ABRF 2006 ESRG Study

1
ABRF 2006 ESRG Study
Edman Sequencing as a Method For Polypeptide
Quantitation
2
ESRG Committee Members
Daniel C. Brune (Chair) - Arizona State
Univ. Brian Hampton - Univ. of Maryland,
Baltimore, School of
Medicine Ryuji Kobayashi - Univ. of Texas, M.D.
Anderson Cancer Center Joseph
W. Leone (Chair elect) - Pfizer Klaus D. Linse -
Univ. of Texas, Austin Jan Pohl - Emory
Univ. Richard S. Thoma - Monsanto Co. Nancy D.
Denslow (ABRF liaison) - Univ. of Florida
3
Objectives of the Study

• - Determine how accurate Edman sequencing
• is for the quantitative analysis of
• polypeptides
• - Compare quantitative results obtained by
• Edman sequencing to those obtained using
• mass spectrometry techniques
• - Test the ability of participating laboratories
• to identify a modified amino acid residue

4
Description of the Test Sample
Mixture of 3 Peptides (100 pmol/vial)
lyophilized Peptide C 40 pmol RQAKVLLYSGR Pepti
de C 40 pmol RQA(Ac-Lys)VLLYSGR Peptide B 20
pmol KAQYARSVLLEKDAEPD ILELATGYR - both
unmodified peptides were supplied by the ABRF
Peptide Standards Committee
5
Sample Preparation
1. Peptide C synthesis and purification a.
Synthesis on a Milligen 9050 peptide
synthesizer - FMOC chemistry b. Purified by HPLC
- Jupiter Proteo C12 column
(Phenomenex) c. Mass was verified using
MALDI-TOF and Edman Sequencing d.
Concentration determined by AAA 2. Peptides B
C each solubilized to 1mg/5mL in 10mM TFA/30
Acetonitrile
6
Sample Preparation (cont.)
3. Three peptides were mixed a. 15.25?L C, 59mL
B, 51.6?L C b. 1874?L 10mM TFA/30
Acetonitrile 4. 10?L aliquots placed in 0.6mL
Eppendorf tubes (100 pmol/tube) 5. Dried in
a Speed-Vac 6. Distributed 2 tubes per lab
7
Requested Information
- The amino acid sequence of each peptide - Areas
for peaks in each cycle - Areas, picomolar yields
and retention times for each amino acid peak
in the standard - Instrument information a.
sequencer e. Gradient b. sample loading f.
Solvents c. HPLC equipment g. Flow rate d.
Column - 34 facilities requested the sample - 18
facilities returned sequencing data
8
Sequencer Information
Manufacturer 8 ABI 49X-Ht and Model (5-14
yrs, Ave 8.6) 10 ABI 49X-cLC (0-11 yrs,
Ave. 7.0) Reagents 15 all Manufacturers
reagents 2 used some (R1, R2C, R4, R5
and Premix) 1 S4 was homemade TFA
Cleavage 15 liquid, 3 gas-phase Sample
Support 15 GFF, 3 PVDF DTT in S2 8 yes, 10
no Other additives 1 TCEP to R5
9
Labs Participating in ESRG Studies
Number of Participating Labs
Year
10
Accuracy of Identification
Short Peptide (C C)
K Ac-K
R
R
G
S
Y
L
L
V
A
Q
Correct Calls
Incorrect Calls
Accuracy
No Calls
Y
T
L
L
D
E
D
E
L
S
A
Q
R
G
A
E
I
P
A
K
L
V
R
Y
A
K
Long Peptide (B)
Cycle
11
Initial Yield Calculation
1. Log10 (AA pmol yield) Short
Peptide 3(A),5(V),6(L),7(L),8(Y) Long
Peptide 2(A),4(Y),5(A),8(V),9(L),10(L)
2. Plot data and determine best fit line
Long Peptide
Short Peptide
y - 0.0338x 1.57 R2 0.9284
y - 0.0334x 0.946 R2 0.834
3. I.Y. (antilog of y-intercept)/( of sample
tested) R.Y. (antilog of slope)
12
Initial Yield Performance
Short Peptide (C C)
80 pmol
Ave. 53.6pmol 67.0
Initial Yield (pmol)
Long Peptide (B)
20 pmol
Ave. 13.1pmol 65.6
Lab ID
13
Initial Yield Ratio Short (C C) vs. Long (B)
Peptide
4/1 ratio
Ave. 4.27/1
Ratio
Lab ID
20 of 23 labs were between 3/1 and 5/1 ratios
14
Reasons for Low Initial Yield
- Sample Washout - N-termini blocked a.
interaction with the membrane b. chemical
modification - Quality of reagents a. Biobrene
(age or cycles performed) b. EtOAc (gt 6
months) - Poor solubility during reconstitution
- Specific Low Yielding Amino Acids (e.g. Cys,
Trp, Ser, Arg) - Non-linear Repetitive Yield -
Small amount of sample analyzed
15
Initial Yield - Short Peptide (C C) Slope
Calculations Sequence - RQAKVLLYSGR
200
220
110
300
100
500
550
400
440
330
Log 10 (pmol)
770
800
600
660
700
1
2
990
880
900
3
4
5
ABRF
ESRG2006
Cycle
16
Initial Yield - Long Peptide (B) Slope
Calculations Sequence - KAQYARSVLLEKDAEPDILELATGYR
110
100
220
200
300
550
330
400
440
500
Log 10 (pmol)
800
770
660
600
700
2
1
990
900
880
4
5
3
ABRF
ESRG2006
Cycle
17
Initial Yield vs Linearity
Short Peptide (C C)
Initial Yield (pmol)
Long Peptide (B)
Linearity (R2)
18
Linearity (R2 gt 0.9) Short Peptide - 5 labs
(22) Long Peptide - 10 labs (43) Accuracy (?
50 from expected) Short Peptide (40-120 pmol) -
16 labs (70) Long Peptide (10-30 pmol) - 19 labs
(83)
19
Initial Yield vs Amount Analyzed
Short Peptide (C C)
Initial Yield (pmol)
Long Peptide (B)
Percent of Test Sample Sequenced
20
Initial Yield Edman Sequencing vs Mass
Spectrometry
Mass (m/z)
21
Mass Spec Ratio Short (C C) vs. Long (B)
Peptide
Ave. 22.4/1 Std Dev. 30.6
4/1 ratio
Ratio
Lab ID
3 of 16 labs were between 3/1 and
5/1 (Not enough data to separate MS
instrumentation)
Instrumentation
MALDI 16 ESI or Ion Trap 5 Not reported 3
22
The Modified Amino Acid Acetyl-Lysine
DPTU
Cycle 4
Ac-K
K
Y
PTH Standard
D
N
Q
E
G
S
T
H
A
K
Y
V
W
M
R
P
DPTU
F
L
I
13 of 18 labs correctly identified the modified
peptide
23
Area Ratio Short (C vs C) Peptide
Ave. 1.5/1
1/1 ratio
Ratio
Lab ID
16 of 21 labs were between 0.5/1 and 1.5/1
24
Mass Spec Ratio Short (C vs C) Peptide
Ave. 2/1
1/1 ratio
Ratio
Lab ID
8 of 15 labs were between 0.5/1 and 1.5/1
25
Conclusions

• - Edman sequencing tends to underestimate
• initial yields by approximately 1/3. Most
• lab data was within a factor of 2 from the
• expected accuracy.
• Edman Sequencing produced very accurate
• peptide to peptide ratios. Mass
• Spectrometry produced far less accurate
• ratio data.
• The majority of labs were able to identify
• the modified amino acid Acetylated lysine

26
Acknowledgements
Thanks to all the participating laboratories for
taking the time to analyze the test sample and
sending in their results. Without their
participation, this effort would not have been
successful. Thanks to the Henriette Remmer from
the ABRF Peptide Standards committee for
supplying two of the synthetic peptides. Thanks
also to Renee Schrauben for removing identifiers
from the responding laboratories.
27
Congratulations to Joe Leone On his election
as ESRG Chair!
28
Suggested Future ESRG Studies
1. Deblocking a Protein 2. Sequencing a Very
Large Protein on a PVDF Membrane 3.
Internal Cleavage Analysis 4. Repetitive Yield
Study 5. Protein Mixture or an Antibody 6. Very
Hydrophobic Peptides 7. More Tips on Unusual
Amino Acids 8. Phosphorylated and/or
Glycosylated Proteins 9. Alkylated Cysteine 10.
Evaluation of PVDF Membranes and Transfer
Techniques 11. Low Protein Concentrations (lt 2
pmol) 12. Quantitation of Post-Translational
Modifications