Introduction

MALDI-TOF protein profiling and machine learning were developed to distinguish and classify meat from different species of livestock animals, with the aim of establishing a methodology useful for certifying meat species.

Methods

The samples were beef (filet mignon and sirloin), pork (loin), chicken (breast fillet) and tilapia fish (fillet). Approximately 3 cubic mm were excised, without apparent fat. After maceration with a sterile plastic pestle, proteins were extracted with a mixture of acetonitrile/water/trifluoroacetic acid, 50:49.9:0.1 v/v, followed by centrifugation at 13,000 g for 2 minutes. Supernatants were mixed with alpha-cyano-4-hydroxycinnamic acid, and mass spectra were acquired in a MALDI Biotyper Sirius One (Bruker Daltonics), with external calibration. Peak identification and meat classification were carried out with MALDI Biotyper Compass Explorer 4.1 and ClinProTools 3.0 (Bruker Daltonics) software.

Results

Meat protein extraction methodology successfully produced nice spectra for fresh and frozen/cooked meat. Fresh samples of beef (n=6), pork (n=6), chicken (n=4) and tilapia (n=2) were analyzed, and clearly distinct peaks were observed for all types of meat. Considering peaks with signal/noise equal to or greater than 5, the mean numbers of peaks obtained were 30, 26, 17 and 41, respectively, for beef, pork, chicken and tilapia. Analysis with the Biotyper algorithm allowed correct identification with an average score equal to or greater than 2,000 for beef after freezing or cooking. The mass spectra obtained also allowed the distinction between Nelore and Angus breeds, but so far they have not differentiated between the filet mignon and sirloin cuts. PCA classification revealed possible biomarkers for meat types.

Conclusions

Different meat species can be correctly classified with the MALDI-TOF mass spectrometry and may be used in the future for meat certification. The new aspect presented here comes from the possibility of distinguishing cattle breeds.