Simple Approach for Improved LC–MS Analysis of Protein

Jan 25, 2019 - LC–MS based analysis of protein biopharmaceuticals could benefit from improved data quality, which can subsequently lead to improved ...
0 downloads 0 Views 637KB Size
Subscriber access provided by UNIV OF LOUISIANA

Article

A Simple Approach for Improved LC-MS Analysis of Protein Biopharmaceuticals via Modification of Desolvation Gas Shunhai Wang, Tao Xing, Anita P Liu, Zehong He, Yuetian Yan, Thomas J Daly, and Ning Li Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.8b05846 • Publication Date (Web): 25 Jan 2019 Downloaded from http://pubs.acs.org on January 27, 2019

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 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 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.

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 7 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

A Simple Approach for Improved LC-MS Analysis of Protein Biopharmaceuticals via Modification of Desolvation Gas Shunhai Wang,*† Tao Xing,† Anita P. Liu, Zehong He, Yuetian Yan, Thomas J. Daly, and Ning Li Analytical Chemistry Group, Regeneron Pharmaceuticals Inc., 777 Old Saw Mill River Road, Tarrytown, New York 105916707, United States. ABSTRACT: LC-MS based analysis of protein biopharmaceuticals could benefit from improved data quality, which can subsequently lead to improved drug characterization with higher confidence and less ambiguity. In this study, we created a simple device to modify the desolvation gas on a Q-Exactive mass spectrometer and to demonstrate the utility in improving both peptide mapping analysis and intact mass analysis, the two most routinely and widely applied LC-MS techniques in protein biopharmaceutical characterization. By modifying the desolvation gas with acid vapor from propionic acid (PA) and isopropanol (IPA), the ion suppression effects from trifluoroacetic acid (TFA) in a typical peptide mapping method can be effectively mitigated, thus leading to improved MS sensitivity. By modifying the desolvation gas with base vapor from triethylamine (TEA), the charge reduction effect can be achieved and utilized to improve the spectral quality from intact mass analysis of protein biopharmaceuticals. The approach and device described in this work suggests a low-cost and practical solution to improve the LC-MS characterization of protein biopharmaceuticals, which has the potential to be widely implemented in biopharmaceutical analytical labs.

Liquid chromatography coupled to mass spectrometry (LCMS) technique has already become an indispensable tool for indepth characterization of protein biopharmaceuticals in analytical labs to support their development and license applications.1 Based on a recent retrospective evaluation of the use of mass spectrometry (MS) in Food and Drug Administration (FDA) biological license applications (BLAs), an overwhelming majority of those applications have implemented MS or LC-MS based workflows.2 To provide characterization of different protein attributes, a wide variety of LC-MS based assays can be performed, in which both peptide mapping analysis and intact mass analysis are most routinely and widely applied. To improve the confidence of the analysis and reduce the ambiguity associated with data interpretation, continual efforts have been made to improve the data quality from the LC-MS analysis, including using optimized experimental procedures, fine-tuned instrument parameters as well as more advanced mass spectrometers. Here, we present a novel approach to improve the LC-MS data quality from both peptide mapping analysis and intact mass analysis of protein biopharmaceuticals by modifying the desolvation gas using a simple device. LC-MS based peptide mapping analysis is routinely applied to confirm the primary structure of protein biopharmaceuticals, in which a protein molecule is first hydrolyzed into small peptide fragments using a protease with known specificity (although a non-specific protease can also be applied), then the amino acid sequence of each peptide fragment is determined by LC-MS/MS analysis, taking into consideration the cDNA predicted sequence and the protease specificity.3-5 Data from peptide mapping analysis can also be utilized to identify and quantify post-translational modifications, confirm the disulfide bond linkages and even detect amino acid substitution events

present at very low levels (< 0.1%).6 During peptide mapping analysis of protein biopharmaceuticals, LC-MS is often performed in combination with ultraviolet (UV) detection to generate so-called UV fingerprints, which alone can be used as an identification assay during quality control (QC) and drug release. To effectively separate peptides on a reversed-phase column with good peak shape, trifluoroacetic acid (TFA) is commonly used as a mobile phase modifier due to its excellent ion pairing ability.7 However, TFA is also known for its ion suppression effects during the electrospray (ESI) process due to the increased surface tension as well as its ability to form ion pairs with analytes in the gas phase, thus leading to significantly decreased MS sensitivity.8 Over the past two decades, different strategies have been investigated and implemented to alleviate the decrease in MS sensitivity due to TFA. For example, modifying the TFA-containing mobile phases with acetic acid or propionic acid has demonstrated a significant MS signal enhancement without compromising chromatographic integrity during bioanalysis of some basic compounds.9 Post-column addition of a mixture of propionic acid and isopropanol is another commonly used strategy that does not require the modification of the LC method. However, the experimental setup does require additional pumps, consumes large quantities of reagents, and is not suitable for continuous analysis of large sample sets.10 Acid vapor assisted ESI within an enclosed spray chamber is another elegant solution to counteract the signal suppression effects of TFA.11 However, its utility in protein biopharmaceutical characterization has been limited so far, presumably due to the requirement of a special ESI source. Finally, advances in reversed-phase column chemistry, particularly in the development of charged-surface C18 stationary phases, significantly reduced the dependence of using TFA to achieve good peak shape during peptide mapping

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

analysis.12 Replacing TFA with a MS-friendly mobile phase modifier (e.g. formic acid), however, will inevitably reduce the retention of most peptides, rendering some short and hydrophilic peptides undetectable due to co-elution with the solvent front, resulting in decreased sequence coverage. Nevertheless, until those new columns have been routinely and widely adopted in protein biopharmaceutical characterization, TFA-based LC-MS methods are still the standard in peptide mapping analysis. In this study, we describe a novel approach to counteract TFA ion suppression during LC-MS analysis by modifying the desolvation gas with acid vapor from propionic acid and isopropanol. The modification was conveniently implemented using a device assembled from all commercially available parts. Unlike previous reports in which the acid vaporassisted ESI setups were often realized using nano- or capillaryflow,11,13-15 the developed device overcomes these limitations and can be easily integrated into a standard Ion Max ion source (Thermo Scientific) for routine peptide mapping analysis using analytical flow. We have also demonstrated that the same device can be utilized to modify the desolvation gas with weak base, in order to improve the LC-MS based intact mass analysis of protein biopharmaceuticals via charge reduction. Charge reduction strategies have already been widely exploited in many intact mass applications by either simplifying the MS spectra of complex protein samples or by better preserving non-covalent interactions during native MS experiments.16-19 During evaluation of the developed device, we found that charge reduction strategies can also be utilized to stabilize highly charged protein analytes during the MS analysis and prevent them from undesired fragmentation and decay, thus leading to overall improved spectral quality. We also demonstrated the utility of this approach by successfully deciphering a highly heterogeneous protein molecule that cannot be resolved by the standard method. Again, unlike many previous reports where charge reduction was achieved by either modifying the mobile phases or post-column addition of the weak bases,16,19 the new device enables the charge reduction via the desolvation gas, thus eliminating concerns over chromatographic performance as well as retaining the practicality for routine analysis. EXPERIMENTAL SECTION Materials. Deionized water was provided by a Milli-Q integral water purification system installed with a MilliPak Express 20 filter (Millipore Sigma, Burlington, MA). NIST Monoclonal Antibody Reference Material 8671 (NISTmAb, humanized IgG1κ monoclonal antibody) was purchased from National Institute of Standards and Technology (Gaithersburg, MD). Protein-X is a highly glycosylated protein recombinantly produced at Regeneron (Tarrytown, NY). Rapid Peptide Nglycosidase F (Rapid PNGase F) with 5× Rapid PNGase F buffer was purchased from New England Biolabs Inc. (Ipswich, MA). TCEP-HCl (tris(2-carboxyethyl) phosphine hydrochloride), Tris-HCl pH 7.5 (UltraPure), trifluoroacetic acid (LC-MS grade), formic acid (LC-MS grade), and acetonitrile (Optima LC/MS grade) were purchased from Thermo Fisher Scientific (Waltham, MA), trypsin (sequencing grade) was purchased from Promega (Madison, MI). Triethylamine, iodoacetamide, propionic acid and acetic acid were purchased from Sigma Aldrich, Co. (St. Louis, MO). 2propanol (HPLC grade) was purchased from VWR International, LLC (Radnor, PA).

Sample Preparation. To prepare the tryptic digests for peptide mapping analysis, NISTmAb stock sample (100 µg) was diluted into 5 mM acetic acid and reduced with 5 mM TCEP-HCl at 80 °C for 10 min. After adjusting the pH to 7.5 using 1 M Tris-HCl (pH 7.5), 10 mM iodoacetamide and 5 µg of trypsin (E/S ratio at 1:20) were added, and the sample was incubated at 37 °C for 3 hours. Finally, the solution was acidified by 1% TFA to quench the digestion. To reduce NISTmAb for intact mass analysis, 10 mM TCEP-HCl was added to diluted NISTmAb solution (1 µg/µL), and the sample was incubated at 50°C for 30 minutes. To remove the N-glycans present on reagent Protein-X, the protein sample was first diluted to 0.25 µg/µL using 5×Rapid PNGase F buffer (containing reducing agent), and the solution was then incubated at 80 °C for 10 min. Subsequently, Rapid PNGase F was added to the solution, and the sample was incubated at 50°C for 30 minutes. Modification of the Desolvation Gas. The sheath gas flow from a Q-Exactive mass spectrometer was redirected to a Duran pressure plus bottle (SCHOTT North America, Inc., Elmsford, NY) through a Canary-Safe Cap (Analytical Sales and Services, Inc., Flanders, NJ) using 1/8” TEFLON tubing (Figure 1). The outgoing tubing from the bottle was then connected back to the HESI-II probe in a Thermo Scientific Ion Max ion source. For peptide mapping analysis, 150 mL of propionic acid (PA) and 50 mL of isopropanol (IPA) were mixed and transferred into the bottle. For intact mass analysis, 1% triethylamine (TEA) (v/v) in 200 mL of acetonitrile was transferred into the bottle. The bottle, containing concentrated acid or base, was then placed into a polyethylene secondary container (BEL-ART acid/solvent bottle carrier, Wayne, NJ) with a 16mm opening in the top for insertion of tubing. To disable the modification, the sheath gas tubing was directly connected to the HESI-II probe without passing through the device. Photographs of the actual device and the device in connection to a Q-Exactive mass spectrometer are shown in Figure S1 and Figure S2 (Supporting Information). Safety Considerations. For desolvation gas modified experiments, the sheath gas was set to 15 arbitrary units. A higher setting might be possible, but the pressure within the bottle should be measured and made sure not to exceed the pressure rating of the bottle. A pressure resistant bottle (e.g. Duran pressure plus bottle, pressure rating: +1.5 bar) is highly recommended for this application. A secondary container was also required for this setup to prevent possible acid or base spills. LC-MS Analysis. For peptide mapping analysis of NISTmAb, aliquots (2 µg) of digests were separated using an ACQUITY UPLC Peptide BEH C18 Column (130Å, 1.7 µm, 2.1 mm x 150 mm) (Waters, Milford, MA) for online LCMS/MS analysis on a Q-Exactive mass spectrometer. For formic acid (FA)-based analysis, mobile phase A was 0.1% FA (v/v) in water, and mobile phase B was 0.1% FA in acetonitrile (ACN). For TFA-based analysis, mobile phase A was 0.05% TFA (v/v) in water, and mobile phase B was 0.045% TFA (v/v) in ACN. Detailed LC gradient and MS parameters were included in the Supporting Information. For intact mass analysis of reduced NISTmAb and reagent Protein-X. Aliquots (4 µg) of each sample were separated using a BioResolve RP mAb Polyphenyl column (50 mm x 2.1 mm, 2.7 µm) (Waters, Milford, MA) for online LC-MS analysis on a Q-Exactive mass

ACS Paragon Plus Environment

Page 2 of 7

Page 3 of 7 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 spectrometer. Detailed LC gradient and MS parameters were included in the Supporting Information. RESULTS AND DISCUSSION Modification of the desolvation gas was demonstrated on a Q-Exactive mass spectrometer equipped with a standard Ion Max ion source. The accessibility of both sheath gas and auxiliary gas from this instrument made implementation of the modification straightforward. The device can be conveniently enabled or disabled within 1 minute without any need for system equilibration. Although it should be feasible to implement similar modifications to other mass spectrometers from different vendors, more extensive modifications might be required.

Figure 1. Representation of the device for desolvation gas modification on a Q-Exactive mass spectrometer.

To improve the peptide mapping analysis by counteracting the TFA ion suppression, a mixture of propionic acid (PA) and isopropanol (IPA) at a 3 to 1 ratio was transferred into the device to deliver the acid vapor into the sheath gas. This recipe was directly adopted from a previous study, in which Apffel et al. used a factorial experimental approach and systematically evaluated the effect of different weak acids and organic solvents for the “TFA Fix”.10 Although PA and IPA were delivered into the LC flow via a post-column addition in that study, the proposed mechanism of action and conclusion should be generally applicable. Briefly, in the presence of PA, the deprotonated TFA anions, which are the main culprits for ion suppression by forming ion pairs with analytes, can be reprotonated and removed from the ESI process through evaporation due to the relatively high volatility. As a weak acid, PA does not exhibit a strong ion pairing effect with analytes, thus facilitating the ionization process. IPA mainly serves as a carrier for the acid vapor, but might also contribute to the reduced surface tension of the ESI droplets, thus leading to better desolvation and ionization.20 To evaluate the MS sensitivity improvement as a result of desolvation gas modification, LC-MS based peptide mapping analysis was performed using the tryptic digests of NISTmAb as a testing standard. Both FA- and TFA-containing mobile phases were tested with the desolvation gas modification either enabled or disabled for the TFA-based analysis. Figure 2 shows the base peak chromatograms (BPCs) from the LC-MS analysis of the tryptic digests of NISTmAb using a BEH C18 column. It is evident that the separation and retention of tryptic peptides

on the C18 column were significantly different when FA or TFA were used as the mobile phase modifier. In general, TFA outperforms FA from the chromatographic performance perspective, exhibiting better retention of hydrophilic peptides, better peak shapes and overall higher peak capacity. For example, a small tryptic peptide, EYK (P1, Figure 2), was not retained on the C18 column when FA was used, whereas this peak was retained on the same column when TFA was applied. This feature might improve the sequence coverage of protein biopharmaceuticals in the peptide mapping analysis. On the other hand, TFA-based analysis exhibited a significant loss in MS sensitivity compared to the FA-based analysis. Subsequently, modification of the desolvation gas with acid vapor from PA and IPA led to a significant increase in MS sensitivity for TFA-based analysis.

Figure 2. BPCs from the LC-MS analysis of NISTmAb digests using the FA-based method (top panel), the TFA-based method (middle panel) and the TFA-based method with PA/IPA modified desolvation gas (bottom panel). The MS signal from the TFA control experiment was amplified two times for better visualization. P1 to P6 were six representative tryptic peptides from NISTmAb.

As demonstrated by six representative tryptic peptides (P1 to P6) of different size and retention time (Table S1, Supporting Information), a 4.9-5.8-fold increase in MS sensitivity was readily achieved by this simple approach, regardless of the peptide size and mobile phase composition. Comparable background noise was also observed before and after the desolvation gas modification, which subsequently led to the overall improved signal to noise ratio (Figure S3, Supporting Information). In addition, a subtle increase in charge states was also consistently observed for many tryptic peptides after the desolvation gas modification (Table S1, Supporting Information). This feature might lead to improved spectral quality of the tandem MS due to more efficient fragmentation of higher charged species, resulting in increased identification confidence. Overall, by modifying the desolvation gas with acid vapor using the developed device, a significantly more sensitive peptide mapping method can be readily achieved. The increase in MS sensitivity is of great value to characterize lowabundance attributes present in protein biopharmaceuticals, such as PTMs and sequence variants. As previously discussed, although post-column addition of a weak acid is an effective and straightforward approach to

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

counteract TFA ion suppression, it is not suitable for continuous analysis of large sample sets and consumes a large quantity of reagents. To evaluate if the new approach can be applied to continuous analysis and if the improved sensitivity can be maintained over a long period of time, peptide mapping analysis of NISTmAb digests were repeatedly performed over two consecutive days. Again, after enabling the desolvation gas modification with acid vapor, the increase in MS sensitivity, compared to the control method, was immediately achieved and maintained for at least 37 consecutive runs (~ 3800 min) as tested (Figure 3). The consumption of the PA and IPA mixture, under the applied experimental conditions, was estimated to be 1 mL per hour. As a result, the developed approach is considered highly suitable for routine peptide mapping analysis of protein biopharmaceuticals, in which continuous analysis of large sample sets might be required.

attributed to the decay of the highly charged heavy chain species during the MS analysis, such as neutral losses of water (-18 Da), ammonia (-17 Da) and carboxylic acid (-44 Da). In contrast, the zoom-in view of charge state +21 from the modified method exhibited a greatly improved signal to noise ratio, thus allowing the identification of several low-abundance glycoforms that were not detected in the control method. The deconvoluted mass spectrum (Figure S4, Supporting Information) further demonstrated that the improved method could confidently identify glycoforms present at levels as low as 0.3% on the heavy chain of NISTmAb.

Figure 3. MS intensities of six representative tryptic peptides from NISTmAb digests using control method (run 1-2) and desolvation gas modified method (run 3-39). P1 to P6 were six representative tryptic peptides from NISTmAb. The relative standard deviation (RSD) was calculated based on the MS intensities of each peptide from 37 modified runs.

Besides peptide mapping analysis, the same device can also be utilized to improve the intact mass analysis of protein biopharmaceuticals via charge reduction, which is achieved by modifying the desolvation gas with weak base. By shifting the charge state envelope towards a high m/z region, the MS spectrum can be greatly simplified due to the increased spacing between two adjacent charge states. In a recent study, Ding et al. have found the intact mass analysis of reduced monoclonal antibodies can be greatly improved by adding a trace amount of base additive into mobile phases.16 To evaluate if similar improvements can be achieved via desolvation gas modification, reduced NISTmAb was used as a test article. Triethylamine (TEA), a frequently used charge reducing reagent, was diluted to 1% (v/v) using acetonitrile (ACN) and used to deliver the base vapor into the desolvation gas. After chromatographic separation on a reversed-phase (RP) polyphenol column, the MS spectra of the reduced heavy chain of the NISTmAb acquired from both a control method and from the desolvation gas modified method are shown in Figure 4. Following desolvation gas modification, a significant charge reduction on the NISTmAb heavy chain, from charge states +23 to +55 to charge states +14 to +28, was achieved. Close examination of charge state +39 from the control method revealed a high level of background noise, which was likely

Figure 4. Mass spectra of reduced heavy chain of NISTmAb from (a) the control method and (b) charge reduction method. F: Fucose; G, Galactose; Gc: N-Glycolylneuraminic acid; M: Mannose; Hex: Hexose; A1: biantennary with only one GlcNAc; A2: biantennary with both GlcNAcs.

The dramatic improvement in spectral quality is not only attributed to the highly simplified spectrum resulting from less crowded charge state envelopes, but more importantly is a result of stabilized protein analytes. This is consistent with the wellestablished knowledge that the collision energy associated with a protein ion during MS analysis is proportional to its charge state. Specifically, lower charge states decay slower than the higher charge states of the same analyte during the analysis within the orbitrap. Notably, efforts to mitigate the extensive decay of highly charged heavy chain species from the control method were not successful, even after completely removing the source-induced dissociation (SID) energy. On the contrary, high levels of SID energy (75 eV used in this experiment) can be readily tolerated from the charge reduction method without

ACS Paragon Plus Environment

Page 4 of 7

Page 5 of 7 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 noticeable decay or fragmentation of the heavy chain, which further improves the spectral quality via more efficient desolvation and adduct removal. Similar improvement in spectral quality was also achieved for the reduced light chain of the NISTmAb (Figure S5, Supporting Information). It is worth noting that the application of a charge reduction strategy to improve MS spectral quality is most helpful for fully denatured and reduced proteins, as they are often highly charged (due to increased protein surface areas21) during ESI-MS analysis and are most susceptible to undesired fragmentation and decay. The developed charge reduction method described here, achieved by modifying the desolvation gas with TEA, can also be utilized to tackle the high mass heterogeneity present in complex protein samples. These complex samples might include various protein reagents that are critically important to support different assays during the development of protein biopharmaceuticals. Using intact mass analysis to confirm the identities of those protein reagents, no matter if they are obtained from commercial sources or produced in-house, is frequently required to support the downstream studies supporting clinical development activities. Some protein reagents are highly heterogeneous in molecular weight due to extensive glycosylation, which can present significant challenges for the standard intact mass method. To demonstrate the utility of the developed charge reduction method, a complex reagent protein (Protein-X) with high mass heterogeneity was used as a test article. Prior to the analysis, Protein-X was first treated with PNGase F under both denaturing and reducing conditions, in order to remove the mass heterogeneity introduced by the presence of N-glycans. As shown in Figure 5, even after the PNGase F treatment, Protein-X exhibited a highly convoluted MS spectrum that cannot be deciphered using a regular intact mass method. This was likely attributed to the overlapping charge envelopes from different co-eluting mass forms of this molecule at low m/z region. In contrast, after enabling the desolvation gas modification with TEA, the MS signal of the same sample was immediately shifted to the high m/z region and exhibited much better resolved charge states. The deconvoluted spectrum (Figure 5b) indicated that this protein was extensively modified by O-glycans with possibly 10 different glycosylation sites and 4 different O-glycan forms. Overall, nearly 35 different mass species from this protein were confidently identified and assigned. It is worth noting that the improved spectral quality might also be partly attributed to stabilized protein ions after charge reduction, as the polysaccharide moiety on this molecule could be labile under regular ESI-MS conditions. It is also worth pointing out that the developed approach might be particularly applicable for the characterization of PEGylated protein products, as they typically exhibit a high degree of mass heterogeneity. Finally, for some other LC-MS based intact mass methods, where the use of TFA is inevitable to ensure chromatographic

performance, the developed method could also be tested to counteract TFA ion suppression using PA/IPA modified desolvation gas. For example, hyphenation of size exclusion chromatography (SEC) to MS using mobile phases containing ACN, TFA, and formic acid has been used for reduced mAb analysis.22 Hydrophilic interaction chromatography coupled to MS using TFA-containing mobile phases has been used to study the low molecular weight impurities in mAb samples.23 Application of PA/IPA modified desolvation gas in both methods led to significant improvement in MS sensitivity of mAb fragments (e.g. heavy chain, light chain and smaller fragments), but not the intact mAb (~ 150 kDa) (data not shown). This observation is consistent with a previously published study.10 It is thought that a large protein might accommodate a large number of TFA anions, which cannot be effectively replaced by PA during the ESI process. CONCLUSIONS Improved LC-MS based characterization of protein biopharmaceuticals is of great value to better support the drug development process by providing data with higher confidence and less ambiguity. We demonstrated a novel approach to improve the data quality from both peptide mapping analysis and intact mass analysis via desolvation gas modification using a simple but effective device. By using PA/IPA modified desolvation gas, the TFA ion suppression from a typical peptide mapping method can be effectively mitigated, leading to improved MS sensitivity. We also showed that the developed approach can be easily implemented without changing the LC method and is capable of continuous analysis of large sample sets, making it particularly suitable for routine characterization of protein biopharmaceuticals. By using TEA modified desolvation gas, the new approach could also be utilized to improve the intact mass analysis of proteins via charge reduction. Significant improvement in spectral quality not only allows the detection of minor mass forms otherwise buried in noise, but also enables the mass measurement of highly heterogeneous proteins. Notably, the modification was easily implemented via a simple device that was put together with all commercially available parts at minimal cost and could be enabled or disabled with little effort. Although we have only demonstrated the approach on Q-Exactive series instruments, similar modifications might also be feasible on other instruments. In addition, the versatility of this device might be further exploited for other MS-based applications by changing the ESI conditions. Finally, with the ever-increasing role played by LC-MS methods in protein biopharmaceutical characterization, the developed approach can make a deeper and broader contribution by serving as a low-cost and practical solution to improve the analytical capability and better support drug development.

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 7

Figure 5. LC-MS analysis of a highly heterogeneous protein with multiple O-glycans. (a) raw mass spectra acquired using control method (top) and charge reduction method (bottom). (b) deconvoluted mass spectrum from the charge reduction method.

Notes The authors declare the following competing financial interest(s): S.W., T.X., A.P.L.Z.H., Y.Y., T.J.D., and N.L. are full-time employees and shareholders of Regeneron Pharmaceuticals Inc.

ASSOCIATED CONTENT Supporting Information The Supporting Information is available free of charge on the ACS Publications website. Additional information as noted in the text (PDF)

AUTHOR INFORMATION Corresponding Author *E-mail: [email protected].

Author Contributions †S.W. and T.X. are co-first authors.

ACKNOWLEDGMENT This study was sponsored by Regeneron Pharmaceuticals Inc. The authors would like to thank Ashley Roberts of Scientific Writing Group for her assistance in editing the manuscript.

REFERENCES (1) Kaltashov, I. A.; Bobst, C. E.; Abzalimov, R. R.; Wang, G.; Baykal, B.; Wang, S. Biotechnol Adv 2012, 30, 210-222. (2) Rogstad, S.; Faustino, A.; Ruth, A.; Keire, D.; Boyne, M.; Park, J. J Am Soc Mass Spectrom 2017, 28, 786-794. (3) Dick, L. W., Jr.; Mahon, D.; Qiu, D.; Cheng, K. C. J Chromatogr B Analyt Technol Biomed Life Sci 2009, 877, 230-236. (4) Bongers, J.; Cummings, J. J.; Ebert, M. B.; Federici, M. M.; Gledhill, L.; Gulati, D.; Hilliard, G. M.; Jones, B. H.; Lee, K. R.; Mozdzanowski, J.; Naimoli, M.; Burman, S. J Pharm Biomed Anal 2000, 21, 1099-1128.

ACS Paragon Plus Environment

6

Page 7 of 7 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 (5) Mouchahoir, T.; Schiel, J. E. Anal Bioanal Chem 2018, 410, 2111-2126. (6) Zeck, A.; Regula, J. T.; Larraillet, V.; Mautz, B.; Popp, O.; Gopfert, U.; Wiegeshoff, F.; Vollertsen, U. E.; Gorr, I. H.; Koll, H.; Papadimitriou, A. PLoS One 2012, 7, e40328. (7) Shibue, M.; Mant, C. T.; Hodges, R. S. J Chromatogr A 2005, 1080, 68-75. (8) Annesley, T. M. Clin Chem 2003, 49, 1041-1044. (9) Shou, W. Z.; Naidong, W. J Chromatogr B Analyt Technol Biomed Life Sci 2005, 825, 186-192. (10) Apffel, A.; Fischer, S.; Goldberg, G.; Goodley, P. C.; Kuhlmann, F. E. J Chromatogr A 1995, 712, 177-190. (11) Chen, J.; Liu, Z.; Wang, F.; Mao, J.; Zhou, Y.; Liu, J.; Zou, H.; Zhang, Y. Chem Commun (Camb) 2015, 51, 14758-14760. (12) Lauber, M. A.; Koza, S. M.; McCall, S. A.; Alden, B. A.; Iraneta, P. C.; Fountain, K. J. Anal Chem 2013, 85, 6936-6944. (13) Kharlamova, A.; Prentice, B. M.; Huang, T. Y.; McLuckey, S. A. Anal Chem 2010, 82, 7422-7429.

(14) Kharlamova, A.; McLuckey, S. A. Anal Chem 2011, 83, 431437. (15) Zhu, Y.; Nugent, K. D. US 8227750 B1, July 24, 2012. (16) Ding, W.; Qiu, D.; Bolgar, M. S.; Miller, S. A. Anal Chem 2018, 90, 1560-1565. (17) Pacholarz, K. J.; Barran, P. E. EuPA Open Proteom 2016, 11, 23-27. (18) Hopper, J. T.; Sokratous, K.; Oldham, N. J. Anal Biochem 2012, 421, 788-790. (19) Huang, L.; Gough, P. C.; Defelippis, M. R. Anal Chem 2009, 81, 567-577. (20) Iavarone, A. T.; Jurchen, J. C.; Williams, E. R. J Am Soc Mass Spectrom 2000, 11, 976-985. (21) Kaltashov, I. A.; Mohimen, A. Anal Chem 2005, 77, 5370-5379. (22) Liu, H.; Gaza-Bulseco, G.; Chumsae, C. J Am Soc Mass Spectrom 2009, 20, 2258-2264. (23) Wang, S.; Liu, A. P.; Yan, Y.; Daly, T. J.; Li, N. J Pharm Biomed Anal 2018, 154, 468-475.

Insert Table of Contents artwork here

7 ACS Paragon Plus Environment