Autofluorescence quantifies collagen in sausage batters with a large variation in myoglobin content

Summary

Designed experiments have been used to separate the effect of muscle type from the effect of gross chemical components (total muscle protein minus collagen, collagen and fat) on the autofluorescence emission spectrum obtained from sausage batter. Provided extreme muscles with large variation in myoglobin content were chosen, myoglobin absorbance was the dominating cause of variation in the spectral range normally thought interesting for collagen quantification. Even some information of the types of myoglobin derivatives present in the batter seemed detectable in the spectra. The batters were low in elastin and therefore collagen was the second most important cause of variation in the autofluorescence emission spectra. Collagen could only be modelled from autofluorescence emission spectra provided a multivariate regression model was used, preferable after some type of normalization of spectra. Pork fat and total muscle protein minus collagen had only a minor impact on the variation in autofluorescence emission spectra for excitation wavelengths 332-380 nm. The lowest prediction error obtainable for collagen content of sausage batter through emission spectra was 0.48%. The method of autofluorescence gave lower prediction errors for collagen content than did the method of near infrared reflectance.