High-Efficiency Microflow and Nanoflow Negative Electrospray

Subscriber access provided by TRENT UNIV

Article

High-efficiency micro- and nano-flow negative electrospray ionization of peptides induced by gas-phase proton transfer reactions Marija Nišavi#, Amela Hozi#, Zdenko Hamersak, Martina Radi#, Ana Butorac, Marija Duvnjak, and Mario Cindri# Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b04466 • Publication Date (Web): 04 Apr 2017 Downloaded from http://pubs.acs.org on April 6, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

High-efficiency micro- and nano-flow negative electrospray ionization of peptides induced by gas-phase proton transfer reactions

Marija Nišavić†┴, Amela Hozić‡┴, Zdenko Hameršak‡, Martina Radić‡, Ana Butorac§, Marija DuvnjakϨ and Mario Cindrić‡*

† Vinča Institute of Nuclear Sciences, University of Belgrade, Belgrade, Serbia ‡ Ruđer Bošković Institute, Bijenička cesta 54, Zagreb, Croatia § BIOCentre, Central Lab Services, Zagreb, Croatia Ϩ Faculty of Agriculture, University of Zagreb, Zagreb, Croatia

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract Liquid chromatography coupled to electrospray ionization mass spectrometry is routinely used in proteomics research. Mass spectrometry-based peptide analysis is de facto performed in positive ion mode, except for the analysis of some post-translationally modified peptides (e.g. phosphorylation and glycosylation). Collected mass spectrometry data after peptide negative ionization analysis is scarce because of a lack of negatively charged amino acid side chain residues that would enable efficient ionization (i.e. in average, every tenth amino acid residue is negatively charged). Also, several phenomena linked to negative ionization such as corona discharge, arcing and electrospray destabilization due to the presence of polar mobile-phase solutions or acidic mobile phase additives (e.g. formic or trifluoroacetic acid) reduce its use. Named phenomena influence micro- and nano-flow electrospray ionization of peptides in a way that prevents formation of negatively charged peptide ions. In this work we have investigated the effects of post-column addition of isopropanol solutions of formaldehyde, 2,2-dimethylpropanal, ethyl methanoate and 2-phenyl-2-oxoethanal as the negative ion mode mobile phase modifiers for the analysis of peptides. According to the obtained data, all four modifiers exhibited significant enhancement of peptide negative ionization, while ethyl methanoate showed the best results. The proposed mechanism of action of the modifiers includes proton transfer reactions trough oxonium ion formation. In this way, mobile phase protons are prevented from interfering with the process of negative ionization. To the best of our knowledge, this is the first study that describes the use and reaction mechanism of aforementioned modifiers for enhancement of peptide negative ionization. Key words: nano-electrospray • nano-liquid chromatography • negative ionization•• proteins/peptides • proton transfer • ethyl methanoate (EM)


Page 2 of 31

Page 3 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Introduction A possibility for direct coupling of liquid chromatography (LC) to electrospray ionization mass spectrometry (ESI-MS) has brought ESI-MS to the forefront of proteomics research, enabling a direct analysis of complex peptide mixtures. Further, the introduction of micro- and nano-LC/ESI-MS systems significantly increased the percentage of identified low abundance peptides.1 This increased sensitivity is attributed to higher efficiency ion sampling due to the formation of smaller electrospray droplets. So far, ESI-MS peptide analysis has been performed almost exclusively in positive ion mode, with the exception of post-translationally modified peptides that contain negatively charged groups.2,3 Conversely, peptide analysis in negative ion mode has been neglected due to several reasons, including poor ionizability at low flow rates (lower than 5 µL/min), low sensitivity of tryptic peptides (under acidic conditions at mobile phase pH~3, lysine and arginine residues are positively charged) and strong dependence upon nature of the analyte.4 In this regard, negative ionization has proven to be useful for the analysis of various ions that carry negatively charged groups, such as carboxylic, phosphoric and sulfonic group.4 5 Several literature references describe ionization and analysis of peptides in negative ion mode but they are mostly restricted to specific peptide amino acid composition and are,5–76–8 therefore, useful only for targeted analysis rather than high-throughput proteomics research. Optimal electrospray ionization in positive ion mode requires solvent of different physico-chemical characteristics than electrospray ionization in negative ion mode.8 4 Therefore, solvent system adjusted for positive ion formation can suppress the formation of negative ions (and vice versa) but can also cause electrical discharge,9 after switching the instrument in negative ion mode. Moreover, the fact that the mobile phase adjusted for positive ion formation is also adjusted for high-efficiency chromatography separation of peptides further aggravate the



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

establishment of efficient negative ionization under the same operating conditions. During the ion source discharge, unwanted chemical ionization of the gas phase analyte and solvent molecules occurs, resulting in poor quality spectra with high non-uniform background noise.84 Furthermore, arcing can cause loss of ESI current and the presence of aqueous and highly conductive solutions additionally contribute to the electrospray instability.84 The effect is amplified at low flow rates (< 5 µL). Low flow and the small sprayer tip diameter (< 50 µm) produce a narrow and convergent spray with small volume droplets (e.g. conventional electrospray source produces droplets in the size range 1-2 µm, while nano-electrospray source produces droplets of 100-200 nm).10 Such narrow and convergent small-sized electrospray plume is extremely sensitive to the slightest changes in source geometry, temperature, electrospray voltage, nitrogen flow or solvent composition.10 Since the required electrospray voltage increases with the square root of the surface tension,11 the solvents with high surface tension are more likely to cause electrical discharge that can disrupt the process of electrospray ionization. Therefore, post-column addition of water-miscible, volatile and low surface tension chemically inert solvent would contribute to electrospray stability. Isopropanol (IPA) satisfies all aforementioned requirements and it has already been demonstrated that it can stabilize electrospray by suppressing electrical discharge.12 Unfortunately, acetonitrile or methanol that are commonly used in mobile phase solutions cannot be changed for IPA in liquid chromatography (IPA is a high-viscosity solvent that generates considerably high system backpressure). For this reason, IPA would be an ideal choice as a post-column modifier for enhancing electrospray stability without affecting chromatographic separation. Apart from the application of nonpolar or chlorinated solvents,13–15 attempts have been made towards addressing challenges of negative ionization by using electron-scavenging gases,12,15 but all these attempts did not solve


Page 4 of 31

Page 5 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


the problem of inadequate peptide negative electrospray ionization at micro- and nano-flow rates. Also, several attempts to increase negative ESI response included the addition of mobile phase modifiers. Wu et al. investigated the effects of several bases, neutral salts and weak acids as negative mobile phase modifiers in the analysis of androgen receptor modulators.16 This work showed that, unlike neutral salts and bases, weak acids improved negative response, in accordance with “wrong-way-round” ESI concept.17 Also, this study showed that formaldehyde (FA) at concentrations 10-100 mM increased ESI response of the tested analytes. The main feature attributed to the positive effect of FA, as well as low molecular weight weak acids, is its small anion volume that has lesser suppressive effect on negative ESI response of hydrophilic compounds that accumulate on the surface of ESI droplets. However, the aforementioned explanations should be reevaluated through the gas-phase chemistry, thus they should have more pronounced chemical background with a proposed reaction mechanism. In this work, we tested several compounds with a purpose of finding a suitable postcolumn mobile phase modifier for enhancement of micro- and nano- flow electrospray peptide ionization in the negative ion mode compatible with the positive ion mode and mobile phases optimized for peptide separation. Among various compounds tested in our laboratory, including aforementioned weak acids, bases and neutral salts, we have chosen FA and three other structurally related volatile compounds with a distinct capability for hydration in the gas phase, namely 2,2-dimethylpropanal (2,2-DMP), ethyl methanoate (EM) and 2-phenyl-2-oxoethanal (2P-2-OE), for further testing. Modifiers have been chosen due to their ability to effectively transfer protons, forming the concept of volatile aldehyde/ketone and ester mobile phase additives (modifiers) for optimal positive and negative micro- and nano-flow rates electrospray



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 31

ionization. Their molecular formula and relevant physico-chemical properties are listed in Table 1. Table 1. List of tested negative ion mode modifiers and their physico-chemical properties. Modifier

Formaldehyde

Structural formula

Mr

Boiling point/°C

Density/gcm-3

30.03

-22.1

1.08 (w = 36%)

(FA)A

86.13

74-76

0.793

74.08

52-54

0.92

2-phenyl-2-

134.13

64 (anhydrous)

1.1 (anhydrous)

oxoethanalP-2-OE

152.15

142 (monohydrate)

Solid state

2,2-dimethylpropanal (2,2-DMP)DMP

Ethyl methanoatee (EM)M

monohydrate (2-P-2-OE)

The effect of each selected compound was investigated on bovine serum albumin (BSA) tryptic digest. Detailed mechanism of negative ionization enhancement effect is proposed for each of the modifiers.


Page 7 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


EXPERIMENTAL SECTION Reagents. Bovine serum albumin and agiotensin were purchased from Sigma-Aldrich (St. Louis, MO, USA) and dissolved in 25 mM ammonium bicarbonate buffer of pH 7.8 (1 mg/mL). The protein was subjected to trypsin (Merck, Germany) digestion for 18 h, while the peptide was incubated (3h, 37 ̊ C) with 5 µL of [Ru(4’ Cl-tpy)(dach)Cl]Cl (Cl-tpy = chloro2,2′:6′,2″-terpyridine, dach = 1,2-diaminocyclohexane) solution, obtained after dissolving a small amount of the compound in miliQ water.18,19 An aliquot (10 µg) containing BSA peptides was dried in centrifugal vacuum evaporator (Eppendorf, Hamburg, Germany) and resuspended in 0.1% formic acid (Kemika, Zagreb, Croatia) to a 0.1 mg/mL concentration. Angiotensin was purified using a ZipTip C18 column (Eppendorf, Hamburg, Germany) and further treated in the same manner as BSA tryptic peptides. Formaldehyde was purchased from Kemika (Zagreb, Croatia) and the other modifiers: 2-phenyl-2-oxoethanal, 2,2-dimethylpropanal and ethyl methanoate were all purchased from Sigma-Aldrich (St. Louis, MO, USA). With the exception of formaldehyde, the modifiers were prepared in following concentrations: 0.1 mM, 0.5 mM, 1 mM, 2.5 mM, 5 mM, 7.5 mM and 10 mM, as isopropanol (J.T. Baker, Deventer, The Netherlands) solutions, to enhance ES stability. Formaldehyde was prepared as 5 mM, 10 mM or 20 mM isopropanol solution. Acetonitrile used for LC mobile phase was purchased from Merck (Germany). LC/MS conditions. The obtained BSA peptides were analysed using SYNAPT G2-Si mass spectrometer (Waters, Milford, MA, USA) coupled with a nanoACQUITY UPLC (Waters, Milford, MA, USA) equipped with nanoACQUITY UPLC 2G-V/M Symmetry C18 Trap Column (100Å, 5µm, 180 µm x 20 mm) and ACQUITY UPLC BEH130 C18 Analytical Column (130 Å, 1.7 µm, 100 µm × 100 mm). Mobile phase A was aqueous 0.1% formic acid and mobile phase B



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

was 0.1% formic acid in 95% acetonitrile. Volume of the injected sample was 4 µL and the column temperature 40 °C. Isocratic delivery of mobile phase A to the trap column was performed at 15 µL/min for two minutes. Peptides were eluted from the analytical column at a flow rate of 1 µL/min using the following gradient: 1%-60% B (0-20 min), 60%-99% B (20-22 min), 99%-1% B (22-24 min), 1% B (24-30 min). The modifier solutions were introduced from Synapt A channel by a “T” connector, at a flow rate of 0.4 µL/min. Mass spectra were recorded in negative resolution mode with capillary and cone voltages set at 3.5 kV and 40 V, respectively. Source temperature was 80 °C and nanoflow gas pressure 1 bar. Mass range was set from 500-4000 Da. All measurements were performed in triplicate. RESULTS AND DISCUSSION Effects of the negative ion mode modifiers on ESI response. The chromatogram obtained after recording BSA peptides in the negative ion mode without post-column addition of the modifiers is shown in Figure 1A. The figure clearly demonstrates the inability of the analyte to deprotonate in negative ion mode under the same conditions (except switched polarities) standardly applied in the positive ion mode at micro-flow rates. Adjusting the parameters for capillary and cone voltages or source geometry does not improve negative ionization. The post-column addition of IPA (Figure 1B) improves negative ion formation, due to IPAs suitable physico-chemical properties that ensure low onset potential and enable work under higher voltages without electrical discharge outbreak. Although IPA highly contributes to electrospray stability and the effectiveness of negative ionization, it is not until the addition of proposed negative ion mode modifiers that the sample ionization is achieved, comparable to the one obtained in the positive ion mode. In this regard, all four tested modifiers showed significant


Page 8 of 31

Page 9 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


enhancement of negative ionization of BSA tryptic digest. Representative chromatogram obtained after post-column addition of 1.0 mM EM IPA solution is shown in Figure 1C. It can be seen that synergistic effect of IPA and EM enables efficient peptide negative ionization suitable for high-throughput proteomics experiments. Also, post-column addition of the modifiers proved to be non-interfering with the process of positive pepide ionization, so complementary set of data can be recorded in positive ion mode without additional changes in instrumental setup, including applied voltages and flow rates (Figure S1). However, even though that observed base peak intensity (BPI) chromatogram in positive ion mode is approximately twice the intensity of BPI chromatogram observed in negative ion mode, peptide ionization should be evaluated peak by peak (the effect of negative peptide ionization is found to be more pronounced for peptide ions that carry negatively charged acidic groups). The method for negative peptide ionization with BSA tryptic digest was optimized to the point where peptide sequence coverage for both ion modes was equalized (higher than 80%). Commonly, negative ion mode requires higher voltages for electrospraying the liquid than positive mode, which increases the risk of electrical breakdown. To decrease this risk, voltages can be lowered at the expence of increased flow rates, so stable ES can still be established. Although this approach can be efficient in some cases,20 it can also cause significant decrease in sensitivity. By implementing post-column addition of the modifiers, we have shown that negative ionization can be achieved under the conditions equivalent to the ones applied in the positive ion mode, without unnecessary compromising of sensitivity. To further pronounce this, negative ion mode BPI chromatograms obtained under nanoliter flow rates (300-500 nL/min) after post-column addition of EM are shown in Figure S2a-c.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 31

Several concentrations of each modifier were tested in order to find the optimal one, e.i. the lowest concentration that generates the highest increase in peptide negative ionization (Figure 2, Tables S1 and S2). Formaldehyde concentrations of 5 mM, 10 mM and 20 mM were tested for this purpose. When concentrations below 5 mM were used, stable ES could not be established. Therefore, this concentration can be considered as the lowest possible FA concentration that can be used as peptide negative ion mode mobile phase modifier. Other modifiers were tested in a series of concentrations ranging from 0.1-7.5 mM. For 2,2-DMP, EM and 2-P-2-OE, the lowest concentration needed for stable ES establishment was that of 0.1 mM. When comparing to FA, all three modifiers gave better results with fifty times lower minimal concentration needed for stable ES. This is important because low modifier concentrations decrease the possibility of source contamination and result in increased analyte signal to noise ratio. As seen from Figure 2, the best effect is achieved under 0.5 mM 2-P-2-OE and 1.0 mM 2,2-DMP and EM concentrations. The comparison between determined optimal concentrations of each modifier shows that EM causes the highest increase in negative ionization, followed by 2-P-2-OE and 2,2DMP. Although applied with the highest concentration when compared to the other modifiers, FA proved to be the least effective. The influence of peptide hydrophobicity, size and charge on the negative ionization enhancement effect of the modifiers. A desirable property of an ideal modifier is the enhancement of peptide negative ionization regardless of its hydrophobicity, size and charge. In order to test this feature for each of the modifiers, several control peptides (Table 2) with different characteristics have been selected for more detailed testing.


Page 11 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table 2. Size, hydrophobicity and charge characteristics of the selected peptides. m/z

number of amino acids

amino acid peptide sequence

retention time/min

BB index (hydrophobicity)

663,3619 (1-) 687,3578 (1-) 884,4002 (1-) 920,4729 (1-) 925,4783 (1-) 972,4427 (1-) 1000,5679 (1-) 1140,6992 (1-), 569, 8457 (2-) 1161,6156 (1-), 580,3039 (2-) 1303,7010 (1-), 651, 3466 (2-) 1477,7803 (1-) 718,3944 (2-) 1509,8277 (1-), 754,4099 (2-) 1637,9226 (1-), 818,4574 (2-) 782,3599 (2-)

5 6 8 8 7 8 10 10

KFWGK AWSVAR DDSPDLPK AEFVEVTK YLYEIAR DLGEEHFK LVVSTQTALA KQTALVELLK

11,5 12,0 11.7 12,5 13,2 11,7 13,6 13,7

44,6 34,6 33,6 40,1 59,7 35,0 43,2 46,2

10

LVNELTEFAK

14,0

45,8

11

HLVDEPQNLIK

13,20

42,4

13 14 12

LGEYGFQNALIVR RHPEYAVSVLLR VPQVSTPTLVEVSR RHPEYAVSVLLR

14,6 13,7 13,7

46,3 45,9 40,8

13,4

39,4

16,0

52,1

15 13

KVPQVSTPTLVEVSR DAFLGSFLYEYSR

Non-polar, alifatic and aromatic side chains Polar, charged side chains Polar, neutral side chains

For testing the effect of the modifiers depending on peptide hydrophobicity,21 four control peptides (m/z 884, 972, 920 and 925) of equal charge, similar size and different hydrophobicity were selected. As seen from Figure 3A (Table S3), more hydrophobic peptides (m/z 920 and 925) are detected as higher intensity ions comparing to more hydrophilic ones (m/z 884 and 972), disregarding the used modifier. Differential ionization during ESI is one of the



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 31

major drawbacks in MS-based proteomics research, thatresearch that results in increased limit of detection and reduced percentage of identified peptides. This particularly holds true for hydrophilic peptides that are less retained on RP columns and are, generally, not as easily ionizable as hydrophobic ones.22–25 Namely, hydrophobic peptides are eluted from reversed phase LC column with a high percentage of acetonitrile that, comparing to higher water content, produces more stable electrospray. Also, it has already been demonstrated that the increase in ESI response is observed for peptides with more extensive nonpolar regions.26,27 This feature was attributed to their increased dropplet surface affinity and, consequently, more successful competition for excess charge. However, in this case some discrete differences between the modifiers effect on ionization of more hydrophobic peptides are also notable, suggesting the most prominent effect of EM and the weakest effect of FA. Also, it should be noted that, disregarding the applied modifier, the peptide at 920 m/z shows higher ionization efficiency than the peptide at 925 m/z with higher BB index. This result suggests that there are other factors than hydrophobicity that influence negative ionization efficiency of peptides with similar characteristics (size and charge state). In the positive ion mode, it has been shown that arginine placed at either C- or N-terminus enhances both, ESI and MALDI responsiveness.

28,29

For ESI,

this factor proved to be more important than hydrophobicity. Analogically, one can presume that amino acid composition, particularly the presence and position of acidic residues could effect responsiveness in the negative ion mode. Nevertheless, further investigation shoud be performed in this regard. When testing the influence of peptide size on the enhancement of negative ionization by each of the employed modifier, four control peptides of equivalent charge, similar hydrophobicity and different size (m/z 663, 1000, 1161 and 1477) were selected. Again, EM


Page 13 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


showed significantly higher effect on all tested peptides, followed by 2,2-DMP (Figure 3B, Table S4). Since most of the detected peptides in all recorded spectra are present as singly or doubly negatively charged species, the influence of charge on modifiers ionization enhancing effect was investigated on four singly and four doubly charged peptide ions coming from eight different peptides selected in a wide retention time window from 11.7 to 16 min (Figure 4A, Table S5). Overall, it can be seen that EM causes the most, while FA causes the least effective peptide negative ionization regardless of the peptide charge. When comparing the effect of 2,2-DMP and 2-P-2-OE, 2,2-DMP seems to exhibit more consistent effect on ionization efficiency while the effect of 2-P-2-OE seems to vary regardless of the peptides charge. To further check whether any modifier has a tendency of favorizing a charge state in which peptide ions are appearing, the values of the ratio between signal intensities of singly and doubly charged ions coming from five peptides selected in a narrow retention time window (13.2-14 min) are compared for each modifier (Figure 4B, Table S6). It can be noticed that the prevalence of doubly charged species significantly grows with the increase of peptide size disregarding the modifier applied. This is due to the increased number of potential deprotonating sites in larger peptides.30 28

Simultaneously, the number of potential protonating sites also increases, but the

modifiers’ frequent gas phase proton transfer minimizes protonation and therefore facilitates peptide negative ionization and multiple charging. From the presented results, we can suggest that peptide negative ionization to an extentlargely follows general rules that govern electrospray ionization in the positive ion mode, disregarding the applied modifier. This also supports the idea of modifiers’ shared mechanism of action. The highest enhancement of negative ionization achieved by EM for each investigated



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 31

peptide property is in accordance with highest BPI chromatogram value obtained for the complete tryptic digest, which makes this modifier an optimal choice to be used as peptide negative ion mode mobile phase modifier. The reason for this might be in functional group differences between EM and the other investigated modifiers. As the only ester in this group, its superior effect might be linked to the gas phase proton affinity or the kinetics of the gas phase proton transfer reactions. Further, slightly different mechanism of action of EM comparing to the other modifiers (Scheme 3) doubles the proton transfer ability which makes proton circulation in the system more prominent and increases the probability of peptide deprotonation. Proton transfer reaction mechanisms of the negative ion mode mobile-phase modifiers under the acidic conditions. Based on the selection of negative ion mode mobile phase modifiers used in this study and their functional group similarity, an underlying mechanisms of their action are proposed. In the study of Wu et al.,17 a positive effect of FA on negative ionization of the tested compounds has been attributed to its small anion volume. As seen from this study, larger molecules, such as 2-P-2-OE showed overall better results in enhancing peptide negative ionization comparing to FA. Formaldehyde proton transfer reaction mechanism. Under the acidic conditions, FA undergoes a reaction of methandiol formation, presented in the following three-step Scheme 1. 1) Carbonyl oxygen protonation:

2) Nucleophilic addition on protonated formaldehyde:


Page 15 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


3) Proton transfer from protonated geminal diol to a water molecule:

Scheme 1. Formaldehyde proton transfer reaction mechanism.

The key to positive action of FA as an enhancer of peptide negative ionization is that protons, that normally interfere with the process of negative ionization, undergo a reversible reaction of methandiol formation, which facilitates negative ionization. 2,2-dimethylpropanal proton transfer reaction mechanism. As FA, 2,2-DMP undergoes a reaction of diol formation under the acidic conditions. The mechanism is shown in three-step Scheme 2. 1) Carbonyl oxygen protonation:

2) Nucleophilic addition on protonated 2,2-dimethylpropanal:

3) Proton transfer from protonated geminal diol to a water molecule:



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Scheme 2. 2,2-dimethylpropanal proton transfer reaction mechanism. Ethyl methanoate proton transfer reaction mechanism. Under the acidic conditions, esters undergo a hydrolysis reaction. Ethanol and formic acid are formed from EM in a four-step hydrolysis reaction. The mechanism is shown in Scheme 3. 1) Carbonyl oxygen protonation:

2) Nucleophilic addition of water molecule to an activated carbonyl group:

3) Proton transfer and ethanol formation:

4) Proton transfer to a water molecule and formic acid formation:


Page 16 of 31

Page 17 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Scheme 3. Ethyl methanoate proton transfer reaction mechanism. The described EM hydrolysis reaction is not entirely analogous with the reaction mechanism of the other modifiers, but what they have in common is oxonium ion formation in proton transfer reactions. The last two steps shown in Scheme 3 that include proton transfer reactions followed by ethanol and formic acid formation are considerd to be the key steps responsible for pronounced effect of EM. This fast and frequent proton transfer is crucial for preventing mobilephase protons and formate ions in interfering with the process of peptide negative ionization. 2-phenyl-2-oxoethanal proton transfer reaction mechanism. This modifier undergoes the same reaction as FA and 2,2-DMP. It contains two carbonyl groups so, theoreticaly, twice as much protons is included in the reaction of alcohol formation. Therefore, it could be assumed that 2-P-2-OE would enhance negative ionization at half the concentration of the above mentionioned modifiers. Nevertheless, this assumption was not experimentaly confirmed because the lowest concentration limit for a negative ion mode modifier depends on the ES stability. The reaction mechanism for 2-P-2-OE is shown in Scheme 4.

1) Carbonyl oxygen protonation:



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 31

2) Nucleophilic addition on protonated 2-phenyl-2-oxoethanal:

3) Proton transfer to water molecules:

Scheme 4. 2-phenyl-2-oxoethanal proton transfer reaction mechanism Based on the presented mechanisms, it is clear that the modifiers behave as proton scavenging agents and that their potency is not limited only to peptide negative ionization enhancement at micro- and nano-flow rates. From the experience in our work, the application of the presented negative ion mode mobile-phase modifiers can be used for the analysis of various molecules in the negative ion mode.


Page 19 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Applications of negative ionization induced by the modifiers on metallated peptides. In order to test the full potential of the modifiers effect on peptide negative ionization, we have recorded negative ion mode MS spectrum of metallopeptide formed between a model peptide and a transition metal complex, by using FA as the modifier. We have used angiotensin II (DRVYIHPF) and ruthenium complex, namely [Ru(4’ Cl-tpy)(dach)Cl]+ (Figure 5), that provides the peptide with an extra positive charge upon binding. The complex binds to histidine nitrogen upon Cl ligand replacement.19,31 Histidine protonation site is therefore blocked, but the peptide is provided with two positive charges coming from rutenium metal center. Apart from that, the peptide contains ariginine that is positively charged, while aspartic acid and tyrosine residues are being protonated at pH~3. The obtained peptide-complex adduct is shown in Figure 5, as a singly charged ion at 1525.59 m/z. Two additional intense peaks can be observed at 1571.61 and 1617.60 m/z. These are attributed to mono- and di- formic acid adducts of ruthenated angiotensin. Since the analyte has excess positive charge, adduct formation with formic acid present in LC mobile phase is observed as a way of further facilitation of negative ionization process. The possibility of negative ionization enhancement through formic acid adduct formation has also been demonstrated on another metallo-based compound, a mannosecontaining ferrocene (Figure S3). These experiments highlight the potency of the described approach for negative ionization enhancement and shedsThis experiment highlight the potency of the described approach for negative ionization enhancement and sheds a light on a broadness of applications for its use in areas of protein research where positive ion mass spectrometry has a dominant position (e.g. metalloproteomics).



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 31

CONCLUSIONS Here, we have presented the application of formaldehyde, 2,2-dimethylpropanal, ethyl methanoate and 2-phenyl-2-oxoethanal as a new class of specific mobile phase modifiers for the analysis of peptides at micro- and nano-flow rates in either positive or negative ion modes. Postcolumn addition of these compounds enabled efficient peptide ionization in the negative ion mode, without influencing their chromatographic separation or the positive ion mode ionization efficiency. All the modifiers were introduced to ESI source as isopropanol solutions to increase micro- and nano-flow ES stability, although isopropanol alone does not provide sufficient ES stabilization or negative ionization enhancement. The lowest formaldehyde concentration to secure stable ES was 5 mM, while the rest of the modifiers exhibited even better effect with fifty times lower concentration (e.g. 0.1 mM). The efficiency of the modifiers at low concentrations defines them as chromatography and analyte-wise non-interfering chemical reagents, driven by proton transfer reaction mechanism. Moreover, effective negative ionization with low modifier concentrations is important for preventing unnecessary source contamination in the long run. The optimal results were obtained for 0.5 mM concentration of 2-phenyl-2-oxoethanal and 1 mM for ethyl methanoate and 2,2-dimethylpropanal, which gave them comparable advantage over formaldehyde. Another feature of an optimal negative ion mode mobile phase modifier is its effectiveness in negative peptide ion formation disregarding peptides physico-chemical properties. In this respect, 1 mM ethyl methanoate solution in isopropanol showed the best results relative to peptide amino acid composition, hydrophobicity, charge and size. Achieving efficient peptide negative ionization can provide a complementary set of data to the one obtained in the positive ion mode, increasing the amount of information obtained from a single sample. This approach can increase protein sequence coverage, providing a step closer


Page 21 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


towards unambiguous protein identification (see Figure S4a-c). The obtained complementary information from negative ion mode can be used for confirmation of tentatively proposed structures of unknown compounds, broadening the possibilities of nano-ESI MS as a tool for structural characterization.

AUTHOR INFORMATION Corresponding Author *E-mail: [email protected]; fax: +38514561010 Author Contributions ┴M. Nišavić and A. Hozić contributed equally to this work. Notes The authors declare no competing financial interest.

ACKNOWLEDGEMENTS This work is supported by the SIIF project ApliMetaFarma RC.2.2.08-0046. The authors wish to acknowledge Ministry of Education, Science and Technological Development of the Republic of Serbia: (grant No. 172011) and HrZZ project PEPTGLYCOSAR IP-2014-09-7899 for participation.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

REFERENCES (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) (26) (27) (28) (29) (30) (31)

Wilm, M. S.; Mann, M. Int. J. Mass Spectrom. Ion Process. 1994, 136 (2), 167–180. Hersberger, K. E.; Håkansson, K. Anal. Chem. 2012, 84 (15), 6370–6377. Deguchi, K.; Ito, H.; Takegawa, Y.; Shinji, N.; Nakagawa, H.; Nishimura, S.-I. Rapid Commun. Mass Spectrom. RCM 2006, 20 (5), 741–746. Cech, N. B.; Enke, C. G. Mass Spectrom. Rev. 2001, 20 (6), 362–387. Desiderio, D. M. Mass Spectrometry: Clinical and Biomedical Applications; Springer Science & Business Media, 2013. Men, L.; Wang, Y. Rapid Commun. Mass Spectrom. RCM 2006, 20 (5), 777–784. Li, Z.; Yalcin, T.; Cassady, C. J. J. Mass Spectrom. JMS 2006, 41 (7), 939–949. Pu, D.; Cassady, C. J. Rapid Commun. Mass Spectrom. RCM 2008, 22 (2), 91–100. Kebarle, P.; Verkerk, U. H. In Electrospray and MALDI Mass Spectrometry; Cole, R. B., Ed.; John Wiley & Sons, Inc., 2010; pp 1–48. Ham, B. M.; MaHam, A. Analytical Chemistry: A Chemist and Laboratory Technician’s Toolkit; John Wiley & Sons, 2015. Smith, D. P. H. IEEE Trans. Ind. Appl. 1986, IA-22 (3), 527–535. Straub, R. F.; Voyksner, R. D. J. Am. Soc. Mass Spectrom. 1993, 4 (7), 578–587. Hiraoka, K.; Kudaka, I. Rapid Commun. Mass Spectrom. 1992, 6 (4), 265–268. Cole, R. B.; Harrata, A. K. J. Am. Soc. Mass Spectrom. 1993, 4 (7), 546–556. Cole, R. B.; Harrata, A. K. Rapid Commun. Mass Spectrom. 1992, 6 (8), 536–539. Wu, Z.; Gao, W.; Phelps, M. A.; Wu, D.; Miller, D. D.; Dalton, J. T. Anal. Chem. 2004, 76 (3), 839–847. Zhou, S.; Cook, K. D. J. Am. Soc. Mass Spectrom. 2000, 11 (11), 961–966. Rilak, A.; Bratsos, I.; Zangrando, E.; Kljun, J.; Turel, I.; Bugarčić, Ž. D.; Alessio, E. Inorg. Chem. 2014, 53 (12), 6113–6126. Nišavić, M.; Masnikosa, R.; Butorac, A.; Perica, K.; Rilak, A.; Korićanac, L.; Hozić, A.; Petković, M.; Cindrić, M. J. Inorg. Biochem. 2016, 159, 89–95. Pu, D.; Clipston, N. L.; Cassady, C. J. J. Mass Spectrom. JMS 2010, 45 (3), 297–305. Osaka, I.; Takayama, M. Rapid Commun. Mass Spectrom. RCM 2014, 28 (20), 2222– 2226. Mirzaei, H.; Regnier, F. Anal. Chem. 2006, 78 (12), 4175–4183. Foettinger, A.; Leitner, A.; Lindner, W. J. Mass Spectrom. JMS 2006, 41 (5), 623–632. Frahm, J. L.; Bori, I. D.; Comins, D. L.; Hawkridge, A. M.; Muddimana, D. C. Anal. Chem. 2007, 79 (11), 3989–3995. Kulevich, S. E.; Frey, B. L.; Kreitinger, G.; Smith, L. M. Anal. Chem. 2010, 82 (24), 10135–10142. Cech, N. B.; Enke, C. G. Anal. Chem. 2000, 72 (13), 2717–2723. Cech, N. B.; Krone, J. R.; Enke, C. G. Anal. Chem. 2001, 73 (2), 208–213. Krause, E.; Wenschuh, H.; Jungblut, P. R. Anal. Chem. 1999, 71 (19), 4160–4165. Abaye, D. A.; Pullen, F. S.; Nielsen, B. V. Rapid Commun. Mass Spectrom. RCM 2011, 25 (23), 3597–3608. Banerjee, S.; Mazumdar, S. Int. J. Anal. Chem. 2012, 2012, e282574. Lazić, D.; Arsenijević, A.; Puchta, R.; Bugarčić, Ž. D.; Rilak, A. Dalton Trans. Camb. Engl. 2003 2016, 45 (11), 4633–4646.


Page 22 of 31

Page 23 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


FIGURES

Figure 1. Comparison of negative ion base peak intensity (BPI) chromatograms of BSA tryptic digest: without post-column addition of mobile phase modifiers (A), after post-column addition of isopropanol (B) and after post-column addition of 1.0 mM ethyl methanoate diluted in isopropanol (C).

Figure 2. Graphical representation of negative ion BPI chromatogram intensities obtained for BSA

tryptic

digest

samples

depending

on

the


modifiers’

concentration.


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3. Histograms presenting the influence of optimal concentrations of the modifiers on the intensity of: four peptide ions of equal charge, similar size and different hydrophobicity (A) and four peptide ions of equal charge, similar hydrophobicity and different size (B).

Figure 4. Histograms presenting: BPI ion intensities of four selected singly and doubly charged peptide ions for each of the modifiers (A) and the values of the ratio between signal intensities of singly and doubly charged ions of five selected peptides for each of the modifier (B).


Page 24 of 31

Page 25 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5. Negative ion mode MS spectrum of angiotensin-[Ru(4’ Cl-tpy)(dach)Cl]+ complex, obtained after post-column addition of 5 mM formaldehyde diluted in isopropanol.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

“ For TOC only ”:


Page 26 of 31

Page 27 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Comparison of negative ion base peak intensity (BPI) chromatograms of BSA tryptic digest: without postcolumn addition of mobile phase modifiers (A), after post-column addition of isopropanol (B) and after postcolumn addition of 1.0 mM ethyl methanoate diluted in isopropanol (C). Figure 1.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Graphical representation of negative ion BPI chromatogram intensities obtained for BSA tryptic digest samples depending on the modifiers’ concentration. Figure 2. 47x26mm (300 x 300 DPI)


Page 28 of 31

Page 29 of 31

1 2 3 4 5 6 7




1 2 3 4 5 6 7 8


Page 30 of 31

Page 31 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Negative ion mode MS spectrum of angiotensin-[Ru(4’ Cl-tpy)(dach)Cl]+ complex, obtained after postcolumn addition of 5 mM formaldehyde diluted in isopropanol. Figure 5. 313x150mm (300 x 300 DPI)


High-Efficiency Microflow and Nanoflow Negative Electrospray

Recommend Documents