Introduction
Metabolite concentrations have been used to evaluate the health status of animals and humans, and as a tool to provide an understanding of the complex interplay of metabolism. However, concentrations of metabolites are static measurements and do not reveal important kinetic movement into (appearance) and out of (disappearance) a particular metabolic pool. Unsurprisingly, plasma amino acid (AA) concentrations are poorly correlated with estimates of AA and protein requirements(Reference Mitchell, Becker and Jensen1,Reference Sohail, Cole and Lewis2) , and thus, are considered an insensitive method to estimate AA requirements(Reference Pencharz and Ball3). The use of stable isotopes in human and animal research has provided a highly sensitive method to measure kinetics of metabolites(Reference Kim, Suh and Lee4). Furthermore, the use of stable isotope tracers together with indirect calorimetry (to quantify the volume of CO2 produced; VCO2) have made it possible to quantify the rate of oxidation of substrates, such as, but not limited to, AA. The measurement of AA oxidation can be used to determine the requirement of AA, where the oxidation of an indicator AA, such as L-[1-13C]-Phenylalanine (L-[1-13C]-Phe), at varying intakes of the test AA is used as the biological outcome(Reference Elango, Ball and Pencharz5). After the invention of IAAO in 1983(Reference Kim, McMillan and Bayley6), which was first applied in pigs, the IAAO was subsequently applied in humans using intravenous 13C-Phe to determine the Lys requirement(Reference Zello, Pencharz and Ball7). The IAAO protocol with oral provision of isotope was then validated(Reference Bross, Ball and Pencharz8), making it less invasive than the intravenous approach. This less invasive approach was further supported in a following study(Reference Kriengsinyos, Wykes and Ball9) in which identical lysine requirement estimates were found in humans repeatedly fed 13C-Phe or intravenously supplied 13C-Phe. Since then, the IAAO technique has been broadly used under different states of health(Reference Courtney-Martin, Bross and Raffi10–Reference Mager, Wykes and Roberts12) and in different species(Reference Tabiri, Bertolo and Ball13–Reference Wei, Chen and Xinmei17) due to its non-invasive and highly sensitive nature.
More recently, we have worked on applying the IAAO technique in adult cats to improve our limited understanding of AA requirements in obligate carnivores and more specifically, the domestic cat. First, we developed a semi-synthetic diet to use in carbon oxidation studies(Reference Pezzali and Shoveller18,Reference Pezzali, Bullerwell and Dancy19) and confirmed that enrichment of 13CO2 can be captured using respiration chambers during an isotope dilution study in cats that received 13C-Phe orally rather than intravenously(Reference Pezzali, Mahroukh and Courtney-Martin20). However, cats failed to achieve a steady state of 13CO2 enrichment in breath using oral priming (4⋅8 mg/kg) and constant (1⋅04 mg/kg) doses of L-[1-13C]-Phe(Reference Pezzali, Mahroukh and Courtney-Martin20), which were provided over a thirteen small meal regimen as reported in dogs(Reference Shoveller, Danelon and Atkinson15). The oxidation of L-[1-13C]-Phe can only be calculated when an equilibrium of 13CO2 enrichment in breath is reached, which is achieved using the constant infusion-isotope dilution approach. Equilibrium, also referred to as a steady state, is achieved when the rate of appearance of a metabolite in a specific body pool is equal to its rate of disappearance. However, isotopic equilibrium may take several hours to be reached if the pool size of the metabolite is large in relation to its turnover rate(Reference Issekutz, Paul and Miller21), which may present practical and ethical concerns. To overcome this challenge, a priming dose of L-[1-13C]-Phe is given in conjunction with a constant infusion of L-[1-13C]-Phe in carbon oxidation studies in humans(Reference Di Buono, Wykes and Ball22), pigs(Reference Moehn, Bertolo and Pencharz23) and dogs(Reference Shoveller, Danelon and Atkinson15, Reference Mansilla, Gorman and Fortener24). However, this approach only reduces the time to reach the isotopic steady state if the prime-to-constant ratio of the tracer is adequate to the pool size and turnover of the substrate(Reference Wolfe25). Furthermore, 13CO2 produced from oxidation of L-[1-13C]-Phe enters the bicarbonate pool before exhalation. However, the rate of exchange between 13CO2 and the unlabelled bicarbonate pool is slow and may delay the time to reach 13CO2 steady state in breath. Thus, priming the bicarbonate pool may be used as an option to reduce the time to reach the steady state of labelled expired CO2(Reference Allsop, Wolfe and Burke26). The ideal priming dose of bicarbonate has yet to be determined in adult cats. Developing an isotope protocol to achieve equilibrium of 13CO2 in breath of cats, when L-[1-13C]-Phe is used as the tracer, is the next step to allow the successful application of carbon oxidation techniques in this species. Therefore, the aim of the present study was to develop an oral isotope protocol for adult cats that would produce steady state in expired 13CO2 during the time frame of carbon oxidation studies.
Materials and methods
The present study was carried out according to the guidelines for animal care and use provided by the Canadian Council on Animal Care. All ethical and animal-related aspects of the pilot trials were approved by the University of Guelph Animal Care Committee (AUP#4424).
Animal and housing
One adult (2 years old) neutered male purpose bred cat (Marshall Biosciences, North Rose, NY, USA) was used. The cat was housed with other purpose bred cats (n 18) in an indoor free-living environment (7⋅1 m × 5⋅8 m) located in the Animal Biosciences Department at the University of Guelph. The room was approved for cat inhabitation by the Chief Veterinary Inspector of the Ontario Ministry of Agriculture, Food and Rural Affairs (OMAFRA) under the Animals for Research Act prior to the arrival of the cats. The environment was enriched with perches, toys, hide boxes, beds, scratching pots and climbing apparatuses. The light (12 h light:12 h cycle), temperature (20 °C) and humidity (40–60 %) were controlled and monitored daily. Cleaning of the litter boxes and exterior surfaces was performed once daily at the same time. Cats were socialised with a familiar individual five days a week, for 2 h each day.
Study design and diet
The cat used in the study transitioned from a commercial dry diet (T22 Total Grain-Free, Nutram Pet Products, Elmira, ON) to a commercial wet diet (Friskies Paté Salmon Dinner, Purina Wet Cat Food, Purina, St. Louis, MO; Metabolisable energy=1151 kcal/kg; moisture, max = 78 %; crude protein, min = 10 %; crude fat, min = 5 %; crude fibre, max = 1 %, ash, max = 3⋅3 %) over a 6-d period, where the intake of the wet commercial diet gradually increased. The phenylalanine (Phe) and the tyrosine (Tyr) content of the commercial diet were determined via hydrolysis (AOAC, 2012; method 994⋅12) using ultraperformance liquid chromatography (Waters Corporation, Milfor, MA, USA). The cat was then fed 100 % of its daily energy intake to maintain body weight (BW; 269 kcal/d), based on historical feeding and BW records. Food was provided in two equal daily feedings (07:30 and 16:00 h) throughout the study. Water was provided ad libitum throughout the study from standing and free-flowing water.
Two separate pilot trials (1 and 2) were conducted. Three isotope protocols (treatments) were tested within each pilot trial. Each treatment was replicated three times using the same cat, totalling three periods. In each period, the order of treatments was randomly assigned. The different isotope and sample collection protocols evaluated in each pilot trial are described below.
Pilot trial 1
The objective of this first pilot trial was to determine whether modifying our original isotope protocol(Reference Pezzali, Mahroukh and Courtney-Martin20), by either adding a priming dose of NaH13CO3 or increasing the priming or constant dose of L-[1-13C]-Phe, would result in a steady-state condition. The cat underwent a 2-d feeding regimen: (1) d 0: regular feeding regimen as described above and (2) d 1: IAAO study day where treatments were tested. The cat was fed the same diet through the study as only the effect of isotope protocol on the enrichment of 13CO2 in breath was being investigated. Thus, the usual 2-d dietary adaption period used in IAAO studies(Reference Moehn, Bertolo and Pencharz23) was not required. BW was measured the morning of each IAAO study day to ensure accurate delivery of the isotope dose. This 2-d feeding regimen was repeated 9 times (3) times within each experimental period to achieve three replicates per treatment. A similar IAAO feeding and breath collection protocol as described in our previous study was applied(Reference Pezzali, Mahroukh and Courtney-Martin20). Briefly, three fasting respiration/indirect calorimetry measurements were collected, followed by the feeding protocol. Thirteen meals were offered, corresponding to 50 % of the cat's food allowance; after completing each IAAO, the cat was fed the remaining 50 % of its daily food intake. The first three meals were fed every 10 min (0, 10 and 20 min) to achieve fed state and the following ones were fed every 25 min. Background enrichment was determined by the collection of CO2 samples over three consecutive 25 min period after fed state was achieved (45, 75 and 90 min) and before the tracer protocol began. A priming dose of bicarbonate was top-dressed on the sixth meal combined with the priming dose of L-[1-13C]-Phe (99 %, Cambridge Isotope Laboratories, Inc., Tewksbury, MA). A constant dose was given simultaneously and continued throughout the remaining meals. Three isotope protocols (treatments) were tested (A, B and C; Table 1). All treatments contained a similar priming dose of NaH13CO3 (0⋅176 mg/kg) (99 %, Cambridge Isotope Laboratories, Inc., Tewksbury, MA). The priming dose of NaH13CO3 was derived based on the priming dose utilised in IAAO studies in humans(Reference Di Buono, Wykes and Ball22). Treatment A followed the priming (4⋅8 mg/kg) and constant (1⋅04 mg/kg) doses of L-[1-13C]-Phe as we previously used(Reference Pezzali, Mahroukh and Courtney-Martin20) to determine whether the failure to achieve 13C steady state in the breath of cats(Reference Pezzali, Mahroukh and Courtney-Martin20) was due to an improper prime to constant ratio of L-[1-13C]-Phe or simply due to the need to prime the CO2 pool prior to L-[1-13C]-Phe provision. In treatments B and C, the priming (9⋅4 mg/kg) and constant (2⋅4 mg/kg) dose of L-[1-13C]-Phe were increased, respectively, based on the doses used for dogs(Reference Shoveller, Danelon and Atkinson15,Reference Mansilla, Gorman and Fortener24) .
Isotope protocol for pilot trials 1 and 2
NaH13CO3: 13C-Sodium bicarbonate; L-[1-13C]-Phe: L-[1-13C]-Phenylalanine.
* Isotope dosing solutions were top-dressed on the respective meals.
† Fed background of 13CO2 (n 3) was determined by collecting breath samples after the third, fourth and fifth meals.
‡ Fasted background of 13CO2 (n 3) was determined by collecting breath samples before meal provision (time: −50, −25 and 0 min) and fed background of 13CO2 (n 1) was determined by collecting one breath sample after the third meal (time: 45 min). The detailed protocol is presented in Fig. 1.
Pilot trial 2
Having established the ideal prime and constant doses of L-[1-13C]-Phe, we proceed to pilot 2, in which we aimed to determine the ideal priming dose of NaH13CO3. The same 2-d feeding regimen was used as described above. During IAAO, the feed regimen was kept similar, but the time of isotope provision and breath sample collection for determination of background of 13CO2 was modified (Fig. 1). Three breath samples were collected prior to feeding to determine the fasted background of 13CO2 (−50, −25 and 0 min), and one breath sample was collected after fed state was achieved (45 min) to determine the fed background of 13CO2. Three isotope protocols were tested (D, E and F; Table 1). Three priming doses of NaH13CO3, top-dressed on the fourth meal, were tested (D: 0⋅264 mg/kg; E: 0⋅352 mg/kg; F: 0⋅44 mg/kg), while the priming (4⋅8 mg/kg) and constant (1⋅04 mg/kg) doses of L-[1-13C]-Phe were kept similar across treatments. The priming dose of L-[1-13C]-Phe was top-dressed on the fifth meal and a constant dose was given simultaneously and continued throughout the remaining meals.
Feeding, isotope and sample protocol (treatment F in pilot trial 2) proposed to be utilised in IAAO studies in cats. aEach meal represented one-thirteenth of half of the daily food intake for the cat. bPriming dose of NaH13CO3 was top-dressed on the fourth meal (time: 45 min). cPriming dose of L-[1-13C]-Phenylalanine (L-[1-13C]-Phe) was top-dressed on the fifth meal. The continuous dose of L-[1-13C]-Phe started on the fifth meal with the priming dose, followed by continuous supply through the remaining meals. dIC: indirect calorimetry. Three 25-min measures of respiratory gases were obtained prior to feeding to obtain the resting volume of CO2 produced (VCO2). Starting at 45 min, VCO2 was measured in 25-min intervals for the duration of the study. eThree 25-min breath samples collection for 13CO2 background were obtained at −50, −25 and 0 min (fasted state) before food and isotope provision. One breath sample was collected at time −45 min before isotope provision for determination of 13CO2 background during fed state. Breath samples were then collected every 25 min for the duration of the study.
Breath samples analysis
Samples of CO2 were collected by trapping subsamples of expired CO2 in 8 ml of 1M NaOH over 25-min periods. The samples were transferred and retained in a 10 ml vacutainer tube (#366430 BD) that was evacuated to prevent dilution of 13CO2 and stored at room temperature until analysis. Analysis of 13C enrichment in breath CO2 samples was done at the Environmental Isotope Laboratory, University of Waterloo (200 University Ave W, Waterloo, ON, Canada) using a Gasbench II interfaced with a Delta V Plus mass spectrometer (Thermo Scientific, Bremen, Germany). Enrichments were expressed above background samples (Atom percent excess, APE).
Statistical analysis
A sample size of one is commonly utilised to assess the dynamics of metabolites in vivo in pilot trials(Reference Rakotoambinina, Marks and Badran27). Thus, a single cat was utilised in pilot trials 1 and 2 to comply with the three Rs principle of animal experimentation(Reference Russell and Burch28). Treatments were replicated using the same cat to account for variation between days. Isotopic steady state was defined as the enrichment of 13CO2, as APE, remaining constant in at least the last three breath samples. The APE was fitted against meal number (offered in 25-min intervals) to determine the number of meals necessary, within each isotope protocol, to achieve steady state of 13CO2. Steady state was evaluated by visual inspection, by regression analysis using add-in Analysis ToolPak in Microsoft Office Excel 2020 and by competing statistical models, namely broken-line linear (BLL) or broken-line quadratic (BLQ) model using PROC NLMIXED in SAS (SAS Inst., Cary, NC). Models were compared based on the Bayesian information criterion (BIC), where the smaller the value, the better the fit to the model(Reference Milliken and Johnson29). Differences between fasted and fed background enrichments in pilot trial were analysed using PROC GLIMMIX with physiological state (fasted v. fed) as the fixed effect. Statistical difference was declared when P < 0⋅05.
Results
The cat remained healthy and maintained BW throughout both pilot trials (data not shown). In every IAAO study day, all meals were consumed immediately after each feeding. In pilot trial 1, the slope of the line for breath 13CO2 enrichment data for the last three samples was not significantly different from zero for treatment A (P = 0⋅14), B (P = 0⋅10) and C (P = 0⋅16). The coefficient of variation (CV) for the last three samples was the lowest for treatment A (5⋅06 %) followed by treatments B (7⋅65 %) and C (11⋅59 %), which are considered high CV for plateau enrichment of 13CO2(Reference Bross, Ball and Pencharz8). Thus, even though the slope was not significantly different from zero, we did not feel confident to declare that steady state was achieved due to the high CV and as enrichment was still rising through numerical and visual inspection (Fig. 2). Thus, BLL and BLQ analysis were performed to provide an additional method to quantitatively assess isotopic plateau in CO2. The model that best fit the enrichment of 13CO2 was the BLL for all treatments in pilot 1 (lowest BIC). The breakpoints estimated occurred at approximately meal 12 for treatments A and C and at meal 11 for treatment B (Fig. 2). In pilot 2, the slope of the line for breath 13CO2 enrichment data for the last three samples was significantly different from zero (P = 0⋅04) in treatment D, but it was not in treatments E (P = 0⋅08) and F (P = 0⋅49). The CV for the last three samples for treatment F was the lowest (1⋅11 %) followed by D (4⋅32 %) and E (7⋅43 %). The model that best fit the enrichment of 13CO2 was the BLL for treatments D and E and BLQ for treatment F. The asymptote occurred at approximately meals 10, 9 and 8 for treatments D, E and F, respectively (Fig. 3). No differences (P = 0⋅30) were observed in fasted and fed background enrichment of 13CO2 (1⋅102 v. 1⋅101 % ± 0⋅001, least square means ± sem) evaluated in pilot trial 2.
Pilot trial 1: (a) visual inspection (values are x̄ ± sd) and (b) fitted broken-line linear model for 13CO2 expressed as atoms percent excess (APE) as a function of meal (25-min intervals). Isotope was provided orally over small meals. The priming dose (0⋅176 mg/kg) of NaH13CO3 remained similar among treatments (Trt), while the priming and constant doses of L-[1-13C]-Phe varied as follows. Trt A: priming dose: 4⋅8 mg/kg; constant dose: 1⋅04 mg/kg. Trt B: priming dose: 9⋅4 mg/kg; constant dose: 1⋅04 mg/kg. Trt C: priming dose: 9⋅4 mg/kg; constant dose: 2⋅4 mg/kg.
Pilot trial 1: (a) visual inspection (values are x̄ ± sd) and (b) fitted broken-line linear (purple) or broken-line quadratic (blue) model for 13CO2 expressed as atoms percent excess (APE) as a function of meal (25-min intervals). Isotope was provided orally over small meals. The priming (4⋅8 mg/kg) and constant (1⋅04 mg/kg) doses of L-[1-13C]-Phe remained similar among treatments (Trt). The priming dose of NaH13CO3 varied across Trt D (0⋅264 mg/kg), Trt E (0⋅352 mg/kg) and Trt F (0⋅44 mg/kg).
Discussion
The present study was conducted to develop an oral isotope infusion protocol in cats that would produce steady-state conditions of expired 13CO2 for subsequent carbon oxidation studies, such as IAAO. In the IAAO methodology, phenylalanine (Phe) meets the criteria to be used as the indicator AA(Reference Elango, Ball and Pencharz30), and thus, L-[1-13C]-Phe is the tracer of choice to measure flux of 13CO2 at varying intakes of test AA. When Phe is used as the indicator AA, Tyr must be provided in excess to ensure that changes in Phe oxidation are solely due to changes in the intake of the test AA and are not being used to obtain the metabolic requirement for Tyr(Reference Shiman and Gray31). The diet contained 0⋅82 and 1⋅46 % of Phe and Phe + Try on a dry matter basis, respectively, supplying almost twice the requirement established by the Association of American Feed Control Officials(32) for adult cats consuming commercial diets (Phe = 0⋅45 % and Phe + Try = 0⋅74 %; dry matter basis). Furthermore, dietary Phe (including the intake of the tracer) also needs to be similar among dietary treatments(Reference Elango, Ball and Pencharz30) or pool size will differ. Unlike our previous isotope dilution study(Reference Pezzali, Mahroukh and Courtney-Martin20), where (in each experimental day) cats were fed thirteen small meals corresponding to their total feed allowance, only half of the daily feed allowance was provided in each IAAO day in the present study to ensure that all small meals were promptly consumed. This feeding regimen is commonly applied in IAAO studies in pigs(Reference Moehn, Bertolo and Pencharz23,Reference Shoveller, Moehn and Rademacher33–Reference Levesque, Moehn and Pencharz35) and does not affect the metabolic outcome of interest because the ratio of indispensable AA consumed is not affected by the meal size. While the flux of Phe is affected by its dietary intake, the % of L-[1-13C]-Phe that is oxidised should be similar whether half or the total daily feed allowance is provided during IAAO studies, not affecting the breakpoint. Thus, the results observed herein can be solely attributed to perturbations in the kinetics of Phe and/or bicarbonate owing to different isotope dosages.
In pilot trial 1, increasing the priming dose (9⋅4 mg/kg) of L-[1-13C]-Phe or increasing the priming (9⋅4 mg/kg) together with the constant dose (2⋅4 mg/kg) did not result in steady state of 13CO2 in breath samples. Although the latter isotope protocol was successful in producing a steady-state condition of 13CO2 in breath of dogs during IAAO studies(Reference Shoveller, Danelon and Atkinson15,Reference Mansilla, Gorman and Fortener24,Reference Mansilla, Fortener and Templeman36–Reference Sutherland, Mansilla and Fortener39) , the enrichment of 13CO2 indicates that the Phe pool was overprimed in the present study. Over-priming results in a negative slope following the initial rise in enrichment. If breath samples had been collected for longer periods, the negative slope would likely be detected, and thus, a longer period would be required to achieve steady state. Thus, we hypothesised that the priming and constant doses of 4⋅8 and 1⋅04 mg/kg of L-[1-13C]-Phe, respectively, are ideal for cats. Likely, cats failed to achieve steady state of 13CO2 in breath when this isotope protocol was used in our previous study(Reference Pezzali, Mahroukh and Courtney-Martin20) due to a lack of priming of the bicarbonate pool. In parentally fed human neonates, the isotopic steady state of 13CO2 in breath was achieved 12 h after the start of L-[1-13C]-Phe infusion without provision of a priming dose of NaH13CO3(Reference Roberts, Ball and Moore40–Reference Chapman, Courtney-Martin and Moore42). Likely, steady state of 13CO2 in breath of cats would have been achieved with a longer feeding regimen and continuous supply of L-[1-13C]-Phe. It would be difficult, however, to apply such a 12 h half-hourly feeding regimen in cats as they are not parentally fed, and we rely on their continuous voluntary food intake to successfully apply the IAAO protocol. Although priming the bicarbonate pool in pilot trial 1 (treatment A) improved the response in 13CO2, the dose applied (0⋅176 mg/kg) [which is the same previously used in children(Reference Elango, Humayun and Ball43), men(Reference Di Buono, Wykes and Ball22) and women(Reference Ennis, Rasmussen and Lim44)] was not ideal for cats as changes in enrichment of 13CO2 at the last data points were still observed, and thus, higher priming doses of NaH13CO3 were tested in the following pilot trial (pilot trial 2). We also administered the NaH13CO3 dose prior to the priming dose of L-[1-13C]-Phe to enrich the bicarbonate pool in advance, allowing for 13CO2 from oxidation of Phe to be detected. Steady-state 13CO2 enrichment in the cat's breath was maintained and achieved faster when the highest priming dose of NaH13CO3 (0⋅44 mg/kg) was provided compared to the other doses (0⋅176, 0⋅264 and 0⋅352 mg/kg). The priming dose considered ideal for cats (0⋅44 mg/kg) is 2⋅5 times higher than the dose used in IAAO studies in humans(Reference Di Buono, Wykes and Ball22,Reference Elango, Humayun and Ball43,Reference Ennis, Rasmussen and Lim44) , indicating a greater retention of CO2 and slower appearance of CO2 into, and transit through, the bicarbonate pool. This finding was unexpected because cats, due to their smaller size than humans, would have a higher flux between labelled 13CO2 and the unlabelled CO2 in the bicarbonate pool(Reference Issekutz, Paul and Miller21), which should have necessitated smaller doses of labelled NaH13CO3 to achieve steady state. As such, the high NaH13CO3 priming dose we observed in the cat indicates a larger bicarbonate pool or greater retention of CO2, separate from the bicarbonate pool, arising from idiosyncrasies inherent to this species and this finding deserves further research.
Cats are obligatory carnivores, and their metabolism has adapted to a diet consisting predominantly of animal tissues, with many of the adaptations relating to the carbohydrate and protein content of the diet(Reference Morris45). Cats have a higher requirement for nitrogen (N) due to their limited ability to regulate AA catabolic enzymes(Reference Rogers, Morris and Freedland46). Furthermore, cats do not reduce the activity of urea cycle enzymes(Reference Rogers, Morris and Freedland46) when fed lower protein diets like omnivorous and herbivorous animals(Reference Payne and Morris47–Reference Chen, Lewis and Miller50). The first step of the urea cycle is to convert ammonia (product of AA oxidation) and bicarbonate to carbamoyl phosphate. When NaH14CO3 was intravenously injected in a cat, 98 % of the urea carbon synthesised was derived from labelled bicarbonate, with 50 % of the bicarbonate incorporation into urea occurring in the first 30 min(Reference Kornberg, Davies and Wood51). This indicates that the high priming dose of NaH14CO3 necessary to achieve a steady state of 13CO2 in the breath of cats may be due to greater utilisation of bicarbonate by the urea cycle (Fig. 4). Measuring enrichment of 13C in urea would indicate losses of 13CO2 via the urea cycle. Unfortunately, enrichment of 13C was only measured in breath samples in the present study due to difficulty in sample collection. Furthermore, at least half of the incorporation of 14C in the urea cycle occurred rapidly, 30 min post-prime, with NaH14CO3 provision(Reference Kornberg, Davies and Wood51). Priming the bicarbonate pool prior to priming the pool of the substrate of interest (e.g., Phe) might, to some extent, allow earlier detection of 13CO2 derived from the oxidation of L-[1-13C]-Phe, and thus, help to achieve 13CO2 steady state in breath faster.
Major metabolic fates of 13CO2 derived from oxidation of L-[1-13C]-phenylalanine.
The lack of ability to control AA catabolic enzymes may be beneficial for cats as AA are used for gluconeogenesis. The activity of the rate-limiting enzyme of gluconeogenesis (e.g., pyruvate carboxylase) in the liver is upregulated in cats compared to dogs(Reference Washizu, Tanaka and Sako52). The conversion of pyruvate to oxaloacetate via pyruvate carboxylase also requires bicarbonate as a substrate, and thus, would be a reason for retention of CO2 to be used in the process of gluconeogenesis (Fig. 4). Indeed, the metabolic status of rapid gluconeogenesis may affect the recovery of labelled CO2. A lower recovery of 14CO2 in NaH14CO3-perfused livers of starved compared to fed rats(Reference Tomera, Goetz and Rand53) was observed and is likely due to a higher rate of gluconeogenesis in the former. The dietary macronutrient content may also affect recovery of labelled CO2, with a greater intake of carbohydrates resulting in reduced gluconeogenesis, and thus, affecting the NaH13CO3 dose. The same authors(Reference Tomera, Goetz and Rand53) also observed a net incorporation of 14CO2 into other compounds, such as urea, protein, and carboxylic and amino acids. The rate of incorporation into each substrate varies widely and it is affected by metabolic state(Reference Barstow, Cooper and Sobel54); thus, animals need to be in the same metabolic state and to maintain their BW throughout IAAO studies. A fraction of CO2 that is not recovered in breath may also be lost in slowly exchanging pools, such as bone (Fig. 4). In the cat, approximately 6 % of injected NaH14CO3 was found in the bone after 4 h, while only 1 % was found in muscles, viscera and blood(Reference Kornberg, Davies and Wood51). It is also important to consider that the bicarbonate pool is not homogeneous and the rate of excretion of labelled carbon after intravenous injection of labelled bicarbonate follows an exponential equation with three major bicarbonate pools with distinct kinetics: a central vascular pool, a rapid turnover pool (bicarbonate in soft tissues) and a slow turnover pool (bicarbonate in bone tissue)(Reference Barstow, Cooper and Sobel54–Reference Irving, Wong and Wong56). Furthermore, CO2 transport and storage dynamics are sensitive to changes in acid-base status(Reference Leese, Nicoll and Vaenier57); thus, dietary electrolyte balance should be considered in future carbon oxidation studies. While it has been recently reported that changes blood pH via dietary supplementation can affect protein kinetics in humans(Reference Moore, Lysecki and Breen58), this has yet to be investigated in cats. As the intake of some AA can influence acid-base balance(Reference Patience59) and in dose-response studies concentrations of the test AA are provided in graded levels, caution should be taken to ensure a consistent electrolyte composition among test diets.
To account for retention of labelled CO2, the bicarbonate retention factor (BRF) is usually determined and used as a correction factor to compute the flux of CO2 in carbon oxidation studies. Although determining the BRF in cats would aid our understanding pertaining the metabolism of bicarbonate in this species, the goal of the present study was to determine the best isotopic protocol to achieve steady state of 13CO2 in breath samples of cats when L-[1-13C]-Phe is used as the tracer in IAAO studies. The BRF used to compute flux of 13CO2 does not influence the achievement of steady state of 13CO2 in breath, only the accuracy of the total 13CO2 excretion. We used the BRF determined in dogs which assumes 100 % recovery of 13CO2, and thus, likely underestimates the total absolute flux of 13CO2 in cats. The BRF, however, does not influence the degree of change of flux of 13CO2, which is more important than the absolute value in oxidation studies. As mentioned above, CO2 retention can be influenced by factors inherent to the animal (e.g., metabolic status) and the diet (e.g., electrolytic composition); thus, ideally, one should determine the BRF under conditions identical to the isotope study and using the same system for breath collection and measurement of respiratory exchange(Reference Hoerr, Yu and Wagner60) as done previously in dogs(Reference Shoveller, Danelon and Atkinson15) and chickens(Reference Tabiri, Bertolo and Ball13). Given the numerous factors influencing BRF, future studies should investigate the effects of properties of the diet (e.g., electrolytic composition and protein content) on the acid balance of the cat and, consequently, on bicarbonate retention.
Conclusions
An isotopic steady state of 13CO2 enrichment in breath can be achieved in cats using a thirteen small meal regimen; wherein a priming dose of NaH13CO3 (0⋅44 mg/kg) and L-[1-13C]-Phe (4⋅8 mg/kg) should be provided in the fourth and fifth meals, followed by a constant dose (1⋅04 mg/kg) of L-[1-13C]-Phe in the next meals. Fasted background of 13CO2 can be used if there are no major differences in the macronutrient composition of dietary treatments. This protocol resulted in an isotopic steady-state condition necessary to successfully use the IAAO technique in cats, which can be used to determine indispensable AA requirements and AA bioavailability in future studies.
Acknowledgements
We would like to acknowledge the cat Perry for his outstanding participation in this study. Perry adapted extremely well to the IAAO protocol. He promptly consumed all small meals and remained calm during calorimetry in all IAAO days.
J. G. P. was responsible for conceptualisation, methodology, data curation, formal analysis, investigation, methodology, visualisation and writing (original draft preparation); J. G. L. was responsible for data curation and writing (review and editing); S. M. P. was responsible for methodology and writing (review and editing); A. K. S. was responsible for conceptualisation, methodology, funding acquisition, resources, supervision and writing (review and editing).
This work was supported by funds from the Natural Sciences and Engineering Research Council of Canada Discovery Program.
The authors declare that there is no conflict of interest regarding the publication. However, it is noteworthy that 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. was previously employed by P&G and Mars Pet Care and has received honoraria and research funding from various commodity groups, pet food manufacturers and ingredient suppliers.
Open access
