Wavelength dependence of light-induced lipid oxidation and naturally occurring photosensitizers in cheese.

Degradation of the potential photosensitizers, riboflavin, chlorophyll, and porphyrin, in Danbo cheese by monochromatic light of wavelength 366, 436, or 546 nm was studied. Three cheeses were investigated, two conventional (16% fat and 25% fat) and one "organic" (25% fat). The effect of illumination was measured by fluorescence spectroscopy and analyzed using multiway and multivariate data analysis. Riboflavin was found to degrade only by 436 nm light, whereas chlorophylls and porphyrins also were influenced by 436 and 546 nm light. The organic cheese had the largest chlorophyll content both before and after similar light exposure, and no change in chlorophyll of this cheese was observed for any of the illumination wavelengths. Upon light exposure of the cheeses, volatile compounds were formed, as analyzed by gas chromatography-mass spectrometry (GC-MS). The relative concentrations of methyl butanoate, 1-pentanol, benzaldehyde, 2-butanone, 2-heptanone, and butyl acetate were found to weakly correlate with the surface fluorescence intensity. 1-Pentanol and the ketones are secondary lipid oxidation products, consistent with a chemical coupling between photosensitizer degradation and formation of volatile lipid oxidation products.

Application of fluorescence spectroscopy in the evaluation of light-induced oxidation in cheese.

Light-induced oxidation of semihard cheese has been evaluated by fluorescence spectroscopy. The cheese was packaged in two packaging materials and exposed to different storage conditions, which included light/dark storage, oxygen availability, and storage time (0, 4, 7, 14, 21, 42, 70, or 84 days). Fluorescence excitation-emission matrices (EEM) were analyzed by PARAFAC, which gave an estimation of the pure excitation and emission spectra of the fluorophores and the concentrations of these. This analysis showed the presence of components such as tryptophan, tyrosine, vitamin A, fluorescent oxidation products, and riboflavin. Effects of packaging material, light or dark storage, and storage time were seen. However, there was no effect of the oxygen availability on the fluorescence measurements. The score values obtained by the PARAFAC models and chemical and physical measurements were analyzed together by principal component analysis (PCA). The loadings showed a separation of the variables into three groups; the first group was related to oxidation, the second group was related to the degradation of both riboflavin and vitamin A, and the third group was linked to the protein structure.

Fluorescence of muscle and connective tissue from cod and salmon.

Autofluorescence of salmon and cod muscle was measured and compared with autofluorescence of collagen type I and type V. Similarities between fluorescence of fish muscle and collagen were found in that the same peaks were obtained around 390, 430, and 480 nm. These similarities are supported by principal component analyses. Texture and gaping score were predicted from the fluorescence spectra by partial least-squares regression. However, the predictions did not perform well. Relating fluorescence to the gaping score gave a prediction error of 0.91 and a correlation of 0.43 when measuring gaping on a scale from 0 to 5. There was no relation between texture and fluorescence spectra. Fluorescence of fish muscle could be related to the storage time. However, this relation seemed not to be induced by changes in collagen.

A chemometric analysis of ligand-induced changes in intrinsic fluorescence of folate binding protein indicates a link between altered conformational structure and physico-chemical characteristics.

Ligand binding alters the conformational structure and physico-chemical characteristics of bovine folate binding protein (FBP). For the purpose of achieving further information we analyzed ligand (folate and methotrexate)-induced changes in the fluorescence landscape of FBP. Fluorescence excitation and emission two-dimensional (2D) spectra were recorded over a wide range of wavelengths on a Perkin-Elmer LS 55 spectrofluorometer at varying pH in different buffers, and the resulting three-dimensional data were subjected to a chemometric analysis, parallel factor analysis (PARAFAC). The most important finding was the occurrence of two maximum intensity emission wavelengths of tryptophan, 350 nm (component one) and 330 nm (component two). In contrast to the first component, the score of the short wavelength component increased with increasing ligation of FBP. Since the emission wavelengths of indole groups in tryptophan shorten with increasing distance from the solvent surface of proteins, an increasing number of the 11 tryptophan residues seem to reorientate from the solvent surface to the interior of FBP with increasing ligation. The sharp decrease in hydrophobicity at pI=7-8 following binding of folate accords fairly well with the disappearance of strongly hydrophobic tryptophan residues from the solvent-exposed surface of FBP. The PARAFAC has thus proven useful to establish a hitherto unexplained link between parallel changes in conformational structure and physico-chemical characteristics of FBP induced by folate binding. Parameters for ligand binding derived from PARAFAC analysis of the fluorescence data were qualitatively and quantitatively similar to those obtained from binding of radiofolate to FBP. Herein, methotrexate exhibited a higher affinity for FBP than in competition with radiofolate. This could suggest a rapid and firm complexation of folate to FBP, blocking access of competing ligands.