1 Universidade Federal Do Rio De Janeiro Instituto De

UNIVERSIDADE FEDERAL DO RIO DE JANEIRO INSTITUTO DE QUÍMICA PROGRAMA DE PÓS-GRADUAÇÃO EM CIÊNCIA DE ALIMENTOS

Ana Beatriz Neves Martins

DEVELOPMENT AND STABILITY OF JABUTICABA

(MYRCIARIA JABOTICABA) JUICE OBTAINED BY STEAM EXTRACTION

RIO DE JANEIRO 2018

Ana Beatriz Neves Martins

DEVELOPMENT AND STABILITY OF JABUTICABA

(MYRCIARIA JABOTICABA) JUICE OBTAINED BY STEAM EXTRACTION

Dissertação de Mestrado apresentada ao Programa de Pós-graduação em Ciência de Alimentos do Instituto de Química, da Universidade Federal do Rio de Janeiro como parte dos requisitos necessários à obtenção do título de Mestre em Ciência de Alimentos.

Orientadores: Prof.ª Mariana Costa Monteiro Prof. Daniel Perrone Moreira

RIO DE JANEIRO 2018

Ana Beatriz Neves Martins

DEVELOPMENT AND STABILITY OF JABUTICABA

(MYRCIARIA JABOTICABA) JUICE OBTAINED BY STEAM EXTRACTION

Aprovada por:

______Presidente, Profª. Mariana Costa Monteiro, INJC/UFRJ

______Profª. Maria Lúcia Mendes Lopes, INJC/UFRJ

______Profª. Lourdes Maria Correa Cabral, EMPBRAPA

RIO DE JANEIRO 2018 4

ACKNOLEDGEMENTS

Ninguém passa por essa vida sem alguém pra dividir momentos, sorrisos ou choros. Então, se eu cheguei até aqui, foi porque jamais estive sozinha, e não poderia deixar de agradecer aqueles que estiveram comigo, fisicamente ou em pensamento.

Primeiramente gostaria de agradecer aos meus pais, Claudia e Ricardo, por tudo. Pelo amor, pela amizade, pela incansável dedicação, pelos valores passados e por todo esforço pra que eu pudesse ter uma boa educação. A vocês, que colocam os meus sonhos na frente dos seus e batalham e vibram pelas minhas conquistas como se fossem suas, os meus maiores e mais sinceros agradecimentos! Obrigada pelo apoio incondicional! Vocês são a minha base!

Ao Felipe, pelo amor, companheirismo, paciência e compreensão. Por ir ao fundão num sábado (ou dois...) ajudar a higienizar 60 quilos de jabuticaba, por entender meus momentos de ausência e ansiedade. Obrigada por sempre seguir de mãos dadas comigo por esse caminho as vezes esburacado.

Aos meus orientadores que, durante esse tempo, foram mais do que orientadores acadêmicos. Agradeço não só pela confiança, pelos ensinamentos e pelo suporte na elaboração deste trabalho mas também, e principalmente, pelo apoio, incentivo, por acreditarem no meu potencial mais do que eu mesma e por terem sido tão compreensivos em um dos momentos que eu mais precisei. Escrevo esses agradecimentos com a certeza da grande oportunidade que foi ser aluna de vocês e de que não eu poderia ter tido orientadores melhores. Vocês são exemplos de profissionais e pessoas e foram fundamentais para essa conquista. A vocês, a minha admiração e o meu muito obrigada!

À Olga, que, fugindo de todo e qualquer estereótipo, foi e vai muito além de sogra. Agradeço por todas as conversas científicas e bate-papos que tivemos e por todo carinho e acolhimento.

À Ellen, uma das melhores pessoas que eu poderia conhecer. Dona de um coração gigante

esteve ao meu lado em todos os momentos que precisei, fossem eles acadêmicos ou pessoais. Sou grata por toda ajuda e conselhos durante esse tempo!

À Kim, que de orientadora de IC virou uma amiga. Obrigada pela paciência, por todas as explicações e pelas muitas vezes que você me ouviu!

À Suellen, um dos grandes presentes que o mestrado me deu. Agradeço pela companhia diária nesse tempo, pelo trabalho compartilhado e pelos muitos conselhos.

À minha aluna de IC, Mariana, que contribuiu com parte da realização deste trabalho. Obrigada pela dedicação e boa vontade!

Aos amigos do LBNA Ana Rafaela, André, Aline, Emília, Fabrício, Genilton, Laís, Nathália Ferrari, Nathália Martins e Vanessa. Cada um teve uma parcela nesse grande feito, sem contar com almoços diários. Obrigada pela ajuda e companhia de sempre! Gostaria de fazer também um agradecimento especial à Desirée e à Talita, que inúmeras vezes salvaram minhas análises gentilmente me cedendo água milli q!

Às professoras e aos alunos do LABAF’s, Lili, Maria Lucia, Denise, Ana, Julia e Luan, que me receberam da melhor maneira possível, como se eu fosse de casa. Obrigada também pelas orientações científicas que me foram dadas durante todo esse tempo de trabalho com vocês! Agradeço especialmente à Iris, à Camila e à Isabelle, companheiras de laboratório que tive a grande oportunidade de conhecer. Vocês tornaram o trabalho mais leve e divertido.

Às amigas Analú, Ju, Carol, Catarina, Dani, Débora, Evelyn, Gabi e Luana e ao amigo Christian, que sempre estiveram na torcida, enviando energias positivas.

Às amigas do intercâmbio Val, Paola e Raquel, que mesmo de longe se fizeram presentes, fosse na ajuda com o inglês ou no apoio e na torcida de sempre.

Agradeço a todos que direta ou indiretamente fizeram parte deste trabalho e dа minha formação.

RESUMO

Martins. Ana Beatriz Neves. Desenvolvimento e avaliação da estabilidade do suco de jabuticaba (Myrciaria Jaboticaba) produzido pelo método de extração por arraste a vapor. Rio de Janeiro, 2018. Dissertação (Mestrado em Ciência de Alimentos) – Instituto de Química, Universidade Federal do Rio de Janeiro, Rio de Janeiro, 2018.

Nativa das regiões central, sudeste e sul do Brasil, a jabuticaba (Myrciaria Jaboticaba) é uma berry globosa de cor escura, com crescimento natural em climas subtropicais. Sua polpa é esbranquiçada, suculenta e gelatinosa, com sabor doce e levemente ácido. A fruta apresenta alto teor de compostos fenólicos, principalmente antocianinas, sendo também considerada uma importante fonte dietética de nutrientes. Apesar de seus desejáveis atributos sensoriais, teor de compostos bioativos e perfil nutricional, é altamente perecível, o que limita sua comercialização e consumo. A técnica de arraste a vapor consiste na extração do suco pela da lixiviação da polpa da fruta pelo vapor d'água. Utilizado no processamento de suco de frutas, o método permite a obtenção de um produto microbiologicamente seguro e que preserva as características nutricionais e sensoriais da fruta. Assim, no presente estudo o suco de jabuticaba foi produzido pela técnica em questão. A reprodutibilidade e as condições do processo de produção do suco foram avaliadas e a composição centesimal, os perfis de açúcar, ácidos orgânicos e compostos fenólicos, a atividade antioxidante e as qualidades microbiana e sensorial do suco foram inicialmente caracterizadas. Também foram realizadas comparações quanto aos teores de compostos fenólicos e aceitação sensorial de dois sucos de jabuticaba produzidos com e sem adição de sacarose. Além disso, avaliou-se o efeito do armazenamento a longo prazo a 25 ºC sobre a composição química e as qualidades microbiana e sensorial do suco de jabuticaba, bem como o efeito do armazenamento em condições aceleradas (40, 50 e 60 ºC) sobre a composição química. Os parâmetros cinéticos de degradação de cianindina-3-O- glicosídeo (C3G) e delfinidina-3-O-glicosídeo (D3G) e formação de ácido gálico foram também determinados. A extração do suco de jabuticaba por arraste a vapor se mostrou reprodutível e 30 min foi o melhor tempo de extração de acordo com os parâmetros pré- estabelecidos. O suco foi composto principalmente por água e carboidratos. O principal composto fenólico encontrado no suco foi a C3G (40%), enquanto a frutose e a glicose foram os principais açúcares (93%) e o ácido cítrico, o principal ácido orgânico (91%). O

suco também apresentou boas qualidade microbiológica e aceitação sensorial. Exceto para o ácido elágico, que foi 1,2 vezes menor no suco com adição de sacarose em comparação ao suco sem adição de sacarose, e da quercetina, com conteúdo ligeiramente superior no suco com adição de sacarose, o perfil de compostos fenólicos bem como o conteúdo total de compostos fenólicos foram semelhantes entres os sucos. Os valores de atividade antioxidante por FRAP, TEAC e Folin-Ciocalteu não foram influenciados pela adição de sacarose. O suco com adição de sacarose, por sua vez, apresentou melhor aceitação sensorial. Após 112 dias a 25ºC, os teores de açúcares e ácidos orgânicos permaneceram estáveis, enquanto os conteúdos de C3G e D3G foram quase completamente degradados e o conteúdo de ácido elágico e o conteúdo total compostos fenólicos foram reduzidos pela metade. A degradação de ambas as antocianinas do suco de jabuticaba seguiu uma reação de primeira ordem, enquanto a formação de ácido gálico seguiu uma reação de ordem zero. Os valores de atividade antioxidante por FRAP e TEAC mostraram uma diminuição significativa. A cor do suco sofreu um grande impacto, como resultado da degradação das antocianinas. Ainda, o suco permaneceu microbiologicamente seguro durante o período de armazenamento e, com exceção do atributo de cor, todos os demais atributos sensoriais não foram modificados. Os resultados indicam que o suco de jabuticaba extraído a vapor apresentou potencial de comercialização e pode ser uma forma de contribuir para a valorização da fruta.

Palavras-chave: Antocianinas; cinética química de compostos fenólicos; açúcares; ácidos orgânicos; vida de prateleira; valorização da fruta

ABSTRACT

Martins. Ana Beatriz Neves. Development and stability of jabuticaba (Myrciaria Jaboticaba) juice obtained by steam extraction. Rio de Janeiro, 2018. Dissertação (Mestrado em Ciência de Alimentos) – Instituto de Química, Universidade Federal do Rio de Janeiro, Rio de Janeiro, 2018.

Jabuticaba (Myrciaria Jaboticaba) is a dark-colored globose berry native to the Brazilian central, southeast and south regions and grows naturally in subtropical climates. Its pulp is whitish, juicy and gelatinous, with sweet and slightly acid flavor. Fruit presents high phenolic compounds content, particularly anthocyanins, and can also be considered an important dietary source of nutrients. Despite its desirable sensory attributes, bioactive compounds contents and nutritional profile, fruit is highly perishable, showing restricted commercialization and consumption. The steam extraction is a method based on raising water vapor, which reaches the fruit, transferring heat and leaching out the pulp. Used in fruit juice processing, it is reported to produce a product with good nutrients, color and flavor retention and long-term microbial safe. So, in the present study, jabuticaba juice was produced by steam extraction. The reproducibility and conditions of juice production process were evaluated and juice proximate composition, sugars, organic acids and phenolic compounds profiles, antioxidant activity, microbial and sensory qualities were initially characterized. Comparisons regarding phenolic compounds contents and sensory acceptance of two jabuticaba juices, produced with and with no added sucrose, were also carried out. Furthermore, the effect of long-term storage at 25 ºC on chemical composition, microbial and sensory qualities of steam extracted jabuticaba juice, as well as the effect of accelerated storage conditions (40, 50 and 60 ºC) on chemical composition were evaluated. The kinetic parameters of cyanindin-3-O-glucoside (C3G) and delphinidin-3-O-glucoside (D3G) degradation and gallic acid formation. Jabuticaba juice steam extraction was shown to be reproducible and 30 min was the best extraction time according to the pre-established parameters. Juice was mainly composed by water and carbohydrate. C3G was the major phenolic compound (40%), while fructose and glucose were the main sugars (93%) and citric acid, the main organic acid (91%). Juice was also microbial safe and sensory well- accepted. Except for ellagic acid, which was 1.2 times lower in juice with added sucrose in comparison to juice with no added sucrose, and quercetin, which was slightly higher in

juice with added sucrose, the phenolic compounds profile and its total content were similar between juices. The addition of sucrose did not have influence on FRAP, TEAC and Folin- Ciocalteu values and juice with added sucrose was better sensory accepted. After 112 days at 25 ºC, sugar and organic acid contents remained stable, whereas C3G and D3G were almost completely degraded. Ellagic acid and total polyphenol compounds contents were reduced by half. Degradation of both anthocyanins from jabuticaba juice followed a first- order reaction while formation of gallic acid followed a zero-order reaction. Antioxidant activity values by FRAP and TEAC showed a significant decrease. Juice color suffered a great impact, as a result of anthocyanins degradation. It remained microbiologically stable during the storage period and, witch exception of color attribute, all the others sensory attributes were not modified. This results indicates that steam extracted jabuticaba juice showed a potential of commercialization and could be a way to aid fruit valorization.

Keywords: Anthocyanins; phenolic compounds chemical kinetic; sugars; organic acids; shelf life; fruit valorization

LIST OF FIGURES

Figure 1. Jabuticaba tree ..………………………………………………………………... 25 Figure 2. Jabuticaba fruit 25 Figure 3. Chemical structures of the main organic acids ………………………...……… 27 Figure 4. Phenolic compounds basic structure …………………………………………... 28 Figure 5. General chemical structures of the major flavonoids subclasses ……………… 30 Figure 6. Chemical structures of the major anthocyanidins ……………………………... 31 Figure 7. Chemical structures of anthocyanins reported in jabuticaba ………………….. 32 Figure 8. General chemical structures of phenolic acids ………………………………… 33 Figure 9. Products from the hydrolysis of hydrolysable tannins ………………………… 34 Figure 10. Chemical structures of a gallotannin and of ellagitannins presented in jabuticaba fruit …………………………………………………………………………… 34 Figure 11. Depsides reported from jabuticaba …………………………………………... 36 Figure 12. Illustrative scheme of juice steam extractor ………………………………….. 40 Figure 13. Study design ………………………...... 48 Figure 14. Phenolic compounds content of steam extracted jabuticaba juices with no added sucrose and with added sucrose …………………………………………………… 66 Figure 15. Antioxidant activity values by FRAP, TEAC and Folin-Ciocalteu of steam extracted jabuticaba juices with no added sucrose and with added sucrose ……………... 67 Figure 16. Sensory acceptance and purchase intent of steam extracted jabuticaba juices with no added sucrose and with added sucrose …………………………………………... 68 Figure 17. Fructose, glucose and sucrose contents of steam extracted jabuticaba juices with no added sucrose over the storage time at 25 ºC ……………………………………. 71 Figure 18. Citric, malic, tartaric and oxalic acids contents of steam extracted jabuticaba juices with no added sucrose over the storage time at 25 ºC …………………………….. 72 Figure 19. Delphinidin-3-O-glucoside and cyanidin-3-O-glucoside contents of steam extracted jabuticaba juices with no added sucrose over the storage time at 25 ºC, 40 ºC, 50 ºC and 60 ºC …………………………………………………………………………... 75 Figure 20. Arrhenius plots for degradation of delphinidin-3-O-glucoside and cyanidin- 3-O-glucoside in steam extracted jabuticaba juice with no added sucrose ………………. 77 Figure 21. Total color difference (ΔE) of steam extracted jabuticaba juice with no added sucrose over the storage time at 25 ºC, 40 ºC, 50 ºC and 60 ºC ………………………….. 80

Figure 22. Gallic acid contents of steam extracted jabuticaba juices with no added sucrose over the storage time at 25 ºC, 40 ºC, 50 ºC and 60 ºC ………………………….. 81 Figure 23. Arrhenius plot for degradation of gallic acid in steam extracted jabuticaba juice with no added sucrose ……………………………………………………………… 82 Figure 24. Ellagic acid contents of steam extracted jabuticaba juice with no added sucrose over the storage time at 25 ºC, 40 ºC, 50 ºC and 60 ºC ………………………….. 83 Figure 25. Total phenolic compounds contents of steam extracted jabuticaba juice with no added sucrose over the storage time at 25 ºC, 40 ºC, 50 ºC and 60 ºC ……………….. 84 Figure 26. Network of degradation/formation of anthocyanins, hydrolysable tannins and related phenolic compounds in the jabuticaba juices during storage …………………….. 86 Figure 27. FRAP, TEAC and Folin-Ciocalteu values of steam extracted jabuticaba juice with no added sucrose over the storage time at 25 ºC temperature ………………………. 88

LIST OF TABLES

Table 1. Yield, anthocyanins, pH, total soluble solids and total color difference values of steam extracted jabuticaba juice ...…………………………………………………….. 59 Table 2. Proximate composition, energy, pH, total soluble solids and titratable acid values of steam extracted jabuticaba juice ...……………………………………………... 60 Table 3. Sugars contents of steam extracted jabuticaba juice with no added sucrose ….... 61 Table 4. Organic acids contents of steam extracted jabuticaba juice with no added sucrose ……………………………………………………………………………………. 62 Table 5. Phenolic compounds contents of steam extracted jabuticaba juice with no added sucrose …………………………………………………………………………….. 63 Table 6. Initial microbiological analysis of steam extracted jabuticaba juice with no added sucrose …………………………………………………………………………….. 64 Table 7. Sensory acceptance and purchase intent of steam extracted jabuticaba juice with no added sucrose ……………………………………………………………………. 65 Table 8. Phenolic compounds contents of steam extracted jabuticaba juice with no added sucrose during storage at 25 ºC …………...………………………………………. 73 Table 9. First-order kinetic model fitting, rate constants and activation energy of anthocyanin degradation in steam extracted jabuticaba juice with no added sucrose …… 76 Table 10. Instrumental color of steam extracted jabuticaba juice with no added sucrose during storage at 25 ºC ……………...……………………………………………………. 79 Table 11. Microbiological analysis of steam extracted jabuticaba juice with no added sucrose during storage ……………………………………………………………………. 89 Table 12. Sensory evaluation of steam extracted jabuticaba juices with no added sucrose and with added sucrose at the beginning (n=118) and the end (n=110) of storage at 25 ºC 91

LIST OF ABREVIATIONS

AA Antioxidant activity AA Ascorbic acid ABTS 2,2’-azino-bis (3-ethylbenzothiazoline-6-sulfonic acid) ANOVA Analysis of variance CADEG Rio de Janeiro`s agricultural trading center CFU Colony-forming units CV Coefficient of variation C3G Cyanidin-3-O-glucoside C3S Cyanidin-3-O-sambubioside DAD Diode-array detector DHAA Dehydroascorbic acid DWB Dry weight basis D3G Delphinidin-3-O-glucoside D3S Delphinidin-3-O-sambubioside ELSD Evaporative light scattering detection FRAP Ferric Reducing Antioxidant Power GA Gallic acid GAE Gallic acid equivalent HAT Hydrogen atom transfer HHDP Hexahydroxydiphenic acid HHP High hydrostatic pressure HPLC High performance liquid chromatography HTST High-temperature short-time LTLT Low-temperature long-time MPN Most probable number SET Single electron transfer TA Titratable acidity TEAC Trolox Equivalent Antioxidant Capacity TPTZ 2,4,6-tri(2-piridyl)-s-triazine TROLOX 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid TSS Total soluble solids

UV Ultraviolet WAS With added sucrose WNAS With no added sucrose

SUMMARY

1 Introduction ...…………………………………………………………………………. 18

2 Justification ………………………………………..………………………………….. 21

3 Literature Review …………………………………………………………………….. 23 3.1 Jabuticaba (Myrciaria spp.) ...……………………………………………...... 24 3.1.1 General characteristics ……………………………………………………………... 24 3.1.2 Chemical composition ……………………………………………………………… 26 3.1.3 Technological applicability ………………………………………………………… 36 3.2 Fruit juice processing ………………………………………………………………… 37 3.2.1 Production and preservation technologies ………………………………………….. 38 3.3 Food products development ………………………………………………………….. 42

4 Objectives ……………………………………………………………………………… 44 4.1 General objective …………………………………………………………………….. 45 4.2 Specific objectives …………………………………………………………………… 45

5 Material and methods ………………………………………………………………… 46 5.1 Juice processing ……………………………………………………………………… 47 5.2 Juice production reproducibility and processing conditions …………………………. 49 5.3 Stability study ………………………………………………………………………… 49 5.4 Proximate composition, pH, titratable acidity and total soluble solids ………………. 49 5.5 Sugar analysis by HPLC-ELSD ……………………………………………………… 50 5.6 Organic acids analysis by HPLC-DAD ………………………………………………. 50 5.7 Phenolic compounds analysis by HPLC-DAD ………………………………………. 51 5.8 Phenolic compounds degradation and formation kinetics ……………………………. 52 5.9 Antioxidant activity by spectrophotometric methods ………………………………... 53 5.10 Instrumental color …………………………………………………………………... 54 5.11 Microbiological analysis ……………………………………………………………. 55 5.12 Sensory analysis …………………………………………………………………….. 55 5.13 Statistical analyses ………………………………………………………………….. 56

6 Results and discussion ………………………………………………………………… 57 6.1 Reproducibility and conditions of the process of production of jabuticaba juices by steam extraction ………………………………………………………………………….. 58 6.2 Characterization of steam extracted jabuticaba juices ……………………………….. 60 6.3 Effect of storage on chemical composition, microbial and sensory qualities of steam extracted jabuticaba juice ………………………………………………………………… 68 6.3.1 Chemical stability: sugars, organic acids, phenolic compounds, color and antioxidant activity stability ..……………………………………………………………. 70 6.3.2 Microbial and sensory qualities ……………………………………………………. 89

7 Conclusion ……………………………………………………………………………... 92

8 References ……………………………………………………………………………... 94

INTRODUCTION

Jabuticaba is a spherical, purple to black colored fruit from the family Myrtaceae and the genus Myrciaria, native to the Brazilian central, southeast and south regions, growing naturally in subtropical climates. The most cultivated and consumed species is M. jaboticaba (Vert) O. Berg, known as jabuticaba sabará (DUARTE AND PAULL, 2015; SALOMÃO et al., 2018; TEIXEIRA et al., 2011). Jabuticaba is highly appreciated due to its whitish, juicy and gelatinous pulp with sweet and slightly acid flavor. Additionally to its desirable sensory characteristics, it not only shows a valuable nutritional profile but also contains considerable contents of bioactive compounds, such as phenolic compounds, which show health properties mainly due to their antioxidant, anti-inflammatory, antimicrobial, anti-mutagenic and anti-proliferative activities. Its phenolic compounds include flavonoids, like anthocyanins and flavonols, phenolic acids and tannins, and depsides as minor constituents (ALEZANDRO et al., 2013; WU, LONG, KENNELLY, 2013). Despite its rich nutritional and functional composition, jabuticaba is not extensively commercialized, being popularly consumed fresh once it possesses a highly perishable nature, easily spoiling. Thus, the fresh fruit has been used to manufacture artisanal products like jams, vinegars, juices and different types of alcoholic beverages such as liquors and “wines”, and others. Jabuticaba based-products represent a way to avoid changes in the quality of the fruit during storage and to extend its shelf life (DUARTE AND PAULL, 2015; SALOMÃO et al., 2018; TEIXEIRA et al., 2011, WU, LONG, KENNELLY, 2013). Over the last few years, consumers have been increasingly looking for healthy, natural food products, also convenient to consume, avoiding those containing artificial flavors, colors and preservatives. In order to meet this new preference, food industry has been focusing on product development and differentiation, using innovative technologies for processing and packaging, as well as incorporating healthier ingredients. Because of that, the global juice market has been in expansion and novel fruit drinks, with functional properties and exotic flavors, have become available (PRIYADARSHINI AND PRIYADARSHINI, 2018; VIDIGAL et al., 2011). Different methods are used on fruit juice production and, usually, there is a need to apply heat treatments post-production to extend shelf life. Steam extraction has been used for small- and medium-scale red grape juice production. It is reported to inactivate enzymes and pasteurize juice during its extraction. As a rapid and low-cost method, it adds

value to raw materials, being of great economic importance to small producers (BATES, MORRIS, CRANDAL, 2001; LOPES et al., 2016). The development of new food products relies on the application of knowledge from different areas. Defining product specifications is just as important as evaluating it after processing and the desired storage period. Thus, food chemistry, processing, packaging, quality assurance, as well as consumer acceptance, that is, its commercialization, show great relevance for this process. (SIRÓ et al., 2008; WINGER, 2006).

JUSTIFICATION

Given this context, the development of jabuticaba juice by steam extraction, an alternative production technique, meets consumers new demand and also contributes to the fruit valorization, while the evaluation of its chemical, microbial and sensorial stability during storage is important to assure the quality of the final product. Moreover, it highlights the social-economic potential of juice production by small producers, as an important source of income.

LITERATURE REVIEW

3.1 Jabuticaba (Myrciaria spp.)

3.1.1 General characteristics

The jabuticaba tree belongs to the family Myrtaceae and the genus Myrciaria. Several species have been described in the literature, however, the total number of existing species is not yet well established and there are controversial information concerning its classification (DUARTE AND PAULL, 2015; PEREIRA et al., 2005; SALOMÃO et al., 2018; TEIXEIRA et al., 2011). Jabuticaba species are native to the Brazilian central, southeast and south regions and grows naturally in subtropical climates. Although mostly cultivated in the states of Minas Gerais, Espírito Santo, Rio de Janeiro and São Paulo, trees from this genus can be also found in other states of the country and in other countries, such as Paraguay and Argentina (DUARTE AND PAULL, 2015; SALOMÃO et al., 2018; TEIXEIRA et al., 2011). Among the species already related, the most common are M. cauliflora (Mart) O. Berg, known as jabuticaba paulista, ponhema or assú, and M. jaboticaba (Vert) O. Berg, known as jabuticaba sabará. These two species are largely found in Brazil and, despite the fact that both produce edible fruits with interesting characteristics for fresh consumption or industry applicability, M. jaboticaba is the most cultivated and appreciated one (DUARTE AND PAULL, 2015; LEITE-LEGATTI et al., 2015; SALOMÃO et al., 2018; TEIXEIRA et al., 2011). Jabuticaba trees (Figure 1) from different species show great variability in its characteristics, including size, ideal climatic conditions, fruiting season. In general, trees are medium sizes, branched, and their trunk has a thin outer bark. Flowers grow directly from the trunk and branches and tress normally flower and fruit once a year, with high productivity. Trees from M. jaboticaba specie are 6 to 9 meters tall and normally flower and fruit between the months of August and November (end of winter and beginning of spring), with its peak in September (DUARTE AND PAULL, 2015; SALOMÃO et al., 2018; TEIXEIRA et al., 2011; WU, LONG, KENNELLY, 2013).

Figure 1. Jabuticaba tree.

Fruits physical characteristics may vary according to the species. Therefore, the term jabuticaba refers to the edible fruit of all Myrcyaria species already described (SALOMÃO et al., 2018). The fruit (Figure 2) is a globose berry, generally up to 3 centimeters in diameter and the number of seeds per fruit can range from 0 to 4. Its peel is thin, fragile and black colored and its pulp, whitish, juicy and gelatinous, with sweet and slightly acid taste when mature. It is a non-climacteric fruit (the ripening process does not take place once the fruit is removed from the tree), so its harvest must be done at fruits full ripening stage, which is usually reached in 40 up to 60 days from flowering, when the peel color becomes dark-purple or black (ALEZANDRO et al., 2013; DUARTE AND PAULL, 2015; SALOMÃO et al., 2018; TEIXEIRA et al., 2011; WU, LONG, KENNELLY, 2013).

Figure 2. Jabuticaba fruit.

3.1.2 Chemical composition

Jabuticaba fruit can be considered an important dietary source of nutrients. It contains high moisture, carbohydrate and fiber contents, low lipids and proteins contents, and different contents of vitamins and minerals. However, varied nutritional composition values may be found in the literature. Carbohydrate is the major nutritional component. Its fraction is mainly composed by the monosaccharides fructose and glucose, but also presents low contents of the disaccharide sucrose. Lipid and protein are found in low contents, ranging from 0.5% to 2% and from 1% to 5%, respectively. Fruit contains from 10% to 40% fiber and around 85% moisture and 3% ash. Several minerals have been quantified in jabuticaba fruit, being potassium the one found in the highest concentration (1180 mg to 700 mg/100 g dry weight basis). Other minerals such as magnesium (72 mg to 100 mg/100 g dwb) and phosphorus (75 mg to 100 mg/100 g dwb) are also found in high contents, while copper, manganese, iron, zinc, calcium, selenium, sulfur, boron and cobalt are found in lower contents. (ALEZANDRO et al., 2013; GURAK et al., 2014; INADA et al., 2015; LIMA et al, 2010). Moreover, INADA et al. (2015) have reported an iron content in jabuticaba higher than in those vegetable foods well known as iron-rich and that the fruit can be considered a source of copper and manganese. In addition to its nutritional value, jabuticaba fruit presents relative amounts of non-nutrients compounds such as organic acids and phenolic compounds. Organic acids are organic compounds with acidic properties. The most common ones are carboxylic acids and its acidity is due to its carboxylic group (-COOH). A number of different organic acids can be found in various fruits (Figure 3), especially citric, malic and ascorbic acids. They are weak acids and exist in their acid or salt forms. Commonly present in greater contents in citrus fruits, citric acid is also the main organic acid in berries (SOUZA et al., 2014, MIKULIC-PETKOVSEK et al., 2012). In certain fruits, oxalic, succinic and/or tartaric acids can be also found, being the last one the main organic acid in grapes (FORD, 2012). Fruit organic acid content is influenced by fruit maturity stage and tends to be higher in immature fruits, generally declining with ripening. Despite the already mentioned variety, most of them are usually found in low content. These compounds have influence on fruit’s pH and acidity and, therefore, a directly contribution to fruit’s flavor (sourness), suitability for processing (jams, for example) and preservation. Besides naturally present in fruits, some of them, such as citric, malic and ascorbic acids,

are largely used by the food industry as food additives, preservatives, acidulants and flavoring agents, in beverages, reducing the pH and so inhibiting microorganisms growth and enhancing flavors (COULTATE, 2009; QUITMANN, FAN, CZERMAK, 2014; SCHERER et al., 2012; WALKER AND FAMIANI, 2018). Citric, succinic, malic, oxalic and acetic acids have already been identified in jabuticaba. According to LIMA et al. (2010), citric acid was the major organic acid present in the fruit, followed by succinic and malic acids, while oxalic and acetic acids were found in low concentrations. However, JHAM et al. (2007) reported a different profile. According to the authors, only three organic acids were identified. Succinic acid was the most abundant, followed by citric acid, while malic acid was found only in trace amounts.

Citric acid Malic acid Ascorbic acid

Oxalic acid Tartaric acid Succinic acid

Figure 3. Chemical structures of the main organic acids (Chemspider).

Phenolic compounds are products of the secondary metabolism of plants. Synthetized by specialized cells, they aid beyond plants nutrition, being important to its survival. These compounds are involved in plants response to biotic and abiotic stress in the environment. They are responsible for plant defense, providing protection against insects, herbivores and pathogens like virus, bacteria and fungus, as well as ultraviolet solar radiation and oxidative stress. Phenolic compounds also play an important role as pigments, assisting plant pollination and fertilization by attracting pollinators and seed

dispersers (SALTVEIT, 2018; SARKAR AND SHETTY, 2014; VERMERRIS AND NICHOLSON, 2006). Besides their impacts on plant defense response, phenolic compounds show biological effects on human health, contributing to prevention and risk reduction of several chronic diseases. Their health properties are mainly related to their antioxidant, anti- inflammatory, antimicrobial, anti-mutagenic and anti-proliferative activities. Thus, the beneficial effects from fruit and vegetable intake, moreover its nutritional value, are linked to the presence of those compounds (SARKAR AND SHETTY, 2014; SAUCEDA, 2018; WU, LONG, KENNELLY, 2013). Structurally, phenolic compounds contain an aromatic ring with one or more hydroxyl groups attached directly to it (Figure 4). This hydroxylated six-carbon benzene ring structure is the basis of all phenolic compounds. Despite the phenolic structure in common, phenolic compounds can differ significantly in chemical structure according to the number of phenol rings (or carbon atoms) and the structural elements bound to the rings. They may be present as simple (a single ring structure with a substitution) or larger and complex molecules (highly polymerized) and also in their conjugated form with one or more sugar residues (mono, di or oligosaccharides linked to hydroxyl groups) and organic acids (ANDRÉS-LACUEVA et al., 2010; MANACH et al., 2004; SALTVEIT, 2018; SARKAR AND SHETTY, 2014; SAUCEDA, 2018; VERMERRIS AND NICHOLSON, 2006).

Figure 4. Phenolic compounds basic structure (VERMERRIS AND NICHOLSON, 2006).

Due to their variety, phenolic compounds can be classified in different ways, based on criteria like its origin, biological function or, more commonly, chemical structure, which includes several classes (or groups) and subclasses. Their classification according to

the chemical structure is based on the number of phenol rings that they contain and on the structural elements that bind these rings. In this sense, there are two main groups, flavonoids and non-flavonoids (ANDRÉS-LACUEVA et al., 2010; CROZIER, JAGANATH, CLIFFORD, 2006). Moreover, some authors even divide them into more classes, as flavonoids, phenolic acids, stilbenes and lignans (MANACH et al., 2004) or as flavonoids, phenolic acids, stilbenes, lignans, courmarins and tannins (SHAHID AND ZHONG, 2015). The occurrence of phenolic compounds in foods is variable; many classes can be widely found, yet, some of them are only present in specific types of foods, for example isoflavones in legumes (especially soy beans and soy products). Flavonoids class is the most common distributed one. Additionally, numerous phenolic compounds from different classes can be contained in a single food. Fruits, especially berries, vegetables, cereals, herbs and plant-based foods like chocolate, fruit juices, tea, coffee and red wine are recognized important dietary sources of phenolic compounds (BELSCAK- CVITANOVIC et al., 2018; LÓPEZ-NICOLÁS AND GARCÍA-CARMONA, 2010; SARKAR AND SHETTY, 2014). Jabuticaba is a fruit that presents a high content of phenolic compounds, which includes flavonoids, like anthocyanins and flavonols, phenolic acids and tannins, and depsides as minor constituents (WU, LONG, KENNELLY, 2013). The type and quantity of phenolic compounds, however, vary among the fruit fractions (peel, pulp and seed) (ALEZANDRO et al., 2013; GURAK et al., 2014; INADA et al., 2015) and can be also influenced by environmental/climatic factors, plant species/variety, ripening stage, harvesting time, post-harvest management and processing and storage conditions (LÓPEZ- NICOLÁS AND GARCÍA-CARMONA, 2010; SAUCEDA, 2018).

Flavonoids (C6-C3-C6) comprise a class of compounds with a fifteen-carbon general structure, with two phenolic rings (A and B) connected by a three-carbon chain forming a heterocyclic ring with an oxygen atom (C), usually found glycosylated. This class can be divided into several subclasses due to structural variations in the degree of hydroxylation, hydrogenation and glycosylation (including the position and type of sugar moieties) and the presence of methyl groups. The main subclasses (Figure 5) are flavones, flavonols, flavan-3-ols (or flavanols), isoflavones, flavanones and anthocyanidins (ANDRÉS- LACUEVA et al., 2010; BELSCAK-CVITANOVIC et al., 2018; CROZIER, JAGANATH, CLIFFORD, 2006).

Figure 5. General chemical structures of the major flavonoids subclasses (CROZIER, JAGANATH, CLIFFORD, 2006).

Anthocyanidins are natural water-soluble pigments, responsible for red, blue and purple colors of many plant tissues. Several compounds have already been identified, but only six are commonly found in the nature: cyanidin, peonidin, pelargonidin, petunidin, malvidin and delphinidin (Figure 6), being cyanidin the most widespread. These compounds differ from each other as they may present hydroxyl (-OH) and/or methoxyl

(O-CH3) groups in their chemical structure. However, as highly instable and susceptible to degradation, anthocyanidins (aglycone form) are found glycosylated, known as anthocyanins. Glucose and rhamnose are the most usual sugars, but others like rutinose, xylose, arabinose and fructose are still frequently found. The sugar moieties may also be acylated by different organic (aliphatic) and aromatic acids. Thus, depending on the nature, the number and the position of glycosylation and the acyl substituents, as well as the hydroxylation and methoxylation patterns, a huge variety of anthocyanins is found. (CASTAÑEDA-OVANDO et al., 2009; CROZIER, JAGANATH, CLIFFORD, 2006; HE AND GIUSTI, 2010; KHOO et al., 2017; PRIOR AND WU, 2012). Two anthocyanins 30

have been identified in jabuticaba fruit, cyanidin-3-O-glucoside (C3G) and delphinidin-3- O-glucoside (D3G) (Figure 7). Both compounds are found in high contents, being C3G the major phenolic compound in the fruit (INADA et al., 2015, PLAZA et al., 2016; WU et al., 2012). Anthocyanin content varies between fruit ripening stages, increasing significantly through maturation (ABE, LAJOLO, GENOVESE, 2012; ALEZANDRO et al., 2013). These compounds are mostly present in the peel, being responsible for its characteristic dark-purple color (GURAK et al., 2014; INADA et al., 2015).

Figure 6. Chemical structures of the major anthocyanidins (CROZIER, JAGANATH, CLIFFORD, 2006).

Figure 7. Anthocyanins reported in jabuticaba (Adapted from WU, LONG, KENNELLY, 2013).

Flavonols are the most common flavonoids in food, being quercetin and kaempferol the main representatives (MANACH et al., 2004). In jabuticaba fruit, different flavonols have been reported. WU et al. (2012) reported the presence of five compounds, quercetin, myricitrin, quercitrin, quercimeritrin and isoquercetin while INADA et al. (2015) identified four compounds, rutin, myricetin, myricitrin and quercetin. On the other hand, according to PLAZA et al. (2016), only quercitrin could be identified. This dissimilarity could be explained by environmental conditions or even by fruit species, as described by ALEZANDRO et al. (2013), who found greater contents of quercetin derivates in Paulista fruits. Phenolic acids are compounds characterized by an aromatic ring with a carboxylic group and one or more hydroxyl (-OH) and/or methoxyl (O-CH3) groups. There are two main types, the derivatives of benzoic acids (or hydroxybenzoic acids), with a seven- carbon (C6-C1) framework, and the derivatives of cinnamic acid (or hydroxycinnamic acids), with a nine-carbon (C6-C3) framework (Figure 8). The degree and the position of the hydroxylation and methoxylation in the aromatic ring create the variety of compounds within each group. Moreover, hydroxybenzoic acids are components of complex structures such as hydrolysable tannins (BELASCAK-CVITANOVIC et al., 2018; SARKAR AND SHETTY, 2014; SHAHIDI AND ZHONG, 2015). Three hydroxybenzoic acids, gallic, protocatechuic (or 3,4-dihydroxibenzoic) and ellagic acids, and two hydroxycinnamic acids, m-coumaric and trans-cinnamic acids have been identified in jabuticaba fruit. According to INADA et al. (2015), gallic and ellagic acids (Figure 9) were found in higher contents than the other phenolic acids reported, mostly present in peel and seed fruit

fractions. Additionally, ABE, LAJOLO, GENOVESE (2012) showed that jabuticaba is one of the main sources of ellagic acid among fruits consumed by the Brazilian population.

Figure 8. General chemical structures of phenolic acids (SHAHIDI AND ZHONG, 2015).

Tannins can be classified in three groups, condensed tannins, complex tannins and hydrolysable tannins. Condensed tannins are oligomers and polymers of flavan-3-ols, also known as proanthocyanidins. Complex tannins comprise compounds in which a monomer of flavan-3-ols (catechin) is linked to a hydrolysable tannin unit through a glycosidic bond. Hydrolysable tannins are highly polymerized compounds, with gallic acid (Figure 9) units and hexahydroxydiphenic acid (HHDP) (Figure 9) units esterified to a polyol core (mainly glucose). There are two types of hydrolysable tannins, gallotannins and ellagitannins (Figure 10), depending on the acid component. Gallotannins are simpler molecules, formed by gallic acid units only, whereas ellagitannins are mostly formed by both gallic acid and hexahydroxydiphenic acid (linkage between nearby galloyl groups in the gallotannin molecule) units. The hydrolysis of gallotannins releases gallic acid molecules, while ellagic acid molecules are product from the ellagitannins hydrolysis (after a spontaneous rearrangement of hexahydroxydiphenic acid) (Figure 10) (ANDRÉS- LACUEVA et al., 2010; CROZIER, JAGANATH, CLIFFORD, 2006; HAGERMAN, 2002; PLAZA et al., 2016; VERMERRIS AND NICHOLSON, 2006).

Figure 9. Products from the hydrolysis of hydrolysable tannins (HAGERMAN, 2002).

A B

Figure 10. Chemical structure of a gallotannin (A) and of ellagitannins (B) presented in jabuticaba fruit (Adapted from HAGERMAN, 2002).

The jabuticaba fruits have been reported to contain gallotannins and ellagitannins, compounds which may be associated with fruit astringency. WU et al. (2012) identified seven ellagitannins, HHDP-galloylglucose, casuariin (di-HHDP-glucose isomer), pedunculagin (di-HHDP-glucose isomer), casuarinin (di-HHDP-galloylglucose),

tellimagrandin I (HHDP-digalloylglucose) and tellimagrandin II (HHDP-trigalloylglucose) and casuarictin (di-HHDP-galloylglucose isomer), which, except for tellimagrandin I, were detected for the first time in the fruit. PLAZA et al. (2016) found the same compounds in M. jaboticaba peel. As noted by the authors, casuarinin and casuaricitin were the main ellagitannins among the detected ones, pedunculagin and casuariin were also found in high concentrations and pentagalloyl hexose was the only gallotannin found. Still according to them, these hydrolysable tannins, besides not being the phenolic group found in higher concentration, were the most contributors to the total antioxidant activity of M. jaboticaba. PEREIRA et al. (2017), in their study, isolated eight hydrolysable tannins from jabuticaba, which included cauliflorin, alnusiin, pedunculagin, strictinin, casuarictin, 1,2,3,4,6-penta- O-galloyl-β-D-glucose, castalagin and vescalagin. According to the authors, it was the first time that the presence of the ellagitannin cauliflorin, was characterized as 4,6-O-tergalloyl- D-glucose, was reported in the literature. Fruit at full-ripe stage has shown great contents of castalagin and vescalagin, with castalagin levels higher than vescalagin ones. The authors also showed that tannins content decreased over the fruit development and maturation. A different ellagitannin profile, however, was reported by ALEZANDRO et al. (2013), at odds with the compounds reported by WU et al. (2012), PLAZA et al. (2016) and PEREIRA et al. (2017). The authors identified sanguiin H-10 isomers, sanguiin H-6 and lambertianin C. Depsides are phenolic compounds composed of two or more monocyclic aromatic units linked by an ester bond, most often found in lichens. RENEYTERSON et al. (2006) reported, for the first time, the presence of two depsides, 2-O-(3,4-dihydroxybenzoyl)- 2,4,6-trihydroxy-phenylacetic acid and jaboticabin (Figure 11), in jabuticaba fruits. Then, WU et al. (2012) identified these two depsides as minor constituents (1.26%) in the fruit extract, could not being detected in jabuticaba commercial juices.

Figure 11. Depsides reported from jabuticaba (WU, LONG, KENNELLY, 2013).

3.1.3 Technological applicability

The jabuticaba fruit harvest is still regional and mostly domestic, done by family farmers, with few commercial plantations in the country, what makes difficult to find information from official agencies regarding its production and commercialization. Additionally, as a result of fruit fragility and high perishability, its commercialization is limited to the areas of cultivation and surroundings (SALOMÃO et al., 2018; TEIXEIRA et al., 2011). Due to its sensory attributes, jabuticaba is a popular and highly appreciated fruit, presenting a great marketing potential. However, it is normally consumed fresh since it spoils easily. Fruit physical and sensory characteristics may rapidly undergo modifications during its postharvest period. At 25 ºC, its quality during storage is affected by water loss, microbial deterioration, pulp fermentation and susceptibility to mechanical injury, which cause undesirable changes in fruit appearance, texture and flavor and reduce its shelf life up to three days. Thus, in order to avoid changes in the quality of the fruit during storage and to extend its shelf life, jabuticaba has been used to manufacture products like jams, vinegar, juices and different types of alcoholic beverages such as liquor and wine and others, mostly artisanal (DUARTE AND PAULL, 2015; SALOMÃO et al., 2018; TEIXEIRA et al., 2011, WU, LONG, KENNELLY, 2013). Owing to fruit high perishable nature, the development of jabuticaba-based 36

products is an alternative to prevent post harvest losses, as well as a way to aid fruit valorization, stimulating its production and consumption. Jabuticaba rich nutritional value and phytochemical composition and pleasant flavor make the fruit favorable for industrial production and commercialization. However, the availability of jabuticaba-based products in the market, although increasing, is still low. Recently, famous and smaller brands have launched products. Heartbrand®,Vigor®, Queensberry® and Myberries® are brands that have included jabuticaba products in their catalogues, an ice pop with fruit juice and peel pieces, a flavored Greek yogurt, a fruit jam and a nectar, respectively. Besides those marketed on a large scale, numerous jabuticaba- based products have been being developed in smaller scales, such as craft products or, yet, in the academic research field, such as cereal bars, cookies, breakfast cereals, gluten-free pasta and muffins, all produced using jabuticaba peel flour (APPELT et al., 2015; GARCIA et al., 2016; MICHELETTI et al., 2018; OLIVEIRA, ALENCAR, STEEL, 2018; ZAGO et al., 2015). These studies showed that the flour is a potential ingredient and could be used in the formulation of these kinds of product without negatively interfering in their physical, sensory and technological characteristics. Moreover, according to GURAK et al. (2014), jabuticaba pomace powder obtained as a co-product of juice extraction could be considered a functional ingredient as a source of bioactive compounds and fiber. Another possible fruit applicability is as natural coloring. According to BALDIN et al. (2016), the addition of microencapsulated jabuticaba extract to fresh sausages as a natural pigment ingredient could be an alternative to colorants, also with antioxidant and antimicrobial activities. INADA et al. (2018a) produced jabuticaba juice by pulping, which was subjected to high hydrostatic pressure processing and showed an improved bioactive potential, by increasing phenolic compound content and antioxidant activity. INADA et al. (2018b) also produced jabuticaba juice by steam extraction and reported that the profile of phenolic compounds was similar to that previously reported for the fruits, indicating that the phenolic compounds found in fresh fruit were also present in the juice.

3.2. Fruit juice processing

Food processing is the set of physical, chemical and/or biological operations that result in the modification of foods from its raw/fresh state into products. It includes production, preservation and packaging techniques and its main goal is to extend food

products shelf life by inhibiting microbial proliferation and deterioration, preventing undesirable changes. Food processing, thus, aids food distribution and commercialization, ensuring adequate availability of products, and is also an alternative for year-round consumption of seasonal foods. It represents a way to increase the convenience and variety of products, making them more attractive and easy to consume (requiring low or no preparation efforts) (FELLOWS, 2017; TOLA AND RAMASWAMY, 2012). Benefits associated to food processing are preservation, safety, quality, availability, sustainability, convenience and health and wellness (FLOROS et al., 2010). Moreover, processed foods can have a direct contribution to nutrients dietary intake (WEAVER et al., 2014). However, processing and storage conditions might have influence on final product quality, deserving attention in order to best preserve fresh food properties. The fruit juice processing has a great importance for both food industry and consumers. It not only adds economic value to fruits through the technological input to the final product but also contributes to fruit valorization and provides a great variety of products, with different flavors and types. Even fruits cultivated in particular regions are being commercialized and consumed throughout the country as fruit juices. As a result, there has been an expansion in the fruit juice market and the beverage has been gaining popularity among the consumers (CASWELL, 2009; SILVA AND ABUD, 2017).

3.2.1 Production and preservation technologies

Juice is the liquid extracted from fruit tissues. Depending on fruit texture and type of juice desired, numerous extraction methods can be employed in order to separate the liquid part from the solid fibrous material of the fruit. In general, fruit juice extraction is conducted by mechanical processing, such as pressing or squeezing (MUSHTAQ, 2018; TAYLOR, 2016). In some cases, further processing, for example, preservation technologies are required to avoid microbial spoilage and biochemical changes. There are several methods applied by food industry to ensure the preservation of fruit juice products (microbial safety and enzyme deactivation). Recently, besides the conventional thermal processing, nonconventional (or alternative) non-thermal techniques have been proposed for juices. Still, thermal processing remains one of the most important and also most cost- effective treatments (PETRUZZI et al., 2017; TOLA AND RAMASWAMY, 2012). Extraction and post extraction steps are both important in the juice manufacturing process

and might interfere on juice quality (MUSHTAQ, 2018). Fruit juice mechanical production relies on the application of two types of forces, compression and attrition, in solo or combined ways (FELLOWS, 2017). For this reason, processing of juices by mechanical extraction methods also demands the application of methods of conservation. There are different juice extractors, ranging from kitchen to industrial scale, with some of them designed specifically to certain types of fruits. Pressers and pulpers are common equipment found in food industry (FELLOWS, 2017; MUSHTAQ, 2018; TAYLOR, 2016). The steam extraction is an alternative method to the mechanical ones. It is conducted by the raising water vapor, which reaches the fruit, transferring heat and leaching out the pulp. This method is reported to inactivate enzymes and pasteurize without drastically compromising juice nutritional profile, producing a juice with good color and flavor retention and also providing a long-term microbial safety, without requiring any further preservation technology to be applied. It is a simple and low-cost method, with short time performance, widely used in small- and medium-scale juice production, especially for red grape juice (BATES, MORRIS, CRANDAL, 2001; LOPES et al., 2016). A steam extractor, or steam juicer (Figure 12), is composed of three pieces, a water pan, a juice-collecting container and a fruit container, arranged one above the other. The bottom piece is the water pan, placed on a heat source in order to boil the water and, thus, produce the steam for the process. The juice-collecting container has a central conical opening in its bottom, which allows the passage of the steam, and an outlet with a drain, through which the juice is bottled. At the end of the drain there is a clamp, which helps the juice packaging. The top piece, the fruit container, is a basket (and a lid), which the bottom and side surfaces have perforations, allowing the steam to reach the fruit and the juice to be collected, while keeping the fruit out of the extracted juice (ANJOS, 1999).

Figure 12. Illustrative scheme of a juice steam extractor (Source: www.bestofjuicer.com/best-steam-juicer).

As preservation technologies, conventional thermal methods are based on the application of heat in order to eliminate pathogenic microorganism, reduce or eliminate undesired spoilage microorganisms and inactivate enzymes, promoting safety and quality to foods during storage (AGÇAM, AKYILDIZ, DÜNDAR, 2018; NGADI, BAJWA, ALAKALI, 2012). The intensity and duration of heating are the most important factors for ensure treatment efficacy, and different time/temperature combinations can be used, what will depend on the food matrix to which it is applied (PETRUZZI et al., 2017; NGADI, BAJWA, ALAKALI, 2012). To date, pasteurization is the most commonly thermal method used by food industry in the processing of juices and beverages. It is a mild heat treatment, with heating temperatures usually below 100 ºC, and includes, specially, both low- temperature long-time (LTLT) and high-temperature short-time (HTST) techniques. LTLT comprises temperatures around 65 ºC for no less than 30 minutes while HTST, temperatures at 75ºC or above, with the duration time ranging from 15 up to 30 seconds (CHEN, YU, RUPRASINGHE, 2013; FELLOWS, 2017; ORTEGA-RIVAS AND SALMERÓN-OCHOA, 2014).

Among the non-thermal preservation technologies, high hydrostatic pressure (HHP), pulsed electric field, ultrasound, short-wave UV light and pulsed light are the mainly innovative approaches to obtain fruit products with extended shelf life without heat application (GÓMEZ, WELTI-CHANES, ALZAMORA, 2011; ORTEGA-RIVAS AND SALMERÓN-OCHOA, 2014; TOLA AND RAMASWAMY, 2012). HHP consists of subjecting the food to elevated pressures, which is homogeneously and instantaneously distributed throughout the product, resulting in microbial inactivation and enzymatic activity reduction (AUGUSTO et al., 2018; ORTEGA-RIVAS AND SALMERÓN- OCHOA, 2014; TOLA AND RAMASWAMY, 2012). Pulsed electric field technology is based on the application of an externally generated electric field with high voltage pulses for short periods of time across the food product, resulting in microbial inactivation (KOUBAA et al., 2018). Ultrasound technology involves the application of sound waves with high power and low frequencies. The high energy generated by sound leads to the formation of cavities, what causes the disruption of cellular structures (SWAMY, MUTHUKUMARAPPAN, ASOKAPANDIAN 2018). The use UV light in food processing for preservation of fruit juices is related to the short-length UV light, or UV-C light. The short-wave UV light radiation is absorbed by different cellular components, especially by the DNA. The absorbed UV light, thus, damages the microbial DNA by promoting the formation of products that inhibit DNA transcription and replication, what results in cell death. The microorganism becomes unable to proliferate and so inactive, reducing the microbial load (BAYSAL, 2018; GÓMEZ, WELTI-CHANES, ALZAMORA, 2011). Pulsed light involves the use of intense and short duration pulses of light of a broad spectrum of wavelength, ranging from the ultraviolet to the near infrared region. Its mode of action is responsible for the microorganism inactivation and, consequently, food product preservation is the same as the short-wave UV light radiation (GÓMEZ, WELTI- CHANES, ALZAMORA, 2011). Despite the great variety of non-thermal preservation technologies, those are high costly and thus not frequently used (GÓMEZ, WELTI- CHANES, ALZAMORA, 2011; ORTEGA-RIVAS AND SALMERÓN-OCHOA, 2014; TOLA AND RAMASWAMY, 2012).

3.3 Food product development

Developing new food products is a complex process. It requires extensive scientific research and involves the interaction between different areas of knowledge, such as food, nutritional, sensory and consumer sciences and food technology. Raw material and/or ingredients characteristics, technological processes, nutritional, sensorial and microbial qualities of the final product (due to processing and storage conditions), product costs and marketing values and consumers preferences are relevant aspects that should be take into account (BIGLIARDI AND GALATI, 2013; DWYER et al., 2012; RAJAURIA AND TIWARI, 2018, WINGER, 2006). There are important stages in the product development process. Defining the product and its characteristics (composition, size, shape, packaging, stability, storage conditions), the processing technologies and conditions that will be employed and the target market is the first one. Then, as an experimental food product development, it should be conducted in a laboratory scale, evaluating if the product characteristics meets the requirement. The final stage corresponds to a larger scale production and products commercialization (WINGER, 2006). Over the past few years consumers have become more concerned about having a healthy lifestyle. Most of all, their eating habits have changed and there has been increasing awareness about their food choice, mainly regarding to foods content and production practices. Consumers believe that food has a direct contribution to their health and thus, their quality of life, what justifies the present demand for products with convenience, pleasant sensory characteristics and shelf stability while, at the same time, less processed, with high nutritional value, fewer or no additives and health benefits from beyond the nutritional profile (BIGLIARDI AND GALATI, 2013; CARRILLO et al., 2011; FERRAREZI, OLBRICH, MONTEIRO, 2012; ROMANO et al., 2015). However, there is still a negative perception of processed foods, which directly impacts on novel product development and acceptance (DWYER et al., 2012; FLOROS et al., 2010). As a consequence of these changes in consumer preferences, the food industry has been facing technical challenges in food processing, looking for alternatives technologies for production and conservation to develop novel and more attractive processed foods, with good retention of fresh product characteristics such as flavor, color, nutrient and bioactive compounds contents and fresh quality and, at the same time, with no negative

impacts on human health (BIGLIARDI AND GALATI, 2013; SARKAR AND COSTA, 2008; WEAVER et al., 2014). Maintaining products quality throughout food chain is an issue to food industry that goes beyond consumer new demand. It includes technological challenges in response to the influence of different processing and storage conditions on chemical and physical characteristics, nutritional value and safety of the final product (FULLER, 2011; WINGER, 2006). Fruit products may present distinct contents and profiles of nutrients and bioactive compounds from the fruit in its raw state, depending on the processing technology adopted. In general, temperature is the main responsible factor for the compounds degradation during food processing, especially if under severe heating conditions (GANCEL et al., 2011). On the other hand, non-thermal methods have less impact on nutrients and bioactive compounds contents then the thermal ones, although some of them, such as ozone treatment, have been reported to cause losses of phytochemicals contents (RAWSON et al., 2011). Food products quality losses are expected over the time. During long periods of storage, microbial, nutritional value, sensorial characteristics and functional properties are susceptible to changes. Food intrinsic factors (moisture, water activity, pH, composition), temperature and exposure to light and/or oxygen are the main causes. In this way, growth of pathogenic and/or spoilage microorganisms, losses of nutrients and bioactive compounds and modifications of sensory characteristics as appearance, odor/flavor, texture/viscosity are important criteria to evaluate products shelf life. As a result, determining a product shelf life will depend on physicochemical and sensory analysis (ASHURST, 2016; FULLER, 2011; KONG AND SINGH, 2016). However, monitoring those criteria is not enough to define how long a product remains stable. There is a need to establish until which degree those modifications are acceptable, without negatively affecting the products overall quality. Identifying which modifications will have less or greater impact on it, since several changes will occur simultaneously, at different rates, is equally important (ASHURST, 2016; FULLER, 2011; KONG AND SINGH, 2016).

OBJECTIVES

4.1 General objective

The general objective of this study was to develop jabuticaba juice by steam extraction, evaluating its chemical, microbial and sensory stability during a storage period of 4 months.

4.2 Specific objectives

To determine reproducibility and conditions of the process of production of jabuticaba juice obtained by steam extraction.

To assess the chemical composition of steam extracted jabuticaba juices with and without sucrose addition.

To evaluate the chemical, microbial and sensory stability of steam extracted jabuticaba juice stored at room temperature (25 ± 2 ºC) for 112 and 168 days, respectively.

To evaluate the chemical stability of steam extracted jabuticaba juice stored at accelerated conditions (40 ºC, 50 ºC and 60 ºC) for 42 days.

MATERIAL AND METHODS

5.1 Juice processing

Fruits (Myrciaria jaboticaba, cv. Sabará), purchased at Rio de Janeiro`s agricultural trading center (CADEG), were manually selected, washed, and sanitized in sodium hypochlorite (100 ppm solution) for 15 min and then, packed in vacuum plastic bags and stored at -20 ºC until use. Juice was produced by steam extraction using a stainless steel steam juicer (Cook N Home, City of Industry, California, United States). Steam extraction was conduced using 4 kg of thawed and crushed fruit and 3 L of water. For the juice extraction with the addition of sucrose, 134 g of table sugar were added to the steam juicer together with fruits. The amount of sucrose (6.05% w/v) was established in accordance with INADA et al. (2018b). Fruits were added to the steam juicer after the water boiling point was reached and the extraction was conducted for the given time. The juice was bottled while hot (extraction temperature approximately 80 °C) in sterilized amber glass bottles (to avoid compounds photo-degradation), which were completely filled (with minimum headspace area), capped and rotated to ensure effectiveness of the heat treatment on the cap and headspace area. Initially, jabuticaba juice was produced with no added sucrose (WNAS) and with added sucrose (WAS), and six different extraction times were performed, 10, 20, 30, 40, 50 and 60 min, in order to verify juice production reproducibility as well as to set its processing conditions. Juice extraction time for the following analysis, then, consisted to 30 min (item 5.2). Study design is ilustrated in Figure 13.

10 – 60 min

Reproducibility Processing 30 min conditions Yield pH Yield Total soluble solids Anthocyanins Instrumental color content Anthocyanins content Initial Stability

characterization study

25 ºC 40, 50 and 60 ºC

Sugars Phenolic compounds Organic acids Instrumental color Phenolic compounds Antioxidant activity Instrumental color Antioxidant activity Microbial quality Sensorial quality

Figure 13. Study design. Analyses in orange boxes ( ) were carried out with both WNAS and WAS juices, while in blue boxes ( ), with only WNAS juice.

5.2 Juice production reproducibility and processing conditions

The reproducibility of the juice extraction process was set through the calculation of the coefficient of variation (%CV) between two batches, within each time of extraction.. Yield (measured with a graduated cylinder), pH (described in item 5.4), total soluble solids (TSS) (described in item 5.4), instrumental color (described in item 5.10) and anthocyanin content (described in item 5.9) were the parameters taken into account. The duration of extraction to be used in the following experiments (juice stability study) was chosen according to the parameters yield and anthocyanin content, by comparison between all the times of extraction.

5.3 Stability study

The stability of steam extracted jabuticaba juices was evaluated under different temperatures of storage, depending on the juice. After bottled, juices were stored at four different temperatures: room temperature (25 ± 1 ºC), 40 ºC, 50 ºC and 60 ºC, except for juice WAS, which was not stored at 25 ºC. Samples at 25 ºC were stored during a period of 112 days, while samples at accelerated stability condition, 42 days. The stability of the juices stored at 25 ºC was evaluated through antioxidant activity, phenolic compounds content and instrumental color, every 14 days, organic acids contents, every 28 days, and sugars content and microbial quality, every 56 days. Sensorial stability was evaluated at the beginning and at the end of the storage period. At the same time, the stability of the juices stored at 40 ºC, 50 ºC and 60 ºC were evaluated through antioxidant activity, phenolic compounds content and instrumental color every 7 days. Juice stability was evaluated in triplicates of process (three batches of juice were extracted without sucrose addition and three batches were extracted with added sucrose).

5.4 Proximate composition, pH, titratable acidity and total soluble solids

Lipids, protein, total dietetic fibers, moisture and ash contents, pH, titratable acidity (TA) and total soluble solids (TSS) of steam extracted jabuticaba juice with no added sucrose were determined in triplicate at the beginning of the storage period, according to official methods (Association of Official Analytical Chemists, 2016). Total carbohydrate

was determined by difference, subtracting moisture, lipid, protein and ash contents from 100%. Energy value was calculated from the contents of lipids, proteins and carbohydrates (excluding dietary fiber) multiplied by the Atwater factors (9 kcal/g, 4 kcal/g and 4 kcal/g, respectively). Results were expressed per 100 mL of juice.

5.5 Sugar analysis by HPLC-ELSD

Sugar content of steam extracted jabuticaba juice WNAS was evaluated in triplicate of process, at the beginning of the storage period and every 56 days of storage. Previously to the analysis, juices were diluted with acetonitrile and centrifuged (13,000 rpm, 10 min) (MiniSpin, Eppendorf, Hamburgo, Alemanha). The supernatant was collected and filtered through a 0.45μm cellulose ester membrane (Millipore, Barueri, Brazil). The liquid chromatography system (Shimadzu, Kyoto, Japan) included a LC-20AT quaternary pump, a 8125 manual injector (Rheodyne) with a 20μL loop, an ELSD-LT II evaporative light scattering detection (ELSD), a CBM-20A system controller and a DGU- 20A5 degasser. Chromatographic separation of sugars was performed according to FARAH et al.

(2006), with slight modifications. A normal phase column (NH2, 5μm, 250mm × 4.6mm, Zorbax) was used and the mobile phase (isocratic elution) consisted of a 85% acetonitrile aqueous solution, with a flow rate of 1.0 mL/min. ELSD was set to gain 4, 40 ºC and 350 kPa nebulizer gas pressure (nitrogen). Identification of analytes was performed by comparison with retention time of the respective standard. Quantification was performed by external calibration. Data were acquired by LC solution software (Shimadzu Corporation, version 1.25, 2009). Results were expressed as g per 100 mL of juice.

5.6 Organic acids analysis by HPLC-DAD

Organic acids content of steam extracted jabuticaba juice WNAS was evaluated in triplicate of process, at the beginning of the storage period and every 28 days of storage.

Previously to the analysis, juices were centrifuged (13,000 rpm, 10 minutes) (MiniSpin, Eppendorf, Hamburgo, Alemanha) and the supernatant was collected and filtered through a 0.45μm cellulose ester membrane (Millipore, Barueri, Brazil). The liquid chromatography system (Shimadzu, Kyoto, Japan) included two parallel pumps LC-20AD, automatic injector SIL-20AHT, diode array detector (DAD) SPD- M20A, system controller CBM-20A and degasser DGU- 20A5. Chromatographic separation of organic acids was performed according to SCHERER et al. (2012), with slight modifications. A reverse phase column (C18, 5 μm, 150 mm × 4.6 mm, Phenomenex) was used and the mobile phase (isocratic elution) consisted of a potassium phosphate buffer solution (KH2PO4) 0.01M, pH 2.6 (adjusted with orthophosphoric acid, H3PO4), with a flow rate of 0.5 mL/min and a 10 μL injection volume. Identification of analytes was performed by comparison with retention time and absorption spectrum of the respective standard. Quantification was performed by external calibration. Data were acquired by Lab Solutions software (Shimadzu Corporation, version 5.82 SPI, 2008-2015). Results were expressed as g per 100 mL of juice.

5.7 Phenolic compounds analysis by HPLC-DAD

Phenolic compounds content of steam extracted jabuticaba juices WNAS and WAS were evaluated in triplicate of process, at the beginning of the storage period and every 7 or 14 days of storage, depending on the storage temperature. The liquid chromatography system used and the sample pre-treatment applied were the same as described in item 4.6. Chromatographic separation of anthocyanins was performed according to INADA et al. (2015), with slight modifications. A reverse phase column (C18, 5 μm, 150 mm × 4.6 mm, Phenomenex) was used and the mobile phases consisted of a gradient of 1% formic acid and 2% acetonitrile in water (eluent A) and 1% formic acid and 2% acetonitrile in methanol (eluent B), with a flow rate of 1.0 mL/min and a 10 μL of injection volume. Prior to the injection, the column was equilibrated with 23%B. After injection, solvent composition was kept constant until 1 minute, increased to 29% B in 2 minutes, to 33% B in 4 minutes, to 48% B in 6 minutes, to 85% B in 8 minutes and to 95% B in 10 minutes. Then, it decreased to 23% B in 11 minutes. Between injections, 10 minutes intervals were

used to re-equilibrate the column with 23% B. Anthocyanins were monitored by DAD at 530 nm. Chromatographic separation of non-anthocyanins phenolic compounds was performed according to INADA et al. (2015), with slight modifications. A reverse phase column (C18, 5μm, 250mm × 4.6 mm, Kromasil) was used and the mobile phase consisted of a gradient of 0.3% formic acid and 1% acetonitrile in water (eluent A) and 1% acetonitrile in methanol (eluent B), with a flow rate of 1.0 mL/min and an 10 μL of injection volume. Prior to the injection, the column was equilibrated with 18.2% B. After injection, solvent composition was increased to 20.2% B in 1 minute, to 43.4% B in 18 minutes, to 85.9% in 23 minutes and kept constant until 30 minutes. Between injections, 10 minutes intervals were used to re-equilibrate the column with 18.2% B. Non-anthocyanins phenolic compounds were monitored by DAD from 190 to 370 nm. Identification of analytes was performed by comparison with retention time and absorption spectrum of the respective standard. Quantification was performed by external calibration. Data were acquired Lab Solutions software (Shimadzu Corporation, version 5.82 SPI, 2008-2015). Results were expressed as mg per 100 mL of juice.

5.8 Phenolic compounds degradation and formation kinetics

Based on the results for phenolic compounds contents during the storage period at different temperatures, kinetic models were fitted to the order of reaction of gallic acid formation and anthocyanin degradation, zero and first-order, respectively. A general expression (Eq. 1) was used to determine the reaction rate

-d[C]/dt = k [C]n (1) where [C] is the concentration of the phenolic compounds under consideration, t the reaction time, k the rate constant and n the order of the reaction. As a result, zero (n=0) and first-order (n=1) reaction rates, at constant temperature, could be expressed as Equations 2 and 3, respectively, after integration between a given time point and t0.

[C]t = [C]0 -kt (2)

[C]t = [C]0 exp (-kt) (3)

Zero and first-order rate constants at each temperature were calculated through linear regression (k = -slope), by plotting [C] against time and ln [C]t/[C]0 against time, respectively. The effect of the temperature on the reaction rate constants was described by the Arrhenius equation (Eq. 4)

k = A exp (-Ea/RT) (4)

-1 -1 where A is the frequency factor (day ), Ea the activation energy (kJ mol ), R the universal gas constant (8.3145 J mol-1 K-1) and T the absolute temperature (K). The activation energy was calculated as a product of the gas constant R and the slope of the graph obtained by plotting ln k against 1/T.

5.9 Antioxidant activity by spectrophotometric methods

Antioxidant activity (AA) of steam extracted jabuticaba juices with no added sucrose and with added sucrose was evaluated in triplicate of process, at the beginning of the storage period and every 7 or 14 days of storage, depending on the storage temperature. AA was evaluated by FRAP (Ferric Reducing Antioxidant Power), TEAC (Trolox Equivalent Antioxidant Capacity) and Folin-Ciocalteu assays. Juices were centrifuged (13,000 rpm, 10 minutes) (MiniSpin, Eppendorf, Hamburgo, Alemanha) and the supernatant was collected for the analysis. FRAP assay was performed according to BENZIE AND STRAIN (1996) with slight modifications. FRAP reagent was prepared by mixing 300 mM acetate buffer (pH 3.6), 10 mM TPTZ (2,4,6-tri(2-piridyl)-s-triazine) solution and 20 mM iron chloride

(FeCl3) solution, proportion 10:1:1, respectively. The reagent was warmed to 37 ºC and maintained at this temperature until the analysis. Aliquots of juice (20 μL) and colorless FRAP reagent (180 μL) were pipetted into a 96-well microplate, which was placed in a multilabel reader (Victor3 1420, PerkinElmer, Turku, Finland). After 4 min for the reaction to take place, that is, the reduction of iron from ferric state (Fe3+) to ferrous state (Fe2+) with the formation of a colored ferrous-TPTZ complex, the absorbance was read at 595nm. Quantification was performed using a calibration curve prepared with 1mM iron sulfate 2+ (FeSO4) solution. Results were expressed as mM of Fe equivalent per 100mL of juice. TEAC assay was performed according to RE et al. (1999) with slight modifications.

The radical cation stock solution (ABTS+) was generated by reacting 2,2'-azino-bis (3- ethylbenzothiazoline-6-sulphonic acid) (ABTS), potassium persulfate (K2S2O8) and water, 12-16h prior the use. At the day of the analysis, the ABTS radical cation stock solution was diluted in water (1:50) to an absorbance of 0.70 ± 0.02 at 720 nm (intensely bluish green colored ABTS reagent). Aliquots of juice (10μL) and ABTS reagent (190μL) were pipetted into a 96-well microplate, which was placed in a multilabel reader (Victor3 1420, PerkinElmer, Turku, Finland). After 6 min for the reaction to take place, that is, the neutralization of ABTS+ by either reduction via electron donation or hydrogen atom donation, the absorbance was read at 720 nm. Sample absorbance was subtracted from solvent blank absorbance (distilled water). The measured decrease in color intensity is proportional to the ability of antioxidant compounds in the sample to scavenge radicals. Quantification was performed using a calibration curve prepared with a 2mM Trolox (6- hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid) solution. Results were expressed as mM of trolox equivalent per 100 mL of juice. Folin-Ciocalteu assay was performed as described by SINGLETON, ORTHOFER, LAMUELA-RAVENTÓS (1999), with slight modifications. 20% sodium carbonate

(Na2CO3) and 500 μL/mL gallic acid stock solutions were prepared before the analysis. First, aliquots of juice (100 μL), distilled water (190 μL) and Folin-Ciocalteu reagent (50 μL) were mixed and homogenized. Then, 150 μL of 20% sodium carbonate solution were added to the mixture, which were another time homogenized. After, the mixture was held at 40 ºC for 30 min to allow its reaction. Aliquots of the mixture (300 μL) were pipetted into a 96-well microplate, which was placed in a multilabel reader (Victor3 1420, PerkinElmer, Turku, Finland). The absorbance was read at 765 nm. Quantification was performed using a calibration curve prepared with a gallic acid. Results were expressed as mg of gallic acid equivalent (GAE) per 100 mL of juice.

5.10 Instrumental color

Instrumental color of steam extracted jabuticaba juices WNAS and WAS was evaluated in triplicate of process, at the beginning of the storage period and every 7 or 14 days of storage, depending on the storage temperature. Instrumental color parameters of juices were measured using a Konica Minolta colorimeter CR-400 (Konica Minolta, Tokyo, Japan). The equipment was set to illuminant

D65 (daylight) and 2° observer angle and calibrated using a standard white reference tile (ceramic color standard). The CIELab color space was used to determine the color components L*, a* and b*, which represents lightness (0, black; 100, white), greenness (negative, -a*) or redness (positive, +a*) and blueness (negative, -b*) or yellowness (positive, +b*), respectively.

The total color difference (ΔE) during storage, between the initial storage time (t0) and any given time (ti), was calculated using the following equation:

∆E* = (�∗ − �∗ )2 + (�∗ − �∗ )2 + (�∗ − �∗ )2 √ �0 �� 0 �� 0 ��

5.11 Microbiological analysis

Steam extracted jabuticaba juice WNAS microbial quality was evaluated according to the Brazilian legislation (RDC nº 12, 02/01/2001, ANVISA), investigating the absence of thermotolerant coliforms bacteria and Salmonella spp. Possible deteriorating microorganisms were also evaluated (heterotrophic bacteria, lactic bacteria, yeasts and molds). Microbiological analysis was conducted according to the analytical methodology described in the Compendium of Methods for the Microbiological Examination of Foods (DOWNES AND ITO, 2001).

5.12 Sensory analysis

The present study received authorization to proceed from the Federal University of Rio de Janeiro ethics committee (approval number: 2.425.898). Sensory acceptance and purchase intent of steam-extracted jabuticaba juices WNAS and WAS were performed at the beginning and at the end of the storage period, 0 and 180 days, respectively. Before the tests, the consumers provided written informed consent and answered a questionnaire to assess the consumer profile. All the consumers reported frequent fruit juice consumption (at least once a week). Juices (25 mL) were presented at 10 ºC in 50 mL plastic cups coded with three-digits numbers and offered monadically in balanced order. Mineral water was offered for palate cleansing between samples. The following acceptance attributes were evaluated: overall impression, aroma, color, flavor and viscosity, using a 9 55

points structured hedonic scale, with intensity ranging from 1 (“dislike extremely”) to 9 (“like extremely”) and the “indifferent” (“neither like nor dislike”) response set at 5. Participants were also asked to report presence or absence of any residual taste. The purchase intent was assessed using a 5 points structured scale, ranging from 1 (“I definitely would not buy”) to 5 (“I definitely would buy”). Immediately after juices preparation, sensory acceptance and purchase intent were carried out by 118 untrained consumers (60 men and 57 women), aged between 17 and 40 years. After juices storage for 180 days, sensory acceptance and purchase intent were evaluated again, this time by 110 untrained consumers (64 men and 46 women), aged between 17 and 58 years. Consumers recruited were, in its majority, students of the Federal University of Rio de Janeiro.

5.13 Statistical analyses

Data are expressed as mean ± standard deviation. The reproducibility of the process of production of jabuticaba juice obtained by steam extraction was evaluated by the coefficient of variation. The effect of the addition of sucrose was evaluated by t-test, comparing the extractions added or not. Differences between times of extraction were evaluated by one-way ANOVA followed by Tukey’s test. To evaluate the differences among the intervals of storage, one-way ANOVA followed by Dunnett’s test was used. Pearson correlation analysis was performed to evaluate associations between variables instrumental color and anthocyanins content. Statistical analyses were performed using GraphPad Prism software for macOS (version 5.0a) and results were considered significant when p < 0.05.

RESULTS AND DISCUSSION

6.1 Reproducibility and conditions of the process of production of jabuticaba juices by steam extraction

Controlling the food manufacturing process is a way to assure the final product quality. There are different control strategies, applied pre-, during and/or post-processing (BERG et al., 2013). So, for a consistent and reliable process performance over time, identifying products critical quality attributes and sources of variability within the manufacturing process are important steps (RATHORE AND KAPOOR, 2017), as well as defining the acceptable range for quality variation (BERG et al., 2013). For this purpose, juice’s production yield, pH, TSS, color and anthocyanin content were chosen to evaluate the reproducibility of the process of production of steam extracted jabuticaba juice. SST and pH are characteristics related to product taste and also used to define identity and quality standards by Brazilian legislation, while color is a sensory attribute linked to consumer perception of the product. Anthocyanins were chosen as these compounds are major bioactives presented in the juice and also directly impact on product color. Except for color, which was evaluated through total color difference (ΔE), the variation of all attributes was evaluated by the coefficient of variation (%CV) between two batches of juice production (each batch was considered as a replicate). Juice production yield, pH, TSS, ΔE and anthocyanin contents of the six extraction times of both types of juices, that is with no added sucrose (WNAS) and with added sucrose (WAS), are presented in Table 1. Juice pH and TSS %CV values were below 2% and 6%, respectively, while yield and anthocyanin contents presented slightly higher %CV values, below 7% and 13%, respectively. Total color differences, except for 10 and 20 min extraction times, were indistinguishable (OBÓN, CASTELLAR, ALACID, 2009). Thus, considering the non-controllable sources of variability within the process of juice steam extraction, such as heat intensity or strength applied during fruit crushing, all parameters chosen to evaluate the juice steam extraction reproducibility were in accordance with pre- established acceptable values (%CV < 15%), demonstrating that the extraction method was consistent, capable of producing juices with similar characteristics. The choice of the juice steam extraction time was made according to only two parameters, yield and anthocyanin contents, once there were no significant differences on juice pH and TSS values between the six extraction times proposed, for both juices, and also considering that color would not be a decisive factor for consumer’s choice, as no

reference color has been previously established. We observed an increase in juices production yield and anthocyanins content (Table 1) along the duration of the extraction, which was more expressive at initial extraction times, especially up to 30 min, progressively decreasing in the following extraction times, regardless of the addition of sucrose. Concerning juice WNAS, the increase in volume produced was of 44%, 19%, 10% and 6% after 20, 30, 40 and 50 min, respectively, while juice WAS volume increase was of 31%, 21%, 12%, and 9%. For both juices, the increase in yield between 50 and 60 min extraction times was not significant. Regarding anthocyanins content, WNAS 10 min extraction presented 31% and 21% less than 20 and 30 min, respectively. Differences between 30 and 40 min, 40 and 50 min and 50 and 60 min extraction were not significant. WAS 10 min extraction presented 80%, 33%, 20%, 14% and 10% less than 20, 30, 40 and 50 min, respectively. So, taking into account the increase in juice production yield and anthocyanins content between the following times, 30 min was the most advantageous extraction time.

Table 1. Yield, anthocyanins content, pH, total soluble solids and total color difference values of steam extracted jabuticaba juices. Extraction Anthocyanins TSS Yield (L) pH ΔE time (min) (mg/100 mL) (ºBrix) With no added sucrose 10 1.28a (7%) 5.70a (13%) 3.42a (1%) 10.8a (0%) 2.58 20 1.84b (1%) 7.45b (5%) 3.41a (1%) 10.4a (2%) 0.42 30 2.19c (1%) 9.00c (2%) 3.41a (0%) 10.6a (1%) 1.13 40 2.41d (0%) 10.09c,d (1%) 3.41a (0%) 10.5a (3%) 1.09 50 2.57e (0%) 10.95d,e (1%) 3.41a (0%) 10.6a (3%) 1.12 60 2.66e (0%) 11.46e (1%) 3.41a (0%) 10.9a (2%) 1.17 With added sucrose 10 1.38a (2%) 2.88a (12%) 3.40a (0%) 13.4a (6%) 3.24 20 2.02b (1%) 5.19b (5%) 3.42a (2%) 14.4a (3%) 5.88 30 2.38c (1%) 6.90c (2%) 3.43a (1%) 14.6a (1%) 1.42 40 2.63d (1%) 8.24d (1%) 3.43a (1%) 14.4a (2%) 0.68 50 2.80e (2%) 9.40e (0%) 3.43a (1%) 14.5a (1%) 1.36 60 2.92e (2%) 10.30f (1%) 3.43a (1%) 14.3a (1%) 0.55 Results are expressed as mean of two process replicates, on fresh weight basis, except for the total color difference; means in the same column with different superscript letters are significantly different, for each type of extraction (One-way ANOVA test followed by Tukeys’ post hoc test; p < 0.05); ΔE values indicate total color difference between two production batches of each type of extraction; % values between parentheses indicates the %CV between two production batches of each extraction time.

6.2 Characterization of steam extracted jabuticaba juices 59

Fruit juices are mainly composed of water and water-soluble compounds, such as sugars, pigments, organic acids, phenolic compounds, vitamins and minerals, pectic substances, containing low amounts of proteins and lipids (LOZANO, 2006). Steam extracted jabuticaba juice WNAS presented high moisture content, low contents of proteins, lipids and dietary fibers, while carbohydrates were the main macronutrient (Table 2), in accordance with the nature of the product and, moreover, with the method of production employed, which is based on aqueous extraction. This proximate composition profile was similar to those showed by ALEZANDRO et al. (2013), GURAK et al. (2014) and INADA et al. (2015) for the fruit, except dietary fiber, which, according to these authors, were found in relative high amounts. However, the fiber fraction of jabuticaba fruit is mainly composed of insoluble fiber, with small amounts of soluble fiber (ALEZANDRO et al., 2013; GURAK et al., 2014), what could justify juice low fiber content, once only the soluble fraction was extracted to the juice.

Table 2. Proximate composition, energy, pH, total soluble solids and titratable acid values of steam extracted jabuticaba juice with no added sucrose. Moisture (%) 94.99 ± 0.12 Ash (g/100 mL) 0.43 ± 0.01 Protein (g/100 mL) 0.11 ± 0.00 Lipid (g/100 mL) 0.01 ± 0.00 Carbohydrate (g/100 mL) 4.47 ± 0.12 Dietary fiber (g/100 mL) 0.02 ± 0.00 Energy value (kcal/100 mL) 18.37 ± 0.51 TSS (ªBrix) 9.9 ± 0.2 pH 3.38 ± 0.01 TA (g citric acid/100 mL) 1.14 ± 0.04 Results are expressed as mean ± standard deviations of three replicates, on fresh weight basis; TA, titratable acidity.

TSS, pH and titratable acidity (TA) values of steam extracted jabuticaba juice WNAS are shown in Table 2. TSS content was lower than TSS fruit content (12.7 ºBrix), evaluated before juice extraction. This difference is related to the exogenous water incorporated to the juice during the steam extraction processing, resulting in a dilution of the soluble compounds (YAMAMOTO et al., 2015). Likewise, LOPES et al. (2016), analyzing grape juices extracted by four different processing methods (steam extractor, domestic blender, masticating juice extractor and centrifugal juicer), found that steam extracted grape juice had a TSS content much lower than the other three juices produced by mechanical methods. The low pH was in accordance with jabuticaba’s fruit pH values previously reported (ALEZANDRO et al., 2013; INADA et al., 2015). Sugar content of WNAS juice (Table 3) was characterized by the presence of two monosaccharides, glucose and fructose, and one disaccharide, sucrose. Fructose was the major sugar, followed by glucose, while sucrose was found in much lower concentration. Fructose and glucose, together, accounted for around 93% of the total sugar content. This profile was in accordance with previous reports of the fruit sugar profile (LIMA et al., 2010; ALEZANDRO et al., 2013).

Table 3. Sugars contents (g/100 mL) of steam extracted jabuticaba juice with no added sucrose. Sugars Contents Fructose 1.79 ± 0.11 Glucose 1.10 ± 0.11 Sucrose 0.20 ± 0.03 Results are expressed as mean ± standard deviations of three process replicates, on fresh weight basis.

Organic acids found in WNAS juice (Table 4) include oxalic, tartaric, malic and citric acids. Among these four organic acids, citric acid was the most abundant, accounting for around 91% of the total organic acids content, followed by malic and oxalic and tartaric acids, which contents were similar. LIMA et al. (2010) reported five organic acids in the jabuticaba fruit. Citric acid was the one found in higher concentrations, followed by succinic, malic, oxalic and acetic acids, the last two appearing in very low concentrations. In contrast, JHAM et al. (2007) presented a different profile. Succinic was the main organic acid in jabuticaba fruit, citric acid was the second most abundant and malic was only detected in trace amounts. Both studies have analyzed the organic acid content of 61

paulista and sabará species, showing no differences between the species. Besides species, other factors might influence fruits organic acids content, such as ripening stage, environmental and cultivation conditions, what may justify the differences found among these studies and ours (FAMIANI et al., 2015; ZHENG et al., 2009).

Table 4. Organic acids contents (g/100 mL) of steam extracted jabuticaba juice with no added sucrose. Organic acids Contents Oxalic acid 0.02 ± 0.00 Tartaric acid 0.02 ± 0.00 Malic acid 0.09 ± 0.01 Citric acid 1.31 ± 0.07 Results are expressed as mean ± standard deviations of three process replicates, on fresh weight basis.

Some authors have reported the presence of ascorbic acid (AA) in jabuticaba fruit pulp (ABE, LAJOLO, GENOVESE, 2011; INADA et al., 2015), yet, neither this compound nor its main degradation product (dehydroascorbic acid, DHAA) were identified in the juice, probably having been degraded during juice production. Heat, oxygen, light, pH and other factors are commonly responsible for AA degradation during food processing. This compound is highly susceptible to oxidation to DHAA, which may be further decomposed to other degradation products, such as 2,3-diketogulonic acid (GREGORY III, 2008). MERTZ et al. (2010) evaluated the impact of thermal treatment on AA and DHAA contents of tamarillo nectar and observed that AA was completely degraded after heated at 80, 90 and 95 ºC during 10 min, whereas DHAA content undergone partially degradation, which was more pronounced in the treatments with Phenolic compounds from different classes were identified in WNAS juice (Table 5), comprising two anthocyanins (C3G and D3G), two hydroxybenzoic acids (gallic and ellagic acids), three flavonols (myricetin-3-O-rhamnoside, quercetin and quercetin-3-O- rutinoside) and one hydroxycinnamic acid (trans-cinnamic acid). From the eight phenolic compounds found in the juice, C3G was found in the highest concentration, accounting for 40% of the juice total phenolic compounds content. Gallic and ellagic acids contents were also high, representing around 28% and 21% of the juice total phenolic compounds 62

content, respectively. D3G was found in lower concentrations, but still significant, whereas myricetin-3-O-rhaminoside, quercetin-3-O-rutinoside, quercetin and trans-cinnamic acid were the minor phenolic compounds, accounting together for about 7% of the phenolic compounds content found in the juice. Previous works have already reported the presence of these phenolic compounds in jabuticaba fruit (WU et al., 2012; ALEZANDRO et al., 2013; INADA et al., 2015; PEREIRA et al., 2017), indicating that the steam extraction method was capable of producing a juice with characteristics similar to that of the fresh fruit in terms of its bioactive compounds. In fact, INADA et al. (2018b) produced jabuticaba juice by steam extraction that presented phenolic compounds content and profile similar to that of the present study, except for m-coumaric and myricetin (minor compounds, according to the authors).

Table 5. Phenolic compounds contents (mg/100 mL) of steam extracted steam extracted jabuticaba juice with no added sucrose. Phenolic compounds Contents Delphinidin-3-O-glucoside 1.13 ± 0.03 Cyanidin-3-O-glucoside 8.06 ± 0.15 Gallic acid 5.59 ± 0.61 Myricetin-3-O-rhaminoside 0.48 ± 0.10 Quercetin-3-O-rutinoside 0.71 ± 0.07 Ellagic acid 4.04 ± 0.25 Quercetin 0.01 ± 0.00 Trans-cinnamic acid 0.13 ± 0.03 Total phenolics compoundsa 20.16 ± 1.12 Results are expressed as mean ± standard deviations of three process replicates, on fresh weight basis; aCalculated as the sum of all phenolic compounds found.

Microbial quality of fresh juice WNAS is presented in Table 6. The absence of thermotolerant coliforms and Salmonella sp. indicates that the juice was produced and bottled under proper sanitary conditions and also that the steam extraction method ensured a juice microbiologically safe for consumption due to the heat applied during the extraction, which could be compared to pasteurization (BATES, MORRIS, CRANDAL, 2001). Low numbers of heterotrophic bacteria, lactic bacteria, yeasts and molds, possible 63

deteriorating microorganisms, were found, showing that sensory quality loss during juice storage due to microorganism spoilage would not probably be a risk. Moreover, these results show that the raw material selection and sanitization, previously to juice extraction, were properly done, possibly lowering its initial microbial load. LOPES et al. (2016) and INADA et al. (2018b) found similar microbial quality results with grape and jabuticaba juices, respectively, both produced by steam extraction.

Table 6. Microbiological analysis of steam extracted jabuticaba juice with no added sucrose. Microorganism Juice value Reference values1 Thermotolerant coliforms (MPN/50 mL)b Absence Absence Salmonella sp. (25 mL) Absence Absence Heterotrophic bacteria (CFU/mL)a 2.0 x 10 ----- Yeasts and molds (CFU/mL) 9.0 x 10 ----- Lactic acid bacteria (CFU/mL) < 1.0 x 10 ----- 1According to Brazilian legislation (RDC nº 12, 02/01/2001). aColony-forming units per milliliter; bMost probable number per 50 milliliters.

Sensory acceptance of juice WNAS is presented in Table 7. The highest scoring attributes were color and viscosity, corresponding to hedonic parameters “like very much” and “like moderately”, respectively. Overall impression and aroma had lower sensory scores, although still satisfactory, between the hedonic parameters “like moderately” and “like slightly”. Flavor presented the lowest score, between the hedonic parameters “like slightly” and “neither like nor dislike”. Many participants reported bitter residual taste and astringent mouth-feel that could explain this result. According to COULTATE (2007), astringency is a sensation associated to bitterness. Therefore, flavor acceptance may be impaired by the presence of jabuticaba peel tannins (PEREIRA et al., 2017; NEVES et al., 2018) in the juice, as well as its acidity. Bitterness and astringency are sensory properties known to be related to the presence of polyphenols in foods and to their acidic profile, and are usually negatively perceived by the consumers (LESSCHAEVE AND NOBLE, 2005; JAEGER et al., 2009). LAAKSONEN et al. (2013), in a study evaluating the sensory quality and the chemical composition of blackcurrant juices, reported the presence of various phenolic compounds and that juices were perceived as sour and astringent. Additionally, they reported that juice’s low pH

(around 3.0) and organic acids content (mainly characterized by citric acid, on average 2.9 g/100 mL) contributed to these sensory properties. The jabuticaba juice developed in the present study showed similar pH values (3.4) and citric acid also as the major organic acid, although at 2.2-fold lower concentration. As a whole, steam extracted jabuticaba juice can be considered an acceptable product as all sensory attributes scores were above the central point in the nine-point hedonic scale. The purchase intent score (Table 7), however, was close to “maybe I would buy, maybe I would not buy” at the five-point scale. This sensory acceptance pattern was similar to the one reported by INADA et al. (2018b) for jabuticaba juice produced by the same extraction method.

Table 7. Sensory analysis of steam extracted jabuticaba juice with no added sucrose (n=118). Attributes Scores Sensory acceptancea Overall impression 6.5 ± 1.6 Aroma 6.6 ± 1.6 Color 7.9 ± 1.1 Flavor 5.6 ± 2.1 Viscosity 7.3 ± 1.5 Purchase intentb 2.9 ± 1.2 Results are expressed as mean ± standard deviation; aNine-point scale (1 = dislike extremely; 2 = dislike very much; 3 = dislike moderately; 4 = dislike slightly; 5 = neither like nor dislike; 6 = like slightly; 7 = like moderately; 8 = like very much; 9 = like extremely). bFive-point scale (1 = I definitely would not buy; 2 = I probably would not buy; 3 = maybe I would buy, maybe I would not buy; 4 = I probably would buy; 5 = I definitely would buy).

There are evidences that the addition of sucrose to tannic solutions attenuates the perception of bitterness and astringency, improving consumer acceptability (JAEGER et al., 2009). Additionally, people tend to prefer sweet-tasting foods and beverages (DREWNOWSKI et al., 2012). INADA et al. (2018b) showed that the addition of sucrose improved consumer acceptability of jabuticaba juice produced by the steam extraction, determining juice ideal sweetness as 6.05% sucrose concentration. In this sense, in the present study, steam extracted jabuticaba juice was also produced with added sucrose (WAS), as described by the authors. Antioxidant activity values, phenolic compounds

content and sensory acceptance of juice WAS were evaluated in comparison to steam extracted jabuticaba juice WNAS. The phenolic compounds profile and its total content (Figure 14) were similar between the juices, except for ellagic acid and quercetin contents, which were significantly different. Ellagic acid was 1.2 times lower in WAS in comparison to WNAS, while quercetin content was slightly higher in WAS. These differences, however, do not seem to be much relevant, especially because quercetin was the minor phenolic compound of both juices, found only in very low concentrations. In contrast, according to NOWICKA AND WOJDYLO (2016), the addition of different sweeteners to sour cherry purees reduced their contents of polyphenols immediately after its processing. The highest content of phenolic compounds was observed in sour cherry puree without any added sweetener, while the addition of sucrose (7%), in particular, led to a decrease of 37%.

20.0

n 15.0 e t

n o 10.0 c

) s L d n

m 5.0 *

u 0 o 0 p 1 / m 1.5 g o c

m ( c

i 1.0 l o

n e 0.5 h P 0.2 0.1 * 0.0

e e id id e e in id s id id c c id id t c ic s s a a s s e a l o o c c o o rc c o c c li i n n e i n lu lu l g i ti u e g g a la m u am h - - G l a -r Q n p E h n l -O -O -r O i ta 3 3 - c o - - -O -3 s- in in 3 n n T d d - ti a i i in e r in an t rc T h y ce e lp C ri u e y Q D M

Figure 14. Phenolic compounds content of steam extracted jabuticaba juices with no added sucrose (WNAS, ) and with added sucrose (WAS, ); results are expressed as mean ± standard deviations of three process replicates, on fresh weight basis; means with * are significantly different (unpaired t-test; p < 0.05).

NOWICKA AND WOJDYLO (2016) have also reported that the addition of sweeteners, including sucrose, led to a reduction in antioxidant activity values of sour cherry purees. In contrast, in the present study, the addition of sucrose did not influence FRAP, TEAC and Folin-Ciocalteu values. None of the assays showed significant differences between steam extracted jabuticaba juices WNAS and WAS antioxidant activity values (Figure 15), what could be justified by the small/no changes in WAS juice phenolic compounds profile and total phenolic compounds content in comparison to WNAS juice.

6 4 350

A ) B C ) L

L 300 m

0 ) 0 0 3 L 0 1 u 250 / m 1 e

/ t 4 x l + 0 o a l 2 0 c e o

1 200 o r / F i

T l E

2 C l o - A o n 150 m i l G m

m o

g ( m

2 F (

P 100 (

C 1 A A R E

F 50 T 0 0 0

Figure 15. Antioxidant activity values by FRAP (A), TEAC (B) and Folin-Ciocalteu (C) of steam extracted jabuticaba juices with no added sucrose (WNAS, ) and with added sucrose (WAS, ); results are expressed as mean ± standard deviations of three process replicates, on fresh weight basis; means with * are significantly different (unpaired t-test; p < 0.05); GAE, gallic acid equivalents.

As expected, steam extracted jabuticaba juice WAS had better sensory acceptance when compared to juice WNAS (Figure 16A). The addition of sucrose resulted in slightly higher overall impression and flavor scores, corresponding to hedonic parameters “like slightly”. Flavor score was around 21% higher, whereas overall impression score was around 11%. Nevertheless, some participants still reported bitter residual taste and astringent mouth-feel. This result suggests that the concentration of sucrose added was not enough to completely mask this disliking taste and sensation. Sensory attributes less related to sucrose addition, such as aroma, color and viscosity did no show significant differences. The purchase intent score (Figure 16B) was also higher for juice WAS (around 21%), but still remained close to “maybe I would buy, maybe I would not buy” at the five-point scale. These results show that concerning jabuticaba juice, flavor is the main

sensory attribute contributing to consumer disliking and, in order to improve consumer acceptability, it would be relevant to develop strategies to attenuate juice bitterness and astringency. Indeed, concentration of added sucrose appears to play an important role in consumer acceptability of polyphenol-rich beverages, as shown by JAEGER et al. (2009). In this study, they reported increasing liking scores of polyphenol-rich beverages containing berry fruit extracts with increasing sucrose level, that is, the more sucrose added the better acceptance of polyphenol-rich beverages. The addition of 3.5, 7.0 and 10.5% of sucrose resulted in more acceptable beverages than that with no added sucrose; yet, beverages containing the two highest concentrations did not show significant differences. According to them, both perceived bitterness and astringency decreased with the addition of sucrose, however, it only resulted in partial masking, suggesting that other approaches should be combined.

9 A * *

8 5 B * 7 6 4 5 3 4 3 2 2 1 1 0 0 r r n a o o ty io m l v si s o o a o es r C l c r A F is Purchase intent p V im l al er v O

Figure 16. Sensory acceptance (A) and purchase intent (B) of steam extracted jabuticaba juices with no added sucrose (WNAS, ) and with added sucrose (WAS, ); results are expressed as mean ± standard deviations; means with * are significantly different (paired t-test; p < 0.05), n = 118.

6.3 Effect of storage on chemical composition, microbial and sensory qualities of steam extracted jabuticaba juice

The shelf life of a food product is the period that it remains suitable for consumption, from both microbial and sensorial points of view. Therefore, safety and

quality are the two main aspects considered (MAN, 2016). It is known that food products characteristics change over time as chemical reactions and microbial growth occur during storage. This may negatively impact food products nutritional value, color, appearance, texture, flavor and bioactive compounds profile and, consequently, consumers acceptability. In this sense, monitoring food products stability and identifying which and how factors affect it can help controlling its quality during storage and estimating its shelf life. Stability studies can be done through experimental storage tests under conditions that simulate those typically found during food products storage, distribution and commercialization, with periodical measurements of quality attributes and the use of kinetic modeling (AMODIO, DEROSSI, COLELLI, 2014; KONG AND SINGH, 2016; BOEKEL, 2008). Kinetic modeling is an important tool for understanding, predicting and controlling the influence of thermal processing and storage on food quality loss. Based on changes in concentrations over time, it is possible to measure how quickly reactions occur and if they are affected by modifications in their conditions. Data is normally obtained experimentally, by monitoring the consumption of a reactant or the formation of a product. Through mathematical models of thermodynamics and chemical kinetics, it is possible to acquire knowledge on kinetics parameters such as reaction order, rate constant and activation energy and on the mechanisms of reactions, as well as to estimate and minimize possible undesirable changes in foods. The Arrhenius equation is the most common relationship used to describe the effect of temperature on chemical reactions, exploring its dependence. Kinetic models in food processing are usually applied to describe, for example, color changes (such as enzymatic and non-enzymatic browning and thermal destruction of pigments), degradation of nutrients and bioactive compounds, formation of undesirable compounds, inactivation of enzymes, microbial growth (LING et al, 2015; PATRAS et al., 2010; BOEKEL, 2008). The present study evaluated the stability of steam extracted jabuticaba juice WNAS stored at 25 ºC for 112 days (chemical) and 168 days (microbial and sensory), in order to simulate juice storage, distribution and commercialization conditions. We also investigated the chemical stability of steam extracted jabuticaba juices WNAS and WAS stored at 40, 50 and 60 ºC. The storage tests at these three different temperatures were performed to minimize the time required to estimate its shelf-life specifically regarding its possible

functional properties as well as to provide phenolic compounds kinetics data and a better understanding of the modifications on these compounds during storage.

6.3.1 Chemical stability: sugars, organic acids, phenolic compounds, color and antioxidant activity

As juices WNAS and WAS showed similar phenolic compounds profile and total content and antioxidant activity values, data of jabuticaba juice WAS chemical stability will not be presented. Sugar content of jabuticaba juice WNAS remained stable during the whole storage period at 25 ºC, except for sucrose, which was found in the lowest concentration (Figure 17). The slight decrease in sucrose content observed at the end of storage period can be possible explained by its chemical hydrolysis to fructose and glucose, the two monosaccharides forming this molecule, catalyzed by the characteristic acidity of the juice. However, no significant increase on fructose or glucose contents was observed, suggesting that the amount of these sugars yielded by the hydrolysis may not have been sufficient to lead to a statistically significant increase in their contents. The stability of sugars is related to their chemistry characteristics. Fructose and glucose show a good stability in acid aqueous solutions, whereas sucrose is more stable in alkaline solutions, being susceptible to hydrolysis at acid pH (CLARKE, EDYE, EGGLESTON, 1997). SANDI et al. (2004) evaluated fructose, glucose and sucrose concentrations of yellow passion fruit pasteurized juices stored at 25 ºC for 4 months and demonstrated that both monosaccharides concentrations increased over storage time while sucrose concentration decreased. Their results are similar to those observed in the present study.

2.5

2.0

t n ) e L t

n 1.5 m

o c

0 r a 1 / g 1.0 g u ( S

0.5 * 0.0 0 56 112

Storage time (days)

Figure 17. Fructose (), glucose () and sucrose () contents of steam extracted jabuticaba juices with no added sucrose over the storage time at 25 ºC; results are expressed as mean ± standard deviations of three process replicates, on fresh weight basis; means with * are significantly different from day 0 (One-way ANOVA test followed by Dunnett post hoc test; p < 0.05).

Except for tartaric acid contents, none of the organic acids of juice WNAS showed significant changes during the storage period at 25 ºC (Figure 18). A slight increase in tartaric acid content was observed at day 84, followed by a decrease at day 112. Still, this was not a relevant difference, as tartaric acid is one of the minor organic acids found in the juice. Among the four organic acids identified, citric acid, as the predominant one, is the one which probably contributes the most to juice flavor characteristics. Changes in citric acid contents during storage would be much more worrisome. According to FÜGEL, CARLE, SCHIEBER (2005), organic acids exhibit low susceptibility to changes during processing and storage, especially if compared to other compounds, such as pigments. This is because their decomposition, resulted from decarboxylation and dehydration reactions, requires heating (CHU AND CLYDESDALE, 1976). Thus, organic acids are susceptible to changes in cases of food thermal over-processing and storage at higher temperatures.

2.0

n e

t 1.5 n ) o c

s m

i 1.0 0

c 0 a

1 / c i g n

( 0.2 a g r O 0.1 * * 0.0 0 28 56 84 112

Storage time (days)

Figure 18. Citric (), malic (), tartaric () and oxalic () acids contents of steam extracted jabuticaba juices with no added sucrose over the storage time at 25 ºC; results are expressed as mean ± standard deviations of three process replicates, on fresh weight basis; means with * are significantly different from day 0 mean (One-way ANOVA test followed by Dunnett post hoc test; p < 0.05).

Phenolic compounds profile of juice WNAS changed during storage at 25 ºC (Table 8). A considerable decrease in anthocyanins contents was observed, with D3G and C3G presenting similar losses (around 97% and 98%, respectively) by the end of the storage period. Ellagic acid content decreased about 51%, while total phenolic compounds content was approximately 50% lower than that observed at day 0. Even though other changes were observed, they were probably not as relevant once they were observed for minor phenolic compounds of the juice. Nevertheless, high intensity unidentified peaks were observed in the chromatogram of juice after 112 days of storage, possibly of phenolic compounds degradation products.

Table 8. Phenolic compounds contents (mg/100 mL) of steam extracted jabuticaba juice with no added sucrose during storage at 25 ºC temperature.

Days of storage

0 112 Delphinidin-3-O-glucoside 1.13 ± 0.03 0.04 ± 0,00* Cyanidin-3-O-glucoside 8.06 ± 0.15 0.20 ± 0.03* Gallic acid 5.59 ± 0.61 6.74 ± 0.51 Ellagic acid 4.04 ± 0.25 2.00 ± 0.08* Myricetin-3-O-rhamnoside 0.48 ± 0.10 0.44 ± 0.02 Quercetin-3-O-rutinoside 0.71 ± 0.07 0.54 ± 0.08* Quercetin 0.01 ± 0.00 0.02 ± 0.00 Trans-cinnamic acid 0.13 ± 0.00 0.19 ± 0.01* Total phenolic compoundsa 20.16 ± 1.12 10.13 ± 0.50* Results are expressed as mean ± standard deviations of three process replicates, on fresh weight basis; means with * are significantly different (One-way ANOVA test followed by Dunnett post hoc test; p < 0.05); acalculated as the sum of all phenolic compounds found.

Anthocyanins are phenolic compounds known to be highly unstable and susceptible to degradation. Several factors affect their stability, accelerating their degradation, such as high pH, high temperature, low concentration, specific chemical structure, light, oxygen, enzymes, sugars, ascorbic acid, metallic ions, among others (CASTAÑEDA-OVANDO et al., 2009; CAVALCANTI, SANTOS, MEIRELES, 2011; PATRAS et al., 2010). Considering this, the expressive decrease in D3G and C3G contents (Table 8) in the present study was, somehow, expected. A similar decrease was reported by INADA et al. (2018b) during the storage of jabuticaba juice at 25 ºC for 90 days. Other previous studies have demonstrated the thermal instability of anthocyanins. SUI et al. (2016) showed that their chemical stability on aqueous solutions directly depends on the storage temperature. According to the authors, solutions stored at 4 ºC showed substantial lower losses than those stored at 25 ºC, 45 ºC and 65 ºC. HELLSTRÖM et al. (2013) and MÄKILÄ et al. (2016) evaluated the anthocyanin content of commercial berry juices stored at 21 ºC for 12 weeks and of blackcurrant juice at 25 ºC for 12 months, respectively. Both studies showed significant reductions on anthocyanins contents, which were attenuated when juices were stored at 4 ºC. In addition, HELLSTRÖM et al. (2013) 73

provided evidence that support the interference of the juice matrix on anthocyanin stability as blackcurrant, crowberry, chokeberry juices and a blend of these three juices presented different losses over the same period and at the same storage temperature. The effect of temperature on the stability of each anthocyanin was investigated. Concentrations of D3G and C3G of jabuticaba juices WNAS stored at 25 ºC, 40 ºC, 50 ºC and 60 ºC decreased over time, at all temperatures, although more rapidly at higher ones (Figure 19). D3G and C3G were almost completely degraded after 112 days at 25 ºC, while at 60 ºC, similar losses were already observed after 7 days. Similarly, SUI et al (2016) have demonstrated that C3G was instable in aqueous solutions, being degraded at higher rates at 65 ºC, leading to a complete loss after six days.

2.0 A

1.5

t )

t m

1.0

( * * 0.5 * * * * * * * * * * * * * * 0.0 0 7 14 21 28 35 42 49 56 63 70 77 84 91 98 105112

Storage time (days)

15 B

)

e 9

0 c

( 6

* *

3 * * * * * * * * * * * * * * * 0 0 7 14 21 28 35 42 49 56 63 70 77 84 91 98 105112 Storage time (days)

Figure 19. Delphinidin-3-O-glucoside (A) and cyanidin-3-O- glucoside (B) contents of steam extracted jabuticaba juice with no added sucrose over the storage time 25 ºC (), 40 ºC (), 50 ºC () and 60 ºC (); results are expressed as mean ± standard deviations of three process replicates, on fresh weight basis; means with * are significantly different from day 0 (One-way ANOVA test followed by Dunnett post hoc test; p < 0.05). 75

Degradation of anthocyanins from jabuticaba juice followed a first-order kinetic model. Apparently, D3G was degraded as fast as C3G. Plots of logarithmically transformed concentrations of both anthocyanins at each temperature against time resulted in straight lines, showing good fitting (R2>0.96; Table 9). First-order reactions are frequently reported in foods, in particular anthocyanins degradation, and indicate that the degradation rate is dependent and proportional to the reagent concentration, showing an exponential behavior (BOEKEL, 2008; HELLSTRÖM et al., 2013; PATRAS et al., 2010; PERON, FRAGA, ANTELO, 2017; SINELA et al., 2017). The rates constants (k) of the degradation reactions increased with the temperature (Table 9), corroborating the fact that the higher temperature, the faster anthocyanins degradation, thus indicating that the storage temperature had a strong influence on D3G and C3G losses. MERCALI et al. (2015) investigated the monomeric anthocyanin degradation from jabuticaba juice storage between 70 and 90 ºC and demonstrated a similar pattern of increasing rate constant values, which was graphically presented (Figure 20) and showed linear behavior, proving the conformity of the model.

Table 9. First-order kinetic model fitting, rate constants and activation energy of anthocyanin degradation in steam extracted jabuticaba juice with no added sucrose. Anthocyanins Delphinidin-3-O-glucoside Cyanidin-3-O-glucoside Coefficient of determination (R2) 25 ºC 0.9658 0.9874 40 ºC 0.9872 0.9939 50 ºC 0.9875 0.9999

Rate constants (k) (mg/100 mL.day) 25 ºC 0.0299 0.0321 40 ºC 0.1196 0.1362 50 ºC 0.2648 0.3149

Activation energy 70.10 73.38 (Ea, kJ/mol)

y = -8826x + 26.17 -1 R² = 0.9997 ) 1 -

s -2 y a d

( y = -8432x + 24.78

k -3 R² = 0.9996 n l -4

-5 0.0030 0.0031 0.0032 0.0033 0.0034

1/T (K)

Figure 20. Arrhenius plots for degradation of delphinidin-3-O- glucoside () and cyanidin-3-O-glucoside () in steam extracted jabuticaba juice with no added sucrose.

The activation energy (Ea) works as a parameter for quantitatively characterizing the effect of temperature on reaction rates. It is the minimum amount of energy that is required so the compounds are able to react and is determined from experimental rate constants (BOEKEL, 2008). A reaction with higher Ea is more sensitive to temperature, what means that substrates in reactions with higher Ea tend to be degraded by minor changes in the temperature (HOU et al., 2013). Experimental Ea values are presented in Table 9. The value for D3G (70.10 kJ/mol) was slightly lower than for C3G (73.38 kJ/mol), suggesting that the former anthocyanin would be less stable than the latter. It is known that degradation rates vary among anthocyanins due to differences in their chemical structure. Generally, the number and the position of hydroxyl and methoxyl groups in the aglycone affect anthocyanins chemical behavior. An increase in hydroxylation decreases their stability while an increase in methoxylation increases their stability (CASTAÑEDA- OVANDO et al., 2009; CAVALCANTI, SANTOS, MEIRELES, 2011). In this sense, since D3G and C3G contain two and one hydroxyl groups, respectively, it would be expected a higher stability of the latter. In fact, SINELA et al. (2017) evaluated the stability of delphinidin-3-O-sambubioside (D3S) and cyanidin-3-O-sambubioside (C3S) from hibiscus extract and demonstrated that D3S tended to be significantly more sensitive to an increase in temperature than C3S. Despite the difference on the nature of glycosylation, both studies evaluated comparatively the degradation of anthocyanins with

the same aglycone moieties, as well as the same position of glycosylation. Beyond anthocyanins contribution to phenolic compounds content of plants, these compounds are natural pigments responsible for the pink, red, violet and blue colors and its hues of fruits, flowers, leaves and some vegetables (CASTAÑEDA-OVANDO et al., 2009; CAVALCANTI, SANTOS, MEIRELES, 2011). Once they are readily degraded, especially during processing and storage of food products, color measurement is important to evaluate food products quality, what can be conducted by visual inspection, trained inspectors using special vocabulary to give the color description, or instruments, such as colorimeters and spectrophotometers. Instrumental measurements have been used in the food industry and provide quantitative measurement, in a more objective way, by simulating the human perception of color under particular conditions as illumination and observer angle, with the use of color indices (or spaces) (WU AND SUN, 2013). Color spaces, although not providing a precise color definition, are useful for evaluating color differences and changes during processing and storage (WROLSTAD, DURST, LEE, 2005). CIELab is the most used color space (WU AND SUN, 2013), in which color is numerically expressed by L*, a* and b* values, lightness and chromaticity coordinates, green–red and blue–yellow color components, respectively. Difference between colors can be described by ΔE values, comprising the total color difference (OBÓN, CASTELLAR, ALACID, 2009). After 112 days of storage, steam extracted jabuticaba juice WNAS lightness values increased around 3%, while green–red and blue–yellow color components values decreased around 67% and 55%, respectively (Table 10). These results indicate that the juice became slightly darker, substantially less reddish and more bluish, losing its characteristic purple coloration. Similar results were described by INADA et al. (2018b) for jabuticaba juice during 90 days of storage at 25 ºC and by REIN AND HEINONEN (2004) for strawberry, raspberry, lingonberry, and cranberry juices after 103 days of storage.

Table 10. Instrumental color of steam extracted jabuticaba juice with no added sucrose during storage at 25 ºC. Days of storage Instrumental colora 0 112 L* 22.44 ± 0.25 23.16 ± 0.06* a* 11.73 ± 0.94 3.92 ± 0.14* b* 3.32 ± 0.35 1.50 ± 0.01* Results are expressed as mean ± standard deviations of three process replicates; means with * are significantly different (One-way ANOVA test followed by Dunnett post hoc test; p < 0.05); aCIELab color space, L* [black (0) to white (100)], a* [green (-) to red (+)] and b* [blue (-) to yellow (+).

Juice total color difference was distinguishable (1.5<ΔE<5) to the human eye already after the first 14 days of storage at 25 ºC and became evident (ΔE>5) to the human eye at day 42, showing a progressive color changing during storage (Figure 21) (OBÓN, CASTELLAR, ALACID, 2009). Positive correlations were observed between total anthocyanins content (sum of D3G and C3G contents) and a* values (r = 0.9122, p < 0.0001) and b* values (r = 0.8837, p < 0.0001) values, indicating that anthocyanins degradation was accompanied by reduction of these chromaticity coordinates values. A negative correlation was also observed between total anthocyanins content and total color difference (r = -0.8963, p < 0.0001), indicating that these variables were inversely associated, as anthocyanins degradation was accompanied by accentuation of ΔE values. Thus, color change could be related to the degradation of anthocyanins observed during juice storage, and was mainly associated to changes in a* and b* chromaticity coordinates values. Total color difference of juices WNAS stored at 40, 50 and 60 ºC followed the same pattern as the juice stored at 25 ºC (Figure 21), with more pronounced changes at the beginning of the storage period. Despite strong correlations between anthocyanin content and chromaticity coordinates and total color difference, juice still presented purple color even after the almost complete degradation of anthocyanins. This fact could be explained by the relation of anthocyanins degradation and color fading rates, which, according to QIAN et al. (2017), seemed to be different, being anthocyanin degradation faster than color fading.

) 8 E D

(

e c

n e

r 6 e f f

i d

o l 4 o c

l a t o T 2

0 0 7 14 21 28 35 42 49 56 63 70 77 84 91 98 105 112 Storage time (days)

Figure 21. Total color difference (ΔE) of steam extracted jabuticaba juice with no added sucrose over the storage time at 25 ºC (), 40 ºC (), 50 ºC () and 60 ºC ().

Although no significant changes were observed in the gallic acid (GA) content of juice WNAS stored at 25 ºC, except at day 14, gallic acid contents of juices WNAS stored under higher temperatures increased significantly from day 7, in a progressive way, which was more pronounced in juices stored at higher temperatures (Figure 22). Contents at the end of the storage period at 40 ºC, 50 ºC and 60 ºC were 1.4-, 1.9- and 3.3-fold higher than at the beginning.

21 * y = 0.3034x + 5.935 *

* 15

n * e ) t y = 0.1198x + 5.349 L n * o

m 12 c

0 * d i

0 c 1 a /

g * * c

i 9 y = 0.018x + 4.669 l

m l y = 0.0414x + 5.340 (

a * * * G *

3 *

0 0 7 14 21 28 35 42 49 56 63 70 77 84 91 98 105 112

Storage time (days)

Figure 22. Gallic acid contents of steam extracted jabuticaba juice with no added sucrose over the storage time at 25 ºC (), 40 ºC (), 50 ºC () and 60 ºC (); results are expressed as mean ± standard deviations of three process replicates, on fresh weight basis; means with * are significantly different from day 0 (One-way ANOVA test followed by Dunnett post hoc test; p < 0.05).

By plotting the gallic acid content at all four temperatures against storage time, the highest slope was observed at 60 ºC (0.3033 mg GA/100 mL.day, R2 = 0.9943), followed by those at 50 ºC (0.1198 mg GA/100 mL.day, R2 = 0.9268), 40 ºC (0.0414 mg GA/100 mL.day, R2 = 0.8927) and 25 ºC (0.0180 mg GA/100 mL.day, R2 = 0.4631). These results suggest that the reaction(s) that led to the formation of this compound was(were) favored by temperature. The increase in gallic acid contents may have been a result of the thermal degradation of delphinidin, as well as the thermal hydrolysis of gallotannins. Gallic acid is known as a degradation product of delphinidin, which, in its turn, is a compound that present low thermal stability (SINELA et al., 2017). Also, gallic acid can be released by the hydrolysis of gallotannins during heating (GONZÁLEZ, TORRES, MEDINA, 2010; TERÁN-HILARES et al., 2017). We can suppose that in our study the second reaction would be more relevant to the increase in gallic acid content once delphinidin-3-O- glucoside contents found in the juices were relatively low and would not be enough to lead

to an increase of this proportion. On the other hand, considerable amounts of gallotannins have already been described in jabuticaba fruit (PLAZA et al., 2016). Formation of gallic acid in jabuticaba juices followed a zero-order kinetic model, meaning that it was independent of the reagent concentration. Therefore, zero-order reactions are normally related to reactions in which the amount of product formed is much smaller than the amount of the precursors present (BOEKEL, 2008). Rates constants (k) values increased by 24-fold between the lowest (0.0126 mg/100 mL.day) and the highest temperatures (0.3033 mg/100 mL.day), indicating that the storage temperature had a strong influence on gallic acid formation. Temperature dependence of the rate of formation was so modeled by the Arrhenius equation, which was graphically presented (Figure 23) and showed linear behavior, proving the conformity of the model. An activation energy value of 75.45 kJ/mol was calculated. To the best of our knowledge, this was the first time that activation energy of gallic acid formation was investigated.

y = -9074x + 25.96 -1 R² = 0.9921

) 1 -

s -2 a i d (

-3 n

l -4

-5 0.0028 0.0030 0.0032 0.0034

1/T (K)

Figure 23. Arrhenius plot for formation of gallic acid in steam extracted jabuticaba juice with no added sucrose (WNAS).

An expressive decrease in ellagic acid content of juice WNAS stored at 25 ºC was observed from day 28 (Figure 24). Its content, however, seemed to have stabilized from this point forward. Among juices WNAS stored at 40, 50 and 60 ºC, ellagic acid contents have undergone different modifications over the storage period. At 40 ºC, initially, there was an increase of 23%. Then, the content started to decrease until day 14. From this 82

moment on, the content remained similar at the following storage times. At 50 ºC, a significant decrease was observed only at day 21. At the highest storage temperature, on the other hand, a significant change in the content was exclusively observed at day 7, comprising a 2.3-fold increase.

12.0 12.0 10.0 10.0

L 8.0 L 8.0 m

0 0 0 6.0 * 0 1

6.0 / 1 / g g m 4.0 * * m 4.0 * * * * * * * * * * 2.0 2.0 0.0 0.0 0 7 14 21 28 35 42 0 14 28 42 56 70 84 98 112 Storage time (days) Storage time (days)

10 12.0 * 10.0 L 8 m

0 L 8.0 0 6 m

1 / 0 g * 0 6.0 1 / m * * * 4 g m 4.0 2 2.0 0 0.0 0 7 14 21 28 35 42 0 7 14 21 28 35 42 Storage time (days) Storage time (days)

Figure 24. Ellagic acid contents of steam extracted jabuticaba juice with no added sucrose over the storage time at 25 ºC (), 40 ºC (), 50 ºC () and 60 ºC (); results are expressed as mean ± standard deviations of three process replicates, on fresh weight basis; means with * are significantly different from day 0 (One-way ANOVA test followed by Dunnett post hoc test; p < 0.05).

In general, these results suggest a tendency of ellagic acid contents to rise at the first days of storage and then to decline. This could be related to the balance between reactions rates of formation and degradation of this compound throughout storage, that is how faster or slower was the hydrolysis of ellagitannin to ellagic acid (HAGER, HOWARD, PRIOR, 2010) in comparison to its action as an antioxidant (HÄKKINEN et al., 2000). Therefore, it is supposed that, at the beginning of storage period, hydrolysis of

ellagitannins presented in the juices overcame ellagic acid oxidation. From some point on, it seems that formation and degradation reactions rates were equivalent. All juices, except for the one stored at 60 ºC, presented final ellagic acid contents lower than the initial contents. Contrary to our results, HAGER, HOWARD, PRIOR (2010) and INADA et al. (2018b) observed increases in ellagic acid contents of blackberry juice, over 6 months, and jabuticaba juice, during 90 days, respectively, both juices were stored at 25 ºC. A considerable decrease (50%) in total phenolic compounds content of jabuticaba juice was observed after 112 days of storage at 25 ºC, similar to that observed in juice stored at 40 ºC after only 42 days (about 51%). Meanwhile, the storage for 42 days at 50 ºC led to a decrease of around 27%. The storage at 60 ºC had no impact on final content of total phenolic compounds (Figure 25).

25 25

20 20 * * L L m m

15 15 0 * 0 * 0 * * 0 * * * * 1 * * 1 / * * / g 10 g 10 m m

5 5

0 0 0 14 28 42 56 70 84 98 112 0 7 14 21 28 35 42

Storage time (days) Storage time (days)

25 25 20 20 * L * L m m

* 15 * * 15

0 * * 0 0 0 1 1 / / g 10 g 10 m m 5 5

0 0 0 7 14 21 28 35 42 0 7 14 21 28 35 42

Storage time (days) Storage time (days)

Figure 25. Total phenolic compounds contents of steam extracted jabuticaba juice with no added sucrose (WNAS) over the storage time at 25 ºC (), 40 ºC (), 50 ºC () and 60 ºC (); results are expressed as mean ± standard deviations of three process replicates, on fresh weight basis; means with * are significantly different from day 0 mean (One-way ANOVA test followed by Dunnett post hoc test; p < 0.05).

These results may be explained by numerous degradation reactions of phenolic compounds, such as deglycosylation, ring cleavage, decarboxylation or hydrolysis, during juice storage period, which may take place simultaneously, consequently leading to the formation of different ones. Beyond that, the degradation of a single compound can result in the formation of more than one degradation product, which could even be degraded into other compounds. The scheme in Figure 26 illustrates possible routes, precursors and products, considering the main phenolic compounds identified in the jabuticaba juices during storage at all temperatures and previous literature data. Cyanidin and delphinidin (aglycone forms) are known as degradation products from the loss of the sugar moiety of the corresponding anthocyanins. These compounds, in its turn, are susceptible to two possible scissions, leading to the formation of protocatechuic acid, phloroglucinaldeyde and/or gallic acid. Protocatechuic acid results from the cleavage of cyanidin B ring, while gallic acid from the cleavage of delphinidin B ring and phloroglucinaldeyde from the cleavage of A ring of both anthocyanins. Furthermore, gallic acid may loose its carboxyl group, leading to the formation of pyrogallol. Cyanidin, delphinidin, phloroglucinaldeyde and pyrogallol were not detected, whereas protocatechuic acid was identified but not quantified due to the co-elution with other peaks. Taking into account the absence of aglycone forms and the presence of degradation products, it is possible to suppose that deglycosylation is the rate determining step, that is, the slowest step in the anthocyanins degradation pathway. It means that the reaction of cyanidin and delphinidin formation requires much more energy to proceed than the subsequent reaction (cleavage of B ring). Meanwhile, reaction of cleavage of A ring seems not be taking place in the juice. Gallic and ellagic acids, known as degradation products of gallotannins and ellagitannins, respectively, were observed. Hydrolysable tannins have not been evaluated in the present study.

1 2 B Protocatechuic acid (3,4-dihydroxybenzoic acid)

3 Cyanidin Cyanidin-3-O-glucoside

A Phloroglucinaldeyde 3 (2,4,6-trihydroxybenzaldehyde)

1 2 4 Pyrogallol B (1,2,3-trihydroxybenzene)

Gallic acid Delphinidin Delphinidin-3-O-glucoside (3,4,5-trihydroxybenzoic acid) 5 1 Deglycosylation

Gallotannins 2 Cleavage of B ring

3 Cleavage of A ring

4 Decarboxylation

5 5 Hydrolysis Ellagitannins

Ellagic acid

Figure 26. Degradation/formation pathways of anthocyanins, hydrolysable tannins and related phenolic compounds in jabuticaba juices during storage (CABRITA et al., 2014; GONZÁLEZ, TORRES, MEDINA, 2010; HAGER, HOWARD, PRIOR,2010; SINELA et al., 2017; TERÁN-HILARES et al., 2017). Compounds marked with the symbols and were detected and not detected, respectively, while compounds in dashed boxes were no investigated.

86 Antioxidant activity values of juice WNAS stored at 25 ºC showed a decrease of about 18% and 25% on FRAP and TEAC values, respectively, after 112 days, while no change was observed for Folin-Ciocalteu values (Figure 27). The divergent behavior of antioxidant activity evaluated by Folin-Ciocalteu, FRAP and TEAC may have resulted from the different mechanisms of each assay. There are two main ways by which antioxidant compounds deactivate free radicals: hydrogen atom transfer (HAT) and single electron transfer (SET). These mechanisms can occur individually or simultaneously, depending on the antioxidant structure and properties, yet, HAT and SET almost always occur together in foods. Because of that, assays for the determination of antioxidant activity can be HAT-based, SET-based or even based on both HAT and SET mechanisms. FRAP assay works via SET mechanisms, while TEAC and Folin-Ciocalteu assays via both HAT and SET mechanisms (PRIOR et al., 2005). In addition, differences in Folin-Ciocalteu values in comparison to FRAP and TEAC values, could also be explained by the lack of specificity of the former assay (SÁNCHEZ- RANGEL et al., 2013). Its response may have suffered interference from non-phenolic compounds, leading to a possible overestimated evaluation. Total phenolic compounds content showed a substantial decrease after 112 days, as already mentioned, mainly driven by the reduction in D3G, C3G and ellagic acid contents. Changes in antioxidant activity values by FRAP, TEAC and Folin-Ciocalteu, however, were not proportional to this decrease. Additionally, apart from the comparison between day 0 and day 112, antioxidant activity values over the storage period showed an unstable behavior, independently of the methods for the determination of antioxidant activity employed, although, results of each method have shown a specific fluctuation (Figure 27). The modifications observed in total phenolic compounds content did not follow what happened to antioxidant activity throughout this same period. These results indicate that both total phenolic compounds content and their types account to antioxidant activity response. Seeing that, they could not be explained just based on the total phenolic compounds content. As stated by RICE-EVANS, MILLER, PAGANGA (1996), there is a relationship between phenolic compounds chemical structure and their antioxidant activity, that is, each compound has a specific antioxidant activity. The phenolic compounds content had a less expressive impact on juice antioxidant activity than the change in its profile. Indeed, PLAZA et al. (2016) observed that C3G, despite being the phenolic compound in greater concentration in jabubitaca peel, was not the major contributor to its antioxidant activity, whereas ellagitannins and gallotannins, although found in lower concentrations, were the class with higher contribution. Furthermore, it is also important to highlight that jabuticaba juice is a complex

matrix, containing several phenolic compounds non-identified in the present study that could have been degraded or formed during the storage, which could have also contributed to these antioxidant activity results.

5.0 A

L 4.5 m

0 0

1 4.0 * / *

+ * 2

e F

l 3.5 o

m m 3.0

2.5 0 14 28 42 56 70 84 98 112 Storage time (days)

4.0 B L

m 3.0

0 * 0 *

1 *

/ * *

x o l 2.0

o * r T

l o

m 1.0

0.0 0 14 28 42 56 70 84 98 112

Storage time (days)

350.0 C L m 300.0 0 0 1 /

E A G 250.0 * g m

200.0 0 14 28 42 56 70 84 98 112

Storage time (days)

Figure 27. FRAP (A), TEAC (B) and Folin-Ciocalteu (A) values of steam extracted jabuticaba juice with no added sucrose (WNAS) over the storage time at 25 ºC; results are expressed as mean ± standard deviations of three process replicates, on fresh weight basis; means with * are significantly different from day 0 mean (One-way ANOVA test followed by Dunnett post hoc test; p < 0.05). 88

6.3.2 Microbial and sensory qualities

Fruit juices are nutrient-rich beverages, presenting both carbohydrates and nitrogen compounds, with high moisture content and, normally, low acidity. Due to it, this kind of product is very liable to be spoiled by microorganism, rapidly fermenting. Thus, its preservation, especially in long-term storage, is challenging and requires careful processing, packaging and storage. After properly produced, fruit juices are, in most cases, subjected to preservation techniques to prevent its microbial deterioration and, sometimes, also added of chemical preservatives or maintained under refrigeration (ASHURST et al., 2017; MUSHTAQ, 2018; RAJAURIA AND TIWARI, 2018). In view of this, a periodic monitoring of microbial quality of juice WNAS was carried out (Table 11). Juice remained microbiologically stable for 168 days at 25 ºC and low numbers of microorganisms were detected during its storage. These results show that the steam extraction followed by hot- filling technique was able to minimize microbial spoilage of the final product, extending its shelf-life without requiring neither the employment of any preservation technology associated nor the addition of chemical preservatives. According to ASHURST et al., 2017, most soft drinks are not completely sterile and may contain very low levels of microorganisms, what do not pose a risk to the shelf life of the product, and, if product is properly pasteurized, there will be no microbiological deterioration during storage. Moreover, these results also suggest that the juice presents unfavorable conditions for the growth of microorganisms, such as its low pH.

Table 11. Microbiological analysis of steam extracted jabuticaba juice with no added sucrose during storage. Days of storage Reference Microorganism 0 56 112 168 values1 Thermotolerant coliforms Absence Absence Absence Absence Absence (MPN/50mL) Salmonella sp. Absence Absence Absence Absence Absence (25mL) Heterotrophic bacteria 2.0 x 10 < 1.0 < 1.0 x 10 < 1.0 x 10 ----- (CFU/mL)a Yeasts and molds 9.0 x 10 < 1.0 < 1.0 x 10 < 1.0 x 10 ----- (CFU/mL) Lactic acid bacteria < 1.0 x 10 < 1.0 < 1.0 x 10 < 1.0 x 10 ----- (CFU/mL)b 1According to Brazilian legislation (RDC nº 12, 02/01/2001). aColony-forming units per milliliter, bMost probable number per 50 milliliters 89

In general, sensory evaluation of juice WNAS at the end of the storage period at 25 ºC evidenced a good sensory stability during 168 days (Table 12). The only attribute that differed significantly was color, which showed lower score. This was, in some way, expected as anthocyanins, compounds responsible for the juice color, were extensively degraded during the storage. However, the difference between means was not so large, corresponding to around 4%, and also much less pronounced than total color difference (instrumental). The purchase intent scores (Table 12) were also equivalent, both corresponding to “maybe would buy, maybe would not buy” at the five-point scale. Therefore, juice sensory characteristics were preserved and probably will not impair on consumers product acceptance in a commercial scale, although color is an important sensory characteristic since it is seen by consumers as a quality indicator. Furthermore, these results suggest that steam extraction was effective in inactivating enzymes naturally present in the fruit, which could have led to physico-chemical changes in the juice during the storage, negatively impacting on its sensory quality. The addition of sucrose did not interfere neither positively nor negatively on juice sensory stability. Similarly to juice WNAS, all sensory attributes, except for color, and purchase intent scores of juice WAS did not show significant differences at the end of the storage period at 25 ºC (Table 12).

Table 12. Sensory evaluation of steam extracted jabuticaba juices with no added sucrose (WNAS) and with added sucrose (WAS) at the beginning (n=118) and the end (n=110) of storage at 25 ºC. Attributes Scores Juice WNAS Juice WAS

Day 0 Day 168 Day 0 Day 168

Sensory acceptancea Overall impression 6.5 ± 1.6 6.6 ± 1.9 7.2 ± 1.4 7.2 ± 1.3 Aroma 6.6 ± 1.6 6.8 ± 1.6 6.6 ± 1.6 7.0 ± 1.5 Color 7.9 ± 1.1 7.6 ± 1.2* 8.0 ± 1.1 7.3 ± 1.4* Flavor 5.6 ± 2.1 5.8 ± 2.2 6.8 ± 1.6 7.1 ± 1.7 Viscosity 7.3 ± 1.5 6.9 ± 1.9 7.5 ± 1.4 7.4 ± 1.5 Purchase intentb 2.9 ± 1.2 3.1 ± 1.2 3.5 ± 0.9 3.7 ± 1.0 Results are expressed as mean ± standard deviation of three process replicates; means of each attribute, in each type of juice, with * are significant different (One-way ANOVA test followed by Dunnett post hoc test; p < 0.05); aNine-point scale (1 = dislike extremely; 2 = dislike very much; 3 = dislike moderately; 4 = dislike slightly; 5 = neither like nor dislike; 6 = like slightly; 7 = like moderately; 8 = like very much; 9 = like extremely). bFive-point scale (1 = I definitely would not buy; 2 = I probably would not buy; 3 = maybe I would buy, maybe I would not buy; 4 = probably I would buy; 5 = I definitely would buy).

CONCLUSION

Steam extraction proved to be reproducible and viable with regard to the production of jabuticaba juice. Overall, the method not only provided a product with acceptable variability of quality aspects, but also with good nutritional, functional and sensorial characteristics. The addition of sucrose improved juice sensory acceptance, contributing to a flavor more appreciated by consumers, while preserving the functional characteristic of the juice. Sugars and organic acids, which contribute to fruit juice sensory characteristics, remained constant during 112 days. These results relate to the unaffected taste in the sensorial analysis of juices stored for 168 days. Additionally, organic acids stability was also important to juice microbiological quality, contributing to unfavorable conditions for the growth of microorganisms during the whole storage period. Despite the almost complete degradation of anthocyanins in the juice and the relevant reduction in ellagic acid and total phenolic compounds contents during storage at 25 ºC, juice still presented, at the end of this period, considerable total phenolic compounds content and antioxidant activity values. Storage experiments at different temperatures provided some insights on the degradation/formation of the main phenolic compounds of steam extracted jabuticaba juice during its storage, such as degradation of anthocyanins and formation of gallic and ellagic acids. In this context, strategies to improve phenolic compounds stability, such as addition of natural conservatives and cold storage, would be of great importance. In summary, steam extracted jabuticaba juice has a potential of commercialization and also represents a way to aid fruit valorization, allowing fruit consumption expansion with regard to fruit harvest period and cultivation place.

REFERENCES

ABE, L. T., LAJOLO, F. M., GENOVESE, M. I. Potential dietary sources of ellagic acid and other antioxidants among fruits consumed in Brazil: Jabuticaba (Myrciaria jaboticaba (Vell.) Berg). Journal of the Science of Food and Agriculture, v. 92, p. 1679–1687, 2012.

AGÇAM, E., AKYILDIZ, A., DÜNDAR, B. Fruit juices: extraction, composition, quality and analysis. Chapter 17, 1st ed., Chennai, Academic Press, 2018, p. 309 – 370.

ALEZANDRO, M. R. et al. Comparative study of chemical and phenolic compositions of two species of jaboticaba: Myrciaria jaboticaba (Vell.) Berg and Myrciaria cauliflora (Mart.) O. Berg. Food Research International, v. 54, p. 468–477, 2013.

AMODIO, M. L., DEROSSI, A., COLELLI, G. Modeling phenolic content during storage of cut fruit and vegetables : A consecutive reaction mechanism. Journal of Food Engenering, v. 140, p. 1–8, 2014.

ANDRÉS-LACUEVA, C. et al. Fruit and vegetable phytochemicals: chemistry, nutritional value, and stability. Chapter 2, 1st ed., New Delhi, Wiley-Blackwell, 2010, p. 53 – 88.

ANJOS, J. B. DOS. Extrator de sucos vegetais a vapor. Comunicado técnico da Embrapa semi-árido. v. 8, 1999, p. 1–3.

APPELT, P. et al. Development and characterization of cereal bars made with flour of jabuticaba peel and okara. Acta Scientiarum. Technology, v. 37, p. 117–122, 2015.

ASHURST, P. The stability and shelf life of food. Chapter 12, 2nd ed., Series in food science, technology and nutrition, n. 297, Woodhead publishing, 2016, p. 348 – 374.

ASHURST, P. R., HARGITT, R. AND PALMER, F. Soft drinks and fruit juice problems solved. 2nd ed., Series in food science, technology and nutrition, Woodhead publishing, 2017.

AUGUSTO, P. E. D., TRIBST, A. A. L., CRISTIANINI, M. Fruit juices: extraction, composition, quality and analysis. Chapter 20, 1st ed., Chennai, Academic Press, 2018, p. 393 – 421.

BALDIN, J. C. et al. Microencapsulated jabuticaba (Mycriaria cauliflora) extract added to fresh sausage as natural dye with antioxidant and antimicrobial acivity. Meat Science, v. 118, p. 15–21, 2016.

BATES, R.P., MORRIS, J.R., CRANDALL, P.G. Principles and practices of small- and

medium-scale fruit juice processing. Chapter 6, 1st ed., Rome, Food and Agriculture Organization, 2001, p. 37 – 58.

BAYSAL, A. H. Fruit juices: extraction, composition, quality and analysis. Chapter 24, 1st ed., Chennai, Academic Press, 2018, p. 463 – 510.

BELSCAK-CVITANOVIC, A. et al. Polyphenols: properties, recovery and applications. Chapter 1, 1st. ed., Chania, Woodhead Publishing, 2018, p.3 – 44.

BENZIE, I. F. F. AND STRAIN, J. J. The ferric reducing ability of plasma (FRAP) as a measure of “antioxidant power”: the FRAP assay. Analytical Biochemistry, v. 239, p. 70– 76, 1996.

BERG, F. VAN DEN et al. Process analytical technology in the food industry. Trends in Food Science & Technology, v. 31, p. 27–35, 2013.

BIGLIARDI, B. AND GALATI, F. Innovation trends in the food industry: The case of functional foods. Trends in Food Science & Technology, v. 31, p. 118–129, 2013.

BOEKEL, M. A. J. S. VAN. Kinetic Modeling of Food Quality: A Critical Review. Comprehensive Reviews in Food Science and Food Safety, v. 7, p. 144–158, 2008.

CABRITA, L., PETROV, V., PINA, F. On the thermal degradation of anthocyanidins :cyanidin. RSC Advances, v. 4, p. 18939–18944, 2014.

CARRILLO, E. et al. Main factors underlying consumers ’ food choice : a first step for the understanding of attitudes toward "healthy eating". Journal of Sensory Studies, v. 26, p. 85– 95, 2011.

CASTAÑEDA-OVANDO, A. et al. Chemical studies of anthocyanins: A review. Food Chemistry, v. 113, p. 859–871, 2009.

CASWELL, H. The role of fruit juice in the diet: an overview. Nutrition Bulletin, v. 34, p. 273–288, 2009.

CAVALCANTI, R. N., SANTOS, D. T., MEIRELES, M. A. A. Non-thermal stabilization mechanisms of anthocyanins in model and food systems — An overview. Food Research International, v. 44, p. 499–509, 2011.

CHEN, Y., YU, L. J., RUPASINGHE, H. P. V. Effect of thermal and non-thermal pasteurisation on the microbial inactivation and phenolic degradation in fruit juice : a mini- review. Journal of the Science of food and Agriculture, v. 93, p. 981–986, 2013.

CHU, N. T. AND CLYDESDALE, F. M. Decomposition of Organic Acids During Processing and Storage. Journal of Milk and Food Technology v. 39, n. 7, p. 477–480, 1976.

CLARKE, M. A., EDYE, L. A., EGGLESTON, G. Sucrose decomposition in aqueous solution, and losses in sugar manufacture and refining. Advances in carbohydrate and biochemistry, v. 52, p. 441–470, 1997.

COULTATE, T. P. Food: the chemistry of its components. Chapter 7, 5th ed., Cambridge, RSC Publishing, 2009, p. 265–311.

CROZIER, A., JAGANATH, I. B., CLIFFORD, M. N. Plant secondary metabolites: occurrence, structure and role in the human diet. Chapter 1, 1st. ed., Chennai, Blackwell Publishing, 2006, p. 1 – 24.

DOWNES, F. P AND ITO, K. Compendium of methods for the microbiological examination of foods. 4th ed., Washington, American Public Health Association, 2001.

DREWNOWSKI, A. et al. Sweetness and food preference. The Journal of Nutrition, v. 142, p. 1142–1148, 2012.

DUARTE, O. AND PAULL, R. E. Exotic fruits and nuts of the new world. Chapter 3, 1st ed., Pondicherry, CABI, 2015, p. 51 – 61.

DWYER, J. T. et al. Is “ Processed ” a four-letter word ? The role of processed foods in achieving dietary guidelines and nutrient recommendations. Advances in Nutrition, v. 3, p. 536–548, 2012.

FAMIANI, F. et al. The organic acids that are accumulated in the flesh of fruits: occurrence, metabolism and factors affecting their contents – a review. Revista Chapingo Serie Horticultura, v. 21, n. 2, p. 97–128, 2015.

FARAH, A. et al. Correlation between cup quality and chemical attributes of Brazilian coffee. Food Chemistry, v. 98, p. 373–380, 2006.

FELLOWS, P. J. Food processing technology: principles and practice. Chapter 4, 4th ed., Series in food science, technology and nutrition, Chennai, Woodhead Publishing, 2017, p. 291–328.

FERRAREZI, A. C., OLBRICH, K., MONTEIRO, M. Consumer interpretation of ready to drink orange juice and nectar labelling. International Journal of Food Science and Technology, v. 48, p. 1296–1302, 2013.

FLOROS, J. D. et al. Feeding the world today and tomorrow: the importance of food science 96

and technology. Comprehensive Reviews in Food Science and Food Safety, v. 9, p. 572– 599, 2007. FORD, C. M. The biochemistry of grape berry. Chapter 4, 1st ed., Bentham Science Publishers, 2012, p. 67–88.

FÜGEL, R., CARLE, R., SCHIEBER, A. Quality and authenticity control of fruit purées preparations and jams — a review. Trends in Food Science and Technology, v. 16, p. 433– 441, 2005.

FULLER, G. W. New food product development: from concept to marketplace. 3rd ed., Boca Raton, CRC Press, 2011.

GANCEL, A.-L. et al. Impact of industrial processing and storage on major polyphenols and the antioxidant capacity of tropical highland blackberry (Rubus adenotrichus). Food Research International, v. 44, p. 2243–2251, 2011.

GARCIA, L. G. C. et al. Preparation of gluten-free noodles incorporated of jabuticaba peel flour. Journal of Food and Nutrition Research, v. 4, n. 2, p. 82–87, 2016.

GÓMEZ, P. L., WELTI-CHANES, J., ALZAMORA, S. M. Hurdle technology in fruit processing. Annual Review of Food Science and Technology, v. 2, p. 447–465, 2011.

GONZÁLEZ, M. J., TORRES, J. L., MEDINA, I. Impact of thermal processing on the activity of gallotannins and condensed tannins from Hamamelis virginiana used as functional ingredients in seafood. Journal of Agricultural and Food Chemistry, v. 58, p. 4274–4283, 2010.

GREGORY III, J. F. Fennema's Food Chemistry. Chapter 7, 4th ed., Boca raton, CRC Press, 2008, p. 441 – 521.

GURAK, D. P. et al. Jaboticaba pomace powder obtained as a co-product of juice extraction: a comparative study of powder obtained from peel and whole fruit. Food Research International, v. 62, p. 786–792, 2014.

HAGER, T. J., HOWARD, L. R., PRIOR, R. L. Processing and storage effects on the ellagitannin composition of processed blackberry products. Journal of Agricultural and Food Chemistry, v. 58, p. 11749–11754, 2010.

HAGERMAN, A. E. The tannin handbook. 2nd ed., Oxford, Miami University, 2002.

HÄKKINEN, S. H. et al. Ellagic acid content in berries : Influence of domestic processing and storage. European Food Research and Technology, v. 212, p. 75–80, 2000.

HE, J. AND GIUSTI, M. M. Anthocyanins: Natural colorants with health-promoting properties. Annual Review of Food Science and Technology, v. 1, p. 163–187, 2010.

HELLSTRÖM, J., MATTILA, P., KARJALAINEN, R. Stability of anthocyanins in berry juices stored at different temperatures. Journal of Food Composition and Analysis, v. 31, p.12–19, 2013.

HOU, Z. et al. Identification of anthocyanins isolated from black rice (Oryza sativa L.) and their degradation kinetics. Food Research International, v. 50, p. 691–697, 2013.

INADA, K. O. P. et al. Screening of the chemical composition and occurring antioxidants in jabuticaba (Myrciaria jaboticaba) and jussara (Euterpe edulis) fruits and their fractions. Journal of Functional Foods, v. 17, p. 422–433, 2015.

INADA, K. O. P. et al. High hydrostatic pressure processing affects the phenolic profile, preserves sensory attributes and ensures microbial quality of jabuticaba (Myrciaria jaboticaba) juice. Journal of the Science of Food and Agriculture, v. 98, p. 231–239, 2018a.

INADA, K. O. P. et al. Jabuticaba (Myrciaria jaboticaba) juice obtained by steam-extraction: phenolic compound profile, antioxidant capacity, microbiological stability, and sensory acceptability. Journal of Food Science and Technology, v. 55, n. 1, p. 52–61, 2018b.

JAEGER, S. R. et al. Polyphenol-rich beverages: insights from sensory and consumer science. Journal of the Science of Food and Agriculture, v. 89, p. 2356–2363, 2009.

JHAM, G. N. et al. Comparions of GC and HPLC for quantification of organic acids in two jabocaticaba (Myrciaria) fruit varieties. Química Nova, v. 30, n. 7, p. 1529–1534, 2007.

KHOO, H. E. et al. Anthocyanidins and anthocyanins: colored pigments as food, pharmaceutical ingredients, and the potential health benefits. Food & Nutrition Research, v. 61, p. 1–21, 2017.

KONG, F. AND SINGH, R. P. The stability and shelf life of food. Chapter 2, 2nd ed., Series in food science, technology and nutrition, n. 297, Woodhead publishing, 2016, p. 43 – 76.

KOUBAA, M. et al. Fruit juices: extraction, composition, quality and analysis. Chapter 22, 1rst ed., Chennai, Academic Press, 2018, p. 437 – 449.

LAAKSONEN, O. et al. Sensory quality and compositional characteristics of blackcurrant juices produced by different processes. Food Chemistry, v. 138, p. 2421–2429, 2013.

LEITE-LEGATTI, A. V. et al. Jaboticaba peel: antioxidant compounds , antiproliferative and antimutagenic activities. Food Research International, v. 49, p. 596–603, 2012. 98

LESSCHAEVE, I. AND NOBLE, A. C. Polyphenols: factors influencing their sensory properties and their effects on food and beverage preferences. American Journal of Clinical Nutrition, v. 81, p. 330–335, 2005.

LIMA, A. DE J. B. et al. Sugars, organic acids, minerals and lipids in jabuticaba. Revista Brasileira de Fruticultura, v. 33, n. 2, 2010.

LING, B. et al. Kinetics of food quality changes during thermal processing a review. Food and Bioprocess Technology, v. 8, p. 343–358, 2015.

LOPES, M. L. M. et al. Original article grape juice obtained using steam extraction and other small-scale extraction methods: phenolic content, antioxidant capacity and stability during storage. International Journal of Food Science and Technology, v. 51, p. 1696–1702, 2016.

LÓPEZ-NICOLÁS, J. M. AND GARCÍA-CARMONA, F. Fruit and vegetable phytochemicals: chemistry, nutritional value, and stability. Chapter 4, 1st ed., New Delhi, Wiley-Blackwell, 2010, p. 101 – 129.

LOZANO, J. E. Fruit manufacturing: scientific basis, engeneering properties, and deteriorative reactions of technological importance. Chapter 2, 1st ed., Springer, 2006, p. 21 – 54.

MÄKILÄ, L. et al. Stability of hydroxycinnamic acid derivatives, flavonol glycosides, and anthocyanins in black currant juice. Journal of Agricultural and Food Chemistry, v. 64, 2016.

MAN, C. M. D. The stability and shelf life of food. Chapter 6, 2nd ed., Series in food science, technology and nutrition, n. 297, Woodhead publishing, p. 171 – 198, 2016.

MANACH, C. et al. Polyphenols: food sources and bioavailability. The American Journal of Clinical Nutrition, v. 79, p. 727–747, 2004.

MERCALI, G. D. et al. Evaluation of non-thermal effects of electricity on anthocyanin degradation during ohmic heating of jaboticaba (Myrciaria cauliflora) juice. Food Chemistry, v. 171, p. 200–205, 2015.

MERTZ, C. et al. Characterization and thermal lability of carotenoids and vitamin C of tamarillo fruit (Solanum betaceum Cav.). Food Chemistry, v. 119, p. 653–659, 2010.

MICHELETTI, J. et al. The addition of jaboticaba skin flour to muffins alters the physicochemical composition and their sensory acceptability by children. Brazilian Journal of Food Techology, v. 21, 2018.

MIKULIC-PETKOVSEK, M. et al. Composition of sugars, organic acids , and total phenolics in 25 wild or cultivated berry species. Journal of Food Science, v. 0, n. 0, p. 1–7, 2012.

MUSHTAQ, M. Fruit juices: extraction, composition, quality and analysis. Chapter 8, 1rst ed., Chennai, Academic Press, 2018, p. 131 – 159.

NEVES, N. D. A. et al. Flavonols and ellagic acid derivatives in peels of different species of jabuticaba (Plinia spp.) identified by HPLC-DAD-ESI/MSn. Food Chemistry, v. 252, p. 61– 71, 2018.

NGADI, M. O., BAJWA, S. S. S., ALAKALI, J. Food biochemistry and food processing. Chapter 39, 2nd ed. New Delhi, Wiley-Blackwell, 2012, p. 746 – 763.

NOWICKA, P. AND WOJDY, A. Stability of phenolic compounds, antioxidant activity and colour through natural sweeteners addition during storage of sour cherry puree. Food Chemistry, v. 196, p. 925–934, 2016.

OBÓN, J. M., CASTELLAR, M. R., ALACID, M. Production of a red–purple food colorant from Opuntia stricta fruits by spray drying and its application in food model systems. Journal of Food Engineering, v. 90, p. 471–479, 2009.

OLIVEIRA, L. C., ALENCAR, N. M. M., STEEL, C. J. Improvement of sensorial and technological characteristics of extruded breakfast cereals enriched with whole grain wheat flour and jabuticaba (Myrciaria cauliflora) peel. LWT - Food Science and Technology, v. 90, p. 207–214, 2018.

ORTEGA-RIVAS, E. AND SALMERÓN-OCHOA, I. Nonthermal food processing alternatives and their effects on taste and flavor compounds of beverages. Critical Reviews in Food Science and Nutrition, v. 54, p. 190–207, 2014.

PATRAS, A. et al. Effect of thermal processing on anthocyanin stability in foods; mechanisms and kinetics of degradation. Trends in Food Science and Technology Technology, v. 21, p. 3–11, 2010.

PEREIRA, L. D. et al. Polyphenol and ellagitannin constituents of jabuticaba (Myrciaria cauliflora) and chemical variability at different stages of fruit development. Journal of Agricultural and Food Chemistryr, v. 65, n. 6, p. 1209–1219, 2017.

PEREIRA, M. et al. Morphologic and molecular characterization of Myrciaria spp species. Revista Brasileira de Fruticultura, v. 27, n. 3, p. 507–510, 2005.

PERON, D. V., FRAGA, S., ANTELO, F. Thermal degradation kinetics of anthocyanins extracted from juçara (Euterpe edulis Martius) and “Italia” grapes (Vitis vinifera L.), and the effect of heating on the antioxidant capacity. Food Chemistry, v. 232, p. 836–840, 2017.

100

PETRUZZI, L. et al. Thermal treatments for fruit and vegetable juices and beverages: a literature overview. Comprehensive Reviews in Food Science and Food Safety, v. 16, n. 4, p. 668–691, 2017.

PLAZA, M. et al. Characterization of antioxidant polyphenols from Myrciaria jaboticaba peel and their effects on glucose metabolism and antioxidant status: A pilot clinical study. Food Chemistry, v. 211, p. 185–197, 2016.

PRIOR, R. L. AND WU, X. Analysis of antioxidant-rich phytochemicals. Chapter 5, 1st ed., Noida, Wiley-Blackwell, 2012, p. 149 – 180.

PRIOR, R. L., WU, X., SCHAICH, K. Standardized methods for the determination of antioxidant capacity and phenolics in foods and dietary supplements. Journal of Agricultural and Food Chemistry, v. 53, p. 4290–4302, 2005.

PRIYADARSHINI AND PRIYADARSHINI, Fruit juices: extraction, composition, quality and analysis. Chapter 2, 1rst ed., Chennai, Academic Press, p. 15 – 32, 2018.

QIAN, B.-J. et al. The effects of gallic/ferulic/caffeic acids on colour intensification and anthocyanin stability. Food Chemistry, v. 228, 2017, p. 526–532.

QUITMANN, H., FAN, R., CZERMAK, P. Acidic organic compounds in beverage, food, and feed production. Advances in Biochemical Engineering/Biotechnology, v. 143, p. 91–141, 2014.

RAJAURIA, G. AND TIWARI, B. K. Fruit juices: extraction, composition, quality and analysis. Chapter 1, 1rst ed., Chennai, Academic Press, 2018, p. 3 – 14.

RATHORE, A. S. AND KAPOOR, G. Implementation of quality by design towards processing of food products. Preparative Biochemistry and Biotechnology, v. 47, n. 5, p. 435–440, 2017.

RAWSON, A. et al. Effect of thermal and non thermal processing technologies on the bioactive content of exotic fruits and their products: Review of recent advances. Food Research International, v. 44, p. 1875–1887, 2011.

RE, R. et al. Antioxidant activity applying an improved ABTS radical cation decolorization assay. Free Radical Biology & Medicine, v. 26, n. 9–10, 1231–1237, 1999.

REIN, M. J. AND HEINONEN, M. Stability and enhancement of berry juice color. Journal of Agricultural and Food Chemistry, v. 52, p. 3106–3114, 2004.

REYNERTSON, K. A. et al. Bioactive Depsides and Anthocyanins from Jaboticaba (Myrciaria cauliflora). v. 57, p. 1228–1230, 2006.

101

RICE-EVANS, C. A., MILLER, N. J., PAGANGA, G. Structure-antioxidant activity relationships of flavonoids and phenolic acids. Free Radical Biology and Medicine, v. 20, n. 7, p. 933–956, 1996.

ROMANO, K. R., ROSENTHAL, A., DELIZA, R. How do Brazilian consumers perceive a non-traditional and innovative fruit juice? An approach looking at the packaging. Food Research International, v. 74, p. 123–130, 2015.

SALOMÃO, L. C. et al. Exotic fruits: reference guide. 1st ed., Chennai, Academic Press, 2018, p. 237 – 244.

SALTVEIT, M. E. Fruit and vetegetables phytochemicals: chemistry and human health. Chapter 5, 2nd ed., v. 1, Noida, Wiley-Blackwell, 2018, p. 115 – 123.

SÁNCHEZ-RANGEL, J. C. et al. The Folin–Ciocalteu assay revisited: improvement of its specificity for total phenolic content determination. Analytical methods, v. 5, p. 5990–5999, 2013.

SANDI, D. et al. Hunter color dimensions, sugar content and volatile compounds in pasteurized yellow passion fruit juice (Passiflora edulis var. flavicarpa) during storage. v. 47, n. 2, p. 233–245, 2004.

SARKAR, S. AND COSTA, A. I. A. Dynamics of open innovation in the food industry. Trends in Food Science and Technology, v. 19, p. 574–580, 2008.

SARKAR, D. AND SHETTY, K. Metabolic stimulation of plant phenolics for food preservation and health. Annual Review of Food Science and Technology, v. 3, p. 395–415, 2014.

SAUCEDA, A. E. Q. Fruit and vetegetables phytochemicals: chemistry and human health. Chapter 6, 2nd ed., v. 1, Noida, Wiley-Blackwell, 2018, p. 115 – 123.

SCHERER, R. et al. Validation of a HPLC method for simultaneous determination of main organic acids in fruits and juices. Food Chemistry, v. 135, p. 150–154, 2012.

SHAHIDI, F. AND ZHONG, Y. Measurement of antioxidant activity. Journal of Functional Foods, v. 18, p. 757–781, 2015.

SILVA, C. E. DE F. AND ABUD, A. K. DE S. Tropical fruit pulps: processing, product standardization and main control parameters for quality assurance. Brazilian Archives of Biology and Technology, v. 60, p. 1–19, 2017.

SINELA, A. et al. Anthocyanins degradation during storage of Hibiscus sabdariffa extract and evolution of its degradation products. Food Chemistry, v. 214, p. 234–241, 2017. 102

SINGLETON, V., ORTHOFER, R., LAMUELA-RAVENTÓS, R. Analysis of total phenols and other oxidation substrates and antioxidants by means of Folin-Ciocalteau reagent. Methods in Enzymology, v. 299, 152–178, 1999.

SIRÓ, I. et al. Functional food. Product development, marketing and consumer acceptance — A review. Appetite, v. 51, p. 456–467, 2008.

SOUZA, V. R. DE et al. Determination of the bioactive compounds, antioxidant activity and chemical composition of Brazilian blackberry, red raspberry, strawberry, blueberry and sweet cherry fruits. Food Chemistry, v. 156, p. 362–368, 2014.

SUI, X., BARY, S., ZHOU, W. Changes in the color, chemical stability and antioxidant capacity of thermally treated anthocyanin aqueous solution over storage. Food Chemistry, v. 192, p. 516–524, 2016.

SWAMY, G. J., MUTHUKUMARAPPAN, K., ASOKAPANDIAN, S. Fruit juices: extraction, composition, quality and analysis. Chapter 23, 1st ed., Chennai, Academic Press, 2018, p. 451 – 462.

TAYLOR, B. Chemistry and technology of soft drinks and fruit juices. Chapter 3, 3rd ed., Pondicherry, Wiley-Blackwell, 2016, p. 31 – 64.

TEIXEIRA, et al Postharvest bilogy and technology of tropical and subtropical fruits volume. Chapter 11, 1st ed., v. 3, Series in food science, technology and nutrition, n. 208, Woodhead publishing, 2011, p. 246 – 274.

TERÁN-HILARES, R. et al. Enhanced antioxidant properties of tara (Caesalpinia spinosa) gallotannins by thermal hydrolysis and its synergistic effects with α-tocopherol, ascorbyl palmitate, and citric acid on sacha inchi (Plukenetia volubilis) oil. Journal of Food Process Engineering, v. 41, p. 1–10, 2018.

TOLA, Y. B. AND RAMASWAMY, H. S. Food biochemistry and food processing. Chapter 38, 2nd ed., New Delhi, Wiley-Blackwell, 2012, p. 725 – 745.

VERMERRIS, W. AND NICHOLSON, R. Phenolic compounds biochemistry. Chapter 1, 1rst ed., Netherlands, Springer, 2006, p. 1 – 34.

VIDIGAL, M. C. T. R. et al. Effect of a health claim on consumer acceptance of exotic Brazilian fruit juices : Açaí (Euterpe oleracea Mart.), Camu-camu (Myrciaria dubia), Cajá (Spondias lutea L.) and Umbu (Spondias tuberosa Arruda). Food Research International, v. 44, p. 1988–1996, 2011.

103

WALKER, R. P. AND FAMIANI, F. Horticultural Reviews. Chapter 8, 1st ed., v. 45, Wiley, 2018, p. 371 – 430.

WEAVER, C. M. et al. Processed foods: contributions to nutrition. American Journal of Clinical NutritionAmerican Journal, v. 99, p. 1525–1542, 2014.

WINGER, R. Food science and technology. Chapter 19, 1st ed., New Delhi, Wiley- Blackwell, 2006, p. 447 – 462.

WROLSTAD, R. E., DURST, R. W., LEE, J. Tracking color and pigment changes in anthocyanin products. Trends in Food Science and Technology, v. 16, p. 423–428, 2005.

WU, D. AND SUN, D. Colour measurements by computer vision for food quality control – A review. Trends in Food Science and Technology, v. 29, p. 5–20, 2013.

WU, S. et al. Metabolite rofiling of jaboticaba (Myrciaria cauliflora) and other dark-colored fruit juices. Journal of Agricultural and Food Chemistry, v. 60, p. 7513–7525, 2012.

WU, S.-B., LONG, C., KENNELLY, E. J. Phytochemistry and health benefits of jaboticaba, an emerging fruit crop from Brazil. Food Research International, v. 54, p. 148–159, 2013.

YAMAMOTO, L. Y. et al. Application of abscisic acid (S-ABA) to cv . Isabel grapes (Vitis vinifera × Vitis labrusca) for color improvement: Effects on color, phenolic composition and antioxidant capacity of their grape juice. Food Research International, v. 77, p. 572–583, 2015.

ZAGO, M. F. C. et al. Jabuticaba peel in the production of cookies for school food: technological and sensory aspects. Ciência e Agrotecnologia, v. 39, n. 6, p. 624–633, 2015.

ZHENG, J. et al. Effects of latitude and weather conditions on contents of sugars, fruit acids, and ascorbic acid in black currant (Ribes nigrum L.) juice. Journal of Agricultural and Food Chemistry, v. 57, p. 2977–2987, 2009.

104