Westermeier R., Naven T., H?pker H.-R. Proteomics in Practice: A Guide to Successful Experimental Design

Подождите немного. Документ загружается.

3 Mass Spectrometry250

3.3.4.2 Precursor Ion Scan

For proteomics applications, a precursor ion scan is employed to

locate sites of protein modifications, such as phosphorylation. In this

experiment the first part of the MS analyzer, MS

, is set to transmit

all the components of the mixture in to the collision cell to undergo

CID. The third part of the analyzer, MS

, is fixed at a specific mass

value, so that only analytes, which fragment to give a fragment ion of

this specific mass will be detected at the detector (see Figure 3.19,

middle panel). In this manner, the precursor ion scan can be used to

selectively identify certain species in a complex mixture, such as pep-

tides containing a phosphorylated residue.

The presence of serine, threonine or tyrosine phosphorylated pep-

tide in a complex peptide digest can be determined using a precursor

ion scan. A phosphorylated peptide fragments during low-energy col-

lisional fragmentation to give a fragment ion of 79 Da (the PO

–

group). Therefore to specifically identify phosphorylated peptides in a

mixture the third part (MS

) of the analyzer is set to 79. Thus, only

species which fragment to give a fragment ion of 79 are detected and

hence indicate the presence of a phosphorylated residue (Wilm et al.

1996; Carr et al. 1996; Annan et al. 2001).

This experiment has been historically restricted to a triple quadru-

pole instrument, but recent applications have been demonstrated

using a hybrid quadrupole TOF instruments (Steen et al. 2001). The

method has been exploited for the specific detection of tyrosine phos-

phorylated peptides in complex mixtures.

3.3.4.3 Neutral Loss Scan

Similarly, a neutral loss scan is employed in proteomics applications

to locate peptides containing a protein modification. In this techni-

que, the 1

and 3

parts of the analyzer are scanned synchronously,

but with a specific m/z offset. Once more the second part of the ana-

lyzer is used as the collision cell and the entire mixture is allowed to

enter the collision cell, only those species which fragment to yield a

fragment ion with the same mass as the offset will be detected (see

Figure 3.20). For instance, serine and threonine phosphorylated pep-

tides readily lose phosphoric acid during low-energy CID. Phosphoric

acid has a mass of 98 Da, thus the offset for a doubly charged peptide

would be set at 49. Hence any species which loses 49 Da from a dou-

bly charged ion would be observed at the detector and be indicative of

a phosphorylated peptide (Covey et al. 1991; Schlosser et al. 2001).

This type of experiment is typically carried out with a triple quad or

Q-TOF type instrument. Phosphopeptide analysis in an ion trap is

Wilm M, Neubauer G, Mann

M. Anal Chem. 68 (1996)

527–533.

Carr SA, Huddleston MJ,

Annan RS. Anal Biochem 239

(1996) 180–192.

Annan RS, Huddleston MJ,

Verna R, Deshaies RJ, Carr SA.

Anal Chem 73 (2001) 393–

404.

Steen H, Kster B, M

Fernandez, Pandy A, Mann M.

Proceedings of the 49

ASMS

conference on mass spectro-

metry and allied topics,

Chicago ( 2001).

Covey TR, Huang EC, Henion

JD. Anal Chem 63 (1991)

1193–1200.

Schlosser A, Pipkorn R, Bosse-

meyer D, Lehman WD. Anal

Chem 73 (2001) 170–176.

3.3 Generating MS Data for Protein Identification 251

typically performed with a MS

approach. Both approaches have dif-

ferent specific benefits, which are not pointed out here.

3.3.4.4 MALDI-TOF Post Source Decay

Though not a genuine MS/MS technique, MALDI-TOF post source

decay (PSD) is capable of generating sequence specific information,

which can be used for protein identification. Described by Spengler

and co-workers in the 1990s (Spengler 1997; Chaurand et al. 1999),

this technique demonstrated some reasonable success as a protein

identification tool. The quality of data is less than that acquired with

conventional MS/MS instruments, as described earlier (Silles et al.

2000; Gevaert et al. 2001), though still sufficient for protein ID from

relatively simple samples such as those derived from a 2D PAGE gel.

In the technique, the peptide of interest is specifically selected

from a complex mixture using an ion gate. In a conventional linear

reflectron, the configuration can only accommodate a small range of

KE differences i.e the reflectron will only be able to focus those ions

that are similar in mass to the parent ion (Figure 3.23). In this

instance, only PSD ions that are close in mass to the precursor ion

will be mass measured correctly. Thus, to perform PSD with a con-

ventional reflectron the voltage on the reflectron has to drop sequen-

tially to allow the fragment ions of lower mass (and lower KE) to

become focussed at the detector. This is termed stepping the reflec-

tron. In a conventional reflectron, several different voltages or seg-

ments, (as many as 12 segments may be required) are required to

focus the entire mass range of product ions (Figure 3.24a). The indi-

vidual segments are then "stitched" together using the instrument

software, generating the complete product ion spectrum. (Figure

3.24b).

The use of a quadratic field reflectron for PSD analysis attempts to

address the issue of stepping the reflectron. It is capable of accommo-

dating ions of widely different kinetic energies in a single spectrum

(Figures 3.22–3.23), because in a quadratic field reflectron an

increased voltage is applied, non-linearly, to create a perfect quadratic

field across the reflectron. Resultantly, all the product ions created

during the post source decay of a particular precursor ion are focused

at the detector, irrespective of their energy, over the entire range of

m/z. Hence, a complete PSD spectrum is obtained under the same

experimental conditions with each pulse of the laser, without the

need for data stitching.

Spengler B. J Mass Spectrom.

32 (1997) 1019–1036.

Chaurand P, Luetzenkirchen F,

Spengler B. J Am Soc Mass

Spectrom 10 (1999) 91–103.

Silles E, Mazon MJ, Gevaert K,

Goethals M, Vandekerckhove,

Lebr R, Sandoval IV. J Biol

Chem. 275 (2000) 34054–

34059.

Gevaert K, Demol H, Martens

L, Hoorelbeke B, Puype M,

Goethals M, Van Damme J, De

Boeck S, Vandekerckhove J.

Electrophoresis 22 (2001)

1645–1651.

3 Mass Spectrometry252

Fig. 3.23: Schematic demonstrating the passage of the precursor ion and

product ion through two types of reflectron. Upper panel: conventional linear

reflectron. Lower panel: quadratic field reflectron.

Under typical timed ion delay MALDI conditions, peptide fragmenta-

tion observed during PSD is reduced (Kauffman et al. 1996). Keough

and co workers described a derivatization method for peptide

sequence analysis by MALDI PSD that facilitated fragmentation dur-

ing PSD, increasing sensitivity significantly. The tag was designed to

promote efficient charge-site initiated fragmentation of the peptide

bonds. Consequently, only a single ion series, the y-ion series, is

observed in the PSD spectrum, simplifying interpretation.

The theory proposed that under MALDI ionization conditions, the

strong basic residue at the C-terminus of tryptic peptides (arginine)

would be protonated and the strong acidic group at the C-terminus

would be deprotonated (Figure 3.25). The additional proton (almost

exclusively MALDI ions in the peptide mass range are singly charged)

would then be free to randomly ionize and subsequently fragment

the peptide amide bonds of the peptide backbone. Hence, when the

fragment ions are formed only the y-ions are observed because, the

formal positive charge typical of a b-ion is neutralized by the negative

charge on the sulfonic acid group, hence b-ions are not observed.

Kauffman R, Chaurand P,

Kirsch D, Spengler B. Rapid

Commun Mass Spectrom 10

(1996) 1199–1208.

Keough T, Youngquist RS,

Lacey MP. Proc Natl Acad Sci

USA 96 (1999) 7131–7136.

Keough T, Lacey MP, Fieno

AM, Grant RA, Sun Y, Bauer

MD, Begley KB. Electrophoresis

21 (2000) 2252–2265.

3.3 Generating MS Data for Protein Identification 253

Fig. 3.24: Schematic of the time focusing of a conventional and

quadratic field reflectron.

Upper panel: only a small segment of the mass range can be

focussed at one time with a conventional linear reflectron.

Middle panel: multiple segments acquired to obtain the whole

mass range of fragment ions using a conventional linear

reflectron.

Lower panel: quadratic field reflectron allows the focussing of

the mass range in a single TOF spectrum.

–

S–A–A–A–A–A–R

(a)

–

S–CHR–CO

NCHR–CO

(b)

ion y

ion

(R = guanidino side-chain of arginine)

Fig. 3.25: The effect of a strongly acidic group at the N-terminus

of an arginine C-terminating peptide.

3 Mass Spectrometry254

In the original method, the peptide digest was derivatized with 2-sul-

fonyl acetyl chloride, labeling the a-amino group at the N-terminus of

each peptide (and the e-amino group of the each lysine side-chain)

with the very strong acid, sulfonic acid (pKa<2).

Labeling of arginine-terminating tryptic peptides greatly facilitates

fragmentation, particularly fragmentation by MALDI PSD. Figure

3.26 demonstrates the extent of fragmentation during a PSD experi-

ment. In the case of this synthetic peptide, 16 consecutive residues

can be readily determined. The derivatization technique is compatible

with protein digests derived from 2D gels (see Sections 3.4.3–3.4.4).

Fig. 3.26: PSD spectrum of a CAF derivatized synthetic peptide

acquired with the Ettan MALDI-TOF Pro (250 fmol applied to the target).

The 16 consecutive residues are readily determined. Though Leucine (L)

is specified in the sequence, this cannot be determined from the

experiment (Isoleucine and leucine are isobaric). dMH

is the precursor

ion. Note the intensity of the precursor ion. In a typical PSD experiment,

fragment ion abundance is a very low percentage of the precursor ion

abundance.

3.3 Generating MS Data for Protein Identification 255

As the derivatization strategy also causes the derivatization of the

e-amino group of the lysine side-chain, therefore labeling the lysine

with a strong acidic group, PSD of lysine C-terminating peptides

becomes inefficient. To extend the applicability of this method to

lysine terminating peptides, the lysine must first be modified to the

homo-arginine derivative. This has recently been reported by a num-

ber of groups to increase lysine tryptic peptide detection in peptide

mass fingerprinting (Brancia et al. 2000; Keough et al. 2000). If this

modification of the lysine peptides is performed prior to the sulfona-

tion reaction then the sulfonation methodology is equally applicable

to lysine C-terminating peptides.

The sulfonation method has further been improved with the devel-

opment of a water stable sulfonic NHS ester reagent (Figure 3.27;

Liminga et al. 2001). Subsequent PSD spectra acquired from peptides

derivatized in this manner is termed chemically assisted fragmenta-

tion (CAF).

Pep

Fig. 3.27: Reaction scheme for the water stable sulfonic NHS ester reagent.

One of the drawbacks of the original method was the hygroscopic nat-

ure of the acetyl chloride reagent, requiring the derivatization proce-

dure to be performed under stringent conditions in a non-aqueous

environment. The novel watersoluble reagent allows the quantitative

derivatization of the peptide digest at sub pmol quantities of starting

material and is compatible with the lysine modification step affording

a one pot derivatization (see Step 10). The derivatization step is amen-

able to automation. This derivatization approach enables peptide

mass fingerprinting data and subsequent sequence information to be

acquired on the same instrument with good sensitivity (see Section

3.5.1).

This derivatization strategy is equally applicable to any MS/MS

instrument, which employs MALDI as the ionization source,

enabling the simple interpretation of the peptide sequence from the

fragmentation of a singly charged ion.

Despite the efforts to improve the performance of MALDI PSD, the

development of MALDI TOF/TOF systems and the performance and

automation of LC-ESI-MS/MS systems, the use of MALDI PSD has

become considerably reduced.

Brancia FL, Oliver SG, Gaskell

SJ. Rapid Commun Mass Spec-

trom 14 (2000) 2070–2073.

Keough T, Lacey MP, Young-

quist RS. Rapid Commun Mass

Spectrom 14 (2000) 2348–

2356.

Liminga M, Born M, strçm

J, Carlsson U, Keough T,

Maloisel JL, Palmgren R,

Youngquist S. Proceedings

ASMS conference on mass

spectrometry and allied topics,

Chicago, USA (2001).

3 Mass Spectrometry256

3.3.4.5 Multi-stage Tandem Mass Spectrometry

Multi-stage tandem mass spectrometry (MS

) involves multiple

stages of mass selection, separated by some form of fragmentation.

This type of fragmentation experiment is restricted to quadrupole ion

traps and FT-ICR MS. A fragment ion produced in the product ion

MS/MS spectrum can be selected, isolated and fragmented a second

time generating even further sequence information in a third spec-

trum. This spectrum is called an MS/MS/MS product ion spectrum,

or MS

(see Figure 3.28; Louris et al. 1989). This type of experiment is

useful for identification of phosphorylated or glycosylated residues if

sufficient material exists. This process can be repeated a number of

times, resulting in a series of MS

spectra where n’ represents the

number of times the fragmentation cycle has been performed.

Fig. 3.28: Schematic demonstrating the workflow in an MS

experiment.

3.3.4.6 Electron Transfer Dissociation

Electron transfer dissociation (ETD) is a new fragmentation techni-

que that fragments peptides by transferring electrons to positively

charged peptides, yielding a strech of sequence ions derived from

Louris JN, Amy JW, Ridley TY

and Cooks RG. Int J Mass

Spectrom Ion Proc 88 (1989)

97–111

3.3 Generating MS Data for Protein Identification 257

cleavage at the amide groups along the peptide backbone. Due to its

non-ergodic nature, amino acid side-chains and important modifica-

tions such as phosphorylation are not fragmented and therefore

important discriminatory information is maintained. The technique

also has the potential to analyze large peptides and proteins and

therefore support the top dpwn approach to proteomics. ETD is now

available in two commercially available ion traps.

3.3.4.7 Electron Capture Dissociation

Electron capture dissociation (ECD) is an alternative fragmentation

method to CID, but it is restricted to FT-ICR instruments (Zubarev

et al. 1998; McLafferty et al. 2000). In the ECD method, multiply

charged ions are irradiated with low-energy electrons produced by an

emitter cathode located behind the ICR cell. Upon capture of elec-

trons, reduced radical cations [M+nH]

(n–1)+

are generated which dis-

sociate by fast and facile fragmentation of the N–C

bond of the pep-

tide chain, producing mainly c and z fragment ions. The ECD

mechanism exhibits extensive peptide cleavage, retention of sites of

modification (Shia et al. 2001) and side-chain specific cleavage

enabling the differentiation of isobaric residues such as leucine and

isoleucine (Kjeldsen et al. 2003) by producing predominantly c and z

type fragment ions (Cooper et al. 2005); again it is non-ergodic in nat-

ure and in that sense similar to ETD.

The technique supports the top down approach to proteomics (Sec-

tion 3.6.2), where intact protein molecular ions are fragmented

within the mass spectrometer enabling protein sequence and PTM

information to be acquired without performing proteolysis.

The top-down proteomics approach is further supported by the

infra red multiphoton dissociation (IRMPD) dissociation technique

(Breuker et al. 2002). This technique is used for ion activation prior to

ECD. Ion activation prior to ECD in this fashion, results in greater

fragmentation of the protein backbone enabling more sequence infor-

mation to be determined.

Generation of the product ion MS/MS spectra alone does not iden-

tify the protein, this requires a database search query using one of the

common search engines. MS experiments for protein identification

rely heavily on computational data analysis and database searching

algorithms. These algorithms have been developed to assign peptide

sequences based on automated interpretation of MS/MS spectra (Eng

et al. 1994; Perkins et al. 1994; Fenyo et al. 2002; Creasy et al. 2002).

Zubarev RA, Kelleher NL,

McLafferty FW. J Am Chem

Soc 120 (1998) 3265–3266.

McLafferty FW, Horn DM,

Breuker K, Ge Y, Lewis MA,

Cerda B, Zubarev RA,

Carpenter BK. J Am Soc Mass

Spectrom. 12 (2001) 245–249.

Shi SDH, Hemling ME, Carr

SA, Horn DM, Lindh I, McLaff-

erty FW. Anal Chem. 73

(2001) 19–22.

Kjeldsen F, Haselmann K F,

Sorensen ES, Zubarev RA..

Anal. Chem. 75 (2003) 1267–

1274.

Cooper HJ, Hkansson K,

Marshall AG. Mass Spectrum.

Reviews 24 (2005) 201–222.

Breuker K, Oh H, Horn DM,

Cerda B, McLafferty, FW. J Am

Chem Soc 124 (2002) 6407–

6420.

Eng JK, McCormack AL and

Yates JR 3

, J Am Soc Mass

Spectrom 5 (1994) 976–989.

Perkins DN, Pappin DJC,

Creasy DM, Cottrell JS. Electro-

phoresis 20 (1999) 3551–3567.

Fenyo D, Beavis RC.Trends

Biotechnol. 20 (2002) 35–38.

Creasy DM, Cottrell JS. Proteo-

mics 2 (2002) 1426–1434.

3 Mass Spectrometry258

Automated analysis of peptide MS/MS spectra is now routine with

hundreds of MS/MS spectra being analyzed in a single LC-MS run.

However, with hundreds of assignments being returned by the search

query for every LC-MS run it is important to be confident that the

assignment is correct and not a false identification. For a comprehen-

sive review of database mining for proteomics applications, please

refer to the monography of Lisacek.

3.4

Protein Characterization

Protein characterization extends further than simply protein identifi-

cation and often refers to the identification and location of post-trans-

lational modifications (PTMs). PTM analysis is one of the pillars on

which proteomics is built, as PTM information cannot be determined

at the DNA level. Post-translational modification of proteins is an

essential route to modify the function/activity of that particular pro-

tein, with the process being governed by a series of enzymes includ-

ing kinases and phophatases. There are several hundred examples of

post-translational modifications (Garavelli, 2004), but the most com-

mon modifications studied are reversible phosphorylation and glyco-

sylation Specifically, reversible phosphorylation of proteins is a key

mechanism for the regulation of major cellular processes such as pro-

liferation, differentiation or apoptosis through highly dynamic and

complex signalling pathways. It has been estimated that 100,000

potential phosphorylation sites exist in the human proteome, with

the large majority poorly characterized (Zhang et al. 2002).

The key challenges in PTM analysis are to isolate/enrich for the

modified protein/peptide from a complex mixture, to quantify the

phospoprotein expression, determine the stoichiometry or occupancy

and to pinpoint the site of modification. To further complicate the

analysis, post-translationally modified proteins are often expressed

with low abundance. Traditionally, PTMs have been identified using

radiolabeling and Edman sequence analysis, amino acid analysis and

immunochemistry. However, due to the low protein expression of tar-

get proteins and the complexity of starting samples methods with

greater sensitivity of detection for PTMs are required.

As we have described previously in this chapter, the evolution of

mass spectrometry instrumentation over the last 20 years has had a

sigificant impact on protein ID and proteomics. Moreover, mass spec-

trometry is well placed to support the identification of PTMs.

Lisacek, F. A guide of good

practices for protein-centric

bioinformatics. WILEY-VCH,

Weinheim (2007).

Garavelli JS. Proteomics 4

(2004) 1527–1533.

Zhang H, Zha X, Tan Y, Horn-

beck PV, Mastrangelo AJ, Alessi

DR, Polakiewicz RD, Comb

MJ. J Biol Chem 277 (2002)

39379–39387.

3.4 Protein Characterization 259

& MS provides the sensitivity needed to study PTMs,

though the sample preparation prior to MS is

absolutely essential to benefit from this sensitivity

capability.

As a post-translational modification will cause a change in the mass

of protein, specifically to the amino acid it is attached to, mass spec-

trometry can detect this. Further, tandem mass spectrometry techni-

ques such as a precursor ion scan and a neutral loss scan can be used

to identify peptides that contain a certain PTM within a complex mix-

ture. Once the PTM containing peptide has been detected, a product

ion tandem mass spectrometry scan can be performed to determine

the location of the PTM within that particular peptide (see Fig. 3.19).

3.4.1

Phosphorylation Analysis

Protein phosphorylation is arguably the most studied PTM. Unlike

N-linked glycosylation, there is no specific sequon which indicates

phosphorylation, though phosphorylation commonly occurs on ser-

ine, threonine and tyrosine and less commonly on histidine residues.

Many methods have been reported and examples exist where the state

of phosphorylation of a single protein has been determined. However,

the mapping of an entire phosphoproteome and relative quantitative

analysis of phosphoproteins could be considered something of a holy

grail in current proteomics method development and here has been

considerable development in pursuit of this goal (Sachon et al. 2006;

Goshe et al. 2001; Blagoev et al. 2004; Gruhler et al. 2004; Tao et al.

2005; Zhou et al. 2001; Reinders et al. 2005).

Strategies that are specifically designed to identify sites of phos-

phorylation, attempt to isolate or at least enrich for phosphopeptides

or phosphoproteins upstream of tandem mass spectrometry.

& Tandem mass spectrometry is the key tool for

identifying the site of phosphorylation.

The isolation/enrichment method is either based on an affinity chro-

matography step or a chemical derivatization approach and most

approaches to date are applied at peptide level (Figure 3.29).

Sachon E, Mohammed S,

Bache N, Jensen ON. Rapid-

Commun Mass Spectrom 20

(2006) 1127–1134.

Goshe MB, Conrads TP,

Panisko EA, Angell NH, Veen-

stra TD, Smith RD. Anal Chem

73 (2001) 2578–2586.

Blagoev B, Ong SE, Kratch-

marova I, Mann M. Nat

Biotechnol 22 (2004) 1139–

1145.

Gruhler A, Olsen JV,

Mohammed S, Mortensen P,

Færgeman NJ, Mann M,

Jensen ON. Mol Cell Proteo-

mics 4 (2005) 310–327.

Tao WA, Wollscheid B, O’Brien

R, Eng JK, Xiao-jun L, Boden-

miller B, Watts JD, Hood L,

Aebersold R Quantitative phos-

phoproteome analysis using a

dendrimer conjugation chem-

istry and tandem mass spectro-

metry. Nat Methods 2 (2005)

591–598.

Zhou H, Watts JD, Aebersold

R. Nat Biotechnol. 19 (2001)

375–337.

Reinders J, Sickmann A.

Proteomics 5 (2005) 4052–

4061.

Though for some simpler

samples, it maybe possible to

employ a shotgun method,

fractionating the sample prior

to tandem mass spectrometry.