Early postmortem determination of porcine meat quality using Raman spectroscopy

The chemical and structural composition of biological tissues can be non-invasively measured using Raman spectroscopy. The resulting Raman spectrum can be used for identification and quantification. In case of multi-component mixtures, this is complicated due to superimposing signals of individual compounds hence multivariate methods are mostly applied for the analysis of Raman spectra. However, these methods provide only little help to understand the physical and chemical processes causing the spectral alterations. Therefore, the aim of this thesis was to determine the changes in the Raman spectra using the example of pork meat in the first hours after slaughter, to identify the underlying mechanisms and to evaluate the potential of early postmortem Raman spectra to predict important meat quality traits. At first, pH and lactate concentration were chosen as indicators as they are known to correlate with the metabolic state of the muscle, which is again closely related to the resulting meat quality. Here, the pH value was shown to be calculable from only two vibrations of the Raman spectra assigned to the terminal phosphate moiety. More accurate predictions are possible with a multiple linear regression model based on signals of lactate, glycogen, creatine (Cr), phosphocreatine (PCr), ATP, IMP, the phosphate and carbonyl group or with a partial least squares regression model based on the whole spectrum. However, the determination of the lactate concentration is complicated due to superimposing signals of other metabolites. To quantify the contribution of individual components, the spectral alterations were simulated in the pre-rigor phase between 1 and 2 h and in the rigor phase between 2 and 8 h after slaughter. The simulation includes difference spectra of the energy metabolites PCr and Cr, glycogen and lactate, ATP and IMP as well as hydrogen and dihydrogen phosphate. Additionally, the Raman signals of alpha-helical proteins, phosphorylated sugars and the difference spectrum of oxy- and deoxymyoglobin add intensity to the observed alterations. The agreement between measured and simulated spectra proves that the Raman spectra indeed provide a detailed fingerprint of the metabolic state of the early postmortem muscle. In parallel, a portable Raman system with control software was developed and successfully tested in two field studies in commercial abattoirs. For the first time, the quality traits pH45, pH24, drip loss, color and shear force were predicted from Raman spectra measured only 1–2 h post mortem. Besides pH45, these quality traits cannot currently be measured early postmortem and are only available 1–3 days after slaughter via partly invasive and time-consuming reference analysis. The promising predictions were confirmed for pH45, pH24 and drip loss in a second study. In both studies, the PLSR models are mostly weighting Raman signals of the energy metabolites. In this thesis, the potential of Raman spectroscopy was proven to rapidly and noninvasively determine important quality traits of pork meat based on early postmortem spectra. Thereby, signals of ubiquitous compounds of the postmortem metabolism are weighted by the prediction models. In future, Raman spectroscopy could therefore allow for an early assessment of quality traits of pork, but also other sorts of meat, at the slaughterline in abattoirs.

Level thesis:degree_level

thesis.doctoral

Grantor dc:publisher

Universität Bayreuth

Year

2014

Author dc:creator

• Scheier, Rico

Contributors dc:contributor

• Köhler, Jürgen

Repository record source_url

https://epub.uni-bayreuth.de/id/eprint/1705/

OAI identifier oai:identifier

oai:epub.uni-bayreuth.de:1705