Evaluation of high-protein diets differing in protein source in healthy adult dogs

Abstract

Given the dynamic market for protein-based ingredients in the pet food industry, demand continues to increase for both plant- and animal-based options. Protein sources contain different amino acid (AA) profiles and vary in digestibility, affecting protein quality. The objective of this study was to evaluate the apparent total tract digestibility (ATTD) of canine diets differing in protein source and test their effects on serum metabolites and fecal characteristics, metabolites, and microbiota of healthy adult dogs consuming them. Four extruded diets were formulated to be isonitrogenous and meet the nutrient needs for adult dogs at maintenance, with the primary difference being protein source: 1) fresh deboned, dried, and spray-dried chicken (DC), 2) chicken by-product meal (CBPM), 3) wheat gluten meal (WGM), and 4) corn gluten meal (CGM). Twelve adult spayed female beagles (body weight [BW] = 9.9 ± 1.0 kg; age = 6.3 ± 1.1 yr) were used in a replicated 4 × 4 Latin square design (n = 12/treatment). Each period consisted of a 22-d adaptation phase, 5 d for fecal collection, and 1 d for blood collection. Fecal microbiota data were analyzed using QIIME 2.2020.8. All other data were analyzed using the Mixed Models procedure of SAS version 9.4. Fecal scores were higher (P < 0.05; looser stools) in dogs fed DC or CBPM than those fed WGM or CGM, but all remained within an appropriate range. Dry matter ATTD was lower (P < 0.05) in dogs fed CBPM or CGM than those fed DC or WGM. Crude protein ATTD was lower (P < 0.05) in dogs fed DC or CGM than those fed WGM. Dogs fed CBPM had lower (P < 0.05) organic matter, crude protein, and energy ATTD than those fed the other diets. Fecal indole was higher (P < 0.05) in dogs fed CBPM than those fed WGM. Fecal short-chain fatty acids were higher (P < 0.05) in dogs fed DC than those fed CGM. Fecal branched-chain fatty acids were higher (P < 0.05) in dogs fed DC or CBPM than those fed WGM. Fecal ammonia was higher (P < 0.05) in dogs fed DC or CBPM than those fed WGM or CGM. The relative abundances of three bacterial phyla and nine bacterial genera were shifted among treatment groups (P < 0.05). Considering AA profiles and digestibility data, the DC diet protein sources provided the highest quality protein without additional AA supplementation, but the animal-based protein diets resulted in higher fecal proteolytic metabolites. Further studies evaluating moderate dietary protein concentrations are needed to better compare plant- and animal-based protein sources.

The current study tested dog diets differing in protein source, including those derived from plants and animals. Dietary differences were noted in nutrient digestibility, fecal metabolites, and fecal microbiota, but gene expression was unaffected.

Introduction

Many pet owners think of their pets as part of the family and are concerned about their health and longevity. Given the important role that nutrition plays on pet health, owners are becoming more aware and selective of the foods they are choosing to purchase for their pets. Diet choices reflect the personal preferences of owners, with different social and cultural factors influencing the decision-making process (Vinassa et al., 2020). One of the most important dietary considerations of consumers and pet food manufacturers is protein source and concentration (Oberbauer and Larsen, 2021).

Because the body requires specific amino acids (AA), dietary AA concentrations are more important than the crude protein (CP) concentration per se (Knight and Leitsberger, 2016). Therefore, well-formulated diets not only provide adequate CP, but also AA in dietary concentrations that meet the needs of the dog or cat population in question. Protein quality has been defined as the ability of dietary protein to meet the needs for regular metabolism and maintenance of body issues (Millward et al., 2008). If an ingredient meets all AA requirements, it is considered a complete, high-quality protein source. However, the use of complementary proteins, by combining two or more sources to meet the AA requirements, is necessary for ingredients that cannot meet the needs on their own. Having the knowledge of how different protein sources can impact the nutritional status of dogs and cats and how they may be applied to the pet food industry may optimize the use of different ingredients to fit unique needs and goals.

Rendered animal co-products provide the majority of protein used by the commercial pet food industry today, with approximately 31% of all rendered proteins and 15% of all rendered fats going into pet foods (Meeker and Meisinger, 2015). Rendering is the process of both physical and chemical transformation using a variety of equipment and processes on animal materials that includes the application of heat, extraction of moisture, and separation of fat (Meeker and Meisinger, 2015). Rendered products are highly sustainable and economical, but excessive heat processing is known to reduce nutrient digestibility and lead to indispensable AA losses (Johnson et al., 1998; Meeker and Meisinger, 2015; Oba et al., 2019). Due to concerns about the effects of processing, which may be true of some rendered ingredients, fresh meats have become a popular alternative in recent years. Fresh mechanically deboned meat refers to meat that has not undergone any treatment except to maintain cold temperature during processing (Meineri et al., 2021). Other alternatives for rendered animal proteins are those derived from plants (e.g., grains, legumes). While plant-based proteins are perceived to be healthier, have a lower environmental impact, and reduce animal welfare concerns (Dodd et al., 2018), they must be used in a complementary manner or require supplementation because they do not contain sufficient concentrations of some indispensable AA (e.g., lysine, tryptophan, methionine) on their own.

The objective of this study was to determine the apparent total tract digestibility (ATTD) of canine diets differing in protein source and to determine the whole blood gene expression and fecal characteristics, metabolites, and microbiota of healthy adult dogs consuming them. We hypothesized that the plant-based protein diets would be less digestible, increase fecal metabolite concentrations coming from protein fermentation (branched-chain fatty acids [BCFA]; phenols and indoles), negatively impact fecal microbiota populations, and alter whole blood gene expression.

Materials and Methods

All animal care procedures were approved by the University of Illinois at Urbana-Champaign Institutional Animal Care and Use Committee prior to animal experimentation.

Animals and housing

Twelve adult spayed female beagles (body weight [BW] = 9.9 ± 1.0 kg; age =6.3 ± 1.1 yr) were used in this study. Dogs were housed individually in pens (1.22 m wide × 1.85 m long) in a humidity- and temperature-controlled room on a 14 h light: 10 h dark cycle. Dogs had ad libitum access to fresh water and were fed once a day (8:00 am) to maintain BW throughout the study. Dogs were weighed weekly, prior to morning feeding. Remaining food was weighed every day to calculate intake. Dogs were weighed and body condition score (BCS) was recorded weekly prior of feeding. A 9-point BCS system was used (Laflamme, 1997). Dogs had access to toys at all times and were socialized at least two times per week where they were given other toys, further enrichment, and socialization with each other and humans.

Experimental timeline and diets

A replicated 4 × 4 Latin square design (n = 12/group) was used to test the four experimental treatments. The experiment consisted of four 28-d periods, with each period consisting of an adaptation phase (day 1 to 22), a 5-d fecal collection phase (day 23 to 27), and 1 d for blood collection (day 28). Four extruded experimental test kibble diets were formulated by Champion Petfoods LP to meet all Association of American Feed Control Officials (AAFCO, 2020) nutrient profiles for adult dogs at maintenance. Diets were formulated to be isonitrogenous and isocaloric, and targets for total dietary fiber (TDF) level and fractions were similar across all treatments. Diets differed in protein source, which included: 1) fresh deboned, dried, and spray-dried chicken (DC), 2) chicken byproduct meal (CBPM), 3) wheat gluten meal (WGM), and 4) corn gluten meal (CGM) (Tables 1 and 2).

Table 1.

Ingredient and analyzed chemical composition of experimental diets differing in protein source¹

	DC	CBPM	WGM	CGM
Ingredient	------------- %, as is --------
Deboned chicken	41.00	0.00	0.00	0.00
Dried chicken	15.70	0.00	0.00	0.00
Spray dried chicken	6.50	0.00	0.00	0.00
Chicken byproduct meal	0.00	45.50	0.00	0.00
Wheat gluten protein	0.00	0.00	46.29	0.00
Corn gluten meal	0.00	0.00	0.00	51.80
Taurine	0.00	0.10	0.10	0.10
L-Lysine	0.00	0.00	0.05	0.01
Potato powder	25.63	37.63	24.17	21.37
Chicken fat	4.30	11.00	17.50	17.00
Liquid chicken palatant	2.50	2.50	2.50	2.50
Dry chicken palatant (dogs)	1.00	1.00	1.00	1.00
Monodicalcium phosphate 21%	1.20	0.00	3.20	3.00
CaCO3	0.00	0.00	2.00	1.50
Miscanthus grass	1.50	1.60	2.00	0.00
Salt	0.35	0.35	0.35	0.35
Potassium chloride	0.00	0.00	0.30	0.32
Choline chloride 60%	0.10	0.10	0.25	0.25
Trace mineral premix	0.07	0.07	0.13	0.12
Vitamin premix	0.10	0.10	0.11	0.11
Natural antioxidant, dry	0.05	0.05	0.05	0.05
Chemical composition
Dry matter, %	88.4	90.3	92.8	89.6
	-------- % dry matter --------
Organic matter, %	92.82	91.81	91.84	93.62
Ash, %	7.18	8.19	8.16	6.38
Acid-hydrolyzed fat, %	23.0	18.0	20.2	22.7
Crude protein, %	41.7	40.1	39.8	41.0
Total starch, %	27.37	27.54	24.24	23.64
Gelatinized starch, %	26.07	25.77	22.86	22.4
Cook, %	95.3	93.6	94.3	94.8
Total dietary fiber, %	7.16	7.62	8.10	6.98
Insoluble fiber, %	6.15	6.09	7.00	6.13
Soluble fiber, %	1.01	1.52	1.09	0.85
Nitrogen-free extract, %2	20.96	26.09	23.74	22.94
Gross energy, kcal/g	5.66	5.38	5.45	5.82
Metabolizable energy kcal/g3	4.15	3.85	3.94	4.17

¹DC, deboned chicken diet; CBPM, chicken byproduct meal diet; WGM, wheat gluten meal diet; CGM, corn gluten meal diet.

²Nitrogen-free extract, % = 100 – (acid-hydrolyzed fat {%} + crude protein {%} + ash {%} + TDF {%}).

³Metabolizable energy = 3.5 kcal/g × crude protein (%) + 8.5 kcal/g × acid-hydrolyzed fat (%) + 3.5 kcal/g nitrogen-free extract (%).

Table 2.

Indispensable and dispensable amino acid (AA) concentrations (% DM) of experimental diets differing in protein source¹

	DC	CBPM	WGM	CGM	AAFCO2
Indispensable
Arginine	2.38	2.33	1.41	1.31	0.51
Histidine	1.01	0.79	0.82	0.81	0.19
Isoleucine	1.73	1.48	1.49	1.68	0.38
Leucine	2.85	2.57	2.78	5.98	0.68
Lysine	2.84	2.18	0.86	0.89	0.63
Methionine	0.91	0.76	0.65	0.85	0.33
Phenylalanine	1.57	1.50	1.97	2.48	0.45
Threonine	1.59	1.40	1.05	1.35	0.48
Tryptophan	0.40	0.32	0.64	0.19	0.16
Valine	1.97	1.84	1.67	1.93	0.49
Dispensable
Alanine	2.30	2.30	1.19	3.30	---
Aspartic Acid	3.71	3.31	1.76	2.72	---
Cysteine	0.38	0.41	0.74	0.61	---
Glutamic Acid	5.76	5.71	12.91	8.40	---
Glycine	2.44	3.14	1.46	1.23	---
Hydroxylysine	0.21	0.24	0.11	0.14	---
Hydroxyproline	0.62	0.94	0.03	0.04	---
Lanthionine	0.01	0.01	0.00	0.00	---
Ornithine	0.06	0.07	0.02	0.04	---
Proline	1.94	2.45	4.39	3.55	---
Serine	1.38	1.40	1.69	1.86	---
Taurine	0.21	0.35	0.19	0.18	---
Tyrosine	1.02	0.94	1.13	1.58	---

¹DC, deboned chicken diet; CBPM, chicken byproduct meal diet; WGM, wheat gluten meal diet; CGM, corn gluten meal diet.

²AAFCO nutrient profiles for adult dogs at maintenance.

Fecal sample collection, scoring, and analysis

From day 23 to day 27, total feces were collected from the pen floor, weighed, composited, and frozen at −20 °C until analyses. All fecal samples collected during the collection phase were scored according to the following scale: 1 = hard, dry pellets, small hard mass; 2 = hard, formed, dry stool; remains firm and soft; 3 = soft, formed, and moist stool, retains shape; 4 = soft, unformed stool, assumes shape of container; 5 = watery, liquid that can be poured. On the first day of the collection phase, one fresh fecal sample (within 15 min of defecation) was collected for measurement of pH, moisture content, microbiota populations, and metabolite concentrations. Fecal pH was measured immediately using an AP10 pH meter (Denver Instrument, Bohemia, NY) equipped with a Beckman Electrode (Beckman Instruments, Inc., Fullerton, CA), and then aliquots were collected. Fecal aliquots for analysis of phenols and indoles were frozen at −20 °C immediately after collection. One aliquot was collected and placed in 2 N hydrochloric acid for ammonia, short-chain fatty acid (SCFA), and BCFA analyses. An additional aliquot was used for fresh fecal dry matter (DM) determination. Finally, three to four aliquots of fresh feces were collected for microbiota analysis. These samples were immediately transferred to sterile cryogenic vials (Nalgene, Rochester, NY), quickly frozen in dry ice, and stored at −80 °C until analysis.

Fecal and dietary chemical analysis

Composited fecal samples were first dried at 55 °C in a forced-air oven. Dietary treatments and dry feces were then ground in a Wiley mill (model 4, Thomas Scientific, Swedesboro, NJ) through a 2-mm screen. Diet and fecal samples were analyzed for DM and ash according to AOAC (2006; methods 934.01 and 942.05), and organic matter calculated. Crude protein (CP) of the diet and feces was calculated using a 6.25 conversion factor from total nitrogen values measured using a Leco (FP2000 and Tru-Mac) according to AOAC (2006; method 992.15). Total lipid content (acid-hydrolyzed fat) of diet and fecal samples was determined according to the methods of the American Association of Cereal Chemists (1983) and Budde (1952). Total dietary fiber of the diets was determined according to AOAC (method 991.43). Gross energy of the diet and fecal samples was measured using an oxygen bomb calorimeter (model 6200, Parr Instruments, Moline, IL). Digestible energy was determined by subtracting the gross energy of feces from the gross energy of the food consumed. Fecal SCFA and BCFA concentrations were determined by gas chromatography according to Erwin et al. (1961) using a gas chromatograph (Hewlett-Packard 5890A series II, Palo Alto, CA) and a glass column (180 cm × 4 mm i.d.) packed with 10% SP-1200/1% H3PO4 on 80/100+ mesh Chromosorb WAW (Supelco Inc., Bellefonte, PA). Nitrogen was the carrier gas with a flow rate of 75 mL/min. Oven, detector, and injector temperatures were 125, 175, and 180 °C, respectively. Fecal ammonia concentrations were determined according to the method of Chaney and Marbach (1962). Fecal phenol and indole concentrations were determined using gas chromatography according to the methods described by Flickinger et al. (2003) analysis.

Calculations

ATTD values were calculated using the equation as follows: ([nutrient intake in grams per dayfecal output in grams per day]/nutrient intake in grams per day) × 100.

Fecal DNA extraction, MiSeq Illumina sequencing of 16S amplicons, and microbiota analysis

Total DNA from fecal samples was extracted using Mo-Bio PowerSoil kits (MO BIO Laboratories, Inc., Carlsbad, CA). Concentrations of extracted DNA were quantified using a Qubit 3.0 Fluorometer (Life Technologies, Grand Island, NY). 16S rRNA gene amplicons were generated using a Fluidigm Access Array (Fluidigm Corporation, South San Francisco, CA) in combination with a Roche High Fidelity Fast Start Kit (Roche, Indianapolis, IN). The primers 515F (5ʹ-GTGCCAGCMGCCGCGGTAA-3ʹ) and 806R (5ʹ-GGACTACHVGGGTWTCTAAT-3ʹ) that target a 252 bp-fragment of the V4 region of the 16S rRNA gene were used for amplification (primers synthesized by IDT Corp., Coralville, IA) (Caporaso et al., 2012). CS1 forward tag and CS2 reverse tag were added according to the Fluidigm protocol. Quality of the amplicons was assessed using a Fragment Analyzer (Advanced Analytics, Ames, IA) to confirm amplicon regions and sizes. A DNA pool was generated by combining equimolar amounts of the amplicons from each sample. The pooled samples were then size selected on a 2% agarose E-gel (Life Technologies, Grand Island, NY) and extracted using a Qiagen gel purification kit (Qiagen, Valencia, CA). Cleaned size-selected pooled products were run on an Agilent Bioanalyzer to confirm appropriate profile and average size. Illumina sequencing was performed on a MiSeq using v3 reagents (Illumina, Inc., San Diego, CA) at the W. M. Keck Center for Biotechnology at the University of Illinois.

Bioinformatics for assessing fecal microbial communities

Forward reads were trimmed using the FASTX-Toolkit (version 0.0.14) and QIIME 2.2020.8 (Bolyen et al., 2019) was used to process the resulting sequence data. High-quality (quality value ≥ 20) sequence data derived from the ­sequencing process were demultiplexed. Data were then denoised and assembled into amplicon sequence variants using DADA2 (Callahan et al., 2016). Sequences were clustered into operational taxonomic units (OTU) using UCLUST through an open-reference OTU picking strategy against the Silva 138 reference database (Quast et al., 2013) with a 99% similarity threshold. An even sampling depth (41,369 sequences per sample) was used for assessing alpha- and beta-diversity measures. Beta-diversity was assessed using weighted and unweighted UniFrac (Lozupone and Knight, 2005) distance measures and presented using principal coordinates analysis (PCoA) plots.

Blood sample collection and analysis

On the final day of each experimental period, fasted blood samples were collected via jugular puncture for serum chemistry, hematology, and gene expression analysis. Samples were immediately transferred to appropriate vacutainer tubes, with some going into BD Vacutainer Plus plastic whole blood tubes with K₂EDTA additive (Becton, Dickinson and Company, Franklin Lakes, NJ), some going into BD Vacutainer Plus plastic serum tubes with clot activator and gel for serum separation (Becton, Dickinson and Company), and some going into PAXgene Blood Tubes (#762165; Qiagen, Valencia, CA). The blood tube for serum isolation was centrifuged at 1,300 × g at 4 °C for 10 min (Beckman CS-6R centrifuge; Beckman Coulter, Inc., Brea, CA). Serum was collected and transported to the University of Illinois Veterinary Medicine Diagnostics Laboratory for serum chemistry analysis. K₂EDTA tubes were cooled (but not frozen) and transported to the University of Illinois Veterinary Medicine Diagnostics Laboratory for hematology analyses.

Total RNA from blood cells was isolated using a PAXgene Blood RNA Kit (#762331; Qiagen, Valencia, CA). RNA concentrations were determined using an ND-1000 spectrophotometer (Nanodrop Technologies, Wilmington, DE). cDNA was synthesized using SuperScript IV reverse transcriptase (Invitrogen, Carlsbad, CA). Gene expression was measured by real-time two-step reverse transcriptase-polymerase chain reaction (RT-PCR) using an Applied Biosystems 7900HT real-time PCR system (Applied Biosystems, Waltham, MA) and was carried out with SYBR Green chemistry (Bio-Rad Laboratories, Hercules, CA) in a QuantStudio 7 instrument (Thermo Fisher Scientific, Waltham, MA) using validated forward and reverse primers (Bio-Rad Laboratories, Hercules, CA). The genes of interest included mammalian target of rapamycin (mTOR, UniqueAssayID: qCfaCID0024417), insulin-like growth factor-1 (IGF-1, UniqueAssayID: qCfaCID0035607), matrix metallopeptidase-3 (MMP-3, UniqueAssayID: qCfaCED0026432), sterol regulatory element-binding transcription factor-1 (SREBP-1, UniqueAssayID: qCfaCED0038260), ribosomal protein S6 kinase A5 (RPS6KA5, UniqueAssayID: qCfaCID0024274), peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α, UniqueAssayID: qCfaCED0028716), heat shock protein (HSP)-A1 (HSP-A1, UniqueAssayID: qCfaCED0035841), and heat shock protein-90AA1 (HSP-90AA1, UniqueAssayID: qCfaCED0026027). All gene expression data were analyzed using the 2^-ΔΔCt method, represented as gene expression relative to the housekeeping gene (RPS5, UniqueAssayID: qCfaCED0028510).

Statistical analysis

All data were analyzed using the Mixed Models procedure of SAS (version 9.4; SAS Institute, Cary, NC). Data normality was confirmed using the univariate procedure and Shapiro-Wilk statistic, with log transformation being used when normal distribution was lacking. If after the logarithmic transformation of the data, the data did not reach normality, the data were analyzed using the npar1way procedure and Wilcoxon statistic. P < 0.05 was considered significant.

Results

Dogs fed the plant-based diets, WGM and CGM, had lower (P < 0.001) food intakes than those fed DC or CBPM (Table 3). The difference in food intake was unexpected, as all test diets were formulated to be isocaloric. Slight changes in BW over time impacted food offerings, resulting in this difference. Fecal output (g/d, as-is and DM) was higher (P < 0.0001) in dogs fed CBPM than those fed the other three diets. Fecal output (g/d, as-is) was also higher (P < 0.0001) in dogs fed the DC or CGM diets than those fed the WGM diet. DM ATTD was lower (P < 0.0001) in dogs fed CBPM or CGM than those fed DC or WGM. Dogs fed CBPM had lower (P < 0.0001) OM and energy ATTD than those fed the other three diets. CP ATTD was lower (P < 0.0001) in dogs fed CBPM than those fed the other three diets. CP ATTD was also lower (P < 0.0001) in dogs fed DC or CGM than those fed WGM. Acid-hydrolyzed fat ATTD did not differ among treatments, with digestibilities >95% in all dietary treatments.

Table 3.

Food intake and fecal output of healthy adult dogs and apparent total tract macronutrient digestibility of experimental diets differing in protein source¹

Item	DC	CBPM	WGM	CGM	SEM	P-value
Food Intake
g food/d (as-is)	161.17a	157.58a	150.33b	145.67b	3.9891	<0.0001
g food/d (DM basis)	142.52a	142.25a	134.70b	135.24b	3.6043	0.0009
Fecal Output
g feces/d (as-is)	59.81b	75.41a	49.54c	61.75b	3.6103	<0.0001
g feces/d (DM basis)	20.30b	26.24a	19.65b	20.83b	1.0268	<0.0001
Digestibility2
Dry matter	85.71a	81.64b	85.44a	84.57b	0.5609	<0.0001
Organic matter	89.7a	85.75b	89.96a	88.33a	0.4538	<0.0001
Crude protein	89.85b	82.59c	93.82a	90.07b	0.5534	<0.0001
Acid-hydrolyzed fat	96.61	96.06	96.01	95.85	0.4029	0.3824
Energy	90.6a	86.76b	91.01a	89.49a	0.4071	<0.0001

¹DC, deboned chicken diet; CBPM, chicken byproduct meal diet; WGM, wheat gluten meal diet; CGM, corn gluten meal diet.

²Digestibility data measured from total feces collected over five consecutive days.

^abcMean values within a row with unlike superscript letters differ (P < 0.05).

Fecal scores were higher (P < 0.01; looser stools) in dogs fed DC or CBPM than those fed WGM or CGM (Table 4), but all remained within an ideal range. Fecal pH was not different (P = 0.07) among diets, but fecal DM was higher (P < 0.0001) in dogs fed WGM than those fed the other three diets. Fecal indole concentrations were higher (P < 0.05) in dogs fed CBPM than those fed WGM, but phenol and total phenol and indole concentrations were not different. Fecal total SCFA concentrations were higher (P < 0.05) in dogs fed DC than those fed CGM, but individual SCFA (i.e., acetate; propionate; butyrate) were not different. Fecal total BCFA concentrations were higher (P = 0.0002) in dogs fed DC or CBPM than those fed WGM. Fecal isobutyrate concentrations were higher (P < 0.0001) in dogs fed CBPM than those fed WGM or CGM. Fecal isobutyrate concentrations were also higher (P < 0.0001) in dogs fed DC than those fed WGM. Fecal isovalerate concentrations were lower (P = 0.0004) in dogs fed WGM than those fed the other 3 diets. Fecal concentrations were higher (P < 0.0001) in dogs fed DC or CBPM than those fed WGM or CGM.

Table 4.

Fecal characteristics and metabolites of healthy adult dogs consuming experimental diets differing in protein source¹

Items	DC	CBPM	WGM	CGM	SEM	P-value
Fecal scores2	2.75a	2.75a	2.25b	2.29b	0.1229	0.0010
pH	6.71	6.93	6.75	6.32	0.1615	0.0695
Dry matter (%)	31.1b	32.3b	35.5a	32.1b	0.7036	<0.0001
	---------- µmol/g (DMB) ----------
Total phenol and indole3	3.52	3.82	2.93	2.82	0.3603	0.0560
Phenol	0.46	0.28	0.39	0.02	0.1384	0.2182
Indole	3.06ab	3.54a	2.54b	2.80ab	0.3300	0.0232
Total SCFA3	422.0a	387.5ab	355.3ab	346.1b	24.957	0.0309
Acetate	238.4	228.1	194.0	195.5	14.850	0.1754
Propionate	119.5	101.5	108	94.7	7.8556	0.0566
Butyrate	64.1	57.8	53.3	55.9	4.9497	0.1966
Total BCFA3	24.0a	28.0a	17.7b	22.8ab	1.9649	0.0002
Isobutyrate	9.04ab	10.69a	7.08c	7.31bc	0.6817	<0.0001
Isovalerate	13.9a	16.2a	9.55b	14.3a	1.2461	0.0004
Valerate	1.13	1.18	1.08	1.25	0.1846	0.8608
Ammonia	207.0a	227.2a	123.6b	162.0b	12.562	<0.0001

¹DC, deboned chicken diet; CBPM, chicken byproduct meal diet; WGM, wheat gluten meal diet; CGM, corn gluten meal diet.

²Fecal scores: 1 = hard, dry pellets; small hard mass; 2 = hard formed, dry stool; remains firm and soft; 3 = soft, formed and moist stool, retains shape; 4 = soft, unformed stool; assumes shape of container; 5 = watery, liquid that can be poured.

³Total short-chain fatty acids (SCFA) = acetate + propionate + butyrate; total branched-chain fatty acids (BCFA) = valerate + isovalerate + isobutyrate; total phenol and indole = phenol + indole.

^abcMean values within a row with unlike superscript letters differ (p<0.05).

Fecal microbiota were shifted among dietary treatments. Alpha-diversity of fecal microbial communities (Figure 1), measured by observed OTU, was higher (P < 0.05) in dogs fed CBPM than those fed DC or WGM. Alpha-diversity analysis assessed by Faith’s PD was higher (P < 0.05) in dogs fed CBPM than those fed WGM. Alpha-diversity analysis assessed by the Shannon diversity index was higher (P < 0.05) in dogs fed CBPM than those fed DC. Although beta-diversity represented by PCoA plots of unweighted (Figure 2) UniFrac distances were not different among diets, the PCoA plots of weighted (Figure 3) UniFrac distances revealed that fecal microbial populations of dogs fed WG or CGM tended to shift away from those fed DC or CBPM. In terms of specific fecal microbiota taxa, the prominent phyla included Firmicutes, Fusobacteria, and Proteobacteria (Table 5). Dogs fed CGM had a higher (P = 0.003) relative abundance of fecal Firmicutes and lower (P < 0.0001) relative abundance of fecal Fusobacteria than dogs fed DC or CBPM. Dogs fed CBPM had higher (P < 0.001) relative abundance of fecal Proteobacteria than those fed WGM and CGM. Relative abundances of fecal unclassified Lachnospiraceae and Blautia were higher (P < 0.0001) in dogs fed WGM than those fed the other three diets. The relative abundance of fecal uncultured Lachnospiraceae was higher (P = 0.007) in dogs fed DC or CGM than those fed CBPM. The relative abundances of fecal Faecalibacterium and Peptoclostridium were higher (P < 0.05) in dogs fed CBPM than those fed WGM or CGM. The relative abundance of fecal Romboutsia was higher (P = 0.002) in dogs fed CGM than those fed DC or CBPM. The relative abundance of fecal Megamonas was higher (P = 0.022) in dogs fed WGM than those fed CBPM. The relative abundance of fecal Fusobacterium was higher (P < 0.0001) in dogs fed DC or CBPM than those fed CGM. Lastly, the relative abundance of fecal Parasutterella was higher (P < 0.0001) in dogs fed DC or CBPM than those fed WGM or CGM.

Figure 1.

Fecal alpha diversity indices of healthy adult dogs consuming experimental diets differing in protein source (DC, fresh deboned, dried, and spray-dried chicken diet; CBPM, chicken byproduct meal diet; WGM, wheat gluten meal diet; CGM, corn gluten meal diet).

Figure 2.

Unweighted principal coordinate analysis (PCoA) plot of healthy adult dogs consuming experimental diets differing in protein source (DC, fresh deboned, dried, and spray-dried chicken diet; CBPM, chicken byproduct meal diet; WGM, wheat gluten meal diet; CGM, corn gluten meal diet).

Figure 3.

Weighted principal coordinate analysis (PCoA) plot of healthy adult dogs consuming experimental diets differing in protein source (DC, fresh deboned, dried, and spray-dried chicken diet; CBPM, chicken byproduct meal diet; WGM, wheat gluten meal diet; CGM, corn gluten meal diet).

Table 5.

Fecal bacterial phyla and genera (relative abundance, %) of healthy adult dogs consuming experimental diets differing in protein source¹

Phyla	Genus	DC	CBPM	WGM	CGM	SEM	P-value
Firmicutes		49.4b	50.4b	56.4ab	65.2a	3.46	0.003
	Unclassified Lachnospiraceae	1.62b	1.78b	3.06a	1.60b	0.36	<0.0001
	Blautia	3.39b	2.93b	7.28a	3.94b	0.75	<0.0001
	Uncultured Lachnospiraceae	1.73a	1.04b	1.65ab	1.87a	0.34	0.007
	Faecalibacterium	6.21ab	7.85a	3.47b	3.54b	1.44	0.017
	Peptoclostridium	9.18ab	10.3a	6.25b	6.65b	1.15	0.003
	Romboutsia	1.14b	1.03b	1.83ab	2.61a	0.41	0.002
	Megamonas	3.37ab	1.15b	3.73a	2.68ab	1.07	0.022
	Allobaculum	3.99	12.8	5.57	4.04	2.92	0.1467
	Uncultured Erysipelotrichaceae	1.01	9.09	3.14	7.06	2.39	0.0658
	Lactobacillus	5.02	4.64	4.35	5.24	3.01	0.1300
Fusobacteriota		27.5a	23.1a	21.2ab	13.6b	2.78	<0.0001
	Fusobacterium	27.5a	23.2a	21.2ab	13.6b	2.78	<0.0001
Proteobacteria		5.53ab	6.83a	3.33b	4.12b	0.63	0.0010
	Parasutterella	2.85a	2.62a	1.25b	1.66b	0.75	<0.0001
Actinobacteriota		1.94	2.15	3.02	1.78	0.491	0.1236
Bacteroidota		15.6	17.43	14.0	17.2	1.909	0.4245
	Bacteroides	9.43	10.19	7.50	13.0	1.56	0.0995
	Alloprevotella	3.01	0.94	1.75	2.23	0.50	0.2900

¹DC, deboned chicken diet; CBPM, chicken byproduct meal diet; WGM, wheat gluten meal diet; CGM, corn gluten meal diet.

^abcMean values within a row with unlike superscript letters differ (p<0.05).

All serum chemistry markers were within the reference ranges except for creatinine concentrations, which were below normal reference range in dogs fed WGM (Table 6). Serum creatinine concentrations were lower (P = 0.0004) in dogs fed WGM than those fed DC or CBPM. Serum creatinine concentrations were also lower (P = 0.0004) in dogs fed CGM than those fed DC. Serum BUN:creatinine ratio was higher (P < 0.05) in dogs fed WGM than those fed DC. Serum chloride concentrations were higher (P < 0.05) in dogs fed WGM than those fed CBPM. Serum bilirubin concentrations were higher (P < 0.05) in dogs fed CBPM than those fed DC. Serum creatine phosphokinase concentrations were higher (P < 0.05) in dogs fed DC than those fed CGM. Hematology values were within the reference ranges for dogs in all treatments (Table 7). However, blood mean corpuscular hemoglobin concentrations were higher (P < 0.05) in dogs fed DC than those fed WGM. Also, blood mean platelet volume was lower (P < 0.01) in dogs fed CBPM than those fed the other three diets. Expression for all measured genes in whole blood was not affected by diet (Table 8).

Table 6.

Serum chemistry of healthy adult dogs consuming experimental diets differing in protein source¹

Item	Reference Range	DC	CBPM	WGM	CGM	SEM	P-value
Creatinine, mg/dL	0.5–1.5	0.63a	0.57ab	0.46c	0.51bc	0.055	0.0004
Blood urea nitrogen (BUN), mg/dL	6–30	15.3	16.0	15.6	15.8	1.295	0.9014
BUN:creatinine ratio		25.6b	30.0ab	35.1a	33.2ab	2.848	0.0145
Total protein, g/dL	5.1–7.0	5.99	6.05	5.96	5.96	0.098	0.5856
Albumin, g/dL	2.5–3.8	3.29	3.28	3.28	3.27	0.067	0.9412
Globulin, g/dL	2.7–4.4	2.70	2.78	2.68	2.69	0.075	0.3485
Albumin:globulin ratio	0.6–1.1	1.23	1.19	1.24	1.23	0.047	0.4051
Ca, mg/dL	7.6–11.4	9.88	10.07	9.90	9.83	0.105	0.1066
P, mg/dL	2.7–5.2	3.13	3.11	2.81	3.08	0.193	0.4023
Na, mmol/L	141–152	144.8	144.9	145.4	145.1	0.656	0.9847
K, mmol/L	3.9–5.5	3.95	3.92	4.00	3.99	0.068	0.6909
Na:K ratio	28–36	36.7	37.2	36.5	36.5	0.634	0.7889
Cl, mmol/L	107–118	110.5ab	109.4b	111.4a	110.8ab	0.729	0.0282
Glucose, mg/dL	68–126	91.1	87.2	88.0	90.1	2.093	0.3926
Alkaline phosphatase, U/L	7–92	36.8	33.8	49.8	38.2	11.42	0.6267
Corticosteroid-induced ALP, U/L	0–40	8.75	7.50	19.8	11.6	9.677	0.4102
Alanine transaminase, U/L	8–65	24.7	22.6	20.7	22.5	1.795	0.1100
Gamma glutamyltransferase, U/L	0–7	2.83	3.08	3.25	3.17	0.256	0.4539
Total bilirubin, mg/dL	0.1–0.3	0.15b	0.21a	0.17ab	0.16ab	0.020	0.0251
Creatine phosphokinase, U/L	26–310	113.0a	102.7ab	91.0ab	86.1b	9.427	0.0454
Cholesterol, mg/dL	129–297	243.8	236.4	250.3	233.1	19.44	0.8940
Triglycerides, mg/dL	32–154	81.5	62.3	108.8	58.8	26.56	0.2346
Bicarbonate, mmol/L	16–24	37.4	22.3	37.1	21.8	11.89	0.2945
Anion gap	8–25	17.4	17.1	17.6	16.5	0.460	0.2718

¹DC, deboned chicken diet; CBPM, chicken byproduct meal diet; WGM, wheat gluten meal diet; CGM, corn gluten meal diet.

^abcMean values within a row with unlike superscript letters differ (p<0.05).

Table 7.

Hematology of healthy adult dogs consuming experimental diets differing in protein source¹

Item	Reference range	DC	CBPM	WGM	CGM	SEM	P-value
Red blood cells, 106/μL	5.50–8.50	7.16	7.08	7.00	7.14	0.164	0.7316
Reticulocytes, %		0.35	0.34	0.43	0.36	0.046	0.0625
Reticulocytes, μL		25,071	24,864	30,944	26,467	3,866	0.1424
Hemoglobin, g/dL	12.0–18.0	16.0	15.7	15.5	15.8	0.303	0.5272
Hematocrit, %	35.0–52.0	47.7	47.0	46.9	47.4	0.846	0.7801
Mean cell volume, fl	58.0–76.0	66.6	66.5	67.0	66.5	0.486	0.1633
MCH2, pg	20.0–25.0	22.3	22.2	22.2	22.2	0.221	0.3468
MCHC2, g/dL	33.0–38.6	33.5a	33.4ab	33.1b	33.3ab	0.175	0.0302
Mean platelet volume, fl		10.6a	10.4b	10.7a	10.6a	0.261	0.0013
Platelets, 103/μL	200–700	282.3	299.8	309.8	277.3	17.56	0.3602
White blood cell count, 103/μL	6.00–17.00	5.51	5.06	5.86	5.41	0.359	0.6118
Lymphocytes, 103/μL		1.08	1.11	1.12	1.11	0.148	0.9996
Monocytes, 103/μL		0.27	0.31	0.28	0.27	0.041	0.7366
Eosinophils, 103/μL		0.13	0.15	0.15	0.18	0.027	0.6037
Lymphocytes, %		19.0	23.1	18.7	20.5	2.039	0.0524
Monocytes, %		4.69	5.20	4.73	4.91	0.428	0.7112
Eosinophils, %		2.03	3.31	2.68	3.33	0.534	0.1496
Basophils, %		0.13	0.25	0.10	0.26	0.054	0.2203

¹DC, deboned chicken diet; CBPM, chicken byproduct meal diet; WGM, wheat gluten meal diet; CGM, corn gluten meal diet.

²MCH = mean corpuscular hemoglobin; MCHC = mean corpuscular hemoglobin concentration.

^abcMean values within a row with unlike superscript letters differ (P < 0.05).

Table 8.

Blood mRNA expression of healthy adult dogs consuming experimental diets differing in protein source¹

Gene Symbol	DC	CBPM	WGM	CGM	SEM	P-value
mTOR	1.01	1.06	1.01	1.04	0.101	0.9802
IGF-1	2.04	3.33	1.78	2.83	0.953	0.8250
MMP3	0.85	1.46	0.62	11.59	4.923	0.1265
HSPA1	1.12	1.04	1.23	1.54	0.174	0.1733
HSP90AA1	1.02	1.05	1.08	0.98	0.104	0.8022
PGC-1α	0.92	1.21	1.22	2.33	0.522	0.3371
RPS6KA5	1.20	1.14	1.14	0.98	0.148	0.5347
SREBP1	1.22	1.07	1.15	2.94	0.747	0.7577

¹DC, deboned chicken diet; CBPM, chicken byproduct meal diet; WGM, wheat gluten meal diet; CGM, corn gluten meal diet.

²Statistics were conducted using ∆∆Ct values to generate P-values; data are reported as fold change in relation to a housekeeping gene and CBPM (2^−∆∆Ct).

³ mTOR, mammalian target of rapamycin; IGF-1, insulin-like growth factor-1; MMP3, matrix metallopeptidase-3; HSPA1, heat shock protein-A1; HSP90AA1, heat shock protein-90AA1; PGC-1α, peroxisome proliferator-activated receptor gamma coactivator 1-alpha; RPS6KA5, ribosomal protein S6 kinase A5; SREBP1, sterol regulatory element-binding transcription factor-1.

Discussion

The pet food industry is continually searching for and testing a variety of protein sources not only to meet the nutritional needs of pets, but also to align with pet owner preferences and beliefs. There is heightened consumer awareness of nutrition and health, and demand for sustainable and natural feeding approaches. In 2020, 41% of dog owners bought “premium” dog foods (Phillips-Donaldson, 2022). “Premium” has no regulatory definition, but in that survey, that term was used and compared to “basic/generic” food that typically contain by-product ingredients, or artificial colors and preservatives. Animal-based protein sources are typically the leading ingredients on premium pet food labels. Some consumers, however, refuse to feed ingredients containing “by-product” in their title because they seem unfit for consumption. Other pet owners are concerned about the sustainability of their pet food choices and may select plant-based options. While it is important that the pet food industry provide options to meet diverse customer demands, it is essential that diets are complete and balanced.

In the current study, two animal-based diets and two plant-based diets were evaluated for nutritional value and impact on the dog. All four experimental diets were formulated to contain similar ingredients and nutrient predictions, except for the primary source of protein and AA. The final diets differed slightly in regard to nutrient composition, but were fairly similar. The dietary protein concentrations ranged from 39.8% to 40.7%, exceeding the AAFCO recommendation for adult maintenance of 18%. Due the rising trend of humanizing pets, consumers are increasingly seeking out premium products made with high-quality, protein-rich ingredients. As a result, premium pet food products tend to contain higher protein and lower carbohydrate concentrations. The high-protein diets tested in the current study contain CP concentrations that are similar to many commercially available diets in this market segment.

The CBPM, WGM, and CGM diets were supplemented with taurine which is a sulfur-containing beta-sulfonic acid present in high concentrations in cardiac and skeletal muscle tissues, and lacking in plants (McCuster, 2014). In the DC diet, no AA supplements were needed to create a complete and balanced formula. Both plant-based protein diets were supplemented with L-lysine to meet the recommended AAFCO minimums for adult dogs at maintenance. Previous research published by Reilly et al. (2021) reported that CGM had low DIAAS-like values for tryptophan (47.3%), with it being the first limiting AA of that ingredient. In the current study, the CGM diet contained a low tryptophan concentration (0.19%), which was just slightly above the AAFCO minimum (0.16%).

In the current study, all dietary treatments were considered well-digested by dogs, above the AAFCO and FEDIAF (FEDIAF, 2021) apparent protein digestibility minimum recommendations of 80%. Comparing just the animal-based diets in the current study, CP digestibility of the DC diet (89.85%) was 7 percentage units more digestible than the CBPM diet (82.59%). The plant-based diets, WGM and CGM, both had high CP digestibilities (93.82% and 90.07%, respectively). The high CP digestibilities of all test diets suggest the use of high-quality ingredients and proper processing conditions. The high CP concentrations present in the diets may have also contributed to such high digestibility values. Both endogenous (e.g., digestive enzymes, mucin, sloughed cells) and exogenous (e.g., dietary) sources of nitrogen contribute to the total amount of nitrogen excreted in the feces (Adeola et al., 2016). As dietary CP/nitrogen concentrations increase, the proportion of nitrogen loss coming from endogenous sources decreases, often resulting in higher ATTD values (Adeola et al., 2016; Yang et al., 2021).

The high CP digestibility of plant-based proteins was likely due to the prior processing needed to separate these protein fractions from the other grain components and using the highly digestible crystalline AA to complete the diets. Both plant proteins used, WGM and CGM, are processed, removing the majority of starch and fiber, and concentrating protein. CGM is produced during the wet-milling of corn, which separates the corn kernel into starch, protein, and dietary fiber fractions (Moniruzzaman et al., 2020). Other plant-based proteins, such as soy protein concentrates and soy protein isolates, have also been shown to perform well in dogs (Clapper et al., 2001; Beloshapka et al., 2016). Based on their cost, consistency, and the positive results reported in the current study and that of past studies, consumers and pet food professionals should continue to consider the use of plant-based proteins in their formulations.

Gross energy digestibility was lower in the CBPM diet, which was primarily due to the lower protein digestibility. Urrego et al. (2017) evaluated poultry meal- and wheat gluten-based diets in brachycephalic dogs. The wheat gluten and poultry meal diets in that study had apparent CP digestibilities of 88% and 82.2%, respectively, which is similar to what was observed in the current study. Fecal scores of dogs fed the animal-based diets were higher than those fed the plant-based diets, but mean scores for all diets were considered ideal (between a score of 2 to 3).

Protein quality is dependent on the AA composition, digestibility, and bioavailability of the ingredient or diet. Apparent total tract CP digestibility is not a true representation of what the host digests because of the microbial metabolism that takes place in the hindgut, as well as endogenous protein losses that interfere with the calculations. Ileal-cannulated dogs or the cecectomized rooster assay provide accurate measures of CP and AA digestibility. Oba et al. (2019) evaluated the true nutrient digestibility and true metabolizable energy of chicken-based ingredients using a precision-based cecectomized rooster assay and reported that chicken meal had a lower DM digestibility (60.1%) compared with a low processed, steamed chicken (76.5%). This difference was thought to have been due to prior processing of the protein sources. CBPM goes through a high heat rendering process that may decrease protein quality. Animal by-product meal ash content and protein quality can vary greatly by temperature at which the original material was processed, as well as the starting composition of the meat and tissues used in the rendering process. Some by-product meals may also contain high amounts of connective tissues, which have constituents that analyze as fiber and are poorly digested by animals (Johnson et al., 1998). This further illustrates the need for pet food formulators to gain insight into ingredients being used for diet formulation.

All dogs remained healthy throughout the study and most serum chemistry and hematology values of dogs fed all diets were within references ranges for adult dogs, except for creatinine. Creatinine was slightly lower than reference range values for dogs fed the WGM diet, but no signs of clinical abnormality were observed during the study. The low creatinine concentrations in dogs fed the WGM diet lead to higher BUN:creatinine ratios compared with dogs fed the other diets. Blood Cl, bilirubin, and creatine phosphokinase differed among treatments, but still remained in healthy references ranges for dogs.

Final utilization of nutrients in the body is moderated by microbiota in the colon, where fermentation occurs. The gastrointestinal microbiome is a complex ecosystem that impacts host health. The production of fecal metabolites by microbiota can be influenced by substances entering the colon. Fecal SCFA are produced primarily from carbohydrate fermentation and are an important source of energy for colonocytes (Morrison and Preston, 2016). Higher SCFA concentrations are generally considered beneficial to the host. Fecal total SCFA concentrations were highest in dogs fed the DC diet and lowest in dogs fed the CGM diet. Because the dietary fiber concentrations were similar across treatments, the reason for this difference is unknown.

Fecal putrefactive compounds, namely ammonia and BCFA, are indicators of increased protein fermentation. Proteolytic fermentation takes place mainly in the distal large intestine, where ammonia is produced from the deamination of AA and hydrolysis of urea, whereas phenols are produced due to decarboxylation of AA (Jha et al., 2019). Ammonia can potentially have a negative impact on intestinal health and can contribute to fecal odor (Lee et al., 2022). Both animal-based diets in the current study resulted in higher fecal ammonia concentrations than the plant-based diets. Dogs fed the CBPM diet had higher fecal indole concentrations compared with those fed WGM. Dogs fed the WGM diet had lower total BCFA concentrations. Due to the lower CP digestibility in the CBPM diet, more protein would have likely reached the colon, increasing proteolytic fermentation by gut microbiota. Previous research reported similar results, with high-protein poultry meal diets resulting in greater fecal concentrations of ammonia, BCFA, and indole compared with a high-protein wheat gluten diet fed to dogs (Nery et al., 2012). Beloshapka et al. (2016) also reported lower fecal BCFA, ammonia, phenol, and indole concentrations in dogs fed bioprocessed soy protein. The fecal metabolite profiles reported in the current study and that of other studies provide another potential benefit of including plant-based proteins in dog foods.

Firmicutes, Bacteroidetes, Proteobacteria, Fusobacteria, and Actinobacteria are the predominant microbial phyla in the gut of dogs (Deng and Swanson, 2015; Rodrigues Hoffmann et al., 2016; Wernimont et al., 2020). Changes to the fecal microbiome have been shown to occur quickly in response to dietary interventions. In the current study, three bacterial phyla, Firmicutes, Fusobacteriota, and Proteobacteria, shifted due to diet. Vegetable fiber content has been reported to increase the overall abundance of Firmicutes and decrease the abundance of Fusobacteria and Proteobacteria (Middelbos et al, 2010; Pilla and Suchodolski, 2020), however, we observed a similar shift in dogs fed the CGM diet that had the least amount of dietary fiber.

It is reported that Megamonas produces enzymes that result in ammonia production (Polansky et al., 2016). However, Megamonas was measured in highest abundance in dogs fed WGM, but had the lowest concentrations of fecal ammonia. Megamonas is also a key SCFA-producing bacteria however, the current study SCFA concentrations did not change in the same direction as change in Megamonas relative abundance. Lee et al. (2022) also reported that increases of relative abundance of Megamonas did not increase SCFA concentrations in healthy dogs, so it is possible unknown factors may be involved. Previous research evaluation of a raw diet, which was high in animal protein, reported high abundances of Proteobacteria and Fusobacteria (Sandri et al., 2017). Proteobacteria have been reported to be more abundant in dogs fed high-protein diets and be more variable among dogs than cats (Moon et al., 2018). Dogs fed the CBPM diet had a higher abundance of Proteobacteria, which could be due to a greater amount of protein that entered the colon. Parasutterella has been shown to play a role in bile acid maintenance and cholesterol metabolism in mice (Ju et al., 2019). CGM fed dogs had the lowest abundance of Fusobacteria. Fusobacterium are associated with IBD and colorectal cancer in humans, but have not been associated with those conditions in dogs. In fact, it is usually the opposite, with healthy dogs and animals fed high-protein diets having high relative abundances of Fusobacterium (Félix et al., 2022). Unexpectedly, bacterial diversity was lower in dogs fed the DC diet than those fed the CBPM diet. Although reduced bacterial diversity is often associated with gastrointestinal disease (Xenoulis et al., 2008; Guard et al., 2015), stool quality was ideal, and no signs of disease were observed during this study. What is deemed “normal” for the canine gut microbiota is highly reliant on the diet being consumed at the time of collection and demonstrates great flexibility (Do et al., 2021). Thus, shifts in microbial communities should be attributed to differences in protein sources, and not as an indication of gut dysbiosis because healthy dogs were used in this study, and these animals remained healthy and without any signs of gastrointestinal intolerance or discomfort in response to experimental diets.

In conclusion, all diets tested in this study were well tolerated and dogs remained healthy when fed both the animal- and plant-based diets. The chicken byproduct meal diet consistently had the lowest apparent total tract energy and nutrient digestibilities, including dry matter, organic matter, and crude protein, and resulted in the highest fecal output. In contrast, the wheat gluten meal diet had the highest apparent total tract crude protein digestibility and resulted in the lowest fecal output. Because the deboned chicken diet was the only diet that did not require AA supplementation and was highly digestible, it had the highest quality protein diet in this study. Although L-lysine supplementation was necessary to make sure that the plant-based diets were complete and balanced, they performed very well, having high nutrient digestibilities and resulting in lower fecal concentrations of proteolytic fermentation metabolites. Three bacterial phyla and nine bacterial genera in fecal samples were shifted among treatments, but fecal scores were maintained by all animals throughout the study so their impact on health are unknown. Because high-protein diets were tested in this study, there was an abundance of AA and impacts of protein quality were likely difficult to measure. Therefore, research on diets containing lower, more moderate concentrations of plant-based versus animal-based protein may be further investigated to more effectively evaluate how protein quality and AA concentrations impact canine health.

Glossary

Abbreviations

AA

amino acids

AAFCO

Association of American Feed Control Officials

ATTD

apparent total tract digestibility

BCFA

branched-chain fatty acids

BCS

body condition score

BW

body weight

CBPM

chicken byproduct meal diet

CGM

corn gluten meal diet

CP

crude protein

DC

fresh deboned, dried, and spray-dried chicken diet

DM

dry matter

FEDIAF

The European Pet Food Industry Federation; HSPA1, heat shock protein-A1

HSP90AA1

heat shock protein-90AA1

IGF-1

insulin-like growth factor-1

MMP3

matrix metallopeptidase-3

mTOR

mammalian target of rapamycin

OTU

operational taxonomic units

PCoA

principal coordinates analysis

PGC-1α

peroxisome proliferator-activated receptor gamma coactivator 1-alpha

RPS6KA5

ribosomal protein S6 kinase A5

RT-PCR

reverse transcriptase-polymerase chain reaction

SCFA

short-chain fatty acids

SREBP1

sterol regulatory element-binding transcription factor-1

WGM

wheat gluten meal diet

Conflict of Interest Statement

J.K. and C. P. are employed by Champion Petfoods. All other authors have no conflicts of interest.