Analysis of Intact Proteins Using Liquid Chromatography

R.P. Freeman, H.J. Wirth, A.A. Gooley

SGE Incorporated, 2007 Kramer Lane, Austin Texas. 78758, USA
Abstract
The demand for separation techniques for intact proteins is increasing with the introduction of a new generation of high resolution mass spectrometers which are able to measure the
mass of small to medium size proteins very accurately. Liquid chromatography is a valuable tool for separating these proteins prior to the MS analysis. Intact protein chromatography is
most commonly used in a top-down approach in proteomics and to determine expression levels during recombinant protein synthesis.
The size of the protein molecule results in very low diffusion coeffcients and therefore slow mass transfer in and out of the pore system. A suffciently large pore diameter is required to
minimise the effects of restricted pore diffusion. We show examples of the separation of intact proteins on a column packed with 3 m C8 silica with 1000 pore size. The molecular
weight of the protein examples reported here cover ribosomal proteins (<40 kDa), monoclonal antibodies (~150 kDa) and intact membrane proteins derived from mouse liver.
Physico-Chemical Background
Diffusion Inside Pores
Renkin Equation (E.M. Renkin, J.Gen.Physio., 38 (1954) 225.)

(
(

|
|
.
|

\
|

|
|
.
|

\
|
+
|
|
.
|

\
|

|
|
.
|

\
|
=
5 3 2
95 . 0 09 . 2 104 . 2 1 1
p
s
p
s
p
s
p
s
f p
r
r
r
r
r
r
r
r
D D
D Diffusion coefficient inside the pores
r = Stokes radius of the analyte
r = Pore radius
p
s
p
=
Sweet spot, where diffusion rates are between 50%and 80%
of the free diffusion. Good compromise between fast diffusion
kinetics and high capacity .
Fast diffusion rates but low capacities -
Pores are larger than needed.
Slow diffusion rates - high capacity -

Pores are too small for effective mass transfer.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Analyte diameter/pore diameter
1000 - 5kDa
1000 - 500Da
1000 - 100kDa
300 - 5kDa
300 - 500Da
300 - 100kDa
200 - 5kDa
200 - 500Da
200 - 100kDa
120 - 5kDa
120 - 500Da
120 - 100kDa
100 - 5kDa
100 - 500Da
100 - 100kDa
D
/D
p
f
Figure 2: Relative diffusion rates in free solution and inside pores as function of molecule size to pore
size ratio
The diffusion rate is described by:

r
T k
D
B
6
=
Einstein-Stokes Equation
Diffusion in Free Solution
0
10
20
30
40
50
60
70
80
90
100
Caffeine Insulin hSA IgG
100%
20.3%
7.1%
3.7%
Compound Size[] Df
Caffeine 6.3 3.89E -10
Insulin 31 7.91E -11
hSA 2.75E -11
IgG 170 1.44E -11
89
Figure 1: Normalized diffusion rates of small, mediumand large molecules in free solution
D = Diffusion rate constant in free solution
k Boltzmann constant
T = Temperature
r = Stokes radius
= Viscosity
f
B=
Figure 3: Design of the ProteCol Range Columns
Accurate Mass Analysis of Intact Ribosomal Proteins
Introduction
Since ribosomal proteins are relatively small (Mw=6,000 to 40,000), they can be rapidly identifed by accurate LC MS analysis of the intact proteins. In the present application ribosomal
proteins isolated from rat liver were separated on a ProteCol C8 HQ1003 column.
Sample Preparation
80S ribosomal proteins were isolated from a rat liver microsomal preparation (Williamson et al; 1997, Eur. J. Biochem. 246: 786-793). One optical density unit at 260 nm of 80S ribosomal
proteins was mixed with two volumes of 6M Guanidine HCl to denature the proteins. 1 % (v/v) formic acid was subsequently added to precipitate the nucleic acids. The mixture was
centrifuged for 15 min at 13,000 rpm and the supernatant was collected into a sample vial ready for LCMS analysis.
Figure 4: Base peak chromatogramof ribosomal proteins
Data Analysis and Results
All data were acquired and reference mass corrected via a dual-spray electrospray ionisation (ESI) source. Each scan or data point on the Total Ion Chromatogram (TIC) is an average
of 15,000 transients, producing a spectrum every second. Mass spectra were created by averaging the scans across each peak and background subtracted against the frst 10 seconds
of the TIC. The resulting base peak chromatogram shows very high peak capacity - 119 discrete protein masses were identifed; 46 of which were identifed as 80S ribosomal proteins.
In some cases several different masses of the same protein were identifed which correlated with known N- and/or C-terminal processing.
Accurate Mass Determination of Intact Monoclonal Antibodies
A 10 mm x 2 mm ID trap column was used for desalting and sample focussing prior to the MS analysis. The trap column was packed with 3 m - 1000 C8 silica. Again, the large pore
size of the stationary phase facilitates a narrow elution profle.
LC System: UltiMate 3000 RSLC-Ti
RP Column: ProteCol C8 HQ1008, 10 mmx 2 mm, 3 m, 1000 (trap column)
Mobile phase A: 100%water + 0.1%Formic Acid
Mobile phase B: 5/95 water/acetonitrile, 0.1%Formic Acid
Gradient: 5-90%B in 10 minutes
Flow-rate: 400 l/min
Temperature: 70C
Sample amount: mAb 20 mg/mL, 5 L injected
MS: Bruker maXis
Analysis of Membrane Proteins
The experiment was performed in three steps:
1. Separation of intact membrane protein sample derived from mouse liver
using a 3 m C8 column with 1000 pore size.
2. Fractions collected from 1. were digested with trypsin off-line.
3. Digested fractions were analyzed by LC-MS/MS using 3 m C18 column with
300 pore size.
Chromatographic conditions for analytical separation
System: Agilent 1100 CapLC with Agilent MSD-iontrap MS
Column: ProteCol-C8 HQ1003 3 m; 1000 150 mmx 300 mID
Sample: 3l AOHUPO-MPI standard
Flow rate: 5.0 l/min
Temperature: 80C
Mobile Phase A: 0.1%formic acid in water
Mobile Phase B: 0.09%formic acid in acetonitrile
1-minute fractions of the intact protein separation were collected and digested
with trypsin off-line. The dried fractions were reconstituted in 100 mM
NH4HCO3 and digested with trypsin overnight, acidifed with 1% formic acid
(FA), concentrated and re-diluted to 10 l with 1% FA.
Chromatographic conditions
System: TSP4000 pump, Surveyor autosampler, Thermo linear ion trap
Column: ProteCol-C18 HQ303 150 mID x 100 mm
Sample volume: 10 l
Mobile phase A: 0.1%FA in 5%acetonitrile
Mobile phase B: 0.1%FA in 90%acetonitrile
MS: 400-1500 mass range; top 6 ions fragmented with 39%collision energy
Figure 5: Base peak chromatogramof mAb and MS traces
Data analysis
Raw data fles were converted to mzXML and searched against the Ensembl
mouse database using Xtandem algorithm (GPM-XE software).
The found proteins were statistically fltered either by number of identifed
peptides (>4) and/or by a cutoff score (10-10). As a result 542 proteins were
identifed with high confdence.
Conclusions
By combining stationary phases with the right surface chemistry and the
appropriate pore size for the analyte it is possible to separate even diffcult
samples such as membrane proteins. Our investigations into the effect of particle
porosity led to the following recommendation for analyte size:
0 10 20 30 40 50 60 70 80 90 100 Time [ min]
0
1
2
3
4
I ntensity x106
AOHU O
M O NAS H
U N I V E R S I T Y
Gradient profile: Time %B
0 20
2 20
42 50
52 80
57 100
72 100
75 20
105 20
Figure 6: Base peak chromatogramof membrane proteins on a capillary column
Figure 7: Example of analysis of digested fractions (fraction 30)
M+1=530.6
M+1=672.7
M+1=432.5
M+1=1181.4
M+1=1420.6
NNGK
LLEACTFHK
LYLGHNYVT AIR
DHMK
10 20 30 40 50 60
0
1
2
3
4
5
6
7
8
9
10
x107
0
1
2
3
4
5
6
7
8
9
10
x106 BPC of fraction 30
EIC for peptides related to peptidase S60
100
120
200
300
1000
30
50
240
800
29,400
800
1400
6,400
21,400
800,000
13
23
100
350
13,000
350
610
2,800
9,500
350,000
Spherical Analyte Cylindrical Analyte
Pore Size lower upper lower upper
IgG Mw ~ 150kDa
hSA Mw ~ 68kDa
Antitrypsin Mw~44kDa
Transferrin Mw~80kDa
1000 pore size
C8 silica
300 pore size
C18 silica
Figure 8: Relative pore and protein sizes for 1000 and 300 pores
Acknowledgements
The ribosomal protein results were obtained from Nicholas Williamson and
Paul ODonnell from Bio21 Mass Spectrometry and Peptide Institute, Bio21
Molecular Science and Biotechnology, Melbourne.
Monoclonal antibody data was provided by Matthias Pelzing, Bruker Biosciences,
Melbourne.
The membrane protein analysis was done in collaboration with Mibel Aguilar,
Monash University and Paul Haynes, APAF, Australia.
TP-0208-H SGE Analytical Science Pty Ltd 02/2011
1237.2 1391.8
1499.1
1665.5
1892.8
2894.2
3208.7
3280.1
3432.5
3514.3
0.0
0.2
0.4
0.6
0.8
x10
4
In
te
n
s
ity
1000 2000 3000 m/
b
+MS 5.02-5.31min
#(136-144)
1.0
x10
4
In
te
n
s
ity
+MS 3.84-4.02min
#(104-109)
a
0
1
2
3
4
1000 1500 2000 2500 m/z
1268.5
1384.7
1449.4
1615.1
1690.8
2 3 4 5 6 Time [min]
a
b
x10
4
In
te
n
s
ity
0
1
2
3
4
BPC 1449.40.2
BPC 2500-3000
1237.2 1391.8
1499.1
1665.5
1892.8
2894.2
3208.7
3280.1
3432.5
3514.3
0.0
0.2
0.4
0.6
0.8
x10
4
In
te
n
s
ity
1000 2000 3000 m/z
b
+MS 5.02-5.31min
#(136-144)
1.0
x10
4
In
te
n
s
ity
+MS 3.84-4.02min
#(104-109)
a
0
1
2
3
4
1000 1500 2000 2500 m/z
1268.5
1384.7
1449.4
1615.1
1690.8
2 3 4 5 6 Time [min]
a
b
x10
4
In
te
n
s
ity
0
1
2
3
4
BPC 1449.40.2
BPC 2500-3000
1237.2 1391.8
1499.1
1665.5
1892.8
2894.2
3208.7
3280.1
3432.5
3514.3
0.0
0.2
0.4
0.6
0.8
x10
4
In
te
n
s
ity
1000 2000 3000 m/z
b
+MS 5.02-5.31min
#(136-144)
1.0
x10
4
In
te
n
s
ity
+MS 3.84-4.02min
#(104-109)
a
0
1
2
3
4
1000 1500 2000 2500 m/z
1268.5
1384.7
1449.4
1615.1
1690.8
2 3 4 5 6 Time [min]
a
b
x10
4
In
te
n
s
ity
0
1
2
3
4
BPC 1449.40.2
BPC 2500-3000
Dedicated Protein Columns - C8, 3 m, 1000
The SGE column range for biological analytes focuses on an optimized pore size for both peptides (200 and 300 ) and proteins (1000 ) and small particle size (3 m) to enable lowest
possible mass transfer restrictions.
Exposure to the analyte to metal surfaces is minimized through the use of glass-lined or PEEK-lined stainless steel tubing. All tubing is PEEK coated fused silica. Great emphasis is put
on the avoidance of void volumes in the design of the capillary column range.