In-Depth Proteome Coverage by Improving ... - ACS Publications

Apr 22, 2017 - Although great achievement has been made in the mapping of human proteome, the efficiency of sample preparation still needs to be impro...
0 downloads 0 Views 1MB Size
Subscriber access provided by HACETTEPE UNIVERSITESI KUTUPHANESI

Technical Note

In-Depth Proteome Coverage by Improving Efficiency for Membrane Proteome Analysis Qun Zhao, Fei Fang, Yichu Shan, Zhigang Sui, Baofeng Zhao, Zhen Liang, Lihua Zhang, and YuKui Zhang Anal. Chem., Just Accepted Manuscript • Publication Date (Web): 22 Apr 2017 Downloaded from http://pubs.acs.org on April 23, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

1

In-Depth Proteome Coverage by Improving Efficiency for Membrane

2

Proteome Analysis

3 4

Qun Zhaoa,1, Fei Fanga,b,1, Yichu Shana, Zhigang Suia, Baofeng Zhaoa, Zhen Lianga,

5

Lihua Zhanga,*, Yukui Zhanga

6 7 8

a

9

Chromatographic R. & A. Center, Dalian Institute of Chemical Physics, Chinese

Key Laboratory of Separation Science for Analytical Chemistry, National

10

Academy of Science, Dalian 116023, China

11

b

12

*To whom correspondence should be addressed. E-mail: [email protected].

13

Phone& fax: +86-411-84379720.

14

1

University of Chinese Academy of Sciences, Beijing 100039, China

These authors contributed equally to this study.

15 16

1

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1

Page 2 of 19

Abstract

2

Although great achievement has been made in the mapping of human proteome, the

3

efficiency of sample preparation still needs to be improved, especially for membrane

4

proteins. Herein, we presented a novel method to deepen proteome coverage by the

5

sequential

6

methylimidazoliumchloride (C12Im-Cl). With such a strategy, the commonly lost

7

hydrophobic proteins by 8 M urea extraction could be further recovered by C12Im-Cl,

8

as well as the suppression effect of high abundance soluble proteins could be

9

decreased. Followed by the in-situ sample preparation and separation with different

10

stationary phases, more than 9810 gene products could be identified, covering 8

11

orders of magnitude in abundance, which was to the best of our knowledge, the

12

largest dataset of HeLa cell proteome. Compared with previous work, not only the

13

number of proteins identified was obviously increased, but also the analysis time was

14

shortened to a few days. Therefore, we expect that such a strategy has great potential

15

applications to achieve unprecedented coverage for proteome analysis.

extraction

of

proteins

using

urea

16 17

2

ACS Paragon Plus Environment

and

1-dodecyl-3-

Page 3 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

1

Introduction

2

Mammalian cell lines are widely used in various studies, and become indispensable

3

for pursuing biological insight.1-3 Mapping the entire cell proteome is of great

4

significance to comprehensively elucidate molecular functions, which is in favor with

5

the accurate diagnosis and treatment of diseases.4-6 Although great achievement has

6

been witnessed in the large scale proteome analysis, great efforts still should be made

7

to improve the identification of membrane proteins, with low abundance and high

8

hydrophobicity.7,8

9

Although chaotropic agents, such as urea and guanidine hydrochloride, are

10

commonly used for cell lysis and protein extraction,9-15 the solubilization efficiency

11

for membrane proteins is very limited.16 To solve this problem, sodium dodecyl

12

sulfate (SDS) has been successfully applied to extract both hydrophilic and

13

hydrophobic proteins.17-20 Unfortunately, the removal of such a detergent is difficult,

14

and the residual SDS in proteome samples, even with low concentration, can suppress

15

the subsequent enzymatic digestion and impede liquid chromatography-tandem mass

16

spectrometry (LC-MS/MS) identification. Although by filter-aided sample preparation

17

(FASP) strategy,21 SDS could be removed with high efficiency, and over 7000 gene

18

products could be identified from mammalian cell line,22-26 the analysis of low

19

abundance membrane proteins is still interfered by the co-existence of soluble

20

proteins with high abundance.

21

In our previous study, 1-dodecyl-3-methylimidazoliumchloride (C12Im-Cl) was

22

successfully used to achieve the large-scale membrane proteome analysis,27 and

23

demonstrated the superiority to other solubilizers, such as SDS and Rapigest. Herein,

24

we developed a new strategy to deepen the coverage of proteome by improving the

25

efficiency for membrane proteome analysis, with the combination of two-step

26

sequential protein extraction with urea and C12Im-Cl, in-situ sample preparation in a

27

filter, and RP-RPLC-MS/MS analysis with different stationary phase during

28

separation at low pH, which demonstrated the great promise to achieve the deep

29

coverage proteome analysis.

30 31

Experimental Section

32

Reagents and materials

33

Protease inhibitor cocktail, urea, dithiothreitol (DTT), iodoacetamide (IAA), and

34

trifluoroacetic acid (TFA) were purchased from Sigma (St. Louis, MO, USA). Trypsin 3

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1

was ordered from Promega (Madison, WI, USA). Acetonitrile (ACN, HPLC grade)

2

was bought from Merck (Darmstadt, Germany). 1-Dodecyl-3-Methylimidazolium

3

Chloride (C12Im-Cl) was obtained from Chengjie (Shanghai, China). Deionized

4

water was purified by a Milli-Q system from Millipore (Milford, MA, USA). Other

5

chemicals were of analytical grade.

6

Durashell C8 particles and XBP C18 particles (5 µm, 100 Å pore) were obtained

7

from Agela (Tianjin, China). Fused-silica capillaries (75 µm i.d./365 µm o.d.) were

8

acquired from Sino Sumtech (Handan, Hebei, China), and Vivacon® 500 microcon

9

filtration devices (molecular weight cutoff 10,000 Da) were purchased from Sartorius

10

(Goettingen, Germany).

11

Protein extraction

12

HeLa cells (7x107) were first lysed and extracted with 4 mL of 8 M urea containing

13

phosphate-buffered saline (1xPBS, pH 8.0) and 2% (v/v) cocktail. After centrifugation

14

at 150,000 g for 40 min, the supernatant was collected and named as 8 M urea fraction.

15

The residual pellet was further extracted with 400 µL of C12Im-Cl containing 50 mM

16

NH4HCO3, and then clarified by centrifugation at 36,000 g for 20 min, named as

17

C12Im-Cl fraction. Protein concentration was determined using a Bradford assay kit

18

from Bio-Rad (Hercules, CA, USA) with BSA as a standard.

19

Protein digestion and peptide fractionation

20

The 8 M urea fraction was transferred to a filter device, and washed with 50 mM

21

NH4HCO3, followed by centrifugation at 14,000 g for 15 min. Subsequently, proteins

22

were denatured and reduced with 100 mM DTT at 95°C for 5 min, and washed with

23

50 mM NH4HCO3. Then, proteins were subjected to alkylation with 20 mM IAA in

24

the dark at the room temperature for 20 min, followed by centrifugation at 14,000 g

25

for 15 min. Afterwards, proteins were digested with the trypsin/protein ratio (m/m) as

26

1:25 at 37°C for 12 h. Finally, the peptides were collected by centrifugation at 14,000

27

g for 15 min.

28

The C12Im-Cl fraction was transferred to a filter device. Proteins were denatured

29

and reduced with 100 mM DTT at 95°C for 5 min, and washed with 8 M urea in 50

30

mM NH4HCO3. Subsequently, proteins were subjected to alkylation with 20 mM IAA

31

containing 8 M urea in 50 mM NH4HCO3 in the dark at the room temperature for 20

32

min, followed by centrifugation at 14,000 g for 15 min. Then proteins were washed

33

with 50 mM NH4HCO3 for three times. Finally, proteins were digested with the

34

trypsin/protein ratio (m/m) as 1:25 at 37°C for 12 h, and the peptides were collected 4

ACS Paragon Plus Environment

Page 4 of 19

Page 5 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

1

by centrifugation at 14,000 g for 15 min.

2

The tryptic peptides obtained from 8 M urea fraction and C12Im-Cl fraction were

3

respectively fractionated by RPLC with a C18 column (5 µm, 100 Å, 250 mm×4.6

4

mm i.d.). Mobile phases A (98% H2O, 2% acetonitrile, adjusted pH to 9.5 using

5

NH3·H2O) and B (98% acetonitrile, 2% H2O, adjusted pH to 9.5 using NH3·H2O)

6

were used to set the gradient as follows: 2%-8% B, 2 min; 8%-18% B, 17 min;

7

18%-30% B, 14 min; 30%-35% B, 1 min; 35%-80% B, 1 min; 80%-80% B, 2 min.

8

The flow rate was 1 mL/min. The eluent was collected every minute. A total of 36

9

fractions were collected, and then concatenated to 18 fractions with equal time

10

interval, vacuum dried and reconstituted in 0.1% (v/v) FA and 2% (v/v) ACN in water

11

for subsequent analyses. Besides, for peptides from C12Im-Cl fraction, the

12

non-retained peptides eluted during sample loading process were collected and

13

individually analyzed.

14

LC-ESI-MS/MS analysis

15

The 19 peptide fractions from 8 M urea extraction (~1.8 µg of each fraction) and 18

16

peptide fractions from C12Im-Cl extraction (~0.8 µg of each fraction) were further

17

analyzed by nanoRPLC-ESI-MS/MS using a high-resolution Triple TOF 5600 mass

18

spectrometer (AB SCIEX, Framingham, MA, USA). Peptides were separated on a

19

capillary column (15cm×75µm i.d.) at the flow rate of 300 nL/min. The mobile phase

20

used was composed of H2O with 2% ACN and 0.1% FA (A), and ACN with 2% H2O

21

and 0.1% FA (B) with the gradient as 5%-22% B, 40 min; 22%-35% B, 15 min;

22

35%-80% B, 5 min; 80%-80% B, 5 min. Mass spectrometry analysis was carried out

23

in a data dependent manner with full scans (350-1,250 m/z). The top 60 precursor ions

24

were selected in each MS scan for subsequent MS/MS scans with charge state of 2-5.

25

MS scans were performed for 0.25s, and 60 MS/MS scans were subsequently

26

performed for 0.04 s of each. The dynamic exclusion for MS/MS was set as 22 s. The

27

CID energy was automatically adjusted by the rolling CID function of Analyst TF 1.6.

28

Database searching

29

Wiff files from Triple-TOF 5600 MS were firstly converted to Mascot generic

30

format (mgf) files using Protein Pilot software (version 4.5). The mgf files were then

31

delivered to Mascot software (version 2.4), and searched against the database IPI

32

human 3.87 (91,464 entries, 22,200 Gene product) with the same parameters as

33

followed: trypsin as the protease with a maximum of two missed cleavages allowed; the

34

carbamidomethylation of cysteine was specified as a fixed modification; the oxidation 5

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1

of methionine and the acetylation of protein N termini were included as variable

2

modifications; the mass error was set to 0.05 Da for precursor ions and 0.1 Da for

3

fragment ions. After database searching, the results of 37 fractions were merged and

4

filtered by pBuild28 to control the false discovery rate (FDR) at peptide spectral level

5

at less than 1%, and reduce the apparent redundancy in protein identification. Besides,

6

the minimum peptide length was specified to 6 amino acids.

7

Bioinformatics analysis

8

The grand average of hydropathicity (GRAVY) values of the identified peptides

9

was calculated using ProtParam program (http://web.expasy.org/protparam/).29,30 The

10

peptides with positive and negative GRAVY values were named as hydrophobic and

11

hydrophilic peptides, respectively. The TMHMM (http://www.cbs.dtu.dk/services/

12

TMHMM/) algorithm was used to predict the transmembrane domains (TMDs) of

13

identified proteins.31 Proteins with at least one predicted TMD were considered as

14

integral membrane proteins (IMPs). The topological structure of transmembrane

15

proteins was displayed with TOPO2 software (http://www.sacs.ucsf.edu/cgi-bin/

16

open-topo2.py). The cellular components and molecular functions based on Gene

17

Ontology (GO) Consortium were assigned with GoMiner.32

18 19

Results and discussion

20

To discover more information of membrane proteins, which were easily lost during

21

commonly used urea extraction method, and further generate the extensive proteome

22

map of HeLa cell line, a two-step sequential extraction-based sample preparation

23

strategy was established. As shown in Figure 1, we firstly extracted the cell lysates

24

with 8 M urea, and then re-extracted the residual pellets with C12Im-Cl, which could

25

supply stronger dissolving capability for the hydrophobic membrane proteins

26

extraction.27 Subsequently, the obtained two protein fractions were respectively

27

treated in situ by our previously developed imFASP method to improve the efficiency

28

of protein digestion process.33 In order to further deepen the coverage of proteins,

29

RPLC at high pH (pH 9.5) was employed to fractionate peptides to decrease the

30

complexity of samples, followed by nanoRPLC at low pH (pH 3) for peptide

31

separation. It should be pointed out that, the collected fractions of peptides from 8 M

32

urea extraction and those from C12Im-Cl extraction were respectively separated on

33

C18 and C8 capillary columns, since our previous study demonstrated that C8 column

34

was beneficial to improve the separation efficiency of hydrophobic peptides.34 6

ACS Paragon Plus Environment

Page 6 of 19

Page 7 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

1

By such a protocol, the totally required MS/MS running time for the single-run

2

analysis was ca. 34 h, much faster than that in previous work (12 days).23 In total,

3

202,362±685 peptide-spectrum matches (PSMs) were obtained (FDR