In vitro digestion assays for estimation of bioaccessibility and bioavailability of minerals and evaluation of NIR and chemometrics viability in adulterant detection in goat milk and its derivatives

Author(s)

José Luan da Paixão Teixeira

• Document type: Doctoral Thesis
• Press: Campinas, SP.
• Institution: Universidade Estadual de Campinas (UNICAMP). Faculdade de Engenharia de Alimentos
• Defense date: 2022-03-07

Examining board members

Juliana Azevedo Lima Pallone; Douglas Fernandes Barbin; Marcelo Antônio Morgano; Patrícia Valderrama; George Rodrigo Beltrão da Cruz

• Advisor: Juliana Azevedo Lima Pallone

Abstract

Data on the bioaccessibility of minerals in goat products and evaluation of the potential of NIR to detect fraud in this type of food are non-existent. Therefore, the objectives of this thesis were to evaluate the bioaccessibility of calcium (Ca), magnesium (Mg) and zinc (Zn) in goat milk and its dairy using static and dynamic models, in addition, to developing multivariate classification models using tools chemometrics to analyze the spectral data obtained in the NIR, to detect adulterants in these products. For milks, yogurts and cheeses, the solubility method showed higher percentages of bioaccessibility of minerals when compared to of dialysis, higher percentages of solubility of Ca (53.0%), Mg (67.8%) and Zn (92.9 %) were presented by Minas frescal goat cheese, pasteurized goat milk and Minas frescal cow cheese, respectively. Although, goat dairy beverages had higher Ca content, when compared to bovine dairy beverages, cow dairy beverages had the highest percentage of bioaccessibility for this mineral, for all static models of in vitro digestion used (solubility, dialysis and INFOGEST). In the INFOGEST method, coupled to Caco-2 cells, the amount of Ca present in fresh cheeses (Minas frescal of goat and cow) and matured (Blue and Pyramid goat cheeses) may be a limiting factor for its solubility. In the dynamic model, where was used a membrane bioreactor system that mimicks the human gastrointestinal tract, the highest percentage of Ca absorption was presented by Minas frescal cow cheese (9.85%), for Mg the Pyramid goat cheese (32.8%) and for Zn the Minas frescal goat cheese (1.08%). In the study using NIR and chemometric to detect adulterants in goat milk, we observed that the PLS-DA model, considering 2 and 5 classes, was developed with 7 latent variables (VL) and variable selection (initial and final section of the spectral data and the second overtone of H-O, were removed) obtained 100% sensitivity and specificity. The PLS-DA model was also able to correctly classify samples of yogurts and goat cheeses, adulterated with cow milk, with different percentages of adulteration (10, 15 and 20%), showing sensitivity and specificity values of 100% in the cross-validation and prevision. The combination of NIR and DD-SIMCA proved to be efficient to control the authenticity of goat milk beverages, in the total spectrum (10.000 - 4000 cm-1) it was verified 95 and 100% of correct attributions for authentic and adulterated samples, respectively. After the variable selection process (5500-4000 cm-1), the results indicated 100% of correct classification for the samples. Thus, we conclude that variations in the gastrointestinal conditions employed by the in vitro digestion simulation methods resulted in different estimates of the bioaccessibility of essential minerals in goat and cow milk and dairy products, however, the general behavior between the absorption potential of minerals in samples did not change, with the use of different models. In addition, the NIR and chemometric studies yield positive results for use in detecting fraud in goat milk and dairy products (AU)

• FAPESP's process: 18/08864-8 - Minerals bioacessibility in goat dairy produts and evaluation of applying NIR and chemometrics to detect adulterations in goat's milk and dairy
• Grantee: José Luan da Paixão Teixeira
• Support Opportunities: Scholarships in Brazil - Doctorate