Assessment of meat authenticity using bioinformatics, targeted peptide
biomarkers and high-resolution mass spectrometry
Alberto Ruiz1, Erik Husby2, Charles T. Yang2, Dipankar Ghosh2 and Francis Beaudry1*
1-Département de biomédecine vétérinaire, Faculté de médecine vétérinaire, Université de Montréal, Saint-Hyacinthe, Québec, Canada
2-Department of Environment & Food Safety, Thermo Fisher Scientific, San Jose, USA
*Corresponding author:
Francis Beaudry, Ph.D.
Associate Professor in Analytical Pharmacology Département de Biomédecine Vétérinaire Faculté de Médecine Vétérinaire
Université de Montréal 3200 Sicotte
C.P. 5000
Saint-Hyacinthe, QC Canada J2S 7C6
Keywords: High Resolution Mass spectrometry, High performance liquid chromatography, Proteomics, Food, Meat, Authenticity, Biomarkers
Abstract
In recent years, we observed a significant increase of food fraud ranging from false label claims to the use of additives and fillers to increase profitability. Recently in 2013, horse and pig DNA were detected in beef products sold from several retailers. Mass spectrometry has become the workhorse in protein research and the detection of marker proteins could serve for both animal species and tissue authentication. Meat species authenticity will be performed using a well defined proteogenomic annotation, carefully chosen surrogate tryptic peptides and analysis using a hybrid quadrupole-Orbitrap mass spectrometer. Selected mammalian meat samples were homogenized, proteins were extracted and digested with trypsin. The samples were analyzed using a high-resolution mass spectrometer. The chromatography was achieved using a 30 minutes linear gradient along with a BioBasic C8 100 × 1 mm column at a flow rate of 75 µL/min. The mass spectrometer was operated in full-scan high resolution and accurate mass. MS/MS spectra were collected for selected proteotypic peptides. Muscular proteins were methodically analyzed
in silico in order to generate tryptic peptide mass lists and theoretical MS/MS spectra. Following
a comprehensive bottom-up proteomic analysis, we were able to detect and identify a proteotypic myoglobin tryptic peptide [120-134] for each species with observed m/z below 1.3 ppm compared to theoretical values. Moreover, proteotypic peptides from myosin-1, myosin-2 and -hemoglobin were also identified. This targeted method allowed a comprehensive meat speciation down to 1% (w/w) of undesired product.
Introduction
In recent years, we have witnessed the internationalization of food production, distribution and retailing. Interestingly, we observed a significant increase of food fraud ranging from false label claims to the use of additives and fillers to increase profitability [Everstine et al. 2013; Ellis et
al., 2012]. Recently in 2013, horse and pig DNA were detected in beef products sold from several
retailers [DG Health and Consumers, European Commission, 2013]. Contaminated meat products are not only misleading consumers but it has ethical and health implications. We believe that certification of meat authentication is very important for consumers health and well-being. It is also an important concern for regulatory agencies (e.g. Food and Drug Administration, Directorate General for Health and Food Safety, Canadian Food Inspection Agency and many others) as well the entire food industry.
In animal science, proteomic studies have emerged over the last decade particularly in the field of allergen and toxin analyses [Rogniaux et al. 2015; Kalb et al. 2015; Attiten et al., 2014; Bao et
al., 2012, Lippolis and Reinhardt 2008]. However, the food industry has an expanded interest in
proteomic research particularly if data can help to provide information on muscle growth and specific meat quality characteristics [Zheng et al., 2015; Gobert et al., 2014]. Thus, proteomic research in the meat science field is generally motivated by the revenue that can be generated by increasing meat production and the improvement in meat quality. Globally, meat consumption is one of the highest when comparing data with all food categories. Food producers and distributors are continuously considering finding more cost-effective ways to produce meat products for human and animal consumption. On the regulatory side, food-testing laboratories are relentlessly looking to find new analytical strategies to test meat product for authenticity and adulteration.
Meat authenticity in food testing laboratories has been traditionally performed using polymerase chain reaction (PCR) and enzyme-linked immunosorbent assay (ELISA) [Kumar et al. 2015; Hsieh and Ofori, 2015]. These methods required intensive customization in order to achieve required sensitivity and accuracy. Another important limitation is molecular information obtained is limited and data mining cannot be performed post-analysis. This is a serious limitation for public health and food safety investigation and research.
Mass spectrometry (MS) has become the workhorse in protein research for many areas in medical and biological sciences [Wood et al. 2014; Lisitsa et al., 2014]. It is now becoming an important asset in foodomics, a relatively new science [Herrero et al., 2012]. A major challenge in proteomics research is related to the analysis of highly complex biological samples. As sample complexity increases, proteomic research requires powerful analytical instruments with high sensitivity, selectivity and a large dynamic range. High-resolution mass spectrometry technology is an important asset to perform such complex assays. New instruments have the speed, resolution, mass accuracy and sensitivity to deliver comprehensive qualitative exploration, rapid profiling, and high-accuracy detection and quantification of proteins in complex matrices. Interestingly, this can now be performed using a single platform. This is, in our opinion, a significant advantage since it will ensure a simplified instrumentation approach for implementation in food analysis laboratories. Recently, a study was published using a shotgun bottom-up proteomic approach using peptide mass fingerprinting (PMF) and mass spectrometry to determine meat authenticity [Bargen et al. 2013; Sentandreu and Sentandreu, 2011]. There are limitations with the PMF approach proposed, including data analysis and pattern complexity that frequently reduce the assay sensitivity to accurately determine meat authenticity. The authors did not present a systematic approach that could be truly useful as an analytical method in a food
chemistry laboratory including regulated environments. Proteomic methods can be easily implemented for routine analysis performed in food chemistry laboratories but we believe it requires a systematic approach based on a proteogenomic mapping and annotation. In our opinion, meat species authenticity can only be determined if a clear and well defined proteogenomic annotation is previously performed and surrogate tryptic peptides are carefully chosen to develop and validate a targeted MS based method.
The global objective of this study was to develop a novel analytical strategy using a state of the art high-resolution Orbitrap mass spectrometer to determine meat authenticity and composition. The objectives of this study were ; (1) To perform in silico characterization of targeted muscular proteins (e.g. myoglobin) of selected mammalian species and identification of proteotypic peptides; (2) To detect and identify targeted proteotypic peptides by mass spectrometry using proteomic bottom-up approaches; and (3) to demonstrate the selectivity, sensitivity and applicability of the proposed analytical strategy for the assessments of meat authenticity.
Materials and Methods Chemicals and Reagents
Proteomic grade trypsin, dithiothreitol (DTT), iodoacetamide (IAA), ammonium bicarbonate (NH4HCO3), hydrochloric acid (HCl), formic acid, water (LC-MS Optima grade) and acetonitrile
(LC-MS Optima grade) were purchased from Fisher Scientific (NJ, USA).
Bioinformatic analyses
Protein alignment and sequence analyses were performed using QIAGEN CLC Sequence Viewer 7.5 (Redwood City, CA, USA). Protein sequences were obtained from the National Center for
fingerprinting and MS/MS fragment ion fragmentation were performed using mMass [Strohalm
et al. 2010]. Additional, thorough protein identification was performed using Thermo Scientific
Proteome Discoverer software (San Jose, CA, USA) along with Matrix Science MASCOT databases (Boston, MA, USA).
Protein extraction from meat products
Briefly, 1 g of raw meat was mixed with 5 mL of distilled water. The mixture was blended at high speed for 3 minutes to obtain a homogeneous suspension. The sample was sonicated for 30 minutes at room temperature, followed by centrifugation at 1500 g for 3 minutes to remove debris. Five hundred µL of the suspension was transferred to a microcentrifuge tube. Proteins were precipitated with 500 µL of acetone. Then acetone was discarded and protein pellet was dried for 20 minutes in a vacuum centrifuge set at 60˚C. The protein pellet was dissolved in 500 µL of 100 mM ammonium bicarbonate (pH 8.5) and the solution was sonicated for 60 minutes at maximum intensity to improve protein dissolution yield. The proteins were denatured by heating at 120˚C for 10 min using heated reaction block. The solution was allowed to cool down 15 minutes and proteins were reduced with 20mM DTT and the reaction was performed at 60 ˚C for 60 minutes. Then proteins were alkylated with 40mM IAA and the reaction was performed at room temperature for 30 min. Two µg of proteomic-grade trypsin was added and the reaction was performed at 40˚C for 24h. The protein digestion was quenched by adding 500 µL of a 1% TFA solution. Samples were centrifuged at 12000 g for 10 min and 200 µL of the supernatants were transferred into injection vials for analysis.
Raw meat mixtures for sensitivity and specificity assessments
Raw pork meat was mixed at several weight % ratios with a mixture of beef : horse : lamb raw meat to verify assay sensitivity and specificity. Precisely, raw pork meat was thoroughly mixed at weight % ratios of 1, 2, 10, 25, 50 and 100 with a mixture of equal weight (i.e. 1:1:1) of raw beef, horse and lamb meat using a laboratory blender for 5 minutes at low speed. All mixtures were performed to obtain a total weight of 100 g. Then, 1 g of raw meat samples was mixed with 5 mL of distilled water and blended at high speed for 3 minutes to obtain a homogeneous suspension. Protein extraction and digestion were performed as previously described above.
Chromatographic conditions
The HPLC system was a Thermo Scientific UltiMate 3000 Rapid Separation UHPLC system (San Jose, CA, USA). The chromatography was achieved using a gradient mobile phase along with a microbore column Thermo Biobasic C8 100 × 1 mm, with a particle size of 5 μm. The initial mobile phase condition consisted of acetonitrile and water (both fortified with 0.1% of formic acid) at a ratio of 5:95. From 0 to 1 minute, the ratio was maintained at 5:95. From 1 to 31 minute, a linear gradient was applied up to a ratio of 50:50 and maintained for 2 minutes. The mobile phase composition ratio was reverted at the initial conditions and the column was allowed to re-equilibrate for 14 minutes for a total run time of 47 minutes. The flow rate was fixed at 75 µL/min and 2 µL of sample were injected.
Mass Spectrometry conditions
A Thermo Scientific Q-Exactive Orbitrap Mass Spectrometer (San Jose, CA, USA) was interfaced with a Thermo Scientific UltiMate 3000 Rapid Separation UHPLC system using a pneumatic assisted heated electrospray ion source. MS detection was performed in positive ion mode and operating in scan mode at high-resolution, and accurate-mass (HRAM). Nitrogen was used for sheath and auxiliary gases and they were set at 10 and 5 arbitrary units. The heated ESI probe was set to 4000 V and the ion transfer tube temperature was set to 300°C. The scan range was set to m/z 500-2000. Data was acquired at a resolving power of 140,000 (FWHM), resulting to a scanning rate of 700 msec/scan when using automatic gain control target of 3.0x106 and
maximum ion injection time of 200 msec. Product ion spectra were acquired a resolving power of 17,500 (FWHM), using automatic gain control target of 1.0x106 and maximum ion injection time
of 100 msec. The collision energy set to 25 and the isolation window was set to 1.5 Da. Instrument calibration was performed prior all analysis and mass accuracy was notably below 1 ppm using Thermo Pierce calibration solution and automated instrument protocol.
Results and Discussion
Bioinformatic analysis
Peptide mass fingerprinting (PMF) is an analytical strategy used to identify and quantify proteins by assigning specific proteolytic fragment masses with in silico peptide masses generated using protein databases and specific digestion algorithms. The method implies that intact proteins are cleaved with a proteolytic enzyme (e.g. trypsin) to generate specific and predictable peptides. The PMF strategy requires database search following HPLC-MS analysis. This method relies on the premise that every unique protein will have a distinctive set of peptides, consequently unique
peptide masses can be found. Nowadays, the PMF workflow is complemented by database matching of MS/MS data for added selectivity and additional peptide sequence characterization. The identification is performed by matching the observed peptide masses with the in silico generated masses derived from protein databases. This process is now highly automated but still requires a skilled expert to increase proteome coverage and accuracy and could be tedious, particularly for routine analysis of food product. Therefore this is not practical for a method that requires extensive validation and added robustness for future implementation into several control laboratories with variable expertise in proteomics. The method proposed in this manuscript is based on a targeted PMF method that specifically relies on upstream identification of species specific proteolytic fragments that would allow meat speciation without having to perform an in-depth PMF analysis using post acquisition data sets. An exhaustive PMF analysis on meat products reveals proteotypic peptides from myoglobin, myosin and hemoglobin [von Bargen at
al., 2013]. Based on these preliminary results, we performed an exhaustive sequence analysis of
myoglobin, myosin and hemoglobin using bioinformatics for four common mammalian species. Figure 1 illustrates the result obtained for myoglobin, a very specific muscular protein. The analysis revealed a specific fragment (myoglobin tryptic peptide 120-134) that can be used for speciation in a targeted method since all four species presented amino acid sequence differences resulting into characteristic precursor masses and fragment ions. Other tryptic peptides were identified for myosin-1, myosin-2 and -hemoglobin and will be discussed subsequently. These results allowed the constitution of a precursor mass list and peptide MS/MS spectrum predictors can be used to identify precursor peptide-to-fragment ion transitions.
High-resolution mass spectrometry analysis
The exhaustive analysis of complex proteomic samples is an important challenge in proteomics research and mass spectrometry. Using PMF strategies, several hundreds or thousands of peptides are generated with a single sample, thus at any given time in a chromatographic run, several peptides are co-eluting. Thus the MS resolution and mass accuracy is a determining factor to improve the accuracy of peptides identification and assignment. This is very important in order to confirm that assigned peptides belong to actual proteins in a biological sample. The analysis of meat samples were performed using a hybrid Quadrupole-Orbitrap mass spectrometer operating in MS at a resolution of 140,000 (FWHM) and in MS/MS at a resolution of 17,500 (FWHM). The total ion chromatogram (TIC) of the meat protein identification experiments illustrated in Figure 2 demonstrates the great sample complexity. Many peptides were detected, identified and assigned using specific doubly and triply charged ions. Peptides from myosin, actin, tropomyosin, creatine kinase, hemoglobin and myoglobin were accurately identified resulting into protein coverage from 23% up to over 50%. However, the objective was to extract information related to our previously in silico identified proteotypic myoglobin peptides. As shown in Figure 2, XIC's of each targeted proteotypic myoglobin peptide was performed and abundant peaks were observed for each selected mammalian species. Detailed analysis of MS spectra revealed that mass accuracies ranged from -0.67 to 1.34 ppm for targeted proteotypic myoglobin peptides as indicated in Table 1. Additionally, the MS peak abundance for these specific peptides were in the top-tier relative to the abundance of each ion observed. This is particularly important since accuracy, sensitivity and robustness are critical in insuring that the method can be used for routine analysis. MS/MS spectra were collected for each peptide at high resolution. Tandem MS spectra of targeted proteotypic myoglobin peptides shown in Figure 3 are
dominated by y-type fragment ions with low abundance b ions, based on the Roepstorff and Fohlman nomenclature [Roepstorff and Fohlman, 1984]. The MS/MS spectra were coherent with the amino acid sequence of each peptide. Interestingly, y14 and y13 product ions were observed
and are specific for each targeted peptide sequence. These product ions can be used to generate very specific product ion XIC's. The acquisitions of MS/MS data were performed using parameters displayed in Table 2. Overlay XICs of chromatogram using y14 and y13 product ions
are shown in Figure 4 and results clearly demonstrate very high specificity to each targeted species. The combination of MS and MS/MS acquired in HR/AM allowed targeted qualitative screening and identification to perform meat authenticity using a specific proteotypic myoglobin peptide.
In silico protein sequence analysis permitted the identification of other proteopytic peptides from
myosin-1, myosin-2 and -hemoglobin based on comprehensive alignments and sequence analyses. Retrospective data analyses permitted the detection and identification of other proteotypic peptide as shown in Table 3. Moreover MS/MS spectra were coherent with the amino acid sequence for each of these targeted peptides. These peptides can be used for confirmatory and complementary measures, to provide maximum sensitivity and accuracy. The peak intensity observed for these proteotypic peptides were high and comparable to myoglobin proteotyptic peptides 120-134. Moreover, these peptides could be used as targeted peptide biomarker for meat speciation.
Method optimization and semi-quantitative analysis
Analytical methods based on PMF strategies heavily rely on the yield and reproducibility of the protein digestion method used. Commonly, extracted proteins are digested in-solution with
trypsin for a period of 24h to obtain maximum protein coverage. This is time consuming and may not be appropriate for rapid analysis of suspected meat samples. We evaluated the reaction kinetics using previously identified myoglobin proteotypic peptides for each species. As shown in Figure 5, we were able to detect the peptides even after only one hour of digestion. Obviously, peak abundances were significantly inferior, but adequate to carefully identify peptides of interest. Maximum peak intensities were observed after 8 h of digestion at 40˚C or 60˚C. A slightly faster reaction was observed at 60˚C resulting in a 2 fold increase of peak intensities after 2 or 4 h of digestion. This is interesting since suspected meat samples could potentially be analyzed within a normal working day. Time is an important element to be considered for food security analysis and risk assessment.
Another advantage of using a targeted method is the ability to perform semi-quantitative and quantitative analysis. Mixtures of pork meat with beef, horse and lamb meats were prepared and analyzed. Pork myoglobin proteotypic peptide precursor ions XIC's ( 5 ppm) were generated and as outlined in Figure 6, peak abundance was directly proportional with the nominal quantity in each tested mixture. Furthermore, a linear regression analysis using myoglobin proteotypic peptide peak areas (y) and nominal quantities (x) resulted in a coefficient of determination (r2) of
0.9676 showing a strong correlation and linearity between observed peak areas and nominal quantities. This result suggests that a quantitative method could be developed and validated using this proposed approach. Myoglobin proteotypic peptide product ions XIC's approach was also tested and provided similar results. Absolute quantitation of targeted myoglobin proteotypic peptides could be achieved using heavy isotope-labeled peptides. Absolute and relative quantitation were achieved by using MS to compare the abundance of a labeled “heavy” (i.e. known concentration) against the in situ “light” targeted peptide. This is a very standard approach
in targeted proteomics and can facilitate method validation, quality control and transferability. Moreover, this approach can be adapted and optimized for analysis using triple quadrupole mass spectrometer operating in Multiple Reaction Monitoring (MRM) mode.
Conclusion
In silico protein sequence alignment and analysis permitted us to discover 4 proteotypic peptides
(i.e. peptides from myoglobin, myosin-1, myosin-2 and -hemoglobin) which could be used to perform meat authenticity using a targeted proteomics approach. Detailed results presented using myoglobin proteotypic peptides (i.e amino acid position 120-134) suggested that accurate meat speciation could be achieved without using a fingerprinting approach as proposed in a recent study [Bargen et al., 2013]. The development of a targeted proteomic strategy based on a systematic proteogenomic annotation is key in our opinion to foster the basis of a robust method applicable for routine analysis of meat sample by MS.
Acknowledgments
This project was funded by the National Sciences and Engineering Research Council of Canada (F.Beaudry discovery grant no. 386637-2010). A. Ruiz received a scholarship from Agència de Gestió d'Ajuts Universitaris i de Recerca (AGAUR) - Generalitat de Catalunya.
References
Attien P, Sina H, Moussaoui W, Zimmermann-Meisse G, Dadié T, Keller D, Riegel P, Edoh V, Kotchoni SO, Djè M, Prévost G, Baba-Moussa L. Mass spectrometry and multiplex antigen assays to assess microbial quality and toxin production of Staphylococcus aureus strains isolated from clinical and food samples. Biomed Res Int. 2014;2014:485620.
Bao KD, Letellier A, Beaudry F. Analysis of Staphylococcus enterotoxin B using differential isotopic tags and liquid chromatography quadrupole ion trap mass spectrometry. Biomed Chromatogr. 2012;26(9):1049-1057.
Bargen, von C, Dojahn J, Waidelich D, Humpf H U, Brockmeyer, J. Journal of Agricultural and Food Chemistry, 2013;61(49), 11986–11994
Everstine K, Spink J, Kennedy S. J Food Prot. 2013;76(4):723-735.
Ellis DI, Brewster VL, Dunn WB, Allwood JW, Golovanov AP, Goodacre R. Chem Soc Rev. 2012;41(17):5706-5727
DG Health and Consumers, European Commission. Horse Meat Issue; DG Health and
Consumers, European Commission: Brussels, Belgium, 2013;
http://ec.europa.eu/food/food/horsemeat/tests_ results_en.htm (accessed December 22, 2014). Gobert M, Sayd T, Gatellier P, Santé-Lhoutellier V. Application to proteomics to understand and modify meat quality. Meat Sci. 2014;98(3):539-543
Herrero M, Simó C, García-Cañas V, Ibáñez E, Cifuentes A. Mass Spectrom Rev. 2012 ; 31(1):49-69
Hsieh YH, Ofori JA. Detection of horse meat contamination in raw and heat-processed meat products. J Agric Food Chem. 2014;62(52):12536-12544.
Kalb SR, Krilich JC, Dykes JK, Lúquez C, Maslanka SE, Barr JR. Detection of Botulinum Toxins A, B, E, and F in Foods by Endopep-MS. J Agric Food Chem. 2015; 63(4):1133–1141 Kumar A, Kumar RR, Sharma BD, Gokulakrishnan P, Mendiratta SK, Sharma D. Identification of species origin of meat and meat products on the DNA basis: a review. Crit Rev Food Sci Nutr. 2015;55(10):1340-1351
Lisitsa A, Moshkovskii S, Chernobrovkin A, Ponomarenko E, Archakov A. Expert Rev Proteomics. 2014;11(1):121-129.
Roepstorff, P., Fohlman, J., 1984. Proposal for a common nomenclature for sequence ions in mass spectra of peptides. Biomed. Mass Spectrom. 11 (11), 601.
Rogniaux H, Pavlovic M, Lupi R, Lollier V, Joint M, Mameri H, Denery S, Larré C. Allergen relative abundance in several wheat varieties as revealed via a targeted quantitative approach using mass spectrometry. Proteomics. 2015 Mar 12. doi: 10.1002/pmic.201400416
Sentandreu MA, Sentandreu E. Peptide biomarkers as a way to determine meat authenticity. Meat Sci. 2011;89(3):280-285.
Strohalm M, Kavan D, Novák P, Volný M, Havlíček V, mMass 3: A Cross-Platform Software Environment for Precise Analysis of Mass Spectrometric Data. 2010; Anal Chem 82 (11), 4648-4651
Woods AG, Sokolowska I, Ngounou Wetie AG, Wormwood K, Aslebagh R, Patel S, Darie CC. Adv Exp Med Biol. 2014;806:1-32.
Zheng A, Luo J, Meng K, Li J, Zhang S, Li K, Liu G1, Cai H, Bryden WL, Yao B. Proteome changes underpin improved meat quality and yield of chickens (Gallus gallus) fed the probiotic Enterococcus faecium. BMC Genomics. 2014;15:1167
Table 1. Specific myoglobin proteotypic peptides for selected mammalian meat species
Species Tryptic peptide sequence
MB (120-134) Theoretical mass (z=2) Observed mass(z=2) Mass accuracy(ppm)
Beef HPSDFGADAQAAMSK 766.8435 766.8436 0.13
Horse HPGDFGADAQGAMTK 751.8383 751.8378 -0.67
Pork HPGDFGADAQGAMSK 744.8304 744.8314 1.34
Table 2. MS/MS parameters used for the acquisition of myoglobin proteotypic peptides (MB 120-134)
Species Targeted peptide Precursor ion
m/z (z1=2) width (Da)Isolation Collisionenergy Product ionm/z (z1=1)
Beef HPSDFGADAQAAMSK 766.8 1.5 25 1298.5681 (y13) 1395.6209 (y14) Horse HPGDFGADAQGAMTK 751.8 1.5 25 1268.5576 (y13) 1365.6103 (y14) Pork HPGDFGADAQGAMSK 744.8 1.5 25 1254.5419 (y13) 1351.5947 (y14) Lamb HPSDFGADAQGAMSK 759.8 1.5 25 1284.5525 (y13) 1381.6053 (y14) 1 z refer to the peptide ion charge state
Table 3. Other specific proteotypic peptides identified for selected mammalian meat species
Species Protein Uniprot
Accession Peptide sequence AA position Theoretical mass (z=2) Observed mass (z=2) Mass accuracy (ppm) Rt (min)
Beef Myosin-1 Q9BE40 TLALLFSGPASG
EAEGGPK
619-637 901.4702 901.4694 -0.89 16.8
Horse Myosin-1 Q8MJV0 TLALLFSGPASA
DAEAGGK
619-637 888.4623 888.4620 -0.34 17.0
Pork Myosin-1 Q9TV61 TLAFLFTGAAGA
DAEAGGGK 619-638 912.9600 912.9594 -0.66 17.4 Lamb Myosin-1 XM_004012 706.1 (RefSeq) TLAFLFSGAASA EAEGGGAK 619-638 927.9652 927.9650 -0.21 17.6
Beef Myosin-2 Q9BE41 TLAFLFSGTPTG
DSEASGGTK
619-639 1022.4971 1022.4968 -0.29 16.4
Horse Myosin-2 Q8MJV1 TLALLFSGAQTA
DAEAGGVK
617-636 960.5073 960.5070 -0.31 17.0
Pork Myosin-2 Q9TV63 TLAFLFSGAQ
TGEAEAGGTK 619-638 978.4891 978.4894 -0.31 17.1 Lamb Myosin-2 XM_004012 707.1 (RefSeq) TLALLFSGTPTAE SEGSGTK 617-636 984.0020 984.0022 0.20 16.5
Beef β- Hemoglobin P02070 FFESFGDLSTAD
AVMNNPK
40-58 1045.4804 1045.4796 -0.77 16.9
Horse β- Hemoglobin P02062 FFDSFGDLSNPG
AVMGNPK
42-60 1000.4646 1000.4637 -0.90 17.2
Pork β- Hemoglobin P02067 FFESFGDLSNAD
AVMGNPK
42-60 1023.4673 1023.4670 -0.29 16.8
Lamb β- Hemoglobin P02075 FFEHFGDLSNAD
AVMNNPK
Figure legends
Figure 1. Bioinformatic analysis of targeted mammalian muscular proteins. Species specific myoglobin (MG) sequences were aligned and thoroughly analyzed. Proteotypic peptides were identified located between amino acid position 120 and 134. Thus a specific precursor ion mass list can be generated and a MS/MS experiments can be performed on species-specific biomarkers.
Figure 2. Total Ion Chromatograms (TIC) and specific Extracted-ion chromatograms (XIC) for each mammalian meat samples are presented. XIC's were generated using the theoretical mass value with a 5 ppm extraction window. TIC's show the high complexity of the samples. The XIC's produced from selected mammalian meat samples suggest high abundance for the myoglobin proteotypic peptide identified following an in-depth in silico investigation.
Figure 3. Product ion spectra of myoglobin proteotypic peptides (120-134). Fragment ion selectivity is preserved for y14 and y13 ions. These products can be used to produce specific
product ion XIC's for meat speciation.
Figure 4. Comparison of y14 and y13 product ion XIC's. Both fragment ions are very specific to
each tested species allowing accurate meat speciation based on MS and MS/MS XIC's.
Figure 5. Kinetic characterization of trypsin catalyzed formation of targeted myoglobin proteotypic peptide (120-134). Reactions were performed at 40˚C and 60˚C for up to 24h. The reaction is almost completed after 8h and a roughly two-fold abundances increase is observed at early time points when the reactions are performed at 60˚C.
Figure 6. Overlay precursor XIC's of targeted pork myoglobin proteotypic peptide (120-134) at m/z 744.8304 5 ppm was achieved using various quantity of pork meat mixed with beef, horse and lamb meat. Precisely, pork meat was mixed at weight % ratios of 1, 2, 10, 25, 50 and 100 with a mixture of beef : horse : lamb meat (weight ratio of 1:1:1). All resulting meat mixture were digested and analyzed in triplicate. Peak abundances were directly proportional with the quantity of pork meat in the mixture.
