Fecal Odor Components in Dogs: Nondigestible Oligosaccharides and Resistant Starch Do Not Decrease Fecal H2S Emission

Fecal Odor Components in Dogs: Nondigestible Oligosaccharides and Resistant Starch Do Not Decrease Fecal H₂S Emission

M. Hesta, DVM, CCVCM Diplomate

G. P. J. Janssens, Prof, PhD, Vet., Sci., Agr., Eng.

J. Debraekeleer, DVM, ECVCN Diplomate

S. Millet, DVM

R. De Wilde, Prof, DVM

Laboratory of Animal Nutrition, Faculty of Veterinary Medicine, Ghent University, Merelbeke, Belgium

KEY WORDS: Dogs, odor, prebiotics

Abstract

Bad fecal odor can be inconvenient for dog owners, but can also indicate decreased gastrointestinal health in the animal. The main fecal odor substances generated by bacterial degradation of endogenous and undigested protein are ammonia, aliphatic amines, branched chain fatty acids, indoles, phenols, and volatile sulfur-containing compounds. Dietary supplementation with a nondigestible trisaccharide, lactosucrose, decreased fecal concentration of certain odor components such as ammonia, indol, ethylphenol, phenol, skatole, and butyric acid in dogs.The goal of the present study was to measure the emission of the most prevalent fecal odor components and to evaluate the effect of the addition of fructo-oligosaccharides (FOS), isomalto-oligosaccharides (IMO), and resistant starch (RS) to a commercial diet on some fecal odor parameters (ammonia and H₂S) in dogs.

Introduction

By oxidative deamination of various amino acids or by nonoxidative lyase reactions and the Stickland reaction (coupled oxidation–reduction between pairs of amino acids), ammonia can be produced.3 Ammonia can also be generated through hydrolysis of urea by urolytic bacteria. In fact, urinary urea is a major source of NH₃ in animal manure (mixture of feces and urine) and wastes.4

Reduction of sulfate and metabolism of sulfur-containing amino acids by anaerobic fermentation produce odoriferous sulfur compounds. In the assimilatory pathway, bacteria produce reduced sulfur for cell biosynthesis, whereas in the dissimilatory pathway, sulfate is used as a terminal electron acceptor and, as a consequence, toxic sulfide is generated.1 Desulfovibrio is the major sulfate reducer in the gut. Sulfate can be provided by the diet (sulfate or sulfur-containing amino acids) or by endogenous sulfated glucoproteins like mucins.4

Pathogenic and less desirable bacteria are responsible for the production of fecal odor components.1 On the other hand, the growth or activity of one or a limited number of desirable bacteria in the colon can be selectively stimulated by prebiotics, which are nondigestible food or feed ingredients.5 In several experiments, fecal concentrations rather than fecal emissions of ammonia and/or other odorous components were measured in dogs.1,2,6–8

The aim of the present study was to measure the emission of the most prevalent fecal odor components and to evaluate the effect of the addition of fructo-oligosaccharides (FOS), isomalto-oligosaccharides (IMO), and resistant starch (RS) to a commercial diet on some fecal odor parameters (ammonia and H₂S) in dogs.

Material and methods

Eight adult beagle dogs (5 neutered males and 3 intact females, 4 to 8 years old) were used to test the addition of 3% FOS, 3% IMO, and 3% RS (Cerestar, Vilvoorde, Belgium) to a commercial diet for dogs on palatability, water intake, urinary and fecal production, fecal consistency, and fecal odor in 2 simultaneous Latin square designs (4 x 4). The basal extruded diet (Royal Canin RCCI Size: Medium Adult 1, Brussels, Belgium) (see Table 1), with no added nondigestible oligosaccharides, was fed to meet the energy requirements (415 kJ/kg BW0.75) based on mean body weight. Sex was not taken into account because body weights were similar. The diet was ground and the supplements were mixed homogeneously in the basal diet. The supplements (FOS, IMO, and RS) were added at the expense of the basal diet. After mixing the supplements and the basal diet, 50% lukewarm water was added to produce a palatable porridge. Between the 4 collection periods, a washout period of 9 days was inserted to prevent carryover effects. During the 7-day adaptation and 5-day collection periods, the dogs were housed in individual digestibility cages. During the collection period, feed and water intake as well as urine and fecal production were recorded daily. A daily score for fecal consistency (1: watery diarrhea; 2: cowpat-like feces; 3: normal; 4: dry, firm) was given. The feces were stored at –18°C. Moisture content of the fecal samples was measured. During the last 2 days of the collection period, fresh fecal samples were collected and immediately stored at –18°C. The dogs were checked for defecation every hour from 8 am to 4 pm. The chance that these samples were contaminated with urine was nearly nonexistent. After thawing and 10% dilution with distilled water, pH was measured using a glass electrode. The defrosted feces were stored in air-closed Erlenmeyers. The feces were incubated during 48 hours at 38°C. An anaerobic milieu was produced by flushing with N₂. H₂S and NH₃ concentrations were measured in the headspace by color reaction, normalized for fecal amount (g). For every group (control, FOS, IMO, and RS), one pooled sample was made. An air sample of 1000 mL was taken from each Erlenmeyer, sent over an absorption tube (Tenax polymer), and analyzed by GC-MS (flame ionization detector [FID]). These results were also normalized for fecal amount (g). Because there was only one pooled sample for each treatment, statistical evaluation was not possible, although the aim was more to get a qualitative idea of the emitted odor components and to increase variability in odor components by the different dietary supplements. For the other determinations, variance analyses were performed using SPSS 10.0 (SPSS, Chicago, IL). The criterion for significance was set at P <0.05.

Results

Palatability was excellent. Water intake, urine and fecal production were not significantly different between treatments (Table 2). The pH of the fresh fecal samples was not different between the 4 treatment groups. Fecal consistency was considered normal in all supplemented groups with a tendency for slightly higher fecal moisture content in the FOS group compared with the control and RS groups (P = 0.065). Large individual variation inhibited significancy of differences in H₂S emission between the 4 groups (Table 3). The ammonia emission was below detection limit (0.2 mg/g) in all fecal samples.

The odor components (GC-MS) of the pooled samples are presented in Table 4. Two of the 5 most abundant components were noticed in all 4 samples, namely 2-butanone and n-propanol. Ethanol and ethylacetate were detected in relatively high concentrations in all samples except for ethanol in RS and FOS. Other components in relatively high concentration were 2-butanol (FOS, RS, IMO), n-butanol (FOS), ethylbutyrate (RS), dimethyldisulfide (DMDS; control) and dimethylsulfide (FOS). The odor detection threshold gives an idea of the sensitivity of humans for odor components. Components with the lowest odor detection thresholds are S-components9; DMDS, dimethylsulfide, and dimethyltrisulfide especially have a very low odor detection threshold. Carbon disulfide was detected by MS but the concentration could not be determined because of a high response factor (concentration/peak area) and, as a consequence, a smaller peak area for the same concentration compared with other substances with lower response factors. Three other S-holding components were butylmethylsulfide, methylpropyldisulfide, and dimethyltrisulfide.

Discussion

In this study, fecal ammonia emission was below detection limit in dogs and did not change after the addition of FOS, IMO, or RS. Previously published data in dogs with NDO did not show an effect on fecal ammonia.1,7,8 Adding 0.5% FOS, MOS (mannanoligosaccharides), or XOS (xylooligosaccharides) did not decrease fecal ammonia in dogs.7 Fecal ammonia concentrations also did not decrease after low inclusion rates of FOS (0.25 and 0.5%) in dogs.1 Adding up to 50% meat and bone meal or Greaves meal to a commercial dog diet increased the ammonia concentration in freshly collected fecal samples, but adding FOS or IMO to these protein-rich diets did not decrease fecal ammonia concentration.8 On the other hand, fecal concentration of odor components decreased after dietary supplementation with lactosucrose (a nondigestible trisaccharide). Lactosucrose decreased ammonia, indol, ethylphenol, phenol, skatole, and butyric acid in dogs.2 The level of Bifidobacterium increased and the level of Clostridia decreased, suggesting a relation between the composition and metabolic activity of the intestinal flora and the production of fecal putrefactive substances.

Fecal concentrations give an idea on the odorous potential but not on the odor production because the odor components are volatile and their emission depends on several environmental components, eg, temperature, pH, moisture of the feces, and atmospheric humidity.9

Dietary factors that might increase the intensity of fecal odor are high dry matter intake and consequently faster rates of digesta passage, large quantities of ingested protein, and less digestible protein.1 These factors result in more undigested protein in the large intestine. Microorganisms from the fecal flora can hydrolyze urea from urine and generate higher ammonia emission. Fecal samples were collected immediately after defecation, minimizing the risk of contamination with urine. Why ammonia emission was below detection limit is not known. In a similar experiment in cats, ammonia emissions were almost always (12 of 15 samples) above detection limit (unpublished data).

Six of 10 compounds with the lowest odor detection thresholds contain sulfur. Hydrogen sulfide (H₂S) is one of them, with a low odor detection threshold of 0.1 µg/m3.9

In our study, H₂S emission did not change after supplementation of FOS, IMO, or RS in dogs. H₂S concentration was highest after Lactobacillus supplementation in dogs in one experiment but not in a second experiment.10 Methanethiol and dimethyl sulfide concentrations tended to be highest when Lactobacillus alone was supplemented to a base diet compared with the basal diet, and to the basal diet supplemented with FOS or a combination of FOS and Lactobacillus. The results were not consistent in these 2 experiments.

The individual variation of the H₂S emission was very high in our study. In an in vitro fermentation test with feces from dogs, similar concentrations of sulfide were noted.11

Because a complete GC-MS analysis was done, relative concentrations can be calculated (percentage of total emission) for the 5 most abundant components. They represented 53.9%, 61.9%, 64.5%, and 74.5% of total emission in FOS, C, RS, and IMO, respectively. Only 2 of these 5 components had a low odor detection threshold, namely dimethylsulfide and dimethyldisulfide. If odor appreciation and interactions between odor components are not taken into account, the importance of individual components can be calculated by dividing the concentration of an odor component by the odor threshold. By this method, the highest numbers, and consequently, the most important odor components, were dimethylsulfide, dimethyldisulfide, and dimethyltrisulfide, as expected. Determining the 5 most abundant components only is probably not the best way to get an idea of fecal odor production, and is not sufficient to determine only components with low odor threshold because components with moderate odor thresholds can be emitted in rather high concentrations also.

In contrast with an earlier study12 in which dimethyltrisulfide was evident by odor port-GC but was detected in only one sample by GC-MS, dimethyltrisulfide was detected in 3 of the 4 samples in our experiment. In this earlier study,12 only peaks comprising at least 0.4% of total area were quantified but the total peak area accounted for 96%.

It is difficult to speculate on the correlation between H₂S and the other S-holding odor components. Although significant differences were not noted for H₂S concentrations, the IMO group showed the highest mean and median value but the sum of all S-holding components in the pooled sample was considerable lower compared with the other 3 groups. It can be hypothesized that H₂S is an S-source to form the other S-holding components or vice versa. If this is true, a low H₂S concentration could indicate a high concentration of the other S-holding components, although a higher H₂S concentration could theoretically give rise to production of more S-holding components. A high H₂S emission could also indicate an active H₂S-metabolizing flora as a consequence production of S-holding components, including H₂S.

The intensity of fecal odor depends on the odor components, their odor threshold, and concentration. Interactions between different odor components are possible and prediction of fecal odor from different emitted odor components is difficult. Therefore, the screening of odor components in this study should be followed by trials to investigate the real contribution of analyzed levels of components to fecal odor, the underlying mechanisms of their metabolism and the impact of nutritional changes, e.g., prebiotic supplementation.

Summary

The aim of the present study was to measure the emission of the main fecal odor components and to evaluate the effect of the addition of fructo-oligosaccharides (FOS), isomalto-oligosaccharides (IMO), and resistant starch (RS) to a commercial diet on ammonia and H₂S emission in dogs.

Eight dogs were supplemented with 0% or 3% FOS, 3% IMO, or 3% RS. Feces were incubated for 48 hours at 38°C (dogs). Flushing with N₂ during fecal incubation simulated an anaerobic environment. Hydrogen disulfide and ammonia concentrations were measured in the headspace. Hydrogen disulfide did not differ significantly between the treatments. The ammonia concentrations in dogs were below detection limit in all fecal samples. For each treatment, a pooled fecal sample was analyzed for other fecal odor components by GC-MS.

Acknowledgments

The authors thank Cerestar for kindly supporting this study financially and Royal Canin for providing the diets.

References

1. Hussein H, Sunvold G: Dietary strategies to decrease dog and cat faecal odour components. In: Reinhart G, Carey D (eds). Recent Advances in Canine and Feline Nutrition. OH: Orange Frazer Press; 2000:153–168.

2. Terada A, Hara H, Kataoka M, Mitsuoka T: Effect of dietary lactosucrose on faecal flora and faecal metabolites of dogs. Microb Ecol Health Dis 5:43–50, 1992.

3. Gottschalk G: Bacterial fermentations. In: Bacterial Metabolism. New York: Springer-Verlag, 1986:269.

4. Mackie R, Stroot P, Varel V: Biochemical identification and biological origin of key odor components in livestock waste. J Anim Sci 76:1331–1342, 1998.

5. Gibson GR, Roberfroid MB: Dietary modulation of the human colonic microbiota: introducing the concept of prebiotics. J Nutr 125:1401–1412, 1995.

6. Martineau B, Laflamme D: Effect of diet on markers of intestinal health in dogs. Res Vet Sci 72:223–227, 2002.

7. Strickling J, Harmon D, Dawson K, Gross K: Evaluation of oligosaccharide addition to dog diets: influences on nutrient digestion and microbial populations. Anim Feed Sci Tech 86:205–219, 2000.

8. Hesta M, Roosen W, Janssens GPJ, Millet S, De Wilde R: Prebiotics affect nutrient digestibility but not faecal ammonia in dogs fed increased dietary protein levels. Br J Nutr. In press, 2003.

9. O’Neill D, Phillips V: A review of the control of odour nuisance from livestock buildings. Part 3, properties of the odorous substances which have been identified in livestock wastes or in the air around them. J Agr Eng Res 53:23–50, 1992.

10. Swanson K, Grieshop C, Flickinger E, et al: Fructooligosaccharides and Lactobacillus acidophilus modify gut microbial populations, total tract nutrient digestibility and fecal protein catabolite concentrations in healthy adult dogs. J Nutr 132:3721–3731, 2002.

11. Giffard C, Collins S, Stoodley N, Butterwic R, Batt R: Administration of charcoal, Yucca schidigera and zinc acetate to reduce malodorous flatulence in dogs. J Am Vet Med Assoc 218:892–896, 2001.

12. Lowe J, Kershaw S, Taylor A, Linforth R: The effect of Yucca schidigera extract on canine and feline faecal volatiles occurring concurrently with faecal aroma amelioration. Res Vet Sci 63:67–71, 1997.

13. Van Gemert L, Nettenbreijer A: Compilation of Odour Threshold Values in Air and Water. Zeist, The Netherlands: Central Institute for Nutrition and Food Research; June 1977 (supplement V, August 1984).

Table 1. Composition of the Basal Diet*†

Moisture 8%

Protein 25%

Methionine 0.5%

Methionine + cysteine 0.85%

Ether extract 12%

Fibre 6.3%

Ash 6.5%

Starch 41.8%

ME 16.4 MJ/kg

Ingredient list: Corn, corn flour, dehydrated poultry meat, beef fat, dehydrated poultry liver, dehydrated beef protein, dehydrated fish, beet pulp, yeast, vegetable oil, fish oil, egg powder, minerals, oligo elements, DL-methionine, vitamins.

*Based on data from the company.

†Percent on as-fed basis.

Table 2. Observations in Dogs

C FOS IMO RS mean SD P n

Feed intake* 1001 1001 1001 1001 1001 0.414 NS 32

Water intake* 1540 1509 1494 1522 1516 181.8 NS 32

Urine production* 1027 975 983 1006 998 226.9 NS 31

Fecal production*† 362.6 389.2 360.3 412.8 381.2 48.13 NS 32

Fecal consistency score‡ 3.01 2.97 3.01 3.06 3.01 0.147 NS 32

Fecal moisture %§ 61.52 63.51 62.08 60.84 61.99 2.42 0.065 32

Fecal pH§ 6.70 6.71 6.82 6.53 6.69 0.338 NS 32

*In grams per 5 days.

†Grams on as-is basis.

‡1: watery diarrhea; 2: cow-pat-like feces; 3: normal; 4: dry, firm.

§Fresh fecal sample.

NS, not significant.

Table 3. Fecal NH₃* and H₂S (µg/g in 1 g feces) Emission in Dogs

Control 3% FOS 3% IMO 3% RS

Mean Median Range Mean Median Range Mean Median Range Mean Median Range P No.

H₂S dogs 1.55 0.79 0.05–4.5 3.57 0.67 0.09–20.5 5.08 1.78 0.69–18.2 1.49 0.98 0.05-5 NS 32

*All NH₃ concentrations in dogs were below detection limit.

Table 4. Main Odor Components of Pooled Freshly Collected Fecal Samples in Dogs (GC-MS) (mg/m3)

Odor Detection

Control FOS IMO RS SEM Threshold*

Aliphatic hydrocarbons

Octane 107 177 35 71,000–710,000

Ketones

Acetone 4049 3444 2427 § 594 940–1,550,000

2-butanone 8287 >16,880 >14,704 >11,269 1893 750–250,000

2-pentanone 362 401 335 19

SUM 12,697 >20,724 >17,466 >11,269 2178

Alcohols

Ethanol >31,337 >2600 >11,637 § 6825 640–1,350,000

n-propanol >22,988 >33,170 >32,632 >34,318 2620

2-butanol 4079 >11,512 8100 20,784 3562 400–80,000

n-butanol 7284 12,624 4941 9362 1630

SUM >65,687 >83,307 >57,311 64,463 5523

S-components

Dimethylsulfide † >23,109 374 7598 4808 0.3–160

Carbondisulfide ‡ ‡ ‡ ‡ 50–100

Butylmethylsulfide 5485 791 732 1009 1162

Dimethyldisulfide (DMDS) 29,379 3408 425 2423 6852 1.1–46

Methylpropyldisulfide 411 57 52 241 82 –

Dimethyltrisulfide 3616 87 580 1109 7.3

SUM 38,891 >27,452 1584 11,851 8256

Esters

Methylacetate † 3318 1247 1581 642 500–550,000

Ethylacetate 8259 9039 7518 >26,904 4668 600–180,000

Methylpropionate 1755 5412 1504 2283 906

Methylisobutyrate 650 780 545 372 86

Ethylpropionate 4097 3995 ¶ 9969 1960

Propylacetate 2622 3456 ¶ 3670 319.7 200–70,000

Methylbutyrate 2056 3281 1223 2430 427

Ethylisobutyrate 334 479 506 584 52

Ethylbutyrate 7254 6909 4194 >14,148 2121

Propylpropionate 1877 2253 1713 4318 603

N-butylacetate 744 723 397 1812 308

Odor Detection

Control FOS IMO RS SEM Threshold*

Ethyl-2-methylbutyrate 474 76 628 137 133

Butylisobutyrate 221 140 180 165 17

Propylbutyrate 3206 3920 1811 5760 822

Butylpropionate 445 764 208 962 167

Sec, butylbutyrate 74 186 65 228 41

Propyl-3-methylbutyrate 319 366 258 254 27

Isobutyl-n-butyrate 144 181 1 55

Methyl-2-methyl-pentanoate 52 39 112 22

Ester 279 917 516 168

Higher ester 753 1165 224 1280 239

SUM >35,615 47,361 23,295 >76,970 11,490

Others

Diethylether 1485 1574 356 392 750–35,000

C₄H₆N₂ 132 212 304 171 37

C₆H₈N₂ 26 35 33 94 16

SUM 1642 1821 336 621 368

Total SUM >161,832 >180,665 >100,169 >166,520 17,831

*Van Gemert and Nettenbreijer, 1977 in O’Neill and Philips, 1992.

†Dimethylsulfide and methylacetate, elution at the same time: sum of both: 7193 mg/m3.

‡Carbondisulphide present but high response factor; concentration could not be determined.

§Ethanol and acetone elution at the same time: sum of both: 1346 mg/m3.

¶Ethylpropionate and propylacetate elution at the same time: sum of both: 518 mg/m3.

Underlined values: 5 most abundant components.

FOS, fructo-oligosaccharides; IMO, isomalto-oligosaccharides; RS, resistant starch; SEM, standard error of the mean .