Zoo Biology 29 : 432448 (2010)

RESEARCH ARTICLE

Evaluation of Nutrient Digestibility

and Fecal Characteristics of Exotic
Felids Fed Horse- or Beef-Based Diets:
Use of the Domestic Cat as a Model
for Exotic Felids
Brittany M. Vester,1 Alison N. Beloshapka,1 Ingmar S. Middelbos,1
Sarah L. Burke,2 Cheryl L. Dikeman,2 Lee G. Simmons,2
and Kelly S. Swanson1,3
1
Department of Animal Sciences, University of Illinois, Urbana, Illinois
2
Henry Doorly Zoo, Omaha, Nebraska
3
Division of Nutritional Sciences, University of Illinois, Urbana, Illinois

The objective of this study was to determine the effects of feeding commercially
available beef- and horse-based diets on nutrient digestibility and fecal
characteristics of large captive exotic felids and domestic cats. Four species of
large exotic felids including cheetahs, Malayan tigers, jaguars, and Amur tigers,
and domestic cats were utilized in a crossover design. Raw meat diets included a
beef-based diet (57% protein; 28% fat) and a horse-based diet (51% protein; 30%
fat). All cats were acclimated to the diet for 16 days followed by a 4 day collection
period, where total feces, including one fresh sample, were collected. All feces
were scored on collection. Intake did not differ due to diet, but fecal output was
greater when cats consumed the horse-based diet. Total tract apparent dry matter
(DM) digestibility was higher (Po0.05) and organic matter (OM) and crude
protein (CP) digestibilities were lower (Po0.05) when cats were fed the beef-based
diet compared with the horse-based diet. CP digestibility was similar in domestic
cats and cheetahs, and greater (Po0.05) than Amur tigers. Fecal scores were
lower and fecal DM was greater (Po0.05) when cats consumed the horse-based
diet compared with the beef-based diet. Domestic cats had lower (Po0.05) fecal

Additional Supporting Information may be found in the online version of this article.
Correspondence to: Kelly S. Swanson, University of Illinois, 1207 W. Gregory Drive, 132 Animal
Sciences Laboratory, Urbana, IL 61801. E-mail: ksswanso@illinois.edu
Received 22 October 2008; Accepted 13 June 2009
DOI 10.1002/zoo.20275
Published online 14 October 2009 in Wiley Online Library (wileyonlinelibrary.com).

r 2009 Wiley-Liss, Inc.

 Horse- vs. Beef-Based Diets in Exotic Felids 433

ammonia concentrations compared with all other species. Fecal ammonia

concentrations were lowest (Po0.05) when cats were fed the horse-based diet.
Fecal total short-chain fatty acid (SCFA), branched-chain fatty acid (BCFA), and
butyrate concentrations were higher (Po0.05) when cats consumed the beef-
based diet. Our results suggest that the domestic cat serves as an appropriate
model for large exotic felid species, but differences among the species exist.
Decreased nutrient digestibility by tigers and jaguars should be considered when
developing feeding recommendations for these species based on domestic cat
data. Zoo Biol 29:432448, 2010. r 2009 Wiley-Liss, Inc.

Keywords: feline; comparative nutrition; diet comparison; digestibility; fecal analysis

INTRODUCTION
Feeding recommendations for large exotic felids are usually extrapolated from
those of domestic cats. Although this is common practice, direct comparisons of
domestic cats and large exotic species fed the same diets have not been done.
Furthermore, complete raw meat diets, various unsupplemented meat sources, or
whole prey are often fed in captivity, but may not be suitable for all species. Previous
digestibility trials have measured the total tract nutrient digestibility in several large
felids, but with very few animals per species, which is due to limited numbers of
animals available in zoos; a problem that persists today [Altman et al., 2005;
Barbiers et al., 1982; Crissey et al., 1997; Dierenfeld, 1993; Morris et al., 1974; Vester
et al., 2008; Wynne, 1989].
Conducting nutritional research in zoos is difcult due to several limitations,
including small population sizes, animal handling difculties, and diet types. Moreover,
the design of animal enclosures, the challenge of feeding animals individually while
respecting formed social groups, and the time required by zoo staff to perform a
digestibility trial often make them very difcult or impossible to conduct.
Previous research in our laboratory measured apparent total tract digestibility
and fecal fermentative end-product concentrations in bobcats, jaguars, cheetahs,
Malayan tigers, and Amur tigers [Vester et al., 2008]. To our knowledge, that study
was the rst to report fecal fermentative end-product concentrations in large felids.
Due to the increased longevity and close proximity of these animals to the public,
minimizing fecal putrefactive compounds is of great interest (odor control). Many
putrefactive compounds (e.g., ammonia, phenols and indoles, sulfur-containing
compounds) are produced in the large bowel via bacterial fermentation or amino
acid degradation that avoids digestion and absorption in the small intestine. As the
digestibility of the protein source decreases, the amount reaching the large bowel
increases. Most protein sources reaching the large bowel including connective tissues
(e.g., collagen) can be fermented by gastrointestinal bacteria [Macfarlane and
Cummings, 1991]. Therefore, as the digestibility of a dietary protein source increases,
the production of putrefactive compounds would be expected to decrease.
The objective of our study was to measure apparent total tract macronutrient
digestibility and fecal fermentative end-product concentrations of four large, exotic
felid species and domestic cats fed raw diets containing either horse- or beef- based
protein sources. We hypothesized that the domestic cat would serve as an
appropriate model for large exotic felids.

Zoo Biology
434 Vester et al.

MATERIALS AND METHODS

Animals
Four species of captive exotic felids and domestic cats were utilized for this
study including, cheetahs (Acinonyx jubatus; n 5 5; mean BW 5 39.673.1 kg),
jaguars (Panthera onca; n 5 4; mean BW 5 47.175.9 kg), Malayan tigers (Panthera
tigris corbetti; n 5 3; mean BW 5 94.875.3 kg), Amur tigers (Panthera tigris altaica;
n 5 5; mean BW 5 126.0716.8 kg), and domestic shorthair cats (Felis sylvestris;
n 5 9; mean BW 5 3.170.1 kg) (Table 1). All exotic felids were individually housed
on concrete oor enclosures maintained by the Henry Doorly Zoo in Omaha, NE.
Exotic felids were allowed access to outdoor enclosures during the study
(JulySeptember). Domestic cats were individually housed in stainless steel
metabolism cages (0.61  0.61  0.61 m3) at the University of Illinois. All animals
were individually housed throughout the study. All animal procedures were
approved by the Henry Doorly Zoo and University of Illinois Institutional Animal
Care and Use Committees (IACUC) before animal experimentation.

TABLE 1. Sex, body weight, and age of exotic and domestic felids

Species Sex BW (kg) BCS Age (y)

Cheetahs F 36.8
F 40.9
F 41.8
M 42.7
M 35.9
Jaguars F 42.3
M 55.0
M 50.9
M 46.4
Malayan tigers F 95.9
F 89.1
M 99.5
Amur tigers F 134.1
F 103.2
F 148.6
F 121.4
M 122.7
Domestic cats F 3.0
F 2.9
F 3.1
F 3.0
F 3.1
F 3.3
F 3.1
F 3.2
F 3.2
3 6.1
3 5.8
3 5.8
3 6.1
3 4.8
3 12.3
3 18.3
3 5.3
4 1.4
3 14.0
2.5 14.0
3 7.3
4 16.4
3 7.2
2 11.5
3.5 14.3
3 12.1
3 1.5
3 1.5
3 1.5
3 1.5
3 1.5
3 1.5
3 1.5
3 1.5
3 1.5

BW, body weight of exotic felids; determined during previous medical evaluation; BCS, body
condition score; determined on a 5-point scale with 1 5 emaciated, 3 5 ideal, and 5 5 obese.
All BCS were determined with special consideration of the species being evaluated.

Zoo Biology
 Horse- vs. Beef-Based Diets in Exotic Felids 435

Diet
Animals were randomized to one of two treatment diets including a beef-based
and a horse-based diet (Table 2). These raw diets were commercially prepared
(Central Nebraska Packing, Inc., North Platte, NE) in two lots of diet and intended
for nondomestic felid species, but were formulated to meet the nutrient requirements
of domestic cats [NRC, 2006]. Exotic cats were fed to maintain BW as determined
from records of the past 12 months. Domestic cats were fed to maintain BW, and
were weighed weekly to allow for an adjustment of food intake as necessary. Food
offered and refused was weighed daily. Water was provided ad libitum throughout
the study. Diet sub-samples were collected daily and stored at 201C until analyses.

Experimental Design
The study consisted of three, 20 day periods. During the study, cats were fed
the same amount each day without fasting periods. Before the start of the study,
all Amur tigers, Malayan tigers, and jaguars were fed the beef-based raw diet for
1.5 years. The previous cheetah diet, which was fed for 1.5 years, included a mixture
of the beef-based raw diet (75%) fed in this study and a commercially prepared horse
meat diet (25%). Domestic cats were maintained on commercially-available dry
extruded kibble diet before this study. All cats were acclimated to the raw diet and
feeding schedule from days 1 to 16 of the study.
From days 17 to 20, total feces were collected to determine apparent total tract
macronutrient digestibility. Feces were weighed, scored, and stored at 201C until
dried for proximate analyses. Scoring was conducted using a 5 point scale as follows:
1 5 hard, dry pellets; 2 5 dry, well formed stools; 3 5 soft, moist, formed stool;
4 5 soft, unformed stool; and 5 5 watery, liquid that can be poured. Upon

TABLE 2. Analyzed chemical composition and ingredients of raw meat diets fed to exotic and
domestic felids

Item Beef Horse

Dry matter 31.0 35.0
Organic matter (% DM) 92.2 91.5
Crude protein (% DM) 58.2 50.9
% Protein as collagen (% DM) 14.7 7.0
Fat (% DM) 26.5 30.3
Gross energy (kcal/g) 6.1 6.1
Calculated ME (kcal/g)a 4.7 4.9
Beef diet ingredients Beef, meat by-products, sh meal, soybean meal,
dried beet pulp, calcium carbonate, dried egg, brewers
dried yeast, Nebraska Brand feline vitamin premix,
salt, Nebraska Brand trace element premix
Horse diet ingredients Horsemeat (USDA inspected and passed), solka oc,
dicalcium phosphate, Nebraska Brand vitamin premix,
Nebraska Brand trace element premix, calcium
carbonate, salt, choline chloride, taurine, stabilized
ascorbic acidRoche Rovimix
a
Calculated using modied Atwater factors (8.5 kcal/g for fat and 3.5 kcal/g for protein and
nitrogen-free extract) and crude ber (maximum) based on the manufacturers label of 2.56 for
the beef-based diet and 3.00 for the horse-based diet.

Zoo Biology
436 Vester et al.

completion of the collection period, feces collected over the 4 days from each animal
were composited.
A fresh fecal sample was collected within 15 min of defecation from individual
animals after day 17. Fresh fecal samples were immediately weighed and frozen to
minimize loss of volatile components. An aliquot of feces was mixed with 5 ml 2N
HCl for ammonia, short-chain fatty acid (SCFA), and branched-chain fatty acid
(BCFA) determination and stored at 201C until analyzed. A second aliquot was
collected and stored at 201C for subsequent determination of phenol and indole
concentrations.

Chemical Analyses
Composited diet and fecal samples were dried at 551C and then ground
through a 2 mm screen in a Wiley mill (model 4, Thomas Scientic, Swedesboro, NJ)
in preparation for chemical analyses. Diet and fecal samples were analyzed for dry
matter (DM) and organic matter (OM) [AOAC, 1984]. Crude protein (CP) was
determined according to AOAC [1995] using a Leco Nitrogen/Protein Determinator
(model FP-2000, Leco Corporation, St. Joseph, MI). Fat concentrations were
determined by acid hydrolysis [AACC, 1983] followed by ether extraction [Budde,
1952]. Gross energy (GE) was determined by use of a bomb calorimeter (Model
1261, Parr Instrument Co., Moline, IL).
Fecal samples were analyzed for ammonia, SCFA, BCFA, phenol, and indole
concentrations as described earlier [Vester et al., 2008]. Fecal ammonia concentra-
tions were measured according to Chaney and Marbach [1962]. Diet samples were
analyzed for long-chain fatty acid (LCFA) concentrations according to Lepage and
Roy [1986]. Briey, internal standards and 0.1 g of diet were hexane extracted to
remove the lipid portion. Individual fatty acids were determined from the extracted
portion using gas chromatography (Hewlett-Packard 5890A Series II, Hewlett-
Packard, Palo Alto, CA) and external standards were used for identication and
quantication. Dietary amino acids were determined using an amino acid analyzer
(model 6300; Beckman Coulter, Fullerton, CA) at the Experiment Station
Laboratory, University of Missouri (Columbia, MO). Collagen concentration was
determined by multiplying hydroxyproline concentrations by 8.

Fecal Collection and Bacterial Genomic DNA Extraction

An aliquot of fresh feces was collected and stored at 801C until DNA
extraction was performed. Genomic DNA was extracted and isolated from fecal
samples (50 mg) using the QIAamp DNA Stool Mini-Kit (Qiagen, Valencia, CA).
Isolated DNA concentration (ng/ml) and purity (260/280) were measured using a
ND-1000 spectrophotometer (Nanodrop Technologies, Wilmington, DE). Template
DNA was diluted to 10 ng/ml and stored at 201C until further analysis.

Quantitative Polymerase Chain Reaction Analysis

Escherichia coli, Bidobacterium genus, Lactobacillus genus, and Clostridium
perfringens were quantied via qPCR using specic primers as described by Lubbs
et al. [2009]. Cycle threshold (Ct) values were plotted against standard curves for
quantication (CFU/g feces) of the target bacterial DNA from fecal samples.

Zoo Biology
 Horse- vs. Beef-Based Diets in Exotic Felids 437

PCR Amplification of 16S rDNA V3 Region

Template DNA was amplied using PCR involving primers (25 pmol of each
per reaction mixture) targeting the conserved regions anking the variable region 3
(V3) of the 16S rDNA gene as described by Muyzer et al. [1996] and Simpson et al.
[1999]. Eubacterial primer 341F (50 -CCTACGGGAGGCAGCAG-30 ) and uni-
versal primer 543R (50 -ATTACCGCGGCTGCTGG-30 ), the number corresponding
to E.coli 16S rDNA [Muyzer et al., 1996], were obtained from Operon Biotechno-
logies, Inc. (Huntsville, AL). To provide a more stable melting behavior of fragments
in DGGE, a GC-rich sequence (GC-clamp) was attached to the 50 -end of 341F
(50 -CGCCCGCCGCGCGCGGCGGGCGGGGCGGGGGGCACGGGGGG-30 ).
Primers 341F and 543R amplify PCR products containing approximately 200 base
pairs. Parameters for PCR amplication were done according to Lubbs et al. [2009].

Removal of Single-stranded DNA (ssDNA) From PCR Product

Single-stranded DNA may remain in samples following PCR and can present
problems for image analysis. Thus, mung bean nuclease (Stratagene, La Jolla, CA)
treatment was performed on PCR products to eliminate ssDNA as described by
Simpson et al. [1999].

Denaturing Gradient Gel Electrophoresis

Mung bean nuclease-treated amplicons were separated on polyacrylamide gels
[nondeionized 40% acrylamide/bis stock solution 37.5:1 (Bio-Rad Laboratories,
Hercules, CA)] on a 29% to 48% gradient according to Lubbs et al. [2009]. Bands
were stained in the gel using SYBR Green Stain I (Invitrogen, Eugene, OR) using a
4 ml 10,000X stain/20 ml 1X Tris-acetate-EDTA buffer dilution and developed for
20 min. Gels were read using a Gel Doc XR scanner (Bio-Rad Laboratories,
Hercules, CA) under UV light with a SYBR green emissions lter and analyzed using
QuantityOne (Bio-Rad Laboratories, Hercules, CA) and Bionumerics 5.0 (Applied
Maths Inc., Austin, TX) software.
Gels prepared for DGGE analysis were analyzed using Bionumerics 5.0
software (Applied Maths Inc., Austin, TX). Lanes and bands were automatically
detected by the program. Phylogenic trees were formed using unweighted pair group
method using Arithmetic Averages (UPGMA) clustering based on Dice coefcient
similarity values to create a dendrogram (Fig. 1).

Calculations and Statistical Analyses

Total tract apparent macronutrient digestibility was calculated using the
following equation: nutrient intake (DM basis)nutrient output (DM basis)/
nutrient intake (DM basis)  100. Data were analyzed using the Mixed Models
procedure of SAS (Cary, NC). The xed effects of species and diet were tested and
the interaction term investigated. Period and cat were considered random effects.
Differences among species were determined using a Fisher-protected least signicant
difference (LSD) with a Tukey adjustment to control for experiment-wise error.
Contrasts were conducted to compare the effect of each treatment within species.
Fecal score data were compared using the GLIMMIX procedure of SAS. A
probability of Po0.05 was accepted as statistically signicant and a probability of

Zoo Biology
438 Vester et al.

Fig. 1. Dendrogram of banding patterns in feces of domestic and exotic felids fed raw meat
diets. Each point represents a single cat fed either a horse- or beef-based diet and relationship
was determined by Dice coefcient values (relatedness of banding patterns).

Zoo Biology
 Horse- vs. Beef-Based Diets in Exotic Felids 439

Po0.10 was considered a trend. Reported pooled standard errors of the mean
(SEM) were determined according to the Mixed Models procedure of SAS.

RESULTS
Dietary chemical composition (Table 2, Supplementary Tables 1 and 2) was
varied between treatments, but met or exceeded all nutrient recommendations
according to NRC [2006]. The beef-based diet contained over twice as much collagen
as the horse-based diet (Table 2).
DM intake differed by species (Table 3). Malayan and Amur tigers consumed a
greater (Po0.05) amount of diet than all other species. Jaguars and cheetahs
consumed a greater (Po0.05) amount of diet than domestic cats. Diet type did not
affect DM intake or intake as expressed on a metabolic BW basis (kcal/BW0.75).
Caloric intake (kcal/kgBW0.75) was lower (Po0.05) in domestic cats as compared
with Malayan and Amur tigers.
Fecal output (g/day DM) was greater (Po0.05) in Amur and Malayan tigers
compared with all other species (Table 3). Jaguars and cheetahs had greater
(Po0.05) fecal output compared with domestic cats. Amur tigers had greater
(Po0.05) fecal output when consuming the horse-based diet. Domestic cats and

TABLE 3. Food intake and fecal output of exotic and domestic felids fed raw meat diets

P-value

Item Beef Horse SEM Trt Species TrtnSpecies

Intake (g/d DM) 71.87 0.38 0.0001 0.95
Domestica 53.3 53.0 0.99
Cheetahb 499.8 522.4 0.79
Jaguarb 506.7 527.4 0.82
Malayanc 970.3 1017.0 0.67
Amurc 1135.2 1222.4 0.31
Intake (kcal/d DM/kg BW0.75) 13.02 0.26 0.002 0.99
Domestica 129.6 134.7 0.41
Cheetahab 165.6 166.8 0.87
Jaguarab 146.3 151.1 0.58
Malayanb 179.8 185.9 0.55
Amurb 173.8 177.3 0.66
Output (g/d DM) 18.73 0.01 0.0001 0.15
Domestica 7.2 6.6 0.98
Cheetahb 57.8 75.7 0.50
Jaguarb 73.2 95.7 0.42
Malayanc 155.7 188.3 0.34
Amurc 175.0 259.9 0.002
Fecal Output (g, as-is)/intake 0.06 o0.0001 0.0002 0.86
(g, DM)
Domestica 0.3 0.2 0.002
Cheetaha 0.4 0.3 0.26
Jaguarb 0.5 0.4 0.06
Malayanb 0.6 0.5 0.12
Amurb 0.6 0.5 0.10

SEM, standard error of the mean.

a,b,c
Items within a column lacking a common superscript differ (Po0.05).

Zoo Biology
440 Vester et al.

cheetahs had lower output to input ratios (fecal output, g as-is/intake, g DM)
compared with all other species. Overall, the fecal output (g, as-is)/intake (g, DM)
was lower (Po0.05) when cats consumed the horse-based diet.
Apparent total tract DM digestibility was greater (Po0.05) in domestic cats
compared with jaguars and Amur tigers (Table 4). Apparent total tract OM
digestibility was greater (Po0.05) in domestic cats compared with Amur tigers.
Differences in apparent total tract OM digestibility between diets were most
pronounced in jaguars (Po0.02) and Amur tigers (Po0.04). Apparent total tract
DM and OM digestibilities were higher (Po0.001) when exotic cats were fed the
beef-based diet compared with the horse-based diet. Domestic cats had higher CP
digestibility when consuming the horse-based diet (Table 4). Apparent total tract CP
digestibility was greater (Po0.05) in domestic cats compared with jaguars and tigers.
Cheetahs had greater (Po0.05) CP digestibility compared with Amur tigers.

TABLE 4. Apparent total tract macronutrient and energy digestibility of exotic and domestic
felids fed raw meat diets

P-value

Item Beef Horse SEM Trt Species TrtSpecies

Dry matter 2.03 0.01 0.006 0.32
Domestica 89.3 90.0 0.69
Cheetahab 89.1 86.4 0.32
Jaguarb 87.3 81.8 0.06
Malayanab 86.1 82.3 0.28
Amurb 86.0 80.9 0.07
Organic matter 0.95 0.002 0.03 0.15
Domestica 96.2 96.5 0.71
Cheetahab 96.2 95.0 0.27
Jaguarab 96.4 93.5 0.02
Malayanab 96.1 93.7 0.10
Amurb 95.2 92.7 0.04
Crude protein 0.89 0.0001 0.002 0.85
Domestica 94.2 96.7 0.006
Cheetahab 94.1 95.8 0.13
Jaguarbc 92.3 94.8 0.04
Malayanbc 91.0 94.7 0.02
Amurc 91.5 93.5 0.08
Fat 0.68 0.19 0.0001 0.22
Domestica 95.2 97.0 0.05
Cheetahab 95.0 94.4 0.71
Jaguarb 92.4 91.9 0.30
Malayanb 92.3 94.1 0.33
Amurb 94.1 92.2 0.21
Energy 0.90 0.82 0.001 0.12
Domestica 95.5 97.3 0.06
Cheetahab 95.2 94.8 0.67
Jaguarb 92.7 92.1 0.61
Malayanb 92.6 94.3 0.24
Amurb 94.3 92.5 0.12

SEM, standard error of the mean.

a,b,c
Items within a column lacking a common superscript differ (Po0.05).

Zoo Biology
 Horse- vs. Beef-Based Diets in Exotic Felids 441

Average apparent total tract CP digestibility was greater (Po0.05) when all cats
consumed the horse-based diet compared with beef. Fat digestibility was not affected
by diet, but domestic cats had greater (Po0.05) fat digestibility compared with
jaguars and tigers. Energy digestibility was not affected by diet, but was higher
(Po0.05) in domestic cats compared with jaguars and tigers.
Fecal scores were lower (Po0.05) and fecal DM was greater (Po0.05) for all
cats consuming the horse-based diet compared with the beef-based diet (Table 5).
Domestic cats had lower (Po0.05) fecal ammonia concentrations compared with all
other species. Cheetahs had lower (Po0.0001) fecal ammonia concentrations when
consuming the horse-based diet. Overall, fecal ammonia concentrations were greater
(Po0.05) when cats consumed the beef-based diet. Phenol concentrations were lower

TABLE 5. Fecal characteristics and protein catabolites of exotic and domestic felids fed raw
meat diets
P-value

Item Beef Horse SEM Trt Species TrtSpecies

Fecal Score 0.22 0.0001 0.0001 0.0001

Domestica 2.9 1.2 0.0001
Cheetahab 3.1 2.2 0.0001
Jaguarb 3.3 2.4 0.0001
Malayanbc 3.5 2.8 0.001
Amurc 3.7 2.8 0.0001
Fecal DM (%) 4.25 0.001 0.02 0.006
Domestica 48.7 39.5 0.06
Cheetahab 28.7 42.8 0.03
Jaguarab 21.7 46.0 0.002
Malayanab 24.4 41.0 0.04
Amurb 24.7 35.7 0.12
Ammonia (mmol/g DM) 42.76 0.03 0.001 0.06
Domestica 179.3 207.2 0.37
Cheetahb 380.0 259.8 0.009
Jaguarb 403.1 326.8 0.12
Malayanab 311.9 333.0 0.68
Amurc 360.1 286.7 0.14
Phenol (mmol/g DM) 0.62 0.52 0.03 0.25
Domestica 0.6 0.5 0.85
Cheetahb 2.8 1.5 0.08
Jaguarab 1.0 1.7 0.35
Malayanab 1.0 1.6 0.47
Amurab 2.3 1.1 0.18
Indole (mmol/g DM) 0.34 0.83 0.045 0.21
Domestica 0.7 0.6 0.79
Cheetahab 1.6 1.9 0.41
Jaguarab 1.4 1.7 0.34
Malayanb 1.6 1.9 0.41
Amurab 2.0 1.3 0.06

SEM, standard error of the mean.

Fecal samples were scored based on the following scale: 1 5 hard, dry pellets; 2 5 dry, well
formed stools; 3 5 soft, moist, formed stool; 4 5 soft, unformed stool; and 5 5 watery, liquid
that can be poured.
a,b,c
Items within a column lacking a common superscript differ (Po0.05).

Zoo Biology
442 Vester et al.

TABLE 6. Fecal short-chain and branched-chain fatty acid concentrations of exotic and
domestic felids fed raw meat diets

P-value

Item Beef Horse SEM Trt Species TrtSpecies

Acetate (mmol/g DM) 70.24 0.0001 0.001 0.63

Domestica 379.2 140.1 0.001
Cheetahb 649.3 299.9 0.0006
Jaguarb 675.3 283.0 0.0006
Malayanb 600.6 306.3 0.01
Amurb 619.6 353.4 0.01
Propionate (mmol/g DM) 27.15 0.0001 0.001 0.86
Domestica 131.2 43.9 0.001
Cheetahb 211.0 106.4 0.003
Jaguarb 246.5 106.9 0.001
Malayanb 211.5 114.8 0.01
Amurb 233.3 117.0 0.008
Butyrate (mmol/g DM) 9.22 0.005 0.0001 0.59
Domestica 32.5 21.8 0.20
Cheetahb 73.3 43.6 0.01
Jaguarb 80.1 57.5 0.07
Malayanb 80.1 67.6 0.47
Amurb 79.9 69.9 0.43
Total SCFA (mmol/g DM) 102.28 0.0001 0.0005 0.67
Domestica 545.4 208.3 0.002
Cheetahb 936.4 452.4 0.0009
Jaguarb 1006.0 457.3 0.0007
Malayanb 892.3 482.4 0.01
Amurb 924.5 543.8 0.01
Isobutyrate (mmol/g DM) 1.98 0.0009 0.03 0.27
Domesticc 7.9 5.7 0.10
Cheetahd 13.7 10.2 0.06
Jaguarcd 12.9 10.0 0.15
Malayancd 11.9 11.0 0.66
Amurd 15.8 8.6 0.003
Isovalerate (mmol/g DM) 2.63 0.05 0.08 0.05
Domestic 6.7 5.9 0.63
Cheetah 14.6 11.5 0.16
Jaguar 13.2 12.7 0.86
Malayan 12.6 15.2 0.32
Amur 17.6 8.8 0.003
Valerate (mmol/g DM) 0.86 0.59 0.0008 0.95
Domestica 4.7 4.5 0.73
Cheetahb 0.7 0.8 0.94
Jaguarb 2.0 1.1 0.44
Malayanb 1.2 0.7 0.69
Amurb 1.0 1.3 0.80
Total BCFA (mmol/g DM) 4.69 0.01 0.25 0.15
Domestic 18.3 14.9 0.27
Cheetah 28.0 21.4 0.12
Jaguar 26.7 22.9 0.42
Malayan 24.5 22.9 0.73
Amur 33.7 17.6 0.004

SEM, standrard error of the mean.

Total SCFA 5 acetate1propionate1butyrate.
Total BCFA 5 isobutyrate1valerate1isovalerate.
a,b
Items within a column lacking a common superscript differ (Po0.05).
c,d
Items within a column lacking a common superscript tend (Po0.10) to differ.

Zoo Biology
 Horse- vs. Beef-Based Diets in Exotic Felids 443

(Po0.05) in domestic cats as compared with cheetahs. Fecal indole concentrations

were lower (Po0.05) in domestic cats as compared with Malayan tigers. Fecal
phenol and indole concentrations were not different between diets.
Fecal acetate, propionate, butyrate, and total SCFA concentrations were lower
(Po0.05) in domestic cats as compared with all other species (Table 6). Fecal
isobutyrate concentrations tended to be lower (Po0.10) in domestic cats compared
with cheetahs and Amur tigers. Fecal isovalerate concentrations were most heavily
inuenced by diet in Amur tigers, with decreasing (Po0.05) concentrations when
consuming the horse-based diet. Fecal valerate concentrations were greater
(Po0.05) in domestic cats compared with all other species, but were not affected
by diet. Fecal concentrations of all SCFA and BCFA, except valerate, were lower
(Po0.05) when cats were fed the horse-based diet as compared with the beef-based
diet. When cats consumed the beef-based diet, the proportion of butyrate was
decreased (Po0.05) as compared with the horse-based diet (7.4 vs. 13.4%,
respectively).
Overall, the fecal E. coli concentrations were greater (Po0.05) in domestic cats
compared with all other species (Table 7). Fecal E. coli concentrations were greater
(Po0.05) when cats consumed the horse-based diet. E. coli was below detectable

TABLE 7. Fecal microbial populations (log CFU/g DM feces) of exotic and domestic felids fed
raw meat diets

P-value

Item Beef Horse SEM Trt Species TrtSpecies

E. coli 1.01 0.0001 0.0001 0.0001
Domesticb 11.1 12.0 0.38
Cheetaha ND2 11.1 0.0001
Jaguara 4.7 11.4 0.0001
Malayana ND 12.6 0.0001
Amura 2.5 11.8 0.0001
Clostridium perfringens 1.45 0.0001 0.0002 0.0008
Domesticb 12.9 13.4 0.68
Cheetahab 7.0 13.4 0.002
Jaguarab 7.5 12.6 0.009
Malayana ND 12.9 0.0001
Amurb 8.7 12.9 0.02
Lactobacillus 1.10 0.0001 0.0001 0.0005
Domesticb 11.0 11.4 0.65
Cheetaha 2.4 10.7 0.0001
Jaguara 4.9 10.0 0.002
Malayana 2.3 10.4 0.0001
Amura 5.6 11.0 0.0008
Bidobacterium 0.22 0.19 0.0001 0.004
Domesticb 8.2 7.1 0.0001
Cheetaha 6.7 7.0 0.37
Jaguara 6.6 6.6 0.81
Malayana 6.5 6.6 0.62
Amura 6.9 6.6 0.45

SEM, standrard error of the mean; ND, not detectable.

a,b
Items within a column lacking a common superscript differ (Po0.05).

Zoo Biology
444 Vester et al.

limits (4.3 log CFU/g feces) in cheetahs and Malayan tigers consuming the
beef-based diet. Overall, the fecal C. perfringens concentrations were greater
(Po0.05) in domestic cats as compared with Malayan tigers. Fecal C. perfringens
was below detectable limits (3.98 CFU/g feces) in Malayan tigers fed the beef-based
diet. Fecal C. perfringens concentrations were greater (Po0.05) when cats consumed
the horse-based diet. Overall, domestic cats had greater (Po0.05) Lactobacillus
concentrations compared with all other species. Fecal Lactobacillus concentrations
were greater when cats consumed the horse-based diet. Domestic cats had greater
(Po0.05) Bidobacterium concentrations compared with all other species, and
decreased (Po0.05) when consuming the horse-based diet. Fecal Bidobacterium
concentrations were not affected by diet. Analysis of fecal banding patterns with
DGGE indicated no signicant clustering due to species or diet groups (Fig. 1).

DISCUSSION
Feeding horse- and beef-based commercially prepared diets to large exotic
felids is common in North American zoos; however, there is a paucity of published
literature on the digestibility of such diets in these species. Because nutrient
requirements are unknown in large felids, nutritionists use known requirements of
domestic cats to establish feeding recommendations of exotic felids. This study
compared domestic cats with exotic felids fed the same diets to test the suitability of
the domestic cat as a model for these species.
In this study, diets contained multiple protein sources and one of the two ber
sources. The beef-based diet contained a moderately fermentable ber source (beet
pulp), while the horse-based diet contained a nonfermentable ber source (cellulose).
Furthermore, the beef diet contained higher collagen content. Thus, differences due
to diet cannot be attributed to protein source, collagen concentration, or ber alone,
but the ingredient combination. Because dietary ber (fermentable vs. nonfermen-
table) may affect the microbial protein concentration in the feces, interpretation of
CP digestibility data in this study is difcult.
Diet inuenced fecal output and total tract macronutrient digestibility in these
species, which may have been due to the different ber sources included in the beef-
and horse-based diets. Caloric intake (kcal/kg BW0.75) was consistent, with higher
nutrient digestibilities by different species, and domestic cats, cheetahs, and jaguars
requiring less food to maintain BW due to their increased utilization of the diets.
Daily metabolizable energy requirements of exotic cats is estimated to be between 55
and 260 kcal  kg BW0.75 [NRC, 2006]. Allen et al. [1995] reported that metaboliz-
able energy requirements could not be extrapolated from domestic cats due to the
large variation within and among exotic felid species. Digestible energy intake,
expressed relative to metabolic BW (kcal/kg BW0.75), was 150185 in cheetahs and
200260 in tigers [Allen et al., 1995]. The authors suggested that, of species evaluated
in that study, young clouded leopards and lions were similar to that reported in
domestic cats, but cheetahs and tigers required more energy to maintain BW. Our
data regarding tigers are similar to Allen et al. [1995], with tigers requiring a higher
caloric intake to maintain BW. Interestingly, however, cheetahs were similar to
domestic cats in this study. This may be due to the fact that digestible energy and
digestive efciency were not directly measured in the domestic cat, but rather a
predictive equation used. Also, activity level, which has a signicant impact on

Zoo Biology
 Horse- vs. Beef-Based Diets in Exotic Felids 445

energy needs, may have differed in cheetahs between studies and contributed to the
conicting results. The differences in energy intake needed to maintain body weight
noted in this study are likely due to the decrease noted in nutrient digestibility in the
larger species when consuming the diets tested.
Digestibility differences were most evident with CP, which was greater when
cats consumed the horse-based diet, and was affected by diet in all species except
Amur tigers. Increased dietary ber has been reported to decrease OM digestibility in
cats [Kienzle et al., 1991]. Diets reported by Kienzle et al. [1991] contained beef,
vitamin and mineral mixes, and differing inclusion levels of low digestible ingredients
(10% wheat bran, 10% cellulose, 15% horn meal, 15% feather meal, 15% rumen
content, and 15% grass meal). In addition, Kienzle et al. [1991] reported a lower CP
digestibility in all ber-supplemented meat diets, except the cellulose-supplemented
diet. Thus, in this study, the decreased CP digestibility of the beef-based diet may
have been due to increased fermentative activity, leading to an increased production
of bacterial protein produced in the large bowel, which will be excreted in the feces
thereby underestimating CP digestibility. Fecal bacterial protein content was not
determined in this study. Because dietary ber source confounds CP digestibility
data, comparing protein sources in diets containing a single ber source would be
most benecial in future studies.
Despite the confounded ingredients in these diets, we noted similar DM
(87.388.9%), CP (92.593.2%), and fat (93.996.2%) digestibilities in an earlier
study evaluating a beef-based diet fed to exotic felids [Vester et al., 2008]. Organic
matter digestibility in this study was greater than that noted in our earlier study. The
beef-based diets tested in both studies contained similar ingredients; however, dietary
composition was different. These diets appear to be highly variable in macronutrient
composition, even when using the same ingredients across batches, which may have
led to the slight differences in nutrient digestibility between studies, as was noted also
by Bechert et al. [2002].
Bechert et al. [2002] reported lower DM digestibilities (75.776.7%) of
commercial horse-, cattle-, turkey-, and deer-based diets fed to cheetahs as compared
with the species evaluated in this study. Diets fed in that study contained greater CP
(5983.8%) and lower crude fat (9.124.7%) concentrations than this study [Bechert
et al., 2002]. Salter et al. [1999] reported similar protein (94.6%) and fat (96.9%)
digestibilities of two cheetahs fed a horse meat-based diet, as compared with
cheetahs in this study. Morris et al. [1974] noted DM, OM, and nitrogen
digestibilities of three tigers were lower (79.2, 82.5, and 88.8%, respectively) than
tigers evaluated in this study. Fat digestibility, however, was consistent with species
measured in this study.
Barbiers et al. [1982] noted tigers (n 5 2) fed a horse meat diet had similar DM
(83.7%), OM (90.0%), and energy (91.8%) digestibilities as tigers fed the horse-
based diet in this study. CP digestibility was lower (88.2%) and fat digestibility was
greater (98.9%) in tigers evaluated in that study as compared with the tigers fed
the horse-based diet in this study. Wynne [1989] noted lower CP (79.2%) and
fat (91.5%) digestibilities for all species than those determined in this study
[Wynne, 1989].
Fecal scores and DM percentage were inuenced by both diet and species in
this study. In general, diet had a greater impact in smaller cats. Similar results have
been reported in small vs. large and giant breed dogs. Large-breed dogs have been

Zoo Biology
446 Vester et al.

reported to have higher fecal output of poorer quality as compared with small breed
dogs fed the same diet [Weber et al., 2004; Hernot et al., 2004, 2005, 2006]. Those
authors suggested that the differences in fecal characteristics were due to longer
transit time, increased intestinal permeability, or increased fermentative activity in
the large bowel of large-breed dogs. Using our 5-point scale, an ideal fecal score
would be a 3 out of 5. Interestingly, tigers had an ideal (2.8) fecal score when
consuming the horse-based diet, while the other species had an ideal fecal score when
consuming the beef-based diet. This is in agreement with water holding capacity of
the beef-based diet, determined by fecal output (as-is)/DM intake, and was likely due
to its soluble ber source (beet pulp) or collagen concentration. These data may
indicate that tigers are better suited to consume a diet containing a nonfermentable
ber source, a diet having a greater CP digestibility, or a diet that contains lower
amounts of collagen, as high collagen supplementation (28% of diet) has been
reported to induced diarrhea in rats [Whitmore et al., 1975].
In zoos, a large number of animals are housed in a small area. Given the air
quality considerations for large felid housing facilities and the link between
putrefactive compounds (produced by dietary protein fermentation in the large
bowel) and gastrointestinal disease, it is important to limit the production of these
compounds if possible. Increased CP digestibility noted with the horse-based diet
and in domestic cats may have led to the reduced ammonia production noted in this
study. Because total tract digestibility was tested in this study, measuring the amount
of CP entering the large bowel was not possible and would require further testing.
Fecal ammonia concentrations in this study were higher than those reported
earlier in domestic cats [Terada et al., 1993]. Terada et al. [1993] tested diets
containing 30.2 or 19.0% CP, which were lower than CP concentrations fed in this
study (50.958.2% CP), and was likely the reason for the differences between the
studies. Other putrefactive compound concentrations, phenol and indole, were
similar to those reported in domestic dogs and cats [Flickinger et al., 2003; Swanson
et al., 2002; Terada et al., 1993].
SCFAs are primarily produced from bacterial fermentation of carbohydrates
in the large intestine. Given the nonfermentable ber source in the horse-based diet,
lower SCFA concentrations were expected. Butyrate is the main fuel source for
colonocytes. Therefore, feeding a diet that increases butyrate concentrations may be
benecial. Hesta et al. [2001] fed cats a diet containing no supplemental ber and
reported similar SCFA ratios to this study. BCFA production can serve as an
indicator of protein breakdown in the large bowel. Fecal isobutyrate, isovalerate,
and total BCFA concentrations were numerically lowest in domestic cats. This may
have been due to increased CP digestibility of domestic cats compared with the other
species. Again, because only total tract digestibility was measured, this remains
speculation. Fecal BCFA were also lowest in cats consuming the horse-based diet,
potentially due to increased CP digestibility.
Interestingly, despite the increase in fermentative end-product concentrations
when cats consumed the beef-based diet, fecal bacteria concentrations measured in
this study were often greater when cats consumed the horse-based diet. This indicates
that while there was an increase in fermentation, diet did not affect E. coli,
C. perfringens, Lactobacillus, or Bidobacterium number. Given the high dietary
protein concentrations, it was expected that C. perfringens concentrations would be
high, regardless of diet. Previous research in our laboratory in adult domestic cats

Zoo Biology
 Horse- vs. Beef-Based Diets in Exotic Felids 447

[Lubbs et al., 2009] noted slightly lower C. perfringens concentrations than those
reported in this study; however, those data were reported on a wet fecal basis, not
DM basis as in this study.
Due to the difference in dietary ber source, large fecal microbial differences
were expected between diets in all species. Contrary to this hypothesis, there was no
clear clustering of banding patterns due to diet. Furthermore, while a unique group
of domestic cats was evident, no other species clustered together. The cluster of
domestic cats was likely to due to large environmental differences that existed
between these two populations (research domestic cat population vs. indoor/outdoor
enclosures of exotic felids). Canine fecal microora analyses by DGGE have
indicated that each individual appears to have a stable and unique banding pattern
regardless of diet fed [Simpson et al., 2002]. Future research in this area may require
more sensitive techniques to determine changes among species.
Most variables measured in this study showed no interaction of diet and
species, indicating that while there were differences between diets, most species
responded in a similar manner. This response suggests that modied dietary feeding
guidelines of exotic felid species may be based on the response of domestic cats. Our
data suggest the large exotic felids may be better suited to a diet containing little
fermentable ber, a more digestible protein source, or less dietary collagen, as fecal
scores improved on the horse-based diet. Measurement of fecal fermentative end-
product concentrations may be an another tool by which zoos can monitor gut
health as an animal ages or during dietary changes. Further evaluation of dietary
ber to maintain intestinal health in large cats is warranted.

REFERENCES
AACC. 1983. Approved methods, 8th ed. St. Paul:
American Association of Cereal Chemists.
AOAC. 1984. Ofcial methods of analysis, 14th ed.
Washington, DC: Association of Ofcial Analy-
tical Chemists.
AOAC. 1995. Ofcial methods of analysis, 15th ed.
In: Cunniff P, editor. Washington: Association
of Ofcial Analytical Chemists.
Allen ME, Oftedal OT, Earle KE, Seidensticker J,
Vilarin L. 1995. Do maintenance energy require-
ments of felids reect their feeding strategies?
Nutr Adivisory Group Proc 1:97103.
Altman JD, Gross KL, Lowry SR. 2005. Nutri-
tional and behavioral effects of gorge and fast
feeding in captive lions. J Appl Anim Welf Sci
8:4757.
Barbiers RB, Vosburgh LM, Ku PK, Ullrey DE.
1982. Digestive efciencies and maintenance
energy requirements of captive wild felidae:
Cougar (Felis concolor); Leopard (Panthera
pardus); Lion (Panthera leo); and Tiger (Panthera
tigris). J Zoo Anim Med 13:3237.
Bechert U, Mortenson J, Dierenfeld ES, Cheeke P,
Keller M, Holick M, Chen TC, Rogers QR.
2002. Diet composition and blood values of
captive cheetahs (Acinonyx jubatus) fed either
supplemented meat or commercial food prepara-
tions. J Zoo Wildl Med. 33:1628.
Budde EF. 1952. The determination of fat in baked
biscuit type of dog foods. J AOAC 35: 799805.
Chaney AL, Marbach EP. 1962. Modied reagents
for determination of urea and ammonia. Clin
Chem 8:130132.
Crissey SD, Swanson JA, Lintzenich BA,
Brewer BA, Slifka KA. 1997. Use of a raw
meat-based diet or a dry kibble diet for sand cats
(Felis margarita). J Anim Sci 75: 21542160.
Dierenfeld ES. 1993. Nutrition of captive cheetahs:
food composition and blood parameters. Zoo
Biol 12:143150.
Flickinger EA, Schreijen EM, Patil AR,
Hussein HS, Grieshop CM, Merchen NR, Fahey
Jr GC. 2003. Nutrient digestibilities, microbial
populations, and protein catabolites as affected
by fructan supplementation of dog diets. J Anim
Sci 81:20082018.
Hernot DC, Weber MP, Biourge VC, Martin LJ,
Dumon HJ, Nguyen PG. 2004. Relationship
between electrolyte apparent absorption and
fecal quality in adult dogs differing in body size.
J Nutr 134:2031S2034S.
Hernot DC, Biourge VC, Martin LJ, Dumon HJ,
Nguyen PG. 2005. Relationship between total
transit time and faecal quality in adult dogs
differing in body size. J Anim Physiol Anim Nutr
89:189193.

Zoo Biology
448 Vester et al.
Hernot DC, Dumon HJ, Biourge VC, Martin LJ,
Nguyen PG. 2006. Evaluation of association
between body size and large intestinal transit
time in healthy dogs. Am J Vet Res 67:342347.
Hesta M, Janssens GPJ, Debraekeleer J,
De Wilde R. 2001. The effect of oligofructose
and inulin on faecal characteristics and nutrient
digestibility in healthy cats. J Anim Physiol Anim
Nutr 85:135141.
Kienzle E, Meyer H, Schneider R. 1991. Investiga-
tions on palatability, digestibility, and tolerance
of low digestible food components in cats. J Nutr
121:S56S57.
Lepage G, Roy CC. 1986. Direct transesterica-
tion of all classes of lipids in a one-step reaction.
J Lipid Res 27:114120.
Lubbs DC, Vester BM, Fastinger ND, Swanson KS.
2009. Dietary protein concentration affects intest-
inal microbiota of adult cats: a study using DGGE
and qPCR to evaluate differences in microbial
populations in the feline gastrointestinal tract. J
Anim Physiol Anim Nutr 93:113121.
Macfarlane GT, Cummings JH. 1991. The colonic
ora, fermentation, and large bowel digestive
function. In: The large intestine: physiology,
pathophysiology, and disease. Phillips SF,
Pemberton JH, Shorter RG, editors. New York,
NY: Raven Press, Ltd. p 5191.
Morris JG, Fujimoto J, Berry SC. 1974. The
comparative digestibility of a zoo diet fed to 13
species of felid and a badger. Int Zoo Yearb
14:169171.
Muyzer G, Hottentrager S, Teske A, Wawer C.
1996. Denaturing gradient gel electrophoresis
(DGGE) of PCR-amplied 16S rDNAA new
molecular approach to analyze the genetic
diversity of mixed microbial communities. In:
Molecular microbial ecology manual. Akkermans A,
van Elsas JD, de Bruijn F, editors. Dordecht,
Netherlands: Kluwer Academic Publishers.
NRC. 2006. Nutrient requirements of dogs and
cats. Washington, DC: The National Academies
Press. 398p.
Zoo Biology
Salter S, Twinney J, Bernal-Soler L, Atkinson J,
Valdes EV. 1999. Evaluation of alternative feline
diets at the Toronto Zoo. Nutritional Advisory
Conference 1999.
Simpson JM, McCracken VJ, White BA,
Gaskins HR, Mackie RI. 1999. Application of
denaturant gradient gel electrophoresis for the
analysis of the porcine gastrointestinal micro-
biota. J Microbiol Methods 36:167179.
Simpson JM, Martineau B, Jones WE, Ballam JM,
Mackie RI. 2002. Characterization of fecal
bacterial populations in canines: effects of
age, breed and dietary ber. Microb Ecol 44:
186197.
Swanson KS, Grieshop CM, Flickinger EA,
Bauer LL, Chow J, Wolf BW, Garleb KA,
Fahey Jr GC. 2002. Fructooligosaccharides and
Lactobacillus acidophilus modify gut microbial
populations, total tract nutrient digestibilities
and fecal protein catabolite concentrations in
healthy adult dogs. J Nutr 132:37213731.
Terada A, Hara H, Kato S, Kimura T, Fujimori I,
Hara K, Maruyama T, Mitsuoka T. 1993. Effect
of lactosucrose (4G-beta-D-galactosylsucrose)
on fecal ora and fecal putrefactive products of
cats. J Vet Med Sci 55:291295.
Vester BM, Burke SL, Dikeman CL, Simmons LG,
Swanson KS. 2008. Nutrient digestibility and
fecal characteristics are different among captive
exotic felids fed a beef-based raw diet. Zoo Biol
27:126136.
Weber MP, Hernot D, Nguyen PG, Biourge VC,
Dumon HJ. 2004. Effect of size on electrolyte
apparent absorption rates and fermentative
activity in dogs. J Anim Physiol Anim Nutr 88:
356365.
Whitmore R, Booth A, Naghski J, Swift C. 1975.
Digestibility and safety of limed hide collagen
in rat feeding experiments. J Food Sci 40:
101104.
Wynne JE. 1989. Comparative digestibility values
in four species of felidae. J Zoo Wildl Med
20:5356.