Phenylalanine requirements using the direct amino acid oxidation technique, and the effects of dietary phenylalanine on food intake, gastric emptying, and macronutrient metabolism in adult cats

Abstract

Despite Phe being an indispensable amino acid for cats, the minimum Phe requirement for adult cats has not been empirically defined. The objective of study 1 was to determine the minimum Phe requirement, where Tyr is in excess, in adult cats using the direct amino acid oxidation (DAAO) technique. Four adult male cats were used in an 8 × 4 Latin rectangle design. Cats were adapted to a basal diet for 7 d, top dressed with Phe to meet 140% of the adequate intake (NRC, 2006. Nutrient requirements of dogs and cats. Washington, DC: Natl. Acad. Press). Cats were randomly assigned to one of eight experimental Phe diets (0.29%, 0.34%, 0.39%, 0.44%, 0.54%, 0.64%, 0.74%, and 0.84% Phe in the diet on a dry matter [DM] basis). Following 1 d of diet adaptation, individual DAAO studies were performed. During each DAAO study, cats were placed into individual indirect calorimetry chambers, and 75% of the cat’s daily meal was divided into 13 equal meals supplied with a dose of L-[1-¹³C]-Phe. Oxidation of L-[1-¹³C]-Phe (F¹³CO₂) during isotopic steady state was determined from the enrichment of ¹³CO₂ in breath. Competing models were applied using the NLMIXED procedure in SAS to determine the effects of dietary Phe on ¹³CO₂. The mean population minimum requirement for Phe was estimated at 0.32% DM and the upper 95% population confidence limit at 0.59% DM on an energy density of 4,200 kcal of metabolizable energy/kg DM calculated using the modified Atwater factors. In study 2, the effects of a bolus dose of Phe (44 mg kg⁻¹ BW) on food intake, gastric emptying (GE), and macronutrient metabolism were assessed in a crossover design with 12 male cats. For food intake, cats were given Phe 15 min before 120% of their daily food was offered and food intake was measured. Treatment, day, and their interaction were evaluated using PROC GLIMMIX in SAS. Treatment did not affect any food intake parameters (P > 0.05). For GE and macronutrient metabolism, cats were placed into individual indirect calorimetry chambers, received the same bolus dose of Phe, and 15 min later received ¹³C-octanoic acid (5 mg kg⁻¹ BW) on 50% of their daily food intake. Breath samples were collected to measure ¹³CO₂. The effect of treatment was evaluated using PROC GLIMMIX in SAS. Treatment did not affect total GE (P > 0.05), but cats receiving Phe tended to delay time to peak enrichment (0.05 < P ≤ 0.10). Overall, Phe at a bolus dose of 44 mg kg⁻¹ BW had no effect on food intake, GE, or macronutrient metabolism. Together, these results suggest that the bolus dose of Phe used may not be sufficient to elicit a GE response, but a study with a greater number of cats and greater food intake is warranted.

The minimum Phe requirement, when Tyr is provided in excess, was higher than the current recommendations indicating that more accurate recommendations for commercial adult cat food are necessary. The oral bolus dose of Phe used in this study was found to have no effect on food intake or macronutrient metabolism but tended to delay the time to peak gastric emptying (GE), suggesting potential functional properties that may influence the GE rate and play a role in promoting satiety.

Introduction

The indispensable amino acid (AA) Phe is an aromatic AA that is used for protein synthesis and serves as a precursor to Tyr, a conditionally indispensable AA that can be further metabolized to produce catecholamines (Raper and Wormall, 1923; Hilton et al., 1998), the pigment melanin (Anderson et al., 2002), or thyroid hormones (Dratman., 1974). Prolonged (11 wk to 9 mo) restriction of dietary Phe in growing kittens resulted in neurological dysfunction, loss of black fur pigmentation, and weight loss (Rogers and Morris, 1979; Yu et al., 2001; Anderson et al., 2002; Dickinson et al., 2004). Currently, the only dose–response studies that empirically estimated Phe requirements were conducted using maximal growth (Anderson et al.,1980), nitrogen retention (Williams et al.,1987), and coat color (Anderson et al., 2002) of growing kittens as the dependent variables. From these three studies, when using purified diets in which AA is assumed to be 100% bioavailable, the minimal requirement (MR) for Phe + Tyr was 13.83 g·kg⁻¹ with Tyr sparing roughly 50% of the Phe requirement (wt:wt). When these studies are considered together, the National Research Council’s Nutrient requirements of dogs and cats (NRC, 2006) defines the aromatic AA (Phe + Tyr) MR for growing kittens to sustain entirely black hair coats and maximum nitrogen retention at 15.3 g-Phe·kg⁻¹ diet with a metabolizable energy value of 4.0 kcal·g⁻¹. To address the lower bioavailability of Phe and Tyr among commercial diets, the National Research Council (NRC) recommended allowance (RA) for Phe in growing kittens is 5.0 g·kg⁻¹ diet when the total AA is supplied at 19.1 g·kg⁻¹ diet.

Although the adequate provision of dietary Phe is important to ensure the health of cats, to date, there are no published studies that have empirically determined the minimum Phe requirement in adult cats at maintenance. Consequently, the minimal Phe requirement for growing kittens determined from the studies has been extrapolated to provide estimates of adequate intake (AI) and RA for adult cats at maintenance (NRC, 2006). Extrapolating AA requirements from different life stages and from studies that applied the nitrogen balance technique, which is insensitive in mature animals, may lead to inaccurate recommendations. Estimates of AA requirements using the nitrogen balance technique are roughly 40% lower than those obtained through carbon oxidation techniques (Zello et al.,1995; Humayun et al., 2007; Elango et al., 2008, 2012). This was observed in dogs, where the MR estimations of AA were higher when assessed using carbon oxidation techniques (Shoveller et al., 2017; Templeman et al., 2019; Mansilla et al., 2020a,b; Sutherland et al., 2020) than when estimated with nitrogen balance or growth. Phenylalanine requirements were assessed in two different cohorts of dogs, one being greater (Shoveller et al. 2017) than and one being less than (Mansilla et al. 2018) the MR in NRC (2006). Thus, there is a need to determine the minimum Phe requirement in adult cats using these more sensitive and accurate techniques. The indicator amino acid oxidation (IAAO) technique has been recently applied in adult cats (Pezzali et al., 2023b) and a steady-state isotope dilution protocol for carbon oxidation studies has been determined (Pezzali et al., 2023c) to allow the successful application of the minimally invasive IAAO in the domestic cat to determine AA requirements or to assess AA bioavailability. In addition to the oxidation of the labeled AA in carbon oxidation studies as an indirect measure of protein synthesis, other secondary physiological outcomes can be considered when establishing AA requirements. While intakes of Phe or Tyr higher than the proposed NRC (2006) RA have been reported to support melanin deposition in hair (Watson et al., 2018), the potential role of Phe in promoting the release of cholecystokinin (CCK), and consequently GE has not been investigated to date.

Considered to be a satiety signal, CCK is a hormone that is released by specific enteroendocrine cells in the upper small intestine, commonly referred to as I-cells, when food reaches the duodenum (Wang et al., 2011). Protein and fat induce CCK release (Hopman et al., 1985), but also some specific AA can stimulate CCK, such as tryptophan and Phe (Gibbs et al., 1973; Hira et al., 2008; Wang et al., 2011; Daly et al., 2013) which then binds to receptors on the pyloric sphincter and is thought to delay GE (Wang et al., 2011). Research in humans (Liddle et al., 1986; Enç et al., 2001) and other species (Debas et al., 1975; Yamagishi and Debas, 1978; Raybould and Tache, 1988) suggests that dietary Phe will result in the release of CCK and delay gastric emptying (GE). Previous findings suggest GE to be a key mediator for hunger and satiety (Hunt, 1980; Di Lorenzo et al., 1988); thus, if CCK stimulation can delay GE, satiety may be induced and ultimately food intake reduced. Additionally, the speed at which GE occurs has a direct impact on macronutrient metabolism, as evidenced by studies conducted by Marathe et al. (2013) and Richards et al. (2021) where the rate of GE influenced postprandial glucose concentrations. For example, rapid GE can cause a rapid increase in postprandial glucose that surpasses the ability of insulin to efficiently move glucose into tissues, and potentially trigger lipogenesis. Ultimately, this can contribute to the accumulation of body fat and result in an elevated risk for obesity. Backus and Rogers (1996) reported that a bolus of Phe (44 mg kg⁻¹ BW) resulted in an increase in plasma CCK in cats but did not evaluate GE. As such, we wanted to understand if dietary Phe will alter the rate of food intake and whether changes in GE rate underpinned these.

Two major studies were carried out using adult cats at maintenance. The objective of study 1 was to empirically determine the minimum Phe requirement in adult cats using the direct amino acid oxidation (DAAO) technique when Tyr is supplied in excess. We hypothesized that the mean Phe requirement is greater than the current recommendations (AI and RA) by the NRC (2006). The objective of study 2a was to evaluate the effects of an oral dose of Phe (44 mg kg⁻¹ BW) on food intake when compared to an isonitrogenous dose of Ala (23.7 mg kg⁻¹ BW). We hypothesized that Phe supplementation would decrease food intake. Lastly, the objective of study 2b was to evaluate the effect of Phe (44 mg kg⁻¹ BW) on GE and macronutrient metabolism, when compared to Ala (23.7 mg kg⁻¹ BW) using the ¹³C-OABT (¹³C-OABT). We hypothesized that Phe would delay GE as well as reduce respiratory quotient (RQ) and energy expenditure (EE).

Materials and Methods

The University of Guelph Animal Care Committee approved all experimental procedures for study 1 (AUP #4640) and studies 2a and 2b (AUP #4680), which were in accordance with national, provincial, and institutional guidelines for the care and use of animals in research. The cats used in all studies were deemed healthy before the start of the study, based on medical history, a physical exam, serum biochemistry profile, and complete blood count.

Animals and Housing

All cats used in studies 1, 2a, and 2b were neutered male domestic shorthairs (Marshall’s BioResources, Waverly, NY, USA), and were group-housed at the Ontario Agricultural College at the University of Guelph (Guelph, ON, Canada) in a free-living environment (7.1 m × 5.8 m). The cats were only kept in individual crates during feeding, which occurred once per day at ~0700 hours. A 12:12 (L:D)-h cycle was maintained with lights turning on at 0700 hours and turning off at 1900 hours. Temperature and humidity were maintained and monitored daily, averaging 23.6 °C and 32.4%, respectively. Environmental enrichment was provided as toys, scratching posts, hide boxes, perches, beds, and climbing apparatuses. The room was cleaned and sanitized once daily, while litter boxes were scooped, and litter was added twice per day as needed. Cats were socialized with familiar individuals 5 d/wk for 2 h daily. Socialization included the following activities: general health assessment, brushing, petting, and voluntary play. Cleaning and socialization were done at the same time each day to ensure the cats had a consistent routine throughout the duration of the studies.

Study 1: The Phe Requirement

Body composition determination

Prior to the study, dual-energy x-ray absorptiometry scans were performed by a trained person 24 h after the last meal to determine lean soft tissue mass (LSTM). Cats were sedated using intramuscular dexmedetomidine hydrochloride (Dexdomitor, Zoetis, Kirkland, QC, Canada; 0.5 mg/mL; 0.015 mg·kg⁻¹). Total body mass, fat mass, LSTM, and bone mass were measured through duplicate scans on Small Animal Mode with the thin setting using a fan-beam dual-energy x-ray absorptiometry (Prodigy Advance GE Healthcare, Madison, WI, USA). Cats were positioned in dorsal recumbency with forelimbs extended cranially and repositioned as necessary between scans, which were completed in ~10 min. Estimates for LSTM were obtained from the system software (enCORE Version 16; GE Healthcare, Madison, WI, USA). Results from both scans were averaged. Sedation was reversed by 0.015 mg·kg⁻¹ atipamezole (Antisedan, Zoetis, Kirkland, QC, Canada) after the completion of the second scan.

Diets and study design

Four cats with a mean body weight (BW) of 5.40 ± 0.87 kg (± SD) and a mean age of 2.98 ± 0.47 yr (± SD) were used. The power calculation was based on a 0.05 probability of type I error using the flux of ¹³CO₂ raw data from dogs from Templeman et al. (2019) (SAS, v. 9.4, SAS Institute Inc., Cary, NC).

Prior to the start of the study, the cats transitioned from a commercial diet (T22 Total Grain-Free, Nutram Pet Products, Elmira, ON; metabolizable energy = 3,835 kcal·kg⁻¹; moisture = 10%, crude protein, min = 36%; crude fat, min = 19%; crude fiber, max = 5.5%) to the experimental basal diet over an 8 d period (transition period), where cats were fed to maintain BW based on historical colony feeding records. The experimental basal diet consisted of a semi-synthetic diet formulated to meet or exceed all nutrient requirements according to the NRC (2006), with the exception of Phe (Table 1). The experimental basal diet was prepared fresh daily as described by Pezzali et al. (2023a). During the transition period, the cats were offered 75% of their daily energy requirement as kibble and 25% as the experimental basal diet for 2 d, followed by 50% kibble and 50% experimental basal diet for 2 d, 25% kibble and 75% experimental basal diet for 2 d, and finally 100% experimental basal diet for the last 2 d of the transition period. Then, cats were offered a 100% basal diet (215.72 ± 4.83 kcal·d⁻¹; mean ± SEM) for 7 d to adapt to the protein intake (adaptation period). Cats were fed twice daily (0800 and 1600 hours), with half of their daily meal offered at each feeding time. Food was available for 1 h during feeding times, and freshwater was provided ad libitum. Over the transition and adaption periods, a Phe solution (6.67 g·L⁻¹ and no Ala) was added to the experimental basal diet to ensure the dietary provision of 140% of the NRC (2006) Phe recommendation for AI. Excess Tyr was provided to ensure Phe is shunted to protein synthesis or oxidation and not to Tyr production (Shiman and Gray, 1998).

Table 1.

Ingredient composition of basal diet on as-fed basis and analyzed nutrient content on a dry matter basis of the basal diet in the direct amino acid oxidation study evaluating the minimum Phe requirement of adult cats (study 1)

Ingredient	g kg−1
Pregelatinized wheat starch1	287.00
Water	210.00
Poultry fat	135.00
Amino acid premix2	134.00
Lamb meal	85.00
Sucrose	50.00
Cellulose	23.00
Dicalcium phosphate	15.00
Sodium bicarbonate	15.00
Palatant3	13.00
Brewer’s yeast	10.00
Potassium chloride	8.00
Calcium carbonate	6.00
Choline	4.00
NaCl	3.00
Vitamin premix4	1.00
Mineral premix5	1.00

Chemical composition	Analyzed content
Dry matter (DM), %	76.77
Metabolizable energy6, kcal kg−1	3224
Crude Protein, % DM	22.81
Fat, % DM	20.26
Ash, % DM	8.96
Nitrogen-free extract, % DM	47.97
Amino acids, % DM
Ala	1.60
Arg	1.79
Asp	0.85
Cys	0.98
Glu	3.75
Gly	1.80
His	0.35
Ile	0.84
Leu	1.69
Lys	1.15
Met	0.33
Met + Cys	1.31
Phe	0.29
Phe + Tyr	2.17
Pro	1.44
Ser	1.03
Thr	0.65
Trp	0.32
Tyr	1.87
Val	0.85

¹PAYGEL 290 Pregelatinized Wheat Starch, Archer Daniels Midland Company, USA.

²Amino acid premix contained per kilogram: 26.00 g L-glutamic acid, 16.00 g L-tyrosine, 10.00 g L-arginine and L-leucine, 8.00 g L-alanine, glycine and L-cysteine, 7.00 g proline, 6.00 g L-lysine, L-threonine, and L-serine, 5.00 g L-isoleucine, 4.00 g L-valine, taurine, and L-asparagine H₂O e, 2.00 g DL-methionine, L-tryptophan, and L-histidine HCl H₂O.

³PALASURANCE C45-140 Dry, Kemin Nutrisurance, Inc., USA.

⁴Vitamin composition of the premix: 105.83 mg/kg biotin, 1,573 mg/kg folic acid, 91,700 mg/kg niacin, 7,020 mg/kg pantothenic acid, 6,350 mg/kg pyridoxine, 4,930 mg/kg riboflavin, 10,100 mg/kg thiamin, 24.94 mg vitamin B12, 13,079,000 IU/kg vitamin A, 412,000 IU/kg vitamin D3, 46,718 IU/kg vitamin E.

⁵Mineral composition of the premix: 1.55% calcium, 0.0348% phosphorus, 0.4658 magnesium, 13.92% potassium, 0.2% sodium, 12.93% chloride, 11.90% sulfur, 15,493.9 ppm copper, 92,272.1 ppm iron, 9,397.4 ppm manganese, 361 ppm selenium, 104,105.8 ppm zinc, and 1,953 ppm iodine.

⁶Metabolizable energy = 3.5 kcal/g × crude protein (%) + 8.5 kcal/g × acid-hydrolyzed fat (%) + 3.5 kcal/g × nitrogen-free extract (%); nitrogen free extract (%) = 100% − (crude protein % + acid-hydrolyzed fat % + crude fiber % + ash %).

After the adaptation period, the study was conducted as an 8 × 4 Latin rectangle design. We conducted the trial as an 8 × 4 Latin rectangle to better randomize the order of treatments, as opposed to a 4 × 4 incomplete Latin square design. All cats received all 8 diets in random order and no diet was duplicated within each DAAO study day. The cats were randomly assigned to one of the eight test diets and fed at 14 g·kg⁻¹ BW. The test diets were prepared by supplementing the basal diet with one of eight L-Phe (99% L-Phe, Thera-Plantes Inc., Saint-Laurent, QC) solutions (0, 0.84, 1.69, 2.53, 4.22, 5.91, 7.60, and 9.28 g·L⁻¹) at 6.30 mL·kg⁻¹. To maintain similar nitrogen content among all solutions, L-Ala (≥99% L-Ala, Sigma-Aldrich, St. Louis, MO) was added to the solutions (5.01, 4.55, 4.10, 3.64, 2.73, 1.82, 0.91, and 0 g·L⁻¹ for solutions 1 to 8, respectively). The final Phe content (% dry matter [DM] basis) in the test diets plus the dressing solution were 0.29%, 0.34%, 0.39%, 0.44%, 0.54%, 0.64%, 0.74%, and 0.84%. The study was carried out using a 2-d feeding regimen: 1) day 0: a 1-d adaptation period to the test diet, where the cats were fed 14 g·kg⁻¹ BW (maintenance of BW); 2) day 1: the DAAO study and indirect calorimetry were conducted simultaneously. Following each DAAO study, cats were switched to a new test diet for 1 d, and the following day another DAAO study was conducted. This 2-d feeding regimen was repeated eight times until all cats consumed each test diet.

DAAO studies

The feeding and isotope protocol for each DAAO study day was conducted according to Pezzali et al. (2023c,d). The total amount of food (75% of the animal’s daily food allowance; 10.5 g·kg⁻¹), priming dose (0.44 mg·kg⁻¹BW) of NaH¹³CO₃ (99%; Cambridge Isotope Laboratories, Inc., Tewksbury, MA) and priming (4.80 mg·kg⁻¹) and constant (1.04 mg·kg⁻¹) doses of L-[1-¹³C]-Phe (99%; Cambridge Isotope Laboratories, Inc.) were based on the BW measured on the morning of each DAAO study day. The timeline for carbon oxidation studies in cats has been described in greater detail by Pezzali et al. (2023c).

Sample Collection and Analysis

Diet analysis

The AA content in the basal diet was analyzed by Ajinomoto Animal Nutrition North America, Inc. (Chicago, IL) (total AA [AOAC 994.12], Tryptophan [IAO 13904:2005 E]). The basal diet was also analyzed for moisture (AOAC 930.15), crude protein (AOAC 990.03), hydrolyzed fat (AOAC 945.16), and ash (AOAC 942.05) at a commercial laboratory (SGS Agri-food Laboratories, Guelph, ON, Canada).

Breath samples and calorimetry data

Calorimetry data were collected automatically using Qubit calorimetry software (Customized Gas Exchange System and Software for Animal Respirometry; Qubit Systems Inc., Kingston, ON, Canada). The flow rate was measured with a mass flow meter (#F250, Qubit Systems Inc.). Concentrations of carbon dioxide (CO₂) and oxygen (O₂) gases within the chambers were measured via CO₂ (QS151, Qubit Systems Inc.) and O₂ (Q-S102, Qubit Systems Inc.) analyzers. Measured volume of CO₂ (VCO₂ during fasting and fed states) were averaged over the collection periods to obtain mean fasting and fed VCO₂ for each cat. Background and enriched samples of CO₂ were collected by trapping subsamples of expired CO₂ in 8 mL of 1 M NaOH. Samples were transferred to serum collection tubes (Vacutainer, Becton Dickinson, Franklin Lakes, NJ, USA; #366430), evacuated, and stored at −20 °C until further analysis. Enrichment of ¹³C in breath samples was performed using a Gasbench II interfaced with a Delta V Plus mass spectrometer (Thermo Scientific, Bremen, Germany). Enrichment of CO₂ samples was expressed above background samples as atom percent excess (APE).

Calculations

The rate of ¹³CO₂ released from Phe oxidation per kilogram of BW (F¹³CO₂, mmol·kg⁻¹·h⁻¹) was calculated as follows:

$${\displaystyle \begin{array}{l}{\displaystyle {\mathrm{F}}^{13}\mathrm{C}{\mathrm{O}}_2(\mathrm{m}\mathrm{m}\mathrm{o}\mathrm{l}\cdot \mathrm{k}{\mathrm{g}}^{-1}\cdot {\mathrm{h}}^{-1})=\frac{(\mathrm{V}\mathrm{C}{\mathrm{O}}_2\times \mathrm{A}\mathrm{P}{\mathrm{E}}_{\mathrm{C}{\mathrm{O}}_2}\times 44.6\times 60)}{(\mathrm{B}\mathrm{W}\times 100\times 1)}}\end{array}}$$

where VCO₂ is the average production of CO₂ during fasting state (mL·min⁻¹), $\mathrm{A}\mathrm{P}{\mathrm{E}}_{\mathrm{C}{\mathrm{O}}_2}$ represents the APE of ¹³CO₂ during isotopic steady state (%), BW is the cat’s BW expressed as total BW (kg) or LSTM (kg), 44.6 (mmol·mL⁻¹) and 60 (min·h⁻¹) convert the VCO₂ to micromoles per hour, the factor 100 changes APE to a fraction, and 1.0 is the retention factor of CO₂ due to bicarbonate fixation based on previous dog report (Shoveller et al., 2017). Fasting VCO₂ was used because of the increased variability associated with fed-state VCO₂ in free-living animals.

Resting energy expenditure (REE) and fed-state energy expenditure (FEE) were calculated based on the abbreviated Weir equation (Weir, 1949):

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{E}\mathrm{E}(\mathrm{k}\mathrm{c}\mathrm{a}\mathrm{l})=3.94\times {\mathrm{O}}_2\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{s}\mathrm{u}\mathrm{m}\mathrm{e}\mathrm{d}(\mathrm{L})+1.11\times \mathrm{C}{\mathrm{O}}_2\mathrm{p}\mathrm{r}\mathrm{o}\mathrm{d}\mathrm{u}\mathrm{c}\mathrm{e}\mathrm{d}(\mathrm{L})}\end{array}}$$

The EE was further expressed in relation to BW and metabolic BW (BW^0.67) for all cats. The Qubit system software calculates the RQ based on the ratio of CO₂ produced and O₂ consumed.

Statistical analysis

The study was designed and conducted as an 8 × 4 Latin rectangle design. The effect of Phe content in the test diet on F¹³CO₂ was analyzed using PROC GLIMMIX of SAS (v. 9.4; SAS Institute Inc.) with diet as a fixed effect and cat as a random effect. The effect of dietary Phe content on F¹³CO₂ was also fitted using competing statistical models, namely broken-line linear and broken-line quadratic, using the NLMIXED procedure of SAS where diet was treated as a fixed effect and cat as a random effect. The estimated mean breakpoint parameter and its corresponding 95% confidence interval were determined in each model. Models were then compared based on the Bayesian Information Criterion, where the lowest value represents the best fit.

Study 2: Measurement of Food Intake (Study 2a), GE Rate and Macronutrient Metabolism (Study 2b)

Study design

For studies 2a and 2b, 12 neutered male domestic shorthair cats with a mean BW (±SD) of 5.21 kg (±0.78) and a mean age (± SD) of 2.83 yr (± 0.39) were used. A sample size calculation for the outcome variable of gastrointestinal transit time was performed to determine the necessary number of cats required for this study using a power of 0.8, an α of 0.05, an effect size of 1.87, and SD of 4.93 using G*Power. This sample size calculation was based on a previous study investigating GE using isotopic markers in cats (Peachey et al., 2000) and indicated a sample size of six cats per treatment would allow for adequate statistical power to maintain a β equal to or greater than 0.80.

In both experiments, cats were randomly allocated into one of two groups (G1, n = 6; G2, n = 6). Each group was randomly assigned to treatment (TRT) or control (CTRL). Cats assigned to TRT received a dose of Phe (44 mg kg⁻¹ BW) mixed with a salmon-flavored lickable cat treats (83 mg kg⁻¹ BW) (Catit Creamy Lickable Cat Treat, Rolf C. Hagen Inc.), while cats on CTRL received an isonitrogenous dose of Ala (23.7 mg kg⁻¹ BW) also mixed into the lickable treat (83 mg kg⁻¹ BW) to ensure the equal molar provision of nitrogen, as protein has satiety-inducing effects. Cats were weighed the day before the experimental period (day −1) for food intake measurement, and GE rate, and macronutrient metabolism assessment to determine their respective dose of TRT or CTRL.

Study 2a: food intake measurement

Study 2a was conducted as a crossover experiment with two periods of 4 d (days 0 to 3). Cats were fed (0700 hours) TRT or CTRL at time = 0 min, and at time = 15 min, the cats were then offered 120% of their daily ration of the commercial diet (T22 Total Grain-Free, Nutram Pet Products) to mimic owner feeding practices. Their daily ration of food was determined based on historical feeding records to maintain BW. The remaining food was periodically weighed to calculate the food intake rate and returned to the cats until all food was eaten. All uneaten food at time 30 min post-feeding was weighed and discarded to measure total food intake. After a 4-d washout period, cats were assigned to the other treatment and a second 4-d experimental period occurred as described above.

Study 2b: Measurement of GE rate and macronutrient metabolism

For study 2b, GE rate and macronutrient metabolism were assessed in a crossover experiment with two periods of 5 d (days 0 to 4) in study 2b. Prior to study 2b, a pilot study was carried out with four neutered male domestic shorthair cats, with a mean BW (±SD) of 5.59 kg (±1.23) and a mean age (± SD) of 2.79 (±0.45) yr, to determine an efficient and accurate sampling technique to measure GE using the ¹³C-OABT. After 24 h of fasting, cats were weighed and placed into individual respiration calorimetry chambers to collect breath samples for GE analysis. GE was assessed using an isotope dilution technique via breath ¹³CO₂ enrichment after ingesting a bolus dose (5 mg·kg⁻¹ BW) of [1-¹³C]-octanoic acid. Once in the chamber and after 30 min of gas equilibration, three fasting respiration/indirect calorimetry measurements and breath samples were taken over three consecutive 20 min periods to determine the resting VCO₂ and volume of O₂ (VO₂), and background enrichment of ¹³C. At time = 0 min, [1-¹³C]-octanoic acid (99%; Cambridge Isotope Laboratories, Inc.; 5 mg kg⁻¹ BW) was added to 50% of their daily metabolizable energy requirement (T22 Total Grain-Free, Nutram Pet Products) and offered to the cat. This meal was consumed within 5 min by all four cats. Expired CO₂ was collected every 20 min for the first 2 h and every 40 min for the next 10 h. Either all (22 breath samples; one breath sample every 20 min from 0 to 2 h and one breath sample every 40 min from 2 to 12 h) or half (11 breath samples; one breath sample every 40 min from 0 to 2 h and every 80 min from 2 to 12 h) of the breath samples collected were used for calculations and then compared to enable an efficient collection approach. Calorimetry data collection (VCO₂ expired and VO₂ consumed) occurred in 4 min periods, every 20 min, and the data over the period were averaged and used for GE analysis. Calorimetry data were collected using Qubit calorimetry software as described previously. Cats spent a total of 12 h within the chambers and were previously adapted to remain in the chambers for 24 h (Hogan et al., 2022).

With the protocol developed from our small pilot trial, study 2b consisted of 5 d (days 0 to 4). During days 0 to 3, cats were fed (0700 hours) their respective dose of TRT or CTRL, and 15 min after treatment consumption, they were fed their daily ration (100%) in amounts to maintain BW based on historical feeding records (T22 Total Grain-Free, Nutram Pet Products). On day 4, after 24 h of fasting, cats were weighed and placed into respiration calorimetry chambers to assess GE and macronutrient metabolism. Cats were then fed (time = 0 min) their corresponding dose of TRT or CTRL. At time = 16 min, [1-¹³C]-octanoic acid (99%; Cambridge Isotope Laboratories, Inc.; 5 mg kg⁻¹ BW) was topically added to 50% of their daily ration and offered to the cat. This meal was usually consumed within 5 min. GE was assessed using half of the breath samples collected as validated in the pilot trial (11 breath samples; one breath sample every 40 min from 0 to 2 h and every 80 min from 2 to 12 h). Samples were collected as described above for both the collection of ¹³CO₂ and determination of VCO₂ and VO₂. Cats stayed in the chamber for an additional 10 h following the completion of breath sample collection, for a total of 22 h of indirect calorimetry. Each calorimetry session consisted of a 1 h fasted period (−60 to 0 min), then the cats received their meal (0 min), followed by a 2-h fed-state (0 to 120 min) and three postprandial periods (6 h each) (120 to 480 min, 480 to 840 min, and 840 to 1,200 min) with the total postprandial state totaling 18 h (120 to 1,200 min) of readings following feeding.

Sample collection, analysis, and calculations

Breath samples (background and enriched samples of CO₂) were collected and stored frozen at −20 °C and analyzed for enrichment of ¹³C as described previously. Enrichment of CO₂ samples was expressed as APE. The rate of F¹³CO₂ expired (mmol min⁻¹) was calculated using the following equation adapted from Peachey et al. (2000) who previously used the ¹³C-OABT to determine GE rate in adult cats:

$${\displaystyle \begin{array}{l}{\displaystyle {\mathrm{F}}^{13}\mathrm{C}{\mathrm{O}}_2(\mathrm{m}\mathrm{m}\mathrm{o}\mathrm{l}{min}^{-1})=\frac{(\mathrm{V}\mathrm{C}{\mathrm{O}}_2\times \mathrm{A}\mathrm{P}{\mathrm{E}}_{\mathrm{C}{\mathrm{O}}_2}\times 44.6)}{100}}\end{array}}$$

The above formula for F¹³CO₂ (mmol min⁻¹) is not adjusted to BW as in study 1. This adjustment was intentionally omitted to facilitate comparisons with previous research on GE in cats using the ¹³C-OABT where BW correction was not applied (Peachey et al., 2000). These authors also calculated the rate of ¹³CO₂ expired per hour, not minute, which explains the absence of a conversion of VCO₂ from mmol min⁻¹ to mmol h⁻¹.

Calculated VCO₂ during fasting, fed, and postprandial states were collected using Qubit calorimetry software and averaged to obtain mean fasting, fed, and postprandial VCO₂ for each cat. Breath samples (background and enriched samples of CO₂) were collected and stored frozen at −20 °C and analyzed for enrichment of ¹³C as described previously. The EE was calculated using the modified Weir equation (Weir, 1949). The production of ¹³CO₂ (mmol min⁻¹) was calculated as done in the pilot trial.

Statistical analysis

Food intake rate, time to finish, and orts data were analyzed using a mixed model via PROC GLIMMIX of SAS (v. 9.4; SAS Institute Inc.), where treatment, day, and their interaction were treated as fixed effects, and cat and period were treated as random effects. Day was treated as a repeated measure within each period. Assumptions about residual normality from the model were assessed through visual inspection and using the Shapiro–Wilk test. A Tukey’s honestly significant difference test was used to separate significant lsmeans.

For the pilot trial, total GE, as assessed by area under the curve (AUC) of ¹³CO₂ (mmol min⁻¹) plotted against time, and time to peak ¹³CO₂ production were analyzed using PROC GLIMMIX in SAS. Treatment (all vs. half) was treated as a fixed effect, and the cat was treated as a random effect. In study 2b, total GE was analyzed for AUC and time to peak using a mixed model via PROC GLIMMIX of SAS, where treatment was treated as a fixed effect, and cat and period were treated as random effects. Assumptions of model residual normality were assessed through visual inspection of residuals and using the Shapiro–Wilk test. A Tukey’s test was used to separate lsmeans when the main effect was significant. Fasted (−60 to 0 min), fed (0 to 120 min), postprandial 1 (120 to 480 min), postprandial 2 (480 to 840 min), postprandial 3 (840 to 1,200 min), and postprandial total (120 to 1,200 min) RQ and EE data were analyzed with PROC GLIMMIX of SAS, using the same aforementioned model. Data are reported as LSM ± SEM. Significance was declared when P < 0.05 and a trend when 0.10 < P ≤ 0.05.

Results

Study 1: the Phe requirement

No effect (P > 0.05) of dietary Phe intake was observed on BW, EE (fed and fasted), and RQ (fed and fasted) (Table 2). The model that best fit the effect of dietary Phe on F¹³CO₂ was the broken-line linear model, both when F¹³CO₂ was expressed as a function of total BW or LSTM. A breakpoint for the mean Phe requirement with excess of Tyr, when F¹³CO₂ as a function of total BW (mmol·kg⁻¹·h⁻¹) was used as the outcome of interest, was identified at 0.32% (34.03 mg·kg⁻¹ BW) with an upper 95% confidence limit (CL) of 0.59% (62.74 mg·kg⁻¹ BW) on a DM basis (Figure 1 and Table 3), on an energy density of 4,200 kcal of metabolizable energy/kg DM calculated using the modified Atwater factors. When expressed on energy density of 4,000 kcal ME/kg DM, the CL would equate to 0.56% on a DM basis. When F¹³CO₂ as a function of LSTM (mmol·kg-LSTM⁻¹·h⁻¹) was used as the outcome of interest, a breakpoint for the mean Phe requirement was found at 0.31% (25.48 mg·kg⁻¹ LSTM) with an upper 95% CL of 0.58% (47.66 mg·kg⁻¹ LSTM) on a DM basis (Figure 1), on an energy density of 4,200 kcal of metabolizable energy/kg DM calculated using the modified Atwater factors. When expressed on energy density of 4,000 kcal ME/kg DM, the CL would be 0.55% on a DM basis.

Table 2.

Body weight, energy expenditure, respiratory quotient, and VCO₂/VO₂ for cats fed diets containing graded levels of Phe (study 1)

Parameters1	Dietary phenylalanine, % DM
	0.29	0.34	0.39	0.44	0.54	0.64	0.74	0.84	Pooled SEM2	P value3
	n = 4	n = 4	n = 4	n = 4	n = 4	n = 4	n = 4	n = 4
BW, kg	5.37	5.36	5.37	5.38	5.38	5.34	5.34	5.36	0.444	0.69
REE, kcal/kg BW	26.75	26.26	29.54	29.46	28.06	29.47	30.52	27.23	2.484	0.73
REE, kcal/kg BW0.67	46.12	45.34	51.13	51.19	48.46	50.92	53.03	47.01	3.917	0.73
FEE, kcal/kg BW	32.15	36.53	35.97	37.08	38.95	38.73	36.07	39.40	3.706	0.41
FEE, kcal/kg BW0.67	55.22	62.89	62.09	64.61	66.42	65.06	62.05	68.16	5.369	0.52
Fasted RQ	0.74	0.76	0.76	0.75	0.74	0.76	0.75	0.76	0.013	0.76
Fed RQ	0.83	0.83	0.83	0.83	0.82	0.83	0.83	0.83	0.007	0.99
Fed VCO2, mL/min	21.26	22.96	22.41	23.39	23.11	23.11	23.89	24.78	1.413	0.42
Fed VO2, mL/min	25.48	27.37	27.07	28.11	28.31	28.06	28.75	29.82	1.720	0.40

¹BW = body weight, FEE = fed-state energy expenditure, REE = resting energy expenditure, RQ = respiratory quotient, VCO₂ = volume of CO₂, VO₂ = volume of O₂.

²Standard error of the mean.

³Significantly different when P ≤ 0.05 and a tendency declared when 0.05 < P ≤ 0.10.

Figure 1.

Influence of dietary Phe on production of ¹³CO₂ from Phe oxidation (F¹³CO₂) on adult cats using the DAAO technique. The F¹³CO₂ is presented as a function of total body weight (A) and lean soft tissue mass (B). The breakpoint (dashed arrows) represents the estimated mean Phenylalanine requirement when Tyr is supplied in excess (0.32% in A and 0.31% in B). Solid arrows represent the upper 95% CL for Phenylalanine requirement (0.59% in A and 0.58% in B). A sample size of n = 4 was achieved for all diets.

Table 3.

Dietary Phenylalanine requirement, in the presence of excess of Tyrosine, for adult cats at maintenance as recommended by AAFCO, FEDIAF, NRC, and estimated in the present study

Units	AAFCO1	FEDIAF2	NRC3 AI4	NRC RA	Present Study
					MR	CL
g/100 g DM5	0.42	0.53/0.40	0.40	0.40	0.31	0.56
g/Mcal ME	1.05	1.33/1.00	1.00	1.00	0.76	1.33
mg/kg BW					34.03	62.74
						(56.89)6

¹Association of American Feed Control Officials Manual. 2023.

²European Pet Food Industry Federation (2023) Nutritional guidelines for complete and complementary pet food for cats and dogs. Values depend on maintenance energy requirements 75 kcal/kg^0.67 or 100 kcal/kg^0.67, respectively.

³Nutrient requirements of dogs and cats; National Research Council, 2006.

⁴AI = adequate intake, RA = Recommended allowance, MR = Minimal requirement, CL = upper 95% confidence limit.

⁵Values for g 100 g DM⁻¹ are determined assuming a dietary energy density of 4,000 kcal ME kg⁻¹.

⁶Values in parentheses represent NRC recommendation for Phe requirement for adult cats at maintenance converted from mg·kg⁻¹ BW^0.67 to mg·kg⁻¹ BW using the average BW of cats in the present study.

Study 2

Study 2a: measurement of food intake

Mean intake rate (g min⁻¹), time to finish (min), orts (g), and intake (% eaten) are reported in Table 4. The mean intake rate did not differ between TRT and CTRL (P > 0.05) but tended (P < 0.10) to be slower for cats on CTRL compared to cats on TRT. There was no treatment-by-day interaction effect (P > 0.05) for intake rate. However, the intake rate was slower on day 2 when compared to day 1 (P < 0.05), but intake rates for days 1 and 2 were similar (P > 0.05) to days 3 and 4. No effects of treatment, day, and their interaction were observed for time to finish, orts (g), and intake (% eaten) (P > 0.05).

Table 4.

Mean daily food intake, time to finish, orts and intake (% eaten) for all cats (n = 12) when evaluating the effects of dietary Phe (44 mg kg⁻¹ BW) on gastric empyting rate (study 2a)

Parameter	Treatment1			Day					P value2
	CTRL	TRT	Pooled SEM	1	2	3	4	Pooled SEM	Treatment	Day	Treatment*day
Intake Rate, g min−1	5.67	6.07	0.518	6.27a	5.62b	5.64ab	5.95ab	0.529	0.08	0.04	0.66
Time to finish, min	14.11	13.21	1.959	13.00	13.83	13.85	13.97	1.969	0.30	0.61	0.16
Orts, g	0.44	0.57	0.484	0.45	0.23	0.35	0.99	0.476	0.80	0.23	0.99
Intake, % eaten	99.36	99.27	0.618	99.41	99.71	99.55	98.59	0.620	0.89	0.20	0.99

¹CTRL = control; TRT = treatment; SEM = standard error of the mean.

²Significantly different when P ≤ 0.05 and a tendency declared when 0.05 < P ≤ 0.10.

^a,bValues in a row with different superscripts are different within Day (P ≤ 0.05).

Study 2b: GE rate and macronutrient metabolism

The enrichment of ¹³CO₂ was successfully captured in all breath samples when using a dose of 5 mg·kg⁻¹ BW of [1-¹³C]-octanoic acid. Baseline ¹³CO₂ enrichment was defined as the ¹³CO₂ enrichment of the first postprandial breath sample. A return to baseline ¹³CO₂ enrichment was defined as an 80% recovery from peak ¹³CO₂ production. In the pilot study, all cats returned to baseline enrichment by 12 h following meal consumption. When evaluating the frequency of breath samples required to capture total GE and time to peak ¹³CO₂ production, reducing sample collection from every 20 min to every 40 min for the first 2 h of breath collection and from every 40 min to every 80 min for the following 10 h of collection produced similar estimates of total GE (AUC) (118.30 vs. 118.76; P > 0.05). Even though all cats returned to baseline (defined as an 80% recovery from peak ¹³CO₂ production) in the pilot trial, only five out of twelve returned to baseline enrichment in study 2b, and thus, 7 out of 12 cats were removed from statistical analysis.

Total GE as assessed by the AUC of F¹³CO₂ (mmol min⁻¹) plotted against time (min) and time to peak ¹³CO₂ production (min) are reported in Table 5. Total GE did not differ between TRT or CTRL (P > 0.05). Time to peak ¹³CO₂ production (min), signifying peak GE did not differ between TRT and CTRL (P > 0.05) but tended (0.05 < P ≤ 0.10) to be slower for cats on TRT than cats on CTRL. In Table 6, data for mean fasted, fed, and three postprandial and total postprandial RQ values are presented for all 12 cats. No differences were detected in RQ values for cats on TRT compared to CTRL for fasted (−60 to 0 min) and all postprandial (120 to 480 min, 480 to 840 min, 840 to 1,200 min, and 120 to 1,200 min) measurements (P > 0.05). However, in the fed state (0 to 120 min), cats on TRT had a higher RQ value than cats on CTRL (P < 0.05). No differences (P > 0.05) for the effect of treatment were detected for EE when based on BW (Table 6). When evaluating the effect of treatment for EE based on LSTM, no differences (P > 0.05) were detected, but EE for cats on CTRL during 840 to 1,200 min of postprandial data collection tended (P < 0.10) to be lower than cats on TRT.

Table 5.

Mean production of ¹³CO₂ calculated by AUC produced from ¹³CO₂ expired (mmol min⁻¹) plotted against time and time to peak for cats that returned to baseline enrichment (n = 5) when evaluating the effects of dietary Phe (44 mg kg⁻¹ BW) on GE rate when using the ¹³C-OABT (study 2b)

GE parameter	Treatment1			P value2
	CTRL	TRT	Pooled SEM
AUC, mmol min−1 min	102	97.0	18.1	0.81
Time to peak 13CO2 production, min	260	468	65.8	0.06

¹CTRL = control; TRT = treatment; SEM = standard error of the mean.

²Significantly different when P ≤ 0.05 and a tendency declared when 0.05 < P ≤ 0.10.

Table 6.

Mean fasted, fed, and postprandial EE and RQ for all cats (n = 12) when evaluating GE rates using ¹³C-OABT. Cats were fed 50% of their daily ration at time 15 min (study 2b)

Calorimetry parameter1	Treatment2
	CTRL	TRT	Pooled SEM	P value2
Fasted EE, kcal kg−1 BW.d−1 (−60 to 0 min)	37.12	40.64	2.799	0.31
Fed EE, kcal kg−1 BW d−1 (0 to 120 min)	38.88	42.04	2.991	0.26
Postprandial EE 1, kcal kg−1 BW d−1 (120 to 480 min)	38.08	39.09	2.010	0.63
Postprandial EE 2, kcal kg−1 BW d−1 (480 to 840 min)	40.73	42.30	2.102	0.37
Postprandial EE 3, kcal kg−1 BW d−1 (840 to 1,200 min)	39.02	42.24	2.109	0.16
Postprandial EE total, kcal kg−1 BW d−1 (120 to 1,200 min)	39.26	40.77	1.922	0.40
Fasted EE, kcal kg-LSTM−1 d−1 (−60 to 0 min)	47.20	49.84	2.832	0.36
Fed EE, kcal kg-LSTM−1 d−1 (0 to 120 min)	49.70	51.65	3.191	0.38
Postprandial EE 1, kcal kg-LSTM−1 d−1 (120 to 480 min)	48.85	47.84	2.067	0.46
Postprandial EE 2, kcal kg-LSTM−1 d−1 (480 to 840 min)	52.04	51.18	2.367	0.41
Postprandial EE 3, kcal kg-LSTM−1 d−1 (840 to 1,200 min)	49.69	51.58	1.961	0.08
Postprandial EE total, kcal kg-LSTM−1 d−1 (120 to 1,200 min)	50.18	48.69	2.343	0.30
Fasted RQ (−60 to 0 min)	0.73	0.74	0.006	0.22
Fed RQ (0 to 120 min)	0.71b	0.73a	0.004	<0.01
Postprandial 1 RQ (120 to 480 min)	0.78	0.77	0.004	0.26
Postprandial 2 RQ (480 to 840 min)	0.79	0.79	0.005	0.85
Postprandial 3 RQ (840 to 1,200 min)	0.76	0.76	0.006	0.86
Postprandial RQ total (120 to 1,200 min)	0.77	0.77	0.004	0.55

¹BW = body weight; EE = energy expenditure; LSTM = lean soft tissue mass; RQ = respiratory quotient.

²CTRL = control; TRT = treatment; SEM = standard error of the mean.

³Significantly different when P ≤ 0.05 and a tendency declared when 0.05 < P ≤ 0.10.

^a,bValues in a row with different superscripts are different (P ≤ 0.05).

Discussion

To our knowledge, this is the first study to empirically determine the minimum Phe requirement, with excess of Tyr, in adult cats. To account for variability within the population, we compared the estimated 95% CL with the level for AI recommended in the NRC (2006). Our 95% CL was ~47% higher than the AI put forth by the NRC (2006) and ~40% higher than the Phe requirements suggested by the Association of American Feed Control Officials (AAFCO, 2023) for adult cats at maintenance. There were no changes in calorimetry responses across all dietary Phe levels evaluated, signifying that differing levels of Phe did not affect energy or macronutrient metabolism and that all changes seen in F¹³CO₂ production were a result of differences in the dietary intake of Phe. These results are similar to those observed in carbon oxidation in dogs (Mansilla et al., 2018, 2020a,b; Templeman et al., 2019; Sutherland et al., 2020), where EE and RQ also remained unchanged across different dietary levels of the test AA, and the requirement of the AA in question was found to be higher than estimates provided by growth and nitrogen balance research. Both growth and nitrogen balance methodologies lack the sensitivity required for the stability of body protein pools in adult animals.

The findings of study 1 suggest that the AI and RA for Phe recommended by the NRC (2006) and by AAFCO (2023) for commercial feline diets are underestimated. Only neutered male cats were used in the present study, so future investigation into Phe requirements in different neuter statuses and sex is warranted. It is important to note that while Phe concentrations in the basal diet were provided from intact protein sources, graded levels of Phe across experimental diets were achieved with crystalline Phe, assumed to have 100% bioavailability. As such, the estimate of the requirement determined in our study cannot be classified as a minimal (physiological) requirement and would fall somewhere between the MR and RA of the NRC (2006). Commercial cat foods utilize intact proteins to fulfill AA requirements and can be highly processed, which can reduce AA bioavailability (van Rooijen et al., 2013). Thus, recommendations for commercial pet food should be higher than the current determination.

As ¹³C-Phe is used as the main indicator AA in carbon oxidation studies, the minimum requirement for Phe must be defined for future IAAO studies to ensure Phe is not limiting protein synthesis and all responses are due to the intake of the test AA. Of the indispensable AA for adult cats, the NRC has only defined MR for Tau, Lys, Met, and total sulfur AA (NRC, 2006). We recently applied the IAAO technique to determine the minimum methionine requirement in adult cats (Pezzali et al., 2024) and also found higher estimates than those proposed by the NRC (2006) and AAFCO (2023). While the Met requirement study (Pezzali et al., 2024) was conducted before the current study, the estimates of the Phe requirement in the present study indicate that Pezzali et al., (2024) provided sufficient amounts of Phe when estimating the methionine requirement. The protocol utilized in the current study was successful due to modifications on the length of the adaption period to the basal diet (7 d) and experimental diets (1 d) from the suggestions proposed by Pezzali et al., (2024) as difficulties were faced in achieving the ideal sample size with longer adaptation periods. Thus, future IAAO studies should follow this adapted protocol.

While we were successful at applying the DAAO technique in adult cats, it is important to emphasize that the breakpoint was identified between diets 1 and 2, making it difficult to determine a clear curve response as a minimum of three points are necessary to satisfactory define a line (Pencharz and Ball, 2003). Thus, a basal diet with a dietary Phe concentration lower than 0.29% DM would be ideal to define a breakpoint with three dietary levels below it. However, developing a palatable semi-synthetic diet lower in Phe presents challenges. Cats have a preferential diet selection for protein (Hewson-Hughes et al., 2011), thus, diets both low in protein content and formulated using crystalline AA, as opposed to intact animal-based proteins, are not very palatable and may result in refusal over time. Our approach represented the best compromise between keeping Phe low and ensuring the cats’ dietary acceptance. Additionally, it is important to point out that the high SEM value for the breakpoint estimate (0.32 ± 0.085%) has had an influence on the wide range observed for the 95% CL. Even though the between-subject variance in amino acid requirements is significantly high, and we attempt to account for it using a crossover experimental design, a larger sample size may be necessary for cats to enhance the accuracy of the breakpoint estimate.

The work presented in study 2, first sought to validate the use of the ¹³C-OABT in cats, to then be utilized for the evaluation of the effects of dietary Phe supplementation on GE. Feline obesity affects upwards of 60% of the global cat population (Cave et al., 2012; Courcier et al., 2012; Wall et al., 2019). With multiple comorbidities resulting from obesity, we investigated the potential use of supplemental Phe for the prevention, rather than treatment of feline obesity, through its potential satiety-inducing effects. The pilot data indicated that a dose of 5 mg·kg⁻¹ BW of ¹³C-Octanoic acid was successful at capturing GE in cats, with a return to baseline ¹³CO₂ enrichment within 12 h of meal consumption. However, under the same sampling protocol as the pilot trial, in study 2a, only 5 of 12 cats returned to baseline. Of the four cats used in the pilot trial, two were used in study 2a, and only one of the cats had a successful return to baseline. While stress can affect GE rates (Gué et al., 1989) this is unlikely in our studies, as the cats had been previously acclimated to the environment, people, and calorimetry chamber system (Gooding et al., 2012) and exhibited no signs of stress throughout both the pilot study and the GE study. This suggests the potential for GE rates to be influenced by day-to-day variation. Future studies should evaluate the recovery of ¹³CO₂ from the administered dose, as this investigation may elucidate the significant variability observed across all cats, and the reasons behind the return to baseline enrichment in only 5 out of the 12 cats.

While Phe supplementation at 44 mg·kg⁻¹ BW did not influence food intake, macronutrient metabolism, or total GE, there was a tendency for Phe to delay time to peak GE suggesting that Phe may reduce the rate of GE. Cats on Phe tended to have peak emptying at roughly 3.7 h later than cats on Ala, signifying that Phe may play a role in regulating peak GE. Nonetheless, despite the limited sample size, the results reveal a trend, suggesting that further exploration is warranted to investigate whether Phe truly has the potential to extend the time to peak GE. Future studies should also consider a different basal diet, as it has been speculated that CCK secretion may be adapted to diet composition; therefore, dietary protein has been suggested as secretagogues of CCK in cats (Backus et al., 1995). Backus et al. (1995) found that intact protein and individual AAs elevate plasma CCK in cats. Thus, it may be warranted to use diets lower in protein in future studies to confirm that a delay in GE is in response to Phe supplementation stimulating CCK release and not a result of dietary protein intake.

The doses of Phe used in studies 2a and 2b were determined based on previous research by Backus and Rogers (1996), where the provision of 44 mg·kg⁻¹ BW of Phe elevated plasma CCK in cats. However, the effects of increased plasma CCK on the satiety response were not evaluated, so it is difficult to conclude if this dose is sufficient to elicit an adequate elevation in CCK concentrations to further cause a delay in GE and thereby reduce feed intake through the promotion of satiation. It is important to note that studies 2a and 2b did not evaluate plasma CCK concentrations of the cats following Phe consumption, as the cats used in our study required sedation to safely collect blood, and blood collection should be done in the fed state when evaluating CCK; thus, future research should measure plasma CCK levels following Phe consumption to determine if Phe truly elevates plasma CCK under these conditions, as well as determine if CCK follows GE patterns in cats. It is important to note that AA seems to impact GE by osmotic mechanism, with concentrations <80 mM being below the predicted threshold range for duodenal osmoreceptors (Stephens et al., 1975). Phe and Tyr differ from the other AA as they are postulated to have these effects when at concentrations <80 mM, however, the dose of Phe used was ~10 mM·kg⁻¹ BW, which may have been too low to see significant effects. It may also be worth investigating additional doses of Phe to evaluate the effects on a satiety response and food intake. Furthermore, while CCK is one of many hormones involved in satiety induction, it is possible that stimulation of multiple hormones together may be necessary to see a reduction in food intake, GE, and macronutrient metabolism in the domestic cat.

Conclusion

The breakpoint for the mean Phe requirement with the presence of excess Tyr, was identified at 0.32% (34.03 mg·kg⁻¹ BW) on a DM basis on an energy density of 4,200 kcal of metabolizable energy/kg DM calculated using the modified Atwater factors. The upper 95% CL (0.59% DM) estimated in this study for Phe in the adult cat is higher than the recommendations made by the NRC (0.40% DM) and AAFCO (0.42% DM). The ¹³C-OABT combined with indirect calorimetry chambers to collect breath samples can be used to evaluate GE in adult cats. A bolus dose of Phe did not influence food intake, GE, or macronutrient metabolism as hypothesized. While there was a trend for cats on Phe to have delayed times to peak GE, our results should be interpreted with caution, and future studies should investigate the effects of alternative doses and explore the length of supplementation of Phe on food intake, GE, and macronutrient metabolism together with satiety hormones.

Glossary

Abbreviations

¹³C-OABT

¹³C-octanoic acid breath test

AA

amino acid

AAFCO

American Association of Feed Control Officials

AI

adequate intake

APE

atom percent excess

AUC

area under the curve

BW

body weight

CCK

cholecystokinin

CL

confidence limit

DAAO

direct amino acid oxidation

DM

dry matter

EE

energy expenditure

F¹³CO₂

flux of ¹³CO₂

FEDIAF

European Pet Food Industry Federation

FEE

fed-state energy expenditure

GE

gastric emptying

IAAO

indicator amino acid oxidation

LSM

least square means

LSTM

lean soft tissue mass

MR

minimal requirement

NRC

National Research Council

RA

recommended allowance

REE

resting energy expenditure

RQ

respiratory quotient

VCO₂

volume of CO₂

VO₂

volume of O₂

Author Contributions

Jocelyn G. Lambie and Júlia G. Pezzali (Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Validation, Visualization, Writing—original draft), Adronie Verbrugghe (Investigation, Resources, Writing—review & editing), Taylor L. Richards (Investigation, Formal analysis, Writing—review & editing), Jennifer L. Ellis (Formal analysis, Writing—review & editing), and Anna K. Shoveller (Conceptualization, Funding acquisition, Methodology, Resources, Supervision, Validation, Writing—review & editing).

Conflict of interest statement.

J.G.L., J.G.P., T.L.R., and J.L.E. have no conflicts of interest. A.K.S. is the Champion Petfoods Chair in Canine and Feline Nutrition, Physiology and Metabolism and additionally consults for Champion Petfoods. A.K.S. has received honoraria and research funding from various pet food manufacturers and ingredient suppliers in addition to provincial and federal granting agencies and declares that they serve on the Trouw Nutrition and Champion Petfoods Scientific Board. A.V. is the Royal Canin Veterinary Diets Endowed Chair in Canine and Feline Clinical Nutrition, serves on the Health and Nutrition Advisory Board for Vetdiet, and has received honoraria and research funding from various pet food manufacturers and ingredient suppliers in addition to provincial and federal granting agencies.