Hens Produce Artificially Enriched 13 C Egg Proteins for Metabolic Tracer Studies

Clinicians and researchers studying protein metabolism in vivo, typically use isotopically-labeled free amino acids as metabolic tracers rather than isotopically-labeled proteins because such proteins are commercially unavailable. However, the use of free amino acids in lieu of protein tracers violates the critical assumption that tracer molecules undergo the identical biochemical reactions as the tracee molecules of interest. To address this problem we synthesized C-labeled proteins using egg laying hens and investigated the relationship between tracer dose and method of delivery on C-protein production. We enriched hens with one of two isotope tracers (C-1-leucine or a uniformly labeled C-amino acid mixture) mixed in their food or dissolved in their drinking water at different dosing levels (86-432 mg day). The recovery of C in egg white proteins of the hen fed C-leucine ranged from 14% to 21%; recovery rates were highest at the lowest dosing level. At the highest dosing level egg whites were enriched more than 150‰ above background levels of C. The time required for half maximal C enrichment depended chiefly on the mode of tracer administration, and ranged from 2.5 days for C-leucine dissolved in water to 4.9 days for C-leucine mixed in food. Relative rates of C recovery in the egg protein were lowest for hens fed the uniformly C labeled amino acid mixture, presumably because of the high proportion of nonessential amino acids. The time required for the C-enrichment in eggs to return to background levels at the end of the enrichment period was about twice that required to initially reach isotopic equilibrium with the diet, indicating significant biochemical discrimination of endogenous C amino acids. We conclude that delivering small amounts of C amino acid tracers in the drinking water of hens is the most effective way to produce C-enriched proteins to for tracer studies that do not require δ C-enrichment above 200‰.


Producing 13 C-Proteins
Although commercially unavailable, 13 C-labeled proteins can be synthesized in the laboratory.Several published studies report synthesizing artificially enriched, 13 C-labeled proteins in sufficient amounts to be used for clinical breath testing (Berthold et al., 1991;Boirie et al., 1995;Evenpoel et al., 1997;Fromentin et al., 2011b;Geboes et al., 2004;Irving et al., 1988).Yet, despite our progress in method development, we still know very little of how simple experimental factors can influence 13 C-protein production (Fromentin et al., 2011b).Such information is necessary if we intend to implement best practices in breath testing (Braden, 2007(Braden, , 2010) ) and metabolic tracer studies (McCue, 2011).
Previous studies describe two general approaches for producing high quality, 13 C-labeled protein by supplementing diets with 13 C-amino acid tracers; one involves egg white proteins (primarily ovoalbumin) and the other involves milk proteins (i.e.whey protein and casein).Here we focus on producing 13 C-labeled egg proteins rather than milk proteins for several reasons.First, maintaining adult dairy cows or lactating women (Irving et al., 1988) for extended periods of time poses unique logistical obstacles for most laboratories such as housing and the ethical concerns with humans.Second, hens are capable of generating 13 C-labeled proteins with significantly lower daily dosages of 13 C-amino acids than required by cows.Additionally, egg proteins are better suited for making relatively small batches of labeled proteins as required.Third, egg whites are virtually lipid free and composed of over 91% protein (dry, ash-free mass; USDA, 2000) thus eliminating additional purification steps required for removing the lipid fraction of milk.Finally, egg proteins are considered the 'gold standard' to which all other dietary proteins are scored with regard to amino acid composition, digestibility, and nutritional value (Chibnall et al., 1943;Mitchell & Block, 1946;Rao et al., 1964;Satterlee et al., 1979;Young et al., 1975) -important characteristics for metabolic tracers.
The immediate goal of this project was to synthesize relatively large amounts of 13 C-protein (e.g., several hundred grams) that can be used in future studies of breath testing and nutrient allocation.The ultimate goal was to increase our understanding of 13 C-protein synthesis in the hen model.The following experiments were designed to allow us to 1) characterize the incorporation and washout kinetics of 13 C-tracers, 2) develop a dose-response curve for a commonly used 13 C-tracer in breath testing (i.e., 13 C-1-L-leucine), 3) compare the efficacy of different dosing techniques (i.e., mixed with food or dissolved in drinking water), and 4) compare 13 C recovery rates between an essential amino acid and a mixture of essential and nonessential amino acids.We hope that by improving the existing protocols for synthesizing 13 C-proteins, these tracer materials can become more accessible to researchers and allow them to track the physiological fates of exogenous proteins in vivo more accurately.

Tracer Selection
We chose to use 13 C-1-L-leucine (hereafter: 13 C-leucine) as the primary tracer molecule in this study for several reasons.First, 13 C-leucine is one of the most commonly used tracers in amino acid breath testing.Second, leucine is an essential amino acid and is therefore less likely to be oxidized during 'first pass' splanchnic oxidation than nonessential amino acids (Berthold et al., 1991;Daenzer et al., 2001;Matthews et al., 1993;McCue et al., 2011b;Yu et al., 1992).Third, leucine is the most abundant essential amino acid and the third most abundant amino acid in egg proteins, comprising 8.4% and 8.8% of the amino acids in egg white and yolk proteins, respectively (USDA, 2000).Finally, 13 C-leucine is widely available from commercial suppliers of stable isotope tracers and can be purchased in bulk for as low as $100 gram -1 .Leucine is a non-glucogenic amino acid (Murray et al., 2003) and losses of the number-1-carbon into carbohydrate and lipid pools should be low (compare with Gruhn & Hennig, 1984).Figure 1 illustrates how the number-1-carbon of a 13 C-1-leucine tracer molecule is converted into 13 CO 2 .The number-1-carbon of leucine is thought to only leave the body as 13 CO 2 (Cheng et al., 1985;Matthews & Bier, 1983;Matthews et al., 1981;Parra & Martinez, 2006;Wetzel & Fischer, 2005), however during metabolic alkalosis, a small amount of HCO 3 can be excreted by the renal system (De Benoist et al., 1984;Yu et al., 1990).Nevertheless, we test the assumption of negligible 'leakage' of the number-1-carbon of leucine in to the lipid pool by analyzing the 13 C of the adipose tissue in hens.

OH
We also conducted 13 C-protein enrichment experiments using mixture of 16 uniformly 13 C-labeled amino acids isolated from algal proteins (CNLM-452, $1250 g -1 , Table 1; Cambridge Isotopes Inc. USA).This trial was conducted to compare the incorporation, kinetics and 13 C recovery rate between a highly enriched, mixed amino acid tracer ( 13 C atom percent = 100%) and a single amino acid tracer containing one 13 C label (i.e., 13 C-leucine; 13 C atom percent = 16.7%).Threonine 5 Tyrosine 4 Valine 5 Total 100

Methods
Three mature egg-laying hens (ISA Brown; ~2 years old) were obtained from a commercial farm (O5-Livestock, Hondo, Texas).Within one week of arriving in the laboratory, the hens were dewormed with three daily doses (60 mg) of Ivermectin (Panacur, Hoechst) dissolved in their drinking water (Carpenter et al., 2001).Before experiments food was offered ad libitum and the mass consumed each day was recorded.Hens were housed individually in stainless steel lagomorph cages, fed a complete, balanced diet (Naturewise Layer; Neutrena; δ 13 C = -17.21‰),and acclimated to constant conditions for six weeks (14L:10D; 27°C).Water consumption was measured each day.

Experiment 1
The feeding experiments consisted of supplementing the pelleted diet of Hen 1 with 13 C-1-L-leucine (Cambridge Isotope Laboratories, Inc.).A pyramid dosing regimen was implemented for Experiment 1 as a means to verify the dose responses of 13 C-1-L-leucine.Because daily food intake ranged from ~115 to 145 grams during the acclimation period, a daily ration of 150 grams of food was used for the experiments to ensure ad libitum availability.Each day, known amounts of 13 C-leucine (i.e., 100, 200, or 300 mg) were mixed with the hen's daily ration.Six sequential dosing treatments were examined over 14-day periods (Table 2).Two of the feeding treatments (i.e., 100 and 200 mg day -1 ) were repeated several weeks apart to explore the effects of long-term tracer integration.In one treatment, fresh drinking water was replaced with a solution of 1 g L -1 of 13 C-leucine and offered ad libitum.Once per week, vitamins and minerals (Avian Super Pack, Animal Science Products) were added to the drinking water according to the manufacturer's suggestion (80 mg L -1 ).Each morning, between 0700-0930 hours, all uneaten food was collected and weighed.The experimental diets correlate with the δ 13 C of egg white and yolk illustrated in Figure 2.
Eggs were collected twice each day and labeled with the time and date before being weighed and refrigerated.
Approximately every two weeks the refrigerated eggs were hardboiled for 20 minutes to halt the protease (e.g.trypsin and chymotrypsin) inhibiting characteristics of raw eggs (Evenpoel et al., 1999;Evenpoel et al., 1998;Liu et al., 1971;Osuga & Feeney, 1968).Eggs were then peeled and the white and yolk fractions were separated, minced, and dried to a constant mass in a convection oven (70°C).The dried tissues were transferred into 50 ml centrifuge tubes and stored at 10°C until further processing.Subsamples (~10 mg) of each egg white and yolk were homogenized using a Wig-L-Bug amalgamator (Dentsply North America, Inc.), packaged into 2 ml centrifuge tubes, and analyzed at the University of Arkansas Stable Isotope Laboratory (Fayetteville, AR).The δ 13 C values (sensu Craig, 1957) and carbon content of samples were determined using a Finnigan Delta Plus continuous-flow IRMS interfaced to a Carlo Erba NC2500 elemental analyzer (Thermo Scientific, Bremen Germany, Milan, Italy).Measured 13 C isotope ratios were corrected to VPDB (Coplen, 1994).Internal standards were run in parallel with the unknown samples and had an analytical error of < 0.1‰.

Calculation of Tracer Kinetics
Rates of 13 C-leucine incorporation into the egg white and yolk fractions were modeled (SigmaPlot 11.0, Systat Software Inc.) using the following 2-parameter, exponential equation (Patterson, 1997): where δ 13 C (x) is the 13 C enrichment at time x, δ 13 C ∞ -δ 13 C Bkg refers to the maximal 13 C enrichment at equilibrium above the initial, background value (Bkg), k is the rate constant for tracer incorporation, and t is time in days.The time required for one-half of the isotopic enrichment to occur (t 50 ) was calculated according to the following equation: We evaluated the possibility that prolonged exposure to artificially enriched 13 C-leucine in the diet could cause gradual, long-term enrichment in egg products using paired t-tests to compare the δ 13 C of egg whites and egg yolks at the same dosing level (i.e.200 mg day -1 ) at different points in the experiment.
Trophic fractionation Δ 13 C for egg white and yolk was calculated only for the background diet according to the following equation: The chemical composition of dried egg whites (NDB No. 01173) and yolks (NDB No. 01137) are well characterized and available through the USDA (National Nutrient Database; http://www.nal.usda.gov/).Values including proximate analysis and amino acid composition were used for subsequent calculations (see Appendix for equations).For example, the amino acid content of egg whites allowed us to calculate the 13 C tracer recovery and the specific 13 C enrichment of individual leucine residues in the whole protein mixture.
Hen 1 was euthanized just prior to the end of the scheduled experiment after it broke its leg on day 104.We subsequently harvested and measured the δ 13 C of several tissues (i.e., pectoral and neck muscle, large and small intestine, liver, lung, cecum, blood, brain, skin, long bone marrow, and abdominal fat depot) from this animal.These tissues were dried to constant mass at 70°C and coarsely ground.One-gram subsamples were further homogenized using an amalgamator as described above.Lipids were separated from all tissues (except whole blood) by vortexing for 30 minutes in an excess of 2:1 chloroform:methanol (Oppel et al., 2010).After centrifugation, the supernatant was discarded, with the exception of adipose tissue and recovered by evaporation under a dry stream of N 2 .Pelleted lean tissue samples were also dried under a stream of N 2 .The δ 13 C of all tissue samples were analyzed as described above for the egg tissues.

Experiment 2
A second experiment was conducted to compare the 13 C integration kinetics at the start of dosing with the 13 C washout kinetics following terminus of 13 C tracer dosing.Hen 2 was fed ad libitum and given 100 mg of 13 C-leucine dissolved in 750 ml of drinking water each day for 21 days.Because we had previously determined that the ad libitum water consumption for Hen 2 was > 800ml day -1 , the use of 750 ml to guarantee optimum utilization of tracer was justified.On day 22, 13 C-leucine dosing was halted and Hen 2 was instead given 750 ml of fresh water each day.The eggs were collected daily and processed as described above.Hen 2 was euthanized 48 days following her first 13 C isotope treatment.We did not harvest Hen 2 tissues because it is likely that many of these tissues would not have had time to reach isotopic equilibrium (Phillips & Eldridge, 2006).

Experiment 3
Hen 3 was fed ad libitum and given 50 mg day -1 of uniformly 13 C-labeled amino acid mixture dissolved in 750 ml of water for 20 days.On day 21 13 C-labeled amino acid dosing was halted and Hen 3 was returned to the to the background diet.The eggs were collected daily and processed as described above.Hen 3 was euthanized 48 days following the first isotope treatment, and like Hen 2 her tissues were not harvested.
Rates of 13 C incorporation into the egg white (not egg yolk) of Hen 2 and 3 were modeled as described above and the rates of 13 C washout from egg whites were modeled (SigmaPlot 11.0, Systat Software Inc.) using the following 2-parameter, exponential equation: where δ 13 C ∞ -δ 13 C Bkg was treated as a shared parameter between the incorporation and washout models.The t 50 for both of these models were calculated using Equation ( 2).Mean values are reported followed by standard deviation, unless indicated otherwise.

Experiment 1: Trophic Fractionation, Incorporation Kinetics, and Dosing Response
Hen 1 (1137 g initial; 1186 g final) consumed an average of 125.0 g day -1 .As such the actual ingested tracer doses were 86.3, 172.7, and 259.0 mg 13 C-leucine day -1 for the three dosing levels, respectively.For the simultaneous dose of tracer mixed with food and dissolved in water, a total of 432 mg of 13 C-leucine day -1 was ingested.A total of 72 intact eggs (mass: 65.2 ± 1.2 g) were collected over the enrichment period yielding a mean net production rate of one egg every 1.25 days.Eight eggs were broken or cracked during laying; the egg white and yolk were salvaged when possible.
On day 56, our concern about diminished egg production in Hen 1 prompted us to offer additional dietary supplements: 100 mg of diatomaceous earth, 200 mg of crushed oyster shells and 15-30 live darkling beetle (Zophobus morio) larvae for one week (i.e.days 57-63; Figure 2).Although we collected and measured the 13 C values in eggs produced during that week, we did not use those eggs (or the eggs laid during the following week) for modeling purposes or statistical analyses.The time divisions refer to different dosing regimens (see Table 1).Note that the period D' refers to a period in the experiments when reduced egg production prompted us to supplement the hen's diet with minerals and live insects.The data points within this period are not used for statistical analyses or modeling purposes.
A significant increase in the δ 13 C of egg white was detected after just one day of dietary 13 C-leucine enrichment (one-tailed t-test; df = 6, p < 0.0001); However a significant increase in the δ 13 C of egg yolk required an additional three days (one-tailed t-test; df = 7, p < 0.0001).The subsequent analyses and modeling of tracer integration into egg tissues accounted for the differences in lag periods between the timing of ovulation (e.g. when yolk is synthesized) and oviposition (e.g. when the egg white protein is synthesized).Exogenous 13 C-incorporation closely followed 2-parameter exponential models (R 2 = 0.90-0.98).The integration models for egg white and yolk had rate constants (k) of 0.141 and 0.179, yielding t 50 of 4.92 and 3.87 days, respectively (Figure 3); neither of these values had a statistically significant difference.  1C atoms into egg white and egg yolks of two hens receiving 13
Egg white exhibited larger increases in δ 13 C than egg yolk, but this difference can be attributed to the high lipid content of the yolk (i.e.55.8% dry mass).According to the 13 C-incorporation models the maximal enrichment at the lowest experimental dose (i.e.86.3 mg day -1 ) was 3.9 times higher for the egg white (i.e.39.2‰ above background) than the egg yolk (i.e.10.0‰ above background).
The mass of 13 C-leucine ingested and the 13 C-enrichment in the egg tissues exhibited a positive, but non-linear response.The relative 13 C-enrichment gradually decreased as the tracer dose in the food tripled from 86.3 mg day -1 to 259 mg day -1 (Figure 4).

Experiment 2: Tracer Delivery Method
Hen 2 (1505 g initial; 1423 g final) consumed food ad libitum and was given access to 100 mg day -1 13 C-leucine dissolved in 750 ml of water for 21 days during which she laid 13 eggs (53.9 ± 9.1 g).Unlike Hen 1, Hen 2 was repeatedly observed spilling water while drinking and we did not attempt to quantify total tracer recovery in Hen 2. Nevertheless, the rate constant (k) for 13 C incorporation was 0.282 resulting in a t 50 of 2.45 days.This value is approximately half of that observed in Hen 1 while consuming a similar dose (86.3 mg day -1 ) of 13 C-leucine mixed into her food (Figure 3).Moreover, the asymptotic enrichment calculated by the tracer integration model (51.0‰) was 30% higher than that calculated for Hen 1 (39.2‰).Assuming a linear dose response over this narrow range of doses (i.e., 86.3 versus 100 mg day -1 ; but see Fig. 4 across a wide range of doses) the egg white of the larger Hen 2 was still approximately 10% more enriched in 13 C than that of the smaller Hen 1.This difference could be attributed to differences in mean egg mass (17% smaller) or the increased bioavailability of dissolved 13 C-leucine.During the subsequent 27-day washout period Hen 2 only laid 2 eggs, thus precluding accurate pharmacokinetic modeling (but see Experiment 3 below).

Experiment 3: 13 C-Labeled Amino Acid Mixture
Hen 3 (2072 g initial, 2010 g final) consumed food ad libitum and was given access to 50 mg of uniformly 13 C-labeled amino acids dissolved in 750 ml of water for 20 days during which she laid 18 eggs.Hen 3 also produced 18 eggs during the 28-day washout period; overall generating over 200 g (dry mass) of 13 C-enriched egg white protein.Eggs had a mean mass (53.4 ± 2.1g) that was 18% lower than of Hen 1, but not significantly different from those of Hen 2 (unpaired t-test, p = 0.951, df = 30).According to the integration model, the asymptotic 13 C enrichment of the egg whites was 78.5‰ above background.The k and the t 50 were 0.3965 and 1.75 days, respectively.The rate constant for loss of 13 C in egg whites after the tracer was removed was 0.227 with a t 50 of 3.05 days.Excessive water spillage was not observed in Hen 3. The mean Δ 13 C white for Hen 2 and Hen 3 was -1.50 ± 1.0‰ (n = 6), a value that differed by only 0.3‰ from that observed in Hen 1.

13 C-1-Leucine Tracers
In the first experiment, Hen 1 produced over 400 g (dry mass) of 13 C-leucine enriched egg white protein over a 90-day period.Because leucine composes 8.4% of whole egg white protein mass, the enrichment of the individual leucine residues in the protein was much greater than that of the whole protein.We calculated that the δ 13 C of leucine residues increased from background levels (-19.1‰;1.08 atom percent) to 1403.6‰ (2.62 atom percent) at the highest tracer dose.In other words, before artificial enrichment, 5.4% of all leucine molecules in the egg white contained at least one 13 C-atom, but after consuming the highest tracer dose (432 mg day -1 ) for 14 days, 25.3% of all of leucine molecules contained one or more 13 C-atoms (see equations in Appendix).
Hen 2 produced over 100 g (dry mass) of 13 C-leucine enriched egg white protein over a 40-day period.The 13 C-atoms from the leucine tracer were incorporated more rapidly into egg tissues when delivered in the drinking water (Figure 3).The asymptotic δ 13 C enrichment in egg proteins was greater than in Hen 1, but because the eggs of Hen 1 were larger it is difficult to make a formal comparison.It is important to note that δ 13 C values of egg protein alone can be misleading since they do not account for the size of the egg which can vary by 66% between 'small' and 'jumbo sizes' (USDA, 2000).Moreover, since hen egg yolk mass exhibits less variation than egg white, the total protein content of eggs can vary by as much as 100%.
We calculated that at equilibrium the 13 C-tracer recovery in the egg white of Hen 1 ranged from 14.3% to 21.5% depending on the dose.As expected the fractional recovery rate (i.e., δ enrichment per mg tracer) was the highest at the lowest dose of the 13 C-leucine tracer (Figure 4).The pattern of relatively diminishing enrichment with increasing dose is likely related to increased oxidation of the exogenous leucine tracer since it was present in increasing excess in the diet.Although the egg white proteins from Hen 2 had higher 13 C enrichment than those of Hen 1 when fed a similar dose of 13 C-leucine (Figure 3), the total fractional rates of 13 C recovery (20.0%) were similar.
In their experiment, Evenepoel et al. (1997) reported total tracer recovery rates as high as 40.2% by using a base diet that was deficient in leucine, which presumably minimized oxidative losses of the tracer (McCue, 2011;McCue et al., 2010).It is noteworthy that the previous study estimated, and did not quantify, the tracer recovered in the egg yolk.The egg yolks in this experiment had dry masses of 7.8 ± 0.4 g; n = 71.If we consider the egg yolks then the 13 C total recovery in this study would be as high as 33.7%, but without requiring a customized, leucine-deficient diet.
In this study, 16.7% of the food labeled with 13 C-leucine tracer was uneaten and was not reused.If this material could be recycled within the food chain, then the total yield of 13 C could be substantially increased.Furthermore, recycling some of the egg yolk back into the food chain (i.e., by combining it with fresh food) could also increase the production of 13 C-proteins by up to 5%, however no studies have examined this possibility.
The aqueous 13 C-leucine was more rapidly incorporated into egg white proteins.This outcome is clear when comparing the t 50 of Hen 1 (4.92 days; crystalline tracer) and Hen 2 (2.45 days; aqueous tracer; Figure 3) at 100 mg doses of 13 C-leucine.This pattern was also evident when Hen 1 switched dosing regimens.Between days 79 and 92, Hen 1 achieved a t 50 of 2.26 when 13 C-leucine was administered via drinking water.

Uniformly 13 C-Labeled Amino Acids
The rate at which Hen 3 (fed a mixture of uniformly 13 C-labeled amino acids) incorporated 13 C into egg white proteins (t 50 = 1.75 days) was similar, if not slightly higher than that documented in Hen 2 (Figure 5).Given that the masses of eggs produced by Hen 2 and 3 did not statistically differ, the differences in incorporation kinetics between the leucine and mixed amino acid tracers could stem from 1) differential incorporation kinetics of the individual amino acids and/or 2) de novo synthesis of nonessential amino acids from recycled carbon skeletons containing 13 C.  13 C integration and washout kinetics for Hen 3 consuming 50 mg day -1 of a uniformly 13 C-labeled mixed amino acid mixture The integration curve was modeled the same as in Figure 3 and the washout curve was modeled using a first-order, 2-parameter, exponential equation: δ 13 C (x) = (δ 13 C ∞ -δ 13 C Bkg )*(1-e -k*t ) where (δ 13 C ∞ -δ 13 C Bkg ) was treated as a shared parameter between the incorporation and washout models.See Methods for details.
Hen 3 consumed a daily dose of artificially enriched 13 C-tracers (50 mg d -1 ) that was half that consumed by Hen 2, but because the concentration of 13 C-atoms in the tracer molecules was higher, the maximal 13 C enrichment of egg white proteins was higher in Hen 3 than Hen 2 (i.e., 78.5‰ versus 51.0‰, respectively).Nonessential, exogenous amino acids are more likely to be oxidized in the postabsorptive phase (Brown-Mason et al., 1981;McCue et al., 2010;Voigt et al., 2008b), and thus the maximal 13 C enrichment of egg white proteins resulting from the uniformly labeled mixed amino acid tracer was lower than expected if solely based on the 13 C content of the two tracers [i.e., uniformly 13 C-labeled mixed amino acid tracer ( 13 C atom percent = 100%) versus the 13 C-1-leucine tracer ( 13 C atom percent = 16.7%)].Consequently, the net cost of 13 C enrichment in the total egg white protein was lower using 13 C-1-enriched leucine than the uniformly 13 C-labeled amino acid mixture.However, if the goal of a future experiment is to produce highly enriched (> 500‰) egg proteins, any dosing level of 13 C-1-labeled leucine (or even uniformly 13 C-labeled leucine) would be insufficient.
The 13 C washout curve for Hen 3 consuming the uniformly labeled 13 C amino acid mixture closely followed a first-order decay model (Figure 5).A near complete isotopic recovery of egg white proteins (i.e., return to background values ~-17‰) was identified 28 days after stopping tracer dosing.The t 50 for the washout period was 3.05 days, nearly twice the value for tracer incorporation (i.e., 1.75 days).Differences in integration and washout kinetics of egg protein were previously documented in hens fed 15 N tracers (Gruhn et al., 1979) and quail fed tritium (Brown-Mason et al., 1981), and most likely result from isotopic discrimination during amino acid transamination and deamination and/or the recycling of 13 C-carbons into de novo synthesis of nonessential amino acids (e.g., Cherel et al., 2005a;Martinez del Rio & Wolf, 2005;Martinez del Rio et al., 2009;McCue & Pollock, 2008;Pearson et al., 2003;Sick et al., 1997).

13 C Enrichment in Somatic Tissues
The tissues of Hen 1 were harvested to compare their 13 C values.The δ 13 C of the lipid fraction of the adipose tissue was -21.18‰, a value that matches the non-enriched values of the egg white and yolk and confirms that none of the 13 C-tracer 'leaked' into the carbon pool of the lipids.Long-term experiments raising mice on 13 C-leucine yielded the same outcome (unpublished observations).An intact egg was removed from the hen's oviduct and the δ 13 C of the egg white was 13.45‰.Because this hen was undergoing a diet switch from high enrichment (432 mg day -1 13 C-leucine) to moderate enrichment (86.3 mg day -1 13 C-leucine), we would expect that the tissues with the most rapid protein turnover would most quickly reflect the changes of 13 C in the diet.
Hens are capital breeders (De Vink et al., 2011;Meijer & Drent, 1999) and synthesize egg products using recently ingested nutrients, but these nutrients mix with endogenous nutrients in circulation.Consequently, the proteins deposited in the egg white can serve as starting point to compare the δ 13 C of the circulating amino acid pool with that of other tissues.When comparing the ratio of the δ 13 C of the lean somatic tissues to the δ 13 C of the egg white, 13 of the 15 tissues examined had values greater than 1.0 (Figure 6).The tissues with the highest apparent rates of protein turnover were smooth muscle tissues (e.g., small intestine.large intestine, and oviduct) whereas the tissues with the slowest rates of protein turnover were the brain, heart, lung, and blood cells.The relative rates of tissue-specific protein turnover suggested by these data closely parallel those reported for non-galliform birds (reviewed in Bauchinger & McWilliams, 2012).Only the lean fraction of the skin and the adipocytes had ratios less than 1.0.Rather than concluding that the skin and adipocytes are synthesized more rapidly than egg whites, we presume these they contained little 13 C because the two-week duration of the previous 432 mg day -1 13 C dosing was insufficient for these tissues to reach isotopic equilibrium (or near equilibrium) with the diet. .Ratio of the δ 13 C values of the lean fractions of different somatic tissues from Hen 1 to that of egg white on day 105.Two weeks prior this hen was switched from 432 mg 13 C leucine day -1 to 86 mg 13 C leucine day -1 and the hen was not at an isotopic equilibrium with its diet.Numbers above the bars represent actual δ 13 C values

Conclusion
This is the first study to compare a) more than two dosing levels of a particular tracer, b) multiple dosing techniques, c) different amino acid tracers, and d) incorporation and washout kinetics during the synthesis of 13 C-enriched egg proteins.The results of these experiments show that increasing the dose of a 13 C-tracer does not cause a concomitant linear increase in the δ 13 C of egg proteins.This study also revealed that delivering tracers dissolved in the drinking water did not necessarily increase the final δ 13 C of proteins, but decreased the time required for the eggs to reach isotopic equilibrium with the diet -a factor that could influence budgets involved with the large scale synthesis of these proteins.The use of highly enriched, uniformly 13 C-labeled mixtures of amino acids may be suitable if desired protein enrichments are above ~500‰ (> 1.5 atom percent), but because of the excessive oxidative losses of 13 C from nonessential amino acids (Fromentin et al., 2011b) and the lack of the positional specificity of 13 C-atoms (Berthold et al., 1991;Fromentin et al., 2011b;Kelly & Martinez del Rio, 2010), a 13 C-1-labeled essential tracer molecule is desirable for applications where desired 13 C enrichments are modest.While both the integration and washout kinetics of 13 C-atoms in egg proteins closely follow first-order models, the half-life for 13 C-atoms leaving the body is much lower than its entry into the body.Additional studies will be necessary to investigate the physiological basis of this isotopic discrimination in vivo.Overall, these experiments provide several new insights that should help make the production of 13 C-labeled proteins more efficient and increase the availability of these molecules for 13 CO 2 breath testing in humans and animals.

Figure 1 .
Figure 1.Series of reactions illustrating the fate of the number-one-carbon in leucine during oxidation

Figure 2 .
Figure 2. The δ 13 C of egg whites and yolks from Hen 1 consuming 13 C-leucine tracers

Figure 3 .
Figure 3. Incorporation curves for 13 C atoms into egg white and egg yolks of two hens receiving 13 C-leucine tracers

Figure 4 .
Figure 4. Dose sensitivity of leucine residues in egg whites for two hens under different dosing regimens.Circles and error bars refer to mean values ± standard deviation

Figure 5 .
Figure5.Illustration of the13 C integration and washout kinetics for Hen 3 consuming 50 mg day -1 of a uniformly13 C-labeled mixed amino acid mixture Figure6.Ratio of the δ 13 C values of the lean fractions of different somatic tissues from Hen 1 to that of egg white on day 105.Two weeks prior this hen was switched from 432 mg13 C leucine day -1 to 86 mg13 C leucine day -1 and the hen was not at an isotopic equilibrium with its diet.Numbers above the bars represent actual δ 13 C values

Table 1 .
Amino acid profiles for the uniformly-labeled 16-amino acid mixture

Table 2 .
Summary of the 13 C-leucine dosing regimen for one hen (Hen 1)