Clinical Chemistry 52: 1223-1237, 2006.
First published April 27, 2006; 10.1373/clinchem.2006.069252
(Clinical Chemistry. 2006;52:1223-1237.)
© 2006 American Association for Clinical Chemistry, Inc.
The MALDI-TOF Mass Spectrometric View of the Plasma Proteome and Peptidome
Glen L. Hortin
1 Department of Laboratory Medicine, National Institutes of Health, Bldg 10, Room 2C-407, Bethesda, MD 20892. Fax 301-402-1885; e-mail ghortin{at}mail.cc.nih.gov.
 |
Abstract
|
|---|
Background: Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) and the related technique, surface-enhanced laser desorption/ionization (SELDI)-TOF MS, are being applied widely to analyze serum or plasma specimens for potential disease markers.
Methods: Reports on the basic principles and applications of MALDI-TOF MS were reviewed and related to information on abundance and masses of major plasma proteins.
Outcomes: MALDI-TOF MS is a particle-counting method that responds to molar abundance, and ranking of plasma proteins by molar abundance increases the rank of small proteins relative to traditional ranking by mass abundance. Detectors for MALDI-TOF MS augment the bias for detecting smaller components by yielding stronger signals for an equivalent number of small vs large ions. Consequently, MALDI-TOF MS is a powerful tool for surveying small proteins and peptides comprising the peptidome or fragmentome, opening this new realm for analysis. It is complementary to techniques such as electrophoresis and HPLC, which have a bias for detecting larger molecules. Virtually all of the potential markers identified by MALDI-TOF MS to date represent forms of the most abundant plasma proteins.
Conclusions: Analyses of serum or plasma by MALDI-TOF MS provide new information mainly about small proteins and peptides with high molar abundance. The spectrum of observed proteins and peptides suggests value for applications such as assessment of cardiovascular risk, nutritional status, liver injury, kidney failure, and systemic immune responses rather than early detection of cancer. Extending analysis by MALDI-TOF MS to lower abundance components, such as markers for early-stage cancers, probably will require more extensive specimen fractionation before analysis.
|
Basic Principles of Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry
|
|---|
Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS), invented in the late 1980s (1)(2), serves as a technique for analyzing peptides and proteins in relatively complex samples. It has even been used for the direct analysis of tissue specimens(3). The importance of MALDI-TOF MS has been recognized by a share of the 2002 Nobel Prize in Chemistry to K. Tanaka for its invention.
In MALDI-TOF MS, a small amount, typically 
1 µL, of specimen containing peptides and protein is dried on a target plate together with a light-absorbing matrix molecule (1)(2). Vaporizing of the mixture of protein and matrix from surface deposits by nanosecond-duration laser pulses releases ionized protein molecules, often with a single charge (z =1), which are accelerated in an electric field within a vacuum. Ions with low mass/charge ratios (m/z) are accelerated to higher velocities and reach the detector before ions with a high m/z, according to the relationship: time = k(m/z)0.5, where k is a constant determined by calibration with protein standards. Analyses are independent of factors other than mass because separations occur in a vacuum. Ions reach the detector over a timespan of 0.01–1 ms; the millisecond timescale to analyze ions produced by a single laser pulse allows acquisition of data from multiple laser pulses for each specimen and provides short analysis time per specimen. Therefore, the technique lends itself to high-throughput analysis.
Two technical advancements improved resolution of MALDI-TOF MS to its current state. First, use of an electronic mirror, or reflectron, to reflect ions substantially increases resolution. A drawback for some analyses is decreased transmission of ions to the detector; therefore, some analyses continue to be performed in a linear ion mode. Second, delayed extraction introduces a delay after the time of sample vaporization before the electric potential is applied (4). Shorter delay times are optimal for small molecules, and longer delay times for large molecules. Although delayed extraction improves resolution, selection of a delay time optimizes analysis for a specific range of m/z. Recent application of orthogonal TOF ion optics may help broaden the optimal range of analysis(5). Advances in MALDI-TOF MS allow high-performance instruments to measure the masses of peptides with a relative molecular mass (Mr) of 1000–3000 with an accuracy approaching 10 parts per million. For a peptide of Mr 2000 and a charge of 1, that corresponds to measuring the mass within 0.02 mass units of the true value. As m/z increases, resolution and mass accuracy progressively decrease. In the low m/z range, <1000, there is higher background from ionized matrix molecules. Therefore, MALDI-TOF MS provides optimal performance for m/z from 1000 to 5000, although it has no absolute upper analytical limit for m/z(6). High-resolution MALDI-TOF MS can determine the monoisotopic masses of peptides with Mr less than 5000. This is the mass of the peptide composed entirely of 1H, 12C, and 14N atoms. Analyses of larger peptides or with low resolution determine the average masses of peptides, which include the contributions of 1.1% of carbon atoms as 13C and 0.37% of nitrogen atoms as 15N. Average masses of peptides and proteins are 0.06% higher than monoisotopic masses. Proteins contain populations of molecules with varying numbers of different isotopic forms of atoms, contributing to broadening of peaks as protein mass increases.
The standard detector for MALDI-TOF mass spectrometers is a microchannel plate, which acts as an electron multiplier for ions reaching the detector. Detector responses relate to the number of ions reaching the detector and ion velocities. Fast ions, with low m/z, produce stronger detector responses than slow ions; therefore, MALDI-TOF MS usually provides more sensitive detection of small molecules. Experiments with a superconducting detector, which detects each ion regardless of the ion velocity (6), demonstrated that the manyfold lower detection sensitivity of most MALDI-TOF mass spectrometers for large proteins is primarily a function of lower detector sensitivity rather than lower ionization efficiency. Broadening of peaks for larger polypeptides and proteins also contributes to lower sensitivity of detection.
Detection characteristics of MALDI-TOF MS and its variant surface-enhanced laser desorption/ionization (SELDI)-TOF MS lead to a strong bias for detection of small peptides and proteins relative to large proteins. As outlined in Table 1
, these detection characteristics differ fundamentally from other widely applied analytical techniques. Most electrophoretic and chromatographic techniques preferentially detect large molecules because absorbance response or the intensity of staining per molar equivalent of protein generally increases in proportion to size. Immunoassay techniques generally yield signals directly related to molar concentration, leading to little bias between measurement of large and small molecules. MALDI-TOF MS thus serves as a complementary technique that provides the most sensitive detection of low–molecular-mass proteins, whereas other traditional methods of analysis have increased or equivalent sensitivity for detecting large proteins. This characteristic helps explain why MALDI-TOF MS has provided new insights into small peptide and protein components.
|
Analysis of Proteins and Peptides by MALDI-TOF MS
|
|---|
MALDI-TOF MS offers rapid determination of molecular masses and the heterogeneity of small amounts of peptides and proteins (7)(8)(9)(10)(11). Usually, intact molecular ions are formed, allowing accurate determination of polypeptide mass. However, some technique-dependent variation in measured mass can occur related to variable loss of labile groups, such as sialic acids or sulfate groups, or association of salts or matrix with protein molecular ions(8)(9)(10)(11)(12)(13). Noncovalently bound subunits of proteins generally dissociate into individual polypeptide chains, whereas peptide chains connected by covalent bonds such as disulfides remain attached. Analyses, therefore, are analogous to electrophoresis under denaturing conditions without reduction of disulfide bonds.
During desorption/ionization of proteins in specimens by laser pulses, both positive and negative molecular ions are formed, usually with singly charged ions predominating (8). The yield of positive ions is greater for most proteins and peptides, which usually are analyzed in the positive ion mode. Although peptides and proteins ionize mainly as singly charged, protonated, molecular ions [m + H]+, small amounts of molecular ions paired with sodium [m + Na]+, matrix, or other ions; protonated dimeric ions [2m + H]+; and doubly charged (z = 2) ions [m + 2H]2+ can form(7)(8)(9)(10)(11). The occurrence of multiple ionic forms complicates the overall pattern and depends on variables of sample composition, sample preparation, and type of matrix used.
The limits of detection for MALDI-TOF MS depend on many variables that may change detection sensitivity by orders of magnitude, so that any general statements about limits of detection must be considered as rough approximations. For small peptides, under optimal conditions, limits of detection can extend to <1 fmol, corresponding to a 1 nmol/L solution in 1 µL. However, the ionization efficiency of peptides is highly variable, with preferential ionization of peptides containing arginine (14). Ionization of peptides is enhanced by chemical conversion of amines to guanidinium groups(15), reflecting the effects of peptide structure on ionization efficiency. Usually, larger peptides and proteins are detected with lower sensitivity, as noted previously. Analysis of large proteins may require concentrations in a specimen of >1 µmol/L. This is particularly true for complex mixtures of proteins such as plasma. There is competition among different proteins and salts in a specimen for ionization, a general phenomenon in MS termed ion suppression, which lowers the yield of specific ions and reduces the sensitivity of detection(16)(17). Therefore, as will be discussed in later sections, most plasma proteins detected by MALDI-TOF MS have concentrations >1 µmol/L unless there is extensive sample fractionation.
The major challenge for many analyses of physiologic fluids is to identify optimal procedures for sample preparation that remove salts and enrich for the protein or peptide of interest. Common approaches for sample preparation include solid-phase extractions, dilution, chromatography, fractionation directly on specialized target surfaces in SELDI-TOF MS, or capture on magnetic particles (18)(19)(20)(21)(22)(23)(24)(25)(26). Capture of proteins by antibodies serves as a rapid and specific method to select and concentrate specific components(27)(28)(29)(30). In SELDI-TOF MS, retentate chromatography is performed directly on the surface of the target plate. The target surface (chip) is modified to contain ion-exchange, hydrophobic, normal-phase, or metal chelate functional groups, and proteins are selectively captured on the surface depending on the protein properties and selection of binding and wash buffers. Because the target surface has a small surface area and binding capacity, only a small proportion of proteins in an applied plasma specimen is bound. Consequently, in SELDI-TOF MS, competition among many plasma components for low-capacity binding sites influences the spectrum of proteins and peptides detected(22)(23)(24). Use of chromatographic columns or magnetic particles with higher surface areas and capacities provides fractionation with less influence from competition for binding sites.
Ionization efficiency and detection sensitivity for different peptides and proteins vary with different matrix molecules. Detection of small peptides generally is favored by use of 
-cyano-4-hydroxycinnamic acid as matrix, small and medium-sized proteins by sinapinic acid, and heavily glycosylated and large proteins by 2,5-dihydroxybenzoic acid or ferulic acid (7)(8)(9)(10).
|
Abundance Ranking of Plasma and Serum Proteins
|
|---|
The number of proteins identified in plasma has expanded over the past few years into the thousands (31)(32)(33)(34). Recent efforts by multiple laboratories for the Plasma Proteome Project have led to the tentative identification of 9504 proteins in plasma by detection of at least 1 tryptic peptide component of the protein(34). Confirmed identifications with 2 or more peptides were obtained for 3020 proteins. This set of 3020 was considered to have confirmed identification and serves as the Core Protein Dataset for the Human Proteome Organization (HUPO) Plasma Proteome Project(34). Although there are thousands of different proteins in plasma, the abundances of individual proteins differ by many orders of magnitude, with 20 abundant proteins constituting >98% of the total mass of protein(32)(33)(34). The ranking of proteins generally has been considered on the basis of mass abundance using units such as grams per liter. However, for MALDI-TOF MS, a particle-counting method, the relevant concentration is molar abundance, with units such as micromoles per liter. Ranking of the top 30 proteins in the plasma of healthy person by molar rather than mass abundance (Table 2
) substantially changes the perspective on relative abundance of many plasma proteins, particularly the small proteins. In this ranking, note also that concentration is expressed as the concentration of noncovalently bound subunits, so that a tetrameric protein such as transthyretin has a 4-fold higher concentration. Some of the components in Table 2
, such as apolipoproteins C-I, C-II, and C-III, are relatively minor plasma components according to mass abundance. Seven of the 30 most abundant plasma components in normal plasma specimens are proteins yielding singly charged ions with m/z <30 000 in MALDI-TOF MS: apolipoproteins A-I, A-II, C-I, C-II, and C-III; transthyretin; and retinol-binding protein.
View this table:
[in this window]
[in a new window]
|
Table 2. Plasma proteins ranked in order of molar concentrations of noncovalently bound subunits in healthy persons.
|
|
Proteins with Mr <30 000 are cleared within hours by renal filtration (35)(36). Therefore, to avoid rapid clearance by the kidney, these plasma components either have multiple subunits, such as transthyretin (also known as prealbumin), which is composed of 4 identical subunits, or are bound to larger carriers, such as apolipoproteins bound to lipoprotein particles or retinol-binding protein associated with transthyretin. A few high–molecular-mass proteins that are relatively major components based on mass abundance, such as IgM and apolipoprotein B, did not quite rank in the top 30 by molar abundance. The rankings in Table 2
should be considered approximations, however, because proteins concentrations can vary in different populations and because well-established reference materials are available only for some of the listed proteins(37). For all of the proteins in Table 2
, 1 µL of plasma contains sufficient protein (>1 pmol) to allow potential detection by MALDI-TOF MS, if the protein can be ionized efficiently. Proteins with high mass or high proportions of carbohydrate generally are most challenging to detect by MALDI-TOF MS(1)(2)(3)(4)(5)(6)(7)(8)(9)(10).
Several additional peptides and proteins become major components depending on how specimens are processed or on physiologic or pathologic processes (Table 3
). If plasma is allowed to clot to form serum, coagulation components such as fibrinogen are depleted, and activation peptides and platelet proteins are released (38)(39)(40)(41)(42). On a molar basis, the most significant changes are increases of fibrinopeptides A and B and removal of most of the fibrinogen. Each fibrinogen molecule releases up to 2 fibrinopeptide A and B molecules(38)(39), giving molar concentrations of fibrinopeptides among the most abundant serum components. Fibrinopeptide A is released faster than fibrinopeptide B, and the fibrinopeptides are trimmed by exopeptidases; hence, the amounts and molecular forms of these short peptides depend on the extent and duration of clotting of an individual specimen(38)(39)(40). Because fibrinogen is an acute-phase reactant, increasing more than 2-fold during inflammation(43)(44), fibrinopeptides can increase correspondingly. Other fragments of fibrinogen also can accumulate in serum or plasma depending on physiologic and specimen-processing variables, but limited information is available about the concentrations of most fragments(45).
In response to inflammation, 2 proteins, C-reactive protein and serum amyloid A, increase in concentration more than 100-fold (43)(44)(46)(47). C-reactive protein is a pentameric protein with noncovalently bound subunits of Mr 23 000, and serum amyloid A is an HDL-associated protein, with multiple forms of Mr 11 000. In response to inflammation or cytokine therapies, these proteins are observed as major peaks by SELDI-TOF MS(30). Reported concentrations of serum amyloid A in severe inflammation(46)(47) rank second only to albumin on a molar basis. Multiple molecular forms of serum amyloid A, related to trimming of the peptide chain by exopeptidase, have been noted by MALDI- or SELDI-TOF MS(30)(46)(47)(48)(49)(50).
Inflammatory responses also produce changes in other proteins. Concentrations of 1-proteinase inhibitor, 1-acid glycoprotein, 1-antichymotrypsin, fibrinogen, and haptoglobin increase more than 2-fold, whereas other components, such as albumin, transferrin, transthyretin, and apolipoproteins, decrease (30)(43)(44). The concentrations of some cytokines increase dramatically in response to inflammation, but cytokines have such low concentrations that these proteins are difficult to detect by SELDI- TOF MS(30).
In renal failure, small proteins that usually are cleared by renal filtration accumulate. One prominent example is ß2-microglobulin, a small protein of Mr 11 729 that builds up in plasma during renal failure (51)(52)(53). Proteolytically modified forms of ß2-microglobulin are observed as the half-life of the protein is extended in patients with renal failure(54)(55). Other small proteins that are cleared substantially by renal filtration accumulate during renal failure to concentrations severalfold above normal; examples of such proteins are cystatin C, complement factor D, free 
and 
immunoglobulin light chains, and retinol-binding protein(53). Peptide fragments of fibrinogen also accumulate(56).
During standard procedures for collection and processing of blood specimens, there is some breakage of erythrocytes and release of hemoglobin. For a plasma specimen collected in evacuated blood collection tubes with good technique, hemoglobin concentrations range from 9 to 120 mg/L (57), which correspond to 0.3–4 µmol/L each of the hemoglobin - and ß-chains. Concentrations are slightly higher in serum(57). Peaks attributed to hemoglobin chains have been observed by MALDI-TOF or SELDI-TOF MS(30)(58)(59). Specimens with a visible red tint attributable to hemolysis have a hemoglobin concentration >200 mg/L, corresponding to >6 µmol/L of - and ß-chains. The abundance of hemoglobin will depend on the technique and skill of the person collecting the blood, the processing procedures, and on pathologic conditions that can lead to increased hemolysis in vivo(60).
Several additional proteins selectively increase in immune responses and disorders. Responses to infection or expansion of leukocyte populations in leukemia can substantially increase plasma concentrations of the Mr 14 693 protein lysozyme, which usually has concentration well below 1 µmol/L (61)(62). Plasma cell disorders can markedly increase concentrations of free immunoglobulin light chains(63)(64).
|
MALDI-TOF MS Profiles of Unfractionated Serum
|
|---|
Unfractionated serum has been analyzed by MALDI-TOF MS [Ref. (19), and Hortin and Remaley, submitted for publication]. The observed profiles of peaks for unfractionated serum specimens are largely in agreement with the molar abundances of plasma proteins. The profiles are dominated by the most abundant components. Major peaks are observed corresponding to IgG, albumin dimer, transferrin, albumin, doubly charged albumin, apolipoprotein A-I, transthyretin, apolipoprotein C-III, apolipoprotein C-II, and apolipoprotein C-I. Serum or plasma must be diluted for this analysis to provide an adequate ratio of matrix to protein. Analysis of serum or plasma by SELDI-TOF MS offers some fractionation of components before MS, but unless immunocapture is applied, the efficiency of fractionation generally is modest, so that the major serum or plasma components predominate(30).
|
Mass Assignments of Plasma Components
|
|---|
There is a problem in assigning the masses of many plasma proteins from protein sequence databases because the databases often provide calculated masses for precursor forms of the proteins and do not correct for posttranslational modifications. Glycoproteins represent a challenge in mass assignment in that there may be heterogeneity in the numbers and structures of oligosaccharides structures; structures also may vary in different physiologic or pathologic states. In MALDI-TOF MS, there also can be variable and method-dependent loss of sialic acids or other labile groups (7)(8)(9)(10)(11)(12)(13). The accuracy of mass assignments by MALDI-TOF MS, therefore, generally is greater for proteins with little or no carbohydrate. Mass estimates of proteins with many sialic acids, such as 1-acid glycoprotein, by MALDI-TOF MS may be lower than values calculated from detailed knowledge of the peptide and oligosaccharide structures and may depend on method of analysis. Therefore, glycoprotein masses in MALDI-TOF or SELDI-TOF MS analysis should be determined for the specific method of analysis.
The masses of nonglycosylated plasma proteins can be predicted accurately from knowledge of the primary amino acid sequence and posttranslational processing, including formation of disulfides. Heterogeneity is observed for many components because of exopeptidase action on the peptide chain, sequence variations, and variable modifications of free sulfhydryl groups of unpaired cysteines (19)(20)(28)(65)(66)(67). In polypeptides with unpaired cysteine residues, such as subunits of transthyretin, apolipoprotein A-II, 1-proteinase inhibitor, and albumin, disulfides may form with a variety of compounds, such as cysteine, cysteinylglycine, glutathione, or other polypeptides (dimerization of apolipoprotein A-II), or the sulfhydryl may undergo oxidation(19)(20)(28)(65)(66)(67). Proteins with unpaired cysteines also may link via disulfides to other proteins, as in the case of 1-microglobulin, such that a mixture of different covalently bound forms of the protein is observed(68). Therefore, identification of unpaired cysteine residues is of major importance in establishing the masses and structural variations of proteins.
Single-base variations in the genes encoding proteins introduce variable mass changes, depending on the resulting amino acid substitution. Substitutions of leucine for isoleucine do not change protein mass, and substitutions of glutamine for lysine or the reverse substitutions change protein mass by only 0.04 mass units, so that they generally would not be detected by MALDI-TOF MS. Approximately 90% of possible amino acid substitutions from single-base variations will lead to mass changes greater than 8 mass units (69). Detection of gene-encoded variations or posttranslational modifications by MALDI-TOF MS depends on the magnitude of the mass change, the mass of the peptide or protein of interest, and the resolution that can be achieved. The frequency of mass variation for 25 different proteins analyzed by MALDI-TOF MS recently has been described(28). Specimens from 96 individuals were analyzed. Most of the reported variation was related to variation in posttranslational processing. Variants were noted for 4 proteins: transthyretin, apolipoprotein E, cystatin C, and serum amyloid A.
|
The Low–Molecular-Mass Proteome or Peptidome
|
|---|
Ultrafiltrates of plasma have been recognized for several years to contain complex mixtures of thousands of peptides (70) that are smaller than apolipoprotein C-I, the smallest major plasma protein. However, these peptides have very low concentrations and were characterized through the analysis of thousands of liters of plasma ultrafiltrate. Application of SELDI-TOF MS to analysis of serum or plasma detected low–molecular-mass components from specimens of a few microliters(71)(72), and great interest in this technique was stimulated by the potential for patterns of small peptides to serve as diagnostic markers(5)(22)(23)(24)(25)(26)(71)(72)(73)(74)(75)(76). Most studies applying MALDI-TOF or SELDI-TOF MS to analyze small amounts of serum or plasma have resolved 100 to 1000 peaks in the m/z range <20 000. A comparison of different analytical platforms showed that the number of peaks that can be resolved depends on the resolution of the mass spectrometer and the degree of specimen fractionation(76).
Apparently, concentrations of peptides bound to carrier molecules are much larger than concentrations of free peptides originally analyzed in plasma ultrafiltrates (70). The complete set of small peptides has been termed the peptidome, or the fragmentome because it has been recognized that most small peptide components of plasma are derived from proteolytic degradation of larger proteins. Peptide fragments of many major plasma proteins, such as fibrinogen, apolipoproteins, transthyretin, and complement factors, have been detected(70)(73)(74)(75)(76)(77)(78)(79). Families of overlapping peptides have been observed in many studies, suggesting the release of peptides from larger parent proteins by endoproteolytic cleavage followed by variable trimming of released peptides by aminopeptidases and carboxypeptidases(70)(74)(75)(76)(77)(78)(79). Polypeptides smaller than Mr 7000 are not detected by the usual techniques of 2-dimensional electrophoresis because they are below the limits of size resolution, and small components may not be fixed in gels and produce lower staining intensity per mole of polypeptide. Consequently, MALDI-TOF MS has served as a path-breaking tool to open up peptidomic analysis in the mass range from Mr 1000 to 7000. This mass range might be considered as an approximate, operationally defined mass range (within optimal detection by MALDI-TOF MS but below usual detection by 2-dimensional gel electrophoresis) for the peptidome; there is no current consensus on what constitutes mass limits for the peptidome or a size cutoff that distinguishes peptides from proteins.
Because peptide fragments, such as the fibrinopeptides, are released in large amounts and an entire cascade of proteases is activated during the clotting of plasma to form serum, the suitability of serum for analysis of peptide profiles has been questioned (79)(80)(81)(82)(83). Recently, a counterargument has been raised that pathophysiologic changes might be reflected in patterns of fragments released during clotting, so that serum may contain greater diagnostic information(74)(75). However, it is hard to envision how the conditions for clot activation and processing of blood specimens can be standardized sufficiently to control the method-dependent variation in the generation of peptides in serum. Evaluations of specimen-handling variables have identified several factors that should be considered for proteomics studies, including the potential value of addition of protease inhibitors to specimens and potential interferences from additives to blood collection tubes(34)(40)(79)(80)(81)(82)(83)(84)(85). One study notably found that MS profiles had a better ability to distinguish sample sources than disease vs control populations, and possible influences of specimen collection variables have called into question the validity of some early reports on the use of mass spectrometric profiles for cancer diagnosis(83). At this point, it is clear that control of preanalytical variables is a critical element in the design of studies analyzing the peptide components of serum or plasma.
There is limited quantitative information about the complex mixture of bound peptides in plasma, but albumin has a potential capacity to bind more than 500 µmol/L, if it binds 1 peptide per molecule. To date, albumin-bound peptides have been the major set examined (5)(73)(74)(75). Analysis of the detection limits for a peptide in serum specimens suggests that peptides can be detected down to 20 nmol/L by use of limited fractionation on SELDI-TOF MS(86). A reported analysis of a complex mixture of peptides bound to HDLs estimated that recovered small peptides had an aggregate molar concentration one fifth that of apolipoprotein A-I(77), so that lipoproteins may be another significant carrier of small peptides, just as they are carriers of several small plasma proteins. Detection of small peptide components in plasma specimens has not been achieved in unfractionated plasma(20) or lipoproteins(77), indicating that, without some fractionation of plasma, detection of peptides is suppressed by the more abundant larger proteins. Strategies to increase the sensitivity of MALDI-TOF MS to detect small peptides have relied on more extensive sample fractionation. Sequential removal of protein components by ultrafiltration of milliliter-volume specimens and separation by HPLC into 96 fractions provided detection of peptides such as insulin and C-peptide, whose concentrations are well below nanomolar in plasma specimens(87). This approach resolved an estimated 1500–3000 peptides in the combined analysis of the 96 fractions(85). Another alternative, immunocapture, concentrates and purifies low-abundance components that would otherwise be difficult to detect(27)(28)(29)(30)(66)(67)(88). Many efforts at sequence characterization of small peptide components have relied on HPLC-electrospray MS rather that MALDI-TOF MS because of the superior sensitivity and high-throughput sequencing provided by HPLC-electrospray MS(70)(73)(74)(77), which has been widely applied for analysis of tryptic peptides for bottom-up proteomics(33)(34).
|
Circulating Half-Lives of Proteins and Peptides
|
|---|
The circulating half-lives of proteins and peptides are important factors in the steady-state concentrations of these components in blood. As evident from several examples listed in Table 4
, the half-lives of individual proteins and peptides vary from days to a few minutes, and protein binding to carriers is a general mechanism for extending the circulating half-lives of plasma proteins. Most larger proteins have half-lives of several days. Proteins circulating as a molecule or complexes with Mr <30 000 are cleared by filtration in the kidneys (35)(36), giving half-lives of a few hours for proteins such as ß2-microglobulin and immunoglobulin light chains. The half-lives of other proteins of similar mass, such as apolipoproteins A-I and A-II and retinol-binding protein, are extended severalfold by binding to larger carriers. Notably however, the half-life of retinol-binding protein is significantly shorter than that of its carrier protein, indicating that some is lost through exchange off of its carrier. Many small peptides, as exemplified by parathyroid hormone(89), have half-lives of only a few minutes because of proteolytic degradation or other clearance mechanisms besides glomerular filtration, which would require hours. Fibrinopeptide B has a half-life of only a few minutes in a test tube when clotting is activated(38). Pharmacologically, linking small peptides or proteins to carriers such as polyethylene glycol or adding tags that bind to albumin are recognized as strategies to greatly extend the circulating half-life(90)(91). The concentrations of small peptide fragments of proteins on carrier molecules ultimately depend on several factors, including the rate of formation, the half-life of the carrier molecule, and the affinity for the carrier molecule.
|
Identification of Potential Biomarkers
|
|---|
Initial reports of the discovery of potential serum or plasma biomarkers by MALDI-TOF or SELDI-TOF MS relied on analysis of profiles of peaks without identifying which molecules corresponded to specific peaks (22)(23)(71)(72). With progress in approaches to identifying peptide sequences, an increasing number of studies have identified the protein corresponding to peaks that are potential biomarkers. Table 5
summarizes proteins identified in 24 different studies, including 18 that analyzed serum as a specimen and 6 that analyzed plasma. In 20 of these reports, the only markers identified were derived from abundant plasma proteins listed in Tables 2
or 3
. Of the 4 reports identifying lower abundance components, 1 identified 2 small platelet proteins that are released during clotting to form serum to yield a serum concentration of 0.1–1 µmol/L(92). The report by Villanueva et al.(74) identified more than 60 peptides serving as potential cancer biomarkers. Most of the peptides were derived from abundant proteins such as fibrinogen; C3; inter--trypsin inhibitor; apolipoproteins A-I, A-IV, and E; and transthyretin listed in Table 2
. Only a few peptides were derived from less abundant proteins such as C4, clusterin (also known as apolipoprotein J), high–molecular-mass kininogen, and coagulation factor XIII. Of these, factor XIII subunits, with a plasma abundance of 0.1 µmol/L, represent the lowest abundance component(31). A study of peptide changes during penile erection also noted increased amounts of an activation peptide from coagulation factor XIII(93). The only report noting peptides that were not derived from relatively abundant large plasma proteins identified changes in insulin and C-peptide expression; both of these peptides had concentrations of 0.1–1 nmol/L(87).
View this table:
[in this window]
[in a new window]
|
Table 5. Biomarkers identified in 24 published studies using MALDI-TOF or SELDI-TOF MS (18 studies examined serum specimens and 6 studies examined plasma).
|
|
The results summarized in Table 5
show that most studies to date have analyzed forms of only 50 of the most abundant proteins in plasma. In addition, every protein and peptide identified in the studies above, except for hemoglobin subunits, was a secretory protein. Therefore, all of the proteins other than hemoglobin would be subjected to processing along the secretory pathway. Most protein sequence databases would yield incorrect calculated masses for the intact proteins because of failure to account for protein processing, although calculations of the masses for small peptide fragments may be correct depending on the occurrence of posttranslational modifications within the peptide fragment. The shortcomings of existing protein databases underscore the need to develop databases of actual measured masses for components of interest.
Considering that most studies of serum and plasma proteins and peptides by MALDI-TOF MS have analyzed forms of only 50 of the most abundant proteins from among the several thousand that have been detected in plasma (32)(33)(34), there remain great unexplored opportunities for probing the proteome and peptidome. Efforts to probe more deeply into the plasma proteome and peptidome, to detect lower abundance components by MALDI-TOF MS, are likely to rely on increased sample fractionation of specimens by chromatographic or immunocapture techniques as well as advances in instrumentation(5)(27)(28)(29)(30)(69)(73)(74)(85)(87)(88). Immunodepletion of several of the most abundant plasma protein components has assisted with the proteomic analysis of lower abundance components in many studies(33)(34). However, in the analysis of the peptidome, capture of high-abundance proteins may serve a different role—identification of which major proteins serve as carriers of specific peptides and for fractionation and enrichment of peptides from the complex mixture of plasma components.
|
Controversies Relating to Reproducibility and Informatics
|
|---|
MALDI-TOF MS analyzes a layer of dry crystals composed of matrix and specimen on a target surface. The inhomogeneous specimen on the target surface, the many variables in sample preparation, and selective sampling by vaporization of small portions of the specimen with laser pulses present challenges in obtaining high reproducibility and attaining standards of practice appropriate for clinical laboratory tests (94)(95)(96). Addressing the many controversial issues relating to MALDI-TOF, such as informatics, marker discovery, quantitative analysis, and reproducibility, are beyond the scope of this review, and the reader is referred to other recent reports(23)(74)(75)(76)(94)(95)(96)(97)(98)(99)(100)(101).
|
Conclusions
|
|---|
The bias of MALDI-TOF MS for detection of low–molecular-mass peptides and proteins has helped to open a new realm of plasma components, comprising the fragmentome or peptidome, that were largely overlooked by traditional techniques. Studies of peptide fragments should advance basic understanding of pathways for protein degradation and clearance. The high complexity of protein fragments in plasma and their interactions with carrier proteins potentially offer large amounts of information of pathophysiologic significance that may be applied for diagnostic purposes.
Future efforts at quantitative analysis of proteins and peptides by MS should consider molar concentrations, as this is the relevant quantity for particle-counting methods and allows clearer expression of the relative concentrations of components that differ greatly in molecular size. Expression of the concentrations of small peptide fragments in molarity also allows better comparison of the quantitative relationship between the peptides, their parent proteins, and their carrier molecules. Development of expanded databases about the molar concentrations and measured masses of the most abundant proteins and peptides in the circulation would expedite future discovery efforts.
In analyzing complex mixtures by MALDI-TOF MS, components of highest molar abundance generally dominate the spectrum and tend to suppress detection of lower abundance proteins. Studies using MALDI- and SELDI-TOF MS to identify potential biomarkers for disease or physiologic processes to date have primarily identified qualitative or quantitative changes in a few abundant plasma proteins. As expressed by Diamandis et al. (100)(101) and Koomen et al.(79), this raises questions about the ability of current approaches for protein profiling by MALDI-TOF MS to detect any low-abundance cancer markers analogous to prostate-specific antigen or carcinoembryonic antigen. Changes of major plasma components related to cancer that have been detected to date may reflect relatively nonspecific systemic responses to cancer(100)(101). Application of improved fractionation techniques before MALDI-TOF MS may be necessary before this approach can probe low-abundance components that are likely to serve as more specific markers of cancer.
Considering that MALDI-TOF and SELDI-TOF MS assess primarily the most abundant small plasma proteins and peptides, logically these techniques are likely to be of greatest value as indicators of systemic processes such as atherosclerosis, nutrition, stroke, major organ injury, hemostatic disorders, acute-phase responses, and kidney dysfunction rather than small, localized disease processes such as early-stage cancer. Useful applications of MALDI-TOF MS to study systemic disorders will not necessarily require the analysis of low-abundance plasma protein components. The ability to identify qualitative and quantitative changes and degradation pathways of multiple abundant plasma proteins is likely to be of great diagnostic value for many systemic physiologic and pathologic processes.
|
Acknowledgments
|
|---|
This work is supported by the intramural program of the NIH Clinical Center. I thank Drs. Alan Remaly and Steven Drake for comments.
|
References
|
|---|
- Tanaka K, Waki H, Ido Y, Akita S, Yoshida Y, Yoshida T. Protein and polymer analyses up to m/z 100,000 by laser ionization time-of-flight mass spectrometry. Rapid Commun Mass Spectrom 1988;2:151-153.[CrossRef]
- Karas M, Hillenkamp F. Laser desorption ionization of proteins with molecular masses exceeding 10,000 daltons. Anal Chem 1988;60:2299-2301.[Medline]
[Order article via Infotrieve]
- Reyzer ML, Caprioli RM. MALDI mass spectrometry for direct tissue analysis: a new tool for biomarker discovery. J Proteome Res 2005;4:1138-1142.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
- Brown RS, Lennon JJ. Mass-resolution improvement by incorporation of pulsed ion extraction in a matrix-assisted laser desorption/ionization linear time-of-flight mass spectrometer. Anal Chem 1995;67:1998-2003.[Medline]
[Order article via Infotrieve]
- Lopez MF, Mikulskis A, Kuzdzal S, Bennett DA, Kelly J, et al. High-resolution serum proteomic profiling of Alzheimer disease samples reveals disease-specific, carrier-protein-bound mass signatures. Clin Chem 2005;51:1946-1954.[Abstract/Free Full Text]
- Twerenbold D, Gerber D, Gritti D, Gonin Y, Netuschill A, Rossel F, et al. Single molecule detector for mass spectrometry with mass independent detection efficiency. Proteomics 2001;1:66-69.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
- Chait BT, Kent SBH. Weighing naked proteins: practical, high-accuracy mass measurement of peptide and proteins. Science 1992;257:1885-1894.[Abstract/Free Full Text]
- Karas M, Gluckmann M, Schafer J. Ionization in matrix-assisted laser desorption/ionization: singly charged molecular ions are the lucky survivors. J Mass Spectrom 2000;35:1-12.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
- Sottani C, Fiorentino M, Minoia C. Matrix performance in matrix-assisted laser desorption/ionization for molecular weight determination in sialyl and non-sialyl oligosaccharide proteins. Rapid Commun Mass Spectrom 1997;11:907-913.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
- Belgacem O, Buchacher A, Pock K, Josic D, Sutton C, Rizzi A, et al. Molecular mass determination of plasma-derived glycoproteins by ultraviolet matrix-assisted laser desorption/ionization time-of-flight mass spectrometry with internal calibration. J Mass Spectrom 2002;37:1118-1130.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
- Sekiya S, Wada Y, Tanaka K. Derivatization for stabilizing sialic acids in MALDI-MS. Anal Chem 2005;77:4962-4968.[Medline]
[Order article via Infotrieve]
- Wolfender JL, Chu F, Ball H, Wolfender F, Fainzilber M, Baldwin MA, et al. Identification of tyrosine sulfation in Conus pennaceus conotoxins a-PnIA and a-PnIB: further investigation of labile sulfo- and phosphopeptides by electrospray, matrix-assisted laser desorption/ionization (MALDI) and atmospheric pressure MALDI mass spectrometry. J. Mass Spectrom 1999;34:447-454.
- Onnerfjord P, Heathfield TF, Heinegard D. Identification of tyrosine sulfation in extracellular leucine-rich repeat proteins using mass spectrometry. J Biol Chem 2004;279:26-33.[Abstract/Free Full Text]
- Krause E, Wenschuh H, Jungblut PR. The dominance of arginine-containing peptides in MALDI-derived tryptic mass fingerprints of proteins. Anal Chem 1999;71:4160-4165.[Medline]
[Order article via Infotrieve]
- Cockrill SL, Foster KL, Wildsmith J, Goodrich AR, Dapron JG, Hassell TC, et al. Efficient micro-recovery and guanidination of peptides directly from MALDI target spots. Biotechniques 2005;38:301-304.[ISI][Medline]
[Order article via Infotrieve]
- Annesley TM. Ion suppression in mass spectrometry. Clin Chem 2003;49:1041-1044.[Abstract/Free Full Text]
- Kratzer R, Eckerskorn C, Karas M, Lottspeich F. Suppression effects in enzymatic peptide ladder sequencing using ultraviolet-matrix assisted laser desorption/ionization mass spectrometry. Electrophoresis 1998;19:1910-1919.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
- Gilar M, Bouvier ESP, Compton BJ. Advances in sample preparation in electromigration, chromatographic, and mass spectrometric separation methods. J Chromatogr A 2001;909:111-135.[Medline]
[Order article via Infotrieve]
- Kiernan UA, Black JA, Williams P, Nelson RW. High-throughput analysis of hemoglobin from neonates using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Clin Chem 2002;48:947-949.[Free Full Text]
- Jin Y, Manabe T. Direct targeting of human plasma for matrix-assisted laser desorption/ionization and analysis of plasma proteins by time-of-flight-mass spectrometry. Electrophoresis 2005;26:2823-2834.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
- Hortin GL, Meilinger B, Drake SK. Size-selective extraction of peptides from urine for mass spectrometric analysis. Clin Chem 2004;50:1092-1095.[Free Full Text]
- Tang N, Tornatore P, Weinberger SR. Current developments in SELDI affinity technology. Mass Spectrom Rev 2004;23:34-44.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
- Simpkins F, Czechowicz JA, Liotta L, Kohn EC. SELDI-TOF mass spectrometry for cancer biomarker discovery and serum proteomic diagnostics. Pharmacogenomics 2005;6:647-653.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
- Aivado M, Spentzos D, Alterovitz G, Out HH, Grall F, Giagounidis AA, et al. Optimization and evaluation of surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS) with reversed-phase protein arrays for protein profiling. Clin Chem Lab Med 2005;43:133-140.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
- Villaneuva J, Philip J, Entenberg D, Chaparro CA, Tanwar MK, Holland EC, et al. Serum peptide profiling by magnetic particle-assisted, automated sample processing and MALDI-TOF mass spectrometry. Anal Chem 2004;76:1560-1570.[Medline]
[Order article via Infotrieve]
- Baumann S, Ceglarek U, Fiedler GM, Lembcke J, Leichtle A, Thiery J. Standardized approach to proteome profiling of human serum based on magnetic bead separation and matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Clin Chem 2005;51:973-980.[Abstract/Free Full Text]
- Xiao Z, Adam BL, Cazares LH, Clements MA, Davis JW, Schellhammer PF, et al. Quantitation of serum prostate-specific membrane antigen by a novel protein biochip immunoassay discriminates benign from malignant prostate disease. Cancer Res 2001;61:6029-6033.[Abstract/Free Full Text]
- Nedelkov D, Kiernan UA, Niederkofler EE, Tubbs KA, Nelson RW. Investigating diversity in human plasma proteins. Proc Natl Acad Sci U S A 2005;102:10852-10857.[Abstract/Free Full Text]
- Zhu Y, Valdes R, Jr, Jortani SA. Application of bioaffinity mass spectrometry for analysis of ligands. Ther Drug Monit 2005;27:694-699.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
- Rossi L, Martin B, Hortin GL, White R, Foster M, Stroncek D, et al. Inflammatory protein profile during systemic high dose interleukin-2 administration. Proteomics 2006;6:709-720.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
- Putnam FW. Alpha, beta, gamma, omega—the roster of the plasma proteins. Putnam FW eds. The Plasma Proteins, 2nd ed 1975;Vol. 1:57-130 Academic Press New York. .
- Anderson NL, Anderson NG. The human plasma proteome: history, character, and diagnostic prospects. Mol Cell Proteomics 2002;1:845-867.[Abstract/Free Full Text]
- Anderson NL, Polanski M, Pieper R, Gatlin T, Tirumalai RS, Conrads TP, et al. The human plasma proteome: a nonredundant list developed by combination of four separate sources. Mol Cell Proteomics 2004;3:311-326.[Abstract/Free Full Text]
- Omenn GS, States DJ, Adamski M, Blackwell TW, Menon R, Hermjakob H, et al. Overview of the HUPO Plasma Proteome Project: results from the pilot phase with 35 collaborating laboratories and multiple analytical groups, generating a core dataset of 3020 proteins and a publicly-available database. Proteomics 2005;5:3226-3245.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
- Heinemann HO, Maack T, Sherman RI. Proteinuria. Am J Med 1974;56:71-81.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
- Brenner BM, Hostetter TH, Humes HD. Molecular basis of proteinuria of glomerular origin. N Engl J Med 1978;298:826-833.[ISI][Medline]
[Order article via Infotrieve]
- Dati F, Schumann G, Thomas L, Aguzzi F, Baudner S, Bienvenu J, et al. Consensus of a group of professional societies and diagnostic companies on guidelines for interim reference ranges for 14 proteins in serum based on the standardization against the IFCC/BCR/CAP reference material (CRM 470). Eur J Clin Chem Clin Biochem 1996;34:517-520.[ISI][Medline]
[Order article via Infotrieve]
- Brummel KE, Butenas S, Mann KG. An integrated study of fibrinogen during blood coagulation. J Biol Chem 1999;274:22862-22870.[Abstract/Free Full Text]
- Profumo A, Turci M, Damonte G, Ferri F, Magatti D, Cardinali B, et al. Kinetics of fibrinopeptide release by thrombin as a function of CaCl2 concentration: different susceptibility of FPA and FPB and evidence for a fibrinogen isoform-specific effect at physiological Ca2+ concentrations. Biochemistry 2003;42:12335-12348.[CrossRef][Medline]
[Order article via Infotrieve]
- Banks RE, Stanley AJ, Cairns DA, Barrett JH, Clarke P, Thompson D, et al. Influences of blood sample processing on low-molecular-weight proteome identified by surface-enhanced laser desorption/ionization mass spectrometry. Clin Chem 2005;51:1637-1649.[Abstract/Free Full Text]
- Coppinger JA, Cagney G, Toomey S, Kislinger T, Belton O, McRedmond JP, et al. Characterization of the proteins released from activated platelets leads to localization of novel platelet proteins in human atherosclerotic lesions. Blood 2004;103:2096-2104.[Abstract/Free Full Text]
- Lundblad RL. The Evolution from Protein Chemistry to Proteom
Top
Abstract
Basic Principles of Matrix...
