Longitudinal assessment of taurine and amino acid concentrations in dogs fed a green lentil diet

Abstract

A recent association between the inclusion of pulses in canine diets and taurine deficiency has become a prevalent issue in the pet food industry. Although dogs do not currently have a nutritional requirement for taurine, taurine deficiencies that do occur can result in serious health issues, such as dilated cardiomyopathy. The objective of this study was to determine the circulating concentrations of plasma and whole blood taurine, indispensable and dispensable amino acid concentrations in the plasma, and taurine and creatinine concentrations in the urine of adult dogs fed a green lentil diet. Twelve adult, intact, female beagles were randomly assigned to a diet containing 45% green lentils (GLD) or a poultry byproduct meal diet (CON) for 90 d. Fresh urine samples were collected every 30 d and analyzed for taurine and creatinine concentrations. A blood sample was also collected every 30 d and analyzed for amino acids including taurine. Animal procedures were approved by the University of Illinois Institutional Animal Care and Use Committee. All diets were formulated to meet or exceed the nutrient requirements for adult dogs at maintenance. The concentrations of taurine in the plasma and whole blood showed no differences (P > 0.05) between dietary treatments or across time points. Similarly, no differences (P > 0.05) in plasma methionine concentrations were observed between treatments or across time points. A treatment effect (P < 0.05) showed dogs fed GLD had higher total primary fecal bile acid excretion compared with dogs fed CON. The differential abundance of fecal microbial communities showed Firmicutes as the predominant phyla in dogs fed both GLD and CON, with Bacteroidaceae, Erysipelotrichaceae, and Lactobacillaceae as predominant families in dogs fed GLD. The α-diversity of dogs fed GLD (P < 0.05) was lower than in dogs fed CON. These data suggest that the inclusion of 45% green lentil in extruded diets does not lower whole blood and plasma taurine concentrations during a 90-d period and is appropriate for use in a complete and balanced formulation for dogs.

Introduction

In July 2018, the Food and Drug Administration announced in a press release that diets containing pulses (i.e., peas, beans, and lentils) and potatoes as main ingredients are associated with the development of dilated cardiomyopathy (DCM) in dogs (FDA, Center for Veterinary Medicine, 2018). DCM is a progressive disease of the cardiac muscle, commonly caused either by genetic mutation (Ingles et al., 2005; McNair et al., 2011) or by taurine deficiency (Moise et al., 1991). Specific genetic mutations in genes encoding for proteins responsible for muscle contraction and proper function of the myocardium have been associated with DCM development (Kamisago et al., 2000; Hanson et al., 2002). Currently, there is no taurine requirement in dogs due to sufficient cysteine sulfinic acid decarboxylase activity that efficiently synthesizes taurine from its precursors, methionine and cysteine (Jacobsen et al., 1964). The association between diets containing these ingredients and taurine-deficient DCM is yet to be definitively established.

Taurine deficiency in dogs could be due to insufficient taurine synthesis from methionine and cysteine (Fascetti et al., 2003), disrupted bile acid (BA) recycling (Ko and Fascetti, 2016), or bacterial degradation (Backus et al., 2006). As reviewed in Marinangeli et al. (2017), although pulses have high concentrations of lysine, making them beneficial additions to many diets, they generally are low in methionine, a precursor for taurine synthesis, relative to lysine. Additionally, unlike most mammals that can conjugate BA with glycine or taurine, dogs conjugate BA with taurine, exclusively (Anantharaman-Barr et al., 1994; Herstad et al., 2018). Normally, BAs are recycled with 99% efficiency. However, disruption of this recycling system causes increased BA excretion in the feces, leading to increased BA synthesis in the liver, potentially depleting taurine stores in the body (Ko and Fascetti, 2016).

Because the mechanism of taurine deficiency in diets containing pulse ingredients is unknown, the objectives of this study were to determine the circulating concentrations of plasma and whole blood taurine, indispensable and dispensable amino acid concentrations in plasma, and taurine and creatinine concentrations in the urine of adult dogs fed a diet containing 45% green lentils (GLD) compared with a poultry byproduct meal control (CON). It was hypothesized that feeding a complete and balanced diet with a high inclusion of green lentils over a long-term feeding trial would not have negative health effects or detrimental decreases of taurine status in dogs, assuming that methionine is provided at sufficient amounts to fulfill the requirement of adult dogs and that taurine is not an indispensable amino acid for these animals.

Materials and Methods

Experimental design

All animal care protocols used in this study were approved by the Institutional Animal Care and Use Committee at the University of Illinois at Urbana-Champaign. All methods were performed in accordance with the U.S. Public Health Service Policy on Humane Care and Use of Laboratory Animals.

Twelve adult, intact, female Beagles (average age: 5.5 ± 1.1 yr, average weight: 12.5 ± 1.1 kg) were used in a completely randomized design. Every 30 d, a 7-mL fasted blood sample was collected for all dogs and allocated for whole blood, serum, and plasma samples. The samples were then analyzed for a serum chemistry analysis, complete blood count (CBC), whole blood taurine analysis, and plasma amino acid analysis. Serum metabolites and CBC were analyzed by the University of Illinois Veterinary Medicine Diagnostics Laboratory (Urbana, IL). Whole blood taurine and plasma amino acid profiles were analyzed by the University of California School of Veterinary Medicine Amino Acid Laboratory (Davis, CA). The dogs were housed in a temperature-controlled room at the Edward R. Madigan Laboratory at the University of Illinois at Urbana-Champaign on a 14:10 (L:D) h schedule. The dogs were individually housed in kennels (1.2 m wide by 1.8 m long) with nose-to-nose contact with dogs in adjacent runs and visual contact with all dogs in the room. The dogs were fed twice daily (0800 and 1600 hours) with ad libitum access to water.

Dogs (n = 6 per trt) were randomly assigned to the GLD or CON treatment group for 90 d (Table 1). Dogs had 15 d of diet adaptation to the CON diet prior to the start of the trial (day −15 to 0). The green lentil was included at the expense of poultry byproduct meal and rice to formulate the diets with similar nutrient profiles. Diets were formulated to be complete and balanced according to AAFCO (2018) recommended values for adult dogs at maintenance. Food intake was calculated to maintain body weight based on historical metabolizable energy requirement of each dog for maintenance. Any food refusals were measured after each meal throughout the duration of the study. Body weight and body condition score were measured weekly and intake was adjusted accordingly to maintain body weight, if necessary.

Table 1.

Ingredient composition of canine diets containing green lentils or poultry byproduct meal

	Dietary treatment1
Ingredient	CON	GLD
Green lentil	—	44.65
Poultry byproduct meal	33.50	19.15
Rice	42.96	10.00
Poultry fat	8.47	10.14
Corn	10.00	10.00
Dried beet pulp	2.50	2.50
Palatant	1.00	1.00
Ca carbonate	0.78	0.57
Dicalcium phosphate	—	1.20
Salt	0.30	0.30
Vitamin premix2	0.18	0.18
Mineral premix3	0.18	0.18
Choline chloride	0.12	0.12
Butylated hydroxytoluene (antioxidant)	0.02	0.02

¹CON, poultry byproduct meal control; GLD, green lentil diet.

²Provided per kilogram of diet: 10.8 mg copper (CuSO₄), 0.36 mg selenium (Na₂SeO₃), 150 mg zinc (ZnSO_4, ZnO), 2,562.8 IU vitamin A, 254 IU vitamin D3, and 32.1 IU vitamin E.

³Provided per kilogram of diet: 17.4 mg manganese (MnSO₄), 284.3 mg iron (FeSO₄), 17.2 mg copper (CuSO₄), 2.2 mg cobalt (CoSO₄), 166.3 mg zinc (ZnSO₄), 7.5 mg iodine (KI), and 0.2 mg selenium (Na₂SeO₃).

Chemical analyses

The chemical composition of GLD and CON was analyzed in a previous study (Reilly et al.,2021). Diets were ground through a 2-mm screen using a Wiley mill (model 4, Thomas Scientific, Swedesboro, NJ) and were analyzed for dry matter (DM), ash, organic matter (OM), acid hydrolyzed fat (AHF), crude protein (CP), gross energy (GE), and total dietary fiber (TDF). DM, ash, and OM were determined according to AOAC (2006; methods 934.01 and 942.05). Total nitrogen values were determined according to AOAC (2006; method 992.15) with CP calculated from Leco (TruMac N, Leco Corporation, St. Joseph, MI). AHF, used to measure total fat content, was analyzed according to AACC (1983) and Budde (1952). GE was analyzed through bomb calorimetry (Model 6200, Parr Instruments Co., Moline, IL). TDF was analyzed according to Prosky et al. (1992).

Amino acid analysis

A 7-mL fasted blood sample was collected via jugular venipuncture from all dogs and analyzed for serum chemistry, CBC, and a complete amino acid profile. Serum chemistry was analyzed using 3 mL of blood collected in a serum separator vacutainer tube, whereas CBC was analyzed using 1 mL of blood collected in an ethylenediaminetetraacetic acid (EDTA) vacutainer tube (Becton, Dickinson and Company, Franklin Lakes, NJ). Blood was then allocated to two lithium heparin vacutainer tubes for whole blood (1 mL) and plasma (2 mL) amino acid analysis (Becton, Dickinson and Company, Franklin Lakes, NJ). Plasma samples were centrifuged (1,300 × g at 4 °C), and supernatant was pipetted into 2-mL cryovials. Whole blood and plasma samples were stored at −20 °C until analysis. A fresh urine sample was collected within 15 min on days 30, 60, and 90. Urine samples were analyzed for taurine and creatinine concentrations. Urine samples were stored at −80 °C until analysis. Urine creatinine was analyzed by the University of Illinois Veterinary Medicine Diagnostics Laboratory (Urbana, IL).

Complete amino acids were analyzed by the University of California School of Veterinary Medicine Amino Acid Laboratory (Davis, CA) by adding 6% sulfosalicylic acid (1:1) to the samples for deproteinization. The mixture then was centrifuged at 14,000 rpm for 25 min. The supernatant was filtered through a 0.45-mm syringe drive polytetrafluoroethylene (PTFE) filter. The pH was adjusted to 2.2 prior to loading 50 mL of the mixture onto the Biochrom 30 amino acid analyzer. Whole blood samples analyzed for taurine were frozen and thawed twice to break the cells, releasing the taurine, and then processed using the amino acid analysis described previously.

BA analysis

A fresh fecal sample was collected from each dog within 15 min of defecation on days 30, 60, and 90. Fecal samples were collected into duplicate 2 mL cryovials. Fecal samples were stored at −80 °C until analysis.

BA analysis was performed according to Batta et al. (2002) to measure unconjugated cholic acid (CA), chenodeoxycholic acid (CDCA), lithocholic acid (LCA), deoxycholic acid (DCA), and ursodeoxycholic acid (UDCA). Fecal samples were lyophilized for 1 wk, pulverized using a spatula (Smart Spatula, USA Scientific, Ocala, FL) and stored at −20 °C until analysis. Aliquots of 10 to 15 mg of lyophilized feces were weighed into disposable glass centrifuge tubes (5 mL, Kimble-Chase, Rockwood TN). A total volume of 200 μL of butanol containing 0.25 mg/mL of internal standards, undecanoic acid, was added to each fecal sample. About 40 μL of HCl (37% American Chemical Society reagent) was added and vortexed for 30 s. Samples were then incubated at 65 °C for 4 h. Following incubation, samples were dried under nitrogen gas at 65 °C. About 200 μL of TMS-derivatization agent (Supelco’s Sylon HTP HMDS + TCMS + Pyridine, 3:1:9 Kit) was added to the sample and incubated for an additional 30 min at 65 °C. Sample was again dried under nitrogen gas at 65 °C for 25 min. Samples were then resuspended in 300 μL of hexane, vortexed briefly, and centrifuged for 10 min at 3,200 × g. The supernatant was then transferred to a Gas Chromatography (GC)/Mass Spectrometer (MS) vial insert (300 μL glass with polymer feet, Agilent, Santa Clara, CA) and capped. A Thermo TRACE 1310 Gas Chromatography coupled with ISQ LT single quadrupole Mass Spectrometer was used for all separations. The GC/MS used a capillary column (DB-1ms Ultra Inert, Agilent, Santa Clara, CA) using the following dimensions: 30 m, diameter: 0.250 mm, film thickness: 0.25 μm. A 20:1 split was used after a 1-μL sample was injected with an inlet temperature of 280 °C. After injection, oven temperature was held at 50 °C for 1 min, then ramped at 20 °C per minute to a final temperature of 300 °C, and held for 12 min. Helium was used as the carrier gas at a nominal flow rate of 1.5 mL/min.

DNA extraction, amplification, sequencing, and bioinformatics

Total DNA was extracted from fresh fecal samples using a Mo-Bio PowerSoil kit (MO BIO Laboratories, Inc., Carlsbad, CA). Quantification of DNA concentration was completed using a Qubit 2.0 Fluorometer (Life Technologies, Grand Island, NY). A Fluidigm Access Array (Fluidigm Corporation, South San Francisco, CA), in combination with Roche High Fidelity Fast Start Kit (Roche, Indianapolis, IN), was used for amplification of the 16S rRNA gene. Full-length 16S PacBio (Pacific Biology, Menlo Park, CA)-specific primers, forward (AGRGTTYGATYMTGGCTCAG) and reverse (RGYTACCTTGTTACGACTT) tags, were added in accordance with the PacBio protocol. The quality of amplicons’ regions and sizes was confirmed by Fragment Analyzer (Advanced Analytics, Ames, IA). A DNA pool was generated through the combination of equimolar amounts of the amplicons from each sample. The pooled samples were selected by size on a 2% agarose E-gel (Life Technologies, Grand Island, NY) and extracted using a Qiagen gel purification kit (Qiagen, Valencia, CA). The pooled, size-selected, and cleaned products were then run on an Agilent Bioanalyzer in order to confirm appropriate profile and mean size. The Roy J. Carver Biotechnology Center at the University of Illinois performed PacBio sequencing. The 16S amplicons were generated with the barcoded full-length 16S primers from PacBio and the 2x Roche KAPA HiFi Hot Start Ready Mix (Roche, Willmington, MA). The amplicons were pooled and converted to a library with the SMRTBell Express Template Prep kit 2.0 (Pacific Biology, Menlo Park, CA). The library was sequenced on 1 SMRTcell 8M in the Sequel II using the CCS sequencing mode and a 10-h movie time. Analysis of CCS was done using SMRTLink V8.0 using the following parameters: minimum length 1,200, maximum length 2,000, minimum passes 3, and minimum rq 0.99. An average of 25,897 reads was obtained, with a total of 932,315 reads. The number of reads ranged from 15,691 to 43,168 per sample.

Analysis of sequences was completed using DADA2 (version 1.14; Callahan et al., 2016). From this analysis, a total of 2,427 taxa were imported into the phyloseq R package (McMurdie and Holmes, 2013). Taxa were removed for not having an assigned phylum, for being mitochondrial DNA, or for having zero counts. The phyla Campilobacterota, Deferribacterota, and Spirochaetota were removed for having very low prevalence (<0.01% of total reads) resulting in 1,655 taxa. Sequences were then agglomerated based on the distribution of inter-taxa phylogenetic distances independent of a reference database, using a threshold of 0.03. This agglomerated very tight clusters of taxa near the tips of the tree. A total of 178 operational taxonomic units (OTUs) were assigned. Prevalence filtering was performed after agglomeration. Based on the observed distributions, OTUs were only retained for further analysis if they were observed in two or more samples. After singleton filtering, the number of assigned OTUs was reduced to 160.

Bray and Curtis (1957) dissimilarity, UniFrac distance (Hamady et al., 2010), and weighted UniFrac distance (Hamady et al., 2010) were calculated between samples after converting OTU abundances to proportions. Nonmetric multidimensional scaling was then used to visualize the distance matrices in two dimensions. Alpha-diversity was assessed by observed OTU, Chao1, Shannon, and Simpson and Inverse Simpson indexes. The DESeq2 R package (Love et al., 2014) was used to identify taxa that were differentially abundant between treatments; a false discovery rate (FDR; Benjamini and Hochberg, 1995) lower than 0.05 was used to declare statistical significance. Canonical correspondence analysis (CCA) was performed using the vegan (Oksanen et al., 2019) R package, with whole blood taurine, plasma taurine, CA, LCA, and CDCA as constraining variables.

Statistical analyses

All data were analyzed in SAS (SAS Institute, Inc., version 9.4, Cary, NC) using repeated measures in the MIXED models procedure. The model was run with a fixed effect of diet and a random effect of dog. Day was used as the repeated variable. The interaction of treatment by day and the main effects of day and treatment were reported using a Fisher-protected least significant difference test with a Tukey adjustment to control for type 1 experiment-wise error. Means were considered to be statistically significant using a probability of P < 0.05. The reported SEMs were determined from the MIXED models procedure in SAS. Urine taurine concentrations were square-root transformed to normalize data.

Results

Diet composition, food intake, and serum metabolites

Ingredient composition (Table 1) of both diets was targeted to be similar with the exception of the inclusion of green lentils. Diets were formulated to be as isocaloric and isonitrogenous as possible (Table 2). Macronutrient composition of the dietary treatments is reported on a dry matter basis (DMB). DM content was 92.2% for GLD and 91.8% for CON. CON had a higher CP content (31.2%) compared with GLD, which had a CP content of 27.4%. The TDF content was higher for GLD (11.2%) than CON (8.7%). The diets had similar GE content at 5.0 and 5.1 kcal/g for GLD and CON, respectively.

Table 2.

Analyzed chemical composition and gross energy content of canine diets containing green lentils or poultry byproduct meal as the primary protein source

	Dietary treatment1
Item	CON	GLD
Dry matter, %	91.8	92.2
Crude protein, %	31.2	27.4
Acid hydrolyzed fat, %	15.9	14.5
Total dietary fiber, %	8.7	11.2
Soluble, %	3.4	4.2
Insoluble, %	5.3	7.0
Ash, %	7.2	7.1
Gross energy, kcal/g	5.1	5.0

¹CON, poultry byproduct meal control; GLD, green lentil diet.

Food intake (g/d) was not significantly different (P > 0.05) between dietary treatments on an as-is basis or DMB (Figure 1). On an as-is basis, the food intake for dogs fed GLD was 192.0 and 196.3 g/d for dogs fed CON. Dogs fed GLD consumed 177.1 g/d DMB compared with dogs fed CON consuming 180.3 g/d DMB. Both diets were considered to be well accepted and resulted in adequate daily food intake to fulfill the nutrient requirements of the dog.

Figure 1.

Food intake of dogs fed green lentil or poultry byproduct meal diets. Abbreviations: CON, poultry byproduct meal control; GLD, green lentil diet.

Serum chemistry and a CBC were analyzed throughout the study to monitor health of the dogs and ensure that dietary treatments did not cause any negative health effects. Serum metabolites were within normal ranges at baseline (Table 3) and remained within normal ranges for the duration of the study. All dogs remained healthy throughout the study. Additionally, the analyzed CBC was determined to be normal for healthy adult dogs (data not shown).

Table 3.

Serum metabolites of dogs fed green lentil or poultry byproduct meal diets

		Day 0		Day 30			P-value
Item	Reference range2	CON	GLD	CON	GLD	SEM	Trt	Day	Trt * Day
Creatinine, mg/dL	0.5 to 1.5	0.7	0.6	0.7	0.6	0.087	0.6621	0.1515	0.8538
BUN3, mg/dL	6.0 to 30.0	15.5	15.0	17.0	14.2	1.927	0.7580	0.1984	0.2681
Total protein, g/dL	5.1 to 7.0	5.5	5.6	5.8	6.0	0.102	0.6252	<0.0001	0.4638
Albumin, g/dL	2.5 to 3.8	3.2	3.2	3.2	3.3	0.051	0.3557	<0.0001	0.7009
Globulin, g/dL	2.7 to 4.4	2.3	2.4	2.7	2.7	0.072	0.9306	<0.0001	0.1535
Ca, mg/dL	7.6 to 11.4	9.9	9.7	9.9	9.8	0.204	0.6742	0.4559	0.7077
P, mg/dL	2.7 to 5.2	4.1	3.6	3.8	3.4	0.274	0.4763	0.1620	0.2901
Na, mmol/L	141 to 152	145.5	142.5	144.8	144.2	1.141	0.2507	0.0683	0.3642
K, mmol/L	3.9 to 5.5	4.2	4.1	4.0	4.0	0.090	0.5652	0.0634	0.1974
ALT (SGPT)	8.0 to 65.0	18.2	23.3	21.0	24.2	2.702	0.1451	0.0787	0.7630
Cl, mmol/L	107 to 118	111.7	108.5	110.7	110.0	1.067	0.2005	0.8752	0.2610
Glucose, mg/dL	68 to 126	97.0	100.7	96.7	96.2	3.466	0.8918	0.0039	0.1295
Total bilirubin, mg/dL	0.1 to 0.3	0.2	0.2	0.2	0.2	0.023	0.8636	0.3301	0.2445
Triglycerides, mg/dL	32 to 154	65.5	74.0	61.2	54.8	3.993	0.4602	<0.0001	0.0102
		Day 60		Day 90			P-value
Item	Reference range2	CON	GLD	CON	GLD	SEM	Trt	Day	Trt * Day
Creatinine, mg/dL	0.5 to 1.5	0.8	0.7	0.7	0.7	0.087	0.6621	0.1515	0.8538
BUN,3 mg/dL	6.0 to 30.0	17.3	16.8	15.8	16.5	1.927	0.7580	0.1984	0.2681
Total protein, g/dL	5.1 to 7.0	5.9	5.9	5.9	5.9	0.102	0.6252	<0.0001	0.4638
Albumin, g/dL	2.5 to 3.8	3.2	3.2	3.3	3.4	0.051	0.3557	<0.0001	0.7009
Globulin, g/dL	2.7 to 4.4	2.8	2.7	2.6	2.5	0.072	0.9306	<0.0001	0.1535
Ca, mg/dL	7.6 to 11.4	9.9	9.8	10.1	9.9	0.204	0.6742	0.4559	0.7077
P, mg/dL	2.7 to 5.2	4.0	4.0	3.8	3.8	0.274	0.4763	0.1620	0.2901
Na, mmol/L	141 to 152	144.0	142.5	145.8	144.2	1.141	0.2507	0.0683	0.3642
K, mmol/L	3.9 to 5.5	4.1	4.2	3.9	4.1	0.090	0.5652	0.0634	0.1974
ALT (SGPT)	8.0 to 65.0	21.3	27.3	21.5	27.8	2.702	0.1451	0.0787	0.7630
Cl, mmol/L	107 to 118	110.7	109.0	110.8	109.2	1.067	0.2005	0.8752	0.2610
Glucose, mg/dL	68 to 126	99.7	96.2	101.3	104.2	3.466	0.8918	0.0039	0.1295
Total Bilirubin, mg/dL	0.1 to 0.3	0.2	0.2	0.3	0.2	0.023	0.8636	0.3301	0.2445
Triglycerides, mg/dL	32 to 154	68.7	57.8	57.7	52.5	3.993	0.4602	<0.0001	0.0102

¹CON, poultry byproduct meal control; GLD, green lentil diet.

²References ranges were provided by the University of Illinois Veterinary Diagnostic Laboratory.

³BUN, blood urea nitrogen.

Whole blood taurine concentrations

No significant interaction between treatment and day or their main effects (P > 0.05) were observed in whole blood taurine concentrations (Figure 2). Whole blood taurine in adult healthy dogs is 200 to 350 nmol/mL, according to the reference ranges determined by University of California at Davis. Taurine concentrations over 150 nmol/mL equate to no known risk for developing taurine deficiency (Delaney et al., 2003) On day 0, the dogs assigned to the GLD and CON treatments had baseline whole blood taurine concentrations of 228.0 and 241.3 nmol/mL, respectively. On day 90, the dogs fed GLD had taurine concentrations of 223.0 nmol/mL and dogs fed CON had 250.3 nmol/mL. Dogs on both treatments remained within normal reference ranges throughout the study.

Figure 2.

Whole blood taurine concentrations of dogs fed green lentil or poultry byproduct meal diets. Dashed line indicates minimum taurine concentration (150 nmol/mL) where deficiency is not observed. Abbreviations: CON, poultry byproduct meal control; GLD, green lentil diet.

Plasma amino acid concentrations

Plasma taurine concentration was similar (P > 0.05) between treatments and over time (Figure 3). The reference range for plasma taurine concentration in adult healthy dogs is 60 to 120 nmol/mL, as determined by University of California at Davis. No known risk for taurine deficiency has been observed for dogs with plasma taurine concentrations greater than 40 nmol/mL (Delaney et al., 2003). Dogs at baseline (day 0) had plasma taurine concentrations of 90.7 nmol/mL for GLD and 83.2 nmol/mL for CON. On day 90, the dogs fed GLD had plasma taurine concentrations of 105.7 nmol/mL, and dogs fed CON had plasma taurine concentrations of 96.8 nmol/mL. All dogs remained within the normal reference ranges throughout the study and were not at risk for taurine deficiency.

Figure 3.

Plasma taurine concentration of dogs fed green lentils or poultry byproduct meal diets. Dashed line indicates minimum taurine concentration (40 nmol/mL) where deficiency is not observed. Abbreviations: CON, poultry byproduct meal control; GLD, green lentil diet.

Plasma amino acid profiles (Table 4) showed no differences (P > 0.05) in methionine concentration (Figure 4) at any time point in the study. At baseline, dogs fed GLD at 63.8 nmol/mL and dogs fed CON at 63.8 nmol/mL. At the end of the study (day 90), the dogs fed GLD had plasma methionine concentrations of 64.8 nmol/mL and dogs fed CON had 55.7 nmol/mL, with minimal variation from these values on days 30 and 60. Mean plasma methionine concentration of 57 nmol/mL has been reported in healthy adult dogs, with a wide range being observed from 45 to 136 nmol/mL (Delaney et al., 2003).

Table 4.

Plasma amino acid concentrations of dogs fed green lentil or poultry byproduct meal diets

	Day 0		Day 30		Day 60		Day 90			P-value
Indispensable amino acid, nmol/mL	CON1	GLD1	CON	GLD	CON	GLD	CON	GLD	SEM	Trt	Day	Trt * Day
Arginine	110.0	119.2	127.0	131.0	127.3	142.2	150.5	154.3	10.931	0.5284	0.0004	0.8782
Histidine	95.3	101.0	90.7	89.7	89.5	96.7	93.5	101.8	5.213	0.4583	0.0208	0.3563
Isoleucine	62.0	65.0	58.7	52.5	59.8	55.0	58.7	54.8	5.149	0.6258	0.1294	0.5716
Leucine	131.8	130.2	121.3	99.3	118.7	106.2	102.2	101.8	9.689	0.4475	0.0004	0.2559
Lysine	212.8	198.2	215.8	196.0	221.5	233.7	223.0	252.0	21.319	0.9478	0.0682	0.2749
Methionine	60.3	63.8	60.0	59.2	63.3	64.3	55.7	64.8	3.487	0.4401	0.2888	0.1897
Phenylalanine	67.3	69.0	66.2	69.3	69.0	71.0	69.7	75.2	3.659	0.4755	0.2535	0.8692
Threonine	83.2	90.7	197.7	221.2	194.7	245.7	176.7	255.3	23.086	0.1372	0.3689	0.0830
Tryptophan	77.7	80.0	77.8	93.8	80.0	103.3	69.0	96.7	12.399	0.2363	0.5356	0.5137
Valine	182.3ab	191.8a	189.8ab	155.7b	183.8ab	160.8ab	163.8ab	156.7b	12.649	0.3915	0.0133	0.0386
Dispensable amino acid, nmol/mL
Alanine	530.8a	574.8a	468.8ab	383.8b	449.3b	410.8b	421.2b	420.0b	26.357	0.5321	0.0001	0.0044
Aspartate	6.0bc	6.0bc	5.3c	8.7b	6.7bc	11.3a	6.8bc	10.7a	0.6630	0.0020	0.0001	0.0003
Cystathionine	3.8bc	3.8bc	5.3ab	3.0bc	6.3a	2.5c	4.5abc	3.2bc	0.5666	0.0013	0.5834	0.0049

¹CON, poultry byproduct meal control; GLD, green lentil diet.

^a–cMeans within a row with different superscript letters are different (P < 0.05).

Figure 4.

Plasma methionine concentration of dogs fed green lentil or poultry byproduct meal diets. Dashed line indicates mean canine plasma methionine concentration (57 nmol/mL) in healthy dogs. Abbreviations: CON, poultry byproduct meal control; GLD, green lentil diet.

In addition, most plasma concentrations of dispensable amino acids did not differ (P > 0.05) between dogs fed CON and GLD or overtime. A significant treatment by time interaction was observed for only alanine, aspartate, and cystathionine (Table 4). Plasma alanine concentration was greater at day 0 for both CON (530.8 nmol/mL) and GLD (574.8 nmol/mL) treatments and decreased overtime in both treatments (421.2 and 420.0 nmol/mL at day 90 for CON and GLD, respectively). Plasma aspartate concentration did not differ on day 0 between CON and GLD; however, on days 60 and 90, it was increased in dogs fed GLD (11.3 and 10.7 nmol/mL, respectively) in contrast with dogs on the CON diet (6.7 and 6.8 nmol/mL, respectively). In contrast, plasma concentration of cystathionine remained stable for dogs fed GLD, whereas dogs on CON (6.3 nmol/mL) had a transient increase (P > 0.05) of this nonprotein amino acid on day 60. However, this effect was no longer observed on day 90, CON (4.5 nmol/mL) and GLD (3.2 nmol/mL) did not differ (P > 0.05) in the concentration of this plasma cystathionine, and these values were also not different (P > 0.05) from baseline concentrations, both at 3.8 nmol/mL.

Urine taurine and creatinine concentrations

Urine taurine concentrations (Table 5) were highly variable throughout the study. Because of the variation, no significant differences (P > 0.05) were observed between treatments at any time point. On day 30, the dogs fed GLD and CON had urinary taurine concentrations of 4,343 and 5,027 nmol/mL, respectively. On day 60, GLD was at 5,665 nmol/mL and CON was 7,469 nmol/mL. At the end of the study, GLD was at 3,509 nmol/mL and CON was 8,960 nmol/mL. Urine creatinine concentrations (Table 5) were not significantly different (P > 0.05) between treatments at any time point. On day 30, GLD had creatinine concentrations of 13,413 nmol/mL and CON had 12,443 nmol/mL. On day 90, GLD had a creatinine concentration of 14,620 nmol/mL and CON had 16,493 nmol/mL. The taurine:creatinine ratio remained unchanged (P > 0.05) throughout the study. On days 30, 60, and 90, dogs fed GLD had a taurine:creatinine ratios 0.33, 0.34, and 0.34, respectively. Dogs fed CON had ratios of 0.39, 0.52, and 0.54, respectively.

Table 5.

Urine taurine and creatinine concentrations of dogs fed green lentil or poultry byproduct diets

	Day 30		Day 60		Day 90			P-value
Item	CON1	GLD1	CON	GLD	CON	GLD	SEM	Trt	Day	Trt * Day
Taurine, nmol/mL	5,027	4,343	7,469	5,665	8,960	3,509	2,314.88	0.1328	0.7009	0.5387
Creatinine, nmol/mL	12,443	13,413	14,629	12,789	16,493	14,620	2,483.51	0.7625	0.2886	0.6236
Taurine:creatinine	0.39	0.33	0.52	0.34	0.54	0.34	0.1272	0.2103	0.7784	0.8170

¹CON, poultry byproduct meal control; GLD, green lentil diet.

Fecal BAs

Fecal BAs (Table 6) were analyzed for primary (CA and CDCA) and secondary (DCA, LCA, and UDCA) BAs. A treatment effect (P < 0.05) showed that dogs fed GLD had higher CA, CDCA, and total primary fecal BA concentrations compared with dogs fed CON. For the secondary BA, a treatment effect (P < 0.05) showed that dogs fed CON had higher fecal LCA concentrations than dogs fed GLD. No significant treatment-by-day effects (P > 0.05) were observed for any of the analyzed BA. Additionally, there were no significant main effects or treatment by day effects (P > 0.05) for the total fecal BA concentrations between treatment groups.

Table 6.

Fecal bile acid excretion of dogs fed green lentils or poultry byproduct meal diets

	Day 30		Day 60		Day 90			P-value
Bile acid, μg/mg	CON1	GLD1	CON	GLD	CON	GLD	SEM	Trt	Day	Trt * Day
Primary, total	0.46	0.71	0.46	0.66	0.41	0.51	0.063	0.0137	0.1014	0.4285
CA	0.21	0.38	0.21	0.34	0.18	0.28	0.046	0.0280	0.2354	0.4665
CDCA	0.27	0.38	0.25	0.32	0.23	0.26	0.031	0.0304	0.0550	0.3980
Secondary, total	3.30	3.46	7.02	4.41	3.15	3.15	1.088	0.4907	0.0164	0.2349
DCA	1.98	2.54	5.04	3.36	2.05	2.28	0.931	0.7708	0.0212	0.2938
LCA	1.16	0.76	1.86	0.86	0.96	0.71	0.183	0.0138	0.0087	0.0651
UDCA	0.15	0.16	0.15	0.17	0.14	0.16	0.009	0.0914	0.1633	0.4660
Total bile acids	3.77	4.81	7.66	5.06	3.56	3.65	1.154	0.6959	0.0217	0.1575

¹CON, poultry byproduct meal control; GLD, green lentil diet.

Fecal microbiome

Fecal microbiota was analyzed and 160 different taxa were identified in dogs fed GLD or CON. The predominant phyla (Figure 5) that were identified in the current study were Actinobacteria, Bacteroidota, Firmicutes, Fusobacteriota, and Proteobacteria. The Firmicutes phyla comprised the majority of the microbial composition at the phyla level in both dogs fed CON and dogs fed GLD. Microbial composition at the family level is shown in Figure 6. In dogs fed CON, Bacteroidaceae, Erysipelotrichaceae, Peptostreptococcaceae, and Lactobacillaceae were predominant families. In dogs fed GLD, Bacteroidaceae, Erysipelotrichaceae, Lactobacillaceae, Peptostreptococcaceae, and Selenomonadaceae were predominant families. At the genus level (Figure 7), dogs fed CON had Allobaculum, Bacteroides, Fusobacterium, and Peptoclostridium as major genera. Dogs fed GLD had Allobaculum, Lactobacillus, and Megamonas as major genera. Analysis of microbial communities (Table 7) showed 14 different taxa that increased and 15 taxa that decreased in differential abundance in dogs fed GLD compared with dogs fed CON (P < 0.05). The difference is characterized as a log 2-fold change, with a P-value and FDR cutoff of 0.05.

Figure 5.

Fecal microbial composition at the phyla level of dogs fed either green lentil or poultry byproduct meal diets. Abbreviations: CON, poultry byproduct meal control; GLD, green lentil diet.

Figure 6.

Fecal microbial composition at the family level of dogs fed either green lentil or poultry byproduct meal diets. Abbreviations: CON, poultry byproduct meal control; GLD, green lentil diet.

Figure 7.

Fecal microbial composition at the genera level of dogs fed green lentil or poultry byproduct diets. Abbreviations: CON, poultry byproduct meal control; GLD, green lentil diet.

Table 7.

Differential abundance of microbial communities in dogs fed green lentils or poultry byproduct meal diets

Phylum	Family	Genus	Species	Fold change	P-value	FDR1
Actinobacteriota	Bifidobacteriaceae	Bifidobacterium	Bifidobacterium pullorum	4.54	0.01	0.04
	Coriobacteriaceae	Collinsella	Collinsella sp.	4.41	0.00	0.00
Bacteroidota	Bacteroidaceae	Bacteroides	Gut metagenome	−3.45	0.00	0.00
		Bacteroides	Uncultured bacterium	−4.00	0.00	0.01
		Bacteroides	Gut metagenome	−3.59	0.00	0.01
		Bacteroides	Bacteroides stercoris	4.23	0.00	0.03
		Bacteroides	Bacteroides caecigallinarum	−4.32	0.01	0.04
	Prevotellaceae	Prevotellaceae_Ga6A1_group	Uncultured bacterium	−8.65	0.00	0.00
Firmicutes	Butyricicoccaceae	Butyricicoccus	Butyricicoccus pullicaecorum	3.61	0.00	0.00
	Clostridia_UCG-014	Clostridia_UCG-014	Uncultured organism	−5.70	0.01	0.04
	Erysipelatoclostridiaceae	Erysipelatoclostridium	Uncultured organism	−6.13	0.01	0.04
	Erysipelotrichaceae	Allobaculum	Uncultured bacterium	5.77	0.00	0.01
		Faecalibaculum	Faecalibaculum rodentium	5.43	0.00	0.01
		Turicibacter	Turicibacter sp.	2.42	0.01	0.04
		uncultured	Uncultured bacterium	−2.31	0.00	0.01
	Lachnospiraceae	[Ruminococcus]_gnavus_group	Unclassified	2.43	0.00	0.00
		Blautia	Blautia hansenii	2.06	0.01	0.04
		Blautia	Unclassified	2.60	0.01	0.05
		Lachnoclostridium	Clostridium glycyrrhizinilyticum	5.55	0.00	0.00
		Lachnoclostridium	Unidentified	−4.65	0.00	0.01
		Unclassified	Unclassified	−3.95	0.00	0.00
		uncultured	Faecalimonas umbilicata	2.83	0.01	0.04
	Peptostreptococcaceae	Romboutsia	Unclassified	3.09	0.00	0.00
	Ruminococcaceae	Faecalibacterium	Unclassified	−1.86	0.00	0.02
	Selenomonadaceae	Megamonas	Uncultured organism	4.54	0.00	0.00
		Megamonas	Uncultured bacterium	−5.00	0.00	0.03
Fusobacteriota	Fusobacteriaceae	Fusobacterium	Unclassified	−1.53	0.01	0.04
Proteobacteria	Sutterellaceae	Parasutterella	Unclassified	−6.31	0.00	0.00
		Sutterella	Unclassified	−2.31	0.00	0.02

¹FDR, false discovery rate.

The α-diversity (measured as observed, Chao1, Shannon, Simpson, and InvSimpson analyses) is shown in Figure 8. The α-diversity of dogs fed GLD was lower (P < 0.05) in observed and Chao1. The β-diversity is represented as weighted (Figure 9a) and unweighted UniFrac distance (Figure 9b). Dogs within each treatment group clustered more closely together than those between treatments. CCA of taxa abundance constrained by CA, CDCA, and LCA, and whole blood and plasma taurine as metabolic variables can be found in Figure 10. Dogs fed CON were more strongly correlated with fecal LCA concentrations and Streptococcus alactolyticus. Additionally, dogs fed CON were also strongly correlated to whole blood taurine concentrations, with Tyzzerella and Clostridium Sensu Stricto as corresponding taxa. Dogs fed GLD were more strongly correlated with fecal primary BA (CA and CDCA) than dogs fed CON. Romboutsia and Turicibacter sp. were strongly correlated with CA, while Blautia hansenii and Clostridium paraputrificum were strongly correlated with CDCA.

Figure 8.

Fecal microbial α-diversity of dogs fed either green lentil or poultry byproduct meal diets. Abbreviations: CON, poultry byproduct meal control; GLD, green lentil diet.

Figure 9.

Principal component analyses of weighted (A) and unweighted (B) distances of fecal microbial communities of dogs fed green lentil or poultry byproduct meal diets. Abbreviations: CON, poultry byproduct meal control; GLD, green lentil diet; NMDS, nonmetric multidimensional scaling.

Figure 10.

Canonical correspondence analysis (CCA) of taxa abundance constrained by bile acid and taurine metabolic variables. Abbreviations: CON, poultry byproduct meal control; GLD, green lentil diet; WB, whole blood.

Discussion

Diet composition and food intake

The diets were formulated to have similar chemical and ingredient compositions, with the exception of the inclusion of green lentils, which were added at the expense of poultry byproduct meal and rice. By design, all diets were formulated to meet AAFCO (2018) requirements for adult dogs at maintenance with no further adjustments on specific dietary amino acid content, or methyl supply (e.g., folate, creatine, l-carnitine, cobalamine, and pyridoxine). Recently, Banton and et al. (2021) reported that dietary supplementation of methionine was effective in increasing the concentration of this amino acid in plasma and whole blood of dogs fed grain-free diets. The same authors also suggested a sparing effect on methionine requirement of dogs fed grain-free diets supplemented with creatine, carnitine, or choline. At 45%, the inclusion of the lentils in the diet is higher than typically observed in commercial grain-free diets. The diets were well accepted by the dogs throughout the study with intake only differing by 4 g/d (as-is) between treatments. While the food acceptance largely depends on the preference of the individual dogs, this corroborates with a previous study that determined that there were no differences in rate of consumption, distraction during meal times, or level of anticipation before meal times for dogs fed an animal protein-based diet compared with dogs fed a diet containing 120 g/kg chickpeas and lentils (Callon et al., 2017).

Whole blood taurine concentrations

Analyzing blood and plasma taurine concentrations is an important aspect in determining if diet can impact taurine biosynthesis and cause deficiency in dogs. Typically, analysis of whole blood taurine concentration more accurately determines taurine status of the animal compared with plasma taurine concentrations (Heinze et al., 2009). This is due to the fact that whole blood taurine concentration better reflects the taurine concentration present within skeletal muscle (Pacioretty et al., 2001). However, there is not a definitive association between the inclusion of legumes and a decrease in whole blood taurine concentration in dogs. A previous study was conducted evaluating the taurine status of Beagles fed either a grain-free diet or a grain-based diet for 28 d (Pezzali et al., 2020). The whole blood taurine concentration was 228 nmol/mL at baseline and remained unchanged through day 14 (235 nmol/mL) and day 28 (253 nmol/mL) in dogs fed the grain-based diet. Similarly, in dogs fed the grain-free diet, the whole blood taurine concentrations on days 0, 14, and 28 were 240, 255, and 271 nmol/mL, respectively (Pezzali et al., 2020). Similar results were observed herein. Dogs on GLD had numerically lower whole blood taurine concentration, but this was also observed at baseline, when animals were still being fed the CON diet. Over time, whole blood taurine concentration was consistent among dogs within treatment, factor depicted by the high P-values for treatment and day interaction and the main effect of day. For this reason, this observation should be attributed to individual differences among dogs rather than a potential treatment effect.

It should be acknowledged that certain breeds, typically large and giant breeds, are genetically predisposed to taurine deficiency and development of DCM (Ko et al., 2007). Large breed dogs have been shown to have lower plasma taurine concentrations compared with small breed dogs due to decreased taurine synthesis rates when excess precursor amino acids (i.e., methionine and cysteine) are unavailable (Ko et al., 2007; Ontiveros et al., 2020). However, only Beagles were used in the current study. This breed is not known to be genetically prone to taurine deficiency or the development of DCM.

Plasma amino acid concentrations

While plasma taurine concentrations are commonly measured in dogs suspected to have diet-associated DCM, it should be noted that plasma taurine concentrations in healthy dogs do not vary considerably with dietary taurine intake (Chesney et al., 2010). This could account for the lack of differences observed between treatments in this study. For example, in healthy Labrador Retrievers that had been fasted for 48 h, no changes in baseline plasma taurine concentrations were observed until after 47 h (Gray et al., 2016). Therefore, the lack of correlation between plasma taurine concentrations in dogs and dietary taurine intake should be considered when establishing an association between grain-free diets and the development of DCM, as not all dogs with DCM show lowered plasma taurine concentrations. In 76 dogs diagnosed with DCM, only 17% were considered to have low plasma taurine concentrations, with concentrations in dogs ranging from 1 to 247 nmol/mL (Kramer et al., 1995).

In dogs with taurine-deficient DCM, it is common to have plasma taurine concentrations lower than 40 nmol/mL, although these findings are inconsistent and may be dependent on genetic propensity for taurine deficiency. In Newfoundlands predisposed to taurine-deficient DCM, plasma taurine concentrations ranged from 3 to 228 nmol/mL, with 63% of the dogs having plasma taurine concentrations below 40 nmol/mL (Backus et al., 2006). It is unclear if the inclusion of legumes in diets contributes to the lowered plasma taurine concentrations in dogs. In healthy Beagles fed grain-based or grain-free diets for 28 d, the plasma taurine concentrations did not change from baseline (121 and 111 nmol/mL, respectively) for both dogs fed grain-based (139 nmol/mL) and grain-free (169 nmol/mL) diets by day 28 (Pezzali et al., 2020).

There is currently no dietary taurine requirement for dogs due to their ability to endogenously synthesize taurine from its precursors, methionine and cysteine, due to sufficient cysteine sulfinic acid decarboxylase activity (Ko and Fascetti, 2016). Plasma methionine concentrations within normal ranges indicate that sufficient sulfur amino acids are present as precursors for biosynthesis of taurine to occur (Ripps and Shen, 2012). The plasma methionine concentrations measured in the current study are similar to the plasma methionine concentrations analyzed in previous studies. The plasma methionine concentration measured in Pezzali et al. (2020) showed that, at baseline, the dogs fed a grain-based diet had a methionine concentration of 58 nmol/mL, whereas dogs fed a grain-free diet had a methionine concentration of 56 nmol/mL. The methionine concentrations remained the same throughout the study regardless of diet, with methionine concentrations of 66 nmol/mL for dogs fed a grain-based diet and 63 nmol/mL for dogs fed a grain-free diet by day 28 (Pezzali et al., 2020). Because plasma methionine concentrations were within normal ranges for both treatment groups in the current study, sufficient methionine was present to adequately synthesize taurine.

Similar to methionine, the measurement of plasma cyst(e)ine is an important indicator of available sulfur amino acids that can act as a precursor for taurine synthesis in dogs (Ripps and Shen, 2012). The association of diet to the development of DCM was previously attributed to dogs fed lamb and rice diets (Fascetti et al., 2003). Tôrres et al. (2003) demonstrated that dogs fed a lamb and rice diet had lower intakes of methionine and cysteine (1.4 g/d) than dogs fed a commercial dog chow (2.2 g/d) after 22 wk. It was later discovered that lamb meal had low bioavailability of cysteine, leading to inadequate synthesis of taurine and development of taurine-deficient DCM in some dogs (Fascetti et al., 2003). Plasma cystine concentrations were not measured in the current study due to degradation of the cyst(e)ine present in the sample.

The accumulation of cystathionine in dogs fed CON on day 60 may indicate that higher concentrations of methionine were converted to cysteine through cystathionine β-synthase, the enzyme responsible for converting homocysteine to cysteine in the transsulfuration pathway, as described in Banerjee et al. (2003). However, without measuring the cysteine concentrations, the authors are unable to definitively comment. The increased plasma aspartate concentrations observed in dogs fed GLD compared with those fed CON could be due to the high concentration of aspartic acid (10.7% DMB) present in lentils (Khazeaei et al., 2019).

Urine taurine and creatinine concentrations

Urine taurine concentrations were measured as a way to acutely reflect changes in taurine status caused by changes in renal homeostatic regulation of taurine excretion (Chesney et al., 1985; Ko et al., 2007). The variation in urinary taurine excretion observed in the current study could be due to the collection of a single fresh urine sample rather than compositing samples over an extended period of time. However, the variation is consistent with previous studies evaluating urine taurine concentrations in dogs. In one study analyzing the urine taurine excretion in healthy Beagles, the total amount of taurine ranged from 0 to 197,037 nmol/kg/24 h (Sanderson et al., 2001).

Creatinine, a product of muscle catabolism, is often used as a marker of renal function (Baum et al., 1975). The kidney is an important regulator of taurine homeostasis, responsible for reabsorbing or excreting taurine based on intake and availability (Chesney et al., 2010). Because creatinine is excreted at a rate that is independent of urine output (Baum et al., 1975), a taurine:creatinine ratio was calculated to correct for differences in output of the fresh urine samples. Similar taurine:creatinine ratios have been reported. Pezzali et al. (2020) had baseline values of 0.35 for both dogs fed grain-free and dogs fed grain-based diets. The values for dogs fed grain-based diets increased to 0.52 on day 14 and 0.58 on day 28, whereas the values for dogs fed grain-free diet increased to 0.56 on day 14 and 0.54 on day 28 (Pezzali et al., 2020). In dogs fed a commercial weight management diet, the taurine:creatinine ratio was lower than that of the current study at 0.07. The dogs were then fasted for 48 h, and the taurine:creatinine ratio after re-feeding was 0.27 (Gray et al., 2016). In Labrador Retrievers fed a commercial grain-free diet for 26 wk, the taurine:creatinine ratio was 0.25 at baseline and remained unchanged at 0.28 by week 26 (Donadelli et al., 2020).

Fecal BAs

Unlike most mammals that conjugate BA with glycine or taurine, dogs exclusively conjugate BA with taurine (Anantharaman-Barr et al., 1994; Herstad et al., 2018). These BAs are recycled through enterohepatic circulation in the distal ileum with 95% efficiency (Ridlon et al., 2006). Disruption of the recycling system can cause increased fecal BA excretion and, therefore, depletion of taurine stores due to increased BA synthesis to compensate for the loss (Ko and Fascetti, 2016). Therefore, excessive BA excretion in the feces could be a potential mechanism for taurine loss in dogs with taurine-deficient DCM, as reviewed in Mansilla et al., 2019. It was hypothesized in the current study that the majority of fecal BA would be secondary, indicating a healthy gut microbiota, with no differences between treatments.

The ratio of primary and secondary BA found in fecal BA profiles can be indicative of gastrointestinal health (Molinero et al., 2019). Guard et al. (2019) analyzed the fecal unconjugated BA in healthy dogs and in dogs with chronic inflammatory enteropathy (CE). Healthy dogs had higher concentrations of secondary BA (3.3 μg/mg) than primary BA (0.4 μg/mg) in the feces compared with dogs with CE which had an inverted ratio, with 1.7 μg/mg primary BA and 0.5 μg/mg secondary BA (Guard et al., 2019). Higher concentrations of primary BA in the feces may be an indicator of gastrointestinal dysbiosis. BAs are highly damaging to bacterial cells due to the ability of BA to solubilize bacterial cell membranes, acidify the cytoplasm, and cause protein misfolding (Ruiz et al., 2013; Molinero et al., 2019). Therefore, bacteria must have protective measures against cell damage or death in the form of bile salt hydrolases, which deconjugate and convert primary BA to secondary BA (Ruiz et al., 2013; Ridlon et al., 2006).

The effect of diet on fecal BA excretion has been well documented. High-fiber diets are often associated with increased fecal output and disruption of the enterohepatic recycling of taurine-conjugated BAs (Story and Kritchevsky, 1978; Mansilla et al., 2019). When lamb and rice diets were associated with DCM, in addition to the low bioavailability of cysteine in lamb meal, rice bran also demonstrated deleterious effects on taurine status (Stratton-Phelps et al., 2002).

Legumes and pulses are high-fiber ingredients that may increase BA excretion in dogs. Pezzali et al. (2020) measured fecal BA excretion in dogs fed a grain-free or grain-based diet for 28 d. The total fecal primary BA content on days 0, 14, and 28 for dogs fed a grain-based diet was 0.28, 0.29, and 0.28, respectively, whereas dogs fed a grain-free diet had fecal primary BA concentrations of 0.34, 0.54, and 0.51 μmol/g, respectively. The total secondary BA concentrations (DCA + LCA) on days 0, 14, and 28 were 1.5, 1.7, and 2.5 μmol/g for dogs fed grain-based diets and 1.4, 1.9, and 2.5 μmol/g for dogs fed a grain-free diet (Pezzali et al., 2020). Donadelli et al. (2020) analyzed the fecal BA profile of Labrador Retrievers fed a commercial grain-free diet for 26 wk. The total primary BA increased from 0.06 μg/mg at baseline to 0.2 μg/mg after 26 wk. Additionally, the total secondary BA also increased from 4.2 μg/mg at baseline to 6.5 μg/mg after 26 wk (Donadelli et al., 2020).

Our findings did not support our hypothesis, as dogs fed the GLD have greater concentration of primary BA and lower concentration of LCA. However, the data gathered herein do not support that changes in fecal BA concentrations will be the cause for taurine deficiency in dogs. These findings are in agreement with Pezzali et al. (2020) and Donadelli et al. (2020).

Fecal microbiome

The gastrointestinal microbiota is a complex system that is increasingly recognized as an important metabolic regulator relating to animal health (Wernimont et al., 2020). The differential abundance of the dogs fed GLD demonstrated increases in Erysipelotrichaceae Allobaculum and Erysipelotrichaceae Faecalibaculum compared with dogs fed CON. Prevotellaceae were decreased in dogs fed GLD compared with dogs fed CON. However, the presence of Prevotellaceae has been variable in dogs (Pilla and Suchodolski, 2019). In overweight dogs fed a low-protein/high-carbohydrate diet, Prevotellaceae was 4.6 times more abundant than in dogs fed a high-protein/low-carbohydrate diet, with no differences in lean dogs fed the same diets (Li et al., 2017).

As mentioned previously, excess excretion of BA has been implemented as a method of taurine depletion in dogs fed grain-free diets (Mansilla et al., 2019). The bacteria present in the large intestine influence BA metabolism through self-protective modifications, such as deconjugation and 7α-dehydroxylation of primary BA (Ridlon et al., 2006). Through these biotransformative mechanisms of primary BA to secondary BA, the gut microbiota effectively alters the BA pool and fecal BA composition (Staley et al., 2017).

The BA profile found in fecal samples is typically reflective of the microbiota’s impact on BA metabolism, with higher concentrations of secondary BA than primary BA (Molinero et al., 2019). Inversions of this ratio have been associated with dysbiosis (Ridlon et al., 2006). One study examined the differences in fecal microbiota and BA profiles in healthy dogs and dogs with exocrine pancreatic insufficiency (Blake et al., 2019). Dogs with treated or untreated exocrine pancreatic insufficiency had a lower fecal secondary BA (0.98 and 0.21 μg/mg, respectively) compared with healthy dogs (5.11 μg/mg). The fecal primary BA concentrations in treated and untreated dogs were 1.4 and 1.2 μg/mg, respectively, whereas healthy dogs had primary fecal BA concentrations of 0.4 μg/mg. The dysbiosis index was higher for both untreated (6.5) and treated (0.8) dogs compared with the dysbiosis index of −3.2 in healthy dogs (Blake et al., 2019).

Bile salt hydrolases, the enzymes responsible for deconjugation of BA, are bacterial enzymes found in species of Bacteroides, Clostridium, Lactobacillus, Bifidobacterium, and Listeria (Ridlon et al., 2006). Additionally, a separate metagenomics study found bile salt hydrolase activity in Firmicutes and Actinobacteria taxa (Jones et al., 2008). Wang et al. (2019) analyzed the microbiome of dogs with canine enteropathy fed a hydrolyzed protein diet. Thirty-one species were identified, belonging to Lactobacillus, Streptococcus, and Eubacterium genera that were positively correlated with fecal LCA and DCA concentrations in dogs (Wang et al., 2019). Similarly, in the current study, Eubacterium and Streptococcus alactolyticus were strongly correlated with LCA concentrations.

The effects of orally administered taurine on the modulation of the gut microbiota have also been studied. Although studies in dogs are minimal, in mice fed diets containing either saline, natural taurine, or synthetic taurine, the primary phyla identified were Firmicutes, Bacteroidetes, and Proteobacteria (Yu et al., 2016). The relative abundance of Proteobacteria decreased by 8.4% and 5.5% in mice given natural taurine or synthetic taurine, respectively, relative to the 3.2% decrease in mice given saline. Additionally, the dominant families were Lachnospiraceae, Rikenellaceae, Porphyromonadaceae, and Helicobacteraceae. A higher relative abundance of Helicobacteraceae (7.3% and 4.1%) and a lower relative abundance of Rikenellaceae (0.6% and 1.3%) were measured in mice given natural or synthetic taurine, respectively, compared with the saline group (Yu et al., 2016).

Although the gut microbiota composition is heavily influenced by diet, further research is required to determine if the intestinal bacteria have a major role in taurine depletion in dogs fed grain-free diets. However, the relationship between microbiota and BA metabolism is an important potential mechanism to consider as a cause of taurine deficiency. Additionally, microbiota composition varies based on the individual, age, and breed of the animal which should also be considered.

Conclusions

Inclusion of green lentils up to 45% did not have any negative effects on health or taurine status in dogs over a 90-d period. Taurine concentrations in the whole blood and plasma were within normal ranges for adult dogs for both diets. No change in plasma methionine concentrations was observed throughout the course of the study, indicating sufficient precursors for taurine synthesis. An important caveat to the potential association of grain-free diets to DCM is that not all dogs with diet-related DCM are taurine deficient. Fecal BA profiles showed higher concentrations of total primary BA in dogs fed GLD compared with dogs fed CON. Fecal microbial communities of dogs fed CON were more strongly correlated with fecal LCA concentrations. The relationship between intestinal microbiota, BA metabolism, and whole-body taurine status needs to be further investigated in order to implement this mechanism as a route of taurine deficiency. While no detrimental effects were observed in this study, the authors acknowledge the limitations of the study. Beagles, a small breed dog not prone to development of DCM, were used in the study and fed diets that are not reflective of dogs fed grain-free diets. With only n = 6 per treatment, the small sample size is also recognized as a limitation in the study. Additionally, the absence of baseline data for fecal and urine samples was based on the assumption that all dogs had similar BA and amino acid profiles at the start of the study but is considered to be a limitation of the study. Further investigation to determine the cause of taurine depletion in dogs fed grain-free diets is needed before establishing a definitive link between legumes and DCM.

Glossary

Abbreviations

AHF

acid hydrolyzed fat

BA

bile acid

CA

cholic acid

CBC

complete blood count

CCA

canonical correspondence analysis

CDCA

chenodeoxycholic acid

CE

chronic inflammatory enteropathy

CON

control

CP

crude protein

DCA

deoxycholic acid

DCM

dilated cardiomyopathy

DM

dry matter

DMB

dry matter basis

FDR

false discovery rate

GC

gas chromatography

GE

gross energy

GLD

green lentil diet

LCA

lithocholic acid

MS

mass spectrometer

OM

organic matter

OTU

operational taxonomic unit

TDF

total dietary fiber

UDCA

ursodeoxycholic acid

Conflict of interest statement

All authors have no conflict of interest to declare.