Protein and milk-casein contents have great influence on rennet properties and cheese yield. The selection of dairy cattle with genetic characteristics suitable for milk transformation is of great interest for dairy farms and firms. The amount and the type of casein in milk strictly depend on genetic factors, also related to some phenotypic aspects as casein micelles dimension and number, and sub-fractions distribution. The aim of this work was to determine protein and casein contents by NIRS and to test its ability for studying casein structure and interactions by comparing the performance of different NIR equipments and sample presentation modes. Raw bulk milk samples were collected from different farms in the Asturia region (Spain) during one month period. Chemical analyses (in duplicates) of protein (TP%) and Non Caseinic Nitrogen (NCN%) contents were determined by Kjeldahl’s method. Casein content was calculated as the difference between NCN and TP contents. Milk samples were split in two parts: the first one was ultra-centrifuged (native casein); the second one was acidified (pH=4,6) and centrifuged (acid casein). The two types of casein were then reconstituted in phosphate buffer (pH=6.8). Milk and casein spectra were collected at 37°C with two equipments, an FT-NIR (Perkin-Elmer, USA) in transflectance mode and a Foss-NIRSystem 6500 (Foss, Denmark) in reflectance mode. Data were processed by the Unscrambler software v.9.2 (Camo Inc., Norway). As expected, PLSR analysis performed on milk transflectance spectra showed a very satisfactory results. Good prediction ability in validation was highlighted for TP and casein contents (Rval=0,817 and 0,847 respectively). An explorative PCA, performed on casein reconstituted spectra, showed good separation between native and acid casein samples suggesting the possibility to investigate casein interactions and structure. NIRS confirmed the ability to quantify TP and casein contents in raw milk. Electrophoretic analyses are in progress in order to separate casein sub-fractions and to evaluate inter and intra molecular interactions among them. The same approach will be applied on individual milk samples.
Marinoni, L., Cattaneo, T. M. P., Soldado Cabezuelo, A., Gonzalez Arrojo, A., De La Roza Delgado, B., Aleandri, R., NIRS ability in the determination of casein content and in the study of micelles interactions, Paper, in Proceedings of 15th International Conference on Near Infrared Spctroscopy, (Cape Town, 13-20 May 2011), ICNIRS, Cape Town 2012: 462-465 [http://hdl.handle.net/10807/57396]
NIRS ability in the determination of casein content and in the study of micelles interactions
Marinoni, Laura;
2012
Abstract
Protein and milk-casein contents have great influence on rennet properties and cheese yield. The selection of dairy cattle with genetic characteristics suitable for milk transformation is of great interest for dairy farms and firms. The amount and the type of casein in milk strictly depend on genetic factors, also related to some phenotypic aspects as casein micelles dimension and number, and sub-fractions distribution. The aim of this work was to determine protein and casein contents by NIRS and to test its ability for studying casein structure and interactions by comparing the performance of different NIR equipments and sample presentation modes. Raw bulk milk samples were collected from different farms in the Asturia region (Spain) during one month period. Chemical analyses (in duplicates) of protein (TP%) and Non Caseinic Nitrogen (NCN%) contents were determined by Kjeldahl’s method. Casein content was calculated as the difference between NCN and TP contents. Milk samples were split in two parts: the first one was ultra-centrifuged (native casein); the second one was acidified (pH=4,6) and centrifuged (acid casein). The two types of casein were then reconstituted in phosphate buffer (pH=6.8). Milk and casein spectra were collected at 37°C with two equipments, an FT-NIR (Perkin-Elmer, USA) in transflectance mode and a Foss-NIRSystem 6500 (Foss, Denmark) in reflectance mode. Data were processed by the Unscrambler software v.9.2 (Camo Inc., Norway). As expected, PLSR analysis performed on milk transflectance spectra showed a very satisfactory results. Good prediction ability in validation was highlighted for TP and casein contents (Rval=0,817 and 0,847 respectively). An explorative PCA, performed on casein reconstituted spectra, showed good separation between native and acid casein samples suggesting the possibility to investigate casein interactions and structure. NIRS confirmed the ability to quantify TP and casein contents in raw milk. Electrophoretic analyses are in progress in order to separate casein sub-fractions and to evaluate inter and intra molecular interactions among them. The same approach will be applied on individual milk samples.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.