| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ANIMAL NUTRITION |
* Department of Animal Sciences, University of Illinois, Urbana 61801 and
and
Nestle Purina Petcare Research, St. Joseph, MO 65403
Abstract
A bacterial marker can be used to determine the portion of fecal N that is of bacterial origin, as well as the effect of dietary factors on the bacterial N in feces of the dog. Two experiments were conducted to determine the efficacy of diaminopimelic acid (DAPA) and purines as bacterial markers in dogs. In Exp. 1, five adult female dogs were fed the same commercial diet. In Exp. 2, 50 dogs were fed one of four test diets (as-fed basis): a prebiotic-free control or diets containing either 1% chicory, 1% mannanoligosaccharide (MOS), or 1% chicory plus 1% MOS. Fresh feces were collected in both experiments and used to isolate a bacteria-rich sample (BRS) by differential centrifugation. In Exp. 1, the BRS had a N:purine ratio of 0.66 and N:DAPA ratio of 18.9. The CV for the N:purine ratio (20.7%) was much higher than that for the N:DAPA ratio (6.1%), indicating that DAPA resulted in a less variable estimate of fecal bacterial N. Using either marker, approximately 50% of the fecal N was estimated to be of bacterial origin. In Exp. 2, the N:DAPA ratio of the BRS did not differ (P = 0.14) among treatments. The BRS from dogs fed prebiotic-containing diets had treatment averages for N:DAPA ratios ranging from 16.9 to 18.5, whereas BRS from dogs fed the control diet had a ratio of 15.9. Averaged across all dogs, approximately 46% of fecal N was of bacterial origin. When calculating fecal bacterial concentrations using the average N:DAPA ratio for all dogs, little difference existed in the estimation compared with using individual values. The value resulting from use of the average ratio was approximately 13% higher than when using the individual ratios for dogs fed the control diet, which was due to the lower N:DAPA ratio for dogs fed the control diet compared with dogs fed the other treatments. Based on the consistency of the N:DAPA ratio of the BRS, DAPA seems to be a suitable marker for estimation of bacterial N in the feces of dogs.
Key Words: Bacteria • Diaminopimelic Acid • Dog • Nitrogen
Introduction
In humans, the microbial content of feces was found to be 55% of DM when using gravimetric techniques (Stephen and Cummings, 1980
). Based on these data, it is estimated that approximately 50% of the DM of dog feces is of bacterial origin. However, we are unaware of any published data available based on studies conducted with dogs. In order to determine the microbial N content of dog feces, a marker found only in bacteria must be identified. Two microbial markers used in other species are purines and diaminopimelic acid (DAPA). Purines, found in microbial RNA, are commonly used to estimate ruminal microbial protein synthesis (Broderick and Merchen, 1992). A key concern with the use of purines as a bacterial marker in dogs is that animal proteins in many dog food formulations are high in purines. Diaminopimelic acid is an AA used almost exclusively in bacterial cell walls (Dugan et al., 1992), and it has been used as a microbial marker in ruminants and swine (Broderick and Merchen, 1992; Sauer et al., 1991). When pigs were fed varying amounts of fiber, 72 to 86% of fecal N was estimated to be of bacterial origin using DAPA (Sauer et al., 1991).
The microbial population in the large intestine of nonruminants can have significant effects on the AA and N contents of feces (Low, 1980). If a suitable method to quantify the proportion of microbes in the feces of dogs is determined, a better evaluation of the protein and AA digestibilities of foodstuffs in the dog can be made. The microbial marker can also be used to determine the effect of dietary ingredients on the microbial N content found in the feces of the dog. This study was conducted to determine whether 1) DAPA or purines can be used as microbial markers to quantify the bacterial N content of canine feces; 2) one of the markers is more appropriate for use in canine studies; and 3) a constant value for N:DAPA of bacteria-rich samples isolated from dog feces can be established.
Materials and Methods
Experiment 1
Animals.
Five adult purpose-bred female dogs with hound bloodlines were used. Dogs were housed individually in 1.2 x 3.1-m, solid-floor pens in a temperature-controlled room (21°C) at the animal care facility in the Edward R. Madigan Laboratory, University of Illinois (Urbana). A 16-h light:8-h dark schedule was used. All dogs were allowed free access to water. The University of Illinois Institutional Animal Care and Use Committee approved the use of animals before the start of the experiment.
Experimental Design.
All dogs were fed the same commercial diet, the main ingredients of which were cornmeal, chicken, ground whole-grain sorghum, and chicken by-product meal. The diet contained approximately 20% CP and 10% crude fat (as-fed basis). All dogs had been fed this diet for a minimum of 1 mo before fecal collection. Fresh feces were collected from the pen floor during a 5-d collection phase. Feces were weighed at the time of collection, frozen at –20°C, and composited by dog at the end of each period.
Fecal samples were divided into two aliquots. One aliquot was freeze-dried, after which samples were ground to pass a 2-mm screen in a Wiley mill (Thomas-Wiley, Swedesboro, NJ) in preparation for chemical analyses. The second aliquot was used to isolate a bacteria-rich sample (BRS). This was done by initially blending the sample in a volume of 0.9% saline equal to three times the sample weight. The BRS then was isolated via differential centrifugation according to Firkins et al. (1984). The BRS was freeze-dried and ground to pass a 2-mm screen.
In addition, five commercial diets were obtained and analyzed for purine and DAPA content. These diets included one canned diet and four dry diets ranging in CP and crude fat concentrations from 18 to 22% and 8 to 12% (as-is basis), respectively. The DM content of the canned diet was 18.6%, whereas the DM of the dry diets ranged from 91.5 to 92.4%.
Chemical Analyses.
Bacteria-rich and fecal samples were analyzed for DM, OM, and ash concentrations according to AOAC (1995). Nitrogen was determined according to AOAC (1995) using a Leco nitrogen/protein determinator (model FP-2000, Leco Corp., St. Joseph, MI). Bacteria-rich and fecal samples were prepared for AA analysis by acid hydrolysis (Spitz, 1973). Amino acid concentrations were then determined using ion-exchange chromatography (Spackman et al., 1958) on a GoldDV711 chromatograph (Beckman Instruments, Inc., Fullerton, CA). Samples were analyzed for purine content using the method of Zinn and Owens (1986).
Calculations.
The proportion of bacterial N expressed as a percentage of fecal N was calculated using the following equation:
 |
The proportion of bacterial N was calculated using both purines and DAPA as microbial markers. The proportion of bacterial N also was calculated using either the N:marker ratio of the BRS of the individual dogs or the average N:marker ratio for all of the BRS. The two different N:marker ratios were used to determine the difference in the variation associated with either marker when a constant value was used for the marker ratio in the BRS.
Experiment 2
Animals.
Fifty dogs were used in this experiment. The dogs were separated into two blocks. Each block consisted of nine beagles, which were at least 10 yr old, and 16 pointers, which were at least 6 yr old. Dogs were housed in individual indoor/outdoor kennels, which were 1.2 x 1.5 m inside and 1.2 x 3 m outside, at Kennelwood, Inc. (Champaign, IL). All dogs had free access to water.
Experimental Design.
Dogs were fed one of four test diets. Twelve dogs were fed each treatment diet, and 14 dogs were fed the control diet. The ingredient and chemical composition of the diets is presented in Table 1. Diets included a control diet, and diets containing 1% chicory, 1% mannanoligosaccharide (MOS), or 1% chicory + 1% MOS. Both chicory and MOS may alter the concentrations of specific bacterial populations in the large intestine; therefore, the addition of these ingredients to the diet may result in a change in the proportion of bacterial concentrations in feces. All diets were analyzed for DM and OM concentrations according to AOAC (1995). Crude protein was determined according to AOAC (1995) using a Leco nitrogen/protein determinator (Leco Corp.). Fat content was determined by acid hydrolysis (AACC, 1983), followed by ether extraction according to Budde (1952). The diets contained approximately 91% DM, 24% CP and 12% acid-hydrolyzed fat.
|
Calculations.
The proportion of bacterial N expressed as a percentage of fecal N was calculated using the same equation as described in Exp. 1. The following values for the N:DAPA ratio of the BRS were used in the equation: 1) the ratio for the individual dog, 2) the average ratio for dogs in each treatment group, and 3) the average ratio for all dogs used in the experiment. The three different ratios assessed the amount of variation associated with using the different ratios, and ascertained that the treatment differences and interrelationships were not altered by the different calculation methods. These different ratios were used to further examine the variation associated with the N:DAPA ratio in BRS and to determine if the estimated proportion of bacterial N changed when a standard N:DAPA ratio of the BRS was used.
Statistics.
Data were analyzed as a 2 x 2 factorial arrangement of treatments in a randomized complete block design by the MIXED models procedure of SAS (SAS Inst., Inc., Cary, NC). The statistical model included the fixed effects of chicory, MOS, and the chicory x MOS interaction, and the random effects of block and breed. Means are reported as least squares means to account for any missing observations.
Results and Discussion
Experiment 1
The composition of the bacteria-rich and fecal samples is presented in Table 2. The composition is presented on an OM basis in order to account for differences in the ash content of the bacterial samples caused by the presence of residual ash from the saline used during bacterial isolation. In the case of BRS isolated from dog feces, the initial sample had to be blended with saline in order to provide the proper sample consistency for isolation. The N, purine, and DAPA concentrations of the BRS were 8.51, 13.4, and 0.45%, respectively, resulting in respective N:purine and N:DAPA ratios of 0.66 and 18.9. The N:purine ratio is somewhat lower than that found in BRS isolated from ruminal fluid. Cecava et al. (1990) reported N:purine ratios of 0.77, 0.65, and 0.74 for mixed bacterial, fluid-associated bacterial, and particle-associated bacterial isolates, respectively, in the ruminal fluid of steers fed diets containing alfalfa hay, corn silage, and ground corn. Titgemeyer et al. (1989) found a N:purine ratio of approximately 1.1, which is higher than that found for dog feces. The SEM of the N:purine ratio (0.06) in the BRS was less than that of the N:DAPA ratio (0.52); however, the CV was much higher for the N:purine ratio (20.7%) than for the N:DAPA ratio (6.1%). This indicates that DAPA is a less variable marker than purines for use in the BRS of dogs. It is important to find a marker with the least variation in the BRS so that a reliable value can be established.
|
). This indicates that either marker estimates a similar amount of bacterial N. When comparing the estimates of bacterial N resulting from using the individual N:purine ratio with the average N:purine ratio, the SEM decreased from 5.09 to 2.25 when using the average value. This indicates that variation in the individual N:purine ratio results in an increase in the variation of the bacterial N estimation. However, when comparing the bacterial N estimations using the N:DAPA ratios, there is little difference in the SEM when using either the individual or average ratios, indicating that the N:DAPA ratio for the BRS was less variable than was the case for the N:purine ratio. The purine content of five commercial diets averaged 1.5% of OM, with a standard deviation of 0.39. This is approximately 10% of the value of the purine concentration found in the BRS. Dietary purines may not be completely digested and may confound the concentration of purines measured in feces. The DAPA content of these diets was analyzed to be 0.1% of the diet.
Based on data from this experiment, and given the significant purine concentrations of canine diets, DAPA was considered the more appropriate marker for use with dogs and was therefore used in Exp. 2.
Experiment 2
Composition of the bacteria-rich and fecal samples is presented in Table 3. For the BRS, there was little variation in N (range 8.5 to 8.8%) content among treatments. There was some variation in DAPA (range 0.48 to 0.60%) content. These values are similar to the values in Exp. 1. The N:DAPA of the BRS calculated for each dog was not different (P = 0.14) among treatments. The dogs fed the prebiotic-containing diets had N:DAPA ratios ranging from 16.9 to 18.5. These values correspond to the ratio calculated in Exp. 1. Bacteria-rich samples from dogs fed the control diet had a numerically, but not significantly, lower mean value for N:DAPA (15.3). This is due to a slightly higher DAPA concentration in the BRS than for the other treatment groups. The variation in the N:DAPA ratio found for dogs fed the different treatments could be due to variation among individual dogs. There always will be some variation among animals in large intestinal bacterial composition. However, the N:DAPA ratio of the BRS from this study is fairly consistent. This finding indicates that DAPA is an appropriate marker for use with dogs.
|
Bacterial N concentrations calculated using different N:DAPA ratios are presented in Table 4. Differences in the number of observations for each treatment are due to the fact that we were unable to obtain an appropriate fecal sample for bacterial isolation from some dogs. A minimum sample of 500 g was required for both bacterial isolation and analysis of a whole fecal sample. Because a fresh fecal sample was required for bacterial isolation, it was not possible to collect a large enough sample from some dogs. No differences in bacterial N concentrations were observed among treatments (P = 0.32). This implies that the addition of chicory and/or MOS had no effect on total fecal bacterial concentrations. Dogs fed diets containing either chicory or MOS had significantly higher fecal concentrations of bifidobacteria than the control (10.5 vs. 10.1 log10 cfu/g of dry feces; unpublished). Dogs fed the diet containing MOS also had lower fecal concentrations of Escherichia coli than those fed the control diet. Because the addition of MOS resulted in the increase in one species and the decrease of another species of bacteria, no difference in total bacterial populations as measured by the use of DAPA would be expected. Also, the shifts in specific bacterial genera and species may be too small to result in a change in the total bacterial N proportion that can be measured through the use of DAPA.
|
Bacteria-rich samples isolated from dog feces in this study were compared with samples isolated from ruminal fluid in Table 5. Titgemeyer et al. (1989) used similar isolation methods as were used for dog feces in this study, except for the initial mixing with saline that was used for the dog fecal samples. There is a large difference in the OM content of the samples. This could be due to differences in the amount of saline used in isolation of the BRS.
|
Implications
A consistent bacteria-rich sample similar to that isolated from ruminal fluid was isolated from dog feces. Diaminopimelic acid seems to be a valuable marker that can be used to estimate bacterial nitrogen expressed as a percentage of total fecal nitrogen of dogs. Although differences exist in the diaminopimelic acid concentration of bacterial species, a nitrogen:diaminopimelic acid ratio of 18 could be used in future studies, allowing for a simple means of estimating differences in the bacterial nitrogen proportion of total fecal nitrogen. Future studies should test the use of diaminopimelic acid to estimate fecal bacterial nitrogen with diets that induce large changes in colonic bacterial populations.
1 Correspondence: 132 Animal Sciences Laboratory, 1207 W. Gregory Dr. (phone: 217-333-2361; fax: 217-244-3169; e-mail: gcfahey{at}uiuc.edu).
Received for publication September 24, 2003. Accepted for publication February 18, 2004.
Literature Cited
AACC. 1983. Approved Methods. 8th ed. Am. Assoc. Cereal Chem., St. Paul, MN.
AOAC. 1995. Official Methods of Analysis. 15th ed. Assoc. Offic. Anal. Chem., Arlington, VA.
Budde, E. F. 1952. The determination of fat in baked biscuit type of dog foods. J. Assoc. Offic. Agric. Chem. 35:799–805.
Broderick, G. A., and N. R. Merchen. 1992. Markers for quantifying microbial protein synthesis in the rumen. J. Dairy Sci. 75:2618–2632.[Abstract]
Cecava, M. J., N. R. Merchen, L. C. Gay, and L. L. Berger. 1990. Composition of ruminal bacteria harvested from steers as influenced by dietary energy level, feeding frequency, and isolation techniques. J. Dairy Sci. 73:2480–2488.[Abstract]
Dugan, M. E. R., W. C. Sauer, K. A. Lien, and T. W. Fenton. 1992. Ion-pair high-performance liquid chromatography of diaminopimelic acid in hydrolysates of physiological samples. J. Chromatogr. 582:242–245.[Medline]
Firkins, J. L., L. L. Berger, G. C. Fahey, Jr., and N. R. Merchen. 1984. Ruminal nitrogen degradability and escape of wet and dry distillers grains and wet and dry corn gluten feeds. J. Dairy Sci. 67:1936–1944.
Low, A. G. 1980. Nutrient absorption in pigs. J. Sci. Food Agric. 31:1087–1130.[Medline]
Sauer, W. C., R. Mosenthin, F. Ahrens, and L. A. den Hartog. 1991. The effect of source of fiber on ileal and fecal amino acid digestibility and bacterial nitrogen excretion in growing pigs. J. Anim. Sci. 69:4070–4077.[Abstract]
Spackman, K. H., W. H. Stein, and S. Moore. 1958. Automatic recording apparatus for use in the chromatography of amino acids. Anal. Chem. 30:1190–1206.
Spitz, H. D. 1973. A new approach for sample preparation of protein hydrolysates for amino acid analysis. Anal. Biochem. 56:66–73.[Medline]
Stephen, A. M., and J. H. Cummings. 1980. The microbial contribution to human faecal mass. J. Med. Microbiol. 13:45–56.[Abstract]
Titgemeyer, E. C., N. R. Merchen, and L. L. Berger. 1989. Evaluation of soybean meal, corn gluten meal, blood meal, and fish meal as sources of nitrogen and amino acids disappearing from the small intestine of steers. J. Anim. Sci. 67:262–275.
Zinn, R. A., and F. N. Owens. 1986. Rapid procedure for quantifying nucleic acid content of digesta. Can. J. Anim. Sci. 66:157–166.
This article has been cited by other articles:
|
|
 |
|
  T. F. Gressley and L. E. Armentano Effects of Low Rumen-Degradable Protein or Abomasal Fructan Infusion on Diet Digestibility and Urinary Nitrogen Excretion in Lactating Dairy Cows J Dairy Sci, March 1, 2007; 90(3): 1340 - 1353. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. K. Spears, L. K. Karr-Lilienthal, C. M. Grieshop, E. A. Flickinger, B. W. Wolf, and G. C. Fahey Jr Pullulans and {gamma}-Cyclodextrin Affect Apparent Digestibility and Metabolism in Healthy Adult Ileal Cannulated Dogs J. Nutr., August 1, 2005; 135(8): 1946 - 1952. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
