The world leader in serving science
Michal Godula Ph.D. Thermo Fisher Scientific
Meat Authenticity Testing using Thermo Scientific Orbitrap HRAM MS Technology
2
Meat Substitution
A real scandal !!
3
How testing is done?
• Two-dimensional polyacrylamide gel
electrophoresis and western-blot analysis
• Qualitative Real-Time PCR
• Enzyme-linked immunosorbent assay
(ELISA)
Challenges
• These methods are mostly qualitative
• Molecular information obtained is limited
• Data can’t be revisited post-acquisition
for data mining
• They are not generic approaches and
need to be heavily customized
Motivation : $$$
• Addition of meat from undeclared species
to a specific meat product in order to
lower production cost and increase
profitability
Cost per kg: Horse meat << Beef meat
An international issue
• It is economic fraud
• It represents health issues due to specific
dietary restrictions
• It is an ethical problem
• It is also an important cultural and
religious issues
Meat Substitution
4
Unit mass resolution MS • Is targeted (triple quadrupole)
• Selectivity provided by tandem MS/MS (SRM transition needed)
• False positives are reality
• Need to setup instrument (SRM) before analysis
• Realistic breakpoint is 200-300 compounds in a run
• Time consuming data processing
Can perform the same level of quantitation as MS/MS
Selectivity obtained by accurate mass measurement (only m/z needed)
No false positives!
No need to setup instrument (SRM) before analysis
Unlimited number of compounds in a run – perfect for screening
Automated data processing
High Resolution Accurate Mass
What About Mass Spectrometry Options
5
How Accurate Is Your Mass?
• Mass accuracy usually expressed as ppm or mDa error
610/
true
truemeas
m
mmzm
• Quadrupole MS
• Orbitrap MS
ppmzm 20010500
0.5001.500/ 6
ppmzm 21010314.500
10314.50010414.500/ 6
100 mDa error
1 mDa error
6
•Resolution
Mass Resolution FWHM
m
mR
m (FWHM)
• Quadrupole MS • Orbitrap (HR)MS
10004.0
400R 100000
004.0
400R
7
Resolution vs. m/z
0
50000
100000
150000
200000
250000
300000
350000
100 200 300 400 500 600 700 800 900 1000
Resolu
tion (
FW
HM
)
Q Exactive HF Q Exactive
Q-TOF
m/z
Much higher resolution
at low m/z range
8
Thermo ScientificTM MS Portfolio
HR/AM
MS, MSn
Ap
plie
d
Mark
ets
Researc
h
Mark
ets
Non-targeted Analysis
Targeted Analysis
Quantitative Qualitative
• Biomarker Discovery
• Proteomics
• Metabolism
• Metabolomics
• Proteomics
• Bioanalysis
• Food Safety
• Environmental
• Clinical/Toxicology
• Metabolomics
• PTM Analysis
• Lipidomics
Transform Your Science
Ion Traps Triple Quads
Tribrid Orbitrap
MS
(Q) Exactive Series MS
9
Principle Of OrbitrapTM Operation
12
2
R
Rmz 2
2
R
Rmzr
zm
kz
/
Makarov A. Anal. Chem. 2000, 72, 1156-1162.
Characteristic frequencies: • Frequency of rotation ωφ
• Frequency of radial oscillations ωr
• Frequency of axial oscillations ωz
Inte
nsity
m/q
10
• Fundamental difference to other HRAM instruments
• Parameter measured is frequency, not time/voltage/current
• Resolution allows more accurate m/z determination
• Less prone to ambient conditions changes
• Usually stable within <2 ppm during several days
• No need for lock mass in “routine work”
• Small footprint
• Easy to setup
Orbitrap Mass Analyzer Features
11
Q ExactiveTM MS - a 3D View
Quadrupole Mass Filter
HCD
Cell
Orbitrap Mass
Analyzer
C-Trap
RF-Lens
12
Why Bottom-up Proteomics Workflow Is An Interesting Option To Develop An MS Based Assay?
All life forms are related by common
ancestry and descent. The construction
of phylogenies provides explanations of
the diversity seen in the natural world.
Today, phylogenies are usually
constructed using DNA sequence data.
Relationship between genes and
species is central for meat speciation
13
Traditional Peptide Fingerprinting Approach Using MS
0 5 10 15 20 25 30 35 40 45
Time (min)
Porcine TICSus scrofa
Generate a mass list (MS and MS/MS
data)
Database search (MASCOT)
MS and MS/MS matching
Peptide fingerprinting
7 429 common
unique peptides
identified
Beef Horse
Lamb Pork
17 specific peptides
9 specific peptides 14 specific peptides
23 specific peptides
Using a relatively high
abundance threshold
…
14
PMF
Bargen et al. (2013), Journal of Agricultural and Food Chemistry, 61:11986-
11994
Peptide Mass Fingerprinting
15
• It relies heavily on the quality of the MS and MS/MS data
• It strongly relies on bioinformatics and parameterizations
• It requires highly skilled scientists to obtain comprehensive
results
Is this really appropriate for implementation in a
routine food analysis laboratory ?
Can we propose alternative strategies ?
What Are The Main Limitations Of This Analytical Approach?
16
Targeted Bioinformatics Analysis Example: Myoglobin
Extraction of gene
information
In silico sequence
alignment
In silico protein
translation and
alignment
Sequence analysis
and In silico tryptic
digestion
Generate
proteotypic mass
lists
Proteotypic peptides can be
identified
(120-134)
YLEFISDA IIHVLHAKHP SDFGADAQAA MSKALELFR
YLEFISDA IIHVLHSKHP GDFGADAQGA MTKALELFR
YLEFISEA IIQVLQSKHP GDFGADAQGA MSKALELFR
YLEFISDA IIHVLHAKHP SDFGADAQGA MSKALELFR
120 134
Beef
Horse
Pork
Lamb
Myoglobin is the primary oxygen-carrying protein of muscle tissues
It is a highly abundant protein
17
How Bottom-up Proteomics Can Be Used For Meat Speciation
Phase 1
Proteome mapping
Targeted
Meat products
Protein Extraction
Proteolysis
Peptide enrichment
MS and MS/MS
Protein / peptide
identification
Genome annotation
18
Bottom-up Proteomics Sample Preparation
1. Meat sample mixed with water (1:5) is homogenized and the mixture is sonicated
2. Proteins in the suspension are precipitated with acetone (1:1)
3. Acetone is discarded and the generated protein pellet is dryed to remove all traces of
acetone.
4. Protein pellets are dissolved in ammonium bicarbonate (pH 8.5).
5. Proteins are denatured by heating at 120˚C
6. Reduced with Dithiothreitol (DTT) and alkylated with Iodoacetamide IAA
7. Proteomic grade trypsin is added and the reaction is performed at 40ºC for 24h.
Trypsin cleavage occurs after basic amino acids : Lys (K) & Arg (R)
+
H2O
I. II. III. IV.
19
Targeted Analysis Of Myoglobin Proteotypic Peptides
Strategy
Survey MS Scan
Select ions
of interest
Acquired MS/MS
spectra
Generation of post-analysis XICs for
selected proteotyping peptides
Beef Horse Pork Lamb
766.8435
(z=2)
751.8383
(z=2)
744.8304
(z=2)
759.8357
(z=2)
140 000 FWHM
m/z 500-2000
17 500 FWHM
Isolation width = 1.5 AMU
Extracted ion chromatograms with mn ± 5 ppm
20
749.0 749.5 750.0 750.5 751.0 751.5 752.0 752.5 753.0 753.5 754.0 754.5 755.0
m/z
751.8378
752.3393
752.8404
753.3409
∆m is - 0.7 ppm
Very complex TIC’s
Over 7000s of unique
peptides can be found.
Each Targeted Peptides Can Be Detected And Extracted From Tics
Need for very high
resolution data (>100 000).
21
Identification Of Specific Product Ions – MS/MS
22
Survey MS Scan
Select ions
of interest
Acquired MS/MS
spectra
Generation of post-analysis XICs for
selected proteotyping peptides
Beef Horse Pork Lamb
766.8435
(z=2)
751.8383
(z=2)
744.8304
(z=2)
759.8357
(z=2)
140 000 FWHM
m/z 500-2000
17 500 FWHM
Isolation width = 1.5 AMU Beef Horse Pork Lamb
1298.5681
(y13)
1268.5576
(y13)
1254.5419
(y13)
1284.5525
(y13)
1395.6209
(y14)
1365.6103
(y14)
1351.5947
(y14)
1381.6053
(y14)
Extracted ion chromatogram at y14 or y13 ± 5 ppm
Data Dependent MS/MS Used For Targeted Work
23
766.8 1298.5681 (y13) 766.8 1395.6209 (y14)
751.8 1268.5576 (y13) 751.8 1365.6103 (y14)
744.8 1351.5947 (y14)
759.8 1381.6053 (y14)
Beef
Horse
Pork
Lamb
Very Specific Xics Allow Species Differentiation
759.8 1284.5525 (y13)
744.8 1254.5419 (y13)
24
Q Exactive Analysis At 140 000 FWHM Resolution.
0 5 10 15 20 25 30 35 40 45
Time (min)
100%
50%
25%
10%
1%
Peak areas observed were
proportional with the relative
concentration
0 20 40 60 80 1000
1000000
2000000
3000000
% W/W
Pe
ak A
rea
R2 = 0.9676
Pork meat was mixed at weight % ratios of 1, 2, 10, 25, 50 and 100 with a mixture of 1:1:1
beef:horse:lamb meat
Digestion was performed in triplicate.
25
Sam
ple
1
Sam
ple
2
Sam
ple
3
Sam
ple
40.0
5.0×106
1.0×107
1.5×107
2.0×107
2.5×107
Pe
ak A
rea
Assessment Of The Apparent Intra- And Inter Method Reproducibility
Sam
ple
1
Sam
ple
2
Sam
ple
3
Sam
ple
40.0
5.0×107
1.0×108
1.5×108
2.0×108
2.5×108
3.0×108
3.5×108
Pe
ak A
rea
Beef (Myoglobin peptide 120-134)
% CV < 20%
Sam
ple
1
Sam
ple
2
Sam
ple
3
Sam
ple
40.0
5.0×107
1.0×108
1.5×108
2.0×108
Pe
ak A
rea
Horse (Myoglobin peptide 120-134)
% CV < 16%
Pork (Myoglobin peptide 120-134)
% CV < 22%
Sam
ple
1
Sam
ple
2
Sam
ple
3
Sam
ple
40
1×107
2×107
3×107
4×107
Pe
ak A
rea
Lamb (Myoglobin peptide 120-134)
% CV < 19%
26
Tryptic Digestion Optimization
After 1h, peptides can be detected. At 4h, we observed 30-40% of the maximum
abundance.
27
Chromatograms From Beef Meat Sample Spike With 1 % Pork Meat
MS1
XIC 744.8304 ± 5 ppm
7 8 9 10
Time (min)
Rela
tive a
bundance (
%)
Pure beef are in red
Pure beef are in blue Pure beef are in blue
28
Protein
Proteotypic peptides
MS/MS spectra
Fractionation
Digestion
LC-MS
Homogenization/Lysis
Targeted MS
We can use labeled peptides
as internal standards !
1. Select precursor ion
MS
2. Precursor fragmentation
MS/MS
3. Use Precursor-Fragment
pairs for identification
Quality
Control ! Introduction of internal standards
( 13C, 15N)
This is a must for methods to
be used for routine analysis in
multiple sites
Routine Practive
29
Conclusions
Targeted bottom-up proteomics approach was applied for meat species
detection down to 0.1% w/w
Quick and simple workflow for any laboratory
High resolving power (140.000 FWHM) was needed to obtain sufficient
selectivity (not suitable for QQQ)
Isotopically labelled peptides recommended for routine control
Tested in routine – applying HPLC- Q Exactive
30
Acknowledgments
Special Thank you to dr. Francis Beaudry and dr. Alberto Ruiz from the
Université de Montréal, Canada for providing the data
DOI: 10.1080/19440049.2015.1064173
31
Webinar: Thermo Fisher Scientific Meat Adulteration Resource