http://www.elsevier.com/locate/bba

Biochimica et Biophysica Ac

Proteomic analysis of succinate dehydrogenase and ubiquinol-cytochrome

c reductase (Complex II and III) isolated by immunoprecipitation

from bovine and mouse heart mitochondria

Birgit Schillinga,1, James Murrayb,c,1, Chris B. Yooa, Richard H. Rowa, Michael P. Cusacka,

Roderick A. Capaldib,c, Bradford W. Gibsona,d,*

aBuck Institute for Age Research, Novato, CA 94945, USAbDepartment of Molecular Biology, University of Oregon, Eugene, OR 97403, USA

cMitoSciences LLC, Eugene, OR 97403, USAdDepartment of Pharmaceutical Chemistry, University of California, San Francisco, CA 94143, USA

Received 28 May 2005; accepted 12 July 2005

Available online 3 August 2005

Abstract

The oxidative phosphorylation system (OXPHOS) consists of five multi-enzyme complexes, Complexes I–V, and is a key component of

mitochondrial function relating to energy production, oxidative stress, cell signaling and apoptosis. Defects or a reduction in activity in

various components that make up the OXPHOS enzymes can cause serious diseases, including neurodegenerative disease and various

metabolic disorders. Our goal is to develop techniques that are capable of rapid and in-depth analysis of all five OXPHOS complexes. Here,

we describe a mild, micro-scale immunoisolation and mass spectrometric/proteomic method for the characterization of Complex II (succinate

dehydrogenase) and Complex III (ubiquinol-cytochrome c reductase) from bovine and rodent heart mitochondria. Extensive protein sequence

coverage was obtained after immunocapture, 1D SDS PAGE separation and mass spectrometric analysis for a majority of the 4 and 11

subunits, respectively, that make up Complexes II and III. The identification of several posttranslational modifications, including the covalent

FAD modification of flavoprotein subunit 1 from Complex II, was possible due to high mass spectrometric sequence coverage.

D 2005 Elsevier B.V. All rights reserved.

Keywords: Mitochondria; Succinate dehydrogenase; Ubiquinol-cytochrome c reductase; Posttranslational modification; Neurodegenerative disease

1. Introduction

The contribution of mitochondria to a variety of

metabolic, neurodegenerative, and age-related diseases is

still controversial, but evidence implicating mitochondrial

dysfunction in these and other pathologies has increased

significantly in recent years. Mitochondrial dysfunction has

multiple effects, initiated as reduced energy production, that

results in the formation of reactive oxygen and nitrogen

0925-4439/$ - see front matter D 2005 Elsevier B.V. All rights reserved.

doi:10.1016/j.bbadis.2005.07.003

* Corresponding author. Buck Institute for Age Research, 8001 Redwood

Blvd., Novato, CA 94945, USA. Tel.: +1 415 209 2032; fax: +1 415 209

2231.

E-mail address: bgibson@buckinstitute.org (B.W. Gibson).1 These authors contributed equally to this work.

species [1], altered bioenergetic interactions within the

organelle, and subsequent changes in cellular homeostasis

that can lead to apoptosis or necrotic cell death [2].

The five enzyme complexes that constitute the mito-

chondrial oxidative phosphorylation system (OXPHOS)

contain over 90 different protein subunits, making direct

and specific correlations of these to disease cause or

progression difficult at best. High-resolution structures are

available for all the OXPHOS complexes except Complex I

(NADH:ubiquininone oxidoreductase), which contains the

largest number of subunit proteins (¨46), and has a MW of

950,000 Da. Mutations in nuclear (and mitochondrial) genes

of the OXPHOS complexes can cause specific human

diseases and these have been reviewed recently by van den

Heuvel and Smeitink [3]. Those in Complexes II and III

ta 1762 (2006) 213 – 222

B. Schilling et al. / Biochimica et Biophysica Acta 1762 (2006) 213–222214

have been associated with clinical phenotypes such as optic

atrophy, tumor formation, myopathy, and encephalopathy

depending on the specific subunits involved [4,5].

OXPHOS defects are also observed in a number of late

onset diseases, such as Parkinson’s disease [6], Huntington’s

disease [7], and amyotrophic lateral sclerosis [8]. In these, it

is thought that certain polymorphisms in the genes for the

proteins of the OXPHOS system predispose to higher levels

of accumulated environmental insult thereby increasing

oxidative stress which, in turn, triggers cell dysfunction

and ultimately cell death. However, a mitochondrial

etiology of the above neurodegenerative diseases is not

universally accepted.

To better investigate the role of the mitochondrial

OXPHOS complexes in mitochondrial disease, better

methods are needed to rapidly purify, isolate and character-

ize the proteins that constitute this system. As a first

approach, we have used a sucrose gradient fractionation

centrifugation method to isolate and partially separate the

OXPHOS complexes from human heart and mouse tissue

(Hanson et al. [9] and Taylor et al. [10]). This approach led

to an initial proteomic analysis of the components in each

complex as well as other proteins of the mitochondrion.

More recently, our laboratories have focused on immu-

noisolation as a simple method for the one-step purification

of the OXPHOS complexes. Isolation of Complex I from

human heart [11] and from rodent tissue [12], by this

method has been combined with mass spectrometric

methods for the protein identification and to evaluate

sequence coverage of the individual polypeptides resolved

as a necessary prerequisite for post-translational modifica-

tion studies. Analogous methods have also been demon-

strated for the isolation of human Complex V (ATP

synthase) [13] and the pyruvate dehydrogenase complex

[14]. Moreover, an immunoisolation protocol for bovine

cytochrome c oxidase (Complex IV) has been developed

and the enzyme studied extensively by mass spectrometry

(Murray and Capaldi, unpublished data).

In this current work, we describe a rapid and mild

isolation and characterization of intact Complexes II and III

from both bovine and rodent tissue followed by detailed

mass spectrometric-based proteomic analysis. Both

MALDI-MS peptide mass fingerprinting (PMF) and tandem

mass spectrometry (nano-HPLC-MS, MS/MS) were used to

identify and characterize individual protein subunits from

both Complexes II and III, and this has led to the

identification of several posttranslational modifications.

2. Materials and methods

2.1. Materials

All tissue culture materials were procured from Life

Technologies/Invitrogen (Carlsbad, CA) or Cellgro (Kansas

City, MO). Protease inhibitor cocktail was obtained from

Roche Diagnostics (Indianapolis, IN). Antibodies coupled

to protein G agarose beads for OXPHOS complex isolation

were purchased from MitoSciences (Eugene, OR). NuPAGE

4–12% Bis-Tris (MES) gels were obtained from Invitrogen

(Carlsbad, CA). Materials related to proteomics, such as

sample buffers, 1D SDS PAGE gels were obtained from

Bio-Rad Laboratories (Hercules, CA). Gel stains, such as

Coomassie Brilliant Blue R was purchased from Sigma (St.

Louis, MO). For proteolysis, sequencing grade, modified

trypsin (porcine) was purchased from Promega (Madison,

WI). Non-ionic detergent, n-dodecyl-h-d-maltoside, and

additional reagents for protein chemistry including iodoa-

cetamide and dithiothreitol were obtained from Sigma (St.

Louis, MO). HPLC solvents such as acetonitrile and water

were obtained from Burdick and Jackson (Muskegon, MI).

For immunoprecipitation experiments, protein G agarose

beads and antibody crosslinking reagent, dimethylpimeli-

midate, were purchased from Sigma (St. Louis, MO). For

MALDI-MS a matrix solution of a-cyano-4-hydroxycin-

namic acid in acetonitrile/methanol was purchased from

Agilent Technologies (Palo Alto, CA).

2.2. Preparation of bovine and mouse heart mitochondria

Bovine heart mitochondria were prepared essentially as

described by Smith [15]. Briefly, ventricles in isotonic

buffer (10 mM Tris–HCl pH 7.8, 0.25 M sucrose, 0.2 mM

EDTA, 0.5 mM PMSF) were homogenized in an ultraturrex

blender followed by a polytron tissue disruptor. Particulate

material was removed filtering through cheesecloth fol-

lowed by centrifugation at 1000�g. Mitochondria were

collected from the supernatant by spinning down at

12,000�g and resuspended in the isotonic buffer. Mouse

heart mitochondria were prepared by dissecting 5 mouse

hearts followed by homogenization in 10 mL of the same

isotonic buffer in a Potter–Elvehjem homogenizer. Partic-

ulate material was removed by centrifugation at 1000�g.

Mitochondria were then isolated by centrifugation at

12,000�g. Protein concentration was determined by the

BCA method (Pierce).

2.3. Immunocapture of Complex II and Complex III using

antibody-coupled protein G agarose beads

A monoclonal antibody against Complex II (4H12B-

G12AG2) was coupled to protein G agarose beads for the

capture of Complex II (now available through MitoScien-

ces, MS201). A monoclonal antibody against Complex III

(1A11BC12AB9) was coupled to protein G agarose beads

for the capture of Complex III (now available through

MitoSciences, MS301). Immunocapture was performed as

previously described [16]. Briefly, 2.5 mg of mitochondria

in 0.5 mL PBS were solubilized by addition of 20 mM

dodecyl-h-d-maltoside (Calbiochem) and incubation for 30

min on ice. Insoluble material was removed by centrifuga-

tion at 70,000�g for 30 min. The recovered supernatant

B. Schilling et al. / Biochimica et Biophysica Acta 1762 (2006) 213–222 215

was incubated with 10 AL antibody-conjugated beads

overnight at 4 -C while turning. Beads were washed 3

times in 1 mL PBS, 0.1 mM dodecyl-h-d-maltoside.

Immunocaptured protein complexes were eluted by addi-

tion of 40 AL SDS-PAGE sample buffer (50 mM Tris pH

6.8, 2% SDS, 10% glycerol, 0.001% bromophenol blue).

After the elution and discarding of the beads, the super-

natant was supplemented with 50 mM DTT before

electrophoresis. For electrophoresis, a 20-AL sample was

resolved by SDS-PAGE on a NuPAGE 4–12% Bis-Tris gel

(Invitrogen) in an X-Cell Sure Lock Mini Cell Apparatus

(Invitrogen) using MES running buffer (50 mM 2-

morpholinoethanesulfonic acid, 50 mM Tris base, 0.1%

SDS, 1 mM EDTA) at 100 V for 1.5 to 2 h. Alternatively,

gels were run on a Tris–HCl 10–22% acrylamide gel

according to Laemmli [17]. All SDS gels were fixed with

10% methanol, 7% acetic acid for 30 min, and subse-

quently stained with Coomassie Brilliant Blue followed by

destaining in 10% methanol/7% acetic acid.

2.4. In-gel tryptic digestion of mitochondrial proteins

Protein spots of interest were manually excised and

processed with an automatic in-gel digester Robot, ProGest

(Genomic Solutions, Ann Arbor, MI). The gel spots were

destained and dehydrated with acetonitrile. Proteins con-

tained within these gel spots were reduced with 10 mM DTT

at 60 -C for 30 min, alkylated with 100 mM iodoacetamide

(37 -C, 45 min) and proteolyzed by adding 125–250 ng

sequencing grade trypsin (Promega, Madison, WI) at 37 -Cfor 4 h. The resulting tryptic peptides were then extracted

from the gel by aqueous/10% formic acid extraction and

analyzed by mass spectrometry. In addition to tryptic

digestion, gel bands were separately proteolyzed with

chymotrypsin after reduction and alkylation (125 ng chymo-

trypsin, 25 -C, 4 h). The resulting chymotryptic peptides were

extracted with 50% acetonitrile/5% formic acid, concentrated

to near dryness by vacuum centrifugation (Savant) and

analyzed by mass spectrometry as described below.

2.5. Mass spectrometry

Mass spectra of digested gel spots were obtained by

matrix-assisted laser desorption ionization time-of-flight

(MALDI-TOF) mass spectrometry on a Voyager DESTR

plus (Applied Biosystems, Framingham, MA). All mass

spectra were acquired in positive-ionization mode using

reflectron optics. The instrument was equipped with a 337

nm nitrogen laser and operated under delayed extraction

conditions; delay time 190 ns, grid voltage 66–70% of full

acceleration voltage (20–25 kV). All peptide samples were

prepared using a matrix solution consisting of 33 mM a-

cyano-4-hydroxycinnamic acid in acetonitrile/methanol (1/

1; v/v); 1 AL of analyte (0.1–1 pmol of material) was mixed

with 1 AL of matrix solution, and then air-dried at room

temperature on a stainless steel target. Typically, 50–100

laser shots were used to record each spectrum. The obtained

mass spectra were externally calibrated with an equimolar

mixture of angiotensin I, ACTH 1–17, ACTH 18–39, and

ACTH 7–38.

All proteolytic peptide extracts were analyzed by reverse-

phase nano-HPLC-MS/MS. Briefly, peptides were separated

on an Ultimate nanocapillary HPLC system equipped with a

PepMapi C18 nano-column (75 Am I.D.�15 cm) (Dio-

nex, Sunnyvale, CA) and CapTrap Micro guard column (0.5

AL bed volume, Michrom, Auburn, CA). Peptide mixtures

were loaded onto the guard column and washed with the

loading solvent (H2O/0.05% formic acid, 20 AL/min) for 5

min, transferred onto the analytical C18-nanocapillary HPLC

column and then eluted at a flow rate of 300 nl/min using

the following gradient: 2% B (from 0 to 5 min), and 2–70%

B (from 5–55 min). Solvent A consisted of 0.05% formic

acid in 98% H2O/2% acetonitrile and solvent B consisted of

0.05% formic acid in 98% acetonitrile/2% H2O. The column

eluant was directly coupled to a FQSTAR Pulsar i_ quadru-pole orthogonal TOF mass spectrometer (MDS Sciex,

Concorde, Canada) equipped with a Protana/ProXeon

nanospray ion source (ProXeon Biosystems, Odense, Den-

mark). The nanospray needle voltage was typically 2300 V

in the HPLC-MS mode. Mass spectra (ESI-MS) and tandem

mass spectra (ESI-MS/MS) were recorded in positive-ion

mode with a resolution of 12,000–15,000 FWHM. For

collision-induced dissociation tandem mass spectrometry,

the mass window for precursor ion selection of the quadru-

pole mass analyzer was set to T1 m/z. The precursor ions

were fragmented in a collision cell using nitrogen as the

collision gas. All ESI-MS and MS/MS spectra were

externally calibrated in static nanospray mode using MS/

MS fragment-ions of a renin peptide standard (His

immonium-ion with m/z at 110.0713, and b8-ion with m/z

at 1028.5312) providing a mass accuracy of �50 ppm.

2.6. Database searches

Mass spectrometric data were analyzed with licensed

Mascot (Matrix Sciences, London, UK) bioinformatics

database search engine [18]. Before submission, MALDI-

MS data were first processed with Mascot Wizard v. 1.1.2

(Matrix Sciences, London, UK). The MALDI-MS Peptide

Mass Fingerprint (PMF) data were matched against peptides

from known protein sequences searching in-house custom-

designed databases (for details, see below) using the

following parameters: 100 ppm mass accuracy, 2 missed

proteolytic cleavages allowed (data were internally cali-

brated, i.e., using trypsin autolysis masses, such as m/z

842.5100 and 2211.1046). In all cases, proteolytic digestion

extracts of proteins were also analyzed by nano-HPLC-ESI-

MS and MS/MS. For ESI-MS/MS data sets, spectra were

submitted using Mascot Daemon to our in-house Mascot

server searching in-house, custom-designed databases.

Mascot uses a probability based FMowse Score_ to evaluate

data obtained from tandem mass spectra, e.g., for a score

Table 1

Proteins identified by ESI-MS/MS from immunocaptured bovine heart Complex II

Complex II

protein subunitaTheor. No. tryptic

peptides

Trypsin sequence

coverage (M score)bNo. chymo.

peptides

Chymo. sequence

coverage (M score)bTotal sequence

coverage (%)MW�10�3 pIc

C II-1 68.0 6.4 94 53% (774) 18 17% (226) 60

C II-2 29.4 8.5 72 73% (751) 4d 6% (28) 76

C II-3 15.3 9.2 9 34% (94) 1d 8% (18) 35

C II-4 11.1 7.8 3 16% (37) – – 16

a Complex II protein subunits (descriptive subunit names): II-1 (Fp subunit), II-2 (Ip subunit), II-3 (cytochrome b560), and II-4 (subunit d).b Amino acid sequence coverage of mature (processed) proteins in %, Mascot database search score according to Perkins et al. [18].c Molecular weight and pI were calculated for the mature protein subunits, i.e., after processing of the precursor protein sequences (if applicable) that

contained mitochondrial import signals.d MS/MS spectrum and fragmentation pattern was inspected ‘‘manually’’ and evaluated.

B. Schilling et al. / Biochimica et Biophysica Acta 1762 (2006) 213–222216

>37, protein matches are considered significant [18]. For

LC-MS/MS data, a minimum of two peptides were required

to confirm protein identification; in the few cases where

only one peptide per protein was selected for MS/MS, the

MS/MS spectrum was inspected ‘‘manually’’ using similar

criteria for evaluation as published by Link et al. [19], and

thus confirmed or deleted from the identification list

(manual inspection is annotated in the identification Tables

1–4). This was particularly important for selected chymo-

trypsin digestion, as chymotryptic peptides do not always

obtain a preferable search engine score due to their atypical

fragmentation pattern.

To display the mass spectrometric data (ESI-MS/MS

results), a Java program was developed, ‘‘SeqDisp’’, as

previously described [12]. SeqDisp is used to graphically

display the observed protein sequence coverage of peptides

obtained from different proteolytic digestions performed per

protein subunit, as well as to calculate overall protein

sequence coverage.

2.7. Custom-designed databases and blast searches/

sequence alignment tools

Custom-designed protein databases for bovine and

mouse Complexes II and III were constructed and incorpo-

rated into our in-house Mascot server to enable more in

depth searches (for a complete list of accession numbers for

bovine and mouse databases see Supplementary Table S1).

Only mature OXPHOS complex protein subunits (if

Table 2

Proteins identified by ESI-MS/MS from immunocaptured mouse heart Complex

Complex II

protein subunitaTheor. No. tryptic

peptides

Tryptic sequ

coverage (MMW�10�3 pIc

C II-1 69.0 6.3 118 74% (1434)

C II-2 29.5 8.7 62 71% (723)

C II-3 15.2 9.6 7 26% (74)

C II-4 11.0 7.8 – –

a Complex II protein subunits (descriptive subunit names): II-1 (Fp subunit), IIb Amino acid sequence coverage of mature (processed) proteins in %, Mascotc Molecular weight and pI were calculated for the mature protein subunits, i.

contained mitochondrial import signals as identified by homology to the mature

applicable) were entered into the databases, i.e., after

processing of the precursor protein sequences that contained

mitochondrial import signals as identified by homology to

the mature bovine sequences. For homology studies,

sequence alignment calculations were performed with the

program ‘‘SIM Alignment Tool’’ [20] using the following

parameters: comparison matrix: BLOSUM62; number of

alignments computed: 20; gap open penalty: 12; gap

extension penalty: 4.

3. Results

In this study, intact and functional Complex II and

Complex III were isolated from bovine and mouse heart

mitochondria using a one-step purification method [16].

After immunoprecipitation of the intact complexes, the

individual subunits were separated by 1D SDS-PAGE,

visualized by Coomassie Blue staining, and select bands

were subjected to mass spectrometric-based proteomic

analysis.

3.1. Immunocapture of Complexes II and III from bovine

and mouse tissue

The monoclonal antibodies MS201 and MS301 (Mito-

Sciences) were used to immuno-capture and purify Com-

plexes II and III, respectively, from bovine heart and mouse

heart mitochondria. The results of these immunocapture

II

ence

score)bNo. chymo.

peptides

Chymo. sequence

coverage (M score)bTotal sequence

coverage (%)

32 36% (315) 87

17 45% (135) 80

11 25% (66) 46

– – –

-2 (Ip subunit), II-3 (cytochrome b560), and II-4 (subunit d).

database search score according to Perkins et al. [18].

e., after processing of the precursor protein sequences (if applicable) that

bovine sequences.

B. Schilling et al. / Biochimica et Biophysica Acta 1762 (2006) 213–222 217

experiments and the subsequent separation of Complexes II

and III into their protein subunits by 1D SDS PAGE are

shown in Fig. 1. Complex II was successfully immuno-

purified from both bovine and mouse sources before

undergoing final separation by 1D SDS-PAGE (Fig. 1,

panels A and B). All four protein subunits of bovine

Complex II could be identified by mass spectrometry as

visualized by Coomassie staining, while only 3 out of 4

were identified from mouse. Similarly, bovine and mouse

Complex III were immunocaptured from bovine and mouse

Fig. 1. Immunopurified bovine and mouse Complexes II and III separated

by 1D SDS-PAGE. Complexes II and III were immunopurified from bovine

heart mitochondria (2.5 mg) and mouse heart mitochondria (2.5 mg)

mitochondria using Complex II monoclonal antibody MS201 (MitoScien-

ces) and Complex III monoclonal antibody MS301 (MitoSciences),

respectively. Eluted proteins were resolved on a NuPAGE 4–12% BisTris

(MES) gel, and stained with Coomassie Brilliant Blue. Panel A shows

immunoprecipitated bovine heart Complex II, Panel B shows immunopre-

cipitated mouse heart Complex II. Four out of the four (4/4) Complex II

subunits were identified by mass spectrometry (see annotation). Panels C

and D show immunoprecipitated Complex III isolated from bovine heart

mitochondria (Panel C) and from mouse heart mitochondria (Panel D),

respectively. Ten out of eleven (10/11) Complex III subunits were identified

by mass spectrometry. A few protein bands were observed that contained

interacting proteins that co-immunoprecipitated as marked with a circle (&).For more details about these interacting proteins, see Supplementary Tables

S3 and S4.

heart mitochondria, and subsequently separated by 1D SDS-

PAGE (Fig. 1, panels C and D). Overall, 10 out of 11

subunits of Complex III were identified by mass spectrom-

etry from both bovine and mouse mitochondria, and their

migration position indicated in Fig. 1C and D (for

comparison, see previous report Sun et al. [21] for the

analogous separation of Complex III subunits from rat

heart).

To obtain a more complete sequence coverage of the

immunopurified Complexes II and III from mouse and

bovine heart mitochondria, a combination of MALDI-MS

and nano-HPLC-ESI-MS/MS analysis was employed. The

mass spectrometric results of these latter studies are listed in

Tables 1–4. Following excision from the 1D SDS-PAGE

gel, protein bands were digested with trypsin and analyzed

by MALDI-MS prior to analysis by HPLC ESI-MS/MS. In

addition, all protein bands (except for the mouse Complex

III samples) were subjected to incubation with chymotrypsin

to provide overlapping sequence coverage. All the major 1D

SDS gel bands visualized by Coomassie staining obtained

were assigned as Complex II and Complex III subunits,

demonstrating the specificity of the monoclonal antibodies

and the effectiveness of the immunocapture protocol (see

Fig. 1 and Tables 1–4).

Although HPLC ESI-MS/MS provided the best overall

sequence coverage, in many cases the peptide mass finger-

print data acquired by MALDI-MS was sufficient to identify

the individual Complex II and III protein subunits.

Presumably, these identifications were possible due the

efficiency of the immunoprecipitation protocol combined by

1D SDS-PAGE separation that greatly decreased the overall

sample complexity. For example, Fig. 2 shows two MALDI-

MS spectra that contain molecular ions of peptides obtained

after in-gel tryptic digestion of selected mouse and bovine

protein bands from the 1D SDS-PAGE gels. Similarly, Fig.

2A shows 33 peptide mass fingerprints resulting from

tryptic digestion of mouse Complex II protein subunit 1 that

yielded a protein sequence coverage of 69% and a

corresponding Mascot Mowse (M) score of 310 (a Mowse

score of 25 or higher was considered significant for PMF

data in this study [18]). Likewise, Fig. 2B displays 30

peptide mass fingerprints resulting from tryptic digestion of

bovine Complex III subunit 1 (69% sequence coverage;

M=253). In both cases (Panels A and B), a single unique

protein subunit of Complex II or Complex III was found to

be the major component present in each corresponding gel

band. Several other subunits from Complexes II and III

could also be assigned after tryptic digestion by MALDI-

MS screening alone, with a maximum coverage of 73% (18

peptides, M=129) for bovine Complex III-6 (see Supple-

ment, Table S2).

The initial screening of the tryptic and chymotryptic

samples by MALDI-MS was then followed by more

detailed analyses using online nano-HPLC-ESI-MS/MS.

As expected, the HPLC-MS/MS approach provided a

significant increase in sequence coverage for both Complex

Table 3

Proteins identified by ESI-MS/MS from immunocaptured bovine heart Complex III

Complex III

protein subunitaTheor. No. tryptic

peptides

Tryptic seq.

coverage (M score)bNo. chymo.

peptides

Chymotryp. seq.

coverage (M score)bTotal sequence

coverage (%)MW�10�3 pIc

C III-1 49.9 5.5 32 41% (1000) 52 43% (424) 63

C III-2 46.6 7.8 28 49% (927) 61 55% (503) 74

C III-3 42.8 7.8 1d 2% (49) 1d 2% (22) 5

C III-4 27.6 6.5 16 52% (435) 7 24% (89) 65

C III-5 21.8 7.0 8 27% (238) 28 47% (246) 64

C III-6 13.5 9.1 18 64% (435) 3 19% (13) 66

C III-7 9.6 10.3 9 46% (242) – – 46

C III-8 9.5 4.5 3 19% (112) – – 19

C III-9 8.0 11.5 1d 17% (48) 4d 42% (86) 47

C III-10 7.2 9.5 3 51% (130) – – 51

C III-11 6.5 10.2 – – – – –

a Complex III protein subunits (descriptive subunit names): III-1 (core protein I), III-2 (core protein II), III-3 (cytochrome b), III-4 (cytochrome c1), III-5

(Rieske iron-sulfur subunit), III-6 (14 kDa subunit), III-7 (9.5 kDa subunit), III-8 (11 kDa subunit), III-9 (chain u), III-10 (7.2 kDa subunit), and III-11 (6.4 kDa

subunit).b Amino acid sequence coverage of mature (processed) proteins in %, Mascot database search score according to Perkins et al. [18].c Molecular weight and pI were calculated for the mature protein subunits, i.e., after processing of the precursor protein sequences (if applicable) that

contained mitochondrial import signals.d ESI-MS/MS spectrum and fragmentation pattern was inspected ‘‘manually’’ and evaluated (also see Fig. 3).

B. Schilling et al. / Biochimica et Biophysica Acta 1762 (2006) 213–222218

II and III subunits. Overall, all four protein subunits were

observed for bovine Complex II, and 3 of 4 subunits for

mouse Complex II. For Complex III, 10 out of the expected

11 protein subunits were identified both the bovine and

mouse immunocapture experiments. The smallest protein

subunit from Complex III, subunit 11 (III-11, 6.4 kDa

subunit), was not identified. Tables 1–4 summarize the

Table 4

Proteins identified by ESI-MS/MS from immunocaptured mouse heart

Complex III

Complex III

protein subunitaTheor. No. tryptic

peptides

Tryptic sequence

coverage

(M score)bMW�10�3 pIc

C III-1 49.8 5.3 36 65% (585)

C III-2 46.6 9.0 33 37% (612)

C III-3 43.3 7.8 1d 3% (30)d

C III-4 27.5 6.5 8 39% (122)

C III-5 21.7 7.0 12 27% (161)

C III-6 13.6 9.1 16 54% (206)

C III-7 9.8 10.3 9 40% (133)

C III-8 9.3 4.8 3 32% (72)

C III-9 7.9 11.6 1d 16% (14)d

C III-10 7.4 9.2 7 50% (130)

C III-11 6.5 9.8 – –

a Complex III protein subunits (descriptive subunit names): III-1 (core

protein I), III-2 (core protein II), III-3 (cytochrome b), III-4 (cytochrome

c1), III-5 (Rieske iron-sulfur subunit), III-6 (14 kDa subunit), III-7 (9.5 kDa

subunit), III-8 (11 kDa subunit), III-9 (chain u), III-10 (7.2 kDa subunit),

and III-11 (6.4 kDa subunit).b Amino acid sequence coverage of mature (processed) proteins in %,

Mascot database search score according to Perkins et al. [18].c Molecular weight and pI were calculated for the mature protein

subunits, i.e., after processing of the precursor protein sequences (if

applicable) that contained mitochondrial import signals as identified by

homology to the mature bovine sequences.d ESI-MS/MS fragmentation pattern was inspected manually, fragmenta-

tion was compared to identical peptide observed for homologous bovine

subunit.

mass spectrometric details of protein identification, includ-

ing information as to the number of peptides observed, the

Mascot Mowse scores, and protein sequence coverages

calculated for the mature (processed) protein subunits.

Mascot database search results of the ESI-MS/MS peptide

data are listed for both tryptic and chymotryptic digestions.

As can be seen from these data, high protein sequence

coverages were obtained for the majority of the subunits

from ESI-MS/MS data of the tryptic digests alone (average

coverage ¨40%). This trypsin-based protein sequence

coverage was improved further when combined with the

chymotryptic peptide data set, yielding a combined MS/MS

sequence coverage ranging from a high of 87% (II-1,

mouse) to a low of 5% (III-3, bovine), with a combined

average coverage of ¨59% (see Tables 1–4).

For two of the protein subunits from Complex III (III-3

and III-9, bovine, and mouse), we observed only one tryptic

peptide in each case by tandem mass spectrometry. The

individual Mascot scores for bovine III-3 and III-9 were 49

and 48, respectively, which corresponds to a confident

peptide assignment. Nevertheless, we manually inspected

these MS/MS spectra and fragmentation pattern of these 2

peptides to confirm the identity. The CID (collision-induced

dissociation) tandem mass spectrum of peptide SGPFAPVL-

SATSR (residues 8–20) from bovine Complex III subunit 9

(chain u), for example, is shown in Fig. 3. The molecular

ion, [M+2H]2+ at m/z 645.312+ (M=1288.61 Da) was

selected for collision-induced dissociation and the resulting

MS/MS spectrum revealed distinctive fragmentation,

including a nearly complete y-fragment ion series at m/z

175.1, 262.1, 434.3, 521.3, 634.3, 733.4, 830.5, 901.5, and

1048.6 (corresponding to fragment ions y1, y2, and y4–y10),

and several b-fragment ions at m/z 242.1, 389.2, and 460.2

(corresponding to fragment ions b3–b5). Similarly, the ESI-

MS/MS fragmentation pattern of the ‘‘one-peptide-hit’’

Fig. 2. MALDI-TOF mass spectra of Complex II subunit 1 (A) and

Complex III subunit 1 (B) obtained after immunoprecipitation from bovine

heart mitochondria and 1D SDS-PAGE separation. The MALDI-MS

peptide mass fingerprint spectra display molecular ions of peptides obtained

after in-gel tryptic digestion of the selected protein bands from a 1D SDS

PAGE gel. The observed masses are labeled and annotated with starting and

ending amino acids. Panel A shows 33 peptides resulting from tryptic

digestion of mouse mature/processed Complex II subunit 1 (mature Fp

protein subunit). Overall, a protein sequence coverage of 69% was observed

(Mowse score 310). Panel B shows 30 peptide mass fingerprints resulting

from tryptic digestion of bovine mature/processed Complex III subunit 1

(mature Core 1 subunit). Overall, a protein sequence coverage of 69% was

observed (Mowse score 253). Several more peptide mass fingerprints were

observed as marked with #1–#4, and correspond to the following sequences

in C II-1: #1=475–484 with m/z 1160.6; #2=594–604 with m/z of 1332.7;

#3=69–77 with m/z at 998.5; and #4=180–188 with m/z at 1067.5.

Fig. 3. ESI-MS/MS spectrum of peptide SGPFAPVLSATSR (residues 8–

20) after immunoprecipitation and tryptic digestion of bovine Complex III

subunit 9 (chain u). The molecular ion, [M+2H]2+ at m/z 645.312+

(M=1288.61 Da) was selected for collision-induced dissociation and a

distinctive series of y-fragment ions and several b-fragment ions were

observed. Fragment ions marked with * indicate internal fragment ions;

peaks marked with # indicate fragment ion loss of H2O.

B. Schilling et al. / Biochimica et Biophysica Acta 1762 (2006) 213–222 219

(DVNYGWIIR, residues 72–80) resulting from bovine

Complex III subunit 3 (cytochrome b) was also inspected

manually and was judged to be the correctly assigned based

on an overall quality MS/MS spectrum, the presence of

several b-ions and a significant y-ion series (y1–y7), also

compare similar evaluation criteria as reported by Link et al.

[19]. In addition, chymotryptic digestion of these specific

proteins generated additional peptides that confirmed these

subunit assignments and increased the overall coverage.

Interestingly, the corresponding homologous mouse proteins

Complexes III-9 and III-3 obtained after immunoisolation

from mouse heart, generated identical tryptic peptides,

SGPFAPVLSATSR and DVNYGWIIR, as the bovine

counterpart. The ESI-MS/MS spectra and fragmentation

patterns of these two mouse peptides were weaker but could

be compared to the fragmentation patterns obtained from the

identical bovine peptides and properly validated.

As described above, all protein subunits from both

bovine and mouse Complexes II and III were identified by

mass spectrometry, with the exception of subunit 11 from

Complex III. All of these subunits could be matched to the

major bands as visualized by 1D SDS PAGE (see Fig. 1)

after immunoprecipitation. However, some ‘‘co-precipitat-

ing/interacting proteins’’ were also identified at low to very

low abundance levels in regions of the gel lane where

the protein dye stain was barely visible. Most of these

low abundance co-precipitating proteins were found to be

subunits from other electron transport chain complexes, and

may, in fact, be components of so-called Fsupercomplexes_[22–24]. A complete list of these co-precipitating proteins

can be found in the Supplementary Tables S3 and S4.

3.2. Posttranslational modifications of Complexes II and III

protein subunits

As part of our efforts to purify and identify the subunits

of Complexes II and III, we also examined these separated

proteins for the presence of posttranslational modifications

(PTMs). The mass spectrometric dataset obtained from the

immunoprecipitation experiments was therefore exhaus-

tively searched against custom-designed databases that

included all proteins for the bovine and mouse Complexes

II and III (for database information see Materials and

methods and Supplementary Table S1). Searching custom

databases using our in-house Mascot server allowed us to

perform less restrictive searches, essentially allowing for all

possible PTMs. Using this search, we observed modifica-

B. Schilling et al. / Biochimica et Biophysica Acta 1762 (2006) 213–222220

tions of several protein subunits including N-terminal

acetylation, as well as several oxidative and non-enzymatic

modifications, such as deamidation of Asn, oxidation of Trp

to formylkynurenine, oxidation of Met to the sulphone, and

oxidation of Cys to cysteic acid (for a more detailed list of

modified subunits see Supplementary Table S5). An

example of a modified peptide is shown in Supplementary

Fig. S1, where the tandem mass spectrum (ESI-MS/MS) of

N-acetylated peptide AGRPAVSASSR is shown, confirm-

ing previous observations [25] that the bovine Complex III

protein subunit 6 (14 kDa subunit) is processed by the

removal of the N-terminal methionine and N-acetylated at

alanine. The analogous acetylation event was also observed

for mouse Complex III subunit 6 (see Supplementary Table

S5), which had not previously been identified.

As another example of a protein modification present

in this MS data set, a tryptic peptide from mouse Complex

II subunit 1 (Fp/flavoprotein subunit) corresponding to

residues Ser-55 –Arg-77 (SHTVAAQGGINAALGN-

MEEDNWR) was identified. This peptide, however, was

shifted in mass by 783 Da corresponding to the presence of

a covalently bound flavin adenine dinucleotide (FAD)

group. This FAD-containing peptide was observed at m/z

1075.43+ (M=3223.2 Da) and after collision-induced

dissociation the spectra shown in Fig. 4 was obtained.

Although an extensive y-ion series was observed (y1–y16),

Fig. 4. ESI-MS/MS spectrum of FAD-modified peptide isolated following

immunoprecipitation, 1D SDS-PAGE separation and tryptic digestion of

mouse Complex II. Peptide SHTVAAQGGINAALGNMEEDNWR (resi-

dues Ser-55–Arg-77) originates from the Fp subunit and contains an intact

flavin adenine dinucleotide (FAD) covalently bound to His-56. FAD

contributes 783.1 Da to the mass of the peptide. The molecular ion,

[M+3H]3+ at m/z 1075.43+ (M=3223.2 Da) gave an extensive series of

y(n)-fragment ions after collision-induced dissociation. The b-fragment ions

exhibited loss of adenosine monophosphate (AMP) from the FAD

modification, but still showed the flavin moiety. For example, the b(n)*-

ions show a mass shift of 436 Da compared to the corresponding values

expected for the unmodified peptide. Abbreviations A: adenine; AMP:

adenosine monophosphate. A more complete description of the fragmenta-

tion mechanism of FAD modified peptides can be found in a report by

Halada et al. [30] for pyranose 2-oxidase.

this series gave no information as to the location of the FAD

group as this group appears to be located close to the N-

terminus of the fragmented peptide. Caused by gas phase

reactions, two abundant ions resulting from FAD itself were

observed at m/z 136.1 (adenine; ‘‘A’’) and 348.1 (AMP). In

contrast to the y-ion series, an extensive b*-ion series

exhibited a mass shift that could be directly correlated to the

position of the FAD moiety. These b*-fragment ions showed

a loss of adenosine monophosphate (AMP) from the FAD

modification, but with the flavin moiety attached (+436 Da)

when compared to the expected masses of the unmodified b-

ions. Overall, the fragmentation pattern clearly proved the

FAD group to be covalently attached to residue His-56 as

one might expect from ‘‘prediction by similarity’’ for the

mouse Complex II protein subunit 1.

4. Discussion

In this current study, we have provided a thorough

proteomic analysis of the two smallest OXPHOS compo-

nents, Complexes II and III, each obtained by selective

immunocapture from both mouse and bovine mitochondria.

Mass spectrometric sequence analysis of the immunopuri-

fied proteins led to the identification of 4 out of the 4

subunits of bovine Complex II (3/4 for mouse Complex II),

and to the identification of 10 out of the 11 subunits of

bovine and mouse Complex III. The only protein subunit

that was not identified in this study was Complex III subunit

11, a small 6.4 kDa protein (56 residues) that contains a 21-

amino acid transmembrane domain. Although the large

majority of subunits were identified with a high sequence

coverage, some subunits were more difficult. For example,

although bovine Complex II subunit 4, a small membrane-

anchoring cytochrome b associated subunit, was identified

by mass spectrometry, only 3 peptides were assigned by

MS/MS and these peptides were located outside the large

transmembrane domain for a total sequence coverage of

16%. As the mature bovine II-4 subunit contains 103

residues (11.1 kDa) and residues 15–35 and 70–86 form

transmembrane domains, this is not surprising. Indeed, Yang

et al. [26] also described similar difficulties in identifying

this subunit from E. coli that was compounded by low

staining intensities with Coomassie Blue.

The monoclonal antibodies used in these studies for the

immunocapture of Complexes II and III were initially

described by Murray et al. [16]. Interestingly, these two

antibodies were originally developed against bovine Com-

plexes II and III, but worked equally well against the two

analogous mouse complexes. This high degree of cross-

reactivity is not unexpected given the high sequence identity

between the bovine and mouse subunits, ranging from a

high of 96% to a low of 79% (see Supplementary Table S1).

These antibodies have also proven effective in immunopre-

cipitating Complexes II and III from human mitochondria

(data not shown) suggesting they may prove effective

B. Schilling et al. / Biochimica et Biophysica Acta 1762 (2006) 213–222 221

against other mammalian species, and possibly some non-

mammalian species such as C. elegans where a high level of

protein subunit sequence identity exists [27].

The ability to obtain a high sequence coverage is a

generally a necessary requirement for detecting post-trans-

lational modifications. To graphically present sequence our

coverage data (see Tables 1–4), a Java program named

‘‘SeqDisp’’ was developed in-house as described previously

[12]. A specific case is shown in Supplementary Material

(Fig. S2) where the sequence coverage of mouse Complex II

subunit 2 is displayed for individual experiments using

trypsin (71% coverage) and chymotrypsin (45% coverage),

yielding a combined coverage of 80%. Such graphic

displays of experimental sequence data were performed

for each individual subunit (data not shown) and are useful

to optimize experimental conditions to target specific

posttranslational modifications.

With the rather extensive proteomic/mass spectrometric

data set obtained in our studies, we were able to identify

several posttranslational modifications, including N-termi-

nal acetylation (see Supplementary Fig. S1), as well as

several oxidative and non-enzymatic modifications, such as

deamidation of Asn, oxidation of Trp to formylkynurenine,

oxidation of Met to form a sulphone, and oxidation of Cys

to form a cysteic acid (see Supplementary Table S5). These

latter oxidative (non-enzymatic) modifications, however,

need to be evaluated under more carefully controlled

conditions before any biological significance can be

established. Nonetheless, we were able to identify the

covalent FAD modification of mouse Complex II subunit

1, the Fp (flavoprotein) subunit, and determine the site of

modification as residue His-56. This histidine residue was

‘‘predicted by similarity’’ to carry the FAD modification (see

Swiss-Prot entry for accession number Q8K2B3; gene name

DHSA_MOUSE), our data now validates this prediction.

The FAD moiety (D =783 Da) is quite a large covalent

protein modification consisting of several components, the

flavinmononucleotide (FMN) and adenosine monophos-

phate (AMP) [28,29]. Previously, mass spectrometric

studies have been performed on proteins and peptides

containing FAD [30,31]. It is often challenging to find

conditions that maintain a covalently bound FAD group,

suggesting that the immunopurification and sample han-

dling protocols are mild enough to maintain this and other

protein modifications.

Studies are now underway to identify other posttransla-

tional modifications in these and other OXPHOS com-

plexes, including both enzymatic (e.g., phosphorylation)

and non-enzymatic modifications, especially those that

might arise from oxidative stress [1,32]. To address this

latter point, we have recently developed a stable isotope

methodology [33] using differential alkylation to determine

the oxidative status of cysteine residues in proteins, such as

glutathionylation and nitrosylation. The application of this

latter method in the context of the immunopurification and

proteomic characterization strategy of OXPHOS complexes

described here should hopefully allow one to assess changes

in the oxidation status at both the subunit and the amino acid

level. In general, the ability to isolate highly purified

Complexes II and III (and other OXPHOS enzymes) from

rodent and other mammalian sources provides a powerful

means to examine a wide variety of potential defects or

modifications in specific subunits that make up these

enzymes.

Acknowledgement

This work was supported by NIH R21 NS 043620-01

grant to B.W.G.

Appendix A. Supplementary data

Supplementary data associated with this article can be

found, in the online version, at doi:10.1016/j.bbadis.2005.

07.003.

References

[1] I. Dalle-Donne, A. Scaloni, D. Giustarini, E. Cavarra, G. Tell, G.

Lungarella, R. Colombo, R. Rossi, A. Milzani, Proteins as biomarkers

of oxidative/nitrosative stress in diseases: the contribution of redox

proteomics, Mass Spectrom. Rev. 24 (2005) 55–99.

[2] D.G. Nicholls, Mitochondrial function and dysfunction in the cell: its

relevance to aging and aging-related disease, Int. J. Biochem. Cell

Biol. 34 (2002) 1372–1381.

[3] L. van den Heuvel, J. Smeitink, The oxidative phosphorylation

(OXPHOS) system: nuclear genes and human genetic diseases,

Bioessays 23 (2001) 518–525.

[4] P. Rustin, A. Rotig, Inborn errors of complex II—Unusual human

mitochondrial diseases, Biochim. Biophys. Acta 1553 (2002) 117–122.

[5] P. Rustin, A. Munnich, A. Rotig, Succinate dehydrogenase and human

diseases: new insights into a well-known enzyme, Eur. J. Hum. Genet.

10 (2002) 289–291.

[6] A.H. Schapira, J.M. Cooper, D. Dexter, J.B. Clark, P. Jenner, C.D.

Marsden, Mitochondrial complex I deficiency in Parkinson’s disease,

J. Neurochem. 54 (1990) 823–827.

[7] M.F. Beal, Neurochemistry and toxin models in Huntington’s disease,

Curr. Opin. Neurol. 7 (1994) 542–547.

[8] C. Jung, C.M. Higgins, Z. Xu, Mitochondrial electron transport chain

complex dysfunction in a transgenic mouse model for amyotrophic

lateral sclerosis, J. Neurochem. 83 (2002) 535–545.

[9] B.J. Hanson, B. Schulenberg, W.F. Patton, R.A. Capaldi, A novel

subfractionation approach for mitochondrial proteins: a three-

dimensional mitochondrial proteome map, Electrophoresis 22

(2001) 950–959.

[10] S.W. Taylor, D.E. Warnock, G.M. Glenn, B. Zhang, E. Fahy, S.P.

Gaucher, R.A. Capaldi, B.W. Gibson, S.S. Ghosh, An alternative

strategy to determine the mitochondrial proteome using sucrose

gradient fractionation and 1D PAGE on highly purified human heart

mitochondria, J. Proteome Res. 1 (2002) 451–458.

[11] J. Murray, B. Zhang, S.W. Taylor, D. Oglesbee, E. Fahy, M.F.

Marusich, S.S. Ghosh, R.A. Capaldi, The subunit composition of the

human NADH dehydrogenase obtained by rapid one-step immuno-

purification, J. Biol. Chem. 278 (2003) 13619–13622.

[12] B. Schilling, M.M.S. Bharath, R.H. Row, J. Murray, M.P. Cusack, R.A.

B. Schilling et al. / Biochimica et Biophysica Acta 1762 (2006) 213–222222

Capaldi, C.R. Freed, K.N. Prasad, J.K. Andersen, B.W. Gibson, Rapid

purification and mass spectrometric characterization of mitochondrial

NADH dehydrogenase (Complex I) from rodent brain and a dopami-

nergic neuronal cell line, Mol. Cell. Proteomics 4 (2005) 84–96.

[13] R. Aggeler, J. Coons, S.W. Taylor, S.S. Ghosh, J.J. Garcia, R.A.

Capaldi, M.F. Marusich, A functionally active human F1F0 ATPase

can be purified by immunocapture from heart tissue and fibroblast cell

lines. Subunit structure and activity studies, J. Biol. Chem. 277 (2002)

33906–33912.

[14] M. Lib, A. Rodriguez-Mari, M.F. Marusich, R.A. Capaldi,

Immunocapture and microplate-based activity measurement of

mammalian pyruvate dehydrogenase complex, Anal. Biochem. 314

(2003) 121–127.

[15] A. Smith, Preparation, properties, and conditions for assay of

mitochondria: slaughterhouse material, Methods Enzymol. 10 (1967)

81–86.

[16] J. Murray, M.F. Marusich, R.A. Capaldi, R. Aggeler, Focused

proteomics: monoclonal antibody-based isolation of the oxidative

phosphorylation machinery and detection of phosphoproteins using a

fluorescent phosphoprotein gel stain, Electrophoresis 25 (2004)

2520–2525.

[17] U.K. Laemmli, Cleavage of structural proteins during the assembly of

the head of bacteriophage T4, Nature 227 (1970) 680–685.

[18] D.N. Perkins, D.J. Pappin, D.M. Creasy, J.S. Cottrell, Probability-

based protein identification by searching sequence databases using

mass spectrometry data, Electrophoresis 20 (1999) 3551–3567.

[19] A.J. Link, J. Eng, D.M. Schieltz, E. Carmack, G.J. Mize, D.R. Morris,

B.M. Garvik, J.R. Yates 3rd, Direct analysis of protein complexes

using mass spectrometry, Nat. Biotechnol. 17 (1999) 676–682.

[20] X.Q. Huang, W. Miller, A time-efficient, linear-space local similarity

algorithm, Adv. Appl. Math. 12 (1991) 337–357.

[21] G. Sun, M.T. Kinter, V.E. Anderson, Mass spectrometric character-

ization of mitochondrial electron transport complexes: subunits of the

rat heart ubiquinol-cytochrome c reductase, J. Mass Spectrom. 38

(2003) 531–539.

[22] H. Schagger, K. Pfeiffer, Supercomplexes in the respiratory chains of

yeast and mammalian mitochondria, EMBO J. 19 (2000) 1777–1783.

[23] C.M. Cruciat, S. Brunner, F. Baumann, W. Neupert, R.A. Stuart, The

cytochrome bc1 and cytochrome c oxidase complexes associate to

form a single supracomplex in yeast mitochondria, J. Biol. Chem. 275

(2000) 18093–18098.

[24] N.V. Dudkina, H. Eubel, W. Keegstra, E.J. Boekema, H.P. Braun,

Structure of a mitochondrial supercomplex formed by respiratory-

chain complexes I and III, Proc. Natl. Acad. Sci. U. S. A. 102 (2005)

3225–3229.

[25] S. Wakabayashi, T. Takao, Y. Shimonishi, S. Kuramitsu, H. Matsu-

bara, T. Wang, Z. Zhang, T.E. King, Complete amino acid sequence of

the ubiquinone binding protein (QP-C), a protein similar to the

14,000-dalton subunit of the yeast ubiquinol-cytochrome c reductase

complex, J. Biol. Chem. 260 (1985) 337–343.

[26] X. Yang, L. Yu, C.A. Yu, Resolution and reconstitution of succinate-

ubiquinone reductase from Escherichia coli, J. Biol. Chem. 272

(1997) 9683–9689.

[27] W.Y. Tsang, B.D. Lemire, The role of mitochondria in the life of the

nematode, Caenorhabditis elegans, Biochim. Biophys. Acta 1638

(2003) 91–105.

[28] O. Dym, D. Eisenberg, Sequence-structure analysis of FAD-containing

proteins, Protein Sci. 10 (2001) 1712–1728.

[29] D.E. Edmondson, P. Newton-Vinson, The covalent FAD of mono-

amine oxidase: structural and functional role and mechanism of the

flavinylation reaction, Antioxid. Redox Signal. 3 (2001) 789–806.

[30] P. Halada, C. Leitner, P. Sedmera, D. Haltrich, J. Volc, Identification

of the covalent flavin adenine dinucleotide-binding region in

pyranose 2-oxidase from Trametes multicolor, Anal. Biochem. 314

(2003) 235–242.

[31] L.J. Chlumsky, A.W. Sturgess, E. Nieves, M.S. Jorns, Identification of

the covalent flavin attachment site in sarcosine oxidase, Biochemistry

37 (1998) 2089–2095.

[32] B.W. Gibson, Exploiting proteomics in the discovery of drugs that

target mitochondrial oxidative damage, Sci. Aging Knowledge

Environ. 2004 (2004) pe12.

[33] B. Schilling, C.B. Yoo, C.J. Collins, B.W. Gibson, Determining

cysteine oxidation status using differential alkylation, Int. J. Mass

Spectrom. 236 (2004) 117–127.