UNIVERSIDADE ESTADUAL PAULISTA FACULDADE DE CIÊNCIAS AGRÁRIAS E VETERINÁRIAS CAMPUS DE JABOTICABAL FIBRA PARA CÃES: EFEITOS SOBRE O PROCESSO DE EXTRUSÃO, DIGESTIBILIDADE, FERMENTAÇÃO MICROBIANA, TEMPO DE RETENÇÃO INTESTINAL E PALATABILIDADE DE RAÇÕES PARA CÃES Mariana Monti Médica Veterinária 2015 UNIVERSIDADE ESTADUAL PAULISTA FACULDADE DE CIÊNCIAS AGRÁRIAS E VETERINÁRIAS CAMPUS DE JABOTICABAL FIBRA PARA CÃES: EFEITOS SOBRE O PROCESSO DE EXTRUSÃO, DIGESTIBILIDADE, FERMENTAÇÃO MICROBIANA, TEMPO DE RETENÇÃO INTESTINAL E PALATABILIDADE DE RAÇÕES PARA CÃES Mariana Monti Orientadora: Profa. Dra. Cecilia Villaverde Coorientador: Prof. Dr. Aulus Cavalieri Carciofi Dissertação de Mestrado apresentado à Faculdade de Ciências Agrárias e Veterinárias –Unesp, Campus de Jaboticabal, como pré-requisito para a obtenção do título de Mestre em Medicina Veterinária (Clínica Médica Veterinária). 2015 DADOS CURRICULARES DO AUTOR MARIANA MONTI – Nascida em 13 de Setembro de 1988, em Ribeirão Preto –SP, graduada em Medicina Veterinária pela Faculdade de Ciências Agrárias e Veterinárias da Universidade Estadual Paulista “Júlio de Mesquita Filho” (Unesp), Campus de Jaboticabal em Dezembro de 2012. Foi bolsista PET veterinária (2009-2010) nesta mesma instituição e bolsista de iniciação científica da Fundação de Amparo à Pesquisa do Estado de São Paulo (2011- 2012) na área de nutrição de cães e gatos com ênfase em imunologia, nutrição e envelhecimento de cães Beagles sob a supervisão do Prof. Aulus Cavalieri Carciofi e da Profa. Márcia de O.S. Gomes. Realizou Estágio Curricular na empresa de alimento para cães e gatos Selecta Pet Care Company e no Serviço de Nutrição Clínica da Universidade Autônoma de Barcelona, Espanha, supervisionado pela Profa. Cecília Villaverde. Fez mestrado na Faculdade de Ciências Agrárias e Veterinárias da UNESP, campus de Jaboticabal, Laboratório de Pesquisa em Nutrição e Doenças Nutricionais na UNESP, Campus Jaboticabal na área de nutrição de cães e gatos com ênfase em fontes de fibras para cães e processo de extrusão. Foi bolsista do CNPQ e teve orientação dos Professores Aulus C Carciofi e Cecília Villaverde. Atualmente é pesquisadora na empresa Premier Pet, traballhando na área de desenvolvimento de produtos. Dedico Àos meus pais Eliana Aparecida Varanda e Rubens Monti, Àos meus pets: Neni, Nâna, Filho, Fio, Tita, Bel, Sara, Sansão e Totó AGRADECIMENTOS Á Deus, pela vontade, coragem, paciência e perseverança para a realização deste trabalho. Àos meus amados pais, Eliana e Rubens, por serem meus verdadeiros orientadores, pelo apoio, amor, presença, influência e determinação na minha formação. À Nat e ao Rê, pelos amados sobrinhos Pedro e Maria Isabel. Ao grande Mestre Prof. Dr. Aulus Cavalieri Carciofi pela orientação, oportunidade, confiança e amizade. Serei sempre grata a tudo, muito obrigada! À querida Profa Cecília Villaverde pela orientação, amizade e experiência proporcionadas a mim durante o mestrado. Á todos os colegas de Barcelona que juntos fazem parte dessa equipe. Ao professor Gener Tadeu Pereira pela ajuda prestada sempre que preciso e competência em seu trabalho. Ao excelente colega de trabalho Michael Gibson, que tanto contribuiu para meu crescimento sobre extrusão e língua inglesa. Aos meus queridos amigos: Aninha, Thailinha, Péti, Fabi (e Jú), Fer, Ferdi, Katiani, Chay + Flávio, Bruna P., Mayara’s, Érico, Lud, Michele, Diego e Claudinha, pela amizade, respeito, e por todos os ótimos momentos vividos no lab nesses anos. Já estou sentindo falta! As minhas grandes amigas Bruna, Raquel e Roberta pela amizade excepcional. Á Elaine, pela amizade, confiança, dedicação e amor verdadeiro aos animais. Aos estagiários: Amanda, Aline, Larissa, Mariana, Lorraine, Fran, Hugo, Thiago, Matheus, Marcela, Adriane e Janine. À Universidade Estadual Paulista, Campus de Jaboticabal e ao Programa de Pós Graduação pela obtenção do título. Agradeço ao CNPQ pela bolsa de estudos durante o mestrado. À Mogiana Alimentos S/A, pela manutenção financeira do Laboratório de Pesquisa em Nutrição e Doenças Nutricionais de Cães e Gatos ¨Prof. Dr. Flávio Prada¨. À DILUMIX e ao Luíz Guilherme Marcondes, pela parceria, financiamento de todo o projeto e confiança em nosso trabalho. Aos cães e gatos do LabNutri, pelos ensinamentos, carinho e companhia. Em especial, ao Filho, gatinho querido adotado por mim que fez parte desta jornada. À todos que, direta e indiretamente, contribuíram para realização desta tese. vii Página SUMÁRIO CAPÍTULO 1- Considerações Gerais 1. INTRODUÇÃO ........................................................................................................ 2 2. REVISÃO DE LITERATURA .................................................................................. 4 Fibra para cães ................................................................................................... 4 Fibra de cana ...................................................................................................... 7 Farelo de trigo ..................................................................................................... 7 Fibra de goiaba ................................................................................................... 8 Fibra e o Processo de Extrusão .......................................................................... 9 Moagem ............................................................................................................ 10 Força de corte (avaliação de textura) ............................................................... 11 3. HIPÓTESES DA DISSERTAÇÃO ........................................................................ 12 4. REFERÊNCIAS .................................................................................................... 13 CAPÍTULO 2- FIBER INFLUENCE ON MACROSTRUCTURE AND PROCESSING TRAITS OF EXTRUDED DIETS FOR DOGS .......................................................... 19 Abstract .................................................................................................................... 21 INTRODUCTION ......................................................................................... 22 MATERIAL AND METHODS ....................................................................... 23 RESULTS .................................................................................................... 31 DISCUSSION .............................................................................................. 38 REFERENCES ............................................................................................ 43 CAPÍTULO 3- GUAVA FIBER: MATERIAL CHARACTERIZATION AND EFFECTS ON DIGESTIBILITY, FERMENTATION, GASTROINTESTINAL TRANSIT TIME AND PALATABILITY IN DOGS .............................................................................. 46 Summary ................................................................................................................... 48 INTRODUCTION ........................................................................................ 49 MATERIAL AND METHODS ...................................................................... 51 RESULTS AND DISCUSSION ................................................................... 61 REFERENCES ........................................................................................... 79 CAPÍTULO 4- FIBER SOURCES AND PARTICLES SIZE ON NUTRIENT DIGESTIBILITY, FERMENTATION PRODUCTS, DIET PALATABILITY, AND GASTROINTESTINAL RETENTION TIME OF DOGS FED KIBBLE DIET...... ........ 87 Abstract ..................................................................................................................... 89 INTRODUCTION ........................................................................................ 90 MATERIAL AND METHODS ...................................................................... 91 RESULTS AND DISCUSSION ................................................................... 99 LITERATURE CITED ................................................................................ 113 viii ix FIBRA PARA CÃES: EFEITOS SOBRE O PROCESSO DE EXTRUSÃO, DIGESTIBILIDADE, FERMENTAÇÃO MICROBIANA, TEMPO DE RETENÇÃO INTESTINAL E PALATABILIDADE DE RAÇÕES PARA CÃES RESUMO-. Existe um grande interesse atual no desenvolvimento de alimentos para cães com alta concentração de nutrientes e baixa densidade energética. A adição de fibra em rações para pet é uma maneira de controlar as calorias do alimento propiciando benefícios, e ao mesmo tempo, promover a utilização mais eficiente das fontes de fibra que não são destinadas à alimentação humana. Desta forma, foram formuladas 8 dietas experimentais: Controle (CO), sem adição de ingredientes fibrosos; dietas com fibra de goiaba (GF3, GF6, GF12), com níveis de inclusão de 3%, 6%, e 12%; dietas com fibra de cana (SC), com 9% de inclusão e dois diferentes tamanho de partícula (grande - SCL e pequeno- SCS) e dietas com farelo de trigo (WB), com 32% de inclusão e dois diferentes tamanhos de partícula (grande- WBL e pequeno- WBS). Objetivou-se neste trabalho avaliar os efeitos de inclusões crescentes de fibra de goiaba, bem como o efeito da fibra de cana e do farelo de trigo moídos em diferentes tamanhos sobre o processo de extrusão, digestibilidade dos nutrientes, fermentação no intestino, tempo de retenção gastrointestinal (TRGI) e palatabilidade das dietas. Foi utilizado o programa SAS para análise estatística e as médias foram comparadas por contrastes polinomiais e ortogonais (P<0,05). No Capítulo 2, a adição da fibra de goiaba resultou em aumento linear da amperagem (P<0,001), temperatura (P<0,001) e pressão (P<0,001) na saída da extrusora. Níveis crescentes de goiaba trouxeram maior implemnto de energia mecânica específica (EME) (P<0,001), redução na expansão radial (ER) (P<0,001), aumento na densidade aparente (DA) e menor cozimento do amido. A fibra de cana, em comparação com a fibra de trigo reduziu a amperagem (P<0,001), EME (P=0,013) e DA, mas aumentou o comprimento específico (P<0,001). A inclusão de fibra em menor tamanho reduziu a amperagem (P<0,001), EME (P<0,001) e DA, mas aumentou a ER (P=0,008). No capítulo 3, as rações CO, GF3, GF6 e GF12 foram fornecidas a 24 Beagles adultos por 15 dias de adaptação e após esse período os animais foram alojados em gaiolas metabólicas para acessar a digestibilidade e fermentação. Os animais receberam por via oral uma pílula contendo 10 marcadores radiopacos para determinação do TRGI. O teste de palatabilidade foi realizado com 38 cães utilizando-se teste versus por comparação. A adição da fibra de goiaba não alterou a ingestão de nutrientes, exceto para fibra dietética total (FDT) (P<0,001). A inclusão de fibra resultou em menor digestibilidade para matéria seca (MS) (P<0,001), matéria orgânica (MO) (P<0,001), proteína (PB) (P<0,001), energia bruta (EB) (P<0,001) e energia metabolizável do alimento (EM) (P<0,001). A fibra de goiaba não alterou a concentração fecal de amônia, ácido láctico, pH fecal e ácidos graxos de cadeia ramificada (AGCR), porém, reduziu a concentração dos ácidos graxos de cadeia curta (AGCC) acético e propiônico (P=0,007 e P=0,006). A inclusão de 6% de goiaba não alterou o TRGI, mas 12% de inclusão resultou em menor TRGI (P=0,046) comparado com a dieta CO. No capítulo 4, as dietas CO, SCL, SCS, WBL e WBS foram fornecidas a 30 Beagles adultos seguindo as mesmas metodologias do estudo com fibra de goiaba. Em comparação com a dieta CO, as dietas com fibra de cana e trigo aumentaram a ingestão de todos os nutrientes (P<0,001) e diminuíram a digestibilidade da MS, MO e ME (P<0,001), sem diferenças para PB, gordura, FDT e amido. Cães alimentados com farelo de trigo, em relação aos alimentados com fibra de cana, obtiveram aumento na concentração de ácido láctico (P<0,001) e diminuição da amônia (P<0,001) e pH (P<0,001) fecal, ácido isovalérico e ACCR total (P<0,001). A inclusão de ambas as fontes de fibra reduziram o TRGI comparado com a dieta CO (P<0,001). Os cães apresentaram preferência alimentar pela dieta sem fibra (P<0,01) sem efeitos claros quanto ao tamanho de partícula. No geral, concluiu-se que a adição de fibras limita o cozimento do amido e aumenta o gasto de energia na extrusão; torna as dietas duras, densas e exigem mais cautela para serem recobertas. A fibra de goiaba, por sua vez não causa alterações na fermentação intestinal e no TRGI de cães até 12% de inclusão. A fibra de cana e trigo com pequenos tamanhos de partículas utilizado nesse estudo não resultaram em melhoras quanto à digestibilidade de dietas com alta fibra e não alteraram o TRGI; entretanto, interferiram na formação de AGCC e preferência alimentar dos animais. Palavras-chave: fibras, nutrição de cães, preferência alimentar, resposta fermentativa, tamanho de partícula, tempo de retenção intestinal. x FIBER FOR DOGS: EFFECTS ON EXTRUSION PROCESSING, NUTRIENTS DIGESTIBILITY, MICROBIAL FERMENTATION, GASTROINTESTINAL RETENTION TIME AND DIET PALATABILITY . ABSTRACT- There is a great interest in developing dog food with high nutrient concentration and low energy density. The addition of fibers to commercial pet food is one way to moderate the energy density of these foods, provide health benefits to dogs and cats and, at the same time, to promote a more efficient utilization of feed resources that are not used in human foods. For this, a total of 8 fiber diets were manufactured: Control (CO), with no fibrous ingredient addition; guava fiber diets (GF3, GF6, GF12), at the inclusion levels of 3%, 6%, and 12%; sugarcane fiber diets (SC), with 9% inclusion and two different particle size (large - SCL and small - SCS) and wheat bran fiber diets (WB), with 32% inclusion and two different particle size (large - WBL and small - WBS). This study aimed to evaluate the effect of growing inclusion of guava fiber, as the effect of sugarcane and wheat bran with different particle size on extrusion processing, nutrients digestibility, intestinal fermentation, gastrointestinal retention time (GIRT) and palatability. The software SAS was utilized to statistical analysis and the meanings were compared by polynomial and orthogonal contrasts (P<0.05). In chapter 2, guava fiber inclusion resulted in linear increase of the amperage (P<0.001), temperature (P<0.001) and pressure (P<0.001) at the extruder die was verified. A higher implementation of specific mechanical energy (SME) (P<0.001), a reduction in radial expansion (P<0.001), and a greater specific piece density was verified for guava fiber supplemented diets. The sugarcane fiber in comparison with wheat bran reduced amperage (P<0.001), SME (P=0.013) and piece density, but increased specific length (P<0.001). The inclusion of fibers with smaller size, reduced the engine amperage (P<0.001), SME (P<0.001) and extruded piece density, but increased the radial expansion (P=0.008). In chapter 3, the diets CO, GF3, GF6 and GF12 were fed to 24 adult Beagle dogs during 15 days of adaptation and afterwards the dogs were housed in cages for fecal collection to assess digestibility and fermentation end products. The dogs received an oral pill containing radiopaque markers to determine gastrointestinal retention time (GIRT). Diet palatability was evaluated by the two-pan test using 38 dogs. The addition of guava fiber did not change nutrient intake except for dietary fiber (P<0.001). Fiber inclusion resulted in lower total tract apparent digestibility for dry matter (P<0.001), organic matter (P<0.001), crude protein (P<0.001), crude energy (P<0.001) and food ME (P<0.001). Guava fiber addition did not change the fecal concentration of ammonia, lactic acid, fecal pH, and branched chain fatty acids but it decreased short chain fatty acid concentrations for acetic and propionic acids (P=0.007 and P=0.006). The inclusion of 6% guava fiber did not alter the gastrointestinal transit time, but 12% inclusion did result in a reduced GIRT (P=0.046) compared to the control diet. The chapter 4, the Diets CO, SCL, SCS, WBL e WBS were offered to 30 adult Beagles following the same methodologies used in guava fiber study. In comparison with the CO diet, the diets with sugarcarne and wheat bran fiber increased the nutrient intake (P<0.001) and decreased the coefficients of total tract apparent digestibility (CTTAD) of DM, OM and food ME (P<0.001) without differences for CP, fat, TDF and starch. Dogs fed WB diets, comparing fiber sources, had greater lactic acid concentration (P<0.001) and lower fecal pH and fecal ammonia (P<0.001), isovaleric and total branched chain fatty acids concentration (P<0.001). The inclusion of both fibers decreased the gastrointestinal retention time compared to CO diet (P<0.001). Palatability testing results indicated that the CO treatment was preferred over the fiber diets without clear effects about particle size. In general, fibers addition limited the starch cooking, increases processing cost, turn the kibbles hard, and could bring difficulties for coating. The guava fiber not occasioned alterations in the intestinal fermentation and in the GIRT until 12% inclusion. The sugarcane and wheat bran with small particles did not result in improved digestibility of a high fiber diet and GIRT; however, may have interfered on BCFA and palatability. Keywords: fiber, fermentative response, food preference, gastrointestinal retention time, particle size. 1 . CAPÍTULO 1 CONSIDERAÇÕES GERAIS 2 CAPÍTULO 1- Considerações gerais 1. INTRODUÇÃO A população de cães domésticos no Brasil em 2014 foi estimada em 58,4 milhões de animais, presentes em quase 50% dos lares brasileiros. Essa população impulsiona mercado especializado, formado por mais de 140 mil pontos de venda, segundo os dados da Associação Brasileira de Indústria de Produtos para Animais de Estimação (ABINPET, 2014). A estimativa para o ano de 2015 é de crescimento de 7% sobre 2014, o que significa R$17,9 bilhões de faturamento dos segmentos “Pet”, o que representa em torno de 0,31% do PIB nacional (EUROMONITOR, 2013). Deste faturamento, 67,5% abrange as indústrias de “Petfood” (ABINPET, 2014). A indústria de alimentação animal brasileira, em 2014, produziu mais de 67 milhões de toneladas de ração, de acordo com Sindirações, tendo como principais fatores propulsores a ascensão das classes sociais, o aumento da renda dos brasileiros e a ampliação da posse responsável, com maior dedicação e cuidado aos animais. A busca de alimentação saudável para cães e gatos é frequente por parte dos proprietários, visando aumentar a longevidade e prevenir o desenvolvimento de doenças, como a obesidade (BONTEMPO, 2005). Em virtude do aumento desta enfermidade (GERMAN et al., 2012), atualmente, a inclusão de fibras na dieta tornou-se realidade por parte das indústrias de “petfood” (KAWAUCHI et al., 2011). Versões light, diet ou de redução de calorias estão presentes nos produtos de praticamente todas as indústrias. Fibra tem sido empregada para se reduzir o valor energético da dieta e favorecer o ajuste entre ingestão e gasto calórico, favorecendo a manutenção de condição corporal saudável (FISHER et al., 2012). Já é estabelecida a influência da fibra na digestibilidade da energia. Segundo o Nutrient Requirements of Dogs and Cats (NRC, 2006), para cães cada ponto percentual de fibra adicionada ao alimento ocasiona redução de 1,43% da digestibilidade da energia da ração. Adicionalmente, fibra é incluída devido à sua influência na manutenção da saúde do trato gastrintestinal (REINHART; SUNVOLD, 1996), 3 formação e consistência das fezes (BURKHALTER et al., 2001), diluição da energia do alimento (BISSOT et al., 2010), regulação do apetite e saciedade (BOSCH et al., 2009) e devido a alegações de melhorar o metabolismo de carboidratos (CARCIOFI et al., 2005). Os estudos até o momento para cães limitaram-se a avaliar inclusões de fibra, da forma como estas foram obtidas industrialmente e seus efeitos nos animais, incluindo digestibilidade dos nutrientes, formação de fezes, tempo de trânsito intestinal, produtos de fermentação e respostas metabólicas (FAHEY et al., 1990a, 1990b, 1992; SUNVOLD et al., 1995a; SWANSON et al., 2001; SA et al., 2013; MONTI et al., 2015). Com exceção de uma única publicação com celuloses purificadas (WICHERT et al., 2002), estudos sobre o processamento das fontes de fibra em si, especificamente quanto à influência do tamanho geométrico de suas partículas sobre as respostas induzidas nos animais não estão disponíveis para cães. O processamento industrial da fibra, no entanto, muda suas características físico-químicas, podendo aumentar sua fermentabilidade, solubilidade e viscosidade, alterando assim os efeitos que esta ocasiona nos animais (ZHANG et al., 2009; REDGWELL et al., 2011; ROBIN et al., 2012). Outro aspecto relevante da fibra é indução de importantes alterações no processo de extrusão. Por se tratar de material altamente polimerizado e estruturado, esta não é expansível e apresenta variável absorção de água, interferindo na geração de viscosidade, cozimento, fluxo da massa no interior do tubo de extrusão, formação da estrutura celular e taxa de expansão do extrusado (KARKLE, 2011). Verifica-se, assim, que a inclusão de fibra nas formulações, apesar de justificada pelo aspecto nutricional, traz importantes desafios de processamento, com aumento do consumo de energia elétrica e produção de extrusados poucos expandidos, densos e duros (KARKLE, 2011; MORARU et al., 2003; CAMIRE et al., 2007). Estas informações, infelizmente, somente estão disponíveis para extrusão de alimentos para o homem, produzidas em estudos com cereais matinais. Cereais matinais são basicamente formulados com cerais (arroz, milho, trigo, aveia) e açúcares (glicose, sacarose e frutose), com muito pouca proteína e gordura (BRENNAN et al., 2008). Sendo assim, as grandes diferenças de matérias primas e composição química dos cereais matinais em relação às formulações para cães e 4 gatos fazem com que não seja possível se extrapolar diretamente estes dados para os alimentos comerciais para estes animais, apesar dos equipamentos de extrusão serem similares. Tendo em vista o exposto anteriormente, esta Dissertação teve por objetivos gerais avaliar os efeitos de inclusões crescentes de fibra de goiaba, bem como o efeito da fibra de cana e do farelo de trigo moídos em diferentes tamanhos sobre o processo de extrusão, digestibilidade dos nutrientes, fermentação no intestino, tempo de trânsito intestinal e palatabilidade de alimentos extrusados para cães. Pretendeu-se, como objetivos específicos: - Estudar o efeito da adição de teores crescentes de fibra de goiaba, em formulação para cães, sobre o processo de extrusão, formação e macroestrutura dos extrusados, digestibilidade dos nutrientes e da energia, tempo de retenção do alimento no trato digestório, formação de produtos de fermentação e palatabilidade das dietas. - Avaliar o efeito de dois tamanhos geométricos de fibra de cana e do farelo de trigo sobre o processo de extrusão, formação e macroestrutura dos extrusados, digestibilidade dos nutrientes e da energia, tempo de retenção do alimento no trato digestório, formação de produtos de fermentação e palatabilidade de dietas contendo teores moderados de fibra dietética alimentar. 2. Revisão de literatura Fibra para cães A fibra alimentar é composta por partes comestíveis das plantas, carboidratos resistentes à digestão e absorção que podem ser solúveis ou insolúveis em água, fermentados ou não fermentados pela microbiota do intestino grosso (AACC, 2001). Características como solubilidade e fermentabilidade definem a funcionalidade da fibra para cães (NRC, 2006; CARCIOFI, 2008). A fermentabilidade diz respeito à velocidade e extensão de degradação bacteriana da fibra e correspondente produção de ácidos graxos de cadeia curta (AGCC). Estes últimos são benéficos ao trato gastrintestinal, mas em elevadas concentrações promovem aumento do 5 peristaltismo, do teor de água das fezes e diminuição da digestibilidade da proteína e gordura (SUNVOLD et al., 1995; CARCIOFI, 2008). Fibra de baixa fermentação leva à diminuição da digestibilidade da matéria seca, mas ocasiona baixa interferência na digestibilidade dos demais nutrientes, no teor de água e na qualidade das fezes produzidas (CARCIOFI, 2005). Fibra solúvel pode ter elevada capacidade de retenção de água, formando géis que aumentam a viscosidade luminal interferindo na cinética de digestão e absorção. Esta, geralmente é mais rapidamente degradada pela microbiota intestinal, resultando em concentrações significativas de AGCC (SWANSON et al., 2001). No entanto, nem sempre isto ocorre, pois algumas fibras de elevada solubilidade, como goma arábica e de Psillium são pouco fermentáveis pela microbiota de cães (SUNVOLD, 1995b). Em contraste, a fibra insolúvel tem menor capacidade de retenção de água e é, geralmente, pouco fermentável pela microbiota do intestino. Isto, no entanto, também é bastante variável entre as fontes de fibra insolúvel e algumas apresentam considerável capacidade de serem fermentadas, de modo que este aspecto deve sempre ser estudado e considerado nas formulações (BOSH et al., 2008; CALABRO et al., 2013). A forma de processamento, produção, extração e manufatura das fibras muda sua funcionalidade. Neste contexto, característica importante da fonte de fibra é o tamanho geométrico médio de suas partículas. No único estudo localizado para cães, celulose microcristalina e com fibras longas foram comparadas. Foi verificado que a qualidade das fezes melhorou com o aumento do tamanho da fibra, a celulose em menor tamanho piorou a qualidade das fezes (WICHERT et al., 2002). Verifica- se, assim, que mesmo fibra insolúvel não fermentável como a celulose tem sua funcionalidade dependente da forma como foi previamente processada antes de fornecida aos cães. O maior ou menor tamanho de partículas da fonte de fibra pode alterar não somente a formação de fezes, como demonstrado anteriormente, mas também a interação da fibra com a microbiota (levando em conta que fibras com pequeno tamanho geométrico têm maior área exposta para colonização e degradação microbiana), sua fermentabilidade e a digestibilidade dos nutrientes. A digestibilidade dos nutrientes pode ser afetada de modo diferente em função das características físico-químicas e da quantidade de fibra adicionada à 6 dieta (NRC, 2006). SUNVOLD et al., (1995b) verificaram maior digestibilidade da matéria seca de rações para cães suplementadas com fibra solúvel fermentável, mas esta reduziu a digestibilidade da proteína bruta e da gordura em relação à fibra insolúvel não fermentável. Em estudo recente com felinos, FISCHER et al. (2012), verificaram que apesar do farelo de trigo e fibra de cana-de-açúcar serem constituídos por fibras insolúveis, a inclusão de farelo de trigo resultou em maior digestibilidade dos nutrientes, formação de fezes com maior umidade e maior produção de AGCC que a fibra de cana, que por sua vez reduziu a resposta glicêmica pós-prandial dos animais. Dependendo de sua quantidade e fonte, a fibra pode acarretar diminuição da palatabilidade dos alimentos (CARCIOFI, 2005). Esta possibilidade vem extrapolada da nutrição humana, pois estudos demonstraram que cereais matinais extrusados enriquecidos com fibra tornaram-se mais laxativos, mais densos, menos crocantes, mais duros e compactos, além de diferenciar-se em textura, aparência e sabor, o que reduziu sua aceitação pelas pessoas (KTENIOUDAKI et al., 2012; ROBIN et al., 2012; FOSCHIA et al., 2013;). Além das características sensoriais odor e palatabilidade, alteradas pela inclusão da fonte de fibra, a profunda alteração por ela induzida na extrusão muda a macroestrutura dos kibbles, característica também relacionada à mastigação, palatabilidade e consumo do alimento (CHALLACOMBE, 2011). O aumento da densidade específica e redução da expansão dos kibbles induzidas pela fibra os torna mais duros, tornando necessário empregar-se mais força durante sua mastigação, aumentando o tempo de ingestão dos alimentos. Estas características, no entanto, necessitam ser convenientemente caracterizadas para cães pois não foram estudadas para estes animais. O tempo de retenção da digesta no trato gastrintestinal é outro parâmetro influenciado pela ingestão de fibra. O tempo de retenção se refere ao período compreendido entre a ingestão alimentar e a eliminação das fezes (BURROW et al., 1982). Este não retrata o tempo de permanência da ingesta nos diferentes compartimentos do trato digestório, como estômago, intestino delgado e intestino grosso. Estudos com cães avaliaram o efeito da fibra no tempo de retenção (FAHEY Jr et al., 1990a; 1992; HERNOT et al., 2005). Dissertação de mestrado de nosso grupo de pesquisa utilizou o método de marcadores radiopacos, com sucesso, 7 demonstrando redução do tempo de retenção mediante inclusão de fibra de cana na dietas de cães (SILVA, 2013). Fibra de cana-de-açúcar Na safra de 2013/2014, o Brasil realizou a moagem de 653 mil toneladas de cana-de-açúcar, o que originou 37.713 mil toneladas de açúcar e 27.543 mil m3 de Etanol. Isto posiciona o bagaço de cana como coproduto agrícola de maior volume nacional, segundo a União da Indústria de Cana-de-Açúcar (ÚNICA). A cana-de-açúcar (Saccharum officinarum L.) é uma das espécies mais cultivadas no Brasil. Do bagaço da cana, co-produto da indústria sucroalcooleira, por meio de processos de lavagem, purificação e micromoagem extrai-se a fibra de cana. Esta apresenta 53,5% de celulose, 31,3% de hemicelulose, 6,4% de lignina, 2,6% de proteína bruta, 2,6% de matéria mineral e conteúdo inexpressivo de gordura (VELOSO, 2011). Com cerca de 90% de fibra alimentar, a fibra de cana possui praticamente 100% de fibra insolúvel (PINTO, 2007). O ingrediente praticamente não é fermentado pela microbiota de cães (CALABRO et al., 2013) e quando adicionado em rações extrusadas reduz a digestibilidade da energia e promove formação de fezes adequadas (SILVA, 2013). Este é ingrediente nacional acessível, muito mais barato que a celulose de madeira purificada, tradicionalmente incluída em formulações especiais para cães. Farelo de trigo O farelo de trigo é um dos principais co-produtos da alimentação humana, obtido da moagem do trigo para produção da farinha de trigo, com largo emprego na alimentação animal. Este é constituído basicamente pelo pericarpo da semente de trigo, onde se concentram a maior parte de sua fibra e minerais (BUTOLO, 2010). Apresenta entre 16% e 19% de proteína e 32% e 40% de fibra alimentar, sendo esta constituída por 97% de fibra insolúvel. Para cães, sua inclusão no alimento induz redução da digestibilidade dos nutrientes e da energia, aumento da produção de 8 AGCC e da umidade das fezes, embora as fezes produzidas sejam adequadas (SÁ et al., 2013). Devido a seu baixo custo e teor relativo de proteínas, o ingrediente é largamente empregado em alimentos econômicos. É possível que o farelo de trigo, à semelhança do verificado para a fibra de cana, possa ser reprocessado industrialmente e tornar-se fonte de fibra mais interessante, a fibra de trigo, mediante conveniente redução do tamanho geométrico de suas partículas. A inclusão de farelo de trigo está associada à redução do cozimento na extrusora, aumento do consumo de energia elétrica no processamento e formação de extrusados rugosos e de aspecto grosseiro (CASE et al., 1992). Estas características indesejáveis poderiam ser revertidas, ou minimizadas, mediante adequado preparo e processamento desta fonte de fibra. Fibra de Goiaba O Brasil se destaca no cenário mundial como terceiro maior produtor de frutas, comercializadas tanto in natura como processadas na forma de sucos, concentrados, polpa e doces (MAPA, 2012). Apesar disso, nenhum dos potenciais co-produtos desta importante indústria é atualmente empregado na indústria de alimentos para animais de estimação. A goiabeira (Psidium guajava) é espécie pertencente à família Myrtaceae e originária das regiões tropicais americanas. O Brasil ocupa o lugar de segundo maior produtor mundial de goiabas, com cerca de 275.000 toneladas produzidas por ano. São Paulo é o estado que se destaca no cenário brasileiro, produzindo 65% desse total (POMMER; MURAKAMI, 2006). Durante seu processamento industrial, até 47% da goiaba pode ser descartada (DURIGAN et al., 2002). Considerando-se as estimativas de processamento anual de 202 mil toneladas de goiaba no Brasil, tem-se dimensão do potencial e disponibilidade deste co-produto (SILVA et al., 2009). Esses resíduos de frutas, ao saírem da indústria, apresentam alto teor de umidade. A polpa de goiaba pode chegar a apresentar mais de 55% de água (SILVA et al., 2006), o que tem limitado seu emprego na alimentação animal (PRASSAD; AZEEMODDIN, 1994). Resíduos de indústrias alimentícias apresentam normalmente baixo custo, estão 9 disponíveis e são caracterizados por apresentar elevada fibra alimentar (SERENA; BACH-KNUDSEN, 2007). O uso destes co-produtos é tendência crescente, agregando valor em descartes de processos e reduzindo a geração de resíduos pelas indústrias (YAGCI; GOGUS, 2010). De modo geral, a fibra das frutas apresenta elevada concentração de pectina e hemicelulose em relação ao teor de celulose presente, com baixa concentração de gordura e proteína (FISHER, 2009). A composição química e as características fermentativas de várias polpas de frutas foram estudadas para cães utilizando um modelo in vitro (SWANSON et al., 2001). De acordo com os autores, as fibras de frutas podem fornecer balanço na proporção solúvel e insolúvel da fibra, colaborando para a saúde do trato gastrointestinal. Fibra e o processo de extrusão A extrusão é tecnologia amplamente utilizada para produção de diversos tipos de alimentos para o homem e animais. Umidade, temperatura, pressão e cisalhamento são combinados proporcionando mistura, cozimento, plasticização, texturização e formatação dos alimentos em curso espaço de tempo (RIAZ, 2007). O uso generalizado da extrusão termoplástica na indústria pet food deve-se ao fato dela promover mudanças físicas e químicas nos ingredientes, alterando sua qualidade e propriedades físicas, aumentando seu valor nutricional com eficiência e baixo custo relativo (TRAN, 2008). Dentre os carboidratos, o amido é o principal substrato para que a extrusão ocorra de forma apropriada (CRANE et al., 2000). Durante a extrusão, grânulos de amido são umedecidos e recebem calor, atrito mecânico, corte e pressão, sofrendo o fenômeno de gelatinização: incham, derretem e perdem sua estrutura cristalina (RATNAYAKE; JACKSON, 2009). A fibra, por outro lado, não apresenta funcionalidade na extrusão e sua inclusão resulta em aumento do custo e piora relativa do processamento (GIBSON; ALAVI, 2013) Não foram encontradas publicações sobre os efeitos da fibra na extrusão de alimentos para cães. Para o homem, adição de frutas e legumes em salgadinhos extrusados tem sido estudada desde 1980 (MAGA; KIM, 1989). Desde então, 10 estudos com poupa, pó, bagaço, cascas e sementes foram publicados para produtos extrusados para o homem (KARKLE, 2011; ALTAN et al., 2008; UPADHYAY et al., 2010). De modo geral, a adição de fibra parece induzir aumento da resistência ao fluxo da massa no tubo extrusor, o que resulta em aumento do consumo de energia elétrica e da aplicação de energia mecânica específica. Este aumento da aplicação de energia não resulta em ganho de cozimento, ao contrário a fibra pode limitar o cozimento do amido e a qualidade de processamento, com possível influência no aproveitamento e digestibilidade da dieta. Em relação à macroestrutura do extrusado, a fibra atrapalha a formação de adequada estrutura celular, pois conduz o vapor de água sem que ocorra formação de células. Isto gera aumento da expansão longitudinal e redução da expansão radial, aumento da densidade específica e aparente e formação de extrusados mais duros (ROBIN et al, 2012). Todos estes aspectos podem ter reflexo importante, tanto no custo de processamento como na digestibilidade e palatabilidade de alimentos para cães suplementados com fibra. Moagem A eficiência do processo de produção dos alimentos está diretamente relacionada à eficiência na moagem dos ingredientes. Estudos realizados em codornas, peixes, frangos, suínos e seres humanos (OWSLEY et al., 1981; HEALY et al., 1994; DAHLKE et al., 2001; LEANDRO et al., 2001; SOARES et al., 2003;) demonstraram a relação entre a moagem dos ingredientes e o aproveitamento dos nutrientes. Entretanto, não foram encontrados, na literatura científica, trabalhos que estudaram a influência de diferentes moagens de fontes de fibras para cães. Em relação à moagem de cereais, Bazolli et al. (2015), utilizando-se de formulações para cães contendo milho, arroz ou sorgo, moídos em 3 diferentes tamanhos de partícula (aproximadamente 300, 450 e 600 μm) reportaram redução na digestibilidade dos nutrientes para as dietas contendo milho e sorgo com maiores tamanhos de partícula. Owsley et al. (1981) estudaram o efeito do tamanho das partículas de sorgo na digestibilidade dos nutrientes para suínos. Os resultados indicaram relação inversa entre o tamanho da partícula e a digestibilidade da matéria 11 seca, amido, energia bruta, proteína e aminoácidos. Healy et al. (1994) comparam a moagem de milho e sorgo em diferentes tamanhos de partícula, em ração peletizada para suínos. Dois dos parâmetros checados foram o custo de processamento e a digestibilidade dos nutrientes. Os autores verificaram aumento na digestibilidade dos nutrientes com a redução da granulometria do alimento para valores abaixo de 500 µm. Houve aumento no custo de produção da ração, mas este foi compensado pela melhora no ganho de peso dos animais. O método utilizado nessa dissertação para a redução do tamanho das partículas dos ingredientes fibrosos foi a moagem com moinho de rolos. Esse moinho é mais utilizado na moagem em pequenas quantidades e fornece produto de textura mais uniforme. Dois ou mais cilindros pesados giram em direções contrárias, a velocidades iguais ou diferentes. A alimentação de material é feita na parte superior do moinho. As partículas são comprimidas entre os rolos, que têm ranhuras longitudinais que promovem o corte do material, submetido a forças de compressão. A distância entre os rolos é regulável, determinado o tamanho final das partículas produzidas (BELLAVER; NONES, 2000). Força de Corte (avaliação da textura) O Teste de Textura ou Força de Cisalhamento, segundo Silva (2013), reproduz condições similares às de mastigação pelos animais. Ainda não se tornou prática comum na avaliação de alimentos para animais de companhia, mas é muito utilizado para avaliação de carnes e alimentos para o homem. A análise de textura tem importância na indústria de alimentos, favorecendo tanto a padronização e controle do processos de fabricação, como o desenvolvimento de novos produtos. O teste torna possível se explorar a influência dos ingredientes e do processamento na força de mastigação e aceitação do produto final. A produção de alimentos com diferentes forças de cisalhamento implica em rações com diferentes resistências à mastigação (ISHIZAKI et al., 2006), o que abre oportunidades para se investigar seu efeito na cinética de consumo, tempo de mastigação, palatabilidade e até mesmo na saúde oral dos animais, pelo efeito mecânico em limpeza dos dentes. 12 3. Hipóteses estabelecidas na presente Dissertação Considerando as informações disponíveis na literatura científica, levou-se em conta para esse estudo que a fibra de goiaba, proveniente da indústria de sucos para alimentação humana, convenientemente desidratada, moída e preparada, poderia ter valor como fonte de fibra para cães. As características organolépticas do ingrediente deveriam resultar em bom odor e aceitação do alimento. Considerando- se a presença de importante quantidade de hemicelulose, deveria apresentar fermentação moderada, agregando funcionalidade na geração de ácidos graxos voláteis, relacionados à saúde do intestino. Foi considerado, também, que a moagem fina da fibra de cana de açúcar e do farelo de trigo seria capaz de alterar as propriedades funcionais destas matérias primas. A utilização de fibras de menor tamanho geométrico geraria menor interferência no processo de extrusão e iria favorecer o cozimento adequado, menor consumo de energia elétrica durante o processamento e a formação de extrusados com melhor macroestrutura. A melhor macroestrutura dos extrusados poderia, por sua vez, favorecer a aceitação e palatabilidade do alimento. Esta redução do tamanho de partículas poderia alterar a degradação microbiana dos produtos, aumentando sua fermentabilidade e alterando sua influência na digestibilidade, tempo de retenção gastrointestinal, formação e qualidade das fezes. Diante destas hipóteses, a presente Dissertação de mestrado foi dividida em 4 capítulos: O primeiro contém as considerações gerais, aqui apresentadas. Os capítulos de 2 a 4 foram desenvolvidos em formato de artigo, em língua inglesa, seguindo as normas dos periódicos escolhidos para publicação. O capítulo 2 intitulado “Fibra, macroestrutura e processamento de rações extrusadas para cães”, foi escrito de acordo com as normas da revista “Animal Feed Science and Technology” e avaliou o efeito das fibras nos parâmetros do processo de extrusão; o capítulo 3 intitulado “Fibra de goiaba como um novo ingrediente alimentar: caracterização e efeitos na digestibilidade, fermentação, tempo de retenção e palatabilidade em cães” seguiu as normas da revista “Journal of Animal Physiology and Animal Nutrition” e pretendeu-se avaliar e caracterizar a fibra de goiaba; o capítulo 4 intitulado “Efeito da fibra de cana e do farelo de trigo em diferentes tamanhos de partícula na digestibilidade, produtos da fermentação, preferência alimentar e tempo de retenção gastrointestinal de cães Beagles adultos” foi escrito de acordo com as normas da revista “Journal of Animal Science” e estudou-se o efeito de diferentes tamanho de partículas da fibra de cana e farelo de trigo para cães. 13 1 Elaborada de acordo com as normas da NBR – 6023/2002 (ABNT) 4. REFERÊNCIAS1 AACC. AMERICAN ASSOCIATION OF CEREAL CHEMISTS. Approved methods of the American Association of Cereal Chemists. ed. Saint Paul, v.9, 1990. AACC. AMERICAN ASSOCIATION OF CEREAL CHEMISTS. The definition of dietary fibre. Cereal foods World,v. 46, p.112-126, 2001. ABINPET (Associação Brasileira de Indústria de Produtos para Animais de Estimação). Setor de Pet Food (alimentos para animais de companhia). Disponível em: http://abinpet.org.br/informe-abinpet/. Acesso em: 19 jun. 2014. ALTAN, A.; MCCARTHY, K.L.; MASKAN, M. Twin-screw extrusion of barley-grape pomace blends: Extrudate characteristics and determination of optimum processing conditions. Journal of Food Engineering, v.89, n.1, p.24-32, 2008. doi:10.1016/j.jfoodeng.2008.03.025 BELLAVER, C.; NONES, K. A importância da granulometria, da mistura e da peletização da ração avícola. In: IV Simpósio Goiano de Avicultura, 2000. Goiânia Embrapa, Documentos. Disponível em: < http://www.cnpsa.embrapa.br/sgc/sgc_publicacoes/publicacao_s3f21x6f.pdf> Acessado em: 05 de Dezembro de 2015. BISSOT, T.; SERVET, E.; VIDAL, S.; SERGHERAERT, R.; EGRON, G.; HUGONNARD, M.; HEATH, SE.; BIOURGE, V.; GERMAN, AJ. Novel dietary strategies can improve the outcome of weight loss programmes in obese client- owned cats. Journal of Feline Medicine and Surgery, v.12, p.104-112, 2010. doi: 10.1016/j.jfms.2009.07.003 BONTEMPO, V. Nutrition and health of dogs and cats: evolution of petfood. J. Vet. Res. v.29, p.45-50, 2005. BOSCH G.; VERBRUGGHE A.; HESTA M.; HOLST, J.J.; van der POEL, A.F.; JANSSENS, G.P.; HENDRIKS, W.H. The effects of dietary fibre type on satiety- related hormones and voluntary food intake in dogs. British Journal of Nutrition, v.102, p.318-325, 2009. doi: 10.1017/S0007114508149194 BRENNAN, M.A.; MERTS, I.; MONRO, J.; WOOLNOUGH, J.; BRENNAN, C.S. Impact of guar gum and wheat bran on the physical and nutritional quality of extruded breakfast cereals. Starch, v. 60, p. 248–256, 2008. doi: 10.1002/star.200700698 BURKHALTER, T.M.; MERCHEN, N.R.; BAUER, L.L.; MURRAY, S.M.; PATIL, A.R.; BRENT, J.L.Jr.; FAHEY, G.C.Jr. The ratio of insoluble to soluble fiber components in soybean hulls affects ileal and total tract nutrient digestibilities and fecal characteristics of dogs. The Journal of Nutrition. 131: 1978–1985, 2001. BURROWS, C.F.; KRONFELD, D.S.; BANTA, C.A.; MERRITT, A.M. Effects of fiber on digestibility and transit time in dogs. Journal of Nutrition, v.112, p.1726-1732, 1982. http://abinpet.org.br/informe-abinpet/ http://www.cnpsa.embrapa.br/sgc/sgc_publicacoes/publicacao_s3f21x6f.pdf 14 BUTOLO, J.E. Qualidade de ingredientes na alimentação animal. In: Simpósio Sobre Animais de Estimação. Anais. Campinas:CBNA, 430p. 2010. CALABRO, S.; CUTRIGNELLI, M.I.; BOVERA, F.; CARCIOFI, A.C.; TUDISCO, R.; GUGLIELMELLI, A.; PICCOLO, G. In vitro evaluation of different fiber sources and potential prebiotics for dogs. In: Congress Of The European Society Of Veterinay And Compararative Nutrition, Vienna, Austria. Anais: University of Veterinary Medicine Vienna. v.1, p.63. 2008. CAMIRE, M.E.; DOUGHERTY, M.P.; BRIGGS, J. L. Functionality of fruit powders in extruded corn breakfast cereals. Food Chemistry, v.101, n.2, p.765-770, 2007. doi:10.1016/j.foodchem.2006.02.031 CARCIOFI, A.C. Emprego de fibras em alimentos para cães e gatos. In: Simpósio Sobre Nutrição de Animais de Estimação, 2005. Anais. Campinas: CBNA, 2005, p.95-108. CARCIOFI, A.C.; TAKAKURA, F. S.; OLIVEIRA, L.D.; TESHIMA, E.; JEREMIAS, J.T.; BRUNETTO, M.A.; PRADA, F. Effects of six carbohydrate sources on dog diet digestibility and postprandial glucose and insulin response. Journal of Animal Physiology and Animal Nutrition. v.92, p. 326-336, 2008. doi: 10.1111/j.1439- 0396.2007.00794.x. CASE, S.E.; HAMANN, D.D.; SCHWARTZ, S.J.. Effect of starch gelatinization on physical properties of extruded wheat – and corn based products. Cereal Chemistry v.69, n.4, p.401-404, 1992. CHALLACOMBE, C.A.; SEETHARAMAN, K.; DUIZER, L.M. Sensory characteristics and consumer acceptance of bread and cracker products made from red or white wheat. Journal of Food Science. v.76, n.5, p.337-46, 2011. doi: 10.1111/j.1750- 3841.2011.02200.x. CRANE, S.W.; GRIFFIN, R.W.; MESSENT, P.R. Introduction to commercial pet foods. In: Hand, Thatcher, Remillard, Roudebush. 2000. Small Animal clinical nutrition, 4ed. Mark Morris Institute, P.O. Box 2097, Topeka, Kansas 66601-2097. DAHLKE, F.; RIBEIRO, A.M.L; KESSLER, A.M.; LIMA A.R. Tamanho da partícula domilho e forma física da ração e seus efeitos sobre o desempenho e rendimento de carcaça de frangos de corte. Revista Brasileira de Ciência Avícola, v. 3, p. 211- 217, 2001. DURIGAN, J. F.; SARZI, B.; MATTIUZ, B. PINTO, S.A.A.; DURIGAN, M.F.B. Tecnologia de processamento mínimo de abacaxi, goiaba e melancia. Embrapa, Documentos. Disponível em: Acessado em: 30 de Julho de 2014. Euromonitor® International. The Global Dog and Cat Food Market: Opportunities to Maximise Performance. Disponível em . http://www.cnph.embrapa.br/novidade/eventos/semipos/texto14.pdf 15 GERMAN, A. J.; HOLDEN, S. L.; WISEMAN-ORR, M.L.; REIDB, J.; NOLAN, A.M.; BIOURGEC, V.; MORRIS, P.J.; M, SCOTT, E.M. Quality of life is reduced in obese dogs but improves after successful weight loss. Veterinary Journal, v.192, p.428- 434, 2012. doi:10.1016/j.tvjl.2011.09.015 GIBSON, M.; ALAVI, S. Pet Food Processing- Understanding Transformations in Starch during Extrusion and Baking. Cereal Foods World, v.58, n.5, p.232-236, 2013. doi: 10.1094/CFW-58-5-0232 FAHEY, G.C.JR.; MERCHEN, N.R.; CORBIN, J.E.; HAMILTON, A.K.; SERBE, K.A.; LEWIS, S.M.; HIRAKAWA, D.A. Dietary fiber for dogs: I. Effects of graded levels of dietary beet pulp on nutrient intake, digestibility, metabolizable energy and digesta mean retention time. Journal of Animal Science, v.68, p. 4221-4228, 1990a. FAHEY, G.C.JR.; MERCHEN, N.R.; CORBIN, J.E.; HAMILTON, A.K.; SERBE, K.A.; HIRAKAWA, D.A. Dietary fiber for dogs: II. Iso-total dietary fiber (TDF) additions of divergent fiber sources to dog diets and their effects on nutrient intake, digestibility, metabolizable energy and digesta mean retention time. Journal of Animal Science, v.68, n.12, p.4229-4235, 1990b. FAHEY, G.C.JR.; MERCHEN, N.R.; CORBIN, J.E.; HAMILTON, A.K; BAUER, L.L.; TITGEMEYER, E.C.; HIRAKAWA, D.A. Dietary fiber for dogs: III. Effects of beet pulp and oat fiber additions to dog diets on nutrient intake, digestibility, metabolizable energy, and digesta mean retention time. Journal of Animal Science, v. 70, p. 1169-1174, 1992. FISHER, J. Fruit Fibers. In: Fiber Ingredients: Food Applications and Health Benefits; Eds. CRC Press, Taylor & Francis Group: Boca Raton, FL, USA, p. 427– 438, 2009. FISCHER, M.M.; KESSLER, A.M.; SÁ, L.R.M.; VASCONCELLOS, R.S.; ROBERTI FILHO, F.O.; NOGUEIRA, S.P.; OLIVEIRA, M.C.C.; CARCIOFI, A.C. Fiber fermentability effects on energy and macronutrient digestibility, fecal parameters, postprandial metabolite responses, and colon histology of overweight cats. Journal of Animal Science, v.90, n.7, p. 2233-45, 2012. doi: 10.2527/jas.2011-4334 FOSCHIA, A,M.; PERESSINI, D.; SENSIDONI, A.A.; BRENNAN, C.S. The effects of dietary fibre addition on the quality of common cereal products. Journal of Cereal Science,v.58, p.216-227, 2013. doi:10.1016/j.jcs.2013.05.010 HEALY, B.J.; HANCOCK J. D.; KENNEDY G.A.; BRAMELCOX, P. J.; BEHNKE, K. C.; HINES R.H. Optimum particle size of corn and hard and soft sorghum for nursery pigs. Journal of Animal Science, v.72, n.9, p. 2227-2236, 1994. HERNOT, D.C.; DUMON, H.J.; BIOURGE, V.C.; MARTIN, L.J.; NGUYEN, P.G. Evaluation of association between body size and large intestinal transit time in healthy dogs. American Journal of Veterinary Research, v.67, p.342-347, 2006. ISHIZAKI, M.H.; VISCONTE, L.L.Y.; FURTADO, C.R.G.; LEITE, M.C.A.M.; LEBLANC, J.L. Caracterização Mecânica e Morfológica de Compósitos de 16 Polipropileno e Fibras de Coco Verde: Influência do Teor de Fibra e das Condições de Mistura. Polímeros: Ciência e Tecnologia, v.16, n. 3, p.182-186, 2006. KARKLE, E.N.L. Carbohydrate components of pomace in corn-based extrudates: interactions, expansion dynamics, and structure-texture relationships. Tese de Doutorado (Philosophy Department of Grain Science and Industry). College of Agriculture Kansas State University, Manhattan, Kansas, 2011. KAWAUCHI, I. M.; SAKOMURA, N.K.; VASCONCELLOS, R. S.; de-OLIVEIRA L.D.; GOMES, M.O.S.; LOUREIRO, B.A.; CARCIOFI, A.C. Digestibility and metabolizable energy of maize gluten feed for dogs as measured by two different techniques. Animal Feed Science and Technology, v. 169, p. 96-103, 2011. doi:10.1016/j.anifeedsci.2011.05.005 KTENIOUDAKI, A. and GALLAGHER, E. Recent advances in the development of high-fibre baked products. Trends in Food Science & Technology, v. 28, p. 4-14, 2012. doi:10.1016/j.tifs.2012.06.004 LEANDRO, N.S.M.; STRINGUINI, J.H.; CAFÉ, M.B.; ORSINE, G.F.; ROCHA, A.C. Efeito da granulometria do milho e do farelo de soja sobre o desempenho de codornas japonesas. Revista Brasileira de Zootecnia, Viçosa, v. 30, n. 4, p. 1266- 1271, 2001. MAGA, J.A.; KIM, C.H. Co-Extrusion of rice flour with dried fruits and fruit juice concentrates. Lebensmittel-Wissenschaft & Technologie, v.22, n.4, p.182-187, 1989. MAPA. Fruticultura - Análise da Conjuntura Agropecuária. Acessado em: 30 de Julho de 2014. Disponível em: http://www.agricultura.pr.gov.br/arquivos/File/deral/Prognosticos/fruticultura_2012_13 .pdf MORARU, C.I. & KOKINI, J.L. Nucleation and expansion during extrusion and microwave heating of cereal foods. Comprehensive Reviews in Food Science and Food Safety, v.2,n.4, p.120-138, 2003. doi: 10.1111/j.1541-4337.2003.tb00020.x NATIONAL RESEARCH COUNCIL - NRC. Nutrient requirements of dogs and cats. Washington, D.C: National Academy Press, 2006. OWSLEY, W.F.; KNABE, D.A.; TANKSLEY, T.D.JR. Effect of sorghum particle size on digestibility of nutrients at the terminal ileum and over the total digestive tract of growing-finishing pigs. Journal of Animal Science. v, 52, n.3, p.557-66, 1981. PRASSAD N.B.L.; AZEEMODDIN, G. Characteristics and composition of guava (Psidium guajava) seed and oil. Journal of the American Oil Chemists’ Society, v.71, n.4, p.457-458, 1994. PINTO, M.V.P. Utilização digestiva de dietas com diferentes fontes de fibrosas e determinação de curvas glicêmicas em cães adultos. Dissertação (Mestrado em Zootecnia) Escola de Veterinária da Universidade Federal de Minas Gerais, Belo Horizonte, 2007. http://www.ncbi.nlm.nih.gov/pubmed/?term=Owsley%20WF%5BAuthor%5D&cauthor=true&cauthor_uid=7263529 http://www.ncbi.nlm.nih.gov/pubmed/?term=Knabe%20DA%5BAuthor%5D&cauthor=true&cauthor_uid=7263529 http://www.ncbi.nlm.nih.gov/pubmed/?term=Tanksley%20TD%20Jr%5BAuthor%5D&cauthor=true&cauthor_uid=7263529 17 POMMER, C. V.; MURAKAMI, K. R. N. A goiaba no mundo. O agrônomico, Campinas, p. 22-26, 2006. RATNAYAKE, W.S.; JACKSON. D.S. Starch: sources and processing. In: Encyclopedia of Food Science. Food Technology and Nutrition. 2nd ed. Rev. New York: John Wiley & Sons, p.5567-5572, 2003. REDGWELL, R.J.; CURTI, D.; ROBIN, F.; DONATO, L.; PINEAU, N. Extrusion- Induced Changes to the Chemical Profile and Viscosity Generating Properties of Citrus Fiber. Journal of Agricultural and Food Chemistry. v.59, p.8272–8279, 2011. doi: 10.1021/jf201845b. REINHART, G.D.; SUNVOLD, G.D. In vitro fermentation as a predictor of fiber utilization. In: Recent advances in canine and feline nutritional research; Iams International Nutrition Symposium, 1996, Ohio. Proceedings. Wilmington, Ohio; Orange Frazer, p.15-24, 1996. Riaz, M. N., 2000. Extruders in food applications, In: Riaz M. N. Introduction to extruders and their principles. CRC Press, pp.1-23. RIAZ, M.N. 2007. Extruders and Expanders in Pet Food, Aquatic and Livestock Feeds. Agrimedia, Clenze, p.400. ROBIN, F.; SCHUCHMANN, H.P.; PALZERC, S. Dietary fiber in extruded cereals: Limitations and Opportunities. Trends in Food Science & Technology ,v. 28, p.23- 32, 2012. doi:10.1016/j.tifs.2012.06.008 SILVA, D.A.T.; RABELLO, C.B.V.; SILVA, E.P. et al. Efeito de dois métodos de pré- secagem na composição bromatológica do resíduo do farelo de goiaba para frango de corte In: Jornada de Ensino, Pesquisa e Extensão da UFRPE – Congresso de Iniciação Científica. Anais do Congresso. Recife: Universidade Federal Rural de Pernambuco, 2006. SILVA, F.L. Emprego de fibra de cana-de-açúcar na alimentação de cães: controle da digestibilidade, tempo de retenção intestinal, efeito de saciedade e interferência na saciedade e respostas glicêmicas, insulínicas, colesterol e triglicerídeos pós-prandiais. 2013. Tese de Mestrado. Medicina Veterinária. Departamento de Clínica e Cirurgia Veterinária, FCAV, UNESP, Jaboticabal, São Paulo. SÁ, F.C.; VASCONCELLOS, R.S. BRUNETTO, M.A.; FILHO, F.O.R.; GOMES, M.O.S.; CARCIOFI, A.C. Enzyme use in kibble diets formulated with wheat bran for dogs: effects on processing and digestibility Journal of Animal Physiology and Animal Nutrition, v.97, p.51–59, 2013 25. doi: 10.1111/jpn.12047. SILVA, E.P.; SILVA, D.A.T.; RABELLO, C.B.V.; LIMA, R.B., LIMA, M.B.; LUDKE, J.V. Composição físico-química e valores energéticos dos resíduos de goiaba e tomate 60 para frangos de corte de crescimento lento. Revista Brasileira de Zootecnia, v.38, n.6, p.1051-1058. 2009. 18 SOARES, C.M.; HAYASHI, C.; BOSCOLO, W.R. et al. Diferentes graus de moagem dos ingredientes em dietas peletizadas para a tilápia-do-nilo (Oreochromis niloticus L.) em fase de crescimento. Desempenho e digestibilidade aparente. Zootecnia Tropical, v.21, p.275-287, 2003. SUNVOLD, G.D.; FAHEY Jr, G.C.; MERCHEN, N.R.; REINHART, G.A. In vitro fermentation of selected fibrous substrates by dog and cat fecal inoculum: Influence of diet composition on substrate organic matter disappearance and short-chain fatty acid production. Journal of Animal Science, v.73, p.1110-1122, 1995a SUNVOLD, G.D.; FAHEY JR, G.C.; MERCHEN, N. R. et al. Dietary fiber for dogs: IV. In vitro fermentation of selected fiber sources by dog fecal inoculum and in vivo digestion and metabolism of fiber-supplemented diets. Journal of Animal Science, v. 73, p. 1099-1109, 1995b. SWANSON, K.S.; GRIESHOP, C.M.; CLAPPER, G.M.; SHIELDS, R.G.; BELAY, T. Jr., Merchen, N.R.; Fahey, Jr.G.C. Fruit and vegetable fiber fermentation by gut microflora from canines. Journal of Animal Science v. 79, p. 919–926, 2001. TRAN, Q.D. Extrusion processing: effects on dry canine diets. Tese (Doctorate in Feed Tecnology). Wageningen University. The Netherlands. 2008. UPADHYAY, A.; SHARMA, H.K.; SARKAR, B.C. Optimization of carrot pomace powder incorporation on extruded product quality by response surface methodology. Journal of Food Quality, v.33, p.350-369, 2010. doi: 10.1111/j.1745- 4557.2010.00323.x VELOSO JUNIOR, R.R. Nível de fibra e tipo de processamento na digestibilidade, ingestão e parâmetros bioquímicos da arara canindé (Ara ararauna L.). 2011. Tese (Doutorado Zootecnia) Faculdade de Ciências Agrárias e Veterinárias, Universidade Estadual de São Paulo, Jaboticabal, 2011. ZHANG, M.; LIANG, Y.; PEI, Y.; GAO, W.; ZHANG, Z. Effect of Process on Physicochemical Properties of Oat Bran Soluble Dietary Fiber. Journal of Food Science. v.74, n. 8, 2009. doi: 10.1111/j.1750-3841.2009.01324.x. WICHERT B.; SCHUSTER, S.; HOFMANN, M.; DOBENECKER, B.; KIENZLE, E. Influence of Different Cellulose Types on Feces Quality of Dogs. American Society for Nutritional Sciences. Journal of Nutrition. v.132, p.1728–1729, 2002 19 CAPÍTULO 2 FIBRA, MACROESTRUTURA E PROCESSAMENTO DE RAÇÕES EXTRUSADAS PARA CÃES1 1 Escrito de acordo com as normas da revista Animal Feed Science and Technology 20 Running head: Fiber and extrusion of dog foods. Fiber influence on macrostructure and processing traits of extruded diets for dogs M. Monti a , M. Gibson b , B.A. Loureiro a , F. C. Sá a , T. C. Putarov a C. Villaverde c , S. Alavi b , A. C. Carciofi a,* a Departamento de Clínica e Cirurgia Veterinária, Campus de Jaboticabal, UNESP - Univ Estadual Paulista, São Paulo, Brasil. b Department of Grain Science and Industry, Kansas State University, Manhattan, KS, USA. c Departament de Ciència Animal y dels Aliments, Universitat Autònoma de Barcelona, Campus UAB, Bellaterra, Barcelona, Spain. Presented as an oral communication at the 18 th Congress of the European Society of Veterinary and Comparative Nutrition, Utrecht, Netherlands, 11-13 September 2014 Abbreviations: AAFCO, American association of feed control officials; NRC, National research council; SAS, Statistical Analysis Systems; SME, Specific mechanical energy; STE, Specific Thermal Energy; TSE, Total Specific Energy; WBL and SCL, wheat bran and sugarcane fiber large particles; WBS and SCS; wheat bran and sugarcane fiber small particles * Corresponding author: Tel.: +55 16 3209-2631; fax.: +55 16 3203-1226. E-mail address: aulus.carciofi@gmail.com (A.C. Carciofi). http://www.uab.cat/departament/ciencia-animal-aliments/ 21 Abstract Fiber is currently used in dog food formulations due to its nutritional properties. However, few studies have evaluated the influence of fiber on the extrusion traits and kibble formation. The present study evaluated the effect of fiber type and particle size on extrusion processing parameters and kibble macrostructure of dog foods. In experiment 1, guava fiber was added to a control diet (CO) at different inclusion levels: 3% (GF3), 6% (GF6), and 12% (GF12). In experiment 2, two fiber types (sugarcane and wheat bran) and two sizes were compared to a control (CO) diet. Diets were manufactured using a single screw extruder. Each food was processed in two separated days and samples were collected four times per day, for a total of eight replications per diet. The processing conditions were not changed for any treatment. Data were analyzed via analysis of variance, and the α level of significance was set at 0.05. Guava fiber inclusion resulted in a linear increase on temperature, pressure, and specific mechanical energy (SME) implementation (P<0.001) during extrusion, whereas starch cooking and longitudinal expansion decreased linearly (P<0.001). Piece density and kibble cutting force increased linearly (P<0.001) with guava fiber inclusion. On experiment 2, fiber addition also increased SME implementation (P<0.001). Specific length and cutting force was higher (P<0.001), and radial expansion lower (P=0.008) for fiber supplemented diets. Starch gelatinization was lower for the large particle size fiber diets (P<0.05). Sugarcane fiber induced higher longitudinal expansion, with less dense and harder kibbles than wheat bran (P<0.001). The finely grind fibers lead to formation of kibbles with lower piece density (P=0.018). Fiber source addition increased resistance of the dough to flow inside the extruder, explaining the increased SME. Kibble macrostructure was less acceptable on fiber-supplemented diets, showing negative effects of fiber on kibble formation and macrostructure. In conclusion, fiber inclusion in dog diets increased the electric energy required to extrude and may reduce starch cooking and kibble expansion, leading to the production of denser and harder kibbles. Keywords: extrusion, fiber sources, guava, particle size, processing, sugarcane 22 1. Introduction During the extrusion process to make dog food a combination of moisture, shear, temperature, and pressure are applied to the mix of ingredients, creating a continuous and short process forcing the material through a specifically designed opening (Riaz, 2000). The process induce changes on food ingredients, resulting in extensive cooking and in a plasticized food dough (Altan et al., 2008; Yağci and Göğüş, 2008). These modifications are directly linked and depend of on the total energy transferred to the dough, composed by the implemented specific mechanical energy (SME) and the specific thermal energy (STE). The combination of these two types of energy promotes starch gelatinization, protein denaturation, lipid modification, inactivation of enzymes, and microbe viability reduction. At the end of the extruder barrel, the plasticized dough expands in contact with atmosphere, creating a particular kibble macrostructure that affects shape and texture (Griffin, 2003; Challacombe et al., 2011). Fiber supplemented extruded diets are produced nowadays by most pet food companies. Fiber is used to dilute energy density and promote specific benefits on gut and general health (Kawauchi, et al., 2011; Fischer et al., 2012). The inclusion of fiber sources, however, influences the processing parameters and SME implementation, potentially altering the final product characteristics (Mendonça et al., 2000). Fiber is a highly structured not expandable material with variable water absorption capacity; it can affect viscosity, mass flow inside the barrel, and the cellular arrangement formation of the extrudates (Karkle et al., 2012). Due to this, kibble expansion ratio and important texture characteristics as hardness, and crispness may be altered by fiber (Karkle et al., 2012), thus changing food sensory attributes (Koppel et al., 2015). 23 Unfortunately, detailed information about the impact of fiber on extrusion traits and kibble macrostructure is only available for human foods (Brennan et al., 2008; Baik et al., 2004; Karkle, et al., 2012b). The characteristics of the specific fibrous material included in the recipe have also an impact on extrusion traits, and thus fiber effects cannot be generalized. Fibrous fruit and vegetable processing by-products has been studied in human extruded foods (Upadhyay et al., 2010; Karkle et al., 2012), as a way to add value to these foods and to minimize the environmental impact of these residues (Altan et al., 2008). Besides fiber type, fiber particle size is also important. It is possible to change the material dynamics into the extruder changing the fiber particle length. Understanding these effects could be important to achieve better cooking, reduce extrusion cost, and overcome some negative effects of fiber on extrudate macrostructure. However, there is no data regarding the effect of fiber particle size in the extrusion of canine diets. The present study aimed to evaluate the influence of the addition of increasing amounts of guava fiber (experiment 1) and the addition of two different particle sized sugarcane and wheat bran fiber on the extrusion traits and kibble macrostructure of canine diets. 2. Material and Methods 2.1. Fiber ingredients and diet formulation Two separate experiments were conducted, the first experiment was designed to study the inclusion level of guava fiber (Psidium guajava) and the second one to assess the impact of sugarcane fiber and wheat bran particle sizes. The company Dilumix (Leme, Sao Paulo, 24 Brazil) provided the fiber ingredients: guava fiber (68.9% of insoluble fiber and 1.4% of soluble fiber), sugarcane fiber (87.0% of insoluble fiber and 0.4% of soluble fiber), and wheat bran (37.8% of insoluble fiber and 1.7% of soluble fiber). A control diet (CO) containing maize and poultry by-product meal was formulated for adult dogs, according to the European Pet Food Industry Federation nutritional guidelines (FEDIAF, 2013). The fiber sources were added to the control diet, replacing maize to create four formulations in experiment 1, the control diet plus three addition amounts of guava fiber (GF): 3% (GF3), 6% (GF6), and 12% (GF12). The guava fiber used had a mean particle size of 213 μm. On experiment 2 the same control formulation was used, and sugarcane fiber or wheat bran was added (Table 1), for a total of 5 dietary treatments. The same lot of raw fiber material was ground to obtain two different particle sizes: large sugarcane fiber (SFL), small sugarcane fiber (SFS), large wheat bran (WBL), and small wheat bran (WBS); with a diameter of 395μm, 197μm, 345 μm, and 143μm respectively. The particle sizes were determined using laser diffraction particle size analysis (Boac et al., 2009). The total dietary fiber content of all ingredients was determined before diet manufacture and the diets containing sugarcane fiber and wheat bran were formulated to provide approximately 16% of total dietary fiber. 2.2. Diet preparation The ingredients, with the exception of fiber sources, were weighed, mixed, and ground using a hammer mill fitted with a screen sieve size of 0.8 mm (Sistema Tigre de Mistura e Moagem, Tigre, Sao Paulo, Brazil). The fiber source was then added to the ingredients and mixed again, compounding the final diet. 25 Diets were extruded in a single screw extruder (MEX 250, Manzoni, Campinas, Brazil), with a processing capacity of 250 kg/h, screw diameter of 15.9 mm, L/D ratio of 29.3 and die open area of 15.9 mm 2 . The extruded screw have five sections: initial - single flight and no steam lock; second – single flight and small steam lock; third – double flight uncut and small steam lock; fourth - double flight uncut and medium steam lock; fifth - double flight cut cone. The preconditioner shaft speed was set to obtain an average residence time of 3.5 minutes. Thermal energy was implemented on the dough at preconditioner by direct steam infusion. For all treatments, the extruder screw speed was set to 465 rpm. Each food was processed separately on two different days for replicates. Each day the production was started with a high fiber diet to stabilize the equipment and establish the basal processing conditions. After the stabilization of the equipment (minimum of 45 min), the processing was kept constant and no alterations was done on any software conditions (feed rate, extruded screw speed, water and steam injection, and cutting knife speed). After this moment, production parameters were registered at each 15 minutes, with at least four measurements per diet and day, totaling eight observations per diet. The parameters registered were: preconditioner temperature, engine amperage, dough temperature before die, dough temperature out extruder, dough pressure before die, extruder output mass, water infusion, bulk density after extruder, and bulk density after dryer. Other parameters registered included ambient temperature, working water temperature, mash feed temperature, and steam pressure. At each observation time, samples of food were collected from the preconditioner, the extruder and the dryer and stored at -20ºC for further analysis. After extrusion, the kibbles were dried in a forced air dryer at 105°C for 20 minutes, and coated with fat and liquid palatability agents. 26 Table 1. Ingredients and chemical composition of the experimental diets used on experiment 1 and experiment 2. Item Diets a CO GF3 GF6 GF12 SC WB Ingredients, (g/kg, as-fed basis) Maize 578.2 546.8 516.4 449.4 474.6 304.4 Poultry by product meal 318.0 318.6 318.2 325.2 325.6 261.2 Poultry Fat 64.4 65.2 66.0 66.0 73.2 76.4 Guava Fiber b - 30.0 60.0 120.0 - - Sugarcane Fiber c - - - - 90.0 - Wheat Bran d - - - - - 320.0 Fish oil 1.5 1.5 1.5 1.5 1.5 1.5 Palatant e 20 20 20 20 20 20 NaCl 5 5 5 5 5 5 KCl 5 5 5 5 5 5 Vitamin and Mineral mix f 3 3 3 3 3 3 Choline Chloride 2 2 2 2 2 2 Antioxidant g 0.4 0.4 0.4 0.4 0.4 0.4 Mold inhibitor h 1 1 1 1 0.1 0.1 Analyzed chemical composition of the final product (g/kg, DM-basis) Moisture 59.0 68.0 60.0 73.0 58.0 58.6 Ash 60.0 55.7 66.2 61.0 56.8 61.0 Crude Protein 252.1 257.7 259.5 251.7 254.8 252.1 Crude Fat 153.0 159.0 152.0 146.1 148.7 150.2 Total dietary fiber 107.5 131.3 166.3 183.6 166.1 180.0 Insoluble fiber 107.5 128.5 163.7 176.9 166.1 171.7 Soluble fiber 0.00 2.8 2.7 6.7 0.0 8.4 Starch 412.0 387.0 357.0 357.0 379.1 344.0 a CO-Control diet, without added fiber source, GF3 – addition of 3% guava fiber, GF6 – addition of 6% guava fiber, GF12 – addition of 12% guava fiber, SC – addition of sugarcane fiber, WB – addition of wheat bran. b Dilufiber Guava, Dilumix, Leme, SP, Brazil. c Vit2be Fiber, SPF do Brazil, Descalvado, Brazil d Wheat bran fiber, Dilumix, Leme, SP, Brazil. e Liquid Palatant, SPF do Brazil, Descalvado, Brazil. f Added per kg of diet: iron, 120 mg; copper, 15 mg; magnesium, 75 mg; zinc, 150 mg; iodine, 2 mg; selenium, 0.3 mg; vitamin A, 18,000 IU; vitamin D3, 1,000 IU; vitamin E, 100 IU; vitamin K, 2 mg; biotin, 0.6 mg; thiamine. 20 mg; riboflavin, 10 mg; pantothenic acid, 50 mg; niacin, 75 mg; vitamin B6, 6 mg; folic acid, 4 mg; vitamin B12, 0.1 mg. g Banox: BHA-butylated hydroxyanisole, BHT-butylated hydroxytoluene, propyl gallate and calcium carbonate. Alltech do Brasil Agroindustrial Ltda. h Mould Zap: Ammonium dipropionate, acetic acid, sorbic acid and benzoic acid. Alltech do Brasil Agroindustrial Ltda. 27 2.3. Laboratory Analyses Food samples were grounded in a cutting mill (Mod MA-350, Marconi, Piracicaba, Brazil) fitted with a 1mm screen. The samples were analysed by oven-drying of the sample for dry matter (DM) (method 934.01), by muffle furnace incineration for ash content (method 942.05), by the Kjeldahl method for crude protein (method 954.01), and with a Soxhlet apparatus extraction for acid hydrolysed ether extract (method 954.02), following the methods described by the Association of Official Analytical Chemists (AOAC, 1995). Organic matter (OM) of the samples was calculated as DM minus ash. Dietary fibre (total, soluble, and insoluble) was measured by using a combination of enzymatic and gravimetric procedures (method 991.43, AOAC, 1995), the total amount of starch was measured according to described by Hendrix (1995), and the degree of gelatinization of the starch was determined by the amyloglucosidase method as described by Sá et al (2013). 2.4. Kibble traits and macrostructure The extrudate cutting force was measured in 20 kibbles per diet. Kibbles were first dried to achieve the same moisture in an oven (ETS Modelo 532, Systens Eletro-Tech, Inc., Glenside, PA, EUA) at 55°C during 24 hours. Individual kibbles were weighted and the cutting force determined with a texturometer (Texture Analyser TAX/T2I - Stable Micro Systems Ltda, Godalming, UK) equipped with a load cell of 50kg.f. The samples were compressed with a 10mm penetration distance and speed of 2.0mm/s, using a Warner Bratzler Knife. To ensure that the samples were representative, the diameter of scanned kibble should correspond to the treatment average diameter. 28 Electronical micrographs of the kibbles were obtained in a scanning field emission scanning electron microscope (JEOL, JSM- 7500F; Miaka, Tokyo, Japan), adjusted to 20 kV. Images were evaluated by a trained examiner, blinded to the experimental groups. The analyses were performed on the Chemistry Institute, UNESP, campus of Araraquara. All image analysis was qualitative, not submitted to statistical evaluation. Images of the external and internal surface were conducted. For the internal evaluation, kibbles were cut along the medial direction. Pictures of the internal surface were processed with standardized filter sets using the software Adobe Photoshop CC 2015 (Adobe Systems, California, USA) and the total area of inner cells measured. The results was expressed as the percentage of cell areas in relation to mean kibble diameter area, with the following formula: Percentage of cell areas = Total area of inner cells measured mean kibble diameter area For each treatment, the length (le), diameter (de) and mass (me) of 20 extrudates were measured and used to obtain the radial expansion ratio (RE), specific length (lsp) and piece density (ρ), as described below. RE= de 2 dd 2 lsp= le me (mm/g) ρ= 4 me π× de 2 2 ×le (kg/m 3 ) were: dd= die open area. 29 2.5. Specific mechanical energy and Specific thermal energy calculations The SME was calculated for each treatment in accordance with Riaz (2007), with the following formula: SME= (√3×Voltage×(WA-EA)×(cosFi÷1000))×1000 Production Where; Voltage = (220V) WA = working amperage, the motor amperage during processing EA = empty amperage, the motor no load amperage The STE is the sum of the thermal energy contributions of raw materials, water and process water vapor divided by the feed rate (kg/h). This was calculated after the mass balance according to Riaz (2000): -Preconditioner energy balance: QR + QW + QS = QPc + QSL + QHL + ∑∆h - Extruder energy balance: QP + QW + Qsme + Qbarrel = Qex + QSL + ∑∆h Where; QR = Raw material heat capacity QHL = Preconditioner heat loss by convection QBarrel = Extruder heat loss by convection QW = Water input heat capacity; Qsme = Mechanical energy amount QS = Steam input heat capacity 30 Qte = Thermal energy amount Qp = Preconditioner Product heat capacity Qex = Extruder Product heat capacity QSL = Steam loss heat capacity ∑∆h = Reaction energy The amount of heat (Q) was obtained from the formula: Q = m.c.T Where: m = mass; c = specific heat capacity T = temperature - Preconditioner mass balance equation (preconditioner output): Mraw + Mw + MS = Msl + Mpc - Extruder balance mass equation: Mpc + Mw = Msl + Mf Where: Mraw = raw material mass Mw = water mass; Ms = steam mass; Msl = steam loss mass Mpc = preconditioner mass; Mf = final mass The Total Specific Energy (TSE) was obtained by the sum of the SME and STE. 31 2.6. Statistical analysis Each experiment was analyzed separately. Both studies follow a randomized block design, with two blocks (days of extrusion) and four repetitions per block (each 15 min interval of sampling), totaling eight repetitions per treatment. The experimental unit was considered the time sampling, except for kibble macrostructure and deforming force, when the experimental unit was one kibble (with 20 repetitions per treatment). Data were submitted to analysis of variance, model sums of squares were separated into treatment (diet) and block (days of extrusion) effects. When differences were found on F test, on experiment 1 polynomial contrasts were used to evaluate the inclusion levels of guava fiber, and on experiment 2 orthogonal contrasts were used to detect differences between: control and fiber supplemented foods (CO versus SFL + SFS + WBL + WBS); sugarcane and wheat bran fibers (SFL + SFS versus WBL + WBS); fiber particle sizes (SFL + WBL versus SFS + WBS). The mixed procedure of SAS statistical software (version 9.1; SAS Institute, Cary, NC, USA) was used to perform the analysis. All data were found to comply with ANOVA assumptions. Values of P<0.05 was considered significant. The α level of significance was set at 0.05. 3. Results 3.1. Experiment 1 During extrusion the preconditioner temperature (P<0.001) and output moisture (P=0.005) decreased linearly as guava fiber inclusion increased (Table 2). The mass flow in the extruder fluctuated randomly. Starch gelatinization after preconditioner was similar 32 among diets, with a mean value of approximately 33% (P>0.05). The motor load (engine amperage), and die temperature and pressure increased linearly with guava fiber inclusion (P<0.001). Extruder output mass varied according to the preconditioner mass flow (P=0.003). Regarding energy balance, the SME implementation increased approximately 84% when comparing CO with GF12 (P=0.047). The STE implementation increased linearly (P=0.015), which together resulted in linear increase on TSE (P=0.001). Because the SME presented greater proportional increase than STE, the STE:SME ratio decreased quadractically (P=0.002). The addition of guava fiber reduced linearly the kibble radial expansion rate and starch gelatinization (P<0.001), and increased the piece density (P<0.001). Fiber inclusion resulted in a quadratic reduction on kibble specific length (P<0.001), whereas it resulted in an increased kibble hardness, as verified by the linear increase on cutting force (P<0.005). Figure 1 illustrates internal and external pictures of the kibbles produced on experiment 1. It is possible to note that the cellular structure in the CO diet is composed by larger cells compared to the GF12 diet. The percentage of cell area in relation to total kibble diameter area was 15.3% for CO, and decreased to 13.4% for GF3, 13.0% for GF6, and 12.6% for GF12. It is possible to see bigger pores on the kibble surface of the CO diet compared to GF3 and GF6 diets, whereas the kibbles of the GF12 diet showed an uneven and rough surface. 3.2. Experiment 2 The parameters measured at preconditioner did not change on experiment 2, with the exception of a lower output mass for fiber supplemented diets in comparison with the CO diet (P=0.007; Table 3). The engine amperage was lower for sugarcane fiber than wheat bran diets, and lower for the small particle size than large particle size fibers (P<0.001). Pressure at 33 die was lower for fiber supplemented diets than for CO diet (P=0.046). Extruded output mass varied accordingly to the preconditioner output, with a higher value for the CO diet (P=0.046) compared to fiber diets. The SME implementation was higher for fiber diets compared to CO diet, also higher for wheat bran compared to sugarcane fiber diets and for large particles compared to small particle fibers (P<0.01). Fiber supplemented diets presented higher STE implementation than CO (P<0.001), and so did the sugarcane fiber compared to the wheat bran diets (P=0.013). These variations resulted in higher TSE addition and lower STE:SME ratio for the fiber supplemented diets than for the CO diet (P<0.01), and higher STE:SME ratio to sugarcane than wheat bran supplemented diets (P<0.001). Regarding kibble traits, diets with small particle size fiber had higher starch gelatinization than large particle size fiber diets (P=0.051). Piece density was higher for the wheat bran than for sugarcane fiber diets, and for large fiber particle than for small fiber particle diets (P<0.02). Fiber inclusion reduced the kibble radial expansion compared to the CO diet (P=0.008). Longitudinal expansion (specific length) was increased after fiber addition to diets (P<0.001), and was higher for sugarcane in comparison with wheat bran diets (P<0.001). Fiber inclusion resulted in a significant increase on cutting force (P<0.001) compared to CO diet, and this was higher for the sugarcane than for wheat bran diets (P<0.001). The internal and external pictures of the kibbles produced on experiment 2 are shown in Figure 2. The cellular structure of the CO and the diets produced with small particle fibers (SFS and WBS) are composed by bigger cells compared to large particle fiber diets (SFL and WBL). The percentage of cell area in relation to total kibble diameter area was 15.6% for CO, 13.2% for SFL, 14.6% for SFS, 17.0% for WBL, and 16.1% for WBS. There were no differences on pore formation of the kibble surface among treatments. 34 Table 2. Processing parameters and kibble macrostructure traits of extruded dog diets with different inclusions of guava fiber. Experiment 1. a CO= control diet, without added fiber source, GF3= addition of 3% guava fiber, GF6= addition of 6% guava fiber, GF12= addition of 12% guava fiber. b S.E.M= standard error of the mean (n= 32). c Linear and quadratic effect of guava fiber additions. d SME= specific mechanical energy. e STE= specific thermal energy. f TSE= total specific energy. g n=20 kibbles per diet. Item Diets a S.E.M. b Contrast c CO GF3 GF6 GF12 Linear Quadratic Preconditioner Temperature (°C) 85.6 84.4 85.7 82.2 0.59 <0.001 0.301 Output moisture (%) 27.7 26.1 25.7 25.2 0.01 0.005 0.577 Output mass (as-fed, kg/h) 181.9 170.2 162.8 190.8 2.31 0.004 0.005 Starch gelatinization (%) 35.6 30.3 33.6 33.5 1.28 0.987 0.756 Extruder Engine amperage (A) 42.1 42.80 46.6 49.18 0.70 <0.001 0.160 Die temperature (°C) 125.8 130.3 141.3 134.3 1.46 <0.001 0.137 Die pressure (bar) 61.7 61.7 70.3 70.6 0.88 <0.001 0.909 Output mass (as-fed, kg/h) 170.4 161.2 149.6 153.2 3.32 0.043 0.003 Energy balance (kW-h/ton) SME d 15.3 19.5 25.1 28.1 0.81 <0.001 0.047 STE e 74.7 74.4 84.6 87.2 2.27 0.015 0.682 TSE f 90.0 93.9 109.6 115.3 3.52 <0.001 0.815 STE:SME 4.88 3.81 3.37 3.10 0.21 <0.001 0.002 Kibble traits (After Dryer) Starch gelatinization (%) 92.8 91.1 90.5 88.3 0.93 <0.001 0.823 Piece density (kg/m 3 ) 0.37 0.40 0.42 0.43 0.01 <0.001 0.190 Radial expansion rate 4.1 3.9 3.5 3.2 0.05 <0.001 0.263 Specific length (mm/g) 47.4 40.8 41.6 46.3 0.37 0.097 <0.001 Cutting force (kg.f) g 2.4 2.9 3.1 3.4 0.06 <0.005 0.263 35 Figure 1. Field scanning electron micrograph of diets with guava fiber inclusion. A, B, C and D – Internal area (increased 25x) of CO, GF3, GF6 and GF12 diets, respectively. A1, B1, C1 and D1- External area (increased 50x) of CO, GF3, GF6, and GF12 diets, respectively. CO= control diet, without added fiber source, GF3= addition of 3% guava fiber, GF6= addition of 6% guava fiber, GF12= addition of 12% guava fiber. A A1 B B1 C C1 D D1 36 Table 3. Processing parameters and kibble macrostructure traits of extruded dog diets with different types and size of fiber. Experiment 2. a CO= control diet, without added fiber source, SFL= sugarcane fiber, large particle; SFS= sugarcane fiber, small particles; WBL= wheat bran, large particles; WBS= wheat bran, small particles. b S.E.M.= standard error of the mean (n= 40). c Sugarcane fiber (SFL + SFS) versus wheat bran (WBL+WBS). d Large fiber particles (SFL + WBL) versus small fiber particles (SFS + WBS). e Control (CO) versus fiber supplemented diets (SFL + SFS + WBL + WBS) f SME= specific mechanical energy. g STE= specific thermal energy. h TSE= total specific energy. i n=40 kibbles per diet Item Diets a S.E.M. b Contrast CO SFL SFS WBL WBS SC x WB c Large x Small d CO X Fiber e Preconditioner Temperature, (°C) 85.6 84.1 84.1 82.9 84.3 0.48 0.616 0.522 0.1879 Output moisture, (%) 27.7 26.7 27.9 27.0 28.3 0.004 0.675 0.102 0.8124 Output mass (as-fed, kg/h) 181.9 168.86 154.34 166.09 164.61 3.01 0.465 0.136 0.007 Starch gelatinization, (%) 35.6 30.5 31.6 29.9 33.9 1.07 0.572 0.124 0.431 Extruder Engine amperage (A) 42.1 41.0 41.0 45.9 40.9 0.49 <0.001 <0.001 0.936 Die temperature (°C) 125.8 126.3 118.7 123.9 124.9 1.16 0.443 0.170 0.314 Die pressure (bar) 61.7 53.4 52.5 56.2 60.2 1.57 0.110 0.569 0.047 Output mass (as-fed, kg/h) 170.4 152.0 141.3 150.0 152.0 4.03 0.479 0.237 0.046 Energy balance (kW-h/ton) SME f 15.3 16.7 16.9 24.0 17.07 0.81 0.014 <0.001 <0.001 STE g 74.7 110.3 120.1 108.4 90.6 0.68 0.013 0.209 <0.001 TSE h 90.0 127.0 136.1 132.4 108.3 5.06 0.058 0.071 <0.001 STE:SME 4.88 6.60 7.06 4.51 5.12 0.28 <0.001 0.104 0.007 Kibble traits (After Dryer) Starch gelatinization (%) 92.8 93.8 94.6 91.9 94.0 0.48 0.673 0.051 0.545 Piece density (kg/m 3 ) 0.37 0.38 0.36 0.45 0.41 0.008 <0.001 0.018 0.103 Radial expansion rate 4.1 2.9 3.0 2.8 3.1 0.04 0.578 0.701 0.008 Specific length (mm/g) 47.4 57.0 58.7 51.4 51.7 0.69 <0.001 0.317 <0.001 Cutting force (kg.f) i 2.4 4.0 4.1 3.4 3.1 0.06 <0.001 0.2122 <0.001 37 Figure 2. Field scanning electron micrograph of diets with sugarcane and wheat bran fiber. A, B, C, D, and E – Internal area (increased 25x) of CO, SCL, SCS, WBL, WBS, diets, respectively. A1, B1, C1 and D1- External area (increased 25x) of CO, SCL, SCS, WBL, WBS, diets, respectively. CO= control diet, without added fiber source, SFL= sugarcane fiber, large particle; SFS= sugarcane fiber, small particles; WBL= wheat bran, large particles; WBS= wheat bran, small particles. A B C D E A1 B1 C1 D1 E1 38 4. Discussion This study evaluated the effect of the type and particle size of fiber sources on extrusion parameters in diets for dogs. Although processing conditions were kept constant, some fluctuation on preconditioner temperature, output moisture, and output mass were noted on experiment 1. The reductions in moisture and temperature could be related to a reduced capacity of steam absorption of the dough after guava fiber inclusion. It is possible that the fiber conducted the steam, reducing its condensation on the dough, as discussed by Robin et al. (2012) about kibble expansion. On experiment 2 we observed a tendency for lower preconditioner moisture for the large particle size fibers, reinforcing this hypothesis and suggesting that the fiber structure might play an important role for this effect. An increased resistance of the dough to the flow inside the extruder could explain the differences in extruder amperage, temperature and pressure shown on experiment 1. Guava fiber addition resulted in an increased motor work, electric energy consumption, and shear, increasing the temperature, pressure, and SME implementation. On experiment 2 this effect of fiber was not so evident, probably due the higher feed rate (preconditioner output mass) of the CO diet, interfering on the outcomes. However, when the mass production was corrected, the SME implementation was also higher for the fiber supplemented diets compared to the CO diet. Altan et al. (2008), Upadhyay et al. (2010), and Karkle et al. (2012) already demonstrated this effect of fiber for human extruded foods. At least partially, this can be related to a reduction in the available water during the extrusion when fiber is added to the recipe, thus reducing dough fluidity (Hill, 2003). In experiment 1, the treatments with higher guava fiber inclusion also showed less in-barrel moisture (preconditioner output moisture). This was an unexpected outcome, and could have contributed to the increase in dough flow 39 resistance. Nevertheless, the differences in moisture are small, less than 10% between CO and GF12. For experiment 2, in-barrel moisture did not vary between diets, and the water holding effect of fiber was more evident increasing dough resistance. An increase in shear might explain the higher STE, and TSE implemented after fiber addition, effect shown in both experiments. The reduction on STE:SME ratio happened because the mechanical energy increased more than the thermal energy addition. Wheat bran and sugarcane fiber differ regarding their influence on processing traits. Wheat bran inclusion resulted in higher SME implemented, suggesting a higher resistance to flow in comparison with sugarcane fiber diets. An effect of fiber particle size was also observed, but only the WBL resulted in increased SME, and no differences were seen between SFL and SFS. These differences are likely related to particular physical characteristics and influences of each ingredient on the flow dynamic inside the extruder. The increased mechanical energy consumption has implications on the production cost of the food. The utilization of electric power increased 85% after 12% addition of guava fiber in experiment 1, and wheat bran and sugarcane inclusion resulted in a mean elevation of 22% in experiment 2. The increase was particularly important for the WBL, with a value 57% higher than for CO. These increases in electric power add cost to the production of fiber supplemented diets, and alternatives to a more rational processing condition should be studied. Kibble macrostructure was worse on fiber-supplemented diets on both experiments, showing the negative effects of fiber on kibble formation and macrostructure. A reduction in starch cooking was observed in experiment 1 with guava fiber addition, and in experiment 2with large particle size fibers. The starch fraction works as a thermoplastic polymer during extrusion. When water, energy, and time are sufficient in the processing, the starch granule 40 loses its crystallinity, swells, and disrupts, forming an amorphous mass that binds all food components forming a continuous structure (Ding et al., 2005). Energy implementation was higher for fiber supplemented diets, and does not explain the lower starch gelatinization and worse kibble formation on fiber supplemented diets in both experiments. A possible explanation could be a limitation on available water for starch cooking with fiber inclusion (Hill, 2003; Nelson, 2001). It is possible that the added fiber retained the water (Santos et al., 2008; Khanna and Tester, 2006), that then became less available for hydration, dough lubrication, and starch gelatinization. High extrudate density with a shift toward longitudinal instead of radial expansion were verified on kibbles of both experiments after fiber inclusion. This effect was already demonstrated after cellulose inclusion on human food (Chinnaswamy and Hanna, 1991). Usually these are the result of the formation of kibbles with small size and thicker cell walls (Robin et al, 2012). Imaging from the diets from our study suggest that this is the case with high guava fiber and large particle size fiber inclusion, although this was only qualitatively evaluated. These structural changes could be responsible for the increased of kibble hardness (cutting force) with fiber addition. The higher cutting force is explained by the smaller cells with thicker walls, and by the strengthen effect of the fiber particles against the rupture of the molten starch mass. Fiber effect in expansion and cell formation has been also attributed to its water binding capability (Moraru and Kokini, 2003; Jin et al., 1994; Camire and King, 1991). In addition, due to its structure, fiber can conduct the water vapor out of the kibble, reducing the flash off effect on cell formation (Lue et al., 1991; Robin et al., 2012), directly affecting expantion. We also noted differences between the effect of wheat bran and sugarcane on kibble formation, with sugarcane inclusion resulting in lower piece density and higher specific length and cutting force. This means that sugarcane fiber induce higher longitudinal 41 expansion, with lighter and harder kibbles than wheat bran. This diverse structure forming traits should be explored in specific diets when differences in density or specific influences on chewing dynamics are of interest. Regarding particle size, the finely grind fibers favoring formation of kibbles with lower piece density, signaling that fiber particle reduction might be a means of increase kibble expansion. Several implications for this interference of fiber addition on expansion, cell structure formation and increase in hardness should be considered. Absorption of fat added by surface coating could be less efficient in fiber enriched diets. The lesser internal open space and external porous may impede the internal migration of fat, which will stay on surface. Higher residence time on coating equipment, or even special coating systems with vacuum may be required for high fat addition. The diets from the present experiment were evaluated by an expert human panel to describe their sensory traits (Koppel et al., 2014). Fracturability, and initial crispness were physical attributes that differentiated the CO with the fiber supplemented diets. This could affect acceptability of the kibble by the dogs, although there are very few studies that evaluate the influence of the physical attributes of the kibble on canine diet palatability. It is possible that these physical alterations change the mastication kinetics or sensation, altering the acceptability of the food by the dog. The interference of fiber supplementation of chewing sensation, reducing food acceptance For is well described in human beings (Karkle et al., 2012a, Martin et al., 2013). 5. Conclusions The addition of fiber sources to dog food formulations increases electric energy required to extrude, and may reduce starch cooking and kibble expansion, leading to the production of denser and harder kibbles. 42 Acknowledgments The authors acknowledge the financial and technical support of Guabi PetCare to the Laboratory of Research in Nutrition and Nutritional Diseases of Dogs and Cats “Prof. Dr. Flávio Prada”, to Dilumix, Leme, SP, Brazil for the donation of the fiber sources and the partial financial support of the study, and Manzoni Industrial Ltda, Campinas, SP, Brazil for the extruder. 43 References Altan, A., Mccarthy, Kathryn L., Maskan, M., 2008. Evaluation of snack foods from barley– tomato pomace blends by extrusion processing. J. Food Eng. 84, 231-242. Baik, B.K., Powers, J., Nguyen, L.T., 2004. Extrusion of regular and waxy barley flours for production of expanded cereals. Cereal Chem. 81, 94–99. Boac, J.M., Maghirang, R.G., Casada, M.E., Wilson, J.D., JUNG, Y.S., 2009. Size distribution and rate of dust generated during grain elevator handling. Appl. Eng. Agric. 25, 533–541. Brennan, M.A., Merts, I., Monro, J., Woolnough, J., Brennan, C.S., 2008. Impact of guar gum and wheat bran on the physical and nutritional quality of extruded breakfast cereals. Starch. 60, 248–256. Calabrò, S., Cutrignelli, M.I., Bovera, F., Carciofi, A.C., Tudisco, R., Guglielmelli, A., Piccolo, G., 2008. In vitro evaluation of different fiber sources and potential prebiotics for dogs. In: Congress of the European Society of Veterinary and Compararative Nutrition, Vienna, Austria. Anais: University of Veterinary Medicine Vienna. 1, 63. Camire, M.E., Dougherty, M.P., Briggs, J.L., 2007. Functionality of fruit powders in extruded corn breakfast cereals. Food Chem. 101, 765-770. Camire, M.E., King, C.C. 1991. Protein and fiber supplementation effects on extruded cornmeal snack quality. J. F