:‘WNV'QK

"n "u v."- "v“.—

i

I ..
a)... z. :3...

I. ԤL-\t fit
35 waxguu
ariévleQ.” :23! . .

duh. :lhm

. .53). g"

'- i 1.»
.mri. hunk...

. 7r. 3:310:15...
"HE-353.54.
(512:1)...

 

 

 

LIBRARY
Michigan State
University

 

 

 

This is to certify that the
thesis entitled

MASS SPECTROMETRIC CHARACTERIZATION OF ZINC
ACTIVATED PROINSULIN C-PEPTIDE AND C-PEPTIDE
MUTANTS

presented by

ZACHARY KELTNER

has been accepted towards fulfillment
of the requirements for the

MS. degree in Chemistry

 

 

&K//7§7/

Major Professor’s §ignature

muons, 21"”, 200‘]

 

Date

MSU is an Afi‘irmalive 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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:IProj/Aoc8. Pres/C IRCIDateDue, indd

 

 

MASS SPECTROMETRIC CHARACTERIZATION OF ZINC ACTIVATED
PROINSULIN C-PEPTIDE AND C-PEPTIDE MUTANTS

By

Zachary Keltner

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
Chemistry

2009

 

ABSTRACT

MASS SPECTROMETRIC CHARACTERIZATION OF ZINC ACTIVATED
PROINSULIN C-PEPTIDE AND C-PEPTIDE MUTANTS

By
Zachary Keltner
Proinsulin C-peptide has recently been demonstrated to ameliorate some chronic
complications caused by diabetes. Recent experiments have shown that this activity is
dependent upon activation with certain metals, such as Fe(II), Cr(III) or Zn(II). In an
effort to gain a greater understanding of the peptide-metal interactions responsible for this

activity, “top-down” tandem mass spectrometry characterization of proinsulin C-peptide

. . . 2+ .
and various C-peptide mutants in the presence or absence of Zn was earned out. Gas-

phase fragmentation of various protonated precursor ion charge states of C-peptide, as
well as Zn-activated C-peptide, yielded extensive sequence coverage. These results
indicate multiple sites for Zn binding, making it difficult to localize the Zn to a specific
residue within the peptide sequence. However, certain product ions indicated a likely role
for acidic residues in Zn binding. Five single-amino acid C-peptide mutants were
synthesized, by substitution of each acidic residue with alanine. These mutants did not
significantly affect the fragmentation behavior for either the unbound or Zn bound
precursor ions, suggesting that any one acidic residue is not critical for binding. However,
a 50% decrease in the relative Zn binding for the mutants was observed compared to the
wild type sequence. Additional studies on a C-terminal pentapeptide sequence EGSLQ,
as well as an alanine substituted mutant, AGSLQ, were also performed. Substitution of
the Glu residue in this peptide was found to affect both the binding and activity,

implicating a role for the Glu27 residue.

ACKNOWLEDGMENTS

Throughout the time I have spent compiling my research, several people have helped and
supported me. Without them, this would not have been possible. I would first like to
acknowledge my research advisor, Dr. Gavin Reid. His guidance and support was
invaluable to my development as a scientist. He has been there from day one, and has
always provided much needed direction. I would like to thank the members of my
committee, Dr. Gary Blanchard, Dr. David Weliky, and Dr. Daniel Jones for their help
and guidance. I would also like to thank Ms. Jennifer Meyer and Dr. Dana Spence for
their help in conducting the ATP release experiments, as well as for providing sound
advice when it was needed most. Mr. Matt Ruiz has been an invaluable resource for his
assistance in acquiring MALDI spectra. In addition, I would like to thank the other
members of my research group, namely Ms. Amanda Palumbo and Dr. Jennifer Cox for
helping me throughout my time here and listening to my questions, all while becoming
good fi'iends along the way. My friends outside of the Chemistry program have served as
a great support system to me through the entire process. Finally, I would like to
acknowledge my family who has been a great asset in this and everything in my life.
Their patience in helping me deal with my frustration and stress throughout my graduate

career is much appreciated.

iii

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................... vi
LIST OF FIGURES ......................................................................................................... vii
1. CHAPTER ONE: A Biological Role for Proinsulin C-peptide? ................................. 1
1.1 Diabetes Mellitus and its Complications .......................................................... 1
1.2 Insulin and its Roles in Diabetes ....................................................................... 2
1.3 Proinsulin C-peptide ......................................................................................... 5
1.4 Metals and their Roles in Diabetes ................................................................. 10
1.5 Metal-Activated C-peptide .............................................................................. 13
1.6 Metal-Binding Analysis by Mass Spectrometry ............................................. 17
1.7 Aims of This Thesis ........................................................................................ 20
2. CHAPTER TWO: Strategies for the Synthesis and Characterization of
Proinsulin C-peptide, its Mutants, and Metal-Binding .............................................. 21
2.1 Solid-Phase Peptide Synthesis (SPPS) of the C-peptide Mutants .................. 21
2.1.1 Materials .............................................................................................. 21
2.1.2 Synthesis .............................................................................................. 22
2.1.2.1 Deprotection ........................................................................... 23
2.1.2.2 Coupling ................................................................................. 23
2.1.2.3 Ninhydrin Test ....................................................................... 23
2.1.2.4 Final Deprotection/Cleavage ................................................. 24
2.1.2.5 Workup/Extraction ................................................................. 25
2.2 High Performance Liquid Chromatography (HPLC) for the Purification
of C—peptide and its Mutants ........................................................................... 25
2.3 Mass Spectrometry for the Analysis of Peptides and Proteins ....................... 27
2.3.1 Introduction ......................................................................................... 27
2.3.2 Ionization Sources ............................................................................... 27
2.3.2.1 Electrospray Ionization (ESI) ................................................ 28
2.3.2.2 Matrix-Assisted Laser Desorption-Ionization (MALDI) ....... 30
2.3.3 Mass analyzers .................................................................................... 31
2.3.3.1 Quadrupole Mass Analyzer .................................................... 32
2.3.3.2 Quadrupole Ion Trap .............................................................. 37
2.3.3.3 Tandem Mass Spectrometry .................................................. 44
2.3.3.4 Linear Ion Trap (LIT) ............................................................ 45
2.3.4 Peptide Fragmentation Nomenclature ................................................. 47
2.3.5 Instrumentation .................................................................................... 48
2.3.4.1 MALDI-LIT analysis ............................................................. 48
2.3.4.2 ESI-LIT analysis .................................................................... 49

iv

3. CHAPTER THREE: Characterization of Proinsulin C-peptide, its Mutants,

and Binding with Zn2+ ............................................................................................... 51
3.1 Top-down Characterization of Proinsulin C-peptide ...................................... 51
3.1.1 Bottom-up and Top-down Proteomics ................................................ 51
3.1.2 Characterization of Proinsulin C-peptide ............................................ 53
3.2 Characterization of Proinsulin C-peptide Bound to Zn2+ ............................... 60
3.3 Top-down Characterization of the Proinsulin C-peptide Mutants .................. 67
3.4 Characterization of the Zn2+-bound C-peptide Mutants ................................ 70

3.5 Effect of Amino Acid Mutation on the Relative Binding of Zn2+ to C-
Peptide ............................................................................................................ 78

3.6 Characterization and Activity Studes of the C-tenninal Pentapeptide of
C-peptide ......................................................................................................... 81
3.7 Conclusions and Future Directions ................................................................. 88
SUPPORTING INFORMATION .................................................................................... 89

REFERENCES ................................................................................................................ 130

LIST OF TABLES

TABLE PAGE

2.1 Sequences of the single amino acid mutants of C-peptide that were
synthesized, along with their respective abbreviations. Locations of the
alanine substitutions are labeled in bold and underlined on the sequences ............. 22

2.2 HPLC gradient used for the purification of proinsulin C-peptide mutants ............. 26

LIST OF FIGURES

FIGURE PAGE

1.1

1.2

2.1

2.2

2.3

2.4

2.5

2.6

2.7

3.1

3.2

3.3

Structure of proinsulin, showing locations of cleavage into insulin and C-
peptide. Sequences of the A) A-chain, B) B-chain and C) C-peptide are
shown below the structure. ........................................................................................ 3

Sequence homology of C-peptide for 20 mammalian species. The sequence
of human C-peptide is given in the top line, and upper-case letters below
represent sequence conservation in more than one species, while lower-case
letters represent conservation in one species. Letters in boxes indicate
conservation in all or all but one species, while dashes represent gaps in the
sequence. ................................................................................................................... 6

Representation of droplet formation in ESI. ........................................................... 29

Schematic of a quadrupole mass spectrometer, showing applied AC potential

((D0) .......................................................................................................................... 33
Stability diagram for a quadrupole mass analyzer. ................................................. 37
Schematic of a quadrupole ion trap ......................................................................... 38

Mathieu stability diagram for a 3D ion trap showing stability of ions of
different m/z values. Larger balls indicate larger m/z values ................................. 4O

Diagram of Dehmelt pseudopotential well in a 3D ion trap. ERE is the

energy where resonance ejection typically is carried out. ....................................... 42
Nomenclature for peptide fragmentation. ............................................................... 48
E81 mass spectrum of 10 uM proinsulin C-peptide in water. ................................. S4

CID MS/MS spectrum of the [M+2H+]2+ charge state (m/z = 1543.3) of C-
peptide from Figure 3.1. .......................................................................................... 58

Product ion and sequence coveragg summaries obtained from the CID spectra
+ + + + + + +

of the A) [M+H ]+, B) [M+2H ] , C) [M+3H ] , D) [M+4H ] , and E)

combined product ion and sequence coverage summary for all charge states

observed for C-peptide. The [M+H+]+ were acquired using MALDI while
all others were acquired using ESI. Ions with a relative abundance greater
than 25% are labeled on the summary ..................................................................... 59

vii

3.4 A) Mass spectrum of 10 uM C-peptide and 10 uM ZnC12 B) High resolution
UltraZoom scan of the [M+H++Zn2+]3+ charge state C) High resolution
UltraZoom scan of the [M+Zn2+]2+ charge state. .................................................. 61

3.5 CID MS/MS Spectrum ofthe [M+Zn2+]2+ charge state ( m/z = 1542.7) of C-

peptide from Figure 4A. Ions observed to have Zn bound are labeled with T
in bold ...................................................................................................................... 65

3.6 Product ion and sequence coverage summaries obtained fi'om the CID

spectra of the A) [M+Zn2+]2+, B) [M+Zn2++H+]3+, C) [M+Zn2++2H+]4
charge states of C-peptide as well as D) the overall product ion and sequence
coverage summary. \ indicates an unbound b or y ion, \ indicates only

the Zn-bound fragment was found, and \ indicates that both bound and
unbound ions were identified .................................................................................. 66

3.7 CID MS/MS Spectrum of the [M+2H+]2+ charge state (m/z = 1482.6) of
E27A ........................................................................................................................ 68

3.8 Product ion and sequence coverage summaries obtained from the CID

spectra of the [M+2H+]2+ charge state of A) E27A, B) E1 1A, C) D4A, D)
E3A, and E) ElA mutants of C-peptide. Ions with relative abundance greater
than 25% are labelled on the summary. .................................................................. 69

3.9 ESI mass spectra of 10 uM ZnClz and 10 uM A) E27A, B) E1 1A, C) D4A in
H20 ......................................................................................................................... 73

3.10 E81 mass spectra of 10 uM ZnClz and 10 uM A) E3A, B) EIA in H20 ............... 74

3.11 CID MS/MS Spectrum of the [M+Zn2+]2+ charge state (m/z = 1514.44) of
the E27A mutant of C-peptide from Figure 9A. ..................................................... 75

3.12 Product ion and sequence coverage summaries obtained from the CID

spectra of the A) [M+Zn2+]2+, B) [M+Zn2++H+]3+, C) [M+Zn2++2H+]4
charge states of E27A as well as D) the overall product ion and sequence
coverage summary. \ indicates an unbound b or y ion, \ indicates only

the Zn-bound fragment was found, and \ indicates that both bound and
unbound ions were identified. ................................................................................. 76

3.13 Product ion and seqzuegce coverage summaries obtained fi'om the CID
+ +

spectra of the [M+Zn ] charge state of A) E27A, E) E1 1A, C) D4A, D)

BSA, and E) ElA mutants of C-peptide. \ indicates an unbound b or y ion,

viii

3.14

3.15

3.16

3.17

81

S2

S3

S4

SS

S6

S7

\ indicates only the Zn-bound fragment was found, and \ indicates that

both bound and unbound ions were identified ........................................................ 77

Plot showing binding ratios of the [M+Zn2++nH+](n+2)+ to [M+nH]11+ for C-

peptide and the 5 mutants A) Accounting for [M+Na]+ [M+K]+ adducts and
B) not accounting for the salt adducts. Intensities for the 2+, 3+, and 4+
charge states were summed. Standard deviations for the triplicate
measurements are shown on the plot. .................................................................... 80

ESI mass spectra for 10 pM 2n2+ and 10 pM A) EGSLQ and B) AGSLQ ........... 82

CID-MS/MS spectra of the [M+Zn2+]2+ precursor ions of A) EGSLQ and B)
AGSLQ .................................................................................................................... 86

Bar graph showing normalized ATP release in erythrocytes for C-peptide as
well as the EGSLQ pentapeptide and the AGSLQ mutant. Measurements
were taken with n = 7, and error bars are plotted as the standard error of the
mean. A * indicates a statistically significant difference was calculated using
a single sided students t-test (p = 0.025). ................................................................ 87

MALDI CID MS/MS Spectrum of the [M+H+]+ charge state (m/z = 3021.3)
of C-peptide ............................................................................................................. 90
CID MS/MS Spectrum of the [M+3H+] 3+ charge state (m/z = 1007.6) of C-
peptide ..................................................................................................................... 91

CID MS/MS Spectrum of the [M+4H+] 4+ charge state (m/z = 756.0) of C-
peptide. .................................................................................................................... 92

CID MS/MS Spectrum of the [M+Zn2++H+]3+ charge state (m/z = 1007.6)
of C-peptide. ............................................................................................................ 93

CID MS/MS Spectrum of the [M+2H++Zn2] 4+ charge state (m/z = 772.0)
of C-peptide. ............................................................................................................ 94

MALDI CID MS/MS Spectrum of the [M+H+]+ charge state (m/z = 2962.3)
of E27A ................................................................................................................... 95

CID MS/MS Spectrum of the [M+3H+]3+ charge state (m/z = 988.6) of
E27A ........................................................................................................................ 96

ix

88

S9

$10

$11

$12

813

$14

$15

816

$17

818

819

S20

S21

CID MS/MS Spectrum of the [M+4H+]4+ charge state (m/z = 741.5) of
E27A ........................................................................................................................ 97

MALDI CID MS/MS Spectrum of the [M+H+]+ charge state (m/z = 2962.5)
of E1 1A. .................................................................................................................. 98

CID MS/MS Spectrum of the [M+2H+]2+ charge state (m/z = 1482.6) of
EllA ........................................................................................................................ 99

CID MS/MS Spectrum of the [M+3H+]3+ charge state (m/z = 988.6) of
EllA ...................................................................................................................... 100

CID MS/MS Spectrum of the [M+4H+]4+ charge state (m/z = 741.5) of
EllA ...................................................................................................................... 101

MALDI CID MS/MS Spectrum of the [M+H+]+ charge state (m/z = 2976)
of D4A ................................................................................................................... 102

CID MS/MS Spectrum of the [M+2H+]2+ charge state (m/z = 1489.5) of
D4A. ...................................................................................................................... 103

CID MS/MS Spectrum of the [M+3H+]3+ charge state (m/z = 993.5) of
D4A. ...................................................................................................................... 104

CID MS/MS Spectrum of the [M+4H+]4+ charge state (m/z = 745.5) of
D4A. ...................................................................................................................... 105

MALDI CID MS/MS Spectrum ofthe [M+H+]+ charge state (m/z = 2962.3)
of E3A ................................................................................................................... 106

CID MS/MS Spectrum of the [M+2H+]2+ charge state (m/z = 1488.5) of
E3A ........................................................................................................................ 107

CID MS/MS Spectrum ofthe [M+3H+]3+ charge state (m/z = 988.5) ofE3A ...108
CID MS/MS Spectrum of the [M+4H+]4+ charge state (m/z = 741.6) ofE3A ...109

MALDI CID MS/MS Spectrum of the [M+H+]+ charge state (m/z = 2962.4)
ofElA. .................................................................................................................. 110

S22

$23

$24

$25

826

S27

$28

829

S30

S31

S32

S33

S34

CID MS/MS Spectrum of the [M+2H+]2+ charge state (m/z = 1488.5) of
ElA ........................................................................................................................ 111

CID MS/MS Spectrum of the [M+3H+]3+ charge state (m/z = 988.5) of
EIA ........................................................................................................................ 112

CID MS/MS Spectrum of the [M+4H+]4+ charge state (m/z = 741.5) of
EIA ........................................................................................................................ 113

Overall product ion and sequence coverage summaries of the [M+H+]+,
[M+2H+]2+, [M+3H+]3+, and [M+4H+]4+ charge states of the A)B27A, B)
E1 1A, C) D4A, D) E3A, and E) ElA mutants of C-pcptide ................................. 114

CID MS/MS Spectrum of the [M+Zn2++H+] 3+ charge state (m/z = 1008.9)
of E27A. ................................................................................................................ 115

CID MS/MS Spectrum ofthe [M+Zn2++2H+] 4+ charge state (m/z = 758.0)
of E27A ................................................................................................................. 116

CID MS/MS Spectrum of the [M+Zn2+] 2+ charge state (m/z = 1514.4) of
E1 1A ...................................................................................................................... 1 17

CID MS/MS Spectrum of the [M+H ++Zn2+] 3+ charge state (m/z = 1008.9)
of E1 1A ................................................................................................................. 118

CID MS/MS Spectrum of the [M+2H++Zn2+]4+ charge state (m/z = 758.0)
of E1 1A. ................................................................................................................ 119

CID MS/MS Spectrum of the [M+Zn2+]2+ charge state (m/z = 1521.6) of
D4A. ...................................................................................................................... 120

CID MS/MS Spectrum of the [M+H++Zn2+]3+ charge state (m/z = 1014.0)
of D4A. .................................................................................................................. 121

CID MS/MS Spectrum of the [M+2H++Zn2+]4+ charge state (m/z = 760.7)
of D4A. .................................................................................................................. 122

CID MS/MS Spectrum of the [M+Zn2+]2+ charge state (m/z = 1513.6) of
E3A ........................................................................................................................ 123

xi

335 CID MS/MS Spectrum of the [M+H++Zn2+]3+ charge state (m/z = 1008.9)
of E3A. .................................................................................................................. 124

S36 C1D MS/MS Spectrum of the [M+2H++Zn2+]4+ charge state (m/z = 757.0)
of E3A. .................................................................................................................. 125

S37 CID MS/MS Spectrum of the [M+Zn2+]2+ charge state (m/z = 1513.38) of
EIA ........................................................................................................................ 126

S38 CID MS/MS Spectrum of the [M+H++Zn2+]3+ charge state (m/z = 1009.9)
ofElA ................................................................................................................... 127

S39 CID MS/MS Spectrum of the [M+2H++Zn2+]4+ charge state (m/z = 757.8)
of EIA ................................................................................................................... 128

S40 Overall product ion and sequence coverage summaries obtained from the
2+ 2+ + 2+ 3+ + 2+ 4+

CID spectra of the [M+Zn ] , [M+H +Zn ] , and [M+2H +Zn ]
charge states of A) EllA, B) D4A, C) E3A, D) ElA, mutants of C-

peptide. \ indicates an unbound b or y ion, \ indicates only the Zn-

bound fragment was found, and \ indicates that both bound and unbound
ions were identified ............................................................................................... 129

xii

CHAPTER ONE

A biological role for proinsulin C-peptide?

1.1 Diabetes Mellitus and Diabetic Complications

Diabetes mellitus is a disease that affects an estimated 180 million people
worldwide, with this number projected to double by the year 2030 (World Health
Organization). In 2005, an estimated 1.1 million deaths were reported from this disease.
Diabetes is characterized by deregulation of glucose in the body causing hyperglycemia,
i.e., elevated blood glucose levels, due to either a lack of insulin production in the B-cells
of the pancreas or a bodily resistance to the insulin being produced. Diabetes is generally
classified based on the means by which hyperglycemia is caused.

Type 1 diabetes, also called juvenile diabetes or insulin-dependent diabetes, is
characterized by a lack of insulin production due to auto-immune destruction of the B-
cells. This type of diabetes can become fatal but is managed by regular insulin injections.
Type 2 diabetes, also called adult-onset diabetes, results from an inability of the body to
effectively use the insulin that is produced. This form of diabetes is largely diagnosed in
middle-aged adults, but recently has been also found in overweight children. Key factors
associated with the onset of type 2 diabetes are obesity and lack of exercise. This form of
diabetes may largely be managed by diet and exercise. The third category of diabetes is
gestational diabetes which is characterized by hyperglycemia similar to type 2 diabetes

during pregnancy. This form usually ends after pregnancy.

 

Although the disease may be managed using the methods stated above, there is no
cure for diabetes. Additionally, many chronic complications may arise even with proper
treatment. Elevated blood glucose levels lead to glycation of lipids and proteins by an

irreversible Amadori reaction, leading to advanced glycation endproducts (AGES). It has

been shown that type 2 diabetic patients exhibit elevated levels of AGES,1 which act to

deregulate normal enzyme functions. These AGES are also a key source of free radicals

Wthh act to Increase ox1dat1ve stress In d1abet1c patients.

Macrovascular complications, including heart disease and stroke, are another
serious category of complications associated with diabetes. In fact, it has been estimated
that 50% of diabetic patients die of cardiovascular diseases (World Health Organization).
Hypertension, caused by a constriction of the blood vessels which leads to high blood
pressure, can lead to many of these complications.

Diabetic patients are also subject to several microvascular complications, which
can result in vision disorders (retinopathy), kidney failure (nephropathy) and nerve
damage (neuropathy). Diabetic retinopathy is the leading cause of blindness in adults,
and affects 80% of long term (>20 years) diabetic patients. Diabetic nephropathy is one
of the leading causes of kidney failure, and an estimated 10—20 % of diabetic patients die
of kidney failure. Nerve damage due to neuropathy is caused by lack of blood flow to the

extremities, which results in a high rate of foot amputations among diabetic patients.

1.2 Insulin and its Role in Diabetes

Proinsulin is a prohorrnone that is produced in the B-cells of the Islets of

Langerhans in the pancreas. Its structure has been well-defined3, and it consists of 3

subunits, the A and B chains of insulin, and C-peptide, which connects the A and B
chains. The structure and sequence of human proinsulin is shown in Figure 1.1. Once
produced, disulfide bonds between the A and B chains of insulin are formed, which looks
the structure of insulin into place. Proinsulin is then cleaved by two different proteases to

form C-peptide and insulin in equimolar amounts.

  

C-peptide
69 Gho
0
do (b
9 G
0 O
8 G Exopeptidase
Exopeptidasc o Endoprotease
0
Endoprotease g
0
99°“ 00
c?
0

.A GIVDQCCWSSICSLYQLENYCE
B FVEQHLCGSHLVNALYLVSGNRGFFYTPKT
(IEAEDLQVGQVELGGGPGAGSLQPLALEGSLQ

Figure 1.1: Structure of proinsulin, showing locations of cleavage into insulin and c-
peptide. Sequences of the A) A-chain, B) B-chain and C) C-peptide are shown
below the structure. (Adapted and modified from Reference 4)

One protease behaves as an endoprotease and cleaves at the basic residues on each end of

the C-peptide sequence, and another acts as an exopeptidase to remove the basic residues

at the ends of the peptide. The cleavage sites of the proteases are also shown in Figure

1.1. Once insulin is formed, it is stored in secretory vesicles. Upon an increase in

glucose concentration, insulin is released from the B-cells through ATP dependant K+

channel and Ca2+ channels. Upon release of Ca2+, insulin is released from the secretory

vesicles into the blood stream.
Insulin’s main function is to aid in the transport of plasma glucose into cells,
where it is then used as fuel when needed, or stored as glycogen. This acts to regulate the

concentration of glucose in the blood stream when glucose levels are high. Glucose
transport operates mainly through 2 glucose transporters, GLUTl and GLUT4.5 GLUTl
is a ubiquitous carrier that is located in erythrocytes and endothelial cells. It is a large
transmernbrane protein that transports glucose and other molecules such as Vitamin C by
undergoing conformational changes.6 GLUT4 is an insulin-sensitive transporter found in
muscle and adipose tissue, and is located in intracellular vesicles. Insulin binds to the

insulin receptor on the muscle or fat cells, which translocates the GLUT4 transporter to

the cell membrane, where glucose is then transported inside the cell. Once in the cell,

glucose is phosphorylated to glucose-6-phosphate where it undergoes glycolysis.7

In type 1 diabetes, the absence of B-cells does not allow for the production of
insulin, such that glucose transport is inhibited. Type 2 patients suffer fiom insulin
resistance, where the produced insulin is not used effectively, causing the B-cells to
produce even more insulin in response to increased blood glucose levels.

Along with glucose transport, insulin has also been shown to play a role in

endothelium-derived nitric oxide (N 0) production.8 NO is produced by endothelial nitric

oxide synthase (eNOS) in response to increased blood flow and has been shown to
promote vasodilation within cells. Insulin has been shown to play a role in the activation

of eNOS, resulting in an increased production of NO, via phosphorylation at the Ser1179

reSIdue of eNOS. Under hyperglycemic conditlons, where Insulin 1s elther not present

or not effective, NO production is hindered not allowing the endothelial cells to respond

to an increase in the shear stress within the cells. Both type 1 and type 2 diabetic patients

. . . 10 . . .
have been shown to have decreased availability of NO, causrng complications such as

hypertension and atherosclerosis.

1.3 Proinsulin C-peptide
While the effects and mechanisms of insulin on diabetes are fairly well-
understood, recent attention has Shifted towards the other by-product of proinsulin

cleavage, i.e., C-peptide. As mentioned above, C-peptide is a variable-length peptide that

bridges the A and B-chains of insulin. Since its sequencing in 1971,11 it has been

thought to aid only the correct folding of insulin, and have no biological activity.12
Because it is secreted in equimolar amounts along with insulin, detection of C-peptide
has been used as a diagnostic marker for diabetes.13 Evidence supporting a lack of
activity for C-peptide has been provided by the lack of primary structure conservation

among different species, while both the A and B chains of insulin are highly conserved. 14

A summary of the sequence homology of C-peptide from 20 mammalian species, is

shown in Figure 1.2. The human C-peptide sequence is listed at the top, while variable

amino acids are listed below. Capitalized letters represent sequence variants observed in
more than one species, while lower-case letters represent sequence variants observed in
only a single species. Letters in boxes indicate conservation in all, or all but one, of the

mammalian species. Dashes indicate possible gaps in the sequences.

EIIIIDLEJVGQVGGGPGAGSLQPLIGSLE]

GVdGP aPAL AaSleLiGgsAGG aVPP
L n mAeA ma sdDerTSp M19
9 Edt 1e e t 1p 11R

r m

Figure 1.2: Sequence homology of C-peptide from 20 mammalian species.
(Adapted and modified from reference 14)

It can be seen from Figure 1.2 that very few amino acids are conserved, while
deletions are present in the internal segments of some species. It has also been

demonstrated that human C-peptide lacks any secondary structure under physiological

. . . . . . 15 . . .
condltrons or In 11p1d env1ronments, further supportmg the contentlon that C-peptrde

plays no biological role in diabetes.
Despite the lack of sequence conservation and secondary structure, recent
experiments have shown that C-peptide exerts favorable effects among patients with type

1 diabetes. In vivo experiments have shown that administration of C-peptide can improve

red blood cell deformability,16 renal and nerve function,l7.21 as well as improve

22, 23

microvascular blood flow. Other in vitro studies have shown that C-peptide can

24, 25 2

activate mitogen-activated protein kinases (MAPK) as well as stimulate eNOS.2

As mentioned before, eNOS plays a key role in chronic diabetic complications such as

hypertension. C-peptide has also been observed to specifically bind G protein—coupled

26 . . . .
receptors on human cell membranes , demonstrating a basrs for the act1v1ty of C-

peptide.

While it has been Shown that C-peptide has the ability to bind to cell membranes,

the fate of the bound peptide was not known. Recently, an experiment conducted by

Lindhal et. al.27 using labeled C-peptide demonstrated that C-peptide is internalized

within the cytosol of mouse fibroblast cells as well as human embryonic kidney cells.
This study gives yet more evidence to the biological activity of C-peptide, giving
evidence for intracellular activity.

C-peptide appears to be an unusual case of a peptide which lacks sequence

conservation yet exhibits biological activity. When looking at the C-peptide sequences

. . . . . . 14
for various specres, only 7 resrdues are highly conserved across marnmalran Specres.

These residues mainly consist of Glu and Gln residues, and cover both the N-terminal
and C-terminal regions of the peptide. This may suggest some significance to these

residues. Thus, several studies have been conducted to relate structural elements of C-

14, 18, 24-26, 28-31

peptide to specific biological activities. These experiments have

resulted in conflicting viewpoints about which part of the peptide is active. One

+
experiment studied different segments of rat C-peptide and their ability to stimulate Na ,

K+-ATPase activity.3O Na+, K+-ATPase is involved in the regulation of Na+ and K+

transport into and out of skeletal cells, and is the cause of some muscular dysfunctions in

diabetic patients.32 This study found that the C-terrninal portion of the peptide was

active, with the fragments EVARQ (residues 27-31) and VARQ residues (28-31)
showing 103% and 92% activity over the full length peptide, respectively. The
remaining N-terrninal portion of the peptide sequence only showed 2% activity, while
segments in the middle showed between 3 and 80% activity, with the section consisting
of residues 11-19 returning the most activity.

Another study examined the regions of human C-peptide and its ability to displace
human membrane-bound full length C-peptide.31 This study found that the C-terminal
pentapeptide (EGSLQ) could fully displace the membrane bound peptide, while the C-
terminal tetrapeptide (GLSQ) had no effect on the displacement. Free Glu showed about
50% displacement, while the pentapeptide EVARQ corresponding to the pentapeptide of
rat C-peptide was not able to displace human C-peptide. The middle segment (residues

11-19) which showed some activity in rat C-peptide showed no activity for the human

case. The results fiom this study led to a suggestion that the Glu27 residue is key in

receptor binding,31 which is also consistent with the previous experiment showing that
. . . . + + . . 31
the C-termrnal pentapeptrde rs crucral to Na ,K -ATPase act1v1ty.

A conflicting study conducted by Ido et. al. 18 examined albumin permeation

upon addition of insulin and C-peptide. Increased albumin permeation among diabetic

patients leads to vascular dysfunction. It was found that albumin permeation decreased

60-70% in the aorta, uvea, retina and nerves upon administration of both C-peptide and
insulin. It was also found that administration of retro C-peptide (i.e., a reversed
sequence) retained almost full activity, as did rat C-peptide. However, a scrambled full-
length C-peptide was found to be inactive, suggesting that the sequence, but not direction
of the amino acids is critical. C-peptide fragments were also analyzed, and it was found
that fragments containing the N-terminal residues of the peptide were critical for activity,
while the C-terminal region was not necessary. This result is in stark contrast to the
previous results described above that showed the C-terminal pentapeptide was the active
region of the peptide.

Secondary structure analysis was also conducted on human C-peptide by

. l4 . . .
Hennksson et. al. WthI'I demonstrated that secondary structure can be mduced In C-

peptide upon addition of a high concentration (80%) of 2,2,2-trifluoroethanol (TFE),
though the peptide retains no secondary structure in aqueous solution. Circular dichroism
(CD) was used to identify the secondary structure present. In these experiments it was
found that the N-terminal portion of the peptide (residues 1-11) formed an (it-helical
structure. The propensity for helical formation further increased when a tetra-alanine
mutant (D4A/V 7A/G8A/V 10A) was examined. Activity was also demonstrated using
MAPK phosphorylation, and it was found that substitution of the 3 highly conserved
glutarnic acid residues with alanine resulted in a dramatic decrease in the MAPK
phosphorylation. It was also demonstrated that inducing structure in the peptide also
increased MAPK phosphorylation, which increased further with the tetra-alanine mutant,

suggesting that secondary structure is crucial to its activity. Note that while secondary

structure may be induced in C-peptide, the extremely high concentrations of TF E

required are far fi'om physiological conditions.

Other conflicting secondary structure analyses using molecular dynamics33 and

CD18 have shown that the C-terminal region of the peptide exists as a turn-like structure

upon addition of high concentrations of TFE. These studies, compared to the previous

. . l4 . . . . . .
expenment discussed above, also Indicate drfferrng percentages of a-helrcal formatlon

in the N—terminal region of the peptide. The middle segment of the peptide has been
shown to adopt hairpin secondary structure upon addition of TFE. However, consistent
with previous reports, C-peptide adopted no secondary structure under physiological
conditions.

From these studies, it can be seen that C-peptide can have several beneficial
effects on the chronic complications caused by diabetes. However, no clear relationship
has been drawn between the structure of C-peptide and its biological function. Although
several reports Show that the C-terminus of the peptide is the active region, other
conflicting studies Show that the N-terminus or middle segments of the peptide can be

active for certain experiments.

1.4 Metals and their Roles in Diabetes

Metals and regulation of their levels in the body plays a crucial role in many

bodily processes.34 One of the most important roles that metals play is to regulate the

levels of reactive oxygen Species which increase oxidative stress. As mentioned before

advanced glycation endproducts are one cause of oxidative stress among diabetic

10

 

patients, which has been shown to decrease insulin secretion levels as well as being a

factor in increased insulin resistance. A variety of metals may act as antioxidants;

. . . . 34 .
however Iron, chromrum and zrnc have proven to be very Important. Although their

effects are well known, including in diabetes, the mechanisms by which these effects
occur has not yet been fully characterized for some of these metals.
Iron is a metal whose effects on the body have been well established. An increase

in the absorption of iron by the body, called hemochromatosis, has been implicated in
diabetes}5 It was found that diabetes developed in 63% of patients who had
hemochromatosis. Furthermore, 22% of these patients developed diabetic complications

such as nephropathy and neuropathy, as well as retinopathy.35

The mechanism by which diabetes caused by iron overload occurs is not yet

understood. However, studies have been conducted to demonstrate that treatment with

. . . . . . . . . . . . 36
chelatron or blood donation results In Increased Insulin sensrt1v1ty and insulin secretion.

Iron has also been linked to cardiovascular disease, a key complication of diabetes.

Diabetic patients exhibiting high iron levels have been shown to have an increased risk of
. . . 37 . . .
developrng cardiovascular disease. A gain, some of the cardiovascular ailments may be
. 2+ . .
relieved through Fe chelatron or blood donation.

Chromium is another essential element whose commercial use has been
extensively promoted as being beneficial to diabetes. As mentioned above, diabetic

patients exhibit decreased glucose uptake into cells either because of lack of insulin

11

production or resistance to insulin. Experiments on rats have shown stimulation of

glucose uptake upon addition of Cr3+.38

In experiments performed by Schwarz and co-workers in the 1950s, it was found

that rats fed with a certain diet consisting of chromium-deficient Torula yeast developed

. . . 39 . .
impaired glucose tolerance in response to exogenous glucose. This led to the discovery

of a new compound termed the glucose tolerance factor, which aided in proper glucose

metabolism. The structure of this factor has not been defined, but it has been determined

3+ . . . . 40
that Cr 18 essential to its function.

When investigating the mechanism by which chromium plays a role in diabetes,

the existence of a small chromium-binding molecule, termed chromodulin, was reported
in the 1980’s.41 Chromodulin is a peptide consisting of only four types of amino acids,
namely glycine, cysteine, glutamic acid and aspartic acid, with a molecular weight of

1500 Da.42 Chromodulin has the ability to bind up to four Cr3+ ions, even though it has

such a small molecular weight. Its mechanism has been investigated, and it is believed to

play a role in increasing the efficacy of the conversion of glucose to carbon dioxide or

. . . . 38, 43 . . . . .
lipid in rat adipocyte cells. This process is essential to cellular respiration.

Chromodulin has also been shown to be active in the processes of signal transduction, as
well as insulin signaling.

Zinc is another important metal whose link to diabetes and insulin has been long

established. It has been shown that 2112+ exists in the B-cells along with proinsulin in

millirnolar concentrations. Once proinsulin is cleaved to form mature insulin and C-

12

. . . . . . 2+ . 44 2+
peptide, the mature insulin exrsts as a hexamer which 18 bound to two Zn ions. Zn

. . 45 .
enters the secretory vesrcle through a transmembrane protein called ZnT—8, where it

binds and forms the insulin hexamer. The vesicles then reach the membrane of the B-cell,

where insulin, C-peptide, and Zn2+ are all released.

Along with the conformational role played by Zn2+ in the organization of the

insulin hexamer, it has also been found that patients with diabetes exhibit elevated levels

of Zn in urine,46 indicating that the Zn is not being used properly within the secretory

vesicles. Destruction of B-cells in type 1 diabetes results in a lack of Zn to perform other
functions, such as reducing oxidative stress. As mentioned before, oxidative stress is a
key problem associated with tissue damage in diabetic patients.

Metals play a key role in the development and progression of diabetes and
diabetic complications. As shown above, metals such as Zn are in high concentrations in
the pancreas where insulin is produced and secreted, and other metals such as chromium
have shown to interact with insulin and facilitate glucose metabolism. As C-peptide is
secreted along with insulin, it would be reasonable to hypothesize that C-peptide could
have an ability to interact with these metals, which may help explain some of the

inconsistencies obtained from the C-pcptide experiments.

1.5 Metal-Activated C-peptide

Recently, Meyer et. a1.47 demonstrated that human C-peptide has the ability to

release ATP fiom rabbit erythrocytes. ATP is a known stimulant of endotheliumoderived

l3

 

nitric oxide, and it has been shown that erythrocytes have the ability to release ATP in
response to different stimuli. It has also been shown that diabetic erythrocytes release
50% ATP with respect to non-diabetic erythrocytes. It was shown that ATP release was
increased three-fold over a period of 8 hours when C-peptide was added to a sample of
erythrocytes. From these results, it was also demonstrated that upon addition of C-
peptide, ATP levels in human patients with type 2 diabetes could be restored to those
observed in healthy control patients.

It was observed however, that the increased ATP release was time dependent.
After 25-36 hours following preparation of C-peptide in water, ATP release was
dramatically decreased to baseline levels compared to erythrocyte controls. Following
analysis of the C-peptide sample using electrospray ionization (ESI) mass spectrometry,

it was observed that the peptide was intact and no degradation products were found.

However, in a freshly prepared solution of C-peptide, adducts formed with F e2+, as well

as K+ and Na+, were observed. Solutions of HPLC purified C-peptide and water were
. . 2+ . . . 2+

then spiked With Fe solutions and it was found that a major Fe adduct was formed,

indicating a preferential binding of C-peptide to F e2+.

ATP release experiments revealed that upon addition of fresh solutions of Fe2+ to

solutions of C-peptide, ATP release was increased. As a control, a peptide solution

which had been stored for 30 days was incubated with erythrocytes and demonstrated no

activity. C-peptide purified by HPLC also showed not activity. Solutions of F e2+ were

then added to previously inactive C-peptide samples, and the activity was restored. The

14

effect of the oxidation state of iron was studied by incubating the erythrocytes with

samples of C-peptide and Fe3+, which displayed no increase in ATP release. These

samples were subjected to ESI-MS analysis where it was found that minimal F e3+ adduct
. 2+ . . +

was formed. It was determined that upon exchange of Fe With ions such as Na and

+ . . . . . 2+ 3+
K in solution over time, or ox1dation of Fe to Fe , the ATP release was observed to

. . 2+
decrease. Thus, the increase in ATP release was due to the presence of an Fe

‘activated’ C-peptide.

To determine whether other metals could ‘activate’ C-peptide with respect to an

increase in ATP release fiom erythrocytes, the effect of Cr3+ was also examined by

Meyer et. al. Following addition of Cr3+ to a C-peptide sample, a dominant Cr3+ adduct

with C-peptide was observed by mass spectrometry analysis. Furthermore, an increase in

ATP release was observed for a period of 72 hours for the Cr3+ activated sample, as
opposed to 24 hours for that of Fe2+. Again, inactive C-peptide samples, where fi'esh

Cr3+ was added displayed an increase in the ATP release. The same results were

obtained when applied to glucose uptake measurements. There was a 31% increase in the
+

amount of glucose entering the erythrocytes upon incubation of C-peptide and Cr3 .

These results were significant in that this was the first demonstration of metal
interactions with C-peptide, and that metal interactions with C-peptide are responsible for
a specific biological activity. As mentioned before, various metals play integral roles in

the body as well as complications due to diabetes. Zn is an extremely important metal

15

and is observed in high concentrations (mM) within the pancreas, so it is not
unreasonable to believe that C-peptide-metal interactions could occur with Zn. Thus,

ATP release and glucose uptake experiments were conducted with C-peptide activated

.1.
with an .48 It was found that ATP release increased by 31% over control levels (i.e.,

erythrocytes alone) upon addition of Zn2+ activated C-peptide, while C-peptide addition

without added metal or the addition of Zn only did not produce any increase in ATP

release. Similar results were obtained for glucose uptake measurements, measured using

1A'C-labeled glucose, where it was found that glucose transport into diabetic rabbit

erythrocytes increased by 36%, although this was a smaller effect than that of control
non-diabetic rabbit erythrocytes, which increased transport by 64%.

It is clear from these studies that C-peptide activated with certain metals plays a
crucial role in certain biological activities. When attempting to gain a complete
understanding of a biological molecule, both its function and structure must be
understood. The functional significance of metal activated C-peptide has now been
established, as described above. However, a greater understanding of the specific metal-
C-peptide interactions that give rise to these functions remains to be determined.
Examination of which residues in the C-peptide sequence are involved in metal binding
and the relative binding strength of these residues needs to be understood. Such studies
may also aid interpretation of the multiple conflicting prior studies described above that
suggested different regions of the C-peptide sequence were important to activity. One

technique that has been successful in these type of analyses is mass spectrometry.

l6

1.6 Metal-Binding Analysis by Mass Spectrometry
Early studies on peptide-metal associations were performed using fast atom

bombardment (FAB) ionization which produced singly charged peptide-metal

49-55 . . .
complexes. The more recent development of electrospray ionization (ESI) has

subsequently enabled the analysis of peptide- or protein-metal complexes involving much

53, 56-59

larger biomolecules. Many different transition metal-peptide or protein

complexes have been investigated by ESI-MS. However, to obtain information regarding
the site of binding on the peptide or protein, collision-induced dissociation tandem mass
spectrometry (CID-MS/MS) is typically used to fragment the ion along the peptide
backbone while maintaining the non-covalent metal-peptide interaction. The details of

this process will be discussed in the next chapter. These studies allow for the binding

. . . . . . . 49,
interactions of complexes to be determined, and have been investigated With alkali, 50

. 51, 52 . . 53, 54 .
alkaline earth, as well as transrtion metal elements. Generally, the Site of the

metal interaction has a strong influence on directing the fragmentation reactions. For

example, abundant [bn_1+Cat+OH]+ product ions from monovalent alkali-containing

peptide ions are observed due to coordination of the metal to two amide carbonyl

groups.50 Other reports have demonstrated alkali metals binding to the carboxylate
. . 55 . . . . 2+
terminus of the peptide. The binding of transrtion metals such as Ca have been

. . . . 53
shown to occur at deprotonated carboxylate groups of ac1d1c ammo acrds. It was

17

determined by Yu et. al.56 that the gas-phase complexes are similar to their solution-
phase counterparts.

One study on the binding properties of angiotensin peptides with Zn2+ showed
that CID-MS/MS of the doubly protonated precursor ions of both angiotensin I and II

resulted in extensive b- and y-type sequence ions,60’ 61 while CID-MS/MS of the

[M+Zn2+]2+ precursor ions showed abundant Zn-bound fragment ions of the form

7 This allowed localization of the er2+ to the His6 and His9

[M+aZn2++(c-2a)H+]C+.5
residues.

Another more recent experiment explored the Zn2+-binding properties of the

amyloid fragment AB(1-16).58 ESI-MS/MS was used to locate the Zn binding sites to 3

histidine and 1 arginine residues which are known to direct the peptide into a hairpin

conformation. Zn has also been shown to bind to cysteine as well as acidic residues such

. . . . 62
as glutamic acrd and aspartic acrd.

The binding properties of Cr3+ to the synthetic acidic peptides AAADA,

EAAAG, AGGAAAA, GGDDAA, AAGEEAA, DGDAAA, EAEAAA, DAADAA,

EAAEAA, EAAAEA, EAAAE, EEE, DDDD, AND EEEEEEE have also been

explored.59 These peptides are similar to C-peptide in that C-peptide has 5 acidic

. . . . . . . . 3+
resrdues and no histidine or cysteine resrdues. This study determined that the Cr was

binding to the carboxylate groups of the acidic amino acids. Sequential losses of water

from the Cr3+ bound molecular ion indicated that not all acidic residues within the

18

peptide are involved in the binding. It was also found that a peptide such as

AGGAAAA-OCH3 which contains no acidic residues was still able to bind Cr3+, even

though this occurred at low abundances. This suggests that carboxylate groups may not

necessarily be the only groups involved in binding.

19

Aims of This Thesis:

AS described above, MS and MS/MS methods have been shown to be useful tools
for the analysis of non-covalent peptide-metal complexes when ESI is used as the
ionization source. Therefore, in order to obtain further insights into the binding

interactions of metal ions to proinsulin C-peptide, the following specific aims were

formulated.

1. Characterization of proinsulin C-peptide using ‘top down’ MS/MS

+
2. Characterization of Zn2 binding Sites within activated proinsulin C-peptide

3. Characterization of proinsulin C-peptide mutants formed by single amino acid

substitution using ‘top down’ MS/MS

. . 2+ . . . . . . . .
4. Characterization of Zn binding Sites Within prornsulin C-peptide mutants formed

by single amino acid substitution using ‘top down’ MS/MS

5. Determination of the effect of single amino acid substitutions or C-peptide segments

on the relative binding of Zn2+ to proinsulin C-peptide.

6. Characterization of the Zn2+ binding sites of the C-terminal pentapeptide of C-

peptide (EGSLQ) and its mutant AGSLQ, and examination of their biological

activity.

20

 

CHAPTER TWO:

Strategies for the Synthesis and Characterization of Proinsulin C-peptide, its

Mutants and Metal Binding

2.1 Solid Phase Peptide Synthesis (SPPS) of the C—peptide Mutants

2.1.1 Materials

Synthesis was conducted from the C- to N-terminus on a Wang resin preloaded
with an N-a-Fmoc protected glutamine residue purchased from EMD Biosciences (San
Diego, CA). N-a-Fmoc protected amino acid building blocks Fmoc-G1n(Trt), Fmoc-Leu,
Fmoc—Ser(tBu), Fmoc-Gly, Fmoc-Glu(OtBu), Fmoc-Ala, Fmoc-Pro, Fmoc-Val, Fmoc-
Asp(OtBu) were purchased from EMD Biosciences (San Diego, Ca) as well. Sequenal
grade trifluoroacetic acid (TFA) was purchased fiom Pierce (Rockford, ll). Reagent
grade dimethyl formamide (DMF) was purchased from Jade Scientific (Canton, MI) and
dried under 3A molecular sieves (Spectrum Chemicals, Gardena, CA). Reagent grade
O-(Benzotriazol-l-yl)-N,N,N’,N’-tetramethyluronium tetrafluoroborate (TBTU) and l—
Hydroxybenzotriazole hydrate (HOBt) were purchased from Peptides International
(Louisville, KY). Reagent grade piperidine, diisopropylethylamine (DIPEA), phenol,
potassium cyanide, ninhydrin, ethanedithiol (EDT), triisopropylsilane (T18) and pyridine
were purchased from Sigma Aldrich (St. Louis, MO).

Resins were weighed into 5 mL Toriq fiitted-syringes (Fisher Scientific,

Pittsburgh, PA) fitted with a 3-way stopcock (VWR, West Chester, PA). Washing was

21

achieved using a Vac-Man laboratory vacuum manifold (Promega, Madison, WI), which

also allowed multiple syntheses to be conducted at once.

2.1.2 Synthesis

SPPS was achieved through a stepwise procedure consisting of two steps. The
first step is deprotection of the Fmoc protecting group from the N-terminal end of the
resin bound amino acid. The second step involves activation of the next amino acid
followed by coupling to the resin. The final step, carried out after the sequence is
completed, involves final deprotection of the N-a protecting group at the N-terminus,
followed by cleavage of the peptide from the resin as well as any orthogonal protecting
groups on the Side chains of any amino acids. These steps will be outlined below. The
sequences of the C-peptide mutants and C-peptide segments that were prepared in this

study are listed below in Table 2.1

 

Sequence Abbreviation

 

 

" " ‘”BAEDIL‘QVGQVBLGGGPGAGSLQPLALAGSIJQ'“'“" " " "

,.‘ , , ‘ ‘ , .
Sf: - 1‘n'f'";"i1ll:‘ wawfifififtffri's'h "'\\‘L

EAEDLQVCovEIGCGPGAGSLQPILAIBCSES"“ EllA

 

EAEZLWQVGQVELGGGPGAGSLQPLALEGSLQ " ”543”
I}; ‘.-:"-f-l+;:s:i',1‘.:-;-"- , ,. " .‘ ' . ' 7 . ‘ ;. ‘37:.‘77.’ ; .. .7,”..:.:-...:.ar:s,.-at... 3
EAADLQVGQVELGGGPGAGSLQPLALEGSLQ E3A
AAEDLQVGQVELGGGPGAGSLQPLALEGSLQI ma

 

Table 2.]: Sequences of the single amino acid mutants of C-peptide that were
synthesized, along with their respective abbreviations. Locations of the alanine
substitutions are labeled in bold and underlined on the sequences.

22

Synthesis of the E27A mutant was performed on a 0.15 mmol scale. Synthesis of the
BIA, E3A, D4A, E1 1A mutants, the C-terrninal peptide segment EGSLQ and the alanine

substituted C-terminal peptide segment AGSLQ, were performed on a 0.1 mmol scale.

2.1.2.1 Deprotection

After swelling the resin in 3 mL DMF for 15 minutes, the resin was then washed
once with 30% piperidine in anhydrous DMF. 3 mL of the piperidine solution was then
added and the resin was thoroughly mixed for 25 minutes. The resin was then filtered

and washed resin 8 times with 5 mL DMF until the pH was < 8.

2.1.2.2 Coupling

For each coupling step, at least a 5:1 excess of amino acid to resin was used. The
amino acid was pro-activated by dissolving in 1.5 ml activator solution (0.5 M TBTU, 0.5
M HOBt in DMF) and 300 11L of DIPEA. The solution was vortexed to completely
dissolve the amino acid, then added to the resin. The resin was then mixed for 15

minutes, filtered and washed 5x with SmL of DMF.

2.1.2.3 Ninhydrin Test

To determine if the coupling reaction was successful, a ninhydrin color test was

used.63 Ninhydrin reacts with primary amines to form a deep blue color. Therefore, in

the event of an incomplete coupling, deprotected primary amines will be present on the

residual N-terminus of the initial peptide, which will cause the resin to turn blue when the

23

test is conducted. If the coupling is successful, all resin-bound peptide will be coupled
with a protected amino acid.

This test was performed by taking a small amount of resin (2 mg) from the
syringe and placing it in a glass test tube. The resin was then washed with 2 mL of
ethanol. Then, 150 uL each of 0.28 M ninhydrin in ethanol, 42.5 M phenol in ethanol,
and 20 11M (aq) KCN in pyridine were added to the resin, and the solution was mixed
thoroughly. After incubation at 100 °C for 10 minutes, the color was inspected. If the
solution and resin were blue in color, the coupling process was repeated using the same
amino acid as described above. If the solution was yellow, the synthesis was continued.

During the synthesis, it was observed that for particular residues, such as
glutamine, more than two coupling events were required in order to completely couple
the amino acid. In this event, a 10:1 excess of amino acid: resin was used and the
coupling time was increased from 15 to 20 minutes. The dissolved amino acid was also

left to activate in solution for 5 minutes prior to addition to the syringe.

2.1.2.4 Final Deprotection/Cleavage

Once the target amino acid sequence was completed, a final round of deprotection
was performed. After the deprotection, the resin was then washed with 5 x 5 mL of
ethanol and allowed to dry. The resin was observed to shrink as the methanol
evaporated. A cleavage solution containing 95% TFA, 0.5% water, and 0.5% TIS was
freshly prepared, then added to the resin. The stopcock of the syringes were removed and

replaced with a cap, and the solution was left to mix thoroughly for 3 hours.

24

 

2.1.2.5 Work up/Extraction

Once the cleavage was completed, the contents of the syringe were emptied into a
100 mL round-bottom flask, and then washed with 2 mL of TFA. The TFA was then
evaporated under nitrogen until about 100 11L of solution remained. The crude peptide
was then precipitated by adding 10 mL of cold ether. The precipitate was then vacuum-
filtered over a 30 mL Buchner funnel attached to a 125 mL side-arm flask. The
precipitate was then put into another 100 ml round-bottom flask and reconstituted in
water and, if necessary, a minimal amount of 25% acetic acid solution. The peptides

were then lyophilized overnight.

2.2 High Performance Liquid Chromatography (HPLC) Purification of C-peptide
and its Synthetic Mutants

All peptides were purified by Reverse Phase-High Performance Liquid
Chromatography (RP-HPLC) using a Shimadzu (Columbia, MD) LC-20AB solvent
delivery system coupled with a SPD-ZOAV UV/V IS detection system. Samples were
dissolved with vortexing in a minimal amount of water, and up to 20% ACN was added
in cases where the peptide was not completely soluble. Samples were injected using a
Rheodyne model 9215 injection valve (Oak Harbor, WA) and a 2 mL stainless steel
sample loop onto a Vydac C18 Protein and Peptide column purchased from Grace
(Deerfield, IL). For purification of human proinsulin C-peptide (Genscript, Piscataway,
NJ), a linear gradient from 100% buffer A to 100% buffer B over 60 minutes was used.

A variety of gradient conditions were evaluated to achieve optimal separation of the

25

synthesized C-peptide mutant peptides. Ultimately, peptides were eluted using the non-
linear gradient conditions shown in Table 2.2

Solvent A consisted of 0.1% HPLC-grade TFA (Pierce, Rockford, IL) in water
purified using a Nanopure Diamond filtration system from Barnstead (Dubuque, IA) with
a resistance of 18.2 MO. Solvent B consisted of 60% HPLC-grade acetonitrile (ACN)
purchased from EMD biosciences (Gibbstown, NJ), containing 0.089% HPLC-grade
TFA. Solvents were sparged with helium for 15 minutes prior to running the HPLC.
Peptides were monitored at 215 nm and individual fractions were collected into 15 mL
centrifuge tubes. The purity of these fractions were then determined by mass

spectrometry analysis. Pure fractions containing the desired peptide were pooled and

 

 

 

 

lyophilized overnight.
Time (min) % A % B
:1
. ._ 2: , ,g 5.1""; - , . . .. . t _ 29.63;!» 3.7.15.1- L- l .13]
20 66 34
__100- f,” -- -—< , 33- -A +~~e ---,....--..67.. *~~*~‘ i
, Ti
W12}, 0 100
a --..::~;;n ante-.1; a. - .. . . . . . ‘ . _ . ‘ . »; Itt‘IT‘JC‘ ‘."'-?'"."=i~‘="'— 51'3- 4"‘-‘”‘-?“"-'"~’7-1t' "a?!
. 3.: a“: évfsi-Lfé-siéxsy .- . ., . , . . . .1 7 ,»_. . » ‘us'w, '. -- :2»? a 53-“:
127 100 0
135 . Stop Stop ;
F!“ I {A ’ ‘ £ , , ‘ , q . ., -. i - - mfg ..rycyaekflr‘a‘s ”‘2’ 22" 3'12

 

Table 2.2: Gradient used for C31 mutant purification

26

 

2.3 Mass Spectrometric Analysis

2.3.1 Introduction

Mass Spectrometry involves the formation of gas-phase ions that are then
separated by various mass analyzers on the basis of their individual mass to charge (m/z)
ratios. The separated ions are then detected and a spectrum is output as a function of both
the rn/z and the abundance of the ions present in the sample. Additionally, selected ions
may be isolated and subjected to dissociation, followed by mass analysis of the resultant
product ions, to allow for structural information to be obtained. The two crucial
components of a mass spectrometer are the ionization source and the mass analyzer. An
in-depth discussion of the theory behind the mass spectrometry methods employed in the

studies performed in this thesis is given below.

2.3.2 Ionization Sources

A variety of ionization methods have been developed over the years, including
electron impact ionization (EI), chemical ionization (CI), fast atom bombardment (FAB),
electrospray ionization (E81) and matrix-assisted laser desorption-ionization (MALDI).
Both ESI and MALDI have gained wide use in the analysis of biological molecules
because of their ability to transfer these large molecules intact into the gas-phase. Both
MALDI and ESI were used in examination of the structure and binding of proinsulin C-

peptide and its mutants, and these ionization methods are described in more detail below.

27

2.3.2.1 Electrospray Ionization (ESI)

ESI is an atmospheric pressure ionization technique that was introduced by John

. 64 . . . . . . .
Fenn and coworkers in the 1980’s . This is in contrast to prevrous lomzation methods

such as CI and EI where the ions were required to be formed under vacuum. Formation
of ions under higher pressures allows for “soft” ion formation, where the ion is formed
with minimal internal energy, thereby undergoing minimal fragmentation. ESI also
produces multiply-charged ions, which allow for the analysis of very large biological
molecules in mass analyzers with a limited m/z range. The dissociation of different
multiply charged ions often results in different fragmentation pathways, often giving rise
to different structural information for each charge state.

In ESI, the sample is dissolved in a solution typically consisting of methanol,
acetic acid and water. The sample is pumped through a narrow (100 um i.d.) silica
capillary tubing, with the tip of the capillary placed near the entrance to the mass
Spectrometer (~2 cm). A schematic of a typical ESI interface is shown in Figure 2.1.
The tip of the capillary is maintained at atmospheric pressure. A high voltage (3-5 kV) is
applied to the capillary with respect to ground which is normally the entrance to the mass
spectrometer, and the solution is pumped at flow rates in the 1-10 uL/min range. The
high voltage in combination with the short distance from the counter electrode creates a
large electric field, resulting in charge buildup at the tip of the capillary. The exiting
solvent therefore forms a Taylor cone at the tip to alleviate the charge density. When the
charge density on the Taylor cone becomes too great, a continuous series of electrostatic

droplets form, which then undergo a series of coulombic fission events whereby solvent

28

is evaporated from the droplet to form the ions. Figure 2.1 also shows a schematic of

droplet formation and evaporation.

in? . a - '.- ”(.590

“54"13'13y

.‘ 5. '25;
Eg-hfibfig 0

. -. . "..v" 1.."

 

 

 

 

 

Electrons
—-——->
(4'; TDC Electrons
High Voltage
Power Supply

 

 

Figure 2.1: Representation of droplet formation in ESI. (Adapted from reference
65)
The mechanism for the formation of ions by ESI has been a subject of debate over

many years. Two main theories have been proposed: the charge residue model (CRM)

67,68

66 and the ion evaporation model (IEM) . The CRM mechanism proposes that as the

droplets lose solvent molecules through evaporation, Coulombic fission occurs when the

surface charge density on the droplet becomes greater than the surface tension of the

29

liquid. This will occur when the charge is greater than a limiting value, termed the

Rayleigh Limit, and is given by Equation (1)

£12 = 84280703 (1)
where q is the charge on the droplet, y is the surface tension, and D is the diameter of the
spherical drop. When this limit is reached, the droplet will explode into smaller droplets,
and will continue until a droplet is formed with only a single analyte molecule. As the
remaining solvent evaporates, a gas-phase ion is formed. The IBM suggests that the
droplet undergoes a series of Rayleigh breakdowns similar to the CRM to form
successively smaller droplets. However, in contrast to the CRM, the IBM proposes that
ions are evaporated directly from the surface of the droplet as the electric field increases
due to solvent evaporation. Once the electric field becomes stronger than the solvation

forces, the ion will be evaporated directly into the gas phase. Even though both theories

. . . . . . 67 .
offer competing Views for gas-phase ion formation, it has been shown that ions can be

formed by either mechanism, where smaller molecules are formed by IBM, and larger

biological molecules are formed by CRM.

2.3.2.2 Matrix-Assisted Laser Desorption Ionization (MALDI)

Matrix-assisted laser desorption ionization is a technique developed by Karas and

co-workers in the late 1980’s 69 for the analysis of large molecular weight biomolecules.

This ionization method involves the use of a laser for desorption of the analyte as well as

for ionization. In this technique, the sample is dissolved (typically in water and organic

30

solvents such as acetonitrile (ACN), along with small amounts of trifluroacetic acid
(TFA)), and is then mixed on a sample stage with a matrix compound containing a UV-
absorbing chromophore. The sample is then allowed to evaporate, co-crystallizing the
analyte and matrix together. The selection of the matrix component is crucial to a
MALDI experiment, in that the compound must absorb at the wavelength of the laser to
be used, and it also must have a low vapor pressure that it will not evaporate once
inserted into the vacuum chamber. The sample is then inserted into the source region of
the mass spectrometer, which is maintained at vacuum. The laser is then pulsed onto the

sample, and the matrix will absorb the laser radiation. The ionization process of MALDI

. . 70, 71 . . .
is still not well understood , but both analyte and matrix ions are formed. This

causes interference ions due to clusters of matrix ions in the low-mass region of the
Spectrum.

MALDI has become a very effective technique for analyzing large molecular-
weight and polar molecules, such as biomolecules. Because this process forms primarily
singly-charged ions, the ranges of mass analyzers that are effective for this technique are
limited. Typically, TOF mass analyzers are used as they have the advantage of unlimited
mass range. However, in the experiments conducted, due to the relatively low mass of

the C-peptide and its mutants, MALDI was able to be used to obtain tandem mass spectra

for the singly protonated [M+H+]+ ions of C-peptide and its mutants, using a linear

quadrupole mass analyzer.

31

2.3.3 Mass Analyzers

In order to obtain a mass spectrum, the gas-phase ions that have been formed
must be separated according to their m/z values. Ion formation allows for manipulation
of the ions using various fields. A number of mass analyzers have been developed over
the years including magnetic sector, time of flight (TOF), quadrupole (Q), quadrupole ion
trap (QIT) (ineluding both 3D and linear configurations) and Fourier transform ion
cyclotron resonance (FT-ICR). In the experiments conducted here, quadrupole ion trap
and linear ion trap mass analyzers were used. Ion traps, coupled with ionization methods

such as ESI are particularly attractive for use in the analysis of both small and large

biomolecules, as they are capable of multiple stages of tandem mass spectrometry MS“)

to obtain structural information.

2.3.3.1 Quadrupole Mass Analyzer

Ideally, a quadrupole mass analyzer consists of 4 hyperbolic-shaped rods aligned
parallel to one another as shown below in Figure 2.2. However, cylindrical rods are
typically used for cost effectiveness and practicality. Ions entering through the z-
dimension of the mass analyzer experience a quadrupole electric field which consists of
an alternating current (AC) signal and a direct current (DC) signal. The DC Signals on
one opposing pair of rods is of opposite Sign with respect to the other pair, while the AC
signal on each pair of rods are 180 degrees out of phase with respect to each other. When
positive ions enter the trap, they are attracted to a pair of rods when a net negative
potential is observed and repelled from a pair of rods (i.e., focused toward the center of

the assembly) when a net positive potential is observed. Because an AC signal is applied,

32

the polarities on the rods will change rapidly, which will focus the ions to the center of

the quadrupole as the ions are passing through along the z-axis.

 

Applied Potential ((195.7

Figure 2.2: Schematic of a quadrupole mass spectrometer, showing applied AC and
DC potentials ((D0).

The applied potential is given the symbol (D0, and is described by the following

equations

(Do = +(U-Vcos cot) and -(I)0 = -(U-Vcos wt) (2)

33

where U is the applied DC signal, V is the zero-to-peak amplitude of the AC signal in the

RF region, and 0) is the angular frequency, which can be given in terms of the applied RF

frequency, f ((0 = 27rf). Ions entering the quadrupole assembly along the z-axis

experience forces in the x and y directions. These forces can be expressed by equations

(3) and (4) below

2
Fx — mi—x = —-ze£32
dtz 5x
2
Fy=mi—;-=— 862
dt 5y

where (I) is a function of both x and y and can be written in terms of (D0

(x2 —y2)(U — Vcos (at)

 

2 2 2
(ny=(D0(x ‘y )/’0 = 2
r0

(3)

(4)

(5)

Equation (5) can be differentiated and rearranged to achieve the equations of ion motion

in the x and y directions, given by equations (6) and (7) respectively

dzx 22e
—+

alt2 mro2

 

(U — V cos 600x = 0

34

(6)

d2y _ 2ze
dtz mroz

 

(U — V cos (ot)y = O (7)

When comparing these equations to the equation solved by Mathieu72 given below in

equation (8), the working equations for the quadrupole mass analyzer can be determined

231+(a —2q c052§)u-0 (8)
déz U u

In equation (7), u represents motion in either the x or y direction, and 6 = tot/2. The
variables a and q are dimensionless parameters. When equations (6) and (7) are
rearranged in terms of the parameters a and q, the working equations of ion motion in the

quadrupole are obtained as shown in equations (9) and (10).

82eL/
a. =a. =—ay =—7—, (9)
ma) r0
4zeV
qu=qx:—qy=—2—2— (10)
mar r0

These equations state that ions of a given mass-to-charge ratio in the quadrupolar

field will pass through the assembly as long as the motion along the x and y directions

does not exceed the value of r0 (the distance between the center of the quadrupole and the

35

 

rods). It can be noted that equations (9) and (10) are a function of both m/z, the applied

DC amplitude U, and the applied RF amplitude V. For a given experiment, r0 and w are

fixed, and e is the elementary charge (1.602 x 10'19 C) which remains constant.

Therefore, the stability of an ion is determined as a function of the amplitudes of the RF
and DC potentials applied to the rods. Equations (9) and (10) can be rearranged to solve

for U and V, respectively, giving equations (11) and (12).

 

 

2 2
U: H mm r0 (11)
826
2 _2
V u ma) )0 (12)
426

A stability diagram plotting the a vs. q parameters can be constructed which will
Show values of a and q, and therefore values of U and V, where the ions in the
quadrupole will be stable and pass through the trap. This is shown in Figure 2.3. In this
figure, the area under the curve represents where ions will pass through the instrument.
Because a and q are dependent upon m/z, U, and V, fixed values of U and V will result in
differing different a and q values for ions of various m/z. The apex of the curve
occurring at a = 0.237 and q = 0.706 represents the point where only one m/z value will
remain stable, creating a mass filter. By adjusting the parameters U and V over time so
ions of increasing rn/z will successively pass through the apex of the diagram, the

quadrupole may be operated as a mass analyzer.

36

 

0.3 '-

(q = 0.706, a = 0.237)

0.2 -'

0.1-

 

 

 

0.0 1
0.0 0.2 0.4 0.6 0.8 1.0

Figure 2.3: Stability diagram for a quadrupole mass analyzer.

In contrast, by not applying a DC potential (i.e., a = 0) to the quadrupole, ions
will lie along the x-axis of Figure 2.3, where many different m/z values may be passed
through the assembly, only being limited by a low-mass-cutoff imposed by the RF
amplitude which results in ions with q values higher than 0.908 being unstable. The
application of this will be discussed in more detail in the next section in terms of the

quadrupole being used as an ion trap as opposed to a mass filter.

2.3.3.2 Quadrupole Ion Trap Mass Analyzer
A quadrupole ion trap consists of a hyperbolic-shaped ring electrode capped by
two hyperbolic end cap electrodes. Each end cap electrode has a hole drilled in the center

for ion injection/ejection. A schematic of a quadrupole ion trap is shown in Figure 2.4,

37

 

where r0 is the distance from the center of the trap to the ring electrode and 20 is the

distance from the center of the trap to the end cap electrodes.

Source

    

End Cap Electrode

Ring Electrode

Detector

Figure 2.4: Schematic of a quadrupole ion trap (Adapted and modified from

Reference 72)

A variable-amplitude radio frequency (RF) is applied to the ring electrode which follows

the same relationship as equation (1). The end cap electrodes are typically held at ground.
Ions entering the trap will experience the RF potential in three dimensions (r and

z), and stable ions will be bound inside the trap. Much like the quadrupole, the working

38

equations for ion motion in the trap may be derived using the Mathieu equation (8), to

obtain equations (13) and (14)

— l6zeU
771(7‘02 + 223 )a)2

 

(13)

au =aZ=—2ar :

828V
qu =_2qr =q2= 2 2 2 (14)
m(rO +220 )a)

 

Again, r0, 20, and a) are all fixed for a given instrument. For an ion of a given m/z ratio,

the parameters a and q are a function of the RF and DC potentials as in the quadrupole

mass analyzer. An ion will be stable in the trap as long as the trajectories along the r and

2 dimensions do not equal r0 and 20. A stability diagram can be constructed by plotting a

vs. q, and is shown in Figure 2.5

39

 

Unstable ions

.,,q,= 0.908

 

T’

 

I

-0.2 4

I

—O.3 ‘

-0.4 .

U

I

-0.5 -

-0.6 s

I

 

 

Figure 2.5: Mathiew stability diagram for a 3D ion trap showing stability of ions of
different m/z values. Larger balls indicate larger m/z values. (Adapted and
modified from Reference 72)

,6 is a fundamental stability parameter and will be discussed in more detail later.
This diagram Shows the regions of a and q, and therefore the values of U and V where
ions will be stable in the trap. In commercial instruments, no DC potentials are applied,

therefore a = 0 and ions will be aligned along the x-axis of the diagram. The boundary of

the diagram where ,6 = 1 is at q = 0.908. This means that an ion having a q-value

approaching 0.908 will become unstable and travel farther than 20. This point is termed

40

the stability limit. Looking at equation (14), the value of q is proportional to the
amplitude of the applied RF potential, and inversely proportional to the m/z ratio of the
ion. Therefore, ions of higher m/z will have a lower q-value, and ions of small m/z will
have a larger q-value, bringing them closer to the stability limit. This imposes a low-
mass-cutoff value for the ion trap.

Ions are injected into the trap, where they encounter an inert bath gas, typically
helium. This gas is held at a constant pressure of 1 mTorr, and ions will collide with the
gas and transfer kinetic energy to the Helium. This “collisional cooling” focuses the ions
into the center of the trap, which improves the trapping efficiency. Another factor in the
trapping efficiency of the ions is the Dehmelt pseudopotential well, which is given by
equation (15)

— V zeV2

Dzzqz§=

 

(15)

2 2 2
m(rO + 220 )a)

A graphical representation of equation (15) is shown below as a function of q in Figure
2.6. The potential well is a measure of how stable the ion is in the trap. Ions that lie in

the q region around 0.8 are held at the lowest potential and are therefore most stable.

41

   

 

 

 

Figure 2.6: Diagram of Dehmelt pseudopotential well in a 3D ion trap. ERE is the

energy where resonance ejection typically is carried out. (Adapted from Reference
73)

To acquire a mass spectrum, the RF amplitude is increased, which brings the ions

closer and closer to the stability limit. As an ion approaches a q value of 0.908, its

motion along the z-axis will increase and the ion will travel farther than the dimension 20.

When doing this, the ion will pass through a hole in the end cap electrode where it will
strike the detector. Thus, as the RF potential is ramped, ions of increasing m/z will reach
the stability limit and become ejected from the trap through the z-axis. This method is

termed ejection at the stability limi‘ This, however, is not the most effective way of

42

acquiring a mass spectrum. The pseudopotential well depth near the stability limit
becomes very steep, which can cause more than one m/z value to be ejected at a certain
RF potential. This results in a loss of resolution in the mass spectrum.

Another method called resonance ejection is used instead. In this method, a high
amplitude RF potential is applied to the end-cap electrodes, termed the supplemental
voltage, with a frequency that corresponds to a q ~ 0.8. As mentioned before, ions at this
q value are in the deepest part of the potential well. Due to intertia, ions in the trap will
be oscillating at a secular frequency, f, which is less than the frequency, v, of the RF field
applied to the ring electrode. The relationship between the applied fi'equency and the

secular frequency is shown in equation (I 6)

 

f.= ; <19

where [32 is the stability parameter mentioned above, and is related to a and q by equation

(17), when q < 0.4.

fizz az+ L (17)

As shown in Figure 4, ,6 will have a maximum value of 1 at q = 0.908, the stability
boundary. Therefore, ions will be oscillating in the trap at a maximum of one-half the
fundamental frequency. As the RF potential on the ring electrode is ramped, the secular

frequencies of the ions in the trap will approach that of a supplemental frequency that can

43

be applied to the end caps. When the ions become in resonance with this supplemental
frequency, their kinetic energy will increase and they will begin to move farther along the
z-axis until they exit the trap and reach the detector. The main benefit of resonance
ejection is an increase in resolution. Ions of the same m/z coming into resonance with the
supplemental frequency will become coherent, resulting in a “packet” of ions which
allows for the ions of same m/z to be ejected in a smaller range of RF potentials.
Resolution is inversely proportional to the rate at which the RF potential is increased,
called the scan rate. Therefore, greater resolution may be obtained at the cost of analysis

time.

2.3.3.3 Tandem Mass Spectrometry (MS/MS)

One of the main benefits of ion trap instruments is their ability to perform multi-
stage mass spectrometry. This consists of consecutive steps of trapping ions, isolation of
a desired precursor ion, fragmentation of the precursor ion resulting in a series of product
ions, followed by mass analysis. A single fragmentation event is termed tandem-mass

Spectrometry, or an MS/MS experiment. Multiple stages of isolation and fragmentation

can be conducted, and are called MSn experiments.

The process of ion injection and storage was discussed previously. Once ions
have been trapped, a precursor ion must be selected for isolation. This is done by
applying a high amplitude “notched” broadband supplemental RF potential to the end cap
electrodes. This potential will ideally contain all frequencies except for that of the
desired ion. Application of this potential will excite all other ions and eject them out of

the trap, leaving only the desired ion. This ion is then subjected to fragmentation by the

44

application of a low amplitude supplemental RF potential corresponding to the secular
frequency of the isolated ion stored at a particular q value. The ion will become
translationally excited, and collide with the helium bath gas, whereby translational energy
is converted to vibrational energy, causing the ion to fragment. Newly-formed product
ions will fall out of resonance with the supplemental RF signal as they will have new
secular fi'equencies. These ions will become focused in the center of the trap through
‘cooling’ collisions with the background gas, following which mass analysis by
resonance ejection can be performed. If the fragmentation is conducted at a high q-value,
many of the low-mass fragment ions may be below the low-mass cutoff and will not
remain in the trap. To circumvent this, ion activation is performed at a lower q-value,
(typically q = 0.25) to ensure that fragment ions remain in the trap while allowing enough

energy to be input into larger molecules to allow them to fragment without ejecting them.

2.3.3.4 Linear Ion trap

While quadrupole ion traps have their advantages, a few disadvantages arise due
to their design. First, as mentioned above, ions entering the trap experience the
quadrupolar field in all three dimensions. Positively charged ions entering the trap during
the positive cycle of the applied RF potential, when the amplitude is high, will be
repelled away fiom the trap and be lost. During high negative amplitudes, positively
charged ions will be accelerated in the z-dimension and pass through the trap undetected.
Only a small range of RF values will therefore allow ions to be stably trapped. This
property is termed ion injection efficiency and is a major limit to sensitivity in ion trap

MS.

45

Another disadvantage arises from the ability to eject ions out of the trap, termed
ejection efficiency. In 3D traps, ions are excited along the z-axis in both directions.
Therefore, when performing resonance excitation, half the ions will be ejected out of the
entrance endcap electrode trap in the direction opposite to the detector. To overcome
this, a positive potential barrier can be applied to the source-end of the ion trap which
will repel ions out to the detector, thereby increasing ejection efficiencies and therefore
sensitivity.

The third disadvantage that 3D ion traps exhibit arises from the number of ions
that the trap can store. Ions are stored at the center of the trap. Therefore, as the number
of ions in the trap increases, positively charged ions of the same m/z that are oscillating
with the same fiequency will repel one another, causing the distribution of fi'equencies to
increase, which causes peak broadening and a loss of resolution. This phenomenon is

termed space-charge.

. . . . . . . 74, 75
To overcome each of these limitations, a 2-dimen310nal linear ion trap may

be used. This instrument employs a normal quadrupole mass analyzer as shown in Figure
2.2 operated in RF only mode to trap ions in the x- and y-dimensions. A positive DC
potential is applied to electrodes positioned at the ends of the quadrupole, which
effectively traps ions in the z-dimension. In the linear ion trap, ions are stored along the
z-axis of the trap, which decreases space-charge effects and allows for greater ion storage
capacity. Also, the ions feel very little of the applied quadrupolar field in the z-axis,
which greatly increases ion injection efficiencies. For mass analysis, ions are ejected
radially in the x dimension, through holes cut into the rods, by resonance ejection similar

to that in the 3D ion trap. Dual detectors placed on each Side of the trap allow for greater

46

detection efficiencies. Linear ion traps therefore allow for much better sensitivities and
are capable of higher resolutions than 3D ion traps. Therefore, the linear ion trap was
employed for experiments requiring hi gli resolution, such as resolving isotopic

distributions of precursor and product ions.

2.3.4 Peptide Fragmentation Nomenclature

Peptides or proteins subjected to CID-MS/MS will proceed to fragment along
various bonds within the peptide backbone. Cleavage at the same bond on succesive
amino acids will result in the loss of the mass of the amino acid. These are called
sequence ions, and comparing ions formed due to sucessive cleavages along the peptide
or protein will allow the sequence to be determined. Depending on whether the charge is
retained on the N- or C-terminus along with the location of the cleavage will result in
different different sequence ions. A summary of the nomenclature for peptide
fragmentation is shown in Figure 2.7. Peptides or proteins subjected to CID will

typically undergo fiagmentation at the amide bond, forming b- and y- type ions. When

the charge is retained on the C-terminus, a yn —type ion is formed, where n represents the

number of amino acids contained in the ion. If the charge is retained on the N-terminus,

a bn-type ion is formed.

47

a] bl CI 02 b2 C2 03 b3 C3

\zm

/

\

O—

/
\

\

2m

/
/O=O

 

 

 

 

 

 

 

 

 

\
HZN T" T/ 1? OH
O R' H O R'"
L. __ __ _. _. ._ g __ __
x 3 y 3 23 3‘2 Y2 Z2 3‘1 Yr 21
Figure 2.7: Nomenclature for peptide fragmentation
2.3.5 Instrumentation

Two different mass spectrometers were used in the current study; i) a Thermo
Fisher Scientific (San Jose, CA) model LTQ linear ion trap equipped with nanospray
(nanoESI) ionization, and ii) a Thermo Fisher Scientific (San Jose, CA) model vMALDI

LTQ XL linear ion trap equipped with a MALDI ionization source.

2.3.5.1 MALDI-LIT Analysis

For MALDI experiments, 1 pM peptide samples were prepared in milli-Q water,
as well as 50 mg/mL solutions of DHB matrix in 50% acetonitrile and 0.5% TFA. 0.5 11L
each of the peptide and matrix solutions were spotted onto a stainless steel 96-well

MALDI plate from Thermo (W altham, MA), and allowed to crystallize. The plate was

48

then inserted into the instrument, and the correct spot was located- The laser power was
adjusted to achieve maximum signal, but still obtain optimal resolution. This was found

to be 40%. For MS spectra, 15 scans were acquired and averaged. Singly protonated

+ . . . . .
([M+H] ) precursor ions were isolated usrng a wrndow of 4 mass units to ensure that the

entire isotopic distribution was isolated. The peptide was then fragmented using a
normalized collision energy of 25% for a period of 30 msee at an activation q value of
0.2, and 200 scans were collected to achieve maximum signal to noise. As the sample
was being ablated, the signal began to drop, so the laser spot was relocated to an area

where more signal was obtained.

2.3.5.1 ESI-LIT Analysis

For MS and MS/MS analysis of C—peptide, C-peptide mutants or segments,
samples were prepared in water by dilution from 33.7 uM (0.1 mg/ mL) peptide stock
solutions. In order to reduce the observation of alkali metal adducts, the Barnstead water
purification system was left to run for thirty minutes before making any solutions. Then,
all samples were prepared using centrifuge tubes that had been thoroughly washed before
filling to remove any excess salts from the tubes. For examination of Zn-activated C-

peptide, C-peptide mutants or segments, samples of 10 11M C-peptide, C-peptide mutants

or segments containing 10 uM ZnC12 were prepared by dilution from 1 mM ZnC12 and

33.7 11M (0.1 mg/ mL) peptide stock solutions. Samples were injected using a 250 uL
syringe (SGE, Ringwood, Australia) into a nESI source using a silica capillary transfer

tubing (360 um OD. x 100 um I.D.) purchased from Polymicro Technologies (T uscon,

AZ). For Zn-binding experiments, 250 pL of a 10 11M ZnClz solution was first flushed

49

through the lines in order to minimize the observation of sodium and potassium adducts
in the mass spectra. This was done both at the beginning of the experiment as well as
between samples. The sample was then pumped through a PicoTip emitter purchased
from New Objective (Wobum, MA) with a tip diameter of 30 pm at a flow rate of 0.5
uL/rnin where it entered the mass spectrometer using a spray voltage of 1.8 kV and a
capillary temperature of 250 °C. Other instrument settings included a capillary voltage of
20 V and a tube-lens offset voltage of 120 V. For MS spectra, 200 scans were acquired
and averaged. For MS/MS, precursor ions were isolated using a window of 4 mass units
to ensure that the entire isotopic distribution was isolated, fragmented using a normalized
collision energy of 25% for a period of 30 msec at an activation q value of 0.2. 100 scans
were collected to achieve maximum signal to noise. All spectra were acquired using the
high resolution ZoomScan mode. MS and MS/MS spectra of solutions containing 1 11M
C-peptide with 1 uM Zan and 1 uM E27A with 1 11M ZnClz were also acquired using
the same instrument conditions. Instead of flushing with 20 uM ZnClz, the lines were
flushed with a solution of 1 11M ZnClg at the beginning of the experiment as well as in

between samples.

50

CHAPTER THREE

Characterization of Proinsulin C-peptide, C-peptide Mutants, and Their Binding

, 2+
wrth Zn

3.1 Top-down Characterization of Proinsulin C-peptide

3.1.1 Bottom-up and Top-down Proteomics

Currently, two methods for sequencing proteins are being employed, “bottom-
up” and “top-down” proteomics. In “bottom-up” up proteomics, a protein sample is first
digested, separated by HPLC and injected into the mass spectrometer. The individual
peptides may be subjected to MS/MS analysis to sequence the individual peptide ions.
The product ion masses along with the precursor ion mass can be input into a database

where the protein may be identified. However, proteolytic digestion greatly increases

9

. . . . . . 7 77 .
mixture complexrty, making the analySIS more difficult. “Top-down” proteomics is

a developing method which involves injection of a whole protein sample directly into the
mass spectrometer. Similar to the “bottom up” approach, MS/MS analysis may be
conducted, except the analysis is performed on the whole protein as opposed to individual
peptides. The problem of mixture complexity is minimized using this technique,
however other problems arise. Large molecules do not fi'agment well by CID due to the
large number of vibrational modes available for energy redistribution. Therefore, higher
charge states must be selected to provide the molecule with sufficient energy following

collisional activation.

51

It has also been found that greater sequence coverage may be obtained by

. . . 78,79 . . .
fragmenting multiple precursor ion charge states , each gIVing access to different

fragmentation pathways. This results in the formation of complementary sequence
information amongst the various MS/MS spectra, resulting in an increase in sequence
coverage. The differing fragmentation patterns for the different charge states has been
determined to be due to differences in the ability of the ionizing proton(s) to move along

the peptide backbone, causing fragmentation at each amino acid. This is termed “proton

mobility”.80 When the number of protons is less than or equal to the number of arginine

and lysine residues within the protein sequence, limited fragmentation is typically
observed. These are termed “non-mobile” and “partially-mobile” conditions. When the
number of protons is greater than the combined number of basic residues (arginine, lysine
and histidine), the ionizing proton(s) are considered to be “mobile” and thus may move
along the peptide backbone to yield more extensive sequence information. The

fragmentation pathways for different proteins and peptides under different charge state

.. . . 78,81,82
conditions have been extensrvely studied.

Although C-peptide, with a length of 31 amino acids and molecular weight of
3021 Da is not a protein per se; its analysis was performed on the intact biological
peptide, without prior use of a digesting agent. Therefore, a top-down characterization

strategy can be considered as being applied.

52

3.1.2 Characterization of proinsulin C-peptide
The ESI mass spectrum of HPLC-purified proinsulin C-peptide in water, obtained

using a linear quadrupole ion trap mass spectrometer, is shown in Figure 3.1. The

[M+2H+]2+, [M+3H+]3+ and [M+4H+]4+ charge states are all observed. Note that two

. . . . . . . + +
major adduct ions corresponding to ionization of C-peptide wrth K and Na were also

observed, consistent with the previously reported C-peptide metal activation

. 47 . . + + .
experiments. These ions are most likely formed due to the presence of Na and K in

the water used to dilute the samples, as well as on the insides of the fused-silica capillary
tubing that was used for introduction of the sample to the instrument. While not critical

to the characterization of the protonated C-peptide, minimization and control of these

adduct levels will prove important in the Zn2+ - binding experiments (see below).

Analysis by MALDI yielded the singly protonated precursor ion (data not shown).

53

coca

_ _

1‘

com: 32

lr____

82

ea
CONE ooou

 

Il‘

 

‘

  

can...
.u

+NH+M++I+SE
+NH+IN+$Z

\

Barbra

 

arias

cow

 

 

loo—

(%) 901191311an 94119193

Figure 3.1: ESI Mass spectrum of 10 11M C-peptide in water.

54

To obtain maximum sequence information, the [M+H+]+, [M+2H+]2+,
+ 3+ + 4+ .
[M+3H ] , and [M+4H ] charge states were subjected to CID-MS/MS. The MS/MS
spectrum obtained for the [M+2H+]2+ charge state is shown in Figure 3.2. The MS/MS

spectra for the [M+H+]+, [M+3H+]3+, [M+4H+]4+ are given in the Supporting

Information section as Figures S1, S2, and S3, respectively. Extensive fragmentation for
each charge state was observed.

To correlate a product ion in the mass spectrum with a sequence ion fiom the
peptide, the charge of the ion was first determined. This was accomplished by examining
the spacing of the peaks in the isotopic distribution, which follows the relationship given

in equation (18)

z = l/Am (18)

where z is the charge of the ion and Am is the spacing between adjacent isotopic peaks.
Once the charge state and therefore the mass of the ion was determined, it was then

matched with a theoretical sequence ion mass. In the case of ions where a loss of 18

. . 0
mass umts was observed corresponding to a loss of water, a symbol was used.

Sequence ions observed to have a loss of 17 mass units corresponding to a loss of

*
ammonia have a label. The charge state of the ion, if greater than one, is designated on

the label.

55

A summary of the product ions that were assigned and the sequence coverage that
was obtained for each of the individual charge states, as well as an overall summary of
the product ions and sequence coverage, are given in Figure 3.3. Product ions that were
above 25% abundance in the mass spectra are labeled on the product ion and sequence
coverage summaries.

From an examination of Figures 3.2, Figures 81, 82, and S3 and Figure 3.3, it can
be seen that almost complete sequence coverage was obtained for nearly every charge
state. Indeed, the only ions that are not observed are the low-mass b- and y-type ions,
which is expected due to the limited low-mass cutoff of the ion trap. The overall
sequence coverage in Figure 3.3 indicates that complete sequence coverage was obtained.

This high degree of sequence coverage is unusual, particularly for large peptides. One

. . . . . + +
rationale to explain this extensrve sequence coverage, espeCially for the [M+H ]

precursor ion, is the lack of basic residues within the peptide. As mentioned above, the
presence of basic residues such as lysine or arginine will sequester ionizing protons,
limiting their ability to move along the peptide backbone to facilitate fragmentation. For
the C-peptide sequence however, which lacks basic residues, even the singly-charged ion
is categorized as having a mobile proton. Thus, the proton is free to move along the

backbone, allowing for fragmentation at nearly every bond. A few common cleavage

patterns are present across the various charge states. For every charge state, the y9 ion

was observed as an abundant ion, and in the case of the [M+3H+]3+ and [M+4H+]4+

charge states, it was observed as the base peak. This product ion is formed by cleavage

N-terminal to the Proz3 residue. Enhanced proline cleavage has previously been observed

56

. . . . . . 81 . 78 .
to be a dominant process in prollne-contalnlng peptides and proteins under partially-
mobile and mobile protonation conditions. Other dominant ions such y19 and y24 ions

. . . . 64
are also consrstent wrth known peptide fragmentation patterns. The wealth of sequence

information obtained by fragmentation of C-pcptide will be useful when comparing to the

fragmentation of C-peptide bound with Zn2+ (see below).

57

 

 

 

 

 

 

 

 

 

 

 

 

 

 

a...
N>bfl mm
ooom oomfi oofi ooE cog co: cog oomfi F
_ - t t w - ,ii t i -r m
m _ A u
m _ w 2% m...
.. ow—D NLD \W—Q m “ m—D 62AM 0N T m; “Mb
3 + n
Moo—AA +902 +~omn _ £+~o~£ .// N I m firm
/ +~oo5 ./+NoA.~n 9 +IHI
M33 2% ED I H
3 LBS W
82 8: 8.2 cos 0% 8A 0% 8m ..nn
. A _ f. _ r17. A _ _ _ _ _ _ AL . _ _ _ _ _ A A _ A _ _
444.4 it t 1 114 t i A. A. l-..1 41.4 l“. m
/ /_ 4W3? so...» 3 m :e_ 4‘11 . J ./£ M m
:A /2A_ . AA _ A. a.» _ a . c
/ as; an ea.A A" AA.» . an m ..m .a l V W
.2925 Am a 6 can so» can A» m E m. w
:9 2M 1 U M
SD I W: m
3
. .m.

 

 

a.» rem—No.—

Figure 3.2

of C-peptide from Figure 3.1

58

02+
b30

03+
b30

c .. has an arena )4

b3o

Y6

b27

anemia

Y6

b21
Ylo Y9
b22
Ylo Y9

i
B EARhtiittttlttthfittfidt

Yl3

Yl3

8
2+
bl8

l

b
Ylo

bIS
Ylsl2+

E tttlttlttlttttlfihhf that Q

111
blO b12

ELQL

Y25 24

Yzl Y20 Yl9

bIZ
b 2+ 2411142+ 2+
bll '2 bi3 bIS

. Hannahmanna

b9

b7

27
b6 U)

A it a

Figure 3.3: Product ion and sequence coverage summaries obtained from the CID
spectra of the A) [M+H+]+, B) [M+2H+]2+, C) [M+3H+]3+, D) [M+4H+]4+, and E)
combined product ion and sequence coverage summary for all charge states

observed for C-peptide. The [M+H-F]+ were acquired using MALDI while all others
were acquired using ESI. Ions with a relative abundance greater than 25% are
labeled on the summary.

59

3.2 Characterization of Proinsulin C-peptide Bound to Zn2+

As mentioned in Chapter 1 above, C-peptide has previously been observed to bind

to Zn2+. Figure 3.4A shows the ESI mass spectrum of 10 uM C-peptide containing 10

uM ZnClz, obtained using the linear quadrupole ion trap mass spectrometer. Abundant

ions corresponding to the doubly ([M+Zn2+]2+), triply ([M+H++Zn2+]3+) and quadruply
+ 2+ 4+ . .

([M+2H +Zn ] ) charged precursor ions of Zn-bound C-peptrde were all observed.

Notably, ions corresponding to Na+ and K+ adducts were not observed. This was

achieved by equilibrating the transfer line and ESI spray tip with 250 1.1L of a 10 11M

ZnClz solution before the sample was introduced.

In order to unambiguously identify the ions at m/z 1029.7 and m/z 1542.7 Da in

Figure 3.4A as the Zn-bound [M+H++Zn2+]3+ and [M+Zn2+]2+ ions, respectively, high

. -l . . 79 .
resolution ‘UltraZoom’ (28 m/z sec resonance ejection scan rate) scans were acquired

for the two charge states, and are shown in Figure 3.48 and 3.4C, respectively. By
observing the isotopic distributions for these ions and by comparison with the theoretical
distributions for the Zn-bound C-peptide, it was found that the experimental results
agreed with the theoretical distributions, thereby confirming the identity of the Zn-bound

C-peptide.

60

as
New A 32 SA: 32 RA:

_ _

ArchiA/a U +MTN2N++E+H>E m -2:
ooom oowfi 002 003 com coo com com cow

24m A 3m _

L _

 

 

 

 

 

 

 

 

 

 

 

 

 

    

 

 

 

 

 

_ _ _ _ _ . thiil‘AffilIW_ A At Att_I_ _ _ _ 1.1.— Ail
..\ M... MUTEV+2H r
Lafaarmfia ..a. 335%: .
+mm+$m+$a t

affirfé - 2:

(%) eouepuan 911913193
61

Figure 3.4: A) Mass spectrum of 10 uM C-peptide and 10 11M ZnClz B) UltraZoom
scan of the [M+H-'-+Zn2+]3+ charge state C) UltraZoom scan of the [M+Zn2+]2+

charge state.

To characterize the metal-binding site(s) within the Zn-bound C-peptide, CID-MS/MS

was then performed on the [M+Zn2+]2+, [M+H++Zn2+]3+ and [M+2H++Zn2+]4+ charge

states. Figure 3.5 shows the MS/MS spectrum of the [M+Zn2+]2+ charge state. Figure

3.6 shows the product ion and sequence coverage summary for the [M+Zn2+]2+,

[M+H++Zn2+]3+ and [M+2H++Zn2+]4+ charge states, as well as a combined product ion

and sequence coverage summary. Ions above a relative abundance of 25% in Figure 3.3

are labeled in Figure 3.4. MS/MS spectra for the [M+H++Zn2+]3+ and

+
[M+2H +Zn2+]4+ charge states are shown in the Supporting Information section as

Figures S4 and S5, respectively.

For each of these spectra, the presence of Zn within a given product ion was
assigned based on the observation of its characteristic isotope distribution. Similar to the
spectrum of protonated C-peptide, the charge state of the product ions was first
determined in order to determine the mass of the ion. Ions bound to Zn were identified
by their characteristic isotopic distribution in the same manner as identifying the intact

Zn-bourid C-peptide precursor ions. Product ions that contain Zn were given in bold
along with a T label.

The observation of product ions in both bound and unbound forms indicates that
multiple populations of the Zn-bound C-pcptide are present, suggesting that the Zn is not
preferentially bound at any one Specific site or region within the peptide sequence,
making localization of the binding site difficult. This is also confirmed by the

observation of Zn-bound product ions at both the N- and C-terminal regions of the

62

peptide. Another factor further confounding the ability to localize the metal binding Site

to a particular location is the number of fragment ions observed. Abundant product ions

such as the y9 and y19 ions present in the unbound MS/MS spectrum in Figure 3.2 are

also present in Figure 3.5. As mentioned before, previous studies have found that
fragmentation of metal-peptide complexes resulted in a decreased number of fragment
ions, as the metal acts as a ‘non-mobile’ charge, which can direct the fragmentation to
occur at the binding site(s). From the MS/MS spectra in Figure 3.5 and Figures S4 and

85, as well as Figure 3.6, however, almost complete sequence coverage was achieved,

indicating a lack of a specific binding site for 2112+ to C-peptide.

Despite the similarity of the two MS/MS spectra, some conclusions may be
reached when examining the distribution of unbound, “mixed” (i.e. bound and unbound),

and bound ions. Unbound sequence ions are observed to occur for small sequence ions,

where binding is not observed to occur before the y5 ion. Ions corresponding to cleavage

in the middle region of the peptide appear to be mixed, while high mass sequence ions
corresponding to cleavages near the termini of the peptide are observed as only Zn-bound

ions. This would suggest roles of multiple residues in the binding of Zn.
, + 2+ 3+ + 2+ 4+ ,
Note that in the [M+H +Zn ] and [M+2H +Zn ] charge state product ion
and sequence coverage summaries in Figure 3.6 the presence of y-type ions containing
Zn were only observed for y5 and above . The y5 ion corresponds to cleavage at the N-

terrninal side of the Glu27 residue, which as discussed above has shown to have potential

biological activity. The observation of bound y5 and unbound y4 ion would therefore

63

suggest that some binding is occurring at the Glu27 residue, which is consistent with the
studies discussed previously showing that transition metals bind to deprotonated

carboxylates of acidic amino acids.

64

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  
 

ooucpunqv

 

rmmmee

 

 

 

 

 

 

as
ooom cog 83 85 cog 82 83 co?
5» _ ON A:
L La: LA LA; L La L AA :n
LAL. 2% L3» in a» LLL \ J 2.3% .2 n\ 29
L. a— _ a a—
n L a La» L»... A» 62% LA; .2: \ M +22% 2% AL...»
aALL—Am + an
2 ”3%: LSDA+~Afi£A+NAauAm +9 «Am
+~_+~:N+—>: : .2 +23%
Lag. 31:3. An
+252
82 2: L oooL coo cow och coo com
L A L A _ LL hL L4 _e__ rf__ 1L LILL l .lllLiltL Tr_lt_.L
a..- L L. a- L,
_ A.» / LL LLLA
a . J
L La eA L aw
2.3a aD o L. LLOAA
...aAh cab on— 063 m: mAm
GA:AA 6 09A _ own
2 2»
+32» L A— _ an «A. eAAA
2 L
E o D a%
a.»

oouepunqv

 

lam—wow

CID MS/MS Spectrum of the [M+Zn2+]2+ charge state ( m/z = 1542.7)

of C-peptide from Figure 4A. Ions observed to have Zn bound are labeled with 1* in

bold.

Figure 3.5

65

cweaningSLWQ

A E Humansapaiaawaaaaa
B annular a an L n6 a
D an} an an: a no

Figure 3.6: Product ion and gequence coverage summaries obtained from the CID
+ +
spectra of the A) [M+Zn2 ] , B) [M+Zn2++H+]3+, C) [M+Zn2++2H+]4 charge

states of C-peptide as well as D) the overall product ion and sequence coverage
summary. \ indicates an unbound b or y ion, \ indicates only the Zn—bound

fragment was found, and \ indicates that both bound and unbound ions were

identified.

3.3 Top-down Characterization of the Proinsulin C-peptide Mutants

To gain a greater understanding of the effects of the potential residues involved in
metal-binding to C-peptide, several single amino acid mutants were synthesized using
solid-phase peptide synthesis. Each of the acidic glutamic acid and aspartic acid residues
were individually replaced with alanine (termed E27A, E1 1A, D4A, E3A, and BIA) then
the mutant peptides were first characterized in their protonated forms by CID-MS/MS,

and compared to the data obtained from protonated C-peptide. A representative CID-

MS/MS spectra for the [M+2H+]2+ charge state of the E27A mutant is given in Figure

3.7. The CID-MS/MS spectra acquired for the [M+H+]+ through [M+4H+]4+ charge

states for each of the five mutants are provided in Figures S6 through 824. As with the
wild type C-peptide, CID-MS/MS spectra for the singly protonated precursor were

acquired using MALDI. A product ion and sequence coverage summary from the

[M+2H+]2+ charge states for each of the five mutants is given in Figure 3.8. The overall

product ion and sequence coverage summaries from the 3 observed charge states for each
of the 5 mutants for is given in Figure 825. Product ions whose relative abundances were
observed above 25% are labeled on the summaries. When compared to the MS/MS
spectra of protonated C-peptide, the spectra for the mutants were all very similar,
demonstrating that the presence of the single mutations did not significantly affect the
fragmentation behavior. As with C-peptide, complete sequence information was obtained

from each of the mutants.

67

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NE
oooN 83 82 oo: 82 82 oov L 82
_ L p L L _ L L i L L — L
. L . 11;... I . L
oLND /
NNLA oumn / ooNLA V
ow» m.
LN an L: \ u
LL 5 LL .203 $2» 55:2» I m.
are. a... L
2L 2 ram;
o
82 oo: ooo. ooo oow ooh oom
L L L _ L L L L L L L L _ b L L L L L L L
ItIIrI- r. L . J44. 1.— it 11 «21: J<L44Lfilr1ii0mwt1fiLLLri 1_ LI. .
Law.» ‘ L can a .L/ J J IL: M.»
N 0 LA on on FA 8» I.
L L. V
: L L
e LL co; 5 I m.
0..
L5 3 o; I M
So— I 9
LL ISLE

CID MS/MS Spectrum of the [M+le+]2+ charge state ( m/z = 1482.6) of
68

Figure 3.7
E27A.

 

.LLLLLL... fig. LLLLLLLLLLLEL... .
.LLLLLLLL ...LLLfiLLLLLLwLLLLLe. .
..LLLLL...OLLLLLOLLLLLHLELL... .
...LLLLL LLLLLLLLLLLLWLLLSL... .
ea... ..LfifiLLLLLLag...

69

3.4 Characterization of the Zn2+-bound C-peptide Mutants

Identical to the method employed for analysis of the Zn-bound C-peptide, each of
the five C-peptide single amino acid mutants were analyzed by ESI-MS in the presence
of Zn. Figures 3.9 and 3.10 show the resultant ESI mass spectra obtained for the E27A,

E1 1A, D4A, E3A, and ElA peptides.

Similar to the Zn-bound C-peptide spectrum in Figure 3.4A, the doubly
([M+Zn2+]2+), triply ([M+H++Zn2+]3+) and quadruply ([M+2H++Zn2+]4+) charged
precursor ions were all abundant. Note that while the spectrum in 3.4A did not contain

+ +
Na and K adducts, these adducts were all present in the C-peptide mutant spectra,

albeit at variable abundance. This will be discussed later in this thesis in regard to the
relative binding experiments.

CID-MS/MS spectra were then acquired to determine if the fragmentation
behavior of the Zn-bound C-peptide was altered upon substitution of a single acidic
residue, and whether this allows information to be obtained in regards to the metal
binding. Ions were labeled using the same process as for the Zn-bound C-peptide

experiment.
’7 ’7
The MS/MS spectrum for the [M+Zn"+]"+ charge state of the E27A mutant is

shown in Figure 3.11. The MS/MS spectra for the [M+H++Zn2+]3+

and
+ 2+ 4+ . . . .
[M+2H +Zn ] charge states are shown in Figures 826 and 327, respectively. Figure

+ +
3.12 shows the product ion and sequence coverage summary for the [M+Zn2 ]2 ,

[M+H++Zn2+]3+ and [M+2H++Zn2+]4+ charge states, as well as a combined product ion

70

and sequence coverage summary for the E27A mutant. Figures 828 through S39 contain
the MS/MS spectra for the [M+Zn2+]2+ through [M+2H++Zn2]4+ charge states of the
remaining four mutants. A product ion and sequence coverage summary for the

[M+Zn2+]2+ charge state for the remaining four mutants is given in Figure 3.13. Overall

product ion and sequence coverage summaries for the four remaining mutants are given
in Figure S40.

When comparing the MS/MS spectra for C-peptide in Figure 3.5 and that for the

E27A mutant in Figure 3.11, there are many similarities. The y9 ion corresponding to

cleavage N-terminal to proline is dominant in both cases, as is the Zn-bound b30 ion. The

overall sequence coverages are similar, and when examining the distribution of bound
and unbound sequence ions, a similar pattern was observed to that of C-peptide.

Another measure of the similarity of the binding was obtained by examination of

the ratio of bound: unbound intensities for dominant ions. When looking at the y9 ion,

for example, the unbound ion is dominant in all cases, with the Zn bound ion occurring at
low intensities. These ratios remain consistent for similar charge states of C-peptide and

E27A.

One main difference however, is the lack of a Zn-bound y5 ion in any of the
. . . + 2+ 3+
spectra of the E27A mutant. This ion was observed in the [M+H +Zn ] and

[M+2H++Zn2+]4+ MS/MS spectra of C-peptide, although at low abundance. The lack of

binding on the N-terminal region of E27A suggests a role for Zn binding of the Glu27

residue.

71

 

When comparing the mutant spectra with each other, again, there were again very

few differences. Similar dominant product ions as well as binding ratios of the y9 ions

indicated that one amino acid does not play a crucial role to the binding of Zn. Again,
‘bound’ sequence ions at both the N- and C-tcrminal region of the sequences were

observed, indicating non-specific binding.

72

 

 

 

 

 

 

 

 

 

 

 

 

 

 

as
ooom com: cog cog cog oooL com com oov
Pi.......r-< 1P1-4u..h.#
.NL..=N+E
arm +3: Lafifmmifi ..
/. l
U frat);
armmeiE m
..L..aN+.m+§ 2:
_ _ L L _1 r _-_ _ _l_ t_._ L L _ 1.1 L L- L blw r1? _ ...L.iw:l_ _-_:_ _
+NH+NGN+ ”— L. Mv+mv+2
grit): LL/ ...Laamefié i
m L...L.:m+2_ i
harmfmmifi m
+nm+N§N++m+EH On:
_ . _ r r _ _ . _ . r _ r L . _ _ . h . _ _
«J i i l. i l l 1
+NH+NGN+E
Liza ..L+.=N+.:~+§ ..
< /+nr=m+a .l
armrmfié w 2:
+n~+~nN++m+a

(%) aouepunqv aAncpH

Figure 3.9: ESI mass spectra of lOuM ZnClz and lOuM A) E27A, B)E11A, C) D4A

in H20

73

 

 

 

 

 

 

 

 

 

 

 

 

 

 

as
ooom com: 82 83 82 02: cow coo o9c
TL _ hLiL _ L _ “##le _ FL._4_._.1__+_ %4-r.L._ _/_ _J_w_ -_4_._
L+vm+mv+EH l
+~L+~=N+$a L.
.LLLNLLNrmNEL u
magi): 1
m M._....L..L.:m+2L 1
+~L+M++IN+SL
.LFLNLLELLEL i 2:
____LLrerLL_LL ______L
31 i 1 . i .41 - . -
L...
$752 I
LNFLNEL ..L..=N.+.mm+2_ .
+~H+=N+$E /
/+mm+:m+zu_
< armrmfia i
+mm+~cN++m+2H I oo—

74

(%) aouepunqv sixtielag
Figure 3.10: ESI mass spectra of lOuM ZnC12 and IOuM A) E3A and B) EIA in

H20

 

 

 

 

 

 

 

 

 

 

    

   

 

 

 

 

 

 

ES
coca 8o— oof oo: 82 82 co: co:
_ . .-
M. x m ./ . m. . IL .
m m 5» x. m .. so:h m .. .. M. . . w . .
. m . . . . .. .. . . . . .. m I
Em .3. m .2.» .2: 2w... .3. £2: ... TI an“... m .. H W.
.5 an 8.» 2%.. .. e; .\ +~ .3... £52.» I w.
+ w; t .. =. m
8% NA 3.2 ._. a— t I +~+c~% I. O
+~+ +~+Fn .13 O
3 Rafi; $23.3 .25 I
:22 2 h +N..w£.+~..w£ I
com. co: 2:: com com con coo com
. . . . I. I-I._..I J; z . ..
.. MS“... ... mm M WM... .M/ 55 § ...p. M... I
Mr" ... m .33.: Ms; M M Z: : ma no; i M on; .39 V
.m .N m.— .roA—m o M. % I .fl 0 N. . ..a rl 0..
n x r. a ‘ tan _ n a £3 a o.» 0% , m
.. an as» I m
9
9A Immmw

CID MS/MS Spectrum of the [M+Zn2+]2+ charge state (m/z = 1514.44)

of the E27A mutant of C-peptide from Figure 9A.

Figure 3.1 1

75

EIE‘ELIEISLQ

GlS LQ
\\

A“testersIIIIIILIIIII
aEAEIIIIIIIIIIIIIIIIIIII
I

CIAEIIQIIIIIIII
we new It IIIIII «I I

Figure 3.12: Product ion and sequence coverage summaries obtained from the CID

spectra of the A) [M+Zn2+]2+, B) [M+Zn2++ll+]3+, C) [M+Zn2++2H+]4 charge
states of E27A as well as D) the overall product ion and sequence coverage
summary. \ indicates an unbound b or y ion, \ indicates only the Zn-bound

fragment was found, and \ indicates that both bound and unbound ions were
identified.

76

 

- “teammateanew

AEItewelwaaaae
C13%)lele)IIIIGQGIIGEIEIURIGIIQ
nmamaNew

Figure 3.13: Product ion2 and sequence coverage summaries obtained from the CID
spectra of the [M+Zn2 ”1+charge state of A) El 1A, B) D4A, C) E3A and D) ElA
mutants of C- -peptide. \ indicates an unbound l or y ion, \ indicates only the Zn-
\

bound fragment was found, and indicates that both bound and unbound ions

were identified.

77

3.5 Effect of Amino Acid Mutations on the Relative Binding of Zn2+ to C-peptide

While it has been shown that substitution of a potentially metal-binding amino
acid has little to no effect on the fragmentation behavior of C-peptide, it is possible that
elimination of one of these residues could play a role in the amount of Zn that the peptide
can bind, thereby providing information regarding the residues that are important to C-

peptide metal interactions. Therefore, ESI-MS spectra were acquired in triplicate for

. . . 2+
each of the C-peptide mutants in the presence of one molar equivalent of Zn .

Representative spectra for each peptide are shown in Figure 3.4A for the C-peptide and in

Figures 3.9A-C and 3.10 A-B for the mutants. Compared to the wild-type C-peptide

. . + + . .
spectrum in Figure 3.4A, where no Na or K adduct ions were observed in the spectrum,
+ + , ,
Na or K adducts of variable abundance were observed in the spectra from the mutants

in Figures 3.9and 3.10, indicating a lower affinity for Zn2+ for these mutants.

To quantitatively measure the amount of Zn binding for the C-peptide versus the
mutants, a ratio of bound: unbound can be calculated from each spectrum. This was
performed by summing the intensities of each of the charge states observed for the ions

containing Zn, and dividing by the summed intensities of the charge states observed for

+ +
the unbound ions (Figure 3.14A). Also, to account for the presence of the Na and K

salt adducts in the mutant peptide spectra, a modified bound: unbound ratio was
calculated after summing the intensities of the salt adduct peaks with the unbound ions

Figure 3.143). It can be seen from Figure 3.14 that an approximate 50% decrease in the

amount of Zn binding was observed in each of the mutants compared to the wild-type C-

78

peptide. Thus, while mutation of any individual acidic residue results in a decrease in Zn
binding, it appears that no specific acidic residue is critical. This is consistent with the
MS/MS data which demonstrated that the fragmentation behavior did not change upon
substitution of a single acidic residue.

Previous activity studies conducted on C—pcptide have been performed using

. . . 2+ .
equimolar concentrations of C-peptide and Zn in the nanomolar range. The mass
spectrometry studies conducted here were performed using 10 uM concentrations of both

. 2+ . . .
C-peptide and Zn . In an effort to examine any concentratlon dependence on the Site of

binding in C31 and the E27A mutant, as well as the relative binding of Zn2+, ESI-MS

and ESI-MS/MS experiments were conducted on the full length C—peptide and E27A
mutants at 1 pM concentrations (data not shown). The MS spectra were essentially

identical, suggesting that a 10-fold decrease in the concentration did not significantly

. 2+ . . . . . .
affect the srte of Zn b1nd1ng. Results from a relative binding experiment (data not

shown) showed an overall decrease in the amount of Zn2+ binding at the 1 pM

concentration, although the relative binding ratios of C-peptide : E27A between the 10
pM samples and 1 11M samples remained relatively constant. This suggested that a

decrease in concentration did not affect the relative binding.

79

 

'8

 

N
8

 

go
8

 

E(Bound)/E(U nbound)
'8 '8

p
8

 

 

p
8
1

C31 EIA E3A

.N
8

 

 

'8

 

'8

 

 

 

2(Bound)IE(Unbound)
.0 r‘
8 8

 

 

p
8

i
C31 EIA E3A

Figure 3.14: Plot showing binding ratios of the [M+Zn2++nH+] to [M+nH]n+
for C-peptide and the 5 mutants A) Not accounting for [M+Na]+ and [M+K]+

adducts and B) accounting for the [M+Na]+ and [M+K] adducts. Intensities for
the 2+, 3+, and 4+ charge states were summed. Standard deviations for the
triplicate measurements are shown on the plot.

D4A EllA E27A

(n+2)+

80

3.6 Characterization and Activity Studies of the C-terminal Pentapeptide of C-
peptide

In order to more closely examine the effect of a single acidic residue on Zn

binding and activity, a c-peptide segment containing residues 27-31 (i.e. in the absence of
the other four acidic residues) was examined. As discussed in Chapter 1, this fragment of

the C-peptide sequence has been shown to exhibit biological activity, and requires the

. 14, 31 . .
presence of the Glu27 reSIdue. Thus, as an extensmn of the mass spectrometric

binding characterization studies described above, the EGSLQ peptide was synthesized
and subjected to MS and MS/MS analysis in both its unbound and Zn-bound forms.
Similar experiments were also conducted on a AGSLQ peptide sequence in which the
previously reported critical Glu27 residue was substituted with an alanine.

ESI-MS spectra for the EGSLQ and AGSLQ peptides in the presence of 10 uM
Zn are shown in Figures 3.15A and 3.158, respectively. From these spectra, it can be
seen that there is a dramatic decrease in the amount of binding as compared with C-
peptide and the full length single amino acid mutants. This is likely a result of a
decreased number of locations for Zn binding, as the EGSLQ peptide contains only a
single acidic residue. While triplicate experiments were not conducted, decreased binding
was observed for the AGSLQ peptide, consistent with that expected due to the lack of an
acidic residue. Note however, that the C-terminal carboxylate group is available for

binding in both peptides.

81

 

 

l l
I l l l l T l
800 1000 1200 1400 1600 1800 2000
m/z

l
l
600

 

 

[M+H?"

 

[M+H?"

--L- 4+.ll. llLlL -

[M+Zn2+] 2+
400

 

--.1

200

 

 

_ [M+Zn2+] 2+

1

(%) aouepunqv 9A11€[9}{

Figure 3.15: ESI mass spectra for 10 pM Zn2+ and 10 11M A) EGSLQ and B)
AGSLQ.

1001

82

 

CID-MS/MS spectra from the [M+Zn2+]2+ precursor ions of the EGSLQ and

AGSLQ peptides are shown in Figures 3.16A and 3.168, respectively. Limited sequence

information was observed for both peptides. Dominant product ions corresponding to the

102+ To

neutral losses of water (labeled as M and M 02+) were observed in both spectra.

This is similar to the full-length peptides.
Examination of the other dominant ions in the CID-MS/MS spectra in Figure 3.16

allow for insights into the binding properties of this peptide. In the MS/MS spectrum of

EGSLQ, the Zn-bound b4 and unbound y1 ion are dominant. These two ions correspond

to cleavage at the same bond: the Leu/Gln amide bond. The Zn—bound b4 ion indicates

that Zn is retained on the N-terminal region of the peptide. This end contains the
potentially-critical Glu27 residue, which from earlier experiments on the full-length
single amino acid mutants was implicated in the Zn-binding. In the spectrum of AGSLQ

in Figure 3.16B, this ion pair is much less dominant. Instead, there is a dominant bound-

y3/ unbound-b2 ion pair. This corresponds to cleavage of the Gly/Ser bond, where the Zn

is retained on the C-terminal region. This could indicate that binding is occurring at the

C-terminal carboxylate of the peptide or at the side chain hydroxyl group of the Ser or at

a carbonyl oxygen of the amide backbone. Also of note is the presence of the bound-b4

/unbound-y1 pair in the spectrum of AGSLQ is observed, albeit at much lower abundance

compared to their abundance in the spectrum of EGSLQ. As there are no acidic residues

to bind Zn in the b4 fragment, which would again implicate Zn binding involving the

carbonyl oxygens or the Ser residue.

83

 

The ratio of the b4/y1 pair to the y3/b2 pair indicates a measure of the competition

for the Zn at the N- and C- termini for each peptide. The spectrum of EGSLQ contains

the y3/b2 pair at only minor abundance, indicating a much weaker competition of the C-

terminal carboxylate for Zn. The y3/b2 pair for AGSLQ is present at a much higher

abundance, indicating a greater affinity of the Zn to the C-terminal carboxylate.

2+ . . . . . . 83
Zn has been shown to exhibit tetrahedral coordination geometry in proteins

These structures were based on solution-phase calculations where water molecules aid in
the stabilization of the metal. In the case of EGSLQ, there are only two main sites for the

Zn to interact, and because a gas-phase ion is being measured, no water molecules are

. . . . . 3+ . . .
available to interact wrth the Zn. Studies conducted usmg Cr on ac1d1c peptides have
. 59 . . .
proposed two-coordinate gas-phase complexes, which would suggest a Similar

geometry for the gas-phase EGSLQ-Zn2+ complexes. The AGSLQ peptide only has one

carboxylate group, and as mentioned before the Ser residue or possibly the amide
carbonyl groups could aid in the stabilization.

While the binding data on the single amino acid mutants and the results from the
EGSLQ fragment indicate a role for the Glu27 residue in the binding of Zn2+ to C-
peptide, it is necessary to correlate these results with a specific biological activity. Thus,
similar ATP release experiments to those performed on Zn-activated full length C-

peptide described in Chapter 1, were performed using the EGSLQ and AGSLQ peptides

(these data were provided by Ms Jennifer Meyer and Prof. Dana Spence).

84

 

ATP release experiments were performed using a luciferin/luciferase

chemiluminescence assay.47 Rabbit erythrocytes were incubated for two hours with an
appropriate volume of 8.3 uM stock solution of peptide and ZnClz in a physiological salt

solution to yield a final peptide and 2112+ concentration of 10 nM. Chemiluminesence

was measured using a photomultiplier tube (PMT). F i gurc 3.17 shows the summary of
the ATP release data with n = 7, and statistical significance was calculated using a single-
sided students t-test with p = 0.025

Incubation of the erythrocytes with the EGSLQ peptide in the presence of Zn
resulted in a greater than 100% increase in ATP release. Notably, this was statistically
equivalent to the ATP released upon incubation of the erythrocytes with the full-length
C-peptide. The AGSLQ mutant, however, showed no increase in ATP release activity
over the control level. These data are therefore consistent with the mass spectrometry

results described above, that indicate a possible role of the Glu27 residue in Zn binding.

85

600

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

_. O __
O
[\
O
V)
'— V)
' O’
O
_/ _ 8 o
E! \0 V1
1
_ C
V)
U ..
———/ g
in ‘2
to a. o
10 ha O
x +' v
.G
3‘
,. .a/
+fl/‘lb O 3" 6112’
+- /J
8/ o
>-. _ A— o
‘ m
+7 J
+
t O + 8 3
l. g _ O t. z 4
r g E .- 2 in! N
E 'h on “-1:
" i —=I
—-t J 8
J ;i— N
l o
~ " 8
ID ‘1
x O Q ‘2 o
N '" — ll)
Jam-«I ._.
s: ___. >: “““ d
O xx“... 0
L S n1-.. .-.------"""" — O
_.t ..D H
-———a
O O
O O
F... '—

(%) oouepunqv QAHQIQH

Figure 3.16: CID-MS/MS spectra of the [M+Zn2+]2+ precursor ions of A) EGSLQ
and B) AGSLQ.

86

 

 

 

 

 

 

 

 

 

Normalized ATP Release

 

 

RBC C31 EGSLQ AGSLQ

Figure 3.17: Bar graph showing Normalized ATP relase in erythrocytes for C-
peptide as well as the EGSLQ pentapeptide and the AGSLQ mutant.
Measurements were taken with n = 7, and error bars are plotted as the standard
error of the mean. A * indicates a statistically significant difference was calculated
using a single sided students t-test (p = 0.025).

87

3.7 Conclusions and Future Directions

+
It has been demonstrated that proinsulin C-peptide has the ability to bind an ,

and that Zn-activated C-peptide may play a key role in alleviating certain
microvasculature complications of diabetes. From the results outlined in this thesis, using
MS and MS/MS analysis of Zn bound C-peptidc and various single amino acid mutants,
as well as C-peptide segments, no specific Zn binding site is apparent. However, the
results indicate that elimination of any one of the acidic residues within the C-peptide
sequence results in a decrease in the overall binding to approximately 50% compared to
that of the wild type full length C-peptide sequence. Activity studies demonstrate a
critical role for the Glu27 residue in the ability of the Zn activated EGSLQ peptide to
release ATP from erythrocytes, where AGSLQ did not return any activity.

Future work should include activity measurements on the single amino acid
mutations synthesized for the binding experiments. This will allow for a correlation
between the activity and Zn-binding for full-length peptides, as opposed to the C-terminal
pentapeptide fragment. While the Glu27 residue has shown to be active in the segment, it
has not yet been demonstrated that elimination of this residue in the full length sequence
has any effect on the activity of the peptide. Since it is clear fi'om the experiments
performed that substitution of one acidic residue does not eliminate the Zn-binding,
synthesis of peptides containing multiple mutations of the acidic residues will need to be
performed. This will allow for a greater understanding of the peptide-metal interactions.
Synthesis of peptides containing four mutations will allow for the effect of a single acidic

residue in the full length sequence to be studied.

88

SUPPORTING INFORMATION

89

 

 

 

 

 

 

 

   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

an.
comm cccm cocm cocm comm comm ocom comm ccvm comm comm co_m cccm
EL_______ :___:_;__._:I___M_a_t. ___:._u_1:__lr+c::_
._ L. _._ __ M : i“: .Z..
w . . . . m I
a» _ m “can map M. _ em» ... M ... .. was M M V
. _ .. a... . a... a. a. ...a... a. ...a ...s... I m
. _ w- . .. I . .
m 3» 5A 5.3 _ _.._ 2% on”.A NE 8% w.
. - o omn . l w
cos c mmn 9
5
:2 Own l
82 Immmm _
coco com: oo: 82 oofl ooE oofl oomc co: oooc ooc cow com
Luca... C . ..t.:_.._..___:Ml_IC .1::F:_L:qw_:_l__:::::__4_:clt___te::
. M .. . .m .. E . .1 M.
w u a .. .. n I u a
.. m .. .. . m5 .. mm. .a . oi .. I
. . c . . . c N; :x a» _.
\ . . M05 .. 2% ...o5 ..3» / .. V
. c. 8o c M .. . /. . I a.
so . c M a; so .... as» m
o_m% cmcn M M 2% mi m.
m l u
5% ocmNA «5 2% m
3 IE:

charge state (m/z =

++

MALDI CID MS/MS Spectrum of the [M+H

Figure 81

3021.3) of C-peptide

90

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

m}:
cocm 83 83 cc: 82 82 ccE ocmM comM
/ M M \ M . .
M. .. M of a; a; _ of M . . I u M. ... l
MN» 3.» .. .. .. 2c M. M. M m» M W» F. W
mo .5 to o; +N93.. ... MN 2:; I m
M .. M a. . p
M E M ... I“ 8
Mo 2 m5 ...... cameos :nM r. m
v.5 .... mm .
om M+m Q +mmmD l
..Ne chaos; so
[ammo
2% ..Nmfi :maamo
oo: 82 ooc com com coo com ooMo com
_ p — _ p h _ - — b — b M _ _ _ — _ _ p p u n _
II. I . 4:.II- . III, fl: .:-
.. a M an a I
. M M . M of... a: ...n V
..N m 8». c a o I a.
a» .M M +~ M 5 so can m
+mMmD M p
\ 33 +mc5 .l M
m5 +mmmn +m_m~m %
33%.;ch
Immw:
a»

charge state (m/z = 1007.6) of

3+

+

CID MS/MS Spectrum of the [M+3H ]

Figure S2

C-peptide

91

 

as.
88 82 8.: 8: 88 82 8... 8: 82 8:

____f_______rt.__

 

 

 

 

 

 

 

 

 

 

 

 

 

 

/ / ._ \
a 8 c . M .
o c a; M ..
2D ...x 1% M m5 ... m; M... M; x. W.
a .. .. I
an a... _. .. 0.; m.
2.25 ..m
+mmm£ 1.. MW
Immmow
In
oo: 82 com com com coo com oov com
. .. . _ _:_ _ _ _ _ _ woo .MI_ _I.
_ 1 1.1 .4 : . d
..\ . ../ mm ....... on M u M w; M M... E ....... 3» 1
2m M +N QM. M erl +N MM . .m MM :nM 3% “D
M Fran. . JML+~2nMFNI +~ l W
8» +N N _ to. M n
+mocmp M +N 5 .2 M. o» u
889:2» a: x \ +~~£ I m.
ca 0
a
+mNND PA +NE~A +NV—D m— l
_ +N n .
§\ $8588»..ch ImmwM m

CID MS/MS Spectrum of the [M+4ll+] 4+ charge state (m/z = 756.0) of
92

Figure S3
C-peptide

 

 

  

 

 

 

 

 

 

 

 

 

    
 
 

 

 

 

 

 

 

 

93

 

__
NE. 2
cocm 8% com: 8: 82 82 so: 82 82 W
.II_J_ . —WLI«»1. »_ 4 .. ... 4 . e
. . M. 1. ,. _ I u
M M w M m u
w w; .. .. M M _ .l..L Q ..s W. m.
a; :8 o; M in M M I m. .m
2 u m .2 , +
n . +282 +883 ...: .MI W H]
.. . .. 9
. . m ... H
428.» v; .5 38» +2 £\\ .... I H+
.8 . n
+8.. ._ - .8 .5 z
8 23:15.5 :8 +2 I83... W
+9. 3 I.
m
8: cog coo cow o2. coo com oov com ..m
r — FF.I_ 1p _ _.1 .4 _ II . «_.IJ.1I1. . m
. an... :1... 8 . .. m
.4... M .. ”M m M m M .... w
j... .1: .__. a v .w
ova wm cc.» ’ ...u PA «D M 3% I. ”On... m
fl. 5 n 0% m» n“ v M. S
.3» . o a . o n I w M
+~ +283; 85 m m
8.. anamzzfi on a 2 I C
.8 c +~+ a+~+ .— 4
+93% +N+¢N~h +N+ ND afiN fl S
Fm I e
..fiunn 29 We
n

 

 

 

1007.6) of C-peptide

 

 

   

 

 

 

 

 

 

 

 

 

 

cocm 82 com: cot coE oofl cc: ccfl oofi co:
r..___..____..__..P____I_I_
\ 2.» \
2n / I
N; . W
x r m
023 W.
I w
9
r
I82.
:9 m
co: coc_ coo com com ooc com coo com
13L _I.____4I_I.I+_ _1_II__IP.I_I _
W .. 1 ...... ./
8c ...m... .. .. ....../ / .... i I
2.» Max... 2 . M. . ...... .122.» E ... V
+2. M .M . ....2. a...“ . M+~2£ .t. / ..vn I Q.
is. M IT. .5 a. 8 M m/ a; o m
a .. +~ . c P
+2 c .. _ .... 8.» . I m
+2...» +9. 5: as» m
......_....... / a...
caching +22.— 82922.»

+ 2 4+
CID MS/MS Spectrum of the [M+2H +Zn ] charge state (m/z =

772.0) of C-peptide

Figure SS

94

 

 

 

 

 

 

 

 

 

 

a...
comm oc.m cocm cocm ooc_m comm cocm comm ccvm comm comm cc_m
....__....:... ...._.4.I.. ..-. «..rr...._._IC____.1fr
. .... .. ...: ...... .o fix...
8.. a ...... 8M 8.. o c .. no as: N» .... ONE I
8c. a 5 8c we. 8% W
....E o no I m
m.
mm
o n 1 u
as. . c
omn 3% In
.2 law:
cocm coo. co... co: coo. com. cc: com. com. co: coo. coo
2.. .. ...» 9.
.N 8.. c .... . ...» 2 . cm; .S
n .m% n _ .n V
mm.» o c; 2% m n m.
o. u
. .. .... m.
o
. .A a
a... BS...

charge state (m/z

++

MALDI CID MS/MS Spectrum of the [M+H ]

Figure S6

2962) of E27A

95

 

 

 

 

 

 

 

   

 

 

 

 

 

ca
ooom 82 com. co: ooo. com. ooE com.
.. m .. . _.
/..3 5m .\ o.\ .... .... ... :MM. ... /. . . _
8 c . ...» ...» z .. m I
c a... 2.. .. c ......Nac av.
8; .. .. I n
2.. ....A 88c 8.3 w.
8 ...
c . . m; I
+m rm... +mom% m
88.. «8» so I
Ivmcm.
Looms
com. co: coo. coo coo coo coo com
_ . _ . . . . _ . . . . _ . . . .
I14}! 111114 444 dlijW 4.11114
m .. MI fl
. M .. ... no ... I
r... .... .. 8 V
. % o3 D 0% .I m.
w
.. - ..
O
a
9A Ivmmom

gc state (m/z = 988.6) of

char

3+

+

CID MS/MS Spectrum of the [M+3H ]
96

Figure S7
E27A

 

 

 

 

 

 

 

 

 

 

a...
cocm coo. coco coo. cco_ coo. cow. com.
{14. . .IMI.II_«. I.I_I .1. .I.+— Fr W HIM...
com. co_M coo. coo coo coo coo coo
.... . M ..M . . _ . .... _
. M M .m o M M M mm» m
2% 8xwxs~ / . M M MNSo M; . \ o
:c 2» 2MM . w .o. M . m
.. .... . . . : .. .
8.3 M; man $8: 3:» ..IMM.LM m. ox
. + c . o
8.385 88.. M 8....A .... oM +~
.. 2
.5585 .M +N .—
9» +22; +mmmp +mzn

 

souepunqv

 

memo

aouepunqv

 

Lag

CID MS/MS Spectrum of the [M+4H+]4+ charge state (m/z = 741.5) of

Figure S8
E27A

97

 

NE

  

 

 

 

 

 

    

 

 

 

 

 

 

 

 

4

«ma

comm cc.m cocm ccom cocm comm ccom comm ccvm comm
Trp.._...._.t. ..I..qf.t...a_..1._ :.I.L._._.
.... Mow .... ..NMM M... “M
M . .. 2.. M .. ......
o.» o o MN mm» M om»
oomo c coon
comp
2.). .
con
02
cocm coo. co... co: coo. com. cc... com. com. cc:
£.t..a.-...£1t .1. .
.. M ... —<
M.. M .... m ..... / ./ ...;
mm» .3 M .. ./ ...o w; M .5 N... m. :o
.nmo M M .c. 8.. M o o
omo M ..ox 8.» .o
. M.
M c
of
.mo
2%

. _
M-..

 

comm cc.m
...... M.
8..
.5.
coo. coo

..

mmn

 

omma

 

 

 

aouepunqv

Immomo

9 OUBPUI‘I (1V

Immo;

MALDI CID MS/MS Spectrum of the [M+H+]+ charge state (m/z =
98

2962.5) of E1 1A

Figure S9

 

 

 

N...
cocm coo. och co: coo.
I - III I II--- I‘ll i - I ..
.N . / . ...o .....o .
mm.» c a com.» Wm c to 8.5
3% cm% AM ...... /.
.. ..o
_moM M .A
of
o...
82 co: co? coo cow
Flu—[LlaIn—II—ll—ILIr—II—ILLLIL
M M_
2%..
...
......
_

 

 

 

 

8»

 

coo—

 

 

 

 

ccE com.
M .. . .
+882 ....x 8.3 $53. I
.. .. . v
23M 3 cmn m5 .. +mom£ I m;
cm +m +mmm£ U
+mo o 2% P
. e
.. I. u
.. 9
“cs 9
PM I
.5 I838
ocM. cco coo
. _ifim4flWW41MIk » w
.. ..
w
I n
u
D.
B
I u
0
a
fi
Immcmo

 

CID MS/MS Spectrum of the [M+2H-F]2+ charge state (m/z = 1482.6)

Figure 810
of E1 1A

99

ooom

oooo

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NE W
8.
oofi oo: oof oomo ooE oofl m
-. --....iiiiIIl-l!ll|iiilI IIIlIII|lI z
\ / o. , _ .t _ m
o ~ (
2» CE \23 kw5 to M; 2 . m K zoom I m
a; w; o... ith :3» m. :03 W m
2.. 1 . k I m we
+N p a J1 +~ % m. ..m
l c
+~oomn $03 m «... +~o~> m mm]
$an m5 2% ......N I W
E 3 IE: H
h
oooo owo oow oom oao owm W
W 5 g m
M - V w
a
o 9% l W m
p S
e M
l w m
a C

 

 

Ivmm:
a»

Figure 811
El 1A

100

ooom

oom—

 

$2;

 

ooM: oo:
\ M.

of” Sn—

oooo ooo

+
N
0.----
N
>5
_-u-qf---
m ,,
OO

 

a»

—
‘0

a“-

+

N
O

45"

NE

ooE

N
-

 

oom _

0"...

 

 

 

0»

 

AME

$23 on Lin

 

 

+-5

ooE

v5+moomn+~o~£

2%;

+~_ 5o
+~

----------_-_-..i::

r
3:.

O

 

M ------

oom

\ a
5

 

mi

N ------..-.-

N
o
..D

souepunqv

 

[SEN

souepunqv

 

I334”

4+ charge state (m/z = 741.5) of

l

+

CID MS/MS Spectrum of the [M+4H

Figure 812
El 1A

101

NE
oomm oo_m ooom ooom oowm ooR ooom owmm oovm 83 8% SK

:_____:___ _.______‘____ I___:__

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

_
4 ‘1 4 1 4 4‘ d
4M. M. M M ,5» M My
/ as ”No SM ..fio . ..
... no x fine 2% EM ..... m» V
a a» a ... o.
o a an .. I m
05 w
NND ~N~A I "a...“u
82 9
03 v-m l.
O: imam
ooom oooM oowM oo: 82 82 ooE 82 82 oo: 82 ooo
_ _I_ _ _ _ _ _ _ _ . _ M M _ _I_ — _ _ F
44M1441‘1114 Mdti ‘ 41
M ofin . In . /., / o; M.
ME. M. /. w. M N; _S M a» I
M M”; i 23 z 2o I W.
. 2 2% u
oz» 9“ _ /M: x I m.
of n— w
5»
Immune.—
a»

charge state (m/z

++

MALDI CID MS/MS Spectrum of the [M+H ]

102

Figure 813
2976) of D4A

 

 

 

 

 

 

 

 

 

 

 

 

 

 

9,
NE &
4
oooN oooM oowo oo: ooE oomM ooE 82 A...
II- - I .- ,- if ,, - . Ill . III! I - .IIII, ,1---11II£ I!-1I!I III 2
../ , a M. v W
.5 as T. :2 ..., 5 L M - m
.323 ... w; to $82 3% 33%.. Fm... .... V M
a 2% z .. N .. a. e
n a o M ... :33 u n We
om» +~o n 3% R u a
in o; +~ +N 9 m. .m
5» an 2n m; M. I m H]
. +
\— l m
o; N; . W
IvmoM M H.
E m
82 oo_: oo_o_ owo oww 2.x. 2% own .m
_ _ _ — _ _ p pp p _ _ — _ — h _ _ _ _ h _ _ — _ m — _ m
4 «I A II II :1 I <«4I ‘11 I 111 1.4414 I u
4 :4 can m \ / M a m m m
o ... o D 5 c% on— 1l 0..
g S
5 / W. S
w T n M
a; . u S
o n on D. M
B
I w m
a C
I m
S
. e
a» Immwmo me
n

of D4A

103

 

 

 

 

 

 

 

 

 

 

 

 

 

 

E8
ooom ooa oowM oo: 82 co? co: oom—
M I
+~o~£ murky» I WV:
so; _u _ xxzos m.
.v v ....
p N; I
. $5M
Looms +~w§+mwwn m5 2% vaomM
82 oo: 83 ooo oow ooh ooo oom
I— _ -< —<1_ L r p PI.
.w M a
. a s
M .. u. .I
m M In W
x . +~ M NN \ a I m
an» a, 3M a o M Maofisofi 35:53 I m.
Ann as; an $23 3 m
i Ivmmwé

3+ charge state (m/z = 993.5) of

l

+

CID MS/MS Spectrum of the [M+3H

Figure SIS
D4A

104

 

NE
ooom 83 83 oo: cog
_ . a 4.14.4— . . .I .41._1.4.Ip #4
/ ....
52 an in
82 oo: 82 ooo oow
. . _ , _ .M . M
. / / .. / ..
..M an 2» Lap M /+§\ M M /. .FL... 5
. : +N 2 51.. M M
... a a ... MM :2: ....3
ago M :2»
+229 o.... +~2o~
w—
n
+~ +mn~n+~£n+~
a.»

 

 

 

 

 

82 oo: cog
. M
6,5 M. . .. $5
“x rIML r... M. M
o a M $39
EPNaMPNoNoM W;
232% N5
....5
ooo

/+~~ 5

+~o~£

 

-[:

 

5 ----...

M

: .
+~m

2% +~En+m~NDM

329%; e» on

 

 

 

souepunqv

 

Immoms

 

 

 

aouepuan

 

I mmowd

CID MS/MS Spectrum of the [M+4H+]4+ charge state (m/z = 745.5) of

Figure 816
D4A

105

ea

 

 

 

 

 

 

 

 

 

 

8% oo: ooom ooom 83 83 ooom oomm oovm oomm oomm ooMN
_._____:___._ ..-:r _p:_<_.______4__I_1_I_w___._4_ 1
M4 ... 1 M 1 MM M M A

M. can ... /. / MN \

5 fix @3025 “£29 ...... ... x 2.x
3. ...
Rs 9 a... NN
.... ..
002
of 3.» ONE
02
ooom oooM ooM: oo: 88 82 ooE 82 82 oo : oooM ooo
: .... ...... ..., / 2.»
so a 8 M, / /. m5 ... _S .A
a o ...... 2.» so 2» £5
2 M M...»
as» m 35 E
an—
a.»
a;

aouepunqv

 

Immo;

souepuan

mmfid

MALDI CID MS/MS Spectrum of the [M+H+]+ charge state (m/z

Figure 817

2962.3) of E3A

106

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

15»
NE W
ooom 83 com: oo: 83 so: 8.1 82 F
1 ,111 111111'1 1. _
m .. . .. ~ .. .. . .. W
on» an m ... mo; owi .. .. _. m .. ... m
o o8» We m5 1 2 3..» +N 3.... MN»... .... I a
” ~ . .. ” I
2% .. MEN zoofi +~w£ ... :5. +N m. .m
.u 2 o; .. 2
5n 5. 05 a an EMS ... fl w ml
... a H
N; I m
0; N5 Immmvd m
0011.. ..oo_: oo_o_ coo .oow com core ©me M.
111.11 .... 41 ‘11 :11 .14.4+1..5§_H11J1 #:1111 1 4111 11 m
.5. ... .. so; can 0 n c m» E u
u. / an; 3 o» a I V nw.
: S
o a
\:D a: in I W: W
+~m~£ % ”I M
I u
m m

Figure 818
of E3A

107

cocm
_

 

 

 

 

 

 

NE
83

82

co:

 

 

 

co?

    

 

 

 

 

 

w ...
v
. - m
Ea
Syn +~o~n m.
+~oM£ EL . .+~ \ I w
.13; . 9
a; I
Sa
3 +Nwmn m5 2% [mm—m6
+~o n
002 oo_: 002 com oww own owe owm
lad. \Afl hmj MM... fix. A.“ Man»
“.5... 2» .. .. m ....+N 3 w 5 .. w I
. TN . N.

ix. / o5 zap... a. ... a a .3 .S can I W.
. +~v§+~zn or» N n $.95 w.
2% c

+N “—3 +-~n . cm I w
+~ +m D 9

.. [mafia

CID MS/MS Spectrum of the [M+3H+]3+ charge state (m/z = 988.5) of

Figure 819
E3A

108

 

 

 

 

 

 

   

 

 

 

 

 

 

..m
N\E m,
4
cch ccc_ ccw_ ccS cog ccfl co: co: ._w
_ . _1—1__I.1. .11.11_11W‘1—1. .11 ... Z
/. .2. .... .. W
_ND .... c _.D .... En— xx .... . m I m
a; mi 0; . .. V ..m
..... FIE . . +~Z£ u. $89] m. We
2 m ...... . m u a
.. o m m .. . m. m
3 am a . ‘52; I w H
ca ca om :w I a +m.
So?» an Ego W
m n m
ooN_ cc: ooc_ coo com com coo com .m
_ . . ., m
_ a
w 1 m
.c M v m
:2 o 3 . 5 . 7 I W. W
+moomn S» c N.» _. I . inc; W. M
+~ «03:2» AVE “ I m w
1...... ........_.... .. .. m
S
[mm—cc; m
a» 9A
.n m

109

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

._
NE. W
comm co: cocm comm comm comm cocm comm cog comm 8% SS 9.
..__._._.r_._._.1...1. _11..__:_L1r__1__....1_11_____1._.1_1_1_.111__.. m
1 . ...- . . .. .... .... . ..
.. .. 1.. R .... .... a .. .. .... u
To c o 2.. 8 Comp 1 x 1 .1 x I mm
002 0N% OND NINA Q ch VN... .... mnxm mmxh V +
.... .. .. 1 m .I
a 5 NE u H
3.» oz» P +
1 m m.
.5. 029 m m
I .m
m
.... Immvw. m
C
cocm coo. co... co: coo. com. co... com. com. co: coo. coo W
TL. .1. .4... .1... .. ....1...1_1..11.I m
m ed: M ...... u 2 W
. .3 o .. in w; .. o 2.25 V W
. 2% So .. a. I
. ...» m» I m C
..5n 5 ...» w. m
T u
.. . m
Immmcw

....»

Figure 821

2962.4) of EIA

110

 

 

    

      

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

c).
NE. W
cocm cco_ ccc__ cc: cog co? co: com. F
_."rIIIJILIFIFIW. .11. . .IflLLIIFLIrILLLIIfLI. VLI_I.IL|I_|LI W
1|qu .__...t11. .. .. (
.... ...... . .. .. .. 1
a»; c.m£ W .. _L +mcc~2 ....Nmm%.._.. .... V M
.. .. .. Z I a. w...
QN~A fl. wrm +NwN~A m_D+NNND +N D. M: ..m
5. .3 .. .3 ....A 2 acfil e +c
03 o a» ... w H]
.. a

a .. H
.. .. ... I n
. m.
...» N c FEM: . m
com. 8.: coo. coo cam of oac own .m...
P p h _ r h _ r _ o _ _ _ _ _ _ 1. _ _ _ _ _ _ _ _ o p r 1.
11 .11 4111 m 411 4111 1 1:111 41411 #111 111.1411fi11 n
....M . ... o....» ..3 «03 3\ / 1‘13 no 3 m
.. ...... . . .... ... .. .. 1 m
...... .. a w. m
.... I .m. M
or» D. S
SD I m M

Ivmmvg

 

 

a»

Figure 822
of EIA

111

     

 

 

 

 

 

 

 

 

 

 

 

......
ccoN coo. 82 con. com. com. ccE com.
\ . .. .... .. \ _ .. .
5. m.“ em... 03 ...... ...... t. .. .. . I
. ... .... . . ... .... .. .. .. .1
.3 ......N o .. . F m.
.. 2.1.2.. . u
... +N 2.. . +N 3 0..
... 2.. +~ o... I w
+~ c ...... .. m
. N...» I
23
lvmowg
+~ccm£ +~w~£ m5 2%
com. 8.: one cmm
.414111 ..1 1 1 _1q11
...... . .. .. 1
\ . n .. . V
... . .. 1 M
£3» +N..3 ...» w.
52% _ l m
+~nnn 9
Ivmwcw

 

..

 

CID MS/MS Spectrum of the [M+3H+]3+ charge state (m/z = 988.5) of

Figure 823
ElA

112

 

cocm cco— ocwfi
1— p 4. .1 .fi—1 _ . h _
.//
_Nn
ccmo cc._ coco

 

    

+
N
W
N ‘u--..-
.D ‘
..D

 

 

 

 

 

 

 

 

 

 

 

 

 

NE
cc: ccE com. ccE com—
?» ....» ....» W.
n
u
p
Du
u
3
9
. .. ram:
cwo cow coo occ cwm
_1. Jif.
. .. ,,
. . .... / I
... a. m +~o5 V
.o on . Ex w. M l m.
.. ... _ .. ...... . 1... ..
+~ ... :o 2.; I u
T3: +~ a _ M...
+~c§+~w5 +~2D+~~5 0% on I
a» Immo~.~

Figure 824
ElA

CID MS/MS Spectrum of the [M+4H+]4+ charge state (m/z = 741.5) of

113

.. a. W. .5. egg. @fifiggfig.
.. g @ Egg. @fififigfiggg ..
o g F. g. 3%. @gégfio .. .
o a. £5.55. fig??? ......
.. g. gogéfigggfio .. ..

<

of the [M+H+]+,
of the A)E27A, B) EllA, C)

d sequence coverage summaries

+ 4+
nd [M+4H ] charge states

Figure 825: Overall product ion an

[M+2H+]2+, [M+3H+]3+, a

of C-pcptide

D4A, D) E3A, and E) ElA mutants

114

 

cocm ooofi
/
23
co: coo.
.__
.. .. .. _
.. \ «mi
. a» .
c— ....
.. .— of 231
+5.3— 1...;
+9.3.h

 

 

 

 

 

....

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

oo... oo: coo. com. ooE com. com.
_ . _ . . , .1 _
. .. \ . .. .
... ... ...... ... . _ I
D orA m F.L WV
".5... m.
.. .... ... 1
122...... ...... .. I m
+2.. 3 2W . +2.3: ..
.. ... my»
/ 3 +9. I
+2 . .
...... ...... ......1...... 1......
ooo ooo ooh ooo com ooo oom
r _ _ VF - _ _ F _ _ _ — _ _ _ _ _
_ . ... 11111 1‘1fl 414141 141:1 4.1.1111/4 J 1
m w .... o i 3 ...... m»
. n . r
o/ 3...... EL ...... o .... ... ......
a... . . I w
135.12.... ... ...... w.
Zoo.» m
£2.53 I
s. Ivmoo;

 

CID MS/MS Spectrum of the [M+Zn2++H+] 3+ charge state (m/z =

1008.9) of E27A

Figure 826

115

Non

 

 

 

 

 

 

 

 

 

 

 

 

 

z
w
cocm coo. co... co... coo. com. cc: com. com. cc: (
Tr......+..............r._.1..11 m
A / m.
m. r.
.. ...
. +
a
l n
Z
w
: Immm.w 11
n e
m
cc: coo. cco com com coo cco cc... com ..m
1.5.1.11. .1. .1.1_ . .. m
. . ./ _m
.. . i a
...n ...... I W
OS I W m
n
u S
P M
I m... m
m C
Immc:

 

Figure 827

116

758.0) of E27A

 

 

 

 

 

 

 

 

      

 

  

  

 

 

 

 

 

 

NE
oooN ooo. oom. oo: ooo. oom. oov. 82
m .... x/ . ...,r/ a... a . r.“ m u
E. .. ,, .... . .....t. .
u. .2. E e3 . .... 31...... e; a... c. e E...“ . .. v
..Nmn .Nn ...cfih ......Q m. % .... A c... .. can .... .. .. .. m.
o .. a. "2.2.. .. .N I u
a. .. . of». . w
. .. 3 u
N... c .12.. I 9
+2 3 .2?me ... n e
e a P. ..
+332 +N+wNQ+N+wN% lva # . ~
82 oo: ooo. ooo owm own oawo owm
u . . t .__ ....ietjjtfifiw ; _.a: .
,. e.
.53.; ., f . .m o“ / .... ...... m ... 3,... t
.. .. m .m o .3 o c... u.. ... 0V.
5 . or. or. . u m I
:o ,. .3 .3. . or. of. c... We. m.
E. ...... oi ...» I m
9
a [age

117

CID MS/MS Spectrum of the [M+Zn2] 2+ charge state (m/z = 1514.4)

Figure 828
of El 1A

as
88 82 8.: 8: c8. 82 8: 8m. 82

— .

.__—.b_...—_....__.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A I .1 .I VII/I/ / __ _
a a . . . "n .m .. m
. o a; c an”; F. or. W... "a . m “r V
a .. . .. .... .u o.
.r i e .I m
.. 4+" 1.5L \ , .... P
is. , .... 2. Ea. :. w
. .2» .. e ..o. T m
, 52.. £3. £5... u.“
a . . I
+8. o . m
3.3%....
+2.5. .2. $5.255. E N.. Ivmfi.
oo: ooo. ooo oow ooo ooo oom oov oom
-11 4_ .4; _4_I_14.11_th 1.1.F4_1.IJ_ _J~_1_ I. r _I_
.. . .n A J». 44 mg: WI /
u. r. .. r. e... i I
_ s
w - w
. .... m m e n
3.3+: .... e. a
$.35 +~+~3 +fio£ $23533..— cn +~ I w.
u
.. +22; is. .3. I a
Ivmmo.

 

...»

charge state (m/z

23+
1

+

CID MS/MS Spectrum of the [M+H +Zn

1008.9) of El 1A

118

Figure 829

 

NE
88 82 8M: 8: 8o. 82 8: 8m. 82 8:

_._L__..r__Lpp_....__._r___.__tr.__._.

 

I
souepuan

 

 

 

 

 

 

CID MS/MS Spectrum of the [M+2H++Zn2]4+ charge state (m/z =

 

 

 

:D ImmHN.~ .01;
cc : coo. coo cow coo coo coo cow com
.4 . . _I_ _ . .
V
I 0..
n
..u
0..
1| 9
w
e . m
I w m
s .m
Immovo e )
+2...» am.» o>+-5 W M
m .h

 

cocm

com.

#33.—

  

cco.

   
 

  

«TI
.0

.2:

.3:

cc:

----‘
c.”'

'v“.--.
”...-O-
.5.—
e
— -
In:

 

    

    

 

 

 

 

 

 

 

 

 

 

ca
cow. coo. coo. coo. co... com.

.5. . our... 2...... .. .. 2. 3w... m :2... .. ...... ......I V
m .. so. .. o... . . .2... q
2...... ...? ... n

...... . .... w... in. .4822. ...... .. I m.
amfiamzcms I m
+~+w-h2£mrh _I
2323.. 2o 55%|
+332 vmmmd
ccc. cco com coo cco com
itLI. .JFrarw f. . _ . .11.. 4.
. \e M. 44 _ / m e.
m . . .. MWe 2. ..
.. x . ..
o5 +31. /oo% +£— . PA a I W.
on I e
of m
9. Immoo.

2]2+ charge state (m/z = 1521.6) of

CID MS/MS Spectrum of the [M+Zn

Figure S31
D4A

120

as
..ch 8o. 8.: 8: 8o. ...... cc: ...... ......

 

 

 

 

 

 

 

 

  

 

 
       

 

 

 

 

 

 

F_I..I__...._—III.1PI_J__4I.17_ .. . .. .. .I .
\ m / /. a... u.
o... M E. w. o... m .... . m I
M.... o N. ammo... N... .... W.
.. . .... I n
. 4.: z. I u
. 3.. . re a. . P
..+~+ in on.» . B
.. +~o x m u
.. 3% x m .I w
.. o... +9.3: e. .m
+8. .. \ .... I
+1.3 o: ......
2 a m. . .
+2.... .a. . .. ......Imae
cc: ccc. coo cow cco cco coo ccv ccm
_ . _ . P Fr _ p p . _ _ — . _ _ _ L — .L’F _ . . m _
,, 1 ..-. I... .11 .1 II to .1 J .51 I
.. / . m . ...IIII... . . . ....o... ...
. . . . .. . .. . .. e . I
. .2. e. ... .te. ...... ”.52.. .0... .... . ...... v
MN.” . o m .. ...h m 0
+2 /c... +3.5. c . .... ... .... co .... I m.
a... 3.29:2...5 +2.... I m
a. / £233....» m
+13: I
oo Immmo...

121

CID MS/MS Spectrum of the [M+H-1+an]3+ charge state (m/z =

1014.0) of D4A

Figure S32

 

 

 

 

 

 

 

 

 

 

 

 

cocm 82 8M: 8: 82 82 83 82 SQ 8:
_.__._..__g__.r__.__L.____.._.__.___+r_
.4 “1.. I “I 41.1:
\x \x w ~ 3 xx . m

m5 in in M W.

_ I m

. P

as m

I o

9
:9 ISSN

8: 82 com com con 80 com cow 8m
____I_.I_I_ _I_ P _I_ _I_I_I_ _ __
:4 a: m
, “a... M ,

a; m. E. 03 I W

ELF: m» n m

+$n§+~m£ — u. I m.

I . w

+12.h +N 3 _ I 9

+~+~£§n£ $2;

is» as» $33 Immmv ~

CID MS/MS Spectrum of the [M+2H-l-+Zn2]4+ charge state (m/z =

Figure S33

760.7) of D4A

122

 

 

 

 

 

 

     

 

 

 

£8
cocm com. ocfi oo: 83 82 ooE oom—
.. w. .. . .. .
... m. a... .... arm .... “......y. ...m; L \m /, ...
...: . ... a, .a f..._....... n. «Ix asst
....z a» . ....... a... .. .12
35 +5. W a +2» +2.. 4:» N. wife; 2% an i am .... in fifnfifi my.
a» m; f 2.. , flu x. z» ... / a I u
a / an ,. , a: +25 55925 m.
+=£
:5: $5 +2.. 2% Iago
8.: oo: 82 coo com com coo com
. m
m S m \
rag—INS» :nM 23 of?» + z 3 FA V
+8» an a I m.
Shams: I W.
m
9
fl Immmvd

+ charge state (m/z = 1513.6) of

2

123

CID MS/MS Spectrum of the [M+Zn2]

Figure S34
E3A

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NE
88 82 8M: 8: 82 82 83 82 82
.__—.__r.__p._.
I II I I 11 _. ; . :
m _ w % ..,, x... m m r
”5“”; o; m5 5». L: W.
. v g . . u n
afieismfi a M £3— : _SI w.
a ..F . / B
E S M ; rnaizmfi I w
x 2% m _ 9
/ + n. +~+m£
+8.25 2 w; an I
a +1
E5. m; 5523 +2: Immmmx
8: 88 o8 28 con 8o 8m 8w 8m
_I_.._I_ . _.L _ _ _ _:_..I_;_._ :x. _ f: , L
d4 .I 4 .mn .... « .... ‘ M ‘ _. F ..H m —.
u M: ..a M" 3 . \ u x
“+2 a 2 ._ o 3n x in 3 / I
u E a ”on n 3
an EL M. on— o WV;
aaébéan i .5 m.
I B
U
0
9
5 F» +53.— o:—
9» IE8;

+
213 charge state (m/z =

+

CID MS/MS Spectrum of the [M+H +Zn

1008.9) of E3A

Figure S35

124

 

 

 

 

 

 

 

 

 

 

 

 

 

 

as
88 83 88 8: 82 82 8: 82 82 8:
______.____rr.__._b___________4..4_I_I_I.I fr? _
Md \. .
... .... .. - .V.
m n
I m.
02$ W
l o
9
.5 $me
8: 82 o8 _o_8_ 08 08 0% 8w 8m
1w:1_ _un :_ {_qu _I .__«Fjfi_sdrr_ ..1? _.h _ _
... m A .. T... 3. .... ./ ...,
m. w .... ...”. W .w / i l
2.21;».5 if. .. ,. a
u E; .. own V
a.» +~ . m» I a.
/+n._. _ w / hm. m
a; in: M. +$ m , I w.
+~._. ..a 3% 0
+2» 2.» 2 a a E m
N +2. +N D n— _ en— +NN_£ I
2 2
Buff a f p Imufi
+23.h

 

CID MS/MS Spectrum of the [M+2H++Zn2]4+ charge state (m/z =
125

Figure S36
757.0) of E3A

 

 

 

 

 

 

 

 

 

 

   

 

 

 

 

 

Ea
cocm 82 com: 82 82 co: so: 82
. . , n ..
...; 2:. E 1...... 25......» e; M m.
5» a; w; M? m P
9ND .... M— E . ..../ m ..... W
2% +23 D an %+~+w~%cmt / 2%.29.“ I %
+1. a $12: ..
m = I
+21%: M .. a
2...: $232321; Kg:
82 oo: 88 com com com coo oom
_II_I_ _Z: _ _LLEI. _1FI_I_I_
.. _. ,. 4 h
m up +5 E M £3
m 3 2 .u m. I
m/; s m» e v
3 .. I m
u
D.
I B
u
3
9
9A Immvws

CID MS/Ms Spectrum of the [M+Zn2]2+ charge state (m/Z = 1513.38)
126

Figure S37
of EIA

he
88 82 8.: 8: 82 82 8: 82 82

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

— .I- n —1— p p — — FPI- ——I.{P - f— I—|Ir.;I—II—IP‘I|—I —AI.”I‘I_ j
...... ... ...... .. . ...: ... m w
..— . s .. u . n
eeiiefsa ... +22: 35“.. / .. _;I w.
.. a ...... . u e
..m... a $23.... I w
E: a m E a
.n E. 3 I
2%; E a
is: aching; +2: ism—02
8: 82 08 28 CE 0% 8m 8w 8m
H _ _ _ L _ _ _ _ _ _ _ _ _ _ _ _ _ . _ r
In 4.114.: -..... I
. m .. ..
/. Moon Sn 5 5A 3 T
”+22.— ... V
M e» on I m.
... I w.
+25 +22: +22: m
a w I53...

Z112]3+ charge state (m/z

+

CID MS/MS Spectrum of the [M+H +

1009.9) of EIA

127

Figure S38

NE

 

 

 

 

 

       

 

  

 

 

 

 

 

 

cocm 82 8M: 8: 82 82 83 82 82 8:
_.____._____._._..___FL_T.__4__..____..___._
4‘1 4x4 4 «4 1.41% 11.411444 t
2 e x .
a 5 E w W
M n
- w
as m
I o
9
_ n [83.x
oo: 82 com cow OE o8 com oov com
_rtr.__4_1,1t_1_:1_11_11_1‘._7_1_w..
. L1,. T
.. M .... L M. x/ i
.L wh— ... 2.3% m ..M M» mn— . I
a , v . m
a» t 4,, a E g a. I m
cm . h .. .. M
2.. a .t M u
+ 2» “in; ..... it» ... m.
i .. .u ... a» 1 u
. M ... +2 MW
ix; iviamsfi on ads I
3.2
. i=5; a a a law:
+22%

CID MS/MS Spectrum of the [M+2H++Zn2]4+ charge state (m/z =
128

Figure S39
757.8) of EIA

LLLLL

\

LLL
L LLL LLLLL LLLLLL LLLLL

L
LLLLLLLL LLL LLL LLLLL

AELLLLLLLLLILL‘LL LL LLLLLLLLL
B L LLLLLLL LL

Figure S40: Overall Product ion and sequence coverage summaries obtained from
2+ 2+ + 2+ 3+ + 2+ 4+

the CID spectra of the [M+Zn ] , [M+H +Zn ] , and [M+2H +Zn ] charge

states of A) El 1A, B) D4A, C) E3A, D) ElA, mutants of C-peptide.\ indicates an

unbound b or y ion, \ indicates only the Zn-bound fragment was found, and \
indicates that both bound and unbound ions were identified.

._a
\O

2

REFERENCES

130

10.

11.

12.

REFERENCES

Ahmed, N., Advanced glycation endproducts - role in pathology of diabetic
complications. Diabetes Research and Clinical Practice 2005, 67 (1), 3-21.

Baynes, J., Role of oxidative stress in development of complications in diabetes.
Diabetes 1991, 40(4), 405—412.

Fullerto.Ww; Potter, R.; LOW, B. W., Proinsulin - crystallization and preliminary
X-Ray diffraction studies. Proceedings of the National Academy of Sciences of
the United States of America 1970, 66 (4), 1213-&.

Hua, Q. X.; Liu, M.; Hu, S. Q.; Jia, W. H.; Arvan, P.; Weiss, M. A., A conserved
histidine in insulin is required for the foldability of human proinsulin - Structure
and function of an A1a(BS) analog. Journal of Biological Chemistry 2006, 281
(34), 24889-24899.

Klip, A.; Paquet, M., Glucose transport and glucose transporters in muscle and
their metabolic regulation. Diabetes Care 1990, 13 (3), 228-243.

Baldwin, S. A., Mammalian passive glucose transporters - members of an
ubiquitous family of active and passive transport proteins. Biochimica Et
Biophysica Acta 1993, 1154 (1), 17-49.

Watson, R. T.; Kanzaki, M.; Pessin, J. E., Regulated membrane trafficking of the
insulin-responsive glucose transporter 4 in adipocytes. Endocrine Reviews 2004,
25 (2), 177-204.

Steinberg, H. 0.; Brechtel, G.; Johnson, A.; Fineberg, N.; Baron, A. D., Insulin-
Mediated Skeletal-Muscle Vasodilation Is Nitric-Oxide Dependent - a Novel

Action of Insulin to Increase Nitric-Oxide Release. Journal of Clinical
Investigation 1994, 94 (3), 1172-1179.

Kim, F.; Gallis, B.; Corson, M. A., TNF-{alpha} inhibits flow and insulin
signaling leading to NO production in aortic endothelial cells. Am J Physiol Cell
Physiol 2001, 280(5), C1057-1065.

Traub, 0.; Vanbibber, R., Role of nitric-oxide in insulin-dependent diabetes-
mellitus related vascular complications. Western Journal of Medicine 1995, 162
(5), 439-445.

Arthur S. C. K0, D. G. 8., Jan Markussen, Finn Sundby,, The amino acid
sequence of the C-peptide of human proinsulin. European Journal of
Biochemistry 1971, 20(2), 190-199.

Dodson, G.; Steiner, D., The role of assembly in insulin's biosynthesis. Current
Opinion in Structural Biology 1998, 8(2), 189-194.

131

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

Bantle, J. P.; Laine, D. C.; Hoogwerf, B. J .; Goetz, F. C., The potential usefulness
of postprandial urine C-peptide measurement in classifying diabetic-patients.
Diabetes Care 1984, 7 (2), 202-203.

Henriksson, M.; Nordling, E.; Melles, E.; Shafqat, J .; Stahlberg, M.; Ekberg, K.;
Persson, B.; Bergman, T.; Wahren, J .; Johansson, J.; Jomvall, H., Separate
functional features of proinsulin C-peptide. Cellular and Molecular Life Sciences
2005, 62 (15), 1772-1778.

Henriksson, M.; Shafqat, J.; Liepinsh, E.; Tally, M.; Wahren, J .; Jomvall, H.;
Johansson, J., Unordered structure of proinsulin C-peptide in aqueous solution

and in the presence of lipid vesicles. Cellular and Molecular Life Sciences 2000,
57 (2), 337-342.

Kunt, T.; Schneider, S.; Pfutzner, A.; Goitum, K.; Engelbach, M.; Schauf, B.;
Beyer, J.; Forst, T., The effect of human proinsulin C-peptide on erythrocyte
deformability in patients with Type I diabetes mellitus. Diabetologia 1999, 42 (4),
465-471.

Femqvistforbes, 13.; Johansson, B. L.; Wahren, J., C-peptide improves nerve
function in type-l diabetic-patients with autonomic neuropathy. Diabetologia
1995, 38, A238-A238.

Ido, Y.; Vindigni, A.; Chang, K.; Stramm, L.; Chance, R.; Heath, W. F.;
DiMarchi, R. D.; DiCera, E.; Williamson, J. R., Prevention of vascular and neural
dysfimction in diabetic rats by C-peptide. Science 1997, 277 (5325), 563-566.

Johansson, B. L.; Borg, K.; Femqvist-Forbes, E.; Kemell, A.; Odergren, T.;
Wahren, J., Beneficial effects of C-peptide on incipient nephropathy and
neuropathy in patients with Type 1 diabetes mellitus. Diabetic Medicine 2000, 17
(3), 181-189.

Johansson, B. L.; Borg, K.; FemqvistForbes, E.; Odergren, T.; Remahl, S.;
Wahren, J., C-peptide improves autonomic nerve function in IDDM patients.
Diabetologia 1996, 39(6), 687-695.

Samnegard, B.; Jacobson, S. H.; Jaremko, G.; Johansson, B. L.; Sjoquist, M.,
Effects of C-peptide on glomerular and renal size and renal function in diabetic
rats. Kidney International 2001, 60(4), 1258-1265.

Forst, T.; Kunt, T., Effects of C-peptide on microvascular blood flow and blood
hemorheology. Experimental Diabesity Research 2004, 5 (1), 51-64.

Johansson, B. L.; Wahren, J .; Pemow, J ., C-peptide increases forearm blood flow
in patients with type 1 diabetes via a nitric oxide-dependent mechanism.
American Journal of Physiology-Endocrinology and Metabolism 2003, 285 (4),
E864-E870.

132

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

Kitamura, T.; Kimura, K.; Jung, B. D.; Makondo, K.; Okamoto, S.; Canas, X.;
Sakane, N.; Yoshida, T.; Saito, M., Proinsulin C-peptide rapidly stimulates
mitogen-activated protein kinases in Swiss 3T3 fibroblasts: requirement of protein

kinase C, phosphoinositide 3-kinase and pertussis toxin-sensitive G-protein.
Biochemical Journal 2001, 355, 123-129.

Kitamura, T.; Kimura, K.; Jung, B. D.; Makondo, K.; Sakane, N.; Yoshida, T.;
Saito, M., Proinsulin C-peptide activates cAMP response element-binding
proteins through the p38 mitogen-activated protein kinase pathway in mouse lung
capillary endothelial cells. Biochemical Journal 2002, 366, 737-744.

Rigler, R.; Pramanik, A.; Jonasson, P.; Kratz, G.; Jansson, 0. T.; Nygren, P. A.;
Stahl, S.; Ekberg, K.; Johansson, B. L.; Uhlen, S.; Uhlen, M.; Jomvall, H.;
Wahren, J., Specific binding of proinsulin C-peptide to human cell membranes.
Proceedings of the National Academy of Sciences of the United States of America
1999, 96 (23), 13318-13323.

Lindahl, E.; Nyman, U.; Melles, 13.; Sigmundsson, K.; Stahlberg, M.; Wahren, J.;
0brink, B.; Shafqat, J .; Joseph, B.; Jomvall, H., Cellular internalization of
proinsulin C-peptide. Cellular and Molecular Life Sciences 2007, 64 (4), 479-486.

Ekberg, K.; Rigler, R.; Pramanik, A.; Kratz, G.; Johansson, B. L.; Uhlen, M.;
Jomvall, H.; Wahren, J ., Specific binding of proinsulin C-peptide to human cell
membranes. Diabetologia 1999, 42, A200-A200.

Forst, T.; De la Tour, D. D.; Kunt, T.; P‘futzner, A.; Goitom, K.; Pohlmann, T.;
Schneider, S.; Johansson, B. L.; Wahren, J .; Lobig, M.; Engelbach, M.; Beyer, J.;
Vague, P., Effects of proinsulin C-peptide on nitric oxide, microvascular blood
flow and erythrocyte Na+,K+-ATPase activity in diabetes mellitus type I. Clinical
Science 2000, 98 (3), 283-290.

Ohtomo, Y.; Bergman, T.; Johansson, B. L.; J omvall, H.; Wahren, J ., Differential
effects of proinsulin C-peptide fragments on Na+, K+-ATPase activity of renal
tubule segments. Diabetologia 1998, 41 (3), 287-291.

Pramanik, A.; Ekberg, K.; Zhong, Z.; Shafqat, J .; Henriksson, M.; Jansson, 0.;
Tibell, A.; Tally, M.; Wahren, J.; Jomvall, H.; Rigler, R.; Johnansson, J., C-
peptide binding to human cell membranes: Importance of Glu27. Biochemical and
Biophysical Research Communications 2001, 284 (1), 94-98.

Koc, B.; Erten, V.; Yilmaz, M. I.; Sonmez, A.; Kocar, I. H., The relationship
between red blood cell Na/K-ATPase activities and diabetic complications in
patients with type 2 diabetes mellitus. Endocrine 2003, 21 (3), 273-278.

Mares-Guia, T. R.; Maigret, B.; Martins, N. F.; Maia, A. L. T.; Vilela, L.; Ramos,
C. H. 1.; Neto, L. J.; Juliano, M. A.; dos Mares-Guia, M. L.; Santoro, M. M.,
Molecular dynamics and circular dichroism studies of human and rat C-peptides.
Journal of Molecular Graphics and Modelling 2006, 25 (4), 532-542.

133

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

Valko, M.; Rhodes, C. J.; Moncol, J.; Izakovic, M.; Mazur, M., Free radicals,
metals and antioxidants in oxidative stress-induced cancer. C/zemico-Biological
Interactions 2006, 160(1), 1-40.

Dymock, I. W.; Cassar, J .; Pyke, D. A.; Oakley, W. G.; Williams, R.,
Observations on the pathogenesis, complications and treatment of diabetes in 115
cases of haemochromatosis. The American Journal of Medicine 1972, 52 (2), 203-
210.

Femandez-Real, J. M.; Lopez-Bermejo, A.; Ricart, W., Iron stores, blood

donation, and insulin sensitivity and secretion. Clinical Chemistry 2005, 51 (7),
1201-1205.

Schafer, A. 1.; Cheron, R. G.; Dluhy, R.; Cooper, B.; Gleason, R. E.; Soeldner, J.
S.; Bunn, H. F., Clinical Consequences of Acquired Transfusional Iron Overload
in Adults. New England Journal of Medicine 1981, 304 (6), 319-324.

Mertz, W.; Roginski, E. E.; Schwarz, K., Effect of trivalent chromium complexes
on glucose uptake by epididymal fat tissue of rats. Journal of Biological
Chemistry 1961, 236(2), 318-&.

Schwarz, K.; Mertz, W., A Glucose tolerance factor and its differentiation from
factor-3. Archives of Biochemistry and Biophysics 1957, 72 (2), 515-518.

Mertz, W.; Schwarz, K., Relation of glucose tolerance factor to impaired
intravenous glucose tolerance of rats on stock diets. American Journal of
Physiology 1959, 196 (3), 614-618.

Yamamoto, A.; Wada, 0.; Ono, T., A Low-molecular-weight, chromium-binding
substance in mammals. Toxicology and Applied Pharmacology 1981, 59 (3), 515-
523.

Yamamoto, A.; Wada, 0.; Ono, T., Isolation of a biologically-active low-
molecular-mass chromium compound from rabbit liver. European Journal of
Biochemistry 1987, 165 (3), 627-631.

Davis, C. M.; Vincent, J. B., Chromium in carbohydrate and lipid metabolism.
Journal of Biological Inorganic Chemistry 1997, 2 (6), 675-679.

Emdin, S. 0.; Dodson, G. G.; Cutfield, J. M.; Cutfield, S. M., Role of zinc in
insulin-biosynthesis - some possible zinc-insulin interactions in the pancreatic B-
cell. Diabetologia 1980, 19(3), 174-182.

Chimienti, F.; Favier, A.; Seve, M., ZnT-8, a pancreatic beta-cell-specific zinc
transporter. Biometals 2005, 18 (4), 313-317.

Cunningham, J. J.; Fu, A. Z.; Mearkle, P. L.; Brown, R. G., Hyperzincuria in
individuals with insulin-dependent diabetes-mellitus - concurrent zinc status and

134

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

the effect of high-dose zinc supplementation. Metabolism-Clinical and
Experimental 1994, 43 (12), 1558-1562.

Meyer, J. A.; Froelich, J. M.; Reid, G. E.; Karunarathne, W. K. A.; Spence, D. M.,
Metal-activated C-peptide facilitates glucose clearance and the release of a nitric
oxide stimulus via the GLUTl transporter. Diabetologia 2008, 51 (1), 175-182.

Jennifer A. Meyer, W. S., Anders A. F. Sima, Zachary Keltner, Gavin E. Reid,
David Daleke and Dana M. Spence, hyperglycemia-induced phosphatidylserine
exemalization is associated with zinc-activated C-peptide resistance to the type 2
diabetic erythrocyte (manuscript in preparation).

Teesch, L. M.; Adams, J ., Fragrnentations of gas-phase complexes between alkali-
mtal ions and peptides - metal-ion binding to carbonyl oxygens and other neutral
functional-groups. Journal of the American Chemical Society 1991, 113 (3), 812-
820.

Teesch, L. M.; Orlando, R. 0; Adams, J ., Location of the alkali-metal ion in gas-
phase peptide complexes. Journal of the American Chemical Society 1991, 113
(10), 3668-3675.

Hu, P. F.; Gross, M. L., Strong-interactions of anionic peptides and alkaline-earth
metal-ions - metal-ion-bound peptides in the gas-phase. Journal of the American
Chemical Society 1992, 114 (23), 9153-9160.

Hu, P. F.; Gross, M. L., Strong-interactions of anionic peptides and alkaline-earth
metal-ions - bis(peptide) complexes in the gas-phase. Journal of the American
Chemical Society 1992, 114 (23), 9161-9169.

Nemirovskiy, O. V.; Gross, M. L., Determination of calcium binding sites in gas-
phase small peptides by tandem mass spectrometry. Journal of the American
Society for Mass Spectrometry 1998, 9 (10), 1020-1028.

Nemirovskiy, O. V.; Gross, M. L., Gas phase studies of the interactions of Fe2+
with cysteine-containing peptides. Journal of the American Society for Mass
Spectrometry 1998, 9 (12), 1285-1292.

Grese, R. P.; Cemy, R. L.; Gross, M. L., Metal-ion peptide interactions in the gas-
phase - a tandem mass-spectrometry study of alkali-metal cationized peptides.
Journal of the American Chemical Society 1989, 111 (8), 2835-2842.

Yu, X. L.; Wojciechowski, M.; Fenselau, 0, assessment of metals in
reconstituted metallothioneins by electrospray mass-spectrometry. Analytical
Chemistry 1993, 65 (10), 1355-1359.

Loo, J. A.; Hu, P. F.; Smith, R. D., Interaction of angiotensin peptides and zinc

metal-ions probed by electrospray-ionization mass-spectrometry. Journal of the
American Society for Mass Spectrometry 1994, 5 (11), 959-965.

135

58.

59.

60.

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

Zirah, S.; Rebuffat, S.; Kozin, S. A.; Debey, P.; Foumier, F.; Lesage, D.; Tabet, J.
C., Zinc binding properties of the amyloid fragment A beta(1-16) studied by
electrospray-ionization mass spectrometry. International Journal of Mass

Spectrometry 2003, 228 (2-3), 999-1016.

Pu, D.; Vincent, J. B.; Cassady, C. J ., The effects of chromium(lll) coordination

on the dissociation of acidic peptides. Journal of Mass Spectrometry 2008, 43 (6),
773-781.

Roepstorff, P.; Fohlman, J ., Proposal for a common nomenclature for sequence
ions in mass-spectra of peptides. Biomedical Mass Spectrometry 1984, II (11),
601-601.

Biemann, K.; Papayannopoulos, I. A., Amino-acid sequencing of proteins.
Accounts of Chemical Research 1994, 27 (1 1), 370-378.

Jemigan, R.; Raghunathan, G.; Bahar, 1., Characterization of interactions and

metal-ion binding-sites in proteins. Current Opinion in Structural Biology 1994, 4
(2), 256-263.

Kaiser, E.; Colescot.Rl; Bossinge.Cd; Cook, P. 1., Color Test for detection of free

terminal amino groups in solid-phase synthesis of peptides. Analytical
Biochemistry 1970, 34 (2), 595-&.

Fenn, J .; Mann, M.; Meng, C.; Wong, S.; Whitehouse, C., Electrospray ionization
for mass spectrometry of large biomolecules. Science 1989, 246 (4926), 64-71.

Kebarle, P., A brief overview of the present status of the mechanisms involved in
electrospray mass spectrometry. J Mass Spectrom 2000, 35 (7), 804-17.

Gomez, A.; Tang, K., Charge and fission of droplets in electrostatic sprays.
Physics of Fluids 1994, 6 (1), 404-414.

Iribame, J. V.; Thomson, B. A., Evaporation of small ions from charged droplets.
Journal of Chemical Physics 1976, 64 (6), 2287-2294.

Thomson, B. A.; Iribame, J. V., Field-induced ion evaporation from liquid
surfaces at atmospheric-pressure. Journal of Chemical Physics 1979, 71 (11),
4451-4463.

Karas, M.; Baehmann, D.; Bahr, U.; Hillenkamp, F., Matrix-assisted ultraviolet-
laser desorption of nonvolatile compounds. International Journal of Mass
Spectrometry and Ion Processes 1987, 78, 53-68.

Karas, M.; Gluckmann, M.; Schafer, J., Ionization in matrix-assisted laser
desorption/ionization: singly charged molecular ions are the lucky survivors.
Journal of Mass Spectrometry 2000, 35 (1 ), 1-12.

136

71.

72.

73.

74.

75.

76.

77.

78.

79.

80.

81.

82.

83.

Zenobi, R.; Knochenmuss, R., Ion formation in MALDI mass spcctrmnetry. Mass
Spectrometry Reviews 1998, 1 7 (5), 337-366.

March, R. E., An introduction to quadrupole ion trap mass spectrometry. Journal
of Mass Spectrometry 1997, 32 (4), 351-369.

de Hoffman, E. D. S., V., Mass Spectrometry Principles and Applications. John
Wiley and Sons: New York, 1992.

Hager, J. W., A new linear ion trap mass spectrometer. Rapid Communications in
Mass Spectrometry 2002, 16 (6), 512-526.

Schwartz, J. C.; Senko, M. W.; Syka, J. E. P., A two-dimensional quadrupole ion
trap mass spectrometer. Journal of the American Society for Mass Spectrometry
2002, 13 (6), 659-669.

Reid, G. E.; McLuckey, S. A., 'Top down' protein characterization via tandem
mass spectrometry. Journal of Mass Spectrometry 2002, 37(7), 663-675.

Scherperel, G.; Reid, G. E., Emerging methods in proteomies: top-down protein
characterization by multistage tandem mass spectrometry. Analyst 2007, I32 (6),
500-506.

Reid, G. E.; Wu, J.; Chrisman, P. A.; Wells, J. M.; McLuckey, S. A., Charge-
state-dependent sequence analysis of protonated ubiquitin ions via ion trap
tandem mass spectrometry. Analytical Chemistry 2001, 73 (14) 3274-3281.

Scherperel, G.; Yan, H. G.; Wang, Y.; Reid, G. 13., 'Top-down' characterization of
site-directed mutagenesis products of Staphylococcus aureus dihydroneopterin

aldolase by multistage tandem mass spectrometry in a linear quadrupole ion trap.
Analyst 2006, I31 (2), 291-302.

Dongre, A. R.; Jones, J. L.; Somogyi, A.; Wysocki, V. H., Inllucnee of peptide
composition, gas-phase basicity, and chemical modification on fragmentation
efficiency: Evidence for the mobile proton model. Journal of the American
Chemical Society 1996, 118 (35), 8365-8374.

Kapp, E. A.; Schutz, F.; Reid, G. E.; Eddes, J. S.; Moritz, R. L.; O'l-lair, R. A. J.;
Speed, T. P.; Simpson, R. J ., Mining a tandem mass spectrometry database to
determine the trends and global factors influencing peptide fragmentation.
Analytical Chemistry 2003, 75 (22), 6251-6264.

Paizs, B.; Suhai, S., Fragmentation pathways of protonated peptides. Mass
Spectrometry Reviews 2005, 24 (4), 508-548.

Dudev, T.; Lim, C., Tetrahedral vs octahedral zinc complexes with ligands of

biological interest: A DFT/CDM study. Journal of the American Chemical
Society 2000, 122 (45), 11146-11153.

137

”lltlt’lltiiillittflilittlililttl“