(A) (B)
Mannose
11.7
13.1
13.7
10 11 12 13 14 15 16 17 18 19 Time [min]0
2
4
6
5x10Intens.
Glucose Galactose Glucose
Mannose
Galactose Hexose Standards
A coagulation Factor Protein
permethylated alditols
Mannitol (15%) in formulation
CHO derived
Profiling and Heightened Characterization of O-linked Glycans from Therapeutic Glycoproteins Elaine Sun1, Mario DiPaola1, Marshall Bern2, Andrew Hanneman1 1Charles River Laboratories, Woburn, MA; 2Protein Metrics Incorporated, San Carlos, CA
4 LC-MS wi th Charged Aerosol Detect ion (CAD) 1 Key Points • Biopharma requires robust methods for routine profiling and detailed structural characterization of O-linked glycans
• Monitoring the degradation reactions associated with maintaining an intact reducing end may dictate the most suitable approaches
• Advantageous if: the same method used to profile oligosaccharides includes monosaccharide O-glycans in the same chromatogram
• The exact methods for pursuing detailed structural analysis (accurate mass, MS/MS, and MSn) is an important consideration, especially for complex glycoproteins such as blood derived protein products
3 MS-based Methods used to Prof i le and Character ize O - l inked Glycans
2 O-glycan Divers i ty
Front. Genet., 19 March 2015 | Louise E. Tailford†, Emmanuelle H. Crost†, Devon Kavanaugh† and Nathalie Juge*
*
* Common in CHO-expressed proteins
Mucin Type O-glycans
Non-CHO Glycoproteins 1. Emerging Expression Systems: O-glycosylation in Spodoptera frugiperda (Sf9) and Trichoplusia ni (Hi-5) insect cell lines is complex and include abundant hexuronic acid (Sf9 and Hi-5) and O-linked phosphocholine (Sf9) Glycobiology vol. 22 no. 11 pp. 1593 Abstract #217 Stefan Gaunitz, Chunsheng Jin, Anki Nilsson, Jining Liu, Niclas G. Karlsson and Jan Holgersson
2. Glycoengineering: e.g. GlycoSwitch® Jacobs PP, Geysens S, Vervecken W, Contreras R, Callewaert N. Engineering complex-type N-glycosylation in Pichia pastoris using GlycoSwitch technology. Nat Protoc. 2009; 4:58–70. Hamilton SR, Cook WJ, Gomathinayagam S, Burnina I, Bukowski J, Hopkins D, Schwartz S, Du M, Sharkey NJ, Bobrowicz P, Wildt S, Li H, Stadheim TA, Nett JH. Production of sialylated O-linked glycans in Pichia pastoris. Glycobiology. 2013;23:1192–1203
3. Various therapeutic glycoproteins are derived from human blood
Profiling and Characterization of O-glycans from Diverse/Complex O-glycoproteins 1) A wide variety of O-glycans exist in nature
*
* Mucin-type Therapeutic glycoprotein with a mucin-type domain
1. Analyze O-glycans while they still reside on the protein or peptide backbone (may need to de-N-glycosylate, de-sialylate, etc.)
De-N-glycosylated and de-sialylated (CHO expressed glycoprotein) Note “nascent” O-glycan monosaccharide moieties (GalNAc )
2. Release by non-reductive β-elimination for subsequent chromophore or fluorophore labeling (important to monitor degradation)
3. Release using “classical” reductive β-elimination (minimal degradation “peeling” product)
“classical RBE” reductive beta elimination Carlson, 1966
non-reductive beta elimination for chromophore or fluorophore labeling
acidic α-proton
peptide backbone
A) Permethylated alditols (RP-LCMS/MS) • Release and clean-up • Permethylation • C18 RP LC-MS/MS (Ion Trap CID) • Extracted Ion Chromatograms (XICs) for quantification
A
“pee
ling”
prod
uct
Online MS/MS
B) Native Alditols (PGC-LC-CAD-MS/MS)
Characterization
Profiling
Flow splitter Thermo Scientific™ Vanquish™ UPLC system
Thermo Scientific Charged Aerosol detector
Thermo Scientific™ Hypercarb™
HPLC Column (PGC)
Dual detection data streams (CAD, and MS/MS) analyzed in collaboration with Protein Metrics
CHO derived
1. Configuration
fresh
mV
0.00
20.00
40.00
60.00
mV
0.00
20.00
40.00
60.00
mV
0.00
20.00
40.00
60.00
Minutes6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 22.00 24.00 26.00 28.00 30.00 32.00 34.00 36.00 38.00 40.00 42.00 44.00 46.00 48.00 50.00 52.00 54.00 56.00
mV
0.00
20.00
40.00
60.00
mV
0.00
20.00
40.00
60.00
mV
0.00
20.00
40.00
60.00
Minutes6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 22.00 24.00 26.00 28.00 30.00 32.00 34.00 36.00 38.00 40.00 42.00 44.00 46.00 48.00 50.00 52.00 54.00 56.00
mV
0.00
20.00
40.00
60.00
mV
0.00
20.00
40.00
60.00
mV
0.00
20.00
40.00
60.00
Minutes6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 22.00 24.00 26.00 28.00 30.00 32.00 34.00 36.00 38.00 40.00 42.00 44.00 46.00 48.00 50.00 52.00 54.00 56.00
mV
0.00
20.00
40.00
60.00
mV
0.00
20.00
40.00
60.00
mV
0.00
20.00
40.00
60.00
Minutes6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 22.00 24.00 26.00 28.00 30.00 32.00 34.00 36.00 38.00 40.00 42.00 44.00 46.00 48.00 50.00 52.00 54.00 56.00
Pro-Thr-ol
mV
0.00
100.00
200.00
mV
0.00
100.00
200.00
mV
0.00
100.00
200.00
mV
0.00
100.00
200.00
Minutes4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 22.00 24.00 26.00 28.00 30.00 32.00 34.00
50 mM NaOH 0.5 M NaBH4
100 mM NaOH 0.75 M NaBH4
100 mM NaOH 0.5 M NaBH4
Carlson 1966
1. Buffer exchange samples first 2. Use fresh sodium borohydride 3. Check peeling product (*) to optimize release 4. Other species may be present, so use MS/MS to characterize the peaks
*
2. Release optimization
mV
5.00
10.00
15.00
20.00
25.00
30.00
35.00
mV
5.00
10.00
15.00
20.00
25.00
30.00
35.00
40.00
Minutes2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 22.00 24.00 26.00
Method Blank
Receptor-Fc (de-N-glycosylated)+ β-elimination and SPE PGC RP-HPLC/CAD/MS
background
(trace)
sucrose
37.3%
62.7%
C)Buffer exchange to remove formulation components.
Buffer blank
Include to help identify unknowns.
3. Applications A) Biopharma samples by PGC-CAD-MS (CHO derived)
B) Blood derived O-glycosylated products (MSn characterization of permethylated O-glycans)
• Extended signal averaging (v. HPLC) • Product/precursor relationships • Spectrum matching and libraries
5 Conclusions 1. Currently, we prefer traditional reductive beta elimination due to the diversity of structures potentially presented in a contract
research organization (CRO) environment (potential degradation reactions cannot be tolerated) 2. We closely monitor peeling reactions during method development 3. Place monosaccharides within the same chromatogram as oligosaccharides 4. CAD detection may be promising for routine batch-to-batch comparability 5. Prefer MS/MS for structural confirmation (vs. accurate mass) 6. Permethylation and MSn are highly useful for detailed or de novo structure analysis
Acknowledgements: We would like to thank the entire CRL Woburn Mass Spectrometry Team: Kurt Morgenstern, Le Meng, Art Cansizoglu, and Michael Chi; as well as St. John Skilton, Rose Lawler and Ilker Sen from PMI.