Glycosylation analysis of immunoglobulin G Maurice Selman Leiden University Medical Center Department of Parasitology Biomolecular Mass Spectrometry Unit Leiden – The Netherlands
Structure IgG
Structure of both N-glycans determine Fc conformation
R. Jefferis, Expert Opin. Biol. Ther. (2007) 7(9), 1401-13
Lack of core-fucosylation enhances IgG1 binding to Fcγ-receptors N-acetylglucosamine galactose mannose fucose sialic acid
antibody-dependent cellular cytotoxicity (ADCC)
macrophage
Shields et al, 2002, J. Biol. Chem., Iida et al, 2006, Clin. Cancer Res., Ferrara et al, 2011, PNAS
Research goal
•
Set up robust and fast analysis methods for IgG (subclass specific) Fc Nglycopeptide profiling – Sample preparation (miniaturized) – Analysis – Data processing
•
Study various IgG Fc N-glycosylation features such as fucosylation, sialylation, galactosylation and the incidence of bisecting GlcNAc in clinical samples
Tryptic peptides of the 4 human IgG subclasses
G0F E293EQYNSTYR301
IgG1
2634.0 Da
E293EQFNSTFR301
IgG2, IgG3
2602.1 Da
E293EQFNSTYR301
IgG4
2618.1 Da pep
IgG glycosylation profiling: sample work-up
sample work-up in the 96-well-plate format: up to 4 plates a day (384 samples)
purify IgG’s from plasma with Protein A or G affinity chromatography
overnight digestion with trypsin
nanoLC-ESI-MS
Desalt and purify digest (RP or HILIC)
(detailed picture, 110 injections a day)
MALDI-MS less detailed picture, 384 samples a day
Wada et al, 2007, Glycobiology
NanoLC sheath-flow ESI-MS analysis tryptic IgG glycopeptides BPC nanoLC capillary
sheath liquid
EIC IgG1
G0F G1F G2F G2FS
EIC IgG4
EIC IgG2
7
8
9
10
11
Time [min]
Selman et al, 2012, J. of Proteomics
dry gas (nitrogen)
Robustness nanoLC sheath-flow ESI-MS IgG1 100 Average preparation 1 90 Average preparation 2 80 Average preparation 3 70 Average all preparations 60 50 40 30 20 10 0
IgG2 100 90 80 70 60 50 40 30 20 10 0
G0F
G1F
G2F
G0FN
G1FN
pep
pep
pep
pep
pep
G2FN pep
G1FS pep
G2FS pep
G1FNS pep
G2FNS pep
G0 pep
G1 pep
G2 pep
G0N pep
G1N pep
G2N pep
G1S pep
G2S pep
Pregnancy related IgG Fc N-glycosylation changes
IgG2
Relative abundance (%)
IgG1
Relative abundance (%)
Galactosylation
Sialylation
SA/Gal
35
90 80
25 70 60
15
50
Bisecting GlcNAc
23
20
21
18
19
16
17
14
15
12
13
10
11
8
40
5
9
6
76
38
27
15
23
13
19
11
15
9
66 28 56 18 46 36
8
1
2
3
4
5
6
Pregnancy time point
11
1
2
3
4
5
6
Pregnancy time point
7
1
2
3
4
5
6
Pregnancy time point
1
2
3
4
5
6
Pregnancy time point
nanoLC sheath-flow ESI-MS conclusions
•
Sheath-flow ESI sprayer allows robust zero dead volume nanoLC-MS interfacing
•
In-spray mixing of sheath-liquid with LC eluent allows the use of TFA containing mobile phases
•
Fast and repeatable IgG subclass specific Fc N-glycopeptide profiles can be obtained (total analysis time = 13 minutes, 110 injection a day)
•
NanoLC sheath-flow ESI-MS is highly suitable for the analysis of large clinical and biopharmaceutical sample cohorts
IgG glycosylation profiling: sample work-up sample work-up in the 96-well-plate format: up to 4 plates a day (384 samples)
purify IgG’s from plasma with Protein A or G affinity chromatography
overnight digestion with trypsin
nano-HPLC-ESI-MS
Desalt and purify digest (RP or HILIC)
(detailed picture, 110 injections a day)
MALDI-MS less detailed picture, 384 samples a day
High throughput IgG glycopeptide desalting and purification
RP and HILIC desalting in the 96-well-plate format: up to 4 plates a day (384 samples)
reversed phase purification (C18): sample elution with acidified ACN (peptides and glycopeptides)
HILIC purification (Sepharose): Sample elution with water (glycopeptide specific)
RN-mode MALDI-TOF-MS with DHB pep
pep
pep
pep
RN-mode MALDI-TOF-MS with ClCCA
pep
pep
pep
pep
pep
pep
Pep Peptide moiety IgG1 glycopeptide
0.6
2762.10 pep
0.4 pep
3215.24 3247.24
0.2
*
0.0
pep
3215.24 3247.23
2956.14
2632.01
pep
pep
0.2
*
2396.96
0.4
2924.14
pep
2924.15 2956.16
0.6
2632.04
0.8
2794.09
0.8
2794.10
1.0
2396.96
Relative intensity
1.0
2600.04
2600.05
IgG 2 glycopeptide
0.0 2400
2500
2600
2700
2800
m/z
2900
3000
3100
3200
2400
2500
2600
2700
2800
m/z
2900
3000
3100
3200
40000
*
2762.55
40000
2761.98
* 30000
30000 pep Intensity [a.u.]
pep
20000
20000
10000 10000
0
2761
2762
2763
2764 2765 m/z
2766
2767
2768
2769
0
2761
2762
2763
2764 2765 m/z
2766
2767
2768
2769
Theoretical mass IgG2 G1F: 2762.09 Da Matrix
Measured mass (Da)
Error (ppm)
95% confidence interval (Da)
S/N ratio
Cl-CCA (A)
2761.98*
40*
0.05
175*
11083*
DHB (B)
2762.55*
167*
0.16
85*
6490*
THAP/diammonium citrate
2762.25¥
68 ¥
0.09
40¥
7710 ¥
*Values represent average of 10 replicates ¥Values represent average of 5 replicates
Selman et al, 2012, Proteomics
Resolution
Conclusions MALDI-MS
•
Entire workflow (IgG capturing, digestion and glycopeptide desalting) is suitable for robotized platforms
•
High throughput sample preparation and MALDI-MS analysis is possible (384 samples in less than 36 hours)
•
Cold matrixes allow the registration of intact sialylated species by negative ion mode MALDI-TOF-MS
•
ClCCA has a high potential for high throughput glycopeptide profiling
•
Data evaluation (automated peak picking and integration)
Acknowledgments
LUMC, Biomolecular Mass Spectrometry Unit • Rico Derks • Carolien Koeleman • Bart Schoenmaker • Liam Mcdonnell • Magnus Palmblad • Gerhild Zauner • André Deelder • Manfred Wuhrer
Roche • Dietmar Reusch • Niklas Engler • Markus Haberger