| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Proteomics and Protein Markers |
1 BioVisioN AG, Hannover, Germany.2 Institute of Pathology, Medical School Hannover, Hannover, Germany.
aAddress correspondence to this author at: BioVisioN AG, Feodor Lynen Strasse 5, 30625 Hannover, Germany. Fax 49-511-53889666; e-mail h.tammen{at}biovision-discovery.de.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
Methods: We used differential peptide display (DPD) to compare comprehensive peptide maps from pairs of serum samples from healthy volunteers. Peptides were separated by liquid chromatography, and fractions were subjected to mass spectrometry. Mass spectra of all fractions were combined, giving a peptide map representing a two-dimensional display of peptide masses. After comparison of peptide mass maps, peptides that differentiated FXIIIA phenotypes were identified by mass spectrometry.
Results: Val34Leu polymorphisms of the activation peptide of FXIIIA were identified in 20 serum samples from 10 volunteers by DPD, and their sequences were confirmed by nanoelectrospray-ionization quadrupole time-of-flight mass spectrometry. Analysis of three (V34V, V34L, and L34L) phenotypes was confirmed by allele-specific genotypic analysis in all (n = 10) volunteers.
Conclusion: DPD provides a simple and easy-to-use phenotype assay with advantages over PCR-based assays in being faster and directly analyzing the compound of interest.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
Differential peptide display (DPDTM; BioVisioN) (7)(8)(9) is a technique that generates comprehensive peptide maps of 
3000 peptides covering a mass range of 950–15 000 Da from a biological sample. The sample is separated by reversed-phase HPLC (RP-HPLC; Fig. 1A
), and the peptides eluting from the HPLC column are collected in 96 fractions. Each fraction is subjected to matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS; Fig. 1B
), and the mass spectra of all 96 fractions are combined, producing a two-dimensional display of peptide masses in which the abscissa displays the mass-to-charge ratio, the ordinate is determined by the retention time on the RP-HPLC column, and the signal intensity is depicted by color saturation (Fig. 1C
). This display method allows the processing and analysis of large amounts of data. Peptide maps of individual samples can be superimposed, facilitating the detection of differences in the resulting subtractive peptide maps (Fig. 1D
). For the identification of peptides, peaks from individual HPLC fractions can be subjected to nanoelectrospray ionization quadrupole time-of-flight mass spectrometry (nESI-qTOF-MS) sequencing, which gives peptide fragment spectra (Fig. 1E
). These spectra serve to identify the corresponding peptide sequences by remote database searching.
|
In the present study, we identified the activation peptide (amino acids 1–37) cleaved from the NH2 terminus of FXIIIA and several truncated variants in serum samples by DPD; the exchange of valine to leucine was detected by a mass difference of 14 Da and a chromatographic shift of four fractions. The peptide sequences were confirmed by nESI-qTOF-MS. We also identified this polymorphism by direct MALDI-TOF-MS measurement in unfractionated serum samples.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
1 h), the tube was centrifuged at 2000g for 10 min at 4 °C. The serum was separated from the clot and stored at -80 °C until analysis.
sample preparation
Before analysis, the protein load in the serum samples is reduced by precipitation with trichloroacetic acid. Serum (0.5 mL) was added to 1 mL of ice-cold distilled water. For protein precipitation, 0.5 mL of a solution containing 200 g/L trichloroacetic acid was added, and the tubes were vigorously mixed for 30 s. After incubating for 30 min on ice, samples were centrifuged at 18 000g for 30 min at 4 °C. The supernatants were transferred to sample vials.
allelic genotyping
Genotyping for the FXIII V34L polymorphism was performed at a professional laboratory for genetic analyses (IMBA, Weinitzen, Austria) by an allele-specific amplification assay (6).
dpd analysis
Serum extracts were loaded on a RP-HPLC column. The peptides were eluted by an acetonitrile gradient (4–40%), and 96 fractions were collected. The lyophilized fractions were resuspended in a mixture of 
-cyano-4-hydroxycinnamic acid (matrix) and 6-desoxy-L-galactose (comatrix) in acetonitrile containing 1.5 g/L trifluoroacetic acid (TFA) and subjected to MALDI-TOF-MS (10) in a Voyager DE STR (Applied Biosystems). Data were analyzed, including peak recognition and visualization, with the in-house-developed software package SpectromaniaTM (BioVisioN). MS signals were quantified after baseline correction by integration of absolute signal intensities in 1-Da bins. The Spectromania software package also transformed the mass spectrum of each fraction to a virtual one-dimensional gel; the molecular weight of each peptide is indicated by its position within the virtual gel lane, whereas the MALDI signal intensity for each peptide is indicated by the color intensity of the corresponding bar. The converted mass spectra of all 96 fractions were combined to give a two-dimensional display of peptide masses, termed a peptide map (Fig. 2
). Peptide maps of individual samples were superimposed, generating mean peptide maps. Differentially expressed peptides were detected by calculation of subtractive peptide maps.
|
identification by nESI-qTOF-tandem MS
Selected peptides were identified by sequencing with a nESI-qTOF tandem mass spectrometer (QSTAR pulsar; Sciex) with subsequent protein database searching. The resulting peptide fragment spectra were produced in the product-ion scan mode (spray voltage, 950 V; collision energy, 28 eV). A total of 200 scans per sample were accumulated. Data processing before database searching included charge state deconvolution (Bayesian reconstruct tool of the BioAnalyst program package; Sciex) and deisotoping (customized Analyst QS macro; Sciex). The resulting spectra were saved in MASCOT (Matrix Science) generic file format and submitted to the MASCOT search engine. Cascading searches including several posttranslational modifications in SwissProt (Ver. 41; www.expasy.ch) and MSDB (Ver. 030212; EBI) were performed by the MASCOT DEAMON client software (Ver. 1.9; Matrix Science).
maldi-tof-ms analysis of unfractionated serum samples
For sample preparation before MALDI-TOF-MS, ZipTip devices (Millipore Corporation) were equilibrated with 10 µL of 1 g/L TFA. Deproteinized serum (5 µL) was aspirated into the device, and the device was washed three times with 1 g/L TFA. Peptides were eluted directly onto the MALDI target by use of a mixture of -cyano-4-hydroxycinnamic acid (matrix) and 6-desoxy-L-galactose (comatrix) in acetonitrile (580 mL/L) containing 1 g/L TFA. MALDI-TOF-MS analysis was performed with a Voyager STR mass spectrometer (Applied Biosystems).
surface-enhanced laser desorption/ionization ms analysis
Surface-enhanced laser desorption/ionization (SELDI)-TOF-MS (11)(12) was performed with WCX chips (weak cation-exchange capacities), SAX chips (anion-exchange capacity), and immobilized metal-affinity capture (IMAC) chips (metal-chelating capacity). Because of the low detection potential of hydrophobic chips and the high costs of the expendable SELDI targets, H4 chips were not used in this study. Approximately 5 µg of protein was diluted 1:10 in the respective binding buffers and applied to the chip by a bioprocessor (Ciphergen Biosystems Inc.). After 30 min, samples were removed, and the chips were washed with the respective washing buffers followed by two quick washes with water. Two 0.7-µL portions of freshly prepared matrix were then added to each spot. Cyano-4-hydroxycinnamic acid (10 g/L) as matrix and 6-desoxy-galactose (10 g/L) as comatrix in 500 mL/L acetonitrile containing 5 g/L TFA were used. WCX chips were preactivated by the addition of 10 mmol/L HCl to the spots followed by three washes with H2O. As binding and washing buffer, 20 mmol/L ammonium acetate, pH 4.0, was used. SAX chips were activated by the addition of 5 µL of 50 mmol/L HEPES buffer, pH 7.5, containing 1 mL/L Triton X-100, which was also used as binding and washing buffer. IMAC chips were pretreated with water and loaded with Ni2+ by use of 20 µL of 100 mmol/L NiSO4. Excess Ni2+ was removed by rinsing with water. As binding and washing buffer, 50 mmol/L Tris, pH 7.5, containing 300 mmol/L NaCl and 0.2 mL/L Triton X-100 was used. Additionally, all samples were analyzed with a gold chip. Extracts were mixed with matrix (see above), applied to the chip, dried, and analyzed in the Ciphergen mass spectrometer.
chemicals
All laboratory chemicals were obtained from Merck AG if not otherwise indicated. All laboratory chemicals for SELDI-MS analysis were obtained from Sigma-Aldrich GmbH.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
2500 peptide sequences are stored, to predict the identities of the peptides. The mass of 3950 Da corresponded to the deduced mass of the FXIIIA activation peptide with valine at position 34; the mass peak at 3964 Da corresponded to a similar peptide with leucine at position 34. In addition, signals with m/z ratios of 3349/3363, 2759/2773, and 2602/2616 with a similar pattern corresponding to the above-mentioned subgroups were detected. nESI-qTOF-MS analysis confirmed the identities of the detected mass peaks. The peak with a mass-to-charge ratio of 3950 Da belonged to the activation peptide of FXIIIA (amino acids 1–37), whereas the peak with a mass-to-charge ratio of 3964 Da also belonged to the activation peptide of FXIII containing leucine at position 34 (Table 1
). The other differentially detected signals belonged to truncated forms of the activation peptide: 3350/3364 belonged to a peptide comprising amino acids 6–37, 2759/2773 to a peptide comprising amino acids 12–37, and 2602/2616 to a peptide comprising amino acids 13–37 (Table 1
). In each case, the 14-Da higher mass corresponded to the Leu-34 variant. These peptides might be generated by action of proteases such as thrombin and plasma kallikrein (Table 1
). Our observation with 10 volunteers shows that in all cases in which the VV phenotype was detected, no peptide signals were detected at the positions of the LL phenotype and vice versa (Fig. 3
). This was true for all four peptides that were derived from the activation peptide. This means that, in all cases of VV phenotype, only peaks with m/z values of 3950, 3349, 2759, and 2602 were detected from the respective pairs. For the LL phenotype, only peaks with m/z values of 3964, 3363, 2773, and 2616 were detected from the respective pairs.
|
|
In Caucasian individuals, 53% are homozygous for valine (VV), 8% are homozygous for leucine (LL), and 39% are heterozygous (VL). In the group we investigated, a similar distribution was determined. As expected, the two heterozygous individuals had lower peak intensities, which indicates lower concentrations of the activation peptide of both phenotypes in the blood of these individuals.
To confirm our results, all 10 blood samples were analyzed by allele-specific genotyping for the FXIIIA Val34Leu polymorphism. The results obtained validated the results obtained by DPD analysis.
detection of the val34leu polymorphism of fxiiia by maldi-tof-ms and seldi-tof-ms
The Val34Leu polymorphism was also detectable in serum samples by a rapid procedure with solid-phase extraction before subsequent MALDI-TOF-MS analysis. With this approach, peptide masses of 3950 and 3964 Da were identified, which corresponded to the Val-34 and Leu-34 variants of the FXIIIA activation peptide, respectively. In addition, masses of 3349/3363 Da and 2759/2773 Da were detected, which corresponded to the truncated forms of the activation peptide that had also been observed by DPD.
The polymorphism of FXIIIA identified by DPD was also observable in the spectra produced by the Ciphergen mass spectrometer. Protein-depleted serum samples from four selected individuals were applied to SELDI chips. Peptides derived from FXIIIA that were identified by DPD were not observed by SELDI-MS on chips with cation-exchange (WCX chip), anion-exchange (SAX chip), or metal-chelating (IMAC chip) surfaces. On the standard gold surface chip, peaks with m/z ratios of 3954 and 3967 were readily detected, which corresponded to the DPD-identified activation peptides (amino acids 1–37). The truncated forms of the activation peptide were not detected by the SELDI chromatographic or standard gold chips.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
). No other peptides were detected or sequenced at the respective positions (chromatographic elution plus mass-to-charge ratio). We believe that it is very unlikely that isobaric peptides derived from other proteins led to identical patterns because this pattern occurred in four distinct molecular forms and at a strongly pronounced intensity corresponding to the high concentration of FXIIIA-derived fragments.
|
In addition, a direct MALDI-TOF-MS approach accurately detected this polymorphism, enabling fast and simple routine diagnostics. After solid-phase extraction of peptides from serum samples, the activation peptide of FXIIIA and degradation products of this peptide were easily detected.
MALDI-MS instrumentation has matured over recent years, and fast and reliable analysis is possible. This procedure presented here is well suited for the determination of the identified polymorphism in larger cohorts and in routine diagnostics. The identification is facilitated by the high serum concentration of FXIIIA (30 mg/L of serum) (13)(14).
This example shows a possible use for mass spectrometric analyses in routine laboratories. The sensitive DPD method allows the detection of biologically relevant peptides and their identification. SELDI-TOF-MS or MALDI-TOF-MS measurement can then be optimized for detection of this specific compound. This is achieved by variation of chip surfaces (e.g., cation- or anion-exchange surfaces), by use of antibodies immobilized on surfaces, or as described here, by use of simple gold chips or by use of solid-phase extraction before MALDI-TOF-MS. These approaches are easily established in clinical laboratories and are user friendly, allowing the analysis of large numbers of samples in a short period of time.
The Val34Leu polymorphism is directly detected by the described mass spectrometric techniques. No time-consuming amplification step is necessary, and the risk of contamination leading to false-positive results is relatively low compared with molecular analysis tools such as allele-specific PCR and PCR with restriction fragment length polymorphism analysis, for which a separate laboratory for sample preparation is also needed.
The described method can also be used in retrospective studies in which the phenotype can be determined in samples from patients with known follow-up data for whom no DNA was available. This approach can also be used when ethical aspects prohibit the use of DNA.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  M. Kinter Toward Broader Inclusion of Liquid Chromatography-Mass Spectrometry in the Clinical Laboratory Clin. Chem., September 1, 2004; 50(9): 1500 - 1502. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
