Alkylated Dihydroxybenzoic Acid as a MALDI Matrix Additive for

Apr 16, 2012 - Hydrophobic peptides are generally difficult to detect using matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) b...
0 downloads 0 Views 3MB Size
Download PDF
Technical Note pubs.acs.org/ac

Alkylated Dihydroxybenzoic Acid as a MALDI Matrix Additive for Hydrophobic Peptide Analysis Yuko Fukuyama,*,† Ritsuko Tanimura,† Kazuki Maeda,‡ Makoto Watanabe,† Shin-Ichirou Kawabata,† Shinichi Iwamoto,† Shunsuke Izumi,‡ and Koichi Tanaka† †

Koichi Tanaka Laboratory of Advanced Science and Technology, Shimadzu Corporation, 1, Nishinokyo-Kuwabaracho, Nakagyo-ku, Kyoto 604-8511, Japan ‡ Department of Mathematical and Life Sciences, Graduate School of Science, Hiroshima University,1-3-1 Kagamiyama, Higashi-Hiroshima 739-8526, Japan ABSTRACT: Hydrophobic peptides are generally difficult to detect using matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) because the majority of MALDI matrixes are hydrophilic and therefore have a low affinity for hydrophobic peptides. Here, we report on a novel matrix additive, o-alkylated dihydroxybenzoic acid (ADHB), which is a 2,5-dihydroxybenzoic acid (DHB) derivative incorporating a hydrophobic alkyl chain on a hydroxyl group to improve its affinity for hydrophobic peptides, thereby improving MALDI-MS sensitivity. The addition of ADHB to the conventional matrix α-cyano-4-hydroxycinnamic acid (CHCA) improved the sensitivity of hydrophobic peptides 10- to 100-fold. The sequence coverage of phosphorylase b digest was increased using ADHB. MS imaging indicated that hydrophobic peptides were enriched in the rim of a matrix/analyte dried spot when ADHB was used. In conclusion, the addition of ADHB to the standard matrix led to improved sensitivity of hydrophobic peptides by MALDI-MS.

M

atrix-assisted laser desorption/ionization (MALDI)1,2 and electrospray ionization (ESI)3 mass spectrometry (MS) are indispensable analytical tools for peptide and protein analysis. The recent progress of MS in sensitivity and accuracy has driven the development of proteomics.4,5 However, MS is often problematic in the analysis of insoluble hydrophobic peptides and proteins, such as membrane proteins.6−12 MALDI-MS is the most suitable for analyzing hydrophilic peptides,11 because conventional MALDI matrixes have hydrophilic properties and thus have a low affinity for hydrophobic peptides. The limited solubility of hydrophobic peptides in aqueous solvent is also problematic. To address these issues, two approaches using either a solvent6,12−14 or a detergent9,15−18 have generally been reported. The use of aqueous formic acid as a solvent for the solubilization of hydrophobic peptides reportedly led to the degradation of acidlabile peptides.13 The use of nonaqueous chloroform/ methanol solvent was suitable for hydrophobic peptides but inapplicable for hydrophilic peptides.6 A silicone prestructured plate with hexafluoroisopropanol as an analyte solvent was reported to increase the sensitivity of hydrophobic membrane protein digest by a focusing effect on matrix/analyte droplets at the picomole (pmol) level.12 An alternative is in situ liquid− liquid extraction preparation, using ethyl acetate as a waterimmiscible organic solvent.14 This technique is an on-target preparation protocol, and the operation is cumbersome because aqueous and organic phases must be deposited separately on a target plate. Additionally, sensitivity of the peptides is at a low pmol level. The addition of n-octyl glucopyranoside (n-OGP) detergent in a peptide mixture has been reported to improve peptide ion © 2012 American Chemical Society

peak intensity and amino acid sequence coverage.9,15−17 It promotes peptide solubilization during in-gel digestion due to its properties as a nonionic detergent with a critical micelle concentration. Functional cleavable detergents consisting of a conventional matrix structure and hydrophobic moiety (e.g., 2-cyano-3-(4-dodecyloxyphenyl)acrylic acid, which is structurally related to α-cyano-4-hydroxycinnamic acid (CHCA), and 3-(4dodecyloxymethyloxy-3,5-dimethoxyphenyl)acrylic acid, which is structurally related to 3,5-dimethoxy-4-hydroxycinnamic acid (SA)) have been reported.18 These detergents solubilize proteins, then generate MALDI matrixes after cleavage under an acidic condition, and finally increase the signal intensity of hydrophobic proteins. In principle, n-OGP and the functional cleavable detergents act as a detergent to improve the solubility of peptides, and sensitivity improvement is an additional effect. No systematic study of sensitivity improvement has been performed; thus, the precise mechanism remains unclear. In fact, previous approaches using either a solvent or a detergent to improve the solubility of hydrophobic peptides still present several issues. The use of a solvent involves many complicated processes before measurement, and sensitivity is at pmol levels. The use of a detergent involves the possibility of forming micelles that may cause deterioration of spectrum quality. Both approaches lack systematic study on sensitivity, and the mechanism for sensitivity improvement is not yet understood. Received: February 23, 2012 Accepted: April 4, 2012 Published: April 16, 2012 4237

dx.doi.org/10.1021/ac300540r | Anal. Chem. 2012, 84, 4237−4243

Analytical Chemistry

Technical Note

On the basis of the findings of previous studies, we launched a study to develop a novel matrix or a matrix additive for sensitive detection of hydrophobic peptides, which have hardly been detected by MALDI-MS. Here, a novel matrix additive, o-alkylated dihydroxybenzoic acid (ADHB), was synthesized for highly sensitive analysis of hydrophobic peptides. ADHB is a 2,5-dihydroxybenzoic acid (DHB) derivative incorporating a hydrophobic alkyl chain on the hydroxyl group and thus is expected to have affinity for hydrophobic peptides. The addition of ADHB to the conventional matrix CHCA improved the sensitivity of hydrophobic peptide 10- to 100-fold, which is at the femtomole to subfemtomole level, without decreasing mass spectral quality. This report discusses details of the sensitivityenhancing effect with MS imaging to clarify the distribution of an analyte, a matrix, and a matrix additive on a matrix/analyte dried spot.

Sample Preparation for MALDI-MS. ADHB solution was prepared in 50/50 (v/v) ACN/0.1% aqueous TFA at 5 mg/mL. ADHB solution and matrix solution were mixed at ratios of 1:1, 1:10, 1:100, and 10:1 (v/v). The analyte solution (0.5 μL) and the matrix solution containing ADHB (0.5 μL) were mixed on a stainless-steel plate (sample plate 2.8 mm ring × 384 well, Shimadzu/Kratos, UK) to be analyzed by MALDI-TOFMS. MALDI-MS. MALDI-TOFMS measurement was performed using an AXIMA Performance (Shimadzu/Kratos, UK) mass spectrometer equipped with a nitrogen UV laser (337 nm) in linear mode, in both positive and negative ion modes. All spectra were obtained at the threshold laser power. MS imaging was carried out by Intensity Mapping software functionality supplied by the MALDI-TOFMS instrument manufacturer or by BioMap software (copyrighted by Novartis, available at http://maldi-msi.org/). All MS images were constructed from raster-scanned data sets composed of two shots of laser irradiation at each of the 6561 data points in square regions of 4000 μm × 4000 μm with 50 μm measurement pitch (i.e., 81 × 81 lattice). The detection limit was defined as the lowest quantity of analyte detected as [M + H]+ or [M − H]− with S/N ≥ 2 when each series of analyte solution at a different concentration made by 10-fold serial dilutions was evaluated. Sensitivity improvement was determined by dividing the detection limit using CHCA by that using CHCA with ADHB.

■

EXPERIMENTAL SECTION Peptides and Proteins. Temporin A amide, NF-kB inhibitor SN50, OVA-BIP hybrid peptide, melittin honey bee, β-amyloid 22-42, MPGΔNLS, β-amyloid 1-11, β-amyloid 1-28, GPHRSTPESRAAV, and β-conglycinin hydrolysate 165-178 FAS inhibitor thioesterase antagonist soy were purchased from AnaSpec, Inc. Humanin, [Gly14]-humanin, and β-amyloid 1-42 were purchased from Peptide Institute, Inc. ACTH 18-39 was obtained from Sigma. MassPREP phosphorylase b digestion standard was obtained from Waters Corporation. The peptides and digests were dissolved in 50/50 acetonitrile (ACN)/0.1% aqueous trifluoroacetic acid (TFA) (v/v) at appropriate concentrations. All water for the aqueous solutions was deionized using a Milli-Q ultrapure water system (Millipore, USA). Matrixes. α-Cyano-4-hydroxycinnamic acid (CHCA), 2,5dihydroxybenzoic acid (DHB), and 3,5-dimethoxy-4-hydroxycinnamic acid (SA) were purchased from LaserBio Laboratories. 1,5-Diaminonaphthalene (DAN) was purchased from Sigma-Aldrich. Each matrix was dissolved in 50/50 ACN/0.1% aqueous TFA (v/v) at 10 mg/mL. 2-Hydroxy-5-octyloxybenzoic Acid (o-Alkylated DHB, ADHB). ADHB incorporating a C8 alkyl chain (Figure 1) was

■

RESULTS AND DISCUSSION

Optimization of ADHB Usage as a Matrix Additive. CHCA solution and ADHB solution were mixed at ratios of 1:1, 10:1, 100:1, and 1:10 (v/v), as described in the Experimental Section, to compare the detection limits of the hydrophobic peptide humanin. Compared with the use of CHCA alone, the detection limit of humanin peptide was improved 10-fold in positive ion mode using the 1:10 (v/v) mixture, 100-fold in positive and 10-fold in negative mode using the 1:1 (v/v) mixture, and 100-fold in both positive and negative modes using the 10:1 (v/v) mixture (data not shown). The 10:1 (v/v) mixture ratio had the highest sensitivity. ADHB did not function alone as a matrix to ionize the peptides (see next paragraph). Comparative evaluation of different alkyl chain lengths (C1, C4, C6, C8, C10, and C16) in the ADHB structure indicated that the octyl (C8) chain was the most effective for sensitivity improvement in both positive and negative ion modes (Table 1). Hence, ADHB incorporating

Figure 1. 2-Hydroxy-5-octyloxybenzoic acid (o-alkylated DHB, ADHB) as a novel matrix additive.

Table 1. Detection Limit of Hydrophobic Peptide Humanina humanin (fmol/well)

synthesized as follows. A mixture of DHB (1.54 g, 10 mmol), 1-bromooctane (2.32 g, 12 mmol), potassium carbonate (3.0 g), and anhydrous acetone (200 mL) was refluxed for 8 h, and the inorganic salts were then separated by filtration. The solvent was dried over MgSO4 and evaporated under reduced pressure. The residue was subjected to flash chromatography on silica gel (hexane−ethylacetate, 8:2, v/v) to yield ADHB (2.02 g, 76%) as a light-yellow solid: 1H NMR (400 MHz, CDCl3): δ = 7.29 (d, J = 3.2 Hz, 1H), 7.00 (dd, J = 8.9, 3.2 Hz, 1H), 6.88 (d, J = 8.9 Hz, 1H), 4.32 (t, J = 6.7 Hz, 2H), 1.76 (tt, J = 7.4, 6.7 Hz, 2H), 1.25 (m, 10H), and 0.89 (t, J = 6.9 Hz, 3H). Highresolution mass spectrometry: Calculated for C15H23O4+ [M + H]+ 267.1591; Found. 267.1591. ADHB possessing different alkyl chain lengths (C1, C4, C6, C10, and C16) was synthesized in a similar way.

C16-ADHB + CHCA C10-ADHB + CHCA C8-ADHB + CHCA C6-ADHB + CHCA C4-ADHB + CHCA C1-ADHB + CHCA CHCA

positive mode

negative mode

1 1 0.1 1 10 10 10

10 10 1 10 100 100 100

Detection limit of humanin (BB Index −5800, HPLC Index 117.4) using CHCA with ADHB possessing different alkyl chain lengths of C1, C4, C6, C8, C10, or C16 as CHCA + ADHB or CHCA alone (see Experimental Section).

a

octyl chain (C8-ADHB) was used in all the experiments described below, and the abbreviation “ADHB” is used 4238

dx.doi.org/10.1021/ac300540r | Anal. Chem. 2012, 84, 4237−4243

Analytical Chemistry

Technical Note

Figure 2. Positive-ion mass spectra of 10 fmol equimolar mixture of β-amyloid 1-11 and humanin using (a) CHCA + ADHB, (b) CHCA alone, or (c) ADHB alone using the raster scanned data sets (see Experimental Section).

hereafter in this report to denote C8-ADHB. The highest sensitivity of humanin was obtained when ADHB was added to CHCA, compared to the ADHB mixture with DHB, SA, or DAN (data not shown). CHCA + ADHB exhibited 100-fold greater sensitivity than CHCA alone. CHCA + ADHB made by mixing CHCA solution and ADHB solution at a ratio of 10:1 (v/v) was the optimal condition for highly sensitive hydrophobic peptides analysis. The peptide was detected at 0.1 fmol/well using CHCA + ADHB (Table 1). This optimized mixture ratio of ADHB was employed in all the experiments in this study. Analysis of Hydrophobic and Hydrophilic Peptide Mixture Using CHCA + ADHB. An equimolar mixture of hydrophobic and hydrophilic peptides (humanin and β-amyloid 1-11) was analyzed using CHCA + ADHB, CHCA alone, or ADHB alone. Figure 2 presents the mass spectra of a mixture of 10 fmol of each peptide in positive-ion mode. A clear difference between CHCA + ADHB and CHCA alone was observed. Humanin was detected with higher intensity than β-amyloid 1-11 using CHCA + ADHB, whereas β-amyloid 1-11 was detected with higher intensity than using CHCA alone. These results indicate that ADHB contributes to an intensity increase of hydrophobic peptide ions. The optimal laser power using CHCA + ADHB was almost the same as or slightly lower than that using CHCA alone. However, neither hydrophobic nor hydrophilic peptides were detected using ADHB alone, indicating that ADHB did not function alone as a matrix to ionize peptides, as discussed above. Figure 3 presents MS images of humanin, β-amyloid 1-11, CHCA, and ADHB. With the use of CHCA + ADHB, humanin was detected mainly in the rim of the matrix/analyte dried spot, and β-amyloid 1-11 was detected throughout the entire dried spot, with the exception of the rim. In contrast, with the use of CHCA alone, these peptides were detected throughout the entire dried spot. Additionally, ADHB itself was detected mainly in the rim of the dried spot, and CHCA itself was detected throughout the entire spot. These results imply that the use of ADHB enriches hydrophobic peptides. The enrichment observed for humanin using ADHB was expected to extend for a range of hydrophobic peptides. Sensitivity Improvement by ADHB and Peptide Hydrophobicity. The sensitivity of 14 peptides with a range of hydrophobicity (BB Index −9210 to +9030) was evaluated using CHCA + ADHB or CHCA alone in both positive- and negative-ion modes. The BB Index is a typical hydrophobicity scale proposed by Bull and Breese, based on the free surfacevapor transfer energy of each amino acid, determined from the surface tension of their isoelectric solutions.19 Table 2 indicates their sensitivities in positive-ion mode. Improvement of sensitivity using ADHB was confirmed for peptides analytes

Figure 3. Matrix/analyte crystal photos (top rows) on a sample plate and MS images of (a, e) humanin, (b, f) β-amyloid 1-11, (c, g) CHCA, and (d) ADHB for a 10 fmol equimolar mixture of humanin and β-amyloid 1-11 using CHCA + ADHB or CHCA alone, using the raster scanned data sets (see Experimental Section). The same data set as for Figure 2 was used for the MS imaging.

No. 1−7 (BB Index −9210 to −1760) and analyte No. 10 (BB Index +310). Thus, sensitivity improvement with ADHB was related to peptide hydrophobicity as indicated with the BB Index, with the exception of analyte No. 10. A similar tendency was observed in negative-ion mode (data not shown). From an alternative point of view, improvement in sensitivity using ADHB was observed for all peptides with a molecular weight (MW) exceeding 2000, except for analyte Nos. 1, 9, and 12. Hence, MW or the surface area of a peptide molecule may also be related to sensitivity improvement. Sensitivity improvement of hydrophobic peptides using ADHB may be related to the localization of hydrophobic 4239

dx.doi.org/10.1021/ac300540r | Anal. Chem. 2012, 84, 4237−4243

Analytical Chemistry

Technical Note

Table 2. Sensitivity Improvement for Peptides with the BB Index from −9210 to +9030a analytes

detection limit (fmol/well)

no.

BB index

MW

name

CHCA + ADHB

CHCA

sensitivity improvement rate (fold)b

1 2 3 4 5 6 7 8 9 10 11 12 13 14

−9210 −7070 −5800 −5410 −5130 −4470 −1760 −1500 −380 +310 +2510 +2690 +4650 +9030

1396.8 2781.5 2687.2 2657.2 2290.5 2846.5 4514.1 1999.3 2465.7 2766.2 1325.3 3262.5 1364.5 1847.8

temporin A, amide NF-kB inhibitor humanin [Gly14]-humanin OVA-BIP hybrid peptide melittin, honey bee β-amyloid 1-42 β-amyloid 22-42 ACTH 18-39 MPGΔNLS β-amyloid 1-11 β-amyloid 1-28 GPHRSTPESRMV β-conglydnin 165-178

0.1 0.1 0.1 0.1 0.1 0.1 0.1 1 0.1 0.1 1 0.1 1 0.1

1 10 10 1 10 1 10 1 0.1 1 1 0.1 1 0.1

10 100 100 10 100 10 100 1 1 10 1 1 1 1

Detection limit and sensitivity improvement for 14 peptides with a BB Index from −9210 to +9030 using CHCA + ADHB, compared with CHCA alone in the positive-ion mode (see Experimental section). Boldface indicates that the detection limit improved. bSensitivity improvement was calculated by dividing the detection limit using CHCA alone by that using CHCA + ADHB. a

Table 3. Correlation of Sensitivity Improvement, HPLC Index, and MS Imaging of Peptidesa

a

Sensitivity improvement of 14 peptides using CHCA + ADHB in positive-ion mode (see Experimental Section). The number on the left-hand side of the table is the same as in Table 2 based on the BB Index. Boldface indicates that the detection limit improved. bSensitivity improvement was calculated by dividing the detection limit using CHCA alone by that using CHCA + ADHB for differentially analyzed peptides (see Table 2). cMS images were created for each peptide in the 100 fmol, 14 peptide mixture analyses.

peptides in the rim region of the matrix/analyte dried spot. ADHB was also predominantly detected in the rim; hence, a strong relationship was hypothesized between hydrophobic peptides and ADHB. The peptides that were detected with higher sensitivity may strongly interact

with ADHB. The hydrophobic alkyl chain of ADHB can interact with hydrophobic peptides. Therefore, the HPLC Index was next used to investigate the correlation between sensitivity improvement and hydrophobicity of peptides. 4240

dx.doi.org/10.1021/ac300540r | Anal. Chem. 2012, 84, 4237−4243

Analytical Chemistry

Technical Note

Figure 4. Positive-ion mass spectra of 100 fmol phosphorylase b digests using (a) CHCA + ADHB or (b) CHCA alone in m/z 700 to 2500 (top) and m/z 2500 to 6000 (bottom). The digest ion peaks are annotated with asterisks. ND indicates that ion peaks were not detected. Insets are MS images of ion peaks at m/z 3715.2 and m/z 4899.3 using CHCA + ADHB.

The HPLC Index is a hydrophobicity scale proposed by Browne et al. based on predictive retention coefficients of peptides by RP-HPLC.20−22 This index indicates the interaction between the peptide and the C18 alkyl chain attached to the stationary phase of a reverse-phase column. This interaction may apply to the relationship between the alkyl chain of ADHB and hydrophobic peptides. The peptides listed in Table 2 were sorted in the order of the HPLC Index (Table 3). The results clearly indicated sensitivity improvement using ADHB for hydrophobic peptides with an HPLC Index of 100 or higher. Additionally, these peptides were all detected in the rim of the matrix/analyte dried spot. Consequently, it was concluded that sensitivity improvement using ADHB was more closely related to the HPLC Index than the BB Index described above. These results imply that the sensitivity improvement for hydrophobic peptides by the enrichment effect may be due to interactions between peptides and the alkyl chain of ADHB. Analysis of Phosphorylase b Digests. The CHCA + ADHB mixture was applied to the analysis of phosphorylase b trypsin digests. Figure 4 depicts the mass spectra of the digests using CHCA + ADHB or CHCA alone. In the m/z 700 to 2500 range, identical peptide ions were detected with both matrixes. However, in the m/z > 2500 range, ions were observed more strongly using ADHB, especially for ions at m/z 3715.2, which has a relatively high HPLC Index of 128.2, and m/z 4899.3, which has a relatively high HPLC Index of 116.5. Additionally, these two peptides were detected in the rim of the matrix/ analyte dried spot (MS images in Figure 4). As a result, sequence coverage of phosphorylase b increased from 53.9%

(CHCA alone) to 63.2% (CHCA + ADHB) (10% increase). These results indicate that the sensitivity improvement of ADHB was applicable for protein digest analysis. Peak intensity distribution of hydrophobic peptide ions. Figure 5 depicts mass spectra and heat maps color-coded according to absolute peak intensity, indicating a peak intensity distribution of [M + H]+ or [M + Na]+ for the hydrophobic peptide OVA-BIP (HPLC Index 100.8; see Table 3) using CHCA + ADHB or CHCA alone. The heat map was calculated using the BioMAP function presented in the Experimental Section. Comparison of these MS images clearly demonstrates that [M + H]+ of the peptide was strongly detected in the rim of a matrix/analyte dried spot using CHCA + ADHB. Using CHCA alone, the peptide was detected as [M + H]+ and [M + Na]+ in the whole area of a matrix/analyte dried spot with a relatively lower intensity than CHCA + ADHB. The lower signal of sodiated ions observed using CHCA + ADHB should be due to the hydrophilic property of sodium salts (i.e., they tend to be in aqueous phase and stay away from the hydrophobic ADHB). Thus, hydrophobic peptides were predominantly detected as [M + H]+ with small amounts of sodium salts using CHCA + ADHB, which achieved simplification of mass spectra and sensitive detection of hydrophobic peptides. This method using ADHB may be similar to that using a silicon wafer modified plate (Zeng et al.), in which the sensitivity of peptides is improved by selectively capturing them and salts in different regions of the inside hydrophilic layer and the outer hydrophilic layer on a plate.23 Both are on-plate selective enrichment and self-desalting 4241

dx.doi.org/10.1021/ac300540r | Anal. Chem. 2012, 84, 4237−4243

Analytical Chemistry

Technical Note

Figure 5. Mass spectra and heat maps color-coded according to absolute peak intensity, indicating the peak intensity distribution of [M + H]+ or [M + Na]+ for the hydrophobic peptide OVA-BIP (HPLC Index 100.8; see Table 3) using CHCA + ADHB or CHCA alone. The heat map was calculated by the BioMAP function discussed in the Experimental Section. Individual mass spectra were recorded once across the whole surface and then accumulated within regions of interest using the BioMap software. Mass spectra for the entire surface are depicted at the top; the second row represents each mass spectrum accumulated along the rim of the matrix/analyte dried spot (red trace) or in the entire dried spot, with the exception of the rim (blue trace).

Thus, a larger amount of the ratio of ADHB to CHCA in the matrix/additive solution causes lower detection sensitivity of hydrophobic peptides because a smaller amount of CHCA cannot ionize a majority of the peptides enriched with ADHB. This result agrees with the fact that the sensitivity improvement rate was higher in the order of the use of the 10:1, 1:1, and 1:10 (v/v) mixtures of CHCA and ADHB solutions (see the optimization paragraph). The effect of the alkyl chain length of ADHB on the detection limit of humanin (Table 1) is to increase sensitivity by 100-fold from C4 to C8; however, no change occurs with CHCA alone to C4 or from C10 to C16. This result is explained as follows. ADHB possessing C1 to C4 is not highly hydrophobic; thus, they only slightly enrich hydrophobic peptides in the rim of the matrix/analyte dried spot. ADHB with C10 to C16 is too hydrophobic to interact with peptides and thus cannot enrich hydrophobic peptides in the rim. Hence, the sensitivity of hydrophobic peptides is not particularly improved using ADHB with C4 or C10 to C16. In contrast, ADHB with the octyl (C8) chain may enrich itself in the rim of the spot, interact with peptides, and enrich hydrophobic peptides in the rim, effectively improving sensitivity of the hydrophobic peptides.

techniques of peptides; however, the method using ADHB separates only hydrophobic peptides from hydrophilic peptides (and salts). In addition, this enrichment and desalting effect is accompanied by the self-assembly of the matrix (CHCA) and the matrix additive (ADHB). Accordingly, CHCA + ADHB achieves simple concentration of hydrophobic peptides without any special modification on a plate. Mechanism of Sensitivity Improvement of Hydrophobic Peptides by ADHB. In this study, the following observations were made. (1) Sensitivity of hydrophobic peptide analysis was improved using the matrix additive ADHB, which is a DHB derivative incorporating a hydrophobic alkyl chain. (2) With MS imaging, hydrophobic peptides were detected in the rim of a matrix/analyte dried spot using CHCA + ADHB, and hydrophilic peptides were observed in the inner entire area of the spot. Using CHCA alone, both hydrophobic and hydrophilic peptides were detected throughout the dried spot. (3) Sensitivity improvement using ADHB closely correlated with the HPLC Index of individual peptides. These observations suggest that ADHB enriches hydrophobic peptides maintaining an affinity with CHCA, whereas ADHB itself does not ionize peptides. In this case, the ionization properties of CHCA + ADHB as a matrix depend on CHCA. 4242

dx.doi.org/10.1021/ac300540r | Anal. Chem. 2012, 84, 4237−4243

Analytical Chemistry

■

With MS imaging, ADHB itself was detected in the rim of the spot, whereas CHCA was detected throughout the spot (Figure 3). ADHB can be amphipathic because its structure consists of a hydrophobic alkyl chain and a hydrophilic hydroxybenzoic moiety; however, ADHB is practically insoluble in water alone. When mixed with water, ADHB does not dissolve; it gathers at the air−water interface, arranging itself in a single orientation where the alkyl chain protrudes into the air and the hydroxybenzoic acid moiety faces the inside of the interface. An analyte droplet on a target plate contains CHCA, ADHB, and analyte peptides in 50/50 aqueous ACN. In a small droplet on a plate, convection flow occurs outward from the interior, where solutes can be conveyed with the flow to the rim of the droplet. With a 50/50 aqueous ACN droplet, ACN evaporates more rapidly than water, due to high volatility, and leaves a droplet of higher water content behind. As ACN vaporizes, ACN should be provided from the solution to the surface, and ADHB is carried with ACN to the surface rim region (“coffee-ring effect”).24 Therefore, the ADHB-rich phase is localized at the bottom surface of a droplet. ADHB and hydrophobic peptides interact positively; hence, ADHB and the peptides are coconcentrated in the rim of the spot. This is still a hypothesis that must be clarified.

REFERENCES

(1) Tanaka, K.; Waki, H.; Ido, Y.; Akita, S.; Yoshida, Y.; Yoshida, T. Rapid Commun. Mass Spectrom. 1988, 2, 151−153. (2) Karas, M.; Hillenkamp, F. Anal. Chem. 1988, 60, 2299−2301. (3) Fenn, J. B.; Mann, M.; Meng, C. K.; Wong, S. F.; Whitehouse, C. M. Science 1989, 246, 64−71. (4) Aebersold, R.; Mann, M. Nature 2003, 422, 198−207. (5) Matta, A.; Ralhan, R.; DeSouza, L. V.; Siu, K. W. M. Mass Spectrom. Rev. 2010, 29, 945−961. (6) Green-Church, K. B.; Limbach, P. A. Anal. Chem. 1998, 70, 5322−5325. (7) Breaux, G. A.; Green-Church, K. B.; France, A.; Limbach, P. A. Anal. Chem. 2000, 72, 1169−1174. (8) Bird, G. H.; Lajmi, A. R.; Shin, J. A. Anal. Chem. 2002, 74, 219− 225. (9) van Montfort, B. A.; Canas, B.; Duurkens, R.; GodovacZimmermann, J.; Robillard, G. T. J. Mass Spectrom. 2002, 37, 322− 330. (10) Redeby, T.; Emmer, Ǻ . Anal. Bioanal. Chem. 2005, 381, 225− 232. (11) Yang, Y.; Zhang, S.; Howe, K.; Wilson, D. B.; Moser, F.; Irwin, D.; Thannhauser, T. W. J. Biomol. Tech. 2007, 18, 226−237. (12) Redeby, T.; Roeraade, J.; Emmer, Ǻ . Rapid Commun. Mass Spectrom. 2004, 18, 1161−1166. (13) Schey, K. L. Methods Mol. Biol. 1996, 61, 227−230. (14) Kjellström, S.; Jensen, O. N. Anal. Chem. 2003, 75, 2362−2369. (15) Cohen, S. L.; Chait, B. T. Anal. Chem. 1996, 68, 31−37. (16) Gobom., J.; Schuerenberg, M.; Mueller, M.; Theiss, D.; Lehrach, H.; Nordhoff, E. Anal. Chem. 2001, 73, 434−438. (17) Katayama, H.; Nagasu, T.; Oda, Y. Rapid Commun. Mass Spectrom. 2001, 15, 1416−1421. (18) Norris, J. L.; Porter, N. A.; Caprioli, R. M. Anal. Chem. 2005, 77, 5036−5040. (19) Bull, H. B.; Breese, K. Arch. Biochem. Biophys. 1974, 161, 665− 670. (20) Browne, C. A.; Bennett, H. P. J.; Solomon, S. Anal. Biochem. 1982, 124, 201−208. (21) Horváth, C.; Melander, W.; Molnár, I. J. Chromatogr. 1976, 125, 129−156. (22) Purcell, A. W.; Aguilar, M. I.; Hearn, M. T. W. J. Chromatogr. 1992, 593, 103−117. (23) Zeng, Z.; Wang, Y.; Shi, S.; Wang, L.; Guo, X.; Lu, N. Anal. Chem. 2012, 84, 2118−2123. (24) Deegan, R. D.; Bakajin, O.; Dupont, T. F.; Huber, G.; Nagel, S. R.; Witten, T. A. Nature 1997, 389, 827−829.

■

CONCLUSION In this study, we demonstrated that a novel matrix additive ADHB improved the sensitivity of hydrophobic peptides through their enrichment in the rim of a matrix/analyte dried spot. ADHB significantly increased the sequence coverage of protein. These sensitivity improvements were systematically studied using CHCA with/without ADHB. Hydrophobic peptides were detected at the subfemtomole level. The protocol is simple: CHCA is mixed with ADHB and then used as the matrix. Peptides in a range of hydrophobicity can be detected with this matrix system. The sensitivity improvement using ADHB will be useful for analyzing proteins with hydrophobic regions, such as membrane proteins. This technique may uncover novel posttranslational modifications through detecting hydrophobic peptides that have not been detected by MALDI-MS. The combination of this ADHB technique with gel or HPLC-based proteomics has the potential for future comprehensive analyses and biomarker discovery. A study to develop a more effective matrix (or matrix additive) based on the findings obtained is underway.

■

Technical Note

AUTHOR INFORMATION

Corresponding Author

*Phone: +81-75-823-2897. Fax: +81-75-823-2900. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research is granted by the Japan Society for the Promotion of Science (JSPS) through the “Funding Program for World-Leading Innovative R&D on Science and Technology (FIRST Program),” initiated by the Council for Science and Technology Policy (CSTP). The authors thank Dr. Hiroki Kuyama for preparing the manuscript and Mr. Kenichi Taniguchi and Dr. Takashi Nishikaze for helpful comments. 4243

dx.doi.org/10.1021/ac300540r | Anal. Chem. 2012, 84, 4237−4243