J. Anim Sci. 2008. 86:702-711. doi:10.2527/jas.2007-0146
© 2008 American Society of Animal Science
Changes in rumen microbial fermentation are due to a combined effect of type of diet and pH1
S. Calsamiglia*,2,
P. W. Cardozo*,
A. Ferret* and
A. Bach
1 Animal Nutrition, Management and Welfare Research Group, *Departament de Ciència Animal i dels Aliments Universitat Autònoma de Barcelona 08193-Bellaterra, Spain; and and
Institució Catalana de Recerca i Estudis Avançats (ICREA) and Institut de Recerca i Tecnologia Agroalimentàries, IRTA-Unitat de Remugants Barcelona, Spain
 |
Abstract
|
|---|
Low ruminal pH may occur when feeding high-concentrate diets. However, because the reduction in pH occurs at the same time as the amount of concentrate fed increases, the changes observed in rumen fermentation may be attributed to pH or the type of substrate being fermented. Our objective was to determine the contribution of pH and type of substrate being fermented to the changes observed in rumen fermentation after supplying a high-concentrate diet. Eight dual-flow, continuous culture fermenters (1,400 mL) were used in 4 periods to study the effect of pH and type of diet being fermented on rumen microbial fermentation. Temperature (39°C), solid (5%/h), and liquid (10%/h) dilution rates, and feeding schedule were maintained constant. Treatments were the type of diet (FOR = 60% ryegrass and alfalfa hays and 40% concentrate; CON = 10% straw and 90% concentrate) and pH (4.9, 5.2, 5.5, 5.8, 6.1, 6.4, 6.7, and 7.0). Diets were formulated to have similar CP and ruminally undegradable protein levels. Data were analyzed as a mixed-effects model considering the linear, quadratic, and cubic effects of pH, the effects of diet, and their interactions. Semipartial correlations of each independent variable were calculated to estimate the contribution of each factor to the overall relationship. True digestion of OM and NDF were affected by pH, but not by type of diet. Total VFA were reduced by pH and were greater in CON than in FOR. Acetate and butyrate concentrations were reduced by pH but were not affected by diet. Propionate concentration increased as the pH decreased and was greater in CON than in FOR. Ammonia-N concentration decreased with decreasing pH and was lower in CON than in FOR. Microbial N flow was affected by pH, diet, and their interaction. Dietary N flow increased as pH decreased and was greater in CON than in FOR. The degradation of CP followed the opposite pattern, increasing as pH increased, and was less in CON than in FOR. The efficiency of microbial protein synthesis (g of N/kg of OM truly digested) was slightly reduced by pH and was less in CON than in FOR. These results indicate that the effects of feeding a high-concentrate diet on rumen fermentation are due to a combination of pH and substrate. Furthermore, the digestion of OM in high-concentrate diets is likely limited by the pH-induced effects on the microbial population activity.
Key Words: microbial fermentation • pH • diet
|
INTRODUCTION
|
|---|
In intensive production systems, acidosis affects from 14 to 40% of the animals in the herd, generating over $9 million losses each year in the United States (Oetzel et al., 1999
; Kleen, 2004). The supply of high-concentrate diets reduces DMI and fiber digestion, and the proportions of VFA in the rumen are modified. These changes have been attributed to the reduction in rumen pH (Erfle et al., 1982; Mould and Ørskov, 1983; Mould et al., 1983; Hoover et al., 1984). However, because in vivo the change in pH occurs as a consequence of feeding high-concentrate diets, the observed results are confounded between the effects of rumen acidosis and the effects of dietary concentrate level. High-concentrate diets tend to ferment toward propionate, and low rumen pH also results in greater ruminal propionate molar proportions (Kaufman et al., 1980). Which factor is responsible for the increase in rumen propionate? Is it the change in diet or the parallel change in rumen pH? This question is not trivial because if the effects depend on rumen pH, the development of strategies or technologies to control pH, such as the use of buffers, is essential, but if the effects are dependent on the diet, then we may need to choose between giving a high-concentrate diet and accepting the consequences, or limiting the intake of concentrate and reducing performance. Mould and Ørskov (1983) and Mould et al. (1983) were the first to suggest that the effects observed in acidosis were the combination of a pH and a substrate of fermentation effects. However, the design and execution of in vivo experiments to test this hypothesis is difficult because, in vivo, maintaining a high rumen pH with a high-concentrate diet or a low rumen pH with a high-forage diet may proof unfeasible. In vitro models that simulate rumen microbial fermentation are a useful tool in the development of this type of study because they allow for controlling the effects of diet type and pH independently. Russell (1998) demonstrated in a simple in vitro study that the change in acetate:propionate ratio observed after feeding concentrates was attributed mainly to the diet (75%) and to a lesser extent to the pH (25%). The separation of these 2 factors involved in the control of rumen fermentation is important to fully understand rumen function, to develop more accurate mathematical models, and to design strategies to prevent and control rumen acidosis. The objective of this study was to determine the effects of rumen pH and the type of diet on rumen microbial fermentation with the aim of quantifying the degree of the contributors of each factor on modifications of rumen microbial fermentation, and to develop regression equations to describe these effects.
|
MATERIALS AND METHODS
|
|---|
The research protocol was approved by the Campus Laboratory Animal Care Committee of the Universitat Autònoma of Barcelona (Spain).
Apparatus and Experimental Design
Eight 1,400-mL, dual-flow, continuous culture fermenters (Hoover et al., 1976) were used in 4 periods to study the effects of rumen pH and type of diet on microbial fermentation and nutrient flow. All other fermentation factors, including solid (5%/h) and liquid (10%/h) dilution rates and temperature (39°C), were maintained constant. The pH was maintained at each target level by infusion of 3 N HCl or 5 N NaOH. Fermentation parameters were monitored and controlled by a computer and a programmable linear controller, and fermentation conditions were programmed with LabView (FieldPoint, National Instruments, Austin, TX). Anaerobic conditions were maintained by infusion of N2 at a rate of 40 mL/min. Artificial saliva (Weller and Pilgrim, 1974) was continuously infused into the flasks and contained 0.4 g/L of urea to simulate recycled N.
The treatments were arranged in a 2 x 8 factorial design with 2 types of diets (Table 1; FOR = 60:40 forage:concentrate ratio typically fed to dairy cattle; CON =10:90 forage:concentrate ratio typically fed to feedlot cattle) and 8 pH levels (4.9, 5.2, 5.5, 5.8, 6.1, 6.4, 6.7, and 7.0), and were randomly assigned within periods. When the fermenters were fed the FOR diet, they were inoculated with a composited rumen fluid from 2 lactating dairy cattle fed the same FOR diet for at least 2 mo before the beginning of the trial. When the fermenters were fed the CON diet, they were inoculated with a composited rumen fluid from 2 heifers fed a similar CON diet for at least 2 mo before the beginning of the trial. These treatments were designed to provide 2 fermentation environments typically found in the rumen of cattle fed different substrates of fermentation. A total of 100 g of DM were fed continuously throughout the day. Diets were designed to meet or exceed current nutrient recommendations for a lactating Holstein cow (NRC, 2001) and a growing beef heifer (NRC, 1996) for the FOR and CON diets, respectively. Each experimental period consisted of 6 d for adaptation and 3 d for sampling.
During sampling days, collection vessels were maintained at 4°C to impede microbial action. Solid and liquid effluents were mixed and homogenized for 1 min, and a 500-mL sample was removed via aspiration. Upon completion of each period, effluent from the 3 d of sampling was composited and mixed within fermenter and homogenized for 1 min. Subsamples were taken for total N, ammonia-N, and VFA analyses. The remainder of the sample was lyophilized. Dry samples were analyzed for DM, ash, NDF, ether extract, and purine contents. Bacteria were isolated from the fermenter flasks on the last day of each period. Fermenter contents were homogenized for 1 min to dislodge solid phase bacteria, and strained through 2 layers of cheesecloth. Bacterial cells were isolated within 4 h by differential centrifugation at 1,000 x g for 15 min to eliminate feed particles, and at 10,000 x g for 15 min to isolate the bacterial pellet. Pellets were rinsed twice with saline solution and re-centrifuged at 10,000 x g for 15 min. The pellet was washed from the tubes with distilled water to prevent contamination of bacteria with ash. Bacterial cells were lyophilized and analyzed for DM, ash, N, and purine contents. Digestion of DM, OM, NDF, and CP, and flows of total N, NAN, microbial N, and dietary N were calculated as described by Stern and Hoover (1990).
Chemical Analyses
Effluent DM was calculated after lyophilizing 300-mL aliquots in triplicate with subsequent drying at 103°C in a forced-air oven for 24 h (AOAC, 1990). The DM content of diets and bacterial samples were determined by drying samples for 24 h in a 103°C forced-air oven (AOAC, 1990). Dry samples were ashed overnight at 550°C in a muffle furnace. Ether extract and total N were determined as described by AOAC (1990) procedures. The NDF of diets and effluents were analyzed sequentially by the detergent system (Van Soest et al., 1991) using a heat-stable alpha-amylase (Ankom Technology, Macedon, NY) and sodium sulfite. Samples for VFA were prepared as described by Jouany (1982) using 4-methylvaleric acid (Aldrich Chemical Company, Milwaukee, WI) as the internal standard. The analysis was performed by GLC (model 6890, Hewlett Packard, Palo Alto, CA) using a polyethylene glycol nitroterephthalic acid-treated capillary column (BP21, SGE, Europe Ltd., Buckinghamshire, UK). A 4-mL subsample of filtered fluid was acidified with 4 mL of 0.2 N HCl and frozen. Samples were centrifuged at 25,000 x g for 20 min, and the supernatant was analyzed for ammonia-N (Chaney and Marbach, 1962). Effluent and bacterial cells were analyzed for purine content by HPLC (Balcells et al., 1992).
Statistical Analysis
There was no need for data conversions because all data were normally distributed. Equations were developed with a value of 0 for FOR and 1 for CON. However, to avoid generating large correlations between the intercepts and the slopes, a new variable for pH was created, with values centered to the average pH (5.95) before fitting a mixed-effects regression model (Pinheiro and Bates, 2000). Therefore, when the pH = 5.95, the value was set to zero, and the corresponding change of pH as a positive or a negative value was referred to 5.95. Data were analyzed with a mixed-effects regression model, with random intercepts and random slopes of pH (mean centered) within each period, and diet and the linear, quadratic, and cubic pH effects, and the interaction between the diet and pH, as fixed effects. Thus, the mathematical expression of the model was
 |
where i indexes the pH levels and j the period numbers; xij represents the diet effect; zij, zij2, and zij3 represent the linear, quadratic, and cubic pH effects, respectively; wij represents any continuous variable that could affect yij (for example, CP degradation for NH3-N concentrations), whereas 
0j and 1j represent the corresponding intercept and random coefficient. The selected variance-covariance matrix was unstructured, which estimates a correlation for each pair of data. Independent variables with a P value > 0.10 were manually removed, and the resulting model was fitted again until obtaining a model with only coefficients with P values < 0.10. The F-test statistic was used to test that all the coefficients of the regressors (pH, diet) were all jointly zero, which would indicate that the model was significant.
To assess that the contribution to the changes observed in the dependent variables were either due to factors inherent to the diet or due to changes in pH, the contributing semipartial correlations (the percent of variance in the dependent variable which is not estimated by the other independent variables in the model that is explained by the variable under study) of each of the independent variables were calculated. The contribution of pH and diet to the explained variation was calculated by adding the contribution of the main effects (pH + pH2 + pH3; or diet) and 50% of the interaction between main effects, divided by the overall variation explained by the model. The sum of individual semipartial correlations may be different than the total coefficient of correlation because the residual colinearity between variables may be added twice and the random variable is not accounted for when conducting partial correlations.
|
RESULTS
|
|---|
Diets were formulated to be isonitrogenous and with the same rumen degradable protein content as estimated by the Cornell Net Carbohydrate and Protein System model (v. 5.0.3, Cornell University, Ithaca, NY). Differences in composition between diets were as expected based on their ingredient composition (Table 1
).
The effect of pH, diet, and their interactions explained 79% of the variation in the OM truly digested (OMTD), with an average value at mean pH (5.95) of 49.4% (Table 2, Figure 1a). Semipartial correlations were significant for pH (semipartial R2 = 0.26) and pH2 (semipartial R2 = 0.22), being less important for pH3 (semipartial R2 = 0.09). The largest change was associated with a linear effect of pH (14.9 unit reduction in OM digestion for each unit of change in pH). Diet had no effect on OMTD, but the significant pH x diet interaction indicates that the effect of pH was larger in FOR than in CON when pH fell below 6.0 (Table 2, Figure 1a). Average NDF digestion at mean pH (5.95) was 24.2%, ranging from less than 5% at pH 4.9 to 39% at pH 6.7 (Table 2, Figure 1b). The effect of pH, diet, and their interactions explained 60% of the observed variation, all being attributed to a pH effect. Semipartial correlations for pH (semipartial R2 = 0.11), pH2 (semipartial R2 = 0.11), and pH3 (semipartial R2 = 0.04) were significant, but in contrast with OMTD, the pH x diet interactions were not.
View this table:
[in this window]
[in a new window]
|
Table 2. Intercept, coefficients, SEM and semipartial correlations (pR2) for the significant variables in the model describing the effects of pH and type of diet on true OM digestion (OMTD) and NDF digestion (NDFD)
|
|
The data on VFA are presented as concentrations rather than as total production or molar proportions. In continuous culture fermenters, the flow of digesta is constant and there is no absorption; therefore, VFA concentrations are proportional to their production, and molar proportions can be easily calculated from total and individual VFA concentrations. The effect of pH, diet, and their interactions accounted for 81% of the variation observed in total VFA concentrations, which were affected by both pH and diet (Table 3, Figure 2a). Average total VFA concentration at mean pH (5.95) was 101.3 mM for FOR and was 20.4 mM greater in CON than in FOR (semipartial R2 = 0.31; Table 3). Total VFA concentration increased with pH (sum of semipartial R2 = 0.40). The effect of pH, diet, and their interactions explained 84% of the observed variation in acetate concentration (Table 3, Figure 2b), which was mostly affected linearly by pH (semipartial R2 = 0.81) with a 23.7 mM increase for each unit increase in pH. The model explained 79% of the overall variation in propionate concentration, and at mean pH (5.95) was 24.2 and 47.9 mM in FOR and CON, respectively (Table 3, Figure 2c). The concentration of propionate was mostly affected by a diet effect (semipartial R2 = 0.58), with the pH having a minor effect (sum of semipartial R2 = 0.19). The major effect of pH was linear (semipartial R2 = 0.10), increasing 19.5 mM per each unit decrease in pH. The effect of pH, diet, and their interactions explained 40% of the observed variation in butyrate concentration, which was 11.8 mM at the mean pH (5.95; Tale 3). Butyrate concentration was only affected by pH (semipartial R2 = 0.36) and pH3 (semipartial R2 = 0.22), and diet had no effect. The acetate:propionate ratio could be modeled from the combination of the decrease in acetate production due to pH, and the increase in propionate production due to the combined effect of pH and type of diet, and these factors explained 89% of the variation (Table 3, Figure 2d). This ratio was mainly affected by a linear effect of pH (semipartial R2 = 0.12, decreasing 1.93 per unit change of pH) and was 1.66 units lower in CON than in FOR (semipartial R2 = 0.40).
View this table:
[in this window]
[in a new window]
|
Table 3. Intercept, coefficients, SEM, and semipartial correlations (pR2) for the significant variables in the model describing the effects of pH and type of diet on total and individual VFA concentrations
|
|
Results of N metabolism are shown in Table 4. The effect of pH, diet, and their interactions explained 96% of the observed variation in ammonia-N concentration, which was affected by pH and diet (Figure 3a). Diet had the largest effect (semipartial R2 = 0.57), and at mean pH (5.95) ammonia-N concentration was 9.02 and 1.55 mg/100 mL in FOR and CON, respectively. Ammonia-N concentration decreased linearly as pH decreased (semipartial R2 = 0.11, decreasing 6.87 mg/100 mL for each unit reduction in pH), and quadratic (partial R2 = 0.04) and cubic (semipartial R2 = 0.06) effects were minor. As expected (total volume and passage rates were similar across fermenters), similar trends were found for the flow of ammonia-N (for which the model explained 96% of the observed variation). The effect of pH, diet, and their interactions explained 77% of the variation of microbial N flow (Figure 3c), which was affected mostly by diet (semipartial R2 = 0.37) and to a lesser extent by pH (sum of semipartial R2 = 0.23). The effect of pH, diet, and their interactions explained 82% of the variation in dietary N flow (Figure 3b), which increased as pH decreased (semipartial R2 = 0.14) and was greater in CON than in FOR (semipartial R2 = 0.45). The model explained 87% of the variation in CP degradation (Figure 3d), and followed the same patterns as dietary N flow, being affected mostly by diet (semipartial R2 = 0.35) and by a linear effect of pH (semipartial R2 = 0.13). On average, CP degradation of CON was 27.6 percentage units lower than FOR, and 1 unit of pH change resulted in a 24.5 percentage unit decrease in CP degradation. The measures of efficiency of microbial protein synthesis were calculated for energy (EMPS; g of bacterial N/kg of OMTD) and protein (ENU; g of bacterial N/100 g of rumen available N) as described by Bach et al. (2005; Figure 4a, b). The model explained 71% in the EMPS, but only 23% of the ENU. Variation in EMPS (average of 35.4 g of N/kg of OMTD at pH 5.95) was mostly explained by a diet effect (semipartial R2 = 0.63), whereas pH had a small effect (sum of semipartial R2 = 0.09). The ENU (average of 61.1 g of N/g of available N at pH 5.95) was mostly affected by a pH2 x diet interaction (semipartial R2 = 0.22), whereas the quadratic effects of pH (semipartial R2 = 0.16) and diet (semipartial R2 = 0.15) were less important.
View this table:
[in this window]
[in a new window]
|
Table 4. Intercept, coefficients, SEM, and semipartial correlations (pR2) for the significant variables in the model describing the effects of pH and type of diet on N metabolism
|
|
|
DISCUSSION
|
|---|
The pH was the major determinant of the changes in OMTD (pH contributed to 85% of the variation explained by the model). The pH x diet interaction for OMTD confirms that when pH falls below 6.0, the reduction in OMTD was much larger in FOR than in CON, but there was no direct diet effect. Average degradation of FOR (49.4%) agrees with results reviewed by Illg and Stern (1994), who reported that the average OMTD was 52.4% (n = 268) and 53.0% (n = 80) in dual flow continuous culture fermenters fed high-forage diets and in vivo trials, respectively. At first sight, it may seem surprising that the high-concentrate diet would have similar rumen OMTD compared with the high-forage diet because the CON diet was based on corn and barley (Table 1), and the degradability of these 2 ingredients in the rumen has been estimated to be around 70 to 80% using in situ techniques (Cerneau and Michalet-Doreau, 1991). However, OMTD of high-concentrate diets based on barley and corn tested in continuous culture have also resulted in values lower than expected (54 to 57%, Rotger et al., 2006a; 35%, Devant et al., 2001). Several trials have been conducted with rumen and duodenal canulae in cattle fed 90% concentrate for ad libitum intake (Cole et al., 1976a,b; Galyean et al., 1979a,b; Veira et al., 1980; Rahnema et al., 1987; Zinn et al., 1995; Choat et al., 2002). In these studies, the average OMTD of unprocessed diets was 53.0%, ranging from 43% (Veira et al., 1980) to 65% (Galyean et al., 1979a), and very close to the values reported in the present experiment. Therefore, the lower than expected OMTD observed with CON could not be considered an odd result. If the main ingredients of the CON diet were expected to be highly degradable (as suggested by in situ tabular values), the low degradation can only be attributed to a change in the microbial ecosystem. Then, there should be some differences in the microbial ecosystem of cattle fed a 90% concentrate diet that result in lower than expected OM digestion. We hypothesize that degradation of OM in fermenters was limited by microbial activity. This limitation may result from the bacteria reaching a limit in their growth, where an additional supply of energy would not improve growth or activity. Under these circumstances, bacteria reduce efficiency of growth and the degradation of nutrients is limited by either the number or the activity of bacteria. Additional results from this trial will provide further support for this hypothesis.
The fermenter fluid pH explained all the variation observed in NDF digestion accounted by the model, with diet having no effect. The pH x diet interaction observed in OMTD was not significant for NDF. The effect of pH was relatively small above pH 6.0, but degradability of NDF decreased sharply below this pH threshold. To our knowledge, few papers have addressed the contribution of pH and substrate of fermentation to the changes in fiber digestion. Mould and Ørskov (1983) and Mould et al. (1983) concluded that pH was a major determinant of the reduction of fiber and OM degradation, with a pH threshold around 6.0. Furthermore, the greater fiber content of the FOR diet compared with CON (33.4 vs. 24.9% NDF) justifies the interaction in OMTD, where the effects were larger for FOR compared with CON. The effect of pH on fiber digestion in the rumen has been extensively documented (Erfle et al., 1982; Mould and Ørskov, 1983; Mould et al., 1983; Hoover, 1986). The reduction in fiber digestion at low pH results from the reduction in the fibrolytic microbial population, and it has been attributed to a reduction in the ability of fibrolytic bacteria to attach to feed particles (Cheng et al., 1980), to the slow replication rate of fibrolytic bacteria under low pH (Russell and Dombrowski, 1980), or to both. Mould and Ørskov (1983) also suggested that the type of substrate being fermented could affect cellulolytic activity (the so-called carbohydrate effect), which had a significant, but small, contribution to the depression of fiber degradation. Although in general terms our data agree with those reported by Mould and Ørskov (1983) and Mould et al. (1983), we found no diet effect under the in vitro conditions of this experiment.
In contrast, the variation in total VFA concentration explained by the model was the result of the combined effect of pH and diet (contributing 56 and 44% to the variation explained by the model, respectively). Total VFA concentration decreased as pH decreased, consistent with the reduced OMTD. It is important to realize that in vivo, the VFA concentration is the driving force for reducing ruminal pH, and results presented herein may seem contradictory. However, the current experiment was designed to determine the effect of pH on rumen microbial fermentation, and the pH was changed without modifying diet fermentability. The reduction in pH resulted in a reduction in OMTD and VFA production, and may be part of the self-regulating mechanism of the rumen ecosystem against ruminal acidosis. Therefore, in vivo, the supplementation with highly degradable diets results in an increase in OMTD and VFA production and a reduction in pH, but at the same time, as the pH is reduced, OMTD and VFA production are also reduced. It is important that rumen models incorporate these 2 interrelated processes. On the other hand, a higher total VFA concentration in CON may be inconsistent with the lack of effect of diet on OMTD. However, total VFA are the results of their production after degradation of OM and the utilization of carbons for microbial protein synthesis (Dijktra et al., 1992). We hypothesize that the greater total VFA concentration without concomitant changes in OMTD in CON in a system where flow is constant and absorption does not exist is due to a change in the VFA profile (1 mol of butyrate is equivalent to 2 mol of acetate) or to a reduced use of carbon skeletons by microbes for growth, therefore increasing release as VFA. Because butyrate production was not affected by diet (Table 2) and microbial N flow was lower in CON than in FOR (Table 4), the latter hypothesis may seem more plausible. Furthermore, this provides additional evidence that microbial growth was limited in CON, which may further support the hypothesis that OMTD observed with CON was limited by microbial growth.
The fermenter fluid pH was the major determinant of the concentrations of acetate (it contributed to 98% of the explained variation). This strong effect of pH may be directly related to the decrease in fiber digestion with low pH because most fibrolytic bacteria are acetate producers (Kaufman et al., 1980). Furthermore, some bacteria (i.e., S. bovis) ferment starch into acetate, formate, and ethanol at high pH, but at low pH the pyruvate formate lyase is inhibited and fermentation turns into lactate (Asanuma et al., 1999). Therefore, the reduction in pH may explain the reduction in acetate production independently of the diet fed. In contrast, propionate concentration was the result of the combined effects of pH and diet that contributed to 55 and 45% of the variation explained by the model, respectively. The maximal production of propionate occurred, as expected, at pH 5.5, because at pH lower than 5.5, lactic acid producing bacteria proliferate at the expense of propionate producers and lactic acid accumulates (Nocek, 1997). The acetate:propionate ratio can be modeled from the pH effects on acetate and the combined effects of pH and diet on propionate concentrations. Overall, of the explained variation, 76% was attributed to a diet effect and 24% to a pH effect. These results are very similar to those of Russell (1998), who reported that diet and pH contributed 75 and 25% to the changes in the acetate:propionate ratio, respectively. Unfortunately, no other fermentation parameters were reported by Russell (1998), and we are not aware of other research that has provided additional data on the contribution of these factors to changes in rumen microbial fermentation.
Changes in N metabolism were explained more by the effect of diet (77% for ammonia-N concentration; 73% for dietary N flow; and 65% for protein degradation) than by the effect of pH. The effect of pH on protein degradation has been demonstrated previously (Hoover et al., 1984; Shriver et al., 1986; Calsamiglia et al., 2002). However, most of the reduction in ammonia-N concentration, dietary N flow, and CP degradation was due to a diet effect, although it could not be explained by the expected degradability of proteins using tabular values. In fact, diets were formulated to have similar CP degradabilities using the CNCPS (Sniffen et al., 1992) and NRC (1996, 2001) models, suggesting that our knowledge of protein degradation in high-concentrate diets is limited. In fact, Devant et al. (2000) already observed that in situ CP degradation of SBM was lower in the rumen of cattle fed a 90% concentrate diet (41%) compared with the same SBM incubated in the rumen of cattle fed a 60% forage diet (58%) even when rumen pH was always above 6.0 in both types of animals. Rotger et al. (2006b) reached similar conclusions. The reduced protein degradation in CON is also inconsistent with the fact that amylolytic bacteria tend to be more proteolytic than cellulolytic bacteria (Wallace and Cotta, 1989). Therefore, more research appears to be required to understand the processes involved in protein degradation in the rumen of cattle fed high-concentrate diets.
The variation in microbial N flow was explained by a combination of pH and diet effects (39 and 61% of the explained variation of the model, respectively). Microbial N flow decreased as pH decreased. At low pH, bacteria spend part of the available energy in maintaining the proton-motive force across the cell membrane increasing maintenance requirements at the expenses of growth (Wallace and Cotta, 1989). Microbial N flow was, on average, 26% lower in CON compared with FOR. The limited growth of bacteria under CON diets may be associated to either the greater maintenance requirements of amylolytic bacteria (Russell et al., 1992), or to a potential limitation of available N in the fermenter fluid. Ammonia-N concentration in CON was below the 5 mg of N/ 100 mL suggested as the minimum required for optimal microbial growth (Satter and Slyter, 1974), and microbial protein synthesis may have been compromised. Efficiency of microbial protein synthesis was explained mostly by a diet effect (78% of the explained variation), with the pH having a smaller effect (22% of the explained variation). The limited impact of pH on the EMPS agrees with data by Hoover and Miller (1992) in vitro and Bach et al. (2005) in vivo, who summarized several studies where pH was modified and observed that the efficiency of microbial protein synthesis was only affected when pH decreased below 5.5. The EMPS was 10.9 units lower in CON than in FOR. Because OMTD was not affected by type of diet, the reduced efficiency is attributed to a decreased synthesis of microbial protein, which is confirmed by the lower bacterial N flow in CON. In fact, as discussed above, the lack of effect of type of diet on OMTD together with greater VFA concentration in CON suggests that VFA accumulated due to lack of utilization of carbon skeletons for microbial protein synthesis. The lower microbial N flow and efficiency of microbial cell synthesis provides further support to our hypothesis that growth or activity of microbes in CON was limited in spite of energy availability, which resulted in an OMTD lower than expected and an accumulation of VFA.
Results indicate that changes in nutrient digestion and microbial fermentation are due to the combined effects on pH and diet in different proportions depending on the measure. Changes in true OM and fiber degradation, and acetate and butyrate concentrations are mostly dependent on pH, whereas the acetate:propionate ratio, ammonia-N concentration, the flow of ammonia, dietary and microbial N, and efficiency of microbial protein synthesis are mostly affected by the type of diet. Total VFA and propionate concentrations are affected by both pH and diet in similar proportions. These effects should be incorporated into future mathematical models of rumen fermentation to help better understand rumen microbial metabolism. The microbial degradation of OM in high-concentrate diets appears to be limited by the growth or activity of microbes. This hypothesis and its implication for animal performance deserve further research.
|
Footnotes
|
|---|
1 This research was partially funded by the project AGL2002-01642 of the Spanish government. 
2 Corresponding author: Sergio.Calsamiglia{at}uab.es
Received for publication March 7, 2007.
Accepted for publication October 31, 2007.
|
LITERATURE CITED
|
|---|
AOAC. 1990. Official Methods of Analysis. 15th ed. Assoc. Off. Anal. Chem., Arlington, VA.
Asanuma, N., M. Iwamoto, and T. Hino. 1999. Structure and transcriptional regulation of the gene encoding pyruvate formate–lyase of a ruminal bacterium, Streptococcus bovis. Microbiol. 145:151–157.[Abstract/Free Full Text]
Bach, A., S. Calsamiglia, and M. D. Stern. 2005. Nitrogen metabolism in the rumen. J. Dairy Sci. 88:E9–E21.[Abstract/Free Full Text]
Balcells, J., J. Guada, J. M. Peiró, and D. S. Parker. 1992. Simultaneous determination of allantoin and oxypurines in biological fluids by high performance liquid chromatography. J. Chromatogr. 575:153–157.[Medline]
Calsamiglia, S., A. Ferret, and M. Devant. 2002. Effects of pH and pH fluctuations on microbial fermentation and nutrient flow from a dual-flow continuous culture system. J. Dairy Sci. 85:574–579.[Abstract]
Cerneau, P., and B. Michalet-Doreau. 1991. In situ starch degradation of different feeds in the rumen. Reprod. Nutr. Dev. 31:65–72.[Medline]
Chaney, A. L., and E. P. Marbach. 1962. Modified reagents for determination of urea and ammonia. Clin. Chem. 8:130–132.[Abstract]
Cheng, K. J., J. Fay, and R. Howarth. 1980. Sequence of events in the digestion of fresh legume leaves by rumen bacteria. Appl. Environ. Microbiol. 40:613–619.[Abstract/Free Full Text]
Choat, W. T., C. R. Krehbiel, M. S. Brown, G. C. Duff, D. A. Walker, and D. R. Gill. 2002. Effects of restricted versus conventional dietary adaptation on feedlot performance, carcass characteristics, site and extent of digestion, digesta kinetics, and ruminal metabolism. J. Anim. Sci. 80:2726–2739.[Abstract/Free Full Text]
Cole, N. A., R. R. Johnson, and F. N. Owens. 1976a. Influence of roughage level on the site and extent of digestion of whole shelled corn by beef steers. J. Anim. Sci. 43:483–489.[Abstract/Free Full Text]
Cole, N. A., R. R. Johnson, and F. N. Owens. 1976b. Influence of roughage level and corn processing method on the site and extent of digestion by beef steers. J. Anim. Sci. 43:490–496.[Abstract/Free Full Text]
Devant, M., A. Ferret, S. Calsamiglia, R. Casals, and J. Gasa. 2000. Effects of protein concentration and degradability on performance, ruminal fermentation, and nitrogen metabolism in rapidly growing heifers fed high-concentrate diets from 100 to 230 kg body weight. J. Anim. Sci. 78:1667–1676.[Abstract/Free Full Text]
Devant, M., A. Ferret, S. Calsamiglia, R. Casals, and J. Gasa. 2001. Effect of nitrogen source in high-concentrate, low-protein beef cattle diets on microbial fermentation studied in vivo and in vitro. J. Anim. Sci. 79:1944–1953.[Abstract/Free Full Text]
Dijktra, J., H. D. S. C. Neal, D. E. Beever, and J. France. 1992. Simulation of nutrient digestion, absorption and outflow in the rumen. Model description. J. Nutr. 122:2239–2256.[Abstract/Free Full Text]
Erfle, J. D., R. J. Boila, R. M. Teather, S. Mahadevan, and F. D. Sauer. 1982. Effect of pH on fermentation characteristics and protein degradation by rumen microorganisms in vitro. J. Dairy Sci. 65:1457–1464.[Abstract/Free Full Text]
Galyean, M. L., D. G. Wagner, and F. N. Owens. 1979a. Corn particle size and site and extent of digestion by steers. J. Anim. Sci. 49:204–210.[Abstract/Free Full Text]
Galyean, M. L., D. G. Wagner, and F. N. Owens. 1979b. Level of feed intake and site and extent of digestion of high concentrate diets by steers. J. Anim. Sci. 49:199–203.[Abstract/Free Full Text]
Hoover, W. H. 1986. Chemical factors involved in ruminal fiber digestion. J. Dairy Sci. 69:2755–2766.[Abstract/Free Full Text]
Hoover, W. H., B. A. Crooker, and C. J. Sniffen. 1976. Effects of differential solid-liquid removal rates on protozoa numbers in continuous cultures of rumen contents. J. Anim. Sci. 43:528–534.[Abstract/Free Full Text]
Hoover, W. H., C. R. Kincaid, G. A. Varga, W. V. Thayne, and L. L. Junkins. 1984. Effects of solid and liquid flows on fermentation in continuous cultures. IV. pH and dilution rates. J. Anim. Sci. 58:692–699.[Abstract/Free Full Text]
Hoover, W. H., and T. K. Miller. 1992. Rumen digestive physiology and microbial ecology. Bull. 708T, Agric. Forestry Exp. Sta., West Virginia University, Morgantown, WV.
Illg, D. J., and M. D. Stern. 1994. In vitro and in vitro comparisons of diaminopimelic acid and purines for estimating protein synthesis in the rumen. Anim. Feed Sci. Technol. 48:49–55.[CrossRef]
Jouany, J. P. 1982. Volatile fatty acids and alcohol determination in digestive contents, silage juice, bacterial cultures and anaerobic fermentor contents. Sci. Aliments 2:131–144.
Kaufman, W., H. Hagemeister, and G. Dirksen. 1980. Adaptation to changes in dietary composition, level and frequency of feeding. Page 587 in Digestive Physiology and Metabolism in Ruminants. Y. Ruckebusch and P. Thivend, ed. AVI Publishing, Westport, CT.
Kleen, J. L. 2004. Prevalence of subacute ruminal acidosis in Dutch dairy herds – A field study. PhD Thesis Tierärztliche Hochschule, Hannover, Germany.
Mould, F. L., and E. R. Ørskov. 1983. Manipulation of rumen fluid pH and its influence on cellulolysis in sacco, dry matter degradation and the ruminal microflora of sheep offered either hay or concentrate. Anim. Feed Sci. Technol. 10:1–14.[CrossRef]
Mould, F. L., E. R. Ørskov, and S. O. Mann. 1983. Associative effects of mixed feeds. I. Effect of type and level of supplementation and the influence of the rumen fluid pH on cellulolysis in vivo and dry matter digestion on various roughages. Anim. Feed Sci. Technol. 10:15–30.[CrossRef]
Nocek, J. E. 1997. Bovine acidosis: Implications on laminitis. J. Dairy Sci. 80:1005–1028.[Abstract]
NRC. 1996. Nutrient Requirements of Beef Cattle, 7th ed. Natl. Acad. Press, Washington, DC.
NRC. 2001. Nutrient Requirements of Dairy Cattle, 7th rev. ed. Natl. Acad. Press., Washington, DC.
Oetzel, G. R., K. V. Nordlund, and E. F. Garrett. 1999. Effect of ruminal pH and stage of lactation on ruminal lactate concentrations in dairy cows. J. Dairy Sci. 82(Suppl. 1):38–39. (Abstr.)
Pinheiro, J. C., and D. M. Bates. 2000. Mixed-Effects Models in S and S-PLUS, Statistics and Computing. Springer Ltd., Berlin, Germany.
Rahnema, S. H., B. Theurer, J. A. Garcia, W. H. Hale, and M. C. Young. 1987. Site of protein digestion in steers fed sorghum grain diets. II. Effect of grain processing methods. J. Anim. Sci. 64:1541–1547.[Abstract/Free Full Text]
Rotger, A., A. Ferret, S. Calsamiglia, and X. Manteca. 2006a. Effects of nonstructural carbohydrates and protein sources on intake, apparent total tract digestibility, and ruminal metabolism in vivo and in vitro with high-concentrate beef cattle diets. J. Anim. Sci. 84:1188–1196.[Abstract/Free Full Text]
Rotger, A., A. Ferret, S. Calsamiglia, and X. Manteca. 2006b. In situ degradability of seven plant protein supplements in heifers fed high concentrate diets with different forage to concentrate ratio. Anim. Feed Sci. Technol. 125:73–87.[CrossRef]
Russell, J. B. 1998. The importance of pH in the regulation of ruminal acetate to propionate ratio and methane production in vitro. J. Dairy Sci. 81:3222–3230.[Abstract]
Russell, J. B., and D. B. Dombrowski. 1980. Effect of pH on the efficiency of growth by pure cultures of rumen bacteria in continuous culture. Appl. Environ. Microbiol. 39:604–610.[Abstract/Free Full Text]
Russell, J. B., J. D. O’Connor, D. G. Fox, P. J. Van Soest, and C. J. Sniffen. 1992. A net carbohydrate and protein system for evaluating cattle diets: I. Ruminal fermentation. J. Anim. Sci. 70:3551–3561.[Abstract]
Satter, L. D., and L. L. Slyter. 1974. Effect of ammonia concentration on rumen microbial protein production in vitro. Br. J. Nutr. 32:199–208.[CrossRef][Medline]
Shriver, B. J., W. H. Hoover, J. P. Sargent, R. J. Crawford Jr., and W. V. Thayne. 1986. Fermentation of a high concentrate diet as affected by ruminal pH and digesta flow. J. Dairy Sci. 69:413–419.[Abstract/Free Full Text]
Sniffen, C. J., J. D. O’Connor, P. J. Van Soest, D. G. Fox, and J. B. Russell. 1992. A net carbohydrate and protein system for evaluating cattle diets: II. Carbohydrate and protein availability. J. Anim. Sci. 70:3562–3577.[Abstract]
Stern, M. D., and H. W. Hoover. 1990. The dual flow continuous culture system. Pages 17–32 in Proc. Continuous Culture Fermenters: Frustration or Fermentation. Northwest ADSA-ASAS Regional Meeting, Chazy, NY.
Van Soest, P. J., J. B. Robertson, and B. A. Lewis. 1991. Methods for dietary fiber, neutral detergent fiber and nonstarch polysaccharides in relation to animal nutrition. J. Dairy Sci. 74:3583–3597.[Abstract]
Veira, D. M., G. M. MacLeod, and J. H. Burton. 1980. Nutrition of the weaned Holstein calf. II. Effects of dietary protein level on nitrogen balance, digestibility and feed intake. J. Anim. Sci. 50:945–951.[Abstract/Free Full Text]
Wallace, R. J., and M. A. Cotta. 1989. Metabolism of nitrogen-containing compounds. Pages 217–250 in The Rumen Microbial Ecosystem. P. N. Hobson, ed. Elsevier Applied Science, New York, NY.
Weller, R. A., and A. F. Pilgrim. 1974. Passage of protozoa and volatile fatty acids from the rumen of a sheep and from a continuous in vitro fermentation system. Br. J. Nutr. 32:341–351.[CrossRef][Medline]
Zinn, R. A., C. F. Adam, and M. S. Tamayo. 1995. Interaction of feed intake level on comparative ruminal and total tract digestion of dry-rolled and steam-flaked corn. J. Anim. Sci. 73:1239–1245.[Abstract]
Top
Abstract
INTRODUCTION
; and 10:90 forage:concentrate, ----3 
) on the true digestion of (A) OM and (B) NDF.
