Automated Immunoaffinity-based Proteomic...

Automated Immunoaffinity-based Proteomic Methods for the Study of Post-translational ModificationMatthew P. Stokes1, Jeffrey C. Silva1, Steven Murphy2, Jason Russell2, Xiaoying Jia1, Jian Min Ren1, Kimberly Lee1

1: Cell Signaling Technology, Inc., Danvers MA 01923 2: Agilent Technologies, Inc., Santa Clara, CA 95051

Post-translational modification (PTM) of proteins, including phosphorylation, acetylation, methylation, and ubiquitina-tion, are critical events in all aspects of cellular signaling. Antibody-based enrichments of post-translationally modified peptides combined with LC-MS/MS have proven to be powerful methods for the study of PTMs in a wide variety of cells and tissues, and in profiling various disease states (1-4). These antibody-based methods involve complex pro-tocols that necessitate great care to achieve optimal results and reproducibility. Here, manual (batch mode) versus automated protocols have been compared with respect to the number of post-translationally modified peptides iden-tified and the corresponding relative abundance of those peptides between the two sample preparation procedures. The AssayMAP Bravo Platform (Agilent Technologies, Inc.) allows automation of antibody-based peptide enrichments, simplifying the enrichment protocol and providing results that can be superior to manual methods.

The AssayMAP Bravo Platform outperformed a traditional batch mode method for immunoaffinity purification of post-translationally modified peptides with all antibodies tested. In each case, the Bravo method resulted in a higher number of modified pep-tides identified along with a lower number of non-specific, unmodified peptides. The improved performance was likely due to the decrease in unmodified peptides in the Bravo samples, as relative abundance of the modified, target peptides changed little between methods (Figures 4F, 5E, 6C), and nearly all modified peptides were present in both methods at the MS1 feature level (Figure 5D).

Matthew P. Stokes Cell Signaling Technology3 Trask Lane, Danvers, MA 01923email: mstokes@cellsignal.comwww.cellsignal.com

Introduction

Conclusions

Contact Information

Phospho-Serine/Threonine Ubiquitin Branch Acetyl-lysinePhospho-tyrosine

© 2014 Cell Signaling Technology. Inc. Cell Signaling Technology®, CST™, PTMScan® and XP® are trademarks of Cell Signaling Technology, Inc. Orbitrap Velos™ is a trademark of Thermo Fisher Scientific Inc. Progenesis® is a Registered trademark of Nonlinear Dynamics Limited. SORCERER™ is a trademark of Sage-N Research.

14PSTSHOWPTMS0165ENG_00

Figure 5: Ubiquitin Branch Motif Antibody. Number of ubiquitinated (Green, A) and unmodified (Grey, B) peptides identified in extract from mouse embryo using the Bravo method and the stan-dard batch mode method. Venn diagrams of overlap between Bravo method and standard method at the MS2 (identification, C) and MS1 (feature, D) levels. E. Log2 ratio versus intensity plot comparing Bravo method to batch mode method. Green = ubiquitinated peptides, Grey = unmodified peptides.

Figure 6: Acetyl-lysine Antibody. Number of acetylated (Red, A) and unmodified (Grey, B) pep-tides identified using the Bravo method or the standard batch mode method. C. Log2 ratio versus intensity plot comparing Bravo method to batch mode method. Red = acetylated peptides, Grey = unmodified peptides.

Figure 4: Phospho-tyrosine (p-Tyr-1000) Antibody. A. Gel Stain and western blot analysis on extract from Jurkat cells treated with pervanadate (PV: high p-Tyr signal) and mouse liver extract (low p-Tyr signal). Western blot performed using phospho-tyrosine (P-Tyr-1000) Rabbit mAb #8954. Num-ber of tyrosine phosphorylated (Blue) and unmodified (Grey) peptides identified with the AssayMAP Bravo Platform or the standard batch mode method in Jurkat cells treated with pervanadate (B & C) or mouse liver (D & E). F. Log2 ratio versus intensity plot comparing Bravo method to batch mode method. Blue = phosphopeptides, Grey = unmodified peptides. The median log2 ratio for each is indicated.

Figure 3: Ser/Thr Motif Antibody Mixture. A. List of antibodies included in the Ser/Thr motif antibody mixture with consensus phosphorylation motifs. B. Venn diagram of peptide identifications using the motif antibody mixture and IMAC. Number of phosphorylated (Orange, C) and unmodified (Grey, D) peptide identifications from mouse embryo for the Ser/Thr Motif Antibody mixture using the AssayMAP Bravo Platform or the standard batch mode method.

Figure 2: The AssayMAP Bravo System. Antibodies were bound to Protein A cartridges and washed. Peptides were then bound to antibody/Protein A beads, washed, eluted, C18 purified, and analyzed using LC-MS/MS.

Figure 1: The Batch Mode PTMScan® Method. Samples are digested to peptides, run over reverse phase columns, immunoaffinity purified with the appropriate motif antibody, C18 purified, and subjected to LC-MS/MS.

1. Bind Antibody to Protein A

2. Bind Peptides to Protein A/Antibody

Antibody Description Motif

Akt Substrate RXX(s/t)

Akt Substrate RXRXX(s/t)

AMPK Substrate LXRXX(s/t)

ATM/ATR Substrate (s/t)Q

ATM/ATR Substrate (s/t)QG

Cdk Substrate (K/R)(s/t)PX(K/R)

CK Substrate (s/t)(D/E)X(D/E)

MAPK Substrate PX(s/t)P

PKA Substrate (K/R)(K/R)X(s/t)

PKC Substrate (K/R)X(s/t)X(K/R)

PKD Substrate LXRXX(s/t)

PLK Binding Motif S(s/t)P

tP Motif (s/t)P

tPE Motif (s/t)PE

tXR Motif (s/t)XR

14-3-3 Binding Motif (R/K)XX(s/t)XP

Human cell lines or mouse tissues were lysed, digested with trypsin, and desalted over C18 columns. Peptides from 2 mg of samples were processed using the standard batch-mode PTMScan® protocol (Figure 1) or the Assay-MAP Bravo system (Figure 2) using Protein A cartridges and the antibody purification application. Motif antibodies (100 μg) were loaded onto the cartridges and washed with PBS. Peptides resuspended in immunoaffinity purification (IAP) buffer were loaded onto Protein A/Antibody cartridges, washed using IAP buffer and water, and eluted in 0.15% TFA. Enriched peptides were purified on StageTips and analyzed by LC-MS/MS on an Orbitrap Velos™ mass spec-trometer using a top 20 data-dependent analysis method. MS/MS spectra were assigned to peptide sequences using SORCERER™ (5), and label-free quantification was performed using Progenesis® (Nonlinear Dynamics). Two indepen-dent immunoprecipitation reactions were performed for all antibodies using the AssayMAP Bravo Platform (Bravo 1 and Bravo 2). Replicate injections were run for each sample. Bars represent average number of identifications across replicate injections, and error bars are –/+ 1 standard deviation.

Methods

IMAC6,931IMAC6,931

Motif AbMixture3,184

Motif AbMixture3,184

687

3,391

4,000

AssayMAPBRAVO 1

AssayMAPBRAVO 2

Batch Mode

Pho

spho

pep

tide

Iden

tifica

tions 3,500

3,000

2,500

2,000

1,500

1,000

500

0

3,397

2,784

11,709

14,000

AssayMAPBRAVO 1

AssayMAPBRAVO 2

Batch Mode

Unm

od

ified

Pep

tide

Iden

tifica

tions

12,000

10,000

8,000

6,000

4,000

2,000

0

11,800 12,282

MW

Gel Stain p-Tyr-1000 #8954

Jurk

at +

PV

Live

r

MW

Jurk

at +

PV

Live

r

5,5836,000

AssayMAPBRAVO 1

AssayMAPBRAVO 2

Batch Mode

5,000

4,000

3,000

2,000

1,000

0

5,507

4,349

Pho

spho

pep

tide

Iden

tifica

tions

7,234

12,000

AssayMAPBRAVO 1

AssayMAPBRAVO 2

Batch Mode

Unm

od

ified

Pep

tide

Iden

tifica

tions

10,000

8,000

6,000

4,000

2,000

0

6,831

10,593

475

600

AssayMAPBRAVO 1

AssayMAPBRAVO 2

Batch Mode

500

400

300

200

100

0

494

184

Pho

spho

pep

tide

Iden

tifica

tions

6,290

12,000

AssayMAPBRAVO 1

AssayMAPBRAVO 2

Batch Mode

Unm

od

ified

Pep

tide

Iden

tifica

tions

10,000

8,000

6,000

4,000

2,000

0

6,396

11,104

Jurkat + PV

Mouse Liver

Complimentary Methods

Maximum Abundance

10,000 100,000 1,000,000 10,000,000 100,000,000 1,000,000,000

Log2

Rat

io (B

ravo

: Ba

tch

Mod

e)

-10

-

5

0

5

Median Phospho = -.09 Median Unmodified = -1.71

2,9743,500

AssayMAPBRAVO 1

AssayMAPBRAVO 2

Batch Mode

Ub

iqui

tinat

ed P

eptid

e Id

entifi

catio

ns

3,000

2,500

2,000

1,500

1,000

500

0

2,903

1,181

3,789

9,000

AssayMAPBRAVO 1

AssayMAPBRAVO 2

Batch Mode

Unm

od

ified

Pep

tide

Iden

tifica

tions 8,000

7,000

6,000

5,000

4,000

3,000

2,000

1,000

0

4,082

8,196

Bravo Method2,608

Bravo Method2,608

Batch Mode368

Overlap1,206

Overlap1,206

Batch Mode36

Bravo Method84

Overlap3,701

Overlap3,701

Maximum Abundance

100,000 1,000,000 10,000,000 100,000,000 1,000,000,000

Log2

Rat

io (B

ravo

: Ba

tch

Mod

e)

-15

-1

0

-5

0

5

10

1

5

Median Ubiquitin = 0.32 Median Unmodified = -2.27

2,3132,500

AssayMAPBRAVO 1

AssayMAPBRAVO 2

Batch Mode

Ace

tyla

ted

Pep

tide

Iden

tifica

tions

2,000

1,500

1,000

500

0

2,220

1,942

3,919

6,000

AssayMAPBRAVO 1

AssayMAPBRAVO 2

Batch Mode

Unm

od

ified

Pep

tide

Iden

tifica

tions

5,000

4,000

3,000

2,000

1,000

0

4,008

5,368

Maximum Abundance

10,000 100,000 1,000,000 10,000,000 100,000,000 1,000,000,000

Log2

Rat

io (B

ravo

: Ba

tch

Mod

e)

-10

-5

0

5

1

0

Median Acetyl = 0.82 Median Unmodified = 0.20

1. Rush, J. et. al. (2005) Nat Biotechnol. 23, 94–101. 4. Guo, A. et. al. (2013) Mol Cell Proteomics. 13, 372–387.

2. Lee, K.A. et. al. (2011) J Biol Chem. 286, 41530–41538. 5. Lundgren, D. H. et. al. (2009) Curr Protoc Bioinformatics. Chapter 13, Unit 13 13.

3. Stokes, M.P. et. al. (2012) Mol Cell Proteomics. 11, 187–201.

References

A A

B

B

C

C

A A

D

D

F E

C

B B

E

C D

Presentation Posters, Case Studies and Publications