Title Evaluation of column hardware for chelating compounds in liquidchromatography-mass spectrometry( 本文(Fulltext) )

Author(s) 坂牧, 寛

Report No.(DoctoralDegree) 博士(工学) 甲第472号

Issue Date 2015-03-25

Type 博士論文

Version ETD

URL http://hdl.handle.net/20.500.12099/51030

※この資料の著作権は、各資料の著者・学協会・出版社等に帰属します。

i

Evaluation of column hardware for chelating compounds

in liquid chromatography-mass spectrometry

Hiroshi Sakamaki

March 2015

i

Evaluation of column hardware for chelating compounds

in liquid chromatography-mass spectrometry

液体クロマトグラフィー質量分析法におけるキレート化合

物のためのカラム材料の評価

Hiroshi Sakamaki

Material Engineering Division

Graduate School of Engineering

Gifu University

JAPAN

March 2015

i

Evaluation of column hardware for chelating compounds

in liquid chromatography-mass spectrometry

A dissertation submitted to the Gifu University

in partial fulfillment of the requirements for

the degree of Doctor of Philosophy in Material Engineering

By

Hiroshi Sakamaki

March 2015

i

Contents

page

Contents i

List of abbreviations iv

Summary vi

Chapter 1 Introduction 1

1.1 High performance liquid chromatography 1

1.2 Liquid chromatography-mass spectrometry 2

1.3 Kind of column hardware 3

1.4 Chelating compounds on liquid chromatography-mass spectrometry 4

1.5 Carryover in bioanalysis 5

1.6 Objectives of present research 6

1.7 References 7

Chapter 2 Evaluation of column hardware on liquid chromatography-mass

spectrometry of phosphorylated compounds 10

2.1 Introduction 10

2.2 Experimental 12

2.2.1 Reagents and chemicals 12

2.2.2 Sample treatment 15

2.2.3 Chromatographic and mass spectrometric conditions 15

2.2.3.1 FAD 16

2.2.3.2 LPA 16

2.2.3.3 AMPs 16

2.2.3.4 Phosphorylated peptides 17

ii

2.2.4 Combinations of chromatographic tubes and frits for column 17

2.3 Results and discussions 20

2.3.1 Evaluation of the prepared columns and the liquid chromatography- mass spectrometry system with non-phosphorylated compounds

20

2.3.2 Comparison of chromatographic tubes 23

2.3.3 Effect of frits 29

2.4 Conclusions 37

2.5 References 38

Chapter 3 Evaluation of column carryover of phosphorylated peptides and

fumonisins by duplicated solvent gradient method in liquid chromatography-tandem mass spectrometry

40

3.1 Introduction 40

3.2 Experimental 43

3.2.1 Reagents and chemicals 43

3.2.2 Sample treatment 45

3.2.3 Chromatographic and mass spectrometric conditions 45

3.2.4 The column hardware 48

3.2.5 Carryover by duplicated solvent gradient method 48

3.2.6 Calibration curve of phosphorylated peptides in bioanalysis 49

3.2.7 Injection carryover 49

3.3 Results and discussion 50

3.3.1 Evaluation of the prepared columns 50

3.3.2 Improvement of adsorption of phosphorylated compounds in the liquid chromatography-tandem mass spectrometry system

52

3.3.3 Carryover by duplicated solvent gradient method 54

iii

3.3.4 Calibration curve of standard samples in phosphorylated peptides 59

3.3.5 Injection carryover 63

3.4 Conclusions 65

3.5 References 66

Conclusions 68

Figure list 70

Table list 72

List of publications 74

List of presentations 75

Curriculum vitae 77

Acknowledgements 78

iv

List of abbreviations

HPLC; High performance liquid chromatography

LC-MS; Liquid chromatography-mass spectrometry

LC-MS/MS; Liquid chromatography tandem mass spectrometry

ESI; Electrode spray ionization

SIM; Selected-ion monitoring

SRM; Selected-reaction monitoring

C18; Octadecylsilyl-silica gel

ICP-MS; Inductively coupled plasma mass spectrometry

S/N; Signal to noise ratio

TF; Tailing factor

SD; Standard deviation

tR; Retention time

N; Number of theoretical plates

ULOQ; Upper limit of quantitation

LLOQ; Lower limit of quantitation

GL; Glass lined stainless steel

Ti; Titanium

PE; Polyethylene

PEEK; Polyetheretherketone

LPA; Lysophosphatidic acid

T18p; NVPLYpK

T19p; HLADLSpK

v

FB1; Fumonisin B1

FB2; Fumonisin B2

FB3; Fumonisin B3

FAD; Flavin adenosine dinucleotide

AMP; Adenosine monophosphate

dAMP; 5-Deoxyadenosine monophosphate

cAMP; Cyclic denosine monophosphate

USP; United States Pharmacopeia

vi

Summary

In recent years, the biological samples are widely analyzed using liquid

chromatography-mass spectrometry (LC-MS) and liquid chromatography tandem mass

spectrometry (LC-MS/MS). However, some compounds, in particular the analysis of

phosphorylated compounds of peptides and small molecules are interacted with a metal

material such as tubings and valves, and the severe peak tailing are caused. As these

improvements, the metal-free and cleaning and masking of the system have been

performed. In addition, metal-free is achieved by monolith column and fused silica

capillary column that does not use the frit in capillary LC. On the other hand, the

common columns for semi-micro LC-MS(/MS) are not achieved the metal-free. In

addition, the column hardware, such as chromatographic tube and frit are not also

evaluated. In this paper, we evaluated two types of chromatography tube (stainless steel

and glass lined stainless steel; GL) and four type of the frits (stainless steel, titanium; Ti,

polyether ether ketone; PEEK, and polyethylene; PE) by LC-MS analysis of chelated

compounds.

The phosphorylated compounds were analyzed by LC-MS, the

chromatography tubes and the frits were evaluated by the value of peak shape, signal to

noise ratio (S/N) and peak intensity. In comparison in 2 kinds of the tubes, the S/N of

flavin adenosine dinucleotide (FAD) with GL tube and stainless steel frits column

(GL-S column) was 6-times higher than that of stainless steel tube and frits column (S-S

column), and the intensity was 4-times larger for the former than the latter. The intensity

and S/N of lysophosphatidic acid (LPA) were 1.7-times higher for GL-S column than

S-S column and the peak tailing factor (TF) of FAD and LPA with GL-S column were

vii

improved, compared to S-S column. In comparison in 4 kinds of the frits, the intensity

and the S/N with GL and PE frits column (GL-PE column) were higher than those of the

other columns. The intensity and the S/N of FAD for GL-PE column were 19-times and

30-times higher than that for GL-S column, respectively. We suggest that the

mechanisms of retention in our system were affected more by the Fe ions present on the

inner surface of the S-S column. Since the specific surface area of a pair of frits was 70

times larger than that of a chromatographic tube (150 mm x 2.1 mm), the frits were

found to have more effective improvement of the S/N as well as the intensity than the

chromatographic tubes, when phosphorylated compounds were analyzed by LC-MS.

When the phosphorylated compounds were analyzed by LC-MS(/MS) using a GL-PE

column, the intensity and S/N were increased.

The different columns of three materials were evaluated by the carryover of

phosphorylated peptides and fumonisins in LC-MS/MS. In order to eliminate carryover

caused by the injection operation in the autosampler, calculated carryover by the

duplicated solvent gradient method was developed. GL tube and PE frits column

(GL-PE column) showed the most effective improvements in peak shape and the

carryover compared to the other columns. Carryovers of fumonisin B1 (FB1) with

GL-PE column could be reduced to 1/10, and lower limit of quantitation of

phosphorylated peptides as well as the range of the calibration curve was also improved.

If the carryovers of FB1 and NVPLYpK (T18p) were not caused by the GL-PE column,

those carryovers calculated by the duplicated solvent gradient method were

corresponded to those of the flow path from the injection port to the inlet frit of the

column. The carryover values of FB1 with stainless steel tube-stainless steel frits (S-S)

column were 1.67% and those of the flow path were 0.23%. On the other hand, those

viii

carryovers of T18p with S-S column were 0.82% and those of the flow path were 0.49%.

We found that the majority carryover has occurred by S-S column in our system and the

duplicated solvent gradient method.

1

Chapter 1

Introduction

1.1. High performance liquid chromatography

High performance liquid chromatography (HPLC) is today the premier

technique for chemical analysis and related applications, with an ability to separate,

analyze, and/or purify virtually any sample.1 It is one of several analytical methods for

the separation and analysis of the mixture. And it is one of the most useful and widely

applied analytical techniques.

In HPLC the mobile phase is liquid and in contrast to gas chromatography,

HPLC is a suitable technique for the analysis of compounds with a wide range of

polarities, high molecular weights, and thermally unstable or could ionize in solution.

The popularity of HPLC came from its flexibility, both in terms of the diverse nature of

the components that can be separated and the different modes that can be used (reverse

phase, ion exchange, analytical and etc.), and also a reliable and reproducible qualitative

and quantitative analytical method. In the development of analytical methods with

HPLC we must determine many variables to have a good separation.2

A column is the heart of HPLC system. Porous microspheres were used in

column in 1969, the high performance and high pressure of LC began.3-5 Alkyl

bonded-silica gels were developed in 1971, has become the prototype of the modern

column.6 The most widely used packed column in HPLC is utilized octadecylsilyl-silica

gel (C18) with the particle size ranges of 2-5 μm.

2

1.2. Liquid chromatography-mass spectrometry

LC-MS has been achieved in 1970s, and one of the most technical

improvements was the interface, such as moving belt, direct liquid introduction,

continuous-flow or frit-terminated fast atom bombardment, electrode spray ionization

(ESI) or atmospheric-pressure chemical ionization, for removing the mobile phase and

ionizing the analyte. The ESI was appeared as a practical atmospheric pressure

ionization method in 1989,7 it was spread rapidly. The Nobel Prize in Chemistry 2002

was awarded "for the development of methods for identification and structure analyses

of biological macromolecules" with one half jointly to John B. Fenn.

Mass spectrometry is performed using one or two mass analyzers (tandem

mass spectrometry: MS/MS), which consist of ion traps, time of flight or predominantly

of quadrupoles. They can operate in the full scan mode or in the more sensitive

selected-ion monitoring mode (SIM). The MS/MS provides some detection mode. The

selected-reaction monitoring (SRM) is most selectivity and quantitative detection

mode.8 As a powerful quantitative and qualitative instrument, the LC-MS(/MS) is used

in the wide fields of medicine, pharmacy, engineering, agricultural science and etc.

LC-MS(/MS) is an excellent analysis method, however, there is a disadvantage

that non-volatile mobile phase is not available. The phosphate, typical buffer of HPLC,

cannot be used, therefore, some applications could not be used.

3

1.3. Type of column hardware

In HPLC, columns are packed with minute packing materials under high

pressure, and in order to avoid the outflow of the packing materials, filters which are

called “frits” are used at both ends of the column. As for the column hardware materials,

stainless steel, PEEK, glass, fused-silica capillary and etc. are normally used. In

reversed-phase HPLC, stainless steel columns are widely used while fused-silica

capillaries are used for capillary columns with inner diameter smaller than 0.5 mm.9 On

the other hand, PEEK columns are normally used in ion chromatography; however,

PEEK or glass columns have low pressure resistance, they are difficult to be used as

separation columns without further modification. Aluminum sleeve PEEK column

packed with 3-μm C18 particle under a pressure of 60 MPa obtained 130000 plates per

meter.10 A GL capillary column (300 mm x 0.3 mm) packed with 3-μm C18 particle

obtained 40000 theoretical plates.11 For the frit materials, sintered stainless, mesh

stainless, Ti, glass, Teflon, PEEK and etc. are used. However, no frit is needed for

monolithic columns.12 The frits in fused silica capillary for capillary LC and

electrochromatography were formed in silica monolith and sintered stainless steel.13-17

In addition, for the case of “integrated spray columns”, which are normally used in nano

LC, self-assembled particle frits are normally utilized.18

4

1.4. Chelating compounds on liquid chromatography-mass spectrometry

The chelating compounds are organic compounds, and are called chelants,

chelators, chelating agents, or sequestering agents. The chelation involves the formation

or presence of two or more separate coordinate bonds between a polydentate ligand and

a single central atom.19

Due to samples adsorption, the detection of chelating compounds on LC-MS is

normally difficult; this is caused by the adsorption of the silanols groups and metal

ions20. Thus, three methods of improvement could be used; (1) Deactivation of

chromatographic system such as valves, needles and tubes21-25; (2) Addition of chelate

regeants in pretreatment and sample solvent20, 26-28; (3) Using alkaline mobile phase as

in immobilized metal affinity chromatography.26, 29 The improvements with methods (1)

and (2) were due to prevention from interaction between chelated compounds and metal

ions on the surface of flow path. Deactivation of the inner wall of fused silica capillary

could be carried out by treatment with dimethyldichlorosilane to prevent hydrogen

bonding between the silanol groups on the inner wall surface of the fused silica

capillary and phosphorylated compounds, thus the adsorption could be minimized25.

Kim et al. improved peak intensity of phosphorylated peptides by adding phosphoric

acid to sample solutions.20 On the other hand, Wakamatsu et al. improved peak intensity

of nucleic acids by flowing 400 mM phosphoric acid to the chromatographic system

involving stainless steel components as a pretreatment.23 Asakawa et al. used carbonate

mobile phase for nucleic acids, which are phosphorylated compounds, in order to reduce

the adsorption of the target compounds.26 It was concluded that this effect would be due

to the desorption of nucleic acid in alkaline mobile phase and the strong buffer capacity

of carbonate mobile phase.

5

1.5. Carryover in bioanalysis

Carryover is a phenomenon in which the sample from the previous injection is

observed or detected during the analysis of the subsequent injections. Carryovers are

often observed when injecting high concentration samples using LC-MS/MS. 30, 31 One

of the criteria for the carryover in bioanalysis is that the peak area of the analyte in a

blank sample that follows an injection of a highest standard sample (i.e. upper limit of

quantitation, ULOQ) should be less than 20% of the peak area of the lower limit of

quantitation (LLOQ) of the sample. 30, 31 Researchers often spend a lot of time and effort

in order to satisfy with this criterion. 32, 33 Therefore, in order to solve this problem, the

ULOQ is set lower while maintaining the LLOQ.34

The carryover is caused by residual materials as well as the irreversible

adsorption occurred in the flow path of the LC-MS/MS system. They occur in the

column35-37 and autosampler38 mainly. In the case of carryover caused by the

autosampler, a peak of the sample is eluted at the same decided retention time, whereas

a carryover in the column does not elute at the same retention time, and phenomena

such as peak tailing or leading are often observed.35 Carryovers are very often observed

for phosphorylated compounds and also basic compounds, in which the former is

adsorbed on the metallic material in the flow path by interacting with the metal ion. On

the other hand, the latter is adsorbed on the residual silanol in the packing material.

6

1.6. Objectives of present research

In recent years, the biological samples are widely analyzed using LC-MS(/MS).

However, some compounds, in particular the analysis of phosphorylated compounds of

peptides and small molecules are interacted with a metal material such as tubings and

valves, the severe peak tailing are caused. In this paper, in order to solve these problems,

we proposed a column which does not contain the metal in the column hardware, and

evaluated two types of chromatography tube (stainless steel and GL) and four type of

the frit (stainless steel, Ti, PEEK, and PE) by LC-MS(/MS) analysis of phosphorylated

compounds.

The objective in chapter 2 is to find the optimum combination of column

hardware for LC-MS analysis of the phosphorylated compound. The phosphorylated

compounds were analyzed by LC-MS, the chromatography tubes and the frits were

evaluated by the value of peak shape, S/N and peak intensity. As comparison of the

chromatographic tube, 2 columns were prepared with a common stainless steel and GL

tube. These frits were made from stainless steel. As comparison of the frit, 4 columns

were prepared with 4 kinds of frits (stainless steel, Ti, PEEK, and PE) and GL tube.

The objective in chapter 3 is to evaluate the carryover effect of column

hardware. The different columns of three materials were evaluated by the carryover of

phosphorylated peptides and fumonisins in LC-MS/MS. In order to eliminate carryover

of injection operation in autosampler, the calculated carryover by the duplicated solvent

gradient method was developed39. It is possible to calculate the carryover of a column

that has never been determined particularly. Calibration curves for the quantification of

phosphorylated peptides are compared, as adapting the criteria of carryover in

bioanalysis.

7

1.7. References

1. L. R. Snyder and J. J. Kirkland, Introduction to modern liquid chromatography, 2nd

ed.,Wiley-Interscience, New York, 1979.

2. J. J. Kirkland, Anal. Chem., 1969, 41, 218.

3. J. J. Kirkland, J. Chromatogr. Sci., 1969, 7, 7.

4. J. J. Kirkland, J. Chromatogr. Sci., 1969, 7, 361.

5. R.F. Venn, Principles and Practice of Bioanalysis, Taylor and Francis, London,

2000.

6. J. J. Kirkland, J. Chromatogr Sci., 1971, 9, 206.

7. J. B. Fenn, M. Mann, C. K. Meng, S. F. Wong, and C. M. Whitehouse, Science,

1989, 246, 64.

8. H. H. Maurer, J. Chromatogr. B, 1998, 713, 3.

9. T. Takeuchi and D. Ishii, J. Chromatogr., 1981, 213, 25.

10. T. Huang and P. T Kissinger, Curr. Sep., 1996, 14, 114.

11. M. Konishi, Y. Mori, and T. Amano, Anal. Chem., 1985, 57, 2235.

12. K. K. Unger, R. Skudas, and M. M. Schulte, J. Chromatogr. A, 2008, 1184, 393.

13. Y. P. Zhang, Y. J. Zhang, W. J. Gong, N. Chen, A. L.Gopalan, and K. P. Lee,

Microchem. J., 2010, 95, 67.

14. J. Park, H. Oh, and I. S. Jeon, J. Chromatogr. A, 2011, 1218, 7895.

15. X. Zhang and S. Huang, J. Chromatogr. A, 2001, 910, 13.

16. M. Kato, M. T. Dulay, B. D. Bennett, J. P. Quirino, and R. N. Zare, J. Chromatogr.

A, 2001, 924, 187.

17. B. Behnke, J. Johansson, S. Zhang, E. Bayer, and S. Nilsson, J. Chromatogr. A,

1998, 818, 257.

8

18. Y. Ishihama, J. Rappsilber, J. S. Andersen, and M. Mann, J. Chromatogr. A, 2002,

979, 233.

19. A. D. McNaught and A. Wilkinson, Compendium of Chemical Terminology, 2nd ed.,

Oxford, Blackwell Science, 1997.

20. J. Kim, D. G. Camp, and R. D. Smith, J. Mass Spectrom., 2004, 39, 208.

21. Behnke, J. Johansson, S. Zhang, E. Bayer, and S. Nilsson, J. Chromatogr. A, 1998,

818, 257.

22. P. C. Sadek, P. W. Carr, L. D. Bowers, and L. C. Haddad, Anal. Biochem., 1985,

144, 128.

23. A. Wakamatsu, K. Morimoto, M. Shimizu, and S. Kudoh, J. Sep. Sci., 2005, 28,

1823.

24. R. Zhao, S. J. Ding, Y. Shen, D. G. Camp II, E. A. Livesay, H. Udseth, and R. D.

Smith, J. Chromatogr. B, 2009, 877, 663.

25. T. De Vijlder, J. Boschmans, E. Witters, and F. Lemière, Inter. J. Mass Spectrom.,

2011, 304, 83.

26. Y. Asakawa, N. Tokida, C. Ozawa, M. Ishiba, O. Tagaya, and N. Asakawa, J.

Chromatogr. A, 2008, 1198, 80.

27. D. Winter, J. Selder, Y. Ziv, Y. Shiloh, and W. D. Lehmann, J. Proteome Res., 2009,

8, 418.

28. D. Siegel, H. Permentier, and R. Bischoff, J. Chromatogr. A, 2013, 1294, 87.

29. R. Tuytten, F. Lemiere, E. Witters, W. Van Dongen, H. Slegers, R. P. Newton, H.

Van Onckelen, and E. L. Esmans, J. Chromatogr. A, 2006, 1104, 209.

30. EMA, Guideline on the validation of bioanalytical methods,

EMA/CHMP/EWP/192217/2009 Rev.1 Corr., 21 July, 2011.

9

31. US FDA, Guideline for Industry; Bioanalytical Method Validation, Draft,

September, 2013.

32. M. Tamura, K. Matsumoto, J. Watanabe, J. Iida, Y. Nagatomi, and N. Mochizuki, J.

Sep. Sci., 2014, 00, 1.

33. T. Gazzotti, E. Zironi, B. Lugoboni, A. Barbarossa, A. Piva, and G. Pagliuca, Food

Chemistry, 2011, 125, 1379.

34. C. Chassaing, J. Luckwell, P. Macrae, K. Saunders, P. Wright, and R. Venn,

Chromatographia, 2000, 53, 122.

35. S. Dolman, S. Eeltink, A. Vaast, and M. Pelzing, J. Chromatogr. B, 2013, 912, 56.

36. R. ter Heine, C. G. Alderden-Los, H. Rosing, M. J. Hillebrand, E. C. van Gorp, A. D.

Huitema, and J. H. Beijnen, Rapid Commun. Mass Spectrom., 2007, 21, 2505.

37. N. C. Hughes, E. Y. Wong, J. Fan, and N. Bajaj, The AAPS Journal, 2007, 9, 353.

38. J. W. Dolan, LC GC North America, 2001, 19, 164.

39. D. H. Vu, R. A. Koster, A. M. A. Wessels, B. Greijdanus, J. W. C. Alffenaar, and D.

R. A. Uges, J. Chromatogr. B, 2013, 917, 1.

10

Chapter 2

Evaluation of column hardware on liquid chromatography-mass spectrometry of

phosphorylated compounds

2.1. Introduction

In HPLC, columns are packed with minute packing materials under high

pressure, and in order to avoid the outflow of the packing materials, filters which are

called “frits” are used at both ends of the column. As for the column hardware materials,

stainless steel, PEEK, glass, fused-silica capillary and etc. are normally used. In

reversed-phase HPLC, stainless steel columns are widely used while fused-silica

capillaries are used for capillary columns with inner diameter smaller than 0.5 mm.1 On

the other hand, PEEK columns are normally used in ion chromatography; however,

PEEK or glass columns have low pressure resistance, they are difficult to be used as

separation columns without further modification. Aluminum sleeve PEEK column

packed with 3-μm C18 particle under a pressure of 60 MPa obtained 130000 plates per

meter.2 A GL capillary column (300 x 0.3 mm) packed with 3-μm C18 particle obtained

40000 theoretical plates.3 For the frit materials, sintered stainless, mesh stainless, Ti,

glass, Teflon, PEEK and etc. are used. However, no frit is needed for monolithic

columns.4 The frits in fused silica capillary for capillary LC and electrochromatography

were formed in silica monolith and sintered stainless steel.5-9 In addition, for the case of

“integrated spray columns”, which are normally used in nano LC, self-assembled

particle frits are normally utilized.10

Due to samples loss, the intensity of phosphorylated compounds on LC-MS is

normally low; this is caused by the adsorption due to the silanols groups and metal

11

ions.11 Thus, three methods of improvement could be used; (1) Deactivation of

chromatographic system such as valves, needles and tubes9, 12-15; (2) Addition of chelate

regeants in pretreatment and sample solvent11, 16-18; (3) Using alkaline mobile phase as

in immobilized metal affinity chromatography.16, 19 The improvements with methods (1)

and (2) were due to prevention from interaction between phosphate compounds and

metal ions on the surface of flow path.

Deactivation of the inner wall of fused silica capillary could be carried out by

treatment with dimethyldichlorosilane to prevent hydrogen bonding between the silanol

groups on the inner wall surface of the fused silica capillary and phosphorylated

compounds, thus the adsorption could be minimized.15

Kim et al. improved peak intensity of phosphorylated peptides by adding phosphoric

acid to sample solutions.11 On the other hand, Wakamatsu et al. improved peak intensity

of nucleic acids by flowing 400 mM phosphoric acid to the chromatographic system

involving stainless steel components as a pretreatment.13 Asakawa et al. used carbonate

mobile phase for nucleic acids, which are phosphorylated compounds, in order to reduce

the adsorption of the target compounds.16 It was concluded that this effect would be due

to the desorption of nucleic acid in alkaline mobile phase and the strong buffer capacity

of carbonate mobile phase.

Our approach was to develop a new column by optimizing the column

hardware for LC-MS analysis of phosphorylated compounds. To the best of our

knowledge, the column hardware was not evaluated on semi-micro flow LC-MS. In this

paper, we evaluated the influence of chromatographic tubes and frits on LC-MS analysis

of some phosphorylated compounds.

12

2.2. Experimental

2.2.1. Reagents and chemicals

Adenosine monophosphate (AMP), 5-deoxyadenosine monophosphate (dAMP)

and cyclic adenosine monophosphate (cAMP) were purchased from Sigma-Aldrich

(Tokyo, Japan). FAD, LPA, formic acid and ammonium acetate were purchased from

Wako Pure Chemical (Osaka, Japan). Phosphorylated peptide standard kit was

purchased from Funakoshi (Tokyo, Japan). Three phosphorylated peptides

(SFVLNPTNIGM-pS-KSSQGHVTK, TRDIYETD-pY-YRK,

DLDVPIPGRFDRRV-pS-VAAE) contained in the kit were used in this paper. Iron

standard solution of 1000 mg/L, phosphoric acid, acetonitrile and methanol were

purchased from Kanto Chemicals (Tokyo, Japan). All other reagents were of analytical

grade and were used without further purification. Water was purified by a Milli-Q

purification system obtained from Millipore (Bedford, MA, USA).

13

O

N

N

N

N

NH2

OH OH

O

P

O

OH

OH

AMP

O

N

N

N

N

NH2

OH

O

P

O

OH

OH

dAMP

O

N

N

N

N

NH2

O

P

O

OHO

cAMP

OH

14

CH3

CH3

O

PO

PO

N

O

NH

O

OH

OHOH

OOH

OH

O

O

OH OHNN

NN NH2

FAD

O

OH

O

P OH

O

OH

O

CH3

LPA

Fig. 2-1 Chemical Structures of AMPs, FAD and LPA.

15

2.2.2. Sample treatment

AMP, dAMP, cAMP and FAD were dissolved separately in water at a

concentration of 100 mg/L as stock solutions. The mixtures composed of AMP, dAMP,

and cAMP were dissolved in water at a concentration of 500 μg/L each as sample

solution. FAD was dissolved in water at a concentration of 10 mg/L as sample solution

while LPA was dissolved in methanol at a concentration of 100 mg/L as a stock solution.

LPA was then diluted to a concentration of 10 mg/L using methanol and used as sample

solution.

3 phosphorylated peptides were dissolved separately in a mixture of 20 mM

phosphoric acid and acetonitrile (1/1) at a concentration of 20 mg/L as stock solutions.

The mixtures composed of phosphorylated peptides at a concentration of 2 mg/L were

dissolved in 0.1% formic acid as sample solution. All stock solutions were stored at

-20 °C, just before analysis, the sample solutions were prepared.

2.2.3. Chromatographic and mass spectrometric conditions

The chromatographic system used was a Waters 2695 Alliance System

(Milford, MA). The stainless steel tubing from autosampler to column and from column

to MS were replaced with PEEK tubing of 400 x 0.13 mm and 600 x 0.13 mm,

respectively. Methanol was used as the needle washing solvent. The sample vials used

were Waters Polypropylene vials for preventing any possible adsorption of

phosphorylated compounds.

The MS used was a Waters ZQ single quadrupole mass spectrometer with ESI

performed in positive and negative mode. Instrument control and data acquisition were

achieved using Waters Masslynx Software. The processing was achieved using Waters

16

Empower Software. MS conditions were optimized by post-column infusion of the

standard solutions.

2.2.3.1. FAD

The chromatographic mobile phase was a mixture of 0.1% formic acid and

acetonitrile (10:90), at a flow rate of 0.3 mL/min. The column temperature was kept at

40 °C throughout the experiment. The injection volume was 5 μL. The MS detection

was carried out by SIM mode at m/z 348 in positive mode. The capillary sprayer voltage

was 4.0 kV and the sample cone voltage was 30 V. The source temperature was 100 °C

and the desolvation temperature was 400 °C. The desolvation and cone gas flow-rates

were set to 400 and 50 L/h, respectively.

2.2.3.2. LPA

The chromatographic mobile phase was a mixture of 5 mM ammonium acetate

and methanol (5:95), at a flow rate of 0.3 mL/min. The injection volume was 1 μL. The

MS detection was carried out by SIM mode at m/z 435.5 in negative mode. The

capillary sprayer voltage was -3.5 kV and the sample cone voltage was -35 V. The

source temperature was 100 °C and the desolvation temperature was 400 °C. The

desolvation and cone gas flow-rates were set to 500 and 50 L/h, respectively.

2.2.3.3. AMPs

The chromatographic mobile phase was a mixture of 0.1% formic acid and

acetonitrile (99:1), at a flow rate of 0.3 mL/min. The injection volume was 5 μL. The

MS detections were carried out by SIM mode at m/z 348 (AMP), 332 (dAMP) and 330

(cAMP) in positive mode. The capillary sprayer voltages were 3.5 kV and the sample

17

cone voltages were 20 V. The source temperature was 100 °C and the desolvation

temperature was 400 °C. The desolvation and cone gas flow-rates were set to 400 and

50 L/h, respectively.

2.2.3.4. Phosphorylated peptides

The chromatographic mobile phase A and B were 0.1% formic acid and 0.1%

formic acid in acetonitrile, respectively. The gradient flow started with 95% A and 5% B,

and linearly reached 50% A and B in 10 min, at a flow rate of 0.3 mL/min. Next, the

washing of chromatographic system including a column was carried out by passing

through 5% A and 95% B for 2 min, at a flow rate of 0.4 mL/min. The injection volume

was 1 μL. In order to minimize carryover of the phosphorylated peptides, a mixture of

50% 20 mM phosphoric acid and 50% acetonitrile was used as the needle washing

solvent.

The MS detections were carried out by SIM mode at m/z 688

(SFVLNPTNIGM-pS-KSSQGHVTK), 568 (TRDIYETD-pY-YRK) and 731

(DLDVPIPGRFDRRV-pS-VAAE) in positive mode. The capillary sprayer voltages

were 3.5 kV. The sample cone voltage for DLDVPIPGRFDRRV-pS-VAAE and

TRDIYETD-pY-YRK were 30 V, and that for the SFVLNPTNIGM-pS-KSSQGHVTK

was 20 V. The source temperature was 100 °C and the desolvation temperature was

400 °C. The desolvation and cone gas flow-rates were set to 400 and 50 l/h,

respectively.

2.2.4. Combinations of chromatographic tubes and frits for column

Chromatographic tubes used were stainless steel tube (150 mm x 2.1 mm

18

I.D.) and GL tube (150 mm x 2.0 mm I.D.) without deactivated treatment. The

sintered frits were made of stainless steel, Ti, PEEK and PE. Table 2-1 shows

combination of the tubes and the frits for the column. All columns were packed with

L-column2 ODS 5 μm particle (Chemicals Evaluation and Research Institute, Tokyo,

Japan) under a pressure of 50 MPa. The numbers of theoretical plates of naphthalene

as a sample were measured for these columns by an optimized semi-micro LC

Shimadzu Prominence 20A series, and we used the columns that produced more than

11000 plates.

Comparisons of S-S and GL-S column mean difference of chromatographic

tubes. FAD, LPA and AMPs were analyzed on LC-MS for the comparisons. In

addition, comparisons of GL-S, GL-Ti, GL-PEEK and GL-PE columns mean

difference of frits. FAD, LPA, AMPs and phosphorylated peptides were analyzed on

LC-MS for the comparisons.

19

Table 2-1 The types of chromatographic tube and frit used in each specific column.

Column Chromatographic tube Frit

S-S column Stainless steel Sintered stainless steel

GL-S column GL Sintered stainless steel

GL-PEEK column GL Sintered PEEK

GL-PE column GL Sintered PE

GL-Ti column GL Sintered Ti

20

2.3. Results and discussions

2.3.1. Evaluation of the prepared columns and the liquid chromatography-mass

spectrometry system with non-phosphorylated compounds

We needed to measure the number of theoretical plates of the five kinds of

columns made from different materials for quality control before evaluating the column

material. These were tested by using an optimized Shimadzu Prominence 20A series

semi-micro LC. Naphthalene was used as a sample for measuring the plates and the TF,

where uracil was used for determining void volume. The reproducibility for 6 columns

of the same material was confirmed, and all columns had good values (Table 2-2). The

plates of S-S column were the highest among all columns. Retention time of uracil for

S-S column was 1.55 minutes, and that for GL-S column was 1.45 minutes. These were

caused by the difference in column volume. Since the retention times in the all GL-

columns were same, we concluded that the volume was same.

The LC-MS system used has large extra column volume for 2.1 mm I.D.

column. Therefore, we evaluated the system by the peak tailing and the number of

theoretical plates of copper oxinate that was a non-phosphorylated compound. The

isocratic mobile phase of acetonitrile / 0.1% formic acid (5/95) was used, at a flow rate

of 0.3 ml/min. As a result, all columns showed 1.01 to 1.11 of TF and 6600 to 8200 of

the plates (Table 2-3). We confirmed that the system did not affect the peak shape and

all columns had similar column performance. The retention time of all GL- columns

was similar and eluted approximately 6% earlier compared with that of the S-S column.

It means that the linear velocity was different. If the materials had same interactions

with the compounds, the retention time had to be shifted 6-7%.

21

Table 2-2 The chromatographic parameters of the prepared 6 columns for quality

control.

to/min tR/min N TF

mean mean SD mean SD mean SD

S-S column 1.55 8.15 0.03 14000 50 1.02 0.01

GL-S column 1.45 7.58 0.03 12300 280 1.05 0.02

GL-Ti column 1.45 7.59 0.03 12000 370 1.12 0.02

GL-PEEK column 1.45 7.56 0.04 12400 410 1.11 0.03

GL-PE column 1.45 7.58 0.03 12900 190 1.11 0.02

The isocratic mobile phase 60% acetonitrile was used for column testing, at a flow

rate of 0.2 mL/min. t0 is retention time of uracil. tR is retention time of naphthalene.

The number of theoretical plates (N) and TF of naphthalene are calculated by United

States Pharmacopeia (USP) method. The values are mean and standard deviation (SD)

of 6 columns.

22

Table 2-3 Comparison of the analysis of copper oxinate (0.4ng) as a

non-phosphorylated compound for prepared 5 kinds of columns.

tR/min) Intensity/cps N TF

mean SD mean SD mean SD mean SD

S-S column 4.09 0.04 1.38E+06 3.95E+04 8200 320 1.01 0.03

GL-S column 3.81 0.05 1.20E+06 4.95E+04 7100 140 1.01 0.02

GL-Ti column 3.84 0.02 1.35E+06 2.76E+04 6600 310 1.11 0.01

GL-PEEK column 3.87 0.04 1.39E+06 2.48E+04 6900 380 1.08 0.01

GL-PE column 3.82 0.04 1.38E+06 2.57E+04 7800 210 1.10 0.01

The isocratic mobile phase of acetonitrile / 0.1% formic acid (5/95) were used, at a

flow rate of 0.3 mL/min. The values are mean and SD of 5 injections.

23

2.3.2. Comparison of chromatographic tubes

Table 2-4 shows the intensity, S/N and TF of FAD, LPA and AMPs analyzed

using GL-S and S-S columns on LC-MS as comparison of chromatographic tubes. The

SIM chromatograms are shown in Fig. 2-2. The S/N of FAD with GL-S column was

6-times higher than that of S-S column, and the intensity was 4-times larger for the

former than the latter. The intensity and S/N of LPA were 1.7-times larger for the former

than the latter and the peak TFs of FAD and LPA with the former were improved,

compared to the latter. Metal ions in the mobile phase, the metal material used and the

silanol groups present on the GL tube and packing material could be considered as the

three reasons of improvements.

First, Kim et al. considered that the silanol groups on C18 stationary phase

caused interaction with phosphate compounds as same as metal ions.11 The silanol

groups were present in minor amounts in the packing materials and the inner surface

of the GL tube. When the 0.1% formic acid as a mobile phase was used for FAD and

AMPs, hydrogen bonding was occurred between the dissociated silanol groups and

the undissociated phosphate groups23. When GL-S column was used for FAD and

AMPs, the peak tailing had to be caused larger than that of S-S column, however,

GL-S column showed good performance clearly. We suggest that the interactions of

metal on the inner surface of the S-S column were greater than those of the silanol and

metal of the inner surface of the GL-S column. Since FAD had 2 phosphate groups,

the improvement effect was larger than that of AMPs. When 5mM ammonium acetate

was used as the mobile phase for LPA, the interactions were suppressed by ion

repulsion caused by between dissociated silanol groups and dissociated phosphate

groups.23

24

Second, the interactions were caused by metal ion eluted from

chromatographic pump, inline frits and column.25 The metal ions in mobile phase of

0.1% formic acid eluted from the LC-MS (including the column) were determined by

inductively coupled plasma mass spectrometry (ICP-MS). The selected metal ions

were Al, Fe, Ti, Ni, Cr and Mo. Fe and Al can cause formation of chelating with

phosphorylated compounds.17, 19, 21, 22 The SUS316 were composed of Fe, Ti, Ni, Cr

and Mo. Fe ions eluted from the S-S column and GL-S column were detected at a

concentration of 1.8 μg/L and 1.3 μg/L, respectively. The concentration of Mo, Al, Cr,

and Ni eluted from those columns were sub-ppb or less.

We investigated whether μg/L level of Fe would affect the chromatographic

separation of LPA or not. We expected to change height and shape of the peak by

mixing Fe3+ and phosphorylated compounds. This test was conducted for a LPA sample

solution that was prepared to include Fe3+ at concentrations of 0, 2, 10, 50 and 200 μg/L.

These peak areas, heights and TF were measured by flow injection-mass spectrometry.

However, because of the matrix effect of NO3-, the distorted peak of LPA could not be

satisfied. Next, LC-MS method using GL-PE column was adopted. The matrix effect

could be avoided by the chromatographic separation. Fig. 2-3 shows a relationship with

the concentration of Fe3+ and peak height and TF of LPA. Fe3+ in the sample solution

was found to have no effect on the peak height and TF at 10 μg/L or less in this testing.

The tailing and the broad peak were detected in chromatogram of LPA sample solution

including 200 μg/L Fe3+. This peak was delayed than the main peak. Fe3+ of 1.8 μg/L

were detected in the mobile phase eluted from the LC system connecting a column, but

these concentrations would not affect the retention behavior. We suggest that chelating

formation of Fe3+ and LPA was performed quickly since samples prepared was so stable

25

from first injection (within 1 minute after prepared). This meant that injected LPA

formed chelating with Fe3+ in the mobile phase.

Third, we considered that the metal ions existing on S-S column inner surface

were richer than GL-S column. The Fe and Al ions caused interaction with the

phosphate groups.17, 19, 21, 22 The S-S column was made from SUS316 which included

50% or more of iron. Glass of GL-S column was made from borosilicate glass which

included 2% aluminum. Since Fe ions on the stainless steel were richer than Al ion on

the glass surface, the S/N and the intensity with GL column were improved. In addition,

the chelating formation between phosphorylated compound and Fe ion was more stable

than Al ion.25

As a result, we suggest that the behaviors of retention in our system were

affected more by the Fe ions present on the inner surface of the S-S column than the

influence of Fe ions contained in the mobile phase and the silanol groups.

26

Table 2-4 Comparison of chromatographic parameter between S-S and GL-S column

with FAD (50 ng), LPA (10 ng) and AMPs (2.5 ng) used as the samples.

(Chromatographic tubes were compared, i.e. between S and GL, while the same S-frit

was used)

tR/min Intensity/cps S/N TF

mean SD mean SD mean SD mean SD

FAD

S-S column

3.69 0.13 4.21E+03 3.27E+02 4 0.3 -* -

GL-S column

3.11 0.02 1.96E+04 2.83E+02 24 3 7.15 0.80

LPA

S-S column

8.23 0.10 5.47E+04 1.06E+04 60 11 2.56 0.18

GL-S column

8.04 0.08 8.90E+04 2.77E+03 104 14 2.21 0.06

cAMP

S-S column

6.92 0.01 1.28E+05 6.06E+03 104 22 1.05 0.03

GL-S column

6.53 0.01 1.50E+05 1.01E+04 137 10 1.04 0.07

dAMP

S-S column

2.82 0.01 5.14E+04 3.05E+03 36 4 2.34 0.32

GL-S column

2.67 0.00 8.60E+04 5.16E+03 83 10 1.62 0.12

AMP

S-S column

2.02 0.02 9.55E+04 2.08E+04 91 29 3.88 0.31

GL-S column 1.91 0.02 1.45E+05 3.21E+04 181 41 2.34 0.06

The values are mean and SD of 5 injections.

* Not be calculated.

27

Time2.00 4.00 6.00 8.00 10.00 12.00

%

0

100

2.00 4.00 6.00 8.00 10.00 12.00

%

0

100

2.00 4.00 6.00 8.00 10.00 12.00STD column LPA 10ppm 1uL 010 SIR of 1 Channel ES-

TIC (LPA)9.87e4

8.10

0.44

SUS LPA 10ppm 1uL 010 SIR of 1 Channel ES- TIC (LPA)

9.87e47.93

0.32

Time2.00 4.00 6.00 8.00 10.00

%

0

100

2.00 4.00 6.00 8.00 10.00

%

0

100

2.00 4.00 6.00 8.00 10.00STDfrit E4506 VB2 010 SIR of 2 Channels ES+

348 (FAD)2.28e4

3.821.190.860.03

3.621.25 2.37 3.06

4.184.64 5.175.79 6.21 7.28 7.75

SUSfrit E4506 VB2 010 SIR of 2 Channels ES+ 348 (FAD)

2.28e43.11

1.11

0.94

0.531.26

2.25

3.18

3.283.493.73

4.15

(a)

(b)

(a)

(b)

Fig. 2-2 SIM chromatograms of FAD (left) and LPA (right) using S-S column (a),

GL-S column (b). A comparison of chromatographic tube is shown.

28

0

50

100

150

200

250

300

2 10 50 200

HeightTF

Hei

ght a

nd T

F(%

of c

ontro

l)

Concentration of Fe ion in LPA sample solution (μg/L)

Fig. 2-3 Effect of concentration of Fe3+ in LPA sample standard solution on the peak

area, peak height and TF of LPA.

Concentration of Fe ion in LPA sample solution /μg L-1

Hei

ght a

nd T

F, %

of c

ontro

l

29

2.3.3. Effect of frits

The specific surface areas of the frits were measured by the Brunauer,

Emmett and Teller method.26 The areas of SUS, Ti, PEEK and PE frits per a column

were 6.4*10-2 m2, 1.1*10-1 m2, 1.2*10-1 m2 and 6.6*10-2 m2, respectively.

The GL-PE column improved intensity, S/N and TF of phosphorylated compounds.

Table 2-5 shows the intensity, the S/N, the retention time and the TF of FAD, LPA and

AMPs analyzed on LC-MS with GL-S, GL-Ti, GL-PEEK and GL-PE columns as

comparison of the frits. The SIM chromatograms of FAD (Right) and LPA (Left) are

shown in Fig. 2-4 (a), (b), (c) and (d), respectively.

The intensity and the S/N for GL-PE column were higher than those for the

other columns. The intensity and the S/N of FAD for GL-PE column were 19-times and

30-times higher than those for GL-S column, respectively. We considered that the

difference were due to the interactions of metal ions in mobile phase and the specificity

of frit materials. The metal ions in mobile phase of 0.1% formic acid eluted from the

LC-MS system were determined by ICP-MS. Fe ions and Al ion were eluted from the

GL-PE column at a concentration of 1.8 μg/L and 0.17 μg/L, respectively . The

concentration of Mo, Cr, and Ni eluted from those columns were sub-ppb or less. Since

the concentration of the metal ions eluted were low and the difference between the

columns were low, we concluded the metal ions contained in the mobile phase could be

excluded. We assumed that reducing the interaction between the phosphorylated

compounds and the residual metal ions in flow path of our system was more effective in

order to improve the S/N and the intensity. We considered that the phosphorylated

compounds were adsorbed onto stainless steel frit of GL-S column as well as the

chromatographic tubes. On the other hand, the metal ions hardly exist on the frits in

30

GL-PEEK column and GL-PE column, however, good performance was clearly

obtained when the GL- PE column was used. We considered that the polarity and the

specific surface area were different between PEEK and PE frits. The latter could not

entrap the FAD, since the difference of polarity between FAD and PE frits is bigger than

that between FAD and PEEK frits. PE has high hydrophobicity, and the powders were

used as packing materials in reversed-phase LC.27 In addition, the specific areas of

PEEK frits were larger than those of PE frits.

The results of the AMPs were similar to those of FAD. However, there was

no difference between the GL-PE column and the GL-PEEK column. High polarity of

AMPs as compared to the FAD caused weak retention. Since the difference in polarity

between the frits was larger, the intensity and S/N could not be improved.

The intensity and S/N of LPA with GL-PE column showed the best

performance among the others, and those were 7-times higher than GL-S column,

respectively. Those with GL-PEEK column were as same as the GL-Ti column, and

they were less than half of the GL-PE column. However, the retention times were

shorter than GL-S columns, and that with GL-Ti column was the shortest in the all.

We considered that the Ti frit had slight or no interactions and other frits had some

interactions. The Ti frit might have a slight chelating interaction or ion repulsion

under neutral mobile phase.28-30 PE and PEEK frit had hydrophobic interactions with

low polar compounds such as LPA. SUS frit had larger chelating interaction.

The GL-PE column improved intensity and S/N of phosphorylated peptides.

Table 2-6 shows the intensity, S/N and the TF of the peptides analyzed on LC-MS

with GL-S, GL-Ti, GL-PEEK and GL-PE columns as comparison of the frits. The

SIM chromatograms of phosphorylated peptides (right: TRDIYETD-pY-YRK, center:

31

SFVLNPTNIGM-pS-KSSQGHVTK, left: DLDVPIPGRFDRRV-pS-VAAE) are

showed in Fig. 2-5. The intensity and S/N of phosphorylated peptides with GL-PE

column were higher than those of the other columns. The symmetric peaks of

phosphorylated compounds were obtained by using the GL-PE column. These results

were the same as the tested phosphorylated compounds.

Since the specific surface area of PE frits (approximately 6.6*10-2 m2 per

column) was ca. 70-fold larger than the chromatographic tube (approximately

9.4*10-4 m2 per column), the improvement for replacement of the frits were more

effective than that of the tube.

As a confirmation of durability, by using the GL-PE column, injection of 500

times or more of riboflavin phosphate was performed. There was no change in the

chromatogram and the chromatographic parameters (not shown data).

32

Table 2-5 Comparison of chromatographic parameters for FAD (50 ng), LPA (10 ng)

and AMPs (2.5 ng). (4 types of frits were compared while the same kind of

GL-chromatographic tube was used)

tR/min Intensity/cps S/N TF

mean SD mean SD mean SD mean SD

FAD

GL-S column

3.11 0.02 1.96E+04 2.83E+02 24 3 7.15 0.80

GL-Ti column

3.04 0.03 4.85E+04 3.47E+03 46 7 6.11 0.43

GL-PEEK column

3.06 0.01 2.96E+05 1.03E+04 551 50 2.76 0.21

GL-PE column

2.97 0.02 3.56E+05 3.81E+03 701 36 2.59 0.15

LPA

GL-S column

8.04 0.08 8.90E+04 2.77E+03 104 14 2.21 0.06

GL-Ti column

3.70 0.02 2.55E+05 5.10E+03 442 22 1.55 0.05

GL-PEEK column

5.06 0.05 2.50E+05 4.76E+03 368 43 2.02 0.07

GL-PE column

4.70 0.00 6.25E+05 1.33E+04 696 88 1.49 0.02

cAMP

GL-S column

6.53 0.01 1.50E+05 1.01E+04 137 10 1.04 0.07

GL-Ti column

6.51 0.03 1.77E+05 1.22E+04 265 41 1.06 0.07

GL-PEEK column

6.48 0.01 1.72E+05 1.00E+04 257 24 0.96 0.04

GL-PE column

6.58 0.03 1.97E+05 5.30E+03 261 22 1.02 0.02

dAMP

GL-S column

2.67 0.00 8.60E+04 5.20E+03 83 10 1.62 0.12

GL-Ti column

2.66 0.00 1.86E+05 5.30E+03 242 23 1.13 0.05

GL-PEEK column

2.67 0.01 1.90E+05 5.80E+03 292 51 1.05 0.02

33

GL-PE column

2.68 0.01 2.15E+05 4.00E+03 331 72 1.05 0.03

AMP

GL-S column

1.91 0.02 1.45E+05 3.20E+04 181 41 2.34 0.06

GL-Ti column

1.90 0.00 3.45E+05 1.70E+04 686 30 -* -

GL-PEEK column

1.92 0.00 4.07E+05 1.10E+04 792 100 1.57 0.01

GL-PE column 1.91 0.01 3.81E+05 1.40E+04 825 65 1.10 0.13

The values are mean and SD of 5 injections.

* Since chromatographic separation of impurity peak was insufficient, this value

could not be calculated.

34

Table 2-6 Chromatographic parameters of phosphorylated peptides.

tR/min Intensity/cps S/N TF

mean SD mean SD mean SD mean SD

TRDIYETD-pY-YRK

GL-S column 6.17 0.00 2.24E+05 5.30E+03 237 5 1.11 0.03

GL-Ti column 6.26 0.00 3.90E+05 8.80E+03 529 43 1.15 0.01

GL-PEEK column 6.33 0.02 4.18E+05 3.50E+03 522 10 1.13 0.01

GL-PE column 6.16 0.02 4.15E+05 1.30E+04 620 30 1.07 0.01

SFVLNPTNIGM-pS-KSSQGHVTK

GL-S column 6.91 0.00 7.70E+03 2.90E+03 10 4 1.29 0.23

GL-Ti column 6.98 0.00 2.19E+05 2.70E+04 448 91 1.21 0.06

GL-PEEK column 7.03 0.02 2.95E+05 5.40E+03 423 97 1.22 0.05

GL-PE column 6.92 0.01 3.95E+05 1.30E+04 590 121 1.18 0.02

DLDVPIPGRFDRRV-pS-VAAE

GL-S column 8.32 0.01 4.34E+04 2.40E+03 53 10 1.03 0.08

GL-Ti column 8.41 0.00 9.94E+04 5.00E+03 234 37 1.05 0.01

GL-PEEK column 8.44 0.02 1.09E+05 6.30E+03 162 12 1.05 0.04

GL-PE column 8.32 0.01 1.11E+05 3.10E+03 230 10 1.05 0.05

The values are mean and SD of 5 injections.

35

Time2.00 4.00 6.00 8.00 10.00

%

0

100

2.00 4.00 6.00 8.00 10.00

%

0

100

2.00 4.00 6.00 8.00 10.00

%

0

100

2.00 4.00 6.00 8.00 10.00

%

0

100SUS LPA 10ppm 1uL 010 SIR of 1 Channel ES-

TIC (LPA)6.28e5

7.93

TiLPA 10ppm 1uL 010 SIR of 1 Channel ES- TIC (LPA)

6.28e5

3.68

PEEK LPA 10ppm 1uL 010 SIR of 1 Channel ES- TIC (LPA)

6.28e5

5.01

PE LPA 10ppm 1uL 010 SIR of 1 Channel ES- TIC (LPA)

6.28e54.70

Time2.00 4.00 6.00 8.00 10.00

%

0

100

2.00 4.00 6.00 8.00 10.00

%

0

100

2.00 4.00 6.00 8.00 10.00

%

0

100

2.00 4.00 6.00 8.00 10.00

%

0

100SUSfrit E4506 VB2 010 SIR of 2 Channels ES+

348 (FAD)3.57e5

3.11

Tifrit E4506 VB2 010 SIR of 2 Channels ES+ 348 (FAD)

3.57e5

3.071.11

PEEKfrit E4506 VB2 010 SIR of 2 Channels ES+ 348 (FAD)

3.57e53.07

1.12

PEfrit E4506 VB2 010 SIR of 2 Channels ES+ 348 (FAD)

3.57e52.98

1.100.99

(a)

(b)

(c)

(d)

(a)

(b)

(c)

(d)

Fig. 2-4 SIM chromatograms of FAD (left) and LPA (right) using GL-S column (a),

GL-Ti column (b), GL-PEEK column(c) and GL-PE column(d). A comparison of frit

is shown.

36

Time2.00 4.00 6.00 8.00

%

0

100

2.00 4.00 6.00 8.00

%

0

100

2.00 4.00 6.00 8.00

%

0

100

2.00 4.00 6.00 8.00

%

0

100test E4506 SUS 6 p pep 2000ug_L 003 SIR of 5 Channels ES+

568 (Kinase Insulin Receptor3)3.97e5

6.17

test E4506 Ti 6 p pep 2000ug_L 003 SIR of 5 Channels ES+ 568 (Kinase Insulin Receptor3)

3.97e56.26

test E4506 PEEK 6 p pep 2000ug_L 003 SIR of 5 Channels ES+ 568 (Kinase Insulin Receptor3)

3.97e56.35

test E4506 PE 6 p pep 2000ug_L 003 SIR of 5 Channels ES+ 568 (Kinase Insulin Receptor3)

3.97e56.14

(a)

(b)

(c)

(d)Time

2.00 4.00 6.00 8.00

%

0

100

2.00 4.00 6.00 8.00

%

0

100

2.00 4.00 6.00 8.00

%

0

100

2.00 4.00 6.00 8.00

%

0

100test E4506 SUS 6 p pep 2000ug_L 003 SIR of 5 Channels ES+

688 (B-casein p-pep)4.33e5

test E4506 Ti 6 p pep 2000ug_L 003 SIR of 5 Channels ES+ 688 (B-casein p-pep)

4.33e5

6.97

test E4506 PEEK 6 p pep 2000ug_L 003 SIR of 5 Channels ES+ 688 (B-casein p-pep)

4.33e5

7.03

test E4506 PE 6 p pep 2000ug_L 003 SIR of 5 Channels ES+ 688 (B-casein p-pep)

4.33e56.90

Time2.00 4.00 6.00 8.00

%

0

100

2.00 4.00 6.00 8.00

%

0

100

2.00 4.00 6.00 8.00

%

0

100

2.00 4.00 6.00 8.00

%

0

100test E4506 SUS 6 p pep 2000ug_L 003 SIR of 5 Channels ES+

731 (PKA Regulatory Subunit2)1.14e5

8.33

test E4506 Ti 6 p pep 2000ug_L 003 SIR of 5 Channels ES+ 731 (PKA Regulatory Subunit2)

1.14e5

6.98

test E4506 PEEK 6 p pep 2000ug_L 003 SIR of 5 Channels ES+ 731 (PKA Regulatory Subunit2)

1.14e5

7.04

test E4506 PE 6 p pep 2000ug_L 003 SIR of 5 Channels ES+ 731 (PKA Regulatory Subunit2)

1.14e5

6.90

(a)

(b)

(c)

(d)

(a)

(b)

(c)

(d)

6.00 8.00

6.165.97

8.096.96 8.39

SIR of 5 Channels ES+

Fig. 2-5 SIM chromatograms of 1 μmol TRDIYETD-pY-YRK (left) ,

SFVLNPTNIGM-pS-KSSQGHVTK(center) and DLDVPIPGRFDRRV-pS-VAAE

(right) using GL-S column (a), GL-Ti column (b), GL-PEEK column (c) and GL-PE

column (d).

37

2.4. Conclusions

We evaluated the influence of column hardware on LC-MS analysis of

phosphorylated compounds. In order to reduce of the interaction with the

phosphorylated compounds, using a GL-PE column could improve the S/N as well as

the detection intensity and the peak tailing. Replacing to PE frits has more effective

improvement of the intensity as well as the S/N than replacing the tubes of the column.

Since the metal ions eluted from our system were less than μg/L level, the

improvements were hardly affected. Thus, we believe that to use GL-PE column is

one of the best method for analysis of phosphorylated compounds.

38

2.5. References

1. T. Takeuchi and D. Ishii, J. Chromatogr., 1981, 213, 25.

2. T. Huang and P. T. Kissinger, Curr. Sep., 1996, 14, 114.

3. M. Konishi, Y. Mori, and T. Amano, Anal. Chem., 1985, 57, 2235.

4. K. K. Unger, R. Skudas, and M. M. Schulte, J. Chromatogr. A, 2008, 1184, 393.

5. Y. P. Zhang, Y. J. Zhang, W. J. Gong, N. Chen, A. L.Gopalan, and K. P. Lee,

Microchem. J., 2010, 95, 67.

6. J. Park, H. Oh and I. S. Jeon, J. Chromatogr. A, 2011, 1218, 7895.

7. X. Zhang and S. Huang, J. Chromatogr. A, 2001, 910, 13.

8. M. Kato, M. T. Dulay, B. D. Bennett, J. P. Quirino, and R. N. Zare, J. Chromatogr.

A, 2001, 924, 187.

9. B. Behnke, J. Johansson, S. Zhang, E. Bayer, and S. Nilsson, J. Chromatogr. A,

1998, 818, 257.

10. Y. Ishihama, J. Rappsilber, J. S. Andersen, and M. Mann, J. Chromatogr. A, 2002,

979, 233.

11. J. Kim, D. G. Camp, and R. D. Smith, J. Mass Spectrom., 2004, 39, 208.

12. P. C. Sadek, P. W. Carr, L. D. Bowers, and L. C. Haddad, Anal. Biochem., 1985,

144, 128.

13. A. Wakamatsu, K. Morimoto, M. Shimizu, and S. Kudoh, J. Sep. Sci., 2005, 28,

1823.

14. R. Zhao, S. J. Ding, Y. Shen, D. G. Camp II, E. A. Livesay, H. Udseth, and R. D.

Smith, J. Chromatogr. B, 2009, 877, 663.

15. T. De Vijlder, J. Boschmans, E. Witters, and F. Lemière, Inter. J. Mass Spectrom.,

2011, 304, 83.

39

16. Y. Asakawa, N. Tokida, C. Ozawa, M. Ishiba, O. Tagaya, and N. Asakawa, J.

Chromatogr. A, 2008, 1198, 80.

17. D. Winter, J. Selder, Y. Ziv, Y. Shiloh, and W. D. Lehmann, J. Proteome Res., 2009,

8, 418.

18. D. Siegel, H. Permentier, and R. Bischoff, J. Chromatogr. A, 2013, 1294, 87.

19. R. Tuytten, F. Lemiere, E. Witters, W. Van Dongen, H. Slegers, R. P. Newton, H.

Van Onckelen, and E. L. Esmans, J. Chromatography. A, 2006, 1104, 209.

20. F. Gaucheron, D. Molle, J. Leonil, and J. L. Maubois, J. Chromatogr. B, 1995, 664,

193.

21. V. Cael, D. Champmartin, and P. Rubini, J. Inorg. Biochem., 2003, 97, 97.

22. S. Liu, C. Zhang, J. L. Campbell, H. Zhang, K. K. Yeung, V. K. Han, and G. A.

Lajoie, Rapid Comm. Mass spectrum., 2005, 19, 2747.

23. J. Zhang, Q. Wang, B. Kleintop, and T. Raglione, J. Pharm. Biomed. Anal., 2014,

98, 247.

24. R. W. Slingsby, A. Bordanov, and M. Grimes, J. Chromatogr. A, 2001, 913, 159.

25. E. Gumienna-Kontecka, R. Silvagni, R. Lipinski, M. Lecouvey, F. C. Marincola, G.

Crisponi, V. M. Nurchi, Y. Leroux, and H. Kozłowski, Inorg. Chim. Acta, 2002, 339,

111.

26. S. Brunauer, P. H. Emmett, and E. Teller, J. Am. Chem. Soc., 1938, 60, 309.

27. Y. Shioi and S. I. Beale, Anal. Biochem., 1987, 162, 493.

28. Y. Ikeguchi and H. Nakamura, Anal. Sci., 1997, 13, 479.

29. I. Kuroda, Y. Shintani, M. Motokawa, S. Abe, and M. Furuno, Anal. Sci., 2004, 20,

1313.

40

30. Q. C. Lee, M. H. Roper, H. Terrisse, and B. Humbert, Colloids Surf. B, 2014, 123,

150.

41

Chapter 3

Evaluation of column carryover of phosphorylated peptides and fumonisins by

duplicated solvent gradient method in liquid chromatography-tandem mass

spectrometry

3.1. Introduction

Carryover is a phenomenon in which the sample from a previous injection is

observed or detected during the analysis of subsequent injections in HPLC. This is a

common problem that can compromise the accuracy of HPLC. Carryover is often

observed when injecting highly concentrated samples using LC/MS/MS. One of the

criteria for carryover in bioanalysis is that the peak area of the a blank sample that

follows an injection of a highest standard sample (i.e., ULOQ) should be less than 20%

of the peak area of the LLOQ of the sample.1, 2 Researchers often spend a significant

amount of time and effort to meet this criterion. Therefore, in order to solve this

problem, the ULOQ is set lower, while the LLOQ is maintained.3

Carryover is caused by residual material as well as irreversible adsorption in

the flow path within the LC/MS/MS system. It primarily occurs within the column4-6

and autosampler.7, 8 In the case of carryover caused by the autosampler, a peak of the

sample is eluted at the same decided retention time, whereas a carryover in the column

does not elute at the same retention time, and phenomena such as peak tailing or leading

are often observed.6 Carryover is often observed with phosphorylated and basic

compounds, in which the former is adsorbed on metallic material in the flow path via

interactions with metal ions. On the other hand, the latter is adsorbed on residual silanol

in the packing material.

42

Carryover in the autosampler can be improved by changing the sampler needle

and valve materials,3, 9-12 minimizing the contact area between the needle and seal, and

optimizing the solvents used to wash the needle.9, 11-13 In general, needle materials are

made of PEEK, stainless steel, and fused silica. Carryover could be minimized by using

modified platinum-coated, stainless steel needles. Carryover is caused by residual

samples between the needle and seal in the autosampler. The outside of the needle is

immersed in the sample in the vial, and the sample adsorbs on the needle. Therefore, the

carryover can be improved by using needle materials that do not adsorb the sample,

and/or by reducing the contact area between the needle and seal. In addition, washing

the needles with certain solvents, such as 100% organic solvents for hydrophobic

compounds and counter ionic solvents for ionic compounds, could be effective.14

Carryover in the column is mainly due to the adsorption of samples on the

packing materials.4-6, 15 The carryover could be improved by changing the washing time

of the column as well as the gradient profile.16 The carryover of fumonisins was

improved by using mobile phase containing 0.3% formic acid.17

On the other hand, in order to eliminate carryover of peptides, Dolman et al.

examined the column washing step in four columns containing different packing

materials.4 The carryover was eliminated by using a poly (styrene-co-divinylbenzene)

monolithic column without mesopores.

In HPLC, columns are packed with minute packing materials under high

pressure, and in order to avoid the outflow of the packing materials, filters which are

called “frits” are used at both ends of the column. As for the column hardware materials,

stainless steel, PEEK, glass, fused-silica capillary etc. are typically used. For the frit

materials, sintered stainless steel, mesh stainless steel, Ti, glass, Teflon, PEEK etc., are

43

used. The samples are adsorbed on the column hardware, and carryover occurs.

Mochizuki et al. eliminated the carryover of fumonisins by using a metal-free PEEK

column.15

The evaluated samples in this study were fumonisins and phosphorylated

peptides. Severe carryover was observed in preliminary studies. Fumonisins produced

by Fusarium species are called mycotoxins, which are toxic to both humans and

animals and may contaminate food and animal feed. Fumonisins contain four carbonyl

groups. Phosphorylated peptides have one or more phosphate groups, which lend a high

affinity towards metals. They are adsorbed on the metal parts of the flow pass in HPLC,

thus causing poor peak symmetry and carryover. The primary carryover occurs in the

column and autosampler, and identifying the carryover in partial HPLC systems is

difficult. Conventional carryover was calculated using the ratio of the peak area of the

previous sample to that of the blank sample. This method included all carryover, such as

in the needle and valves of the autosampler as well as in the column. Therefore, in order

to exclude the carryover of the injection valve operation in the autosampler, the

duplicated solvent gradient method18 was used. Since the injection carryover in the

autosampler was not included in this method, it was possible to limit the carryover in

the column and flow path. This method is ideal for calculating carryover in the column

only. In this paper, the carryover of three columns made of different materials was

evaluated using the duplicated solvent gradient method.

44

3.2. Experimental

3.2.1. Reagents and chemicals

A mixture of phosphorylated peptides, Mass PREP phosphorylated peptide

Sample Kit–Enolase, was purchased from Nihon Waters (Tokyo, Japan). This kit

included four phosphorylated peptides (T18p, T19p, T43p, and T43pp); however, only

two phosphorylated peptides (T18p: NVPLYpK and T19p: HLADLSpK) were used in

this work. A standard solution containing 50 mg/L of Fumonisin B1 (FB1) and B2 (FB2),

and a standard solution containing 50 mg/L Fumonisin B3 (FB3), formic acid,

acetonitrile, methanol, and isopropanol were purchased from Kanto Chemicals (Tokyo,

Japan). FAD was purchased from Wako Pure Chemical (Osaka, Japan). All other

reagents were of analytical grade and were used as received without any further

purification. Water was purified with a Milli-Q purification system obtained from

Millipore (Bedford, MA).

45

CH3

CH3

CH3

O

OOHO

OH

O

O

OH

O

OH

O

R1CH3 NH2

OHR2

Fig. 3-1 Chemical structures of fumonisin B1 (R1=OH, R2=OH), fumonisin B2 (R1=H,

R2=OH), fumonisin B3 (R1=OH, R2=H).

46

3.2.2. Sample treatment

FAD was dissolved in water at a concentration of 100 mg/L as stock solutions,

and the sample solution was dissolved in water at a concentration of 10 mg/L. The

mixture of FBs was dissolved in acetonitrile at a concentration of 10 mg/L to prepare

the stock solution. The sample solution containing the 3 FBs at a concentration of 1

mg/L was dissolved in water. A mixture of phosphorylated peptides was dissolved in

water at a concentration of 1000 fmol/μL to prepare the stock solution. A sample

solution of the mixture of phosphorylated peptides (500 fmol/μL) was dissolved in 10

mM phosphoric acid. All stock solutions were stored at -20 °C, and the sample solutions

were prepared immediately prior to analysis.

3.2.3. Chromatographic and mass spectrometric conditions

The chromatographic system was a Thermo Scientific Ultimate 3000 BioRS

(Bremen, Germany) with a 10 μL mixer. A mixture of

methanol/acetonitrile/H2O/isopropanol 1:1:1:1 (v/v/v/v) was used as the needle washing

solvent.

Fumonisins: The chromatographic mobile phases A and B were 0.1% formic

acid and acetonitrile, respectively. Using a gradient flow, the separation began with 80%

A and 20% B and reached 30% A and 70% B in 10 min, at a flow rate of 0.2 mL/min.

Next, the chromatographic system, including the column, was washed by passing

through 5% A and 95% B for 4 min. Next, the system was equilibrated by passing

through 80% A and 20% B for 6 min. The second gradient flow of the duplicated

solvent gradient method was the same as the first gradient condition without an

injection. The injection volume was 3 μL.

47

Phosphorylated peptides: The chromatographic mobile phases A and B were

0.1% formic acid and acetonitrile, respectively. Using a gradient flow, the separation

began with 95% A and 5% B and reached 50% A and B in 10 min, at a flow rate of 0.2

mL/min. Next, the chromatographic system, including the column, was washed by

passing through 5% A and 95% B for 4 min, and the system was equilibrated by passing

through 95% A and 5% B for 6 min. The injection volume was 2 μL.

The MS/MS was an AB Sciex 3200QTRAP triple quadrupole ion trap hybrid

mass spectrometer equipped with a Turbo V Ionsource. The measurements were

performed in positive mode with an ionization voltage of 5500 V at a temperature of

550 °C. The curtain gas setting was 30 psi. Gas 1 and Gas 2 were set to 60 psi. The

MS/MS detection was carried out in the SRM mode at m/z 722.4 → 352.4 for FB1 and

at m/z 706.35 → 336.3 for FB2 and FB3. The SRM transition for FB2 and FB3 could not

be independent. In regard to the peptides, the MS/MS detection was carried out in SRM

mode at m/z 407.2 → 186.2 for T18p and at m/z 432.2 → 383.2 for T19p. The other

MS/MS parameters and values are shown in Table 3-1. The MS/MS conditions were

tuned and optimized by the infusion of the sample solutions with a syringe pump.

Instrument control and data acquisition were carried out using Analyst 1.6.1 Software,

and the data was processed using MultiQuant 2.1.1 Software.

48

Table 3-1 MS/MS parameters and values of samples

Parameter Fumonisins Phosphorylated peptides

FB1 FB2 and FB3 T18p T19p

Declustering potential/V 101 66 40 40

Entrance potential/V 10 9.5 4 3

Collision energy/V 39 39 20 20

Collision cell exit potential/V 62 56 25 23

Dwell time per transition/ms 100 100 100 100

49

3.2.4. The column hardware

The chromatographic tubes made of stainless steel (150 mm x 2.1 mm I.D.)

and GL tubes (150 mm x 2.0 mm I.D.) were utilized without any deactivation treatments.

The sintered frits were made of stainless steel, PEEK, and PE. The S-S column was

composed of a stainless steel tube and stainless steel frits. The GL-PEEK column was

composed of a GL tube and PEEK frits. The GL-PE column was composed of a GL tube

and PE frits. Three columns were packed with L-column2 ODS 3 μm particles

(Chemicals Evaluation and Research Institute, Tokyo, Japan) from the same lot.

3.2.5. Carryover by duplicated solvent gradient method

The carryover values were calculated using the peak area ratio that eluted by

the second gradient to the first gradient in the duplicated solvent gradient method.

This peak area ratio was defined as the relative carryover. The use of relative

carryover had two advantages in the evaluation. First, the ratio was not directly related

to the absolute sample concentration. Second, the amount of the absolute carryover in

the analyte could be estimated by the relative carryover and concentration of the

preceding sample.19

The average relative carryovers were calculated by 6 repeated injections

using the duplicated solvent gradient method. The standard sample solutions

contained 1 mg/L fumonisins and 500 fmol/μL phosphorylated peptides. The samples

were evaluated in the three columns made of different materials, and the carryover

values were calculated.

3.2.6. Calibration curve of phosphorylated peptides in bioanalysis

50

The effect of the column hardware on the calibration curve of the standard

solution of phosphorylated peptides was investigated. These investigations involved the

carryover. The ULOQs of T18p and T19p were set to 500 fmol/μL. First, the standard

solutions of the phosphorylated peptides were prepared at concentrations of 5, 10, 25,

50, 100, 250, and 500 fmol/μL. Each standard sample solution was injected 6 times,

from low to high concentrations in each column; the calibration curves of the average

area were determined, and the weighting was found to be 1/x. Next, the relative

carryover of each column was calculated using the blank sample solution and the

ULOQ standard sample solution.

3.2.7. Injection carryover

Samples that were eluted by injection operation were not included in the

carryover calculated by the duplicated solvent gradient method. The injection

carryover could be calculated by subtracting the carryover values calculated by the

duplicated solvent gradient method from the carryover values calculated by the blank

injection. Thus, the method might be useful to compare the carryover in autosamplers.

The carryover of the injection operation in the autosampler was investigated.

The average relative carryovers were calculated for 6 repeated injections of the

standard sample solution and blank sample solution. The standard sample solutions

were the same as those described in the previous section. The GL-PE column was

used for these experiments.

51

3.3. Results and discussion

3.3.1. Evaluation of the prepared columns

The theoretical plates of the three kinds of columns made from different

materials were measured as a quality control measure, before evaluating the column

material, using an optimized Shimadzu Prominence 20A series semi-micro LC.

Naphthalene was used as the sample to measure the plates and TF, and uracil was used

for measuring the void volume. The relative standard deviation of the TF of naphthalene

in 6 columns containing the same material ranged from 1.0% to 2.7%, and all columns

exhibited a TF ranging from 1.02 to 1.11 with 12400 to 14000 plates (Table 3-2). The

number of theoretical plates in the S-S column was the greatest among all the columns.

The retention time of uracil in the S-S column was 1.55 min, and that in the GL-PE and

GL-PEEK column was 1.45 min. This difference was caused by the column volume,

since the retention times in all GL-columns were the same.

52

Table 3-2 The chromatographic parameters of the prepared 6 columns for quality

control.

to/min tR/min N TF

Mean Mean SD Mean SD Mean SD

S-S column 1.55 8.15 0.03 14000 50 1.02 0.01

GL-PEEK column 1.45 7.56 0.04 12400 410 1.11 0.03

GL-PE column 1.45 7.58 0.03 12900 190 1.11 0.02

The isocratic mobile phase 60% acetonitrile was used for column testing, at a flow

rate of 0.2 mL/min. t0 is retention time of uracil. tR is retention time of naphthalene. N

and TF of naphthalene are calculated by USP method. The values are mean and SD of

6 columns.

53

3.3.2. Improvement of adsorption of phosphorylated compounds in the liquid

chromatography-tandem mass spectrometry system

When configuring the MS conditions upon flow injection, tailing peaks of the

phosphorylated compounds were detected. Since this was considered to affect the

interaction between the metal and them, the tubing between the autosampler and column

was replaced with 650 mm x 0.075 mm nanoViper tubing (Thermo Scientific, Bremen,

Germany), and the tubing between the column and MS/MS was replaced with 650 mm

x 0.075 mm PEEKsil tubing. In addition, the stainless steel electrode in the Turbo V

Ionsource was replaced with a PEEKsil hybrid electrode. The inner surface of the

hybrid electrode was made from fused silica capillary, and the peak shape was

significantly improved as compared to that of the stainless steel (Fig. 3-2). The

phosphorylated compounds were reported to interact with stainless steel commonly

used in LC/MS systems, which likely caused the severe peak tailing.20

54

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.80%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

55%

60%

65%

70%

75%

80%

85%

90%

95%

100% 0.20

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.80%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

55%

60%

65%

70%

75%

80%

85%

90%

95%

0.19

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.80%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

55%

60%

65%

70%

75%

80%

85%

90%

95%

100% 0.14

Fig. 3-2 SRM chromatogram of improvement of FAD peak tailing by the LC-MS/MS

system without a column. (a) Standard system, (b) Replaced Nanoviper tubing and

PEEKsil tubing, (c) Replaced connecting tubes and Hybrid electrode.

(a) (b) (c)

Time/min Time/min Time/min

Rel

ativ

e in

tens

ity, %

Rel

ativ

e in

tens

ity, %

Rel

ativ

e in

tens

ity, %

55

3.3.3. Carryover by duplicated solvent gradient method

The carryovers observed from all samples in the duplicated solvent gradient

method did not include those of the blank sample injection. These corresponded to the

sum of the carryover of the tubing in the flow path and column, including the packing

materials and hardware.

The carryover values of FB1 in the S-S column were about 10 times greater

than those in the other columns (Table 3-3a). If the peak separation was not enough,

calculating the peak area was difficult due to the identical SRM transition of FB2 and

FB3. Fig. 3-3 shows the expanded SRM chromatogram of the FBs. The leading peaks

occurred in the carryover peak with the S-S column, and the separation of FB2 and

FB3 could not be achieved and the peak areas could not be calculated. Thus, we

calculated the carryover values using the sum of the peak areas of FB2 and FB3.

Leading peaks were not observed with the other columns. The S-S column showed a

tailing peak that eluted in the first solvent gradient, while the other columns showed

sharp peaks. These results were mainly due to the difference in metal ions on the inner

surface of the column hardware.15 Since metal ions hardly exist on the inner surface of

the GL-PEEK and GL-PE columns, the FBs were not adsorbed. The S/N of FB1 in the

GL-PE column was highest among all columns. Metal ions hardly exist on the inner

surface of the GL-PEEK and GL-PE columns; however, a good performance was

obtained when the GL-PE column was used. The polarity of the PEEK and PE frits

likely differed. The latter could not entrap FB1, since the difference in polarity

between FB1 and the PE frits was larger than that between FB1 and PEEK frits.21 PE is

highly hydrophobic, and powders were used as packing materials in reversed-phase

LC.22

56

The retention times of the carryover peak of FBs were in agreement with

those obtained by adding the start time of the second solvent gradient to the retention

time of the peak in the first solvent gradient. This indicated that FBs were adsorbed in

the tubing and/or in the inlet of the stainless steel frit.6 In addition, the carryover peaks

of FBs showed severe leading, indicating that the samples eluted at shorter retention

times. Therefore, the carryover occurred in the column.6 We considered that the

carryover in the leading peaks was due to the adsorption of FBs on the packing

materials, chromatographic tube, and/or outlet frit of the column made of stainless

steel. However, since the other columns did not show the leading peak, the packing

materials could be excluded. The carryover was caused by the chromatographic tube

and inlet frit because sharp peaks were obtained with the GL column.

In regard to the carryover observed with phosphorylated peptides, the values

and peak shape of the carryover peaks were similar to those of the FBs. The carryover

of T19p with the GL-PE column was reduced to one-tenth compared to that with the

S-S column (Table 3-3b). Fig. 3-4 shows the carryover peaks for T19p, which were

more asymmetric than those of T18p in all columns, likely because of stronger

interactions between T19p and metal ions in the columns.14, 23, 24 The carryover peak

in the S-S column showed leading. The carryover in the leading peaks was likely due

to adsorption of T18p and T19p on the chromatographic tube and/or outlet frit of the

column made of stainless steel.

57

Table 3-3 Comparison of mean and SD of carryover by duplicated solvent gradient

method and S/N using different column hardware.

Table 3-3a 1 mg/L fumonisins

FB1 FB3 FB2 FB2+FB3

Carryover, % S/N Carryover, % Carryover, % Carryover, %

Mean SD Mean SD Mean SD Mean SD Mean SD

S-S column 1.90 0.09 527 14 - - - - 2.76 0.16

GL PEEK column 0.18 0.04 693 12 0.36 0.04 0.56 0.09 0.49 0.07

GL PE column 0.20 0.06 792 15 0.47 0.02 0.68 0.02 0.67 0.05

Table 3-3b 500 fmol/μL phosphorylated peptides

T18p T19p

Carryover, % S/N Carryover, % S/N

Column Mean SD Mean SD Mean SD Mean SD

S-S column 1.31 0.06 396 16 4.28 0.44 323 26

GL PEEK column 0.24 0.02 392 52 0.67 0.18 370 33

GL PE column 0.21 0.04 483 43 0.42 0.12 417 32

58

5 10 15 20 250

2

4

6

8

10

Time/min

Rel

ativ

e in

tens

ity, %

5 10 15 20 250

2

4

6

8

10

Time/min

Rel

ativ

e in

tens

ity, %

5 10 15 20 250

2

4

6

8

10

Time/min

Rel

ativ

e in

tens

ity, %

5 10 15 20 250

2

4

6

8

10

Time/min

Rel

ativ

e in

tens

ity, %

5 10 15 20 250

2

4

6

8

10

Time/min

Rel

ativ

e in

tens

ity, %

5 10 15 20 250

2

4

6

8

10

Time/min

Rel

ativ

e in

tens

ity, %

(a)

(f)

(d)

(e)

(c)

(b)

FB1

Carryover of FB1

Carryover of FB1

FB1

FB1

Carryover of FB1

Carryover of FB2

Carryover of FB2

Carryover of FB2

Carryover of FB3

Carryover of FB3

Carryover of FB3

FB2

FB3

FB2

FB2

FB3

FB3

Fig. 3-3 Expanded SRM chromatograms of fumonisins by the duplicated solvent

gradient method. (a) SRM chromatogram of FB1 using S-S column. (b) SRM

chromatogram of FB3 and FB2 using S-S column. (c) SRM chromatogram of FB1 using

GL-PEEK column. (d) SRM chromatogram of FB3 and FB2 using GL-PEEK column.

(e) SRM chromatogram of FB1 using GL-PE column. (f) SRM chromatogram of FB3

and FB2 using GL-PE column.

59

5 10 15 20 250

2

4

6

8

10

Time/min

Rel

ativ

e in

tens

ity, %

5 10 15 20 250

2

4

6

8

10

Time/min

Rel

ativ

e in

tens

ity, %

5 10 15 20 250

2

4

6

8

10

Time/min

Rel

ativ

e in

tens

ity, %

5 10 15 20 250

2

4

6

8

10

Time/min

Rel

ativ

e in

tens

ity, %

5 10 15 20 250

2

4

6

8

10

Time/min

Rel

ativ

e in

tens

ity, %

5 10 15 20 250

2

4

6

8

10

Time/min

Rel

ativ

e in

tens

ity, %

(a)

(f)

(d)

(e)

(c)

(b)

Carryover of T19p

Carryover of T19p

Carryover of T19pCarryover of T18p

Carryover of T18p

Carryover of T18p

T18p

T18p

T18p

T19p

T19p

T19p

Fig. 3-4 Expanded SRM chromatograms of phosphorylated peptides by the duplicated

solvent gradient method. (a) SRM chromatogram of T18p using S-S column. (b) SRM

chromatogram of T19p using S-S column. (c) SRM chromatogram of T18p using

GL-PEEK column. (d) SRM chromatogram of T19p using GL-PEEK column. (e)

SRM chromatogram of T18p using GL-PE column. (f) SRM chromatogram of T19p

using GL-PE column.

60

3.3.4. Calibration curve of standard samples in phosphorylated peptides

In bioanalysis, the peak area of a blank sample that follows an injection of

ULOQ sample should be less than 20% of the peak area of the LLOQ sample.1, 2

Therefore, the carryover should be reduced in order to expand the range of the

calibration curve. As such, we investigated whether or not the column affected the

calibration curves of the phosphorylated peptides.

When the S/N was less than 10, the concentration was excluded from the

range. As a result, the calibration curve of T18p with each column could be prepared

in the range of 10 fmol/μL to 500 fmol/μL. The range of T19p in the S-S column was

from 50 fmol/μL to 500 fmol/μL, that in the GL- PEEK column was from 25 fmol/μL

to 500 fmol/μL, and that of the GL-PE column was from 10 fmol/μL to 500 fmol/μL.

Table 3-4 shows the parameters for each column in the calibration range. The

parameters of T18p were slightly different from those of T19p. However, the slope of

T19p with the GL-PE column was the greatest among those with all other columns,

indicating a high sensitivity. The S/N of 50 fmol T19p in the S-S column was 11, that

in the GL-PEEK column was 45, and that in the GL-PE column was 82. The

differences were likely related to interactions with the metal, as noted for the FBs.

Next, the criteria of the carryover in bioanalysis were added to the results.

When the ULOQ was set to 500 fmol, the LLOQ was calculated from the calibration

curve of each column using the blank sample peak area. Table 3-5 shows the results.

The carryover of T18p in the S-S column was 1.46%, and the LLOQ was calculated to

be 36 fmol/μL. However, the LLOQ of the GL column was less than the value of the

S/N. The carryover of T19p in the S-S column was 4.47%, and the LLOQ was

calculated to be 126 fmol/μL. The adsorption of T19p by metal ions was estimated to

61

be very high. Therefore, the carryover was high, and the LLOQ also increased. In

contrast, the LLOQ of the GL-PE column was one-tenth that of the S-S column; the

value was 13.4 fmol/μL. Since the GL-PE column and GL-PEEK hardly contain metal

in the flow path, the LLOQ was low, and the range of the calibration curve was

expanded.

62

Table 3-4 Comparison of calibration curves obtained using different column hardware.

T18p T19p

Column Slope Intercept R2 Slope Intercept R2

S-S column 61.3 26 0.992 34.1 -167 0.995

GL-PEEK column 66.8 152 0.990 51.3 -3 0.995

GL-PE column 72.3 119 0.991 72.9 15 0.994

63

Table 3-5 Comparison of carryover and LLOQ of T18p (upper) and T19p (bottom)

obtained using different column hardware.

Area

Column ULOQ/cps Blank/cps Carryover, % LLOQ/fmol μL-1

S-S column 30630 448 1.46 35.8

GL PEEK column 32510 205 0.63 10>*

GL PE column 34450 170 0.49 10>*

Area

Column ULOQ/cps Blank/cps Carryover, % LLOQ/fmol μL-1

S-S column 16940 757 4.47 126.3

GL PEEK column 25250 206 0.82 25>*

GL PE column 35080 211 0.60 13.4

*Less than the range of the calibration curve.

64

3.3.5. Injection carryover

The duplicated solvent gradient method and the blank injection method were

used. The carryover values in the former method were calculated using the value

without the blank sample injection, and the latter included the carryover of the blank

sample injection. The differences corresponded to carryover in the injection operation.

Table 3-6 shows the carryover values of the injection operation calculated by these

methods using the GL-PE column. The calculated carryover of the injection operation

was not dependent on the type of column.

Carryover of FB2 by the blank injection method was 0.79%, and that by the

duplicated solvent gradient method was 0.68%. This difference (0.11%) was due to

carryover in the injection operation. Therefore, improvements in the flow path,

including the inlet frit of the column, were more effective than those in the washing of

the autosampler needle. Using the biocompatible-UHPLC and the conventional needle

washing step, carryover of FBs hardly occurred. The carryover in the injection

operation of T18p and T19p were 0.27 and 0.18%, respectively. The carryover might

be improved by washing the autosampler.

Even though numerous reports indicated that carryover occurs in the

autosampler and column, the percentages were not shown. If the carryover of FB1 and

T18p were not caused by the GL-PE and GL-PEEK columns, the carryover was

calculated by the duplicated solvent gradient method and corresponded to those of the

tubing within the flow path. The carryover value of FB1 with the S-S column was 1.67%

and that of the system was 0.23%. The carryover of T18p with the S-S column was

0.82% and that of the system was 0.49%. We found that the majority of the carryover in

our system was due to the S-S column and the gradient method.

65

Table 3-6 Comparison of mean of carryover (%) obtained using PE column.

Carryover FB1 FB3 FB2 T18p T19p

By the blank injection 0.23 0.53 0.79 0.49 0.60

By the duplicated solvent gradient 0.20 0.47 0.68 0.21 0.42

Injection carryover 0.03 0.06 0.11 0.28 0.18

These values are the average of the percent carryovers calculated by the blank sample

and duplicated solvent gradient method.

66

3.4. Conclusions

The use of a GL-PE column improved the peak shape and carryover more

effectively than the other columns. The GL-PEEK and GL-PE columns greatly

improved the carryover of fumonisins and phosphorylated peptides, and the carryover

of FB1 could be reduced to 1/10. In addition, the duplicated solvent gradient method

could be used to evaluate the column carryover. The use of GL-PEEK and GL-PE

columns also improved the LLOQ of phosphorylated peptides, and the range of the

calibration curve could be expanded. Since carryover peaks in the GL-PE column were

symmetrical peak of the samples, carryover did not occur in the column. The carryover

calculated using the duplicated solvent gradient method corresponded to that within the

tubing within the flow path. The carryover value of FB1 in the S-S column was 1.67%

and that of the system was 0.23%. That of T18p in the S-S column was 0.82% and that

of the system was 0.49%. We found that the majority of the carryover in our system

occurred due to the S-S column and the method.

67

3.5. References

1. EMA, Guideline on the validation of bioanalytical methods,

EMA/CHMP/EWP/192217/2009 Rev.1 Corr., 21 July, 2011.

2. US FDA, Guideline for Industry; Bioanalytical Method Validation, Draft,

September, 2013.

3. C. Chassaing, J. Luckwell, P. Macrae, K. Saunders, P. Wright, and R. Venn,

Chromatographia, 2000, 53, 122.

4. S. Dolman, S. Eeltink, A. Vaast, and M. Pelzing, J. Chromatogr. B, 2013, 912, 56.

5. R. ter Heine, C. G. Alderden-Los, H. Rosing, M. J. Hillebrand, E. C. van Gorp, A. D.

Huitema, and J. H. Beijnen, Rapid Commun. Mass Spectrom., 2007, 21, 2505.

6. N. C. Hughes, E. Y. Wong, J. Fan, and N. Bajaj, The AAPS Journal, 2007, 9, 353.

7. J. W. Dolan, LC GC North America, 2001, 19, 164.

8. J. W. Dolan, LC GC North America, 2001, 19, 478.

9. P. T. Vallano, S. B. Shugarts, E. J. Woolf, and B. K. Matuszewski, J. Pharm. Biomed.

Anal., 2005, 36, 1073.

10. R. P. Grant, C. Cameron, and S. Mackenzie-McMurter, Rapid Commun. Mass

Spectrom., 2002, 16, 1785.

11. K. W. Dunn-Meynell, S. Wainhaus, and W. A. Korfmacher, Rapid Commun. Mass

Spectrom., 2005, 19, 2905.

12. A. Schellen, B. Ooms, D. van de Lagematt, R. Vreeken, and W. D. van Dongen, J.

Chromatogr. B, 2003, 788, 251.

13. Y. Asakawa, C. Ozawa, K. Osada, S. Kaneko, and N. Asakawa, J. Pharm. Biomed.

Anal., 2007, 43. 683.

14. M. Tamura, K. Matsumoto, J. Watanabe, J. Iida, Y. Nagatomi, and N. Mochizuki, J.

68

Sep. Sci., 2014, 00, 1.

15. Shimadzu Technical Report, Vol. 17.

http://www2.shimadzu.com/applications/lcms/Shimadzu_TechReport_Vol17_LCMS

.pdf

16. M. Tamura, A. Uyama, and N. Mochizuki, Anal. Sci., 2011, 27, 629.

17. T. Gazzotti, E. Zironi, B. Lugoboni, A. Barbarossa, A. Piva, and G. Pagliuca, Food

Chemistry, 2011, 125, 1379.

18. D. H. Vu, R. A. Koster, A. M. A. Wessels, B. Greijdanus, J. W. C. Alffenaar, and D.

R. A. Uges, J. Chromatogr. B, 2013, 917, 1.

19. W. Zeng, D. G. Musson, A. L. Fisher, and A. Q. Wang, Rapid Commun. Mass

Spectrom., 2006, 20, 635.

20. A. Wakamatsu, K. Morimoto, M. Shimizu, and S. Kudoh, J. Sep. Sci., 2005, 28,

1823.

21. H. Sakamaki, T. Uchida, L. W. Lim, and T. Takeuchi, J. Chromatogr. A, 2015, in

press.

22. Y. Shioi and S. I. Beale, Anal. Biochem., 1987, 162, 493.

23. S. Liu, C. Zhang, J. L. Campbell, H. Zhang, K. K. C. Yeung, V. K. M. Han, and G.

A. Lajoje, Rapid Commun. Mass Spectrom., 2005, 19, 2747.

24. T. De Vijlder, J. Boschmans, E. Witters, and F. Lemiere, Int. J. Mass Spectrom.,

2011, 304, 83.

69

Conclusions

We evaluated the influence of column hardware on LC-MS analysis of

phosphorylated compounds. In comparison with 2 kinds of the tubes, the S/N of FAD

with GL-S column was 6 times higher than that of S-S column, and the intensity was 4

times larger for the former than the latter. The intensity and S/N of LPA were 1.7 times

higher for GL-S column than S-S column and the TF of FAD and LPA with GL-S

column were improved, compared to S-S column. In comparison with 4 kinds of the

frits, the intensity and the S/N with GL-PE column were higher than those of the other

columns. The intensity and the S/N of FAD for GL-PE column were 19 times and 30

times higher than that for GL-S column, respectively. We suggest that the mechanisms

of retention in our system were affected more by the Fe ions present on the inner surface

of the S-S column. Since the specific surface area of a pair of frits was 70 times larger

than that of a chromatographic tube (150 mm x 2.1 mm), the frits were found to have

more effective improvement of the S/N as well as the intensity than the

chromatographic tubes, when phosphorylated compounds were analyzed by LC-MS.

When the phosphorylated compounds were analyzed by LC-MS(/MS) using a GL-PE

column, the intensity and S/N were increased.

In order to eliminate carryover of injection operation in autosampler, the

calculated carryover by the duplicated solvent gradient method was developed.GL-PE

column improved most effectively the peak shape and carryover compared to the other

columns. The use of GL-PEEK and GL-PE columns greatly improved the carryover of

fumonisins and phosphorylated peptides, and carryovers of FB1 could be reduced to

1/10. In addition, the duplicated solvent gradient method could be used for evaluating

70

the column carryover. The use of GL-PEEK and GL-PE columns also improved the

LLOQ of phosphorylated peptides, and the range of the calibration curve was able to

expand. If the carryovers of FB1 and T18p were not caused by GL-PE and/or GL-PEEK

columns, those calculate by using the duplicated solvent gradient method were

corresponded to those of the tubings of flow path. The carryover values of FB1 with S-S

column were 1.67% and those of the system were 0.23%. Those of T18p with S-S

column were 0.82% and those of the system were 0.49%. We found that the majority of

carryover has occurred by S-S column in our system and the method.

71

Figure list

Chapter 2

Fig. 2-1 Chemical Structures of AMPs, FAD and LPA.

Fig. 2-2 SIM chromatograms of FAD (left) and LPA (right) using S-S column (a),

GL-S column (b). A comparison of chromatographic tube is shown.

Fig. 2-3 Effect of concentration of Fe3+ in LPA sample standard solution on the peak

area, peak height and TF of LPA.

Fig. 2-4 SIM chromatograms of FAD (left) and LPA (right) using GL-S column (a),

GL-Ti column (b), GL-PEEK column(c) and GL-PE column(d). A comparison of frit

is shown.

Fig. 2-5 SIM chromatograms of 1 μmol TRDIYETD-pY-YRK (left) ,

SFVLNPTNIGM-pS-KSSQGHVTK(center) and DLDVPIPGRFDRRV-pS-VAAE

(right) using GL-S column (a), GL-Ti column (b), GL-PEEK column (c) and GL-PE

column (d).

72

Chapter 3

Fig. 3-1 Structure of fumonisin B1 (R1=OH, R2=OH), fumonisin B2 (R1=H, R2=OH),

fumonisin B3 (R1=OH, R2=H).

Fig. 3-2 SRM chromatogram of improvement of FAD peak tailing by the LC-MS/MS

system without a column. (a) Standard system, (b) Replaced Nanoviper tubing and

PEEKsil tubing, (c) Replaced connecting tubes and Hybrid electrode.

Fig. 3-3 Expanded SRM chromatograms of Fumonisins by the duplicated solvent

gradient method. (a) SRM chromatogram of FB1 using S-S column. (b) SRM

chromatogram of FB3 and FB2 using S-S column. (c) SRM chromatogram of FB1

using GL-PEEK column. (d) SRM chromatogram of FB3 and FB2 using GL-PEEK

column. (e) SRM chromatogram of FB1 using GL-PE column. (f) SRM chromatogram

of FB3 and FB2 using GL-PE column.

Fig. 3-4 Expanded SRM chromatograms of phosphorylated peptides by the duplicated

solvent gradient method. (a) SRM chromatogram of T18p using S-S column. (b) SRM

chromatogram of T19p using S-S column. (c) SRM chromatogram of T18p using

GL-PEEK column. (d) SRM chromatogram of T19p using GL-PEEK column. (e)

SRM chromatogram of T18p using GL-PE column. (f) SRM chromatogram of T19p

using GL-PE column.

73

Table list

Chapter 2

Table 2-1 The types of chromatographic tube and frit used in each specific column.

Table 2-2 The chromatographic parameters of the prepared 6 columns for quality

control.

Table 2-3 Comparison of the analysis of copper oxinate (0.4ng) as a

non-phosphorylated compound for prepared 5 kinds of columns.

Table 2-4 Comparison of chromatographic parameter between S-S and GL-S columns

with FAD (50 ng), LPA (10 ng) and AMPs (2.5 ng) used as the samples.

(Chromatographic tubes were compared, i.e. between S and GL, while the same S-frit

was used)

Table 2-5 Comparison of chromatographic parameters for FAD (50 ng), LPA (10 ng)

and AMPs (2.5 ng). (4 types of frits were compared while the same kind of

GL-chromatographic tube was used)

Table 2-6 Chromatographic parameters of phosphorylated peptides.

74

Chapter 3

Table 3-1 MS/MS parameters and values of samples.

Table 3-2 The chromatographic parameters of the prepared 6 columns for quality

control.

Table 3-3 Comparison of mean and SD of carryover by duplicated solvent gradient

method and S/N using different column hardware.

Table 3-4 Comparison of calibration curves obtained using different column hardware.

Table 3-5 Comparison of carryover and LLOQ of T18p (upper) and T19p (bottom)

obtained using different column hardware.

Table 3-6 Comparison of mean of carryover (%) obtained using PE column.

75

List of publications

1. Hiroshi Sakamaki, Takeharu Uchida, Lee Wah Lim, and Toyohide Takeuchi,

“Evaluation of column hardware on liquid chromatography-mass spectrometry of

phosphorylated compounds”, Journal of chromatography A, 2015, 1381, 125.

2. Hiroshi Sakamaki, Takeharu Uchida, Lee Wah Lim, and Toyohide Takeuchi,

“Evaluation of column carryover of phosphorylated peptides and fumonisins by

duplicated solvent gradient method in liquid chromatography/tandem mass

spectrometry”, Analytical Science, 2015, 31, 91.

76

List of presentations

1. Hiroshi Sakamaki, Yoshihisa Sudo, Lee Wah Lim, and Toyohide Takeuchi,

“Influence of Column Materials on LC/MS analysis of Phosphate Compounds”, The

23rd Conference of the Society for Chromatographic Science, Gifu, November,

14-16, 2012 (Poster Presentation)

2. Hiroshi Sakamaki, Yoshihisa Sudo, Lee Wah Lim, and Toyohide Takeuchi,

“Reduction in the adsorption of phosphorylated compounds in LC-MS”, The 133ed

Annual Meeting of the Pharmaceutical Society of Japan, Yokohama, March, 27-30,

2013 (Poster Presentation)

3. Tomoki Obata, Hiroshi Sakamaki, Yoshihisa Sudo, Lee Wah Lim, and Toyohide

Takeuchi,

“Investigation of analytical conditions for utilizing the characteristic of the

Phenylhexylcolumn effectively” The 133rdAnnual Meeting of the Pharmaceutical

Society of Japan, Yokohama, March, 27-30, 2013 (Poster Presentation)

4. Hiroshi Sakamaki, Yoshihisa Sudo, Lee Wah Lim, and Toyohide Takeuchi,

“Influence of Column Materials on Biocompatible LC/MS/MS Analysis Indicated

by Phosphorylated Compounds” The 24th Conference of the Society for

Chromatographic Science, Tokyo, November 11-13, 2013 (Oral Presentation)

5. Hiroshi Sakamaki, Takeharu Uchida, Lee Wah Lim, and Toyohide Takeuchi,

77

“Evaluation and applications of metal free column in LC/MS/MS “ The 134th

Annual Meeting of the Pharmaceutical Society of Japan, Kumamoto, March, 27-30,

2014 (Poster Presentation)

6. Yuta Nakano, Hiroshi Sakamaki, Takeharu Uchida, Lee Wah Lim, and Toyohide

Takeuchi,

“Effect of Metal Free Column on LC/MS/MS Analysis” The 27th Symposium on

Biomedical-Analytical Sciences, Tokyo, August 20-21, 2014 (Poster Presentation)

7. Hiroshi Sakamaki, Takeharu Uchida, Lee Wah Lim, and Toyohide Takeuchi,

“Reduction of carryover using metal-free column on Biocompatible LC/MS/MS”,

The 25th Conference of the Society for Chromatographic Science, Tokyo, December

11-13, 2014 (Oral Presentation)

78

Curriculum vitae

Hiroshi SAKAMAKI

Born on June 17, 1976 in Aichi prefecture, Japan.

1992 – 1995: Okazaki-nishi High School, Aichi, Japan.

1995 – 1999: Bachelor at Department of Applied Chemistry, faculty of Engineering,

Gifu University, Gifu, Japan.

1999 – 2001: Master Degree at Applied Chemistry Division, Graduate School of

Engineering, Gifu University, Gifu, Japan.

2001 – now: Chemicals Evaluation and Research Institute, Japan (CERI).

2012 – 2015: Doctoral Degree at Material Engineering Division, Graduate School of

Engineering, Gifu University, Gifu, Japan.

79

Acknowledgements

I would like to express my sincerest gratitude and appreciation to my advisor

Professor Toyohide Takeuchi, for great support to me throughout my study with his

wide knowledge, patience and understanding. His precious suggestions, encouragement

and effort have provided a good basis for my dissertation.

I would like to express my heartfelt gratitude to Dr. Lee Wah Lim for each

inspirational, supportive, and helped me in all the time of research and writing

dissertation.

In order to obtain a doctorate, it is deeply grateful to the CERI that had been

provided me a great deal of time and financial support.

Sincere thanks to all members in Takeuchi’s laboratory for their kind support,

and all members of my division in CERI.

I would like to express gratitude for Dr. Tadokoro Hiroshi, Mr. Takeharu

Uchida and late Dr. Yoshihisa Sudo for be giving me the opportunity and many

supports in order to begin this study.

Last but not least, I would like to thank my father and mother, mother in law.

Special word of gratitude to my wife Yoshie and my sons Hodaka, Eita, and Ryowa for

their patience, deep understanding, great support, help and their unconditional love.

Hiroshi Sakamaki

March 2015