H. LEE MOFFITT CANCER CENTER & RESEARCH INSTITUTE, AN NCI COMPREHENSIVE CANCER CENTER – Tampa, FL

1-888-MOFFITT (1-888-663-3488) www.MOFFITT.org

© 2010 H. Lee Moffitt Cancer Center and Research Institute, Inc.

Introduction

Proteome profiling experiments must satisfy two primary requirements: maximizing depth of protein sampling as well as scaling to clinical cohorts. The approach presented here integrates discovery experiments used to create reference databases and rapid data independent acquisition (DIA) methods to examine large clinical cohorts using < 3 hours of instrument time per individual patient’s resected tumor specimen.1-2 The combination facilitates data mining for protein identification and relative quantitation across larger sample sets within 1 month.

Scaling Discovery Proteomics to Large Lung Cancer Cohorts using Data Independent Acquisition

Bin Fang,1 Melissa A. Hoffman,1 Amol Prakash,2 Scott M. Peterman,3 Paul A. Stewart,1 Guolin Zhang,1 Richard Z. Liu,1 Matthew A. Smith,1 Joseph Johnson,1 Steven A. Eschrich,1 Eric B. Haura,1 & John M. Koomen1

1. H. Lee Moffitt Cancer Center, Tampa, FL 2. OPTYS Tech 3. Thermo Fisher Scientific

Experimental Details

Comparison of Discovery Methods

Sample Preparation: FFPE Tissue was processed using the method reported by Sprung et a l[3]. Twelve fractions from basic pH reversed phased liquid chromatography (bRPLC) were analyzed using LC-MS/MS. Unfractionated digests were analyzed using pSMART and hybrid DIA acquisition methods.

Spectral Library Construction: Sequest and Mascot searches were used for spectral library generation (Proteome Discoverer 1.4, Thermo Fisher Scientific).

DIA and Data Processing: Samples were analyzed with MS1 followed by narrow window DIA (n = 18) in each cycle. The data were matched to the spectral libraries and qualitative and quantitative data processing was performed using Pinnacle (OPTYS Tech). In addition, hybrid acquisition with DIA and targeted MS2 is examined using cancer biomarkers.

References

Acknowledgments This work used Analytic Microscopy, Proteomics, and Tissue Cores funded in part by Moffitt’s Cancer Center Support Grant (NCI P30-CA076292). Project funding provided by the American Lung Association (Lung Cancer Discovery Award LCD-257857-N), the NCI (R21-CA169980), Moffitt Lung Cancer Center of Excellence, and Moffitt’s Innovation and Technology Pilot Project Funding.

1. Prakash A, et al. J Proteome Res. 2014, 13, 5415-30. 2. Gillet LC., et al. Mol Cell Proteomics 2012, 11, O111.016717 3. Sprung RW Jr, et al. Mol Cell Proteomics 2009, 8, 1988-98. 4. Stewart et al. PLoS ONE 2015, 10, e0142162.

Conclusions 1. DIA balances a sufficient depth of coverage with the ability to analyze large cohorts of patients (n =

100-400) within 1-2 months on a single instrument. 2. Hybrid data acquisition / processing method developed, which includes spectral library generation,

scans combining MS1, asymmetric narrow window DIA, targeted MS2 scans, spectral library matching and dual MS1/MS2 quantitation using Pinnacle software.

3. Applicable to Tumor Microarray Specimens.

m/z m/z

16 Highest Intensity Peptides for MS/MS

DDA

DIA

?

MS/MS across m/z Range with User-Defined Windows

m/z m/z

Undersampling: What Peptides/Proteins are Missed?

Maximize ID and Quantification from Single Sample

Lung Tissue Sample Type # Squamous Cell Carcinoma (SCC) 50 SCC Adjacent Lung Tissue 50 Adenocarcinoma (ADC) 46 ADC Adjacent Lung Tissue 13

ADJ Lung- Pools

SCC- Pool

ADC-Pool

812

491

3,661 41%

1,494

1,173 322

868

CTR

L1

CTR

L2

CTR

L3

CTR

L4

AD

C1

A

DC

2

SCC

1

SCC

2

High

Low

TMT (6)

DIA (1)

Label Free (24)

DIA TMT Label Free

Ctrl Tumor Ctrl Tumor Ctrl Tumor

Selected Cancer-Related Proteins Not Observed in DIA, but in Spectral Library: 4E-BP1, AKT(PKB), ASK1 (MAP3K5), Bcl-2, Bcl-XL, Bcl-XS, Bim, Collagen IV, COX-2 (PTGS2), c-Src, Frizzled, GIPC, GSK3 alpha/beta, GYS1, HGF receptor (Met), IBP, IGF-1 receptor, IGF-2, I-kB, IKK-beta, IKK-gamma, Insulin receptor, IRAK1/2, IRAK4, K-RAS, Matrilysin (MMP-7), MEK1/2, MEK3(MAP2K3), MEK6(MAP2K6), MELC, Midkine, MMP-1, MNK1, MRLC, MyD88, Neurofibromin, NF-κB, N-WASP, p130CAS, PARD6, PDK (PDPK1), PI3K cat class IA, PI3K cat class IA (p110-alpha), PI3K cat class IB (p110-gamma), PI3K reg class IA, PI3K reg class IA (p85), PI3K reg class IA (p85-alpha), PI3K reg class IA (p85-beta), PKC-zeta, PLAUR (uPAR), PP2A catalytic, PTEN, RHEB2, Shc, SMAD2, SP1, TAB1, Talin, TGF-beta 1, Tryptase, VCAM1, VEGF-A, WNT, XIAP

Prot

eins

in D

IA

Proteins in Spectral Library

Top 50 Enriched Pathways y = 0.69x + 4.5

R² = 0.80

0

20

40

60

0 20 40 60 80 100 Apoptosis Signaling, GeneGO, Metacore

Proteins Peptides

1

9

17

0 50 100 150 200 Qua

ntifi

catio

ns

(x 1

03 )

Sorted Sample Order

ADC Adj Ctrl

ADC ADC Pool SCC SCC

Pool SCC Adj Ctrl

AD

C

Ctrl

(A)

SCC

C

trl (S

) Po

ol (A

)

Pool

(S)

0

3E3

6E3

Qua

ntifi

able

Pro

tein

s

AD

C

Ctrl

(A)

SCC

C

trl (S

) Po

ol (A

)

Pool

(S) 0

1E4

2E4

Qua

ntifi

able

Pep

tides

y = 0.2462x + 1030.3 R² = 0.9644

0

3E3

6E3

0 1E4 2E4 Qua

ntifi

able

Pro

tein

s

Quantifiable Peptides

m/z 0

50

100

Inte

nsity

486.76 483.76

485.5 SIS Endogenous

R² = 0.998 0

50

100

150

0 20 40 60 80 100 Amount (fmol)

L/H

Rat

io

ERBB2_ELVSEFSR Heavy: 1.25 fmol/injection

Light: 0.04 to 80 fmol CV < 5.5% across range

MS

Targeted MS2 m/z 485.5

DIA m/z 487.50

28.6 28.8 29.0 29.2 29.4 29.6 29.8

Time (min)

0

50

100

0

50

100

Rel

ativ

e Ab

unda

nce

0

50

100

450

1440

485.5

Time (s)

m/z

0 19.2

NL: 6.01E6

NL: 9.02E6

NL: 4.43E6

7

8

9

10

11

12

18

13

14

15

16

17 3,605

3,585

3,097 3,144

3,576

3,598

1mm 0

200

0 120 Area (mm2)

Prot

ein

(µg)

DIA LC-MS/MS proteome profiling data were acquired from lung squamous cell carcinoma, adenocarcinoma, and adjacent lung tissue. DIA analysis of cores from a tissue microarray produced quantitative data for > 3,000 proteins each, indicating that the technology can be applied to minimal amounts of formalin fixed paraffin embedded tumor tissue.

LB-140

Tissue Microarray Analysis

Molecular Classification using Lung Cancer Proteomes

Targeted Quantification in DIA

Pilot Project Comparing Squamous Tumors to Adjacent Lung Tissue using Label-Free, TMT, and DIA Discovery Proteomics Enables Investigation of Detectable Biology.

Adenocarcinoma (EGFR WT, KRAS WT) Squamous Cell Carcinoma

Assessment of 159 Total Tumor and Lung Tissue Proteomes using 200 DIA-LC-MS/MS Analyses over 25 Days Enables Molecular Classification of Tumors.

Protein Quantification from Single TMA Core Sections Builds on Whole Slide Imaging (Aperio), Histological Maps (Genie) , and DIA-LC-MS/MS Analysis.

Targeted Peptide-Based Quantitative Assays for Known Biomarkers (e.g. HER2) can be Built into DIA Proteome Analysis Experiments.