”QM
?

r:

.

3w...

. . H...
smwmefln-w

a...
fin

 

 

a
Last
.442:

44.4...

:1 u"...
. .. . . .L .
Iii—3:01.". I .'1. .

 

 

 

L'QQARY
N'liiil ”you: State

Universitymm

 

 

This is to certify that the
dissertation entitled

TOWARD PHOSPHORYLATED PEPTIDE ENRICHMENT
BASED ON ORGANOFUNCTIONALIZED
MESOSTRUCTURED SILICA

presented by

Dong-Keun Lee

has been accepted towards fulfillment
of the requirements for the

Doctoral degree in Chemistry

 

 

    

i/ZM. !

 

Major Pro ;. o ' Signature

    

MSU is an Affinnative Action/Equal Opportunity Employer

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5I08 K:IProj/Acc&Pres/ClRC/DateDue indd

 

TOWARD PHOSPHORYLATED PEPTIDE ENRICHMENT BASED ON
ORGANOFUNCTIONALIZED MESOSTRUCTURED SILICA

By

Dong-Keun Lee

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Chemistry

2009

ABSTRACT

TOWARD PHOSPHORYLATED PEPTIDE ENRICHMENT BASED ON
ORGANOFUNCTIONALIZED MESOSTRUCTURED SILICA

By
Dong-Keun Lee

Reversible protein phosphorylation is a key regulatory process in
controlling many cellular events, such as cell cycle, cell growth, cell differentiation,
and metabolism."2 In order to achieve detailed insight into the regulation of these
reversible phosphorylation processes, it is often necessary to characterize the
phosphorylation sites of specific proteins. Nevertheless, the identification of
phosphopeptides including phosphorylation sites by analytical technique, such as
mass spectrometric method, remains challenging. Therefore, specific isolation
and enrichment of phosphorylated protein and phosphorylated peptides are often
prior to analytical process. So far the most common enrichment method is
IMAC? (immobilized metal affinity chromatography) Due to the lack of selectivity
for containing carboxylate groups in IMAC enrichment, however, this method still
has a limitation. Recently, to overcome this poor selectivity, a few covalent
binding methods have been studied, such as covalent binding of modified
phOsphopeptides by B-elimination, covalent binding by a—diazo functionalized

polymer resin, and covalent binding by oxidation-reduction condensation reaction

of amine functionalized polymer resin.”6 Those methods show better selectivity,
but still have limitations in the process of enrichment, such as increment of
protein complexity by unwanted reactions, limitation of phosphotyrosine species
enrichment, swelling problems upon solvent. Therefore, in this study I designed
a-diazo functionalized mesoporous silica solid resins, which have rigid open
framework structures, very high surface areas, and their wide ranges of pH and
solvent stability in the application of the covalent immobilization and separation of
phosphorylated peptides and proteins. In addition, for the facilitating enrichment
process and minimizing sample loss, new types of separation media, for example

a-diazo functionalized mesoporous thin film, were prepared.

1. Graves, J. D.; Krebs, E. G. Pharmacol. Ther. 1999, 82, 111-121

2. Hunter, T. Cell 2000, 100, 113-117

3. Posewitz, M. C.; Tempst, P. Anal. Chem. 1999, 71, 2883-2898

4. Oda, Y., Nagasu, T., Chait, B. T. Nat. Biotechnol. 2001, 19, 379-382
5. Lansdell, T. A.; Tepe, J. J. Tet. Let. 2003, 45, 91-93

6. Warthaka, M., Karwowska-Desaulniers, P., Pflum, M. ACS Chemical Biology,
2006, 1, 11, 697-701

Copyright
By
Dong-Keun Lee

2009

ACKNOWLEDGEMENTS

I want to express my appreciation to Dr. Pinnavaia for his support and
guidance during my graduate studies. I greatly appreciate the intellectual and
inspirational discussions we have had over the years. I also would like to convey
my gratitude for the flexibility he has allowed me so that I could complete my
research while taking care of my family.

I am grateful to Center for Advance Microscopy, especially Dr. Xudong
Fan, Allica and Ewa for all of their support. I thank for their teaching the basic
SEM and TEM, which have been very helpful to my research.

I would like to express my thanks to current and former group members
for their friendship. Especially, I am thankful to Dr. Seong-Su Kim for his
intellectual discussions as well as friendship.

l am grateful to my family for their support. My parents, Myung-Hun and
Hye-Suk, have helped and scarified for me to achieve my goal during all their
lives. Their unlimited loves have always encouraged me to endure and
overcome from the time of fmstration during my life. I also want to thank my
sister, parents-in-law, and sister-in-law.

Most importantly, I would like to express my deepest gratitude to my son
and wife, Benjamin and Eun-Sook, for all of the unconditional love and
encouragement they have given over the years. Without their support and

sacrifices, I would have not been able to accomplish any of this success.

TABLE OF CONTENTS

LIST OF TABLES ................................................................................ ix
LIST OF FIGURES ............................................................................. x
LIST OF SCHEME .............................................................................. xiv
ABBREVIATIONS ................................................................................ xv
1. Chapter 1
Introduction .................................................................................. 1
1.1 Research Objects and Significance ................................................ 1
1.2 Enrichment method of phosphorylated protein ................................. 4
1.2.1 Immobilized Metal Affinity Chromatography (IMAC) ..................... 4
1.2.2 Immunoprecipitation (IP) ......................................................... 6
1.2.3 Metal Oxide/hydroxide Affinity Chromatography (MOAC)............... 7
1.2.4 Specific chemical modification methods ..................................... 8
1.2.5 Previous method applied in nanoscience ................................... 14
1.2.5.1 Dendrimer .................................................................... 14
1.2.5.2 Nanoparticle .................................................................. 17
1.2.5.3 Zeolite .......................................................................... 17
1.2.5.4 Mesoporous (MCM-41) .................................................... 18
1.3 Organo-functionalized mesostructure ............................................. 18
1.3.1 Post-synthesis pathway..." ..................................................... 19
1.3.2 Co-condensatioin pathway ..................................................... 22
1.4 References ............................................................................... 25
2. Chapter 2
Direct Assembly of Large Pore Organofunctionalized Silica Mesostructure from
Sodium Silicate and bis(2-hydroxyethyl)«3-aminopropylsilane ..................... 29
2.1 Introduction ............................................................................... 29
2.2 Experimental ............................................................................. 32
2.2.1 Regents .............................................................................. 32
2.2.2 Synthesis of a BHAPS-functionalized hexagonal MSU-H mesostructu-
re ....................................................................................... 33
2.2.3 Synthesis of BHAPS-functionalized mesocellular silica foam, MSU-F..
......................................................................................... 33
2.3 Physical Characterization ............................................................ 36
2.4 Results and discussion ............................................................... 37
2.4.1 BHAPS-functionalized MSU-H hexagonal structure ..................... 37
2.4.2 BHAPS-functionalized MSU-F foam structure ............................. 46
2.5 Conclusion ................................................................................ 47
2.6 References ............................................................................... 51

vi

3. Chapter 3
Diazo-functionalization of mesoporous silica for potential use in phosphorpeptide

enrichment ......................................................................................... 52
3.1 Introduction .............................................................................. 52
3.2 Experimental section of NHNMPTS functionalized mesostructure........ 58

3.2.1 Reagents ........................................................................... 60
3.2.2 30% NHNMPTS functionalization of SBA-15 silica by grafting reaction
........................................................................................ 60

3.2.3 10% NHNMPTS functionalization of MSU-F silica by direct assembly
method .............................................................................. 61

3.2.4 30% NHNMPTS functionalization of MSU-F silica by grafting method
.......................................................................................................... 62

3.2.5 Diazotization of NHNMPTS mesostructure ................................. 64
3.3 Physical Characterization ............................................................. 67
3.4 Results ..................................................................................... 68
3.5 Conclusion ................................................................................ 76
3.6 References ............................................................................... 88

4. Chapter 4

Improved Diazo-functionalization of mesoporous silica and the preparation of an
IMAC-type mesoporous silica for potential use in phosphopeptide

enrichment ....................................................................................... 89
4.1 Introduction ............................................................................... 89
4.2 Experimental ............................................................................. 94

4.2.1 Reagents ............................................................................ 94
4.2.2 GPTS-, APTS-, and SATS-functionalized MSU-F mesocellular silica

foam by direct assembly ................................................ 95
4.2.3 30% SATS functionalized MSU-F silica by grafting method ........... 98

4.2.4 Diazotization reaction of 30% GPTS-functionalized MSU-F silica... 98
4.2.5 Diazotization reaction of 30% APTS-functionalized MSu-F silica... 100

4.2.6 Absorption of organophosphate ............................................. 100
4.2.7 Preparation of Fe3+ immobilized on diacetate-functionalized MSU-F
....................................................................................... 101
4.3 Physical Characterization ........................................................... 101
4.4 Results and Discussion ............................................................. 103
4.4.1 30% GPTS functionalized MSU-F and its diazotization ............... 103
4.4.2 Diazotization of Aminophenyl functionalized MSU-F silica ........... 109
4.4.3 Immobilization of Fe3+ on a succinate-functionalized MSU-F silica
foam ................................................................................. 119
4.5 Conclusion .............................................................................. 125
4.6 References ............................................................................. 126
5. Chapter5
Phosphorylated protein enrichment based on diazo functionalized mesoporous
silica film ............................................................................................ 127

vii

5.1 Introduction ................................................................................ 127

5.2 Experimental .............................................................................. 129
5.2.1 Reagents ............................................................................. 129
5.2.2 Synthesis of diazo functionalized mesoporous film ....................... 130
5.2.3 Protein digestion ................................................................... 131
5.2.4 Methyl esterification of protein .................................................. 131
5.2.5 Phosphorylated protein enrichment ........................................... 132

5.3 Physical Characterization .............................................................. 132

5.4 Result and Discussion ................................................................. 133
5.4.1 X-ray diffraction analysis ......................................................... 133
5.4.2 Reflectance infrared spectroscopy ............................................ 133
5.4.3 Transmission electron microscopy ............................................ 134
5.4.4 Matrix assist laser desorption ionization mass spectrometry analysis

....................................................................................... 134

5.5 Conclusion .................................................................................. 136

5.6 Reference .................................................................................. 143

viii

LIST OF TABLES

Table 2.1.Reaction stoichiometries used for the supramolecular assembly of
BHAPS-functionalized MSU-H silica ........................................... 35

Table 2.2.Textural properties of BHAPS-functionalized hexagonal MSU-H meso-
structure ............................................................................... 42

Table 2.3.298i solid state NMR cross-linking parameters for BHAPS-functional-
ized MSU-H ............................................................................. 43

Table 2.4.Textural properties of BHAPS-functionalized mesocelluar MSU-F foam
mesostructure ......................................................................... 49

Table 2.5.29Si solid state NMR parameters for BHAPS-functionalized MSU-F...49

Table 3.1.Reaction stoichiometries used for the supramolecular assembly of
NHNMPTS-functionalized SBA-15, MSU-F, and MSU-F silica .......... 63

Table 3.2.The amounts of reagents used in the diazotization of NHNMPTS-
functionalized mesostructured silica ............................................ 66

Table 3.3. UV analysis data from fmoc titration and their deprotected fmoc
concentration by UV analysis .................................................... 75

Table 3.4.Textural properties of NHNMPTS functionalized mesoporous materials
determined from N2 adsorption isotherm ...................................... 82

Table 3.5.lntegral intensity of resonances observed in the 29Si solid state NMR of
NHNMPTS-functionalized mesostructu res ................................ 86

Table 3.6.Degrees of NHNMPTS functionalization determined by solid state
NMR analysis and UV analysis of Fmoc titration ........................... 87

Table 4.1.Reaction stoichiometries used for the syntheses of GPTS-, APTS- and
SATS-functionalized MSU-F foam silica ...................................... 97

Table 4.2.”Si solid state MAS NMR data for 30% GPTS functionaized MSU-F
foam silica. The degree of functionalization is determined by the ratio
between the integral of the T3 resonance intensity to the total integral
intensity (Ta/(Q‘+Qa+T3)) ...................................................... 107

LIST OF FIGURES

Figure 1.1.Regulatory reversible mechanism by phosphorylation and dephos-
phorylatoin process in signaling pathway ..................................... 1

Figure 1.2.Enrichment strategies using B—elimination reaction. (a) B-elimination
reaction under basic condition. (b) Chemical modification based on B-
elimination reaction ................................................................ 12

Figure 1.3.Enrichment strategies using chemical modification of carbodiimide
condensation reaction ............................................................ 13

Figure 1.4. Schematic illustration of phosphorylated peptide isolation by using
dendrimer ............................................................................ 16

Fiture 1.5. Schematic representation of the post-synthesis method of grafing
organic groups on the mesostructure walls ................................. 21

Figure 1.6.Schematic representation of co-condensation pathway of incorpo-
rating organo-fucntional groups onto mesostructure silica .............. 24

Figure 2.1.XRD spectra of BHAPS-functionalized hexagonal MSU-H samples
after soxhlet extraction with ethanol. The product were formed from
reaction mixture in which (A) 10%, (B) 20%, and (C) 30% of the total
silicon ................................................................................. 40

Figure 2.2.N2 adsorption-desorption isotherm of BHAPS-functionalized MSU-H
obtained from reaction mixture containing (A) 10%, (B) 20% and (C)
30% BHAPS. The inserts provide the BJH pore size distribution
obtained from the adsorption branch of the isotherm .................... 41

Figure 2.3.29Si solid state NMR spectra of BHAPS-functionalized MSU-H silica.
The products were formed from reaction mixtures in which BHAPS
represented (A) 10%, (B) 20%, and (C) 30% of the total silicon. The
relative integrals of the 0‘, Q3, and T3 resonances were used to
determine the amount of BHAPS incorporated into the mesostructure
.......................................................................................... 44

Figure 2.4.TEM images of BHAPS-functionalized MSU-H containing 17 mole %
BHAPS in the pore walls. The white circles are intended to identify
the pore openings. ............................................................................ 45

Figure 2.5.N2 adsorption-desorption isotherm and BJH pore size distribution for
7.5 mole % BHAPS-functionalized MSU-F formed from a reaction

mixture containing 10 mole % BHAPS. The pore size distributions
were obtained from both the adsorption and desorption isotherms

.......................................................................................... 48
Figure 2.6.TEM images of BHAPS-functionalized MSU-F containing 7.5% BHA-
PS in the pore walls ................................................................ 50

Figure 3.1.Fluorene derivatives ............................................................... 55

Figure 3.2.Schematic illustration of phosphorylated peptide isolation by reaction
with polymer-immobilized o-diazo groups ................................... 56

Figure 3.3.Schematic illustration of a-diazo groups synthesis on solid support..57
Figure 3.4.lmmobilized NHNMPTS functional group on a silica surface .......... 58

Figure 3.5. Possible low angle X-ray diffractions on hexagonal mesoporous silica
......................................................................................................... 69

Figure 3.6.N2 adsorption—desorption isotherms and BJH pore size distribution for
SBA-15 silica calcined at 500 °C as determined from the adsorption
isotherm ............................................................................... 78

Figure 3.7.Powder X-ray diffraction pattern for calcined SBA-15 silica .......... 79

Figure 3.8.IR spectra obtained after each reaction step for the diazo functional-
ization of 30% NHNMPTS SBA-15 by the silane grafting method.....80

Figure 3.9.N2 adsorption-desorption isotherm for 10% NMNMPTS functionalized
MSU-F prepared by direct assembly and the BJH pore sizes
distribution obtained from the adsorption and desorption isotherms..81

Figure 3.10.N2 adsorption-desorption isotherms for 30% NMNMPTS functionali-
zed MSU-F prepared by grafting method and the BJH pore sizes
distribution obtained from the adsorption and desorption isotherms..83

Figure 3.11.IR spectra obtained after each reaction step in the diazo functional-
ization of 30% NHNMPTS MSU-F by the silane grafting method......84

Figure 3.12.Comparison of N2 adsorption and desorption isotherms for 30%
NHNMPTS functionalized MSU-F, glycine-functionalized MSU-F, and
diazo functionalized MSU-F silica. The isotherms are offset by 500
cm3/g for clarity ...................................................................... 85

Figure 4.1.N2 adsorption—desorption isotherm of 30% GPTS functionalized M-SU-
F prepared by direct assembly method (upper panel) and the cell size

xi

and window size distributions (lower panel) obtained by fitting the
adsorption desorption data, respectively, to the BJH model .......... 106

Figure 4.2.29Si solid state MAS NMR spectrum of 30% GPTS functionalized M-
SU-F foam silica .................................................................. 107

Figure 4.3.IR spectra for the step-wise conversion of 30% GPTS functionalized
MSU-F foam (made by direct synthesis) to diazo functionalized MSU-
F ..................................................................................... 108

Figure 4.4.N2 adsorption-desorption isotherms for 30% aminophenyl functional-
ized MSU-F silica prepared by direct assembly (upper panel) and the
cell and window size distributions obtained from the adsorption and
desorption branches of the isotherms, respectively (lower panel)...111

Figure 4.5.lnfrared spectra for aminophenyl functionalized MSU-F silica and the
corresponding diazotized derivative made by reaction with sodium
nitrite ................................................................................. 114

Figure 4.6.3‘P Solid state NMR spectrum of methyl phosphate covalent bonded
to MSU-F foam silica wall by Sandmeyer-Gattermann aromatic
substitution reaction after hydrochloric acid washing. The two peaks
separated by 4 kHz from the line at -7 ppm are spinning side bands

.............115

Figure 4.7.3‘P Solid state NMR spectrum of methyl phosphate on 30% ami-
nopheyl functionalized MSU-F. The organophosphate anion was
introduced into aminophenyl-functionalied MSU-F. silica with the
cyclohexyl ammonium salt of [P03(OCH3)]2' in methanol solution
........................................................................................ 116

Figure 4.8.3‘P Solid state NMR spectrum of methyl phosphate electrostatically
bonded to cyclohexyl ammonium on pure MSU-F. The
organophosphate anion was introduced into MSU-F silica with the
cyclohexyl ammonium salt of [P03(OCH3)]2' in methanol solution
......................................................................................... 117

Figure 4.9.3‘P Solid state NMR spectrum of methyl phosphate covalent bonded
to MSU-F foam silica wall by Sandmeyer-Gattermann aromatic
substitution reaction. The four peaks separated by 4kHz as shown
above is the spinning side bands of covalently bonded methyl
phosphate species at -7 ppm .................................................. 118

Figure 4.10. N2 adsorption-desorption isotherms for 10% succinic anhydride (SA-
TS) functionalized MSU-F silica prepared by direct assembly method

xii

(upper panel) and the cell and window size distributions obtained form
the adsorption and desorption isotherm, respectively .................. 120

Figure 4.11. N2 adsorption-desorption isotherms for pure MSU-F silica (upper pa-
nel) and the cell and window size distributions obtained from the
adsorption and desorption isotherms, respectively. This silica was
used to prepare a 30% succinic anhydride functionalized derivative by
the grafting method ............................................................... 121

Figure 4.12. IR spectra for succinate-functionalized MSU-F silica prepared by
grafting reaction of SATS and for the corresponding derivative
containing complexed Fe(ll|) cation .......................................... 123

Figure 4.13.TEM image (upper panel) and EDX spectrum (lower panel) for Fe(lll)
succinate functionalized MSU-F silica prepared by grafting method
........................................................................................ 124

Figure 5.1.Powder X-ray diffraction pattern for HMS film on silicon wafer. (Str-
ucture direct agent was removed by soxhlet ethanol extraction.)....137

Figure 5.2.IR spectra obtained in each reaction step for the diazo functionalized
mesoporous silica film synthesis. a) nitrile functionalized mesoporous
silica film, b) carboxylic acid functionalized mesoporous silica film, c)
diazo functionalized mesoporous silica film. Circles indicate the
characteristic vibration of functional group. Each spectrum is offset by
15 ..................................................................................... 138

Figure 5.3.TEM image for organo-functionalized (COOH) mesoporous silica
film .................................................................................... 139

Figure 5.4.MALDI mass spectra of 10 pmole of tryptic digested ovalbumin. a)
conventional MALDI spectra of ovalbumin (no enrichment), b)
ovalbumin enrichment on diazo mesoporous silica film. (The asterisks
indicate the phosphorylated peptide fragments.) ......................... 140

Figure 5.5. MALDI mass spectra of 10 pmole of tryptic digested methyl-esterified
ovalbumin. a) conventional MALDI spectra of methyl esterified
ovalbumin (no enrichment), b) methyl esterified ovalbumin enrichment
on diazo mesoporous silica film. (The asterisks indicate the
phosphorylated peptide fragments.) ......................................... 141

Figure 5.6. MALDI mass spectra of 10 pmole of tryptic digested methyl esterified
B-casein. a) conventional MALDI spectra of B-casein (no enrichment),
b) methyl esterified B-casein enrichment on diazo mesoporous silica
film. (The asterisks indicate the phosphorylated peptide fragments.)
........................................................................................ 142

xiii

LIST OF SCHEMES

Scheme 3.1. Organosilane Grafting pathway .............................................. 54
Scheme 3.2. Direct assembly pathway (co-condensation pathway) ................. 54
Scheme 3.3. The immobilization of Fmoc-glycine on mesostructure ................ 59
Scheme 3.4. Fmoc deprotecting process and analyte for UV analysis ............. 74
Scheme 4.1. Elimination of Glycidyl group by cyclization reaction ................... 91

Scheme 4.2. Possible elimination mechanism of glycidyl group by nucleophilic
cycliczation ........................................................................ 92

Scheme 4.3. Nucleophiles which possibly react with terminal epoxide group....94

xiv

APTS
BET
BHAPS
BJH
Boc
DCM
DHB
DMF
DMSO
EDC
EDX
Fmoc
FT-IR
GPTS
HMS

HONO

IDA

IMAC

IUPAC

ABBREVIATIONS

p-aminophenyltrimethoxysilane
Brunauer-Emmett-Teller
Bis(2-hydroxyethyl)-3-aminopropyltriethoxysilane
Barrett-Joyner-Halenda

t-butyl carbamate

Dichloromethane

Dihydroxybenzoic acid

N, N-Dimethylfonnamide

Dimethyl sulfoxide

Ethyl carbodiimide

Energy—dispersive X-ray spectroscopy
9H-fluoren-9-ylmethoxycarbonyl

Fourier transform Infrared
(3-Glycidoxypropyl)trimethoxysilane

Wormhole mesostructured silica synthesized with amine surfactant
using hydrogen bonding interaction

Nitrous acid

lminodiacetic acid

Immobilized metal affinity chromatography
Immunoprecipitation

International Union of Pure and Applied Chemistry

LC

MALDI
MAS NMR
MCF
MCM-41
Mmole
MOAC
MRI
MSU-F

MSU-H

MSU-X

NHNMPTS
NTA

PEO

P123
PTMs

02

Liquid chromatography

Matrix assisted laser desorption ionization

Magic angle spinning nuclear magnetic resonance

Mesostructured cellular foam

Mobil composition of matter 41

Millimoles

Metal affinity chromatography

Magnetic resonance imaging

Mesostructured cellular foam synthesized with triblock copolymer
surfactant, trimethylbenzene and water soluble silicate at near
neutral assembly conditions

Hexagonal mesostructured silica synthesized with triblock
copolymer surfactant and water soluble silicate at near neutral
assembly conditions

Wormhole mesostructured silica synthesized with triblock
copolymer surfactants and TEOS under neutral (N°l°) assembly
conditions

N-hydroxyethyI-N-methylpropyltriethoxysilane

Nitrilotriacetic acid

polyethylene oxide

Pluonic 123 ((EO)20(PO)7o(EO)20)

Post-translational modifications

lncompletely condensedsilica sites Si(OSi)2(OH)2

T2

TED
TEM
TEOS
TFA
THF
TMB
UV-Vis
XRD

MALDI

lncompletely condensedsilica sites Si(OSi)3(OH)

Completely condensed silica sites Si(OSi)4
3-(Triethoxysilyl)propylsuccinic anhydride

Large pore hexagonal mesostructured silica assembled under high
acid low pH conditions with TEOS as the inorganic precursor and
triblock copolymer surfactant

Functionalized 02 site RSi(OSi)2(OH)

Functionalized 03 site RSi(OSi)3
Tris-(carboxymethyI)-ethylendiamine

transmission electron microscopy

tetraethylorthosilicate

Trifiuoroaceticacid

Tetrahyd rofuran

Trimethylbenzene

Ultraviolet-Visible

X-ray diffraction

Matrix assisted laser desorption ionization

Chapter 1
Introduction

1.1 Research Objects and Significance

Reversible protein phosphorylation is a key regulatory mechanism
involved in major cellular events, like proliferation, differentiation, and apoptosis
through complex signaling processes. As shown below, this regulatory
reversible mechanism is mainly controlled by combining two reactions of different
classes of enzymes: protein kinases, which catalyze the transfer of a phosphate
group to an aminoacid side-chain of proteins and phosphatases which catalyze a
hydrolysis of phosphoester bonds.“3 Therefore, the defects or alterations of
two types of enzymes, protein kinase and phosphatase, might result in serious
diseases such as cancer and neurodegeneration diseases.

Protein "
., - " ATP

  
 

p

   

, Klnase
Phosphatase

. _ _ ADP
j Protein, 2

Cellular events

 

Figure 1.1. Regulatory reversible mechanism by phosphorylation and dephos-

phorylatoin process in signaling pathway

To understand these detailed biological processes and signaling
pathways in a system. requires information regarding the phosphorylated proteins
involved in these processes, and how, where and when these phosphorylation
processes take place. These reversible protein modifications by phosphorylation
on serine, threonine and tyrosine residue occur in at least one-third of all proteins
at any one time in a life cycle. It also is estimated that there are more than
100,000 potential phosphorylation sites in the human proteome.“'5
Unfortunately, however, only a few thousand phosphorylation sites are currently
known, and even fewer are well characterized because of several limitations in
detection. First of all, phosphorylated protein abundances are much lower than
the amount of proteins, because only a small fraction of the proteins within cells
is phosphorylated at a given time during a signaling pathway. Only 1~2% of all
proteins exist in phosphorylated form.6 Moreover, as mentioned, variation in
phosphorylation sites enhances the complexity of phosphorylation patterns.
Secondly, the phosphorylation process is a labile and highly dynamic event
regulated by kinases and phosphatases on a very short timescale.5 Thus, one
of the huge tasks in proteomics studies of signal transduction is the development
and optimization of strategies suitable for the investigation of low-abundant
phorphorylated proteins and sensitive detection of post-translational
modifications (PTMs). Recently, to overcome these limitations, a faster, and
highly sensitive phosphorylated protein analysis method, mass spectrometry
analysis, has been developed in proteomics.7'8 However, the phosphorylated

protein peaks are often suppressed by unphosphorylated proteins in a mass

spectrometry analysis of complex protein mixtures. Therefore, despite mass
spectrometry’s huge improvement in proteomics, there is no routinely available
method that allows the simple and straightfonrvard analysis of phosphorylated
protein in complex protein mixtures.9 Recently, the exceptional physical and
chemical properties of mesostructured metal oxides, such as rigid open
framework structures, very high surface areas, pH and solventstability ranges
exceeding the stability ranges of proteins, and narrow pore size distributions,
have received attention for applications in biomolecule separations. My
research is aimed at developing such novel mesostructured materials to have
high loading capabilities and high selectivities toward phosphorylated proteins.

The specific objectives of this research are as follows:

1. To develop the methodology for the supramolecular assembly of
organofunctional mesostructured molecular sieves having suitably
large pores via the co-condensation of organosilicon and inorganic
silicate precursors.

2. To extend the known synthesis strategies for organofunctionalized
meso-structures and the chemically modify the surface organic
groups for use in phosphorylated protein enrichment.

3. To investigate new types of organofunctional mesostructures having
a high percentage of surface organic groups for phosphorylated

protein enrichment.

1.2 Enrichment method of phosphorylated protein

Phosphoproteins have been analyzed through the use of total protein
lysates. However, due to the small fractions of the total proteins in cellular
lysates. the identification of phosphoproteins relevant to signaling processes from
complex protein mixtures still can be challenging. Therefore, enrichment
strategies are essential to identify the very low abundance of phophoproteins.5
Here, I introduce some enrichment methods which have been developed to purify
phosphoproteins as well as phosphopeptides from complex protein mixtures,
namely, i) Immobilized Metal Affinity Chromatography (IMAC); ii)
Immunoprecipitation (IP); iii) Metal Oxide Affinity Chromatography (MOAC); iv)
Specific chemical modification method. Furthermore, the state-of-the art

enrichment techniques, in combination with nano—science will be also discussed.

1.2.1 Immobilized Metal Affinity Chromatography(lMAC)1°'19

Immobilized metal affinity chromatography is presently the most popular
and frequently utilized enrichment technique, originally introduced by Porath in
1986.1°'11 This IMAC technique is based on the high binding constant of
phosphoserine for Fe3+ (>103).12 The binding involves the electrostatic
interaction of two components, such as a negatively charged phosphate group
and a positively charged metal ion species immobilized onto iminodiacetic acid
(IDA), nitrilotriacetic acid (NTA), or tris-(carboxymethyI)-ethylendiamine(TED)
chelating ligand.13 Various metal ions other than Fe3+ have been investigated

for better selectivity and phosphoprotein recovery. Generally, the metal ions

can be categorized into three kinds of acids, hard, intermediate, and soft.
These characteristics are related to the specific anion bindings governed by the
hard-soft-acid-base rule. Hard metal ions, such as Fe”, and Al“, show better
binding affinity with oxygen. Soft metal ions, such as CU+, and Hg”, show their
preferences for sulfur. lnterrnediate metal ions, such as R”, M”, and Co”,
can be favorably bound to nitrogen, oxygen, and even sulfur. Therefore, the
oxygen atoms on phosphate groups attaching to serine, threonine, and tyrosine
show a high affinity to a hard metal ion such as Fe“, which has been the most
common binding between metal ion and phosphate group in IMAC
phosphoprotein separation.

Another method using Zirconium for enrichment can be described as a
direct method without the need for an elution process. Originally this method14
used a porous silica wafer as a substrate. The porous silica wafers were
modified by a phosphonate group followed by binding of Zr“. The zirconium
phosphonate (ZrP) modified surface had a strong interaction with
phosphopeptides and specifically captured the phosphopeptides from complex
peptide mixtures. In addition, the captured phosphopeptides on the zirconium
phosphonate-modified wafer were directly placed on a MALDI target for further
analysis by MALDI MS.

Ga(|ll) has been also studied in IMAC phosphoprotein separation. In
1999, Tempst15 and a colleague claimed Ga(lll) showed even better selectivity
and elution-off properties than Fe(lll) for phosphoprotein separation. In their

studies, they compared a Fe(lll), Zr(lV), and Ga(lll) IMAC columns, all of which

captured phosphorylated proteins very well. However, in the elution process,
Fe(lll) and Zr(IV) IMAC columns still retained the acidic residues of various
peptides after an extensive washing process. Also, the columns did not
efficiently elude phosphorylated proteins at dilute basic condition. At present,
although this method is most powerful in phosphorylated protein binding, it has
been problematic in its specificity. Due to its electrostatic binding characteristic,
this IMAC method sometimes binds to non-phosphorylated proteins or peptides
having high number of acidic residues, such as aspartate and glutamate, which is
the significant limitation of this method. For this reason, another IMAC

approach has been discussed.“17

In 2002, Ficarro et. al."3 suggested that .
methyl esterification of the C-terminus and of acid residue such as glutamate and
aspartate in protein mixture before the enrichment process can reduce the non-
specific binding in an IMAC application. This specific chemical modification on
carboxylic acid is able to increase the selectivity in phsophorylated protein
enrichement. In this study, more than 200 peptide sequences, and 380
phosphoylation sites are determined by mass spectrometry analysis after the
enrichment of tryptic phosphopeptides by IMAC. However, due to the relatively
strong and restricted reaction conditions, such as the use of concentrate HCI and

dried methanol in an anhydrous reaction environment, esterification reactions

might lead to sample complexity depending on an experimental conditions.

1.2.2 Immunoprecipitation (IP)

lmmunoprecipitaion (IP) is the enrichment method for precipitation of an

antigen by an antibody specific to that antigen. In a phospho—specific antibody,
anti-phosphotyrosine antibodies are commonly used for enriching a tyrosine-
phosphorylated protein from a protein mixture. Although these antibodies have
been relatively effective at enriching low-abundance tyrosine phosphorylated
proteins, there are still no immunopurification protocols having high selectivity for
the phosphotyrosine of peptides."’°'21'22 Moreover, anti-phosphoserine and anti-
phosphothreonine antibodies are not currently available for the enrichment of
proteins containing phosphorylated serine and threonine residue. Therefore,
the immunoprecipitation method is not generally suitable for enriching

phosphoproteins from a complex protein mixture

1.2.3 Metal Oxide/hydroxide Affinity Chromatography (MOAC)

Recently, a new alternative method to IMAC has been reported. This
approach is based on the specific affinity of phosphate group to a metal oxide
surface. Basically, three metal oxide or hydroxide particles have been widely
studied, titania (1102,2328 zrconia (ZrO2),28 and aluminum hydroxide (AI(OH)3).29
Larsen and colleagues” recently reported the use of titania for phosphorylated
peptides enrichment. In their study, dihydroxybenzoic acid (DHB) was used for
the selective enhancement of phosphorylated peptides in competition with non-
phosphorylated peptides on TiO2. Due to the thermodynamically stable bond
between dihydroxybenzoic acid and TiO2. named chelating bidentate bond,
dihydroxynbenzoic acid effectively inhibits the single weak binding of

nonphosphorylated peptides. Dihydroxybenzoic acid doesn’t affect or less

affects the binding of the phosphorylated peptides, but it retards binding of non-
phosphorylated peptides. In the comparison with IMAC, this novel methodology
showed superior results in terms of the selectivity and sensitivity of
phosphorylated peptide binding. In addition, the simplicity of this procedure and
ease of materials handling can expedite the analysis process, which typically
required less than 5 min per sample. Zirconia has been also applied to
phosphopeptides enrichment. Kweon et al.28 first demonstrated the
phosphopetides enrichment by using the ZrO2 microtip syringe. In her study,
she compared the selectivity and sensitivity of the ZrO2 microtip with those of the
TiO2 microtip, which showed similar result to each other in overall performance.
In selectivity, however, singly phosphorylated peptides were better enriched with
ZrO2 microtips. On the other hand, TiO2 microtips favorably enriched multiply
phosphorylated peptides.

In another MOAC study of proteomics, a novel AI(OH)3 based enrichment
method also has been shown to have a good phosphorylated peptide selectivity
from a complex peptide mixture by ligand exchanges of OH groups with
phosphate groups. Wolshin et al.29 reported this method is even more selective
and cost effective than the commercially available IMAC and other

phophoprotein-enrichment kits.

1.2.4 Specific chemical modification methods
To achieve phosphoprotein enrichment, a site-specific modification of the

phosphate moiety in a protein has been applied. To date, two methods have

been reported. The first method uses the chemical modification of the
phosphorylation sites of serine and threonine by j3-elimination30'31 under strong
alkaline conditions, which results in a dehydroalanine and dehydroaminobutyric
acid residue. This unsaturated residue readily reacts with a nucleophile (in this
case ethanedithiol) and subsequently a thiol terminal group can be linked to a
biothin affinity tag or other immobilizing agent, which is illustrated in Figure 1.2.
These biotinylated peptides can be bound to avidin or streptavidin bead and
separated. Through this process, the labile phosphate groups can be also
substituted by various stable marker molecules which give better ionization
efficiency in mass spectrometry analysis. However, there are some drawbacks
to this application. First of all, unprotected cystein and methionine residues can
become involved in unwanted side reactions with the unsaturated
dehydroalanine and dehydroaminobutyric aicd residue. In a Michael-type
addition reaction, the non bonding electrons on surfur groups of cystein and
methionine residue act as a necleophile and can produced side products. To
overcome this problem, the cystein residue of the sample should be oxidized by
performic acid, thereby inactivating it. Secondly, O-Iinked sugar moieties can
generate an additional complication in this enrichment process. This O-linked
sugar moiety might also undergo a B—elimination reaction and change to a
dehydroalanyl residue, which is the same result as the phosphate residue.
Removing this glycosylated peptide prior to the modification reaction can be a
solution for this problem.

The second method is a multi-step derivatizaton method reported by

Zhou et al. in 2001.32 As shown in Figure 1.3, ethyl carbodiimide (EDC) catalyzes
the addition of cystamine to a phosphate group on the protein via
phosphoroamidate-bonds, which can thereby be covalently bound to a solid
glass bead containing an immobilized iodoacetyl group. Elution of
phosphorylated peptides is accomplished by cleavage of phosphoroamidate
bonds by trifluoroaceticacid (TFA). One major advantage of this method is it
can be applied all types of phosphorylated proteins, unlike the first method.
However, the second method has a drawback in that all aminO- and carboxy-
groups must be blocked prior to enrichment to prevent intramolecular and
intermolecular condensation. At present, these two methods are quite novel
and promising techniques for isolating and enriching phosphorylated peptides
from a mixture. However, these methods still have some problems, in that they
need a significant amount of phorphorylated protein or peptide for successful
subsequent mass spectrometry analysis. Also, the selectivities of these
methods are still questionable. Instead of the chemical derivatization of a
protein mixture in solution, solid support has been chemically modified for
isolating phosphorylated peptides.

Tepe et al.33

established a new type of solid support that contains an a-
diazo functional group on the surface. In his approach, the a-diazo functional
group can be easily substituted by a nucleophile such as the phosphate group
of a phosphorylated peptide. This leads to a covalent bond between the solid

support and the phosphate group. Tepe also claimed this method enhances

selectivity, compared with electrostatic IMAC binding. Moreover, elution of the

10

protein can be conducted by simple filtration. However, this method also has
difficulties. In order to confirm the selective binding of phosphorylated peptides
onto a-diazo functional group, the carboxylic acid moiety must be blocked
through esterification reaction. The unprotected carboxylic acid can be a
nucleophile and then attack the d-diazo functional group, which can reduce the

selectivity of this this approach.

11

eqsuedozomm: co..mc_E__mfa :o comma cosmoEuoE .moEwco

3. cozficoo 9me Eva: cozomo. co..mc.E._oIn Am. dozomo. :o..mc.E..an 9.6: 8.335 EoEcotcm NV 059”.

O I 1
ti... 4:. 3..
: j mIOI wlzIm 4 MIWIJIZ
I O Emu O I 5.0 cow—Eco: m. 11%
./\/\/\/\2 w 0 m ...m ...0 :00...
£505 _ a N
WNIUNT—U :_uo_m\/\/\/\/\Z Im ID ID
1000 0 3V
Eon o...b=noc.Emo._u>zon. 0559.52.32... ass—23.6300 octomocnwocn.
O I I O I I
f... o ._ _. Li...
W .m— J. omnm n . .. _ _ swam N .
Iololzl :olwl: Iolwlzl :w
.. AI|| N A|I
nIUIUII «On—n... - mu IO 40%... - mu
Iolmlo... . ...Ol.n..IO...
o e.

O

12

Esconomo. cozmmcovcoo oEEzuoEmo u.o cosmocfioE _mo_Eo..o msma $5.996 .coEcoEm. m... 9:9...

:oszszozzlor$lor$lzleem. J :offozzlolwwlolwwizluom.

 

«:0 N...c
O <E. O
IOIW IO... _m. O"n..IO_._
o eeemelemezuozz
o _.

J ..oezszozzlmiavlorwmlzleem.
:

«:0
_ o
eeumtsloiol .
__ . onelo:
o z _
...mszuozz
.. T__
:oezszozzlo o zleem. ..orw o zl...
kw: . km... crawl. :i$i
N10 :0
“MEES 0.< n.u SEE-«.2553 com. O
m::0:z _ 1051012
ondlo: + ...oldFo:

 

_
8m. zofaozz 8.. o

1.2.5 Previous method applied in nanoscience

During last few decades, nanoscience has been one of the most active
fields in science and technology. Especially, due to their various practical
applications, such as absorbents,34 catalysts,35 heavy metal ion traps,3“‘39
optics,40 sensors41 and etc., nanomaterials have stimulated studies of their
synthesis and properties. Recently, these nanomaterials have been also
applied in biological science, as drug delivery carriers,42 proteolysis reactors,“44
and MRI contrasting agents.45 In addition, because their important physical
properties (99. high surface area) are not owned by bulk materials, these
nanomaterials have been studied as a affinity probe toward a specific target in
the field of bioanalytical chemistry, especially as a selective enrichment medium
of low-abundance phosphopeptides from enzymatic digest products. Here I
introduce some nanomaterials which have been applied to the enrichment
process in phosphoproteomics, namely i) dendrimer, ii) nanoparticle, iii) zeolite,

iv) mesostructure.

1.2.5.1 Dendrimer

Aebersold et al.46 have established a new phosphopeptide enrichment
method by using dendrimer support, which is shown in Figure 1.4. One of the
notable aspects of this approach is that he covalently coupled phosphorylated
peptides to a synthetic polyamine (dendrimer) in a single step by adding ethyl
carbodiimide (EDC) and imidazole. The covalently immobilized

phosphopeptides on dendrimer were separated from nonphosphorylated

14

peptides by simple filtration, recovered via acid hydrolysis, and subsequently
characterized by LC/MS/MS. With this method, Aebersold could identify
phosphorylation sites. He also produced an esterification reaction prior to
binding phosphopeptides to dendrimer, which enhanced the selectivity of this
method. Finally, Aebersold and colleagues applied this method to the study of
tyrosine phosphoproteins in human T cells, which showed the isolation of a total
of 97 tyrosine phosphoproteins and interacting partners and also the
identification and quantification of 75 tyrosine phosphorylation sites as well as 80

serine and threonine sites.

15

es.oE_.ucmu 95: >9 c258. 0252. no.m.>.o..amoca .0 8.6.6:... oszczow v... 2.6.“.

       

 

 

 

 

 

« : I o .m. a.
”zooioI«IzI: «I IT 1 eem
N...o A :o A.
. 5.... $3 . ..eEEEon.
Aw Aw toumEEE... o:_E<
...OImlo: on In... lo...
0 ..
35.5592
o : z o I z
... __ _ . ._ _ _
...oolo oIzI... t ...oio oIzI:
a _ 5......er u _
:« :«
._. ._.
...OIdIo: ...olelo:

16

1.2.5.2 Nanoparticle

To date, microsized metal oxide particles have been used for enrichment
of phosphorylated peptides from a complex mixture. However, in the
enrichment process of low abundant phosphopeptides, relatively low binding
capacities will have low sensitivity toward target species. Therefore, high
surface area materials such as nanoparticles will surely have an advantage for a
potentially higher trapping sensitivity in enrichment process. Here, a new type
of core/shell nanoparticles has been reported. Chen et al. in 200547 introduced
TiO2-coated Fe304 magnetic nanoparticles (ca.50nm) to isolate and identify the
phosphorylated peptides from the peptides mixture. As mentioned above in
MOAC, selectivity of TiO2 toward phosphorylated peptides is quite satisfying.
Moreover, Fe304 magnetic nanoparticles in a core allow phophopeptide-bound
nanoparticles to isolate readily from the mixture solution by applying a magnetic
field. In this paper, Chen et al. claimed the lowest detectable concentration of
phosphopeptides by using this approach is 500pM for 100uL, which is a much

lower detection limit than any other current method.

1.2.5.3 Zeolite

Recently, a porous nanoparticle also has been applied to phosphorylated
peptide enrichment. Zhang et al. in 200448 reported Fe3T-imm0bilized zeolite-
beta nanoparticles (ca.100nm) have been used, for the first time to enrich and
identify phosphopeptides from a tryptic B—casein digested peptide mixture. In

this study, he claimed the zeolites’ large external surface areas make a large

17

quantity of Fe3+ groups on the surface and also produce a sufficient number of
effective bindings with the target species. Moreover, high dispersibility can
easily facilitate the chelation process of phosphorylated peptides. Although this
method still needs further development, it opens up a new possibility for the

enrichment of phosphopeptides.

1.2.5.4 Mesostructure (MGM-41)

The first synthesized Fe3”- immobilized mesostructure has been reported
for the phosphorylated peptides enrichment application by Pan et al. in 2006.49
In this study, he prepared ca. 600nm particle size and 3nm pore size
mesostructure with a Fe3“ modified surface and then it was used as the
adsorbent to phosphorylated peptides in tryptic o-casein and B-casein digests.
After the separation by Fe3“- immobilized MGM-41, the peaks of
phosphorpeptides were enhanced by the reduction of non-phosphorylated
peptides’ peaks, which illustrated Fe3*- immobilized MCM-41 was successfully
applied in proteomics research. However, it still has the certain amount of
nonspecific bindings with peptides mixture as IMAC has. Moreover, given the
small size of pore, I assume the high degree of functionality originated from its
high surface is not totally used for this application, and thus it still needs further

development.

1.3 Organo-functionalized mesostructure

Since mesostructures received attention with their prominent features,

18

such as their high surface area, narrow pore size distribution, and large pore
volume etc, mesostructures have been widely studied in various applications.
They are used extensively as adsorbents,34 ion exchangers,5° heterogeneous

1 and sensory materials.41 Moreover, due to

catalysts for petroleum refining,5
their pH and solvent stability and easily accessible void spaces, they have been
researched as chromatographic separation media.52 Recently, to expand
utilization of mesostructures, like trapping heavy metal ions,3‘5“°’9 and delivering
drugs,42 they have been functionalized with various organic groups. The
specific affinity and reactivity of these mesostructures can be tuned by
appropriate organic moiety on their surfaces. Thiol functionalization is best for

mercury adsorption,36'38

and amine moiety on the surface prefers arsenate
adsorption.”39 Lin et al.42 also reported new type of spherical MCM-41 having
amine functional groups as a drug delivery carrier. In his research, due to
holding drug molecules inside the pores, the amine functional groups are
covalently bound to capping materials by amidaton reaction. To prepare these
materials, generally two chemical approaches are used, which are the post-

synthesis pathway (grafting pathway) and co-condensation pathway (direct-

assembly pathway)

1.3.1 Post-synthesis pathway
Post-synthesis pathway is the functionalization method accomplished by
the condensation reaction between an organosilane and the silanol group on the

silica mesostructure. The first grafting functionalization was reported by Beck et

19

al. by anchoring chlorotrimethylsilane on the mesostructure MCM-41.53 In 1997,
for heavy metal ion binding, hexagonal MOM-41 and worrnhole-like HMS silica
were also functionalized by the grafting of 3—mercaptopropyltrimethoxysilane,
which shows unprecedented high loading capacities for mercury (2.5mmole/g
and 1.5mmolelg respectively).36'38 However there are some drawbacks to the
post-synthesis pathway. First of all, due to the differences in geometrical
conformations between the silanol group on the mesostructured surface and the
hydrolysable alkoxide group in organosilane, the functional groups are not linked
completely to the surface. Therefore, some amount of organosilane groups on
the mesostructure materials can be easily removed through hydrolysis during
uses. In addition, the inability to control the anchoring liquid organic moiety on
the solid mesostructure make inhomogeneous dispersion of organosilane, which
leads to the limitation of functionality by pore blockages. Figure 1.5 shows the

schematic representation of the post-synthesis pathway.

20

OSi

SiO l '
I

NH , . / \ on /"
m“ N" 2 N82$I03 or TEOS S'Oxsro 0" H0 °\ ,OSi

 

Si
H NV‘N‘ '\ H N NH, /
2 QSEWNH 4) o/ OHHZN 22 21-10 B
2 SiO I. ”Li“ N‘r NHZ I
HZN NH2 \Sl-OH HENW’ “It/L. Ho-Si\08i
I) ,9 NH2
, H.N
Surfactant Micelle S'O’\S'/OH 2 NHZ Ho‘saI
' \ / \ .
o\S/OH c?" HO\S_/O 03.
I I
. / ‘0~Si-v0’\
SIC I OSI
SiO

(IDS‘ /
SIO OSI
S

gi,ors'i~o\s,/ urfactant Removal via

 

SiO /0/ \OH on HO/ '\o\ 03' Calcination or Solvent Extraction
\ , l >
SI SI
/ ‘on Ho’ \
f i
Sio- _ -
S\l OH HO §"osr
o o
\ ,OH HO\ I
3.0” t s.
,/S'\o.sli..o/S\I 8.0 (RSI . .
SIO / OSi I OSI
. \fo’Sho- /
$10 /3.\ (I) /Si\
Mesostructure Silica SiO\Sl/O q / HO O\S./OSI
Si I
l \o/ \L HO/ \
$10. I._ I. _-_I.
> S‘I o.Si/ L\ /o S"osa
O /\ SI\ 0
Grafting of (RO)3Si L \S./0 OH I 0\S./
. SiO" '\ OH / l\ .
(L=organtc group) 0\ / 0H C\ /o OSI
Si\ | Si
SIO/ OTSITO I
/ OSl
SiO

Organo-functionalized Mesostructure Silica

Figure 1.5 Schematic representation of the post-synthesis method of grafing

organic groups on the mesostructure walls

21

1.3.2 Co-condensation pathway

The co-condensation pathway is a one-pot synthesis based on the co-
condensation of siloxane and organosilane precursors during the assembly of the
mesostructures, which is illustrated in Figure 1.6. Due to this co-condensation
reaction, organosilane group should be more uniformly distributed on the
mesostructure surface. Moreover, in comparison to the grafting pathway, direct
assembly pathway leads to a more complete crosslinking of the organosilane to
the mesostructure framework. Therefore, it might be a much more promising
pathway to the preparation of stable organo-functionalized mesostructured
molecular sieves. However, one of the problems associate with the co-
condensation pathway is the loss of structural order at high loading percentages,
like 15 mol% or more. Moreover, if the organic moiety is hydrophilic, such as
hydroxyl group, functional group might be buried inside the framework, which
results in low degree of valid functionalization. The first co-condensation
synthesis has been done on MCM-41, reported by Burkett et al.54 In his few
papers, he reported functionalized MGM-41 with various functional groups
including aminophenyl, mercaptopropyl, phenyl and octyl. However, due to the
strong interaction between templating surfactant and framework, in some cases
the functionalized MCM-41 decomposed upon removing the templating
surfactant from the pores by acid. The functionalized mesostrUctures are also
prepared by hydrogen bonding assembly, namely functionalized HMS and MSU—
X. In this non—electrostatic technique, functionalized mesostructures are stable

to the removal of surfactant by extraction techniques. In 2001, Pinnavaia and

22

colleagues reported up to 50mol% thiol functionalized HMS.55 In his research, the
materials retained well expressed mesostructures after removing surfactants with
about 3.0 nm of pore sizes, 0.7 cm3/g of pore volumes, and 1230 mzlg of surface
areas. Recently, in order to incorporate bio-molecules into ordered meso-
structure silica materials, large pore materials like SBA—15 and MCF are very
attractive because a suitable pore size required for internal adsorption of large
molecules.”57 Moreover, the surface modification on silica by direct assembly
also plays an important role in binding protein as this will influence the strength of
the interaction between the biomolecules and the internal surface of the
mesostructure silica. Zhao et al.56 reported functionalized SBA-15 with various
functional groups by co-condensation pathway. In this paper, he prepared 5
different functional groups on the mesostructures, such as amine, nitrile, vinyl,
phenyl and thiol. He also claimed the level of disorder of the mesostructures

depended on the type and amount of the organosilanes in the synthesis.

23

HQN NH2

Nab} 3310‘ NH2
HZNM HZNQS 321%
OSI
SIO I IOSi

Surfactant Micelle g/(O’Sjl‘o\Si/\
R /
SIO\ /0 \OH HO 0\ (Bi

H2

Na2Si03 or TEOS + (RO)3Si L (L=organic group)

  

SI //SI
[\OI'I-iN HzN NHzHO I)
sro- 2N ‘12. ..NNH” I
8(i-R H2 N ”5:35:12 R- S’"OSi
O NH2 0
\ ,OH H2N NH2 Ho\/
SiO’S' Si
Surfactant Removal >Sl\o | o’SI/
I - SiO/ TS" ,
by Solvent Extraction SiOI os.
(IDSI
SiO OS,
\.,0’Sl‘o\ /
’S'\ R /S'\
SIO\S(0 OH HO \i/o\S OS,
I
/ \OH Ho’ \
f i
Sic- ._ _
' S(' R R §"08i
o o
\ ,or-t HO\ /
SiO’SK /Si\
O\ /OH R HO\ /O OSI
Si\ l ,Si
SiO/ O~—Si—ro \ .
I OSI
SiO

Organo-functionalized Mesoporous Silica

Figure 1.6 Schematic representation of co-condensation pathway of

incorporating organo-fucntional groups onto mesostructure silica

24

1.5 References

(1)

(2)

(3)

(4)

(5)

(6)
(7)
(8)

(9)

Venter, J. 0., Adams, M. D., Myer, E. W., Li, P. W., Mural, R. J., Sutton, G.
G., Smith, H. O., Yandell, M., Evans, C. A., Holt, R. A., et al. Science 2001,
291, 1304-1351

Manning, 6., Whyte, D. B., Martinez, R., Hunter, T., Sudarsanam, S.
Scinece 2002, 298, 1912-1934

Alonso, A., Susan, J., Bottini, N., Friedberg, I., Osterman, A., Godzik, A.,
Hunter, T., Dixon, J., Mustelin, T. Cell 2004, 117, 699-711

Zhang, H., Zha, X., Tan, Y., Hombeck, P. V. et al. J. Biol. Chem. 2002, 277,
39379-39387

Kalume, D. E., Molina, H., Pandey, A. Curr. Opin. Chem. Biol. 2003, 7, 64-
69

Schlessinger, J., Harvey Lect. 1993, 89, 105-123
Sickmann, A., Meyer, H. E. Proteomics 2001, 1, 200-206

Mann, M., Ong, S. E., Gronborg, M., Steen, H., Jensen, 0. N., and Pandey,
A. Trends Biotechnol. 20, 261-268

Arnott, D., Gawinowicz, M. A., Grant, R. A., Neubert, T. A., Packman, L. C.,
speicher, K. 0., Stone, K., Turck, C. W., J. Biomol. Tech. 2003, 14, 205-215

(10) Andersson, L., Porath, J. Anal Biochem 1986, 154 250-254

(11) Stensballe, A., Andersen, S., Jensen, 0. N. Proteomics 2001, 1, 207-222

(12) Porath, J., Carlsson, J., Olsson, l., and Belfrage, G. Nature, 1975, 258, 598-

599

(13) Gaberc—Porekar, V., Menart, V. J. Biochem. Biophys. Methods 2001, 49,

335-360

(14) Zhou, H., Xu, 8., Mingliang, Y, Feng, 8., Pan, 0., jiang, X., Li, X., Han, G.,

Fu, Y., Zou, H. J. Proteome. Res. 2006, 5, 2431-2437

(15) Posewitz, M. C., Tempst, P. Anal Chem. 1999, 71, 2883-2892

(16) Ficarro, S. B., McCleland, M. L., Stukenberg, P. T., Burke, D. J. et al. Nat.

Biotechnol. 2002, 20, 301

25

(17) Brill, L. M., Salomon, A. R., Ficarro, S. B., Mukherji, M. et al. Anal. Chem.
2004, 76, 2763-2772

(18) Blacken, G. R., Gelb, M. H., Turecek, F. Anal. Chem. 2006, 78, 6065-6073

(19) Kokubu, M., Ishihama, Y., Sato, T., Nagasu, T., Oda, Y. Anal. Chem. 2005,
77, 5144-5154

(20) Pandey, A., Podtelejnikov, A. V., Blagoev, B., Bustelo, X. R., Mann, M.,
Lodish. H. F. Proc. Natl. Acad. Sci 2000, 97, 179-184

(21) Kratchmarova, I., Blagoev, B., Haack-Sorensen, M., Mann, M., Science
2005, 308, 1472-1477

(22) Zhang, G., Neubert, T. A. Proteomics 2006, 6, 571
(23) Sano, A., Nakamura, H. Anal. Sci. 2004, 20, 565-566

(24) Kimura, Y., Shibasaki, S., Morisato, K., lshizuka, N., Minakuchim H.,
Nakanishi, K., Matsuo, M. et al. Anal. Biochem. 2004, 326, 262-266

(25) Pinkse, M. W., Uitto, P. M., Hilhorst, M. J., Ooms, B., Heck, A. J. R. Anal.
Chem. 2004, 76, 3935-3943

(26) Miyazaki, S., Morisato, K., lshizuka, N., Minakuchi, H., Shintani, Y., Furuno,
M. Nakanishi, K., J. Chromatogr. A2004, 1043, 19-25

(27) Larsen, M. R, Thingholm, T. E., Jensen, O. N., Roepstorff, P., Jorgensen, T.
J. D. Mol Cell Proteomics 2005, 4, 873- 886

(28) Kweon, H. K., Hakansson, K. Anal. Chem. 2006, 78, 1743-1749

(29) Wolshin, F., Wienkoop, S., Weckwerth, W. Proteomics 2005, 5, 4389-4397

(30) Oda, Y., Nagasu, T., Chait, B. T. Nat. Biotechnol. 2001, 19, 379-382

(31) Meyer, H. E., Eiserrnann, B., Heber, M., Hoffmann-Posorske, E., Korte, H.,
Weigt, C., Wegner, A., Hutton, T., DoneIIa-Deana, A., Perich, J. W. Faseb. J.
1993, 7, 776-782

(32) Zhou, H., Watts, J. D., Aebersold, R. Nat. Biotechn012001, 19, 375-378

(33) Lansdell, T. A., Tepe, J. J. Tetrahedron Lett. 2004, 45, 91-93

(34) Davies, M. E. Chemistry & Industry 1992, 137

(35) Clark, J. H., Macquarrie, D. J. Chem. Commun. 1998, 853-860

26

(36) Mericier, L., Pinnavaia, T. J. Adv. Mater. 1997, 9, 500-&

(37) Fryxell, G. E., Liu, J., Hauser, T. A., Nie, Z. M., Ferris, K. F., Mattigod, S.,
Gong, M. L., Hallen, R. T. Chem. Mater. 1999, 11, 2148-2154

(38) Feng, X., Fryxell, G. E., Wang, L. 0., Kim, A. Y., Liu, J., Kemner, K. M.
Science 1997, 276, 923-926

(39) Yoshitake, H., Yokoi, T., Tatsumim T. Chem. Mater. 2002, 14, 4603-4610
(40) Trindade, T., O’Brien, P., Pickett, N. L. Chem. Mater. 2001, 13, 3843-3858

(41) Xu, X., Tian, B., Zhang, S., Kong, J., Zhao, D., Liu, B. Analytica Chimica
Acta 2004, 519, 31-38

(42) Girl, 8., Trewyn, B. G., Stellmaker, M. P., Lin, V. S.-Y. Angew. Chem. Int. Ed.
2005, 5038- 5044

(43) Shur W. Fan, J. Yang, P., Liu, C., Zhai, J., Lei, J., Yan, Y., Zhao, D. ,Chen,
X. Anal Chem. 2006, 78, 4811-4819

(44) Zuo, 0., Yu, W., Zhou, X., Zhao, D., Yang, P. Rapid Commun. Mass.
Spectrom. 2006, 20, 3139-3144

(45) Lee, J., Huh, Y., Jun, Y., Seo, J., Jang, J., Song, H., Kim, S., Cho, E., Yoon,
H., Suh, J., Cheon, J. Nat. Medicine 2006, 13, 95-99

(46) Tao, W. A., Wollscheid, B., O'Brien, R., Eng, J. K., Li, X., Bodenmiller, 3.,
Watts, J. 0., Hood, L., Aebersold, R. Nat. Methods 2005, 2, 591

(47) Chen, C., Chen, Y. Anal. Chem. 2005, 77, 5912-5919

(48) Zhang, Y., Yu, X., Wang, X., Shan, W., Yang, P., Tang, Y. Chem. Comm.
2004, 2882-2883

(49) Pan, C., Ye, M., Liu, Y., Feng, S., Jiang, X., Han, G., Zhu, J., Zou, H. J.
Proteome. Res. 2006, 5, 31 14-3124

(50) Lim, M., Blanford, C. F., Stein, A. Chem. Mater. 1998, 10, 467-470

(51) Corma, A., Martinez, A., Martinez-Sofia, V., Monton, J. B. J. Catal. 1995,
153, 25-31

(52) Boissiere, C., Ktimmel, M., Persin, M., Larbot, A., Prouzet, E. Adv. Funct.
Mater. 2001, 11, 129-135

27

(53) Beck, J. S., Vartuli, J. C., Roth, W. J., Leonowicz, M. E., Kresge, C. T.,
Schmitt, K. O, Chu, C. T. W., Olson, D. H., Sheppartd, E. W., McCullen, S.
B., Higgins, J. B., Schlenker, J. L. J. Am. Chem. Soc. 1992, 114, 10834-
10843

(54) Burkett, S. L., Sims, S. 0., Mann, S. Chem. Commun. 1996, 11, 1367-1368

(55) Mori, Y., Pinnavaia, T. J. Chem. Mater. 2001, 13, 2173-2178

(56) Chong, A. S. M., Zhao, X. S., Kustedjo, A. T., Qiao, S. Z. Micoropor.
Mesopor. Mater. 2004, 72, 33-42

(57) Han, Y., Stucky, G. 0., Butler, A. J. Am. Chem. Soc. 1999, 121, 9897-9898

28

Chapter 2

Direct Assembly of Large Pore Organofunctionalized Silica Mesotructure

from Sodium Silicate and bis(2-hydroxyethyI)-3-aminopropylsilane

2.1 Introduction

In order to extend the materials applications of mesostructures,
organofunctionalized mesostructures containing various organic groups have
been studied. These materials can be tuned for specific use through the
incorporation of appropriate organic moieties, such as thiol, amine, carboxylic
acid, and vinyl. Consequently, organofunctionalized mesostructure has been an
important topic for research in areas, such as catalysts, and heavy metal ion
trappings. Moreover, if we combined the chemical specificity of an organic
group and the large porosity of a mesostructure, it might suggest further
applications, such as large biomolecule separations.

In general, two methods have been used for the synthesis of
organofunctionalized mesostructures, namely, i) grafting of organosilanes onto
the surface silanol groups of a pre-assembled mesostructure; and ii) direct
assembly involving a one-step co-condensation reaction of TEOS
(tetraethylorthosilicate) or sodium silicate with an organosilane.

However, the grafting method has some drawbacks. First, due to the
limited number of surface silanol groups on the walls of the mesopores, the

loading level of organosilanes can be restrained. Second, in a grafting

29

synthesis, controlling the uniformity of organofunctional group distribution within
the mesostructure is very difficult. Third, because of the heterogeneous
distribution of surface silanos, organosilanes cannot be fully cross-linked into the
mesostructure, and thus might lead to the weak linking of the organic groups to
the mesostructure. Finally, the grafting method requires more processing steps
than the direct assembled pathway for organofunctionalized mesostructure
synthesis.

The first direct assembly of a large pore organofunctionalized
mesostructure was accomplished by Stucky et al. in 2000. The hexagonal
structural order was maintained up to 15 mole percent incorporation of
organosilane group. However, because of the strongly acidic synthesis
conditions, the direct assembly pathway of SBA-15 has some drawbacks. For
example, for an amine functionalized SBA-15 synthesis, the structural order was
preserved up to only 5 mole percent functionalization. It was suggested that the
protonated silyl amine under acidic condition could hydrogen bonded with
surface silanol groups of silica and this could lead to the destruction of the
ordered mesostructure. In addition, the functionalization of SBA-15 with an
unstable organic group in acid media, such as t-butyl carbamate (Boo), might be
a problem for the direct assembly pathway. The Boc group has been known as
a good protecting group under basic conditions. However, the Boc group can
be hydrolyzed to an amine group under acid conditions, which can be a limitation
in preparing organofunctionalized SBA-15 by the co-condensation route.

Therefore, we need another pathway to synthesize large-pore

30

organofunctionalized mesostructures, particularly one that can be carried out
under pH-neutral condition pathway. Also, instead of costly TEOS as a silica
source, a cheaper silicon source, like sodium silicate, is desirable for economic
reasons. (Sodium silicate is almost 300 times cheaper than TEOS) Although
sodium silicate is an economically favorable silicon source, only a few studies
have been focused on the synthesis of large pore organofunctionalized
mesostructures by using sodium silicate. The objective of the work described in
this chapter is to the development synthetic techniques for the assembly of large
pore organo-functionalized MSU-H and F silica and to overcome the limitation for
the synthesis of these materials from water-soluble silicate sources.

In 2000, Pinnavaia and co-workers synthesized large pore silica
mesostructures using cost-effective sodium silicate under pH-neutral conditions,
designated MSU-H and MSU-F. The objective of the present work is to use this
procedure to prepare, novel large pore organofunctionalized mesostructures.
For the synthesis of organofunctionalized hexagonal MSU-H ' structures, the
approach is focused on preventing condensation reaction of the organosilane
itself in the presence of the water as a reaction medium. The presence of a
water medium causes the organosilane to precipitate before it anchors to the
mesostructured framework during direct assembly synthesis. To circumvent this
problem, a non-aqueous solvent was used for the synthesis of
organofunctionalized MSU-H.

In previously reported MSU-F mesostructured foam synthesis, an “oil in

water" microemulsion was used as the structure-directing agent. The

31

microemulsion was prepared by mixing of . water as a solvent,
(EO)2o(PO)7o(EO)2o(PIuronic P123) as a surfactant, and 1,3,5-trimethylbenzene
(TMB) as a co-surfactant. However, this aqueous emulsion template causes
precipitation of the organosilanes itself, which results in little or no
functionalization of the mesostructure. To succeed in the preparation of non-
aqueous microemulsion droplets, polar organic solvent formamide has been
used. This solvent has previously been used to form titania foam structures.
On the basis of this approach, bis(2-hydroxyethyl)-3-aminopropyltriethoxysilane
(BHAPS) functionalized derivatives of hexagonal MSU-H and mesocelluar MSU-

F were successfully prepared.

2.2 Experimental
2.2.1 Reagents

The non-ionic surfactant, Pluonic 123 ((EO)2o(PO)7o(EO)2o), was obtained
from BASF for the synthesis of large pore functionalized MSU-H and MSU-F
derivatives. As silicate sources, bis(2-hydroxyethyI)-3-aminopropyltriethoxysiI-
ane (BHAPS) was purchased from Gelest Inc., and sodium silicate solutions
(NaOH 14%, SiO2 27%) was obtained from Aldrich. Glacial acetic acid and
formamide were purchased from Spectrum. Absolute ethanol was purchased
in-house. All the above reagents were used without further purification. Water
used for the hydrolysis reaction was purified by a double-exchanged Millipore

filter apparatus to remove cations and anions.

32

2.2.2. Synthesis of a BHAPS-functionalized hexagonal MSU-H meso-
structure

The functionalized MSU-H structure was prepared by a direct assembly
pathway from bis(2-hydroxyethyI)-3-aminopropyltriethoxysilane, along with cost
effective sodium silicate. In order to prevent self-condensation reaction of the
organosilane in the following step, ethanol was used as a solvent instead of
water. Pluronic 123 (0.809, 0.013mmole) was dissolved in glacial acetic acid
(0.69, 10mmole) and ethanol (1.3~3.5 g, 28~76 mmole). (Then, BHAPS
(0.605~1.85 g, 1.21~3.64 mmole) was added to the mixture at an ambient
temperature. After the addition of aqueous sodium silicate (2.43 g (11.1 mmole)
~ 1.89 g (8.63 mmole) sodium silicate and 30 9 H2O (1.7mole)), the mixture was
stirred at 60 °C for 1 day. The surfactant was then removed from the air-dried
product by soxhlet extraction with ethanol. The overall reaction stoichiometry

for obtaining the desired derivatives was in Table 2.1.

(1-x) SiO2 : X BHAPS : 0.011 P123 : 0.81 Glacial acetic acid : 2.3~6.2 Ethanol :
138 H2O, where x=0.1 to 0.3

2.2.3 Synthesis of BHAPS-functionalized mesocelluar silica foam, MSU-F
The synthetic procedure used to prepare organofunctionalized MSU-F

foam structures was similar to the one used to prepare MSU-H. In order to

prepare the microemulsion template without hydrolysis while allowing for

condensation reaction of the organosilane, formamide was used as a solvent.

33

For the preparation of MSU-F, 0.6 g (5.0 mmole) of TMB (1, 3, 5 — trimethyl-
benzene) was added to the surfactant solution which is prepared by mixing
Pluronic 123 (0.8 g, 0.013 mmole), glacial acetic acid (0.69, 10mmole), and
formamide (11.34 g, 252 mmole). Then a microemulsion template forming,
BHAPS (0.605 g, 1.21 mmole) was added to the previous solution mixture,
followed by the addition of sodium silicate solution (2.43 g (11.1 mmole) sodium
silicate and 30 g (1.7 mole) H2O). The reaction mixture was allow to age at
25°C for 1 day followed by one day at 100 °C. The procedure for removing the
surfactant was the same as MSU-H. The molar compositions of the reaction

mixtures used to prepare the MSU-F derivatives were as follows

(1-x) SiO2 : X BHAPS : 0.011 P123 : 0.405 TMB : 0.81 Glacial acetic acid : 10~20

Forrnamide : 138 H2O, where x=0.1

34

”:22 ms... 9% 28 _ms .3 855.2er

 

 

 

 

 

2.6 so... an 2. 2.6 36 mm.” x. on
...o no... mu 3 2.6 «in and .3. cu
mod so... an 2. ”rd 3... v.2. ex. 2.
anon—99:.
EOE. .e.oEE. .e.oEEv GEES. EoEE. Ao_oEE. 23...: Eczema”. c.
.2... E mm<zm
.855 .2593. 35... an ..n. mn.<...m 00m... 9.5.5 ..o .x. 0.2...
be 5.82“.

 

 

 

 

 

 

 

 

 

mo___m 1-20.2 oon..mco..oc2.wn.<...m .o >5Eommm .m_:oo.oEm.a:m e... .8 com: we...eEoEo_o.m cozomom .FN .2an

35

2.3 Physical Characterization

The mesostructured BHAPS-functionalized hexagonal MSU-H and foam-
like MSU-F were characterized by powder x-ray diffraction (XRD) analysis,
nitrogen adsorption-desorption measurement, 29Si solid state magic angle
spinning nuclear magnetic resonance (MAS NMR) spectroscopy, and
transmission electron microscopy (T EM).

Low angle X-ray diffraction patterns were obtained by a Rigaku Rotaflex
Diffractometer using CuKa radiation (l.=1.542 A). Nitrogen adsorption-
desorption isotherms were taken at -196 °C on a Micromeritics Tristar 3000
sorptometer. The samples were outgassed at 100°C under 1045 torr for about
12 hours prior to analysis. Surface area was calculated from the BET plot
according to IUPAC recommendations. The Barret-Joyner-Halenda (BJH)
model was used to derive pore size distribution from adsorption branch of the
isotherms. TEM images were obtained on a JOEL 2200FS instrument with an
accelerating voltage of 200 kV. TEM samples were prepared by sonicatiing
mesostructured products in ethanol for 30 minutes, which is followed by droping
of the suspension onto carbon coated copper grid. 2S’Si solid state MAS NMR
spectra were taken on a Varian 400 solid state NMR spectrometer at a 4 kHz

spinning frequency with 400 seconds pulse delay.

36

2.4 Results and discussion
2.4.1 BHAPS - functionalized MSU-H
Powder X-ray diffraction analysis

Figure 2.1 illustrates the low angle XRD reflections for three different mole
BHAPS-functionalized MSU-H mesostructures formed from reaction mixture in
which 10, 20, 30% of the total silicon in the reaction mixture was in the form of
BHAPS. The products were assembled from sodium silicate in the presence of
a nonionic structure directing agent (P123) under near-neutral pH conditions
(pH~6 or 6.5) and at different levels of BHAPS concentration. The three
functionalized products exhibit (100), (110), (200) reflections consistent with
hexagonal framework order. The XRD patterns are consistent with previously
reported results for hexagonal MSU-H or SBA-15 structure types. Long-range
hexagonal order is retained up to 30 mole percent BHAPS in the reaction
mixture, as indicated by the presence of higher order duo and d2oo reflections, in

addition to the d100 peak in the XRD patterns.

N2 adsorption desorption isotherm

N2 adsorption-desorption isotherms for the organofunctionalized MSU-H
products formed from reaction mixture containing 10%, 20% and 30% BHAPS
are shown in Figure 2.2. All three products show the well-expressed adsorption
steps at a partial pressure of about 0.8, which confirms the presence of large
uniform framework pores for each of these mesostructures. In addition, the

shapes of adsorption and desorption hysteresis loops between partial pressure of

37

0.6 to 0.8 indicate the presence of well-ordered pores. Table 2.2 summarizes
the physical properties obtained from the N2 isotherms. The surface areas
range from 650 ~ 720 m2/g. Pore volumes are in the range of 0.8 ~1.4 cm3lg,

and pore size varied from 7.1 ~ 8.7 nm, respectively.

29Si solid state MAS NMR spectroscopy

Figure 2.3 shows representative 29Si MAS NMR spectra for the presence
of BHAPS functionalized MSU-H silica. Resonances around -110 ppm and -102
ppm correspond to fully cross-linked Q4 silicon sites and to 03 sites containing
one OH terminal site group. The third peak around -70 ppm is assigned to T3
LSiO3 centers where L is BHAPS. Also, 29Si MAS NMR allows us to quantify
the level of organosilane functionalization for each reaction product. The
increase in the T3 band intensity in Figure 3 is correlated with the amount of the
organosilane incorporated into the mesostructure. Table 2.3 provides the
degree of functionalization obtained by 29Si solid state MAS NMR analysis. The
degree of functionalization achieved was less the level theoretically expected on
the basis of the initial reaction mixture. It is assumed that some of hydroxyl
group of the BHAPS organosilanes are hydrogen-bonded to PEO groups of the
structure directing agents and are extracted when the structure directing agents
are removed by soxhlet extraction. 29Si MAS NMR analysis shows that
approximately half of the initial amounts of organosilane in the reaction mixture is

incorporated directly into the assembled mesostructures.

38

Transmission Electron microscopy (TEM)

Figure 2.4 provides the TEM micrograph for the 17% BHAPS-
functionalized MSU-H product obtained from the reaction mixture containing 30%
BHAPS. Even though there is a high degree of functionalization, the product
exhibits a well ordered hexagonal structure. In addition, the pore size estimated
from the image corresponds to the value obtained from the N2 isotherm analysis.
In the image of Figure 2.4 the dark and light contrast regions represent the pore

and walls of the mesostructure, respectively.

39

.82.... as. o... .o <98 6. can seem .m. as... 2. 5...; c. 23...... coeooo. 52. noes. so;

.260... 0...... .9555 5.2. 5.8996 5.58 .ocm mo.aEmm 1.305. .2398... emu..mco_.oc2.wn.<1m “.o 980% omx .. PM 9:9“.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

@N @N on
u 3 F a." a 3 3 I 3 p ...e u 3 p
M I
z L
m u w. t o o l
o 0 If u
0 W
u
s
..M.
(on can <3 can (on can ..W
<3 2.. <3 2. <3 3. W
<2. 2: <2. 2: <9. 2: .u
2... e... 2... a... 2.... e... m
m
-.\ .m. n
o W m < 0

 

 

 

4O

AS 9.58:8 233:. 5.68. EB. 35990 rims. 3N..mco_.o:2.mm<rm .0 £550.... co..&ommu.co.§omum

.8}; 9532. 3.5.3.
H md md ed ~.o

IDDDbI

IDFDDbbDDII

6%... 23mm... m>_.m.m~.

 

nu

 

 

 

 

 

 

 

 

o . m.. m.. «A. .wo o
bblo >b>h>btbP>$bb+Pbbpro
8. A m .8.
e
.oo~ mw .oo~
m
.oom 5 team
W
too. _ .ooe
Mm
room 1‘ room
com com

H

 

.mn.<_._m.§m 8. .25 $8 a. ...xa.

Nz Na 2:9...

3a.}: 8332.. m>..m_mm
md md ed Nd o

p?hpp>».uanbLnrb#?Po

4‘

1com A

r cow

 

 

41

.QOdHOntQ um DmEthumv mE:_O> 0.0m 50505 pwmv ..0:0.—.ufi$EEm¢3mmcam Or: >0 Bum—30300

u

.395... v.3 9.. new @50QO 86 c353 3:295“. 0... Eat 35390.. .39.. :3 seem... 5.3.83 9.. 80.. 858.2QO

 

 

 

 

Nad so QM v.8 0.3. .x. O»
N06 ONm QN Nd 9:. o\e ON
mné hrs MN Rd N.—.—. o\o OF
5.9....3 oat-5 nAEE «.55 A55 9532 gateway. :.
2.5.0.. 93.... no.2 coat-5 305.25 :55 .8053“. 20.— 9:23» u 9.33 x. 20:

 

 

 

 

 

 

 

 

6.202.303... rims. _mcommxm... um~._mco..oc2.wn.<rm .0 35.2.0... 5.3.8.. .~.~ 03m...

42

..fitotaK. 3.2.2:. 2.5.2 52. 8.53.8...

 

 

 

 

 

N.» #2. m...~ ado" as: .\. on
in $2. ...3 9; En» ..\. on
Q. .x.» m5 mfic «NS. ..\. 3
oceans—acetone 2:35 5.331 :.
«Qatva n... «G <0
.0 0250a 9.5.3 .x. 30:

 

 

 

 

 

 

 

rims. omuacoaocaénzrm 6. 29393 9:57.83 «=22 29m 38 _me .3 man.

43

4
Q A Q4 B

Q3

T3

 

 

Chemical Shift (ppm) Chemical Shift (ppm)

 

Q4 C
03
T3
Chemical Shift (ppm)

Figure 2.3. 298i solid state NMR spectra of BHAPS-functionalized MSU-H
silica. The products were formed from reaction mixtures in which BHAPS
represented (A) 10%, (B) 20%, and (C) 30% of the total silicon. The
relative integrals of the Q4, Q3, and T3 resonances were used to determine
the amount of BHAPS incorporated into the mesostructure.

44

 

Figure 2.4. TEM images of BHAPS-functionalized MSU-H containing 17 mole %
BHAPS in the pore walls. The white circles are intended to identify the pore
openings.

45

2.4.2 BHAPS- functionalized MSU-F foam structures
N2 isotherms

Figure 2.5 shows the N2 adsorption-desorption isotherms for the
mesocelluar foam structure formed from a reaction mixture in which 10% of the
silicon is in the form of BHAPS. The well-expressed adsorption step at relative
pressure above 0.9 reflects the very large cell size of this material, which is about
49 nm. In addition, the derivative contains a secondary pore size centered near
14 nm. From the desorption loop, we could also calculate the window size,
which is about 11 nm. The ratio of cell to window size for the primary pore
structure is 4.45, corresponding to a “closed cell” foam structure. The
secondary pore structure has a cell to window size of 1.3 correspond to an open
cell structure. The summary of physical properties obtained from the N2
isotherms for the mesocelluar BHAPS functionalized MSU-F foam structure is in
Table 2.4. The surface area is 385 mzlg and pore volume- is 2.20 cm3/g,

respectively.

29Si solid state MAS NMR

Table 2.5 provides the ”Si solid state MAS NMR 'for a BHAPS-
functionalized mesocelluar MSU-F foam formed from a reaction mixture in which
10% of the silicon was in the form of BHAPS. From this 29Si solid state MAS
NMR analysis, we can quantify the amount of organic moieties on
mesostructures. Based on these data, we find that 7.5% BHAPS is

incorportated onto the mesocelluar foam walls. Thus, about 75% of the

46

BHAPS available in the reaction mixture in incorporated into the final reaction

product.

TEM (Transmission electron microscope)

Figure 2.6 shows the TEM micrograph of 7.5% of a BHAPS functionalized
MSU-F foam. The image is typical of a mesocellular foam structure. The
imaged pores have a diameter of about 50 nm, which agrees with the large pore

size found by nitrogen adsorption.

2.5 Conclusion

By using a non-aqueous reaction medium for direct supramolecular
assembly, we successfully synthesized large pore (7 to 50 nm) and well ordered
BHAPS-functionalized hexagonal and celluar foam mesostructures, denoted
MSU-H and MSU-F, respectively. In the case of BHAPS-functionalized
hexagonal MSU-H the use of ethanol in place of water prevents self-
condensation reaction of the organosilane. Up to 17 mole percent BHAPS
could be incorporated into the pore walls of the mesostructure with approximately
50% efficiency. In the organofunctionalized mesocelluar MSU-F synthesis, a
non-aqueous microemulsion template was formed using formamide as a solvent,
(EO)20(PO)70(EO)20 (Pluronic P123) as a surfactant, and 1,3,5—trimethylbenzene
(TMB) as a co-surfactant. The resulting microemulsion was highly effective in
producing a mesocellular foam containing 7.5 mole percent BHAPS in the pore

walls with 75% efficiency.

47

$2.553. 5.3.080 ucn coneomum

o... 58 EB. 3:350 295 9.0.59.6... mum 0.0.. on... .wn.<_..m .x. o.oE o. 9.59:3 2:3... corona. m ES. Becca.

.-nms. oo~__oeo_.o§..wo<:m s as... m.» .o. 8.59%... one so. ram use 5.2.8. ..o_a.oaoo.co_§oaoa N2 .3 2:9.

:5: 3.2.5.0 Ben. A0n=n= 2.53.5 diam—om

 

 

 

 

 

 

 

 

 

 

 

 

 

..m; 8. o. m . a... a... a... N... o A
Ma 01. o m s . - p s n n r t p . s t . . . . . . ni— ml
I m . . n
( u o n
d «.9 . «o o w W
n e . a V s
.m .m . mod p. P O
.m. o «6.. P m J
r. S n . .. S I. q
o d . m d 6
w a #6.. a. nu." 0-
D m ” mad W W \W
..m E... v 8... 1w W...
V . .. E: . we 6
D m on av B o .W 9
d . .
w .. a. 8... m u
d It It

48

.mgmfieo. .3 85328.. ..n.+m0+eo..n. .3 85328..

 

fimfi

a. ..N

Nam 53

$3.

 

 

n0.n0+<0

 

«5.35.3285. ..o 3.300

 

n0 ..0

 

 

 

958...: .5308.— :. awn—(rm .x. 0.05.

 

 

..-sw: oo~__oeo_.oca-wn_<zm 5. 222.23 «:22 Boa 28 _ms .m.~ ass.

um:.c..o.ou oE:_o> Eon...v .850... Cum. .o__o......oEEm-.:omc:.m o... .3 85.32.50 ..ouoE :3 30cm... 5.3.88 9.. E9. 325.200

.mmduomi .m

ad

 

 

 

 

 

 

 

.3 m8 .. a. .2. so.
oars... oars. age. «.5...
2:25 8.88. c. mo<xm .\. 2o:
oE=.o> 20... no; 385m on; Sou—.5. 3.2.5... Eon.

 

 

6.303.389: Ema. ".592 83:839.. uoN._mco..oca.mn.<Im .0 $3.30... .93on .vN 03....

49

.m__m; 28 9: c_ 92%

 

$3 359:8 “Sm: Buacozocefiaém 3 835 5m: .3 2:9...

 

50

2.6 Reference

(1) Chong, A. S. M., Zhao, X. S., Kustedjo Angeline T., Qia, 8. Z. Microporous
Mesoporous Mater. 2004, 72, 33-42

(2) Fowler, C. E., Burkett, S. L., Mann, S. Chem. Commun. 1997, 1769-1770

(3) Huh, 8., Wiench, J. W., Yoo, J.-C., Pruski, M., Lin, V. S. Y. Chem. Mater.
2003, 15, 4247-4256

(4) Carvalho, W. A., Wallau, M., Schuchardt, U. J. MOI. Catal. A 1999, 144, 91-99

(5) Corma, A., Martinez, A., Martinez-Sofia, V., Monton, J. B. J. Catal. 1995, 153,
25-31

(6) Beck, J. S., Vartuli, J. C., Roth, W. J., Leonowicz, M. E., Kresge, C. T.,
Schmitt, K. D., Chu, C. T. W., Olson, D. H., Sheppartd, E. W., McCullen, S. B.,
Higgins, J. B., Schlenker, J. L. J. Am. Chem. Soc. 1992, 114, 10834- 10843

(7) Burkett, S. L., Sims, S. 0., Mann, S. Chem. Commun. 1996, 11, 1367-1368

(8) Margolese, D., Melero, J. A., Chritiansen, S. C., Chmelka, B. F., Stucky, G. D.
Chem. Matter. 2000, 12, 2448-2459

(9) Chong, A. s. M, Zhao, x. 3., J. Phys. Chem. B 2003, 107, 12650-12657

(10) Kim, S. S., Pauly, T. R., Pinnavaia, T. J. Chem. Commun. 2000, 17, 1661-
1662

(11) Kim, S. S., Karkamkar, A., Pinnavaia, T. J. Phys. Chem. B 2001, 105, 32,
7663-7670

51

Chapter 3

Dino-functionalization of mesoporous silica for potential use in

phosphopeptide enrichement

3.1 . Introduction

As described in chapter 1, many enrichment methods, particularly
immobilized metal affinity chromatography (IMAC)1, selective adsorption on metal
oxide column (T i02, Al203, and Zr02)2'3", and immunoprecipitation of
phosphoprotein by specific antibodies5, have been used to enhance the
separation and identification of phosphorylated peptides. Among these
strategies, IMAC is the most widely used technique as an enrichment method.
However, nonspecific binding between the acidic carboxylate moieties in
nonphosphorylated peptides and the metal cations on the IMAC column often
results in low specificity and sensitivity for target phosphorylated peptides.

Recently, Tepe et al.6 invented a new type of polymer supported reagent
containing o-diazo groups for phosphoprotein enrichment. This method is based
on the reaction of surface immobilized a-diazo groups with the phosphate group
of peptides. The resulting covalent binding between the polymer immobilized a-
diazo group and the phosphorylated peptide results in improved selectivity
compared with the electrostatic binding mechanism of IMAC. This general
reaction scheme between immobilized a-diazo groups and phosphorylated

peptides is illustrated in Figure 3.2. However, this method has a drawback.

52

Polymer resins are generally densely entwined and lacking in porosity and
surface area. In order to improve reactions with immobilized organic groups of
the polymer, the polymer should be swellable. However, in order to achieve the
swollen state of the polymer during reaction, only certain solvents can be used.
This limits the usefulness of polymer matricies. In order to better implement the
Tepe’s method for phosphoprotein enrichment, there is a need for
orgnofunctionalized rigid framework solids having a high surface area and
porosity.

During the past 15 years, mesostructured forms of silica have drawn
attention in the catalyst area because of their large surface areas and uniform
pore sizes.7 Furthermore, their surface silanol groups can be used to link metal
oxides or organic groups on the pore surfaces for catalytic purposes. These
same attractive physical and chemical properties of mesoporous silica are also a
great attraction to overcome the weakness of polymer resins for improving
phosphorylated peptide enrichment. In order to apply Tepe’s method to
mesostructured silicas, surface modifications by organo silanes is needed. As
described previously in chapter 1 and 2, there are two approaches to the surface
modification of mesostructured silica, namely the grafting method and the direct
assmebly method. In the grafting methods, specific organic groups are
introduced by condensation reaction of an organosilane with the silanol groups of
pre-assembled mesoporous silica. Surface modification by the direct assembly
pathway9 is achieved by condensation reaction of an organosilane with silica

precursors during supramolecular assembly process. The grafting and direct

53

assembly pathways are represented in the following schemes (Scheme 3.1 and

3.2):

Scheme 3.1. Organosilane Grafting pathway

9H
2i'OH + LSi(OR)3 ——) :i'O'Sfi-L

f I 0..

Silica surface Silica surface

where L is a hydroxyl-functional organo group.

Scheme 3.2. Direct assembly pathway (co-condensation pathway)

Silicate source + LSi(OR)3 + Surfactant
(sodium silicate, TEOS, etc.)

——-—> ———> {Si07_]1.,.[LSiO1,fi]x - mesostructure

where L is a hydroxyl-funtional organo group, and x is the degree of

functionalizatoin

After modification of the mesostructured silica surfaces by organosilanes,
further reactions are needed to achieve phosphorylated protein enrichment by
the Tepe’s method.6 As the first step toward diazo functionalizaton, fmoc-glycin is
coupled to hydroxide-functionalized mesostructure through the use of 2,6-

dichloro benzoyl chloride. 2,6-Dichloro benzoyl chloride is a good coupling

54

reagent to produce the anhydride form with fmoc-glycine, which facilitates ester
bond formation between hydroxyl group on mesostructures and the fmoc-glycine.
After formation of the fmoc-glycine ester on the mesostructure surfaces, a
deprotection reaction of fmoc group is carried out on the fmoc group. As shown
below, fluorene is a polycyclic hydrocarbon having an ionizable proton at the 9
position with a pKa of 22.6 in DMSO. Proton dissociation is facilitated by the
delocalization of electron over the aromatic rings. Therefore, under base
conditions, as in the presence of piperidine in DMF solvent, fmoc-glycine
modified mesostructures dissociate to form 9-methylene—9H-fluorene, 002, and
amine groups on mesostructure as shown in Figure 3.3. In addition, the 9-
methylene-QH-fluorene generated in the reaction can be quantified by UV
analysis, which allows determination of the amount of glycine ester immobilized

on the silica surface.10

8 9 1 CH3
70.02 0.0 0.0
6 5 4 3

Fluorene 9-methyl-9H-fluorene 9-methylene-9H-fluorene

Figure 3.1. Fluorene derivatives

In the last step, HONO (nitrous acid) gas is used as the nitrosating agents
for diazotization. When sodium nitrite solutions are acidified by sulfuric acid,

nitrous acid is generated, which reacts with amine residues grafted to the

55

.8305

39.6.5 umN___noEE_¢oE>_oa 53> couomo. >9 c258. 0233 oofiEocamoca .0 20:27.2: osmEmcom N...” 659....

 

 

m a,
’0 w
IOIMIO: T ...Oldlo .5523. A
m ...of: . S: . m /=\

O

 

+ 2m: .ofizod Al ..I u ...l
...Olklo. +|/=\ m... Hal/=\._oE>.on. Alv .2la43820n
o

W O _ O
IOInIO...

m l. L

 

 

56

.toaazm Eom co £322? 332.. o~m_u.o .o con—262.. oszocow .m.m 2:9“.

occ.o:c.:o.o:o.>5ofi.m

Nout. 0.0
.52 . VII

/
z/|I\l. m \|\ W “:3 352.35
I ominous: .u «:2
«1+ ozuz .. \/=\o
. H

fo

 

Wmmmm

eo_._o_u.o.~ P..._

o z/|\L.M Aflsozvooeu o,z/Il\l.m
0. our... .5 W
O

W 02.5.5 33:2. W

 

57

mesostructure and generates o-diazo groups which are needed as functional
groups for phosphorylated peptide enrichement applications.

In this chapter, I will describe the synthesis of large pore silica
mesostructures having N-hydroxyethyl-N methyl amino groups as the precursors
to o-diazo group functionalized mesostructures. I also will propose a new
diazotization synthesis for large pore silica mesostructures for improved

specificity and efficiency in the enrichement of phosphopeptides.

3.2 Experimental section of NHNMPTS functionalized mesostructure

The diazotization reactions on the silica surfaces were carried out on three
different samples prepared either by the direct synthesis method ‘or by the silane
grafting method. The initial immobilized functional mesostructures contained
30% N-hydroxyethyl-N-methylamino—propylsilyl groups (NHNMPS) immobilized
on hexagonal SBA-15 silica by grafting reaction, a 10% NHNMPTS
functionalized foam-like MSU-F silica made by direct assembly, and a 30%
NHNMPTS functionalized foam-like MSU-F silica made by silane grafting

reaction. The surface immobilized NHNMPTS group is illustrated in Figure 3.4.

iflNf—PH
‘CH3

/

Silica surface

Figure 3.4. Immobilized NHNMPTS functional group on a silica surface

58

This functional group has a hydroxyl group at its terminal position. The hydroxyl
group is a good nucleophile, which can react with the anhydride form of fmoc-
glycine derivative to form the immobilized fmoc-glycine ester. The anhydride
form of fmoc-glycine was prepared by in situ reaction of fmoc-glycine with 2,6-

dichlorobenzoylchloride as illustrated in the following scheme.

Scheme 3.3. The immobilization of Fmoc-glycine on mesostructure

2,6 dichlorobenzoyl chloride

0
0' “TM 0

CI

OLN Togbé -NOH—/_\/—/
3 /_/° NH):

§,_/_\N

CH3 NHNMPTS functionalized

mesostructure

W9

59

3.2.1. Reagents

The non-ionic surfactant, Pluonic 123 ((EO)20(PO)70(EO)20), was obtained
from BASF for the preparation of large pore SBA-15 and MSU-F mesopoorous
silica. As an organosilicate source, N-hydroxyethyl-N-methylpropyl triethoxy-
silane (NHNMPTS) was purchased from Gelest Inc. The silicate sources,
tetraethyl orthosilicate (T EOS, reagent grade 98%) and sodium silicate solution
(NaOH 14%, SiOz 27%), were purchased from Aldrich. Glacial acetic acid,
hydrochloric acid and formamide were obtained from Spectrum. Absolute
ethanol was obtained in-house. Water used in the synthesis was obtained from
a double-exchanged Millipore filter apparatus. For the diazotization synthesis,
Fmoc-glycine was purchased from Fluka. 2,6-dichlorobenzoyl chloride,
piperidine, and dried N,N-dimethyl formamide (DMF) were purchased from Alrich.
Pyridine was purchased from Jade scientific. Dichloromethane was also
purchased from Jade scientific, and was dried for the reaction. Sodium nitrite
was obtained from Spectrum. All the above chemicals except dichloromethane

were used without further purification.

3.2.2. 30% NHNMPTS functionalization of SBA-15 silica by grafting reaction.

The NHNMPTS functionalized SBA-15 mesoporous silica was prepared
by a grafting pathway from N-hydroxyethyl-N-methylpropyltriethoxysliane.
Pluronic 123 (4 g, 0.7 mmole) was dissolved in 2 M hydrochloric acid solution

(120 g, 0.24 mole) and water (30 g, 1.7 mole). Then, tetraethyl orhtosilicate

60

(8.50 g, 0.04 mole) was added to the mixture with stirring at ambient
temperature. After the addition of TEOS, the mixture was stirred at 40°C for 20
hours followed by aging in an oven at 100 °C for 2 days. The surfactant was
then removed by calcination at 500 °C for 4 hours. During calcination the
temperature was heated to 500 °C at a rate of 2 °Clmin. After the calcinations, it
was cooled to room temperature in 4 °Clmin decrements. The overall reaction
stoichiometries used for the preparation of the SBA-15 mesoporous silica is
given in Table 3.1. After pure SBA-15 silica was prepared, 0.5 g of pure SBA-15
mesoporous silica (7.7 mmol) was mixed in 15 mL tolune solvent with NHNMPTS
(1.05 g, 3.3 mmol) for 3 hours at 25 °C and 4 hours at 110 °C for reflux. The
30% NHNMPTS functionalized SBA-15 material was then filtrated, washed with

ethanol and dried at 25 °C.

3.2.3. 10% NHNMPTS functionalization of MSU-F silica by direct assmbly

method.

The direct assembly synthesis of NMNMPTS functionalized MSU-F silica
was prepared by same procedure as described on chapter 2. First of all, the
surfactant solution was prepared by mixing Pluronic 123 (0.8 g, 0.13 mmol),
glacial acetic acid (0.6 g, 10 mmol), and formamide (11.34 g, 252 mmol). Then,
0.6 g (5.0 mmol) of TMB (1, 3, 5 — trimethyI-benzene) was added to the
surfactant solution to foam a micro-emulsion template. NHNMPTS (0.384 g, 1.21

mmol) was added to the micro-emulsion template solution, followed by the

61

addition of sodium silicate solution (2.7 g (11.1 mmol) sodium silicate and 30 g
(1.7 mol) H20). The reaction mixture was allowed to age at 25 °C for 1 day
followed by one day at 100 °C. The removal of surfactant was carried out by
soxhlet extraction method using ethanol as the solvent. Table 3.1 provides the
reaction stoichiometries for the preparations of NMNMPTS functionalized MSU-F

silica by direct assembly method.

3.2.4. 30% NHNMPTS functionalization of MSU-F silica by the silane

grafting method.

30% NHNMPTS functionalized MSU-F silica was prepared by anchoring
the NHNMPTS organosilane on pre-assembled MSU-F foam silica. In the
synthesis of pure MSU-F foam silica, Pluronic 123 (1.2 g, 0.206 mmol) was
stirred with 1.0 M of acetic acid (10 ml, 10 mmol) and water (10 ml) until it was
dissolved. To make the micro-emulsion template, TMB (1, 3, 5 - trimethyl-
benzene) (1.0 g, 8.3 mmol) was mixed with surfactant solution for 30 min. Then,
sodium silicate solution (2.43 g (12.1 mmol) of sodium silicate and 30 g (1.7
mole) of H20) was added to the micro-emulsion template for making MSU-F
mesostructure. The final reaction mixture was stirred for 1 day at 25 °C and kept
in an oven for 1day at 100 °C. After the reaction, this sample was filtered, dried
and calcined at 500 °C for 4 hours. The surface of this calcined MSU-F silica
was then functionalized by NHNMPTS organosilane. This procedure of the
functionalization was the same as that described for SBA-15, which was provided

in section 3.2.2.

62

.0 nE0x0 .o .6800. 00am
.0: h+mo+vovluh+mb _ u. .

m

E22 m<_>. 0.0.0 0.6.0 .mmm .3 00550.00 00.5 .0000... .00... 0... c. 0.0.000 :00._.0 000.000.2090 .0 c0000.. 0020000 0:..0

.6582. 0.8.0 5208 mm;
858 02.... 9.. ..-nws. 80.22.83. 05 a“. 98......95 $509.0. 0. 0300 .o 858 82.0 0,: 0-00.2 0:0 0300 .o 00.0
.5 0:00 2.00.00 0. 02.000. 9.0.0.0. .030. :0 00.000 c00._.0 .0 .caoEm 0... :0 00000 00.032000. 0. .5080 0:0..000008 .0050...

0:50... 0... .0 0000 0... 0. 202.0300. .00050E 20E0000 82.0 0;. 0:0 9....05 0:0..0 0:. o. .0.0. .0. 0:0 .0: 0000:9000 0.... n

co..0N._0co..0::.o:09o .o .00. 00.090. 0... 00.00.05 5:03.000 0.0.0.0... 0;. :. 00>... 0002.00.00 0c... 0

 

 

 

 

0N... QN - - 3 0.0 3.0 00.0 ...N. 0.0. ".522 .x. on
3.0 0.0 00.0 - 2 m 2.0 .5 t: ..E. ".50: .\. 2
2.0 5.0 - 8.0 - no t 3 0.0. 0330 0. 00
0.0300... .05. .2955
.20.... .0.oE. .0.0EE. .0.oEE. .0.oEE. .0.oEE. .0.oEE.
:. 9.3.5.2...2 00.500 0 0.0.0.0:
0.0.5 00.E0&.o. .0... 0(01 mi... mm X. aha—52:2
.0 02.00.... 00...m

 

 

 

 

 

 

 

 

 

 

 

 

“TDwS. 0:0 £wa _m_.-<mw 00N._0co..0c2-w...n=>.ZIZ .0 0.00:.:>m 0... .0. 000: 000.02.250.90 c2800”. .—..m 0.90...

63

3.2.5. Diazotization of NHNMPTS mesostructures

The diazo functionalized mesostructures were obtained by a three-step
route. First, esterification of the alcohol groups on the mesostructured
NHNMPTS-silica surface was accomplished by reaction with the with anhydride
formed from fmoc-glycine and 2,6—dichlorobenzoylchloride. In this esterification
reaction, the vacuum-dried mesostructure having NHNMPTS functional groups
was dispersed in DMF solvent. (100 mL DMF per mmol NHNMPTS) In a
separate flask under a nitrogen atmosphere, fmoc-glycine (1O eq. per 1 eq. of
NHNMPTS) was dissolved in an equal volume of dried DMF, followed by the
addition of 2,6-dichlorobenzoyl chloride (10 eq). After 30 minutes, pyridine (15
eq) was added to the stirred reaction mixture and the mixture was aged for an
additional 15 minutes. The resulting was transferred to the NHNMPTS
functionalized MSU-F dispersed in DMF solvent, and mixture was stirred for
overnight under nitrogen condition. After formation of the fmoc-glycine ester
bond in the mesostructure, the fmoc protecting group was detached by the
addition of 20% piperidine/DMF solution (2:8 by volume). The volume of
piperidine/DMF solution was 50ml per mmonl NHNMPTS. As a last step, the
amine group, on the immobilized glycine residue was diazotized by the addition
of a nitrosolating agent, acidified NaNOz. First, sodium nitrite (4.5 g, 55 mmol)
was dissolve in H20 (10 mL). Through the use of a canula, the flask containing
sodium nitrite solution was connected to a second flask containing the glycine-
functionalized mesostructures (100 mg) suspended in dichloromethane (10 mL).

Then, the sodium nitrite solution was purged with nitrogen gas overnight. After

54

being purged with nitrogen gas, 10% sulfuric acid (0.5 mL in 4.5 mL H20) was
added to the sodium nitrite solution resulting in the generation of nitrous acid gas.
The gas was transferred by canula for reaction with the amine group of the
glycine-functionalized mesostructure for a period of 20-30 minutes. The actual
amounts of reagents used for synthesizing a—diazo group of three different

mesostructure materials are listed in Table 3.2.

65

 

0 0 .. .o. 0.00....

 

 

 

 

0 30
.E 2 0 3 0 0.0 a . 0 0.0. 0 8.. 0 3.0 a «0.0
. o 3 05.2212 «.8
0 0.00 0 0 r .0. 0-00....
.0. 2 0 2 0 0... a . 0 0.00 0 3... 0 8... 0 8."
. 8 .. 0.3.2222 $2
.0. 3 a 2 a 0... 0 . 0 0.0.. a 8.. 0 8.0 0 «a...
. o 3 0522...: 38
ion. 00.3.00 ...o > a
h _: G—hu
25 a .30.... 3. .. ...E .55 2.2.... 2320...»... . . 0.0020...
0. 0.0.09.0 30.2.2.0... . .30.“.
0.22005 .0 N

 

 

 

 

 

 

 

 

 

 

.00...0 00.30:..00000. 005.002.002.902212 .o 00:00:30.0 0.... 0. 0000 0.00000. .0 0.0.50.0 00... N...” 0.00...

66

3.3. Physical characterization

The physical properties of NHNMPTS-functionalized mesostructures and
their derivaties were determined by reflectance-Infrared spectroscopy, UV-Vis
spectroscopy analysis, 29Si solid state magic angle spinning nuclear magnetic
resonance (MAS NMR) spectroscopy, powder x-ray diffraction (XRD) analysis,
and nitrogen adsorption-desorption analysis.

For the chemical transformation of NHNMPTS groups to diazo groups,
each step was confirmed by observing the characteristic vibration of the
functional groups on the mesostructure surfaces using Nicolet Protégé 400
Magna reflectance infrared spectrophotometer equipped with a Barnes
analytical/spectra-tech diffuse reflectance accessory. Each spectrum was
collected in the 700 - 4000 cm'1 spectral region. The number of scan was 32
and data spacing was1.928 cm".

For the determination of the degree of functionalization, UV-Vis spectra in
the range 200—500nm were obtained on a Hitachi U-4001 spectrophotometer.
The scan speed was 300 nm/min, and the data collection interval was 0.5 nm.
For the quantification of functionalization, the absorbance at 301 nm for the
piperidine-dibenzylfulvene adduct formed in the reaction allowed quantification of
the degree of functionalization. Samples for the UV—Vis studies were prepared
by stirring 5 mg of fmoc-glycine functionalized sample in 0.4 ml dichloromethane
(DCM) and 0.4 ml piperidine for 30 minutes, followed by the addition of 1.6 mL of
MeOH and 7.6 mL of DCM. For the baseline correction, the reference sample

was also prepared by mixting 0.4 mL piperidine, 1.6 mL MeOH, and 8 mL DCM.

67

In order to further verify the degree of the functionalization, 29Si MAS solid
state NMR was used. 29Si MAS solid state NMR spectra were obtained at 79
MHz on a Varian VXR-4OOS solid state NMR spectrometer operating at 4 kHz
spinning frequency. The pulse delay was 400 seconds, which was enough time
to fully relax the magnetization of the 29Si nuclei before another pulse was
applied. Samples were contained in 6mm zirconia rotors. Talc was used as a
chemical shift reference with a value of -98.1 ppm.

Powder XRD data were collected on a Regaku Rotaflex Diffractometer
using CuKo radiation (A=1.542 A) generated at 45 kV and 100 mA. Data were
collected from 0.7 degree to 5 degree in 0.02 degree increments.

N2 adsorption-desorption isotherms were obtained at -196 °C on a
Micromeritics Tristar 3000 sorptometer. Prior to analysis the samples were
degassed at 90 °C and 10'6 torr for about 12 hours. Surface area were
calculated from a BET plot of the adsorption data between 0 and 0.30 P/Po.
From the adsorption branch of the isotherm pore size distributions were derived

by using the Barret-Joyner-Halenda (BJH) model.

3.4 Results

The N2 adsorption-desorption isotherm and framework pore size
distribution of SBA-15 silica are shown in Figure 3.6. The characteristic type IV
isotherm and hysteresis loop indicates capillary condensation takes place. At a
partial pressure of about 0.8, the isotherm exhibits a sharp adsorption step which

is indicative of the presence of uniform cylindrical pores of 9 nm. (Figure 3.6)

68

Another characteristic of hexagonal SBA-15 silica is the presence of low
angle reflections in the X-ray powder diffraction pattern. As shown below, a well
ordered hexagonal array of pores is expected to show three major reflections,
namely the (100), (110), and (200) reflections. The powder XRD pattern in

Figure 3.7 contains these reflections.

 

Figure 3.5. Possible low angle X-ray diffractions on hexagonal mesostructure

Figure 3.8 compares IR spectra obtained after each reaction step for the
diazo functionalization of 30% NHNMPTS SBA-15 by the silane grafting method.
First of all, the bottom spectrum shows the typical IR spectrum of mesoporous
silica. The peak around 1000 ~ 1200 cm'1 represents the stretching vibration of

the silicon-oxygen-silicon bonds. The stretching vibration of isolated surface

69

silanol group occur at 3750 cm", while the broad band centered at 3400 cm'1
represent the hydrogen bonding interaction between silanol groups in the
mesostructure. After the NHNMPTS group has been grafted to the mesoporous
silica surface, carbon-hydrogen stretching and bending vibrations appear in the
spectrum. The three adsorption bands near 2800 ~ 2900 cm'1 are assigned to
the stretching vibrations of the C-H centers on the NHNMPTS group. The
carbon-hydrogen bending vibrations are at 1450 ~1550 cm“. In next reaction
step, fmoc-glycine is attached to the hydroxyl group of NHNMPTS to form fmoc-
glycine ester. The reaction was confirmed by the presence of two major IR
adsorption vibrations, namely, the vibration of the carbonyl bond (1730 cm“) and
the bending vibration of the aromatic C-H bond (800 cm"). After the fmoc group
was removed, the band below 800cm'1 clearly disappeared. However, due to
unexpected side reactions (like mustard gas type reactions or cyclic amide
reactions which will be discussed in Chapter 4), the stretching frequency of the
carbonyl group is weaker than expected. Also, because of the overlapping of the
stretching bands of the hydrogen bonded O-H bonds of the silanol groups and
the NH bonds of the amine, the primary amine band at was not observed at
3300 ~ 3500 cm". However, the band at 1640 cm" can be assigned as the
bending vibration of the N-H bond, verifying its presence. Following the diazo
functionalization step a weak but clearly expressed band for the diazo group is
found at 2112 cm".

Diazo function nalization also was applied to 10% NHNMPTS

functionalized MSU-F silica foam prepared by direct assembly synthesis. Due to

70

its large pore feature, the mesocellular foam structure can be characterized by X-
ray analysis only at very low angle which makes normal Powder XRD analysis
impossible. Therefore, mesocellular foam silica generally is characterized by the
nitrogen adsorption-desorption isotherms and framework pore size distributions
derived from the adsorption and desorption branches. As shown in Figure 3.9,
the sharp adsorption step above P/Po = 0.9 is representative of the extremely
large cell size of the structure. The large cell size might be caused by the
formamide solvent acting as a co-surfactant. From the adsorption branch, the
cell sizes of the MSU-F mesocellular foam are found to be bimodal with the
dominant size at 74nm and the lesser at 15.4 nm. The window size as
determined from the desorption branch is 15.4 nm. Detailed isotherm data are
given in Table 3.4.

For comparison of efficiency between grafting method and direct assembly
synthesis, 30% NHNMPTS functionalized MSU-F foam silica was prepared by
the grafting method. The 30% NHNMPTS functionalized MSU-F foam silica also
was characterized by nitrogen adsorption-desorption isotherm analysis. Figure
3.10 shows typical N2 isotherms of a mesocellular foam structure with a 17 nm
cell size calculated from the adsorption branch and 15 nm and 6 nm window
sizes calculated from desorption branch. Two different window sizes can be
explained by the inhomogeniety in functionalization by the grafting method. The
texture properties are given in Table 3.4.

Figure 3.11 compares the IR spectra obtained after each reaction step in

diazo functionalization of 30% NHNMPTS MSU-F foam silica prepared by the

71

grafting method. The assignments of major IR absorption peaks for each step in
the functionalization process are given in. Figure 3.11. The 30% NHNMPTS
MSU-F silica foam shows much better IR spectral resolution than 30%
NHNMPTS SBA-15 prepared by the same pathway. In the spectrum for the
mesostructure having attached fmoc groups, the intensity of the carbonyl
vibration is much stronger than that of functionalized SBA-15. Also, this
spectrum shows the aromatic C-H vibration at 3050 cm'1 which was not found for
the SBA-15 derivative. After deprotection of fmoc group, the IR spectrum still
shows the sharp carbonyl stretching .vibration at 1740 cm“, while the
corresponding 30% NMNMPTS SBA-15 did not. Presumably, the larger pore
mesocelluar foam silica derivative has a high degree of organo-functioalization
than hexagonal SBA-15. A more detailed discussion of the degree of
functionalizatoin will be provided later along with 29Si solid state MAS NMR data.
In Figure 3.11, the diazo vibration at 2110 cm'1 and the carbonyl vibration at
1740cm’1 are clearly evident. These bands verify the successful diazo
functionalization of the structure.

In order to use the high surface area and large pore volumes of diazo
functionalized SBA-15 and MSU-F materials for the future proteomic
applications, the accessibility of the framework pore structure after diazo
functionalization is very important. In order to determine mesoporosity, the
nitrogen adsorption-desorption isotherms for the following three samples were
collected, namely 30% NMNMPTS MSU-F made by grafting method, the 30%

glycine functional derivative, and the 30% diazo derivative. As shown in Figure

72

3.12, diazo functionalized MSU-F retains an isotherm typical of mesocellular
foam silica structures with a high surface area and a large framework pore
structure. However, in comparison with the 30% NHNMPTS, there was some
loss of pore volume, which may be caused by two possible reasons. Some
amount of degradation may occur for the pore structures during synthesis,
particularly through the use of the base, piperidine. The pore volume is inversely
related to the weight of the sample. Therefore, increased weight caused by
organofunctionalization contributes to a lowering of the specific pore volume.

As shown in table 3.5, the degree of functionalization was determined by
298i solid state MAS NMR and uv analysis after fmoc titration. From 298i solid
state MAS NMR analysis, we can quantify the fraction of organo-functional silicon
centers in the silica mesostructure. The ratio of organic groups to the total
amount of silicon species provides the degree of functionalization, upon
NHNMPTS functionalization at the initial reaction stage. However, fmoc titration
method, as followed by UV-Vis analysis, shows the degree of functionalization at
a different reaction step. In this reaction step fmoc groups are removed from the
mesostructure using 20% piperidine/DMF, resulting in glycine-functionalized
MSU-F. The piperidine also generates 9-methylene-9H-fluorene and carbon
dioxide as a side product. In this process, this 9-methylene-9H-fluorene reacts
with extra piperidine in the DMF solution and forms 1-((9H-fluoren-g-
yl)methyl)piperidine (piperidine-dibenzylfulvene adduct) which is UV-active at 301
nm. The fmoc deprotecting process and the analyte for UV analysis are

illustrated in the following reaction schemes. (Scheme 3.4)

73

Scheme 3.4. Fmoc deprotecting process and analyte for UV analysis

0 H
i [_PLHIO .0 + [NJ

._/—\N

I
\
3 CH3 piperidine

 

1-((9H-fluoren-g-yl)methyl)piperidine

74

For the determination of the degree of functionalization by fmoc titration,

the concentraton of fmoc was calculated using Beer’s law:

Absorbance = molar absorbance (7800 mol'1 cm”) x concentration of absorbing

material x cell length(1 cm)

Table 3.3. UV analysis data from fmoc titration and their deprotected fmoc

concentration by UV analysis

 

 

 

 

 

Concentration
Sample Absorbance
(mmolelg)
30% NHNMPTS SBA-15 (G) 1.000 0.26
10% NHNMPTS MSU-F (D) 0.072 (16 times diluted) 0.30
30% NHNMPTS MSU-F (6) 0.131 (16 times diluted) 0.53

 

 

 

A plot of absorbance versus molar concentration will be straight line
because molar absorbance and cell length are constant values. In reality, the
plot is not linear over the entire concentration range, particularly beyond an
absorbance of 1.0. Therefore, for the better accuracy, the concentrations of fmoc
from the 10% NHNMPTS MSU-F (D) and 30% NHNMPTS MSU-F (G) samples

were calculated using a 16 fold dilution of the deprotection solution.

75

 

3.5 Conclusion

In this experiment, I reported diazo-functionalized mesostructured silica for
potential use in proteomic analysis have been successfully synthesized.
However, as revealed by IR spectral analyses, the degree of diazo
functionalization is still low compared to the initial degree of NHNMPTS
functionalization obtained by grafting or direct assembly synthesis (see Table
3.6). In order to increase the level of diazo funtionalization, here is some aspect

to consider for the preparation of functional mesostructure materials.

Graftlng and direct methods

On the basis of the NMR and UV-VIS analytical data for the 10%
NHNMPS MSU-F sample made by the direct assembly method and for the 30%
NHNMPS MSU-F made by silane grafting synthesis (c.f., Table 3.6), we conclude
that the methods result in formation of products in which 8 % and 20 % of the
silicon centers are actually functionalized. Thus, the grafting method is less
efficient than direct assembly in providing functionalized derivatives. It is likely
that the organosilane used in grafting synthesis are not evenly distributed on the
surface and are clustered in domain rich in NHNMPTS group. Under these
condition steric constraints will lead to low level of fmoc glycine coupling. Direct
assembly pathway is the best synthesis method to prepare a functionalized
mesostructure for better efficiency. But efforts to increase the level of NHNMPTS
functionalizatoin to values greater than 10% did not provided foam

mesostructures.

76

w

rt.” ~

II

he

30% NHNMPS SBA-15 (grafting) and 30% NHNMPS MSU-F (grafting)

The NMR data in Table 3.5 shows that 30% NHNMPS SBA-15 prepared
by the grafting method and 30% NHNMPS MSU-F prepared by grafting method
have similar degrees of functionalization. However, the UV-VIS data for the
fmoc-glycine derivatives show the large pore MSU-F (74 nm pore size) to have
twice the functionalization than the relatively small pore SBA-15 derivative (10
nm pore size). Therefore, large pore mesostructures are the best candidates for
higher degree of functionalization, probably because they allow more uniform

distribution of functional groups on the silica surface.

77

SBA-15

Volume adsorbed (cm’lg, STP)

   

0 02 Q4 03 08 1

Relatlve Pressure (PIPo)

Pore size distribution - SBA-15
0.04

0.035 8.9 nm

F’
o
u

0&25

Adsorption
P
S

0015

dVIdD pore volume (cm’lg - nm)
9
3

P
a
c»
a

 

O

0 100 200 300 400 500 600
Pore slze (nm)

Figure 3.6. N2 adsorption-desorption isotherms and BJH pore size distribution for

SBA-15 silica calcined at 500 °C as determined from the adsorption isotherm.

78

XRD of SBA-15

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

E i I E i
d spacing = 10.5 nm
100 """""""""
3
S.
E
to
t:
.2
5
110
1‘- 200
E L

 

 

 

 

 

 

1 1.5 2 2.5 3 3.5 4 4.5 5
2theta

Figure 3.7. Powder X-ray diffraction pattern for calcined SBA-15 silica.

79

.859: mcESm mcm=m

es 3 3-5m szzzz $8 so eoeodfieozoea ones as .2 new 8:82 some are .8598 28% E on 939d

\|l\2/|\.Im

\|\2/|\l_m.

:olb
w
W

. 3110.5

25.:

 

Aon. cone—5.3225

3:...0

=3ch

 

cow coup ccwv ooou OQVN oowN ooun comm ooov

33:95

soueulwsueu %

80

1 400

1 200

1 000

800

600

400

200

Volume adsorbed (cm’lg, STP)

   

0 0.2 0.4 0.6 0.8 1
Relative Pressure (PIPo)

15.4 nm

uongosea

Adsorption
(wu - 61.3110) ewnlOA 910:! op/Ap

dVIdD Pore Volume (ems/g - nm)

 

0 50 100 150 200

Pore slze (nm)
Figure 3.9. N2 adsorption-desorption isotherm for 10% NMNMPTS functionalized

MSU-F prepared by direct assembly and the BJH pore sizes distribution obtained

from the adsorption and desorption isotherms.

81

 

 

 

«3 2v 3. do. 2 6. use: mEEzzz «.8
s3 3.... can: 5...: .E a: use: mhdszzz $2
5.. mus a.» 6. 3.3m «5:222 $8

 

A215: 2.5.3 20.".

 

 

ASN.5 «Ea coat—5

 

EE. .8255 20m

 

oEEwm

 

 

..Eosofl c8983 Nz Eon 355233 £3.39: 393035 uoN__mcouoca 90.22123 mmanoa .9238... in min...

82

1200

   

a“: 1000

.—

U)

E" 300

E

8.

8

e 600

O

i

0 400

E

2

g 200
0
0 02 0.4 0.6 0.8 1

Relative Pressure (PIPo)

Adsorption
dVIdD pore volume (cm3lg - nm)
(tuu - Blgwo) amnlmt eJod cpl/(p
uondJosag

 

0 1O 20 30 40 50 60

Pore Size (nm)
Figure 3.10. N2 adsorption-desorption isotherms for 30% NMNMPTS functional-

ized MSU-F prepared by grafting method and the BJH pore sizes distribution

obtained from the adsorption and desorption isotherms.

83

.uofioE 9535 mcm=m

9: E “.592 wedszzz $8 so eoeouaeoeoea ones as 5 ass 8:89 eons eta 8598 oedema... E is 2:9“.

 

:EofenEscon;
3110.5 can 83 o2... ooou can coca comm 8mm ocow
335.95.
3.1.0 c326
nT—U/
z 5 3310.6
Ionk /I\l
ecu—33:0
3:3 l /o
f»: m SmVOlo 331.0 owe—=95 H.
o \|\ /|\I m
x \_F o w
0 . 3.1 n
O a... m...
z.
\IKz/I\l w
o
”10’
z E
\|\ /|\l
2.H2\ O

 

1800

 

 
 
 
  
   

+30% NHNMPTS-MSU-F
1600 -

mus-- 30% glycine-MSU-F

+30% diazo—MSU—F
1200

1000

 

Volume adsorbed (cm3lg, ST P)

 

 

    

 

   

0 0.2 0.4 0.6 0.8 1

Relative Pressure (P/Po)

Figure 3.12. Comparison of N2 adsorption and desorption isotherms for 30%
NHNMPTS functionalized MSU-F, glycine-functionalized MSU-F, and diazo

functionalized MSU-F silica. The isotherms are offset by 500 cm3/g for clarity.

85

 

. _mQmemS .mm 80 2 oo oases—Ea w 02
Amh+mh+m0+vglmh+mb _ n. «_u 5 .39 «U . ”.5 .0m

 

 

 

 

 

 

 

 

 

 

 

 

U0N=mcozoc97wkm<¢ZIz

ho KEZ 23w

.\. on 8 «2 8n 5... 8v ”saws. mEEzzz so...

.\. a - S 2" Se 6. dim: mEEzzz $2

.a 2 2 S m: S» 6. 3-3mm mEEzzz $8
a gang—accrues.—

h ._. O O oEEww
N n a e
9590 n.0 daemon.
$9309.50me

28 58 05 5 Badge moocmcomm: co 36:25 _memQE .m.m 03m...

86

 

 

.329: 2:33 noN___noEE_ ho :ozmzcmocoo or: 05890 9

038mm 05 he Ego; 2:52 05 B 3220 mm; 955 2590 uoN___noEE_ B 0.9: m7: dock AdeEEo—e mm; 0353
he 20:. .5: :36 2390 nmN___noEE_ ,6 mw_oE 0:. ucm 6053:2382 10 02mm: 05 m_ x 20:5 .xAmaofidxiNQwv :2
35:28 331:9“; wage 05 amen. Emu E22 99m 38 _waN .3 85320 :o=m~_m:ozo:2 B «.09me 05 :o :32

2m 22:; _mmaEmm _o Ema B: wo=m_oE 2:35 uwN__BoEE_ S mcozmbcmocoo 05 9m $35923 :_ mw:_m> 0F...

 

 

 

 

we> 8.: a 33.25: v.3 $3 a. ”Tam: 955.212 .88
mo> and a BEBEE «4.1.x.» An: dim: who—2212 Axx...
oo> mm... a .9225: «.5 $2 612-com mEszzz $8
333.80% 522.5 33.22: 5:25 Eeficoflua 00.5 283
.3 55522225 025 3 ue:_E._3mu siege E22 28» u=ow 3 2:an
ans: .0 cow—35.; 5582350 9390 nos—522. :o_uau__a:o=o:a ex.

 

 

 

 

 

 

.5595 SE“. ho m_w>_m:m >3 _ucm 1.22 83m 28 B nchtoEU :o:m~=m:o=o:2 w._.n=2212 10 $0300 .m.m 038.

87

3.6 Reference

(1 ) Andersson, L., Porath, J. Anal Biochem 1986, 154 250-254

(2) Sano, A., Nakamura, H. Anal. Sci. 2004, 20, 565-566

(3) Kweon, H. K., Hakansson, K. Anal. Chem. 2006, 78, 1743-1749

(4) Wolshin, F ., Wienkoop, S., Weckwerth, W. Proteomics 2005, 5, 4389-4397

(5) Pandey, A., Podtelejnikov, A. V., Blagoev, B., Bustelo, X. R., Mann, M.,
Lodish. H. F. Proc. Natl. Acad. Sci 2000, 97, 179-184

(6) Lansdell, T. A., Tepe, J. J. Tetrahedron Lett. 2004, 45, 91-93

(7) Corma, A., Martinez, A., Martinez-Sofia, V., Monton, J. B. J. Catal. 1995, 153,
25-31

(8) Beck, J. S., Vartuli, J. C., Roth, W. J., Leonowicz, M. E., Kresge, C. T.,
Schmitt, K. D., Chu, C. T. W., Olson, D. H., Sheppartd, E. W., McCullen, S. B.,
Higgins, J. B., Schlenker, J. L. J. Am. Chem. Soc. 1992, 114, 10834- 10843

(9) Burkett, S. L., Sims, S. 0., Mann, S. Chem. Commun. 1996, 11, 1367-1368

(10) Chan, W.C., White, P.D. In Fmoc Solid Phase Peptide Synthesis — A
Practical Approach, Oxford University Press, New York. 2000, 62-63

88

Chapter 4

Improved Diazo-functionalization of mesoporous silica and the preparation

of an IMAC-type mesoporous silica for potential use in phosphopeptide

enrichment

4.1. Introduction

In the last chapter, three NHNMPTS functionalized mesoporous silicas
and their diazotization reactions have been discussed. However, a few
questions have been arisen from the research presented in the last chapter.
First of all, there was a big difference between the degree of functionalization
based on the initial reaction rate and the amount of surface functional group
detected by UV analysis. Also, as mentioned in Chapter 3, an unexpected loss
in carbonyl band intensity in the IR spectrum was observed after the deprotection
of the fmoc group. The purpose of this chapter is to inVestigate possible reasons
for these observations and to examine other approaches to diazotized
mesoporous materials with an improved degree of functionalization. Also, a new
IMAC type of mesoporous silica for potential use in phosphopeptide enrichment
is described.

In Chapter 3, we determined the degree of organo-functionalization by two
instrumental analyses techniques, namely, 29Si solid state NMR and UV analysis.
However, there was a big difference in the results provided by these two

techniques. For example, the actual degree of functionalization for formally 10%

89

NHNMPTS MSU-F silica foam by direct assembly synthesis was 8%, as
determined by solid state NMR. This level of functionalization corresponds to 1.2
mmolelg. However, the concentration of functional group on the mesostructure
surface was found to be only 0.3 mmolelg by UV-VIS analysis of the fmoc groups
liberated from the mesostmcture surface. Generally 298i solid state NMR
analysis distinguishes and quantifies the types of silicon centers. The
resonances of immobilized organosilane centers having Si-C bonds can be
distinguished from Si04 silicon centers and the fraction of each type of center
can be quantitatively determined by integration of the resonance lines, provided
that the delay time between pulses allows for the complete relaxation of the
nuclei. This measurement shows how much organic moiety is incorporated onto
the mesostructure surface.

For fmoc titration by UV analysis“ 2, the fmoc groups initially react with the
OH functional groups on mesostructure surface, as shown in the scheme on
page 57 in Chapter 3. Upon complete reaction, as shown on page 74 in Chapter
3, the reaction residues, 9-methylene-9H-fluorene and carbon dioxide, are
generated through deprotect of the fmoc group. During this process, the 9-
methylene-QH-fluorene residue reacts with excess piperidine, and produces 1-
((9H-fluoren-9-yl) methyl) pipen’dine (piperidine-dibenzylfulvene adduct), which
can be quantified by UV-Vis spectroscopy. This measurement determines how
much organic moiety is actively participating in the diazotization reaction.
Therefore, through comparison of the solid state NMR data and UV-VIS fmoc

titration data, we can determine all the hydroxyl ethyl functional groups on the

90

 

silica surface that do not participate in the organo-functionalization reaction. This
allows us to conclude that some of the terminal hydroxyl functional groups of
NHNMPTS interact with the silica framework or is embedded into the framework,
which leads low accessibility toward other organic reagents.

We also anticipated a potential problem in losing an undesirable amount
of carbonyl species during deprotection of the fmoc group. As shown in Figure
3.8, the intensity of the 1730 cm'1 carbonyl stretching band for the fmoc group
attached to the surface of MSU-F silica (third spectrum, two carbonyl group) is
much stronger than that of the fmoc deprotected MSU-F silca product (fourth
spectrum, one carbonyl group). There are a few possible explanations for this
observation. First of all, the nonbonding electrons on the nitrogen atom may
participate in a cyclization reaction leading to the elimination of the glycidyl
groups, as shown on below:

Scheme 4.1. Elimination of Glycidyl group by cyclization reaction
0
pp"...

s§—/_\sz;/Q > iii—[TMB + H2N\/l(l)\.
\CH3 \CH3 0

A similar type of cyclization reaction, called the mustard gas reaction3, has been

 

observed at a sulfur center. Owing to this type of side reaction, we might lose

more glycidyl groups than we expected.

91

A second possible complicating reaction that may occur on the
mesostructure surface involves an internal nucleophilic cyclization to form an
aziridinone species.4 Also, due to the strain of the three membered aziridinone
ring, a more likely reaction is the formation of a dimer by cross reaction between
two functional groups. The possible reactions are represented in the following

schemes.

Scheme 4.2. Possible elimination mechanism of glycidyl group by nucleophilic

cycliczation

7 hkc”, "” + Ei CH3

Aziridinone

In order to circumvent the potential problems associated with the above

reactions, two other organosilanes have been selected for the preparation of

92

TR ...um Effigy

 

diazo-functionalized mesoporous silica materials, namely (3-glycidoxypropyl)
trimethoxysilane, and p-aminophenyltrimethoxysilane.

The (3—glycidoxypropyl) trimethoxysilane molecule has an epoxide ring at
its terminal position, which should make it less likely to embed itself in the walls
of a silica mesostructure in comparison to NHNMPTS. However, to serve as a
precursor for diazotization the epoxy ring needs to be opened to generate a
nucleophilic OH center.

VVltl'l a p-aminophenyltrimethoxysilane, the phenyl group is hydrophobic
and less likely to become embedded in the hydroxylated walls of the
mesostructure. Thus more of the NH2 groups may be accessible for diazotization.
Also, the amine groups can directly react with nitrous acid to easily form a diazo
functionality.

So far we have considered new diazo functionalized silica for the potential
use in phosphorylated peptides separation. However, IMAC is still one of the
most powerful tools for phosphorylated peptide enrichment at the present time.
Recently, Zou et al.5 reported the enrichment of phosphopeptide by a MOM-41
silica that mimicks an IMAC column. He functionalized the MOM-41 silica by
grafting reaction of a (3-glycidoxypropyl) trimethoxysilane that had been coupled
to a metal chelating agent, namely iminodiacetic acid. However, the coupling of
the iminodiacetic acid to the (3—glycidoxypropyl) trimethoxysilane looked very
difficult. For the preparation of the organosilane, the solid iminodiacetic acid was
dissolved in water, followed by an adjustment of the pH to 11 with sodium

hydroxide. Then, this solution is mixed with (3-glycidoxypropyl) trimethoxysilane

93

very slowly at low temperature to limit hydrolysis of the alkoxide groups on silicon
and the polymerization of the organosilane. The hydroxide anion, the
carboxylate oxygen from iminodiacetic acid, and the nitrogen from iminodiacetic
acid can compete for reaction with the epoxide group, whereas only the nitrogen
from iminodiacetic acid is the desired nucleophile. Possible nucleophilic attack of
(3-glycidoxypropyl) trimethoxysilane by the three competing centers is illustrated
below. The two competing reaction by OH' and the carboxylate oxygen centers at
low iminodiacetic acid concentration is likely to lead to low yields of the desired

iminodiacetic acid functionality.

Scheme 4.3. Nucleophiles which possibly react with terminal epoxide group

/ :6”-
o o-
(CH30)3Si/\/\o’\w 12$"
o '0 H

i7"

 

This chapter examines a new pathway for the preparation of an IMAC type
large pore mesoporous silica material for potential use for phosphorylated
peptide enrichment.

4.2 Experimental

4.2.1 Reagents

94

 

The structure directing agent, Pluonic 123 ((EO)20(PO)70(EO)20) was
obtained from BASF for the preparation of large pore MSU-F mesoporous silica.
As organosilane sources, (3—glycidoxypropyl)trimethoxysilane (GPTS), p-amino-
phenyltrimethoxysilane (APTS), and 3-(triethoxysilyl)propylsuccinic anhydride
(SATS,95%) were purchased from Gelest Inc. The silicate source, sodium
silicate solution (14% NaOH, 27% SK): by weight), was obtained from Aldrich.
Lithium aluminum hydride (LiAlH4) in tetrahydrofuran (1.0 M), dichloromethane,
and tetrahydrofuran (THF) were also purchased from Aldrich. The
dichloromethane and tetrahydrofuran were dried by use of a Solvtek drying kit.
Glacial acetic acid, hydrochloric acid, perchloric acid, dried methanol and dired
formamide were purchased from Spectrum. Dimethyl sulfoxide (DMSO) was
purchased from EMD chemicals Inc. For an experiment of binding organo-
phosphate, mono—methylphosphate bis(cyclohexyl ammonium) salt was
purchased from Aldrich. Absolute ethanol was obtained in-house‘. Water used in
the synthesis was from a double-exchanged Millipore filter apparatus. For the
diazotization reaction, all the chemicals used in the synthesis were the same
chemicals described in Chapter 3. All of reagents were used without further

purification except for dichloromethane and tetrahydrofuran (THF).

4.2.2. GPTS-, APTS-, and SATS-functionalized MSU-F mesocellular silica
foam by direct assembly
The GPTS-, APTS-,and SATS-functionalized MSU-F mesocellular silica

foam was prepared by a direct assembly pathway using the organosilane

95

reagents (3-glycidoxypropyl)trimethoxysilane (GPTS), p-aminophenyltrimethoxy—
silane (APTS), and 3-(triethoxysilyl)propylsuccinic anhydride (SATS). Firstly,
Pluronic 123 (0.80 g, 0.13 mmol) was dissolved in formamide (11.34 g, 252
mmol) mixed with glacial acetic acid (0.6 g, 10 mmol). In order to make the
micro-emulsion template 0.6 g (5.0 mmol) of TMB (1, 3, 5 - trimethyl-benzene)
was mixed with the surfactant solution for 30 minutes. After the preparation of
the surfactant solution, GPTS (0.860 g, 3.639 mmole), APTS (0.766 g, 3.639
mmole), or SATS (0.389 g, 1.213 mmole) followed by the addition of sodium
silicate solution. The mixture was stirred at 25°C for 20 hours, followed by static
aging in an oven at 100°C for one day. Then the surfactant Was removed by
soxhlet extraction with ethanol for one day. The overall reaction stoichiometries

used for the preparation of the MSU-F mesoporous silicas are given in Table 4.1.

96

 

 

AmeNmzv 98:7. anow mm; 858 mo___w on... .n_-3ws_ .6 m md :0 So coEmo 2.958 mm; cozomoc

9.5sz 9F .039 9: E :96 858 coo___m Co EzoEm 9: co comma m_ ocm=mocm9o Co EsoEm 9: .859: 9555 9:

Co $8 9: c. $62683. .wuofioE >3Eomwm 696 9: new mcESm 9:26 9: 2 .99 .0: new :0. cozmcmfioo of. n

:osz__m:o=o:Eocho co _o>o_ c.3098 o£ coumoEE :03?ng 23.29: 05 E :96 omficoozoa 05k a

 

 

 

 

 

- ad 2 3 Ed 3.” v.2 a 6. “..ams. wh<m $8
mu... 3 2 m 3... 3.. t: a .9 aims. mh<w $2
w
mu... 3 2 m 35 3.» 9..” a E. ".-ams. 952 38
mm... 3 2 m 35 3." 3d a a. use: wEo $8
A295 A225 308:: EOE—5 EOE—5 ABOEE. 3.9.5:
0958 a «.5355.
23:85.0". cone; o<OI ms... mu .5. ecu—39590 «2.5

 

 

 

 

 

 

 

 

 

 

8% 55 “.592 Beacoficpmem Ea -mE< .-mEo .6 8859a 9: .3 8m: 8:652:2on 8:08”. .3 2an

4.2.3. 30% SATS functionalized MSU-F silica by grafting method“

A 30% SATS-functionalized MSU-F silica was synthesized by grafting
SATS organosilane on to pre-assembled MSU-F foam silica. For the preparation
of pure MSU-F foam silica, Pluronic 123 (1.2 g, 0.206 mmol) was stirred in 1.0 M
acetic acid (10 mL, 10 mmol) and water (10 ml) until it dissolved. To prepare the
micro-emulsion template, TMB (1, 3, 5 - trimethyI-benzene) (1.0 g, 8.3 mmol)
was mixed with the surfactant solution for 30min. Then, sodium silicate solution
2.43 g (12.1 mmol, sodium silicate) and 30 9 H20 (1.7 mole) were added to the
micro-emulsion template. The final reaction mixture was stirred for one day at
25°C and kept under static conditions in an 'oven for one day at 100 °C. The
product was filtered, dried and calcined at 500 °C for 4. hours. The calcined
MSU-F silica was then functionalized by reaction with a SATS/DMF solution
under reflux conditions for 4 hours. The detailed procedure for SATS

functionalization was the same as that described for SBA-15 in section 3.2.2.

4.2.4. Diazotization reaction of 30% GPTS-functionalized MSU-F silica

The epoxide ring of 30% GPTS-functionalized MSU-F silica was opened
using a base (LiAIH4)7 or an acid catalyst (HCIO4)8. In the ring opening reaction
with LiAlH47, 0.5 g of 30% GPTS functionalized MSU-F (1.6 mmole of organic
moiety) was dispersed in the dried THF solvent (10 mL). Then, 10 mL of 1M
LiAlH4 in THF was added dropwise to a mixture of 30 % GPTS-functionalized
MSU-F and THF in an ice bath. After the addition of 1M LiAlH4, the reaction was

allowed to equilibrate at ambient temperature for 2 hours. This reaction mixture

98

was cooled in an ice bath and the mixture was quenched with H2O until the
evolution of H2 gas ended. This solid was filtered and mixed with 300 ml of 3.0 M
HCI solution for 30 minutes to dissolve the aluminum side product. The resulting
solid product was filtered, washed and air-dried.

For the acid catalyzed ring opening reaction”, 0.5 g of 30% GPTS
functionalized MSU-F (1.6 mmole of epoxide) was mixed with 0.1 mL of HClO4 in
50 mL of DMSO/H2O (33 mL/17 mL). Then, this mixture was stirred at 80°C for 3
hours. The final product was filtered, washed with ethanol and air-dried.

After the epoxide ring-opening processes, diazotization reaction was
performed using the three-step route which was described in.Section 3.2.5.9
First, the hydroxyl groups of the ring-opend GPTS-silica was allowed to react
with fmoc-glycine activated by 2,6-dichlorobenzoylchloride. In order to improve
the reactivity of 2,6-dichlorobenzoylchloride, the ring opened GPTS-silica was
air-dried overnight prior to use. After the fi'noc-glycine ester bond was formed,
deprotection of the fmoc group was achieved using 20% piperidine/DMF solution
(2:8 by volume) at a ratio of 0.2 mmole of GPTS functional group to 10 ml of
piperidine/DMF solution. As the last step, the amine group formed in the
deprotecting fmoc reaction, was diazotized by reaction with the nitrosating agent.
(acidified NaNO2) A sodium nitrite solution (4.5 g, 55 mmol/ 10 ml water) and the
mesostructured organosilica (100 mg) suspended in dichloromethane (10 ml)
were placed in two separate flasks. By use of a canula, the flask containing the
sodium nitrite solution was connected to flask containing the mesostructureed

organosilica flask. Then, the sodium nitrite solution was purged with nitrogen gas

99

overnight. To the sodium nitrite solution 5 ml of 10 % sulfuric acid solution was
added to produce nitrous acid gas. These gases were allowed to mix with the
amine functionalized mesostructure for 20-30 minutes. The final diazotized silica

was filtered, washed with dichloromethane and air dried.

4.2.5. Diazotization reaction of 30% APTS functionalized MSU-F silica1o

A 10 mg quantity of 30% aminophenyl functionalized MSU-F silica was
mixed with 5 mL of water and 3 mL of concentrated hydrochloric acid, and the
resulting mixture was heated in a hot water bath (70 °C). After 15 minutes,
another 5 mL of concentrated hydrochloric acid was added. Then the mixture
was stored in an ice bath. A second solution was prepared by dissolving 2.3 g of
sodium nitrite in 6.7 mL of water and cooling the solution in an ice bath. After 30
minutes, the cold sodium nitrate solution was transferred to the protonated
aminophenyl functionalized MSU-F silica mixture one pipetteful at a time. In
order to keep the temperature low (0 °C ~ 5 °C) during transfer, additional pieces
of ice were added to the diazotization flask, as needed. After another 30 minutes,

the final solid product was filtered.

4.2.6. Absorption of organophosphate11

A 20 mL quantity of 100 mM methyl phosphate bis(cyclohexyl ammonium)
salt (2 mmole) in dried methanol was mixed with the diazotized APTS-
functionalized MSU-F silica, and the mixture was heated to reflux temperature
overnight. The product was then filtered and dried at room temperature. Finally,

100

in order to remove the ionically bonding species, the phosphate-bound MSU-F
silica was mixed with 20 mL of LG M HCI for 15 minutes. The final product was

filtered with water, followed by wash with ethanol, and dried at room temperature.

4.2.7. Preparation of Fe3+ Immobilized on diacetate-functionalized MSU-F12
Fey-immobilized MSU-F was prepared from SATS-functionalized MSU-F
foam silica. First, succinic anhydride functionalized MSU-F (1.6 mmole of the
organic group) was suspended in 0.48 ml of concentrated acetic acid. The
mixture was heated at 45 ~ 50 °C for 1~2 hours, followed by the addition of 2.42
mL of water. The white milky suspension was stirred at 60 °C for 3 hours.
Afterward 0.024 mL of cone. HCI was added, and the mixture was stirred at
ambient temperature overnight. The diacetate-product was filtered and dried.
Finally, the diacetate functionalized MSU-F silica was mixed with an excess
amount of 100 mM FeCla solution for 30 minutes, rinsed with ethanol, and air-

dned.

4.3. Physical characterization

GPTS-, APTS-, and SATS-functionalized MSU-F silica and their
derivatives were characterized by reflectance-infrared spectroscopy, 2“’Si, 31P
solid state magic angle spinning nuclear magnetic resonance (MAS NMR)
spectroscopy, and nitrogen adsorption-desorption analysis.

Infrared spectroscopy was performed using a Nicolet Protege 400 Magna

reflectance infrared spectrophotometer equipped with a Barnes analytical

101

lspectra-tech diffuse reflectance accessory. Infrared spectra (32 scans per
spectrum) were collected in the range of 700 to 4000 cm'1 with a resolution of
1.928 cm".

The degree of functionalization achieved for an each mesostructured
organo-silica was determined by quantitative 29Si MAS solid state NMR. 29Si
MAS NMR spectra were collected at 79 MHz on a Varian VXR-400S solid state
NMR spectrometer. The sample spinning rate was 4 kHz, and the pulse delay
was 400 seconds. Talc was used as a chemical shift reference (-98.1 ppm). 31F
MAS NMR was used to verify the incorporation of organo phosphate onto the
mesostructure. The spectra were collected at 162 MHz on a Varian VXR-4008
solid state NMR spectrometer operating in single pulse mode. The spinning
frequency was 4 kHz and a pulse delay was 1 second. H3PO4 was used as the
chemical shift reference. (Oppm)

N2 adsorption-desorption isotherms were collected on a Micrometrics
Tristar 3000 sorptometer. Before analysis, the samples were degassed at 90°C
and 10'6 torr for about 12 hours. BET surface areas were calculated from the
linear part of a BET plot. BJH pore size distributions were derived from the
adsorption branch of the isotherm.

TEM images and EDX data were obtained on a JEOL 200FS transmission
electron microscope which was operated at 200kV. The sample was prepared
by sonification of the materials in ethanol for 30 minutes. This suspension then

was dried on a carbon-coated holey film supported on a Cu grid.

102

P :1.‘ in r. 3:me
I

 

4.4. Results and Discussion

4.4.1 30% GPTS functionalized MSU-F and its diazotization

The graphs in Figure 4.1 provide the nitrogen adsorption—desorption
isotherms of 30% GPTS functionalized MSU-F silica and the pore (cell) and
window size distributions obtained from the adsorption and desorption branches
of the isotherms, respectively. The adsorption—desorption isotherms show typical
type IV behavior, having a hysteresis loop between 0.7 and 0.95 relative
pressure associated with the capillary condensation of nitrogen taking place in a
large pore material. As shown on Figure 4.1, 30 % GPTS functionalized MSU-F
silica shows bimodal 53 nm and 16 nm cell sizes and a 7nm window size. The
BET surface area for these GPTS functionalized foam silica is 460 mzlg, and the
pore volume is 1.85 cm3/g.

29Si solid state MAS NMR was used to determine the degree of
functionalizaton. As shown in Figure 4.2, the 29Si solid state MAS NMR
resonance lines of 30% GPTS functionalized MSU-F silica appear at -70, -104,
and -112 ppm, which are assigned to RSi(OSi)3 (T3), (OH)Si(OSi)3 (Q3), and
Si(OSi)4(Q‘) sites, respectively. The ratio of organo silicon to total silicon, which
is defined the degree of functionalization (Ta/(T3+Q3+Q4)), is 29%. (cf. Table 4.2)
This degree of funtionalization is higher than reported earlier for NHNMPTS- or
BHAPS- functionalized MSU-F foam silica (see chapter 3). Presumably, the less
hydrophilic the terminal functional group of organic moiety (ex. the hydrophilic

hydroxyl group) is, the less it partitions in the surfactant during synthesis, which

103

 

leads to a high degree of framework functionalization. As it was mentioned in
Chapter 2, a hydrophilic organo functional group, such as the hydroxyl group,
can interact with hydrophilic groups of surfactants through H-bonding and have a
better chance of being removed when surfactant is removed by solvent extraction,
which might be the reason of the less functionalization of the product than what
we attempt.

In the stepwise conversion of 30% GPTS-MSU-F to a diazotized analog,
each reaction step was conflnned by IR spectra. The IR spectra in Figure 4.3 for
the aminoacid ester formed by deprotection of the fmoc group shows a much
stronger carbonyl stretching vibration than the derivative made from NHNMPTS
MSU-F (see Chapter 3, Figure 3.8). The incorporation of the initial epoxide by
direct synthesis places more of the organosilane at the pore wall surfaces for
subsequent functionalization. In addition, the 29Si solid state NMR spectrum
(Figure 4.2) verified the incorporation of more GPTS groups in the framework
walls of MSU-F in comparison to NHNMPTS groups. However, the last reaction
step leading to diazotization occurs in low yield. The diazo stretching vibration at
2420 cm'1 is very weak and shifted from the expected frequency of 2110 cm”.
The reason for the shift might arise from a dominant triple bond character instead
of the stabilized resonance double bond character for unknown reasons. Also,
the electron withdrawing effect of the visinal hydroxyl group to the diazo
functional group may make the diazo bond somewhat stronger and result in the

higher frequency adsorption for diazo functional group in the IR spectrum.

104

 

".A

E .
.(q

 

The new peak at 1400 cm‘1 can be assigned to a NO vibration band. In
the diazotization reaction, this NO species reacts with amine group of the
aminoacid ester and generates the diazo functional group. In an effort to
decrease the NO band intensity and increase the diazo band intensity, the
reaction was carried outwith longer reaction time (1hr, 1hrs 30minutes, and 2hrs).
However, this did not increase the diazo band intensity, nor did it decrease the
intensity of the 1400 cm'1 band. Thus, the 1400 cm'1 band is unlikely to be
associated with the formation of a diazo intermediate and may simply be NO

adsorbed in the pores of the silica.

105

1200

 

 

 

 

 

E 1000
p—
in
2’ 800
n
E
3
'u 600
o
g 400
in
'u
in
0 200
E .
2 .
>° 0,
-2001 l ' i J
0.2 0.4 0.6 0.8 1
0.1 0.6

P

o

on
9
0|

9
g

P
8
(wu - Blcwo) awnIOA 910d ap/Ap

.0
O
&
9 P
N (a)
uondiosea

P
o
N

Adsorption
dVIdD pore volume (cm3/g - nm)
9

 

0 20 40 60 80 100 120 140

Pore size (nm)
Figure 4.1. N2 adsorption-desorption isotherm of 30% GPTS functionalized MSU-F
prepared by direct assembly method (upper panel) and the cell size and window size
distributions (lower panel) obtained by fitting the adsorption desorption data, respectively,

to the BJH model.

106

 

 

 

 

W

 

I I I I I I I I

0 -10

Chemical Shift (ppm)

-1

I I I T 1 I I

50 -20

Figure 4. 2. 298i solid state MAS NMR spectrum of 30% GPTS functionalized MSU-F

foam silica.

Table 4.2. 298i solid state MAS NMR data for 30% GPTS functionaized MSU-F foam

silica. The degree of functionalization is determined by the ratio between the integral of

the T3 resonance intensity to the total integral intensity (Ta/(Q4+QS+T3)).

 

 

 

 

Integrated peak areas Degree of
T3 03 Q4 functionalizaton
30% GPTS MSU-F 65.2 79.0 82.7 29%

 

 

 

 

 

107

 

aims. 8.2.2382
cum:V 9 33553 826 E mumEv Enos. “Toms. 85.90382 memo e\eom co 87202.8 833ch 9: Lo.— mbooaw m. .2. 9:9”.
ATE“: ..onE::o>m>>

cow ocNF ccmw occm och och comm coon coo—V

 

o
:0 O\
U/\ _ o/ .50.; a -
:o o/\/\_ m ol 3 E u on
o:_u>_m-uoEu_ b can... co So o/o
3.10 I—
0.00 Jaw 0\u.u@EO.—< COP m
o
o __m / S
O /\/\ severe W.
35253333 1—\ cm_. W.
sax/fie maul s
O/\/\_ mxo a
cow

mam CZEVWI

.JHJQO POU<O/\/\_O_w/WH

cmN

108

4.4.2. Diazotization of Aminophenyl functionalized MSU-F sillca

Up to this point, several approaches to the diazotization of MSU-F foam
silica have been investigated starting from organosilane groups integrated into
the pore walls. However, due to the complexities of the reaction steps, the yields
of diazo functional groups were low. The largest degree of diazotization was
achieved starting with 30% NHNMPTS MSU-F. (0.53mmole/g, Table 3.6 in
Chapter 3, In the latter case, the functional group density was 4.53 mmol per
gram.) A simpler pathway to diazotization is needed. Accordingly, a new
organosilane, namely, p-aminophenyltrimethoxysilane (APTS) group was used
as a reaction precursor to 30% APTS-functionalized MSU-F. The diazotization of
p-aminophenyl group is a well-known simple one-pot synthesis. Also, the
delocalized pi electrons of the diazo group and benzene might help to stabilize
the diazo functionality.

For the characterization of 30% APTS functionalized MSU-F, the presence
of mesopores was verified by N2 adsorption-desorption isotherms. The surface
organic groups involved in each reaction step were characterized by IR
spectroscopy.

Figure 4.4 displays the nitrogen adsorption-desorption isotherms for 30%
p-aminophenyl MSU-F foam silica. The isotherms for the soxhlet-extracted
material shows an abrupt uptake of N2 adsorption at the relative pressure 0.9
~1.0 due to capillary condensation within the mesopores of the silica framework.
This sharp adsorption step at partial pressures (P/Po) greater than 0.9 is also

indicative of presence of large pores, greater than 20nm in diameter. The lower

109

panel of Figure 4.4 illustrates the cell size distribution obtained by fitting the
adsorption isotherm to the BJH model. Most of the cells are 20 — 60 nm in
diameter. Additionally, this material has a relatively small window size (8.4 nm,
obtained from the desorption isotherm) in comparison to the cell size, which can
be classified as a “closed cell” structure. The BET surface area is 349 mzlg, and

pore volume is 0.88 cm3/g.

110

600

 

 

E?

5 500 +Adsorption

$3 +Desorption

E 400

3.

B

_n 300

L

o

%

m 200

d)

E

2 100

o

>

0
0 0.2 0.4 0.6 0.8 1
Relative Pressure (P/Po)
0.04
0.014 +Pore (Cell) size
.. 41 nm *Wi d . 0.035 Q

E 0.012 n owsrze <
C \

. 0.03 a.
{D U
M 0.01 “r '5
E 13 nm 0.025 c

u 'I

5 "’ 0003 m 9

3 g ' 0.02 5 w

a. — O

3 2 g a

.3 g °'°°5 0.015 m g-

< 2 5;; 3
0.004

g 0.01 3...
0 a?
g 0002 0.005 3
'0 i

o o

0 20 40 60 80 100
Pore Size (nm)

Figure 4.4 N2 adsorption—desorption isotherms for 30% aminophenyl functionalized
MSU-F silica prepared by direct assembly (upper panel) and the cell and window size
distributions obtained from the adsorption and desorption branches of the isotherms,

respectively (lower panel).

111

Figure 4.5 compares the infrared spectra of 30% aminophenyl
functionalized MSU-F foam silica and its diazotized reaction product. An intense
diazo stretching vibration is observed at 2273 cm“. This band is much stronger
than previously observed diazo stretching vibrations. The position of the band at
2273cm'1 is consistent with the presence of a positive net charge on the diazo
group with the triple bond character.

The diazo-functionalized MSU-F silica was allowed to react with a simple
organo phosphate (methyl phosphate ammonium salt) in order to verify the
accessibility of the diazo group. The reaction between organo phosphate and
diazo functionalized MSU-F silica is based on Sandmeyer-Gatterman aromatic
substitution reaction.13'“ As shown on Figure 4.6, the 31P solid state MAS NMR
confirms the presence of covalently bonded methyl phosphate. The strong peak
at -7ppm is assigned to the bonded phosphate and the two other peaks
separated by 4 kHz are spinning side bands. In solid state NMR, in order to
reduce the effect of anisotropy, the rate of sample spinning should be faster than
the anisotropy interaction. However, when the anisotropy is large, as with
covalent binding, the sample spinning rate may not be sufficient to remove the
spinning side bands.11 Therefore, we can assume the anisotropy of the
phosphate group is large due to the covalent binding of methyl phosphate to the
mesostmcture, which results in the spinning side bands at -32 ppm and 18 ppm,
in agreement with the spinning rate of 4 kHz. The low signal to noise ratio is
result from the low concentration of phosphate group in the material. The

assignment of the -7 ppm resonance to covalently bonded organophosphate is

112

further confirmed by the chemical shift in comparison to ionically bonded
organophosphate. As shown by the spectrum in Figure 4.7 and 4.8, the shifts for
[P03(OCH3)]2‘ ions character on APTS—MSU-F silica and pure MSU-F silica
occur at 1.5 ppm and 1.7 ppm, respectively. These values are consisted with the
shift (2.2ppm) reported previously for organophosphate dianion."5 In this
electrostatic interaction between phosphate and amine group, methyl phosphate
mostly bound to ammonium species instead of aniline on APTS because of the
low pKa value of anilinium ion species compared to alkylamonium. (pKa of
anilinium ion:4.6, pKa of alkylammonium:10.6) This is the reason why only one
chemical shift of phosphorous was shown in Figure 4.7. The weak NMR
resonances for the covalently formed organophosphate in Figure 4.6 are a
consequence of hydrolysis that occurs when the reaction product is worked with
dilute HCI in order to remove ionically bound organophosphate. If the reaction
product is not washed with dilute HCI, then a much more intense NMR spectrum
is obtained, as shown in Figure 4.9. The spectrum contains intense resonance
lines for both covalently bound and ionically bound organophosphate, along with
a third resonance at about 23ppm, which is unassigned. The dramatic difference
in intensity between the spectra in Figure 4.6 and Figure 4.9 indicates that both
ionically bound and covalently bound organophosphate are removed by the acid

wash.

113

BEE Eavow 53> cozomm.

3 oumE o>=m>tou nouzonE mcficoammtoo 05 van 85m ”.522 “03:98:82 _>cocaoc_Em 02 98on beats .m.v 059n—

rhbv .maEacmSES

 

 

 

 

   
   

OOOH OOON OOOm OOOG
came 0 a . . .
_ s a o s - 8
AninU . / AHMHIIU I.
”52:02 ON... mn- urn—:92 m
s, . ow u
5.. ADV—+2 S
.5 E2535 w.
/ / 4 8 m
u
m
i on Mm»
eastscofioza _
i OOH
2530050
.. ON
3:3

   

gum—=92.

 

was. i _Eufiossi

 

3. _._.2

O
to
(%) asueugwsueu

 

OOH

 

114

 

 

 

 

 

FlTllillllllllllllllllll'lllllll‘lllllllll

250 150 100 50 0 -50 -100 -150 -200
Chemical Shift (ppm)

Figure 4.6. 31P Solid state NMR spectrum of methyl phosphate covalent bonded to MSU-
F foam silica wall by Sandmeyer—Gattermann aromatic substitution reaction after
hydrochloric acid washing. The two peaks separated by 4kHz from the line at -7ppm are

spinning side bands.

115

1.5 ppm

 

 

AA— A; _A A ‘A—A- J
V' — ——" vvvv WV‘vv
' 1

I" V V *f I T~I"’_V Y I y Y 7 V— ' V V

50 0 50
Chemical Shift (ppm)

 

Figure 4. 7. 3‘P Solid state NMR spectrum of methyl phosphate on 30% aminopheyl
functionalized MSU-F. The organophosphate anion was introduced into aminophenyl-
functionalied MSU-F silica with the cyclohexyl ammonium salt of [P03(OCH3)]2' in

methanol solution.

116

1.7 ppm

 

 

 

250 0 250
Chemical Shift (ppm)

Figure 4. 8. 31P Solid state NMR spectrum of methyl phosphate electrostatically bonded
to cyclohexyl ammonium on pure MSU-F. The organophosphate anion was introduced
into MSU-F silica with the cyclohexyl ammonium salt of [P03(OCH3)]2' in methanol

solution.

117

 

 

 

 

 

 

 

 

 

 

 

 

 

4kHz 4kHz 4kHz 4kHz
0.5 ppm '5 ppm
l| .
i
I I l I l l I I l I T I
50 0 - 50
Chemical Shift (ppm)

Figure 4.9. 31P Solid state NMR spectrum of methyl phosphate covalent bonded

to MSU-F foam silica wall by Sandmeyer-Gattermann aromatic substitution

reaction. The four peaks separated by 4kHz as shown above is the spinning side

bands of covalently bonded methyl phosphate species at -7 ppm.

118

4.4.3. Immobilization of F0“ on a succinate-functionalized MSU-F silica

foam

Two succinic anhydride functionalized foams of MSU-F silica were
synthesized by grafting (30% functionalization) and direct assembly (10%
functionalization) methods, using 3-(triethoxysilyl)propylsuccinic anhydride
(SATS) as the organosilane. The physical properties of those foams were
characterized by N2 adsorption-desorption isotherms. Further organic
modification of the grafted derivative, such as the ring opening reaction of
succinic anhydride, were characterized by IR, TEM and EDX.

Figure 4.10 provides the N2 adsorption-desorption isotherms for 10%
SATS functionalized MSU-F silica made by direct assembly. As shown in Figure
4.10, the isotherm exhibits a Type IV isotherm with capillary condensation
occurring between 0.9 and 1.0 relative partial pressure (P/Po), which reflects the
very large and broad cell size (37nm) as illustrated in the lower panel of Figure
4.10. Also, this material shows a small adsorption step around P/Po ~0.9, which
corresponds to secondary cell size having a diameter of 16 nm. The window size
obtained from the desorption isotherm is 6 nm, which is the characteristic ratio for
a “closed cell” type of mesocelluar foam. The BET surface area is 423 mzlg, and
pore volume is 1.47 cm3/g. The textural properties of pure MSU-F silica also
were determined by N2 adsorption-desorption methods. As shown on Figure 4.
11, this silica has very narrow window and cell sized distributions, and very large
pore volume. The window and cell size were 11.4nm and 21nm, respectively.

The BET surface area is 517 mzlg, and pore volume is 2.23 cm3/g.

119

 

1000

 

 

 

 

 

 

 

 

 

 

 

 

E
l-
"’_ 800
2’
E f
3 600
'0
0
.D
5 / I
g 400 l
a /
0
g 200
—o M
>
o

0 0.2 0.4 0.6 0.8 1
Relative Pressure (PlPo)

P
a

 

6nm

I

9 .o
w -h
0'!

P
o
00
I
.°
00

- 0.25

P
o
a:

I

P

N

0.04

I l
.° P
d —L
0|
uondiosaa

0.02

O
"o
0|
(uiu - iii/cuts) auimon aiod ap/Ap

 

 

Adsorption
dVIdD pore volume (cm3/g - nm)

O

_ ; ‘0
0 20 40 60 80 100

Pore size (nm)

 

Figure 4.10. N2 adsorption-desorption isotherms for 10% succinic anhydride (SATS)
functionalized MSU-F silica prepared by direct assembly method (upper panel) and the
cell and window size distributions obtained form the adsorption and desorption isotherm

(down panel), respectively.

120

1 500

 

 

8
O
0

§

 

Volume adsorbed (cm‘lg STP)

 

 

 

 

 

 

#4)
WM”

0
0 0.2 0.4 0.6 0.8 1
Relative Pressure(P/Po)

 

Pore Size Distribution

9
s

Desorption (11.4 nm)

9
a

P
N
in

Adsorption (21 nm)

Pore Volume (cmalg)
O
L. 9
0| N

P
.n

P
o
on

 

0 20 40 60 80 100 120 140

Pore Diameter (nm)
Figure 4.11. N2 adsorption-desorption isotherms for pure MSU-F silica (upper panel) and

the cell and window size distributions obtained from the adsorption and desorption
isotherms, respectively. This silica was used to prepare a 30% succinic anhydride

functionalized derivative by the grafting method.

121

 

Each reaction step in the modification of the surface was monitored by FT-
IR spectroscopy. The FT-IR spectra in Figure 4.12 confirm the presence of the
acidic carbonyl group formed by the ring opening of succinic anhydride. The
immobilized Fe3" species in the MSU-F silica foam. The spectrum of succinic
acid-MSU-F (upper panel in Figure 4.12) shows a sharp and intense absorption
peak of the acid carbonyl group at 1722 cm". When this material was mixed with
aqueous ferric chloride, the obtained solid exhibited a new carbonyl stretching
band at 1590cm", which is due to the coordination of Fe (ill) to the carboxylate

finkages.

In order to verify the stability of the mesostructue during the grafting
process and the presence of Fe(lll), the transmission electron microscopy (T EM)
image and the energy dispersive X-ray spectrum (EDX) shown in figure 4.13
were obtained. The upper panel in Figure 4.13 shows the bright field TEM
images of the mesoporous foam silica. The image shows a very uniform cell size
(20nm), in accord with size obtained by N2 adsorption. From this image region,
EDX analysis was performed. The bottom panel in Figure 4.13 shows the
characterized peaks of Fe, which verifies immobilization of Fe (ill) on the succinic

acid functionalized MSU-F silica.

122

"1‘

§ .._

.cozmo Econ noxwano mEEmEoo o>=m>tou

mcficoammtoo 05 .60 ucm w._.<w ho genome. mcEEm E umsmama mo___w ”saws. uoN__mco:ocE.$mc_oo:m ..os. 86on m: .96 0.39”.

$.83 33.235535

 

 

 

 

occa ooo~ ooom ooov
..._>/ a . . 2 u
\ ,.._ 8.:Emperoemocsot m
s. 8 s
.. 1 w
LC m...
g/A/ NNNH um?_umo__>xo€mu. . on m
a
K?\.. - s w
u- as. $5583.- :58 _///
i}. 2:
L EEO “3:0 - om mA..
- as u
m.
- cm W
m
- Ow Mm
”Tam—2 I Sum 3583 l.

 

 

 

OOH

123

 

 

7 .. .. _i.-- ,i.
0 2 4 6 8keV
Figure 4.13. TEM image (upper panel) and EDX spectrum (lower panel) for Fe(lll)

succinate functionalized MSU-F silica prepared by grafting method.

124

 

 

4.4.4. Conclusion.

In this chapter, three different mesoporous organosilica syntheses have
been carried out for potential proteomic separations in the future. Unexpectedly,
the first GPTS functionalized mesostnicture showed a low concentration of
immobilized diazo species as judged by the IR spectra in Figure 4.8. This result
reflects low yield involved in the multi—step reaction sequence. However, the
simple one-pot synthesis of immobilized diazo-phenyl groups and Fe3+
carboxylated groups were successfully performed. The one pot synthesis of
diazophenyl group showed a very intense diazo vibration at 2273cm". in
addition, the resulting derivative provided good covalent coupling to methyl
phosphate. Lastly, the Fe3+ foam silica, containing immobilized Fe (III)
carboxylate groups may be suitable for phosphoprotein enrichment by the most
popular method IMAC. Both materials may prove to be good candidates for

future studies of phosphorylated peptide enrichment.

125

ET". - $9!!an
4‘

 

4.5. Reference

(1) Novabiochem Catalog and Peptide synthesis Handbook,1997l1998, method
16.

(2) MilliGen Techniccal Note 3.10.

(3) Veterans at Risk: The Health Effects of Mustard Gas and Lewisite, institute of
Medicine 2005, 72

(4) Tanabe Seiyaku Company Limited, 681336460 1973

(5) Pan, C., Ye, M, Liu, Y., Feng, S., Jiang, X., Han, G., Zhu, J. and Zou, H.
Journal of Proteome Research 2006, 5, 11, 3114-3124

(6) Mier—Vinue, J., Montana, A. M., Moreno, V., and Prieto, M Z. Anorg. Allg.
Chem. 2005, 631, 2054-2061

(7) McCarthy, J. R., mcCowan, J., Zimmerman, M. B., Wenger, M. A., Emmert, L.
W. J. Med. Chem. 1986, 29, 1586-1590

(8) Hoffmann, H. M. R., Brandes, A. Tetrahedron, 1995, 51, 1, 155-164

(9) Lansdell, T. A., Tepe, J. J. Tetrahedron Lett. 2004,45, 91-93

(10) Lehman, J. W. Operational Organic Chemistry 1999

(11) Hodgson, H. H., Clifford, K. Journal of the Chemical Society 1942, 581-583

(12) Dunn, J. 0., Watson, J. T., Bruening, M. L. Anal. Chem. 2006, 78, 1574-
1580

(13) Sandmeyer T., Ber., 1884, 17, 1633; 1890, 23, 1880.
(14) Gatterman L., Ber., 1890, 23, 1218.
(15) Duer, M. J. Introduction to Solid-State NMR spectroscopy 2004, 62-66

(16) Benoit-Marquie, F., de Montety, C., Gilard, V., Martino, R., Maurette, M. T.,
Marlet-Martino, M. Environ. Chem. Lett. 2004, 2, 93-97

126

 

 

Chapter 5
Phosphorylated protein enrichment based on diazo funtionalized meso-

porous silica film

5.1 . introduction

Protein phosphorylation is one of key mechanisms in biological regulatory
process, such as cell proliferation, differentiation, and apoptosis in a signaling
process."3 In order to identify phosphorylated peptides and their sequences,
mass spectrometry has emerged as a powerful analytical technique.4 However,
due to the low abundance of phosphorylated proteins and variations of
phosphorylation sites, the analysis of phosphorylated proteins is quite
challenging. Therefore, to improve the analysis of phosphorylated proteins, a
numbers of recent studies have been focused on separation and enrichment
methods of phosphorylated proteins combined to mass spectrometry analysis.
So far the most common technique in the enrichment of phosphorylated proteins
is IMAC (Immobilized Metal Affinity Chromatography).5'6'7' 8 However, its lack of
selectivity for peptides containing acidic amino acid residue, namely Glu and Asp,
and lots of variables of experimental conditions result in impurity with non-
phosphorylated proteins.

For more effective method for the enrichment, recently, covalent bonding
enrichment strategies have been addressed. In this early study, the B-
eliminations of phosphate groups followed by the conjugation of thiol groups can

(I 9,10

lead to covalent bindings on solid phase bea However, there are few

127

limitations. First of all, unprotected cystein and methionine residue can be
involved in unwanted side reaction. O-linked sugar moiety can undergo the B-
eliminations reaction and lead to complication of enrichment result. In addition,
this method excludes the enrichment of phosphotyrosine species. Therefore,
alternative method and strategies were still needed.

In order to overcome this limitation, Tepe et al.11 in 2004 established a
new type of solid support that contains d-diazo functional group on the Wang
resin surface. In this method, o-diazo groups can be easily replaced by
phosphate group in phosphorylated protein enrichment process.

In 2006, Pflum et al.12 also published new covalent binding enrichment
method. By using glycine conjugate Wang resin, she bound phosphate groups in
phosphorylated peptide to amine on Wang resin, which leads to
phosphoramidates on Wang resin.

Above two cases, they have advantages in the direct and simple
enrichment of phosphorylated protein without any phosphate group modification.
However, due to the swelling characteristic of polymer resin, Surface functional
group can be only accessible in a certain solvents.

In the study of new enrichment strategies to enhance the selectivity of
phosphortylate enrichment, we applied new d-diazo functionalized high surface
area silica films which contain nanometer size pores. This q-diaz'o functionalized
mesoporous silica films have some advantages in this application. First of all,
due to the considerably large surface area, it generates high binding capacity.

These materials also have open framework structures, which lead to free access

128

rm“:

of phosphorylated protein on the surface of a-diazo functional group without any
solvent limitation. In addition, because this method is a direct on-plate
enrichment method, the enrichment process can be facilitated with minimizing
the sample loss during process. In this chapter, I will describe how to prepare
new types of materials and show the very efficient, selective enrichment method
of phosphorylated proteins, which can have strong potential in phosphorylated

protein analysis.

5.2 Experimental
5.2.1 Reagents

The amine surfactant, dodecylamine, was purchased from Aldrich for the
synthesis of mesoporous film. As silicate source, tetraethylorthosilicate (T E08)
and triethoxysilylpropionitrile were also obtained from Aldrich. For the tryptic
digestion, trypsin was obtained from Promega. Bovine B-casein and chicken egg
ovalbumin were purchased from Sigma. Other reagents used in the tryptic
digestion, like Tris-HCI, urea, 1, 4-dithio-DL-threitol, iodoacetamide and
ammonium bicarbonate, were purchased from lnvitrogen, J.T. Baker,
BioChemika, Sigma, and Columbus Chemical Industries, respectively. HPLC
grade acetonitrile, anhydrous methanol, trifluoroacetic acid, acetyl chloride, 2,5-
dihydroxybenzoic acid, hydrochloric acid, and 1% aqueous phosphoric acid were
obtained from Aldrich. Acetic acid, ammonium hydroxide, hydrogen peroxide

and sulfuric acid were obtained Spectrum. All the above reagents were used

129

without further purification. Absolute ethanol was purchased in-house, and water

was purified by a double-exchanged Millipore filter apparatus.

5.2.2 Synthesis of diazo functionalized mesoporous film

The diazo functionalized mesoporous silicate film was prepared by direct
assembly method from triethoxysilylpropionitriIe. Dodecylamine (1.2 g) was
dissolved in the mixture of ethanol (13.3 g) and water (11.2 9). Then, the mixture
of 4.0 g tetraorthosilicate and 1.065 g triethoxysilylpropionitrile were added to the
surfactant solution at an ambient temperature. After 1 minute, additional 111 g of
ethanol were added to the previous prepared solution.

Before functionalized mesoporous silica film synthesis, silicon substrates
(1.5 cm x 1.5 cm) were cleaned with 100 mL of 75 % sulfuric acid with hydrogen
peroxide for 2 hours at 98 °C. These cleaned silicon wafers were spin-coated by
the previous prepared silicate solution. These films on the wafers were aged for
3 days at room temperature.

For the further modification from nitrile to carboxylic acid, nitrile
functionalized mesoporous silicate wafer was added to 36 g of 55% of sulfuric
acid solution and heated 92~95 °C for 1.5 hours. These wafers were washed
with copious amount of water and dried with nitrogen. For the diazo
functionalization, 0.085 mL of anhydrous N,N-dimethyl formamide in 30 mL of
dried dichloromethane was added on the previous prepared vacuum dried
carboxylic acid functionalized mesoporous film, which was followed by addition

of 1.42 mL of oxalyl chloride. This reaction mixture were stirred at 0 °C for 30

130

minutes and then stirred at 25 °C for 1 hour. After this reaction, the mesoporous
film on silicon substrate was transferred to another container under the nitrogen
condition. On this film, 6 mL of tetrahydrofuran was added, which was followed
by 3 mL of trimethylsilyldiazomethane at 0 °C. After 2 hours, the diazo

functionalized mesoporous film on silicon wafer was dried under nitrogen.

5.2.3 Protein digestion

100 pg of proteins were dissolved in 20 pL urea (6M)/Tris-HCI (50mM). 5
p L of 10mM 1,4-dithio-DL-threitol was added on the protein solution and heated
at 65 °C for 1 hour. After cooling, 160 uL of 50mM ammonium bicarbonate was
added to the protein solution, which was followed by addition of 10 pL of 100 mM
iodoacetamide. The result solution was kept for 1 hour in the dark.
Subsequently, 10 uL of 0.5 ugluL trypsin was added to the protein solution, and
the protein solution was incubated for 16 hours at 37 °C. The digestion solution
was quenched by the addition of 11 pL of acetic acid, which was stored in a -70

°C freezer until use.

5.2.4 Methyl esterification of protein

Tryptic digested protein was dried by using Speedvac. 200 uL of
methanolic HCI (320 uL of acetyl chloride in 2 mL of anhydrous methanol) was
added to the protein digested. The final solution was placed for 2 hours at room

temperature, and dried by Speedvac.

131

5.2.5 Phosphorylated protein enrichment

For the enrichment of phosphorylated protein, the dried methyl esterified
and non-methyl esterified phosphorylated peptides were diluted with the mixture
of 5% acetic acid and H20 (vzv, 1:1) to make 10 uM of tryptic digested protein
solution. The 1 uL of prepared digested protein solution was applied in 2 mm
circular trench on diazo functionalized mesoporous film. For 1 hour, water was
kept adding to prevent from drying of protein solution. After 1 hour, resulting
silicon wafer was rinsed with 20 mL of acetic acid, acetonitrile, and H20 mixture
solution.(v:v:v, 3:30:67) For the dissociation of phosphate bond from diazo
functionalized silicon wafer, 2 uL of NH4OH (pH11) was droped on the film in the
process of methyl esterified phosphorylated peptide enrichment. 0.7 uL of matrix
solution (2mg of 2.5-dihydroxybenzoic acid, 100 uL of 1% phosphoric acid, and
100 uL of acetonitrile) was applied on the 2mm circular trench. The final wafer

was attached onto the modified MALDI plate by double side tape.

5.3 Physical Characterization
The organo-functionalized mesoporous silica films were characterized by
powder x-ray diffraction analysis, reflectance-infrared spectroscopy, and
transmission electron microscopy. In addition, the enrichment ofphosphorylated
protein is confirmed by the matrix assist laser desorption ionization. (MALDI)
Powder XRD data were collected on a Regaku Rotaflex Diffractometer
with CuKo radiation (A = 1.542A) which is generated at 45kV and 100mA. The

diffraction data were collected from 1.5 degree to 10 degree of 29 with an

132

 

 

 

increment of 0.02 degree. To confirm the chemical transformations in each step
from nitrile to diazo groups on mesoporous film surface, Nicolet Protg 400 Magna
reflectance infrared spectrometer equipped with a Barnes analytical/spectra-tech
diffuse reflectance accessory was used. Each spectrum was collected in the 700
to 400 cm'1 spectral region. The number of scan was 16 and data spacing was
1.928 cm".

To verify the pore morphology on organo-functionalized mesoporous silica
film, TEM images were obtained on a JOEL 2200FS instrument with an
accelerating voltage of 200kV. TEM samples were prepared dipping the cupper
grid in the solution used in the film synthesis.

For the identification of enriched phosphorylated pepetides, MALDI linear

ion trap mass spectrometer (Therrno vMALDI LTQ XL) has been used.

5.4 Result and discussion
5.4.1. X-ray diffraction analysis

Figure 5.1 shows the low angle XRD reflections of nitrile functionalized
mesoporous silica film on silicon wafer. The presence of a low angle diffraction,
which is 3.8 nm in d spacing, indicates the average pore-pore distance for a

worrnhole structure.

5.4.2. Reflectance infrared spectroscopy
Figure 5.2 illustrates the IR spectra between 4000 and 700 cm'1 of nitrile,

carboxylic acid, and diazo functionalized mesoporous silica films. In all spectra,

133

the peaks around between 1000 and 1200 cm’1 represent the stretching vibration
of the Si-O-Si bonds. The broad bands at 3400 cm'1 show the hydrogen bonding
interaction of silanol groups in mesoporous silica films. The stretching vibrations
of the C-H centers on organic moieties of each functional group are shown at
2800~2900cm‘1. In addition, the each reaction step was confirmed by the
presence of specific IR adsorption vibration, namely, nitrile group at 2275 cm",

carboxylic acid group at 1720cm", diazo group at 2107 cm".

5.4.3. Transmission electron microscopy

The worrnhole framework structure is confirmed by TEM shown on Figure
5.3. The pore to pore distances for wonnhole structure from TEM correspond to
the value obtained from the d spacing value from the XRD analysis. In the image
of Figure 5.3, the dark and light regions well represents the walls and pores of

the typical worrnhole mesoporous materials, respectively.

5.4.4. Matrix assist laser desorption Ionization mass spectrometry analysis
In order to examine the selective binding of phosphorylated peptides onto
diazo funtionalized mesoporous silica thin films, three different protein samples
has been studied to verify the enrichment, such as tryptic digested ovalbumin,
methyl esterified tryptic digested ovalbumin, methyl esterified tryptic digested B-
casein. Because of their well know phosphorylation sites of three samples, we

were able to confirm them easy on MALDI mass spectra.

134

Tryptic digested ovalbumin

Tryptic digested ovalbumin has three well know phosphorylated peptides
fragments, namely, mlz 2089 (EWGpSAEAGVDAASVSEEFR), 2512
(LPGFGDpSlEAQCGTSVNVHSSLR), and 2903 (FDKLPGFGDpSlEAQCGTSV—
NVHSSLR).

As shown on Figure 5.4a, the tryptic digested ovalbumin without
enrichment showed three monophosphorylated peptides fragemtns at mlz 2089,
2512, and 2903 among complex peptide fragments. After the enrichment by
diazo functionalized mesoporous silica thin films, phosphorylated peptide
fragments of ovalbumin are successfully enriched down to 10 pmole level. In
Figure 5.4b, it showed the strongest peak at mlz 2089, which represents
selective enrichment of phsophorylated peptide. However, due to the low
concentration of binding or difficulty in dissociation of covalent binding after
enrichment, the signal to noise ratio is quite high. Moreover, the peaks at mlz
2512 and mlz 2903 were too small to detect it. The reason of the enrichment
difficulty of these fragments might be low ionization efficiency in detection and

unfavorable hydrophobic character of high weight peptide fragments.

Methyl esterified tryptic digested ovalbumin

To eliminate the possible binding with carboxylic acid moiety, tryptic
digested ovalbumin was methyl esterified. As shown on figure 5.5a, the methyl
esterified tryptic digested ovalbumin without enrichment represents three

monophosphorylated peptides fragments at m/z 2173 (EWGpSAEAGV-

135

DAASVSEEFR), 2554 (LPGFGDpSlEAQCGuTSVNVHSSLR), and 2960
(FDKLPGFGDpSIEAQCGTSVNVHSSLR). In Figure 5.5b, the spectra showed
better selective enrichment compared to non-methylated sample. However, it is

still difficult to find high mass fragments, such as mlz 2554 and mlz 2960.

Methyl esterified tryptic digested B-casein

In order to generalize the covalent enrichment method, methyl esterified
tryptic digested B-casein were also tested. In Figure 5.6a, it didn’t show clearly
three possible phosphorylated peptides fragements, namely, mlz 2160
(FQpSEEQQQTEDELQDK), mlz 3078 (ELEELNVPGEIVEpSLpSpSpSEESITR),
mlz 3235 (RELEELNVPGEIVEpSLpSpSpSEESITR). As shown on Figure 5.6b,
however, three possible fragements are shown distinctly by comparing to non
enrichment methyl esterified tryptic digested B-casein spectra. This result
represents the successful enrichment of methyl esterified tryptic digested [3-

casein on diazo functionalized mesoporous silica thin film.

5.5 Conclusion

we firstly and successfully applied high surface area mesoporous silica
thin films with selective covalent binding of phosphorylated proteins as a new
phosphorylated peptides enrichment method. This covalent enrichment method
by high surface area thin film showed selective and facile enrichment by
comparing to other enrichment process. The results in this study show the

strong potential in the enrichment and analysis of phosphorylated proteins.

136

1400

 

— A. HMS Film on Silicon wafer (thin)
-- B. HMS Film on Silicon wafer (thick)

1200

1000 '

800 '

600 '

Intensity (A.U.)

200 '

 

 

 

1.5 3.5 5.5 7.5 9.5

29

Figure 5.1. Powder X-ray diffraction pattern for HMS film on silicon wafer.
(Structure direct agent was removed by soxhlet ethanol extraction.)

137

 

45

40 -
.~
U c
35 '
30 -
A
e\"
V
3
‘: 25 -
I0
.35." b
E 20-
,, 0
C
E
'—

15 '

 

 

 

 

0 - . .
3700 2700 1700 700

wavenumber(cm'1)

Figure 5.2. IR spectra obtained in each reaction step for the diazo functionalized
mesoporous silica film synthesis. a) nitrile functionalized mesoporous silica film,
b) carboxylic acid functionalized mesoporous silica film, c) diazo functionalized
mesoporous silica film. Circles indicate the characteristic vibration of functional
group. Each spectrum is offset by 15.

138

 

Figure 5.3. TEM image for organo-functionalized (COOH) mesoporous silica film

139

100 b * 100% = 429

 

 

 

     

100- a 100% = 6606

l
0 ‘ Inn-LMMJ “Agni-d hi.i.|lLflfldI1A.ni .4. IL. A A. .4.

Relative Abundance (%)

50-

 

 

 

 

*
o: l... .

1000 1500 2000 2500 ' 3000 I 3500
mlz

 

 

Figure 5.4. MALDl mass spectra of 10 pmole of tryptic digested ovalbumin. a)
conventional MALDI spectra of ovalbumin (no enrichment), b) ovalbumin
enrichment on diazo mesoporous silica film. (The asterisks indicate the
phosphorylated peptide fragments.)

140

100 b * 100% = 394

 

 

 

100- a 100% = 2537

Relative Abundance (%)

50-

 

 

 

ill

 

o ‘ .. , . '
1000 1500 2000 2500 3000 3500
mlz

Figure 5.5. MALDI mass spectra of 10 pmole of tryptic digested methyl-esterified
ovalbumin. a) conventional MALDI spectra of methyl esterified ovalbumin (no
enrichment), b) methyl esterified ovalbumin enrichment on diazo mesoporous
silica film. (The asterisks indicate the phosphorylated peptide fragments.)

141

 

100.. b 100% = 2996 *

 

 

 

 

 

Relative Abundance (%)

 

 

 

 

10°“ a 100% = 3646
50.
0
1000 1500 2000 2500 3000 3500
mlz

Figure 5.6. MALDI mass spectra of 10 pmole of tryptic digested methyl esterified
B-casein. a) conventional MALDI spectra of Bocasein (no enrichment), b) methyl
esterified fi-casein enrichment on diazo mesoporous silica film. (The asterisks
indicate the phosphorylated peptide fragments.)

142

5.6 Reference

(1) Koch,C. A., Anderson, 0., Moran, M. F., Ellis, C., Pawson, T. Science, 1991
252,668.

(2) Pawson,T., Scott, J. D. Science, 1997. 278, 2075.

(3) Hunter, T. Cell, 2000, 100, 113.

(4) McLachlin, D. T., Chait, B. T. Curr. Opin. Chem. Biol, 2001, 5, 591.

(5) Porath, J., Carlsson, J., Olsson, |., Belfrage, (3. Nature, 1975, 258, 598.
(6) Andersson, L. Porath, J. Anal. Biochem, 1986, 154, 250.

(7) Cao, P. Stults, J. T. J. Chromatogr. A, 1999, 853 225.

(8) Zhou, W., Merrick, B. A., Khaledi, M. G., Tomer, K. B. J. Am. Soc. Mass
Spectrom., 2000, 11, 273.

(9) Oda, Y., Nagasu, T., Chait, B. T. Nat. Biotechnol. 2001, 19, 379-382

(10) Meyer, H. E., Eisermann, B., Heber, M., Hoffmann-Posorske, E., Korte, H.,
Weigt, C., Wegner, A., Hutton, T., Donella—Deana, A., Perich, J. W. Faseb. J.
1993, 7, 776-782

(11) Lansdell, T. A., and Tepe, J. J. Tetrahedron Lett. 2004, 45, 91-93.

(12) Warthaka, M., KanNowska-Desaulniers, P., Pflum, M. ACS Chemical Biology,
2006, 1, 11, 697-701

143

3 1293 03062 944

 

Hu-