Adult dogs of different breed sizes have similar threonine requirements as determined by the indicator amino acid oxidation technique

Abstract

Threonine (Thr) requirements for immature (growing) Beagles have been determined, but little knowledge is available on Thr requirements for maintenance in mature dogs. Moreover, differences of Thr requirements among different breeds or sizes of adult dogs have not been investigated. The objective of the present study was to determine Thr requirements in adult dogs of three different breeds using the indicator amino acid oxidation (IAAO) technique. In total, 13 adult dogs were used, 4 Miniature Dachshunds (5.8 ± 0.4 kg body weight [BW]; 3 spayed and 1 neutered), 4 spayed Beagles (9.3 ± 0.6 kg BW), and 5 neutered Labrador Retrievers (30.5 ± 1.7 kg BW). Dogs were fed a Thr-deficient diet (Thr = 0.23%) and randomly allocated to receiving one of seven concentrations of Thr supplementation (final Thr concentration in experimental diets was 0.23%, 0.33%, 0.43%, 0.53%, 0.63%, 0.73%, and 0.83%; as fed basis) for 2 d. After 2 d of adaptation to the experimental diets, dogs underwent individual IAAO studies. During the IAAO studies, total daily feed was divided into 13 equal meals; at the sixth meal, dogs were fed a bolus of l-[1-¹³C]-Phenylalanine (Phe) (9.40 mg/kg BW), and thereafter, l-[1-¹³C]-Phe (2.4 mg/kg BW) was supplied with every meal. Before feeding the next experimental diet, dogs were fed a Thr-adequate basal diet for 4 d (Thr = 0.80% as fed basis) in known amounts that maintained individual dog BW. Total production of ¹³CO₂ during isotopic steady state was determined by enrichment of ¹³CO₂ in breath samples and total production of CO₂ measured using indirect calorimetry. The mean requirements for Thr, defined as the breakpoint, and the 95% confidence interval (CI) were determined using a two-phase linear regression model. For Miniature Dachshunds, the two-phase model was not significant, and Thr requirements could not be determined. Mean Thr requirements for Beagles and Labradors were 72.2 and 64.1 mg/kg BW on an as-fed basis, respectively. The requirement for Thr between these two dog breeds was not different (P > 0.10). Thus, the data for Beagles and Labradors were pooled and a mean requirement for Thr was determined at 66.9 mg/kg BW, and the 95% CI was estimated at 84.3 mg/kg BW. In conclusion, estimated Thr requirements for Beagles and Labradors did not differ, and these recommendations are higher than those suggested by NRC (2006) and AAFCO (2014) for adult dogs at maintenance.

Introduction

Compared with other monogastric livestock species (i.e., poultry and swine), little is known about amino acid (AA) requirements in the dog. The latest version of NRC (2006) for dogs and cats compiles a series of studies determining AA requirements in the immature (growing) and adult dogs. However, the NRC (2006) states that: “no individual dose-response peer-reviewed reports could be found for the minimal requirements of any of these AA -including Thr- for maintenance of dogs” and suggests a minimal requirement (MR) of 0.34% in the diet for Thr based on one doctoral dissertation (Ward, 1976) and one peer-reviewed report (Sanderson et al., 2001). Sanderson et al. (2001) conducted a 4-yr study where 18 Beagles (twelve 1-yr-old spayed females, three 9-yr-old spayed females, and three 2.5-yr-old castrated males) were fed different low-crude protein diets in amounts necessary to maintain body weight (BW) and reported no clinical signs of deficiency. The NRC (2006) argues that no diet that meets the MR for protein and contains cereal grains, animal byproducts, and plant proteins has been shown to be deficient in those AA (including Thr). The NRC statement may be true but does not encourage research on AA nutrition to reduce excess dietary AA supply and reduction of N excretion. Moreover, studies determining AA requirements for growing dogs have used mainly the Beagle as a representation of all dog breeds and, as such, do not account for potential differences among dog breeds, sizes, different life stages (young vs. senior), or lifestyles (active vs. less active).

Requirements for AA have generally been determined using nitrogen balance and growth performance (Milner, 1979; Burns and Milner, 1982; Morris et al., 2004). These techniques require feeding over a period of time (usually a week or more) to allow adaptation to the test diet. Nitrogen balance and growth performance (mainly muscle growth) are closely related to protein deposition (the main factor for estimating AA requirements during growth) and do not fully consider the utilization of AA for other metabolic pathways (i.e., Thr as a Gly precursor; Met as methyl group donor). Furthermore, the body protein pool does not change considerably at maintenance rendering nitrogen balance and BW gain techniques insensitive for determining AA requirements with increasing concentrations of a single AA. The indicator AA oxidation (IAAO) technique has been validated in multiple species, and it is particularly appealing since it can measure changes in AA oxidation without compromising AA utilization when feeding deficient diets for longer periods such as those used in growth and nitrogen balance studies. Moreover, the IAAO technique is minimally invasive and allows for the determination of AA oxidation over a range of AA consumption in the same subject over a relatively short period of time (Elango et al., 2012). Thus, the IAAO technique is a more sensitive technique for measuring changes in protein synthesis and allows the determination of AA requirements in adult dogs (Shoveller et al., 2017).

This report is one of a series working to determine AA requirements of adult dogs of different breeds. The objective of the present study was to determine the requirements for Thr in adult dogs of different breeds using the IAAO technique. We recently reported no differences among three breeds of adult dogs for phenylalanine (Phe) but found that Beagles had a greater tryptophan (Trp) requirement than Miniature Dachshunds or Labrador Retrievers (Mansilla et al., 2018; Templeman et al., 2019). Given our findings with Trp, we hypothesized that the Thr requirement would be different among breeds. We predicted that actual Thr requirements are higher than those presented in the literature currently based in N balance and maintenance of BW (NRC, 2006; AAFCO, 2014).

Materials and Methods

Animals and housing

The present experiment was approved by the Procter and Gamble Pet Care Animal Care and Use Committee. A total of 13 dogs were used; 4 miniature Dachshunds (5.8 ± 0.4 kg BW; 0.81 ± 0.01 yr of age, mean ± SD; 3 spayed and 1 neutered), 4 spayed Beagles (9.3 ± 0.6 kg; 5.60 ± 0.29 yr of age, mean ± SD), and 5 neutered Labrador Retrievers (30.5 ± 1.7 kg; 3.84 ± 1.54 yr of age, mean ± SD). Miniature Dachshunds, Beagles, and Labrador Retrievers were chosen to represent small, medium, and large dog breeds, respectively. Numerous studies that have estimated indispensable AA requirements for human Elango et al. (2007); Hsu et al. (2006); Tang et al. (2014); Turner et al. (2006); Lazaris-Brunner et al., 1998; Bross et al., 2000; Wilson et al., 2000; Di Buono et al., 2001; Roberts et al., 2001; Courtney-Martin et al., 2002; Kriengsinyos et al., 2002; Mager et al., 2003), swine (Moehn, et al., 2004, 2008), poultry (Tabiri et al., 2002; Coleman et al., 2003), and dogs (Mansilla et al., 2018; Templeman et al., 2019) using the IAAO method have done so using similar numbers of subjects (n = 4 to 6). A power of greater than 0.8 was found for the two-phase linear regression model used for each breed size using a data step of SAS and data from Templeman et al. (2019) (SAS Institute Inc., v. 9.4, Cary, NC). During the study, dogs were housed in groups of 2 in a temperature-controlled room (21 °C) and with a lighting schedule of 12:12 (L:D) h. All dogs were healthy throughout the experiment as per the medical examination by a licensed veterinarian. Dogs received daily socialization, exercise, and regular veterinary care as previously reported (Shoveller et al., 2017).

Diets and study design

A basal diet was formulated to meet or exceed requirements for all indispensable AA according to NRC (2006; Table 1). The extruded kibble basal diet was fed twice daily (0700 and 1300 hours) during 14 d prior to the beginning of the experiment (adaptation period) in amounts known to maintain dog individual BW. During the whole experiment, water was supplied ad libitum via an automatic watering system. After the 14-d adaptation period to the basal diet, a test diet (similar to basal diet but without added crystalline Thr; final Thr = 0.23%) was fed to the dogs at 20 g/kg of BW for Miniature Dachshunds and Beagles, and at 15 g/kg BW for Labradors for 2 d prior to each IAAO study. The test diets were supplemented with one of seven l-Thr dressing solutions (0, 3.33, 6.67, 10.00, 13.33, 16.67, and 20.00 g/L) at 6.0 mL/kg BW for Miniature Dachshunds and Beagle dogs and at 4.5 mL/kg BW for Labradors. To maintain similar nitrogen content among all dressing solutions, l-Ala was added (14.9, 12.5, 10.0, 7.5, 5.0, 2.5, and 0 g/L for solution one to seven, respectively) (l-Thr and l-Ala supplier: Skidmore Sales & Distributing, West Chester Township, OH). The final Thr contents in the test diets plus the dressing solution were 0.23%, 0.33%, 0.43%, 0.53%, 0.63%, 0.73%, and 0.83 % (experimental diets). After the 2-d adaptation period to the experimental diet (Moehn et al., 2004), the IAAO study combined with indirect calorimetry was conducted. After each IAAO study, dogs returned to the basal diet for 4 d before being fed the test diet with a different dressing solution and conducting the next IAAO study. This 7-d feeding regimen was repeated seven times with treatments assigned to dogs in a random order (Microsoft Excel 2010: Random function) based on a complete randomized block (within breed) design. After completion of the study, all dogs were fed each of the seven experimental diets. In addition, blood samples (4 mL) were collected from the jugular vein in serum vacutainers (Becton & Dickinson, Franklin Lakes, NJ) at the end of each IAAO study. Throughout the whole study, all dogs consumed their entire daily diet offerings (basal and test diets).

Table 1.

Ingredient composition and analyzed nutrient contents of the basal on an as-fed basis

Ingredients	g/kg
Corn starch	480.6
Chicken fat	130.6
Chicken meal	63.9
Yellow corn	50.6
Brewer’s rice	50.6
Amino acid premix1	75.9
Beet pulp	30.4
Dicalcium phosphate	29.0
Chicken flavor	20.2
Potassium chloride	13.3
Sodium bicarbonate	10.1
Chicken liver flavor	5.06
Brewer’s yeast	5.06
Ground flax	5.06
Choline chloride	4.47
Vitamin premix2	4.25
Sodium hexametaphosphate	4.05
Calcium carbonate	5.80
Mineral premix D3	3.44
Fish oil	2.91
Sodium chloride	1.82
Monosodium phosphate	2.33
Ethoxyquin	0.51
Nutrient content	Analyzed content
Metabolizable energy, kcal/kg (calculated)4	3,700
DM, %	92.22
CP, %	10.70
Arg, %	1.031
Cys, %	1.030
His, %	0.540
Ile, %	0.534
Leu, %	0.811
Lys, %	0.723
Met, %	0.527
Phe, %	0.674
Thr, %	(0.227)5
Trp, %	0.342
Tyr, %	0.475
Val, %	0.630

¹Provides per kg of final diet: 4.03 g of Arg, Cys, His, Ile, Leu, Lys, Met, Phe, Thr, Trp, Tyr, and Val each. Threonine was removed for the test diets.

²Vitamin premix contained per kg: 6,650 K IU vitamin A, 365,000 IU vitamin D₃, 100,400 IU vitamin E, 4,100 mg thiamine, 2,500 mg niacin, 2,000 mg pyridoxine, 7,750 mg d-pantothenic acid, 115 mg folic acid, 45 mg vitamin B₁₂, 2,500 mg inositol, 13,750 mg vitamin C, 1,200 mg β-carotene.

³Mineral premix contained per kg: 150 mg cobalt carbonate, 4,500 mg copper sulphate, 900 mg potassium iodide, 72,000 mg iron sulfate, 8,000 mg manganese oxide, 5,800 mg manganese sulfate, 60,000 mg sodium selenite.

⁴Calculated metabolizable energy based on modified Atwater values.

⁵Values presented in brackets correspond to the Thr-deficient test diet only; AA premix without crystalline Thr.

Indicator amino acid oxidation studies

On the day when the IAAO study was conducted, dogs were weighed and moved to individual respiration calorimetry chambers. After 30 min of gas equilibration, three fasting respiration/indirect calorimetry measurements were taken over two consecutive 25-min periods to determine the resting volume of CO₂ and O₂ produced (VCO₂; VO₂) for each dog. Dogs were then fed (time = 0) their corresponding feed allowance divided into 13 equal meals based on BW measured that day. The first three meals were fed every 10 min to induce a fed state, and the other 10 meals were fed every 25 min. The total amount of feed fed during the IAAO study was based on BW measured the same day in the morning after 17 h of fasting (20 g/kg of BW for Miniature Dachshunds and Beagles and at 15 g/kg BW for Labradors). Background enrichment was determined by the collection of CO₂ samples over three consecutive 25-min periods. The next meal (95 min after first feeding) contained a priming dose (9.40 mg/kg BW) of l-[1-¹³C]-Phe (99%; Cambridge Isotope Laboratories Inc., Tewksbury, MA). To maintain the supply of l-[1-¹³C]-Phe, the following seven meals contained constant doses (2.40 mg/kg BW) of l-[1-¹³C]-Phe for all dogs. Expired CO₂ was collected over the last eight 25-min periods. Overall during each IAAO study, each dog spent ~6.3 h inside the calorimetry chamber. Additional details regarding the timeline for each IAAO study can be found in Mansilla et al. (2018).

Sample collection and analysis

Nitrogen content of the basal diet was analyzed with a LECO analyzer (LECO Corporation, St. Joseph, MI). Serum was immediately analyzed for serum AA concentrations using the method of Bidlingmeyer et al. (1984). Amino acid content in the test diet was analyzed using the AOAC method 999.12 (AOAC International, 2000). Calorimetry data were collected automatically using Qubit calorimetry software (Customized Gas Exchange System and Software for Animal Respirometry; Qubit Systems Inc., Kingston, ON). Measured VCO₂ during fasting and fed states were averaged over the collection periods to obtain a mean fasting and fed VCO₂ for each dog. Background and enriched samples of CO₂ were collected by trapping subsamples of expired CO₂ in 1 M NaOH. The NaOH solution was subsampled and samples were stored at room temperature until further analysis. The enrichment of ¹³C in breath CO₂ captured in NaOH solution was measured by continuous-flow isotope ratio mass spectrometry (20/20 isotope analyzer; PDZ Europa Ltd., Cheshire, UK). Enrichment of CO₂ samples were expressed above background samples (atom percent excess [APE]).

Calculations

The rate of ¹³CO₂ released per kg of BW per h (F¹³CO₂, mmol.kg⁻¹.h⁻¹) was calculated using the following equation: F¹³CO₂ = (FCO₂)(ECO₂)(44.6)(60)/[(BW)(1.0)(100)], in which FCO₂ is the average production of CO₂ during the isotope steady-state phase (mL/min); ECO₂ is the average ¹³CO₂ enrichment in expired breath at isotopic steady state (APE, %); and, BW is the weight of the dog (kg). The constants 44.6 (mmol/mL) and 60 (min/h) convert the FCO₂ to micromoles per hour; the factor 100 changes APE to a fraction; and the 1.0 is the retention factor of CO₂ in the body due to bicarbonate fixation as reported previously (Shoveller et al., 2017). Similar calculations were performed to determine the rate of ¹³CO₂ released per kg of lean body mass (LBM). Resting and fed energy expenditure (EE) was calculated based on VO₂ and VCO₂ based on the modified Weir equation (Weir, 1949). EE (kcal/d) was expressed in relation to BW, BW^−0.75, and LBM for all dog breed sizes.

Body composition determination

LBM was determined at the end of the 7-wk study using an X-Ray Bone Densitometer (Model Delphi A, Hologic Inc., Marlborough, MA) on Beagles and Labrador dogs. Accurate prediction of LBM was not possible in Miniature Dachshunds due to their unique conformation. Dogs were fasted overnight (18 h since last meal) and sedated by an experienced technician using Dexmedetomidine (Dexdomitor, Pfizer, New York, NY) at a dose of 0.02 mg/kg, and Carprofen (Rimadyl, Pfizer, New York, NY) at a dose of 2 to 4 mg/kg administered i.m. Propofol (PropoFlo, Abbott, Chicago, IL) at a dose of 5 to 7 mg/kg was administered i.v. for induction of anesthesia. During scans, anesthetized dogs were positioned in sternal recumbency with the cranial aspect of ante brachium placed on the table and phalanges pointing caudally. Body mass composition (i.e., mineral, fat, lean, and water contents) was determined in the left and right arms and legs, trunk, and head (data not shown). Whole-body composition was determined by the sum of all regions measured on individual dogs. Scans were done in triplicate for each dog and the median value of the three scans was recorded. Following the scan, atipamezole (Antisedan, Pfizer, New York, NY) was administered to each dog at a dose of 0.2 mg/kg. Dogs were placed in a heated cage until fully recovered and monitored for 1 wk.

Statistical analysis

The effect of Thr content in the test diet on F¹³CO₂, BW, and calorimetry data was analyzed using PROC MIXED of SAS (v. 9.4; SAS Institute Inc., Cary, NC) with diet as a fixed effect and dog as a random effect. The estimate of the mean Thr requirement and the upper 95% confidence interval (CI) for individual dog breeds and combining Beagles and Labradors were derived by breakpoint analysis of the F¹³CO₂/kg and F¹³CO₂/kg LBM primary data using a two-phase linear regression model as previously reported (Shoveller et al., 2017). Differences in AA concentration in blood were determined using orthogonal contrast comparing each dietary Thr content against the lowest Thr diet (0.23%). Linear contrasts were also performed relating serum AA concentration to the amount of Thr consumed. The mean Thr requirements were also calculated with Thr concentration in serum data using the two-phase linear regression model. BW, LBM, and the calorimetry data were also pooled per dog and were analyzed among breeds using breed as a fixed effect and dog as a random variable. Results were considered statistically significant at P ≤ 0.05 and a trend when P ≤ 0.10.

Results

All dogs had similar body condition score and BW throughout the study (P > 0.10). During the IAAO study, all dogs consumed all their meals immediately after feeding. Isotopic steady state for all dogs on each test diet was confirmed by the last four analyzed breath samples (data not shown). Using a two-phase linear regression relating F¹³CO₂ to the concentrations of Thr in the test diets, the breakpoint for Thr could not be determined for Miniature Dachshunds (Supplemental Figure 1). Thr requirement, based on the F¹³CO₂·kgBW⁻¹·h⁻¹ for Beagles (P = 0.0425, R² = 0.16), was estimated to be 0.361% (as-fed basis) and with a 95% CI of 0.549% (as-fed basis) (72.2 and 109.8 mg/kg BW, respectively; Figure 1). Similar results were obtained for Thr requirement as function of LBM (F¹³CO₂·kgLBM⁻¹·h⁻¹) for Beagles (P = 0.0197, R² = 0.22). For Labradors Retrievers, F¹³CO₂ at the lowest Thr concentration (0.23%) was lower than any other concentration and was not considered for the two-phase model. The Thr requirement was estimated to be 0.427% with a 95% CI of 0.518% on an as-fed basis when estimated by F¹³CO₂·kgBW⁻¹·h⁻¹ (P = 0.182, R² = 0.33) (64.1 and 77.7 mg/kg BW, respectively [Figure 2]) and 0.430% with a 95% CI of 0.518% on as-fed basis when estimated by F¹³CO₂·kgLBM⁻¹·h⁻¹ (P = 0.0152, R² = 0.36) (64.5 and 77.7 mg/kg BW, respectively). The mean requirements for Thr determined in Beagles and Labradors were not different (P > 0.10), and these data were pooled to obtain a single estimated requirement (Figure 3). For Beagles and Labrador Retrievers together, the mean requirement for Thr was 0.463% with a 95% CI of 0.628% based F¹³CO₂·kgBW⁻¹·h⁻¹ (69.5 and 94.2 mg/kg BW, respectively). Similarly, the requirement based on F¹³CO₂·kgLBM⁻¹·h⁻¹ was 0.473 with 95% CI of 0.631% (71.0 and 94.7 mg/kg BW, respectively). To account for different feed intakes between Beagles and Labradors, F¹³CO₂ was also analyzed in relation to AA intake (mg/kg BW). The mean requirement determined by this method was 66.9 mg/kg BW with an upper 95% CI of 84.3 mg/kg BW and relatively similar to requirements determined without consideration of AA intake as a function of BW or LBM.

Figure 1.

Relationship between Thr content in the experimental diets and production of ¹³CO₂ per kg BW per h (A) or per kg LBM per h (B) from the oxidation of orally administered l-[1-¹³C]-Phe in Beagles analyzed by a two-phase regression model. Dashed line: estimated Thr requirement (0.361% for A, 0.366% for B); dotted line: 95% upper CI for Thr requirement (0.549% for A, 0.547% for B). Data points represent mean + SE of samples (n = 4).

Figure 2.

Relationship between Thr content in the experimental diets and production of ¹³CO₂ per kg BW per h (A) or per kg LBM per h (B) from the oxidation of orally administered l-[1-¹³C]-Phe in Labrador Retrievers analyzed by a two-phase regression model. Dashed line: estimated Thr requirement (0.427% for A, 0.430% for B); dotted line: 95% upper CI for Thr requirement (0.518% for A, 0.518% for B). Data points represent mean + SE of samples (n = 5).

Figure 3.

Relationship between Thr intake (mg/kg BW) and production of ¹³CO₂ per kg BW per h from the oxidation of orally administered l-[1-¹³C]-Phe in Beagles and Labradors analyzed by a two-phase regression model. Dashed line: estimated Thr requirement is 66.9 mg/kg BW; dotted line: 95% upper CI for Thr requirement (0.84.3 mg/kg BW). Data points represent mean + SE of samples (n = 4 or 5).

Neither EE (resting, REE; or fasted, FEE) nor respiratory quotient (RQ; fasted or fed) was affected by the different Thr concentrations in the diet (P > 0.10) as expected. BW, LBM, and the indirect calorimetry data were pooled for comparison among breeds (Table 2). As expected, BW of animals was different among the three breeds of dogs (P ≤ 0.05). Absolute LBM (kg) was lower for the Beagles compared with the Labradors (P ≤ 0.05); LBM relative to BW (%) was similar for both breeds (P > 0.10). Resting and fed EE per kg BW^0.75 was similar for all breeds (P > 0.10). There was no difference in fasting RQ (P > 0.10); fed RQ was greater for Beagles compared with Miniature Dachshunds (P ≤ 0.05), but Labradors did not differ from Beagles or Dachshunds (P > 0.10). Rate of CO₂ and O₂ was different among all dog breeds with the highest for Labradors and lowest for Miniature Dachshunds (P ≤ 0.05).

Table 2.

BW, LBM, and indirect calorimetry data (± SE¹) of the dogs used

	Miniature Dachshunds	Beagles	Labradors Retrievers	Pooled ANOVA
	n = 4	n = 4	n = 5	P-value
BW, kg	4.62 ± 0.13c	9.56 ± 0.41b	31.5 ± 0.62a	<0.001
Metabolic BW, kg0.75	3.85 ± 0.07c	5.44 ± 0.23b	13.31 ± 0.35a	<0.001
LBM, kg	—	7.51 ± 0.40b	24.91 ± 1.25a	<0.001
LBM, % BW	—	78.81 ± 2.1	79.84 ± 3.2	0.793
Resting EE, kcal/BW0.75	59.31 ± 2.34	69.27 ± 6.09	75.36 ± 9.26	0.101
Fed EE, kcal/BW0.75	98.46 ± 2.99	82.45 ± 8.88	98.07 ± 13.62	0.238
Fast RQ	0.769 ± 0.009	0.803 ± 0.026	0.780 ± 0.040	0.456
Fed RQ	0.823 ± 0.004b	0.875 ± 0.008a	0.850 ± 0.011ab	<0.001
Fed VO2, L/min	2.34 ± 0.14c	3.79 ± 0.36b	11.06 ± 0.54a	<0.001
Fed VCO2, L/min	1.93 ± 0.14c	3.32 ± 0.33b	9.41 ± 0.51a	<0.001

¹Standard error of mean, n = 4 for Miniature Dachshunds and Beagles, n = 5 for Labrador Retrievers. ^a–cDifferent superscripts in the same row differ significantly, P ≤ 0.05.

For Dachshunds and Labradors, a breakpoint could not be determined in the two-phase linear regression when using Thr concentration in serum, but serum Thr concentration (Table 3) increased linearly with dietary Thr concentrations. For Beagles, the breakpoint for the two-phase linear regression was 0.455% with a 95% CI at 0.641% (91 and 128.2 mg/kg BW, respectively). Results from serum AA that are directly related to Thr or Ala metabolism are presented in Table 4. In Dachshunds, Thr concentration in serum was greater only for dogs fed 0.73% dietary Thr content compared with 0.23% (P ≤ 0.05); Ala decreased with all dietary Thr concentration (P ≤ 0.05). For Beagles, Thr and Gly increased at the highest four dietary Thr concentrations (P ≤ 0.05), while Ala and Glu concentrations decreased with the highest three dietary Thr concentrations (P ≤ 0.05); Asp decreased at 0.63% and 0.73% dietary Thr compared with 0.23% dietary Thr (P ≤ 0.05). For Labrador Retrievers, Thr increased with 0.63% and 0.83 % dietary Thr, and Ala decreased at 0.83% Thr compared with 0.23% dietary Thr (P ≤ 0.05).

Table 3.

Serum concentration of selected AA in adult Miniature Dachshunds, Beagles, and Labrador Retrievers fed diets containing increasing concentrations of Thr

Miniature Dachshunds
	Dietary Thr content, %							SEM1	Linear2
	0.23	0.33	0.43	0.53	0.63	0.73	0.83
Thr, μM	64	91	136	214	207	311*	236	68	0.008
Gly, μM	100	137	123	153	125	146	78	36	0.823
Ser, μM	889	377	310	519	613	707	913	204	0.350
Ala, μM	784	595*	550*	618*	558*	654*	550*	49	0.016
Glu, μM	194	153	78	113	89	168	161	39	0.789
Asp, μM	98.9	95.0	24.4	57.2	36.8	89.2	85.7	26.4	0.786
Beagles
	Dietary Thr content, %							SEM	Linear
	0.23	0.33	0.43	0.53	0.63	0.73	0.83
Thr, μM	31	71	95	225*	317*	488*	613*	62	<0.001
Gly, μM	59	135	67	183*	209*	177*	198*	41	0.010
Ser, μM	687	705	524	509	403	548	384	198	0.218
Ala, μM	766	659	703	723	578*	610*	582*	55	0.054
Glu, μM	173	112	172	154	69*	73*	89*	31	0.028
Asp, μM	107.3	89.7	92.9	91.7	20.8*	27.4*	46.3	26.0	0.013
Labrador Retrievers
	Dietary Thr content, %							SEM	Linear
	0.23	0.33	0.43	0.53	0.63	0.73	0.83
Thr, μM	74	131	236	220	325*	213	510*	69	0.003
Gly, μM	202	157	245	171	209	129	188	50	0.609
Ser, μM	440	426	519	502	618	535	616	155	0.311
Ala, μM	894	774	835	780	764	757	624*	67	0.009
Glu, μM	129	145	97	128	116	172	79	29	0.627
Asp, μM	67.5	76.3	47.1	66.4	63.7	108.7	35.6	23.7	0.910

¹Standard error of the mean, n = 4 at each concentration of dietary Trp for Miniature Dachshunds and Beagles, n = 5 for Labrador Retrievers.

²Linear regression relative to the amount of Thr in the final experimental diet.

*Significantly different (P ≤ 0.05) when compared with the lowest concentration of dietary Thr.

Table 4.

Recommended dietary Thr inclusions for adult dogs at maintenance by AAFCO, FEDIAF, NRC, and the present study

			NRC3		Beagles + Labrador		Beagles		Labrador Retrievers
	AAFCO1	FEDIAF2	MR	RA	MR	CI	MR	CI	MR	CI
g/100 g DM4	0.48	0.60/0.52	0.34	0.43	0.502	0.681	0.391	0.595	0.463	0.562
g/Mcal ME	1.205	1.51/1.30	0.85	1.08	1.251	1.697	0.976	1.484	1.154	1.400
mg/kg BW					66.9	84.3	72.2	109.8	64.1	77.7
								(62.6)6		(46.4)

¹ AAFCO (2014).

² European Pet Food Industry Federation (2013) Nutritional guidelines for complete and complementary pet food for cats and dogs. Values depend on maintenance energy requirements 95 kcal/kg^0.75 or 110 kcal/kg^0.75, respectively.

³Nutrient requirements of dog and cats NRC (2006).

⁴Values for g/100 g DM are determined assuming a dietary energy density of 4,000 kcal ME/kg.

⁵Assuming dietary energy density of 4,000 kcal ME/kg.

⁶Values in parentheses represent NRC recommendation for Thr requirement for adult dogs at maintenance converted from mg/kg BW^0.75 to mg/kg BW using the average BW of individual breeds.

Discussion

Threonine is an indispensable AA and cannot be synthetized endogenously de novo in mammals. Despite the exclusive dependency of dietary Thr supply to meet Thr requirements, there is a paucity of information about minimal concentrations of Thr in the diet for adult dogs (NRC, 2006). The proposed upper 95% CI for Thr of 0.628% on an as-fed basis (84.3 mg/kg BW) was calculated including data from Beagles and Labrador Retrievers. To our knowledge, this is the first dose–response study determining Thr requirements for adult dogs at maintenance. It is worth noting that during the analysis of data, estimated requirements were mainly obtained with only two points in the linear phase of the breakpoint analysis and estimates of the breakpoint may be more precise with additional dietary concentrations of Thr.

The NRC (2006) currently suggests an MR and recommended allowance (RA) of Thr for adult dogs at maintenance of 0.34% and 0.43%, respectively (dry matter basis). These recommendations are based on the lowest concentrations of Thr reported by Sanderson et al. (2001) and Ward (1976) in which dogs were fed low-crude protein diets for extended periods of time (upwards of 4 yr) without showing any signs of clinical deficiency. The 95% CI determined for Beagles, Labradors, and the two breeds together are, on average, ~30% higher than the RA put forth by the NRC (2006) (Table 4). As well, the 95% CIs estimated in this study are ~25% higher than the Thr requirements suggested by AAFCO (2014) for adult dogs at maintenance (Table 4). However, it must be acknowledged that different ingredients have different nutrient bioavailabilies. In the present study, half of the total Thr content of the estimated requirement was supplied from crystalline Thr with an estimated bioavailability of 100%. As such, it must be considered that estimates presented herein may underestimate Thr needs for adult dogs fed commercial diets formulated with natural intact protein-based ingredients (with an assumed bioavailability of 80%; NRC, 2006).

To date, the only study that has determined Thr requirement in a dose–response manner was conducted by Burns and Milner (1982) with 5 to 6-wk-old Beagles. That study estimated a mean Thr requirement of 0.52%. This estimate does not account for population variability or the reduced AA bioavailability in natural protein sources. As well, Burns and Milner (1982) used measures of nitrogen balance and growth performance estimate Thr requirements, a common technique for empirically estimating AA requirements in growing animals (Milner, 1979; Burns and Milner, 1982; Morris, 2004). The sensitivity of these techniques for determining AA requirements in adulthood, when the protein pools are not growing, is decreased resulting in underestimation of AA requirements (Elango et al., 2012; Wu, 2014). Moreover, when using such techniques, AA utilization for other metabolic pathways different than protein synthesis are compromised while feeding lower concentrations of the test AA for longer periods of time. Therefore, nitrogen balance or growth performance prioritize protein deposition over other metabolic pathways in which the test AA is utilized (Elango et al., 2012). The impact of overlooking AA utilization for other reactions not related to protein deposition may have a greater effect on the maintenance AA requirements of adults compared with growing animals as protein deposition is not the main driver of AA utilization in mature animals (Moughan, 1999). The IAAO technique, in contrast, determines AA utilization when pool sizes of secondary metabolites have not likely been impacted and, consequently, incorporates both AA utilization for protein synthesis and alternative oxidative pathways (Penchartz and Ball, 2003). Overall, the IAAO is a more appropriate technique for determining AA requirements in adult dogs (Shoveller et al., 2017).

The Thr requirement expressed in relation to BW or LBM was not different. This is likely related to the main role Thr plays in whole-body metabolism. Threonine is the primary AA used for intestinal mucin production (Lien et al., 1997; Mantle and Allen, 1981), and in some monogastric animals (i.e., growing pigs), it is estimated that >80% of dietary Thr is utilized by the gastrointestinal tract (Scharrt et al., 2005). Therefore, higher utilization and requirements for Thr will increase with a higher visceral size relative to live BW and not with LBM. Moreover, compared with growing dogs, protein deposition in adult animals is fairly static, further increasing the importance of Thr utilization for mucin production relative to the estimated requirement. In the semi-purified test diet used for the present experiment, the fiber content was increased with beet pulp to simulate fiber content of commercial diets. However, by changing the content of vegetable protein, compared with animal protein in commercial diets, the total fiber content will vary and this will potentially increase intestinal Thr needs for higher mucin production (Mathai et al., 2016). It is necessary to include a maximal fiber concentration or accommodate Thr concentration in the diet, relative to the dietary fiber content, to ensure an adequate supply of Thr.

Concentrations of AA in serum were presented only for those AA directly related to Ala and Thr metabolism. Thr increased and Ala decreased with higher Thr and lower Ala supplementation, respectively for all breeds. Only in Beagles, Gly increased, and Ala and Glu decreased with higher concentrations of Thr. In other monogastrics (i.e., pigs), after feed consumption, Ala is almost entirely utilized for glucose production in the liver; Ala-N can be then incorporated into Glu, the main AA for transamination reactions (Brosnan, 2000), and later to Asp (Nissim et al., 2003) explaining the increase in Asp and Glu concentrations with higher Ala consumption. Excess Thr is mainly catabolized in the liver producing de novo Gly (Ballevre et al., 1990). It is likely that dogs have similar AA metabolism to pigs and can be used as a rationale for explaining the present results.

Few studies have focused on breed differences as the only source of variation for calorimetry measurements in dogs. In a review of the literature, Bermingham et al. (2014) propose that maintenance energy requirements are 126.2 and 127.2 kcal/BW^0.75 for Beagles and Labradors, respectively, higher than the REE values estimated in the present study. However, maintenance energy requirements are not the same as EE, explaining the difference between these data sets. Resting and fed EE for the Dachshunds, Beagles, and Labrador Retrievers were similar to our previous studies (Mansilla et al., 2018; Templeman et al., 2019). The fasting RQ coefficient was not different among the studied breeds and was analogous to what occurred in our previous reports (Mansilla et al., 2018; Templeman et al., 2019). Fed RQ increased compared with fasting RQ implying a preferential shift to increased carbohydrate oxidation in the postprandial period. Fed RQ was lower for Miniature Dachshunds compared with Beagles, and both were similar to Labrador Retrievers. The difference in fed RQ among breeds may indicate difference in substrate utilization for energy metabolism and potentially AA requirements in very small dog breed sizes. More research exploring the effect of breed or dog size on AA requirements is needed.

The present study is part of a series of experiments where the requirements for Phe and tryptophan (Trp) have been previously estimated for adult dogs of different breeds using the direct and indirect AA oxidation techniques, respectively (Mansilla et al., 2018; Templeman et al., 2019). Direct AA oxidation results indicated that the estimated requirements of Phe were similar across different breed sizes and that the requirements generated for Dachshunds and Beagles were lower than those presented by the NRC (2006), while the requirements estimated for Labradors were higher (NRC, 2006; Mansilla et al., 2018). Dietary Trp requirements estimated with IAAO techniques were higher for Beagles compared with Labrador Retrievers or Miniature Dachshunds, and the requirements estimated were all three breed sizes were higher than the NRC recommendations for adult dogs at maintenance (Templeman et al., 2019). Although we determined Thr requirements for adult dogs at maintenance, a longer-term dose–response study is needed in order to practically validate the presented estimations. It is also worth mentioning that in such longer-term studies, maintenance of BW or LBM should not be considered the sole final response for determining AA requirements. AA are used for multiple functions that could be negatively affected without disturbing BW or LBM during marginal low consumption. Finally, the authors acknowledge that no conclusions can be drawn regarding how Thr requirements would differ based on hormonal status, as for this study, only spayed and neutered dogs were used. Additional research is necessary to determine whether or not AA requirements differ between intact or spayed/neutered adult dogs.

Conclusions and Implications

The present paper is part of a series of studies determining AA requirements in adult dogs of different breeds using the IAAO technique. According to the results from this study, estimated maintenance requirements for Thr in the adult dog have been underestimated by the NRC (2006) and AAFCO (2014) recommendations. The recommendations presented are derived from diets using highly bioavailable Thr, potentially underestimating Thr requirements. Therefore, when formulating adult dog diets using intact protein-based sources with the intention of meeting or exceeding the requirements estimated in this report, attention must be paid to the difference in nutrient bioavailability between semi-purified diets and commercial diets with natural protein ingredients. Moreover, long-term studies determining essential AA requirements for adult dogs at maintenance are necessary before changes are adopted by the regulatory entities and further investigation in protein and energy metabolism in small dog breeds and among different dog breeds is warranted.

Supplementary Data

Supplementary data are available at Journal of Animal Science online

Supplementary Figure S1. Relationship between Thr content in the experimental diets and production of ¹³CO₂ per kg BW per hour from the oxidation of orally administered l-[1-¹³C]-Phe in Miniature Dachshunds. The two-phase linear-plateau model was not significant.

Glossary

Abbreviations

AA

amino acid

APE

atom percent excess

BW

body weight

CI

confidence interval

EE

energy expenditure

FEE

fasted energy expenditure

IAAO

indicator amino acid oxidation

LBM

lean body mass

MR

minimal requirement

Phe

phenylalanine

RA

recommended allowance

REE

resting energy expenditure

RQ

respiratory quotient

Thr

threonine

Trp

tryptophan

VCO2

volume of CO2 produced

VO2

volume of O2 produced

Conflict of interest statement

A.K.S. and L.F. were employees of the Procter & Gamble Co.; L.F. is now employed by Mars, Pet Care. W.D.M. and J.R.T have no conflicts of interest.