HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Supplemental Data All Versions of this Article: M600279-MCP200v1 6/2/181    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Glossary Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Osés-Prieto, J. A. Articles by Burlingame, A. L. Search for Related Content PubMed PubMed Citation Articles by Osés-Prieto, J. A. Articles by Burlingame, A. L. Social Bookmarking                      What's this?

---

Research

Formation of -Formyllysine on Silver-stained Proteins

Implications for Assignment of Isobaric Dimethylation Sites by Tandem Mass Spectrometry^*,S

Juan Antonio Osés-Prieto, Xin Zhang and Alma L. Burlingame

From the Department of Pharmaceutical Chemistry, University of California, San Francisco, California 94143-0446

	ABSTRACT

TOP ABSTRACT EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Considerable effort is focused presently on the detection and comprehensive assignment of post-translational modifications of proteins. Obviously attention must be paid to the possibility of chemical modifications that may occur to protein samples during sample handling and manipulation prior to analysis by tandem mass spectrometry. This is of particular concern when a modification is isobaric with the mass differential in common with a known post-translational analog. Here we provide evidence that silver staining protocols that use formaldehyde can result in -formylation of lysine residues. This modification is in fact isobaric with the important product of methyltransferases, ,-dimethyllysine. Without exercising proper caution the analysis of silver-stained protein samples by mass spectrometry looking for dimethylation of lysine will yield a significant number of misassigned sites of modification. High accuracy measurements of the mass of the precursor ions and their fragments are required to eliminate this uncertainty. The occurrence of dimethylation of the -amino function of lysine residues has been reported often in histones. For histone samples excised from silver-stained gels, we found that most sites initially assigned to be dimethylated by automatic search engines under standard search parameters (100 ppm error tolerance) are actually in fact formylated. Caution must be exercised when data obtained from instruments unable to perform high accuracy mass measurements (better than 5 ppm) are to be interpreted.

Among all the amino acids, the lysine residues (the -amino groups) are particularly subjected to modifications. The reported modifications on lysines include acetylation, methylation, ubiquitination, sumoylation, and biotinylation (1, 2). Lysine residues represent 5.94% of all amino acids in the total entries in the Swiss-Prot database (data from ca.expasy.org/sprot/, Release 50.3 of July 11, 2006 of UniProtKB/Swiss-Prot), ranking seventh in the classification of the amino acids by their frequency of appearance. Lysines are even more abundant in some families of proteins, such as DNA-binding protein histones. For instance, 15.4% of all residues in Saccharomyces cerevisiae histone H2B.1 are lysines.

Basic residues are essential elements for the function of these proteins, which regulate DNA structure and function. Lys and Arg can establish either sequence-specific interactions with the DNA by hydrogen bonds with groups in the border of the pairs of bases, thus allowing for the recognition of specific sequences in the DNA, or sequence-unspecific interactions by salt bridges with the negatively charged sugar-phosphate backbone of the DNA.

In eukaryotes and Archaea, the DNA associates with histones in the chromatin (5, 6), which is tightly packed in a bead stringlike structure whose individual components are called nucleosomes. These are made of a core of histones around which the DNA double helix is wrapped. Histones are small basic proteins with a very high percentage of lysine and arginine residues. Their high isoelectric point allows them to interact electrostatically with the DNA. Chromatin structure is essential to regulate DNA function because access of other proteins to DNA will depend on its ability to maintain a partially open conformation (6). Post-translational modifications of histones are very important elements that modulate and/or regulate chromatin structure and function, and they result in epigenetic regulation of gene expression (6–8). Phosphorylation, ubiquitination, methylation, and acetylation of different residues have been described (9): some have been associated with active euchromatin (10), whereas others have been associated with transcriptionally inert heterochromatin (11, 12).

Reagents used during silver staining have been reported to introduce unwanted modifications on functional groups of proteins (13–17). However, due to its high sensitivity, silver staining is still one of the most popular staining methods for SDS-PAGE gels prior to mass spectrometry analysis (18). The aim of this study was to highlight how some of the modifications that silver staining introduces (i.e. formylation of lysine residues) can actually interfere with the discovery of isobaric PTMs such as dimethylation (19). Dimethylation, trimethylation (20, 21), and their recently discovered enzymatic di- and tridemethylases (22–25) are the subject of intense biological interest presently. Here in our study of silver-stained gel samples using a high mass accuracy mass spectrometer (an FT-ICR instrument), we found that numerous lysines with a +28-Da mass shift are actually formylated lysines that would otherwise be assigned to dimethylation without high mass accuracy measurement. These modifications are not detected in Coomassie-stained gel or in-solution digested preparations of the same protein samples.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Protein Sources—
Yeast (S. cerevisiae) alcohol dehydrogenase, human (Homo sapiens) hemoglobin, and bovine (Bos bovis) glutamate dehydrogenase were purchased from Sigma. Rabbit (Oryctolagus cuniculus) creatine kinase was from Roche Applied Science. The yeast (S. cerevisiae) histone proteins were acid-extracted and reverse-phase purified by Dr. C. David Allis’s group at Rockefeller University. The peaks corresponding to different histones were analyzed by SDS-PAGE and by Western blot with specific antibodies to confirm the presence of the histone by a band of the right molecular weight. Histones H2B and H3 were analyzed.

SDS-PAGE—
Proteins (5 µg) were separated on 4–20% polyacrylamide gels (ReadyGels, Bio-Rad) under denaturing conditions. Gels were stained with silver (18) or colloidal Coomassie staining (26).

In-gel Digestion—
Protein bands were excised from gels and in-gel digested with trypsin as described previously (27). The extracted digests were vacuum-evaporated and resuspended in 10 µl of 0.1% formic acid in water.

In-solution Digestion—
Proteins (5 µg in 25 mM NH₄HCO₃) were denatured and reduced by incubating for 5 min at 95 °C in the presence of 2.5 mM DTT and then alkylated by incubation with 10 mM iodoacetamide for 2 h in the dark at room temperature. Remaining iodoacetamide was quenched by adding DTT to a final concentration of 10 mM and incubating at 37 °C for 15–30 min. Acetonitrile was then added to the samples to a final concentration of a 10% (v/v) and digested overnight at 37 °C using 80 ng of sequencing grade modified trypsin (Promega, Madison, WI). Samples were then vacuum-evaporated and resuspended in 10 µl of 0.1% formic acid in water.

Reverse-phase LC-MS/MS Analysis—
The digests were separated by nanoflow liquid chromatography using a 100-µm x 150-mm reverse-phase Ultra 120-µm C18Q column (Peeke Scientific, Redwood City, CA) at a flow rate of 350 nl/min in an Ultimate high performance liquid chromatography system equipped with a FAMOS autosampler (both Dionex-LC Packings, San Francisco, CA). Mobile phase A was 0.1% formic acid in water, and mobile phase B was 0.1% formic acid in acetonitrile. Following equilibration of the column in 2% solvent B, approximately one-tenth of each digest (1 µl) was injected, and then the organic content of the mobile phase was increased linearly to 40% over 30 min and then to 50% in 3 min. The liquid chromatography eluate was coupled to a hybrid linear ion trap-Fourier transform mass spectrometer (LTQ-FT, Thermo Scientific, San Jose, CA) equipped with a nanoelectrospray ion source. Spraying was from an uncoated 15-µm-inner diameter spraying needle (New Objective, Woburn, MA). Peptides were analyzed in positive ion mode and in information-dependent acquisition mode to automatically switch between MS and MS/MS acquisition. MS spectra were acquired in centroid mode using the ICR analyzer in the m/z range between 300 and 2000. For each MS spectrum, the most intense multiple charged peak over a threshold was selected to perform a survey MS scan in profile mode in a 10-Da window using the ICR cell. This was followed by two CID experiments. Product ions were analyzed sequentially on both the linear ion trap and the ICR trap in centroid mode. The CID collision energy was automatically set to 25%. A dynamic exclusion window of 0.5 Da was applied that prevented the same m/z from being selected for 90 s after its acquisition. Typically the ICR cell performs with 25,000 resolution and less than 5 ppm mass measurement accuracy in both MS and CID spectra.

Peak lists were generated using Mascot Distiller version 2.1.0.0 (Matrix Science, Boston, MA). Parameters for MS processing were set as follow: peak half-width, 0.02; data points per Da, 100. Parameters for MS/MS data were set as follows: peak half-width, 0.02; data points per Da, 100. The peak list was searched against the Swiss-Prot database as of April 18, 2006 (containing 216,380 entries) using in-house ProteinProspector version 4.21.4 (a public version is available on line). A minimal ProteinProspector protein score of 22, a peptide score of 15, and a minimal discriminant score threshold of 0.0 were used for initial identification criteria. Carbamidomethylation and acrylamide modification of cysteine; acetylation of the N terminus of the protein; oxidation of methionine; mono-, di-, and trimethylation of lysine; and +12 or +30 adducts on cysteine, arginine, tryptophan, histidine, lysine, asparagine, glutamine, and tyrosine were allowed as variable modifications. Peptide tolerance in searches was 100 ppm for precursor and 0.8 Da for product ions, respectively, to simulate performance of inferior but most commonly available mass spectrometers. Peptides containing one miscleavage were allowed. The number of PTMs was limited to one per peptide. The CID spectra with possible post-translational modifications were further inspected manually.

	RESULTS

TOP ABSTRACT EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Silver Staining of Different Proteins Introduces Formylation on Lysine Residues—
To investigate chemical modification artifacts of silver staining, several unrelated proteins from different sources (yeast alcohol dehydrogenase (AdH), human hemoglobin (Hb), bovine glutamate dehydrogenase (GdH), and rabbit creatine kinase (CK)) were subjected to electrophoresis on polyacrylamide gels and then stained following standard protocols either with silver or with colloidal Coomassie (see "Experimental Procedures"). The protein bands were excised from the gels and in-gel digested with trypsin, and the resulting peptide mixtures were analyzed by LC coupled on line to electrospray MS/MS. Mass spectrometry was performed in an LTQ-FT instrument. This instrument, combining a linear ion trap (LIT) with an FT-ICR cell, is able to perform high accuracy mass determinations often in the low ppm error range. Data were acquired using a method that first analyzes CID-generated fragments in the ion trap for better sensitivity and then analyzes the fragments from the same precursor ion in the FT-ICR cell for maximum accuracy of the mass determinations. Aliquots of the same proteins were also in-solution digested and analyzed by MS in the same way. We searched the acquired data against a subset of the Swiss-Prot protein database (see "Experimental Procedures") to identify potentially modified peptides.

Errors of at least 50 ppm on the mass determinations of peptides are common in most available MS systems. Tolerances on the precursor ion of 100 ppm are frequently used for initial searches in quadrupole time-of-flight instruments. In ion trap experiments, this parameter is typically set several daltons wide. These two families of instruments are present in most laboratories. Although our data were acquired in an instrument able to obtain a better performance (error on mass determination better than 5 ppm, see Fig. 1 for the distribution of mass deviations of the peptides identified from the four analyzed proteins in in-solution digest), the data were first searched with highest tolerance (100 ppm for the precursor ion on the survey scan) to mimic the performance of those instruments more commonly available.

View larger version (25K): [in this window] [in a new window]   FIG. 1. Distribution of mass deviations of the peptides identified in in-solution tryptic-digested AdH, CK, Hb, and GdH (data for the four proteins combined). Mass deviations of the peptides were binned in 0.25 ppm windows. Inset, scatter plot of calculated error versus peptide mass.

LC-MS/MS analysis of the tryptic digest of the silver-stained hemoglobin band showed 15 CID spectra matching the predicted sequence of the chain and 22 CID spectra matching the predicted sequence of the ß chain. ProteinProspector reported that two spectra were potentially of dimethylated peptides when a tolerance of 100 ppm was allowed for the mass of the precursor ion (see Table I). The exact m/z values of the precursor ion for these peptides determined in the FT-ICR cell are 600.3355⁺² and 609.3314⁺³. The MS/MS spectrum acquired in the LIT for the ion at 600.3355⁺² is shown in Fig. 2A. The mass values of the C-terminal sequence ions up to y₄ (y₁ to y₄) and N-terminal sequence ions up to b₆ (b₂ to b₆) matched the anticipated values of the unmodified peptide Val¹–Lys¹¹ on the Hb chain. The masses of C-terminal sequence ions starting from y₅ (y₅ to y₁₀) and N-terminal sequence ions starting from b₇ (b₇ to b₁₀) displayed a mass change of +28 amu. These data indicated the modification on Lys⁷. On the CID spectrum obtained from the same precursor ion on the ICR cell we observed most of these sequence ions (Fig. 2B). According to the deviation in the measurement of the m/z values of these peaks (experimental values versus the predicted values for the fragment ions of the Lys⁷ dimethylated or formylated sequence), we can conclude that the peptide is actually formylated because the mass differences for C-terminal sequence ions starting from y₅ (y₅ to y₁₀) and N-terminal sequence ions starting from b₇ (b₇ to b₁₀) were over 35 ppm for the dimethylated and were in contrast less than 5 ppm for the formylated sequence. For example, the observed mass for y₅, 617.3611, is within 1 ppm of the calculated mass for the formylated structure, m/z = 617.3617. The deviation would be –60 and –18 ppm relative to the calculated values of the alternatively dimethylated fragment or a peptide carrying the substitution Lys to Arg, respectively. Besides the error in the m/z measured of the precursor ion is 0.5 ppm when compared with the predicted mass for the formylated peptide, –29.7 ppm when compared with the predicted mass for a dimethylation peptide sequence, and –8.8 ppm for a peptide sequence with a Lys to Arg substitution. Although unlikely, we had to consider the possibility that these proteins could carry a mutation Lys to Arg, which would also result in a 28-Da increase on the mass of the peptide. Based on measurement errors of the precursor and fragment ions, formylation is the only acceptable choice for the +28-Da mass shift.

View larger version (36K): [in this window] [in a new window]   FIG. 2. Structure and low energy CID spectra of selected formylated peptides found in silver-stained proteins. A and B, MS/MS spectra of a peptide with m/z value 600.3355+2 from human Hb (VLSPADK#TNVK). C and D, MS/MS spectra of a peptide with m/z value 823.9235+2 from yeast AdH (VLGIDGGEGK#EELFR). A and C were acquired on the LIT; B and D were acquired on the ICR cell. The LIT spectra are annotated with the experimental m/z values and assigned identification of the observed sequence ions. On the ICR spectra, peaks corresponding to sequence ions have been labeled with the experimental m/z values and the errors in ppm relatives to the assigned sequences. Fragment ions containing the +28-Da modified residues have been labeled with a pair of values showing the errors comparing the experimental mass with the exact masses of the dimethylated (first number) and formylated sequences (second number). # in the sequences indicates that the preceding residue is modified. Ions corresponding to water or ammonium losses are labeled with an asterisk. Internal fragments are labeled with a cross. RA, relative abundance.

Similarly LC-MS/MS analysis of the tryptic digest of silver-stained aldehyde dehydrogenase showed 25 CID spectra matching the predicted peptide sequence. ProteinProspector reported that two CID spectra potentially belonged to dimethylated peptides, Val¹⁹⁷–Arg²¹¹ carrying a +28-Da modification on Lys²⁰⁶ (Fig. 2, C and D) and Met³³²–Arg³⁴⁰ carrying a +28-Da modification on Lys³³⁴ (Supplemental Fig. 1, C and D). A study of the measurement error of the precursor and fragment ion m/z values clearly indicated a formylation instead of dimethylation in both cases. For example, the mass differences for the C-terminal sequence ions starting from y₆ (y₆ to y₁₄) and N-terminal sequence ions starting from b₁₀ (b₁₀ to b₁₄) for peptide Val¹⁹⁷–Arg²¹¹ are over 28 ppm for a dimethylation but less than 5 ppm for a formylation peptide sequence. The error for the precursor ion (m/z 823.9235, 2+) is –20.6 and 1.5 ppm, respectively, for dimethylation and formylation modification.

In addition, we identified Lys⁹⁰ and Lys¹⁹¹ formylation in glutamate dehydrogenase (peptide Leu⁸⁵–Arg⁹² and peptide Ile¹⁸⁸–Lys²⁰⁰, see Supplemental Fig. 2, A, B, C, and D) and Lys¹⁵ formylation in silver-stained creatine kinase (peptide Leu¹²–Lys²⁵, see Supplemental Fig. 3, A and B, for the CID spectra).

A manual search of the LC-MS traces obtained for Coomassie Blue-stained or in-solution digested samples looking for ions with the specific m/z and retention times of the above described formylated peptides did not reveal the presence of any of these formylated peptides. This indicates that their presence correlates with the process of silver staining.

Formylation on Silver-stained Histone Samples—
We found that lysine residues seem to be readily formylated during the silver staining process. This is a very disadvantageous event or a real handicap for the characterization of potential dimethylation modification in silver-stained protein samples. Methylation (including mono-, di-, and trimethylation) is one of the most common modifications for histone proteins. This type of modification has been shown to play important roles in transcription, in particular transcription elongation. Therefore we set out to study the potential interference on the characterization of PTMs of histones caused by silver staining. Samples of histones H2B and H3 from S. cerevisiae were run in polyacrylamide gels and subjected to silver staining. After tryptic digestion of the protein bands, we found that indeed formylated peptides are present. The peptides found to be formylated are listed in Table II. Two examples for H2B are shown in Fig. 3, and two for H3 are shown in Fig. 4, and the rest of the spectra are shown in Supplemental Figs. 3, 4, and 5. In Fig. 3 we show some spectra demonstrating the presence of formylation on peptides Leu¹⁰³–Arg¹¹⁹ and Lys³⁷–Lys⁴⁶ of H2B.1. Sequence ions containing Lys¹¹¹ and Lys³⁷ show a mass increase of 28 Da (Fig. 3, A and C) indicating that these two sites are modified respectively. Errors on the high accuracy mass determinations of the precursor and sequence ions on the ICR cell reveal that both of the modifications are formylation (Table II and Fig. 3, B and D). Likewise Fig. 4 shows spectra demonstrating the presence of formylation instead of dimethylation on histone H3 Lys⁴² and Lys⁵⁶ (peptides Tyr⁴¹–Arg⁴⁹ and Phe⁵⁴–Arg⁶³).

View larger version (35K): [in this window] [in a new window]   FIG. 3. Structure and low energy CID spectra of selected formylated peptides found in silver-stained H2B of S. cerevisiae. A and B, MS/MS spectra of a peptide with m/z value 910.0076+2 (LILPGELAK#HAVSEGTR). C and D, MS/MS spectra of a peptide with m/z value 655.3193+2 (K#ETYSSYIYK). A and C were acquired on the LIT; B and D were acquired on the ICR cell. Spectra annotations are the same as Fig. 2. RA, relative abundance.

View larger version (25K): [in this window] [in a new window]   FIG. 4. Structure and low energy CID spectra of selected formylated peptides found in silver-stained H3 of S. cerevisiae. A and B, MS/MS spectra of a peptide with m/z value 516.7951+2 (YK#PGTVALR). C and D, MS/MS spectra of a peptide with m/z value 631.8591+2 (FQK#STELLIR). A and C were acquired on the LIT; B and D were acquired on the ICR cell. Spectra annotations are the same as Fig. 2. RA, relative abundance.

View larger version (24K): [in this window] [in a new window]   FIG. 5. Structure and low energy CID spectra of dimethylated peptides found in silver-stained H3 of S. cerevisiae. A and B, MS/MS spectra of a peptide with m/z value 682.3645+2 (EIAQDFK#TDLR). A was acquired on the LIT; B was acquired on the ICR trap. Spectra annotations are the same as Fig. 2. RA, relative abundance.

View larger version (35K): [in this window] [in a new window]   FIG. 6. Structure and low energy CID spectra of selected peptides found in silver-stained proteins containing formaldehyde adducts. A and B, MS/MS spectra of a peptide with m/z value 663.8367+2 from human Hb (VN#VDEVGGEALGR). C and D, MS/MS spectra of a precursor ion with m/z value 647.3221+2 corresponding to a mixture of two formaldehyde-modified species arising from the sequence Gly216–Lys223 from yeast H2B (KETY#SSYIYK and KETYSSY#IYK). A and C were acquired on the LIT; B and D were acquired on the ICR cell. The LIT spectra are annotated with the experimental m/z values and assigned identification of the observed sequence ions. In C, sequence ions with different masses for any of the two modified species are labeled with a number indicating which one they belong to (1, KETY#SSYIYK, and 2, KETYSSY#IYK). On the ICR spectra, peaks corresponding to sequence ions have been labeled with the experimental m/z values and the errors in ppm relatives to the assigned sequences. # in the sequences indicates that the preceding residue is modified. Ions corresponding to water or ammonium losses are labeled with an asterisk. Internal fragments are labeled with a cross. RA, relative abundance.

	DISCUSSION

TOP ABSTRACT EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
It is striking that all the proteins analyzed show some peptides that, as a result of the chemical treatment during silver staining, present +28-Da modifications on Lys residues corresponding to formylation. Formylation is not a common PTM that is routinely included as a variable modification in database searches. However, due to its isobaric nature with another important and much more common PTM, dimethylation, formylation introduced by silver staining must not be ignored. Without an instrument able to perform high accuracy mass measurement it is very difficult to differentiate between these two modifications (the shift in mass of the modified peptides is 28.0313 and 27.9949 Da for dimethylation and formylation, respectively). This is particularly dangerous for studies whose goal is characterizing PTMs of certain groups of proteins, such as histones, where dimethylation is an expected and important PTM. In our analysis of silver-stained histone H3, we detected Lys⁷⁹ dimethylation, which agrees with previous works (28). However, most of the +28-Da modified lysine residues we found (Lys³⁷, Lys⁴⁶, Lys⁴⁹, Lys¹¹¹, and Lys¹²³ in H2B and Lys²⁷, Lys⁴², Lys⁵⁶, and Lys⁶⁴ in H3) correspond actually to formylated lysines instead. We feel that for a MS study looking for dimethylation, silver staining may need to be avoided; or it has to be performed on a high mass accuracy instrument to prevent significant misassignment.

We believe the formylation artifact introduced in the silver staining process is caused by the formaldehyde that is used as a reducing agent in most of the commonly used silver staining protocols. Formaldehyde has been blamed previously for the general lower sequence coverage for MS analysis of silver-stained compared with Coomassie-stained gel samples (16). The exact mechanism of peptide losses is still unknown. Some formaldehyde-free silver staining protocols have been published (16) but still present some problems of background and staining homogeneity.

Formaldehyde has been reported to introduce modifications of +12 and +30 Da on various residues (16, 17) in an MS analysis of modification of formaldehyde on synthetic peptides or small proteins. It has been shown to be able to formylate and methylate lysine residues (13, 14). We did not detect any +30-Da modified peptides in this study. However, we observed several +12-Da modified peptides (on His, Asn, and Tyr; see Fig. 6 and Supplemental Fig. 6) as well as a number of formylated lysines.

The +12-Da modification observed may correspond to putative formaldehyde adducts or intramolecular cross-linking. In their study of the reactivity of formaldehyde with synthetic peptides with blocked N termini and short proteins, Metz et al. (15, 17) have shown that reaction with formaldehyde may introduce a +12-Da modification corresponding to a Schiff base in some amino acids (Cys, Arg, Trp, His, and Lys). This may explain our observation of +12-Da modifications on His (Supplemental Fig. 6) as simple formaldehyde adducts. However, this is not possible in the case where we observed +12 Da on Asn and Tyr (Fig. 6) because the above mentioned studies (15, 17) show that these amino acid did not react primarily with free formaldehyde. The authors proposed that free formaldehyde reacts with amino acids with thiols or primary amino groups forming a methylol adduct (which can be detected as a +30-Da modification) and then undergoing dehydration, yielding labile Schiff bases, which can then react with Cys, Arg, Trp, His, Lys, Asn, Gln, and Tyr, forming intra- or intermolecular cross-links through methylene bridges. On peptides containing cross-linking involving a Tyr residue, the cross-linking breaks during MS analysis, but a +12-Da tag stays on Tyr (17). This finding explains how we can observe these +12 Tyr-modified peptides. As for the peptides we found modified in Asn, because direct reaction with formaldehyde does not seem to occur, the +12 tag in Asn also must be an indication of its involvement in a methylene bridge, which upon rupture leaves this residual mass over the Asn residue. In Val¹⁸–Arg³⁰ from Hb ß carrying a +12-Da modification on Asn¹⁹ (Fig. 6, A and B), the methylene bridge could be formed with the terminal Arg. In Lys³⁷–Lys⁴⁶ from H2B (Fig. 6, C and D) carrying a +12-Da modification tag on Tyr⁴⁰ or Tyr⁴³, both the -amino in the N-terminal Lys or the -amino in the C-terminal Lys could be the cross-linked partners of the Tyr residues.

The reported modifications seem to be to some point selective to particular residues in the studied proteins. The reactivity of the functional groups of the side chains of the amino acids of a polypeptide is greatly influenced by their environment. This is also true for lysine. One possible source of these differences is solvent accessibility. However, big differences in reactivity for different Lys residues with similar solvent exposure in a single polypeptide have been reported. This has been shown for the non-catalyzed reaction of lysine -amino groups with different small molecules, like biotin (29).

The pK_a of the particular Lys residues will affect their reactivity toward a small electrophile such as formaldehyde. Due to its partial positive charge, the strongly electrophilic carbon atom of the carbonyl group of formaldehyde is prone to nucleophilic addition. Lysine residues with lower pK_a will have a population of non-protonated forms (susceptible to suffer an electrophilic attack from the formaldehyde) significantly higher than that of the other lysines. The resulting higher contribution of the unprotonated state to the species population would result in greater reactivity toward formaldehyde. It has been reported for calmodulin that pK_a values, rather than structural and steric effects, play the dominant role in determining the reactivity of Lys side chains toward small electrophilic reagents such as formaldehyde (30).

In this study, we show that some amino acids in a protein gel sample can be modified during the silver staining process. These modifications include a +12-Da formaldehyde adduct or intramolecular cross-linking and a more frequent lysine formylation modification. The importance of this work arises from the fact that the existent studies on the "one-carbon" modifications that formaldehyde can inflict are not performed under the conditions used in the context of silver staining. Those studies focus on samples incubated with high concentrations of formaldehyde for long times. Our study actually proves that even the relatively mild conditions of the silver staining method of Shevchenko et al. (18) do not avoid some collateral problems. Using MS this method allows analysis of silver-stained protein samples with a decent coverage, but it is necessary to be aware that the analysis of very specific problems in these samples, such as the search for dimethyllysine residues, can be compromised. Formylation can be easily misidentified as the isobaric, important, and more common modification ,-dimethylation. Wrong identifications can result from the presence of formylated lysine residues as a consequence of the manipulation of the sample. Using mass accuracies in the low ppm range for the mass determinations can eliminate this ambiguity. The accessibility to equipment able to perform high accuracy mass measurements (under 5 ppm) is then critical to assure proper confidence on the findings.

	ACKNOWLEDGMENTS

 
We are very grateful to Dr. C. David Allis’s group for the yeast histone samples.

	FOOTNOTES

 
Received, July 31, 2006, and in revised form, November 22, 2006.

Published, MCP Papers in Press, November 30, 2006, DOI 10.1074/mcp.M600279-MCP200

¹ The abbreviations used are: PTM, post-translational modification; LIT, linear ion trap; Hb, hemoglobin; AdH, aldehyde dehydrogenase; CK, creatine kinase; GdH, glutamate dehydrogenase.

^* This work was supported by the Biomedical Research Technology Program of the National Center for Research Resources, National Institutes of Health (NCRR Grants RR 01614 (to A. L. B.) and RR 19934 (to A. L. B.)). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^S The on-line version of this article (available at http://www.mcponline.org) contains supplemental material.

To whom correspondence should be addressed. Tel.: 415-476-5641; Fax: 415-502-1655; E-mail: alb{at}cgl.ucsf.edu

	REFERENCES

TOP ABSTRACT EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 

1. Walsh, C. T. (2006) Posttranslational Modifications of Proteins: Expanding Nature’s Inventory, Roberts and Company Publishers, Greenwood Village, CO

2. Yang, X. J. (2005) Multisite protein modification and intramolecular signaling. Oncogene 24, 1653 –1662[CrossRef][Medline]

3. Seet, B. T., Dikic, I., Zhou, M. M., and Pawson, T. (2006) Reading protein modifications with interaction domains. Nat. Rev. Mol. Cell. Biol. 7, 473 –483[CrossRef][Medline]

4. Jensen, O. N. (2006) Interpreting the protein language using proteomics. Nat. Rev. Mol. Cell. Biol. 7, 391 –403[CrossRef][Medline]

5. Sandman, K., Bailey, K. A., Pereira, S. L., Soares, D., Li, W. T., and Reeve, J. N. (2001) Archaeal histones and nucleosomes. Methods Enzymol. 334, 116 –129[Medline]

6. Felsenfeld, G., and Groudine, M. (2003) Controlling the double helix. Nature 421, 448 –453[CrossRef][Medline]

7. Parra, M. A., Kerr, D., Fahy, D., Pouchnik, D. J., and Wyrick, J. J. (2006) Deciphering the roles of the histone H2B N-terminal domain in genome-wide transcription. Mol. Cell. Biol. 26, 3842 –3852[Abstract/Free Full Text]

8. Mersfelder, E. L., and Parthun, M. R. (2006) The tale beyond the tail: histone core domain modifications and the regulation of chromatin structure. Nucleic Acids Res. 34, 2653 –2662[Abstract/Free Full Text]

9. Jenuwein, T., and Allis, C. D. (2001) Translating the histone code. Science 293, 1074 –1080[Abstract/Free Full Text]

10. Narlikar, G. J., Fan, H.-Y., and Kingston, R. E. (2002) Cooperation between complexes that regulate chromatin structure and transcription. Cell 108, 475 –487[CrossRef][Medline]

11. Kelly, T. J., Qin, S., Gottschling, D. E., and Parthun, M. R. (2000) Type B histone acetyltransferase Hat1p participates in telomeric silencing. Mol. Cell. Biol. 20, 7051 –7058[Abstract/Free Full Text]

12. Cheung, P., Tanner, K. G., Cheung, W. L., Sassone-Corsi, P., Denu, J. M., and Allis, C. D. (2000) Synergistic coupling of histone H3 phosphorylation and acetylation in response to epidermal growth factor stimulation. Mol. Cell 5, 905 –915[CrossRef][Medline]

13. Tyihak, E., Trezl, L., and Rusznak, I. (1980) Spontaneous N[cepsilon]-methylation of L-lysine by formaldehyde. Pharmazie 35, 18 –20[Medline]

14. Trezl, L., Rusznak, I., Tyihak, E., Szarvas, T., and Szende, B. (1983) Spontaneous N[cepsilon]-methylation and N[cepsilon]-formylation reactions between L-lysine and formaldehyde inhibited by L-ascorbic acid. Biochem. J. 214, 289 –292[Medline]

15. Metz, B., Kersten, G. F., Hoogerhout, P., Brugghe, H. F., Timmermans, H. A., de Jong, A., Meiring, H., ten Hove, J., Hennink, W. E., Crommelin, D. J., and Jiskoot, W. (2004) Identification of formaldehyde-induced modifications in proteins: reactions with model peptides. J. Biol. Chem. 279, 6235 –6243[Abstract/Free Full Text]

16. Richert, S., Luche, S., Chevallet, M., Van Dorsselaer, A., Leize-Wagner, E., and Rabilloud, T. (2004) About the mechanism of interference of silver staining with peptide mass spectrometry. Proteomics 4, 909 –916[CrossRef][Medline]

17. Metz, B., Kersten, G. F., Baart, G. J., de Jong, A., Meiring, H., ten Hove, J., van Steenbergen, M. J., Hennink, W. E., Crommelin, D. J., and Jiskoot, W. (2006) Identification of formaldehyde-induced modifications in proteins: reactions with insulin. Bioconjug. Chem. 17, 815 –822[CrossRef][Medline]

18. Shevchenko, A., Wilm, M., Vorm, O., and Mann, M. (1996) Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels. Anal. Chem. 68, 850 –858[Medline]

19. Wysocka, J., Swigut, T., Milne, T. A., Dou, Y., Zhang, X., Burlingame, A. L., Roeder, R., Brivanlou, H., and Allis, C. D. (2005) WDR5 associates with histone H3 methylated at K4 and is essential for H3 K4 methylation and vertebrate development. Cell 121, 859 –872[CrossRef][Medline]

20. Martin, C., and Yang, Y. (2005) The diverse functions of histone lysine methylation. Nat. Rev. Mol. Cell. Biol. 6, 838 –850[Medline]

21. Peters, A. H., and Schubeler, D. (2005) Methylation of histones: playing memory with DNA. Curr. Opin. Cell Biol. 17, 230 –238[CrossRef][Medline]

22. Shi, Y., Lan, F., Matson, C., Mulligan, P., Whetstine, J. R., Cole, P. A., Casero, R. A., and Shi, Y. (2004) Histone demethylation mediated by the nuclear amine oxidase homology LSD1. Cell 119, 941 –953[CrossRef][Medline]

23. Tsukada, Y., Fang, J., Erdjument-Bromage, H., Warren, M. E., Borchers, C. H., Tempst, P., and Zhang, Y. (2006) Histone demethylation by a family of JmjC domain containing-proteins. Nature 439, 811 –816[CrossRef][Medline]

24. Klose, R. J., Yamane, K., Bae, Y., Zhang, D., Erdjument-Bromage, H., Tempst, P., Wong, J., and Zhang, Y. (2006) The transcriptional repressor JHDM3A demethylates trimethyl histoneH3 lysine 9 and lysine 36. Nature 442, 312 –316[CrossRef][Medline]

25. Cloos, P. A. C., Christensen, J., Agger, K., Maiolica, A., Rappsilber, J., Antal, T., Hansen, K. H., and Helin, K. (2006) The putative oncogene GASC1 demethylates tri- and di-methylated lysine 9 on histone H3. Nature 443, 307 –311

26. Neuhoff, V., Arold, N., Taube, D., and Ehrhardt, W. (1988) Improved staining of proteins in polyacrylamide gels including isoelectric focusing gels with clear background at nanogram sensitivity using Coomassie Brilliant Blue G-250 and R-250. Electrophoresis 9, 255 –262[CrossRef][Medline]

27. Rosenfeld, J., Capdevielle, J., Guillemot, J. C., and Ferrara, P. (1992) In-gel digestion of proteins for internal sequence analysis after one- or two-dimensional gel electrophoresis. Anal. Biochem. 203, 173 –179[CrossRef][Medline]

28. Zhang, K., Tang, H., Huang, L., Blankenship, J. W., Jones, P. R., Xiang, F., Yau, P. M., and Burlingame, A. L. (2002) Identification of acetylation and methylation sites of histone H3 from chicken erythrocytes by high-accuracy matrix-assisted laser desorption ionization-time-of-flight, matrix-assisted laser desorption ionization-postsource decay, and nanoelectrospray ionization tandem mass spectrometry. Anal. Biochem. 306, 259 –269[CrossRef][Medline]

29. Streaker, ED., and Beckett, D. (2006) Nonenzymatic biotinylation of a biotin carboxyl carrier protein: unusual reactivity of the physiological target lysine. Protein Sci. 15, 1928 –1935[Abstract/Free Full Text]

30. Zhang, M., and Vogel, HJ. (1993) Determination of the side chain pKa values of the lysine residues in calmodulin. J. Biol. Chem. 268, 22420 –22428[Abstract/Free Full Text]

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				J. R. Wisniewski, A. Zougman, and M. Mann N{varepsilon}-Formylation of lysine is a widespread post-translational modification of nuclear proteins occurring at residues involved in regulation of chromatin function Nucleic Acids Res., February 2, 2008; 36(2): 570 - 577. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Supplemental Data All Versions of this Article: M600279-MCP200v1 6/2/181    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Glossary Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Osés-Prieto, J. A. Articles by Burlingame, A. L. Search for Related Content PubMed PubMed Citation Articles by Osés-Prieto, J. A. Articles by Burlingame, A. L. Social Bookmarking                      What's this?