw and had free access to water throughout the trial. Cows were allowed to exercise in a dry lot from 0800 to 0900 h and were milked daily at 0500 and 1700 h. Cows were injected with bST (500 mg of Posilac; Monsanto, St. Louis, MO) beginning on d 7 of the trial and thereafter at 14-d intervals until the end of the trial. Cows were fed a TMR ad libitum twice daily as follows: two-thirds of the diet was fed immediately after mixing the feeds at 1730 h, and the remaining TMR was held at 5°C and fed the next day at 0500 h. Orts were collected and weighed once daily at 1700 h, and the feeding rate was adjusted daily to yield orts of about 10% of intake.
The experimental design was an incomplete 4 x 4 Latin square with 4 treatments, 4 periods, and 3 cows. Each experimental period lasted 14 d and consisted of an 8-d adaptation period and a 6-d sample collection period. Each cow was randomly assigned to one of 3 treatment sequences in which a treatment never followed any other treatment more than once. Four dietary treatments (Table 1
) that contained different sources and degradabilities of protein were fed to the cows to compare sampling techniques at the omasum and duodendum. Details and discussion of diet and individual treatment effects on the ruminal outflow of N fractions are provided in the companion paper by Reynal et al. (2007). Diets were adjusted weekly to reflect changes in DM of forages and concentrate mixtures by drying weekly composites of each ingredient overnight in an oven at 105°C.
|
-amylase and Na2SO3 (Hintz et al., 1995).
Ruminal Fluid.
Samples of ruminal fluid were collected at 1800, 1900, 2000, and 2200 h on d 8 and at 0100, 0500, 0600, 0700, 0800, 1000, and 1300 h on d 9 of each period. Ruminal fluid samples were collected from multiple sites in the rumen, and pH of the ruminal fluid was measured immediately by glass electrode.
Duodenal Digesta.
Samples of duodenal digesta were collected 4 times daily at 2-h intervals on d 12, 13, and 14 of each period such that the samples taken represented a 24-h feeding cycle over 3 d. The sampling time was adjusted ahead 2 h daily so that a sample was obtained for each 2-h interval of the day. Following removal of the cap of the duodenal cannula, accumulated digesta were discarded and, when the flow appeared normal, 500 mL of duodenal contents were collected. The pH of duodenal digesta was measured immediately by glass electrode. The volume of sample collected represented less than 2% of the estimated passage of digesta to the duodenum during the 3 d of collection. Samples were pooled by cow and stored at –20°C until analysis. Duodenal samples were thawed and homogenized for 5 min using a propeller-type mixer (Jumbo mixer, model 40; Fisher Scientific International Inc., Hampton, NH) set at high speed. During continuous stirring, a representative subsample (1,000 mL) of digesta was collected by vacuum. Samples were then poured into shallow pans, lyophilized, ground through a 1-mm screen, and analyzed for DM, OM, total N, starch, ADF, and NDF as described above. Extracts for ammonia analysis by flow injection (Dual-Channel QuikChem 8000 FIA; Lachat Instruments, Milwaukee, WI) were prepared by adding 10 mL of pH 2.2 Na-citrate buffer to 0.5 g of sample; after mixing, samples were held at 39°C for 30 min and centrifuged (15,000 x g, 4°C, 15 min), and the supernatant was stored at –20°C until analysis.
Omasal Digesta.
Digesta flow leaving the rumen was quantified using the omasal sampling technique developed by Huhtanen et al. (1997) and modified by Ahvenjärvi et al. (2000). Spot samples were collected from the omasal canal of each cow immediately after collecting the duodenal samples. Each omasal spot sample was divided into 3 subsamples of 125, 200, and 30 mL, respectively. The 125-mL subsamples were stored in an ice bath as they were collected, pooled over the 4 sampling times to yield one 500-mL composite from each cow on each sampling day, and processed later that day to isolate bacteria. The 200-mL subsamples were stored at –20°C as they were collected and pooled over the 12 sampling times to yield one 2.4-L omasal composite from each cow in each period. The 30-mL subsamples were used to study the effects of individual treatments on the flow of soluble N fractions at the omasal canal (Reynal et al., 2007).
The 2.4-L pooled omasal composites were thawed at room temperature and separated into 3 digesta phases as follows. Samples were squeezed through 1 layer of cheesecloth, and the solids retained were defined as the large-particle phase (LP). The filtrate was centrifuged at 1,000 x g (5°C, 5 min), and the supernatant was carefully decanted from the pellet. The supernatant was defined as the fluid phase (FP) and the pellet as the small-particle phase (SP). These separated phases were frozen, freeze-dried, and then ground through a 1-mm screen for later analysis.
Concentrations of Co, Yb, and indigestible NDF (INDF) in LP and SP, and of Co and Yb in FP were used to mix DM from freeze-dried FP, SP, and LP in the correct proportions to reconstitute the omasal true digesta (OTD) flowing out of the rumen based on the triple-marker method of France and Siddons (1986). Aliquots of the SP and LP were mixed in the correct proportions based on the markers to yield a 2-g sample that was ground through a 0.5-mm screen with an Udy Cyclone sample mill (Udy Corporation, Fort Collins, CO) and defined as small plus large particles (SP + LP). The OTD samples were analyzed for DM, OM, ammonia, total N, starch, ADF, and NDF as described above.
Digesta Markers.
Cobalt-EDTA (Uden et al., 1980), YbCl3 (modified from Siddons et al., 1985), INDF (Huhtanen et al., 1994), and Cr2O3 were used as markers to assess the postruminal flow of digesta. Cobalt-EDTA and YbCl3 were dissolved in distilled water and infused into the rumen at constant daily rates of 2.6 g of Co and 2.2 g Yb. Markers were infused for 150 h from d 8 at 0900 h to d 14 at 1500 h using a syringe pump (model no. 33; Harvard Apparatus, Inc., Holliston, MA). The YbCl3 was prepared by adding 230 g of Yb2O3 to 200 mL of distilled water plus 320 mL of concentrated HCl, heating, and stirring until the solution was clear, then diluting to 5 L with distilled water. Samples of LP, SP, FP, and infused markers were analyzed for Co and Yb as described by Reynal and Broderick (2005).
Gelatin capsules containing 10 g of Cr2O3 were administered via the ruminal cannula at 0600 and 1800 h during the last 10 d of each period. The concentration of Cr in duodenal samples was quantified by atomic absorption spectroscopy (air plus acetylene flame; PerkinElmer, Norwalk, CT) after preparation of samples by the procedure of Williams et al. (1962). Indigestible NDF was determined in LP, SP, and TMR samples (but not in FP; Ahvenjärvi et al., 2000), as described by Reynal and Broderick (2005).
Microbial Markers and References.
Total purines and 15N were used to measure microbial NAN flows at the omasal canal and duodenum. Before starting the infusion of markers, a sample of whole ruminal contents was taken from each cow to determine the 15N background (15NB). In the solution containing Co-EDTA and YbCl3, 10% atom excess 15NH4SO4 was dissolved and infused into the rumen at a constant daily rate (182 mg of 15N) according to the procedure described for the infusion of digesta markers.
At the end of each sampling day, the 500-mL omasal composites were separated into phases that were equivalent to the FP and the SP + LP to isolate, respectively, fluid-associated bacteria (FAB) and particle-associated bacteria (PAB). Omasal composites were squeezed through 1 layer of cheesecloth; solids retained on the cheesecloth were washed with 400 mL of 0.85% (wt/vol) NaCl solution and squeezed again; and filtrates (equivalent to FP + SP) were pooled together, stored on ice, and processed later that day for FAB isolation. The solids retained on the cheesecloth (equivalent to the LP phase) were transferred to a 500-mL bottle; 350 mL of 0.85% (wt/vol) NaCl containing 0.1% (vol/vol) Tween-80 and cooled to 5°C was added; and bottle contents were mixed thoroughly and stored on ice until processed for PAB isolation. Filtrates for FAB isolation were centrifuged (1,000 x g, 5°C, 5 min); pellets were saved and supernatants were carefully decanted and recentrifuged (11,300 x g, 5°C, 30 min); supernatants were decanted and discarded; and pellets were resuspended in 100 mL of McDougall’s buffer and recentrifuged (11,300 x g, 5°C, 30 min). The resulting FAB pellets were stored at –20°C until freeze-dried and ground with a mortar and pestle for later analysis. Pellets from the 1,000 x g centrifugation step were mixed with the contents of the 500-mL bottles saved for PAB isolation (equivalent to SP + LP), blended for 20 s using a blender (Waring Products Division, New Hartford, CT), transferred back to the bottles, and stored at 5°C for 24 h. Blended contents were squeezed through 2 layers of cheesecloth and the filtrates were centrifuged (1,000 x g, 5°C, 5 min); supernatants were carefully decanted and recentrifuged (11,300 x g, 5°C, 30 min). Supernatants were decanted and discarded, and pellets were resuspended in 100 mL of McDougall’s buffer and recentrifuged (11,300 x g, 5°C, 30 min). The resulting PAB pellets were stored at –20°C until freeze-dried and were ground with a mortar and pestle for later analysis. Composites were prepared by mixing equal amounts of DM from d 1, 2, and 3, to obtain 1 sample each of FAB and PAB/cow per period. Each bacterial composite sample represented a 24-h feeding cycle. The 15NB samples were frozen, freeze-dried, then sequentially ground through a 1-mm screen and through a 0.5-mm screen with a Udy Cyclone sample mill and stored for later analyses.
In addition, ruminal bacteria (RB) were isolated from samples (1,000 mL) of whole ruminal contents obtained from the reticulum near the reticulo-omasal orifice at 6 separate postfeeding times (0, 2, 4, 6, 8, and 10 h) during the last 3 d of each period. Ruminal samples were blended using a blender (Waring Products Division) at low speed for 1 min and strained through 6 layers of cheesecloth, and the filtrate was centrifuged at 500 x g for 15 min at 4°C. Supernatants were carefully decanted and recentrifuged (18,000 x g, 15 min, 4°C). Supernatants were decanted and discarded, and pellets were washed with 0.9% wt/vol NaCl solution and re-centrifuged (18,000 x g for 15 min at 4°C). The resulting 6 pellets from each cow were pooled and stored at –20°C until freeze-dried and ground with a mortar and pestle for later analysis.
The individual purines adenine and guanine and their metabolites xanthine and hypoxanthine were determined in samples of PAB, FAB, RB, OTD, and duodenal digesta using the HPLC method of Balcells et al. (1992), as described by Reynal et al. (2003). The sum of the purines and their metabolites was defined as total purines (Reynal et al., 2005). Samples of 15NB, FAB, PAB, RB, FP, SP + LP, and duodenal digesta (100 µg of N) were weighed into tin cups and 50 µL of 72-mM K2CO3 solution were added to raise the pH to 12. Capsules were placed in a 60°C oven for 24 h to volatilize ammonia. Samples were then analyzed for NAN and 15N atom% enrichment (APE) of NAN by isotope ratio mass spectrometry (Stable Isotope Facility, Department of Agronomy and Range Science, University of California-Davis, Davis, CA).
Calculations
Flow of Digesta.
Digesta flow to the omasum was calculated from the concentrations of Co, Yb, and INDF in omasal digesta phases using the triple-marker method of France and Siddons (1986). Digesta flow into the duodenum was calculated using Cr, Yb, or Co as an external marker (Van Soest, 1994).
Flow of N Fractions.
The 15N APE (above natural background) was calculated for digesta and microbial samples from each cow in each period as follows:
 |
The 15N APE and total purines and NAN concentration in total omasal bacteria (TB) were calculated from the 15N APE, NAN, and purine contents of FAB and PAB and their proportions in omasal true digesta samples based on the triple-marker method. Microbial NAN flows into the omasal canal and duodenum were calculated using FAB, PAB, TB, and RB as the bacterial reference and 15N or total purines as microbial markers as follows:
 |
 |
Similarly, microbial efficiencies [g of microbial N/kg of OM truly digested in the rumen (i.e., measured at the omasal canal) or whole stomach (i.e., measured at the duodenum)], nonammonia nonmicrobial N (NANMN) flows at the omasal canal or duodenum, and true N digestibilities in the rumen or whole stomach were calculated using microbial NAN flows based on FAB, PAB, TB, or RB as references and 15N or total purines as microbial markers. True N digestibilities in the rumen and whole stomach were calculated as follows:
 |
 |
Statistical Analyses
Data were analyzed using the GLM procedure of SAS (SAS Institute, 1999). The effects of using different bacterial references for calculating microbial and nonmicrobial NAN flows at the omasal canal and duodenum were assessed using the following split-plot model:
 |
where C, P, T, B are cow (random), period (fixed), treatment, and bacterial reference effects, respectively, and 
and 
are the whole-plot and split-plot error, respectively. The same model was used to assess the effects of microbial marker (15N vs. total purines) on microbial and nonmicrobial NAN flows at the omasal canal and duodenum and the effects of digesta sampling site and duodenal digesta flow marker (omasal digesta vs. duodenal digesta calculated using Co, Yb, or Cr as a marker) on the flows and digestibilities of nutrients. Interactions between treatment, digesta sampling site, and microbial marker were assessed using the following split-plot model:
 |
where C, P, T, M, and S are cow (random), period (fixed), treatment, microbial marker, and sampling site effects, respectively, and 
, 
, , and are the errors. Values reported are least squares means that were separated into significant main effects using Fisher’s protected least significant difference. Differences among treatments were considered to be significant when P 
0.05.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
), differences in the amount and source of protein allowed the comparison of the omasal and duodenal sampling techniques under a wide range of dietary conditions, which broadens the scope of our inferences. Diets had similar NDF, ADF, and starch contents, averaging 29.6, 18.2, and 27.4% of DM, respectively. The chemical composition of the individual feeds is presented in Table 2.
|
). These pH values were within normal physiological ranges (Church, 1971), suggesting that there was no apparent contamination of omasal samples with abomasal secretions or of duodenal samples with pancreatic secretions.
|
). To test the effects of microbial marker within digesta sampling site, without the confounding effect of bacterial reference, TB was used as the bacterial reference for both omasal and duodenal calculations (Table 4). Because of the significant interaction between digesta sampling site and microbial marker, the effects of digesta sampling site on total NAN, microbial NAN, NANMN, RDP, and microbial efficiency were reported separately for 15N and total purines. The 3-way interaction among dietary treatment, digesta sampling site, and digesta flow marker was not significant (P < 0.25) for the postruminal flow and digestibility of DM, OM, starch, NDF, and ADF, or for the ruminal outflow of microbial NAN and NANMN and microbial efficiencies.
|
|
). However, 15N APE was not significantly different between TB and RB. Because populations of bacteria and protozoa present in omasal digesta appear to be representative of those inhabiting the rumen (Karnati et al., 2005), these findings suggest that differences in the procedures used to isolate TB and RB altered the species of bacteria harvested, their contribution to each bacterial reference, or both.
When using total purines as a microbial marker, microbial NAN flows at the omasal canal and duodenum were about 5 and 6% higher for RB than TB, respectively (Table 3). Similarly, microbial NAN flows at both sites measured using 15N as a marker were 9% higher when RB rather than TB was used as a reference. When measured at the omasal canal using RB as a reference and 15N as a marker, flows of NANMN were lower and ruminal digestibility of N was higher than those based on TB. Although the ruminal outflow of N fractions and N digestibility followed similar trends across sampling sites, differences between bacterial references were significant only when these variables were measured at the omasal canal (Table 3). Based on these results and the apparently higher standard errors of measurement at the duodenum, it seems reasonable to suggest that omasal sampling was more precise than duodenal sampling to measure the impact of the bacterial reference on the ruminal outflow of N fractions. However, as will be discussed later, it is also possible that omasal sampling might be artificially sensitive, allowing it to reveal differences between bacterial references that otherwise would have required a larger sample size to reach significance.
In agreement with several reports from the literature (Craig et al., 1987; Carro and Miller, 2002; Reynal et al., 2005), N:purine ratios in PAB were 21% higher than those in FAB (Table 3). This resulted in microbial NAN flows at both the omasal canal and duodenum that were 22% higher when PAB, rather than FAB, was used as a reference. Similarly, Carro and Miller (2002) reported that the production of microbial protein in semicontinuous fermenters based on the N:purine ratio in FAB or PAB was either underestimated or overestimated by 16%, respectively, compared with estimations based on the N:purine ratio of a mixed bacterial reference. Differences between FAB and PAB in the N:purine ratio have been attributed to the existence of 1) diverse bacterial species (Dehority and Orpin, 1988), 2) specific growth rates among species, 3) microenvironmental conditions (Bates et al., 1985; Legay Carmier and Bauchart, 1989) in the fluid and particulate phases of digesta, or 4) a combination of these factors.
The 9% lower 15N enrichment of PAB compared with FAB resulted in estimates of microbial NAN flow that were 9% higher than those estimated using FAB enrichment (Table 3). In the literature, 15N enrichments of FAB are reported to be from 6 (Rodriguez et al., 2000) to 15% (Ahvenjärvi et al., 2002) higher than those of PAB. This might be the result of 1) a greater use of ammonia N as a N source by FAB than PAB (Faichney et al., 1997; Carro and Miller, 1999), or 2) differences between the growth rate of FAB and PAB and the rate at which ammonia, AA, and peptides were released in the rumen for microbial utilization, or both.
Although bacteria associated with particulate matter can account for 70 to 75% of the total bacteria in the rumen (Legay Carmier and Bauchart, 1989; Hristov and Broderick, 1996), higher passage rates of liquids than particles may result in similar proportions of PAB and FAB in total bacterial N flows from the rumen (Hristov and Broderick, 1996). In the present study, however, 72% of the bacterial flow at the omasal canal was associated with the particle phase based on triple-marker computations (data not shown). Similarly, PAB accounted for 72% of the bacterial N flow at the omasal canal in the study of Ahvenjärvi et al. (2002). Therefore, when TB are used as the bacterial reference and their enrichment is not directly assessed, it appears that the marker concentrations in FAB and PAB and the relative contributions of these microbial fractions must be used to obtain representative measurements of microbial N flow.
Effect of Microbial Marker
Comparisons between 15N and total purines as microbial markers reported in the literature are inconclusive. Microbial N flows measured in vivo or in continuous cultures using 15N were lower than (Carro and Miller, 2002), higher than (Perez et al., 1996), or similar to (Hristov et al., 2005; Reynal et al., 2005) those estimated using total purines. In the present study, the impact of microbial marker on the ruminal outflow of N fractions and N digestibility depended on the sampling site. When digesta were sampled from the omasal canal, the use of 15N as a microbial marker resulted in higher ruminal outflow of microbial NAN, true N digestibility, and microbial efficiency but lower ruminal escape of NANMN than did total purines (Table 4). In contrast, none of these variables was significantly different between microbial markers when digesta samples were obtained from the duodenum (Table 4).
The proportion of xanthine, a metabolite of the degradation of adenine and guanine, in total purines (i.e., sum of adenine, guanine, xanthine, and hypoxanthine) was 6.2% for FAB, 11.0% for PAB, 24.9% for FP, and 13.5% for SP + LP. In contrast, xanthine concentrations in RB and duodenal digesta samples were 4.4 and 2.6% of total purines, respectively. This suggests that purines in omasal digesta phases were degraded to a greater extent than were purines in their corresponding bacterial references, leading to an underestimation of the omasal flow of microbial NAN relative to 15N. The higher concentration of xanthine in omasal digesta and bacterial references might have resulted from the longer time required for isolation and processing of these samples compared with isolation of RB and immediate freezing of duodenal digesta after collection.
Interestingly, the difference between microbial markers for the flow of microbial NAN at the omasal canal (381 vs. 288 g/d) was as large as the difference between sampling sites for the same variable when 15N, but not total purines, was used as a marker (381 vs. 284 g/d). As proposed earlier, a more extensive degradation of purine bases in omasal samples than in duodenal samples might have prevented detecting differences between sampling sites when total purines were used as microbial markers. This does not explain, however, why the use of 15N resulted in about 100 g/d more microbial NAN passing to the omasum than the duodenum. As proposed later, it is likely that problems associated with the estimation of digesta flows at each sampling site accounted for this difference.
Duodenal Sampling
As shown in Table 5, duodenal DM flows calculated using Co and Yb as digesta markers were significantly correlated (R = 0.99; P < 0.01), with a slope of 0.98 (P < 0.01) and an intercept of 2.39 (P < 0.08). However, flows based on Cr were poorly correlated with those based on Co (slope = 0.69) or Yb (slope = 0.71). Duodenal flows of DM, OM, NDF, ADF, and microbial NAN were significantly lower when calculated using Cr, higher when based on Yb, and intermediate when Co was used as a marker (Tables 6 and 7). Duodenal flows of DM, OM, and starch computed using Co or Yb were always distinctly higher than their intakes, resulting in negative, and therefore biologically improbable, whole-stomach digestibilities.
|
|
|
). Furthermore, mean duodenal DM flows for all cows estimated using Yb and Co were, on average, 59 and 46% higher than Cr-based flows. Cobalt is mainly associated with the fluid phase (Uden et al., 1980) and Yb is mainly associated with small particles (Siddons et al., 1985) that may escape the rumen at rates similar to the fluid phase (Reynal and Broderick, 2003). Therefore, it is possible that at least for the cow with large digesta flow values, enrichment of duodenal digesta samples with particles might have given similar dilutions of the Co and Yb, resulting in the collection of samples that did not represent true digesta. On the other hand, because chromic oxide dosed directly into the rumen does not associate with any phase when flowing through the gut (Merchen, 1988), estimations of digesta flows based on Cr might have been less sensitive to incorrect contributions of digesta phases in duodenal samples because the concentration of Cr was probably affected less than that of Co and Yb.
|
One might speculate that total collection of digesta from animals fitted with reentrant cannulas might provide an alternative to solve this problem. However, reentrant cannulas can cause abnormal propulsion and flow of digesta, they can interfere with animal performance (Wenham and Wyburn, 1980), and data obtained may not accurately represent physiologically normal cattle (Robinson and Kennelly, 1990). Consequently, there is no way to determine with absolute certainty whether digesta samples collected at the duodenum of cattle fitted with closed-T cannulas are representative of true digesta flowing though the duodenum.
Omasal Canal Sampling
Digesta phases are prone to separation when samples are obtained by aspirating through a tube or from a simple-T cannula. To overcome this limitation, France and Siddons (1986) proposed the use of a triple-marker method to reconstitute mathematically the true digesta by combining 3 independent phases (FP, SP, and LP) in the proportions flowing past the cannula. This approach assumes that digesta samples are truly representative of each phase but not of total digesta. Implicitly, this means that the size and the physical and chemical composition of digesta phases in samples that are not representative are considered to be the same as in true digesta (Faichney, 1993). However, when digesta are aspirated from the omasal canal, particles with different functional specific gravities may segregate as they travel through the sample tube. This could result in flows of particles of high specific gravity (e.g., corn kernels) being underestimated. Likewise, sieving of large particulate matter (e.g., long fiber particles) during aspiration through the sampling tube could also result in incorrect mixtures of digesta fractions in the samples. This would not only lead to an underestimation of the contribution of large particles to the flow of total digesta at the omasal canal, but also to the outflow of nutrients that exit the rumen in close association with the LP phase (e.g., starch trapped in whole or partly broken kernels). Consequently, mathematical estimates of the flows of the other 2 phases (i.e., FP and SP) at the omasum and nutrients associated with them would be biased as well (i.e., overestimated).
Ruminal starch digestibilities measured using the omasal sampling technique averaged 86% across diets (Table 6). Data compiled by Reynolds et al. (1997) indicate that the proportion of starch intake that was apparently digested in the rumen of lactating dairy cows fitted with duodenal cannulas and fed 25 different diets containing corn grain either ground, rolled, or flaked ranged from 31 to 69% and averaged 48%. It is therefore likely that in this trial, corn kernels were segregated during sampling, resulting in underestimation of starch flows and overestimation of starch digestibility in the rumen. In addition, ruminal digestibilities of NDF and ADF determined using omasal sampling were higher than total tract digestibilities calculated using Cr as a marker (data not shown), which is in agreement with results from Ahvenjärvi et al. (2001) and suggests that the ruminal outflow of these fiber fractions was probably underestimated. Furthermore, when digesta were sampled from the omasum, the postruminal flow of microbial NAN was about 100 g/d larger for 15N than total purines (Table 4). Interestingly, such a difference is as large as the difference found between the mean of 215 published observations for the flow of microbial N to the duodenum (268 g/d; Ipharraguerre and Clark, 2005) and the mean of 42 reported estimates for the flow of microbial N at the omasal canal (370 g/d; I. R. Ipharraguerre and S. M. Reynal, unpublished data). Because rumen bacteria pass to the lower digestive tract mainly attached to SP and as solutes in the FP (see above discussion), the flow of microbial NAN at the omasal canal might have been overestimated. This hypothesis is supported by the finding that the omasal flow of microbial NAN was strongly correlated with the proportional contribution of SP and FP to the reconstituted omasal digesta, but not with that of the LP (Figure 3). Therefore, it might be possible that at least a part of these differences occurred because reconstituted omasal samples contained larger proportions of small particles and liquid than true digesta, resulting from the loss of large particles during sampling. Independent of the validity of this assertion, the magnitude of the difference between starch and microbial N flows to the omasum and duodenum found in this experiment and in the literature warns that the extrapolation and interpretation of published results as well as the outcomes of their application in model parameterization and predictions should be carefully evaluated to avoid flawed generalizations.
|
and 6) and the cows used (Tables 8 and 9). Assuming that malfunction of the duodenal cannula occurred in one of the animals, data from the other 2 animals were summarized in Tables 8 and 9 using arithmetic means. Starch flows at the duodenum based on Cr were 3.7 times higher than those measured at the omasal canal (2.82 vs. 0.77 kg/d; Table 8). Ruminal digestibility of starch was 86% higher when measured at the omasal canal, compared with the duodenum based on Cr. When estimated using 15N, microbial NAN flows were 44% higher at the omasal canal compared with the duodenum using Cr. However, flows and digestibilities of DM, OM, NDF, and ADF, as well as microbial NAN flow and efficiency based on total purines and RDP supply and NANMN flow based on either microbial marker, were similar between the omasal canal and duodenum based on Cr (data from 2 cows only; Tables 8 and 9). Nevertheless, duodenal flows and digestibilities based on Co and Yb were substantially different from those based on Cr and those measured at the omasal canal. Therefore, omasal canal and duodenal sampling yielded similar results for 2 of 3 cows and only when Cr was used as the duodenal digesta flow marker and total purines as the microbial marker.
|
|
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Collectively, outcomes from this study indicate that more research will be required to determine the accuracy of nutrient flows measured at the omasal canal and duodenum and the agreement between measurements at both sites when different diets and feeding systems are used. Furthermore, published data for the flow into the omasal canal of several nutrients are limited (e.g., starch) or nil (e.g., fatty acids). In the meantime, caution is recommended when extrapolating or interpreting the underlying biology of published results as well as the outcomes from their application (e.g., model parameters and predictions).
|
FOOTNOTES |
|---|
3 Current address: Lucta S. A., P.O. Box 1112, Barcelona 08080, Spain.
4 Current address: US Dairy Forage Research Center, Madison, WI 53706.
5 Current address: Facultad de Agronomía, Universidad de Buenos Aires, Argentina.
Received for publication March 3, 2006. Accepted for publication November 7, 2006.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. J. Shingfield, A. Arola, S. Ahvenjarvi, A. Vanhatalo, V. Toivonen, J. M. Griinari, and P. Huhtanen Ruminal Infusions of Cobalt-EDTA Reduce Mammary {Delta}9-Desaturase Index and Alter Milk Fatty Acid Composition in Lactating Cows J. Nutr., April 1, 2008; 138(4): 710 - 717. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. Reynal, I. R. Ipharraguerre, M. Lineiro, A. F. Brito, G. A. Broderick, and J. H. Clark Omasal Flow of Soluble Proteins, Peptides, and Free Amino Acids in Dairy Cows Fed Diets Supplemented with Proteins of Varying Ruminal Degradabilities J Dairy Sci, April 1, 2007; 90(4): 1887 - 1903. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||
