| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ANIMAL NUTRITION |
,3
* Department of Animal and Veterinary Science, University of Idaho, Moscow 83844-2330 and
and
Agriculture and Agri-Food Canada Research Centre, Lethbridge, AB, Canada T1J 4B1
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
Key Words: Cattle • Fatty Acids • Ruminal Fermentation • Ruminal Protozoa
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
), but because the ruminal environment is anaerobic, catabolism of saturated FA provides little energy to resident microorganisms (Nagaraja et al., 1997). Substituting fat for carbohydrate in diets for ruminants, therefore, decreases the amount of ATP available for microbial growth (Firkins, 1996). Dietary fats often inhibit ruminal microbial activities and fiber digestion (Brooks et al., 1954; Henderson, 1973; Jenkins, 1993).
Ciliate ruminal protozoa possess strong proteolytic capacity (Williams and Coleman, 1992) and are thought to contribute to intraruminal recycling of microbial N (Jouany, 1996). Oils rich in saturated C12 and C14 and unsaturated C16 and C18 FA inhibit ruminal protozoa in vitro and in vivo (Henderson, 1973; Newbold and Chamberlain, 1988; Machmüller and Kreuzer, 1999), but clear evidence linking the effects of oils rich in long-chain unsaturated FA to their high C18:2 or C18:3 content is lacking (Doreau and Ferlay, 1995). Dietary FA may exert a number of effects on bacterial activities and, consequently, on overall ruminal fermentation (Nagaraja et al., 1997). Studies with free saturated medium-chain and unsaturated long-chain FA have noted inhibition of ruminal protozoa but provided little or no information on the effects of these FA on other variables of ruminal fermentation (Matsumoto et al., 1991; Ajisaka et al., 2002).
The objective of this study was to investigate the effects of the free FA present in commonly fed fat supplements for ruminants on protozoal numbers and on protozoal and bacterial fermentative activities in ruminal fluid. We hypothesized that certain FA would decrease protozoal counts and activities without having an overall negative influence on the contribution of ruminal bacteria to fermentation.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
). Group 1 included those FA accounting for over 40% of the total FA in commonly fed fat supplements: lauric (in coconut oil); palmitic (in palm oil); linoleic and linolenic (in linseed and soybean oils); and oleic (in beef tallow or palm oil). They were included in the incubations at 0.25, 0.50, and 1.0% (wt/vol). Group 2 included myristic and stearic acids; these make up 20 to 40% of the fat in coconut oil and mutton tallow, respectively, and were included in the incubations at 0.125, 0.25, and 0.50%. Group 3 included caproic, caprylic, and capric acids, present in fat supplements (e.g., coconut oil, palm kernel oil) at less than 10% and these were included in the incubations at 0.0625, 0.125, and 0.25%. Inclusion levels in inoculum were selected to simulate concentrations that might occur in vivo, assuming immediate hydrolysis of a diet with 4% supplemental fat, 80 kg of total ruminal content, and DM intake of 20 kg/d. For each of the FA tested, the three treatment levels were denoted low (Lo), medium (Med), and high (Hi).
The effect of FA on protozoal predation on bacteria was studied by including 15N-labeled casein in the incubations (for preparation, see Hristov et al., 2002). The 15N-casein used in the study contained 16.0% N (SE = 0.22) with 0.1977 atom % excess 15N (SE = 0.00020). The FA and all other chemicals used in this study were purchased from Sigma Chemical Co. (St. Louis, MO).
Ruminal Inoculum
Inoculum for the in vitro incubations was prepared from ruminal contents of two previously cannulated Hereford heifers (503 ± 8.9 kg BW and 8.7 ± 0.04 kg/d DMI; mean ± SE). The heifers had ad libitum access to a diet that comprised (DM basis) 90% steam-rolled barley grain, 4% barley silage, 5% soybean meal, and 1% mineralized salt (Hristov et al., 1999). Feed was delivered at 0900 and 1600 daily. The donor heifers used in the experiment were managed according to the guidelines of the Canadian Council on Animal Care (CCAC, 1993).
Two incubations were conducted 7 d apart. On incubation days, ruminal contents were collected from the reticulum, ventral sac, and middorsal feed mat of each heifer 2 h after the morning feeding. The samples were combined and filtered by manual compression through two layers of cheesecloth. Filtrate volume was recorded, and the solids retained by the cheesecloth were suspended in a volume of buffer equal to the filtrate volume. The buffer (McDougall, 1948) was amended with DL-glucose (5 g/L) and L-cysteine-HCl (0.5 g/L), gassed with CO2, and warmed to 39°C before use. The suspension, agitated vigorously to detach microorganisms loosely adherent to the solids (Hristov and Broderick, 1994), was then filtered again through cheesecloth. The two filtrates were combined, and the pooled sample was transported immediately to the laboratory (a distance of 0.5 km) in an insulated container. Feed particles were separated from the fluid by flotation (Hristov and Broderick, 1994) and discarded. The remaining fluid was used immediately for the in vitro incubations.
In Vitro Incubations
The in vitro incubations were conducted at 39°C in sealed 150-mL culture bottles. The FA and 15N-casein (286 mg per bottle) were weighed into the bottles before preparation of the ruminal inoculum. All FA except linolenic acid were in the form of sodium salts. Forty milliliters of prewarmed, pregassed buffer (as described above) was added to all vials except those receiving linolenic acid. Linolenic acid was dissolved in 40 mL of buffer in a 100-mL beaker by two 30-s cycles with a sonication probe and then transferred to the bottles.
Once the 40 mL of buffer was added, each culture bottle was flushed with O2-free CO2, infused with 80 mL of inoculum, sealed, and placed in wire baskets fitted onto shaking platforms in an incubator set at 39°C. Start times (addition of inoculum) were staggered and recorded to ensure 4 h of incubation for each bottle. Maximum time from collection of ruminal fluid to commencement of incubation was 30 min. Duplicate bottles, including controls (15N-casein only) were prepared for each level of each FA studied; thus, for each treatment, n = 4 (after two incubations), and for the controls, n = 12.
Sample Analyses
On completion of the 4-h incubation, culture bottles were inserted into shaved ice, and the pH of the incubation liquid was recorded. After thorough mixing, subsamples were withdrawn for enumeration of protozoa, for determination of enzyme activities and concentrations of fermentation products, and for isolation of bacterial and protozoal pellets.
For protozoal enumerations, 4 mL of incubation liquid was added to 4 mL of methyl green:formalin:saline (MFS) solution (Ogimoto and Imai, 1981) and stored at 4°C until counted as described by Hristov et al. (2001). A 30-mL sample of incubation liquid was transferred directly to storage at –40°C for later determination of soluble protein (SP) concentration (by automated Lowry assay; Oosta et al., 1978) and polysaccharide-degrading and deaminative enzyme activities as described by Hristov et al. (1998, 1999). Twenty milliliters of incubation liquid was mixed with 1.7 mL of 65% (wt/vol) trichloroacetic acid and stored on ice for 30 min, and then at 4°C overnight. Following centrifugation (28,000 x g, 20 min, 4°C), concentrations of ammonia, total free amino acids (TFAA), and reducing sugars (RS) were determined in the supernatant fluid (Broderick and Kang, 1980; Hristov et al., 1998).
The remaining incubation liquid was preserved with 5% formalin (final concentration) and centrifuged at 400 x g (5 min, 4°C). The supernatant fluid (40 mL) was combined with 2 mL of saturated HgCl2 for isolation of bacteria. The pellets (protozoal fraction) were washed three times with 10% (vol/vol) formalin in 0.9% (wt/vol) NaCl, and then freeze-dried and analyzed for 15N-enrichment of the total N. Bacteria were isolated from the preserved supernatant by centrifugation (28,000 x g, 20 min, 4°C). The pellets were freeze-dried and analyzed for 15N-enrichment of the total N. Concentrations of lactate and VFA in the supernatant were determined as described by Goodall and Byers (1978) and Hristov et al. (2001), respectively.
Total N and 15N-enrichment of the bacterial and protozoal pellets were determined using a Nitrogen Analyzer 1500 (Carlo Erba Instruments, Milan, Italy) connected to a mass-ratio spectrometer (Optima mass-spectrometer, VG Instruments, Middlewich, U.K.). The proportion of protozoal protein originating from bacterial protein (Pzoa/Bact) was calculated as
 |
where 15Nprotozoa is the corrected 15N-enrichment (atom % excess) of protozoal pellets at 4 h, and 15Nbacteria is 15N-enrichment (atom % excess) of bacterial pellets at 4 h (Hristov et al., 2003). The corrected value for 15N-enrichment of protozoal pellets was calculated as
 |
where 15N(protozoa)t is the enrichment measured in protozoal pellets from treatment, and 15N(protozoa)c is the enrichment measured in the incubations that by light microscopic evaluation had no protozoa (i.e., the capric and lauric acid incubations).
Statistical Analyses
Data were analyzed as a completely randomized design using the GLM procedure of SAS (SAS Institute Inc., Cary, NC) as a split plot, as two separate controls were used in each replicate incubation. Within each replicate, incubation treatments were arranged in a 10 x 3 factorial (FA x level of inclusion), with a control with duplicate vials run for each treatment. When the FA x level interaction was significant (< 0.05), data were analyzed within application level; treatment means were separated by pairwise -test. Correlations between fermentation variables and protozoal numbers and extent of protozoal and bacterial incorporation of the 15N tracer were determined by Pearson’s method.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
). Incubation pH ranged from 6.97 (lauric acid) to 7.67 (capric acid). Across treatments, increasing the FA inclusion level increased (P < 0.01) the pH of the incubation medium.
|
). Lauric acid at all inclusion levels markedly decreased (P < 0.05) ammonia concentrations relative to the control. Ammonia also was or tended to be decreased (P < 0.05 or P < 0.10) to a lesser extent by linolenic and linoleic acids (except linoleic Med), and by the highest levels of caprylic (P < 0.05) and caproic (P < 0.10) acids. A treatment x level interaction (P < 0.001) was also observed for TFAA. Caprylic acid (Hi), and capric and lauric acids (Lo, Med, Hi) caused over fourfold increases (P < 0.05) in free amino acid concentrations in the incubation medium (Table 1). Moreover, TFAA concentrations were doubled (P < 0.05), relative to the controls, by linoleic (Med, Hi) and linolenic (Med) acids, and nearly so by linolenic (Hi) and caprylic (Med) acids.
Treatment-mediated effects on RS concentrations were similar to those observed for TFAA, with a treatment x level (P < 0.001) interaction observed. Capric and lauric acids (Lo, Med, Hi) and caprylic acid (Hi) caused substantial increases (P < 0.05) in RS concentrations (Table 1). The increases effected by capric and lauric acids were linear (P < 0.01 and P < 0.05, respectively). Regression coefficients were 1.078 (r2 = 0.99) for capric acid, and 0.874 (r2 = 0.97) for lauric acid. Linoleic and linolenic acids more than doubled RS concentrations, again reflective of their effects on TFAA, but, in the case of RS, these increases did not attain significance.
The effects of FA on SP concentrations were less extensive than their effects on other fermentation variables with a less pronounced (P < 0.05) treatment x level interaction (Table 1). Capric and lauric acids (Lo, Med, Hi), linolenic and myristic acids (Med and Hi), and linoleic acid (Hi) all increased SP concentrations in the incubation media (P < 0.05; except capric acid [Hi], for which P < 0.10). In contrast, caproic acid (Med) slightly decreased (P < 0.05) SP concentration.
No treatment x level interaction was observed for VFA. Averaged across FA inclusion levels, concentrations of acetic and propionic acids (40.6 and 21.3 mmol/L, respectively) were least (P < 0.001) when capric acid was included in the incubation medium (Table 1). Averaged across treatments (FA types), overall mean acetate concentrations decreased (P < 0.001) as the FA inclusion levels increased. At Lo, Med, and Hi application levels, average acetate concentrations were 60.0, 55.5, and 54.2 mmol/L, respectively (data not shown). Propionate concentrations were not affected (P = 0.11) by treatment level.
Concentrations of branched-chain fatty acids (BCFA) and butyrate were decreased (P < 0.001) relative to controls when capric, lauric, linoleic, or linolenic acids were included in the incubation. Compared with the control, capric and lauric acids also decreased (P < 0.001) total VFA concentrations (by 29 and 22%, respectively). Conversely, myristic and oleic acids increased (P < 0.001) total VFA concentrations (by 9 and 8%, respectively). Averaged across treatments, concentrations of butyric and valeric acids, and total VFA decreased (P < 0.001; P < 0.05; and P < 0.001, respectively) as FA inclusion levels increased (data not shown). Consistent with these and the effects observed on acetate and propionate concentrations, acetate:propionate ratios also decreased (P < 0.001) with increasing FA inclusion, averaging (across treatments) 2.10, 2.10, 2.00, and 1.94 in the control, Lo, Med, and Hi incubations, respectively (data not shown). Individually, capric, linolenic, and linoleic acids decreased (P < 0.001) acetate:propionate ratios compared with the control.
A slight treatment x level interaction (P < 0.05) was observed for lactate concentrations. Consistent with observations of other fermentation variables measured, effects on lactate concentrations (increased to 1.5x to 2x the control; P < 0.05) were exerted primarily by lauric (Lo, Med, Hi), capric (Lo, Hi), caprylic (Hi), and linolenic (Lo, Med; P < 0.10 for Hi) acids, and also by linoleic and oleic acids (Med; P < 0.05).
In general, polysaccharide-degrading enzyme activities in the incubation mixtures increased with increasing concentrations of FA, resulting in a treatment x level interaction (P < 0.01) (Table 2). Four of the 10 FA studied (linolenic, linoleic, oleic, and myristic) substantially increased (P < 0.05) xylanase and amylase activities at all three levels of inclusion. Caprylic acid exerted the same effect (P < 0.05) on xylanase activity, and enhanced amylase activity to a lesser extent (P < 0.05 at Hi; P < 0.10 at Lo; numerical increase at Med). Capric and lauric acids exhibited a different pattern of effect on these enzyme activities. At Lo and Med, these FA markedly increased (P < 0.05) amylase activity, but at Hi they inhibited (P < 0.05) it. Capric acid exerted the same pattern of effect on xylanase activity; lauric acid (Hi) also tended to inhibit (P < 0.10) xylanase activity.
|
).
Treatment x level interactions (P < 0.001) were observed for protozoa numbers, protozoal 15N, bacterial 15N, and Pzoa/Bact. Whereas stearic acid caused slight increases in protozoal numbers (P < 0.05 at Lo and Med), the other FA were inhibitory or did not affect these populations (Table 3). All levels of capric and lauric acids inhibited protozoa entirely (P < 0.05) as did caprylic, linolenic, and linoleic acids at level Hi. Linolenic and linoleic (both at Lo and at Med), and caprylic acid (Med) decreased (P < 0.05), but did not eliminate, protozoa from the incubation medium. The same was true for all levels of myristic and oleic acids. No effects of caproic (P = 0.42 to P = 0.84) or palmitic (P = 0.13 to P = 0.79) acids on protozoal numbers were observed. On average (mean ± SE), the total protozoal populations were made up of 99.8 ± 0.3% Entodinium spp. (data not shown). Dasytricha or Isotricha spp. were detected only in low numbers (1,000 to 5,000 per mL) in individual incubations of control, caproic, caprylic, palmitic, and stearic acids (data not shown). The correlation between protozoal counts and incorporation of 15N into protozoal N was 0.83 (P < 0.001, Figure 1).
|
|
). The residual 15N enrichments (mean ± SE) found in these two treatments (0.0060 ± 0.00069 and 0.0060 ± 0.00044 atom % excess, respectively) were deemed to represent nonprotozoal 15N contamination, and were used to correct 15N enrichments of protozoal N for all other treatments (see Materials and Methods). Incorporation of 15N into protozoal N was unaffected by palmitic and stearic acids but was decreased (P < 0.05) in all other incubations in which protozoa were detected microscopically, except caproic acid (Lo). This effect was most pronounced with the unsaturated long-chain FA (oleic, linoleic, and linolenic). Other than decreases (P < 0.05) by capric and lauric acids (Lo, Med, Hi), caprylic (Med and Hi), and linolenic at Hi only, incorporation of 15N tracer into bacterial N was largely unaffected by FA. Oleic, linoleic, and myristic acids at Lo slightly increased (P < 0.05) 15N incorporation into bacterial N, as did stearic and palmitic (at Med and Hi; P < 0.1 for palmitic).
With no supplementary FA (i.e., in the controls), 50% of the protozoal protein present at the incubation end point, on average, originated from bacterial N (Table 3). None of the treatments increased this proportion (Pzoa/Bact). All levels of oleic and myristic acids decreased (P < 0.05) Pzoa/Bact, and the effect was linear (P < 0.01) for myristic acid (regression coefficient 0.188; r2 = 0.99). Caprylic, linoleic, and linolenic acids at Lo and Med decreased Pzoa/Bact relative to the control. Proportions could not be calculated for the high level of these acids owing to eradication of protozoa. Medium and high levels of caproic acid slightly decreased (P < 0.05) Pzoa/Bact, but the effect was less pronounced than was observed with the unsaturated FA.
Ammonia concentration was highly correlated with the total protozoal numbers (and Entodinium spp., r = 0.67, P < 0.001; data not shown). Positive correlations were observed between ammonia concentration and incorporation of the 15N tracer into protozoal N (r = 0.51, P < 0.001, Figure 1) and into bacterial N (r = 0.28, P < 0.001, Figure 2). Unlike ammonia, concentration of TFAA was negatively correlated to 15N incorporation into protozoal and bacterial cells (r = –0.78 for both, P < 0.001). Concentration of TFAA was negatively related to total protozoal numbers (r = –0.84, P < 0.001). Concentrations of RS and SP correlated negatively with the extent of 15N incorporation into protozoal N (r = –0.59 and –0.66, respectively, P < 0.001), and with total protozoal numbers (–0.59 and –0.60, P < 0.001). Total VFA concentration was positively correlated with tracer incorporation into protozoal N and into bacterial N (r = 0.53 and 0.48, respectively, P < 0.001), with ammonia concentration (r = 0.60, P < 0.001) and with total protozoal numbers (r = 0.66, P < 0.001), but was negatively correlated with TFAA (r = –0.66, P < 0.001) and RS (r = –0.63, P < 0.001) concentrations. Concentration of BCFA was highly correlated with protozoal (r = 0.70, P < 0.001) and bacterial (r = 0.41, P < 0.001) incorporation of the 15N-tracer, and with protozoal numbers (r = 0.75, P < 0.001). Lactate concentration was negatively correlated with protozoal numbers (r = –0.46, P < 0.001). Xylanase and amylase activities were negatively correlated with RS concentration (r = –0.25, P < 0.01; and r = –0.19, P < 0.05, respectively). Xylanase activities correlated positively with the total concentration of VFA (r = 0.23, P < 0.01). Correlations among the three polysaccharide-degrading enzyme activities measured were high and positive (r = 0.74 to 0.79, P < 0.001). Total protozoal numbers correlated well with the 15N-incorporating activities of protozoa and bacteria (P < 0.001; r = 0.83 and 0.69, respectively).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
), the FA may be required in greater concentrations in vivo than were used in vitro to achieve the same effects. Further, Oldick and Firkins (2000) suggested that frequent feeding may offset the effect of oil supplementation; thus, the effect of a single dose of FA may differ from the effect of continuous feeding of dietary fat.
Our approach in determining protozoal activity was based on using 15N-labeled casein as a N source for ruminal bacteria, with the assumption that protozoa would consequently be labeled primarily through ingestion of labeled bacterial protein, along with incorporation of a small portion of the tracer through direct uptake of 15N-casein. This technique was also expected to provide quantitative information on bacterial proteolytic activities. Studies (summarized by Williams and Coleman, 1992) have shown that ciliates do not play a major role in degradation of soluble proteins in the rumen. Whereas Isotricha spp. ingest both insoluble and soluble protein, the entodiniomorphid protozoa, which account for the majority of ruminal protozoal populations, digest only insoluble protein. The ciliate population of the ruminal inoculum used in this study was almost entirely Entodinium spp. (99.8% of the total protozoal count), with Isotricha spp. detected in some treatments at levels below 5,000 cells/mL. Thus, it was assumed that most (if not all) of the tracer (from solubilized 15N-casein) appearing in the protozoa-rich fraction of the incubation medium would be of bacterial origin. The positive correlations observed between 15N-enrichment of protozoal and bacterial N and total protozoal numbers (r = 0.70 and 0.83, respectively), the negative correlation between TFAA concentration and 15N-incorporating activities of protozoa, and the high and positive correlation between total VFA concentration and tracer incorporation into protozoal and bacterial N attest to the validity of this assumption. The usefulness of labeled substrates has been demonstrated by other researchers, in applications such as testing the antiprotozoal effects of surfactants using labeled [Me-14C]-choline (Campbell et al., 1982), and studying bacterial lysis using [3H]-thymidine (Jarvis, 1968) and [14C]-glucose (Hoogenraad and Hird, 1970). Wallace and McPherson (1987) proposed a method for labeling ruminal bacteria with 14C-amino acids that was used successfully to determine protozoal activities in the rumen and to examine the effects of antiprotozoal compounds (Wallace and Newbold, 1991; Newbold et al., 1997).
In the present study, several medium-chain saturated and long-chain unsaturated FA were found to exert dramatic effects on ruminal fermentation. Capric and lauric acids at all three application levels completely eradicated ruminal protozoa, decreased bacterial incorporation of 15N, and significantly shifted the concentrations of fermentation end products, compared with controls. These FA decreased ammonia and butyrate concentrations and increased lactate, effects known to be associated with decreased fauna in the rumen (Williams and Coleman, 1992). With capric and lauric acids, TFAA and soluble protein concentrations were increased, and bacterial proteolytic activity (as measured) was inhibited. Polysaccharide-degrading enzyme activities in the incubation medium were unaffected or increased by these two FA at inclusion levels Lo or Med; only at level Hi (0.25% for capric acid, 1.0% for lauric acid) were xylanase and amylase activities decreased. Observed decreases (14 to 16%) in deaminative activity were numerical only, but it seems that the C10:0 and C12:0 FA inhibited proteolysis and deamination of amino acids. This hypothesis is supported by observation of capric acid- and lauric acid-mediated decreases in BCFA concentrations.
Some BCFA arise in the rumen from the breakdown of branched-chain amino acids (Wallace, 1994). Decreased BCFA concentrations may be indicative of inhibited amino acid catabolism, but could also result from inhibition of BCFA utilization by bacteria; the positive correlation between protozoal and bacterial 15N incorporation and BCFA concentration in the present study is supportive of the first of these options. The increased RS and decreased total VFA concentrations in association with capric and lauric acids also are consistent with decreased carbohydrate assimilation and metabolism by ruminal bacteria. In contrast to our observation that acetate concentration was significantly decreased by capric acid, with no effect on propionate concentration, Ajisaka et al. (2002) reported no effect on total VFA and increased molar proportion of propionate with addition of C8:0 and C10:0 FA. Caprylic acid, another saturated medium-chain FA, also was effective in eliminating ruminal protozoa at inclusion level Hi (0.25% for Group 3). At that concentration, caprylic acid exerted effects similar to those of capric and lauric acids, decreasing ammonia and increasing TFAA and RS concentrations. Both lower levels of caprylic acid decreased the proportion of protozoal protein originating from bacterial protein (Pzoa/Bact), which is an indication of ingestion of bacteria by protozoa. Assessment of caprylic acid (Hi) on Pzoa/Bact was precluded by complete elimination of protozoa by the FA. Myristic acid (C14:0) did not affect ammonia concentrations except at the highest level (0.5%); it increased TFAA and SP concentrations and exerted inhibitory effects on protozoal numbers and activity similar to those of the C8 to C12 saturated FA. Others also have reported a strong in vivo inhibitory effect of C8, C10, C12, and C14 fatty acids on ruminal protozoal in goats (Matsumoto et al. 1991). In that study, protozoa were completely eliminated from the rumen after 2 d of feeding C10 and C12, and after 3 d of feeding C14. Similarly, Ajisaka et al. (2002) reported complete eradication of protozoa in vitro by capric (0.17 to 0.67 mg/mL) and lauric acid (0.33 to 0.67 mg/mL) cyclodextrin complexes. Capric acid (C10:0) is found in small amounts in coconut and palm kernel oils (accounting for 8.4 and 7% of the total FA, respectively), but at 45 to 47% of total FA, lauric acid (C12:0) is a major constituent of these oils, and myristic acid (C14:0) also accounts for a large portion (14 to 18%) of total FA (CRC, 1988). Machmüller and Kreuzer (1999) found a strong inhibitory effect of coconut oil (included at 3.5 and 7.0% of dietary DM) on ruminal ciliates, which was partially responsible for a substantial decrease in methane production. These results clearly show that the effect of short-chain FA on protozoa observed in our in vitro study are similar to those observed when these fatty acids are fed to ruminants.
A severe decrease in protozoal numbers was observed with all three of the unsaturated C18 acids studied, particularly by linolenic (C18:3) and linoleic (C18:2) acids. These acids were not inhibitory to bacterial proteolytic activity, but they substantially decreased the incorporation of 15N into protozoal protein and, consequently, the proportion of protozoal protein originating from bacterial N. Compared with the medium-chain saturated FA, the effects of the C18 unsaturated acids on ruminal fermentation were less pronounced. Increases in TFAA and RS concentrations were of a lower magnitude and total VFA production was not affected, although the acetate:propionate ratio was decreased. As with the C8 to C14 acids, butyrate and valerate concentrations were also decreased by the C18 polyunsaturated FA. Oleic, linoleic, and linolenic acids all enhanced the polysaccharide-degrading (particularly amylase) activities of the incubation medium, suggesting a larger bacterial population compared with the control. These FA also increased lactate concentrations. The significant negative correlation between total protozoal counts and lactate concentration found in the present study supports the proposed role of certain ciliates (i.e., Entodinium spp.) in lactate utilization in the rumen (Newbold et al., 1987). In vivo, however, decreasing the ruminal protozoal population by 42% did not affect the concentration of L-lactate in the rumen (Hristov et al., 2001).
Henderson (1973) reported inhibition of the growth of several ruminal bacteria by C14:0, C16:0, and C18:0 FA. At low concentrations (from 0.01 to 0.1 g/L), C10:0, C12:0, and C18:1 FA had a stimulatory effect, but at higher concentrations also decreased bacterial growth. In contrast, Maczulak et al. (1981) found no effect of C16:0 or C18:0 FA, but C18:1 dramatically inhibited the growth of certain cellulolytic strains. Sutton et al. (1983) reported that coconut oil (high in C12 to C14 saturated FA) possessed stronger antiprotozoal properties than did linseed oil (high in C18 unsaturated FA), and that both oils produced changes in VFA proportions that are typically associated with decreased protozoal numbers in the rumen. In contrast, Newbold and Chamberlain (1988) found a stronger antiprotozoal effect of C18 unsaturated acids (supplied as linseed oil) than was exerted by saturated C12 to C14 acids (from coconut oil). Pantoja et al. (1994) observed that the efficiency of microbial protein synthesis in the rumen increased linearly with the degree of unsaturation in dietary fat for dairy cows. As well, ruminal protozoal populations were shown to decrease linearly with increasing the degree of unsaturation of dietary fats (Oldick and Firkins, 2000). Rapeseed oil, which is high in unsaturated FA, effectively decreased ruminal ammonia and butyrate concentrations and also increased the efficiency of microbial protein synthesis in the rumen, although protozoal numbers were not significantly decreased (Tesfa, 1993). Similar effects of rapeseed oil were reported by Doreau et al. (1991).
The efficiency of utilization of feed N by ruminant animals is low. In the dairy cow, for example, the efficiency of utilization of dietary N for milk protein synthesis has been calculated at 19 to 20% (Tamminga, 1992; MacRae et al., 1995). This inefficiency is due in large part to the wasteful process of N recycling occurring in the rumen. Tamminga (1992) estimated that up to 15% of the dietary N is lost to the dairy cow owing to inefficient N metabolism in the rumen, wasted primarily as ammonia. Ruminal ammonia concentrations can vary greatly (Hristov, 2000); controlling this variability could be an important factor in improving the efficiency of feed N utilization in ruminants and also for reducing excretion of N into the environment. The results from the present experiment, although limited in their direct applicability to in vivo conditions, imply that ammonia concentration is related mostly to total protozoal numbers and to protozoal and bacterial activities in the rumen. Medium-chain saturated FA not only decreased or eradicated ciliate populations, but also inhibited (to different degrees, depending on chain length) bacterial growth, proteolysis, and deamination. Thus, the effects of these FA on ruminal ammonia concentration seem to result both from inhibition of protozoal growth and from inhibition of bacterial activities. Dohme et al. (1999) investigated the effects of coconut oil (rich in medium-chain saturated FA) on ruminal fermentation in faunated and defaunated in vitro systems (RUSITEC). In the faunated system, ammonia concentration was decreased by 43% with coconut oil treatment (compared with protected coconut oil); in the defaunated system, ammonia concentration was 22% of that measured in the faunated, protected coconut oil treatment. Furthermore, bacterial counts were unaffected by type of oil in the faunated system, whereas they were decreased by 30% by the unprotected oil in the defaunated system. Those findings, together with results from the present study, suggest that medium-chain saturated FA have potential for manipulating ruminal proteolysis and ammonia concentrations in the rumen. In reviewing a large set of published data, however, Doreau and Ferlay (1995) concluded that supplemental FA do not affect N metabolism in the rumen (including ammonia concentrations). This discrepancy may be due to the fact that dietary FA are preferentially adsorbed onto the feed particles (Harfoot et al., 1974) and consequently their effect on ruminal protozoa (and bacteria) may be diminished in vivo compared with in vitro. The possibility of using exogenous FA to manipulate ammonia utilization in the rumen must be approached cautiously so that fiber degradation and microbial protein synthesis in the rumen are not impaired (Brooks et al., 1954; Doreau and Ferlay, 1995; Dohme et al., 1999). The other group of treatments in the present study that were found to strongly inhibit protozoal numbers and activity (i.e., the C18 unsaturated FA) were less effective in decreasing ammonia concentration and did not inhibit proteolysis.
|
Implications |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
|
Footnotes |
|---|
2 This study was supported by funds from the Canada Alberta Beef Industry Development Fund. The authors thank the LRC barn staff for their conscientious care of the experimental animals and gratefully acknowledge the technical assistance of L. Neill, C. Barkley, W. Smart, C. Cockwill, and Z. Xu.
3 Correspondence: P.O. Box 3000 (phone: 403-317-2240; fax: 403-382-3156; e-mail: mcallister{at}agr.gc.ca).
Received for publication August 9, 2003. Accepted for publication May 18, 2004.
|
Literature Cited |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. W. Hess, G. E. Moss, and D. C. Rule A decade of developments in the area of fat supplementation research with beef cattle and sheep J Anim Sci, April 1, 2008; 86(14_suppl): E188 - E204. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. N. Hristov, L. R. Kennington, M. A. McGuire, and C. W. Hunt Effect of diets containing linoleic acid- or oleic acid-rich oils on ruminal fermentation and nutrient digestibility, and performance and fatty acid composition of adipose and muscle tissues of finishing cattle J Anim Sci, June 1, 2005; 83(6): 1312 - 1321. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
