Coupling Ambient Ionization Mass Spectrometry with Liquid Chromatography and
Electrochemistry and Their Applications
A dissertation presented to
the faculty of
the College of Arts and Sciences of Ohio University
In partial fulfillment
of the requirements for the degree
Doctor of Philosophy
Yi Cai
December 2016
© 2016 Yi Cai. All Rights Reserved.
2
This dissertation titled
Coupling Ambient Ionization Mass Spectrometry with Liquid Chromatography and
Electrochemistry and Their Applications
by
YI CAI
has been approved for
the Department of Chemistry and Biochemistry
and the College of Arts and Sciences by
Hao Chen
Associate Professor of Chemistry and Biochemistry
Robert Frank
Dean, College of Arts and Sciences
3
ABSTRACT
CAI, YI, Ph.D., December 2016, Chemistry
Coupling Ambient Ionization Mass Spectrometry with Liquid Chromatography and
Electrochemistry and Their Applications
Director of Dissertation: Hao Chen
Ambient ionization methods allow the ionization of untreated samples in the open
environment. In this dissertation, two different ambient ionization techniques, desorption
electrospray ionization (DESI) and probe electrospray ionization (PESI), has been
developed and coupled with liquid chromatograph (LC) and electrochemistry (EC) and
their analytical applications have been explored and discussed.
Liquid sample DESI generally employs a DESI probe to spray solvent with high
voltage to ionize sample as the sample solution is delivered to the ion source by a piece of
fused silica transfer capillary. A new splitting interface, a PEEK capillary tube with a
micro-orifice drilled in the capillary wall, was used to connect with LC column for applying
DESI ionization. A small portion of LC eluent emerging from the orifice can be directly
ionized by DESI with negligible time delay while the remaining analytes can be online
collected. Furthermore, online derivatization using reactive DESI is possible for additional
application such as supercharging proteins.
Since splitting via an orifice introduces negligible dead volume and back pressure,
the performance of the LC/DESI-MS with the focus of using ultra-fast LC for analyzing
sample was further evaluated. Using a monolithic C18 column, metabolites in urine can be
separated within 1.6 min, online monitored by DESI and collected as purified samples.
4
Negative ions can be directly generated for acidic analytes in acidic LC eluent by DESI
during the LC/MS analysis process using a spray solvent with alkaline pH. In addition,
DESI-MS is found to be compatible with ultra-performance liquid chromatography
(UPLC) for the first time. The 45 s separation of drugs can be achieved via UPLC/DESIMS under high temperature.
The combination with EC further broadens LC/MS applications. UPLC-MS
combined with EC via DESI was first developed for the structural analysis of
proteins/peptides that contain disulfide bonds. Using this combined UPLC/EC/DESI-MS
method, peptides containing disulfide bonds can be differentiated from those without
disulfide bonds, as the former are electroactive and reducible. MS/MS analysis of
disulfide-reduced peptide ions provides increased information about the peptide sequence
and disulfide-linkage pattern. In addition, upon online electrolytic reduction and reactive
DESI, supercharged proteins showed increased charges distribution which is of value for
MS/MS sequencing application.
PESI, another ambient ionization technique, employs a conductive solid probe to
ionize samples directly on the probe tip with the aid of applied high voltage. Due to the
high salt tolerance of PESI, the detection of electrochemical reaction products in roomtemperature ionic liquids is realized, for the first time. Furthermore, PESI-MS allows the
detection the electrochemical reaction products on different or multiple electrode surfaces.
In addition, peptides and proteins fractionated through isoelectric focusing (IEF) can also
be directly analyzed by PESI-MS.
5
DEDICATION
I dedicate this dissertation to my friends, parents, and husband
6
ACKNOWLEDGMENTS
I acknowledge my advisor, Dr. Hao Chen, for his great mentoring. His dedication
to science deeply impacts me and encourages me to become a good researcher. He is also
generous to support me for attending conferences.
I acknowledge to all my dissertation committee members, Dr. Peter de B.
Harrington, Dr. Shiyong Wu and Dr. Shigeru Okada for their guidance.
I acknowledge to National Science Foundation Career Award (to Dr. Hao Chen,
CHE-1149367), National Science Foundation Instrument Development for Biological
Research (CHE 1455554), National Natural Science Foundation of China (Grant
21328502), Edison Biotechnology Institute Faculty Fellowship, Merck Research
Laboratories New Technology Review & Licensing Committee, Ohio Third Frontier
Technology Validation and Start-Up Fund, and Center for Intelligent Chemical
Instrumentation, Department of Chemistry and Biochemistry, Ohio University for the
financial support.
I acknowledge to all former and current members in Dr. Hao Chen’s group, Dr.
Zhixin Miao, Dr. Yun Zhang, Zongqian Yuan, Pengyuan Liu, Mei Lu, Dr. Qiuling Zheng,
Si Cheng, He Xiao, Meihong Hu, Chang Xu, Yuexiang Zhang, Najah Almowalad,
Amanda Forni, Sabrina Cramer, Fengyao Li, Denial Adams, David Hu, Prof. Dr. Ping Li,
Dr. Ning Pan, Prof. Dr. Kehua Xu, Prof. Dr. Jun Wang, Prof. Dr. Qiuhua Wu and Prof.
Dr. Zhi Li for their assistance and support.
I acknowledge to all my collaborators, Dr. Howard D. Dewald (Ohio University);
Dr. Michael Held (Ohio University); Dr. Huifang Yao (Merck & Co., Inc., Rahway,
7
New Jersey), Dr. Yong Liu (Merck & Co., Inc., Rahway, New Jersey), Dr. Roy Helmy
(Merck & Co., Inc., Rahway, New Jersey) for their helpful discussion and suggestions.
I acknowledge to Dr. Andrew Tangonan, Bascom French, Paul Schmittauer,
Aaron Dillon, Carolyn Khurshid, Marlene Jenkins, Jackie Bennett-Hanning, Dr.
Zhengfang Wang, Dr. Mengliang Zhang, Xue Zhao, Xinyi Wang, and Dr. Lei Wang for
their generous help.
8
TABLE OF CONTENTS
Page
Abstract ............................................................................................................................... 3
Dedication ........................................................................................................................... 5
Acknowledgments............................................................................................................... 6
List of Figures ................................................................................................................... 11
List of Schemes ................................................................................................................. 16
List of Abbreviations ........................................................................................................ 17
Chapter 1: Introduction ..................................................................................................... 19
1.1 Mass Spectrometry ................................................................................................. 19
1.2 Mass Spectrometer .................................................................................................. 19
1.3 Ionization Methods ................................................................................................. 20
1.3.1 ESI-MS ............................................................................................................ 21
1.3.2 DESI-MS.......................................................................................................... 22
1.3.3 Liquid Sample DESI-MS ................................................................................. 24
1.3.4 Probe Electrospray Ionization (PESI) .............................................................. 25
1.4 EC-MS .................................................................................................................... 26
1.4.1 EC/ESI-MS ...................................................................................................... 28
1.4.2 EC/DESI-MS ................................................................................................... 28
1.5 LC/DESI-MS .......................................................................................................... 30
Chapter 2: A New Splitting Method for Both Analytical and Preparative LC/MS .......... 33
2.1 Introduction ............................................................................................................. 33
2.2 Experimental ........................................................................................................... 35
2.2.1 Chemicals ......................................................................................................... 35
2.2.2 LC Separation Condition ................................................................................. 36
2.2.3 DESI-MS Detection ......................................................................................... 36
2.3 Results and Discussion ........................................................................................... 37
2.3.1 Protein Detection ............................................................................................. 37
2.3.2 Saccharide Detection........................................................................................ 46
2.3.3 Sulfonamides Detection ................................................................................... 49
2.4 Conclusions ............................................................................................................. 51
9
Chapter 3: Coupling of Ultrafast LC with Mass Spectrometry by DESI ......................... 53
3.1 Introduction ............................................................................................................. 53
3.2 Experimental ........................................................................................................... 55
3.2.1 LC/DESI-MS Instrument ................................................................................. 55
3.2.2 LC Separation Conditions ................................................................................ 56
3.3 Results and Discussion ........................................................................................... 57
3.3.1 LC/DESI-MS Analysis of Drugs ..................................................................... 57
3.3.2 LC/DESI-MS analysis of Acidic Compounds ................................................. 61
3.3.3 UPLC/DESI-MS Analysis ............................................................................... 63
3.4 Conclusions ............................................................................................................. 66
Chapter 4: Integration of Electrochemistry with Ultra Performance Liquid
Chromatography/Mass Spectrometry (UPLC/MS)........................................................... 68
4.1 Introduction ............................................................................................................. 68
4.2 Experimental ........................................................................................................... 70
4.2.1 Chemicals ......................................................................................................... 70
4.2.2 Apparatus ......................................................................................................... 70
4.2.3 LC Separation Condition ................................................................................. 71
4.3 Results and Discussion ........................................................................................... 72
4.3.1 Reduction of Somatostatin 1-14....................................................................... 72
4.3.2 Reduction of Insulin ......................................................................................... 75
4.3.3 Reduction of Proteins ....................................................................................... 86
4.4 Conclusions ............................................................................................................. 90
Chapter 5: Coupling Electrochemistry with Probe Electrospray Ionization Mass
Spectrometry ..................................................................................................................... 91
5.1 Introduction ............................................................................................................. 91
5.2 Experimental ........................................................................................................... 92
5.2.1 Chemicals ......................................................................................................... 92
5.2.2 Apparatus ......................................................................................................... 93
5.3 Results and Discussion ........................................................................................... 94
5.3.1 PESI-MS Detection of Electrochemical Reactions in RTILs .......................... 94
5.3.2 PESI-MS Detection of Electrochemical Reaction on Different Electrodes ... 101
5.3.3 PESI-MS Detection of Separated Proteins/Peptides in IEF ........................... 107
10
5.4 Conclusions ........................................................................................................... 111
Chapter 6: Summary and Future Work ........................................................................... 113
References ....................................................................................................................... 115
Appendix: Publications ................................................................................................... 129
11
LIST OF FIGURES
Page
Figure 1-1. Basic components of a mass spectrometer. ................................................... 20
Figure 1-2. Scheme showing an ESI source.[16] Reproduced with permission from
WILEY-VCH, Copyright 2001. ........................................................................................ 22
Figure 1-3. Scheme showing the apparatus of DESI.[17] Reproduced with permission
from Science, Copyright 2004. ......................................................................................... 23
Figure 1-4. Scheme showing the apparatus of liquid sample DESI.[24] Reproduced with
permission from Springer, Copyright 2009. .................................................................... 25
Figure 1-5. Scheme showing the apparatus of PESI.[36] Reproduced with permission from
WILEY-VCH, Copyright 2007. ........................................................................................ 26
Figure 1-6. Timeline for the development of EC/MS with different ionization method.[49]
Reproduced with permission from RSC, Copyright 2013. ............................................... 27
Figure 1-7. Scheme showing the apparatus for online coupling of a thin-layer
electrochemical flow cell with DESI-MS.[28] Reproduced with permission from ACS,
Copyright 2009. ................................................................................................................ 30
Figure 1-8.Scheme showing the apparatus for LC/DESI-MS.[63] Reproduced with
permission from ACS, Copyright 2011. ........................................................................... 32
Figure 2-1. EICs of (a) insulin (+4 ion) and (b) ubiquitin (+6 ion) from LC/DESI-MS
analysis and the corresponding DESI-MS spectra of (c) insulin and (d) ubiquitin. ......... 38
Figure 2-2. EICs of (a) insulin (+4 ion) and (b) ubiquitin (+6 ion) from LC/ESSI-MS
analysis. ............................................................................................................................. 40
12
Figure 2-3. EICs of (a) insulin (+5 ion) and (b) ubiquitin (+7 ion) from LC/reactive
DESI-MS analysis and the corresponding reactive DESI-MS spectra of (c) insulin and (d)
ubiquitin. ........................................................................................................................... 42
Figure 2-4. EICs of (a) insulin (+4 ion) and (b) ubiquitin (+6 ion) from LC/DESI-MS
analysis (100 μm i.d. orifice) and the collected (c) insulin and (d) collected ubiquitin. .. 45
Figure 2-5. Calibration curves of insulin and ubiquitin (UV absorption peak area vs.
protein concentration). ...................................................................................................... 46
Figure 2-6. EICs of (a) N-acetyl-D-glucosamine and (b) maltohexaose eluted out of the
C18 column and detected by DESI-MS. EICs of (c) N-acetyl-D-glucosamine and (d)
maltohexaose eluted out of the C18 column and detected by ESSI-MS. MS spectra of (e)
the collected N-acetyl-D-glucosamine and (f) the collected maltohexaose. The re-analysis
of the collected proteins was performed using ESSI-MS. The peak at m/z 223 in (f) came
from solvent background. ................................................................................................. 49
Figure 2-7. EIC spectra of a mixture of sulfadiazine, sulfamerazine, and sulfaquinoxaline
(a) from LC/DESI-MS analysis and (b) from LC/UV analysis. ....................................... 51
Figure 3-1. EICs of (a) dopamine, (b) 3-methoxytyramine, (c) L-kynurenine, and (d) Ltryptophan acquired by LC/DESI-MS. ............................................................................. 58
Figure 3-2. EICs of (a) dopamine, (b) 3-methoxytryamine, (c) L-kynurenine, and (d) Ltryptophan acquired by LC/ESI-MS. ................................................................................ 59
Figure 3-3. Calibration curves of 3-MT (UV absorption peak area vs. 3-MT
concentration) ................................................................................................................... 61
13
Figure 3-4. EICs of (a) ketoprofen, (b) fenoprofen, and (c) ibuprofen acquired by
LC/negative ion DESI-MS. EICs of (a') ketoprofen, (b') fenoprofen, and (c') ibuprofen
acquired by LC/negative ion ESI-MS. .............................................................................. 63
Figure 3-5. EICs of a) codeine, b) cocaine and c) flunitrazepam acquired by
UPLC/DESI-MS with MS/MS spectra shown in the figure insets. The red lines shown in
the molecule structures indicate the bond cleavage upon CID. ........................................ 64
Figure 3-6. EICs of a) codeine, b) cocaine and c) flunitrazepam acquired by high
temperature UPLC/DESI-MS (80 oC column temperature). Re-analysis of collected d)
codeine, e) cocaine and f) flunitrazepam after high temperature UPLC separation. ........ 66
Figure 4-1. a) The apparatus of UPLC/EC/DESI-MS. b) Extracted ion chromatograms
(EICs) showing the UPLC separation of b) AGCK TFTSC (upper panel) and NFFWK
(lower panel). DESI-MS spectra of AGCK TFTSC c) when cell was off and d) when cell
was on. .............................................................................................................................. 74
Figure 4-2. CID MS/MS spectra of a) [AGCK TFTSC+H]+ (m/z 933), b) reduced
peptide [AGCK+H]+ (m/z 378) and c) reduced peptide [TFTSC+H]+ (m/z 558). ............ 75
Figure 4-3. EICs of the pepsin-digested insulin. .............................................................. 77
Figure 4-4. DESI-MS spectra of the digested insulin peptides carrying no disulfide bonds
YTPKA: a) cell off and b) cell on; FVNQ: c) cell off and d) cell on; GIVE: e) cell off and
f) cell on; YQLEN: g) cell off and h) cell on; YQLE: i) cell off and j) cell on; and VEAL:
k) cell off and l) cell on. .................................................................................................... 78
14
Figure 4-5. DESI-MS spectra of P3 with a) cell off and b) cell on, c) CID MS2 spectrum
of doubly charged P3 at m/z 1037, d) CID MS/MS spectrum of singly charged P3 chain A
at m/z 1311, and d) CID MS/MS spectrum of singly charged P3 chain B at m/z 766. ..... 81
Figure 4-6. CID MS/MS spectra of a) [P1+2H]2+ (m/z 1082) and b) [P4+3H]3+ (m/z 854).
........................................................................................................................................... 83
Figure 4-7. DESI-MS spectra of P2 with a) cell off and b) cell on. CID MS/MS spectra of
[P2+2H]2+ (m/z 769) and [LVCGERGFF+2H]2+ (m/z 514). ............................................ 85
Figure 4-8. EICs of the protein mixture containing insulin, myoglobin and α-lactalbumin.
........................................................................................................................................... 87
Figure 4-9. DESI-MS spectra of the UPLC-separated insulin a) when cell was off without
supercharging, b) when cell was off with supercharging and c) when cell was on with
supercharging. ................................................................................................................... 88
Figure 4-10. DESI-MS spectra of the UPLC-separated myoglobin a) when cell was off
without supercharging, b) when cell was off with supercharging and c) when cell was on
with supercharging. ........................................................................................................... 88
Figure 4-11. DESI-MS spectra of the UPLC-separated α-lactalbumin a) when cell was
off without supercharging, b) when cell was off with supercharging and c) when cell was
on with supercharging. ...................................................................................................... 89
Figure 5-1. a) Scheme showing the PESI-MS and the electrochemical cell configuration.
PESI-MS spectra of b) CoII salen before electrolysis and c) CoII after electrolysis. The
inset in c) shows the oxidation reaction of CoII salen. ...................................................... 96
15
Figure 5-2. CID MS/MS spectrum of the electrochemical oxidation product Co III salen
ion (m/z 325). .................................................................................................................... 97
Figure 5-3. PESI-MS spectra of a) ferrocene before electrolysis and b) ferrocene after
electrolysis. The inset in c) shows the oxidation reaction of ferrocene. ........................... 98
Figure 5-4. PESI-MS spectra of a) TEMPO before electrolysis and b) TEMPO after
electrolysis. The inset in c) shows the oxidation reaction of ferrocene. CID MS/MS
spectrum of the electrochemical oxidation product TEMPO ion (m/z 156). .................. 100
Figure 5-5. a) Equations showing the electrochemical reduction of flutamide and the
electrochemical oxidation of dopamine; EC/PESI-MS spectra of the electrolyzed mixture
sampled from b) the anode surface and c) the cathode surface. ..................................... 102
Figure 5-6. Traditional EC/MS spectrum of the mixture after electrolysis. .................. 104
Figure 5-7. a) Scheme showing the BPE cell configuration; b) the oxidation process of
CLZ. PESI-MS spectra of the electrolyzed CLZ solution sampled from four different
electrode surfaces: c) position 1 (driving electrode anode), d) position 2 (BPE cathode), e)
position 3 (BPE anode), and f) position 4 (driving electrode cathode). ......................... 106
Figure 5-8. a) Scheme illustrating the process of the IEF/PESI-MS; b) PESI-MS
spectrum of the mixture sample in the IEF buffer before electrofocusing; PESI-MS
spectra of c) angiotensin II (fraction #12), d) angiotensin II (1-4, fraction #9), and e)
angiotensin II antipeptide (fraction #6)........................................................................... 109
Figure 5-9. a) PESI-MS spectrum of the mixed protein sample in the IEF buffer prior to
isoelectric focusing; PESI-MS spectra of b) cytochrome c (fraction #20), c) ubiquitin
(fraction #12) and d) β-lactoglobulin A (fraction #6). .................................................... 111
16
LIST OF SCHEMES
Page
Scheme 2-1. Apparatus of a new LC/DESI-MS post column splitting method. ............... 35
17
LIST OF ABBREVIATIONS
CE ........................................................................................................... counter electrode
CI......................................................................................................... chemical ionization
CID ......................................................................................collision-induced dissociation
CLZ ..................................................................................................................... clozapine
CRM.................................................................................................. charge residue mode
CSD .............................................................................................. charge state distribution
DESI............................................................................. desorption electrospray ionization
EC ............................................................................................................ electrochemistry
ECD....................................................................................... electron capture dissociation
EI ........................................................................................................... electron ionization
EIC ........................................................................................ extracted ion chromatogram
EID ....................................................................................electron ionization dissociation
ESI.................................................................................................. electrospray ionization
ESSI ......................................................................................electrosonic spray ionization
ETD ...................................................................................... electron transfer dissociation
FAB ...............................................................................................fast atom bombardment
FAPA .................................................................. flowing atmospheric pressure afterglow
GC ...................................................................................................... gas chromatography
HPLC ................................................................ high performance liquid chromatography
IEF....................................................................................................... isoelectric focusing
IEM .................................................................................................. ion evaporation mode
18
IR............................................................................................................................ infrared
LC .................................................................................................. liquid chromatography
LS-DESI................................................ liquid sample desorption electrospray ionization
MALDI ........................................................... matrix-assisted laser desorption ionization
m-NBA ............................................................................................ m-nitrobenzyl alcohol
MS ......................................................................................................... mass spectrometry
MS/MS ..................................................................................... tandem mass spectrometry
MT.......................................................................................................... methoxytryamine
m/z ...................................................................................................... mass-to-charge ratio
nano-DESI................................................... nanospray desorption electrospray ionization
PESI ..................................................................................... probe electrospray ionization
RE ........................................................................................................ reference electrode
RTIL .................................................................................... room temperature ionic liquid
S/N ..................................................................................................... signal-to-noise ratio
TIC ................................................................................................ total ion chromatogram
TFA ...................................................................................................... trifluoroacetic acid
UPLC ............................................................... ultra performance liquid chromatography
UV ...................................................................................................................... ultraviolet
WE .........................................................................................................working electrode
19
CHAPTER 1: INTRODUCTION
1.1 Mass Spectrometry
Mass spectrometry (MS) is an analytical chemical technique that identifies
chemical compounds based on their mass-to-charge ratios (m/z). Nowadays it has become
a very powerful analytical technique for both qualitative and quantitative analysis. In
addition, tandem mass spectrometry (MS/MS) could be used to produce fragment ions for
structural determination.1-2 With its capability of providing structural and analytical
information, MS has been applied to the characterization of a variety of compounds, from
ranging small organic molecules3 to large biological complexes.4 Coupling other
chromatographic separation methods, such as gas chromatography (GC) and liquid
chromatography (LC), with MS further extends its analytical applications for more
complicated samples. Currently, MS has played an important role in many different
analysis disciplines, including biology,5 medicine,6 pharmacology,7 environmental
sciences,8 etc.
1.2 Mass Spectrometer
Figure 1-1 shows the basic components of a mass spectrometer including a
sample inlet, an ion source, ion optics, a mass analyzer, a detector, and vacuum system. A
sample inlet could be connected with LC or GC for separation purpose. Alternatively,
samples in various physical states (solid, liquid or gas state) could be directly introduced.
An ion source is used to convert analyte to gaseous ions for detection by mass
spectrometer. Ion optics, connecting the ion source and the mass analyzer, focused and
transferred the ions beams generated by ion source to the mass analyzer. A mass analyzer
20
is utilized to measure their m/z of the generated ions. A detector is used to detect the ions
ejected by the mass analyzer and recorded the relative abundance of each of the resolved
ionic species. Finally, a computer is required to control the instrument, acquire and
manipulate data. Among these components, the ion source is one of the most crucial parts
so that the development of novel ionization methods is a highly active research area.
Inlet
Source
Ion Optics
Analyzer
Ion
Detector
Data
System
Vacuum System
Figure 1-1. Basic components of a mass spectrometer
1.3 Ionization Methods
Back to the early 20th Century, classical ionization methods such as electron
ionization (EI)9 and chemical ionization (CI)10 were developed and used. However, these
methods have their limitations. They are only suitable for volatile and non-polar analytes.
In addition, they all work in vacuum condition that makes the instrument operating and
sample handling difficult. These limitations pushed the researchers to develop the new
ionization techniques. With the development of modern ionization methods, such as
atmospheric pressure chemical ionization (APCI),11 electrospray ionization (ESI),12 and
matrix-assisted laser desorption ionization (MALDI),13 the scope of the ionizable
analytes was extended. The MS analytes are no longer limited to volatile molecules; even
larger protein molecule can be ionized. Moreover, the vacuum issue has been solved, the
21
ion source could work under atmospheric pressure. Although these atmospheric
ionization methods expanded the MS application, the limitations, like the complicated
sample preparation process, still exist and encourage the further improvement of the
ionization methods. Recently, the advent of ambient ionization methods allow the
ionization of untreated samples in the open environment.14-15
1.3.1 ESI-MS
ESI was first introduced by John Fenn and Masamichi Yamashita in 1984,12
which brought MS great advance in various application areas. For this seminal work and
contribution, John Fenn won the Nobel Prize in Chemistry in 2002. As shown in Figure
1-2,16 in ESI, the sample solution is usually infused through a piece of metal capillary
with high voltage applied (2~5 kV). With the assistance of nebulizing gas and heating, a
mist of highly charged droplets will be formed at the end of the capillary. As the droplets
travel toward the MS inlet, the solvent evaporates; the charged droplets will shrink into
smaller one; finally the dried analytes ions are formed and enter into the MS for detection.
The ionization mechanism is still under debate: (i) the ion evaporation model (IEM)
suggests that droplets shrink by evaporation until the field strength at their surface is
sufficiently large to assist the field desorption of solvated ions; (ii) the charge residue
model (CRM) assumes that electrospray droplets undergo evaporation and fission cycles,
eventually leading progeny droplets that contain on average one analyte ion or less. In
general, ESI is a powerful ionization technique that can ionize a wide range of analytes,
from small organic compounds to large biomolecules. As a soft ionization method, ESI
produce very little fragments during ionization and works well at atmospheric pressure.
22
Furthermore, ESI as an interface could be used to couple LC with MS based on its liquid
sample analysis characteristics. In this case, ESI become the most commercially equipped
ion source nowadays.
Figure 1-2. Scheme showing an ESI source.16 Reproduced with permission from
WILEY-VCH, Copyright 2001.
1.3.2 DESI-MS
A ambient ionization refers to generate ions directly under ambient condition with
little or no sample preparation.14 In 2004 desorption electrospray ionization (DESI), one
of the most well-known ambient ionization methods, was introduced by Cooks group,
which regarded as a milestone in the development of ionization methods.17 As shown in
Figure 1-3, the DESI apparatus consists of a DESI sprayer, sample surface, and a mass
spectrometer. The DESI sprayer is an electrospray emitter used to create charged
23
microdroplets under high voltage and high pressure nebulization gas. The electrically
charged microdroplets beam is directly onto the solid sample surface to desorb and ionize
sample molecule. The resulting ions travel through the atmospheric pressure interface
into the mass spectrometer. With the capability of surface desorption with little or no
sample preparation, in situ detection, and high throughput analysis, numbers of
applications have been explored with the development of DESI. In general, these can be
classified into the following categories: (I) biological and clinical applications by the
examination of biological surfaces, such as imaging of different tissues, even the intact
bio-specimens, to help disease diagnosis; (II) high-throughput analysis in pharmaceutical
and environmental applications; (III) the forensic security application by the examination
of native surfaces.18-23
Figure 1-3. Scheme showing the apparatus of DESI.17 Reproduced with permission from
Science, Copyright 2004.
24
1.3.3 Liquid Sample DESI-MS
With the advent of DESI, many research groups have made much effort to
improve and modify the DESI apparatus to improve the performance of DESI for
different applications. In typical DESI experiment, liquid samples are died on surface
before ionization. For the purpose of direct analysis a continuous-flowing liquid sample,
our group developed the liquid sample DESI (LS-DESI) in 2008. The prototype
apparatus for liquid sample DESI is shown in Figure 1-4.24 The charged microdroplets
from DESI sprayer are also needed. Instead of analyzing a solid sample, the liquid sample
is continuously infused through a piece of silica capillary onto a surface, the ionization
occurs via the interaction of liquid sample with charged droplets generated by the DESI
sprayer and the generated sample ions are transferred to the MS for detection. Later, the
apparatus was improved by removal of the sample surface. The ionization still occurs
when the charged microdroplets interacts the sample on the tip of the silica capillary.25-27
It has the capability of direct analysis of liquid sample in their native condition, not only
for small molecules but also for large proteins and protein/protein complex. In addition, it
is possible to couple liquid sample DESI-MS with LC,28-29 microfluidics,30 and
electrochemistry (EC).31-35
25
Figure 1-4. Scheme showing the apparatus of liquid sample DESI.24 Reproduced with
permission from Springer, Copyright 2009.
1.3.4 Probe Electrospray Ionization (PESI)
It was well recognized that operating at lower liquid flow rates improves the
detection limits of the sample in electrospray. Since low-flow electrospray uses a
capillary with a small bore diameter of a few μm, care must be taken to avoid capillary
clogging and breaking. To circumvent these problems, the term PESI was introduced by
Hiraoka and his coworkers.36 Unlike ESI using capillaries for sample introduction, a
sample is prepared in a droplet of solution on a surface and loaded by dipping the PESI
probe, a conductive solid probe, into the droplet. Then the droplet of analyte solution is
deposited on the PESI probe. When a high voltage is applied to the probe, the deposited
droplet will deform to form a Taylor cone at the tip of the probe and then emits a spray
26
(Figure 1-5). Compared to ESI or nano-ESI, PESI has the advantages of high salt
37
and
detergent 38 tolerance. In addition, it is rapid and consumes very little of the total sample
solution.39 Several interesting applications of PESI-MS have been found, including
imaging biological tissues,40 monitoring chemical reactions in solution,41 and analyzing
living biological samples in real time.42-43 Very recently, single cell analysis with PESI
for detection of metabolites at cellular and subcellular level has been reported.44
Figure 1-5. Scheme showing the apparatus of PESI.36 Reproduced with permission from
WILEY-VCH, Copyright 2007.
1.4 EC-MS
Both EC and MS play an important role in analytical chemistry. EC coupled with
MS enable identifications of the products or intermediates of electrochemical reactions 45.
27
EC/MS is also used for the mechanistic studies of redox reactions of interest to biology 31,
46
and drug discovery
47-48
. As shown in Figure 1-6, the development of the EC/MS
coupling interface always follows after the progress of different ionization methods.49
The first combination of EC with MS has been introduced in 1971 for in-situ detection of
volatile electrochemical oxidation products by Bruckenstein and Gadde.50 Later, with the
increasing detection requirement of wide scale of sample, such as polar or even charged
analytes that are not suitable for EI-MS, different ionization techniques were continually
developed to combine EC with MS in need, including thermospray (TS),51 fast atom
bombardment (FAB),52 ESI,53 and DESI,31 etc.
Figure 1-6. Timeline for the development of EC/MS with different ionization method.49
Reproduced with permission from RSC, Copyright 2013.
28
1.4.1 EC/ESI-MS
In particular, ESI has been widely used to couple EC with MS for various
analytical applications because ESI could be used for the detection of both small
compounds and large biomolecules. As the high voltage is applied to the capillary and the
MS inlet is grounded, ESI itself is an “electrochemical cell” and could induce
electrochemical reactions.
54
The inherent EC which occurs in ESI source could help to
enhance MS signals of species. The metal-containing analytes, such as metal
porphyrins,55 and ferrocene-based derivatives,56 could undergo an oxidation process
during the ionization to be converted to ionic species and to obtain good signals.
However, the first direct coupling of EC with ESI-MS was reported by Van
Berkel and his co-worker in 1995.53 In their study, three types of EC flow cells were
discussed. The electrochemical cell should be floated or decoupled from the ESI high
voltage.53 The utility of EC/ESI-MS was demonstrated by ionization of neutral
compounds, analysis of EC reaction products and signal enhancement of metal complex
detection.53 With the generation of various types of electrochemical cells to combine with
MS, more applications have been achieved, including monitoring electrochemical
reaction intermediates,57 mimicking metabolic pathway of drug compounds,58 online
chemical derivatization59 , and so on.
1.4.2 EC/DESI-MS
Although EC/ESI-MS has been developed and widely used for years, there are
still some inherent problems of the coupling of EC with ESI-MS. First is voltage conflict.
High voltage is used for ESI to generate charged microdroplets. But for electrolysis, a
29
small voltage is needed for oxidation or reduction. In this case, the small voltage for EC
should be floated decoupled with high voltage for ESI. So, the instrumentation is
complex for EC/ESI-MS. The second problem is the compatibility of solvent and
electrolyte used. MS compatible and volatile solvents such as formic acid, ammonium
hydroxide, and ammonium acetate are used in EC/ESI-MS studies while it is difficult to
employ nonvolatile salts, such as KCl, as an electrolyte due to the possible ion signal
suppression effect of nonvolatile salts and clogging of the capillary.
Once liquid sample DESI is used as the interface for coupling EC with MS, the
major problems mentioned above would be avoided. In EC/DESI-MS (Figure 1-7),
sample solution is introduced by a syringe pump and flows through the EC cell. When
either and oxidation or reduction potential is applied to the working electrode (WE),
analytes will undergo electrolytic reaction. The reaction products will flow out via a short
piece of silica capillary; be desorbed and ionized by charged droplets generated by DESI
probe. In this case, the small potential applied to the electrochemical cell and the high
voltage used for spray ionization are physically separated in EC/DESI-MS, there is no
voltage conflict issue. Moreover, DESI appears to have tolerance for high salt
concentration. Previous results show that a good MS signal can be obtained even when
10 mM NaCl is used as an electrolyte.28 Furthermore, EC/DESI-MS could also avoid the
inherent EC reaction occurring in the ESI source, which can prevent the background
signal from the oxidation of analytes.
With the advantage of combination EC with DESI-MS, several electrochemical
reactions
were
investigated,
including
small
molecule
redox
reaction60
and
30
protein/peptide reduction.32,
34
It could be used for the structural analysis of disulfide
bonding-containing proteins in either top-down or bottom approaches.32, 34-35 Recently,
DESI-MS was also used to capture the fleeting electrochemical reaction intermediates on
the electrode surface by using a rotating working electrode.61-62
Figure 1-7. Scheme showing the apparatus for online coupling of a thin-layer
electrochemical flow cell with DESI-MS.28 Reproduced with permission from ACS,
Copyright 2009.
1.5 LC/DESI-MS
The combination of LC with MS has become one of the most powerful techniques
for the analysis of biomolecules and pharmaceuticals, as it has the capability of LC
separation and the power of mass analysis. Nowadays, ESI is the most commonly
interface for LC/MS. However, there are limitations of LC/ESI-MS. First, the ESI
“friendly” solvent should be used as a mobile phase for ionization. Second, ESI could not
tolerate high flow rate, which would cause a “flood” in the ion source. Normally, under
31
the high mobile phase flow rate, splitting is needed for LC/ESI-MS. Our laboratory has
used DESI for coupling LC with MS which could increase the LC/MS performance.63
Figure 1-8 shows the prototype coupled LC/DESI-MS apparatus. A short piece of fusedsilica capillary was connected to the outlet of the LC column. A liquid jet was formed
and directly ionized by charged microdroplets produced by DESI probe. This is the first
time for DESI to be used as an interface to analyze the LC eluent at the flow rate of 1.8
mL/min. In addition, reactive DESI could also be adopted for online derivatization. The
derivatizing reagent can be doped into the spray solvent to react with the analyte during
the ionization process, which could increase the signal of the analyte. Not only for small
molecules, like LC-separated carbohydrates using N-methyl-4-pyridienboronic acid
iodide as the derivatzing reagent,63 but also for large molecular, separated proteins using
m-nitrobenzyl alcohol (m-NBA) or sulfolane as a supercharging reagent
29
could be
derivatized. Integration of an EC with LC/ESI-MS is also possible, a small peptide,
somatostatin-14, was successfully separated by LC and flowed through a thin-layer flow
cell for electrochemical reduction prior to DESI-MS detection. However, LC/DESI-MS
and LC/EC/DESI-MS still have potential to be modified to achieve better performance
which is what I focused in my PhD study. Now, more and more interesting applications
of liquid sample DESI and other ambient ionization method are studied in our laboratory.
32
Figure 1-8. Scheme showing the apparatus for LC/DESI-MS.63 Reproduced with
permission from ACS, Copyright 2011.
33
CHAPTER 2: A NEW SPLITTING METHOD FOR BOTH ANALYTICAL AND
PREPARATIVE LC/MS
Adapted from Cai, Y. Adam, D. and Chen, H., J. Am. Soc. Mass Spectrom. 2014, 25,
286-292. Copyright 2014, Springer.
2.1 Introduction
Splitting the eluent in LC/MS experiment, in some cases, is needed. Electrospray
ionization, a common ionization method for LC/MS, requires an optimal sample infusion
rate at μL/min level. In addition, post-column splitting is also necessary when preparative
purpose is required in the experiment. Typically, the post-column splitting of LC eluent
can be achieved simply using a Tee splitter, in which one stream channel goes to MS for
detection and the other one goes to the waste or other detector.64-65 However, the
connection Tee splitting would introduce dead volume and backpressure which could
cause the peak broadening or time delay for MS detection.66-67
DESI was originally developed by Cooks and co-workers,17 which has been
introduced to provide direct ionization of analytes with little or no sample preparation.
Besides the analyses of solid samples on surface, in our and other laboratories, we
extended the conventional DESI method for direct liquid sample analysis.24 Our results
show that liquid DESI is a soft ionization method which can be used to ionize high-mass
proteins/protein complexes in solution,68
24
. Furthermore, it is possible to couple liquid
DESI-MS with electrochemical cells33, 69-70 or other separation devices such as microextraction.71 In the previous coupling of LC with MS by liquid DESI,29, 63 LC/DESI-MS
was shown to apply to the analysis of both small organic molecules and large protein
34
molecules. So this LC/DESI-MS is different from LC/DART-MS as DART is typically
limited to small molecule analysis. In these previous experiments,29, 63 LC eluent flowed
out of the LC column as a free jet with a high flow rate, which was ionized via interaction
with a pneumatically assisted DESI spray of a chosen solvent. Therefore, to some extent,
it is hard to operate and easy to cause source “flooding”. Moreover, the collection of
remaining analytes in the eluent after ionization would be difficult due to the dispersion
of the jet sample into a plume during the DESI ionization process. The significance of the
collection of fully separated compounds via LC separation (i.e., preparative liquid
chromatography) is hard to overstate. As a very important protocol to obtain purified
samples, it has been widely adopted in organic and biological research laboratories. Thus,
a new splitting method with fast MS detection is in need to assist both online detection
and online sample collection.
In this study, we present a new splitting interface for LC/MS application based on
fast DESI ionization capability. In this approach, a PEEK capillary tube with a microdrilled orifice on the side wall is used to connect with LC column outlet (Figure 2-1). In
this case, a small aliquot of LC eluent emerges out of the orifice and can be directly
sampled and ionized by DESI, while the remaining analytes can be collected from the
tube outlet. There are several advantages of such a new splitting LC/DESI-MS method.
First, it minimizes the dead volume, because there is no need to use connection tubing to
bridge the splitter and MS. In this case, the peak broadening can be avoided. Second, it is
convenient to collect the purified analytes from the PEEK tube outlet for the preparative
purpose. Third, reactive DESI72-73 can be used for online derivatization without
35
introducing an extra dead volume. Lastly, the application of this proposed splitting
LC/DESI-MS method can tolerate high mobile phase flow rates for shorter run times
DESI spray
probe
N2
LC column
Sample
mixture
DESI
spray
530 µm
510 µm
.
...
.
PEEK tubing
MS
. ...
.. Orifice (i.d. 350 µm)
.
Outlet
PEEK tubing
Scheme 2-1. Apparatus of a new LC/DESI-MS post column splitting method.
2.2 Experimental
2.2.1 Chemicals
Insulin (from bovine pancreas, HPLC grade), ubiquitin (from bovine
erythrocytes), trifluoroacetic acid (TFA), 3-nitrobenzyl alcohol (m-NBA,≥99.5%), Nacetyl-D-glucosamine (≥99%), maltohexaose, sulfadiazine (HPLC grade), sulfamerazine
(HPLC grade, ≥98.8%), and sulfaquinoxaline (HPLC grade, ≥96%) were purchased
from Sigma-Aldrich (St. Louis, MO). Acetic acid was purchased from Fisher Chemicals
(Pittsburgh, PA). HPLC-grade methanol was purchased from Fisher Chemicals
(Pittsburgh, PA).HPLC-grade acetonitrile (ACN) was purchased from EMD Chemicals
Inc. (Billerica, MA).
36
2.2.2 LC Separation Condition
A commercial PerkinElmer HPLC system (Perkin Elmer, Shelton, CT) was used
throughout the experiments. For the protein mixture separation, a Waters XB Bridge TM
300 C4 column (4.6 mm×150 mm) was employed with a trifluoroacetic acid (TFA)containing mobile phase of ACN:H2O:TFA (30:70:0.1 by volume) being used. The
mobile phase flow rate was 1.0 mL/min. For the saccharide mixture separation, an
Agilent ZORBAX ODS C18 column (4.6 mm×250 mm) was used with 0.1% FA in H2O
as the mobile phase at the flow rate of 1.0 mL/min. For the mixture of the sulfonamides, a
monolithic C18 column (Phenomenex, Onyx, 4.6 mm ×100 mm) was adopted. A gradient
elution program from 100% A down to 90% A in 4 min was used (mobile phase
composition A: water and B: ACN) with 4 mL/min flow rate. A 20 μL injection loop was
used for sample loading.
2.2.3 DESI-MS Detection
A Thermo Finnigan LCQ DECA ion trap mass spectrometer was used throughout
the experiments. The commercial ion source of the mass spectrometer was removed to
accommodate a home-built liquid DESI ion source. As shown in Figure 2-1, a short piece
of PEEK tube (i.d. 510 μm; wall thickness: 530 μm; length: 3 cm) with a micro-drilled
orifice (i.d. 350 μm) was connected to the LC column. The orifice was located in the tube
2 cm downstream from the LC column and the tube outlet was slightly bent downward to
facilitate sample collection. The sample eluent flowing out of the orifice underwent
interactions with the charged microdroplets generated from DESI spray for ionization.
Unless specified, the spray solvent for DESI probe was CH3OH/H2O/HOAc (50:50:1 by
37
volume). The injection flow rate is 10 μL/min for DESI probe with 5 kV applied. The
DESI spray probe was placed above the orifice and the distance between the probe and
orifice was about 1-2 mm. The orifice was placed approximately 2 cm away from the MS
inlet.
2.3 Results and Discussion
2.3.1 Protein Detection
To demonstrate the feasibility of the proposed splitting LC/DESI-MS method
(Scheme 2-1), a protein mixture of insulin and ubiquitin was first chosen. For separation,
a C4 column with a mobile phase of ACN:H2O:TFA (30:70:0.1 by volume) was
employed. In peptide and small protein separation, TFA not only serves to adjust pH but
also uses as an ion-pairing agent to promote protein separation. However, during the
ionization process, TFA anions can form strong ion pairs with analytes that could
neutralize or decrease the charges of the protonated analytes. In this case, TFA has been
reported to suppress MS ionization.29 The flow rate is 1.0 mL/min. When the eluent from
LC column flows through the PEEK tube, a small portion of the eluent emerging from the
PEEK orifice was ionized by DESI probe with the assistance of high voltage and
nebulizing gas. As shown in the extracted ion chromatograms (EICs, Figures 2-1a and 21b), two proteins were well separated. The resulting DESI-MS spectra (Figures 2-1c and
2-1d) also clearly display the ionized individual proteins with multiple charge distribution.
In addition, with the help of DESI spray solvent, the TFA suppression effect was
mitigated. It only takes 10 ms for the sample eluent to go through the micro-drilled
orifice (calculation based on the flow rate of eluent through the orifice (300 μL/min) and
38
the orifice dead volume of 50 nL). Therefore, DESI-MS used in this experiment provides
a “near-real time” monitoring of LC eluent-stream in the PEEK tube.
LC/DESI-MS
100
Relative Abundance
0
100
0
+4
a) EIC of insulin
NL: 1.76×106
+6
b) EIC of ubiquitin
NL: 2.69×106
0.0
100
0.5
1.0
1.5
2.0
2.5
3.0
Time (min)
3.5
4.0
4.5
5.0
5.5
+4
c) Insulin
NL : 2.42×105
50
+3
+5
0
100
600
800
1000
d) Ubiquitin
NL : 2.26×105
1400
+6
+7
1600
+8
50
0
1200
+11
+12
600
800
+10
2000
1800
2000
+5
+9
1000
1800
1200
1400
m/z
1600
Figure 2-1. EICs of (a) insulin (+4 ion) and (b) ubiquitin (+6 ion) from LC/DESI-MS
analysis and the corresponding DESI-MS spectra of (c) insulin and (d) ubiquitin.
For comparison purpose, electrosonic spray ionization74 (ESSI, a variant form of
ESI which employs an ESI source with supersonic nebulization gas) was also used to
detect the LC separated analytes. In the experiment, the LC eluent flow rate was reduced
39
to an optimized flow rate at 10 μL/min using the ASI adjustable commercial splitter
(Analytical Scientific Instruments, Richmond, CA), because the high flow rate (1.0
mL/min) could cause the flood for the ESSI ion source. LC separation conditions were
kept the same as mentioned above. As revealed by the resulting EIC spectra (Figures 2-2a
and 2-2b) from ESSI-MS detection, the retention times for two proteins are both delayed
by about 4 min in comparison to DESI detection. This delay is very likely to be caused by
the increased dead volume from the traditional splitting method used in the LC/ESSI-MS
experiment. Also, the two EIC peaks of the proteins are much wider than those recorded
using LC/DESI-MS, each of which lasts about 4 minutes. In addition, the two protein’s
EIC profiles overlap with each other in the retention time window of 8-11 min. Moreover,
Because of the peak broadening and more severe TFA ion suppression effect, the EIC
intensities from ESSI-MS detection (Figures 2-2a and 2-2b) are lower than those from
DESI-MS detection (Figures 2-1a and 2-1b).
40
LC/ESSI-MS
Relative Abundance
100
a) EIC of insulin
NL: 6.42×105
+4
0
100
b) EIC of ubiquitin
NL: 1.65×106
+6
0
0
1
2
3
4
5
6
7
8
9
10
11
12
13
Time (min)
Figure 2-2. EICs of (a) insulin (+4 ion) and (b) ubiquitin (+6 ion) from LC/ESSI-MS
analysis.
Reactive DESI for online derivatization is an additional benefit in this LC/DESIMS method. Reactive DESI is performed simply by doping a chosen chemical reagent
into the DESI sprays solvent for supercharging proteins and for enhancing their signals.
As the supercharging occurs during DESI ionization, the online supercharging would not
introduce an extra dead volume. The charge enhancement for proteins by supercharging
has great value for further structural analysis via top-down approaches for increasing the
fragmentation efficiency.75-76 In this study, DESI spray solvent of CH3OH:H2O:HOAc
(50:50:1 by volume) was doped with 50 mM m-NBA to effect “supercharging” proteins
eluted from LC column. The m-NBA, an effective supercharging reagent,28 was used in
41
the experiment. As m-NBA was doped in DESI spray solvent, the maximum charge state
of insulin shifted from +5 to +6 and the charge with the highest abundance shifted from
+4 to +5 (Figure 2-3c) compared to the regular DESI data (Figure 2-1c). Likewise, the
maximum charge state of ubiquitin shifted from +12 to +13 and the charge with the
highest abundance shifted from +6 to +7 (Figure 2-3d). In addition, compared LC/DESIMS (Figures 2-1a and 2-1b) and ESSI ionization (Figures 2-2a and 2-2b), the intensity of
resulting protein ions in this LC/reactive DESI-MS (Figures 2-3a and 2-3b) are
significantly enhanced. Presumably, m-NBA reduces TFA dissociation77 so that the
decreased concentration of trifluoroacetate anions reduces their ion pairing interaction
with protein cationic sites, a mechanism responsible for signal suppression of TFA.
42
LC/Reactive DESI-MS, with m-NBA
100
+5
a) EIC of insulin
NL: 2.83×107
0
100
+7
b) EIC of ubiquitin
NL: 2.02×107
0
0.0
0.5
1.0
1.5
2.0
c) Insulin
NL : 2.15×106
100
2.5
+5
3.0
3.5
4.0
Time (min)
4.5
5.0
5.5
6.0
+4
50
+6
+3
0
600
100
800
1000
d) Ubiquitin
NL : 1.30×106 +8
1200
1400
0
1800
+6
+10
+11
+12
+13
600
800
2000
+7
+9
50
1600
+5
1000
1200
m/z
1400
1600
1800
2000
Figure 2-3. EICs of (a) insulin (+5 ion) and (b) ubiquitin (+7 ion) from LC/reactive
DESI-MS analysis and the corresponding reactive DESI-MS spectra of (c) insulin and (d)
ubiquitin.
Besides the analytical strength, another important feature of this method is to
collect isolated samples following LC/MS analysis. Indeed, with the aid of online DESI
monitoring as mentioned above, the major portion of insulin and ubiquitin exiting from
the PEEK tube outlet (70%) were collected. The collected samples were re-analyzed by
43
MS (directly infusion) and gave rise to the spectra of isolated insulin and ubiquitin. As
can be seen in Figures 2-4a and 2-4b, there is no cross-contamination. The results
confirm the two samples were completely separated and successfully collected.
The eluent splitting ratio provided by the PEEK tube through the orifice can be
adjusted simply by reducing the orifice i.d.. When the orifice i.d. decreased from 350 μm
down to 100 μm, the splitting ratio of 3:7 can be reduced to 4:96 for which more sample
can be collected. An experiment was conducted to evaluate the performance of the PEEK
tube with the 100 μm i.d. orifice. For this case, the time for the eluent to flow through the
orifice channel was 6 ms (the split flow rate: 40 μL/min; orifice dead volume: about 4
nL). By using a mixture of insulin and ubiquitin as a test sample, the EIC spectra
acquired show good protein separation (Figures 2-4a and 2-4b). In comparison to Figures
2-1a and 2-1b, the signals of both proteins are even higher. Also, the proteins after LC
separation were collected and re-analysis of the collected samples gave pure spectra
(Figures 2-4c and 2-4d). We further quantified the yield of protein collection using UV
spectroscopy (PERKIN ELMER 785A UV/VIS Detector). The UV detection wavelength
was selected at 220 nm. Insulin and ubiquitin standard stock solutions were diluted into
the appropriate concentration ranges (0.625-10.5 μM for insulin and 0.55-8.8 μM for
ubiquitin) using the mobile phase solvent to establish the calibration curves (Figure 2-5).
The peak areas of collected insulin and ubiquitin solution were used to calibrate the
concentrations of the collected protein solutions. Then, based on the collected volume,
the moles of collected samples can be obtained to compare with the original amount of
injected proteins for LC separation (2 nmol) to get the protein collection yields, which
44
was 94.2±0.8% average yield for insulin and 94.6±1.9% average yield for ubiquitin.
These yields are fairly close to the theoretical value of 96% based on the splitting ratio,
suggestion the feasibility and potential of this approach for sample purification.
45
100
+4
c) Collected insulin
100 μM i.d. orifice
+3
LC/DESI-MS, 100 μm i.d. orifice
50
Relative Abundance
100
a) EIC of insulin
NL: 5.11×106
0
0
100
100
b) EIC of ubiqutin
NL: 3.34×106
+5
800
1000
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Time (min)
4.5
5.0
5.5
1400
1600
1800
2000
1800
2000
+6
d) Collected ubiquitin
100 μM i.d. orifice
+7
+5
+8
50
0
1200
+9
6.0
+10
0
+11
+12
800
1000
1200
m/z
1400
1600
Figure 2-4. EICs of (a) insulin (+4 ion) and (b) ubiquitin (+6 ion) from LC/DESI-MS
analysis (100 μm i.d. orifice) and the collected (c) insulin and (d) collected ubiquitin.
46
800000
700000
y = 54901x + 93988
R² = 0.9992
Peak Area
600000
500000
400000
300000
200000
100000
0
0
2
4
6
8
10
12
Insulin Concentration (μM)
700000
y = 62847x + 67220
R² = 0.9993
Peak Area
600000
500000
400000
300000
200000
100000
0
0
1
2
3
4
5
6
7
8
9
10
Ubiquitin Concentration (μM)
Figure 2-5. Calibration curves of insulin and ubiquitin (UV absorption peak area wrt.
protein concentration).
2.3.2 Saccharide Detection
Conventionally, in the preparative LC experiment, the collection of LC-separated
eluent is often enabled by placing a UV detector in between the LC separation column
and the sample collection reservoir. However, such an approach is limited to compounds
with chromophores. For those without chromophores, derivatization is in need, which is
time consuming and troublesome. By using online DESI monitoring in our method, this
problem can be solved as MS is a detector.
47
As a demonstration, a saccharide mixture consisting of N-acetyl-D-glucosamine
(NAG) and maltohexaose (no or weak UV absorption) was chosen to test feasibility of
the LC/DESI-MS. A C18 column was used for separation. As shown in the recorded EIC
spectra, NAG and maltohexaose were well separated (Figures 2-6a and 2-6b). Although
there is no organic solvent used in the mobile phase (only 0.1% FA in H2O was used as
the mobile phase), the saccharides can still be detected by DESI-MS because of the
freedom of DESI probe to choose favorable spray solvent of CH3OH/H2O/FA (50:50:1
by volume) for sample ionization. Again, in contrast, when ESSI was used for detection
(in conjunction with the employment of the commercial splitter to obtain an optimized
flow rate of 20 μL/min for ionization), wider peaks with lower intensities were resulted.
As shown in Figures 2c and 2d, the peaking broadening occurred and both of the
saccharide compounds have a peak width of 2~3 min. As a result, the EIC profiles of
NAG and maltohexaose overlap with each other (Figures 2-6c and 2-6d) and the ESSIMS spectra recorded also show the presence of both NAG and maltohexaose (data not
shown). Lower intensities are also observed in the ESSI-MS detection (Figures 2-6c and
2-6d) in comparison to DESI-MS detection (Figures 2-6a and 2-6b). Two factors might
contribute to the low intensities in the LC/ESSI-MS experiment; one is due to peak
broadening as mentioned above and the other one might be owing to the low ionization
efficiency of analytes in aqueous mobile phase without organic solvent (0.1% FA in H2O
in
this
case).
By
contrast,
in
DESI-MS
detection,
the
spray
solvent
(CH3OH/H2O/HOAc=50:50:1 by volume) yielded favorable for ionization. This
phenomenon is also in agreement with previous reports.29 Following the LC/DESI-MS
48
analysis, the remaining portion of saccharides flowing out of the PEEK tube outlet was
collected and re-tested with ESSI, which show clearly the full separation of two
saccharide compounds (Figures 2-6e and 2-6f).
49
[NAG+H]+
100
[NAG+H]+
a) LC/DESI-MS
[2NAG+H]+
e) Collected N-acetyl-D-glucosamine
100
EIC of N-acetyl-D-glucosamine
NL: 2.70×106
0
Relative Abundance
100
[Maltohexaose+H]+
b) LC/DESI-MS
50
[NAG-H2O+H]+
EIC of maltohexaose
NL: 1.75×106
[3NAG+H]+
[4NAG+H]+
0
0
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
200
5.0
100
[NAG+H]+
100
300
400
500
600
700
800
900
1000
[Maltohexaose+H]+
f) Collected maltohexaose
c) LC/ESSI-MS
EIC of N-Acetyl-D-glucosamine
NL: 2.53×105
50
[Maltohexaose+H]+
0
100
d) LC/ESSI-MS
[Maltohexaose+Na]+
EIC of Maltohexaose
NL: 7.16×105
0
223
0
0
1
2
3
4
5
6
7
8
200
300
400
500
Time (min)
600
700
800
900
1000
m/z
Figure 2-6. EICs of (a) N-acetyl-D-glucosamine and (b) maltohexaose eluted out of the
C18 column and detected by DESI-MS. EICs of (c) N-acetyl-D-glucosamine and (d)
maltohexaose eluted out of the C18 column and detected by ESSI-MS. MS spectra of (e)
the collected N-acetyl-D-glucosamine and (f) the collected maltohexaose. The re-analysis
of the collected proteins was performed using ESSI-MS. The peak at m/z 223 in (f) came
from solvent background
2.3.3 Sulfonamides Detection
Besides the applications to LC separation using regular analytical columns, we
also examined the compatibility of our splitting method to monolithic column-based
ultra-fast LC separation. The monolithic column is a kind of “single-piece column” that
has been developed to tolerate fast high-throughput analysis.78 It allows performing high-
50
throughput and maintaining good separation performance close to that of ultra-high
performance LC. The key factor in such an LC experiment is that an extremely high flow
rate (up to 9 mL/min) can be used with monolithic columns without causing significantly
high back pressure. Thus the separation can be completed in a short period of time. In our
experiment, a mixture of sulfonamides including sulfadiazine, sulfamerazine, and
sulfaquinoxaline, drugs commonly used for eliminating bacteria and treating urinary tract
infections, was chosen as a test sample for demonstration of the application of our
splitting method to the monolithic column-based ultra-fast LC separation. As shown in
EIC spectrum recorded by DESI-MS (Figure 2-7a), the sulfadiazine, sulfamerazine, and
sulfaquinoxaline were all separated and eluted within 3 min. While, when a regular
reversed-phase C18 column was used at 1 mL/min flow rate separation, the total elution
time was 11 min (data not shown).This result shows that the high flow rate does help fast
elution and separation. This result shows that the high flow rate does help fast elution and
separation. We also recorded the chromatogram using an UV detector (detection
wavelength: 254 nm) for comparison (Figure 2-7b). The DESI-MS recorded
chromatogram has slightly better resolution than that recorded by UV detector. In figure
2-6b, there is about 10% overlap of the sulfadiazine and sulfamerazine peaks in the UV
chromatogram, while the two peaks are well resolved in the DESI-MS chromatogram.
However, attempt to split the high flow eluent of 4 mL/min down to 10 μL/min for ESSIMS detection by the commercial splitter caused column failure due to high backpressure
that exceeded the tolerance of the monolithic column used. These results prove that in our
51
split method in conjunction with DESI-MS detection, there is small dead volume and
negligible backpressure.
100
[Sulfamerazine + H]+
a) LC/DESI-MS
[Sulfaquinoxaline + H]+
Relative Abundance
[Sulfadiazine + H]+
0
0.0
0.5
1.0
b) LC/UV
Sulfadiazine
100
0.0
0.5
1.5
1.0
1.5
2.0
2.5
Sulfamerazine
3.0
3.5
Sulfaquinoxaline
2.0
2.5
3.0
3.5
Time (min)
Figure 2-7. EIC spectra of a mixture of sulfadiazine, sulfamerazine, and sulfaquinoxaline
(a) from LC/DESI-MS analysis and (b) from LC/UV analysis.
2.4 Conclusions
This study suggests a new and versatile splitting method for LC/MS coupling, as
an addition example of combining LC with MS using ambient ionization methods. The
LC/DESI-MS method via PEEK tubing with a micro-drilled orifice has the advantages
including premium analytical performance of DESI and successful sample purification
with high collection yield. The presented methodology has striking features involving
narrow peak width, high sensitivity, no back pressure issue and low cost. The freedom to
52
choose DESI spray solvents allows online supercharging separated proteins and for
enhancing their signals. Owing to DESI appears to be compatible with both small organic
molecules and protein/peptide, this LC/DESI-MS method would have wide applications
in bioanalysis.
53
CHAPTER 3: COUPLING OF ULTRAFAST LC WITH MASS SPECTROMETRY BY
DESI
Adapted from Cai, Y. Liu, Y. Helmy, R. and Chen, H., J. Am. Soc. Mass Spectrom. 2014,
25, 1820-1823. Copyright 2014, Springer.
3.1 Introduction
LC/MS is powerful for analysis of complicated samples and there is a growing
demand for fast separation to reduce cost and time. To improve the separation speed and
efficiency, several innovative strategies have been developed, including monolithic or
fused-core columns, high-temperature LC and ultra-performance liquid chromatography
(UPLC).79 Among these innovations, monolithic LC employing a stationary phase
formed from a monolithic porous rod is one of the breakthroughs in the column design.
The porous rod structure provides high permeability, numerous channels, and a high
surface area for interacting with analytes.80 Therefore the column can have a low
hydraulic resistance to allow very high elution flow rate (e.g., 9 mL/min) for ultrafast
separation (in few minutes) or high-throughput analysis. In addition, the column is
tolerant with untreated raw or viscous samples because of its porous structure. In the case
of UPLC, the packing particle size is decreased to less than 2.0 µm, therefore, the
analysis time can be reduced and chromatographic resolution can be enhanced.81
We previously presented a DESI interface to combine LC with MS using a new
LC eluent splitting strategy via a tiny orifice drilled on a capillary tube.82 In the
experiment, a small portion of LC eluent leaking out of the capillary orifice can be
instantaneously desorbed and ionized by DESI while the remaining majority of LC eluent
54
can be collected for preparative purpose. Splitting the eluent in LC/MS experiments is
also necessary under the circumstance when the mobile phase flow rate is too high for
direct MS ionization. In contrast to using tradition splitter, splitting via a capillary orifice
introduces negligible dead volume and back pressure, thus enabling nearly real-time
DESI-MS detection and online MS-directed collection of LC-separation species.82
This study focuses on further investigating the performance of our DESI-MS, in
conjunction with the capillary orifice splitting technique, for coupling with ultrafast LC
based on either monolithic column or UPLC-based separation. The demonstrated
examples include LC/DESI-MS analysis of metabolites in human urine, drugs-of-abuse
in a cola drink, acidic anti-inflammatory drug mixture and insulin peptic digest. Several
advantages of our LC/DESI-MS in this study have been revealed. First, high elution flow
rate used in the monolithic column LC experiments (3 mL/min) led to a very short total
separation time (1.6~3.5 min). Second, DESI-MS-directed purification allowed us to
online collect LC-separated metabolites for further structural elucidation by NMR. Third,
with reducing the capillary orifice to 50 µm, the splitting ratio was lowered down to ca.
1:99 (i.e., only 1% sample was used for DESI-MS detection), offering “nearly nondestructive” MS detection with simultaneous online collection of 99% purified analytes.
Fourth, we demonstrated a novel way to carry out “wrong-way around” ionization for
LC/MS. Acidic compounds that were separated using acidic mobile phase solvent could
be directly ionized into negative ions by DESI using basic spray solvent. It is
advantageous as acidic compounds generally have high ionization efficiency in the
negative ion spray ionization mode rather than in the positive ion mode. In addition, for
55
the first time, DESI was shown to be applicable for combining MS with UPLC. For the
analysis of drugs-of-abuse in cola drink, the total separation process was completed
within 2 min. Comparable sensitivity was obtained with DESI interface in comparison
with using commercial Waters ESI.
3.2 Experimental
3.2.1 LC/DESI-MS Instrument
The schematic of the home-built liquid DESI ion source and the PEEK capillary
tube with a drilled micro-orifice is shown in Scheme 2-1 and details were mentioned
above. Briefly, a short piece of PEEK capillary tube (i.d. 510 μm; wall thickness: 530 μm;
length: 2.5-3 cm) with a micro-orifice (i.d. 100 μm) drilled through the tube wall 0.5~1
cm upstream to the tube outlet was used for LC eluent splitting. In the case of monolithic
column separation, the PEEK tube was directly connected to the LC column. A
commercial PerkinElmer HPLC system (Perkin Elmer, Shelton, CT) was adopted for LC
separation. MS and MS/MS analysis were performed employing a Thermo Finnigan LCQ
DECA-MAX ion trap mass spectrometer (San Jose, CA).
In the case of UPLC separation, a Waters ACQUITY UPLC® System was used
and MS analysis was performed using a Waters Xevo QTOF (Milford, MA). A piece of
extension connection capillary (i.d. 25.4 μm; length: 30 cm), originally used for
connecting UPLC column with ESI source was used to bridge the PEEK capillary tube
with UPLC column due to the distance between UPLC column and the MS inlet. Unless
otherwise specified, the spray solvent for DESI was CH3OH/H2O/HOAc (50:50:1 by
volume) for the positive ion mode and CH3OH/H2O/NH4OH (50:50:2 by volume) for the
56
negative ion mode. The spray solvent for DESI was injected at 10 μL/min with + 5 kV
applied for the positive ion mode or -5 kV applied for the negative ion mode.
3.2.2 LC Separation Conditions
In all monolithic column LC experiments, the mobile phase elution flow rate was
kept at 3.0 mL/min and a 20 μL injection loop was used for sample loading. Under such a
flow rate, the splitting ratio was measured to be 4:96 when the orifice i.d. was 100 μm i.d.
Specifically, Phenomenex Onyx Monolithic C18 column (4.6×100 mm) was employed for
the separation of dopamine, 3-methoxytyramine, L-tryptophan and L-kynurenine in urine.
The urine sample doped with these compounds was diluted 1:1 (v/v) with water to have
the final concentration of 60 μM for each species, and then filtered with 0.2 μm Nylon
Membrane for removing possible particulates before LC injection. Solvent A was 0.1%
FA in H2O, and solvent B was 0.1% FA in ACN with elution program as such: 0-1min, 1%
B; 1-3min, 1% B was ramped to 50%. For the separation of acidic anti-inflammatory
drugs, isocratic mobile phase of H2O: MeOH: FA (30:70:0.05% by volume) was used
and the Phenomenex Onyx Monolithic C18 column (100×4.6 mm) was also chosen.
For the UPLC experiments, ACQUITY UPLC® BEH C18 column (50×2.1 mm)
was employed for the separation. For analysis of drugs in Pepsi, the Pepsi sample
contained codeine, cocaine, and flunitrazepam (diluted using water to 15 μM). The
mobile phase consisted of A: 0.1% FA in H2O and B: 0.1% FA in ACN. A linear gradient
program ran from 23% to 90% solvent B in 3 min. With the mobile phase elution flow
rate kept at 0.3 mL/min, the splitting ratio using the PEEK capillary tube carrying the 100
μm i.d. orifice was 1:1.
57
3.3 Results and Discussion
3.3.1 LC/DESI-MS Analysis of Drugs
First we showed the capability of our LC/DESI-MS for direct analysis of
neurotransmitters, amino acids and their metabolites in biological matrices with
minimum sample clean-up. In this experiment, dopamine, 3-methoxytyramine, Ltryptophan and L-kynurenine were chosen as test samples (note that 3-methoxytyramine
and L-kynurenine are the metabolites of neurotransmitter dopamine and amino acid Ltryptophan, respectively) and dissolved in human urine. As shown in the extracted ion
chromatograms (EICs, Figures 3-1), these compounds were well separated in less than
1.6 min. In the resulting DESI-MS spectra shown in Figure 1 insets, all protonated
molecules are clearly observed. No interference from urine was noted, probably because
salts in urine matrix would elute out without retention. Also, a fragment ion by loss of NH3 group from the protonated ion molecule was observed in both dopamine and 3methoxytyramine DESI-MS spectra (Figure 3-1a and 3-1b insets).
58
100
a) EIC of dopamine in urine
NL: 1.71×106
154
100
+H ]+
[
50 [154-NH3]
0
150
m/z
0
b) EIC of 3-methoxytyramine in urine
NL: 7.11×106
100
Relative Abundance
100
168
0
150
c) EIC of L-Kynurenine urine
100
NL: 1.47×106
[
100
250
+H]+
[
[168-NH3]
50
0
200
m/z
200
250
209
+H ]+
50
0
150
m/z
0
d) EIC of L-tryptophan in urine
NL: 8.70×105
100
200
250
205
100
+H ]+
[
50
0
150
0
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
m/z 200
2.0
250
2.2
Time (min)
Figure 3-1. EICs of (a) dopamine, (b) 3-methoxytyramine, (c) L-kynurenine, and (d) Ltryptophan acquired by LC/DESI-MS.
In comparison, the commercial ESI source was used. The high flow of LC eluent
of 3.0 mL/min that would “flood” the ESI ion source, splitting was in need. A home-built
Tee splitter was used to reduce the flow rate to an optimal value of 100 μL/min for the
ESI detection without causing too much back pressure. ESI-MS data was acquired with
other experimental conditions kept the same as those used in LC/DESI-MS analysis
mentioned above. As revealed by the resulting EIC spectra (Figures 3-2), the signals of
the four compounds recorded by ESI are 20~40 fold lower than those detected by DESI.
It is probably caused by the high proportion (ca. 85~95%) of water in the mobile phase
which disfavored ion desolvation. The regular spray solvent for DESI could help to
59
increase the analyte ionization efficiency under such a circumstance. The sensitivity of
the LC/DESI-MS was good, with using the LCQ DECA MAX mass spectrometer in the
experiment. The evaluation of the LC/DESI-MS sensitivity was performed using single
ion monitoring (SIM) mode. With the injection concentration of 250-620 ng/mL
(corresponding to ca. 0.2-0.5 ng sample injected into the DESI ion source), a signal to
noise ratio (S/N) of 5-14 was obtained, which are similar to the reported results by
Schiavo S. et al.
LC/ESI-MS
Relative Abundance
100
a) EIC of dopamine
NL: 2.40×104
0
100
b) EIC of 3-methoxytyramine
NL: 1.77×105
0
100
c) EIC of L-kynurenine
NL: 4.49×104
0
100
0
d) EIC of L-tryptophan
NL: 3.53×104
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Time (min)
1.4
1.6
1.8
2.0
Figure 3-2. EICs of (a) dopamine, (b) 3-methoxytryamine, (c) L-kynurenine, and (d) Ltryptophan acquired by LC/ESI-MS.
In order to enhance the MS-directed sample collection yield, it would be
beneficious to minimize the sample consumed for MS detection. This can be achieved by
60
reducing the orifice i.d.. In the past, the splitting ratio was adjusted from 30:70 to 4:96 by
reducing the orifice i.d. from 350 μm to 100 μm.82 In this study, with further reducing the
orifice i.d. from 100 μm down to 50 μm, the splitting ratio of 0.9:99.1 (approximately
1:99) was obtained. This ratio would favor improving the sample collection yield up to
99% with only 1% samples being consumed for DESI-MS (i.e., “nearly non-destructive”
MS detection). The experiment was conducted to testify this hypothesis and the mixture
of dopamine, 3-methoxytyramine, L-tryptophan and L-kynurenine wer again chosen as
the test sample. The four samples were well separated by the C18 monolithic column and
successfully detected online by DESI-MS using the smaller orifice (50 μm i.d., data not
shown). During the run of LC/DESI-MS, we collected 3-methoxytyramine from the
PEEK capillary tube outlet and quantified the collection yield using UV absorption
spectroscopy. LC with a UV detector (detection wavelength was selected at 210 nm) was
used for the quantification of the DESI-MS-directed collection yield. 3-Methoxytyramine
(3-MT) standard stock solutions were diluted into the standard solutions with five
different concentrations (1, 2, 5, 10, and 20 μM), which were analyzed three times by
LC/UV using the same LC separation condition described previously for monolithic C18
column procedure to establish the calibration curve (Figure 3-3). The collected 3-MT
solution was also injected for LC/UV analysis in triplicate measurements. The peak areas
of collected 3-MT solution were brought into the regression equation of the calibration
curve to get the concentrations of the collected 3-MT solutions. Then, based on the
collected volume, the moles of collected samples can be obtained to compare with the
original amount of injected 3-MT for LC separation to obtain the collection yield of 98.6%
61
± 0.6 on average. This collection yield is fairly close to the actual splitting ratio value of
99.1%.
100000
90000
y = 4381.8x - 2146.8
R² = 0.9963
80000
Peak Area
70000
60000
50000
40000
30000
20000
10000
0
0
5
10
15
20
25
3-MT Concentration (μM)
Figure 3-3. Calibration curves of 3-MT (UV absorption peak area vs. 3-MT
concentration)
3.3.2 LC/DESI-MS analysis of Acidic Compounds
Acidic mobile phases are often used in LC. However, the acidic mobile phase
represents a dilemma for ionization of acidic analytes by ESI, as acidic analytes tend to
form negative ions rather than positive ions. Typically, the eluent needs to be adjusted to
alkaline pH using a Tee junction to introduce base which would lead to increased postcolumn dead volume. In contrast, DESI-MS can provide a simpler approach to tackling
this challenge, simply by spraying basic solvent to achieve the so-called “wrong-way
around” ionization to generate negative ions. In this study, we investigated LC/DESI-MS
analysis of acidic anti-inflammatory drug mixture consisting of ketoprofen, fenoprofen,
and ibuprofen with LC separation using traditional acidic mobile phase and online DESI-
62
MS detection in the negative ion mode. As shown in EIC spectra recorded by DESI-MS
employing a spray solvent of MeOH:H2O:NH4OH (50:50:2 by volume, Figures 3-4), the
ketoprofen, fenoprofen, and ibuprofen were all separated and eluted within 2.5 min by
isocratic elution using mobile phase of H2O:MeOH:FA (30:70:0.05 by volume). In
comparison, negative ion mode ESI-MS was directly used for the detection of the
separated compounds under the same separation condition except LC eluent flow rate
was reduced using a home-built Tee splitter to an optimal rate of 100 μL/min. The signals
of ketoprofen and fenoprofen were weaker than those of DESI-MS detection and
ibuprofen could not be detected (Figure 3-4), probably because the formic acid present in
the mobile phase suppressed the deprotonation of the analytes in negative mode ESI
ionization process.
63
LC/DESI-MS
100
LC/ESI-MS
100
a) EIC of Ketoprofen
NL: 5.30×105
a’) EIC of Ketoprofen
NL: 2.03×105
Relative Abundance
0
0
100
b) EIC of Fenoprofen
NL: 3.64×105
100
b’) EIC of Fenoprofen
NL: 1.25×105
0
0
100
0
0.0
c) EIC of Ibuprofen 100
NL: 1.15×105
0.5
1.0
1.5
2.0
Time (min)
2.5
3.0
0
0.0
c’) EIC of Ibuprofen
0.5
1.0
1.5
2.0
Time (min)
2.5
3.0
Figure 3-4. EICs of (a) ketoprofen, (b) fenoprofen, and (c) ibuprofen acquired by
LC/negative ion DESI-MS. EICs of (a') ketoprofen, (b') fenoprofen, and (c') ibuprofen
acquired by LC/negative ion ESI-MS.
3.3.3 UPLC/DESI-MS Analysis
Besides monolithic column LC, DESI-MS is also applicable to UPLC
applications. For the analysis of a drug mixture of cocaine, codeine, and flunitrazepam in
cola, the three drugs were well separated in less than 2 min using UPLC (Figure 3-5). The
CID data were also obtained during LC separation, for ion structural confirmation. CID
of the protonated codeine at m/z 300 (Figure 3-5a inset) yielded the fragment ions m/z
243 and 215 by consecutive losses of C3H5O and CO. Direct loss of one molecule of H2O
from m/z 300 produced fragment ion at m/z 282. For cocaine, fragment ion of m/z 182
was generated by CID of the protonated cocaine at m/z 304 via the loss of PhCOOH
(Figure 3-5b inset). In the flunitrazepam MS/MS spectrum (Figure 3-5c inset), CID of the
64
protonated flunitrazepam ion at m/z 314 yielded the fragment ions of m/z 268 and 240 by
consecutive losses of NO2 and CO, in agreement with our previous observation. Direct
loss of CO due to ring contraction was also noted to form a fragment ion of m/z 286. This
result shows the strength of UPLC as it could take over 10 min for the separation using
traditional HPLC.83
300
m/z 300
100
a) EIC of Codeine in Cola
NL: 5.10×103
215
- C3H5O
- CO 243
282
- H2O
200
220
Relative Abundance
0
100
182
240
280
260
300
340
m/z 304
b) EIC of Cocaine in Cola
NL: 7.34×103
304
- PhCOOH
160
0
100
320
200
240
280
320
268
m/z 314
c) EIC of Flunitrazepam in Cola
NL: 2.63×104
- NO2
314
286 - CO
240 - CO
0
0.20
0.40
0.60
0.80
1.00
1.20
1.40
1.60
1.80
200
220
240
260
280
300
320
2.00
2.20
2.40
2.60
2.80
3.00
3.20
340
Time (min)
Figure 3-5. EICs of a) codeine, b) cocaine and c) flunitrazepam acquired by
UPLC/DESI-MS with MS/MS spectra are given in the figure insets. The red lines shown
in the molecule structures indicate the bond cleavage upon CID.
When the temperature of the column was elevated to 80 oC, the elution flow rate
for UPLC separation could be increased to 1.0 mL/min without causing backpressure
65
beyond the system limit. Under this condition, the LC separation was completed in 45 s,
and the analytes were all well detected by DESI-MS (Figure 3-6 a-c). In addition, the
separated drugs were also successfully online collected with the aid of DESI-MS
detection (Figure 3-6 d-f).
66
HTUPLC/DESI-MS
Relative Abundance
100
a) EIC of Codeine in Cola
NL: 2.37×104
0
100
b) EIC of Cocaine in Cola
NL: 4.82×103
0
100
0
Relative Abundance
100
c) EIC of Flunitrazepam in Cola
NL: 3.69×103
0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 2.20 2.40 2.60 2.80 3.00 3.20
Time (min)
300
0
d) Re-analysis of codeine
304
100
e) Re-analysis of Cocaine
0
314
100
0
295
f) Re-analysis of Flunitrazepam
300
305
310
315
320
Figure 3-6. EICs of a) codeine, b) cocaine and c) flunitrazepam acquired by high
temperature UPLC/DESI-MS (80 oC column temperature). Re-analysis of collected d)
codeine, e) cocaine and f) flunitrazepam after high-temperature UPLC separation.
3.4 Conclusions
This study demonstrates the combination of DESI-MS with ultrafast LC
separation using monolithic and UPLC columns. The splitting via a capillary orifice
allows fast LC elution for rapid separation and online MS-directed purification with
recovery yield up to 99%. The DESI also allows direct and efficient ionization of acidic
67
compounds in the acidic eluent. These strengths of DESI-MS would lead to many useful
bioanalytical applications.
68
CHAPTER 4: INTEGRATION OF ELECTROCHEMISTRY WITH ULTRA
PERFORMANCE LIQUID CHROMATOGRAPHY/MASS SPECTROMETRY
(UPLC/MS)
Adapted from Cai, Y.; Zheng, Q.; Liu, Y. Helmy,; R. Loo. J,; and Chen, H., Eur J Mass
Spectrom. 2015, 21, 341–351. Copyright 2015, IM Publications.
4.1 Introduction
Redox-active disulfide bonds are one of the most common protein posttranslational modifications. The disulfide bond can maintain three-dimensional protein
structures and their biological activities.33 However, the presence of disulfide bone
linkages increases the complexity protein structure analysis determined by MS. EC has
been introduced as a new protocol to couple with DESI/MS for the structural analysis of
disulfide bond containing proteins/peptides in either top-down or bottom-up. However, in
such a study, the reduced linear peptide chains are hard to be assigned to their precursors
and sequenced. The EC/DESI-MS method would be further benefited if UPLC could be
coupled for fast separation.
The combination with LC further broadens EC/MS applications. Previous studies
showed that post-column electrochemical conversion in LC/MS could be used to increase
the MS detection sensitivity of the target compounds. In such an experiment, an
electrochemical flow cell is placed in between the LC and MS to convert the LCseparated compounds to more polar or even charged products, which are suitable for MS
detection with increased ionization efficiency.84-86 But no disulfide bond reduction was
studied using LC/EC/MS system. In addition, the mobile phase flow rate should be
69
considered to be compatible with the use of the electrochemical cell.86-87 In this case,
UPLC is a good choice to for this combination. UPLC does not require a high elution
flow rate to achieve fast separation, thus making it a better fit for LC/EC/MS experiments
than HPLC. It also can significantly shorten the separation time by 10 times in
comparison to conventional HPLC. However, UPLC has not been reported to couple with
EC yet.
In this study, we developed the LC/EC/MS method using UPLC for the first time.
We adopted DESI as the interface to couple the electrochemical cell with a mass
spectrometer. There are several advantages of using DESI as the interface. First, the
conflict between the electrochemical redox potential and the high voltage for spray
ionization is avoided using DESI as the interface to combine EC and MS, since the DESI
high voltage is separated from the cell potential. Second, the cell and DESI source can be
connected with a very short piece of the capillary as a conduit, thus minimizing the postcolumn dead volume in this LC/EC/MS apparatus. Third, taking advantage of the
freedom to choose different DESI spray solvents, reactive DESI can be directly
performed in the coupling of UPLC/EC/DESI-MS for post-column derivatization. Using
the new apparatus, we demonstrated the application of LC/EC/MS in protein/peptide
structural analysis. The disulfide bond-containing peptides can be differentiated from
those without disulfide bonds because the disulfide bond-containing peptides are
electroactive and electrochemically reducible. Moreover, the tandem MS analysis of the
reduced peptide ions can provide more information for sequencing and pinpoint the
disulfide bond linkages. In reactive DESI-MS experiment using the spray solvent doped
70
with supercharging reagents, online electrolytic reduction of disulfide-bond containing
proteins in combination with supercharging leads to higher protein charges. In this study,
we focused on its application for the structural analysis of disulfide bond-containing
proteins/peptides, taking advantage of the platform’s capability for fast separation, online
electroreduction and MS detection.
4.2 Experimental
4.2.1 Chemicals
Somatostatin 1-14 was purchased from American Peptide Company (Sunnyvale,
CA). TPCK-treated trypsin from bovine pancreas, pepsin from porcine gastric mucosa,
insulin from the bovine pancreas, α-lactalbumin from bovine milk (type III, calcium
depleted), formic acid (FA) and HPLC-grade acetonitrile (ACN) were all purchased from
Sigma-Aldrich (St. Louis, MO). HPLC-grade methanol was obtained from Fisher
Scientific (Fairlawn, NJ) and deionized water used for sample preparation was obtained
using a Nanopure Diamond Barnstead purification system (Barnstead International,
Dubuque, IA).
4.2.2 Apparatus
A Waters ACQUITY UPLC® System with a Waters Xevo QTOF mass
spectrometer (Milford, MA) was used in this experiment. Our UPLC/EC/DESI-MS
assembly employs a thin-layer μ-PrepCellTM electrochemical flow cell. The cell equipped
with a magic diamond (MD) electrode served as the working electrode (WE), a platinum
electrode served as the reference electrode (RE) and a titanium electrode served as an
auxiliary electrode (AE). The cell was connected to the UPLC column using a piece of
71
PEEK connection tubing (i.d.: 200 μm, length: 25 cm). An ACQUITY UPLC BEH C 18
column (1.7 µm, 2.1 mm x 50 mm) was used for peptide separation. An ACQUITY
UPLC Protein BEH C4 column (1.7 µm, 2.1 mm x 50 mm) were employed for protein
separation. The flow rate of the mobile phase was 200-300 μL/min. After UPLC
separation, compounds flowed through the electrochemical cell and underwent
electrochemical reduction. Then the electrolyzed species flowed out of the thin-layer cell
via a short PEEK tube (i.d. 510 μm; wall thickness: 530 μm, length: 3 cm) carrying a
micro-orifice (i.d. 100 μm) that was located in the tube 2 cm downstream the
electrochemical cell. About one-third of the eluent emerged out of the micro-orifice and
then was subject to DESI ionization. Unless specified, the DESI spray solvent was
CH3OH/H2O/HOAc (50:50:1 by volume). The flow rate was 10 μL/min. A 5 kV potential
was applied to the spray solvent with N2 nebulization (160 psi). A RoxyTM potentiostat
was used to apply a potential (pulsed mode, E1= -2.0 V for 1990 ms, E2 = -1.5 V for 1010
ms and E3= 0 V for 20 ms) to the electrochemical flow cell to trigger redox reaction.
4.2.3 LC Separation Condition
Somatostatin 1–14 was trypsin digested in 25 mM ammonium bicarbonate at a
molar ratio of 1:25 (enzyme/protein) for 12 h at 37 °C. The digest products separation
condition were solvent A was 0.1% FA in H2O, and solvent B was 0.1% FA in ACN; 10%
B was ramped to 40% in 3 min. The mobile phase flow rate was 300 μL/min.
Insulin was pepsin digested in water containing 1% acetic acid at a molar ratio of
1:25 (enzyme/protein) for 12 h at 37 °C . The digested insulin was diluted to 20 μM with
water containing 0.1% FA and 6 μL of the sample was injected into the UPLC system
72
using an autosampler. Solvent A was 0.1% FA in H2O, and solvent B was 0.1% FA in
ACN. Peptides were eluted by using a 5 min linear elution from 5% to 7% B, and from 7%
to 15% B in 1 min, then from 15% to 30% B in 10 min. The flow rate of mobile phase
was 300 μL/min.
For the protein mixture separation, 6 μL of a protein mixture of insulin,
myoglobin, and α-lactalbumin was loaded onto the UPLC C4 column. The elution flow
rate is 200 μL/min. The LC gradient program: solvent A was 0.1% FA in H2O, and
solvent B was 0.1% FA in ACN; 28% B was ramped to 32% in 3 min, and then increased
to 45% in 3 min.
4.3 Results and Discussion
4.3.1 Reduction of Somatostatin 1-14
The home-built apparatus for coupling a thin-layer electrochemical flow cell with
UPLC/DESI-MS was used (Figure 4-1a). A disulfide-containing peptide somatostain 114 (MW 1637.9 Da) was first chosen to test the feasibility of the UPLC/EC/DESI-MS
method. After trypsin digestion, Somatostain 1-14 was produced a peptide mixture,
AGCK TFTSC and NFFWK. The digested mixture (6 μL, 20 μM) was loaded onto the
UPLC for separation. As shown in Figure 1b, the retention times of the two peptides are
1.20 and 2.08 min, respectively. Figure 4-1 c-d shows the DESI-MS spectra of the
peptide AGCK TFTSC. When the cell was off, m/z 933 and m/z 467 were observed
corresponding to the singly and doubly charged peptide AGCK TFTSC ion, respectively.
Tandem MS/MS was performed to elucidate the ion structures. Upon collision induced
dissociation (CID), the fragment ions of the singly charged AGCK TFTSC of m/z 933 are
73
limited, only producing B(b2), A/B(y1) (A/B(y1) refers to a fragment with y1 ion of B
chain linked with an intact A chain; the notation also is applicable to other fragment ions),
A/B(y2), A/B(y3), and B/A(y2) (Figure 4-2a). Once the reduction potential was applied,
two new ions of m/z 378 and m/z 558 were observed, resulting from the reduction of the
AGCK TFTSC. Moreover, CID of the electro-generated ions provides more fragment
ions that cover all of the backbone cleavage sites. CID of the ion at m/z 378 gives rise to
b3, b3-H2O, y2 and y3 fragment ions and CID of the ion at m/z 558 yields b2, b3, b4, b4-H2O,
y3 y3-H2O and y4-H2O fragment ions. According to the CID fragment ions, the two
peptide sequences are assigned as AGCK and TFTSC (Figure 4-2 b-c). In addition, the
sum of the MWs of the two products (378.2 + 558.2 – 2.0 = 934.4 Da) is higher than that
of the precursor peptide (932.4 Da) by 2.0 Da, suggesting that the precursor peptide
AGCK TFTSC has one disulfide bond. The results show that the electrochemical
reduction removes the disulfide bond linkage and produce linear peptides. In this case,
more sequences information can be identified by MS/MS.
74
b)
a)
AGCK
1.20
100
TFTSC
0
0.50
1.00
1.50
0
0.50
AGCK
c) Cell off
%
[
1.00
[
+2H]2+
TFTSC
467.2
AGCK
[
TFTSC
489.6
2.50
3.00
2.08 NFFWK
100
100
2.00
1.50
AGCK
TFTSC
2.00
+H]+
933.4
[
+2Na]2+
Time
3.00
2.50
AGCK
TFTSC
955.4
+Na]+
0
100
d) Cell on
[
+2H]2+
TFTSC
[AGCK+H]+ 467.2
378.2
%
0
AGCK
[
300
350
400
450
AGCK
TFTSC
+H]+
AGCK
[
+Na]+
TFTSC
955.4
[TFTSC+H]+
558.2 [TFTSC+Na]+
580.2
500
550
600
650
933.4
700
750
800
850
900
950
1000
m/z
Figure 4-1. a) The apparatus of UPLC/EC/DESI-MS. b) Extracted ion chromatograms
(EICs) showing the UPLC separation of b) AGCK TFTSC (upper panel) and NFFWK
(lower panel). DESI-MS spectra of AGCK TFTSC c) when the cell was off and d) when
the cell was on.
Furthermore, the result shown above is also helpful for pinpointing the disulfide
bond linkage between the 3rd residue of one chain AGCK with the 5th residue of the other
chain TFTSC. However, the protonated peptide NFFWK containing no disulfide bonds
75
remained unchanged with and without potential applied. This result suggests that
UPLC/EC/DESI could differentiate disulfide bond-containing peptides from others.
100
a)
m/z 933
AGCK
Relative Abundance
0
100
[
TFTSC
B(b2)
250
b)
b2
300
+H]+
350
400
450
500
A/B(y3)
A/B(y2)
A/B(y1)
550
600
650
700
B/A(y2)
750
800
850
900
950
m/z 558
[ T F T S C +H]+
y3
y3-H2O
b4
b4-H2O y4-H2O
b3
-H2O
0 230 250 270 290 310 330 350 370 390 410 430 450 470 490 510 530 550 570
100
y2
c)
m/z 378
b3-H2O
0
200
220
b3
y3
240
260
280
300
[ A G C K +H]+
320
340
360
380
400
m/z
Figure 4-2. CID MS/MS spectra of a) [AGCK TFTSC+H]+ (m/z 933), b) reduced
peptide [AGCK+H]+ (m/z 378) and c) reduced peptide [TFTSC+H]+ (m/z 558).
4.3.2 Reduction of Insulin
After the successful trial with somatostain 1-14 digest, we further tested the
method for the analysis of a protein pepsin digest. The bovine pancreatic insulin was
76
chosen as the test sample. It is known that Insulin is composed of A chain and B chain
linked by two inter-peptide disulfide bonds; and the A-chain has an additional intrapeptide disulfide bond. As shown in the acquired EICs (Figure 3-2), the peptides were
well separated within 15 min. After separation, The MS spectra (Figure 4-4) obtained in
the positive ion show the ions of the first six eluted peptides, including [YTPKA+H]+
(m/z 579), [FVNQ+H]+ (m/z 507), [GIVE+H]+ (m/z 417), [YQLEN+H]+ (m/z 666),
[YQLE+H]+ (m/z 552), and [VEAL+H]+ (m/z 431). When the reduction potential was
applied, there were no new peaks observed (), which indicates that these are peptides
without disulfide bonds. CID MS/MS analysis was applied to gain further structural
information of these peptide ions except GIVE in which the first and second peptide bond
cleavages were missing.
77
%
100
0
YTPKA
2.00
%
2.00
8.00
10.00
12.00
14.00
4.00
6.00
8.00
10.00
12.00
14.00
6.00
8.00
10.00
12.00
14.00
8.00
10.00
12.00
14.00
8.00
10.00
12.00
14.00
10.00
12.00
14.00
12.00
14.00
10.00
12.00
14.00
10.00
12.00
14.00
GIVE
%
100
0
6.00
FVNQ
100
0
4.00
2.00
4.00
YQLEN
%
0
2.00
4.00
6.00
YQLE
%
100
0
2.00
4.00
6.00
100
2.00
4.00
6.00
8.00
100
0
P1:
QCCASVCSL
FVNQHLCGSHL
2.00
4.00
6.00
8.00
2.00
4.00
6.00
8.00
100
P2:
10.00
NYCN
LVCGERGFF
%
0
P3:
100
GIVEQCCASVCSL
HLCGSHL
%
0
2.00
4.00
6.00
8.00
100
P4:
GIVEQCCASVCSL
FVNQHLCGSHL
%
0
Time
VEAL
%
0
%
Relative Abundance
100
2.00
4.00
6.00
Figure 4-3. EICs of the pepsin-digested insulin.
8.00
10.00
12.00
14.00
Time
78
[FVNQ + H]+
507
c) Cell off
[FVNQ + Na]+
529
%
[YTPKA + Na]+
601
0
0
[YTPKA + H]+
579
100
200
250
300
350
400
450
500
550
600
650
%
200
250
300
350
400
[GIVE + Na]+
439
450
500
550
600
m/z
[YQLEN + H]+
666
g) Cell off
100
[YQLEN + Na]+
688
0
0
[GIVE + H]+
417
GIVE+ Na]+
439
300
340
360
380
400
420
440
460
480
500
450
500
100
[YQLE + Na]+
574
100
350
400
450
500
550
600
650
700
700
750
800
m/z
[VEAL + H]+
431
l) Cell on
[VEAL + Na]+
453
%
%
[YQLE + Na]+
300
650
[VEAL + Na]+
453
574
0
600
431
k) Cell off
0
[YQLE + H]+
552
j) Cell on
550
[VEAL + H]+
0
100
[YQLEN + Na]+
688
m/z 0
400
[YQLE + H]+
552
i) Cell off
%
100
320
h) Cell on
100
%
%
f) Cell on
[YQLEN + H]+
666
%
100
0
150
%
e) Cell off
[FVNQ + Na]+
529
0
100
m/z
700
[GIVE + H]+
417
d) Cell on
%
[YTPKA + Na]+
601
150
[FVNQ+ H]+
507
100
b) Cell on
100
Relative Abundance
100
%
a) Cell off
0
100
Relative Abundance
[YTPKA + H]+
579
%
Relative Abundance
100
m/z
0
400
410
420
430
440
450
460
470
480
490
500
m/z
Figure 4-4. DESI-MS spectra of the digested insulin peptides carrying no disulfide bonds
YTPKA: a) cell off and b) cell on; FVNQ: c) cell off and d) cell on; GIVE: e) cell off and
f) cell on; YQLEN: g) cell off and h) cell on; YQLE: i) cell off and j) cell on; and VEAL:
k) cell off and l) cell on.
For the additional four peptides that eluted later (denoted as P1, P2, P3, and P4,
Figure 4-3), new peaks were observed once the potential was applied to the cell. This
79
indicates that the peptides carried disulfide bonds. Since the UPLC separation was used
before electroreduction, these new peptide ions would correspond to the reduced peptides
of the UPLC-separated precursor peptides. Importantly, CID MS/MS analysis of the
reduced peptides provides useful information for the sequencing and disulfide bond
mapping of the precursor peptide. As shown in Figure 4-5, in the case of P3, the sum of
the MWs of the two reduced products GIVEQCCASVCSL and HLCGSHL (2075.9 Da,
calculated from the measured m/z of the corresponding product ions) is higher than that
of P3 (2071.9 Da) by 4.0 Da, which indicates that the precursor peptide P3 has two
disulfide bonds. The singly charged ion of chain B generated from electrolysis (m/z 766)
gave rise to fragment ions b2, b3, b4, b5, b6, y2, y3, y4, y5, and y6 upon CID (Figure 4-5e).
This set of fragment ions covers all of the backbone cleavages and gives the sequence of
the peptide as HLCGSHL with one cysteine residue located at the 3rd amino acid site.
Likewise, CID spectrum of the single charged ion of chain A from P3 reduction (m/z
1311) yielded b3 b4, b5, b6, b6-H2O, b7, b8, b9, b10, b10-H2O, b11, and b12 (Figure 4-5d). The
chain A determines the most of the sequence for XXXEQCCASVCSL (X means the first
three amino acids are unknown). All three cysteine residues of chain A are known to be
located at the 6th, 7th, and 11th amino acid sites. Thus, there are three possible disulfide
bond linkages for P3 (the chain B sole cysteine residue links with one of the three
cysteine residues of chain A). Upon CID, the doubly charged P3 ion (m/z 1037, Figure 45c) dissociated into A(b2), A(b3), A(b4) B/A(b11)2+, B/A(b12)2+, B/A(y8), B/A(y10), B(b2),
B(y2), and B(y4), in which the backbone cleavage between Cys6 and Cys11 of the chain is
missing. This result reveals that the Cys6 in the chain A is paired up with Cys11 of the
80
chain A to form an intra-peptide bond and Cys7 of chain A links with Cys3 of chain B in
P3. The appearance of fragment ions A(b2) and A(b3) in Figure 4-5c shows the 3rd residue
of A is valine. Thus, chain A can be determined as XXVEQCCASVCSL. These results
show that both the locations of disulfide bonds and most of the P3 sequence can be
determined using the information acquired from this UPLC/EC/DESI-MS method.
81
1037
100
100
a) P3 Cell off
[GIVEQCCASVCSL + 3H]3+
HLCGSHL
%
Relative Abundance
[GIVEQCCASVCSL + 2H]2+
HLCGSHL
692
00
1037
100
100
[GIVEQCCASVCSL + 2H]2+
HLCGSHL
b) P3 Cell on
%
[GIVEQCCASVCSL + 3H]3+
HLCGSHL
766 [HLCGSHL+H]+
692
[GIVEQCCASVCSL+H]+
1311
00
700
700
800
800
850
900
900
950
1000
1000
1050
1100
1100
1150
1200
1200
c)
1250
1300
1300
1350
1400
1400
m/z
m/z
×2
x2
m/z 1037
[
GIVEQCCASVCSL
HLCGSHL
+2H]2+
%
Relative Abundance
100
750
B(y2)
B/A(b12)2+
B/A(b11)2+
B(b2)
A(b3)
A(b2)
A(b4) B(y4)
0
200
100
400
600
800
B/A(y10)
B/A(y8)
1000
1200
1400
1600
1800
m/z
m/z
d)
m/z 1311
Relative Abundance
%
[G I V E Q C C A S V C S L +H]+
b3
0
100
200
b4
b8
b6
b5 b6-H2O
400
b7
600
b10-H2O
b9
800
b10 b
11
1000
b12
1200
1400
1600
m/z
m/z
e)
m/z 766
[H L C G S H L +H]+
%
b2
y5
y2
b5
b3
0
100
150
200
250
300
350
y6
y3 b 4
y4
400
b6
450
500
550
600
650
700
750
800
m/z
m/z
Figure 4-5. DESI-MS spectra of P3 with a) cell off and b) cell on, c) CID MS2 spectrum
of doubly charged P3 at m/z 1037, d) CID MS/MS spectrum of singly charged P3 chain A
at m/z 1311, and d) CID MS/MS spectrum of singly charged P3 chain B at m/z 766.
82
Since the low digestion specificity of pepsin, two additional peptides, P1 and P4,
were generated. Their structures (shown in Figure 4-3) are similar to that of P3 and can
be determined by MS/MS analysis. The fragmentation patterns of P1 and P4 ions also
agree with the disulfide bond assignment for P3 (Figure 4-6).
83
100
100
a)
m/z 1082
QCCASVCSL
Relative Abundance
%
[
A(y2)
B(b2)
100
400
400
+2H]2+
B/A(b8) 2+
A/B(y10)2+
B(b )
B(y2) 3
B(y4)
00
200
200
FVNQHLCGSHL
A/B(y6)-NH3
B(b5)B(b )
6
600
600
800
800
1000
1000
1200
1200
A/B(y5)-NH3 A/B(y7)-NH3
A/B(b7)
m/z
m/z
1400
1400
1600
1600
1800
1800
b)
m/z 854
A(b2)
[
GIVEQCCASVCSL
FVNQHLCGSHL
+3H]3+
%
A(y2)
A/B(y9)
A(y1)
A(b3)
B(y4)
orB(y1)
B(b2) A(b4)
B(b3)
A(b5) B(b )
5
0
100
200
300
400
500
600
B/A(y9) B/A(b11)
A/B(y7)
B/A(y10) B/A(y11)
B(b6)
700
2+
800
B/A(y8)
900
1000
1100
1200
3+
1300
m/z
Figure 4-6. CID MS/MS spectra of a) [P1+2H] (m/z 1082) and b) [P4+3H] (m/z 854).
In the case of P2, one of its reduced products (A-chain) was missing in the
detected by UPLC/DESI-MS (Figure 4-7 a-b) It was caused by the lack of the basic
amino acid residues that will decrease the ionization efficiency. Only B-chain
LVCGERGFF was detected. Similar phenomena of missing chain A in the spectra of
reduced insulin has been reported before.88 Even though the chain A was missing, the
84
sequence of P2 and its disulfide bond location still can be inferred based on examination
of MS/MS data and MW analysis. CID spectrum of doubly charged P2 m/z 769 (Figure
4-7c) yields fragment ions B/A(y2), B/A(y2)-NH3, B/A(y3), B/A(b3)2+, A(b2), B(y1), B(y4),
A/B(y7), A/B(y7)2+, and B(b2). CID spectrum of the doubly charged chain B of P2 (m/z 514)
was further examined (Figure 4-7d), and its fragment ions of b2, b3, b6, b7, y1, y4, y5, y6, y7
and y8. In this case, the sequence of the reduced peptide, chain B, was determined to be
LVCGERGFF. Assuming that P2 has one disulfide bond, the MW of chain A is
calculated to be 512.1 Da based on the MWs of P2 and chain B. For sequencing chain A,
the first two amino acids of chain A can be determined as asparagine and tyrosine, which
was confirmed by both the MW of the A-chain and the CID MS/MS analysis of
[P2+2H]2+ (m/z 769) by observing fragment ions B/A(y2) and B/A(y3) (Figure 4-7c). The
appearance of A(b2) and B/A(b3)2+in Figure 4-7c further suggests that the last two amino
acid residues of chain A are cysteine (modified with chain B) and asparagine residues.
Thus the chain A sequence is determined as NYCN. In this case, chain A and B in P2
peptide were connected with one inter-peptide disulfide bond. The Cys3 of chain A links
with Cys3 of chain B in P2, which is in agreement with our assumption above. This result
reveals that the sequence of P2 can also be identified. Note that another peak of m/z 815
was observed with the cell on (Figure 4-7b), which is caused by in-source CID of chain B
ions. Thus, the UPLC/EC/DESI-MS with MS/MS could provide rich information for
insulin sequencing (98% coverage) and disulfide bond location determination (three
disulfide bonds can be located).
85
100
769
a) Cell off
[NYCN+2H]2+
Relative Abundance
%
LVCGERGFF
0
%
815 [CGERGFF+H]+
[NYCN+2H]2+
100
[LVCGERGFF+H]+
1027
769
LVCGERGFF
b) Cell on
[LVCGERGFF+2H]2+
514
0
500
550
600
650
700
750
800
850
900
950
1000
1050
1100
1150
1200
m/z
m/z
100
c)
m/z 769
NYCN
LVCGERGFF
+2H]2+
%
[
0
100
B(b2)
A(b2)
200
300
B/A(b3)2+
B(y4) A/B(y )2+
7
400
500
600
700
800
900
1000
B/A(y2) A/B(y7)
B/A(y2)-NH3
B/A(y3)
m/z
m/z
1100
1200
1300
1400
m/z 514
100
y7
d)
[L V C G E R G F F
+2H]2+
%
Relative Abundance
B(y1)
b2
y1
b3
0
150
200
250
300
y5
y4
350
400
450
500
550
600
b6
650
y6
700
b7
750
y8
800
850
900
950
m/z
m/z
Figure 4-7. DESI-MS spectra of P2 with a) cell off and b) cell on. CID MS/MS spectra of
[P2+2H]2+ (m/z 769) and [LVCGERGFF+2H]2+ (m/z 514).
86
4.3.3 Reduction of Proteins
The UPLC/EC/DESI-MS method is not only for small peptides, but also work for
intact proteins. It can be used for determining the presence of disulfide bonds in proteins.
The higher charging of protein ions can also be obtained using this method via online
electroreduction
combined
with
online
supercharging.
From
previous
work,
electrochemical reduction can remove disulfide bond bridges and then the protein would
unfold with increased charges observed.32, 89-90 Also, supercharging reagents can be used
to increase ion charges.91 It would be interesting that the electroreduction of disulfide
coupled with a supercharging reagent to increase protein charging. So we introduce our
method to supercharge proteins following UPLC separation and electrochemical
reduction. The charge state distributions (CSDs) of protein ions will shift to higher
charges with the appearance of a new population which could indicate a conformational
change after disulfide bond reduction.
In the experiment, we employed a mixture of insulin (containing two inter-peptide
and one intra-peptide disulfide bond), myoglobin (containing no disulfide bonds) and αlactalbumin (containing four inter-peptide disulfide bonds) to test the capability of the
method. The mixture was well separated within 5.5 min (Figure 4-8).
100
insulin
%
100
0
100
0
2.00
4.00
6.00
8.00
10.00
6.00
8.00
10.00
myoglobin
%
100
0
100
0
2.00
4.00
α-lactalbumin
100
%
Relative Abundance
87
0
0
2
2.00
4
4.00
6
6.00
8
8.00
Time
10
Time
10.00
Figure 4-8. EICs of the protein mixture containing insulin, myoglobin and α-lactalbumin.
Comparing the mass spectra recorded of insulin before and after adding the
supercharging reagent m-NBA (50 mM) into the DESI spray, the maximum charge state
shifted from +5 to +6 (Figures 4-9 a-b). Insulin has three disulfide bonds as mentioned
above. Once the reduction potential was applied, two new peaks at m/z 851 and 1134
corresponding to the +4 and +3 charge states of chain B, respectively, could be observed
(Figure 4-9c), suggesting the disulfide bonds existed. The missing of A chain ions in the
spectrum was probably caused by the lack of sufficient numbers of basic amino acid
residues in chain A. As shown in Figure 4-10, for myoglobin, After supercharging
reagent m-NBA (50 mM) doped the DESI spray, the maximum charge state of myoglobin
shifted from +22 to +24, the average charge state increased from +15.1 to +18.9, and the
highest abundance ion was shifted from +18 to +20. But since no disulfide bond was
involved in myoglobin, the mass spectra had no obvious differences with and without cell
on (shown in Figure 4-10 b-c).
88
Insulin
%
100
a) Cell off
without supercharging +5
+4
b) Cell on
with supercharging
+4
+3
100
+5
+3
%
Relative Abundance
0
+6
0
%
100
c) Cell on
with supercharging
+5
Chain B4+
Chain B3+
+4
+3
+6
0
800
900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
m/z
m/z
Figure 4-9. DESI-MS spectra of the UPLC-separated insulin a) when cell was off without
supercharging, b) when cell was off with supercharging and c) when cell was on with
supercharging.
%
Myoglobin
00
+19 +18+17
+20
+16
+15
+21
+22
a) Cell off
without supercharging
+14
+20
+21 +19 +18
+17
+22
+16
+23
+15 +14
+24
+13
+12
+11
100
100
+20
+21 +19 +18
+22
+17
+23
+16 +15
+24
+9
b) Cell off
with supercharging
%
00
+10
+13
100
100
c) Cell on
with supercharging
%
Relative Abundance
100
100
00700
800
800
900
1000
1000
1100
+14
1200
1200
+13
1300
1400
1400
1500
1600
1600
1700
1800
1800
1900
m/z
m/z
Figure 4-10. DESI-MS spectra of the UPLC-separated myoglobin a) when cell was off
without supercharging, b) when cell was off with supercharging and c) when cell was on
with supercharging.
89
For the protein α-lactalbumin, upon adding the supercharging reagent (Figure 411), the maximum charge state of α-lactalbumin shifted from +11 to +14. Moreover,
when the reduction potential was applied, interestingly, a new peak corresponding to +15
ion appeared (Figure 4-11c). Compared the mass spectra of α-lactalbumin before and
after cell on with supercharging reagent (Figure 4-11 b-c), the average charge state
increased from +9.9 to +11.2; and the ion with the highest abundance was shifted from
+8 to +12. α-Lactalbumin has 123 amino acid residues and contains four disulfide crosslinks to maintain and stabilize its structure, so it is likely that the reduction of the
disulfide bonds shall unfold the protein. In this case, unfolded proteins have a greater
capability to carry a larger number of charges on its surface than that of folded proteins.29,
88
This result shows the feasibility of combinng supercharging and disulfide bond
reduction for increasing protein charge states92. Increased charging is useful for
increasing ion dissociation efficiency, especially for proteins constrained by disulfide
bonds.29, 93
α-lactalbumin
100
%
Relative Abundance
100
+8
a) Cell off
without supercharging
+11
0
0
+9
+10
100
%
with supercharging
+14
0
+8
+9
100 b) Cell off
+13
+12
+11
+10
0
100
100
+13
0
+11
+14
%
0
+12
+15
900
1000
1000
1100
1200
1200
1300
c) Cell on
with supercharging
+10
1400
1400
1500
+9
1600
1600
+8
1700
1800
1800
m/z
m/z
Figure 4-11. DESI-MS spectra of the UPLC-separated α-lactalbumin a) when cell was
90
off without supercharging, b) when cell was off with supercharging and c) when cell was
on with supercharging.
4.4 Conclusions
This study suggests a new approach using UPLC/EC/DESI-MS for the structural
elucidation of different types of disulfide-bond containing proteins/peptides. Disulfide
bond-containing peptides in enzymatic digest mixtures can be identified. After UPLC
separation, the linear peptide chains from electrochemical reduction can be easily
assigned to their precursors and sequenced. In addition, upon the CID MS/MS analysis,
sequencing peptides and pinpointing the disulfide linkages are possible. The combination
of electrolysis treatment and online supercharging reactive-DESI could be used to
increase charges for the proteins carrying intra-disulfide bonds. As disulfide bonds play
an important role in protein conformation and function, this UPLC/CE/DESI-MS method
would become a powerful tool in proteomics research.
91
CHAPTER 5: COUPLING ELECTROCHEMISTRY WITH PROBE ELECTROSPRAY
IONIZATION MASS SPECTROMETRY
Adapted from Cai, Y.; Liu, P. Held, M. A.; Dewald, H. D., and Chen, H.,
ChemPhysChem 2016, 17, 1104-1108. Copyright 2016, WILEY-VCH.
5.1 Introduction
EC-MS has become a powerful method for structural identification of
electrochemical reaction products or intermediates.49, 69 Ionization techniques are the key
component to join these two techniques and many ionization methods such as EI,50 TS,51
FAB,52 and ESI53, 94 have been used as the EC/MS coupling interface. Recently, with the
advent of ambient MS, EC/MS coupling has been diversified,49, 69 using techniques such
as DESI,28, 88 nanospray DESI,95 and flowing atmospheric-pressure afterglow (FAPA).96
The direct sampling capability of ambient ionization techniques simplifies the
instrumentation of EC/MS and allows the examination of electrolyzed samples in various
media.34-35,
60, 69, 95
However, although EC/MS techniques have found extensive
applications in proteomics34-35, 88 and drug metabolism simulation,97 no investigation of
the location of the electrolytic redox reaction occurrence in the cell by MS has been
reported to the date. Electrochemical cell MS imaging, telling what electrochemical
reaction takes place and where it occurs, would be of high value in elucidating
electrochemical reaction mechanisms.
PESI is a soft ionization method which operates at atmospheric pressure and
employs a conductive solid probe for a microliter droplet of analyte solution to be
deposited. When a high voltage is applied to the probe, the droplet will be charged and
92
sprayed to produce ions. Early work relevant to PESI was conducted by Shiea and coworkers98-100 and more recently by Hiraoka.36, 40, 101 Compared to ESI or nano-ESI, PESI
has the advantages of higher salt tolerances.37-38 In addition, it is rapid and consumes only
a small fraction of total sample solution (in microliters) for ionization.39 In particular,
PESI-MS is capable of performing chemical imaging analysis of biological tissues,41, 43,
102
providing information about the spatial distribution of analyte of interest. Very
recently, single cell analysis with PESI-MS for the detection of metabolites at cellular
and subcellular levels was reported.44 In addition, PESI was shown to be capable of
directly monitoring solvent-free reactions and organometallic catalysts in roomtemperature ionic liquids (RTILs) in our laboratory.103
In this study, PESI-MS was employed to study electrochemistry. First PESI-MS
was used to monitor directly various electrochemical reactions occurring in RTILs,
including
the
electrochemical
oxidation
of
ferrocene,
N,N’-
bis(salicylidene)ethylenediaminocobalt(II) (Co(II) salen), 2,2,6,6-tetramethylpiperidine
1-oxyl (TEMPO). Second, PESI-MS was introduced to detect electrochemical reactions
on different or multiple electrode surfaces. Furthermore, peptides/proteins separated in an
isoelectric focusing (IEF) cell were also successfully detected by PESI-MS.
5.2 Experimental
5.2.1 Chemicals
Clozapine, 1-butyl-3-methylimidazolium hexafluorophosphate ([BMIM][PF6])
ferrocene, N,N’-bis(salicylidene)ethylenediaminocobalt(II) (Co(II) salen), formic acid
(FA), 2,2,6,6-tetramethylpiperidine 1-oxyl (TEMPO), β-lactoglobulin A from bovine,
93
ubiquitin from bovine erythrocytes, cytochrome c from bovine heart, 3-nitrobenzoic acid
(m-NBA, HPLC) and HPLC-grade acetonitrile (ACN) were all purchased from SigmaAldrich (St. Louis, MO). Angiotensin II (human), angiotensin II (1-4), and angiotensin II
antipeptide were all purchased from American (Sunnyvale, CA). Bio-lyte ampholytes
(pH 3-10) was purchased from Bio-Rad (Hercules, CA). HPLC-grade methanol was
obtained from Fisher Scientific (Fairlawn, NJ) and deionized water used for sample
preparation was obtained using a Nanopure Diamond Barnstead purification system
(Barnstead International, Dubuque, IA).
5.2.2 Apparatus
The mass spectrometer used in this study was either an AB SCIEX Q-trap 2000
triple-quadrupole-linear ion trap mass spectrometer (Concord, Canada) or an LCQ DECA
ion trap mass spectrometer (San Jose, CA). The commercial ion source of the mass
spectrometer was removed to accommodate the PESI ion source. As illustrated in Figure
1a, the PESI probe used was a solid stainless steel needle (Fisher Scientific, San Jose,
CA). Unless specified otherwise, a small amount of sample solution (1-2 μL) from the
electrochemical cell was directly loaded onto the probe tip and then ionized with a 3.5-4
kV high voltage applied to the probe. The sample loading was carried out by inserting the
probe into the specific location of interest in the electrolytic cell for 20s. The distance
between the probe tip and the mass spectrometer inlet varied from 0.5 to 1.0 cm. The ions
generated from PESI were collected and detected using the mass spectrometer mentioned
above. Collision-induced dissociation (CID) spectra were also collected for ion structure
confirmation.
94
5.3 Results and Discussion
5.3.1 PESI-MS Detection of Electrochemical Reactions in RTILs
RTILs has become popular the electrolysis media for electrochemistry because no
additional electrolyte is needed. RTILs have a wide potential window and excellent
solubility for either polar or nonpolar compounds. In this case, many classical
electrochemical reactions can be explored in RTILs. However, RTILs have strong ion
suppression effect for MS examination. In addition heated curtain/desolvation gas was
required to reduce the sample viscosity,104 and the contamination are often the major
problem for MS detection.105 PESI-MS has the advantage of high salt tolerance, so the
challenging experiment of detecting electrochemical reactions in RTILs was explored.
To validate the feasibility of this method, the organometallic compound CoII salen,
a catalyst used for dehalogenation reaction, was first chosen as a test sample. 500 μL of 1
mM CoII salen was dissolved in ionic liquid [BMIM][PF6] and added to a petri dish (i.d.
2.5 cm). Two Pt wires (i.d. 0.4 mm) were served both as the working electrodes (WE)
and the counter electrode (CE) and inserted into the solution for electrolysis (Figure 5-1a).
The PESI probe (a stainless steel needle, Fisher Scientific) was dipped in the sample
solution (about ~30 s) and then ionized with a high voltage (+3 kV) in front of the mass
spectrometer. When no voltage was applied to the cell, there is no electrooxidation
product of CoII salen was observed. The ion m/z 423 was [BMIM]2[PF6]+ (Figure 5-1b).
But as shown in Figure 5-1c, when 5 V was applied across the two Pt wires for 30 V and
the sample solution on the Pt WE was sampled and ionized by PESI, the Co II salen ion at
m/z 325 was detected. Upon CID, m/z 325 gave rise to fragment ions m/z 204, m/z 192
95
and m/z 179 by the loss of C7H5NO, C8H7NO, and C9H8NO, respectively (Figure 5-2),
consistent with its assigned structure. When PESI probe was used to replace Pt WE for
experimental simplicity, a similar result was obtained.
96
Figure 5-1. a) Scheme showing the PESI-MS and the electrochemical cell configuration.
PESI-MS spectra of b) CoII salen before electrolysis and c) CoII after electrolysis. The
inset in c) shows the oxidation reaction of CoII salen.
97
Figure 5-2. CID MS/MS spectrum of the electrochemical oxidation product CoIII salen
ion (m/z 325).
In addition, ferrocene, another organometallic compound, was also tested using
the same apparatus. After electrolyzed, the oxidation product ferrocenium cation at m/z
186 was detected by PESI-MS (Figure 5-3b). As a control, before electrolyzed no
oxidation product of ferrocene was observed (Figure 5-3a).
98
a)
b)
Figure 5-3. PESI-MS spectra of a) ferrocene before electrolysis and b) ferrocene after
electrolysis. The inset in c) shows the oxidation reaction of ferrocene.
Besides the electrolysis of organometallic compounds, the electrochemical
oxidation of TEMPO in RTIL was also examined. TEMPO loses on electron and
undergoes a reversible redox reaction to form the catalytically active oxoammonium
species (Figure 5-4b inset). As shown in Figure 5-4a, only m/z 139 and m/z 157 was
observed which corresponding to [BMIM]+ and [BMIM+H2O]+. Without a voltage
applied on the cell, no oxidation product of TEMPO was observed by PESI-MS. When a
5 V potential was applied to the cell, the oxidation product oxoammonium species at m/z
156 was seen in the PESI-MS spectrum (Figure 5-4b). CID of m/z 156 gave a fragment
99
ion of m/z 139 by the loss of NH2OH. These results suggest that PESI-MS has the
feasibility of studying electrochemical reaction using an ionic liquid as a media.
100
Intensity
7.5e4
156
123
m/z 156
-HONH2
3.5e4
0
100
150
m/z
200
Figure 5-4. PESI-MS spectra of a) TEMPO before electrolysis and b) TEMPO after
electrolysis. The inset in c) shows the oxidation reaction of ferrocene. CID MS/MS
spectrum of the electrochemical oxidation product TEMPO ion (m/z 156).
101
5.3.2 PESI-MS Detection of Electrochemical Reaction on Different Electrodes
For conventional EC/MS experiment, both oxidation and reduction products
would be detected together. But using PESI probe, the products can be detected
separately. To demonstrate the possibility, a mixture of dopamine and flutamide was used
to test. The reduction of flutamide and the oxidation of dopamine process were shown in
Figure 5-5a. In this experiment, the mixture (200 μM each) in MeOH/H2O/FA (50:50:1
by volume) with 10 mM KCl was added into the petri-dish cell. Two gold wires were
used as both the WE and CE. A 5 V voltage was applied across the two electrodes. The
PESI probe was dipped in solution on the cathode surface and on the anode surface to
load sample, respectively. Then the PESI probe was ionized with high voltage. The MS
spectra acquired from the electrolyzed solution sample on the cathode and anode are
given in Figures 5-5 b-c. As shown in Figure 5-5 b for the cathode sample, the nitro
group of flutamide undergoes a two e- reduction process with the loss of one molecule of
water to form nitroso intermediate ion observed at m/z 261, a further two e- reduction
generates the hydroxylamine intermediate ion detected at m/z 263 which further
undergoes a two e- reduction process to yield the final amine product ion appearing at
m/z 247. But only the protonated dopamine (m/z 154) along with its in-source
fragmentation ion at m/z 137 was observed and no oxidation product was detected. In
contrast, as shown in Figure 5-5c, for the anode sample, the oxidized dopamine product,
dopamine quinone, was observed at m/z 152. The ion at m/z 123 results from the
CH2=NH loss from m/z 152 due to in-source dissociation. But only the protonated
flutamide ion at m/z 277 was observed without any reduction products ion.
102
Figure 5-5. a) Equations showing the electrochemical reduction of flutamide and the
electrochemical oxidation of dopamine; EC/PESI-MS spectra of the electrolyzed mixture
sampled from b) the anode surface and c) the cathode surface.
103
As a comparison, the EC/MS experiment was also run in the traditional way. A
commercial thin-layer flow cell was used as the electrochemical cell, and desorption
electrospray ionization (DESI) were used as the ionization method for EC/MS coupling.
The commercial flow cell employed a gold (Au) electrode as the WE (i.d. 8 mm) and a
HyREF electrode as the RF and a titanium electrode the CE. A mixture of dopamine and
flutamide (100 μM each) in MeOH: H2O: FA (50:50:0.1 by volume) was infused into the
cell. The flow rate was 5 μL/min. A -1.5 V was applied to the cell using a Roxy™
potentiostat. The electrochemically oxidized and reduced samples flowed out of the cell
via a short piece of silica capillary and interacted with the charged microdroplets from the
DESI spray for ionization. The spray solvent for DESI was MeOH : H2O : FA (50:50:1,
by volume). The flow rate for DESI spray solvent was 5 μL/ min, and a high voltage + 5
kV was applied to the spray probe. When the mixture sample flowed out of the cell after
electrolysis and was ionized by MS, the acquired DESI-MS spectrum is shown in Figure
5-6. Both the reduced flutamide products and the oxidized dopamine product were
detected simultaneously.
104
Figure 5-6. Traditional EC/MS spectrum of the mixture after electrolysis.
In addition to studying two electrodes, a bipolar electrode (BPE) was employed.
A typical BPE cell involves two driving electrodes and a BPE that is placed between the
driving electrodes. When sufficient potential is applied across the driving electrodes, the
potential differences between the two ends of the BPE built up. In this case, the
electrochemical reaction can be driven. In previous studies, the color indicators or
fluorescent tracer are needed to monitor the occurrence of the electrochemical redox
reaction in BPE cell. But in this case, no chemical structure information can be obtained.
PESI-MS can be a good method to detect the electrochemical reactions in BPE cells
without adding visualization reagents.
Two Pt wires were inserted into the dish serving as two driving electrodes. A Ushaped Pt wire serving as the BPE was placed between the driving electrodes (Figure 57a). 0.5 mM clozapine (CLZ) was chosen as the demonstrated sample. ACN/H2O (30:70,
by volume) containing 20 mM ammonium formate was used as an electrolyte. The CLZ
tends to lose two electrons to form the CLZ cation (Figure 5-7b). So after 30 V was
105
applied to the two driving electrodes, four positions in the cell were sampled and
analyzed by PESI-MS. The acquired spectra are shown in Figure 5-7 c-e, the oxidized
product CLZ cation at m/z 325 was observed in both position 1 and 3 which
corresponding to BPE anode and driving electrode anode. CID of m/z 325 gives a
fragment ion of m/z 277 by the loss of C5H10N2, which is consistent with its structure.
Only protonated unoxidized CLZ (m/z 327) was observed on position 2 and position 4
(Figures 5-7 d-f). No oxidation product was detected. These results suggest that PESI-MS
can be used to detect the electrochemical reaction on different/multiple electrodes.
106
Figure 5-7. a) Scheme showing the BPE cell configuration; b) the oxidation process of
CLZ. PESI-MS spectra of the electrolyzed CLZ solution sampled from four different
electrode surfaces: c) position 1 (driving electrode anode), d) position 2 (BPE cathode), e)
position 3 (BPE anode), and f) position 4 (driving electrode cathode).
107
5.3.3 PESI-MS Detection of Separated Proteins/Peptides in IEF
PESI-MS are not only for detecting small molecular, but also for protein/peptide
analysis. PESI-MS was further employed to analyze IEF separated protein/peptide. IEF
typically uses
the
migration
of
ampholyte
buffer,
a
mixture
of
aliphatic
polyaminopolycarboxylic acids, to establish a pH gradient in the presence of an applied
electric field. It is a technique for separating different molecules (e.g., proteins or
peptides) by differences in their isoelectric points (pIs). A charged analyte starts to
migrate in an IEF cell (configuration is shown in Figure 5-8a) and stops migration when
it reaches the region where the pH is equal to its pI. For IEF experiments, the fractionated
samples are usually examined by LC/ESI-MS and MALDI-MS. However, further timeconsuming preparation is needed as the high concentration of ampholytes suppress the
ion signal. In this study, PESI-MS was attempted to be used for the detection in IEF
experiments, in the consideration of the fact that IEF cell is analogous to the
electrochemical cell and the high salt tolerance of PESI-MS.
Mini Rotofor® cell with 18 mL sample volume (Bio-Rad, Hercules, CA) was
used for liquid-phase IEF to separate a mixture of proteins and peptides. Bio-Lyte
ampholyte (pH 3-10, 40% solution, Bio-Rad) was diluted to 1.5% (for peptides) or 2.0%
w/v (for proteins) by water as the buffer solution. A 700-800 V voltage was applied
across the cell to drive the separation. The total power was remained at 12 W for 2 hours
during the separation. After isoelectric focusing, 20 fractions were collected through
tubings into 20 vials enclosed in the harvesting box driven by a vacuum source. Finally,
the PESI probe was dipped into the sample for 20s and then taken out for detection.
108
In our experiment, a peptide mixture of angiotensin II (pI 6.7), angiotensin II (1-4)
(pI 5.8), and angiotensin II antipeptide (pI 5.2, 0.25 mg/mL each) was separated. After
separation, twenty fractions were numbered and collected and then subjected to PESI-MS
analysis. A drop of the solvent of MeOH: H2O: FA (50: 50: 1 by volume) was added to
the solution adsorbed on the PESI probe tip for dilution and then a 4 kV voltage was
applied to the probe for ionization. Good ion signal was obtained (it turns out that the
dilution helps to obtain stable signal probably due to the reduced viscosity of the sample).
Figures 5-8 c-e show PESI-MS spectra of the fractionated angiotensin II (fraction #12),
angiotensin II (1-4, fraction #9) and angiotensin II antipeptide (fraction #6), respectively.
The ions of all the three separated peptides were clearly seen. In comparison to the
signals of ions detected by PESI-MS before IEF (Figure 5-8b), the intensity of the
separated peptide ions increased by 5-20 folds, showing the enrichment effect of
isoelectric focusing.
109
Figure 5-8. a) Scheme illustrating the process of the IEF/PESI-MS; b) PESI-MS
spectrum of the mixture sample in the IEF buffer before electrofocusing; PESI-MS
spectra of c) angiotensin II (fraction #12), d) angiotensin II (1-4, fraction #9), and e)
angiotensin II antipeptide (fraction #6).
110
Besides peptides, proteins were also tested in this experiment. A mixture of βlactoglobulin A (pI 9.6) ubiquitin (pI 6.8) and cytochrome c (pI 5.1, 0.3 mg/mL each) in
Bio-Lyte ampholyte (pH 3-10, 2.0% w/v, Bio-Rad) aqueous solution was separated in the
IEF cell. After separation, twenty fractions were collected. The PESI probe was dipped
into each fractionated sample solution for 20 s. After adding a drop of MeOH: H2O: FA
(50: 50: 1 by volume) containing 1% m-NBA (a compound found to help obtain protein
ion signal from ion suppressing matrices) to the probe tip for sample dilution and
applying a 4 kV to the probe, protein samples were successfully ionization. As shown in
Figures 5-9 b-d, cytochrome c (fraction #20), ubiquitin (fraction #12) and β-lactoglobulin
A (fraction #6) were clearly detected by PESI-MS. In contrast, no protein signal was seen
from PESI of the mixture sample before isoelectric focusing (Figure 5-9a). Apparently,
these results demonstrate that PESI-MS can serve as a detection technique for IEF
experiments, which would have extensive applications.
111
Figure 5-9. a) PESI-MS spectrum of the mixed protein sample in the IEF buffer prior to
isoelectric focusing; PESI-MS spectra of b) cytochrome c (fraction #20), c) ubiquitin
(fraction #12) and d) β-lactoglobulin A (fraction #6).
5.4 Conclusions
This study presents the development of EC/PESI-MS along with its new
application for electrochemical cell. Benefitting from its high salt tolerance, PESI-MS
could be used to investigate electrochemical reactions taking place in RTILs, a
challenging problem in the past. PESI-MS detection of electrochemical reactions on
112
different electrodes also provides spatial information for the reaction occurrence and the
structural information of the reaction products, which was not achieved by using other
ionization methods. Equally importantly, PESI-MS provides an easy and fast way to
detect the fractionated samples from isoelectric focusing. It can be seen that PESI-MS has
a great potential in the study of electrochemistry.
113
CHAPTER 6: SUMMARY AND FUTURE WORK
Base on the original liquid sample DESI work done in our laboratories, further
modification with the sample introduction part has been carried out, resulting in the
development of the new DESI orifice for LC coupling. In this case, three advantages over
traditional LC/MS could be achieved: 1) The separation can be operated in high flow rate
without causing ion source flooding; 2) Reactive DESI can be achieved for online
derivatization. 3) The continuous MS response could be utilized for further coupling of
LC/MS with EC.
To validate the new orifice of liquid sample DESI, a wide range of samples were
tested from small organic molecules, peptides, protein digests and proteins from which
high quality and reproducible mass spectra were obtained. Versatile application for
coupling with EC has been explored using the LC/DESI-MS interface carrying the new
capillary orifice. For disulfide bond-containing peptides, fragmentation was greatly
enhanced following the reduction. After separation, the reduced linear peptide chains can
be assigned to their precursors and used for sequencing based on ion dissociation. Both
electrolysis treatment and online supercharging reactive-DESI experiments could be used
to increase charges for the proteins carrying intra-disulfide bonds.
For the future works related LC/DESI-MS and EC/LC/DESI-MS, by applying
specific chemistry, the protein reaction could be carried out, such as using selenium
reagent. Then together with LC separation, and online EC reduction, protein or
protein/protein complex 3D structure could be obtained.
114
In the research of PESI, PESI was used as a new interface for coupling EC/MS for
its high salt tolerance. More electrochemical reactions employ a high concentration of
salts as electrolyte could be explored. PESI-MS provides an easy and fast way to detect
fractionated samples from isoelectric focusing. In this case, for further application, real
biological samples can be separated by IEF and analyzed using PESI-MS.
115
REFERENCES
1.
McLafferty, F., Tandem mass spectrometry. Science 1981, 214 (4518), 280-287.
2.
McLafferty, F. W., Tandem mass spectrometry: From infancy to maturity in
twenty-five years. Organic Mass Spectrometry 1993, 28 (12), 1403-1406.
3.
McLafferty, F., Analytical information from mass spectrometry, past and future. J
Am Soc Mass Spectrom 1990, 1 (1), 1-5.
4.
McLafferty, F. W., Tandem mass spectrometric analysis of complex biological
mixtures. International Journal of Mass Spectrometry 2001, 212 (1–3), 81-87.
5.
Aebersold, R.; Mann, M., Mass spectrometry-based proteomics. Nature 2003, 422
(6928), 198-207.
6.
Beck, A.; Sanglier-Cianférani, S.; Van Dorsselaer, A., Biosimilar, Biobetter, and
Next Generation Antibody Characterization by Mass Spectrometry. Analytical Chemistry
2012, 84 (11), 4637-4646.
7.
Bruins,
A.
P.,
Atmospheric-pressure-ionization
mass
spectrometry:
II.
Applications in pharmacy, biochemistry and general chemistry. TrAC Trends in
Analytical Chemistry 1994, 13 (2), 81-90.
8.
Picó, Y.; Blasco, C.; Font, G., Environmental and food applications of LC–
tandem mass spectrometry in pesticide-residue analysis: An overview. Mass
Spectrometry Reviews 2004, 23 (1), 45-85.
9.
Zeleny, J., The Electrical Discharge from Liquid Points, and a Hydrostatic
Method of Measuring the Electric Intensity at Their Surfaces. Physical Review 1914, 3
(2), 69-91.
116
10.
Munson, M. S. B.; Field, F. H., Chemical Ionization Mass Spectrometry. I.
General Introduction. Journal of the American Chemical Society 1966, 88 (12), 26212630.
11.
Carroll, D. I.; Dzidic, I.; Stillwell, R. N.; Haegele, K. D.; Horning, E. C.,
Atmospheric pressure ionization mass spectrometry. Corona discharge ion source for use
in a liquid chromatograph-mass spectrometer-computer analytical system. Analytical
Chemistry 1975, 47 (14), 2369-2373.
12.
Yamashita, M.; Fenn, J. B., Electrospray ion source. Another variation on the
free-jet theme. The Journal of Physical Chemistry 1984, 88 (20), 4451-4459.
13.
Karas, M.; Hillenkamp, F., Laser desorption ionization of proteins with molecular
masses exceeding 10,000 daltons. Analytical Chemistry 1988, 60 (20), 2299-2301.
14.
Ifa, D. R.; Wu, C.; Ouyang, Z.; Cooks, R. G., Desorption electrospray ionization
and other ambient ionization methods: current progress and preview. Analyst 2010, 135
(4), 669-681.
15.
Cooks, R. G.; Ouyang, Z.; Takats, Z.; Wiseman, J. M., Ambient Mass
Spectrometry. Science 2006, 311 (5767), 1566-1570.
16.
Cech, N. B.; Enke, C. G., Practical implications of some recent studies in
electrospray ionization fundamentals. Mass Spectrometry Reviews 2001, 20 (6), 362-387.
17.
Takáts, Z.; Wiseman, J. M.; Gologan, B.; Cooks, R. G., Mass Spectrometry
Sampling Under Ambient Conditions with Desorption Electrospray Ionization. Science
2004, 306 (5695), 471-473.
117
18.
Shin, Y.-S.; Drolet, B.; Mayer, R.; Dolence, K.; Basile, F., Desorption
Electrospray Ionization-Mass Spectrometry of Proteins. Anal. Chem. (Washington, DC, U.
S.) 2007, 79, 3514-3518.
19.
Kauppila, T. J.; Wiseman, J. M.; Ketola, R. A.; Kotiaho, T.; Cooks, R. G.;
Kostiainen, R., Desorption electrospray ionization mass spectrometry for the analysis of
pharmaceuticals and metabolites. Rapid Commun. Mass Spectrom. 2005, 20, 387-392.
20.
Chen, H.; Talaty, N. N.; Takats, Z.; Cooks, R. G., Desorption electrospray
ionization mass spectrometry for high-throughput analysis of pharmaceutical samples in
the ambient environment. Anal. Chem. 2005, 77, 6915-6927.
21.
Rodriguez-Cruz, S. E., Rapid analysis of controlled substances using desorption
electrospray ionization mass spectrometry. Rapid Commun. Mass Spectrom. 2005, 20,
53-60.
22.
Talaty, N.; Takats, Z.; Cooks, R. G., Rapid in situ detection of alkaloids in plant
tissue under ambient conditions using desorption electrospray ionization. Analyst 2005,
130, 1624-1633.
23.
Takats, Z.; Wiseman, J. M.; Cooks, R. G., Ambient mass spectrometry using
desorption electrospray ionization (DESI): Instrumentation, mechanisms and applications
in forensics, chemistry, and biology. J. Mass Spectrom. 2005, 40, 1261-1275.
24.
Miao, Z.; Chen, H., Direct Analysis of Liquid Samples by Desorption
Electrospray Ionization-Mass Spectrometry (DESI-MS). J Am Soc Mass Spectrom 2009,
20 (1), 10-19.
118
25.
Miao, Z.; Wu, S.; Chen, H., The Study of Protein Conformation in Solution Via
Direct Sampling by Desorption Electrospray Ionization Mass Spectrometry. J Am Soc
Mass Spectrom 2010, 21 (10), 1730-1736.
26.
Zhang, Y.; Chen, H., Detection of saccharides by reactive desorption electrospray
ionization (DESI) using modified phenylboronic acids. International Journal of Mass
Spectrometry 2010, 289 (2–3), 98-107.
27.
Pan, N.; Liu, P.; Cui, W.; Tang, B.; Shi, J.; Chen, H., Highly efficient ionization
of phosphopeptides at low pH by desorption electrospray ionization mass spectrometry.
Analyst 2013, 138, 1321-1324.
28.
Li, J.; Dewald, H. D.; Chen, H., Online Coupling of Electrochemical Reactions
with Liquid Sample Desorption Electrospray Ionization-Mass Spectrometry. Anal. Chem.
2009, 81 (23), 9716-9722.
29.
Liu, Y.; Miao, Z.; Lakshmanan, R.; Loo, R. R. O.; Loo, J. A.; Chen, H., Signal
and charge enhancement for protein analysis by liquid chromatography–mass
spectrometry with desorption electrospray ionization. International Journal of Mass
Spectrometry 2012, 325–327, 161-166.
30.
Ma, X.; Zhao, M.; Lin, Z.; Zhang, S.; Yang, C.; Zhang, X., Versatile platform
employing desorption electrospray ionization mass spectrometry for high-throughput
analysis. Anal. Chem. 2008, 80, 6131-6136.
31.
Zhang, Y.; Dewald, H. D.; Chen, H., Online Mass Spectrometric Analysis of
Proteins/Peptides Following Electrolytic Cleavage of Disulfide Bonds. J. Proteome Res.
2011, 10 (3), 1293-1304.
119
32.
Zhang, Y.; Cui, W.; Zhang, H.; Dewald, H. D.; Chen, H., Electrochemistry-
Assisted Top-Down Characterization of Disulfide-Containing Proteins. Analytical
Chemistry 2012, 84 (8), 3838-3842.
33.
Zheng, Q.; Zhang, H.; Chen, H., Integration of online digestion and electrolytic
reduction with mass spectrometry for rapid disulfide-containing protein structural
analysis. International Journal of Mass Spectrometry 2013, 353, 84-92.
34.
Zheng, Q.; Zhang, H.; Tong, L.; Wu, S.; Chen, H., Cross-Linking Electrochemical
Mass Spectrometry for Probing Protein Three-Dimensional Structures. Anal. Chem. 2014,
86 (18), 8983-8991.
35.
Zheng, Q.; Zhang, H.; Wu, S.; Chen, H., Probing Protein 3D Structures and
Conformational Changes Using Electrochemistry-Assisted Isotope Labeling CrossLinking Mass Spectrometry. J Am Soc Mass Spectrom 2016, 27 (5), 864-875.
36.
Hiraoka, K.; Nishidate, K.; Mori, K.; Asakawa, D.; Suzuki, S., Development of
probe electrospray using a solid needle. Rapid Commun. Mass Spectrom. 2007, 21, 31393144.
37.
Mandal, M. K.; Chen, L. C.; Hashimoto, Y.; Yu, Z.; Hiraoka, K., Detection of
biomolecules from solutions with high concentration of salts using probe electrospray
and nano-electrospray ionization mass spectrometry. Analytical Methods 2010, 2 (12),
1905-1912.
38.
Mandal, M. K.; Chen, L. C.; Yu, Z.; Nonami, H.; Erra-Balsells, R.; Hiraoka, K.,
Detection of protein from detergent solutions by probe electrospray ionization mass
spectrometry (PESI-MS). J. Mass Spectrom. 2011, 46, 967-975.
120
39.
Hiraoka, K.; Nishidate, K.; Mori, K.; Asakawa, D.; Suzuki, S., Development of
probe electrospray using a solid needle. Rapid Communications in Mass Spectrometry
2007, 21 (18), 3139-3144.
40.
Chen, L. C.; Yoshimura, K.; Yu, Z.; Iwata, R.; Ito, H.; Suzuki, H.; Mori, K.;
Ariyada, O.; Takeda, S.; Kubota, T.; Hiraoka, K., Ambient imaging mass spectrometry
by electrospray ionization using solid needle as sampling probe. J. Mass Spectrom. 2009,
44, 1469-1477.
41.
Yu, Z.; Chen, L. C.; Erra-Balsells, R.; Nonami, H.; Hiraoka, K., Real-time
reaction monitoring by probe electrospray ionization mass spectrometry. Rapid Commun.
Mass Spectrom. 2010, 24, 1507-1513.
42.
Yu, Z.; Chen, L. C.; Suzuki, H.; Ariyada, O.; Erra-Balsells, R.; Nonami, H.;
Hiraoka, K., Direct profiling of phytochems. in tulip tissues and in vivo monitoring of
change of carbohydrate content in tulip bulbs by probe electrospray ionization mass
spectrometry. J. Am. Soc. Mass Spectrom. 2009, 20, 2304-2311.
43.
Yoshimura, K.; Chen, L. C.; Yu, Z.; Hiraoka, K.; Takeda, S., Real-time analysis
of living animals by electrospray ionization mass spectrometry. Anal. Biochem. 2011,
417, 195-201.
44.
Gong, X.; Zhao, Y.; Cai, S.; Fu, S.; Yang, C.; Zhang, S.; Zhang, X., Single Cell
Analysis with Probe ESI-Mass Spectrometry: Detection of Metabolites at Cellular and
Subcellular Levels. Anal. Chem. 2014, 86, 3809-3816.
121
45.
Volk, K. J.; Yost, R. A.; Brajter-Toth, A., Electrochemistry on line with mass
spectrometry. Insight into biological redox reactions. Anal. Chem. 1992, 64, 21A-26A,
28A, 30A-33A.
46.
Deng, H.; Van Berkel, G. J., A thin-layer electrochemical flow cell coupled online
with electrospray-mass spectrometry for the study of biological redox reactions.
Electroanalysis 1999, 11, 857-865.
47.
Permentier, H. P.; Bruins, A. P.; Bischoff, R., Electrochemistry-mass
spectrometry in drug metabolism and protein research. Mini-Rev. Med. Chem. 2008, 8,
46-56.
48.
Lohmann, W.; Karst, U., Biomimetic modeling of oxidative drug metabolism.
Anal. Bioanal. Chem. 2008, 391, 79-96.
50.
Bruckenstein, S.; Gadde, R. R., Use of a porous electrode for in situ mass
spectrometric determination of volatile electrode reaction products. J. Amer. Chem. Soc.
1971, 93, 793-4.
51.
Hambitzer, G.; Heitbaum, J., Electrochemical thermospray mass spectrometry.
Anal. Chem. 1986, 58, 1067-70.
52.
Bartmess, J. E.; Phillips, L. R., Electrochemically assisted fast atom bombardment
mass spectrometry. Anal. Chem. 1987, 59, 2012-14.
53.
Zhou, F.; Van Berkel, G. J., Electrochemistry Combined Online with Electrospray
Mass Spectrometry. Anal. Chem. 1995, 67, 3643-9.
54.
Blades, A. T.; Ikonomou, M. G.; Kebarle, P., Mechanism of electrospray mass
spectrometry. Electrospray as an electrolysis cell. Anal. Chem. 1991, 63, 2109-14.
122
55.
Vandell, V. E.; Limbach, P. A., Electrospray ionization mass spectrometry of
metalloporphyrins. J. Mass Spectrom. 1998, 33, 212-220.
56.
Dupont, A.; Gisselbrecht, J.-P.; Leize, E.; Wagner, L.; Van Dorsselaer, A.,
Electrospray mass spectrometry of electrochemically ionized molecules: application to
the study of fullerenes. Tetrahedron Lett. 1994, 35, 6083-6.
57.
Castagnoli, N.; Bissel, P.; Jurva, U.; Ashraf-Khorasani, M., Electrochemical mass
spectrometric studies on 1-cyclopropyl-4-phenyl-1,2,3,6-tetrahydropyridine. Rapid
Commun. Mass Spectrom. 2008, 22, 2089-2096.
58.
Faber, H.; Vogel, M.; Karst, U., Electrochemistry/mass spectrometry as a tool in
metabolism studies-A review. Anal. Chim. Acta 2014, 834, 9-21.
59.
Roussel, C.; Rohner, T. C.; Jensen, H.; Girault, H. H., Mechanistic aspects of on-
line electrochemical tagging of free L-cysteine residues during electrospray ionisation for
mass spectrometry in protein analysis. ChemPhysChem 2003, 4, 200-206.
60.
Lu, M.; Liu, Y.; Helmy, R.; Martin, G. E.; Dewald, H. D.; Chen, H., Online
Investigation of Aqueous-Phase Electrochemical Reactions by Desorption Electrospray
Ionization Mass Spectrometry. J. Am. Soc. Mass Spectrom. 2015, 26, 1676-1685.
61.
Brown, T. A.; Chen, H.; Zare, R. N., Detection of the Short-Lived Radical Cation
Intermediate in the Electrooxidation of N,N-Dimethylaniline by Mass Spectrometry.
Angew. Chem., Int. Ed. 2015, 54, 11183-11185.
62.
Brown, T. A.; Chen, H.; Zare, R. N., Identification of Fleeting Electrochemical
Reaction Intermediates Using Desorption Electrospray Ionization Mass Spectrometry. J.
Am. Chem. Soc. 2015, 137, 7274-7277.
123
63.
Zhang, Y.; Yuan, Z.; Dewald, H. D.; Chen, H., Coupling of liquid
chromatography with mass spectrometry by desorption electrospray ionization (DESI).
Chem. Commun. 2011, 47 (14), 4171-4173.
64.
Schiavo, S.; Ebbel, E.; Sharma, S.; Matson, W.; Kristal, B. S.; Hersch, S.; Vouros,
P., Metabolite Identification Using a Nanoelectrospray LC-EC-array-MS Integrated
System. Anal. Chem. 2008, 80, 5912-5923.
65.
Andrews, C. L.; Li, F.; Yang, E.; Yu, C.-P.; Vouros, P., Incorporation of a
nanosplitter interface into an LC-MS-RD system to facilitate drug metabolism studies. J.
Mass Spectrom. 2006, 41, 43-49.
66.
Camenzuli, M.; Goodie, T. A.; Bassanese, D. N.; Francis, P. S.; Barnett, N. W.;
Ritchie, H.; LaDine, J.; Shalliker, R. A.; Conlan, X. A., The use of parallel segmented
outlet flow columns for enhanced mass spectral sensitivity at high chromatographic flow
rates. Rapid Commun. Mass Spectrom. 2012, 26, 943-949.
67.
Andrews, C. L.; Yu, C.-P.; Yang, E.; Vouros, P., Improved liquid
chromatography-mass spectrometry performance in quantitative analysis using a
nanosplitter interface. J. Chromatogr. A 2004, 1053, 151-159.
68.
Ferguson, C. N.; Benchaar, S. A.; Miao, Z.; Loo, J. A.; Chen, H., Direct
Ionization of Large Proteins and Protein Complexes by Desorption Electrospray
Ionization-Mass Spectrometry. Analytical Chemistry 2011, 83 (17), 6468-6473.
69.
Liu, P.; Zheng, Q.; Dewald, H. D.; Zhou, R.; Chen, H., The study of
electrochemistry with ambient mass spectrometry. TrAC, Trends Anal. Chem. 2015, 70,
20-30.
124
70.
Lu, M.; Wolff, C.; Cui, W.; Chen, H., Investigation of some biologically relevant
redox reactions using electrochemical mass spectrometry interfaced by desorption
electrospray ionization. Analytical and Bioanalytical Chemistry 2012, 403 (2), 355-365.
71.
single
Sun, X.; Miao, Z.; Yuan, Z.; Harrington, P. d. B.; Colla, J.; Chen, H., Coupling of
droplet
micro-extraction
with
desorption
electrospray
ionization-mass
spectrometry. Int. J. Mass Spectrom. 2011, 301, 102-108.
72.
Perry, R. H.; Splendore, M.; Chien, A.; Davis, N. K.; Zare, R. N., Detecting
Reaction Intermediates in Liquids on the Millisecond Time Scale Using Desorption
Electrospray Ionization. Angew. Chem., Int. Ed. 2011, 50, 250-254, S250/1-S250/3.
73.
Chen, H.; Cotte-Rodriguez, I.; Cooks, R. G., cis-Diol functional group recognition
by reactive desorption electrospray ionization (DESI). Chem. Commun. 2006, 597-599.
74.
Takats, Z.; Wiseman, J. M.; Gologan, B.; Cooks, R. G., Electrosonic Spray
Ionization. A Gentle Technique for Generating Folded Proteins and Protein Complexes in
the Gas Phase and for Studying Ion-Molecule Reactions at Atmospheric Pressure. Anal.
Chem. 2004, 76, 4050-4058.
75.
Zubarev, R. A.; Kelleher, N. L.; McLafferty, F. W., Electron Capture Dissociation
of Multiply Charged Protein Cations. A Nonergodic Process. J. Am. Chem. Soc. 1998,
120, 3265-3266.
76.
Stutzman, J. R.; McLuckey, S. A., Ion/Ion Reactions of MALDI-Derived Peptide
Ions: Increased Sequence Coverage via Covalent and Electrostatic Modification upon
Charge Inversion. Anal. Chem. 2012, 84, 10679-10685.
125
77.
Streuli, C. A., Titrations in nonaqueous solvents. Anal. Chem. 1964, 36, 363R-
369R.
78.
Sklenarova, H.; Chocholous, P.; Koblova, P.; Zahalka, L.; Satinsky, D.; Matysova,
L.; Solich, P., High-resolution monolithic columns-a new tool for effective and quick
separation. Anal. Bioanal. Chem. 2013, 405, 2255-2263.
79.
Guillarme, D.; Ruta, J.; Rudaz, S.; Veuthey, J.-L., New trends in fast and high-
resolution liquid chromatography: a critical comparison of existing approaches. Anal.
Bioanal. Chem. 2010, 397, 1069-1082.
80.
Guiochon, G., Monolithic columns in high-performance liquid chromatography. J.
Chromatogr. A 2007, 1168, 101-168.
81.
Nguyen, D. T. T.; Guillarme, D.; Rudaz, S.; Veuthey, J.-L., Fast analysis in liquid
chromatography using small particle size and high pressure. J. Sep. Sci. 2006, 29, 18361848.
82.
Cai, Y.; Adams, D.; Chen, H., A New Splitting Method for Both Analytical and
Preparative LC/MS. J. Am. Soc. Mass Spectrom. 2014, 25, 286-292.
83.
Fernandez, P.; Seoane, S.; Vazquez, C.; Bermejo, A. M.; Carro, A. M.; Lorenzo,
R. A., A rapid analytical method based on microwave-assisted extraction for the
determination of drugs of abuse in vitreous humor. Anal. Bioanal. Chem. 2011, 401,
2177-2186.
84.
for
Seiwert, B.; Henneken, H.; Karst, U., Ferrocenoyl piperazide as derivatizing agent
the
analysis
of
isocyanates
and
related
compounds
using
liquid
126
chromatography/electrochemistry/mass spectrometry (LC/EC/MS). J. Am. Soc. Mass
Spectrom. 2004, 15, 1727-1736.
85.
Hayen, H.; Karst, U., Analysis of phenothiazine and its derivatives using
LC/electrochemistry/MS and LC/electrochemistry/fluorescence. Anal. Chem. 2003, 75,
4833-4840.
86.
van Leeuwen, S. M.; Hayen, H.; Karst, U., Liquid chromatography-
electrochemistry-mass spectrometry of polycyclic aromatic hydrocarbons. Anal. Bioanal.
Chem. 2004, 378, 917-925.
87.
Karst, U., Analytical methods: Electrochemistry/mass spectrometry (EC/MS) - a
new tool to study drug metabolism and reaction mechanisms. Angew. Chem., Int. Ed.
2004, 43, 2476-2478.
88.
Zhang, Y.; Dewald, H. D.; Chen, H., Online Mass Spectrometric Analysis of
Proteins/Peptides Following Electrolytic Cleavage of Disulfide Bonds. J. Proteome Res.
2011, 10, 1293-1304.
89.
Kraj, A.; Brouwer, H.-J.; Reinhoud, N.; Chervet, J.-P., A novel electrochemical
method for efficient reduction of disulfide bonds in peptides and proteins prior to MS
detection. Anal. Bioanal. Chem. 2013, 405, 9311-9320.
90.
Nicolardi, S.; Giera, M.; Kooijman, P.; Kraj, A.; Chervet, J.-P.; Deelder, A. M.;
van der Burgt, Y. E. M., On-Line Electrochemical Reduction of Disulfide Bonds:
Improved FTICR-CID and -ETD Coverage of Oxytocin and Hepcidin. J. Am. Soc. Mass
Spectrom. 2013, 24, 1980-1987.
127
91.
Iavarone, A. T.; Williams, E. R., Mechanism of Charging and Supercharging
Molecules in Electrospray Ionization. J. Am. Chem. Soc. 2003, 125, 2319-2327.
92.
Sterling, H. J.; Cassou, C. A.; Trnka, M. J.; Burlingame, A. L.; Krantz, B. A.;
Williams, E. R., The role of conformational flexibility on protein supercharging in native
electrospray ionization. Phys. Chem. Chem. Phys. 2011, 13, 18288-18296.
93.
Zhang, J.; Loo, R. R. O.; Loo, J. A., Increasing fragmentation of disulfide-bonded
proteins for top-down mass spectrometry by supercharging. Int. J. Mass Spectrom. 2015,
377, 546-556.
94.
Bond, A. M.; Colton, R.; D'Agostino, A.; Downard, A. J.; Traeger, J. C., A Role
for Electrospray Mass Spectrometry in Electrochemical Studies. Anal. Chem. 1995, 67,
1691-5.
95.
Liu, P.; Lanekoff, I. T.; Laskin, J.; Dewald, H. D.; Chen, H., Study of
Electrochemical Reactions Using Nanospray Desorption Electrospray Ionization Mass
Spectrometry. Anal. Chem. 2012, 84, 5737-5743.
96.
Smoluch, M.; Mielczarek, P.; Reszke, E.; Hieftje, G. M.; Silberring, J.,
Determination of psychostimulants and their metabolites by electrochemistry linked online to flowing atmospheric pressure afterglow mass spectrometry. Analyst 2014, 139,
4350-4355.
97.
Bischoff, R.; Karst, U., Electrochemistry-mass spectrometry: Fundamentals and
applications in pharmaceutical and environmental sciences. TrAC, Trends Anal. Chem.
2015, 70, 3.
128
98.
Shiea, J.; Huang, M.-Z.; Hsu, H.-J.; Lee, C.-Y.; Yuan, C.-H.; Beech, I.; Sunner, J.,
Electrospray-assisted laser desorption/ionization mass spectrometry for direct ambient
analysis of solids. Rapid Communications in Mass Spectrometry 2005, 19 (24), 37013704.
99.
Jeng, J.; Lin, C.-H.; Shiea, J., Electrospray from Nanostructured Tungsten Oxide
Surfaces with Ultralow Sample Volume. Anal. Chem. 2005, 77, 8170-8173.
100.
Hong, C.-M.; Lee, C.-T.; Lee, Y.-M.; Kuo, C.-P.; Yuan, C.-H.; Shiea, J.,
Generating electrospray from solutions predeposited on a copper wire. Rapid Commun.
Mass Spectrom. 1999, 13, 21-25.
101.
Lee Chuin, C.; Zhan, Y. U.; Hiroshi, N.; Yutaka, H.; Kenzo, H., Application of
Probe Electrospray Ionization for Biological Sample Measurements. Environ. Control
Biol. 2009, 47 (2), 73-86.
102.
Laskin, J.; Heath, B. S.; Roach, P. J.; Cazares, L.; Semmes, O. J., Tissue Imaging
Using Nanospray Desorption Electrospray Ionization Mass Spectrometry. Anal. Chem.
2012, 84, 141-148.
103.
Liu, P.; Forni, A.; Chen, H., Development of solvent-free ambient mass
spectrometry for green chemistry applications. Anal Chem 2014, 86, 4024-32.
104.
Jackson, G. P.; Duckworth, D. C., Electrospray mass spectrometry of undiluted
ionic liquids. Chem. Commun. 2004, (5), 522-523.
105.
Dyson, P. J.; Khalaila, I.; Luettgen, S.; McIndoe, J. S.; Zhao, D., Direct probe
electrospray (and nanospray) ionization mass spectrometry of neat ionic liquids. Chem.
Commun. 2004, (19), 2204-2205.
129
APPENDIX: PUBLICATIONS
1. Yi Cai, Daniel Adams and Hao Chen*, A New Splitting Method for Both Analytical
and Preparative LC/MS, J. Am. Soc. Mass Spectrom. 2014, 25, 286-292.
2. Yi Cai, Yong Liu, Roy Hemly and Hao Chen*, Coupling of Ultrafast LC Separation
with Mass Spectrometry by DESI, J. Am. Soc. Mass Spectrom. 2014, 25, 1820-1823.
3. Yi Cai, Qiuling Zheng, Yong Liu, Roy Helmy, Joseph A. Loo and Hao Chen*,
Integration of Electrochemistry with Ultra Performance Liquid Chromatography/Mass
Spectrometry (UPLC/MS), Eur. J. Mass Spectrom. 2015, 21, 341–351.
4. Yi Cai, Pengyuan Liu, Michael Held, Howard Dewald, Hao Chen*, Coupling
Electrochemistry with Probe Electrospray Ionization Mass Spectrometry (PESI-MS),
ChemPhysChem. 2016, 8, 1104–1108.
5. Si Cheng, Jun Wang,* Yi Cai, Joseph A. Loo,* and Hao Chen*, Enhancing
Performance of Liquid Sample Desorption Electrospray Ionization Mass Spectrometry
Using Trap and Capillary Columns, Int. J. Mass Spectrom. 2015, 392, 73-79.
6. Zhi Li, * Shuaihua Zhang, Yi Cai, Qiuhua Wu, Hao Chen*, Hollow fiber-based solidliquid phase microextraction combined with theta capillary electrospray ionization mass
spectrometry for sensitive and accurate analysis of methamphetamine, Anal. Methods,
submitted.
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
Thesis and Dissertation Services
!