Subscriber access provided by Gothenburg University Library
Article
High-Resolution CZE-MS Peptide Mapping of Therapeutic Proteins: Improved Separation with Mixed Aqueous Aprotic Dipolar Solvents (N,N–Dimethylacetamide and N,N–Dimethylformamide) as the Background Electrolyte Oluwatosin O. Dada, Yimeng Zhao, Nomalie Jaya, and Oscar Salas-Solano Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.7b03405 • Publication Date (Web): 03 Oct 2017 Downloaded from http://pubs.acs.org on October 7, 2017
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 18
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
High-Resolution CZE-MS Peptide Mapping of Therapeutic Proteins: Improved Separation with Mixed Aqueous - Aprotic Dipolar Solvents (N,N– Dimethylacetamide and N,N–Dimethylformamide) as the Background Electrolyte. Oluwatosin O. Dada!, Yimeng Zhao#, Nomalie Jaya!, and Oscar Salas-Solano! Department of Analytical Sciences, Seattle Genetics, Inc. 21823 30th Drive SE, Bothell, WA 98021, USA
Abstract Peptide mapping with mass spectrometry (MS) detection is a powerful technique routinely used for interrogating physicochemical properties of proteins. Peptide mapping benefits from an efficient front-end separation to increase selectivity and reduce complexity prior to MS detection. The most commonly used method for peptide mapping is based on reverse phase liquid chromatography with mass spectrometry (RPLC-MS). Capillary zone electrophoresis with mass spectrometry (CZE-MS) is an orthogonal technique with growing attention for peptide mapping of biotherapeutic proteins due to its high efficiency and sensitivity. However, that growth has been slow due to poorer peptide resolution and method robustness compared to RPLC. Here we present results from optimization of CZE-MS peptide mapping separation using mixed aqueous - aprotic dipolar solvent (N,N–dimethylacetamide (DMA) and N,N–dimethylformamide (DMF), as the background electrolyte (BGE) to improve the separation performance. Addition of DMA or DMF to the BGE impacts separation selectivity through differential change in pKa of the peptides. The CZE-MS peptide mapping method with the modified BGE produced significant improvement in resolution over the conventional CZE-MS methods. The method was evaluated with both sheathless and sheathflow CE-MS ion sources.
!
Department of Analytical Sciences, Seattle Genetics, Inc. 21823 30th Drive SE, Bothell, WA
98021, USA #
Current Address: Weil Cornell Medicine, 1300 York Avenue, NY 10065, USA
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 18
1. INTRODUCTION Monoclonal antibodies (mAb) are the backbone of many biotherapeutic drugs. In the drug development process, from clone selection to release and stability, physico-chemical characterization of the antibody is crucial for in-depth understanding of its structural integrity. Peptide mapping with tandem mass spectrometry (MS/MS), where peptide fragments generated from proteolytic digestion of proteins are separated and analyzed by mass spectrometry (MS), is routinely used for interrogating various characteristics of mAbs including amino acid composition, sequence variants, and post-translational modifications (PTMs), which provide information about the structural components of the antibody that are important for safety and efficacy. Peptide mapping requires an efficient front-end peptide separation to increase selectivity and decrease complexity prior to MS detection. Reverse phase liquid chromatography (RPLC) is a commonly used separation technique for peptide mapping, where peptides are retained and separated based on hydrophobicity on chemically derivatized stationary columns. RPLC-MS peptide mapping has been applied for different characterization purposes in biotherapeutic drug development from confirming cell line stability1 to structural characterization of the antibody.2,3 RPLC provides excellent qualitative and adequate quantitative assessment of the primary sequence, PTMs, and other biochemical properties. However, there are limitations concerning the recovery of small basic and highly hydrophobic peptides. Short basic peptides are poorly retained on the column, eluting in the void volume.4 Large hydrophobic peptides that are strongly retained on the column may be poorly recovered5 Capillary zone electrophoresis (CZE) is a powerful technique that is orthogonal to RPLC. In CZE, analytes are separated based on their charge-to-size ratio in a narrow bore silica capillary filled with a conductive background electrolyte (BGE).6 With recent advances in ion source interfaces, CZE-MS has received attention for several biotech applications such as peptide mapping,7 intact mass analysis,8-10 glycoanalysis,11,12 and quantitation of host cell proteins.13,14 Specifically, CZE-MS peptide mapping offers a means to overcome some of the RPLC-MS limitations indicated above. As an open tubular technique with no stationary phase, CZE-MS can allow complete sequence coverage with good recovery of peptides that tend to be difficult to analyze by RPLC-MS. Its utility for characterizing therapeutic proteins has been reported in several publications. Gennaro et al. published an early demonstration of CZE-MS peptide mapping for characterizing recombinant monoclonal antibodies.4 Other recent publications also
2 ACS Paragon Plus Environment
Page 3 of 18
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
showed its potentials for PTM analysis,5,15 glycopeptides,11,16,17 and structural characterization.7,18 Even though the application of CZE-MS peptide mapping is growing, that growth has been slow arguably due to poorer peptide resolution compared to RPLC. A high-resolution CZE-MS peptide separation is desirable, and most importantly needed as an orthogonal approach to RPLC-MS for biotech applications. Until now, reported separations for CZE-MS peptide mapping have been sub-optimal with many peptides not well-resolved.7,19 Factors that may be contributing to the poor resolution by CZE-MS include the separation speed and in adequate selectivity. Attempts to address this drawback have focused mainly on using neutral coated capillary.20-22 Recently, Chen et al.,23 demonstrated an improvement in peak capacity of CZE-MS separation using a high sample loading on a neutral coated capillary but the separation is still less optimal in comparison with modern RPLC-MS separations. In principle, separation on neutral coated capillary, where electroosmotic flow (EOF) is minimized, can improve peptide resolution. Nevertheless, success with coated capillaries is highly dependent on robust and uniform capillary surface modification, which may be difficult to achieve. In cases where a coated capillary was successfully used, the resolution was still relatively sub-optimal in comparison to RPLC-MS.19,23 Another approach for improving CZE-MS selectivity is to selectively modify the BGE. In principle, the effective mobility of the analyte can be affected by its pKa and the pH of the separation buffer. At a constant pH, the analyte’s pKa can be influenced by using mixtures of aqueous and organic solvents as the BGE. Isopropanol (IPA)24, and acetonitrile24,25 have been previously used in CZE-MS peptide mapping but only with marginal improvement in resolution. In the present work, we explore the potentials of N,N-dimethylacetamide (DMA), and N,N– dimethylformamide (DMF) as a component of the BGE in CZE-MS peptide mapping to improve separation selectivity. DMA and DMF are aprotic dipolar organic solvents which have been used in electrophoretic separation of organic ions26,27, and chromatographic separation of proteins28 but not yet applied to CZE-MS peptide mapping. The interest in these solvents stems from a series of previous publications that described their unique effects on the pKa of organic acids.2933
Kenndler et al. investigated the impact of DMA and DMF on the selectivity of non-aqueous
capillary electrophoresis (NACE) for benzoic acid and its m-substituted derivatives.29 It was shown that DMA and DMF exhibit unique stabilizing effects on the protolytic equilibria of the investigated set of acids, causing a differential change in the pKa values of the acids. These behaviors were attributed to the medium effect of DMA and DMF on protolytic equilibria of the acids. Medium effect is proportional to the reversible work of transferring a mole of the ionizing 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 18
species of a molecular or ionic weak acid from infinite dilution in water to infinite dilution in another solvent.29,34 In this work, our hypothesis is that DMA and DMF should impact the protolytic equilibria of peptides in a fashion similar to organic acids, and therefore improve peptide’s selectivity based on different side groups of their amino acids. This assumption was proven by evaluating the resolving potential of a binary mixed aqueous-organic BGEs consisting of DMA or DMF as an organic modifier for peptide mapping of a monoclonal antibody (mAb1). Optimization of the organic modified BGE was performed with DMA on both bare fused silica (BFS) capillary and neutral linear polyacrylamide (LPA) coated capillary. We used 10% acetic acid (HAc) as the reference BGE15,18 and the baseline comparator for the optimization. The separation with aqueous-organic BGE showed significantly higher resolution than the 10% HAc reference BGE and is at par with chromatographic separation but with added improvement for the recovery of small hydrophilic and large hydrophobic peptides as well as higher sensitivity.
2. EXPERIMENTAL SECTION i.
Materials and Reagents
All reagents were purchased from Sigma Aldrich (St. Louis, MO) unless stated otherwise. Recombinant MAb1 mAb (mAb1) was expressed in Chinese Hamster Ovary (CHO) cells and purified according to established practices.35 3 kDa MWCO filter was purchased from Millipore (Billerica, MA). HPLC grade water, acetonitrile, 2-propanol, and acetic acid were purchased from Fisher Scientific, (Waltham, MA). Lys-C (Lysyl Endopeptidase) was purchased from Wako (Richmond, VA). Both BFS and neutral LPA coated capillary cartridges were purchased from Sciex (Framingham, MA). ii.
Lys-C Peptide Mapping Sample Preparation
Sample digestion with lysyl endopeptidase C (Lys-C) was performed in 50 mM ammonium bicarbonate, pH 8. The details of the sample preparation protocol is presented in the supporting information. iii.
CZE-MS/MS Analysis
The conditions for CZE-MS analysis vary based on the type of ESI ion source used. For sheathless ESI ion source, CESI 8000 (Sciex, Framingham, MA) system was coupled to a 4 ACS Paragon Plus Environment
Page 5 of 18
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
Velos Pro mass spectrometer (Thermo Fisher, Waltham, MA). With this setup, separations were performed with 91 cm long and 30 µm ID Sciex OptiMS Silica Surface and Neutral OptiMS cartridges. The separation voltages were +25 kV and +30 kV for BFS and LPA coated capillaries, respectively. 10% acetic acid (HAc) in water was used as the reference BGE for optimization. Separation on the LPA-coated capillary was performed with pressure assistance to support the electrospray. The final mAb Lys-C digest was injected into the capillary hydrodynamically at 5 psi /60 s. The conductive liquid was maintained at 10% HAc without organic solvent for all experiments. Where formic acid is the aqueous component of the BGE, 1% formic acid is used as the conductive liquid. For sheathflow ESI ion source, CESI 8000 was coupled to a Velos Pro mass spectrometer (Thermo Fisher, Waltham, MA) with EMASS II CE-MS ion source (CMP Scientific, Brooklyn, NY). The separation capillary was a 95 cm long and 50 µm ID neutral LPA coated capillary etched at the outlet end (CMP Scientific, Brooklyn, NY). The ESI spray was delivered with a borosilicate glass emitter (0.75 mm ID, 5 cm long, and 30 µm tip) from CMP Scientific. The reference BGE was 10% acetic acid (HAc) and the ESI sheath liquid was 0.1% formic acid in 10% methanol. The separation was performed with 3 psi/ 7s injection, +30 kV separation voltage, and +2 kV ESI voltage. iv.
RPLC-MS/MS Analysis
Approximately 15 µg of digested samples were injected onto a BEH C18 column (2.1 x 150 mm, 1.7 µm, 130 Å) (Waters Technology, Milford, MA) heated to 55 °C. Separation was performed with an acetonitrile/trifluoroacetic acid (TFA) gradient on a Waters UPLC system where mobile phase A was composed of 0.1% TFA in water. Mobile phase B was composed of 0.08% TFA in acetonitrile. Peptides were eluted using a linear gradient to 33% B. After passing through the UV detector, the eluate was directed into a Velos Pro mass spectrometer (Thermo Fisher, Waltham, MA) fitted with an electrospray ionoization (ESI) source operating in positive ion mode. The capillary voltage was set at 4.5 kV. Desolvation was achieved at a temperature of 250°C and using sheath gas nitrogen at a flow rate of 10 µL/min. MS analysis was carried out on a Thermo Velos Pro mass spectrometer in positive mode at a resolution of 17,500 for MS and MS/MS scans. Data dependent MSMS spectral data were generated from MS survey scans acquired in the m/z range of 200 - 2000 and analyzed with Thermo Xcalibur version 2.2.
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 18
3. RESULTS AND DISCUSSIONS i.
Separation on Bare Fused Silica and Neutral LPA Coated Capillary with 10% Acetic Acid Reference BGE
Most traditional CZE BGEs such as borate, phosphate, and Tris are not suitable for mass spectrometry as they interfere with electrospray ionization. The most common CZE-MS BGEs are HAc15,18 and FA.8,36 In this report, we used 10% HAc as the reference and therefore the baseline comparator for further optimization. Figure 1 shows the Lys-C peptide map of mAb1 as analyzed by CZE-MS on both BFS and LPA coated capillaries using sheathless ESI interface with 10% HAc BGE. The average peak width at half maximum, w, and peak capacity, Cp, for each electropherogram are shown in Table S1. The average peak width was estimated from six reporter peptides based on the extracted ion electropherogram (EIE). EIE was used since absorbance detection was not available. Peak capacity was defined as (1+t)/(1.7w), where t is the separation window. The separation on BFS generated average peak width of 14s and peak capacity of 50. The separation on LPA coated capillary showed similar average peak width of 16s and small improvement in peak capacity to 66. The results resemble most previously reported electropherograms for CZE-MS.7,19,37 Though the separations are completed in less than 30 min, the resolution is less optimal compared to common RPLC separations.38 While the average peak widths for BFS and LPA may indicate good separation efficiency, the fast separation speed resulted in a narrow separation window (~ 15 min) and limited peak capacity with a large number of peptides co-migrating. The sub-optimal resolution in the CZE-MS separations is likely caused by insufficient selectivity within the separation time. To address this drawback, we used mixed aqueous-organic BGEs that improve peptide selectivity.
6 ACS Paragon Plus Environment
Page 7 of 18
BFS (OptiMS) Capillary P5
2 P3 P6 P4
1
Relative Intensity (108)
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
P2 P1
0 2 P3
LPA (Neutral OptiMS) Capillary
P4 P5
1 P2
P6
P1
0 10
12
14
16
18
20
22
24
26
28
30
Migration Time (min)
Figure 1: Base peak electropherogram of Lys-C peptides of mAb1 separated by CZE-MS with 10% acetic acid BGE, 25 kV separation voltage on BFS capillary (Top panel), and 10% acetic acid BGE, 30 kV separation voltage, 2.5 psi pressure on LPA coated capillary (Bottom panel). For all experiments: conductive liquid is 10% acetic acid, ESI voltage is 1.2 kV, and 5 psi / 60s (≈ 25 ng sample) injection. The arrows point to the migration region for the reporter peptides used to calculate average peak width on Table S-1.
ii.
Evaluation of DMA Modified BGE with Sheathless Ion Source
The potentials of DMA and DMF modified BGEs for high-resolution CZE-MS peptide mapping was evaluated. We first performed the BGE optimization with DMA and then applied the result to DMF. Initial comparison of BGEs containing 10% HAc and 10% organic solvent (DMA, DMSO, ACN and IPA) was carried out on a BFS capillary with porous sheathless ion source. A BGE with 20% HAc, no organic additives, was also included in the initial screening to determine whether higher HAc concentration could improve the separation. The conductive liquid was 10% HAc for all BGE types since there was no observable difference with or without organic solvent. The results of the initial BGE screening are shown in Figure S1 (Supporting Information). Compared to the reference BGE (10% HAc), the BGE with 20% HAc generated similar separation. All organic modified BGEs, except for DMA, showed insignificant improvement on the separation (Figure S1). The DMA modified BGE was distinct with improved resolution and the widest separation window, and was further investigated.
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 18
Further optimization of the DMA modified BGE was performed by adjusting the concentration of HAc and DMA. As shown in Figure S2, increasing HAc concentration up to 20% further improved the separation. No additional improvement was seen at HAc concentration higher than 20%. The separation with just 20% HAc as shown earlier in Figure S1 generated a result similar to 10% HAc, suggesting that the improvement in separation is due to the combination of HAc and DMA effects. Increasing the DMA concentration to 20% also yielded further improvement (Figure S2B, Supporting Information). However, increasing both HAc and DMA concentrations resulted in longer separation time. As a result, the final BGE composition of 20%HAc-15%DMA was chosen in order to minimize the separation time while maintaining selectivity. Figure 2 (Top Panel) shows the separation for Lys-C peptides from mAb1 on BFS with 20%HAc-15%DMA BGE. The optimized BGE produced average peak width of 15s, similar to 10% HAc on BFS capillary (Table S2), which suggests that DMA increases peptides selectivity without negative impact on efficiency. The peak capacity also increased significantly from 50 to 94 (Table S-2). The enhanced separation performance is believed to be due to the effect of DMA on the EOF and peptide mobility. Figure S3 shows that the new BGE has good migration time repeatability and linear MS response. A gradual small shift in migration time was observed and is probably due to evaporation of the BGE over time. The relative migration times to an internal reference peak are within
