Standardization of A Sample-Processing Methodology for Stable Isotope Studies in Poultry

Denadai JCI https://orcid.org/0000-0003-0853-1159 Sartori JBI https://orcid.org/0000-0003-2691-8250 Pezzato ACI https://orcid.org/0000-0001-9543-5688 Costa VEII https://orcid.org/0000-0003-3889-7514 Sartori MMPIII https://orcid.org/0000-0003-4119-8642 Petinati BESI https://orcid.org/0000-0001-8760-8083 Gennari RSI https://orcid.org/0000-0002-3633-1712 Silva ETII https://orcid.org/0000-0003-0121-7060 Carvalho MAGI https://orcid.org/0000-0003-4700-4618 Ishizuka ANDI https://orcid.org/0000-0002-1319-5463


INTRODUCTION
The future application of methodologies involving the natural variation of stable isotope abundance in animal nutrition and physiology studies should be stimulated for health, environmental and economic reasons, as well as due to the possibility of developing new models that include details of acquisition rates, synthesis, degradation and destination of nutrients in the various animal tissues.
According to Ducatti (2007), for the progress of this line of research, studies that contribute to the knowledge on carbon isotopic assimilation in different poultry tissues during production phases are needed, in addition of the development of methods that aid sample processing in order to allow faster results.
Considering that the lipid content of the muscular tissues of the animal species, in general, is highly variable, and it is influenced by factors such as size, sex, sexual maturity condition and sampling time, lipid extraction is commonly used to remove depleted carbons from lipids.In addition, it and allows standardization when evaluating and comparing lipid muscle content among different species (Hussey et al., 2008). eRBCA-2019-0687 The yolk is composed of more than 50% lipids in most species of birds (Carey et al., 1980;Sotherland & Rahn, 1987;Burley & Vadehra, 1989), and lipids are depleted in 13-carbon isotope compared with proteins and carbohydrates (DeNiro & Epstein, 1978).Therefore, differences in the lipid content of samples may confound the interpretation of stable carbon isotopes (Post et al., 2007), and many ecological studies chemically remove lipids from tissues or apply an arithmetic correction to explain the greater abundance of isotopes (Larsen et al., 2006).
Anticoagulants are typically used to prevent the coagulation cascade, allowing subsequent centrifugation.Heparin interacts with antithrombin, forming a ternary complex that inactivates several clotting enzymes, such as clotting factors (II, IX and X) and significantly more thrombin (Shuman & Majerus, 1976).EDTA acts as a strong calcium-chelating agent (Rand et al., 1996), and when complexed with magnesium, prevents the coagulation cascade.
In a study evaluating different storage and sample collection methods of elasmobranchs (Kim & Koch, 2012), no statistical differences were found in the isotopic ratio of blood components collected in tubes coated with heparin or with no additives.In that study, muscle lipids and urea were also removed with petroleum ether and deionized water, respectively.Although untreated and treated muscles had similar amino acid compositions, the treated muscle lost 14 N and had a higher C: N ratio, indicating that urea affects isotope ratios and that water treatment removes urea without altering muscle composition.
Blood tissue is often used in stable isotope studies, as it is a relatively non-invasive sampling procedure, and serves as a benchmark for muscle tissue (Wallace et al., 2006;Caut et al., 2008;Dodge et al., 2011).
In order to advance the application of stable carbon and nitrogen isotopes in the field of animal science, a deeper understanding of complicating factors, such as tissue lipid content, the effects of the lipid extraction process on the species in question, and the use of anticoagulants for the collection of blood and plasma samples and their effects on the isotopic signal of the samples is required.
Therefore, in order to standardize sample preparation methods in isotopic studies, the objectives of this study were to determine if the lipid extraction processes alter the isotopic ratio (δ 13 C and δ 15 N) of selected tissues (breast muscle, thigh muscle, and liver) and eggs and determine if the use of anticoagulants interferes with the isotopic ratio (δ 13 C and δ 15 N) of blood and plasma.

MATERIAL AND METHODS
The experiment was conducted at the Department of Breeding and Animal Nutrition, School of Veterinary Medicine and Animal Science, University of the State of São Paulo (UNESP), Botucatu campus, Brazil.
Egg samples were obtained from a layer flock fed the same diet, with δ 13 C and δ 15 N isotopic ratios of -17.28 ± 0.06 ‰ and 2.24 ± 0.08 ‰, respectively.Thirty-two eggs were randomly divided into four treatments with eight replicates of one egg each.The treatments consisted of liquid eggs, dehydrated eggs (oven-dried, or freeze-dried), and eggs submitted to lipid extraction by ether or and chloroform + methanol.
Samples of the pectoral muscle, thigh and liver were obtained from a flock of broilers fed a diet with isotopic ratios of -17.78 ± 0.38 ‰ δ 13 C and 1.03 ± 0.40 ‰ δ 15 N. Twenty-four samples of each tissue were randomly divided into three treatments with eight replicates each.The treatments consisted of dehydrated tissues and tissues submitted to lipid extraction by ether or chloroform + methanol.
Blood samples were collected from layerflock fed the same diet, with δ 13 C and δ 15 N isotopic ratios of -17.28 ± 0.06 ‰ and 2.23 ± 0.07 ‰, respectively.Blood samples were distributed according to a randomized experimental design in a 3x3 factorial arrangement, consisting of three blood physical forms (liquid, ovendried, or freeze-dried), and blood collection with no anticoagulant, with EDTA or with heparin.Plasma results were analyzed according to a 3x2 in a factorial arrangement, consisting of three blood physical forms (liquid, oven-dried, or freeze-dried), and from blood collected with EDTA or heparin.
All samples were stored in a freezer at -20 ° C for up to 2 months until analyses.The dehydrated samples were obtained by drying the samples in a forcedventilation oven (Marconi, model MA 035, SP, Brazil) at 56ºC for 48 hours or freeze-drying in a lyophilizer (model L108, Liobras ® , SP, Brazil).The duration of the lyophilization process was 48 hours and the samples were dried under vacuum at -55°C and pressure around 50 μHg, and then cryogenically ground (Spex-6750 freezer-mill, Metuchen, USA) at -196ºC (Carrijo et al., 2000;Denadai et al., 2008).
Fat was extracted from the dehydrated and cryogenically-ground samples in a Soxhlet apparatus in the ether treatment (Carrijo et al., 2000;Denadai et al., 2008) and according to the method of Folch et al. (1957) in the chloroform + methanol treatment.

Isotopic analysis
The isotopic analyses were carried out at the Center for Stable Isotopes of the Institute of Biosciences of UNESP, Botucatu, SP, Brazil.
Samples weighing approximately 50-120 μg and 500-600 μg were used to measure 13 C/ 12 C, 15 N/ 14 N, respectively, whose CO 2 and N 2 gases were obtained by combustion and measured in a mass spectrometer, which determines the obtainedδ 13 C and δ 15 N enrichment values relative to their respective international standards.
The isotopic composition values were obtained relative to the international standard V-PDB (Vienna Pee Dee Belemnite) for δ 13 C and atmospheric air N 2 for δ 15 N, with analysis error of 0.2 ‰ and calculated by equation 1: where: δX represents the isotope enrichment of the chemical element X ( 13 C or 15 N) of the sample relative to the respective international standard, and R represents the ratio between the least abundant and the most abundant isotope.
Isotopic fractionation was calculated according to Hobson & Clark (1992) as the difference between the δ in the tissue and in the diet (equation 2), both for δ 13 C and for δ 15 N: Negative uppercase delta (∆) values indicate that the tissue is depleted in carbon-13 relative to diet (Hobson & Clark, 1992).

Statistical analysis
The obtained isotopic results were submitted to the multivariate analysis of variance (MANOVA) and univariate (ANOVA, complemented by Tukey's test) using the GLM procedure of the statistical program SAS (1996) or Minitab 16.

Egg samples
The mean δ 13 C and δ 15 N values of eggs were statistically analyzed (MANOVA) and generated regions with 95% confidence (Figure 1).MANOVA showed a statistical difference among the evaluated treatments studied (p<0.01).However, the ANOVA of the δ 13 C and δ 15 N values, separately, complemented by Tukey's test, detected statistical differences only for carbon-13 (Table 1).The results in Table 1 show no statistical differences (p>0.05)carbon-13 and nitrogen-15 between dehydrated and liquid eggs or between eggs submitted to fat extraction either by ether or by chloroform + methanol; however, fat-extracted eggs presented higher carbon-13 isotopic values than liquid and dehydrated eggs (p<0.01).Lipids are depleted in 13-carbon compared with proteins and carbohydrates (DeNiro & Epstein, 1978).However, the lipid-extracted eggs were enriched in approximately 0.52 ‰ of carbon-13 and depleted in 0.06 ‰, on average, ofnitrogen-15 (Table 2).These results conflict with those of Oppel et al. (2010), who examined the effects of the chemical extraction of avian egg yolk lipids on δ 13 C, δ 15 N and δ 34 S and found that lipid extraction by chemicals led to an increase of 3.3 ‰ in δ 13 C, 1.1 ‰ in δ 15 N, and 2.3 ‰ in δ 34 S.
The isotopic fractionation determined dietary δ 13 C and δ 15 N isotopic values of -17.28 ± 0.06 ‰ and 2.24 ± 0.08 ‰, respectively.The calculation of isotopic fractionation (Table 2) showed a decrease of 0.51 ‰ in carbon-13 and of 0.07 ‰ in nitrogen-15 as a result of lipid extraction.These values are much lower than those found by Hobson (1995), who studied the effects of lipid extraction in wild duck, quail, and falcon eggs.According to Denadai et al. (2008), the determination of the isotopic fractionation (Δ tissuediet) aims at reconstructing the diet, that is, finding the isotopic signature of the feed to which the animal submitted, and thus trying to predict the probable components of this diet.Considering the results of isotopic fractionation obtained in the present study (low variation among treatments), it is suggested that the feed and egg samples should be collected from the same batch in order to provide greater certainty of the birds' feeding history (Kennedy & Krouse, 1990).
The multivariate statistical analysis (Figure 2) showed that the isotopic ratios of the isotopic torque of the dehydrated pectoral muscle were statistically different (p<0.05) from that defatted by ether.Isotopic ratios of defatted thigh muscles were statistically different (p<0.05) from the dehydrated ones.The isotopic liver particle defatted with ether was statistically different (p<0.05) from that of defatted with chloroform + methanol.Hussey et al. (2008) did not find any significant difference in δ 13 C between fat and fat-free muscle tissues of two shark species, as observed in the thigh muscle in the present study.Despite the statistical δ 13 C difference in the pectoral muscle and liver between fat removal methods, the numerical difference was small (Δ dehydrated-ether (pectoral m.) = 0.35 ‰ and Δ dehydrated-ether (liver) = 0.48 ‰) between dehydrated and ether treatments, and there was no significant difference between dehydrated and chloroform + methanol samples.These results differ from those reported in several ecological studies, where the carbon-13 isotopic ratio differences of up to3 ‰ due to lipid extraction were found (Kiljunen et al., 2006, Sweeting et al., 2006, Post et al., 2007;Mintenbeck et al., 2008).
Relative to δ 15 N (Table 3), no significant differences (p> 0.05) the pectoral muscle among treatments were detected; however, in the thigh muscle, significantly higher δ 15 N (p<0.05) was detected in samples defatted with chloroform+methanol compared with etherdefatted and dehydrated samples, which were not different from each other (p> 0.05) and contained higher fat concentrations.Lower arithmetic difference in nitrogen-15 isotopic ratio between defatted and dehydrated samples (Δ dehydrated-defatted (average) = 0.12 ‰) was determined compared with carbon-13 ratio (f Δ dehydrated-defatted (average) = 0.27 ‰).In ecological studies, lipid extraction generally enriched the samples in +0.25 to +1.60 ‰ nitrogen-15 values, which are higher than those found in the present study (Pinnegar & Polunin,
The calculation of the isotopic fractionation of broiler tissues aimed at determining the dietary isotopes incorporated into the tissue and whether the lipid extraction method would cause greater differences between the tissues and the diets.The isotopic fractionation data (Table 4) show that, in general, the lipid extraction methods changed the diet-tissue fractionation (Δ TISSUE -DIET ).In tissues that have a high fat content (thigh muscle and liver), the process of fat extraction by ether caused a decrease in the isotopic fractionation of carbon-13.However, the isotopic fractionation of nitrogen-15 in the thigh muscle increased and did not change in the liver.Ecological studies typically evaluatecarbon-13 and nitrogen-15 isotopic ratios in bird tissues and fluids in order to predict their diet composition (Kennedy & Krouse, 1990).The advantage of working with poultry for animal science purposes is that it is possible to determine the actual isotopic fractionation of the tissues in relation to the diet, and not only to estimate this fractionation, with advantages of applying these data in several studies hereafter.
A number of authors have reported that lipid extraction is performed to normalize the ratio of stable carbon isotopes, which, however, has also been shown

Standardization of A Sample-Processing Methodology for Stable Isotope Studies in Poultry
eRBCA-2019-0687 to affect the stable nitrogen isotope ratio in fish and invertebrate tissues (Pinnegar & Polunin, 1999;Murry et al., 2006;Post et al., 2007;Logan et al., 2008).Considering the results of the present study, the possibility of not performing lipid extraction needs to be considered, since lower isotopic variation was found in eggs and broiler tissues in the present study when compared with that reported in other species.This isotopic difference from ecological studies may be caused by a number of factors, such as diet diversity, diet heterogeinity during the day, and lipid deposition, which in wild birds may have different purposes compared with poultry, such as temperature control, energy storage for migratory flight, etc. (Kennedy & Krouse, 1990;Hobson & Clark, 1992), leading led to greater isotopic fractionation for the production of the body lipids of migratory birds compared to laying hens and broilers.

Blood samples
The mean δ 13 C and δ 15 N values of blood samples submitted to different drying methods and anticoagulants are shown in Table 5, and the δ 13 C and δ 15 N isotopic ratios of the plasma are compiled in Table 6.No interaction between drying methods and anticoagulants were detected in either fraction (blood and plasma) for both elements (δ 13 C and δ 15 N), with P values of 0.473 and 0.690 for δ 13 C and δ 15 N in the blood, respectively, and of 0.983 for δ 13 C and 0.802 for δ 15 N in the plasma.Blood δ 13 C isotopic values (Table 5) were not affected either by the drying methods or by the use of anticoagulants (p>0.05);however, the drying process influenced δ 15 N values (p<0.05), which were lower in lyophilized than in liquid blood.
The objective of calculating the isotopic fractionation (Δ tissue -diet) (Tables 7 and 8) of the blood fractions submitted to the experimental treatments was to determine how the dietary isotopes were incorporated into the tissues and if the drying and the fat removal methods would cause greater differences between the tissues and diets.Both lyophilization and EDTA resulted in higher carbon-13 tissue (blood and plasma) to diet ratios.Among blood collection methods, the blood and plasma of laying hens containing the anticoagulant EDTA promoted the highest nitrogen-15 tissue to diet ratio whereas lyophilization resulted in the lowest ratio among blood processing methods.Lemons et al. (2012) studied the effect of the use of anticoagulants for collection of blood samples and its fractions in turtles on δ 13 C δ 15 N isotopic values, and observed 15 N depletion in plasma containing EDTA and 15 N enrichment in heparin-collected blood, which was not found in the present study.

Figure 1 -
Figure 1 -Confidence region determined by the difference between isotopic δ13C and δ15N values of eggs in each treatment compared to the liquid treatment (n = 8).

Figure 2 -
Figure 2 -Principal component analysis carbon-13 and nitrogen-15 in broiler tissues submitted to dehydration and lipid extraction using ether or chloroform + methanol, as determined by multivariate analysis of variance (MANOVA).

Table 1 -
Isotopic δ13C and δ15N values (mean ± SD) of liquid, dehydrated and lipid-extracted (ether or chloroform and methanol) eggs.Different letters in the same column differ statistically by the Tukey's test at p<0.05 *

Table 3 -
Isotopic δ13C and δ15N values (mean ± SD) of broiler tissues submitted to dehydration and lipid extraction by ether or chloroform and methanol.Different letters in the same row differ statistically by the Tukey's test at p<0.05, *isotopic pair every two lines differ statistically by MANOVA p<0.01

Table 4 -
Isotopic fractionation factor (∆tissue -diet) of broiler tissues submitted to dehydration and lipid extraction by ether or chloroform and methanol.

Table 5 -
Isotopic values of δ13C and δ15N (‰) (mean ± SD) of blood samples submitted to different drying methods and collected or not with anticoagulants*.
*Different letters in the same column differ statistically by the Tukey's test at p<0.05.

Table 6 -
Isotopic values of δ13C and δ15N (‰) (mean ± SD) of plasma samples submitted to different drying methods and collected or not with anticoagulants*.
*Different letters in the same column differ statistically by the Tukey's test at p<0.05.

Table 7 -
Isotopic fractionation factor (∆tissue -diet) of blood samples submitted to different drying methods and collected or not with anticoagulants.