Agricultural University of Athens
Εκτίμηση της αλλοίωσης του κρέατος με σύγχρονες μεθόδους
Abstract
dc:descriptionQuality is generally a subjective and sometimes elusive term. Freshness and safety ofmuscle food products are generally considered the most important contributors to quality. Thespoilage of meat may arise from several intrinsic and extrinsic factors including physical damageor chemical changes. However, it is generally accepted that detectable organoleptic spoilage is aresult of decomposition and the formation of metabolites caused by the growth of microorganisms.Muscle foods are described as spoiled if organoleptic changes make them unacceptable to theconsumer. It is therefore crucial to have valid methods to monitor freshness and quality. Onechallenge facing the meat industry is to obtain reliable information on meat quality throughout theproduction process, which would ultimately provide a guaranteed quality of meat products forconsumers. The ideal method for the on-line microbiological analysis of meat would be rapid, noninvasive,reagentless, and relatively inexpensive and these requirements can be met via theapplication of a spectroscopic approach, in combination with any appropriate data analysis strategybased on statistics or machine learning.In the current thesis, the microbial association of meat was monitored in parallel with thechemical changes, pH measurements and sensory analysis. For this purpose, two differentexperiments were designed and several chemical analytical techniques were applied to quantifybiochemical changes in an attempt to monitor meat spoilage. In the first experiment, fresh mincedpork was stored aerobically and under modified atmospheres packaging (MAP) at five differenttemperatures (0, 5, 10, 15 and 20°C) and microbiological analysis in terms of total viable counts,Pseudomonas spp., Brochothrix thermosphacta, lactic acid bacteria, Enterobacteriaceae and yeastand moulds was performed in parallel with Fourier transform infrared (FTIR) spectroscopy,electronic nose (Libra Nose), high performance liquid chromatography (HPLC), gaschromatography/mass spectroscopy (HS/SPME GC/MS), and several molecular techniques (Rep-PCR, Sau-PCR and PFGE). In the second experiment sterile tissue of pork meat fillet wasinoculated with selected spoilage microorganisms at two different temperatures (4 and 10 °C) andtwo packaging conditions (air and MAP), in order to study the metabolic profile produced duringstorage of meat with GC/MS. The first component of the study was designed to evaluate the potential of FTIRspectroscopy as a rapid, reagent-less and non-destructive analytical technique in order to quantifybiochemical changes occurring in fresh minced pork meat in the attempt to monitor spoilage. Forthis reason, partial least squares (PLS) models were constructed to correlate spectral data fromFTIR with minced pork meat spoilage during storage (aerobic and MAP) of meat samples atdifferent storage temperatures (0, 5, 10, and 15°C). Spectral data were collected from the surfaceof meat samples in parallel with microbiological analysis to enumerate the population of totalviable counts, Pseudomonas spp., Brochothrix thermosphacta, lactic acid bacteria andEnterobacteriaceae. Qualitative interpretation of spectral data was based on sensory evaluation,using a three point hedonic scale, discriminating meat samples in three quality classes, namelyfresh, semi-fresh and spoiled. The purpose of the developed models was to classify minced porksamples in the respective quality class, and also to correlate the population dynamics of themicrobial association with FTIR spectra. The obtained results demonstrated good performance inclassifying meat samples in one of the three pre-defined sensory classes. The overall correctclassification rate for the three sensory classes was 94.0% and 88.1% during model calibration andvalidation, respectively for minced pork stored at aerobic conditions. Furthermore, PLS regressionmodels were also employed to provide quantitative estimations of microbial counts during meatstorage. The performance was based on graphical plots and statistical indices (bias factor, accuracyfactor, standard error of calibration, standard error of prediction, and correlation coefficient). Thevalues of the bias factor were close to unity for all microbial groups indicating no systematic biasof the models. Moreover, the calculated values of the accuracy factor showed that the averagedeviation between predictions and observations was 7.5% and 7.9%, 6.0% and 5.9% for totalviable counts and Pseudomonas spp., stored at air and MAP respectively. Finally, correlationsabove 0.80 between observed and estimated counts were observed for both training and test datasets.The performance of a portable quartz microbalance based electronic nose has beenevaluated in monitoring aerobically and under MAP packaged minced pork spoilage at differentstorage temperatures (0, 5, 10, 15, and 20°C). Electronic nose data were collected from theheadspace of meat samples in parallel with data from microbiological analysis for the enumerationof the population dynamics of total viable counts, Pseudomonas spp., Brochothrix thermosphacta,lactic acid bacteria and Enterobacteriaceae. Qualitative interpretation of electronic nose data was based on sensory evaluation discriminating samples in three quality classes (fresh, semi-fresh, andspoiled). First, the data collected from e-nose were subjected on Principal Component Analysis(PCA) in order to minimize the dimensionality of the data. The subjected data were used as inputon Discriminant Function Analysis (DFA) and Support Vector Machine (SVM). DFA was used inorder to classify pork samples in the respective quality class, while Support Vector Machine(SVM) regression models using radial basis kernel function were developed in order to correlatethe population dynamics of the microbial association with electronic nose sensor responses. Theobtained results demonstrated good performance in discriminating meat samples in one of the threepre-defined quality classes. Overall classification accuracies of prediction was 86% and 82% wereobtained for samples stored under aerobic and MAP conditions respectively. Additionally, SVMregression model development, can be efficiently used for the estimation of the population ofvarious microbial groups, which was demonstrated by the correlation for each group ofmicroorganism.Two further studies were conducted using rapid analytical methods such as HPLC analysisof organic acids and GC/MS to monitor the changes in organic acids and volatile compoundsrespectively, which are present in the meat substrate during storage. Minced pork samples storedunder aerobic and MAP at 0, 5, 10, 15 and 20ºC were monitored with both analytical methods, andseveral compounds found to be possible chemical indicators. Moreover, the applied PLS-R modelsgave estimations of the different microbial populations (e.g. TVC: 0.792 for the HPLC and 0.866for the GC/MS models) and the FDA facilitated the discrimination of the samples regarding theirfreshness (correct classification of 97.60% for HPLC and 76.06% for GC/MS). Additionally, theGC/MS was used to identify the volatile compounds present in the sterile meat. Interestinginformation about the evolution of spoilage under different group of inoculated microorganismand different temperature and packaging conditions was demonstrated. The results were analyzedby PLS-DA and one-way-anova. The results provided correlations between volatiles and specificspoilage microorganisms.Finally, a total of 306 colonies were isolated from the selective medium for Brochothrixspp., during the spoilage of minced pork stored at 0, 5, 10 and 15 °C and packed aerobically andunder modified atmosphere packaging conditions (MAP). Brochothrix biodiversity was assessedby Pulsed Field Gel Electrophoresis (PFGE), and representative strains were further analyzed byRep-PCR using primer (GTG)5 and Sau-PCR with primers SAG1and SAG2. Although, different results were obtained from the different methods, a significant diversity among isolates recoveredfrom aerobic conditions was observed. On the contrary, isolates from MAP showed a lower degreeof heterogeneity. The storage conditions affected the Brochothrix diversity, the strains isolated inthe initial stage being different from the ones present at the final stage of storage at chilltemperatures. A representative number of isolates, based on the results of the clustering bymolecular methods, were subjected to 16S rRNA gene sequencing, revealing that all belonged toBrochothrix thermosphacta.
Degree
thesis:*- Grantor dc:publisher
- Agricultural University of Athens
- Year dc:date
- 2013
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Papadopoulou, Olga
- Παπαδοπούλου, Όλγα
Subjects
dc:subject × 22- Αλλοίωση κρέατος
- Χοιρινό κρέας
- Ηλεκτρονική μύτη
- Φασματοσκοπία υπερύθρου με μετασχηματισμό Fourier
- Χημειομετρία
- Μεταβολομική
- Αλυσιδωτή αντίδραση πολυμεράσης
- Εφήμερη αλλοιωγόνος χλωρίδα
- Spoilage of meat
- Pork meat
- Electronic nose
- FTIR
- Chemometrics
- Metabolomics
- PCR
- Specific spoilage microorganism
- Γεωπονικές Επιστήμες και Κτηνιατρική
- Γεωπονική Βιοτεχνολογία
- Άλλες Γεωπονικές Επιστήμες
- Agricultural and Veterinary Sciences
- Agricultural Biotechnology
- Other Agricultural Sciences
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/35030