3
Thermo Scientific Poster Note
•
PN ASMS13_W442_SLin_e 06/13S
tibody (MAb) charge variants
ts is achieved using linear pH gradient
t mass information is acquired on the
pole-Orbitrap mass spectrometer.
h resolution separation of MAb
terizes the structural difference of the
ous due to modifications such as
uncation. Salt gradient cation-
ome success in characterizing MAb
n required to tailor the salt gradient
drug development environment, a
majority of the MAb analyses.
eparate MAb charge variants using
buffer employed to generate the pH
tris, covering a pH range of 6 to 9.5.
of the pH increase was shallow at the
dy, we present a novel pH gradient
t is more linear. This method features
ear gradient was run from 100%
H buffer). Using an online pH meter,
hieved. Furthermore, a plot of
el proteins versus their pI values
te pH elution range of the target MAb
timization of separation can simply be
narrower pH range.
a. Harvest cell culture and
otech company. Proteins and MAb
4
×
250 mm (P/N 074625)
rmo Scientific™ Dionex™
Rapid Separation Pump
artment with two biocompatible
ration Thermostatted Autosampler
Micro Flow Cell
0X buffer A (pH 5.6) and one bottle of
prepared by simply diluting the
d water.
Results
Linear pH gradient
The linear pH gradient was achieved by employing a multi-component buffer system
containing multiple zwitterionic buffer species with pI values ranging from 6 to 10.
Eluent A was titrated to pH 5.6 and eluent B was titrated to pH 10.2. In this pH range,
each buffer species was either neutral or negatively charged. Therefore they were not
retained by cation-exchange column stationary phase and served as good buffers for the
mobile phase and the stationary phase.
Using the gradient method shown in Table 1, six proteins with a range of pI values from
6 to 10 were effectively separated on a MAbPac SCX-10, 10 µm, 4
×
250 mm column.
These proteins were lectin (including three isoforms, lectin-1, lectin-2, and lectin-3),
trypsinogen, ribonuclease A, and cytomchrome C. The chromatogram was shown in
Figure 1. The pH value measured in this experiment as a function of time was plotted in
Figure 2. The pH gradient was essentially linear from pH 5.6 to pH 10.2 over a 30
minute period. The correlation coefficient value R
2
was 0.9996.
An analysis was performed to show that there is a correlation between the elution pH for
the peaks and the corresponding pI values of the protein components. Figure 3 is a
graph comparing the measured pH values for six protein component peaks in Figure 1
as a function of the corresponding pI values. The measured pH values for the six protein
component peaks exhibited a strong linear correlation to the literature based pI values.
Thus, after a calibration procedure, this example supports the fact that linear regression
coupled with the gradient method described here can be used to estimate the pI of a
protein component based on the peak retention time and measured pH.
Table 1. 30 min linear gradient method for the MAbPac SCX-10, 10 µm, 4
×
250
mm, cation exchange columns.
Total run time is 40 min. The linear pH range covers
from pH 5.6 to pH 10.2. UV wavelength was set at 280 nm.
FIGURE 1. Chromatogram of six proteins s
gradient on a MAbPac SCX-10, 10 µm, 4
×
retention time, and corresponding pH values
FIGURE 3. A graph plotting the measured
peaks as a function of the corresponding
six components were exported from the same
Linear pH Gradient Chromatography
The linear pH gradient was generated by running linear gradient from 100% eluent A
(pH 5.6) to 100% eluent B (pH 10.2). For pH gradient analysis carried out on the
MAbPac SCX-10, 10 µm, 4
×
250 mm, cation-exchange columns, the gradient
method in Table 1 was used unless further stated.
y = 0.1548x + 5.0404
R²= 0.9996
5.5
6.5
7.5
8.5
9.5
10.5
0
10
20
Measured pH value
Retention Time [min]
FIGURE 2. A graph showing measured pH
measured pH values were exported from the
The measured pH values are labeled using bl
Lectin-1
Lectin-2
Lectin-3
Trypsinogen
Ribonu
y = 1.6923x - 7.2
R² = 0.9929
5.5
6
6.5
7
7.5
8
8.5
9
9.5
10
10.5
7.5
8.5
9.5
Measured pH value
pI value
0
5
10
15
-5.0
10.0
20.0
30.0
40.0
50.0
60.0
Absorbance [mAU]
Reten
Lectin-1 - 5.87 - 6.04
Lectin-2 - 6.97 - 6.20
Lectin-3 - 8.18 - 6.37
Trypsinogen -15.97 - 7.55
LC-MS
First dimension HPLC: in a scale up purification, 1 mL of IgG was purified from
3.8 mL of HCC using Thermo Scientific™ Pierce Protein A beads. The protein
concentration was determined at ~ 0.5 mg/mL. About 33 µL of the purified IgG was
injected onto a MAbPac SCX-10, 10 µm, 4
×
250 mm column and separated via linear
pH gradient from pH 6.52 to pH 9.28. The column was equilibrated at 40% B. Three
minutes after sample injection, a linear gradient was run from 40% to 100% B in 30
minutes. Fractions were collected onto a 96-well plate at a rate of 0.2 min per fraction
from 10 to 26 min.
Second dimension LC-MS: Thermo Scientific™ ProSwift™ RP-10R monolithic column
(1
×
50 mm) was used for desalting. LC solvents were 0.1% formic acid in H
2
O (Solvent
A) and 0.1% formic acid in acetonitrile (Solvent B). Column was heated to 50 ºC during
analysis. Flow rate was 100 µL/min. After injection of MAbs, a 5 min gradient from 10% B
to 95% B was used to elute the mAbs from the column.
MS: The Q Exactive Orbitrap mass spectrometer was used for this study. Intact MAb was
analyzed by ESI-MS for intact molecular mass. The spray voltage was 4kV. Sheath gas
flow rate was set at 10. Auxiliary gas flow rate was set at 5. Capillary temperature was
275 ºC. S-lens level was set at 55. In-source CID was set at 45 eV. Resolution was
17,500. The AGC target was set at 3E6 for full scan. Maximum IT was set at 200 ms.
Data Processing: Full MS spectra of intact MAbs were analyzed using Thermo
Scientific™ Protein Deconvolution 1.0 software that utilizes the ReSpect algorithm for
molecular mass determination. Mass spectra for deconvolution were produced by
averaging spectra across the most abundant portion of the elution profile for the MAb.
The averaged spectra were subsequently deconvoluted using an input m/z range of 2000
to 4000 m/z, an output mass range of 140000 to 160000 Da, a target mass of 150000 Da,
and a minimum of at least eight consecutive charge states from the input m/z spectrum to
produce a deconvoluted peak.
Time
(minutes)
Flow rate
(mL/min)
%A
%B
0-1
1
100
0
1-31
1
100-0
0-100
31-34
1
0
100
34-40
1
100
0
