A Set of Organelle-Localizable Reactive Molecules for Mitochondrial

May 26, 2016 - Department of Synthetic Chemistry and Biological Chemistry, Graduate School of Engineering, Kyoto University, Katsura, Nishikyo-ku, Kyo...
0 downloads 9 Views 7MB Size
Subscriber access provided by LAURENTIAN UNIV

Article

A set of organelle-localizable reactive molecules for mitochondrial chemical proteomics in living cells and brain tissues Yuki Yasueda, Tomonori Tamura, Alma Fujisawa, Keiko Kuwata, Shinya Tsukiji, Shigeki Kiyonaka, and Itaru Hamachi J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.6b02254 • Publication Date (Web): 26 May 2016 Downloaded from http://pubs.acs.org on May 27, 2016

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.

Journal of the American Chemical Society 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 36

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

Journal of the American Chemical Society

3

A set of organelle-localizable reactive molecules for mitochondrial chemical proteomics in living cells and brain tissues

4

Yuki Yasueda,1 Tomonori Tamura,1 Alma Fujisawa,1 Keiko Kuwata,2 Shinya Tsukiji,3,4

5

Shigeki Kiyonaka,1 and Itaru Hamachi1,5

1 2

6 7

1

8

Engineering, Kyoto University, Katsura, Nishikyo-ku, Kyoto, 615-8510, JAPAN

9

2

Department of Synthetic Chemistry and Biological Chemistry, Graduate School of

Institute of Transformative Bio-Molecules (ITbM), Nagoya University, Chikusa, Nagoya,

10

464-8602, JAPAN

11

3

12

Showa-ku, Nagoya, 466-8555, JAPAN

13

4

14

Showa-ku, Nagoya, 466-8555, JAPAN

15

5

16

Chiyodaku, Tokyo, 102-0075, JAPAN

Frontier Research Institute for Materials Science, Nagoya Institute of Technology, Gokiso,

Department of Materials Science and Engineering, Nagoya Institute of Technology, Gokiso,

CREST(Core Research for Evolutional Science and Technology, JST), Sanbancho,

17 18

Correspondence should be addressed to I.H. ([email protected]).

19 20 21 22 1 ACS Paragon Plus Environment

Journal of the American Chemical Society

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

Page 2 of 36

1

Abstract

2

Protein functions are tightly regulated by their subcellular localization in live cells and

3

quantitative evaluation of dynamically altered proteomes in each organelle should provide

4

valuable information. Here, we describe a novel method for organelle-focused chemical

5

proteomics using spatially limited reactions. In this work, mitochondria-localizable reactive

6

molecules (MRMs) were designed that penetrate biomembranes and spontaneously

7

concentrate in mitochondria, where protein labeling is facilitated by the condensation effect.

8

The combination of this selective labeling and liquid chromatography–mass spectrometry

9

(LC-MS) based proteomics technology facilitated identification of mitochondrial proteomes

10

and the profile of the intrinsic reactivity of amino acids tethered to proteins expressed in live

11

cultured cells, primary neurons and brain slices. Furthermore, quantitative profiling of

12

mitochondrial

13

oxidant-induced apoptotic process was performed by combination of this MRMs-based

14

method with a standard quantitative MS technique (SILAC: stable isotope labeling by amino

15

acids in cell culture). The use of a set of MRMs represents a powerful tool for chemical

16

proteomics to elucidate mitochondria-associated biological events and diseases.

proteins

whose

expression

levels

change

17

2 ACS Paragon Plus Environment

significantly

during

an

Page 3 of 36

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

Journal of the American Chemical Society

Introduction

2

Proteome analysis relying on advanced mass (MS) spectroscopy is now a powerful

3

strategy for large scale profiling of proteins in biological samples, including cell lysates, live

4

cells and in vivo.1 The combination of chemical protein labeling with MS-based proteome

5

techniques has remarkably facilitated the profiling capability, by defining a specific class of

6

proteins with labeling selectivity.2 One of the most successful examples represents so-called

7

activity-based protein profiling (ABPP), which was invented by Cravatt’s group.3 Here, only

8

proteins exhibiting an enzymatic activity-of-interest can be labeled with a bio-orthogonal

9

handle. ABPP employs active site-directed chemical reagents (known as activity-based

10

probes: ABPs) to label proteins with a particular enzyme activity in their active pocket, which

11

can focus on the proteome depending on a particular functional state of enzymes, prior to

12

MS-fingerprinting analysis.

13

In addition to activity, protein localization in a variety of subcellular compartments

14

of live cells should be crucial for proteomics, because the expression level and localization of

15

proteins in each organelle are dynamically varied in response to environmental conditions and

16

tightly involved in their biological functions. The spatio-temporal changes of proteomes in

17

organelles are responsible for the healthy state of organelles, cells and the consequent live

18

bodies. For example, mitochondria, an energy generating organelle, plays crucial roles in

19

regulating redox states of whole cells and mitochondrial dysfunction induced by a perturbed

20

proteome is associated closely with serious diseases,4 such as cancer, Alzheimer’s and

21

Parkinson’s diseases. Therefore, it is crucial to explore what proteins are expressed within a

22

specific organelle and how the proteome changes dynamically with a particular stimulus,

23

thereby addressing a number of fundamental questions that should facilitate medical 3 ACS Paragon Plus Environment

Journal of the American Chemical Society

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 4 of 36

applications.

2

In conventional organelle-focused proteomics, subcellular fractionation followed by

3

a high-resolution 2D-gel and/or LC-MS/MS is a widely accepted approach to characterize

4

organelle proteome.5 However, despite careful preparation, the isolated organelle of interest

5

often suffers from contamination of other organelles and/or potential loss of significant

6

components, which leads to the false-positive and false-negative results. A new method that

7

allows the organelle-focused proteomic mapping without relying on any fractionation was

8

developed recently by Ting and coworkers.6 They selectively over-expressed an

9

organelle-targeting engineered peroxidase enzyme in mitochondria, by which spatially limited

10

oxidation reactions were facilitated to selectively modify mitochondrial proteins with a biotin

11

affinity tag, followed by extensive MS analysis. Although promising, this enzymatic method

12

always requires expression of an engineered peroxidase. Hence it is difficult to apply the

13

technique to cells or tissues7 (such as primary neurons and brain slices) that are not readily

14

amenable to exogenous gene expression.8 In addition, this method uses rather harsh reaction

15

conditions (a high concentration of H2O2), which may detrimentally perturb natural biological

16

environments9 and such phenoxyl radical-mediated labeling may be unsuited for proteins that

17

lack accessible electron-rich residues such as tyrosine on their surface.6 To sidestep these

18

drawbacks, chemistry-based proteomics focusing on an organelle would be attractive.

19

Recently, a few of organelle-directed ABPs have been reported, in which organelle-targeting

20

moieties are conjugated with ABPs.10 However, the ABPs that label a limited enzyme family

21

may not well cover broad-range of proteomes in a target-organelle. For the comprehensive

22

organelle-proteomics, novel approaches not relying on a particular enzyme activity should be

23

required. Toward this end, we recently reported nuclear-localizable reactive molecules 4 ACS Paragon Plus Environment

Page 5 of 36

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

Journal of the American Chemical Society

1

(NRMs) containing a DNA-binding motif and a reactive chloroacetyl group.11 The reagents

2

spontaneously accumulated in the nucleus and selectively reacted with nuclear proteins under

3

intact cell conditions, which allowed the characterization of a range of nuclear proteins by

4

combination with LC-MS/MS analysis.

5

We describe herein a novel chemical tool for mitochondria-focused proteomics that

6

exploits a set of mitochondria-localizable reactive molecules (MRMs) exhibiting a variety of

7

chemical reactivity and selectivity for endogenous proteins. Spontaneous concentration and

8

the resultant high concentration of MRMs in mitochondria allowed for spatially restricted

9

chemical reactions with natural mitochondrial proteins of live cells. By combination of this

10

mitochondria-selective labeling with MS-based protein identification, many labeled peptides

11

can be identified. Thus, the chemo-selectivity of these reactive moieties toward a broad array

12

of side chains of endogenous proteins in live cells can be addressed in detail. Owing to its

13

mild labeling conditions and simple protocol, which need neither genetic manipulation nor

14

organelle fractionation, we succeeded in profiling many mitochondrial proteins in live primary

15

cultured neurons and even in brain slices, as well as HeLa cell lines. Furthermore, coupling of

16

this method with a standard quantitative MS technique (SILAC: stable isotope labeling by

17

amino acids in cell culture) enabled to analyze the dynamic change of mitochondrial proteome

18

induced by an oxidative stress.

19 20

Results and discussion

21

Strategy and molecular designs of mitochondria-localizable reactive molecules (MRMs).

22

For analyzing the mitochondrial proteome, a set of chemically reactive small molecules

23

equipped with mitochondria-targetable function were designed (Figure 1). Since there are 5 ACS Paragon Plus Environment

Journal of the American Chemical Society

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

Page 6 of 36

1

various nucleophilic amino acid residues exposed on the surface of proteins, many

2

electrophilic molecules can potentially react to modify proteins.12,17 Therefore, it was expected

3

that chemical labeling of proteins located in mitochondria should be facilitated by

4

co-localization of MRMs with mitochondrial proteins, when these are efficiently condensed

5

together in the same small spaces like the case of NRMs.11 Proteins modified by such spatially

6

limited reactions are then analyzed and identified after cell lysis to collect data on their

7

localization in live cells. That is, MRMs-based labeling allows the transfer of dynamically

8

altered information on mitochodrial proteome in live cells to a permanent covalent bond.

9

We prepared MRMs composed of both a mitochondria-localizing moiety and a

10

chemically reactive moiety. Tetraethyl-rhodamine (Et4-Rhod) was employed as the

11

localizable moiety for mitochondria13 so that the localization of the MRMs can be readily

12

evaluated by live-cell fluorescent imaging. As a reactive moiety, on the other hand, a variety

13

of reactive functionalities exhibiting distinct amino acid selectivity were used, with the aim of

14

modifying as many amino acids as possible.14 We explored eight reactive moieties here.

15

Among them, chloroacetyl (CA) and maleimide (MI) are expected to preferentially react with

16

the thiol group of cysteines.14a, 15 Amino groups such as the α-amine at the N-terminus of

17

proteins and the ε-amine of lysine side chains are reactive with thiophenyl ester (TE) and

18

thiazolidinethione (TZ), which gives a stable amide bond.16 Epoxide (EP) is reported to react

19

with various nucleophilic amino acids, such as Cys, Lys and His.17 Sulfonyl fluoride (SF) was

20

used previously to label active sites of Ser proteases as ABPP probes and was also expected to

21

react with Lys and Tyr residues.18 To label a wider range of amino acids, photoreactive groups

22

such as phenylazide (PA) and benzophenone (BP), conventional photoaffinity labeling

23

reagents, were also employed.19 Figure 1 summarizes a set of such MRMs and their 6 ACS Paragon Plus Environment

Page 7 of 36

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

Journal of the American Chemical Society

1

(expected) amino acid preferences. Synthesis and characterization of the derivatives by NMR

2

and MS spectroscopy was conducted according to Scheme S1.

3 Living Cell

Mitochondria

LC-MS/MS m/z m/z

Mitochondria-selective protein labeling

Identification of ! Mitochondrial proteome

Mitochondria-localizable ! Reactive Molecules (MRMs)

Localizable moiety!

Reactive moiety

O O

N

Cl

N+

O

O

O N

1b(EP)

Cys

Cys, Lys, His

F

O

N H

O

1a(CA)

O S

O

O

S O

1c(SF)

1d(TE)

Ser, Lys, Tyr

Lys

N O

Et4-Rhod

O N O

N S

5 6 7 8

O

O

1e(TZ)

4

O

N3

S

1f(BP)

Lys

1g(PA)

O

O

1h(MI) Cys

Figure 1. Schematic illustration of mitochondria selective protein labeling and profiling by a set of mitochondria-localizable reactive molecules (MRMs). Et4-Rhod, tetraethyl-rhodamine; CA, chloroacetyl; EP, epoxide; SF, sulfonyl fluoride; TE, thiophenyl ester; TZ, thiazolidinethione; BP, benzophenone; PA, phenylazide; MI, maleimide.

9 10 11

Cell permeability and localization of MRMs.

12

The cell permeability and mitochondrial localization of a set of MRMs in HeLa cells

13

were evaluated initially by confocal laser scanning microscopy (CLSM). Rapid (within 15

14

min) and spontaneous localization in mitochondria were observed for the majority of these

15

molecules. For example, the CLSM image of HeLa cells treated with 1a(CA) for 15 min is

16

shown in Figure 2a. This merged well with the image stained by Rhodamine123, a 7 ACS Paragon Plus Environment

Journal of the American Chemical Society

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

Page 8 of 36

1

representative mitochondrial localization marker.20 Other MRMs such as 1b(EP), 1c(SF),

2

1d(TE), 1e(TZ), 1f(BP) and 1g(PA) gave almost identical CLSM images to the one observed

3

with 1a(CA) (Supplementary Figure S1). The excellent overlap and the resultant high

4

Pearson’s correlation coefficient (more than 0.85) indicated that these MRMs efficiently

5

accumulated in mitochondria of live HeLa cells. Such mitochondrial localization did not occur

6

in HeLa cells pretreated with carbonyl cyanide m-chlorophenylhydrazone (CCCP), a standard

7

uncoupler of the mitochondrial membrane potential21 (Figure S2), revealing that spontaneous

8

accumulation of these MRMs was driven by the high membrane potential of mitochondria.

9

We also spectroscopically determined the intracellular concentration of these MRMs

10

after lysis of the HeLa cells (Figure S3). In the optimized conditions for the below-mentioned

11

protein labeling experiments, the concentrations of 1a(CA)–1g(PA) are in the range of 24 to

12

213 µM inside the cell. These values correspond to a 4 to 21-fold condensation relative to the

13

concentration of the MRMs added extracellularly. Given the volume of mitochondria in live

14

HeLa cells, it is regarded that the condensation effect is further enhanced by ~3-fold,22 which

15

could facilitate the intermolecular labeling reactions in mitochondria. In contrast, the

16

maleimide-appended Et4-Rhod 1h(MI) gave the rather weak intracellular fluorescence,

17

indicating its poor cell permeability (Figure S4). Hence, 1h was omitted from our set of

18

MRMs.

19 20

8 ACS Paragon Plus Environment

Page 9 of 36

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

Journal of the American Chemical Society

1

(a)

(b) 1a(CA)

(kDa) 250 150

Rho 123

Fluorescence

CBB

100 75 50

Merge

Phase

37

25 Reagent 1a! 1b! 1c! 1d! 1e! 1f! 1g!

R = 0.900

2 3 4 5 6 7 8 9 10 11 12

20 µm

(CA) (EP) (SF) (TE) (TZ) (BP) (PA)

Conc.(µM) 10 10 10 Time (h) 6 6 6

5 2

5 5 10 2 0.5 0.5

Figure 2. Mitochondrial proteome labeling in HeLa cells with MRMs. (a) Confocal micrograph images of HeLa cells stained with 1a(CA) (1.0 µM) and Rhodamine 123 (200 nM). Colocalization was quantified using Pearson’s correlation coefficient (R). (b) SDS-PAGE analysis by in-gel fluorescence imaging and Coomassie Brilliant Blue (CBB) staining. Reaction conditions: HeLa cells were incubated with MRMs (5.0–10 µM) in serum-free DMEM at 37 °C for 0.5–6 h. In the case of 1f(BP) and 1g(PA), the cells were then UV irradiated (40 min) on ice. After labeling, the cells were washed, lysed and analyzed by SDS-PAGE. The optimization of labeling conditions with each MRM in living cells was shown in Figure S5 and S6.

13 14

Labeling and identification of proteins in mitochondria of HeLa cells.

15

Having a set of cell permeable MRMs in hand, we then evaluated their reactivity with

16

natural proteins in HeLa cells by SDS-PAGE analysis. After the addition of each MRM

17

(1a(CA)–1g(PA)) to the culture medium of HeLa cells, followed by incubation, the cells were

18

lysed and the resultant lysate was subjected to SDS-PAGE. As shown in Figure 2b, the

19

labeled protein bands were detected by fluorescence due to the Et4-Rhod fluorophore, which

20

indicated that the band pattern was distinct and dependent on the labeling reagent. In contrast

21

to the many bands observed for most MRMs, 1f(BP) gave only a few bands with noticeable

22

band intensity. In addition, the CLSM image after 1a(CA) treatment was not altered by the 9 ACS Paragon Plus Environment

Journal of the American Chemical Society

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

Page 10 of 36

1

addition of CCCP, whereas clear mitochondrial images were not observed in the case of the

2

CCCP-pretreated HeLa cells (Figures S2 and S7). These data additionally support the

3

covalent attachment of 1a(CA) to proteins in mitochondria.

4

Identification of the labeled proteins in detail was subsequently carried out by

5

LC-MS/MS analysis. The Et4-Rhod fluorophore serves as a vehicle for targeting organelles

6

and is also an excellent antigen to elicit antibodies that should enrich labeled proteins and

7

peptides in downstream immunoprecipitation (IP) assays. We thus prepared an anti-Et4-Rhod

8

antibody23 and used this antibody for concentrating the labeled proteins/peptides prior to

9

MS/MS analysis. To directly detect the labeled peptides and to minimize the contamination

10

during IP and gel separation, we exploited the modified “gel-free” method that was originally

11

reported by Cravatt et al. (Figure 3a).24 In our protocol, the labeled peptides rather than

12

proteins were concentrated by IP, which were then analyzed by LC-MS/MS without

13

SDS-PAGE. In the case of 1a(CA), for example, we detected 373 labeled peptides, thus the

14

labeled amino acids were identified (Figure 3b, Table S1 and Supplementary data S1).

15

Clearly, Cys-labeled peptides (290 (78%)) were readily detected as the largest group. The

16

second largest groups were Ser- (8%) and Thr- (6%) labeled peptides, where the amino acid

17

preference is consistent with the intrinsic chemical reactivity of CA. A similar Cys preference

18

was also observed by 1b(EP) with labeling of 83% of the peptides and Ser-labeled peptides

19

were the second largest group (6%). On the other hand, 1d(TE) gave 82% of Lys-labeled

20

peptides and Ser-labeled peptides (12%) formed the second largest group. This preference was

21

almost the same as that of 1e(TZ) (90% for Lys). In the case of 1c(SF), it was characteristic

22

that Lys (46%) and Tyr (40%) were the two predominant amino acids labeled. Peptides

23

labeled by 1f(BP) and 1g(PA) were not detected, which may be due to the instability of the 10 ACS Paragon Plus Environment

Page 11 of 36

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

Journal of the American Chemical Society

1

formed covalent bond and/or the formation of unexpected adducts. Of note, identifying

2

labeled peptides (not proteins) that we conducted here enabled us to address the amino acid

3

preference of these reactive moieties in live cells explicitly, which revealed that the chemical

4

selectivity in live cells roughly reflected those expected in conventional organic reactions.

5

Additionally, this strongly indicated that most of the detected labeled peptides did not come

6

from artifacts of the present MS fingerprinting protocol.

7

The data derived from these labeled peptide sequences were used to identify the labeled

8

proteins. Figure 3c and Table S2 summarizes the identified proteins and their localization.

9

The use of 1a(CA) labeled 244 proteins, and 196 of these proteins are assigned as

10

mitochondrial proteins (80%). Similarly, 85 (among 131), 53 (among 74), 52 (among 65) and

11

25 (among 57) mitochondrial proteins were identified by 1b(EP), 1c(SF), 1d(TE) and 1e(TZ),

12

respectively. Figure 3d and Figure S9 show a representative protein, trifunctional enzyme

13

subunit alpha, mitochondrial (HADHA), identified by 15 labeled peptides, where 12 sites

14

were modified with 5 MRMs. Similarly, 18 sites of 60 kDa heat shock protein, mitochondrial

15

(HSPD1) and 7 sites of Serine hydroxymethyltransferase, mitochondrial (SHMT2) were

16

labeled with 5 and 4 MRMs, respectively (Figure 3e, 3f and Figure S9). These results clearly

17

showed that different labeling reagents reacted with distinct peptide sequences and/or amino

18

acids (despite partial overlap) and more importantly these labeled residues were mainly

19

exposed on the protein surface and not buried in the active pocket. Such surface labeling was

20

observed in most of identified proteins whose 3D-structures were available or postulated25

21

(Figure S9 and S10), suggesting the mitochondria-localized labeling was facilitated by simple

22

intermolecular reactions in limited spaces,26 and not by specific activity/affinity with proteins

23

(Figure S12). It is also evident that the number and sort of identified proteins were largely 11 ACS Paragon Plus Environment

Journal of the American Chemical Society

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

Page 12 of 36

1

dependent on the reactive moiety of the MRMs. More importantly, each MRM can label

2

unique proteins, while some overlap with each other. That is, Figure 3g and Table S3 show

3

that 105 mitochondrial proteins were uniquely identified by 1a(CA), and 10 by 1b(EP), 18 by

4

1c(SF), 14 by 1d(TE), and 4 by 1e(TZ). These results clearly highlight the advantage of using

5

a variety of chemically reactive moieties to cover a wide range of proteomes in mitochondria.

6

The annotation analysis of the whole labeled proteins revealed that 69% are assigned to

7

mitochondria-localizing proteins, 6% of the proteins were not assigned (un-assigned) to a

8

location and 25% were assigned as non-mitochondrial proteins (Figure 3c). The identified

9

mitochondrial proteins were further classified into submitochondrial localizations, namely,

10

“matrix”, “inner membrane (IMM)”, “intermembrane space (IMS)” and “outer membrane

11

(OMM)” (Figure 4a). Among 259 mitochondrial proteins identified in our study, 182 proteins

12

can be annotated by submitochondrial localization on the basis of the Gene Ontology (GO)

13

terms. We found that 163 proteins (66+33+64) were assigned to the matrix or IMM but not

14

IMS or OMM localization, while 19 proteins were assigned to the IMS or OMM localization.

15

That is, our MRM mainly labeled proteins localized in matrix or IMM of mitochondria.

16

Moreover, it was revealed that the coverage of OMM proteins, defined by the percentage of

17

identified OMM proteins over the whole OMM proteins, is lower (13%) than that of matrix

18

(48%), IMM (40%) and IMS (39%) (Figure 4b and 4c). These results clearly show our

19

MRMs mainly react with proteins inside mitochondria, which is consistent with the assumed

20

submitochondrial localization of Et4-Rhod.27 Among the un-assigned proteins, there are some

21

proteins shuttling between mitochondria and ER (or other organelles), such as Reticulon-428

22

and Apolipoprotein C-III29, which may be re-assigned as mitochondria-localized ones. The

23

25% non-mitochondrial proteins detected offers some caution to this method providing false 12 ACS Paragon Plus Environment

Page 13 of 36

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

Journal of the American Chemical Society

1

positive information on protein localization. This is plausibly due to side reactions with highly

2

abundant proteins such as cytoskeletal proteins30, the misdistribution of the MRMs during the

3

labeling process and mischaracterization of the labeled peptides during LC-MS/MS analysis.

4

However, as mentioned above, these MRMs were efficiently concentrated in mitochondria of

5

live HeLa cells. Additionally, negative control experiments using the HeLa lysate and

6

CCCP-pretreated HeLa cells as the labeling sample (1a(CA)) confirmed that the content of the

7

mitochondria-localizing proteins decreased to only 13 and 16%, respectively (Figure 4d,

8

Figure S13 and Supplementary data S5). Given these results, it is concluded that most of

9

MRMs mainly react with mitochondrial proteins by spontaneous condensation in the

10

mitochondrial limited space under live cell conditions and therefore MRMs are suitable tools

11

for mitochondria-selective chemical proteomics.

12 13 14

13 ACS Paragon Plus Environment

Journal of the American Chemical Society

(a)

Digestion

IP

LC-MS/MS m/z

Labeled Proteome

Crude Peptides

Labeled Peptides

(b)

m/z

(c) 400% 350%

S28%

300% 250% 200%

S13%

150%

C290%

100%

C166%

50%

S15%

Y46%

K100%

K53%

K70%

0% 1a(CA)% 1b(EP) 1b(EP)% 1a(CA)

1c(SF)% 1c(SF)

1d(TE)% 1d(TE)

69% 21$

400$

Y% T% S% K% H% E% D% C%

Number of iden-fied proteins

Number of identified peptide

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

Page 14 of 36

350$

Mitochondrial

300$

80%

250$

38$

200$

10$

150$

196$

100$ 50$ 0$

1e(TZ)% 1e(TZ)

95$

Other Organellar

CA$ 1a(CA)

Un-assigned 65% 11$ 72% 35$ 21$ 85$ 53$ EP$ 1b(EP)

80% 9$

259$

44% 4$

52$

SF$ TE$ 1c(SF) 1d(TE)

28$

4$ 25$

TZ$ 1e(TZ)

Y343 (1c)

(d) K334 (1c)

H598 (1c)

C322 (1b) K570 (1d)

K309 (1d) T294 (1d)

(e)

active site

K569 (1c)

K295 (1d)

K646 (1c, 1d, 1e)

(f)

K292 (1d, 1e)

C80 (1a) C91 (1a)

K202 (1c, 1d)

T406 (1d)

K205 (1c, 1d, 1e ) K71 (1c, 1d)

active site

K82 (1c, 1d) K96 (1c, 1d, 1e ) H267 (1c) Y385 (1c)

K133 (1d)

Y106 (1c)

K157 (1d)

(g) Overlapped

Unique Identified 259 mitochondrial proteins

1a(CA) 1b(EP) 1c(SF) 1d(TE)

1

1e(TZ)

2 3 14 ACS Paragon Plus Environment

K409 (1e)

Total$ Total

Page 15 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

Figure 3. Identification and profiling of Et4-Rhod labeled proteins and their modified residues. (a) Workflow of labeled-peptide/protein identification by the gel-free protocol. (b) The number of identified peptides and their labeled residues determined by LC-MS/MS analysis. Y, tyrosine; T, threonine; S, serine; K, lysine; H, histidine; E, glutamic acid; D, aspartic acid; C, cysteine. (c) The number of proteins identified by LC-MS/MS analysis from HeLa cells and the ratio of mitochondrial proteins. The identified proteins were classified by cellular location using the protein ontology information provided in the UniProt database. (d, e, f) 3D structures of (d) HADHA, (e) HSPD1 and (f) SHMT2 with labeled residues (colored). These protein models from SWISS-MODEL were manipulated and rendered in PyMOL. Additional images of these proteins from different viewpoints and other representative proteins are showed in Figure S9 and S10. (g) Color map representation of identified mitochondrial proteins. Proteins that were uniquely or overlappingly identified by each MRM are highlighted in blue or red, respectively. Detailed information on identified proteins is presented in Supplementary data S1.

(a)

(b)

Unclassified 77

(d)

(c)

Mitochondrial

18 19 20 21 22 23 24 25

Submitochondrialspaces

Identified

Whole

Coverage

Matrix

103

215

48%

IMM

109

270

40%

IMS

12

31

39%

OMM

7

53

13%

Number of iden-fied proteins

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

Journal of the American Chemical Society

200"

13%

150" 100"

16% 154! 98!

50" 0"

Other organellar and un-assigned

24! lysate" lysate

18! CCCP" CCCP(+)

Figure 4. Submitochondrial localization of identified proteins and reaction specificity of MRMs. (a) Venn diagram representing the submitochondrial localization of identified proteins from HeLa cells. Based on the GO terms, proteins identified as mitochondrial proteins were further classified into “matrix”, “inner membrane (IMM)”, “intermembrane space (IMS)” and “outer membrane (OMM)”. It should be noted that some proteins are classified into plural submitochondrial regions. The list of identified proteins with their submitochondrial localization can be found in Supplementary data S1. (b) Venn diagram of submitochondrial 15 ACS Paragon Plus Environment

Journal of the American Chemical Society

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 2 3 4 5 6

Page 16 of 36

localization of whole mitochondrial proteins with prior submitochondrial localization data in MitoCarta31 or Uniprot database. (c) Coverage of identified proteins in each submitochondrial spaces. Identified, the number of identified proteins; Whole, the number of mitochondrial proteins registered in the databases. (d) The number of identified proteins and the ratio of mitochondrial proteins in negative control experiments. HeLa cell lysate and CCCP-pretreated HeLa cells were incubated with 1a(CA), followed by LC-MS/MS analysis.

16 ACS Paragon Plus Environment

Page 17 of 36

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

Journal of the American Chemical Society

Application of MRMs to primary cultured neurons and brain slices.

2

Our MRMs-based chemical proteomics method requires neither pre-treatment of cells

3

such as gene transfection, nor careful (and time-consuming) organelle fractionation processes.

4

Thus, this method can be applied to mitochondria-focused proteomics of primary cells and/or

5

more complicated natural samples such as brain tissue. As a representative example, we

6

explored the mitochondrial proteome of primary cultured neurons. Each of the five MRMs

7

were mixed with cultured neural cells isolated from the cerebral cortex of the rat brain,

8

followed by incubation. The localization of each MRM was confirmed by CLSM imaging

9

(Figures 5a, Figure S14 and Supplementary data S1), showing that the MRMs

10

spontaneously accumulated to mitochondria of the rat neuron. Using a similar protocol

11

employed for the HeLa cells, SDS-PAGE analysis and gel-free protein identification were

12

conducted (Figures 5b, 5c and Supplementary data S2). Interestingly, the amino acid

13

preference of each MRM determined by the labeled peptides was almost identical with the

14

experimental data obtained from HeLa cells (Table S4). Additionally, we identified 212

15

proteins that included at least 129 (~61%) mitochondrial proteins using 5 MRMs (Figure 5c

16

and Table S5).

17

Moreover, the mitochondrial proteomics of brain slices was accomplished by mixing

18

each MRM with acutely prepared brain slices of mice cerebral cortexes. As shown in Figure

19

5d, labeled protein bands were observed in all MRMs, while the band intensity of 1b(EP) was

20

quite weak when compared with the labeling in primary neurons. This is probably due to

21

1b(EP) poor tissue permeability. One hundred proteins were assigned as mitochondrial

22

proteins from this brain slice experiments, which corresponded to 64% of the total number of

23

proteins identified (Figure S15, Table S6 and Supplementary data S3). This ratio and amino 17 ACS Paragon Plus Environment

Journal of the American Chemical Society

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

Page 18 of 36

1

acid preference are in agreement with those of HeLa cells and primary cultured neurons

2

(Table S7). We subsequently classified these proteins on the basis of their function using

3

DAVID bioinformatics resources (Figure 5e).32 It was revealed that the majority of the

4

proteins labeled in the brain slices are involved with unique mitochondrial functions, such as

5

oxoacid metabolic process, electron transport chain and energy derivation by oxidation of

6

organic compounds. This labeling pattern is functionally identical with that obtained in

7

labeling the primary neurons with the MRMs. Although the coverage of mitochondrial

8

proteins are not sufficiently high likely due to the low tissue-permeability of MRMs, our

9

MRM-based method clearly has valuable features: (i) simplicity of the experimental protocol;

10

(ii) minimization of unfavourable effects to live samples; and (iii) successful chemical

11

proteomics of mitochondria for sensitive and complicated biological samples that are not

12

easily amenable to exogenous gene manipulation.8

13 14

18 ACS Paragon Plus Environment

Page 19 of 36

(a)

(c)

(b) 1a(CA)

Fluorescence

(kDa) 250 150

Rho 123

250#

Number'of'iden-fied'proteins

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

Journal of the American Chemical Society

100 75 50

Merge

Phase 37

200#

Other Organellar 150#

63% 11#

100#

39# 63%

50#

84#

13# 25#

0#

Reagent 1a! 1b! 1c! 1d! 1e! (CA) (EP) (SF) (TE) (TZ)

Conc.(µM) 1.0 5 Time (h) 6 6

(d) (kDa)

72#

Un-assigned

25 20 µm

61% 11#

Mitochondrial

CA# 1a! (CA)

5 2.5 2.5 6 2 2

83%

2# 7#

EP# 1b! (EP)

68% 15#

69% 18#

34#

32#

40#

SF# 1c! (SF)

TE# 1d! (TE)

TZ# 1e! (TZ)

129#

Total# Total

(e)

Fluorescence

250 150

oxoacid"metabolic"process" oxoacid metabolic process!

100 electron"transport"chain" electron transport chain

75 50

Cultured Neurons Brain Slices

energy"deriva9on"by"oxida9on"of"organic" energy derivation by oxidation ! compounds" of organic compounds!

37

cofactor"catabolic"process" cofactor catabolic process! coenzyme"metabolic"process" coenzyme metabolic process!

Reagent 1a! 1b! 1c! 1d! 1e!

0" 0

(CA) (EP) (SF) (TE) (TZ)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

5" 5

10" 10

15" 15

2020" (%)

Conc.(µM) 5.0 15 15 15 15 Time (h) 6 6 6 2 2

Figure 5. Mitochondrial proteome labeling in primary cortical neurons and brain slices with MRMs. (a) Confocal micrograph images of neurons stained with 1a(CA) (1.0 µM), Rhodamine 123 (200 nM) and Hoechst 33342 (1.0 µM). (b) SDS-PAGE analysis by in-gel fluorescence imaging. Reaction conditions: primary cortical neurons 6 × 105 cells (16-d-old Sprague Dawley rat embryos) were incubated with MRMs (1.0–5.0 µM) in serum-free DMEM at 37 °C for 2–6 h. After labeling, the cells were washed, lysed and analyzed by SDS-PAGE. (c) The number of proteins identified by LC-MS/MS analysis from primary cultured neurons and the ratio of mitochondrial proteins. (d) SDS-PAGE analysis by in-gel fluorescence imaging. Reaction conditions: Brain (cerebrum) slices (250 µm thickness) prepared from P14–21 ICR mice were incubated with MRMs (5.0–15 µM) in the ACSF solution at room temperature for 2–6 h under 95% O2 / 5% CO2. After labeling, the tissues were washed, lysed and analyzed by SDS-PAGE. (e) Functional classification of identified proteins from primary cultured neurons and brain slices by LC-MS/MS analysis (enrichment P-value by DAVID < 0.0002).

18 19

19 ACS Paragon Plus Environment

Journal of the American Chemical Society

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 20 of 36

Chemical profiling of a dynamic proteome during an oxidant-induced apoptosis process

2

We finally applied this MRM approach to explore the dynamics of mitochondrial

3

proteome of live cells in response to a cellular stress. As a proof-of-principle, we combined

4

our MRM-based method with SILAC, a quantitative MS analysis technique,33 in order to

5

investigate the change of mitochondrial proteome of HeLa cells during an apoptosis process

6

induced by paraquat (PQ),34 an oxidative stress inducer. HeLa cells grown in light SILAC

7

media were incubated with PQ for 48 h, followed by 1a(CA) (5.0 µM, 6 h), whereas cells

8

grown in heavy SILAC media were only treated with 1a(CA) in the absence of PQ (Figure

9

6a). The PQ-induced apoptosis was validated by a cleavage of PARP-1, a standard marker of

10

apoptosis (Figure S16). It was also confirmed that 1a(CA) localized in mitochondria of the

11

apoptotic HeLa cells like as normal cells (Figure S17) and the intracellular concentration did

12

not substantially change by PQ treatment (Figure S18). As shown in Figure S19, the

13

fluorescent band patterns of SDS-PAGE gel were well identical between the PQ-treated and

14

PQ-nontreated cells, indicating the PQ-treatment did not largely impact most of mitochondrial

15

proteins of HeLa cells. To obtain the optimal performance for quantitative MS analysis, we

16

carried out a standard protein identification protocol including IP of the labeled proteins (not

17

peptides), followed by MS fingerprinting analysis. 255 proteins were identified and quantified

18

in total from three independent biological replicates (Figure S20), among which 166 proteins

19

(65%) were classified into mitochondrial proteins, showing that our MRM-based labeling

20

works well in apoptotic cells (Figure 6b and Supplementary data S4). Of the quantified

21

proteins, 21 proteins were up-regulated by more than 2-fold in PQ-treated HeLa cells (Table

22

S8), which included “glutathione peroxidase 1 (GPX1)”, “Peroxiredoxin-1 (PRDX1)” and

23

“Peroxiredoxin-2

(PRDX2)” that have been reported to increase under oxidative stress or 20 ACS Paragon Plus Environment

Page 21 of 36

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

Journal of the American Chemical Society

1

apoptotic processes.35 Functional annotation analysis of the 21 up-regulated proteins using

2

DAVID revealed that terms of apoptosis-related biological process mediated by reactive

3

oxygen species (ROS), such as “regulation of apoptosis” (36%, p value: 1.54 × 10-3),

4

“regulation of programmed cell death”

5

peroxide” (18%, p value: 6.41 × 10-3), “cellular response to reactive oxygen species” (18%, p

6

value: 1.21 × 10-2), and “positive regulation of programmed cell death” (23%, p value: 1.25 ×

7

10-2), were significantly enriched (Figure S21). It was also found that 7 mitochondrial

8

proteins were down-regulated by more than 2-fold (Table S9), among which 3 proteins

9

(NDUFS3, NDUFS1 and FASN) are previously reported to decrease under oxidative stress or

10

apoptotic processes.36 These data obtained by MRM-based SILAC technique consistently

11

reflect the dynamic change of mitochondrial proteome of HeLa cells in response to the

12

oxidative stress.

(36%, p value: 1.54 x 10-3), “response to hydrogen

13

In order to further evaluate our quantitative MS data, western blotting (WB) analysis

14

was carried out. We selected 6 representative mitochondrial proteins from the list of identified

15

proteins with substantial (>2-fold) changes in the SILAC experiment. These include three

16

up-regulated proteins (AIFM1, ACO2, GPX1) and three down-regulated proteins (NDUFAF4,

17

NDUFS3, FASN). Among them, three proteins (AIFM1, ACO2, NDUFAF4) are previously

18

unknown to change in the expression levels through apoptosis, whereas substantial change of

19

the expression level for the other three (GPX1, NDUFS3, FASN) are already reported.

20

Besides, two proteins (HADHA, HSPA9) showing not substantial alterations in the SILAC

21

ratios were also evaluated by WB analysis. As shown in Figure 6c, WB data for seven

22

proteins, AIFM1, GPX1, NDUFAF4, NDUFS3, FASN, HADHA and HSPA9 indicated the

23

same tendencies in the change of the expression levels as that determined by the SILAC 21 ACS Paragon Plus Environment

Journal of the American Chemical Society

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

Page 22 of 36

1

experiment. These results clearly validated the data of our quantitative MS analysis exploiting

2

MRM-based labeling strategy. Notably, both up-regulation of AIFM1 and down-regulation of

3

NDUFAF4 in the ROS-mediated apoptosis process are identified for the first time, to the best

4

of our knowledge. AIFM1, apoptosis-inducing factor 1, is an NADH oxidase and implicated

5

in a caspase-independent pathway of programed cell death.37 NDUFAF4, NADH

6

dehydrogenase [ubiquinone] 1 alpha subcomplex assembly factor 4 which is involved in the

7

assembly of respiratory chain complex I, is reported to negatively regulates apoptosis.38

8

Importantly, these proteins are supposed to be associated with some of mitochondrial

9

diseases,39 such as neuropathy and Parkinson disease. Thus, the present new finding may

10

encourage us to facilitate a more detailed study on biological roles of these proteins in

11

dysfunctional mitochondria.

12

On the other hand, WB data of ACO2, mitochondrial aconitate hydratase, did not

13

agree with that of the SILAC experiment. To examine this discrepancy, we focused on the

14

labeling sites of ACO2. MS analysis successfully identified Cys385 and Cys451 as the

15

labeling sites, both of which are involved in binding to an iron-sulfur [4Fe-4S] cluster, an

16

active site of ACO2. It is also reported that this [4Fe-4S] cluster is destructed and removed by

17

ROS,40 leading to a conformational change of ACO2 and its subsequent degradation. Given

18

these reports, it is reasonably considered that the removal of [4Fe-4S] cluster from ACO2 by

19

PQ-induced ROS enhances the accessibility (i.e. reactivity) of the Cys385 and Cys451 with

20

MRMs, resulting in an increase of the SILAC ratio despite of a decline of ACO2 expression

21

level. While this example may point out a false positive case of our MRM-based SILAC

22

approach, it also suggests that the careful examination of our data could give us interesting

23

information on the structure and/or activities changes of a mitochondrial protein in living 22 ACS Paragon Plus Environment

Page 23 of 36

1

cells.

2

Overall, these experiments undoubtedly demonstrate that our MRM-based method

3

provides a convenient and quantitative approach to efficiently focus on representative proteins

4

in the dynamic change of mitochondrial proteins, in combination with a standard quantitative

5

proteomic technique.

6 7 (a)

PQ

PQ-treated SILAC “light”

48 h

labeling

Mix lysate" (1 : 1)

IP"

in-gel" digestion

SDS-PAGE

LC-MS/MS

vehicle

Control SILAC “heavy”

(b)

!mitochondrial

PQ

AIFM1!

Up-regulated " (> 2-fold)

2!

(c)

Other Organellar" and un-assigned"

3!

log2 PQ / Control

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

Journal of the American Chemical Society

AIFM1 ACO2"

1!

GPX1 HADHA!

HADHA

0!

HSPA9 HSP9A!

-1!

FASN

FASN!

Down-regulated " (> 2-fold)

NDUFAF4

-3! 0!

50!

WB SILAC

NDUFS3

NDUFS3! NDUFAF4!

9 10 11 12 13 14 15 16 17 18 19 20 21 22

+

ACO2! GPX1!

-2!

8



100!

150!

200!

-3! -3

250!

-2! -2

-1! -1

0! 0

1! 1

2! 2

33!

log2 (PQ/Control)

Protein

Figure 6. Quantitative MS analysis of the change of mitochondrial proteome under an oxidative stress. (a) Schematic illustration of SILAC experiments coupled with MRMs to quantify mitochondrial proteome changes in paraquat (PQ)-treated HeLa cells. PQ-treated cells grown in the light SILAC medium and non-treated cells grown in the heavy SILAC medium (control) were labeled with 1a (CA) and lysed. Protein extracts from the light and heavy cells were mixed in a 1:1 ratio, and then the Et4-Rhod-labeled proteins were enriched by IP using anti-Et4-Rhod antibody, followed by in-gel tryptic digestion and LC-MS/MS analysis. (b) Log2 ratio plot (PQ-treated (light) cells/control (heavy)) of identified proteins. Proteins assigned as mitochondrial are colored in red in the plot. Proteins without mitochondrial annotation are colored in black. (c) Western blotting analysis of PQ-induced changes in expression levels of the selected proteins (n=3 (±standard deviation)) and their comparision with data from SILAC.

23 23 ACS Paragon Plus Environment

Journal of the American Chemical Society

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 24 of 36

Conclusions

2

We have developed a set of MRMs as novel chemical reagents for organelle-focused

3

proteomics. This gave us a remarkably simple experimental protocol of organelle-selective

4

labeling that is compatible with conventional LC-MS/MS and MS-fingerprinting techniques.

5

This method explicitly addressed the chemical reactivity and selectivity of various reactive

6

groups in live cells and identified the mitochondrial proteome in essentially natural biological

7

habitats, including live primary cultured neurons, brain slices and HeLa cells. Quantitative

8

profiling and characterizing particular proteins have also been accomplished using this method,

9

whose expression levels were found to change dynamically during an oxidant-induced

10

apoptosis.

11

The present MRM-based method was able to label and identify a relatively small

12

fraction of the mitochondrial whole proteome that is roughly estimated to contain 1,000–1,500

13

proteins.4, 41 The limited coverage may be partially attributed to the biased submitochondrial

14

distribution of the Et4-Rhod-based MRMs (in the matrix and inner membrane). This issue

15

would be overcome by varying the localizable moiety to target different areas within the

16

mitochondria.42 Moreover, given the modular structures of MRMs, a rational extension of this

17

strategy targets other organelles by varying the localizable moiety to cover a whole proteome

18

in live cells and tissues. We envision that the sets of organelle-localizable reactive molecules

19

can be widely used against a variety of intact cells involved in organelle-relevant diseases to

20

identify molecular details by profiling differentiating proteomes at high resolution.

21 22 23 24 ACS Paragon Plus Environment

Page 25 of 36

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

Journal of the American Chemical Society

1

Experimental Section

2

Synthesis.

3 4

All synthetic procedures and compound characterizations are described in the Supporting Information.

5 6

General materials and methods for the biochemical/biological experiments.

7

Unless otherwise noted, all proteins/enzymes and reagents were obtained from commercial

8

suppliers (Sigma-Aldrich, Tokyo Chemical Industry (TCI), Wako Pure Chemical Industries,

9

Sasaki Chemical, Bio-Rad, or Watanabe Chemical Industries) and used without further

10

purification.

UV-vis

absorption

spectra

were

acquired

on

a

Shimazu

UV-2550

11

spectrophotometer. SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and western

12

blotting were carried out using a Bio-Rad Mini-Protean III electrophoresis apparatus.

13

Fluorescence gel images and chemical luminescent signals using Chemi-Lumi one (Nacalai

14

Tesque) or ECL Prime (GE Healthcare) were acquired with an imagequant LAS4000 (GE

15

Healthcare). Cell imaging was performed with a confocal laser scanning microscope (CLSM,

16

Olympus, FV10i) using 60× objectives, and images were analyzed using accompanying

17

FV10i-ASW 3.0 Viewer software.

18 19

CLSM analysis of HeLa cells

20

HeLa cells (2.0 × 105 cells) were cultured on 35 mm glass based dishes in Dulbecco’s

21

Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1%

22

Antibiotic-Antimycotic (Anti-Anti, gibco) under a humidified atmosphere of 5% CO2 in air at

23

37 °C for 24 h. The cells were washed twice with serum-free DMEM (HEPES modified, no

24

Phenol Red), and incubated in serum-free DMEM containing Et4-Rhod-type MRM (1.0 µM)

25

for 15 min or Hoechst-type MRM (10 µM) for 60 min. The intracellular localization of these

26

reagents was analyzed by CLSM. As a negative control for mitochondrial-localizable reagents,

27

HeLa cells were preincubated with CCCP (5.0 µM) for 5 min before addition of

28

Et4-Rhod-type MRMs.

29 30

Mitochondrial proteome labeling with Et4-Rhod-type MRMs in HeLa cells 25 ACS Paragon Plus Environment

Journal of the American Chemical Society

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

Page 26 of 36

1

HeLa cells (1.0 × 106 cells) were cultured on 10 cm dishes under a humidified atmosphere of

2

5% CO2 in air at 37 °C for 48 h in DMEM. The cells were washed twice with serum-free

3

DMEM, and incubated in serum free DMEM containing Et4-Rhod-type MRM (5–10 µM) for

4

0.5–6 h. In the case of using photoreactive probes, such as 1f(BP) and 1g(PA), the cells were

5

washed twice with chilled PBS, and then exposed to 365 nm light with a handy UV lamp (4

6

W) at 1 cm distance from the top of cell culture dish for 40 min on ice. After labeling reaction,

7

the cells were washed twice with chilled PBS, lysed by sonication in modified-RIPA buffer

8

(25 mM Tris, 150 mM NaCl, 1% Sodium dodecyl sulfate, 1% Nonidet P-40, 1% sodium

9

deoxycholate, pH 7.6) containing 1% protease inhibitor cocktail set III (Novagen) at 4 °C.

10

Proteins were precipitated with chilled acetone, and the pellets were solubilized by sonication

11

in modified-RIPA buffer. The solubilized proteins were normalized for protein concentration

12

(BCA assay kit, Thermo) to 1.0 mg/mL, and then mixed with 1/4 volume of 5 × SDS–PAGE

13

loading buffer (pH 6.8, 325 mM Tris, 20% sucrose, 15% SDS, 0.03% bromophenol blue)

14

containing 250 mM DTT and vortexed for 30 min. The samples were resolved by SDS-PAGE

15

(8.0 µg protein/lane), and the Et4-Rhod labeled proteins were visualized fluorescence gel

16

imager (LAS4000, GE Healthcare).

17 18

Preparation of LC-MS samples (in-solution digested samples).

19

After mitochondrial proteome labeling with Et4-Rhod-type MRMs, the cells were lysed with

20

RIPA buffer (25 mM Tris, 150 mM NaCl, 0.1% Sodium dodecyl sulfate, 1% Nonidet P-40,

21

1% sodium deoxycholate, pH 7.6). All proteins were precipitated with chilled acetone, and

22

solubilized by sonication in modified-RIPA buffer. Protein concentration was determined by

23

BCA assay and normalized to 2.0 mg/mL (total volume 1 mL). The samples were diluted into

24

1 mL of 12 M urea (final conc. 6 M), reduced with 20 mM DTT at 65 °C for 5 min, and

25

alkylated with 40 mM iodoacetamide at 37 °C for 30 min. After precipitation with 20%

26

methanol containing acetone (8 mL), 500 µg of proteins were treated with following two

27

methods (method A or B).

28

In method A, the pellet of proteins was solubilized by sonication in 500 µL of sodium

29

deoxycholate (SDC) buffer (0.1% SDC, 1M urea, 10 mM ammonium bicarbonate), and

30

digested with 5 µg of trypsin (Thermo, MS Grade) overnight at 37 °C. The mixture was boiled 26 ACS Paragon Plus Environment

Page 27 of 36

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

Journal of the American Chemical Society

1

at 65 °C for 5 min to inactivate trypsin. To remove the surfactant, phase-transfer surfactant

2

protocols were conducted.24 Briefly, the digested solution was mixed with 500 µL of ethyl

3

acetate, acidified with 2.5 µL of TFA, and shaken for 1 min. The mixture was centrifuged at

4

16000 × g for 2 min, and the organic phase was removed. The aqueous phase was washed with

5

500 µL of ethyl acetate again, and neutralized with 200 mM Tris-HCl (pH 7.4).

6

In method B, 500 µg of proteins were solubilized by sonication in 250 µL of urea buffer (4M

7

urea, 100 mM Tris-HCl, pH 8.0). The solution was diluted with 250 µL of 100 mM Tris-HCl

8

(pH 8.0), and incubated with 5 µg of trypsin overnight at 37 °C. The mixture was boiled at

9

65 °C for 5 min and diluted with 1.5 mL of Tris-HCl (pH 8.0).

10

The resulting solutions in method A and B were incubated with 5.0 µL of anti-Et4-Rhod

11

antiserum for 6 h at 4 °C with continuous rotation, followed by addition of 50 µL of nProtein

12

A sepharose 4 Fast Flow (GE Healthcare, 50% slurry in PBS) and further rotation at 4 °C for 2

13

h. The beads were washed 5 times with PBS (1 mL) and water (1 mL). The peptides labeled

14

with Et4-Rhod were eluted with 40 µL of 50% water/CH3CN (0.1% TFA) × 2 and 40 µL of

15

CH3CN (0.1% TFA) × 2.These eluents were collected and centrifuged (12000 × g, 5 min). The

16

supernatant was freeze dried, redissolved with 20 µL of 2% CH3CN/water containing 0.1%

17

TFA, and further centrifuged at 16000 × g for 5 min. This supernatant (2–4 µL) was subjected

18

to LC-MS analysis. The list of identified proteins obtained by method A and B were integrated

19

into an output file. Protein identification experiments using HeLa cells and primary cortical

20

neuronal culture were independently performed twice.

21 22

Preparation of LC-MS samples (SILAC samples)

23

HeLa cells were grown in “heavy” SILAC media (DMEM, Thermo) containing 100 µg/mL of

24

13

25

Laboratories)

26

14

27

humidified atmosphere of 5% CO2 in air at 37 °C. The cells were passaged at least five times

28

in the SILAC media before being used for analysis. The isotope-labeled cells (6.0 × 105 cells)

29

were seeded on 10 cm dishes and incubated for 24 h in the SILAC media. The “light” labeled

30

cells were incubated with light SILAC media containing Paraquat (PQ) (100 µM) for 48 h,

C6 ,

15

N2-Lys (Cambridge Isotope Laboratories) and or “light” SILAC media containing

13 12

C6 ,

C6 ,

15 14

N4,-Arg (Cambridge Isotope

N2-Lys (Thermo) and

12

C6 ,

N4,-Arg (Thermo) supplemented with 10% dialyzed FBS (Gibco) and 1% Anti-Anti under a

27 ACS Paragon Plus Environment

Journal of the American Chemical Society

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

Page 28 of 36

1

while the “heavy” labeled cells were treated with heavy SILAC media containing vehicle

2

(water). Both the light and heavy cells were washed twice with PBS, and incubated in serum

3

free SILAC media containing 1a(CA) (5.0 µM) for 6 h. After labeling reaction, the cells were

4

washed twice with chilled PBS, lysed by sonication in modified-RIPA buffer containing 1%

5

protease inhibitor cocktail at 4 °C. Proteins were precipitated with chilled acetone, and the

6

pellets were solubilized by sonication in modified-RIPA buffer. The solubilized proteins were

7

normalized for protein concentration with BCA assay. Equal amounts of protein from heavy

8

(500 µg/250 µL) and light (500 µg/250 µL) samples were mixed in a 1 : 1 ratio, and the

9

mixture were diluted with PBS into 1.0 mL. The samples were precleared against 100 µL of

10

nProtein A sepharose 4 Fast flow at 4 °C for 30 min. After centrifugation, the supernatant was

11

collected and incubated with anti-Et4-Rhod (10.0 µL) at 4 °C overnight with continuous

12

rotation, followed by addition of 100 µL of nProtein A sepharose 4 Fast flow. After incubation

13

at 4 °C for 2 h, the beads were washed 5 times with RIPA buffer (1.0 mL), and then boiled in

14

2× SDS-PAGE sample buffer (25 µL) containing 100 mM DTT for 3 min. The supernatant

15

(15 µL) was loaded to SDS-PAGE. The samples were resolved 1.5 cm by the running gel and

16

fixed with 45% methanol/water containing 5% acetic acid for 20 min. The fixed gel was

17

washed with 50% methanol/water and water, and manually cut into 8 gel slices without gel

18

staining. The excised gels were subjected to in-gel digestion using trypsin (In-Gel Tryptic

19

Digestion Kit, Thermo Fisher Scientific, trypsin 200 ng/slice). The eluted peptide solution

20

from the gel was freeze dried and redissolved with 18 µL of 2% CH3CN/water containing

21

0.1% TFA. The solutions (4.0 µL) were subjected to LC-MS analyses. Three independent

22

SILAC experiments were performed, in which proteins detected at least twice with paired

23

light- and heavy-labeled peptides were included in the datasets as identified proteins.

24 25

LC-MS analysis

26

After proteome labeling, proteins were digested with trypsin in solution or in gel. The samples

27

for protein identification were analyzed by nano-flow reverse phase liquid chromatography

28

followed by tandem MS, using a LTQ-Orbitrap XL hybrid mass spectrometer (Thermo Fisher

29

Scientific). For LTQ-Orbitrap XL, a capillary reverse phase HPLC-MS/MS system composed

30

of an Agilent 1100 series gradient pump equipped with Valco C2 valves with 150-µm ports, 28 ACS Paragon Plus Environment

Page 29 of 36

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

Journal of the American Chemical Society

1

and LTQ-Orbitrap XL hybrid mass spectrometer equipped with an XYZ nano-electrospray

2

ionization (NSI) source (AMR). Samples were automatically injected using PAL system (CTC

3

analytics, Zwingen, Switzerland) into a peptide L-trap column OSD (5 µm, AMR) attached to

4

an injector valve for desalinating and concentrating peptides. After washing the trap with

5

MS-grade water containing 0.1% trifluoroacetic acid and 2% acetonitrile (solvent C), the

6

peptides were loaded into a nano HPLC capillary column (C18-packed with the gel particle

7

size of 3 µm, 0.1 × 150 mm, Nikkyo Technos) by switching the valve. The eluents used were:

8

A, 100% water containing 0.1% formic acid, and B, 80% acetonitrile containing 0.1% formic

9

acid. The column was developed at a flow rate of 200–500 nl/min with the concentration

10

gradient of acetonitrile. Xcalibur 2.1 system (Thermo Fisher Scientific) was used to record

11

peptide spectra over the mass range of m/z 350–1500, and MS/MS spectra in

12

information-dependent data acquisition over the mass range of m/z 150–2000. Repeatedly,

13

MS spectra were recorded followed by three data-dependent collision induced dissociation

14

(CID) MS/MS spectra generated from three or eight highest intensity precursor ions for the

15

non-quantitative MS or SILAC experiment, respectively. Multiple charged peptides were

16

chosen for MS/MS experiments due to their good fragmentation characteristics. MS/MS

17

spectra were interpreted and peak lists were generated by Proteome Discoverer 1.4 (Thermo

18

Fisher Scientific). Searches were performed by using the SEQUEST-HT (Thermo Fisher

19

Scientific)

20

carbamidomethylation on cysteine and oxidation on methionine were searched with peptide

21

mass tolerance at 10 or 20 ppm for non-quantitative MS or SILAC experiment, respectively;

22

MS/MS mass tolerance of 0.6 Da. Multiple Et4-Rhod modified peptides were omitted from

23

the datasets. The resultant data set was filtered to a maximum false discovery rate (FDR) of

24

0.01. Identified peptides with same sequence were counted as the identical peptides.

against

uniprot

database.

Dynamic

modifications

of

Et4-Rhod,

25 26 27

Acknowledgements

28

We thank Nozomi Ogawa, Karin Nishimura and Youhei Tatsukami (Kyoto University) for

29

experimental support of LC-MS and helpful discussion. We also thank Sho Wakayama (Kyoto

30

University) for preparation of brain slices. Y.Y. (26-4745) is supported by Japan Society for 29 ACS Paragon Plus Environment

Journal of the American Chemical Society

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

Page 30 of 36

1

the Promotion of Science (JSPS) for doctoral fellowships. This work was funded by the Japan

2

Science and Technology Agency (JST) Core Research for Evolutional Science and

3

Technology (CREST) and by JSPS KAKENHI (15H01637) to I.H.

4 5

Competing financial interests

6

The authors declare no competing financial interests.

7 8 9

30 ACS Paragon Plus Environment

Page 31 of 36

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

Journal of the American Chemical Society

References

2 3

(1)

4 5

(a) Yates, J. R. J. Am. Chem. Soc. 2013, 135, 1629. (b) Aebersold, R.; Mann, M. Nature 2003, 422, 198. (c) Iwasaki, M.; Ishihama, Y. Chromatography 2014, 35, 73.

(2)

(a) Weerapana, E.; Wang, C.; Simon, G. M.; Richter, F.; Khare, S.; Dillon, M. B. D.;

6

Bachovchin, D. a; Mowen, K.; Baker, D.; Cravatt, B. F. Nature 2010, 468, 790. (b)

7

Woo, C. M.; Iavarone, A. T.; Spiciarich, D. R.; Palaniappan, K. K.; Bertozzi, C. R. Nat.

8

Methods 2015, 12, 561. (c) Hayashi, T.; Sun, Y.; Tamura, T.; Kuwata, K.; Song, Z.;

9

Takaoka, Y.; Hamachi, I. J. Am. Chem. Soc. 2013, 135, 12252. (d) Li, X.; Foley, E. a;

10

Molloy, K. R.; Li, Y.; Chait, B. T.; Kapoor, T. M. J. Am. Chem. Soc. 2012, 134, 1982.

11

(e) Lin, S.; Zhang, Z.; Xu, H.; Li, L.; Chen, S.; Li, J.; Hao, Z.; Chen, P. R. J. Am. Chem.

12

Soc. 2011, 133, 20581.

13

(3)

(a) Liu, Y.; Patricelli, M. P.; Cravatt, B. F. Proc. Natl. Acad. Sci. U.S.A. 1999, 96,

14

14694. (b) Cravatt, B. F.; Wright, A. T.; Kozarich, J. W. Annu. Rev. Biochem. 2008, 77,

15

383.

16

(4)

Nunnari, J.; Suomalainen, A. Cell 2012, 148, 1145.

17

(5)

Drissi, R.; Dubois, M. L.; Boisvert, F. M. FEBS J. 2013, 280, 5626.

18

(6)

(a) Rhee, H.-W.; Zou, P.; Udeshi, N. D.; Martell, J. D.; Mootha, V. K.; Carr, S. A.;

19

Ting, A. Y. Science 2013, 339, 1328. (b) Hung, V.; Zou, P.; Rhee, H.-W.; Udeshi, N.

20

D.; Cracan, V.; Svinkina, T.; Carr, S. A.; Mootha, V. K.; Ting, A. Y. Mol. Cell 2014,

21

55, 332.

22

(7)

Most recently, APEX strategy has been successfully applied to Drosophila tissue

23

samples: Chen, C.L.; Hu, Y.; Udeshi, N. D.; Lau, T. Y.; Wirtz-Peitz, F.; He, L.; Ting, A.

24

Y.; Carr, S. A; Perrimon, N. Proc. Natl. Acad. Sci. U. S. A. 2015, 112, 12093.

25

(8)

Washbourne, P.; McAllister, a. K. Curr. Opin. Neurobiol. 2002, 12, 566.

26

(9)

Ward, J. F.; Evans, J. W.; Limoli, C. L.; Calabro-Jones, P. M. Br. J. Cancer. Suppl.

27 28

1987, 8, 105. (10)

(a) Wiedner, S. D.; Anderson, L. N.; Sadler, N. C.; Chrisler, W. B.; Kodali, V. K.;

29

Smith, R. D.; Wright, A. T. Angew. Chemie Int. Ed. 2014, 53, 2919. (b) Yang, P.-Y.;

30

Liu, K.; Zhang, C.; Chen, G. Y. J.; Shen, Y.; Ngai, M. H.; Lear, M. J.; Yao, S. Q. Chem.

31

- An Asian J. 2011, 6, 2762. 31 ACS Paragon Plus Environment

Journal of the American Chemical Society

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

Page 32 of 36

1

(11)

Yasueda, Y.; Tamura, T.; Hamachi, I. Chem. Lett. 2016. DOI: 10.1246/cl.151083.

2

(12)

(a) Sletten, E. M.; Bertozzi, C. R. Angew. Chem. Int. Ed. 2009, 48, 6974. (b) Takaoka,

3 4

Y.; Ojida, A.; Hamachi, I. Angew. Chem. Int. Ed. 2013, 52, 4088. (13)

(a) Hoye, A. T.; Davoren, J. E.; Wipf, P.; Fink, M. P.; Kagan, V. E. Acc. Chem. Res.

5

2008, 41, 87. (b) Rin Jean, S.; Tulumello, D. V; Wisnovsky, S. P.; Lei, E. K.; Pereira,

6

M. P.; Kelley, S. O. ACS Chem. Biol. 2014, 9, 323. (c) Wu, S.; Song, Y.; Li, Z.; Wu, Z.;

7

Han, J.; Han, S. Anal. Methods 2012, 4, 1699.

8

(14)

9 10

(a) Weerapana, E.; Simon, G. M.; Cravatt, B. F. Nat. Chem. Biol. 2008, 4, 405. (b) Son, J.; Lee, J.; Lee, J.; Schu, A.; Chang, Y. ACS Chem. Biol. 2010, 5, 449.

(15)

Sadler, N. C.; Melnicki, M. R.; Serres, M. H.; Merkley, E. D.; Chrisler, W. B.; Hill, E.

11

A.; Romine, M. F.; Kim, S.; Zink, E. M.; Datta, S.; Smith, R. D.; Beliaev, A. S.;

12

Konopka, A.; Wright, A. T. ACS Chem. Biol. 2014, 9, 291.

13

(16)

(a) Wang, H.; Koshi, Y.; Minato, D.; Nonaka, H.; Kiyonaka, S.; Mori, Y.; Tsukiji, S.;

14

Hamachi, I. J. Am. Chem. Soc. 2011, 12220. (b) Greenwald, R. B.; Pendri, A; Martinez,

15

A; Gilbert, C.; Bradley, P. Bioconjug. Chem. 1996, 7, 638.

16

(17)

Shannon, D. A.; Weerapana, E. Curr. Opin. Chem. Biol. 2015, 24, 18.

17

(18)

(a) Gu, C.; Shannon, D. A.; Colby, T.; Wang, Z.; Shabab, M.; Kumari, S.; Villamor, J.

18

G.; McLaughlin, C. J.; Weerapana, E.; Kaiser, M.; Cravatt, B. F.; van der Hoorn, R. A.

19

L. Chem. Biol. 2013, 20, 541. (b) Narayanan, A.; Jones, L. H. Chem. Sci. 2015, 6, 2650.

20

(19)

21

Sakurai, K.; Ozawa, S.; Yamada, R.; Yasui, T.; Mizuno, S. ChemBioChem 2014, 15, 1399.

22

(20)

Emaus, R. K.; Grunwald, R.; Lemasters, J. J. Biochim Biophys Acta, 1986, 850, 436.

23

(21)

Gao, W.; Pu, Y.; Luo, K. Q.; Chang, D. C. J. Cell Sci. 2001, 114, 2855.

24

(22)

Campanella, M.; Seraphim, A.; Abeti, R.; Casswell, E.; Echave, P.; Duchen, M. R.

25 26

Biochim. Biophys. Acta. 2009, 1787, 393. (23)

27 28

The selectivity and enrichment efficiency of our homemade anti-Et4-Rhod antibody (antiserum) were revealed in Figure S8.

(24)

(a) Adam, G. C.; Burbaum, J.; Kozarich, J. W.; Patricelli, M. P.; Cravatt, B. F. J. Am.

29

Chem. Soc. 2004, 126, 1363. (b) Masuda, T.; Tomita, M.; Ishihama, Y. J. Proteome Res.

30

2008, 7, 731. 32 ACS Paragon Plus Environment

Page 33 of 36

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

(25)

2 3

Journal of the American Chemical Society

Kiefer, F.; Arnold, K.; Künzli, M.; Bordoli, L.; Schwede, T. Nucleic Acids Res. 2009, 37, 387.

(26)

Since the MRM-based labeling follows the intermolecular reaction mechanism,

4

high-abundance proteins tend to be more labeled and enriched, while low-abundance

5

proteins can also be labeled enough to be identified (See Figure S11).

6

(27)

7 8

(See reference 20). (28)

9 10

It is reported that Rhodamine derivatives mainly distribute into the mitochondria matrix

Marini, E. S.; Giampietri, C.; Petrungaro, S.; Conti, S.; Filippini, A; Scorrano, L.; Ziparo, E. Cell Death Differ. 2014, 22, 1131.

(29)

11

Zhang, H.; Wang, Y.; Li, J.; Yu, J.; Pu, J.; Li, L.; Zhang, H.; Zhang, S.; Peng, G.; Yang, F.; Liu, P. J. Proteome Res. 2011, 10, 4757.

12

(30)

Proteins identified as “non-mitochondrial proteins” include 25 cytoskeletal proteins.

13

(31)

(a) Calvo, S. E.; Clauser, K. R.; Mootha, V. K. Nucleic Acids Res. 2015 [doi:

14

10.1093/nar/gkv1003]. (b) Pagliarini, D. J.; Calvo, S. E.; Chang, B.; Sheth, S. A; Vafai,

15

S. B.; Ong, S.-E.; Walford, G. A; Sugiana, C.; Boneh, A.; Chen, W. K.; Hill, D. E.;

16

Vidal, M.; Evans, J. G.; Thorburn, D. R.; Carr, S. A; Mootha, V. K. Cell 2008, 134,

17

112.

18

(32)

Huang, D. W.; Sherman, B. T.; Lempicki, R. A., Nat. Protoc. 2009, 4, 44.

19

(33)

(a) Ong, S. E.; Mann, M. Nat. Protoc. 2006, 1, 2650. (b) Bantscheff, M.; Schirle, M.;

20 21

Sweetman, G.; Rick, J.; Kuster, B. Anal. Bioanal. Chem. 2007, 389, 1017. (34)

(a) Fei, Q.; McCormack, A. L.; Di Monte, D. A.; Ethell, D. W. J. Biol. Chem. 2008,

22

283, 3357. (b) Takeyama, N.; Tanaka, T.; Yabuki, T.; Nakatani, T. Int. J. Toxicol. 2004,

23

23, 33.

24

(35)

(a) Park, K. J.; Kim, Y. J.; Kim, J.; Kim, S. M.; Lee, S. Y.; Bae, J. W.; Hwang, K. K.;

25

Kim, D. W.; Cho, M. C. Korean Circ. J. 2012, 42, 23. (b) Lu, W.; Fu, Z.; Wang, H.;

26

Feng, J.; Wei, J.; Guo, J. Mol. Cell. Biochem. 2014, 387, 261. (c) De Haan, J. B.;

27

Bladier, C.; Griffiths, P.; Kelner, M.; O’Shea, R. D.; Cheung, N. S.; Bronson, R. T.;

28

Silvestro, M. J.; Wild, S.; Zheng, S. S.; Beart, P. M.; Hertzog, P. J.; Kola, I. J. Biol.

29

Chem. 1998, 273, 22528.

33 ACS Paragon Plus Environment

Journal of the American Chemical Society

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

(36)

Page 34 of 36

(a) Choi, J. W.; Song, M.-Y.; Park, K.-S. Mol. Biosyst. 2014, 10, 1940. (b) Ricci, J. E.;

2

Muñoz-Pinedo, C.; Fitzgerald, P.; Bailly-Maitre, B.; Perkins, G. A.; Yadava, N.;

3

Scheffler, I. E.; Ellisman, M. H.; Green, D. R. Cell 2004, 117, 773. (c) Impheng, H.;

4

Pongcharoen, S.; Richert, L.; Pekthong, D.; Srisawang, P. PLoS One 2014, 9, e107842.

5

(37)

6 7

Hangen, E.; Blomgren, K.; Bénit, P.; Kroemer, G.; Modjtahedi, N. Trends Biochem. Sci. 2010, 35, 278.

(38)

(a) Karp, C. M.; Pan, H.; Zhang, M.; Buckley, D. J.; Schuler, L. A.; Buckley, A. R.

8

Cancer Res. 2004, 64, 1016. (b) Karp, C. M.; Shukla, M. N.; Buckley, D. J.; Buckley,

9

A. R. Oncogene 2007, 26, 1780.

10

(39)

(a) Keane, P. C.; Kurzawa, M.; Blain, P. G.; Morris, C. M. Parkinsons. Dis. 2011, 2011,

11

716871. (b) Keeney, P. M. J. Neurosci. 2006, 26, 5256. (c) Mckenzie, M.; Ryan, M. T.

12

IUBMB Life 2010, 62, 497. (c) Mimaki, M.; Wang, X.; McKenzie, M.; Thorburn, D. R.;

13

Ryan, M. T. Biochim. Biophys. Acta-Bioenergetics 2012, 1817, 851.

14

(40)

15

(a) Bota, D. A; Davies, K. J. A. Nat. Cell Biol. 2002, 4, 674. (b) Goard, C. A; Schimmer, A. D. Oncogene 2014, 33, 2690.

16

(41)

Calvo, S. E.; Mootha, V. K. Annu. Rev. Genomics Hum. Genet. 2010, 11, 25.

17

(42)

Notably, the mitochondrial-surface specific fluorescent probe was recently developed

18

by Uesugi and coworkers: Kawazoe, Y.; Shimogawa, H.; Sato, A.; Uesugi, M. Angew.

19 20

Chem. Int. Ed. 2011, 50, 5478.

34 ACS Paragon Plus Environment

Page 35 of 36

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 2

Journal of the American Chemical Society

TOC

3

35 ACS Paragon Plus Environment

Journal of the American Chemical Society

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

Table of contents 82x26mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 36 of 36