|Home | About | Journals | Submit | Contact Us | Français|
To determine the levels of post-translational modifications, we needed a quantitative technique that would allow comparison of the amounts of acetylated versus mono-, di-, and tri-methylated lysines in histones. One method, IVICAT, generates trimethyl-amines and could be used, but is technically challenging. We have modified this technique to be used with standard laboratory equipment so that this chemistry is accessible to most proteomics laboratories.
As part of our effort to identify and characterize the proteins of Toxoplasma gondii and Cryptosporidium parva,1 a technique to determine the relative amounts of acetylated versus mono-, di-, and trimethylated lysines in histones was needed. As illustrated in Figure 1, by permethylating all available lysine side-chains to the trimethyl quaternary ammonium ion using perdeutero-iodomethane (CD3I), the tryptic peptide isomers would coelute from the liquid chromatography (LC) into the mass spectrometer. There, the isomers would be determined, and MS/MS would determine the location and mass of each trimethylamine (Me3)-lysine within the peptide. Based on the number of perdeuteromethyl (CD3) versus methyl (CH3) residues on a lysine side-chain, the substitution level of the lysine in the protein would be determined. To do this, a reliable way of adding CD3 or CH3 to proteins and peptides was needed. Recently, several laboratories determined relative quantities of peptides using iodomethane and perdeuterated iodomethane,2 and other laboratories showed derivitization of peptides to give a trimethylated amino terminus and/or trimethylated lysines.3,4 These trimethylated peptides showed increased signal intensity in the mass spectrometer. Simons et al.2 applied their method to the analysis of digests of proteins and a proteome. The power of this approach—in vacuo isotope-coded alkylation technique (IVICAT)—is offset by its technical difficulty: A hot torch must be used to seal the glass sample tube while it is under vacuum and cooled by liquid nitrogen. If only the α-amino group is to be derivatized and not the ε-amino groups, the pH of the peptide solution has to be adjusted carefully prior to lyophilization.
Presented here is a modification of the prior vacuum technique, which replaces the blow torch and sealed vial with a glass screw-cap vial equipped with an inert cap and valve. A standard lyophilizer served as the vacuum source.
The permethylation of α- and ε-amino groups of lyophilized or vacuum-concentrated peptides was accomplished. These results were achieved with 500 pmol peptides versus previous experiments, which used hundreds of nmol peptide. Also, permethylation was accomplished on a new substrate: peptides bound to reverse-phase chromatography media packed into pipette tips (ZipTip and OMIX tip), an option that offers some compelling advantages.
Our improved technique makes permethylation through IVICAT accessible to most proteomics laboratories and may be used to provide more complete sequence coverage of proteins and proteomes in addition to quantitation, as well as providing access to other chemistries for modification of peptides and proteins.
Glass vials: clear round-bottom 0.3 mL (6 mm, Agilent Technologies, Santa Clara, CA, USA; 5180-0841); polypropylene (pp) microcentrifuge tubes: BioRad (Hercules, CA, USA) 500 μL; ZipTip: C18 P10 (Millipore, Billerica, MA, USA); OMIX tips: C18 MB (Varian, Lake Forest, CA, USA).
IVICAT vial: a reaction vial assembly consisting of a reaction vial cap with valve and a 40-mL screw-cap vial (Eldex, Napa, CA, USA).
An ABI (Foster City, CA, USA) 4800 Plus MALDI-TOF/TOF mass analyzer was run in positive reflectron and linear mode with a laser intensity of 3500 and 50 shots/subspectrum and 1250 total shots (355 nm; and was set at 20 kV source). The instrument was calibrated before each run.
All peptides were used as received from Sigma Chemical Co. (St. Louis, MO, USA; Figure 2). Peptides were dissolved in 10% acetonitrile (MeCN)/H2O to give ~500 pmole/10 μL. Trifluoroacetic acid (TFA) was avoided except where needed to load the peptide onto a ZipTip or OMIX tip. Glass tubes, pp tubes, and tips were labeled with a tungsten carbide pencil.
Peptide (500 pmole) was added to 100 μL of 6% triethylamine (Et3N)/10% MeCN in a glass tube or pp microtube and mixed. The tube was frozen and then lyophilized.
Peptide (500 pmole) was added to 100 μL of 2% Et3N/10% MeCN in a glass tube or pp microtube and mixed. The tube was dried on a SpeedVac (1 h; no heat).
Addition of 2% Et3N/10% MeCN, mixing, and drying were repeated two times.
Peptide (500 pmole) in 10 μL was added to 30 μL of 0.2% TFA in a pp microtube. A wetted ZipTip was loaded with the peptide by 10 × 10 μL draw/return of the peptide solution. Excess liquid was removed by touching the tip to a KimWipe. The loaded ZipTip was washed by 7 × 10 μL draw/discard with 0.5% Et3N. Air was drawn through the ZipTip several times before drying.
The tubes or tips containing the dried peptides were placed in an IVICAT vial. ZipTip or OMIX tip was placed wide-end down and covered by a plain 200-μL or 1-mL pipet tip. The setup was lyophilized overnight at <30 mT. The vacuum was released to argon flow for 20 min.
While maintaining the argon flow, 200 μL MeI, under argon, was added through the valve into the IVICAT vial at room temperature with care taken to avoid wetting the tubes or tips with MeI. The vial was sealed, and the bottom third of the vial was cooled in liquid nitrogen (N2) for 2+ min to ensure that the MeI was as cold as possible. The valve of the IVICAT vial was attached to a lyophilizer, and the vial was evacuated to ~50 milliTorr. The vial was sealed, removed from the liquid N2, and allowed to warm to 20°C.
The vial was placed in an 80°C oven. After 18–24 h, the vial was removed from the oven and allowed to cool in such a way that the condensate formed away from the tubes or tips. The vacuum was released to an argon flow for 20 min. The IVICAT vial was then placed on a lyophilizer to remove remaining traces of MeI.
The tubes received 100 μL of 60% MeCN, and the contents were dissolved and transferred to labeled 500 μL pp tubes. After drying the tubes in a SpeedVac, 100 μL of 6% Et3N/10% MeCN was added. The contents were heated at 90°C for 45 min to hydrolyze the methyl esters.
The tube was cooled, centrifuged, frozen, and lyophilized. Alternately, the tube could be dried on the SpeedVac without heat. The tube received 20 μL of 1% TFA and was sonicated and then centrifuged. The pH was checked to assure that the solution was acidic. The products were cleaned using a ZipTip with 60% MeCN/0.1% TFA used as the elution buffer.
The reacted ZipTip was wetted by adding 5 μL of MeCN to the top of the plug. The products were eluted from the ZipTip by draw/return of 10 × 10 μL in a tube containing 25 μL of 60% MeCN/H2O or 0.1% TFA/60% MeCN. If needed, a second draw/return could be performed in a tube containing 25 μL of 0.25% TFA/75% MeCN.
After drying the tube in a SpeedVac, 100 μL of 6% Et3N/10% MeCN was added to hydrolyze the methyl esters. The contents were heated at 90°C for 45 min. The tube was cooled, centrifuged, frozen, and lyophilized. Alternately, the tube could be dried on the SpeedVac without heat. The tube received 20 μL of 1% TFA and was sonicated and then centrifuged. The pH was checked to assure that the solution was acidic. The products were cleaned using a ZipTip with 60% MeCN/0.1% TFA used as the elution buffer.
The mechanism of permethylation of an amine is shown in Figure 3. Key steps are the formation of the free amine and the loss of hydriodic acid after MeI addition to the amine. It is noted that during the permethylation reaction, other side-chains are derivatized. Histidine added two methyl groups, Asp and Glu formed the corresponding methyl esters, which were hydrolyzed, and Cys and Met (not present in these peptides) would give methylcysteine and homoserine.
Initially, glass tubes and pp tubes were used for the reaction as in previous efforts. Also, as all amines were to be permethylated, stronger concentrations of Et3N were used, and lyophilization and centrifugal vacuum concentration (SpeedVac) were used. Four peptides were used in this study to optimize reaction conditions (Figure 2). For glass and pp microtubes, the presence of TFA was found to inhibit the reaction. For the glass and pp tube reactions, peptides were dissolved in MeCN/H2O.
The best result for permethylation of peptide α- and ε-amino groups was found using SpeedVac concentration with 3 × 100 μL of 2% Et3N/10% MeCN in glass tubes. Lyophilization of peptides on glass from 6% Et3N gave comparable results. Slightly lower conversions to the permethylated peptides were seen using pp microtubes but with a limited number of comparisons. All of the experiments point to the ε-amino group as the most difficult to methylate.
Finally, permethylation of peptides on common purification matrices (ZipTip and OMIX tip) was examined. Previous studies had shown that a variety of chemistries can be done on peptides bound to these tips.5 Peptides loaded onto these tips with 0.1% TFA could be washed with 0.5% Et3N and were retained. These peptides on tips could then be reacted with MeI to give permethylated peptides as described. Optimization of this method is under way.
Although the expected permethylated peptide products were seen for two of the peptides, each of the four peptides studied demonstrated additional chemistry and mass spectrometric behavior.
The Me3-amino peptide (799 Da) was readily formed (Figure 4). Under reflector and linear mode MALDI-TOF MS, it readily underwent a neutral loss of 59 Da,6 which was likely trimethylamine (NMe3). Bradykinin also underwent backbone cleavage between Phe and Ser during hydrolysis of the methyl esters to give a mass of 615 Da. This product was not present in unhydrolyzed samples (data not shown). The neutral losses gave metastable masses at 739 and 554 Da.
Reactions with this peptide gave the permethylated peptide (Me6: 2550 Da, Figure 5) as well as partially methylated products. The mass for a neutral loss of NMe3 was also present (2491 Da). Other neutral loss masses from the Me5, Me4, and Me3 masses (2536–2477; 2522–2463; and 2508–2449) were present.
The peak shapes indicated that these were metastable ions. ACTH and bradykinin had Arg at the amino terminus. Hirota et al.6 had observed in their study of Me3-lysines that the loss of NMe3 occurred more readily for those peptides that had the Me3-lysine close to the N-terminus.
When cMyc peptide was heated under vacuum without MeI, the peptide readily lost H2O, most likely through formation of pyroglutamic acid (1185 Da). When MeI was present, the predominant product had a mass of 1228 Da along with a minor 1227 Da (Figure 6). The 1228 Da corresponded to addition of three Me groups to the lysine and three Me groups to the N-terminus, followed by loss of NMe3 from the lysine. The 1227-Da mass could arise two ways: formation of the Me6-peptide, followed by loss of HNMe3 through lactone ring formation, or loss of H2O, followed by addition of three Me groups to the lysine. The lactone ring would be expected to open during hydrolysis to afford a 1245-Da mass, as seen. Possible partial structures are shown in Figure 6.
Simons et al.2 reported that angiotensin lost NH3 (−17 Da) when subject to heating under vacuum. No loss of NH3 was seen for angiotensin under our conditions in the absence of MeI. When angiotensin was reacted with MeI, the predominant mass was 1057 Da (Figure 7). This could arise by two mechanisms: initial loss of NH3, followed by addition of two Me to His, or addition of three Me groups to the amino terminus and two Me groups to His, followed by NMe3 loss in the tube or in the MS.
Figures 8 and and99 show the permethylation products of angiotensin loaded on a ZipTip and an OMIX tip, which were reacted subsequently with MeI. The same permethylated product was observed as for reaction on glass and on pp microtubes. Although bradykinin gave the Me3-peptide, ACTH 18–39 and cMyc peptides furnished lower amounts of permethylated products and more products with lower methylation. This is likely a result of the lower concentration of Et3N, which had to be used to retain the peptide on the media. Alternative bases are being examined to generate more complete reactions.
We have described a technique for performing permethylation of peptides on a variety of substrates using equipment readily available in most proteomics laboratories. Especially of interest, we have demonstrated that permethylation by a modified IVICAT can be accomplished on peptides absorbed to ZipTip or OMIX tip. Although more work remains to be done to establish the optimum conditions for the chromatographic media, we are intrigued by the opportunity to use this chemistry to enhance detection of peptides from digests of proteins and proteomes as illustrated in Figure 10.
This work was supported by National Institutes of Health/National Institute of Allergy and Infectious Diseases Contract HSN266200400054C. Structures were drawn using Symyx Draw 3.1 (San Ramon, CA, USA).