Page 1 of 29
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
1
Simultaneous assessment of protein heterogeneity and affinity by capillary
2
electrophoresis-mass spectrometry
3
E. Domínguez-Vega1*, R. Haselberg1, G.W. Somsen1, G.J. de Jong2
4 5
6 7
8 9
10
1
Division of BioAnalytical Chemistry, VU University, de Boelelaan 1083, 1081 HV
Amsterdam, The Netherlands 2
Biomolecular Analysis, Utrecht University, P.O. Box 80082, 3508 TB Utrecht, The
Netherlands *Corresponding author:
[email protected] 11
12
13
14
15
16
17
18
19
20 1
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 29
21
Abstract
22
Most conventional analytical tools for the assessment of protein-protein interactions yield
23
information on the bulk sample. Employing the efficient separation of intact proteins, affinity
24
capillary electrophoresis (ACE) can measure the interaction of components of heterogeneous
25
proteins with a target protein. In this work, the hyphenation of ACE with mass spectrometry
26
(MS) is presented as a novel highly-selective tool for the assessment of protein-protein
27
interactions. The binding of the protease inhibitor aprotinin to trypsinogen was used as
28
protein-protein affinity model. A trypsinogen sample comprising several modifications was
29
analyzed using a background electrolyte of 25 mM ammonium acetate (pH 8.0) containing
30
increasing concentrations of aprotinin (0-300 µM). A capillary coating of Polybrene-dextran
31
sulfate-Polybrene (PB-DS-PB) was employed to prevent adsorption of the proteins to the
32
capillary wall. The trypsinogen variants were separated and could be assigned based on
33
detected molecular masses and relative migration. In presence of aprotinin, both free and
34
aprotinin-bound trypsinogen were detected revealing a 1:1 binding stoichiometry. For most
35
trypsinogen variants, shifts in electrophoretic mobility were observed upon raising the
36
aprotinin concentration, allowing determination of their dissociation constants (Kds). The
37
interacting trypsinogen variants showed similar affinity towards aprotinin (Kds of 3-9 µM),
38
which were not significantly different from the values obtained with ACE-UV and were in
39
agreement with an earlier reported value. The use of the ratio of obtained MS signal
40
intensities of free and protein-protein complex for the determination of Kds was also
41
explored. Derived Kd values (20-104 µM) for the binding variants were similar to those
42
obtained with direct-infusion MS, but higher and less precise as compared with values based
43
on mobility shifts. The suitability of the ACE-MS methodology for the affinity profiling of
44
heterogeneous protein samples was evaluated and components with high, medium or low -
45
affinity towards aprotinin could be successfully discriminated. 2
ACS Paragon Plus Environment
Page 3 of 29
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
46
1. Introduction
47
The bioactivity of proteins is based on their capacity to bind to their targets, which frequently
48
are protein receptors or enzymes. The availability of efficient analytical tools for the study of
49
protein-protein interactions is essential in protein science
50
methods can be used for the determination of protein-protein affinities. Among these are
51
surface plasmon resonance, calorimetry, fluorescence, circular dichroism and nuclear
52
magnetic resonance techniques 3,4. Most of these methodologies, however, are not suitable for
53
the simultaneous determination of the equilibrium dissociation constants (Kds) of multiple
54
sample components, such as protein isoforms and/or impurities. Nevertheless, proteins
55
seldomly are pure, unimolecular compounds and may comprise different post-translational
56
modifications (PTMs) or degradation products. For instance, during production, isolation
57
and/or storage, proteins may undergo modifications that may dramatically influence
58
bioactivity, efficacy and toxicity, thereby compromising their biological properties 5. In other
59
cases, PTMs (e.g. glycosylation) may be essential to maintain proper protein activity.
60
Consequently, the possibility to assess the affinity of all protein variants present in a sample
61
is essential.
62
Affinity capillary electrophoresis (ACE) represents a powerful technique for the assessment
63
of interactions of protein sample components towards a target protein
64
proteins is well established, providing efficient separations under aqueous conditions and in
65
absence of a stationary phase
66
proper capillary coatings. For instance, positively- and negatively-charged capillary coatings
67
have shown very successful in achieving high resolution separation of diverse proteins with
68
various pIs
69
ACE offers the possibility to simultaneously study the interaction of multiple proteins under
70
homogeneous and near-physiological conditions 11. Affinity measurement by CE is based on
1-3
. A number of in-solution
6-11
. CE of intact
12-14
. Protein adsorption can be prevented effectively by use of
15,16
. Employing the efficient separation of intact proteins as provided by CE,
3
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 29
71
the fact that the electrophoretic mobility of the probed protein differs from the electrophoretic
72
mobility of the protein-ligand or protein-receptor complex. When association/dissociation
73
kinetics are slower or similar to the CE separation time, pre-equilibrium or kinetic ACE can
74
be applied, respectively
75
significant time and, subsequently, ligand, receptor and complex are separated by CE
76
assuming that the complex does not dissociate in the course of the analysis. In kinetic ACE,
77
the observed change in electrophoretic mobility and shape (but not area) of the ligand peak
78
when analyzed in presence of protein receptor, may be used for the determination of
79
dissociation constants
80
applying specific experimental settings and data analysis approaches for the determination of
81
dissociation constants under non-equilibrium
82
equilibrium kinetics are fast, a change in effective electrophoretic mobility will be observed
83
for the injected protein(s) when the receptor or ligand is added to the BGE. From the
84
observed mobility shifts at increasing concentrations of receptor or ligand in the BGE, Kds
85
can be calculated. Mobility-shift ACE has the main advantage that Kds of multiple sample
86
components can be determined simultaneously. Using ACE with UV absorbance detection,
87
reliable Kd determination requires components to be fully separated. However, no
88
information on the identity of the sample components is obtained and the stoichiometry of the
89
formed complex cannot be determined.
90
Coupling CE with mass spectrometry (MS) yields a powerful tool for protein analysis, as it
91
combines efficient separation of intact proteins with the mass-selective detection and analyte
92
characterization possibilities provided by MS 13,14,23,24. In addition, resolution of co-migrating
93
compounds can be achieved by MS by monitoring component-specific m/z values. Recently,
94
our group has shown the potential of the parallel use of ACE-UV and CE-MS for the study of
95
the affinity of a nanobody product with the epidermal growth factor receptor 6. The nanobody
10
. In pre-equilibrium ACE, ligand and protein are incubated for a
17,18
. Krylov et al. have developed a series of kinetic ACE methods
19
and equilibrium
20-22
conditions. When
4
ACS Paragon Plus Environment
Page 5 of 29
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
96
comprised of several variants and ACE-UV was used to establish their individual Kd values.
97
CE-MS was employed for characterization of these variants. Instead of using two parallel
98
techniques, the direct coupling of ACE with MS would provide a potentially strong platform
99
for the simultaneous assessment of protein heterogeneity and affinity. Moreover, MS can
100
reveal the stoichiometry of the protein-complex and protein signal intensities potentially can
101
also be used to gain information about protein affinity. In fact, stand alone MS has been used
102
for the determination of the affinity of proteins 10,25-28. Hence, in principle MS data could be
103
used to gain additional affinity information, but some considerations as different ionization
104
efficiencies of unbound and bound protein and/or different stabilities of the complexes in the
105
gas phase should be taken into account 26-28.
106
So far, few studies have been reported employing the concept of ACE-MS for affinity
107
determination by monitoring mobility shifts. The majority of these works focused on the
108
screening of the affinity of di- tri- and tetrapeptides towards antibiotics 29-33. Other mobility-
109
shift based ACE-MS studies deal with the interaction of oligosaccharides with the stromal
110
cell-derived factor-1 receptor 34, and the affinity of small drugs for β-cyclodextrin 25. To our
111
knowledge, protein-protein interactions have never been studied by ACE-MS before. In the
112
present work, the on-line combination of ACE and MS was evaluated as a new selective tool
113
for the simultaneous assessment of protein heterogeneity and protein-protein affinity. The
114
protease trypsinogen (24 kDa) and its inhibitor aprotinin (6.5 kDa) were selected as a model
115
system. These proteins can be efficiently analyzed by CE-MS and have relatively fast
116
equilibration kinetics (koff > 4·10-2 s-1) with a Kd in the low-µM range 35, in principle allowing
117
affinity analysis monitoring changes in electrophoretic mobility. We also studied the
118
usefulness of the developed ACE-MS methodology for the screening of aprotinin-affinity
119
components in heterogeneous protein samples.
120 5
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 29
121
2. Materials and methods
122
2.1.
123
All reagents employed were of analytical grade. Ammonium hydroxide (25% solution), was
124
obtained from Fluka (Steinheim, Germany). Isopropanol (IPA) and acetic acid was provided
125
by Merck (Darmstadt, Germany). Formic acid was supplied by Riedel-De Haen (Seelze,
126
Germany). Trypsinogen, trypsin and α-chymotrypsin from bovine pancreas, aprotinin
127
(bovine lung trypsin inhibitor), cortisone, Polybrene (hexadimethrine bromide, PB) and
128
dextran sulfate (DS) were obtained from Sigma-Aldrich (Steinheim, Germany). Deionized
129
water was obtained from a Milli-Q purification system (Millipore, Bedford, USA).
130
2.2.
131
CE analyses were performed using a Beckman PA 800 plus instrument (Beckman Coulter,
132
Brea, USA). Fused silica capillaries were from Polymicro Technologies (Phoenix, AZ, USA)
133
having an internal diameter of 50 µm and a total length of 60 cm (UV detection) and 80 cm
134
(MS detection). Hydrodynamic injections were performed at 1 psi for 8 s (60-cm capillaries)
135
or 1 psi for 10 s (80-cm capillaries). The separation voltage was 30 kV and the capillary
136
temperature was 20 °C. UV detection was performed at 214 nm after 50 cm on the capillary.
137
The background electrolyte (BGE) was 25 mM ammonium acetate (pH 8.0) (prepared by
138
diluting the appropriate amount of ammonium hydroxide in water and setting the pH with
139
acetic acid) containing 0-300 µM aprotinin. New fused-silica capillaries were rinsed with 1
140
M NaOH for 30 min at 20 psi, and with water for 15 min at 20 psi. After this treatment,
141
capillaries were coated with a PB-DS-PB coating. For this, solutions of 10% (w/v) PB and
142
0.5% (w/v) DS in deionized water were prepared. The solutions were filtered over a 0.45 µm
143
filter type HA (Millipore, Molsheim, France) prior to use. Capillaries were coated by
144
subsequently rinsing 30 min with 10% (w/v) PB solution at 5 psi, 10 min with deionized
Chemicals
Capillary electrophoresis
6
ACS Paragon Plus Environment
Page 7 of 29
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
145
water at 10 psi, 30 min with 0.5% (w/v) DS solution at 5 psi, 10 min with deionized water at
146
10 psi, 30 min with 10% (w/v) PB solution at 5 psi, and 10 min with deionized water at 10
147
psi. After the final coating step, the capillary was rinsed for 10 min with 25 mM ammonium
148
acetate (pH 8.0) at 20 psi. Before each run, coated capillaries were flushed with water for 2
149
min, 25 mM ammonium acetate (pH 8.0) for 2 min and the BGE for 2 min at 20 psi.
150
Overnight, capillaries were filled with water and tips were immersed in vials with water.
151
2.3.
152
MS experiments were carried out on a micrOTOF-QII mass spectrometer (Bruker Daltonics,
153
Bremen, Germany). Source and transfer parameters were optimized by direct infusion of the
154
proteins via the CE capillary. CE-MS coupling was performed using a sheath-liquid
155
electrospray interface from Agilent Technologies (Waldbronn, Germany). The sheath liquid
156
was a mixture of isopropanol-water-25 mM ammonium acetate (pH 8.0) (25:50:25, v/v/v)
157
delivered at a flow rate of 2 µL·min-1 using a syringe pump from Cole-Parmer (Vernon Hill,
158
IL USA). The mass spectrometer was operated in positive-ion mode with an electrospray
159
voltage of 4.5 kV. The nebulizer and drying gas conditions were 0.2 bar and 3 L·min-1
160
nitrogen at 80 ºC, respectively. Transfer parameters were adjusted for optimal detection of
161
non-covalent protein complexes. Quadrupole ion and collision cell energy were 5 and 7 eV,
162
respectively. Transfer and pre plus storage times were set at 200 and 20 µs, respectively. The
163
monitored mass range was 250 to 3000 m/z. Extracted ion electropherograms (EIEs) were
164
obtained with an extraction window of ± 0.5 m/z and using the smooth option of the software
165
(Gaussian at 1 point).
166
2.4.
Mass spectrometry
Affinity CE
7
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 8 of 29
167
BGEs containing increasing concentrations of aprotinin (0–300 µM) were prepared and used
168
for the analysis of trypsinogen (40 and 160 µM for UV and MS detection, respectively). The
169
effective electrophoretic mobility (µeff) of trypsinogen and its components were
170
experimentally determined using Eq. (1):
171
172
(1)
ߤ݂݂݁ =
݀ܮ ݐܮ1 1 ቆ − ቇ [ܿ݉2 ܸ −1 ݏ−1 ] ܸ ݂݁ݐ ܲݐ
173 174
where Lt is the total capillary length, Ld is the capillary length to detector, V is the applied
175
voltage, and tp and teof are the migration times of the protein and EOF marker, respectively.
176
For EOF measurements, cortisone (50 mM) was injected (0.5 psi for 5 s) immediately after
177
trypsinogen injection. For each aprotinin concentration, the difference of the measured µeff of
178
trypsinogen and derived products with the µeff obtained when no ligand was added to the
179
BGE, was determined and plotted versus the ligand concentration. These plots were fitted
180
using Prism 5.0 (GraphPad Software Inc., La Jolla, CA, USA) applying nonlinear regression
181
assuming a 1:1 binding stoichiometry according to Eq. (2):
182
(2)
∆ߤ݂݂݁ =
ݎܥ ݔܽ݉ܤ ݀ܭ+ ݎܥ
183
where ∆µeff is the effective mobility difference of trypsinogen obtained at a specific aprotinin
184
concentration compared to when no aprotinin was added to the BGE, Bmax is the maximum
185
mobility shift, Kd is the dissociation constant, and cr is the ligand concentration added to the
186
BGE. Kd values and their standard deviations were derived by the software using the
187
Levenberg-Marquardt algorithm.
188
For the evaluation of potential viscosity changes upon addition of aprotinin to the BGE, the
189
CE instrument was employed as a viscosimeter. Applying a constant pressure of 1.5 psi, the
8
ACS Paragon Plus Environment
Page 9 of 29
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
190
time needed for a trypsinogen solution (40 µM) to reach the detector was measured
191
repeatedly (n=6) for a BGE containing no or 300 µM of aprotinin.
192 193 194
3. Results and discussion
195
3.1. Set up of ACE-MS system
196
ACE can be performed using near-physiological BGEs containing e.g. phosphate or Tris
197
buffer. However, coupling with ESI-MS requires volatile BGEs to avoid analyte ionization
198
suppression and source contamination. Recent studies show that replacing sodium phosphate
199
or Tris acetate for e.g. ammonium acetate or bicarbonate at medium pH does not significantly
200
alter protein-ligand interactions
201
8.0
202
selected. As trypsinogen (pI 9.3) and aprotinin (pI 10.5) are overall positively charged at pH
203
8.0, a triple layer coating (PB-DS-PB) was employed in order to avoid adsorption of the
204
proteins to the capillary wall
205
by CE-UV. Trypsinogen showed a cluster of peaks migrating in the 4.5-5.5 min range. A
206
single peak was obtained for aprotinin with a significantly longer migration time than
207
trypsinogen (6.8 min). In order to test whether the protein-protein affinity in principle can be
208
monitored by CE, trypsinogen (~40 µM) was injected employing a BGE of 25 mM
209
ammonium acetate (pH 8.0) containing 35 µM aprotinin. A significant shift of the effective
210
electrophoretic mobility of trypsinogen was observed in comparison with the analysis of
211
trypsinogen in absence of aprotinin, indicating measurable protein-protein interaction.
212
Subsequently, infusion experiments were performed in order to determine optimal MS
213
conditions for the detection of free and complexed protein. Trypsinogen (40 µM) was infused
214
via the CE capillary at a flow rate of 300 nL·min−1 into the mass spectrometer using the CE-
35
36,37
. Trypsinogen and aprotinin show optimal affinity at pH
and, therefore, for CE analysis a BGE of 25 mM ammonium acetate (pH 8.0) was
38
. Both trypsinogen and aprotinin were analyzed individually
9
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 10 of 29
215
MS sprayer. Trypsinogen showed an ESI mass spectrum with a charge envelope of four
216
charge states (8 to 11 positive charges). Deconvolution of this mass spectrum yielded a
217
molecular weight of 23,981 Da, which is in good agreement with the expected mass for
218
trypsinogen. Next, trypsinogen (40 µM) was incubated with aprotinin (150 µM) for 10 min
219
(the minimum CE analysis time) and then infused into the mass spectrometer. After
220
deconvolution, the mass spectrum showed three main masses of 6,511 Da, 23,981 Da, and
221
30,493 Da which correspond to aprotinin, trypsinogen and the non-covalent trypsinogen-
222
aprotinin complex (stoichiometry of 1:1), respectively. However, the signal intensity for the
223
protein-protein complex was relatively low, even though significant binding was expected at
224
these relative concentrations. Probably, dissociation of the non-covalent complex occurred
225
during ionization and ion transfer through the ion optics. Therefore, ESI-MS parameters were
226
evaluated in order to achieve efficient detection of the aprotinin-trypsinogen complex.
227
Various dry gas flow rates (2-7 L·min−1), dry gas temperatures (70-200 ºC) and nebulizer gas
228
pressures (0.1-0.6 bar) were studied. Increase of protein-complex signals (relative to free
229
proteins signals) were obtained using relatively soft source conditions (i.e., dry gas of 80 ºC
230
at 3 L·min−1, and nebulizer gas pressure of 0.2 bar) and increased transfer and storage times
231
(200 and 20 µs, respectively). Special attention was paid to the quadrupole ion and collision
232
cell energy, since they are known to affect complex stability 26. Decrease of the collision cell
233
energy to values below 10 eV caused an increase of the complex intensity (Figure 1). A
234
similar trend was observed when lowering the quadrupole ion energy. Optimal aprotinin-
235
trypsinogen signal was achieved using a quadrupole ion energy of 5 eV and a collision cell
236
energy of 7 eV. Values below 5 eV for both collision cell and quadrupole ion energy did not
237
allow effective transfer of protein-protein complex to the detector and, no signals for the
238
protein-complex were observed (Figure 1).
10
ACS Paragon Plus Environment
Page 11 of 29
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
239
For the coupling of ACE with MS, the use of a sheath liquid consisting of IPA and 100 mM
240
acetic acid (75:25, v/v) was initially employed as it provided good ionization of the free and
241
aprotinin-bound proteins. However, in order to achieve a stable CE current over time,
242
addition of BGE to the sheath liquid was needed. Different IPA and BGE ratios were tested.
243
Good sensitivity and CE current stability were obtained with a sheath liquid consisting of
244
IPA-water-ammonium acetate (25 mM, pH 8.0) (25:50:25, v/v/v) and was further employed
245
for CE-MS analysis.
246
3.2. ACE-MS performance
247
3.2.1 Characterization of trypsinogen variants and their aprotinin complexes
248
The trypsinogen sample (including cortisone as EOF marker) was analyzed with the
249
optimized CE-MS method using a BGE without aprotinin. For trypsinogen five partially
250
separated peaks were detected indicating the presence of impurities and/or modified protein
251
forms (Figure 2A). Table 1 lists the masses obtained for each peak after deconvolution of the
252
ESI mass spectra. Peak 4 could be assigned to native trypsinogen. Based on the masses and
253
migration times, peaks 3, 2 and 1 were provisionally assigned to singly-, doubly- and triply-
254
deamidated trypsinogen. Adding one negative charge (at pH 8.0) and an increase of 0.984 Da
255
per modification, deamidation causes a change in the protein charge-to-size ratio, facilitating
256
CE separation. Next to triply-deamidated trypsinogen, peak 1 comprised two other
257
proteinaceous species of relatively low mass, which may be products resulting from auto-
258
proteolytic activity of trypsinogen. Peak 5 corresponded to a protein with a mass that was 18
259
Da lower than trypsinogen, but could not be assigned to an evident modification.
260
In order to study the affinity of the trypsinogen sample components, 25 µM of aprotinin was
261
added to the BGE and trypsinogen was analyzed (Figure 2B). Addition of aprotinin caused a
262
reduction of the EOF. In order to allow proper calculation of the effective electrophoretic 11
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 12 of 29
263
mobility (µeff) of the trypsinogen variants, a neutral EOF marker (cortisone) was included in
264
the samples. Using the CE instrument as viscosimeter (see Experimental section) no
265
significant changes in viscosity were observed for these concentrations of aprotinin in the
266
BGE. Hence, no viscosity correction of the µeff was needed. The reproducibility of the µeff
267
determination was assessed in absence and presence (25 µM) of aprotinin in the BGE on
268
three different days, including installation of capillary. RSDs for µeff were below 2.3% and
269
4.7%, respectively. The extracted-ion electropherograms show trypsinogen and its variants,
270
but with shifted effective electrophoretic mobilites with respect to Figure 2A, due to
271
interaction with aprotinin in the BGE. The peaks of trypsinogen and variants remained nicely
272
symmetric upon addition of aprotinin, confirming that the equilibration kinetics were fast.
273
The mass spectra of trypsinogen and the variants showed the characteristic masses of free
274
proteins (trypsinogen variants and aprotinin), but also clear signals of protein-aprotinin
275
complexes with total masses, indicating a 1:1 binding stoichiometry. Interestingly, the two
276
unassigned proteins comigrating with the triply-deamidated trypsinogen in Figure 2A (peaks
277
1’ and 1’’) showed no shift in effective electrophoretic mobility when aprotinin was added to
278
the BGE. Hence, these proteins exhibit no measurable interaction with aprotinin. This was
279
confirmed by the fact that the mass spectra obtained for these two compounds did not show
280
signals of protein-aprotinin complexes. Overall, these results indicate that ACE-MS can be
281
employed for the simultaneous characterization and affinity assessment of impure proteins
282
samples, including comigrating species.
283
3.2.2 Quantitative assessment of protein-protein interaction
284
In order to determine the Kds of aprotinin towards the trypsinogen variants, the trypsinogen
285
sample was analyzed in BGEs containing increasing concentrations of aprotinin. As
286
trypsinogen and aprotinin have a Kd in the low µM range, the aprotinin concentration in the 12
ACS Paragon Plus Environment
Page 13 of 29
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
287
BGE was varied between 0 and 300 µM. For most trypsinogen variants, increased µeff values
288
were observed when the concentration of aprotinin in the BGE was raised. The µeff was
289
determined for each trypsinogen variant and binding curves were constructed by plotting the
290
change in µeff against the aprotinin concentration (Figure 3A). Subsequent curve fitting using
291
non-linear regression yielded a Kd value of 8 ± 5 µM for trypsinogen, which is in reasonable
292
agreement with the only value reported so far in literature (2 µM) 35. Based on the changes in
293
effective electrophoretic mobility, Kd values of 4 ± 2, 4 ± 2, 9 ± 4 and 3 ± 2 µM were
294
determined for compounds 1, 2, 3 and 5 of the trypsinogen sample respectively (Table 1).
295
These values are not significantly different from the Kd of the native trypsinogen, indicating
296
that the chemical modifications did not affect the aprotinin binding. For the two variants
297
which co-migrated with the triply deamidated trypsinogen in absence of aprotinin, no shifts in
298
the mobility were observed in presence of aprotinin, confirming their lack of affinity.
299
In order to evaluate the accuracy of the ACE-MS method, Kds were also determined for
300
trypsinogen and its variants using ACE-UV employing the same BGEs. As several
301
components co-migrated in peak 1, it was not possible to determine their individual µeffs with
302
ACE-UV. The Kd values obtained for the rest of variants and trypsinogen were between 2 ± 1
303
and 6 ± 3 µM (Table 1). Seemingly slightly lower, these values do not differ significantly
304
from the values obtained by ACE-MS. Potential differences could be attributed to the fact
305
that in CE-MS, in contrast to CE-UV, a large part of the capillary is not thermostated. The
306
resulting somewhat higher temperature may affect protein-protein binding and, thus, the
307
observed Kd value 39,40.
308
ACE-MS in principle also allows assessment of protein-protein binding by using the obtained
309
MS data 10. Notably, the absolute abundances of the protein signals decrease due to ionization
310
suppression when the aprotinin concentration in the BGE is increased. However, as can be
311
seen in Figure 4, with increasing aprotinin concentration in the BGE, the abundance of the 13
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 14 of 29
312
aprotinin-trypsinogen complex ions relatively to the signal of the free trypsinogen ions
313
increases. These relative abundances can be used to determine Kd values, circumventing
314
signal suppression effects by aprotinin
315
the trypsinogen-aprotinin complex and the trypsinogen ions as function of the aprotinin
316
concentration in the BGE indeed provides a binding curve (Figure 3B), yielding a Kd value of
317
92 ± 36 µM. For the variants of trypsinogen, the MS-derived Kd values were between 20 ± 17
318
and 104 ± 71 µM (Table 1). These values are higher and show larger spread than the Kds
319
obtained using electrophoretic mobility shifts.
320
In order to appreciate the MS-based values, the protein-protein affinity was also assessed
321
using stand-alone ESI-MS, i.e., by infusion of pre-incubated mixtures of trypsinogen (40 µM)
322
and aprotinin (0-500 µM) into the mass spectrometer. From the signal intensities measured
323
for free and complexed trypsinogen, a Kd of 96 ± 34 µM was calculated for the aprotinin-
324
trypsinogen interaction. This value represents the overall Kd of trypsinogen, i.e., including its
325
deamidated forms, as it was not possible to reliably discern these variants from the native
326
protein by MS (mass differences of 0.984 Da only) without pre-separation. For the 23,963-Da
327
trypsinogen variant, a Kd of 59 ± 20 µM was determined. These Kds were similar to the MS-
328
based Kds obtained with ACE-MS, indicating that the higher values and spread is caused by
329
MS detection and not by the ACE analysis. However, the MS-based values obtained were
330
higher compared with the electrophoretic-mobility-shift values and the reference value 35. For
331
the MS-based Kd determination it is assumed that the ratio of the measured MS signals of the
332
free protein and the protein complexes reflect the concentration ratio of these species in
333
solution. However, the ionization efficiencies for the free protein and the protein-protein
334
complex most probably are not the same
335
protein complex may vary between gas and liquid phase (e.g. hydrophobic interactions are
336
weaker than noncovalent hydrogen bonds in gas phase) 28. Kd values based on changes in µeff
26
. Plotting the ratio of the cumulative abundance of
26
. Moreover, the stability of the non-covalent
14
ACS Paragon Plus Environment
Page 15 of 29
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
337
are measured in solution and seem more suitable for the determination of protein-protein
338
affinities using ACE-MS.
339
It can be concluded that the proposed ACE-MS method represents a promising means for the
340
simultaneous assessment of heterogeneity and the affinity of individual protein sample
341
constituents. The method, allows determination of Kd values for closely related sample
342
components of similar mass (e.g. deamidated variants) and for co-migrating components.
343
3.3. ACE-MS for affinity screening of heterogeneous proteins
344
The developed ACE-MS system can also be used for the relatively fast affinity screening of
345
heterogeneous proteins, offering qualitative information on the affinity of mixture
346
components/protein variants towards a target protein without the need for repeated
347
experiments using different target concentrations and extensive dissociation constant
348
calculations. In order to demonstrate the usefulness of ACE-MS for this purpose, the proteins
349
trypsin and ɑ-chymotrypsin were analyzed in absence and presence of aprotinin. Aprotinin is
350
known to bind strongly to both trypsin and α-chymotrypsin with Kd values in the pM and nM
351
range, respectively
352
using 0 and 25 µM aprotinin in the BGE. The trypsin sample showed five peaks (Figure 5A1)
353
which on the basis of their mass spectra were assigned to didehydro, acetylated and
354
methylated trypsin (peaks 2-4) and trypsin (peak 5); peak 1 could not be assigned to a
355
common modification of trypsin. The lower mass of the latter component as compared with
356
the intact protein (approx. 6 kDa less) suggests a fragment of trypsin, possibly as a
357
consequence of auto-proteolytic activity. For the ACE-MS analysis of trypsin with 25 µM of
358
aprotinin in the BGE, only the aprotinin-complexed species were observed in the mass
359
spectra of peaks 2-5 (Figure 5A2, red trace), which showed a change in migration time as
360
well. This indicates that next to trypsin itself, also the didehydro, acetylated and methylated
35
. Figure 5 shows the ACE-MS analysis of trypsin and ɑ-chymotrypsin
15
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 16 of 29
361
forms bind strongly to aprotinin. Interestingly, the variant with lower molecular weight (peak
362
1) did not exhibit a significant shift of the migration time (Figure 5A2, blue trace) and its
363
mass spectrum only showed ions corresponding to the free protein (blue trace). This implies
364
that this trypsin fragment has no affinity towards aprotinin.
365
CE-MS analysis of the α-chymotrypsin sample revealed four peaks (Figure 5B1). Peak 3
366
could be assigned to the unmodified protein, whereas peak 1 most probably is a deamidated
367
form. The mass spectra of peaks 2 and 4 showed higher masses compared to the unmodified
368
chymotrypsin (+216 Da for peak 2, and +196 and +246 Da for peak 4). When the α-
369
chymotrypsin was analyzed with aprotinin added to the BGE (Figure 5B2), clear relative
370
differences in affinity were observed. For the +246 Da protein species (peak 4’) only the free
371
protein was observed in the mass spectrum, indicating that this component has no affinity
372
with aprotinin. The rest of the α-chymotrypsin components showed migration time shifts in
373
presence of aprotinin, indicating binding. However, the affinities of these components
374
appeared to be different. For native and deamidated α-chymotrypsin, and the +196 Da
375
species (peak 4’’), only the aprotinin complexes were observed in their mass spectra. This
376
means that these components all have a high affinity to aprotinin. For the +216 Da species
377
(peak 2) both free and aprotinin-complexed protein was observed. This implies that this
378
variant has interaction with aprotinin, but its affinity is lower as compared with α-
379
chymotrypsin itself. Overall, these results demonstrate that the new ACE-MS method can be
380
highly useful for the screening of protein-protein affinities of components of heterogeneous
381
protein samples.
382 383
4. Concluding remarks
16
ACS Paragon Plus Environment
Page 17 of 29
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
384
The feasibility of ACE-MS for the simultaneous characterization of intact proteins and
385
determination of protein-protein affinity was studied using trypsinogen and the protease
386
inhibitor aprotinin as model compounds. Efficient protein separation and determination of
387
affinity-induced changes of protein electrophoretic mobilities was achieved using a PB-DS-
388
DS coated capillary. The use of soft ionization and ion transfer conditions was crucial to
389
provide adequate MS sensitivity while maintaining the formed protein-protein complexes.
390
Both CE and MS data acquired with ACE-MS, i.e. effective electrophoretic mobilities and
391
ratio of signal intensities of free and complexed protein, can be used to determine
392
dissociation constants. By measuring effective electrophoretic mobility shift of trypsinogen
393
as function of aprotinin concentration, similar Kds were found for all the interacting
394
trypsinogen variants, indicating comparable affinity. MS-based Kd assessment suffered from
395
a relative large spread in ESI-MS signal intensities and deviation in ESI efficiencies among
396
free and complexed protein. The ACE-MS approach also showed useful for screening of
397
high, medium and low-affinity components in heterogeneous protein samples. Overall, the
398
present work demonstrates that ACE-MS is a powerful tool for the study of protein-protein
399
interactions exhibiting fast equilibrium kinetics. In a single run, ACE-MS provides (a)
400
assignment of the molecular weight of all protein sample components, including modified
401
forms, other variants and impurities, (b) protein-variant-selective determination of multiple
402
dissociation constants, and (c) establishment of protein-complex stoichiometries. Application
403
of the proposed ACE-MS methodology to larger proteins and biopharmaceuticals is currently
404
being studied in our laboratory.
405
Still there are some limitations of the present method. Fast equilibration is not always the
406
case in protein-protein interactions. Alternative approaches as pre-equilibrium or kinetic ACE
407
may be employed in these cases 10. Use of BGEs at (near)-physiological conditions may not
408
provide optimal CE separation and/or ESI-MS sensitivity. Moreover, determination of Kd’s in 17
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 18 of 29
409
the high µM and mM range may be an issue, as the higher amounts of ligand required in the
410
BGE may result in strong suppression of the MS signal. Sensitivity can be increased and
411
ionization suppression can be reduced by the employment of sheathless interfacing for CE-
412
MS using, e.g.,
413
thermostatting almost the entire CE capillary. We plan to study the feasibility of sheathless
414
ACE-MS in the near future.
a porous tip capillary
41,42
. Sheathless interfacing also provides
415
416
18
ACS Paragon Plus Environment
Page 19 of 29
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
417
Acknowledgment
418
This research was supported by the Dutch Technology Foundation STW, which is part of the
419
Netherlands Organisation for Scientific Research (NWO) and partly funded by the Ministry
420
of Economic Affairs (project number 11056).
421
E. Domínguez-Vega and R. Haselberg contributed equally to this work.
422
19
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 20 of 29
423
Figure captions
424
Figure 1. Effect of the quadrupole ion and collision cell energy on the intensity of the
425
trypsinogen-aprotinin complex ions. Infusion of a pre-incubated mixture of trypsinogen (40
426
µM) and aprotinin (150 µM) at a flow rate of 300 nL·min−1. MS conditions: ESI voltage, -4.5
427
kV; nebulizer, 0.2 bar; drying gas, 3 L/min at 80 ºC. For other conditions, see Experimental
428
section.
429
Figure 2. CE-MS analysis of trypsinogen sample using a PB-DS-PB coated capillary and a
430
BGE of 25 mM ammonium acetate (pH 8.0) containing no (A) or 25 µM (B) aprotinin.
431
Extracted-ion elecropherograms of trypsinogen (2398 m/z), trypsinogen variants (2397-2399,
432
1814.1 and 1798.5 m/z) and cortisone (EOF marker) (361.2 m/z) are depicted. For other
433
conditions, see Experimental section.
434
Figure 3. ACE-MS binding curves for trypsinogen-aprotinin obtained by plotting (A)
435
effective electrophoretic mobility shifts and (B) the ratio between the relative intensity of
436
trypsinogen-aprotinin complex/free trypsinogen against the concentration of aprotinin in the
437
BGE. For experimental conditions, see Experimental section.
438
Figure 4. Mass spectra obtained for peak 4 during ACE-MS of trypsinogen using a BGE of
439
25 mM ammonium acetate (pH 8.0) containing no (A), 2 (B), 15 (C), 50 (D) or 150 µM (E)
440
aprotinin. Arrows indicate trypsinogen-aprotinin complex ions. For other conditions, see
441
Experimental section.
442
Figure 5. CE-MS analysis of trypsin (A) and α-chymotrypsin (B) using a PB-DS-PB coated
443
capillary and a BGE of 25 mM ammonium acetate (pH 8.0) containing no (A1 and B1) or 25
444
µM (A2 and B2) aprotinin. Cumulative extracted-ion electropherograms of free proteins
20
ACS Paragon Plus Environment
Page 21 of 29
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
445
(blue traces) and protein complexes (red traces) are depicted. For other conditions, see
446
Experimental section.
447
21
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 22 of 29
448
References
449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504
(1) Johnsson, N. Biochem Bioph Res Co 2014, 445, 739-745. (2) Heck, A. J. R.; van den Heuvel, R. H. H. Mass Spectrom Rev 2004, 23, 368-389. (3) Berggard, T.; Linse, S.; James, P. Proteomics 2007, 7, 2833-2842. (4) Syafrizayanti; Betzen, C.; Hoheisel, J. D.; Kastelic, D. Expert Rev Proteomic 2014, 11, 107120. (5) Kalman-Szekeres, Z.; Olajos, M.; Ganzler, K. J Pharmaceut Biomed 2012, 69, 185-195. (6) Haselberg, R.; Oliveira, S.; van der Meel, R.; Somsen, G. W.; de Jong, G. J. Anal Chim Acta 2014, 818, 1-6. (7) He, X. Y.; Ding, Y. S.; Li, D. Z.; Lin, B. C. Electrophoresis 2004, 25, 697-711. (8) Albishri, H. M.; El Deeb, S.; AlGarabli, N.; AlAstal, R.; Alhazmi, H. A.; Nachbar, M.; El-Hady, D. A.; Watzig, H. Bioanalysis 2014, 6, 3369-3392. (9) El Deeb, S.; Watzig, H.; Abd El-Hady, D. Trac-Trend Anal Chem 2013, 48, 112-131. (10) Chen, Z.; Weber, S. G. Trends in analytical chemistry : TRAC 2008, 27, 738-748. (11) Galievsky, V. A.; Stasheuski, A. S.; Krylov, S. N. Anal Chem 2015, 87, 157-171. (12) Zhao, S. S.; Chen, D. D. Y. Electrophoresis 2014, 35, 96-108. (13) Haselberg, R.; de Jong, G. J.; Somsen, G. W. Electrophoresis 2011, 32, 66-82. (14) Haselberg, R.; de Jong, G. J.; Somsen, G. W. Electrophoresis 2013, 34, 99-112. (15) Haselberg, R.; Brinks, V.; Hawe, A.; de Jong, G. J.; Somsen, G. W. Anal Bioanal Chem 2011, 400, 295-303. (16) Haselberg, R.; de Jong, G. J.; Somsen, G. W. J Sep Sci 2009, 32, 2408-2415. (17) Avila, L. Z.; Chu, Y. H.; Blossey, E. C.; Whitesides, G. M. J Med Chem 1993, 36, 126-133. (18) Heegaard, N. H. H. J Chromatogr A 1994, 680, 405-412. (19) Krylov, S. N. Electrophoresis 2007, 28, 69-88. (20) Kanoatov, M.; Galievsky, V. A.; Krylova, S. M.; Cherney, L. T.; Jankowski, H. K.; Krylov, S. N. Anal Chem 2015, 87, 3099-3106. (21) Cherney, L. T.; Krylov, S. N. Anal Chem 2011, 83, 1381-1387. (22) Cherney, L. T.; Krylov, S. N. Analyst 2012, 137, 1649-1655. (23) Haselberg, R.; de Jong, G. J.; Somsen, G. W. J Chromatogr A 2007, 1159, 81-109. (24) Monton, M. R.; Terabe, S. Anal Sci 2005, 21, 5-13. (25) Mironov, G. G.; Logie, J.; Okhonin, V.; Renaud, J. B.; Mayer, P. M.; Berezovski, M. V. J Am Soc Mass Spectr 2012, 23, 1232-1240. (26) Liu, J.; Konermann, L. J Am Soc Mass Spectrom 2011, 22, 408-417. (27) Daniel, J. M.; Friess, S. D.; Rajagopalan, S.; Wendt, S.; Zenobi, R. Int J Mass Spectrom 2002, 216, 1-27. (28) Peschke, M.; Verkerk, U. H.; Kebarle, P. J Am Soc Mass Spectr 2004, 15, 1424-1434. (29) Chu, Y. H.; Dunayevskiy, Y. M.; Kirby, D. P.; Vouros, P.; Karger, B. L. J Am Chem Soc 1996, 118, 7827-7835. (30) Chu, Y. H.; Kirby, D. P.; Karger, B. L. J Am Chem Soc 1995, 117, 5419-5420. (31) Lynen, F.; Zhao, Y.; Becu, C.; Borremans, F.; Sandra, P. Electrophoresis 1999, 20, 24622474. (32) Machour, N.; Place, J.; Tron, F.; Charlionet, R.; Mouchard, L.; Morin, C.; Desbene, A.; Desbene, P. L. Electrophoresis 2005, 26, 1466-1475. (33) Dunayevskiy, Y. M.; Lyubarskaya, Y. V.; Chu, Y. H.; Vouros, P.; Karger, B. L. J Med Chem 1998, 41, 1201-1204. (34) Fermas, S.; Gonnet, F.; Sutton, A.; Charnaux, N.; Mulloy, B.; Du, Y. G.; Baleux, F.; Daniel, R. Glycobiology 2008, 18, 1054-1064. (35) Vincent, J. P.; Lazdunski, M. Febs Lett 1976, 63, 240-244. (36) Bao, J.; Krylov, S. N. Anal Chem 2012, 84, 6944-6947. (37) Vuignier, K.; Veuthey, J. L.; Carrupt, P. A.; Schappler, J. Electrophoresis 2012, 33, 33063315. (38) Haselberg, R.; de Jong, G. J.; Somsen, G. W. Anal Chim Acta 2010, 678, 128-134. (39) Musheev, M. U.; Filiptsev, Y.; Krylov, S. N. Anal Chem 2010, 82, 8692-8695. (40) Musheev, M. U.; Filiptsev, Y.; Krylov, S. N. Anal Chem 2010, 82, 8637-8641. (41) Haselberg, R.; de Jong, G. J.; Somsen, G. W. Anal Chem 2013, 85, 2289-2296. (42) Medina-Casanellas, S.; Dominguez-Vega, E.; Benavente, F.; Sanz-Nebot, V.; Somsen, G. W.; de Jong, G. J. J Chromatogr A 2014, 1328, 1-6.
505 22
ACS Paragon Plus Environment
Page 23 of 29
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48
Analytical Chemistry
Table 1. Peak assignment and Kd values obtained by ACE-MS, ACE-UV and DIMS for trypsinogen sample components.
Peak
Mass
Tentative assignment
1’
14379.0 Da
not assigned
1’’
14505.0 Da
not assigned
1
23984.0 Da
2
Kd based on electrophoretic mobility shift ACE-MS ± s.d. ACE-UV ± (µM) s.d. (µM) no affinity n.d.
Kd based on relative MS signal intensity ACE-MS ± s.d. DIMS ± s.d. (µM) (µM) no affinity no affinity
no affinity
n.d.
no affinity
triply deamidated trypsinogen
4±2
n.d.
25 ± 37
23983.0 Da
doubly deamidated trypsinogen
4±2
3±1
3
23982.0 Da
singly deamidated trypsinogen
9±4
4±2
43 ± 22
4
23981.0 Da
trypsinogen
8±5
6±3
92 ± 36
5
23963.0 Da
not assigned
3±2
2±1
104 ± 71
no affinity
20 ± 17 96 ± 34*
59 ± 20
s.d., standard deviation n.d., not determined *overall value for components 1-4.
23
ACS Paragon Plus Environment
Analytical Chemistry
Figure 1
Relative intensity aprotinin-trypsinogen complex (%)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 24 of 29
100 90 80 70 60
Quadrupole ion energy (eV) 3 5 10 15
50 40 30 20 10 0 5 7 10 15 Collision cell energy (eV)
20
24
ACS Paragon Plus Environment
Page 25 of 29
Figure 2
Intensity (x104 cnts)
4
A
8
3
6 2 4 EIE 2398 ± 2 m/z
1’
EIE 1814.1 ± 0.5 m/z
2
5
1
1’’
EIE 1798.5 ± 0.5 m/z EIE 361.2 ± 0.5 m/z (EOF marker) 0
2
4
6
8
10
12
14
16
18
time (min)
4
Intensity (x104 cnts)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
B 4
3 3 2
2
1
0
EIE 1814.1 ± 0.5 m/z
1’
EIE 1798.5 ± 0.5 m/z EIE 361.2 ± 0.5 m/z (EOF marker)
1’’
2
4
5
1
EIE 2398 ± 2 m/z
6
8
10
12
14
16
18
time (min)
25
ACS Paragon Plus Environment
Analytical Chemistry
Figure 3
A
B
X10-5 1.5
Icomplexed protein /Ifree protein
Effective electrophoretic mobility shift (cm2V-1s-1)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 26 of 29
1.0
0.5
0.0 0
100
200
300
Concentration aprotinin in BGE (µ µM)
Peak 1 Peak 2 Peak 3 Peak 4 Peak 5
1.5
1.0
0.5
0.0 0
100
200
300
Concentration aprotinin in BGE (µ µM)
26
ACS Paragon Plus Environment
Page 27 of 29
Figure 4
+10 2399.0
100
A
75
+9 2665.5
50 25
B
100 75
Relative abundance (%)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
50
+12 2543.4
+11 2774.5
25
↓
↓
C
100 75 50
↓
↓
25
D
100
↓
75
↓
50 25 ↓
100
E ↓
75 50 25 2200
2300 2400 2500 2600 2700 2800 m/z
27
ACS Paragon Plus Environment
Analytical Chemistry
Figure 5
A1
2.5 2.0 4
1.5
3
1.0
2 1
0.5 0
B1
3.0 5
Intensity (x103 cnts)
Intensity (x103 cnts)
3.0
2.5
3
2.0 1.5 1.0 1 2
0.5
10
20
30
0
40
10
20
40
30
B2
1.5 3 1.0
Intensity (x103 cnts)
A2
2.0
4
time (min)
time (min)
Intensity (x103 cnts)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 28 of 29
5
2 4
0.5
1
2.0 1.5
3
1.0
12
4’’
4’
0.5
2
0
10
20
30
40
0
time (min)
10
20
30
40
time (min)
28
ACS Paragon Plus Environment
Page 29 of 29
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
85x47mm (300 x 300 DPI)
ACS Paragon Plus Environment