Enhanced dietary reconstruction of Korean prehistoric populations by combining δ13C and δ15N amino acids of bone collagen

Kyungcheol Choy; Hee Young Yun; Benjamin T. Fuller; Marcello A. Mannino

doi:10.1371/journal.pone.0300068

Abstract

Compound specific stable isotope analysis of amino acids (CSIA-AA) is a powerful tool for determining dietary behaviors in complex environments and improving dietary reconstructions. Here, we conducted CSIA-AA on human (n = 32) and animal (n = 13) remains from two prehistoric archaeological sites (Mumun, Imdang) to assess in more detail the dietary sources consumed by prehistoric Korean populations. Results of estimated trophic position (TP) using Δ¹⁵N_Glx-Phe show that the Imdang individuals consumed aquatic resources, as well as terrestrial resources. Principal component analysis (PCA) using δ¹³C and δ¹⁵N essential amino acid (EAA) values show that the Imdang humans closely cluster with game birds and terrestrial herbivores, whilst the Mumun humans closely cluster with C₄ plants. Quantitative estimation by a Bayesian mixing model (MixSIAR) indicates that the Imdang humans derived a large proportion of their proteins from terrestrial animals and marine fish, whereas the main protein sources for the Mumun humans were C₄ plants and terrestrial animals. Additionally, the comparison between the EAA and bulk isotope models shows that there is a tendency to overestimate the consumption of plant proteins when using bulk isotopic data. Our CSIA-AA approach reveals that in prehistoric Korea there were clear differences in human diets through time. This study adds to a growing body of literature that demonstrates the potential of CSIA-AA to provide more accurate estimations of protein consumption in mixed diets than previous bulk isotopic studies.

Citation: Choy K, Yun HY, Fuller BT, Mannino MA (2024) Enhanced dietary reconstruction of Korean prehistoric populations by combining δ¹³C and δ¹⁵N amino acids of bone collagen. PLoS ONE 19(3): e0300068. https://doi.org/10.1371/journal.pone.0300068

Editor: John P. Hart, New York State Museum, UNITED STATES

Received: November 16, 2023; Accepted: February 20, 2024; Published: March 27, 2024

Copyright: © 2024 Choy et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the paper and its Supporting information files.

Funding: This work was supported by Research Funds for Young Faculty Club at Hanyang University (HY-2018-00000001315) and by the National Research Foundation of Korea (NRF) funded by the Ministry of Education NRF-2022S1A5A2A03051382) and also supported by the Aarhus University Research Foundation (AUFF-E-2015-FLS-8-2).

Competing interests: The authors have declared that no competing interests exist.

Introduction

In the last decades, carbon and nitrogen stable isotope ratio measurements of bulk collagen have become a common method to investigate dietary patterns and subsistence activities in ancient populations [1–3]. Nevertheless, dietary reconstructions based on bulk stable isotope ratios can be hampered by ecological and metabolic factors [4–6]. For example, numerous food sources have similar bulk δ¹³C and δ¹⁵N values, which reduces the ability to fully separate dietary sources [4–7]. Therefore, if humans and animals consumed complex diets, bulk stable isotope compositions may not be sufficient to interpret the exact mix of foodstuffs consumed. Moreover, the same plant species at the base of the terrestrial food webs can have a large variability in bulk stable isotope values, owing to the isotopic compositions of their soils of origin and due to possible manuring [8, 9]. Thus, bulk isotopic values of each food source in human diets can vary according to local environments and soil conditions [4, 8, 9]. Furthermore, bulk isotope data can be strongly influenced by metabolic parameters such as dietary routing, and consumer-to-diet isotopic offsets [10]. However, these diverse factors are not often considered in bulk stable isotopic studies. Thus, it can be challenging to determine the isotopic baseline of ancient food webs and discriminate food sources with sufficient accuracy.

In response, new isotopic methods have emerged to reconstruct subsistence practices in past populations [4, 11, 12]. In particular, there has been an increase in the use of compound specific isotope analysis of individual amino acids (CSIA-AA) on bone collagen from archaeological contexts [12–16]. CSIA-AA is a relatively new method that allows more accurate dietary reconstructions in past populations than the bulk isotopic method and consequently provides trophic web reconstructions of past environments [11, 12, 17–19]. Amino acids (AAs) can be divided into two groups on the basis of their synthesis and function: essential amino acid (EAA) and non-essential amino acid (NEAA). While this constitutes the conventional definition of essential (EAA) and nonessential amino acids (NEAA), and this definition is not solely confined to dietary consideration but is also extended to include the role of AAs in supporting protein deposition and growth during different stages of life [20]. The carbon isotope values in AA metabolism are neatly aligned with the conventional distinction between EAA and NEAA. Thus, EAA δ¹³C values can yield more detailed dietary information since they are directly routed from the diet to the consumer’s tissues [21–23]. Therefore, the EAA δ¹³C values in bone collagen provide detailed information on the carbon sources of the foods consumed, and this can be a useful approach to investigate dietary sources in prehistoric populations [5, 11–13].

In contrast to δ¹³C values in AA pathways, δ¹⁵N AA values do not conform to the conventional definition of EAA and NEAA. Instead, there has been the development of the novel definition of source (Tyr, Lys, Phe, Ser, Gly), metabolic (Thr) and trophic AAs (Pro, Ala, Glx, Val, Leu, Asx) in nitrogen AAs [24–28]. However, it is important to be cautious when applying the meaning of source and trophic AAs in nitrogen to dietary studies as these definitions can be variable due to the species studied and developmental age. For instance, among the known source AAs (Tyr, Lys, Phe, Ser, Gly) in nitrogen, glycine and serine have been found to have a diet-to-consumer offset in pigs (Δ¹⁵N _Ser: 3.2‰ in C₃ diets and 2.6‰ in C₄ diets, Δ¹⁵N_Gly: 0.8‰ in C₃ diets and 2.2‰ in C₄ diets) [29]. Only phenylalanine (Phe) appears to exhibit a relative ~0‰ diet-to-consumer offset in many species including humans [26, 30, 31]. Thus, we cannot directly substitute source AAs to EAAs for all animal species and humans, and more research on the definition of source, trophic and metabolic AAs is necessary. However, previous studies have shown that the two δ¹⁵N_AA values (Glx, Phe) in bone collagen are useful for estimating the trophic level of humans and animals in food webs [15, 18, 31, 32]. The δ¹⁵N values of glutamic acid/glutamine (Glx) (trophic AA) become systematically enriched from food sources to consumers, while Phe (source AA) remain relatively unchanged through this process. A comparison of trophic AAs and source AAs (δ¹⁵N_Glx-Phe) can provide a more definitive indicator of trophic position of humans and animals than bulk stable isotope data [33, 34].

Despite the promise of CSIA-AA, most dietary reconstruction research has focused on only one of these measurements, either δ¹³C_AA or δ¹⁵N_AA values of bone collagen due to the complexity of the methodological development and the relatively high cost in terms of time and analytical expenses. Until now, there have been only a few applications of CSIA-AA using both δ¹³C_AA and δ¹⁵N_AA of bone collagen from archaeological sites [14–16, 35]. However, the application of both δ¹³C_AA and δ¹⁵N_AA for dietary reconstructions has increased the possibility of dealing successfully with the issues of food source equifinality, influence of baseline variation, and uncertainties associated with consumer-to-diet isotopic offsets [4]. Thus, it is required to have more applications of both δ¹³C_AA and δ¹⁵N_AA to dietary reconstructions in prehistoric populations from different regions.

In South Korea, the application of bulk isotopic analysis to human remains has recently increased, and covered topics such as crop cultivation, animal domestication, and social status [36–39]. However, there has been only a few studies on prehistoric Korean populations using δ¹³C_AA [6, 36], but no studies using both δ¹³C_AA and δ¹⁵N_AA. In this study, we applied CSIA-AA (both δ¹³C_AA and δ¹⁵N_AA) to humans (n = 32) and animals (n = 13) from two Korean prehistoric sites: Mumun and Imdang. The Mumun sites belong to the Bronze Age (1500–100 BC) when early agriculture was first introduced to the Korean peninsula, and the Imdang site is dated to the Proto-Three Kingdom Period (BC 80–394 AD) when the early ancient states emerged on the Korean peninsula. Thus, the dietary practices from the two different-era sites can offer important clues to the spread of early agriculture and the subsistence economy of ancient states in prehistoric Korea. Previous bulk isotope data showed that C₄ plants such as millets were the main dietary sources in the Mumun sites [36], and there was no consumption of aquatic resources in the diet of the Mumun. On the other hand, at the Imdang site, C₃ plants and terrestrial animals were the main dietary sources [37], and bulk isotope studies suggest that notable proportions of marine animal protein were consumed. However, with bulk stable isotope data, it is difficult to identify minor resources in the millet-dominated Mumun diets or to separate each source in the multiple food groups in the Imdang diets.

Along with the CSIA-AA, we also employ a Bayesian mixing model to estimate the proportional contribution of food sources for each individual. In archaeology, there are two widely used Bayesian models: FRUITS [40] and MixSIAR [41, 42]. The MixSIAR can be used to estimate the contribution of diets in ecological, as well as archaeological studies. In this study, we used the MixSIAR to compare the estimated dietary compositions of two or more groups, provided all groups are directly comparable [10, 42]. For the dietary sources in the model, we additionally measured both δ¹³C_AA and δ¹⁵N_AA of animal dietary sources (marine animals, game birds, terrestrial herbivores) from the archaeological sites, and plant sources (n = 9) from modern farms in South Korea. Overall, the focus of this study is to reveal the significance of the domesticated grains (rice vs. millet) in Mumun diets, and to more accurately distinguish the dietary sources of the two above-mentioned prehistoric Korean populations using the CSIA-AA (both δ¹³C_AA and δ¹⁵N_AA), and to quantify the dietary variability in more detail, using MixSIAR.

Archaeological context

Mumun burials.

The three Mumun burial sites (Hwangsok-ri, Jungdo, Maedun) are all located around the upstream areas of the Han River on the eastern Korean Peninsula [36] (Fig 1). The radiocarbon dates on the human collagen from the three Mumun burials showed that they belong to the Middle Mumun (535–635 BC) period [36]. Korean archaeologists refer to the Mumun period (1500–100 BC) as the Bronze Age in Korean prehistory. The Mumun culture is characterized by crop agriculture with large dry-fields, and settlements with ditch enclosures [43]. Most of the Mumun sites in Korean prehistory are related to human burials, such as cemeteries that do not contain animal remains. Thus, there are very few animal remains recovered from Mumun sites. However, it is suggested that people in the Mumun period continued to fish and hunt wild animals. Archaeobotanical data show that the Mumun people cultivated rice, which was believed to be the most important crop [44–46]. It is assumed that the social and economic complexities during the Mumun period are associated with the spread of rice agriculture. In addition to rice, the Mumun also cultivated millets, barley, wheat, legumes [44, 47]. In particular, studies on charred grains have suggested that millets might also be a dominant crop in the Mumun diets along with rice [44]. However, despite abundant evidence for the cultivation of millets during the Mumun period, it has been difficult to establish the significance of the domesticated grains (rice vs. millet) in Mumun diets.

Download:

Fig 1.

Map of the Korean Peninsula showing the location of the Mumun and Imdang sites studied in this paper (A), and photos showing the two types of burials from the different time periods: exposure of the Mumun burial (Goindol) from Hwangsuk-ri after removing the cover stone (B) and the Imdang burial composed of a double chamber (square and rectangle chamber) (C).

https://doi.org/10.1371/journal.pone.0300068.g001

Imdang burial mounds.

The Imdang burial mounds constitute a well-known archaeological site in South Korea, located on a hillside in the middle of an alluvial plain that was formed by the Kumho River near Gyeongsan City [48] (Fig 1). This burial site contained approximately 1600 burials and a large amount of grave goods such as gilt-bronze crowns and ornaments, pottery, iron weapons and tools, as well as human and animal bones [49, 50]. Radiocarbon dates confirm that this mortuary site was occupied during the Proto-Three Kingdoms period (BC 80–394 AD) [37]. In Korean archaeology, the Proto-Three Kingdom Period is characterized by the occurrence of local polities with intensive agriculture, an increase in iron production, and new pottery production techniques [51, 52]. Archaeobotanical data show that plant resources in Imdang society include domesticated crops (rice, foxtail millet, barnyard millet, perilla) and fruits (peach, persimmon, apricot). In particular, rice grains were found on the floor of the chamber and inside pottery [53]. Further, the Imdang burials also contained the skeletal remains of terrestrial and marine animals. Animal species found at the Imdang site include game birds (pheasant, wild goose, great bustard, swan), terrestrial herbivores (deer, wild boar, hare), and marine fauna (shark, amberjack, sea bream, rockfish, puffer fish, shellfish) [53, 54] (Fig 2). According to the zooarchaeological studies, the most represented terrestrial animals were game birds such as pheasant and wild goose [53–55] and the most common marine animals were shark, amberjack and shellfish [54, 56].

Download:

Fig 2. Example of animal species found in an Imdang burial (Joyeung EII-2ho).

The Imdang burial is composed of a double chamber tomb, with a main chamber (rectangle) and auxiliary chamber (square). Each chamber contains sacrificed humans (sunjang) and a variety of wild birds (pheasant, wild goose, great bustard, duck, crane) and marine fish (shark, amberjack, sea bream, flatfish, puffer fish, shellfish), which were used as ritual offerings in the owner’s funeral.

https://doi.org/10.1371/journal.pone.0300068.g002

Materials and methods

Archaeological humans and animals

We selected well-preserved humans (n = 4) and animals (n = 2) from the Mumun sites and humans (n = 28), and animals (n = 11) from the Imdang site. It should be noted that it is extremely difficult to obtain humans and animals from the Mumun period as these remains are recovered from inland mountain areas of South Korea that are not good for bone preservation due to acidic soil conditions. Thus, this is the reason why so few specimens from the Mumun sites were analyzed in this study. For the Mumun sites, the faunal specimens selected are one deer and wild boar, which were analyzed to provide δ¹³C_AA and δ¹⁵N_AA values of terrestrial sources in the archaeological context. However, numerous well-preserved human and animal specimens were recovered from the Imdang mounds dated to the Proto-Three Kingdom period. We selected Imdang humans (n = 28) to investigate dietary patterns among individuals. As potential dietary sources at the Imdang site, we chose pheasant (n = 1), great bustard (n = 1), wild goose (n = 1) and swan (n = 1) for game birds, and we selected hare (n = 1), wild boar (n = 1), pig (n = 1) and cattle (n = 1) for terrestrial animals. In terms of marine species, sandbar shark (n = 2) and amberjack fish (n = 1) were chosen for the CSIA-AA (both δ¹³C_AA and δ¹⁵N_AA). These marine and terrestrial animals in the Imdang burials were selected as likely food sources for the CSIA-AA (Fig 2).

Modern cereals

In order to reconstruct the human diets in prehistoric Korea, it is necessary to have isotopic data of ancient plants (grains, fruits, and vegetables). However, no plant samples from the two prehistoric Korean sites were available to the CSIA-AA due to the inadequate preservation of floral remains, and there is no published isotopic data on plant remains from the Mumun and Proto-Three Kingdom periods. Thus, we collected C₃ crops (rice, soybean, adzuki bean, wheat, barley, and oat) (n = 6) and C₄ crops (common millet, fox millet, and sorghum) (n = 3) from modern farms in South Korea. These edible cereals are commonly found in archaeological sites dated to the Mumun and Proto-Three Kingdom periods [57], and they were measured for the CSIA-AA as reference materials for potential plant sources in prehistoric Korea.

Collagen extraction

Collagen extraction was completed at the Moesgaard Archaeo-Science Laboratory (Department of Archaeology and Heritage Studies, Aarhus University) according to the protocols outlined in Richards and Hedges [58] with the addition of an ultrafiltration step [59]. Bone samples were cleaned and demineralized in 0.5M HCl at 4°C. Demineralized bones were then rinsed in deionized water and gelatinzed at 70°C in a pH 3 solution for 48 hours. The insoluble fraction was then filtered, first with 5mm Ezee filters and then the remaining solution was filtered to collect purified collagen using >30 kDa filters. The purified solution was then frozen and freeze- dried for 48 hours.

Preparation of amino acids for GC-C-IRMS

For AA hydrolysis, 1 mg of extracted bone collagen was hydrolyzed with 1 mL 6M HCl (110°C, 24 h). Modern cereal samples were homogeneously powdered, washed three times with deionized water, and freeze-dried. Lipids were first extracted from the powdered cereal samples with dichloromethane/methanol (2:1 v/v, 10ml) by ultra-sonication. Approximately, 3 mg of dried powders were hydrolyzed (6M HCl, 2ml, 110°C, 24 h). The AA hydrolysates for the cereal samples were separated from the lipophilic fraction by adding n-hexane/dichloromethane (6:5, v/v, 2mL). A standard mixture of 12 AAs was prepared alongside the samples (Sigma-Aldrich Chemie; Fluka Chemie). An internal standard (14 μl of norleucine, Sigma-Aldrich) was added, and the hydrolyzed collagen and cereal samples were blown down to dryness under N₂ and prepared for AA derivatization [30–32].

For δ¹⁵N_AA measurements, AAs were derivatized to using the N-pivaloyl-i-isopropyl ester (NPIP) method [60]. For preparation of the NPIP derivatives, the hydrolysate was first esterified using a thionyl chloride/2-propanol (v/v, 1:4, 110°C, 2h) and then derivatized using pivaloyl chloride/dichloromethane (v/v, 1:4, 110°C, 2h). Subsequently, the AA derivatives were stored in 250 μL dichloromethane at -20°C in a freezer until analysis. For δ¹³C_AA measurements, a different AA derivatization procedure using N-acetyl methyl ester (NACME) method was employed. [60]. For the preparation of the NACME derivatives, the hydrolysates were methylated by acidified methanol (v/v, 1:4, 75°C, 1h). Each sample was acetylated with acetic anhydride/triethylamine/acetone (v: v: v, 1:2:6) for 10 min at 60°C. Following each reaction, reagents were evaporated under nitrogen. The AA derivatives were stored in ethyl acetate (250 μL) at -20°C in a freezer until GC-C-IRMS analyses.

Gas chromatography-combustion-isotope ratio mass spectrometry (GC-C-IRMS)

Gas chromatography-combustion-isotope ratio mass spectrometry (GC-C-IRMS) was performed at the Department of Marine Sciences, Hanyang University, South Korea. For δ¹⁵N_AA results, these were analyzed with a gas chromatograph (HP 6890N, Agilent Technologies, US) connected to a furnace (GC5 Interface, Elementar, Germany) and an isotope ratio mass spectrometer (Isoprime 100, Elementar, Germany). HP-Ultra 2 (50 m length, 0.32 mm inner diameter, 0.52 μm film thickness) was equipped to the GC. The initial oven temperature was 40°C for first 5 min. The oven temperature was increased at a rate of 15°C/min to 110°C, a rate of 3°C/min to 150°C, a rate of 6°C/min to 220°C (10 min hold), and a rate of 20°C/min to 250°C (8 min hold). As the NPIP derivatization procedure can potentially result in differences between the proline and hydroxyproline [61], the collagen specimens were again derivatized with the NACME protocol to confirm the proline and hydroxyproline δ¹⁵N results. Samples were injected into the GC, equipped with VF-23ms column (30 m length, 0.32 mm inner diameter, 0.25 μm film thickness). The initial oven temperature was 60°C for the first 1 min. The oven temperature was increased at a rate of 15°C/min to 120°C, a rate of 3°C/min to 190°C, and a rate of 15°C/min to 250°C (15 min hold). The initial oven temperature was 60°C for the first 1 min. The oven temperature was increased at a rate of 15°C/min to 120°C, a rate of 3°C/min to 190°C, and a rate of 15°C/min to 250°C (15 min hold). 10 certified AA standards were used, which were purchased from SHOKO-Science (Japan) and Indiana University (USA). In addition, a standard mixture of 12 AAs (alanine, glycine, valine, leucine, isoleucine, proline, aspartic acid, threonine, serine, glutamic acid, phenylalanine, and hydroxyproline) was run for testing instrumental and analytical accuracy. Standard nitrogen isotope ratios were expressed as δ¹⁵N value per mil relative to Ambient Inhalable Reservoir (AIR). Analytical error (SE) of the δ¹⁵N_AA measurement of the standard AAs was checked, and ranged from 0.31‰ for δ¹⁵N glycine to 0.79‰ for δ¹⁵N threonine.

The δ¹³C_AA were analyzed with an IRMS instrument (visION, Isoprime) connected with a Hewlett Packard 7890 N series gas chromatograph (Agilent Technologies) via a combustion interface (glass tube packed with copper oxide, operated at 850°C). Samples were injected into the GC, equipped with VF-23ms column (30 m length, 0.32 mm inner diameter, 0.25 μm film thickness) (at a constant 1.3 mL/min flow rate) as a carrier gas. The initial oven temperature was 60°C for first 1 min. The oven temperature was increased at a rate of 15°C/min to 120°C, a rate of 3°C/min to 190°C, and a rate of 15°C/min to 250°C (15 min hold). Standard carbon isotope ratios were expressed as δ¹³C value per mil relative to Vienna-PeeDee Belemnite (VPDB). Standard deviations of carbon isotope measurements were generally ±0.4‰ as determined by repeated injections of the 10 AA standard mixture and each measured mean AA δ¹³C was corrected relative to the mean AA δ¹³C of standards to account for the exogenous carbon and kinetic fractionation introduced during derivatization [62]. The δ¹³C values of the underivatized AAs in the standard were measured via EA-IRMS, and were used in combination with the measured δ¹³C values of their derivatives to calculate correction factors for each amino acid, as described in Silfer et al. [63]. These correction factors were then used to calculate the δ¹³C values of the AAs in the samples from the δ¹³C values of their derivatives. Analytical error (SE) of δ¹³C_AA measurements was estimated from the AA standards, and ranged from 0.10‰ for δ¹³C valine to 0.45‰ for δ¹³C phenylalanine.

Statistical analysis

Prior to statistical analysis, all isotopic data were tested for univariate normality by Q-Q plots. According to ecological, archaeological and isotopic relevance, potential food sources were categorized into five food groups: terrestrial herbivores (i.e., deer, wild boar, hare), marine fish (i.e., sandbar shark, amberjack), game birds (i.e., pheasant, bustard, wild goose, and swan), C₃ plants (i.e., rice, wheat, barley, oat, adzuki bean, soybean), and C₄ plants (foxtail millet, broomcorn millet, sorghum). For example, wild goose and swan as waterfowl have different habitats from terrestrial birds, such as pheasant and the great bustard in terms of ecological niche. However, due to their terrestrial-based isotopic similarity, we grouped these waterfowl as game birds [37]. Isotopic data on three animal groups (game birds, marine fish, terrestrial herbivores) from the Imdang burials, and two plant groups (C₃, C₄ plants) from modern farms were used for the comparison of dietary sources. For the separation of dietary sources, we used principal component analysis (PCA) to compare EAA δ¹³C and δ¹⁵N values of human collagen in relation to the five food groups. EAAs are those that human and animals cannot synthesize and therefore should only come from dietary intake, and thus, for the PCA analysis, we utilized the five EAA δ¹³C values (Ile, Leu, Phe, Thr, Val) and five EAA δ¹⁵N values (Ile, Leu, Phe, Thr, Val). Along with these five EAA δ¹³C values, we also investigated the effectiveness of three source δ¹⁵N values (Ser, Gly, Phe) in distinguishing each food group.

In conjunction with the PCA analysis, we used a Bayesian mixing model (MixSIAR) to estimate the contribution of sources to the whole diet based on five EAA δ¹³C and δ¹⁵N_Phe proxies. In the MixSIAR, EAA isotopic values of humans were input as raw data and sources were input as means and standard deviations. For the bulk isotope model, it is required to define the trophic discrimination factors (TDF) of each food source for the δ¹³C and δ¹⁵N values (C₃ plant: Δ¹³C = 5.2‰ and Δ¹⁵N = 3.8‰; C₄ plant: Δ¹³C = 4.5‰ and Δ¹⁵N = 3.8‰; animal: Δ¹³C = 1‰ and Δ¹⁵N = 3.8‰) [37]. However, in the EAA isotope model, we used a TDF of 0‰ for the EAA δ¹³C and δ¹⁵N_Phe values due to zero consumer-to-diet isotopic offsets. The EAA isotope model was run at length normal [41, 42]. The model is fit via Markov Chain Monte Carlo, which estimates entire posterior distributions for each variable, given the data. We set error structure as a multiplicative process. Data analyses were conducted using JAGS and R software [64]. All values in the text are given as mean ± standard deviation. The level of significance was set at P ≤ 0.05.

Results

δ¹⁵N_AA values

Nitrogen AA results for the human, animal, and cereal samples analyzed in this study are presented in S1-S3 Tables in S1 File. We obtained δ¹⁵N_AA results for 10 bone collagen AAs from the human and animal samples: seven trophic AAs (Ala, Glx, Hyp, Pro, Val, Ile, Leu) and three source AAs (Phe, Gly, Ser) (S1-C and S2-C Tables in S1 File). For the cereals, 9 AAs were determined: six trophic AAs (Ala, Glx, Pro, Val, Ile, Leu) and three source AAs (Phe, Gly, Ser) (S3-C Table in S1 File). In order to assess quality control (QC), we used the relationship between proline and hydroxyproline since the hydroxylation of proline to form hydroxyproline does not involve exchange of nitrogen atoms [61]. Thus, a plot of δ¹⁵N_Hyp values against δ¹⁵N_Pro values is expected to have a slope close to 1 [65]. Our result revealed that the regression line of the δ¹⁵N_Hyp vs. δ¹⁵N_Pro values is close to y = x, demonstrating the excellent quality of the data (R² = 0.994; Fig 3A).

Download:

Fig 3.

Quality control plot of δ¹⁵N_Pro vs. δ¹⁵N_Hyp values (A) for human and animal collagen and trophic position (TP) plot of compound specific nitrogen isotope values in phenylalanine and glutamic acid (B) of humans and animals from prehistoric Korean sites. The estimated TP lines are defined according to Chikaraishi et al. [17, 31]. Humans are located between TP of C₃ = 3 (black dot-line) and TP of Aquatic = 2 (blue dot-line) estimates suggesting that they might exploit both terrestrial and aquatic resources.

https://doi.org/10.1371/journal.pone.0300068.g003

The δ¹⁵N_AA values were applied to estimate trophic level (TP) of humans and associated animals, and assess which species were consumed by humans. Our results revealed the TP of humans and their food sources (Fig 3B). In TP estimates, we applied a β factor of -8.4‰ and TDF_Glx-Phe value of 7.6‰ for the animals and terrestrial C₃ plants, but used a β factor of -0.4‰ and TDF_Glx-Phe value of 7.6‰ for C₄ plants [17, 31]. The vertical shift between each line represents a different TP in Fig 3B. Estimated TP of modern cereals (C₄ plants) are lower than 1.0 but the TP of terrestrial C₃ plants is 1.5–2.0, and this is relatively higher than the TP of C₃ plants. Estimated TP of terrestrial herbivores (cattle, pig, wild boar, hare, deer) from the Mumun and Imdang sites are in agreement with their feeding habits. The terrestrial herbivores show a TP of 2.1–2.2, consistent with the expected values for pure herbivores (i.e. 2.0). Although livestock (cattle, pig) have ¹⁵N-enriched bulk collagen values, their TP (2.2) are similar to terrestrial herbivores, which suggests that the high bulk δ¹⁵N results of the cattle and pigs is due to high δ¹⁵N values of plants, likely grown by manuring. Game birds (pheasant, great bustard, wild goose, swan) also show similar TP values (1.9 to 2.4) consistent with terrestrial herbivorous diets, even though there is a 4.6‰ difference in the bulk δ¹⁵N values among the game birds.

The humans from the Mumun and Imdang sites presented in this study have a wide range of TP values from 2.5 to 3.3. Bulk stable isotopic data showed that the Mumun humans have low bulk δ¹⁵N values (6.8‰ to 7.6‰), and that the Imdang humans have a wide range of δ¹⁵N values (7.6‰ to 13.3‰) [36, 37]. The Mumun humans had TPs that ranged between 2.5 to 2.9, and this indicates that the low bulk δ¹⁵N values were likely caused by the low bulk δ¹⁵N values of foods in the prehistoric environments. In contrast, the Imdang humans have a wider range of TP values (2.5 to 3.3), which could suggest the additional consumption of marine resources. In general, terrestrial carnivores show TP_AA values ranging from 2.9 to 3.1 [15]. Among the 28 Imdang individuals, 11 individuals have higher TP values (from 3.1 to 3.3) than the terrestrial carnivores (Fig 3B). Zooarchaeological studies reported that marine fish and shellfish were commonly identified at the site of Imdang, but not terrestrial carnivores [54–56]. Thus, we can assume that these 11 Imdang individuals having higher TP values than the carnivores could have consumed marine resources rather than terrestrial carnivores.

δ¹³C_AA values

Carbon AA results of the humans, animals, and cereals are presented in S1-S3 Tables in S1 File. The δ¹³C_AA results of 13 bone collagen AAs, six EAAs (Ile, Leu, Lys, Phe, Thr, Val) and seven NEAAs (Ala, Asx, Glx, Gly, Hyp, Pro, Ser), from the human and animal samples are presented (S1-B and S2-B Tables in S1 File), and the δ¹³C_AA results of 13 plant amino acids, six EAAs (Ile, Leu, Phe, Thr, Val, Tyr) and seven NEAAs (Ala, Asx, Glx, Met, Gly, Pro, Ser), from the cereals are presented (S3-B Table in S1 File). For the quality control (QC) of the data, we tested the regression line of the δ¹³C_Hyp and δ¹³C_Pro values from the Mumun and Imdang sites, and it is close to a 1:1 line, indicating the excellent quality of the data (R² = 0.969) (Fig 4A).

Download:

Fig 4.

Quality control plot of δ¹³C_Pro vs. δ¹³C_Hyp values (A) for human and animal collagen, and Δ¹³C_Gly-Phe vs. Δ¹³C_Val-Phe of human and animal samples from Mumun and Imdang sites (B). All humans overlap with the terrestrial animals in Δ¹³C_Gly-Phe vs. Δ¹³C_Val-Phe, suggesting consumption of terrestrial dietary sources.

https://doi.org/10.1371/journal.pone.0300068.g004

Previous studies have shown that the Δ¹³C_Gly-Phe values are an effective proxy for discriminating between terrestrial and marine sources [6, 7, 11, 66]. The average Δ¹³C_Gly-Phe value for the Mumun humans was 12.4±1.2‰ and the average Δ¹³C_Gly-Phe value for the Imdang humans was 13.3±1.0‰, similar to terrestrial animals. Along with Δ¹³C_Gly-Phe values, it has been suggested that Δ¹³C_Val-Phe values are also effective at discriminating between terrestrial and aquatic sources [5, 11, 66]. Our Δ¹³C_Val-Phe results show that Δ¹³C_Val-Phe proxy can separate terrestrial from aquatic sources (Fig 4B). For the humans, the Mumun individuals have average Δ¹³C_Val-Phe values with a mean of -0.2‰ and the Imdang individuals have average Δ¹³C_Val-Phe values with a mean of 1.7‰, similar to terrestrial animals. In addition, Δ¹³C_Val-Phe can separate terrestrial (C₃ and C₄) plants from terrestrial and aquatic animals. Thus, our Δ¹³C_Gly-Phe and Δ¹³C_Val-Phe proxies indicate that the most of dietary sources in the Mumun and Imdang humans are from terrestrial resources (Fig 4B), although the TP estimates reveals that several Imdang individuals (n = 11) consumed aquatic sources.

Principal component analyses of EAAs

To assess the ability to detect differences in food groups, we conducted principal component analyses (PCA) using three options: 1. Five EAAs δ¹³C values (Phe, Leu, Val, Ile, Thr) and four EAAs δ¹⁵N values (Val, Leu, Ile, Phe); 2. Five EAA δ¹³C values (Phe, Leu, Val, Ile, Thr) and three source AA δ¹⁵N values (Ser, Gly, Phe); 3. Five EAA δ¹³C values (Phe, Leu, Val, Ile, Thr) and one source AA δ¹⁵N value (Phe) (S1 Fig in S1 File). Among the three PCA options tested here, PCA option 2 and 3, utilizing source AAs, exhibit a good separation and clustering of each food group (S1 Fig in S1 File). However, the most effective separation and clustering of each food source (marine, terrestrial, C₃, and C₄ plants) based on metabolic pathways was achieved through the combination of all five EAA δ¹³C values (Phe, Leu, Val, Ile, Thr) and four δ¹⁵N values (Val, Leu, Ile, Phe) (Fig 5). Detailed PCA data of nine EAAs from humans and food sources are shown in S4 Table in S1 File. The δ¹³C values of five EAAs (Phe, Leu, Val, Ile, Thr) were significantly different between C₃ and C₄ plants. The δ¹⁵N values of four EAAs (Val, Leu, Ile, Phe) were significantly different between marine and terrestrial animals. Both δ¹³C and δ¹⁵N values of EAAs can separate the terrestrial animals from other food sources. However, the PCA results showed that the EAA δ¹³C and δ¹⁵N values cannot discriminate between terrestrial herbivores and game birds due to their isotopic similarity in EAAs. To examine the EAA sources in each human, we plotted the Imdang and Mumun humans with each food group. We found that the Mumun humans clustered closely to the C₄ plants (Fig 5A), while the Imdang humans clustered closely to the game birds and terrestrial herbivores (Fig 5B). Thus, our PCA results show that the Imdang humans derived a large proportion of their EAAs from game birds and terrestrial herbivores, while the Mumun human derived a large proportion of their EAAs from C₄ plants.

Download:

Fig 5.

Score plots of principal component analyses (PCA) of δ¹³C and δ¹⁵N values of EAAs from humans and animals at the Mumun (A) and Imdang sites (B). The Mumun humans are clustered closely with C₄ plants and the Imdang humans are clustered closely with most of the terrestrial herbivores. Ellipses represent 95% standard deviation with arrows for EAAs being significant (p<0.05) correlation vectors. Game birds are not shown in these PCA figures due to overlap with terrestrial herbivores.

https://doi.org/10.1371/journal.pone.0300068.g005

MixSIAR model

A MixSIAR model was conducted with five δ¹³C_AA values (Leu, Val, Ile, Thr, Phe) and one source δ¹⁵N_AA value (Phe). These six AAs were selected since they are only found in both the animals and cereals together, and they have little isotopic fractionation between diet and consumers [24, 25, 30]. Thus, as these six AAs are derived mainly from dietary sources, they can provide a more complete picture of the total diets of the Mumun and Imdang individuals. In S5 Table in S1 File, the trophic discrimination factors and the isotopic data of food sources for the MixSIAR model are listed. The EAA isotopic model output indicates that the Mumun humans had a high contribution from the C₄ plants (24–39%) and game birds (17–39%), and a small contribution from C₃ plants (3%) (S6 Table in S1 File). However, the Imdang humans had a high contribution from animal sources such as the game birds (17–63%), marine fish (8–46%), and a small contribution from C₃ (3–16%) and C₄ plants (3–8%). This indicates that the main EAA sources for the Mumun humans were C₄ plants and game birds (Fig 6A), whilst the main EAA sources for the Imdang humans were game birds, terrestrial animals, and marine fish (Fig 6B).

Download:

Fig 6.

Model comparison between the dietary contribution of the five food groups to the Mumun (A) and Imdang humans (B) calculated by MixSIAR model. The dietary estimates on the EAA isotope model show that there is a tendency to reflect more contributions of animal sources (GB, MA), and less contributions of plants sources (C3P, C4P) than the bulk isotope model.

https://doi.org/10.1371/journal.pone.0300068.g006

We examined if there is any difference in dietary estimates in the mixing models using the bulk and EAA isotope data. We generated and compared the MixSIAR model for both bulk and EAA isotopic values, respectively (S5 Table in S1 File). To compare the difference in dietary estimates between the two models, we used paired t-tests to evaluate possible differences in each food group. The t-test results showed that plant sources (C₃ and C₄ plants) are negatively correlated (P < 0.001) between the bulk and EAA isotope model, but animal sources (game birds and marine fish) are positively correlated (P < 0.001). This result indicates that the proportional contributions of plant sources (C₃ and C₄ plants) to both the Mumun and Imdang humans can be elevated in the bulk isotope model, and there is a tendency to overemphasize plant sources in the bulk isotope model (Fig 6). In contrast to plant sources, the contribution of animal sources (game birds and marine fish) is increased when estimated by the EAA isotope model. This result indicates that the EAA isotope model could be more reflective of animal resources in dietary reconstruction than plant resources.

Discussion

Trophic position estimates

Our CSIA-AA study confirms that TP estimates can provide more accurate predictions of ancient human diets within the isotopic baselines of the surrounding environment. The TP values of the Mumun humans show that their low bulk δ¹⁵N values reflect the consumption of δ¹⁵N-low resources from the prehistoric ecosystem. On the other hand, the TP values of the Imdang humans show that the high bulk δ¹⁵N results of 11 of the Imdang individuals (TP 3.1 to 3.3) resulted from the consumption of ¹⁵N-enriched sources like marine fish. However, most of the Imdang individuals below TP 3.0 are associated with the consumption of terrestrial resources. In the trophic position estimates, TP values in humans were generated for theoretically 100% terrestrial and/or marine protein [16]. Practically, as omnivores, humans would be expected to consume multiple food sources consisting of marine fish, terrestrial animals, C₃ and C₄ plants. If humans had a mixed diet, we need to consider each β and TDF value from different food sources. Furthermore, there is a variation in β and TDF values in an ecological niche due to the different primary producers in ecosystems [4]. For example, the β value of C₃ plants is +8.4‰ and the β value of C₄ plants is -0.4‰ and the β value of marine plants -3.4‰ [32]. In addition, the commonly used TDF_Glx-Phe of 7.6‰ is an average value of several observations in ecological studies and can vary between taxa [31]. Thus, it is required to establish more accurate β and TDF values of different food sources (e.g. β and TDF values of marine and C₄ plants). Furthermore, it is considered that the TP value of plants is around 1 [14, 15, 17] and the plant TP value is used as the basis for the TP estimation of humans and animals in the food webs. However, our results show that the TP estimates in grains (rice, soybean, adzuki bean, wheat, oat) are higher than 1, ranging from 1.0 to 2.1. Previously published data also show that the TP estimates in Danish wheat range from 1.4 to 1.7 [67] and German beans and pea range from 2.1 to 2.2 [68]. This discrepancy in plant TP values could be associated with the different lignin content between grains and leaves [69]. In dietary reconstruction, grains are the main crops consumed and could contribute more to human diets than any other plant tissues (i.e. leaves, stems, rachis). Unfortunately, most of the previous TP values of plants are mainly derived from the leaves. Thus, for more accurate TP estimates on human diets, it is necessary to have more detailed studies to calculate the TP values of grains among plant species [69, 70].

Amino acid proxies and PCA patterns

For δ¹³C proxies, previous studies suggested that two AA proxies (Δ¹³C_Gly-Phe and Δ¹³C_Val-Phe) are effective at discriminating between terrestrial and aquatic sources [7, 11, 66]. However, our study shows that the Δ¹³C_Gly-Phe and Δ¹³C_Val-Phe results are incompatible with the estimated trophic position (TP) using Δ¹⁵N AA_Glx-Phe values. This inconsistency indicates that the two δ¹³C_AA proxies are less powerful to fully separate between terrestrial and marine sources when humans have mixed diets. This means that one or two amino acid markers (e.g. Gly, Val, Phe) are less likely to distinguish dietary sources in mixed human diets. Instead, our PCA data with EAAs provide an improved clustering and separation of C₃, C₄, marine and terrestrial protein consumers (Fig 5), and better information on the main source of dietary proteins. The δ¹³C and δ¹⁵N values of EAAs in PCA, instead of source δ¹⁵N_AA values, can well separate all four food sources according to their metabolic pathways. Our PCA patterns show that the Mumun humans are close to the C₄ plants, suggesting that C₄ plants played a significant role in Mumun diets. In contrast, the Imdang humans cluster with the game birds and terrestrial herbivores, indicating that the hunting of wild animals was important in their subsistence activities. Additionally, we found that the EAA patterns in food sources are not able to discriminate the confounding isotope values between terrestrial herbivores and game birds at the Imdang site. This is associated with their isotopic similarity between these two animal groups and their similarity in dietary sources (C₃-based diets) in the terrestrial food webs.

Comparison of mixing models

The output of the two MixSIAR models showed that the model predictions based on both bulk and EAA isotope values yielded different palaeodietary patterns. In the Mumun, the bulk isotope model showed high contributions of C₄ plants and a low contribution of other sources, emphasizing the consumption of C₄ plants (~62%). In contrast, the EAA isotope model finds a contribution of other animal proteins in C₄ plant-dominated Mumun diets (Fig 6). For the Imdang, the bulk isotope model showed the high contribution of game birds (32%) and low contribution of C₄ plants (8%). In particular, the EAA isotope model showed a much higher contribution of animal proteins than the bulk isotope model (Fig 6), emphasizing more the consumption of animal sources such as game birds (41%) and marine animals (21%). Thus, our study confirms that the EAA isotope model can reflect more the consumption of animal proteins in mixed human diets.

This discrepancy between the bulk and EAA isotope models is likely due to variability in the extent of routing of EAAs in foodstuffs. Generally, plants have lower amounts of protein and higher contents of carbohydrates than animal resources. Although there was the same contribution from both plant and animal proteins in human diets, EAAs in human bone collagen could be routed firstly from the EAA in animal proteins rather than plant proteins [71]. This is due to the fact that animal proteins require less metabolic expense in EAA routing than plants proteins. Therefore, the EAA isotope model is more related to the consumption of animal proteins in human diets. On the other hand, the bulk isotope model emphasizes the relative importance of one or two sources, highlighting the dependence on plant proteins. A recent study using the EAA isotope model also suggested that the contribution of animal proteins (marine fish) to total dietary proteins in ancient Romans can be underrepresented in bulk isotope data [14]. Our EAA isotope model in this study shows that the contribution of plant proteins to total dietary proteins can be overrepresented in bulk isotope data. Thus, when humans have mixed diets, the EAA isotope model can be much more suited to distinguishing the consumption of animal proteins in human diets than the bulk isotope model.

Diet of the prehistoric Korean populations

The CSIA-AA approach finds a difference in relative contributions of each dietary source for the two prehistoric Korean populations. In terms of plant sources, the Mumun had high contributions of EAAs from C₄ plants (Fig 7A), but Imdang people had high contributions of EAAs from C₃ plants to their total diets. This indicates that these two prehistoric Korean populations might have cultivated different crops: C₄ plants for the Mumun and C₃ plants for the Imdang, even though the subsistence economy in both populations was based on agriculture. Previous work has suggested that early agriculture on the Korean peninsula was established during the Mumun period (1500–100 BC), but there is no clear evidence which grains (rice vs. millet) were first cultivated as the main crops during that period [38]. However, our study shows that C₄ plants (millets) were the main staples during the Mumun period, supporting the possibility that millets could be the first crops in prehistoric Korea. At the Imdang sites dated to Proto-Three Kingdom period, C₃ plants (rice, barley, soybean) were the dominant plant resources for the human diets (Fig 7A). In Korean archaeology, it has been suggested that rice agriculture was introduced earlier and that it intensified later during the Proto-Three Kingdom period (BC 108–313 AD) [44, 47, 52]. Our EAA isotope data suggest that agriculture was based on C₃ plants (likely rice) during the Proto-Three Kingdom period. Therefore, despite the problem of sample size in this study, it is possible to speculate that there was a change in the main crops cultivated from the Bronze Age to the Proto-Three Kingdom period in Korea.

Download:

Fig 7.

Raincloud plots of dietary estimates of plant (A) and animal (B) proteins generated by the EAA isotope model. There is a clear difference in dietary contributions of plant (C₃ vs. C₄) and animal (game birds vs. terrestrial herbivores) sources between the two prehistoric Korean populations.

https://doi.org/10.1371/journal.pone.0300068.g007

In terms of animals consumed, the Imdang people had high contributions from animal sources to their total diets, while the Mumun people had relatively low contributions from animal sources (Fig 7B). This difference indicates that animal proteins were not mainly consumed until the Mumun period, but in the later Proto-Three Kingdom period, they became significant food items in human diets. It has previously been suggested that most of the domestic animals (i.e. cattle, pig, horse, chicken) became important food sources in human diets during the Proto-Three Kingdom period (BC 108–313 AD) [39]. However, our study reveals that most of the animal resources in the Imdang diets were coming from the hunting of wild birds and herbivores, as well as from fishing (Fig 7B). In relative terms, wild terrestrial birds (e.g. pheasant, wild goose) contributed the most animal protein sources to human diets. Therefore, we can postulate that hunting wild animals and fishing still played a significant role in subsistence activities during the Proto-Three Kingdom period, even though the subsistence economy was based on plant cultivation and animal husbandry.

Conclusions

We applied CSIA-AA to bone collagen to enhance palaeodietary reconstructions in South Korean prehistoric populations. We measured δ¹³C_AA and δ¹⁵N_AA values in human and animal bone collagen from the archaeological sites, and modern cereals from South Korea as reference materials. CSIA-AA provided new information about Prehistoric Korean subsistence strategies, the spread of early agriculture based on the domesticated grains and the subsistence economy of ancient states from the two different-era sites. Results of estimated trophic positions (TP) show that the two prehistoric populations (Mumun and Imdang) obtained most of their dietary protein from terrestrial sources. However, the EAA proxies and PCA data revealed that there were different dietary patterns between the Mumun to the Proto-Three Kingdom periods. The humans from the Mumun sites consumed primarily C₄ plant resources, while the Imdang humans consumed primarily terrestrial animals. Quantitative estimates through the MixSIAR also found a variation in relative contributions of each food source to human diets in these two prehistoric populations. The EAA isotope model revealed that the Mumun had a high dietary contribution from C₄ plants. However, the Imdang had a high dietary contribution of the C₃ plants and a wide range of contributions from animal sources such as game birds, terrestrial mammals and marine animals. In this study, our CSIA-AA approach shows that the two investigated prehistoric Korean cultures differed in terms of the main crops consumed and that hunting and fishing in prehistoric Korea were still important subsistence activities along with agriculture in the Proto-Three Kingdom period. Finally, this study demonstrates that the use of CSIA-AA and MixSIAR to reconstruct palaeodiets allows us to explore more accurately the contribution of each dietary source to whole human diets and enhance our understanding of human adaptability to environments in prehistoric Korea.

Supporting information

S1 File.

https://doi.org/10.1371/journal.pone.0300068.s001

(PDF)

S1 Data.

https://doi.org/10.1371/journal.pone.0300068.s002

(ZIP)

Acknowledgments

We are grateful to In-Seong Cheong, the director at the Yeungnam University Museum and Soyoung Lee, curator at the Chungbuk National University Museum for access to skeletal materials. We are also grateful to Prof. Kyung-Hoon Shin and Dr. Hyuntae Choi for analytical help at the Department of Marine Sciences, Hanyang University. We would like to thank Zuzana Hansen, and Anastasia Brozou for help with sample preparation and collagen extraction at the Moesgaard Archaeo-Science Laboratory (Aarhus University).

References

1. Richards MP. Isotope analysis for diet studies. In: Richards MP, Britton K, editors. Archaeological Science: An Introduction. 2020. pp. 125–143.
2. Szpak P, Metcalfe J, Macdonald R. Best practices for calibrating and reporting stable isotope measurements in archaeology. J Archaeol Sci Rep. 2017;13: 609–616.
- View Article
- Google Scholar
3. Lee-Thorp JA. On isotopes and old bones. Archaeometry. 2008; 50: 925–950.
- View Article
- Google Scholar
4. Larsen T, Frenandes R, Wang YV, Roberts P. Reconstructing Hominin diets with stable isotope analysis of amino acids: New perspectives and future directions. BioSci. 2022; 72: 618–637. pmid:35769500
- View Article
- PubMed/NCBI
- Google Scholar
5. Ma Y, Grimes V, Biesen GV, Shi L, Chen K, Mannino MA, et al. Aminoisocapes and palaeodiet reconstruction: New perspectives on millet based diets in China using acid δ¹³C values. J Archaeol Sci. 2021;125: 105289.
- View Article
- Google Scholar
6. Choy K, Smith CI, Fuller BT, Richards MP. Investigation of amino acid δ¹³C signatures in bone collagen to reconstruct human palaeodiets using liquid chromatography-isotope ratio mass spectrometry. Geochim Cosmochim Acta. 2010; 74: 6093–6111.
- View Article
- Google Scholar
7. Corr LT, Sealy JC, Horton MC, Evershed RP. A novel marine dietary indicator utilizing compound-specific bone collagen amino acid δ¹³C values of ancient humans. J Archaeol Sci. 2005; 32: 321–330.
- View Article
- Google Scholar
8. Szpak P, Longstaffe FJ, Millarire J-F, White CD. Large variation in nitrogen isotopic composition of a fertilized legume. J Archaeol Sci. 2014; 45: 72–79.
- View Article
- Google Scholar
9. Bogaard A, Heaton THE, Poulton P, Merbach I. The impact of manuring on nitrogen isotope ratios in cereals: archaeological implications for reconstruction of diet and crop management practices. J Archaeol Sci. 2007; 34: 335–343.
- View Article
- Google Scholar
10. Cheung C, Szpak P. Interpreting past human diets using stable isotope mixing models, J Archaeol Method Theory. 2021; 28: 1106–1142.
- View Article
- Google Scholar
11. Webb EC, Honch NV, Dunn PJH, Eriksson G, Liden K, Evershed RP Compound-specific amino acid isotopic proxies for detecting freshwater resource consumption. J Archaeol Sci. 2015; 63: 104–114.
- View Article
- Google Scholar
12. Webb EC, Honch NV, Dunn PJH, Linderholm A, Eriksson G, Liden K, et al. Compound-specific amino acid isotopic proxies for distinguishing between terrestrial and aquatic resource consumption. Archaeol Anthropol Sci. 2018; 10:1–18.
- View Article
- Google Scholar
13. Brozou A, Fuller BT, Grimes V, Van Biesen G, Ma Y, Boldsen JL, et al. Aquatic resource consumption at the Odense leprosarium: Advancing the limits of palaeodiet reconstruction with amino acid d13C measurements. J Archaeol Sci. 2022; 141: 105578.
- View Article
- Google Scholar
14. Soncin S. et al. High-resolution dietary reconstruction of victims of the 79CE Vesuvius eruption at Herculaneum by compound-specific isotope analysis. Sci Adv. 2021; 35.
- View Article
- Google Scholar
15. Jaouen K, Richards MP, Cabec AL, Welker F, Rendu W, Hublin J-J, et al. Exceptionally high d15N values in collagen single amino acids confirm Neandertals as high-trophic level carnivores. Proc Natl Acad Sci U S A. 2019; 116: 4928–4933.
- View Article
- Google Scholar
16. Jarman CL, Larsen T, Hunt T, Lipo C, Solsvik R, Wallsgrave N, et al. Diet of the prehistoric population of Rapa Nui (Easter lsland, Chile) shows environmental adaptation and resilience. Am J Phys Anthrol. 2017; 164: 343–361.
- View Article
- Google Scholar
17. Chikaraishi Y, Steffan SA, Ogawa NO, Ishikawa NF, Sasaki Y, Tsuchiya M, et al. High-resolution food webs based on nitrogen isotopic composition of amino acids. Ecol Evol. 2014; 4: 2423–2449. pmid:25360278
- View Article
- PubMed/NCBI
- Google Scholar
18. Naito Y, Chikaraishi Y, Ohkouchi N, Yoneda M. Evaluation of carnivory in inland Jomon hunter-gatherers based on nitrogen isotopic compositions of individual amino acids in bone collagen. J Archaeol Sci. 2013; 40: 2913–2923.
- View Article
- Google Scholar
19. McMahon KW, Newsome SD. Amino acid isotope analysis: a new frontier in studies of animal migration and foraging ecology. In: Hobson KA, Wasenaar LI. editors. Tracking Animal Migration with Stable Isotopes, Academic Press; 2019. pp. 173–190.
20. Leeds PJ. Dispensable and indispensable amino acids for humans, J Nutr. 2000; 130:1835S–1839S. pmid:10867060
- View Article
- PubMed/NCBI
- Google Scholar
21. Choy K, Nash SH, Kristal AR, Hopkins S, Boyer BB, O’Brien DM. The carbon isotope ratio of alanine in red blood cells is a new candidate biomarker of sugar-sweetened beverage intake, J Nutr. 2013; 143: 878–884. pmid:23616504
- View Article
- PubMed/NCBI
- Google Scholar
22. Larsen T, Ventura M, Anderson N, O’Brien DM, Piatkowski U, McCarthy MD. Tracing carbon sources through aquatic and terrestrial food webs using amino acid stable isotope fingerprinting. PLoS One. 2013; 9 e73441. pmid:24069196
- View Article
- PubMed/NCBI
- Google Scholar
23. Larsen T, Hansen T, Dierking J. Characterizing niche differentiation among marine consumers with amino acid δ¹³C fingerprinting. Ecol. Evol. 2020; 10: 7768–7782.
- View Article
- Google Scholar
24. Lübcker N, Whiteman JP, Shipley ON, Hobson KA, Newsome SD. Use of amino acid isotope analysis to investigate capital versus income breeding strategies in migratory avian species. Methods Ecol. Evol. 2023;00: 1–14.
- View Article
- Google Scholar
25. McMahon KW, McCarthy MD. Embracing variability in amino acid d15N fractionation: mechanisms, implications, and applications for trophic ecology. Ecosphere. 2016;7: e01511.
- View Article
- Google Scholar
26. O’Connell TC. ’Trophic’ and ’source’ amino acids in trophic estimation: a likely metabolic explanation. Oecologia. 2017;184: 317–326. pmid:28584941
- View Article
- PubMed/NCBI
- Google Scholar
27. Yun HY, Larsen T, Choi B, Won E, Shin K. Amino acid nitrogen and carbon isotope data: potential and implications for ecological studies. Ecol. Evol. 2022;12: e8929. pmid:35784034
- View Article
- PubMed/NCBI
- Google Scholar
28. Fuller BT, Petzke KJ. The dietary protein paradox and threonine ¹⁵N-depletion: pyridoxal-5’- phosphate enzyme activity as a mechanism for the δ¹⁵N trophic level effect. Rapid Commun. Mass Spectrom. 2017; 31:705–718.
- View Article
- Google Scholar
29. Hare PE, Fogel ML, Stafford TW Jr, Mitchell AD, Hoering TC. The isotopic composition of carbon and nitrogen in individual amino acids isolated from modern and fossil proteins. J Archaeol Sci. 1991; 18: 277–292.
- View Article
- Google Scholar
30. McMahon KW, Polito MJ, Abel S, McCarthy MD, Thorrold SR. Carbon and nitrogen isotope fractionation of amino acids in an avian marine predator, the Gentoo penguin (Pygoscelis papua). Ecol Evol. 2015;5: 1278–1290.
- View Article
- Google Scholar
31. Chikaraishi Y, Ogawa NO, Kashiyama Y, Takano Y, Suga H, Tomitani A, et al. Determination of aquatic food–web structure based on compound-specific nitrogen isotopic composition of amino acids. Limnol Oceanogr Methods. 2009; 7: 740–750.
- View Article
- Google Scholar
32. Chikaraishi Y, Ogawa NO, Ohkouchi N. Further evaluation of the trophic level estimation based on nitrogen isotopic composition of amino acids. In: Ohkouch N, Tayau I, Koba K. editors. Earth, Life and Isotopes. 2010. pp. 37–51.
- View Article
- Google Scholar
33. McClelland JW, Montoya JP. Trophic relationships and the nitrogen isotopic composition of amino acids in plankton. Ecol. 2002; 83: 2171–2180.
- View Article
- Google Scholar
34. Naito Y, Honch NV, Chikaraishi Y, Ohkouchi N, Yoneda M. Quantitative evaluation of marine protein contribution in ancient diets based on nitrogen isotope ratios of individual amino acids in bone collagen: an investigation at the Kitakogane Jomon site. Am J Phys Anthropol. 2010;143: 31–40. pmid:20333711
- View Article
- PubMed/NCBI
- Google Scholar
35. Fontanals-Coll M, Soncin S, Talbot HM, von Tersch M, Gibaja JF, Colonese AC, et al. Stable isotope analyses of amino acids reveal the importance of aquatic resources to Mediterranean coastal hunter-gatherers. Proc Biol Sci. 2023; 22:20221330. pmid:36809804
- View Article
- PubMed/NCBI
- Google Scholar
36. Choy K, Yun HY, Lee J, Fuller BT, Shin K-H. Direct isotopic evidence for human millet consumption in the Middle Mumun period: Implication and importance of millets in early agriculture on the Korean Peninsula. J Archaeol Sci. 2021; 129: 105372.
- View Article
- Google Scholar
37. Choy K, Yun HY, Kim SH, Jung S, Fuller BT, Kim DW. Isotopic investigation of skeletal remains at the Imdang tombs reveals high consumption of game birds and social stratification in ancient Korea. Sci Rep. 2021; pmid:34799611
- View Article
- PubMed/NCBI
- Google Scholar
38. Kim J, Park J. Millet vs rice: an evaluation of the farming/language dispersal hypothesis in the Korean context. Evol Hum Sci. 2020; 2: 1–18.
- View Article
- Google Scholar
39. Kim HS. The diet and breeding system of Dongnae people by isotope analysis at Korea. Kogo Kwangchang. 2015; 20: 89–111(Korean).
- View Article
- Google Scholar
40. Fernandes R, Millard A, Nadeau M, Groote P. Food reconstruction using isotopic transferred signals (FRUITS): a Bayesian model for diet reconstruction. PloS One. 2014; 9: e87436. pmid:24551057
- View Article
- PubMed/NCBI
- Google Scholar
41. Stock BC, Jackson AL, Ward EJ, Parnell AD, Phillips DL, Semmens BX. Analyzing mixing systems using a new generation of Bayesian tracer mixing models. Peer J. 2021;
- View Article
- Google Scholar
42. Stock BC, Semmens B. MixSIAR GUI User Manual. Version 3.1. 2016; https://github.com/brianstock/ MixSIAR.
43. Lee RJ, Bale MT. Social change and household geography in Mumun period South Korea. J Anthropol Res. 2016;72: 178–198.
- View Article
- Google Scholar
44. Ahn S-M. The emergence of rice agriculture in Korea: archeobotanical perspectives. Archaeol Anthropol Sci. 2010;2: 89–98.
- View Article
- Google Scholar
45. Lee G-A. Changes in subsistence patterns from the Chulmun to Mumun Periods: Archeobotanical Investigation. Ph.D. Dissertation, Department of Anthropology, University of Toronto. 2003.
46. Kim M. Rice in ancient Korea: status symbol or community food? Antiq. 2015;89: 838–853.
- View Article
- Google Scholar
47. Crawford GW, Lee G-A. Agricultural origins in the Korean Peninsula. Antiq. 2003; 77:87–95.
- View Article
- Google Scholar
48. Yeungnam University Museum (YUM). Excavation reports on Imdang burial mounds-Imdang 2ho. 2002.
49. Kim D-W. A Study on the Funeral System of Ancient Tombs at the Im-dang Site Unpublished Ph.D. dissertation. Yeungnam University, Gyeongsan, South Korea. 2014.
50. Kim D-W. The possibility of non-burial of bodies on tombs of the province under Silla through burial of Gyeongsan Joyeung E-II-2. Sogang J Early Korean Hist. 2018;28: 395–422.
- View Article
- Google Scholar
51. Davey J. Unreliable narratives: Historical and archaeological approaches to early Silla. J Korean Stud. 2016; 29: 7–32.
- View Article
- Google Scholar
52. Nelson SM, Neu J. Mumun, proto-three kingdoms, and three kingdoms in Korea. In: Habu J, Lape PV, Olsen JW. editors. Handbook of East and Southeast Asian Archaeology. 2017. pp 603–617.
53. Yeungnam University Museum (YUM). Archaeological reports on the animal remains from Imdang site in Kyeongsan I- mammals and birds. 2017.
54. Yeungnam University Museum (YUM). Archaeological reports on the animal remains from Imdang site in Kyeongsan II fish and shellfish. 2018.
55. Ko EB. Birds for the Dead: animal offerings from high-mound tombs at Imdang. J Korean Archaeol Soc. 2018;106: 4–41.
- View Article
- Google Scholar
56. Ko EB. Animal sacrifice rituals in ancient Korea- with focus on Silla tombs of the Three Kingdoms period. Anat Biol Anthropol. 2020; 33: 69–77.
- View Article
- Google Scholar
57. Ahn SM. The periodic change of plant use based on plant remains from archaeological sites, Archaeology of Agriculture. 2013.
58. Richards MP, Hedges REM. Stable isotope evidence for similarities in the types of marine foods used by Late Mesolithic humans at sites along the Atlantic coast of Europe. J Archaeol Sci. 1999;26: 717–722.
- View Article
- Google Scholar
59. Brown TA, Nelson D.E, Vogel J.S. & Southon JR. Improved collagen extraction by modified Longin method. Radiocarbon. 1988;30: 171–177.
- View Article
- Google Scholar
60. Corr LT, Berstan R, Evershed RP. Optimization of derivatisation procedures for the determination of d13C values of amino acids by gas chromatography/combustion/isotope ratio mass spectrometry. Rapid Commun Mass Spectrom. 2007;21: 3759–3771.
- View Article
- Google Scholar
61. O’Connell TC, Collins MJ. Comment on “Ecological niche of Neanderthals from Spy Cave revealed by nitrogen isotopes of individual amino acids in collagen”. J Hum Evol. 2017;117: 53–55.
- View Article
- Google Scholar
62. Fogel ML, Tuross N. Extending the limits of palaeodietary studies of humans with compound specific carbon isotope analysis of amino acids. J Archaeol Sci. 2003;30: 535–545.
- View Article
- Google Scholar
63. Silfer JA, Engel MH, Macko SA, Jumeau EL. Stable Carbon Isotope Analysis of Amino Acid Enantiomers by Conventional Isotope Ratio Mass Spectrometry and Combined Gas Chromatography/ Isotope Ratio Mass Spectrometry, Anal Chem. 1991;63: 370–374.
- View Article
- Google Scholar
64. R Core Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing.2019 https://www.R-project.org/.
65. Roberts P, Fernandes R, Craig OE, Larsen T, Lucquin A, Swift J, et al. Calling all archaeologists: guidelines for terminology, methodology, data handling, and reporting when undertaking and reviewing stable isotope applications in archaeology. Rapid Commun Mass Spectrom. 2018;32: 361–372. pmid:29235694
- View Article
- PubMed/NCBI
- Google Scholar
66. Honch NV, McCullagh JSO, Hedges REM. Variation of bone collagen amino acid δ¹³C values in archaeological humans and fauna with different dietary regimes: developing frameworks of dietary discrimination. Am J Phys Anthropol. 2012;148: 495–511.
- View Article
- Google Scholar
67. Paolini M, Ziller L, Laursen KH, Husted S, Camin F. Compound-specific δ¹⁵N and δ¹³C analyses of amino acids for potential discrimination between organically and conventionally grown wheat. J Agric Food Chem. 2015;63: 5841–5850.
- View Article
- Google Scholar
68. Styring AK, Fraser RA, Bogaard A, Evershed RP. Cereal grain, rachis and pulse seed amino acid δ¹⁵N values as indicators of plant nitrogen metabolism. Phytochem. 2014;97: 20–29.
- View Article
- Google Scholar
69. Kendall IP, Woodward P, Clark JP, Styring A.K, Hanna J.V. & Evershed R.P. Compound-specific δ¹⁵N values express differences in amino acid metabolism in plants of varying lignin content. Phytochem. 2019;161: 130–138.
- View Article
- Google Scholar
70. Styring AK. et al. Refining human palaeodietary reconstruction using amino acid δ15N values of plants, animals and humans. J Archaeol Sci. 2019;53: 504–515.
- View Article
- Google Scholar
71. Jim S, Jones V, Ambrose SH, Evershed RP. Quantifying dietary macronutrient sources of carbon for bone collagen biosynthesis using natural abundance stable carbon isotope analysis Br J Nutr. 2006;95: 1055–1062. pmid:16768826
- View Article
- PubMed/NCBI
- Google Scholar

[ref1] 1. Richards MP. Isotope analysis for diet studies. In: Richards MP, Britton K, editors. Archaeological Science: An Introduction. 2020. pp. 125–143.

[ref2] 2. Szpak P, Metcalfe J, Macdonald R. Best practices for calibrating and reporting stable isotope measurements in archaeology. J Archaeol Sci Rep. 2017;13: 609–616.
View Article
Google Scholar

[3] View Article

[4] Google Scholar

[ref3] 3. Lee-Thorp JA. On isotopes and old bones. Archaeometry. 2008; 50: 925–950.
View Article
Google Scholar

[6] View Article

[7] Google Scholar

[ref4] 4. Larsen T, Frenandes R, Wang YV, Roberts P. Reconstructing Hominin diets with stable isotope analysis of amino acids: New perspectives and future directions. BioSci. 2022; 72: 618–637. pmid:35769500
View Article
PubMed/NCBI
Google Scholar

[9] View Article

[10] PubMed/NCBI

[11] Google Scholar

[ref5] 5. Ma Y, Grimes V, Biesen GV, Shi L, Chen K, Mannino MA, et al. Aminoisocapes and palaeodiet reconstruction: New perspectives on millet based diets in China using acid δ¹³C values. J Archaeol Sci. 2021;125: 105289.
View Article
Google Scholar

[13] View Article

[14] Google Scholar

[ref6] 6. Choy K, Smith CI, Fuller BT, Richards MP. Investigation of amino acid δ¹³C signatures in bone collagen to reconstruct human palaeodiets using liquid chromatography-isotope ratio mass spectrometry. Geochim Cosmochim Acta. 2010; 74: 6093–6111.
View Article
Google Scholar

[16] View Article

[17] Google Scholar

[ref7] 7. Corr LT, Sealy JC, Horton MC, Evershed RP. A novel marine dietary indicator utilizing compound-specific bone collagen amino acid δ¹³C values of ancient humans. J Archaeol Sci. 2005; 32: 321–330.
View Article
Google Scholar

[19] View Article

[20] Google Scholar

[ref8] 8. Szpak P, Longstaffe FJ, Millarire J-F, White CD. Large variation in nitrogen isotopic composition of a fertilized legume. J Archaeol Sci. 2014; 45: 72–79.
View Article
Google Scholar

[22] View Article

[23] Google Scholar

[ref9] 9. Bogaard A, Heaton THE, Poulton P, Merbach I. The impact of manuring on nitrogen isotope ratios in cereals: archaeological implications for reconstruction of diet and crop management practices. J Archaeol Sci. 2007; 34: 335–343.
View Article
Google Scholar

[25] View Article

[26] Google Scholar

[ref10] 10. Cheung C, Szpak P. Interpreting past human diets using stable isotope mixing models, J Archaeol Method Theory. 2021; 28: 1106–1142.
View Article
Google Scholar

[28] View Article

[29] Google Scholar

[ref11] 11. Webb EC, Honch NV, Dunn PJH, Eriksson G, Liden K, Evershed RP Compound-specific amino acid isotopic proxies for detecting freshwater resource consumption. J Archaeol Sci. 2015; 63: 104–114.
View Article
Google Scholar

[31] View Article

[32] Google Scholar

[ref12] 12. Webb EC, Honch NV, Dunn PJH, Linderholm A, Eriksson G, Liden K, et al. Compound-specific amino acid isotopic proxies for distinguishing between terrestrial and aquatic resource consumption. Archaeol Anthropol Sci. 2018; 10:1–18.
View Article
Google Scholar

[34] View Article

[35] Google Scholar

[ref13] 13. Brozou A, Fuller BT, Grimes V, Van Biesen G, Ma Y, Boldsen JL, et al. Aquatic resource consumption at the Odense leprosarium: Advancing the limits of palaeodiet reconstruction with amino acid d13C measurements. J Archaeol Sci. 2022; 141: 105578.
View Article
Google Scholar

[37] View Article

[38] Google Scholar

[ref14] 14. Soncin S. et al. High-resolution dietary reconstruction of victims of the 79CE Vesuvius eruption at Herculaneum by compound-specific isotope analysis. Sci Adv. 2021; 35.
View Article
Google Scholar

[40] View Article

[41] Google Scholar

[ref15] 15. Jaouen K, Richards MP, Cabec AL, Welker F, Rendu W, Hublin J-J, et al. Exceptionally high d15N values in collagen single amino acids confirm Neandertals as high-trophic level carnivores. Proc Natl Acad Sci U S A. 2019; 116: 4928–4933.
View Article
Google Scholar

[43] View Article

[44] Google Scholar

[ref16] 16. Jarman CL, Larsen T, Hunt T, Lipo C, Solsvik R, Wallsgrave N, et al. Diet of the prehistoric population of Rapa Nui (Easter lsland, Chile) shows environmental adaptation and resilience. Am J Phys Anthrol. 2017; 164: 343–361.
View Article
Google Scholar

[46] View Article

[47] Google Scholar

[ref17] 17. Chikaraishi Y, Steffan SA, Ogawa NO, Ishikawa NF, Sasaki Y, Tsuchiya M, et al. High-resolution food webs based on nitrogen isotopic composition of amino acids. Ecol Evol. 2014; 4: 2423–2449. pmid:25360278
View Article
PubMed/NCBI
Google Scholar

[49] View Article

[50] PubMed/NCBI

[51] Google Scholar

[ref18] 18. Naito Y, Chikaraishi Y, Ohkouchi N, Yoneda M. Evaluation of carnivory in inland Jomon hunter-gatherers based on nitrogen isotopic compositions of individual amino acids in bone collagen. J Archaeol Sci. 2013; 40: 2913–2923.
View Article
Google Scholar

[53] View Article

[54] Google Scholar

[ref19] 19. McMahon KW, Newsome SD. Amino acid isotope analysis: a new frontier in studies of animal migration and foraging ecology. In: Hobson KA, Wasenaar LI. editors. Tracking Animal Migration with Stable Isotopes, Academic Press; 2019. pp. 173–190.

[ref20] 20. Leeds PJ. Dispensable and indispensable amino acids for humans, J Nutr. 2000; 130:1835S–1839S. pmid:10867060
View Article
PubMed/NCBI
Google Scholar

[57] View Article

[58] PubMed/NCBI

[59] Google Scholar

[ref21] 21. Choy K, Nash SH, Kristal AR, Hopkins S, Boyer BB, O’Brien DM. The carbon isotope ratio of alanine in red blood cells is a new candidate biomarker of sugar-sweetened beverage intake, J Nutr. 2013; 143: 878–884. pmid:23616504
View Article
PubMed/NCBI
Google Scholar

[61] View Article

[62] PubMed/NCBI

[63] Google Scholar

[ref22] 22. Larsen T, Ventura M, Anderson N, O’Brien DM, Piatkowski U, McCarthy MD. Tracing carbon sources through aquatic and terrestrial food webs using amino acid stable isotope fingerprinting. PLoS One. 2013; 9 e73441. pmid:24069196
View Article
PubMed/NCBI
Google Scholar

[65] View Article

[66] PubMed/NCBI

[67] Google Scholar

[ref23] 23. Larsen T, Hansen T, Dierking J. Characterizing niche differentiation among marine consumers with amino acid δ¹³C fingerprinting. Ecol. Evol. 2020; 10: 7768–7782.
View Article
Google Scholar

[69] View Article

[70] Google Scholar

[ref24] 24. Lübcker N, Whiteman JP, Shipley ON, Hobson KA, Newsome SD. Use of amino acid isotope analysis to investigate capital versus income breeding strategies in migratory avian species. Methods Ecol. Evol. 2023;00: 1–14.
View Article
Google Scholar

[72] View Article

[73] Google Scholar

[ref25] 25. McMahon KW, McCarthy MD. Embracing variability in amino acid d15N fractionation: mechanisms, implications, and applications for trophic ecology. Ecosphere. 2016;7: e01511.
View Article
Google Scholar

[75] View Article

[76] Google Scholar

[ref26] 26. O’Connell TC. ’Trophic’ and ’source’ amino acids in trophic estimation: a likely metabolic explanation. Oecologia. 2017;184: 317–326. pmid:28584941
View Article
PubMed/NCBI
Google Scholar

[78] View Article

[79] PubMed/NCBI

[80] Google Scholar

[ref27] 27. Yun HY, Larsen T, Choi B, Won E, Shin K. Amino acid nitrogen and carbon isotope data: potential and implications for ecological studies. Ecol. Evol. 2022;12: e8929. pmid:35784034
View Article
PubMed/NCBI
Google Scholar

[82] View Article

[83] PubMed/NCBI

[84] Google Scholar

[ref28] 28. Fuller BT, Petzke KJ. The dietary protein paradox and threonine ¹⁵N-depletion: pyridoxal-5’- phosphate enzyme activity as a mechanism for the δ¹⁵N trophic level effect. Rapid Commun. Mass Spectrom. 2017; 31:705–718.
View Article
Google Scholar

[86] View Article

[87] Google Scholar

[ref29] 29. Hare PE, Fogel ML, Stafford TW Jr, Mitchell AD, Hoering TC. The isotopic composition of carbon and nitrogen in individual amino acids isolated from modern and fossil proteins. J Archaeol Sci. 1991; 18: 277–292.
View Article
Google Scholar

[89] View Article

[90] Google Scholar

[ref30] 30. McMahon KW, Polito MJ, Abel S, McCarthy MD, Thorrold SR. Carbon and nitrogen isotope fractionation of amino acids in an avian marine predator, the Gentoo penguin (Pygoscelis papua). Ecol Evol. 2015;5: 1278–1290.
View Article
Google Scholar

[92] View Article

[93] Google Scholar

[ref31] 31. Chikaraishi Y, Ogawa NO, Kashiyama Y, Takano Y, Suga H, Tomitani A, et al. Determination of aquatic food–web structure based on compound-specific nitrogen isotopic composition of amino acids. Limnol Oceanogr Methods. 2009; 7: 740–750.
View Article
Google Scholar

[95] View Article

[96] Google Scholar

[ref32] 32. Chikaraishi Y, Ogawa NO, Ohkouchi N. Further evaluation of the trophic level estimation based on nitrogen isotopic composition of amino acids. In: Ohkouch N, Tayau I, Koba K. editors. Earth, Life and Isotopes. 2010. pp. 37–51.
View Article
Google Scholar

[98] View Article

[99] Google Scholar

[ref33] 33. McClelland JW, Montoya JP. Trophic relationships and the nitrogen isotopic composition of amino acids in plankton. Ecol. 2002; 83: 2171–2180.
View Article
Google Scholar

[101] View Article

[102] Google Scholar

[ref34] 34. Naito Y, Honch NV, Chikaraishi Y, Ohkouchi N, Yoneda M. Quantitative evaluation of marine protein contribution in ancient diets based on nitrogen isotope ratios of individual amino acids in bone collagen: an investigation at the Kitakogane Jomon site. Am J Phys Anthropol. 2010;143: 31–40. pmid:20333711
View Article
PubMed/NCBI
Google Scholar

[104] View Article

[105] PubMed/NCBI

[106] Google Scholar

[ref35] 35. Fontanals-Coll M, Soncin S, Talbot HM, von Tersch M, Gibaja JF, Colonese AC, et al. Stable isotope analyses of amino acids reveal the importance of aquatic resources to Mediterranean coastal hunter-gatherers. Proc Biol Sci. 2023; 22:20221330. pmid:36809804
View Article
PubMed/NCBI
Google Scholar

[108] View Article

[109] PubMed/NCBI

[110] Google Scholar

[ref36] 36. Choy K, Yun HY, Lee J, Fuller BT, Shin K-H. Direct isotopic evidence for human millet consumption in the Middle Mumun period: Implication and importance of millets in early agriculture on the Korean Peninsula. J Archaeol Sci. 2021; 129: 105372.
View Article
Google Scholar

[112] View Article

[113] Google Scholar

[ref37] 37. Choy K, Yun HY, Kim SH, Jung S, Fuller BT, Kim DW. Isotopic investigation of skeletal remains at the Imdang tombs reveals high consumption of game birds and social stratification in ancient Korea. Sci Rep. 2021; pmid:34799611
View Article
PubMed/NCBI
Google Scholar

[115] View Article

[116] PubMed/NCBI

[117] Google Scholar

[ref38] 38. Kim J, Park J. Millet vs rice: an evaluation of the farming/language dispersal hypothesis in the Korean context. Evol Hum Sci. 2020; 2: 1–18.
View Article
Google Scholar

[119] View Article

[120] Google Scholar

[ref39] 39. Kim HS. The diet and breeding system of Dongnae people by isotope analysis at Korea. Kogo Kwangchang. 2015; 20: 89–111(Korean).
View Article
Google Scholar

[122] View Article

[123] Google Scholar

[ref40] 40. Fernandes R, Millard A, Nadeau M, Groote P. Food reconstruction using isotopic transferred signals (FRUITS): a Bayesian model for diet reconstruction. PloS One. 2014; 9: e87436. pmid:24551057
View Article
PubMed/NCBI
Google Scholar

[125] View Article

[126] PubMed/NCBI

[127] Google Scholar

[ref41] 41. Stock BC, Jackson AL, Ward EJ, Parnell AD, Phillips DL, Semmens BX. Analyzing mixing systems using a new generation of Bayesian tracer mixing models. Peer J. 2021;
View Article
Google Scholar

[129] View Article

[130] Google Scholar

[ref42] 42. Stock BC, Semmens B. MixSIAR GUI User Manual. Version 3.1. 2016; https://github.com/brianstock/ MixSIAR.

[ref43] 43. Lee RJ, Bale MT. Social change and household geography in Mumun period South Korea. J Anthropol Res. 2016;72: 178–198.
View Article
Google Scholar

[133] View Article

[134] Google Scholar

[ref44] 44. Ahn S-M. The emergence of rice agriculture in Korea: archeobotanical perspectives. Archaeol Anthropol Sci. 2010;2: 89–98.
View Article
Google Scholar

[136] View Article

[137] Google Scholar

[ref45] 45. Lee G-A. Changes in subsistence patterns from the Chulmun to Mumun Periods: Archeobotanical Investigation. Ph.D. Dissertation, Department of Anthropology, University of Toronto. 2003.

[ref46] 46. Kim M. Rice in ancient Korea: status symbol or community food? Antiq. 2015;89: 838–853.
View Article
Google Scholar

[140] View Article

[141] Google Scholar

[ref47] 47. Crawford GW, Lee G-A. Agricultural origins in the Korean Peninsula. Antiq. 2003; 77:87–95.
View Article
Google Scholar

[143] View Article

[144] Google Scholar

[ref48] 48. Yeungnam University Museum (YUM). Excavation reports on Imdang burial mounds-Imdang 2ho. 2002.

[ref49] 49. Kim D-W. A Study on the Funeral System of Ancient Tombs at the Im-dang Site Unpublished Ph.D. dissertation. Yeungnam University, Gyeongsan, South Korea. 2014.

[ref50] 50. Kim D-W. The possibility of non-burial of bodies on tombs of the province under Silla through burial of Gyeongsan Joyeung E-II-2. Sogang J Early Korean Hist. 2018;28: 395–422.
View Article
Google Scholar

[148] View Article

[149] Google Scholar

[ref51] 51. Davey J. Unreliable narratives: Historical and archaeological approaches to early Silla. J Korean Stud. 2016; 29: 7–32.
View Article
Google Scholar

[151] View Article

[152] Google Scholar

[ref52] 52. Nelson SM, Neu J. Mumun, proto-three kingdoms, and three kingdoms in Korea. In: Habu J, Lape PV, Olsen JW. editors. Handbook of East and Southeast Asian Archaeology. 2017. pp 603–617.

[ref53] 53. Yeungnam University Museum (YUM). Archaeological reports on the animal remains from Imdang site in Kyeongsan I- mammals and birds. 2017.

[ref54] 54. Yeungnam University Museum (YUM). Archaeological reports on the animal remains from Imdang site in Kyeongsan II fish and shellfish. 2018.

[ref55] 55. Ko EB. Birds for the Dead: animal offerings from high-mound tombs at Imdang. J Korean Archaeol Soc. 2018;106: 4–41.
View Article
Google Scholar

[157] View Article

[158] Google Scholar

[ref56] 56. Ko EB. Animal sacrifice rituals in ancient Korea- with focus on Silla tombs of the Three Kingdoms period. Anat Biol Anthropol. 2020; 33: 69–77.
View Article
Google Scholar

[160] View Article

[161] Google Scholar

[ref57] 57. Ahn SM. The periodic change of plant use based on plant remains from archaeological sites, Archaeology of Agriculture. 2013.

[ref58] 58. Richards MP, Hedges REM. Stable isotope evidence for similarities in the types of marine foods used by Late Mesolithic humans at sites along the Atlantic coast of Europe. J Archaeol Sci. 1999;26: 717–722.
View Article
Google Scholar

[164] View Article

[165] Google Scholar

[ref59] 59. Brown TA, Nelson D.E, Vogel J.S. & Southon JR. Improved collagen extraction by modified Longin method. Radiocarbon. 1988;30: 171–177.
View Article
Google Scholar

[167] View Article

[168] Google Scholar

[ref60] 60. Corr LT, Berstan R, Evershed RP. Optimization of derivatisation procedures for the determination of d13C values of amino acids by gas chromatography/combustion/isotope ratio mass spectrometry. Rapid Commun Mass Spectrom. 2007;21: 3759–3771.
View Article
Google Scholar

[170] View Article

[171] Google Scholar

[ref61] 61. O’Connell TC, Collins MJ. Comment on “Ecological niche of Neanderthals from Spy Cave revealed by nitrogen isotopes of individual amino acids in collagen”. J Hum Evol. 2017;117: 53–55.
View Article
Google Scholar

[173] View Article

[174] Google Scholar

[ref62] 62. Fogel ML, Tuross N. Extending the limits of palaeodietary studies of humans with compound specific carbon isotope analysis of amino acids. J Archaeol Sci. 2003;30: 535–545.
View Article
Google Scholar

[176] View Article

[177] Google Scholar

[ref63] 63. Silfer JA, Engel MH, Macko SA, Jumeau EL. Stable Carbon Isotope Analysis of Amino Acid Enantiomers by Conventional Isotope Ratio Mass Spectrometry and Combined Gas Chromatography/ Isotope Ratio Mass Spectrometry, Anal Chem. 1991;63: 370–374.
View Article
Google Scholar

[179] View Article

[180] Google Scholar

[ref64] 64. R Core Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing.2019 https://www.R-project.org/.

[ref65] 65. Roberts P, Fernandes R, Craig OE, Larsen T, Lucquin A, Swift J, et al. Calling all archaeologists: guidelines for terminology, methodology, data handling, and reporting when undertaking and reviewing stable isotope applications in archaeology. Rapid Commun Mass Spectrom. 2018;32: 361–372. pmid:29235694
View Article
PubMed/NCBI
Google Scholar

[183] View Article

[184] PubMed/NCBI

[185] Google Scholar

[ref66] 66. Honch NV, McCullagh JSO, Hedges REM. Variation of bone collagen amino acid δ¹³C values in archaeological humans and fauna with different dietary regimes: developing frameworks of dietary discrimination. Am J Phys Anthropol. 2012;148: 495–511.
View Article
Google Scholar

[187] View Article

[188] Google Scholar

[ref67] 67. Paolini M, Ziller L, Laursen KH, Husted S, Camin F. Compound-specific δ¹⁵N and δ¹³C analyses of amino acids for potential discrimination between organically and conventionally grown wheat. J Agric Food Chem. 2015;63: 5841–5850.
View Article
Google Scholar

[190] View Article

[191] Google Scholar

[ref68] 68. Styring AK, Fraser RA, Bogaard A, Evershed RP. Cereal grain, rachis and pulse seed amino acid δ¹⁵N values as indicators of plant nitrogen metabolism. Phytochem. 2014;97: 20–29.
View Article
Google Scholar

[193] View Article

[194] Google Scholar

[ref69] 69. Kendall IP, Woodward P, Clark JP, Styring A.K, Hanna J.V. & Evershed R.P. Compound-specific δ¹⁵N values express differences in amino acid metabolism in plants of varying lignin content. Phytochem. 2019;161: 130–138.
View Article
Google Scholar

[196] View Article

[197] Google Scholar

[ref70] 70. Styring AK. et al. Refining human palaeodietary reconstruction using amino acid δ15N values of plants, animals and humans. J Archaeol Sci. 2019;53: 504–515.
View Article
Google Scholar

[199] View Article

[200] Google Scholar

[ref71] 71. Jim S, Jones V, Ambrose SH, Evershed RP. Quantifying dietary macronutrient sources of carbon for bone collagen biosynthesis using natural abundance stable carbon isotope analysis Br J Nutr. 2006;95: 1055–1062. pmid:16768826
View Article
PubMed/NCBI
Google Scholar

[202] View Article

[203] PubMed/NCBI

[204] Google Scholar

Figures

Abstract

Introduction

Archaeological context

Mumun burials.

Imdang burial mounds.

Materials and methods

Archaeological humans and animals

Modern cereals

Collagen extraction

Preparation of amino acids for GC-C-IRMS

Gas chromatography-combustion-isotope ratio mass spectrometry (GC-C-IRMS)

Statistical analysis

Results

δ15NAA values

δ13CAA values

Principal component analyses of EAAs

MixSIAR model

Discussion

Trophic position estimates

Amino acid proxies and PCA patterns

Comparison of mixing models

Diet of the prehistoric Korean populations

Conclusions

Supporting information

S1 File.

S1 Data.

Acknowledgments

References

δ¹⁵N_AA values

δ¹³C_AA values