Next Article in Journal
Oligomeric Procyanidin Nanoliposomes Prevent Melanogenesis and UV Radiation-Induced Skin Epithelial Cell (HFF-1) Damage
Next Article in Special Issue
Acorn Isotopic Composition: A New Promising Tool for Authenticity Maps of Montado’s High-Value Food Products
Previous Article in Journal
Interaction of Catechins with Human Erythrocytes
Open AccessArticle

Carbon and Nitrogen Isotope Ratios of Food and Beverage in Brazil

1
Laboratory of Isotope Ecology, Center for Nuclear Energy in Agriculture, University of São Paulo, Av. Centenário, 303, São Dimas, Piracicaba CEP 13416-000, SP, Brazil
2
Ecology Department, Institute of Biological Sciences, University of Brasília, Asa Norte, Brasília CEP 70910-900, Brazil
3
National Institute of Criminalistics, Federal Police, Asa Sul, Brasília CEP 70610-200, Brazil
4
Department of Statistics and Informatics, Rural Federal University of Pernambuco, R. Manuel de Medeiros, 35, Dois Irmãos, Recife CEP 52171-050, Brazil
*
Author to whom correspondence should be addressed.
Academic Editors: Nives Ogrinc and Federica Camin
Molecules 2020, 25(6), 1457; https://doi.org/10.3390/molecules25061457
Received: 27 January 2020 / Revised: 4 March 2020 / Accepted: 6 March 2020 / Published: 24 March 2020
(This article belongs to the Special Issue Isotopic Techniques for Food Science)

Abstract

Several previous studies on targeted food items using carbon and nitrogen stable isotope ratios in Brazil have revealed that many of the items investigated are adulterated; mislabeled or even fraud. Here, we present the first Brazilian isotopic baseline assessment that can be used not only in future forensic cases involving food authenticity, but also in human forensic anthropology studies. The δ13C and δ15N were determined in 1245 food items and 374 beverages; most of them made in Brazil. The average δ13C and δ15N of C3 plants were −26.7 ± 1.5‰, and 3.9 ± 3.9‰, respectively, while the average δ13C and δ15N of C4 plants were −11.5 ± 0.8‰ and 4.6 ± 2.6‰, respectively. The δ13C and δ15N of plant-based processed foods were −21.8 ± 4.8‰ and 3.9 ± 2.7‰, respectively. The average δ13C and δ15N of meat, including beef, poultry, pork and lamb were -16.6 ± 4.7‰, and 5.2 ± 2.6‰, respectively, while the δ13C and δ15N of animal-based processed foods were −17.9 ± 3.3‰ and 3.3 ± 3.5‰, respectively. The average δ13C of beverages, including beer and wine was −22.5 ± 3.1‰. We verified that C-C4 constitutes a large proportion of fresh meat, dairy products, as well as animal and plant-based processed foods. The reasons behind this high proportion will be addressed in this study.
Keywords: processed foods; staple foods; photosynthesis metabolism; isotopes; Brazil processed foods; staple foods; photosynthesis metabolism; isotopes; Brazil

1. Introduction

Rapid industrialization and urbanization together with changes in lifestyles are among the factors responsible for changes in eating habits, especially the consumption of highly processed foods [1]. The access, mainly by urban populations, to an immense variety of processed food products triggered the so-called “supermarket era” [2]. Diversification as a marketing strategy has resulted in a vast range of food products in the last two decades. However, consumers may be duped into buying food items without being aware of their quality and integrity [3].
Brazil is one of the world’s main food consumers which makes the country a vast market for food fraud and adulteration through the large quantity of foods produced, exported, or imported [4]. In the Brazilian territory, complex supply chains of foods with animal origin, such as milk and dairy products, are the main targets of food fraud and adulterations, followed by vegetable oils, especially olive oil, considered as high-value products. Meat and fish, as well as their respective by-products, were also involved in some food fraud and adulteration [4]. Recently there has been a couple of scandals refocusing national attention to food adulteration. The “Ouro Branco” and “Carne Fraca” operations conducted by the Brazilian Federal Police revealed the addition of illegal substances in milk and beef, respectively, to increase business profits. The operations received public attention due to the health harm risks that those adulterations inflicted.
Since sophisticated fraudsters have been generally capable of mimicking the physical and chemical compositions of targeted foods, there is a need to address and prevent food fraud and adulteration [4]. Food safety might be also compromised in these cases. In this context, a current tool of choice in food forensics is stable isotope analysis, which can be a rapid and cost-effective method to detect fraud [3]. Several studies on targeted food items using carbon and nitrogen stable isotope ratios in Brazil revealed that many of the food items investigated are adulterated, mislabeled or even fraudulent [5,6,7,8]. In this regard, carbon isotope ratio analyses have become a useful tool to track the abundances of C3 and C4 food sources in human and animal diets [9]. The primary photosynthetic producers support hundreds of millions of heterotrophic species like humans. The C4 photosynthetic pathway evolved in plants 24–35 million years ago as a response to the decrease in CO2 and increase in O2 atmospheric concentrations [10]. By the competitive advantage of these plants in warm environments, extensive grasslands were formed, mainly in low latitudes [11].
Several large terrestrial herbivores, like bovine, equid and other ungulates, followed by top carnivores, and, finally, hominids, have adapted to these new ecosystems [12]. The presence of hominids in grasslands was remarkable as such in human evolution [10]. The importance of C4 plants became even more evident when humans started agriculture, since in this endeavor, C4 plants like maize (Zea mays L.), sorghum (Sorghum bicolor L.), pearl millet (Pennisetum glaucum L.) and sugarcane (Saccharum spp.) were among the species selected by humans to be cultivated [13]. Maize had such an impact in the Americas that in few centuries, the human populations changed from a C3 to a C4-based diet [14]. Furthermore, with the Columbian exchanges, C4 plants were disseminated throughout the old and new continents [15,16,17].
Even with this competitive advantage, there is still a large dominance of C3 species, almost four times more than C4 species. The European Union (EU) and other temperate regions preferentially grow C3 rather than C4 plants, which modifies their food carbon isotopic composition. As an example, Martinelli et al. [18] observed that hamburgers in countries that feed cattle preferentially with C4 grass or maize (e.g., Brazil, Mexico, and USA) have less negative δ13C values in comparison to countries with cattle preferentially fed with C3 plants (e.g., the EU).
Unlike carbon, of which the main source for terrestrial plants is atmospheric CO2, nitrogen can be acquired from both the atmosphere through biological fixation and through the soil. After the Haber–Bosch process became a viable method at industrial scale, commercial crops also commenced to take up nitrogen from mineral fertilizers. Before the large-scale use of these fertilizers, crops had relied mainly on manures, organic sources of nitrogen generated by animals. Due to the multiple nitrogen sources for plant uptake, interpreting the nitrogen isotopic ratio (15N/14N) in plants is not an easy task [19,20]. However, the use of fertilizers with δ15N values around 0‰ can be distinguished from the use of manures with more positive δ15N values [21,22]. Thus, as a first approach, δ15N of crops may be an indicative of the application of synthetic or organic fertilizers for improving plant growth (e.g., [23]).
After being incorporated into the plant tissue, nitrogen moves up along the food web with preferential loss of 14N related to 15N along the way. Consequently, the organisms positioned higher in the food chain tend to be more enriched in 15N than the organisms positioned on the bottom of the chain, the so-called “trophic isotopic discrimination” [24]. Therefore, the nitrogen isotope ratio has been a useful tool to place organisms along the food chain, including humans. Vegetarians and vegans tend to be depleted in 15N in comparison with omnivorous humans [25].
Carbon and nitrogen isotopic ratios also have the potential to serve as tracers of dietary habits and migration in humans and animals [26,27]. Consequently, the isotopic analysis of human remains such as hair, nails, bones or teeth, allows the retrieval of information regarding a person’s diet to reconstruct dietary habits (e.g., [28,29]) or to determine various regions where the decedents have been fed through time, which gives evidences of past mobility patterns [30].
Previous studies performed in Brazil have found that large urban populations have a relatively homogenous diet regardless of the geographic region of the country [25,31]. Based on the stable isotope composition of fingernails, these authors also found that plants with a C4 photosynthetic metabolism (sugarcane, maize and pastures) constitute the major part of the diet in these large urban centers. Based on this fact, we hypothesize here that there is a pervasive presence of C-C4 in Brazilian food items although the staple foods in Brazil are rice (Oriza sativa L.), beans (Phaseolus spp. and Vigna sp.), and cassava (Manihot esculenta L.), plants that follow the C3 photosynthetic metabolism.
In this context, we developed the first Brazilian baseline assessment using carbon and nitrogen stable isotope ratios of about one thousand food items considered essential to test the above hypothesis. We also generated forensic data supporting food authenticity and traceability. As the assessment included in natura as well as processed foods, it might also become a powerful baseline for tracking human movement under the “supermarket diet” trend in Brazil.

2. Results

The complete dataset containing the δ13C and δ15N values of every single sample of foods and beverages is available in the Supplementary Material.

2.1. Plants Used as Food

Considering all samples of in natura plants, there were two main distinct groups observed: those following the C3 photosynthesis metabolism and those using the C4 pathway (Figure 1a). There were also some plants which follow the crassulacean acid metabolism (CAM), such as pineapple (Ananas comosus L.), and their δ13C is intermediate between C3 and C4 plants (Figure 1a). The average δ13C of C3 plants was −26.7‰ (n = 434), varying from −30.7 to −22.5‰, while the average δ13C of C4 plants was −11.5‰ (n = 47), varying from −14.2 to −10.2‰ (Figure 1a), which is a highly significant difference (p < 0.01). The values in the CAM plants varied from −15.3 to −14.1‰, with an average of −14.8‰ (n = 3; Figure 1a). The δ15N of in natura plants had a large variation from −4.4‰ up to more than 18.4‰ (Figure 1b). However, most of them had positive values. The average for C3 plants was 3.9‰, which does not differ from the δ15N of C4 plants (4.5‰; Figure 1b).
Then, according to the FAO classification, plants were grouped into C3-cereals, C4-cereals, fruits, pulses, tubers and vegetables. Except for C4-cereals, the average δ13C values of these groups ranged from −30.7 to −14.1‰ (Figure 2a). The main δ13C difference was between C4-cereals and the other plant groups (p < 0.01). On the other hand, δ15N values of oilseeds (−0.3‰) and pulses (2.4‰) were significantly less positive (p < 0.01) in comparison with other the groups. In addition, nuts (17.3‰) had more positive δ15N values (p < 0.01) than vegetables, tubers, fruits, and C3-cereals, whose average δ15N values varied between −4.4‰ and 15.9‰ (Figure 2b).

2.2. Meats

Beef had the least negative δ13C values (p < 0.01) among different types of meat, close to the δ13C of C4 plants. The δ13C of lamb, pork, and poultry were also more negative than that of beef (p < 0.01) (Figure 3a). On the other hand, freshwater wild fish had the most negative δ13C values (p < 0.01). In contrast, δ13C values in farmed-raised tilapia (Oreochromis niloticus) were similar to those values of pork and poultry (Figure 3a). In addition, marine fish had a similar δ13C, but it was more negative (p < 0.01) than pork and poultry due to the reasons which will be discussed later.
The δ15N of animal protein consumed in southeast Brazil varied considerably among different products (Figure 3b). The most positive δ15N values were observed in wild freshwater and marine fish (p < 0.01); however, δ15N of farmed raised tilapia was only more positive (p < 0.01) than pork and poultry (Figure 3b).

2.3. Processed Food

The isotope values of processed foods are summarized in the tables due to the large number of food items (Table 1 and Table 2). Plant-based processed foods have a wide range of δ13C values (Table 1). Food items with C3-like values are flours from wheat (Triticum spp.) and cassava, noodles and pasta. Alongside these products, cocoa (Theobroma cacao L.) powder and bitter chocolate had also δ13C values resembling C3 plants, as well as vegetal fat (Table 1). Conversely, sugarcane-derived sugar, fruit, jams, juice-powders, chocolate-drink powders, and industrial puddings had typical δ13C values of C4 plants (Table 1). Other products are clearly a mixture of C3 and C4 plants in different proportions. For instance, chocolate bars (bitter, milk, and white) have different δ13C values according to the contents of cocoa, milk and sugar. Chocolate powder for cooking, with an average δ13C of −19‰, represents a mixture of chocolate and C-C4 sugar.
The δ15N values of plant-based processed foods varied from 2‰ to 6‰ in most of the cases (Table 1). Products derived from wheat like flour, noodles, pasta, and cookies had δ15N values resembling those of in natura wheat grain (see supplementary material). The exceptions are seasonings and stocks that had more negative δ15N (p < 0.01), which differed greatly from their feedstock, suggesting a high degree of isotopic discrimination in these products (Table 1).
Among processed foods of animal origin, dairy products followed the δ13C values of milk, however, some dairy products, such as heavy cream and ice cream, tended to have more negative δ13C values compared to milk, although this difference was not significant (Table 2). Processed meat products, like hot dogs, sausages, salami, ham, and lard followed the δ13C of pork meat (Table 2). The same pattern was observed for jello made mainly of bovine collagen (Table 2). Dehydrated meat stock cubes had an average δ13C value of −19.5‰ (Table 2). By looking at the ingredients of these cubes on their labels, it is difficult to discern whether these δ13C values resemble the presence of animal fat tending to have more negative δ13C values in comparison with the meat itself or if they resemble the addition of vegetable oils of C3 plants with also more negative δ13C values. The same conclusion is applicable to meat-based soup powders, sauces, and seasonings with δ13C values close to the stock cubes (Table 2).
The δ15N values of dairy, processed meat, and jello were similar to those values for milk, pork and beef, respectively, implying that there was a small isotopic discrimination during processing (Table 2). On the other hand, meat-based seasonings, primarily dehydrated meat stock cubes, meat-based seasoning had a significantly less positive δ15N value (p < 0.01) related to other types of meat (Table 2), therefore suggesting a strong isotopic discrimination during processing and/or a significant amount of plant- rather than meat-derived ingredients (Table 1).
As a classical method of data presentation in stable isotope studies, δ13C-δ15N biplot (δ-space) is a bidimensional space, giving information about resources consumed by an organism and also the bioclimatic context in each such organism developed, the so-called “isotopic niche” as defined by Newsome et al. [32]. There is a clear separation in the δ-space between in natura food plants and most animal proteins (Figure 4a). Processed foods of animal origin resemble raw meat and poultry in the δ-space (Figure 4b), while processed foods of plant origin seems to have C4 and C3 plants in variable proportions as ingredients (Figure 4c).

2.4. Beverages

The average δ13C of wine samples was −23.1‰ (n = 291) with a wide range of values, from a minimum of −28.4‰ to a maximum of −14.8‰ (Figure 5a). The average δ13C of beer samples was −20.9‰ (n = 78), having a bimodal frequency distribution, with values concentrating around −27‰ and around −19‰ (Figure 5a). A few samples of soda were also analyzed just to confirm the heavy presence of sugar. The average δ13C of five soda samples was −11.7‰, with a standard deviation of only 0.5‰ (Figure 5b).

3. Discussion

The δ13C of C3 and C4 plants used as food were in the range expected considering other studies with native plants [33,34,35,36,37]. Overall, crops are cultivated in a way to avoid shadow and maximize light exposition [38]. Conversely, leaves in a forest must struggle to maximize their exposition to sun light. Consequently, crops also have high evaporative demands, which prompts stomata closing to avoid water losses [36]. As photosynthesis pumps CO2 out from the stomata, pi/pa (i.e., intercellular to ambient CO2 partial pressure) tend to decrease, leading to an increase in the plant δ13C of crops in relation to forest leaves. This difference is obvious in Figure 6a, showing a frequency distribution of δ13C values of C3 crops in this study in relation to δ13C of tree leaves of the Brazilian Atlantic forest (Martinelli et al. under review). There is not much overlap between δ13C of crops and forest besides a difference of ~6‰ between them (Figure 6a).
There was large variability in the δ15N of domesticated plants, from −4‰ to 18‰ (Figure 6b). This trend was expected since plants take up nitrogen from several sources depending on its availability and the climate, and each source has a different nitrogen isotopic ratio [19]. In domesticated plants, in addition to natural sources, soil amendments, such as mineral and organic nitrogen fertilizers, also have to be considered [21]. The primary source of reactive nitrogen before the advent of the Haber–Bosch process (which allowed the production of synthetic fertilizers) was atmospheric nitrogen which becomes available for organisms through biological fixation [39]. The δ15N of the nitrogen fixed is ~0‰ and as mineral nitrogen fertilizers are also synthesized by using atmospheric nitrogen, the δ15N of such fertilizers is also close to 0‰ [22]. Therefore, plants capable of nitrogen fixation or amended with mineral nitrogen fertilizers tend to have less positive δ15N values compared to other plants. This explains why some domesticated plants like pulses (nitrogen-fixing plants of the Fabaceae family) have less positive δ15N values compared to other domesticated plants in this study [40].
Huelsemann et al. [41], based on a large survey of the literature, showed that the average δ15N of vegetables under the influence of mineral fertilizers was 3.1‰, increasing to 6.8‰ and 9.8‰ in vegetables fertilized with organic and animal-derived manure, respectively. The average δ15N of vegetables found here was 4.2‰, which is more positive than the value found by Huelsemann et al. [41] for plants that received mineral fertilizers (Table 3), but less positive than those fertilized with organic manure. The reasons for this difference are difficult to pinpoint, mainly because several nitrogen sources available for domesticated plants. However, it is likely that the following hypotheses can explain this divergence: (i) the more positive δ15N values of tropical soils relative to temperate soils [42]; and (ii) a higher amount of nitrogen fertilizer is used in more developed countries than in Brazil [25].
Bovine, poultry and processed pork meat, together with dairy products and eggs, are the most important sources of animal proteins for Brazilians [43]. Most cattle herds in Brazil are free-range and C4 grass-fed, especially those of the genus Brachiaria. The average δ13C of 51 samples of these grasses collected in the Amazon region was −12‰, and the average δ15N was equal to 2.0‰ (Martinelli, unpublished data). Poultry and pork in Brazil are raised intensively and fed by a feed composed of maize and soy (Glycine max (L.) Merrill) that have the following isotopic composition: δ13C = −16.7‰, and δ15N = 0.6‰ [9]. Consequently, Brazilian beef, poultry and pork (Figure 4a), as well as dairy products such as milk, heavy cream, yogurt and cheese, and also all types of processed meat products like ham, salami, sausages, hot dogs, among others, are rich in C-C4 (Figure 4b). Although lamb is not widely consumed in Brazil, the C-C4 is also present in this meat, but less than beef, poultry and pork (Figure 4a). In comparison with other countries, the McDonald’s hamburger made in Brazil had the largest amount of C-C4 [18]. The same pattern was found by Osorio et al. [44], who analyzed dry defatted meat from Europe and US. Even in comparison with Chinese defatted beef, which is raised on maize and C3 plants, C-C4 in Brazilian beef is likely higher than Chinese beef [45].
Most of the forage species used to feed dairy cattle in Europe are C3 plants and the addition of maize in their diet is being considered as a deviation from the traditional milk production [46]. Therefore, milk produced in Europe also tends to have more negative δ13C values compared to Brazilian milk, although a direct comparison is not possible because the milk fat was not removed in our study [47].
C4-C is also found in Argentinian milk, but not at the same proportions which exists in Brazilian milk, because in addition to maize, C3 plants like soy, cotton (Gossypium hirsutum L.) seeds, alfalfa (Medicago sativa L.) and oats (Avena sativa L.) are also used to feed dairy cattle in the former country [48]. As milk is the basic dairy ingredient, all dairy products in Brazil tend to have a higher C4 content in comparison with dairy products from other countries. The same is true for processed pork meats like ham, which, in Brazil, has a larger proportion of C-C4 than dry cured ham raised in Spain [49] and Italy [50]. A similar pattern was observed for Brazilian lamb relative to lamb raised in Europe or in some parts of South Africa [51,52].
The presence of C-C4 in animal proteins was even observed in farm-raised fish (Figure 4a), especially tilapia, where the production is growing in the country and is widely available in grocery stores [53]. The fish food which is being used to raise farm-raised fish species is also based on maize and soy, and consequently, the average δ13C of farm-raised tilapia is −18.2‰ (Figure 4a). Marine fish also have δ13C values resembling animals fed with C4 plants (Figure 3a), and the reason is the fact that marine phytoplankton, which are the base of oceanic food chains, have average δ13C values close to −21‰ [54], resulting in δ13C values of oceanic food chains close to the values resembling a mixture of C3 and C4-C in terrestrial ecosystems. In contrast, in freshwater systems, the average δ13C of the phytoplankton in Brazil is more negative than −30‰ [55], which results in the most negative δ13C in freshwater wild fish in comparison with all types of meat (Figure 4a).
Wild fish, either from freshwater or from the ocean, had the most positive δ15N values among all types of animal-derived foods (Figure 4a). This is explained by the fact that the food chain in rivers and in the ocean is complex, including several trophic degrees, which increases the isotopic trophic discrimination regarding nitrogen and, therefore, the 15N enrichment of the food [56]. δ15N in farm-raised fish like tilapia was less positive than wild fish since the available food chain of a farm system is based on fish foods produced from maize and soy (Figure 4a). Coletta et al. [9] also observed that the δ15N of free-range chicken was more positive than barn-raised chicken, probably because free-range chicken feed on soil invertebrates. This trend was also observed here between wild animals (bushmeat) and domesticated ones since bushmeat had a more positive δ15N (p < 0.05) than any domesticated animal (Figure 4). We hypothesize that the animal protein produced from large-scale operations tends to have lower δ15N than wild-animal protein since agriculture is a simplification of natural ecosystems (which have more a complex food chain [57]), therefore leading to a low “trophic discrimination”. In addition, soy, a nitrogen-fixing plant that has a low δ15N value, is extensively used as animal feed [58].
Among plant-based processed foods, excluding those that are wheat-based like pasta, noodles, crackers, and snacks, C-C4 is present in variable amounts in several products, including those for which C-C4 occurrence is somehow unexpected (Figure 4c). For instance, commercial fruit jam and preserves have much more C-C4 related to C-C3 in their compositions. C-C4 can also be found in large quantities in soup powders, juice powders, sugary beverages, baby formulas, and spices like cumin (Cuminum cyminum L.), saffron (Crocus sativus L.), and garam masala (an India-style blend of ground spices), which are mixed with a fine maize flour (termed in Brazil as fubá) (Figure 4b). Maize is also present in large quantities in beers made by the largest brewer companies in Brazil (Figure 5). Most Brazilian wines have at least 25% C-C4 due to the use of sugarcane as an adjunct in the grape (Vitis vinifera) fermentation process [6]. Other products also have suspiciously high contents of C-C4, among them, sauces like mustard and ketchups. Perhaps the most iconic example of the heavy presence of C-C4 in Brazilian processed foods is soy sauce (shoyu), which has a dominant proportion of maize rather than soy or wheat (both are traditional ingredients of the classic recipe) [7].
A question arises about the reasons for such pervasive use of C-C4 by the Brazilian food industry. We mention here that the cost of C-C4 in Brazil is overall cheaper than C-C3. Brazil is mostly a tropical country where the growth of C4 plants is favored by a high light incidence [11,12,59]. Therefore, there is a climatic control in the geographical distribution of C4 plants and it seems that this distribution can be tracked by certain types of foods. This is the case of hamburgers made by a global fast-food company. The country-level δ13C of patties grouped by latitude clearly showed that there was an inverse correlation between the proportion of C-C4 and latitude, meaning that countries in the tropical belt had more C-C4 than countries located in higher latitudes [18].
In Brazil, among agricultural commodities, two C4 plants are very important: sugarcane and maize. Brazil is one of the largest producers of sugar from sugarcane in the world. Brazil was the top sugar producer in 2014, producing ~37 million Mg of sugar [60]. Brazil is also an important producer of maize in the global food market. In 2017, Brazil was the third largest producer of this crop after the USA and China, producing almost 100 million Mg [61]. Therefore, sugar in Brazil is a relatively cheap product which is being widely used by the population and by the food industry. In some remote rural communities of the country, such as those in the Amazon region, sugar is used as a source of energy in some periods of the year during which food is in short supply [62].
The price of maize grain in Brazil is lower than rice, soy, and wheat [58] and consequently, it is widely used to feed animals in Brazil as well as in processed foods. For instance, in 2017, Brazil was the second largest producer of chicken meat (almost 40 million Mg), and beef (almost 10 million Mg); and the fifth largest producer of pig meat, with almost 4 million Mg of meat. Chickens and pigs in the country are fed a combining ratio of maize and soy, with maize as the dominant proportion [9] and most Brazilian cattle herds are fed with C4 forage grasses, mainly of the genus Brachiaria. Therefore, although raw meat is a relatively expensive item for Brazilians, any meat included in their diets would be a source of C-C4 (Figure 4a,b). On the other hand, processed meat, especially pork sausage, termed calabresa in Brazil, has become a cheap alternative for the low-income population, primarily in areas where refrigeration is not available, such as isolated communities of the Amazon region [62]. Coupled with the abundance of cheap C-C4 in the country, there is also the fact that the Brazilian food legislation is usually very flexible regarding the use of ingredients, allowing the use of maize or sugarcane (sources of C-C4) instead of C3-derived plants in products such as wine [6] and soy sauce [7].
By investigating a significant number of food items, we believe that it is fair to conclude that although the traditional staple foods of Brazilians are composed mainly of plants that follow the photosynthetic C3 metabolism (e.g., rice, beans, and cassava), C-C4 plants have a higher degree of pervasiveness in the Brazilian food system (Figure 4), leading to high unaware consumption of these plants by Brazilians (especially processed food [63]), which is ultimately reflected in the more positive δ13C tissue values found in the national population, confirming our initial hypothesis.
Finally, addressing and preventing food fraud and adulteration requires not only enforcement of regulatory systems but also more sampling, monitoring, and development of cost-effective methods contributing to fraud detection [4]. In this sense, understanding and interpreting carbon and nitrogen isotope ratios of both plant and animal based food items together with the use of the raw isotope database would support actions and investigations regarding food adulteration and might assist the prevention of food fraud in the Brazilian territory.

4. Methods

4.1. Sampling Protocol

We analyzed the stable carbon and nitrogen abundances in 1619 samples (804 were composed of in natura vegetal and animal products, while 815 samples were processed foods). Of the total, 1245 and 374 were food items and beverages, respectively. Most of the samples were bought during 2015–2019 in grocery stores located in Piracicaba, a municipality in the Southeast Region of Brazil with a population of ~400,000 inhabitants. In natura animal samples include the following species: cattle, chicken, pig, lamb, freshwater and marine fish, and bushmeat (feral pig and peccary). In natura vegetal samples include cereals, fruits, pulses, tubers, and vegetables grouped according to the FAO classification. For processed food, the classification proposed by Monteiro et al. [64] was followed. Processed foods with animal origin included dairy, cured meat, meat stock, and lard. Dairy products included yogurt, milk cream, ice cream, cheese, and butter. Cured meat included sausages, ham, hot dogs, salami, mortadella, and canned cook pork. Meat stock is defined here as industrially reduced meat stock made from beef, poultry, and pork (Figure 1). Plant-based processed food included: flour, oil and fat, sweeties, chocolates, sauces and soups. “Flour” included the following plants: cassava, maize and soy. “Fats” included margarine and coconut oil. “Sweeties” included a diverse variety of products: corn (maize) flakes, cereal bars, cookies, cakes, pastries, jams, and puddings. “Chocolate” included chocolate bars and chocolate powders mixed with milk (similar to powders used to prepare hot chocolate). “Sauces and soups” included: dehydrated soups, salad sauces, mustard, ketchup, soy sauce (shoyu), and Worcestershire sauce. Finally, in the group defined as “Others”, the following items were included: powder spices, like cumim, urucum, saffron, and sweeties like marshmallow, Japanese soy-based products like miso paste, instant misoshiro powder, baby formula, and pre-cooked polenta.

4.2. Isotopic Analysis

Solid foods were oven-dried at 60 °C to a constant weight. Subsequently, samples were finely ground to facilitate homogenization. Processed foods were composed of several ingredients and cookies with chocolate chips and cookies stuffed with cream were ground together. After homogenization, 1–2 mg of sample was transferred to a tin capsule for further elemental and isotopic analysis. Beverages like soda, beer, and wine were directly placed in tin capsules.
The isotope ratios of carbon (13C/12C) and nitrogen (15N/14N) of each sample were determined using a continuous-flow isotope ratio mass spectrometer (Delta Plus, ThermoFisher Scientific, Bremen, Germany) coupled to an elemental analyzer (CHN-1110, Carlo Erba, Rodano, Italy) at the Laboratory of Isotope Ecology of the Center for Nuclear Energy in Agriculture, University of São Paulo.
Carbon and nitrogen isotope compositions were calculated as
δ   ( ) = [ ( R s a m p l e / R s t a n d a r d ) 1 ] × 1000
where R is the ratio of 13C/12C or 15N/14N. Stable isotope ratios were measured using an internationally recognized standard and relative to a laboratory standard. Thus, we used the 25-(Bis(5-tert-butyl-2-benzo-oxazol-2-yl) thiophene (BBOT; C26H26N2O2S; Fisons Instruments Inc., Danvers, MA, USA) as an international standard while fine-milled sugarcane leaves were used as a laboratory standard.

4.3. Statistical Analysis

Descriptive statistics (mean, standard deviation, median, inter quartile range, and minimum and maximum values) were used to report the δ13C and δ15N values of food and beverage samples. To test the differences between food items, ANOVA was used because carbon and nitrogen isotope values followed a normal distribution. The post-hoc Tukey test was used to pinpoint specific differences between food items. Statistical analyses were performed in R (version 3.6.3, The R Foundation for Statistical Computing, Boston, MA, USA) and RStudio (version 1.2.5019, RStudio, Vienna, Austria) using the “multcomp” package [65].

Supplementary Materials

Spreadsheet: δ13C and δ15N values of every single sample of foods and beverages analyzed in this study.

Author Contributions

Conceptualization, L.A.M. and G.B.N.; Data curation, M.A.Z.P., G.A.J., F.C.F., J.G.G.O., I.S.O., S.H.L., E.A.M. and P.B.C.; Formal analysis, M.A.Z.P., G.A.J., F.C.F., J.G.G.O., I.S.O., S.H.L., P.J.D.-N. and M.Z.M.; Methodology, L.A.M.; Writing—original draft, L.A.M., G.B.N., T.F.G., A.S., A.L.A.F. and E.M.; Writing—review and editing, E.M., T.F.G., A.S., A.L.A.F., F.J.V.C., P.J.D.-N., M.Z.M. and P.B.C. All authors have read and agreed to the published version of the manuscript.

Funding

São Paulo Research Foundation (FAPESP), grant number #2011/50345-9.

Acknowledgments

We would like to thank Julia Gerds for her kind assistance during early stages of the study (FAPESP project, grant number #2011/50345-9).

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Monteiro, C.A.; Moura, E.C.; Conde, W.L.; Popkin, B.M. Public health reviews socioeconomic status and obesity in adult populations of developing countries: A review. Bull. World Health Organ. 2004, 82, 940–946. [Google Scholar] [PubMed]
  2. Popkin, B.M.; Gordon-Larsen, P. The nutrition transition: Worldwide obesity dynamics and their determinants. Int. J. Obes. Relat. Metab. Disord. 2004, 28, S2–S9. [Google Scholar] [CrossRef] [PubMed]
  3. Carter, J.; Chesson, L.A. Food Forensics: Stable Isotopes as a Guide to Authenticity and Origin, 1st ed.; CRC Press: Boca Raton, FL, USA, 2017. [Google Scholar]
  4. Tibola, C.S.; da Silva, S.A.; Dossa, A.A.; Patrício, D.I. Economically motivated food fraud and adulteration in Brazil: Incidents and alternatives to minimize occurrence. J. Food Sci. 2018, 83, 2028–2038. [Google Scholar] [CrossRef]
  5. Mardegan, S.F.; Andrade, T.M.B.; de Sousa Neto, E.R.; de Castro Vasconcellos, E.B.; Martins, L.F.B.; Mendonça, T.G.; Martinelli, L.A. Stable carbon isotopic composition of Brazilian beers-A comparison between large- and small-scale breweries. J. Food Compos. Anal. 2013, 29, 52–57. [Google Scholar] [CrossRef]
  6. Martinelli, L.A.; Moreira, M.Z.; Ometto, J.P.H.B.; Alcarde, A.R.; Rizzon, L.A.; Stange, E.; Ehleringer, J.R. Stable carbon isotopic composition of the wine and CO2-bubbles of sparkling wines: Detecting C4 sugar additions. J. Agric. Food Chem. 2003, 51, 2625–2631. [Google Scholar] [CrossRef]
  7. Morais, M.C.; Pellegrinetti, T.A.; Sturion, L.C.; Sattolo, T.M.S.; Martinelli, L.A. Stable carbon isotopic composition indicates large presence of maize in Brazilian soy sauces (shoyu). J. Food Compos. Anal. 2018, 70, 18–21. [Google Scholar] [CrossRef]
  8. Ducatti, R.; Pinto, J.P.D.A.N.; Sartori, M.M.P.; Ducatti, C. Quantification of soy protein using the isotope method (δ13C and δ15N) for commercial brands of beef hamburger. Meat Sci. 2016, 122, 97–100. [Google Scholar] [CrossRef]
  9. Coletta, L.D.; Pereira, A.L.; Coelho, A.A.D.; Savino, V.J.M.; Menten, J.F.M.; Correr, E.; Frana, L.C.; Martinelli, L.A. Barn vs. free-range chickens: Differences in their diets determined by stable isotopes. Food Chem. 2012, 131. [Google Scholar] [CrossRef]
  10. Sage, R.F. The evolution of C4 photosynthesis. New Phytol. 2004, 341–370. [Google Scholar] [CrossRef]
  11. Still, C.J.; Berry, J.A.; Collatz, G.J.; Defries, R.S. Global distribution of C3 and C4 vegetation: Carbon cycle implications. Global Biogeochem. Cycles 2003, 17, 1006. [Google Scholar] [CrossRef]
  12. Sage, R.F.; Stata, M. Photosynthetic diversity meets biodiversity: The C4 plant example. J. Plant Physiol. 2015, 172, 104–119. [Google Scholar] [CrossRef]
  13. Perry, L.; Sandweiss, D.H.; Piperno, D.R.; Rademaker, K.; Malpass, M.A.; Vera, P.D. Early maize agriculture and interzonal interaction in. Nature 2006, 440, 76–79. [Google Scholar] [CrossRef]
  14. van der Merwe, N.J.; Tschauner, H. C4 Plants and the development of human societies. In C4 Plant Biology; Sage, R.F., Monson, K.M., Eds.; CRC Press: San Diego, CA, USA, 1999; pp. 509–549. [Google Scholar]
  15. Andrews, J. Diffusion of Mesoamerican food complex to southeastern Europe. Geogr. Rev. 1993, 83, 194–204. [Google Scholar] [CrossRef]
  16. Zerjal, C.M.T.; Dumas, V.C.F.; Bedoya, D.M.C. Out of America: Tracing the genetic footprints of the global diffusion of maize. Theor. Appl. Genet. 2013, 126, 2671–2682. [Google Scholar]
  17. Cherniwchan, J.; Moreno-cruz, J. Maize and precolonial Africa. J. Dev. Econ. 2019, 136, 137–150. [Google Scholar] [CrossRef]
  18. Martinelli, L.A.; Nardoto, G.B.; Chesson, L.A.; Rinaldi, F.D.; Ometto, J.P.H.B.; Cerling, T.E.; Ehleringer, J.R. Worldwide stable carbon and nitrogen isotopes of Big Mac® patties: An example of a truly “glocal” food. Food Chem. 2011, 127, 1712–1718. [Google Scholar] [CrossRef]
  19. Högberg, P. 15N natural abundance in soil-plant systems. New Phytol. 1997, 137, 179–203. [Google Scholar] [CrossRef]
  20. Robinson, D. δ15N as an integrator of the nitrogen cycle. Trends Ecol. Evol. 2001, 16, 153–162. [Google Scholar] [CrossRef]
  21. Yoneyama, T.; Kouno, K.; Yazaki, J. Variation of natural N abundance of crops and soils in Japan with special reference to the effect of soil conditions and fertilizer application. Soil Sci. Plant Nutr. 1990, 36, 667–675. [Google Scholar] [CrossRef]
  22. Bateman, A.S.; Kelly, S.D. Fertilizer nitrogen isotope signatures. Isot. Environ. Health Stud. 2007, 43, 237–247. [Google Scholar] [CrossRef]
  23. Nishida, M. Natural 15N abundance can aid the discrimination of organic and conventional rice. JARQ 2018, 52, 173–180. [Google Scholar] [CrossRef]
  24. Minagawa, M.; Wada, E. Stepwise enrichment of N along food chains; Further evidence and the relation between delta 15N and animal age. Geochim. Cosmochim. Acta 1984, 48, 1135–1140. [Google Scholar] [CrossRef]
  25. Nardoto, G.B.; Silva, S.; Kendall, C.; Ehleringer, J.R.; Chesson, L.A.; Ferraz, E.S.B.; Moreira, M.Z.; Ometto, J.P.H.B.; Martinelli, L.A. Geographical patterns of human diet derived from stable-isotope analysis of fingernails. Am. J. Phys. Anthropol. 2006, 131, 137–146. [Google Scholar] [CrossRef] [PubMed]
  26. DeNiro, M.J.; Epstein, S. Influence of diet on the distribution of carbon isotopes in animals. Geochim. Cosmochim. Acta 1978, 42, 495–506. [Google Scholar] [CrossRef]
  27. DeNiro, M.J.; Epstein, S. Influence of diet on the distribution of nitrogen isotopes in animals. Geochim. Cosmochim. Acta 1981, 45, 341–351. [Google Scholar] [CrossRef]
  28. Sponheimer, M.; Alemseged, Z.; Cerling, T.E.; Grine, F.E.; Kimbel, W.H.; Leakey, M.G.; Lee-Thorp, J.A.; Manthi, F.K.; Reed, K.E.; Wood, B.A.; et al. Isotopic evidence of early hominin diets. Proc. Natl. Acad. Sci. USA 2013, 110, 10513–10518. [Google Scholar] [CrossRef]
  29. Minagawa, M. Reconstruction of human diet from σ13C and σ15N in contemporary Japanese hair: A stochastic method for estimating multi-source contribution by double isotopic tracers. Appl. Geochem. 1992, 7, 145–158. [Google Scholar] [CrossRef]
  30. Thompson, A.H.; Chesson, L.A.; Podlesak, D.W.; Bowen, G.J.; Cerling, T.E.; Ehleringer, J.R. Stable isotope analysis of modern human hair collected from Asia (China, India, Mongolia, and Pakistan). Am. J. Phys. Anthropol. 2010, 141, 440–451. [Google Scholar] [CrossRef]
  31. Nardoto, G.B.; Murrieta, R.S.S.; Prates, L.E.G.; Adams, C.; Garavello, M.E.P.E.; Schor, T.; Moraes, A.; Rinaldi, F.D.; Gragnani, J.G.; Moura, E.A.F.; et al. Frozen chicken for wild fish: Nutritional transition in the Brazilian Amazon region determined by carbon and nitrogen stable isotope ratios in fingernails. Am. J. Hum. Biol. 2011, 23, 642–650. [Google Scholar] [CrossRef]
  32. Newsome, S.D.; Maartinez del Rio, C.; Bearhop, S.; Phillips, D.L. A niche for isotopic ecology. Front. Ecol. Environ. 2007, 5, 429–436. [Google Scholar] [CrossRef]
  33. Brescia, M.A.; Di Martino, G.; Guillou, C.; Reniero, F.; Sacco, A.; Serra, F.; Orabona, V. Differentiation of the geographical origin of durum wheat semolina samples on the basis of isotopic composition. Rapid Commun. Mass Spectrom. 2002, 16, 2286–2290. [Google Scholar] [CrossRef] [PubMed]
  34. Kelly, S.; Baxter, M.; Chapman, S.; Rhodes, C.; Dennis, J.; Brereton, P. The application of isotopic and elemental analysis to determine the geographical origin of premium long grain rice. Eur. Food Res. Technol. 2002, 214, 72–78. [Google Scholar] [CrossRef]
  35. Suzuki, Y.; Chikaraishi, Y.; Ogawa, N.O. Geographical origin of polished rice based on multiple element and stable isotope analyses. Food Chem. 2008, 109, 470–475. [Google Scholar] [CrossRef] [PubMed]
  36. Luo, D.; Dong, H.; Luo, H.; Xian, Y.; Wan, J.; Guo, X.; Wu, Y. The application of stable isotope ratio analysis to determine the geographical origin of wheat. Food Chem. 2015, 174, 197–201. [Google Scholar] [CrossRef] [PubMed]
  37. Chung, I.; Kim, J.; Lee, K.; Park, S.; Lee, J.; Son, N.; Jin, Y.; Kim, S. Geographic authentication of Asian rice (Oryza sativa L.) using multi- elemental and stable isotopic data combined with multivariate analysis. Food Chem. 2018, 240, 840–849. [Google Scholar] [CrossRef] [PubMed]
  38. Araus, L.; Ferrio, J.P.; Voltas, J. Agronomic conditions and crop evolution in ancient Near East Agriculture. Nat. Commun. 2014, 5, 3953. [Google Scholar] [CrossRef]
  39. Shearer, G.; Kohl, D.H.; Virginia, R.A.; Bryan, B.A.; Skeeters, J.L.; Nilsen, E.T.; Sharifi, M.R.; Rundel, P.W. Estimates of N2-fixation from variation in the natural abundance of 15N in Sonoran desert ecosystems. Oecologia 1983, 56, 365–373. [Google Scholar] [CrossRef]
  40. Szpak, P.; Longstaffe, F.J.; Millaire, J.; White, C.D. Large variation in nitrogen isotopic composition of a fertilized legume. J. Archaeol. Sci. 2014, 45, 72–79. [Google Scholar] [CrossRef]
  41. Huelsemann, F.; Koehler, K.; Braun, H.; Schaenzer, W.; Flenker, U. Human dietary δ15N intake: Representative data for principle food items. Am. J. Phys. Anthropol. 2013, 152, 58–66. [Google Scholar] [CrossRef]
  42. Martinelli, L.A.; Piccolo, M.C.; Townsend, A.R.; Vitousek, P.M.; Cuevas, E.; McDowell, W.; Robertson, G.P.; Santos, O.C.; Treseder, K. Nitrogen stable isotopic composition of leaves and soil: Tropical versus temperate forests. Biogeochemistry 1999, 46, 45–65. [Google Scholar] [CrossRef]
  43. da Costa Louzada, M.L.; Ricardo, C.Z.; Steele, E.M.; Levy, R.B.; Cannon, G.; Monteiro, C.A. The share of ultra-processed foods determines the overall nutritional quality of diets in Brazil. Public Health 2017, 21, 94–102. [Google Scholar]
  44. Osorio, M.T.; Moloney, A.P.; Schmidt, O.; Monahan, F.J. Multielement Isotope Analysis of bovine muscle for determination of international geographical origin of meat. J. Agric. Food Chem. 2011, 59, 3285–3294. [Google Scholar] [CrossRef] [PubMed]
  45. Zhao, Y.; Zhang, B.; Chen, G.; Chen, A.; Yang, S.; Ye, Z. Recent developments in application of stable isotope analysis on agro-product authenticity and traceability. Food Chem. 2014, 145, 300–305. [Google Scholar] [CrossRef] [PubMed]
  46. Camin, F.; Perini, M.; Colombari, G.; Bontempo, L.; Versini, G.; Wiley, J. Influence of dietary composition on the carbon, nitrogen, oxygen and hydrogen stable isotope ratios of milk. Rapid Commun. Mass Spectrom. 2008, 22, 1690–1696. [Google Scholar] [CrossRef] [PubMed]
  47. Camin, F.; Bontempo, L.; Perini, M.; Piasentier, E. Stable isotope ratio analysis for assessing the authenticity of food of animal origin. Compr. Rev. Food Sci. Food Saf. 2016, 15, 868–877. [Google Scholar] [CrossRef]
  48. Gribo, J.; Baroni, M. V; Horacek, M.; Wunderlin, D.A.; Monferran, M.V. Multielemental + isotopic fingerprint enables linking soil, water, forage and milk composition, assessing the geographical origin of Argentinean milk. Food Chem. 2019, 283, 549–558. [Google Scholar] [CrossRef] [PubMed]
  49. González-Martin, I. Use of isotope analysis to characterize meat fromIberian-breed swine. Meat Sci. 1999, 52, 437–441. [Google Scholar] [CrossRef]
  50. Perini, M.; Camin, F.; Sánchez, J.; Piasentier, E. Effect of origin, breeding and processing conditions on the isotope ratios of bioelements in dry-cured ham. Food Chem. 2013, 136, 1543–1550. [Google Scholar] [CrossRef]
  51. Camin, F.; Bontempo, L.; Heinrich, K.; Horacek, M.; Kelly, S.D.; Schlicht, C.; Thomas, F.; Monahan, F.J.; Hoogewerff, J.; Rossmann, A. Multi-element (H, C, N, S) stable isotope characteristics of lamb meat from different European regions. Anal. Bioanal. Chem. 2007, 389, 309–320. [Google Scholar] [CrossRef]
  52. Erasmus, S.W.; Muller, M.; Butler, M.; Hoffman, L.C. The truth is in the isotopes: Authenticating regionally unique South African lamb. Food Chem. 2018, 239, 926–934. [Google Scholar] [CrossRef]
  53. Castilho-Barros, L.; Owatari, M.S.; Mouriño, J.L.P.; Silva, B.C.; Sei, W.Q. Economic feasibility of tilapia culture in southern Brazil: A small-scale farm model. Aquaculture 2020, 515. [Google Scholar] [CrossRef]
  54. Hardt, F.A.S.; Cremer, M.J.; Tonello Junior, A.J.; Bellante, A.; Buffa, G.; Buscaino, G.; Mazzola, S.; Barreto, A.S.; Martinelli, L.A.; Zuppi, G.M. Use of carbon and nitrogen stable isotopes to study the feeding ecology of small coastal cetacean populations in southern Brazil. Biota Neotrop. 2013, 13, 90–98. [Google Scholar] [CrossRef]
  55. Benedito-Cecilio, E.; Araujo-lima, C.M.; Forsberg, B.R.; Bittencourt, M.M.; Martinelli, L.C. Carbon sources of Amazonian fisheries. Fish. Manag. Ecol. 2000, 7, 305–315. [Google Scholar] [CrossRef]
  56. Newsome, S.D.; Clementz, M.T.; Koch, P.L. Using stable isotope biogeochemistry to study marine mammal ecology. Mar. Mammal Sci. 2010, 26, 509–572. [Google Scholar] [CrossRef]
  57. Phalan, B.; Balmford, A.; Green, R.E.; Scharlemann, J.P.W. Minimising the harm to biodiversity of producing more food globally. Food Policy 2011, 36, S62–S71. [Google Scholar] [CrossRef]
  58. Garrett, R.D.; Lambin, E.F.; Naylor, R.L. Land Use Policy Land institutions and supply chain configurations as determinants of soybean planted area and yields in Brazil. Land Use Policy 2013, 31, 385–396. [Google Scholar] [CrossRef]
  59. Collatz, G.J.; Berry, J.A.; Clark, J.S. Effects of climate and atmospheric CO2 partial pressure on the global distribution of C4 grasses: Present, past, and future. Oecologia 1998, 114, 441–454. [Google Scholar] [CrossRef]
  60. FAOSTAT. Crops Processed. Available online: http://www.fao.org/faostat/en/#data/QD (accessed on 26 February 2020).
  61. FAOSTAT. Crops. Available online: http://www.fao.org/faostat/en/#data/QC (accessed on 26 February 2020).
  62. Murrieta, R. O dilema do papa-chibé: Consumo alimentar, nutrição e práticas de intervenção na Ilha de Ituqui, baixo Amazonas, Pará. Rev. Antropol. 1998, 42, 1–33. [Google Scholar] [CrossRef]
  63. Schor, T.; Azenha, G.S. Ribeirinho Food Regimes, Socioeconomic inclusion and unsustainable development of the Amazonian floodplain. EchoGéo 2017, 41, 1–14. [Google Scholar] [CrossRef]
  64. Monteiro, C.A.; Levy, R.B. A new classification of foods based on the extent and purpose of their processing. Cad. Saude Publica 2010, 26, 2039–2049. [Google Scholar] [CrossRef] [PubMed]
  65. Hothorn, T.; Bretz, F.; Westfall, P. Simultaneous inference in general parametric models. Biom. J. 2008, 50, 346–363. [Google Scholar] [CrossRef] [PubMed]
Sample Availability: Samples of the compounds are available from the authors.
Figure 1. Frequency histogram of δ13C (a) and δ15N (b) of in natura plants used as food in Brazil. The green bars represent C4 plants, the blue bars represent CAM plants, and the red bars represent C3 plants.
Figure 1. Frequency histogram of δ13C (a) and δ15N (b) of in natura plants used as food in Brazil. The green bars represent C4 plants, the blue bars represent CAM plants, and the red bars represent C3 plants.
Molecules 25 01457 g001
Figure 2. Box-whisker of δ13C (a) and δ15N (b) of in natura plants used as food in Brazil grouped according to the FAO classification. C-C3: cereals that follow the C3 photosynthetic metabolism (wheat and rice); C-C4: cereals that follow the C4 photosynthetic pathway (maize); Fru: fruits; Nuts: nuts; Pulse: pulses; S-Oil: oilseeds; Tuber: tubers; Veg: vegetables. See the Supplementary Material for food items included in each of these groups.
Figure 2. Box-whisker of δ13C (a) and δ15N (b) of in natura plants used as food in Brazil grouped according to the FAO classification. C-C3: cereals that follow the C3 photosynthetic metabolism (wheat and rice); C-C4: cereals that follow the C4 photosynthetic pathway (maize); Fru: fruits; Nuts: nuts; Pulse: pulses; S-Oil: oilseeds; Tuber: tubers; Veg: vegetables. See the Supplementary Material for food items included in each of these groups.
Molecules 25 01457 g002
Figure 3. Box-whisker of δ13C (a) and δ15N (b) of raw meats used as food in Brazil grouped as follows: beef, bushmeat (Bush), farm-raised fish (F-Farm), freshwater fish (F-Fres), marine fish (F-Mar), dairy, lamb, pork, and poultry.
Figure 3. Box-whisker of δ13C (a) and δ15N (b) of raw meats used as food in Brazil grouped as follows: beef, bushmeat (Bush), farm-raised fish (F-Farm), freshwater fish (F-Fres), marine fish (F-Mar), dairy, lamb, pork, and poultry.
Molecules 25 01457 g003
Figure 4. Biplot of δ13C vs. δ15N of: (a) plants (yellow triangle), and animal protein (yellow circle). A: C3-cereal; B: C4-cereal; C: fruit; D: pulse; E: tuber; F: vegetable. 1: beef; 2: bushmeat; 3: poultry; 4: egg; 5: freshwater fish; 6: lamb; 7: marine fish; 8: milk; 9: pork; 10: farmed fish (tilapia). (b) plants (yellow triangle), animal protein (yellow circle), processed food of animal origin (small grey circle). (c) plants (yellow triangle), animal protein (yellow circle), processed food of plant origin (small grey circle). In order to facilitate visualization, only processed foods with δ15N > 0‰ were included in (b,c).
Figure 4. Biplot of δ13C vs. δ15N of: (a) plants (yellow triangle), and animal protein (yellow circle). A: C3-cereal; B: C4-cereal; C: fruit; D: pulse; E: tuber; F: vegetable. 1: beef; 2: bushmeat; 3: poultry; 4: egg; 5: freshwater fish; 6: lamb; 7: marine fish; 8: milk; 9: pork; 10: farmed fish (tilapia). (b) plants (yellow triangle), animal protein (yellow circle), processed food of animal origin (small grey circle). (c) plants (yellow triangle), animal protein (yellow circle), processed food of plant origin (small grey circle). In order to facilitate visualization, only processed foods with δ15N > 0‰ were included in (b,c).
Molecules 25 01457 g004
Figure 5. Frequency histogram (a) box-whisker (b) of δ13C of alcoholic beverages in Brazil. The blue bars represent wine samples, the green bars represent cachaça samples, and the red bars represent beer samples.
Figure 5. Frequency histogram (a) box-whisker (b) of δ13C of alcoholic beverages in Brazil. The blue bars represent wine samples, the green bars represent cachaça samples, and the red bars represent beer samples.
Molecules 25 01457 g005
Figure 6. Frequency histogram of δ13C (a) and δ15N (b) comparing native tree leaves of the Atlantic Forest (blue bars) with domesticated plants used as food in Brazil (red bars).
Figure 6. Frequency histogram of δ13C (a) and δ15N (b) comparing native tree leaves of the Atlantic Forest (blue bars) with domesticated plants used as food in Brazil (red bars).
Molecules 25 01457 g006
Table 1. Descriptive statistics* of δ13C and δ15N of plant-based processed foods.
Table 1. Descriptive statistics* of δ13C and δ15N of plant-based processed foods.
δ13C (‰).MeanSDMedianIQRMinMaxn
Cake−20.41.6−20.22.4−22.8−18.211
Chocolate-bitter−26.52.2−26.12.0−31.0−24.319
Chocolate-drink−14.41.1−14.00.9−17.2−13.018
Chocolate-milk−22.00.9−22.21.3−23.4−20.128
Chocolate-powder−18.82.3−19.82.1−20.8−14.86
Chocolate-white−21.90.8−21.70.8−23.8−20.910
Cocoa powder−27.82.3−28.30.7−29.1−20.512
Cookie−23.00.7−23.10.8−23.7−21.412
Fat−28.92.0−30.12.6−30.5−25.48
Flour−26.10.8−26.00.9−27.5−23.920
Jam−15.54.7−13.52.8−26.5−11.633
Juice-powder−13.01.1−12.71.5−14.7−11.810
Noodles−26.90.4−27.10.8−27.3−26.29
Oil−29.51.3−30.01.5−30.8−27.47
Others−17.74.5−17.97.6−24.2−10.919
Pasta−25.70.8−26.21.3−26.4−24.65
Pudding−12.82.3−11.43.2−16.1−11.36
Sauce−20.35.6−20.09.9−29.8−12.327
Seasoning−18.02.8−19.53.3−20.3−13.75
Snack−22.54.8−21.08.8−27.9−17.09
Soda−12.10.5−11.90.4−13.0−11.75
Soup powder−21.82.1−22.61.0−23.3−17.76
Stock−23.82.5−24.62.4−25.7−21.03
Sugar−12.20.3−12.20.4−13.2−11.815
δ15N (‰)MeanSDMedianIQRMinMaxn
Cake3.70.33.60.13.34.311
Chocolate-bitter5.90.75.80.74.47.319
Chocolate-drink5.20.65.40.93.66.018
Chocolate-milk5.80.45.80.55.16.828
Chocolate-powder5.80.85.90.34.36.76
Chocolate-white6.00.65.90.44.96.810
Cocoa powder5.90.96.11.23.86.912
Cookie3.70.93.81.12.45.112
Fat2.52.43.82.1−0.24.08
Flour5.23.43.65.50.910.720
Jam4.52.04.82.80.38.433
Juice-powder−0.42.50.51.1−5.81.810
Noodles3.20.23.20.22.93.59
Oil2.95.12.93.6−0.76.57
Others2.72.32.52.6−0.87.719
Pasta3.30.63.40.72.44.05
Pudding5.90.15.90.15.85.96
Sauce2.22.11.53.6−1.45.827
Seasoning−2.72.4−3.13.2−5.10.75
Snack2.90.52.90.42.13.79
Soda-a------
Soup powder2.01.32.00.80.04.06
Stock−3.13.7−3.73.7−6.40.93
Sugar-------
* SD: standard deviation; IQR: inter quartile range; Min: minimum value; Max: maximum value. a Not determined.
Table 2. Descriptive statistics* of δ13C and δ15N of animal-based processed foods.
Table 2. Descriptive statistics* of δ13C and δ15N of animal-based processed foods.
δ13C (‰)MeanSDMedianIQRMinMaxn
Cheese−16.02.0−15.70.9−22.4−14.114
Cream cheese−18.61.2−18.61.5−20.3−17.45
Dehydrated stock cube−19.55.2−21.49.0−25.8−12.310
Ice cream−21.01.1−21.41.1−21.8−19.16
Jello−12.00.4−12.00.5−12.5−11.56
Lard−19.92.8−19.84.7−23.4−16.68
Processed meat−16.80.9−16.91.0−18.5−14.625
Sauce−19.05.2−19.68.1−27.3−13.511
Seasoning−18.23.7−17.15.8−23.5−14.86
Soup powder−20.12.0−20.53.2−22.6−16.612
Yogurt−17.52.0−17.31.7−21.4−14.313
δ15N (‰)MeanSDMedianIQRMinMaxn
Cheese5.91.25.91.84.27.914
Cream cheese5.40.45.50.34.85.75
Dehydrated stock cube−4.31.2−4.50.9−6.0−2.010
Ice cream5.90.26.00.25.56.16
Jello7.80.98.31.46.58.66
Lard4.31.14.01.03.35.98
Processed meat4.21.73.82.81.86.725
Sauce2.52.72.33.0−1.86.911
Seasoning−2.32.2−3.13.4−4.50.66
Soup powder1.51.81.62.6−2.34.412
Yogurt5.10.75.11.24.16.513
* SD: standard deviation; IQR: inter quartile range; Min: minimum value; Max: maximum value.
Table 3. Comparison of average δ15N (‰) of domesticated plants reported this study and those in the meta-analysis of Huelsemann et al. (2015) [41]. Plants were grouped following the FAO classification. The values represent mean ± standard deviation.
Table 3. Comparison of average δ15N (‰) of domesticated plants reported this study and those in the meta-analysis of Huelsemann et al. (2015) [41]. Plants were grouped following the FAO classification. The values represent mean ± standard deviation.
FAO GroupThis StudyHuelsemann et al. (2015) [41]
δ15N (‰)nδ15N (‰)n
Cereal C33.7 ± 2.8673.2 ± 1.3>128
Cereal C44.5 ± 2.6473.3 ± 1.957
Fruit4.3 ± 3.0814.6 ± 2.4>891
Nuts17.3 ± 0.8102.7 ± 2.1310
Pulse2.4 ± 1.7381.1 ± 1.8109
Tuber4.6 ± 2.8453.1 ± 3.4134
Vegetable *4.2 ± 3.51503.1 ± 3.0>990
* Amended with mineral nitrogen fertilizer.
Back to TopTop