Use of Amino Acids for Signal Enhancement in Mass Spectrometry

Abstract

A method for enhancing mass spectral signals is disclosed. The method includes contacting a sample with a separation column under conditions that allow sample components to bind to a support, applying a first mobile phase gradient to the separation column, the first mobile phase gradient comprising trifluoroacetic acid (TFA) and a small molecule additive (e.g., an amino acid) or formic acid (FA) and a small molecule additive (e.g., an amino acid), applying a second mobile phase gradient to the separation column, the second mobile phase gradient comprising TFA and a small molecule additive (e.g., an amino acid) in acetonitrile (ACN) or formic acid (FA) and a small molecule additive (e.g., an amino acid) in ACN, and performing mass spectrometry on the eluted sample components.

Description

Technical Field

[0001] Reference to Sequence Listing This application incorporates by reference a Sequence Listing submitted in computer-readable form as file 10675WO01-Sequence.txt, created on January 7, 2021, and containing 18,878 bytes.

[0002] The present invention relates to mass spectrometry and to methods for enhancing mass spectrometry signals by the use of amino acids or modified amino acids such as glycine.

Background Art

[0003] Liquid chromatography-mass spectrometry (LC-MS) is used for the characterization of biomolecules including peptide and protein therapeutics. LC-MS is a powerful technique for the characterization of recombinant proteins as well as post-translational and protein modifications (e.g., disulfide bonds, glycosylation and phosphorylation), but has been found to have insufficient separation and sensitivity. For example, glycosylation can be characterized at the peptide level by analyzing glycopeptides generated from tryptic digestion of antibodies. However, glycopeptides having heterogeneous glycoforms are often not well separated by reverse-phase based liquid chromatography (RPLC) conventionally used for peptide mapping. Further, in-source fragmentation of the sugar chains of glycopeptides induced by online mass spectrometry (MS) can occur, resulting in artifacts of cleaved glycoforms, which may impair the accurate quantification of the relative abundance of different glycoforms using MS.

Summary of the Invention

[0004] In one aspect, the present invention provides a method for enhancing mass spectrometry signals, the method comprising contacting a sample with a separation column under conditions that allow sample components to bind to a carrier, applying a first mobile phase gradient to the separation column, wherein the first mobile phase gradient comprises trifluoroacetic acid (TFA) and a small molecule additive (e.g., an amino acid) or formic acid (FA) and a small molecule additive (e.g., an amino acid) or ammonium formate and a small molecule additive (e.g., an amino acid), applying a second mobile phase gradient to the separation column, wherein the second mobile phase gradient comprises TFA and a small molecule additive (e.g., an amino acid) in acetonitrile (ACN), formic acid (FA) and a small molecule additive (e.g., an amino acid) in acetonitrile (ACN), or ammonium formate and a small molecule additive (e.g., an amino acid) in water and acetonitrile (ACN), and performing mass spectrometry on the eluted sample components.

[0005] In some embodiments, the small molecule additive in the first mobile phase is glycine.

[0006] In some embodiments, the small molecule additive in the first mobile phase is glycine and is glycine at a concentration of about 1 mM to about 2 mM.

[0007] In some embodiments, the glycine concentration in the first mobile phase is about 1 mM.

[0008] In some embodiments, the glycine concentration in the first mobile phase is about 2 mM.

[0009] In some embodiments, the small molecule additive in the second mobile phase is glycine.

[0010] In some embodiments, the small molecule additive in the second mobile phase is glycine and is glycine at a concentration of about 1 mM to about 2 mM.

[0011] In some embodiments, the glycine concentration in the second mobile phase is about 1 mM.

[0012] In some embodiments, the glycine concentration in the second mobile phase is about 2 mM.

[0013] In some embodiments, the TFA concentration in the first mobile phase is about 0.05% - 0.1% TFA in H2O, or the FA concentration in the first mobile phase is about 0.1% FA.

[0014] In some embodiments, the TFA concentration in the second mobile phase comprises about 0.05% TFA in 80% ACN and 20% H2O or about 0.1% TFA in 80% ACN and 20% H2O.

[0015] In some embodiments, the ammonium formate concentration in the first mobile phase is 50 mM and the pH is 4.4.

[0016] In some embodiments, the second mobile phase comprises 15% 50 mM ammonium formate (pH 4.4) in H2O and 85% ACN.

[0017] In some embodiments, the sample comprises a peptide, nucleotide, or glycan.

[0018] In some embodiments, the peptide is a glycopeptide.

[0019] In some embodiments, the glycopeptide is obtained from a monoclonal antibody.

[0020] In some embodiments, the monoclonal antibody is of isotype IgG1, IgG2, IgG3, IgG4, or a mixed isotype.

[0021] In some embodiments, the method further comprises preparing the sample under conditions that allow sample components to bind to the support prior to contacting the sample with the separation column.

[0022] In some embodiments, preparing the sample comprises contacting the sample with a denaturing and reducing solution under conditions that allow sample denaturation and reduction, contacting the denatured and reduced sample with an alkylating solution under conditions that allow alkylation of the sample, contacting the alkylated sample with a digestion solution under conditions that allow digestion of the sample, and contacting the digested sample with a quenching solution under conditions that stop digestion of the sample.

[0023] In some embodiments, preparing the sample comprises releasing glycans from the sample using an enzyme or chemical reaction, and labeling the released glycans with a fluorescent label or reducing the released glycans using a reducing agent.

[0024] In some embodiments, the sample is a monoclonal antibody and the digestion solution contains a protease.

[0025] In some embodiments, the protease includes trypsin.

[0026] In some embodiments, the separation column is a liquid chromatography (LC) separation column.

[0027] In some embodiments, the LC separation column includes a hydrophilic interaction (HILIC) liquid chromatography column.

[0028] In some embodiments, performing mass spectrometry on the eluted sample components comprises applying electrospray ionization to generate charged ions from the eluted sample components and measuring the generated charged ions.

[0029] In some embodiments, the method enhances mass spectral signals exhibited by an average of about 5 - 14-fold and / or approximately about 2 - 1000-fold increase at high charge state species (e.g., z≥3).

[0030] In some embodiments, the spectral signal increases by approximately 14-fold and / or approximately 1000-fold in highly charged species.

[0031] In some embodiments, the sample contains glycopeptides or glycans, and the mass spectral signal obtained for the eluted sample components is enhanced 2-fold to 50-fold relative to the mass spectral signal obtained for a control sample in the absence of the small molecule additive. In some cases, the glycopeptide is an O-glycan-containing glycopeptide. In some cases, the glycopeptide is an N-glycan-containing glycopeptide. In some cases, the glycan is an O-glycan. In some cases, the glycan is an N-glycan. In some cases, the O-glycan or N-glycan is linked to a label, optionally procainamide, 2-aminobenzamide, or RapiFluor. In any of these embodiments, the small molecule additive may be glycine.

[0032] In various embodiments, any of the features or components of the above or the embodiments discussed herein may be combined, and such combinations are included within the scope of the present disclosure. Any of the specific values discussed above or herein may be combined with another related value discussed above or herein to list ranges that represent the upper and lower limits of the range, and such ranges and all values within such ranges are included within the scope of the present disclosure. Each of the values discussed above or herein may be represented with a variation of 1%, 5%, 10%, or 20%. For example, a concentration of 10 mM may be represented as 10 mM ± 0.1 mM (1% variation), 10 mM ± 0.5 mM (5% variation), 10 mM ± 1 mM (10% variation), or 10 mM ± 2 mM (20% variation). Other embodiments will be apparent from a review of the forms for practicing the invention described below.

Brief Description of the Drawings

[0033]

Figure 1A

Figure 1B

Figure 1C

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8A

Figure 8B

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21A

Figure 21B

Figure 22

Figure 23

Figure 24

Figure 25

Figure 26A

Figure 26B

Figure 26C

Figure 27

Figure 28A

Figure 28B

Figure 28C

Figure 29

Figure 30A

Figure 30B

Figure 30C

Figure 31A

Figure 31B

Figure 31C

Figure 32

Figure 33

Figure 34

Figure 35

Figure 36

Figure 37A

Figure 37B

Figure 38A

Figure 38B

Figure 38C

Figure 39

Figure 40

Figure 41

Figure 42A

Figure 42B

Figure 42C-1

Figure 42C-2

Figure 42C-3

Figure 42D

Figure 42E

Figure 42F

Figure 42G

Figure 42H-1

Figure 42H-2

Figure 43A

Figure 43B

Figure 43C

Figure 43D-1

Figure 43D-2

Figure 43D-3

Figure 43E

Figure 43F

Figure 43G

Figure 44A

Figure 44B

Figure 44C

Figure 44D

Figure 45A

Figure 45B

Figure 45C

Figure 45D

Figure 45E

Figure 45F

Figure 46A

Figure 46B

Figure 46C

Figure 46D

Figure 46E

[0034] Before the present invention is described, it should be understood that the present invention is not limited to the specific methods and experimental conditions described, since they may vary. It should also be understood that the terms used in this specification are only for describing specific embodiments, and are not intended to be limiting, since the scope of the present invention is limited only by the scope of the appended claims. Any embodiment or the features of the embodiments can be combined with each other, and such combinations are expressly included within the scope of the present invention.

[0035] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention belongs.As used herein, the term "about" when used in relation to a specific recited numerical value means that the value may vary by 1% or less from the recited value.For example, as used herein, the expression "about 100" includes 99 and 101 and all values ​​therebetween (for example, 99.1, 99.2, 99.3, 99.4, etc.).

[0036] Any methods and materials similar or equivalent to those described in this specification can be used in the practice or testing of the present invention, but the preferred methods and materials will be described hereinafter. All patents, applications, and non-patent publications mentioned in this specification are incorporated herein by reference in their entirety.

[0037] Abbreviations used in this specification ACN: Acetonitrile ESI-MS: Electrospray ionization mass spectrometry FA: Formic acid FLR: Fluorescence detection HC: Heavy chain HESI: Heated electrospray ionization HILIC: Hydrophilic interaction liquid chromatography IP-HILIC: Ion-pair hydrophilic interaction chromatography IgG: Immunoglobulin G LC: Light chain LC-MS: Liquid chromatography - mass spectrometry mAb: Monoclonal antibody MPA: Mobile phase A MPB: Mobile phase B MS: Mass spectrometry MW: Molecular weight PROCA: Procainamide 2-AB: 2-Aminobenzamide PTM: Post-translational modification RPLC: Reverse-phase liquid chromatography RPLC-MS / MS: Reverse-phase liquid chromatography tandem mass spectrometry SPE: Solid-phase extraction TFA: Trifluoroacetic acid UV: Ultraviolet

[0038] Definitions As used herein, the term "antibody" is intended to refer to an immunoglobulin molecule (i.e., a "complete antibody molecule") consisting of four polypeptide chains, i.e., two heavy (H) chains and two light (L) chains interconnected by disulfide bonds, as well as multimers thereof (e.g., IgM) or antigen-binding fragments thereof. Each heavy chain consists of a heavy chain variable region ("HCVR" or "V H "), and a heavy chain constant region (C H 1 domain, C H 2 domain and C H 3 domains). In various embodiments, the heavy chain can be of the IgG isotype. In some cases, the heavy chain is selected from IgG1, IgG2, IgG3 or IgG4. In some embodiments, the heavy chain is a heavy chain of isotype IgG1 or IgG4 optionally comprising a chimeric hinge region of isotype IgG1 / IgG2 or IgG4 / IgG2. Each light chain consists of a light chain variable region ("LCVR" or "V L "), and a light chain constant region (C L ). The V H region and the V L region can be further subdivided into hypervariable regions called complementarity determining regions (CDRs) interspersed with more conserved regions called framework regions (FRs). Each V H and V LIt consists of three CDRs and four FRs arranged from the amino terminus to the carboxy terminus in the order of FR1, CDR1, FR2, CDR2, FR3, CDR3, FR4. The term "antibody" includes reference to both glycosylated and non-glycosylated immunoglobulins of any isotype or subclass. The term "antibody" includes antibody molecules prepared, expressed, produced or isolated by recombinant means, for example, antibodies isolated from host cells transfected to express the antibody. For reviews on antibody structure, see Lefranc et al., IMGT unique numbering for immunoglobulin and T cell receptor variable domains and Ig superfamily V-like domains, 27(1) Dev. Comp. Immunol. 55-77 (2003); and M. Potter, Structural correlates of immunoglobulin diversity, 2(1) Surv. Immunol. Res. 27-42 (1983).

[0039] The term "antibody" also encompasses "bispecific antibodies" which include heterotetrameric immunoglobulins that can bind to two or more different epitopes. Half of a bispecific antibody containing a single heavy chain and a single light chain as well as six CDRs binds to one antigen or epitope, and the other half of the antibody binds to a different antigen or epitope. In some cases, the bispecific antibody can bind to the same antigen but at different epitopes or non-overlapping epitopes. In some cases, both halves of the bispecific antibody have the same light chain while retaining bispecificity. Bispecific antibodies are generally described in US Patent Application Publication No. 2010 / 0331527 (December 30, 2010).

[0040] The term "antigen-binding portion" (or "antibody fragment") of an antibody refers to one or more fragments of an antibody that retain the ability to specifically bind to an antigen. Examples of binding fragments included within the term "antigen-binding portion" of an antibody include: (i) a Fab fragment, a monovalent fragment consisting of the VL, VH, CL, and CH1 domains; (ii) an F(ab')2 fragment, a bivalent fragment comprising two Fab fragments linked by disulfide bridges in the hinge region; (iii) an Fd fragment consisting of the VH and CH1 domains; (iv) an Fv fragment consisting of the VL and VH domains of a single arm of an antibody; (v) a dAb fragment consisting of the VH domain (Ward et al. (1989) Nature 241:544-546); (vi) an isolated CDR; and (vii) a scFv consisting of VL and VH, which are two domains of an Fv fragment linked by a synthetic linker so that the VL and VH regions pair to form a single protein chain that forms a monovalent molecule. Other forms of single-chain antibodies, such as diabodies, are also included within the term "antibody" (see, e.g., Holliger et al. (1993) 90 PNAS U.S.A. 6444-6448; and Poljak et al. (1994) 2 Structure 1121-1123).

[0041] Furthermore, antibodies and antigen-binding fragments thereof can be obtained using standard recombinant DNA techniques generally known in the art (see Sambrook et al., 1989). Methods for generating human antibodies in transgenic mice are also known in the art. For example, the VELOCIMMUNE® technology (see, e.g., U.S. Patent No. 6,596,541, Regeneron Pharmaceuticals, VELOCIMMUNE®) or any other known method for generating monoclonal antibodies is used to first isolate a high-affinity chimeric antibody against a desired antigen having human variable regions and murine constant regions. The VELOCIMMUNE® technology involves the generation of a transgenic mouse in which the genome containing the human heavy chain variable region and the human light chain variable region is operably linked to an endogenous murine constant region locus such that the mouse generates an antibody that, in response to antigen stimulation, contains the human variable regions and murine constant regions. DNA encoding the variable regions of the heavy and light chains of the antibody is isolated and operably linked to DNA encoding the human heavy chain constant region and the human light chain constant region. The DNA is then expressed in a cell capable of expressing a fully human antibody.

[0042] The term “human antibody” is intended to include antibodies having variable and constant regions derived from human germline immunoglobulin sequences. The human mAbs of the invention can include amino acid residues not encoded by human germline immunoglobulin sequences (e.g., mutations introduced by in vitro random or site-directed mutagenesis or by somatic mutations in vivo), e.g., in the CDRs, particularly CDR3. However, the term “human antibody” as used herein is not intended to include mAbs in which CDR sequences derived from the germline of another mammalian species (e.g., mouse) have been transplanted onto human FR sequences. The term includes antibodies recombinantly produced in non-human mammals or in cells of non-human mammals. The term is not intended to include antibodies isolated from or generated in a human subject.

[0043] As used herein, the term "glycopeptide / glycoprotein" refers to a modified peptide / glycoprotein having covalently attached carbohydrates or glycans, either during or after their synthesis. In certain embodiments, the glycopeptides are obtained from monoclonal antibodies, for example, from protease digests of monoclonal antibodies.

[0044] As used herein, the term "glycan" refers to a compound containing one or more sugar units. These generally include glucose (Glc), galactose (Gal), mannose (Man), fucose (Fuc), N-acetylgalactosamine (GalNAc), N-acetylglucosamine (GlcNAc), and N-acetylneuraminic acid (NeuNAc) (Frank Kjeldsen, et al. Anal. Chem. 2003, 75, 2355-2361). The glycan moieties of glycoproteins such as monoclonal antibodies are important properties for specifying their function or intracellular location. For example, certain monoclonal antibodies are modified with specific glycan moieties.

[0045] The term "hydrophilic interaction chromatography" or HILIC is intended to include a process using a hydrophilic stationary phase and a hydrophobic organic mobile phase in which hydrophilic compounds are retained longer than hydrophobic compounds. In certain embodiments, the method utilizes a water-miscible solvent mobile phase.

[0046] As used herein, the term "sample" refers to a mixture of molecules that contains at least analyte molecules, such as glycopeptides obtained from monoclonal antibodies, and is manipulated according to the methods of the present invention, including, for example, separation, analysis, extraction, concentration, or "profiling".

[0047] The term "analysis" or "analyzing", as used herein, is used interchangeably and refers to any of a variety of methods for separating, detecting, isolating, purifying, solubilizing, detecting, and / or characterizing a target molecule (such as a glycoprotein). Examples include, but are not limited to, solid phase extraction, solid phase microextraction, electrophoresis, mass spectrometry (e.g., ESI-MS, SPE HILIC, or MALDI-MS), liquid chromatography (e.g., high performance, e.g., reverse phase, normal phase, or size exclusion), ion pair liquid chromatography, liquid-liquid extraction (e.g., accelerated solvent extraction, supercritical fluid extraction, microwave assisted extraction, membrane extraction, Soxhlet extraction), precipitation, clarification, electrochemical detection, staining, elemental analysis, Edman degradation, nuclear magnetic resonance, infrared analysis, flow injection analysis, capillary electrochromatography, ultraviolet detection, and combinations thereof.

[0048] The term "profiling" as used herein refers to any of a variety of analytical methods used in combination to provide the content, composition, or characteristic ratios of glycopeptides in a sample.

[0049] "Electrospray ionization mass spectrometry" or "ESI-MS" is a technique used in mass spectrometry to generate ions using electrospray, where a high voltage is applied to a liquid to generate an aerosol. For example, in electrospray, ions are generated from proteins in solution, enabling the ionization of labile molecules intact and preserving non-covalent interactions. Electrospray ionization is selected as the ion source for combining liquid chromatography and mass spectrometry (LC-MS). The analysis can be performed online by directly feeding the liquid eluting from the LC column into the electrospray or offline by collecting the fractions to be analyzed later on a classical nanoelectrospray mass spectrometer. LC-MS can be used for protein characterization, which includes quantification of biomarkers, analysis of sequence variants, and identification and quantification of glycopeptides.

[0050] As used herein, "contacting" includes bringing together at least two substances in solution or in the solid phase.

[0051] General Description Accordingly, there is a need for a protein characterization method with improved sensitivity. The disclosed invention addresses such a need.

[0052] A new method of LC-MS-based protein characterization that increases mass spectrometry detection sensitivity is disclosed herein. This new method is based on the research reported herein, and the inventors have found that including a small molecule additive (e.g., an amino acid or a modified amino acid) in the mobile phase solution in liquid chromatography results in a significant boost in the mass spectral signal as compared to the signal generated in the absence of such small molecule additives. The inventors have also found that the presence of such additives, e.g., an amino acid (e.g., glycine), when added to the mobile phase buffer, does not affect the retention and chromatographic separation of peptides and glycans on the LC column. Furthermore, the effects of additives such as amino acids, e.g., glycine, on signal boost, charge state shift, and PTM quantification of peptides and glycans were reproducible. Additionally, TFA and glycine buffers improve the lower limit of quantification (LLOQ) in protein quantification, more reliably identify glycopeptides without affecting relative quantification compared to normal TFA buffers in the IP-HILIC-LC-MS method, and generate complementary information compared to FA buffers commonly used in sequence variant analysis while identifying more sequence variants. Accordingly, the disclosed findings have a very wide range of applications for LC-MS-based protein characterization through improving mass spectrometry detection sensitivity. In some embodiments, the disclosed method can be used for biomarker quantification, sequence variant analysis and / or identification and quantification of peptides, e.g., glycopeptides, and / or glycans by LC-MS.

[0053] In some embodiments, the method comprises contacting a sample with a separation column under conditions that allow sample components to bind to the carrier, and applying a migration gradient to the separation column, wherein the migration gradient buffer comprises a small molecule additive (e.g., an amino acid) and TFA, FA, ammonium formate, and / or ACN, and performing mass spectrometry on the eluted sample components.

[0054] The mobile phase used may include a buffer solution containing an ion pair forming agent (e.g., acetonitrile and water) and a buffer solution without it. Examples of the ion pair forming agent include formate, acetate, TFA, and salts. A gradient of the buffer solution can be used. For example, when two buffer solutions are used, the concentration or ratio of the first buffer solution can decrease while the concentration or ratio of the second buffer solution increases over the course of the chromatography run. For example, the ratio of the first buffer solution can decrease from about 100%, about 99%, about 95%, about 90%, about 85%, about 80%, about 75%, about 70%, about 65%, about 60%, about 50%, about 45%, or about 40% to about 0%, about 1%, about 5%, about 10%, about 15%, about 20%, about 25%, about 30%, about 35%, or about 40% over the course of the chromatography run. As another example, the ratio of the second buffer solution can increase from about 0%, about 1%, about 5%, about 10%, about 15%, about 20%, about 25%, about 30%, about 35%, or about 40% to about 100%, about 99%, about 95%, about 90%, about 85%, about 80%, about 75%, about 70%, about 65%, about 60%, about 50%, about 45%, or about 40% over the same run. Optionally, the concentration or ratio of the first and second buffer solutions can return to their starting values at the end of the chromatography run. As an example, the ratio of the first buffer solution can change in five steps from 85% to 63%, 59%, 10%, 85%, while the ratio of the second buffer solution in the same steps changes in five steps from 15% to 37%, 41%, 90%, 15%. The ratio can change gradually as a linear gradient or in a non-linear (e.g., stepwise) manner. For example, the gradient can be multi-phase (e.g., two-phase, three-phase, etc.). In some embodiments, the method described herein uses a decrease in the acetonitrile buffer gradient corresponding to an increase in the polarity of the mobile phase without using an ion pair forming agent.

[0055] In some embodiments, applying a gradient to the separation column comprises applying a first gradient buffer to the separation column, where the first mobile phase buffer comprises TFA and a small molecule additive (e.g., an amino acid), FA and a small molecule additive (e.g., an amino acid), or ammonium formate and a small molecule additive (e.g., an amino acid), and applying a second gradient to the separation column, where the second mobile phase buffer comprises TFA and a small molecule additive (e.g., an amino acid) in ACN, FA and a small molecule additive (e.g., an amino acid) in ACN, or ammonium formate and a small molecule additive in water / ACN.

[0056] In various embodiments, the small molecule additive is selected from glycine, alanine, serine, valine, N-acetylglycine, methionine, β-alanine, aspartic acid, or N-methylglycine. In some cases, the amino acid is selected from glycine, alanine, serine or valine. In some embodiments, the amino acid is alanine. In some embodiments, the amino acid is serine. In some embodiments, the amino acid is valine. In some embodiments, the amino acid in the first mobile phase buffer is glycine. In some embodiments, the amino acid in the second mobile phase buffer is glycine. In some embodiments, the amino acid in the first and second mobile phase buffers is glycine. In some embodiments, the small molecule additive (e.g., an amino acid) in the first and / or second mobile phase buffer is a small molecule (e.g., a modified amino acid) or one of the other amino acids identified above or herein.

[0057] The concentration of the small molecule additive (e.g., amino acid) in the mobile phase buffer is from about 0.5 mM to about 5 mM, such as from about 0.5 mM to about 3 mM, from about 1 mM to about 2 mM, and includes 0.5 mM, 0.6 mM, 0.7 mM, 0.8 mM, 0.9 mM, 1.0 mM, 1.1 mM, 1.2 mM, 1.3 mM, 1.4 mM, 1.5 mM, 1.6 mM, 1.7 mM, 1.8 mM, 1.9 mM, 2.0 mM, 2.1 mM, 2.2 mM, 2.3 mM, 2.4 mM, 2.5 mM, 2.6 mM, 2.7 mM, 2.8 mM, 2.9 mM, 3.0 mM, 3.1 mM, 3.2 mM, 3.3 mM, 3.4 mM, 3.5 mM, 3.6 mM, 3.7 mM, 3.8 mM, 3.9 mM, 4.0 mM, 4.1 mM, 4.2 mM, 4.3 mM, 4.4 mM, 4.5 mM, 4.6 mM, 4.7 mM, 4.8 mM, 4.9 mM or 5.0 mM. In some embodiments, the small molecule additive (e.g., amino acid) is less than 5 mM. In some embodiments, the small molecule additive is glycine at a concentration less than 5 mM. In some embodiments, the amino acid in the first mobile phase buffer is glycine and is glycine at a concentration of about 1 to about 2 mM. In some embodiments, the amino acid in the second mobile phase buffer is glycine and is glycine at a concentration of about 1 to about 2 mM. In some embodiments, the glycine concentration in the first mobile phase buffer is about 1 mM. In some embodiments, the glycine concentration in the first mobile phase buffer is about 2 mM. In some embodiments, the amino acid in the second mobile phase buffer is glycine and is glycine at a concentration of about 1 to about 2 mM. In some embodiments, the glycine concentration in the second mobile phase buffer is about 1 mM. In some embodiments, the glycine concentration in the second mobile phase buffer is about 2 mM. In some embodiments, the amino acids in the first and second mobile phase buffers are glycine and are glycine at a concentration of about 1 to about 2 mM.

[0058] In some embodiments, the concentration of TFA in the first mobile phase is about 0.03% - 0.15% TFA in H2O, such as about 0.03% - 0.1%, or FA is about 0.05% - about 0.15% in H2O, such as about 0.1% FA. In some embodiments, the TFA concentration is about 0.05% - about 0.1% TFA in H2O, or the FA concentration in the first mobile phase is about 0.1% FA. For example, the TFA concentration is about 0.03%, 0.04%, 0.05%, 0.06%, 0.07%, 0.08%, 0.09%, or 0.1% in H2O. In some embodiments, the TFA concentration in the second mobile phase comprises about 0.05% TFA in 80% ACN and 20% H2O or about 0.1% TFA in 80% ACN and 20% H2O. In some embodiments, the concentration of ACN in the second mobile phase is about 60% - 100%, such as 80% - 100% including, for example, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95% or 100%.

[0059] In some embodiments, the ammonium formate concentration in the first mobile phase is 50 mM in H2O. In some embodiments, the second mobile phase is 15% 50 mM and 85% ACN in H2O.

[0060] In some embodiments, the sample comprises a peptide, nucleotide or glycan. For example, the sample may comprise a glycopeptide such as a glycopeptide obtained from a monoclonal antibody. In some embodiments, the monoclonal antibody is of isotype IgG1, IgG2, IgG3, IgG4, or a mixed isotype.

[0061] In some embodiments, the method includes preparing the sample before contacting the sample with the separation column under conditions that allow the sample components to bind to the carrier. In some embodiments, preparing the sample includes contacting the sample with a denaturing and reducing solution under conditions that allow sample denaturation and reduction, contacting the denatured and reduced sample with an alkylating solution under conditions that allow alkylation of the sample, contacting the alkylated sample with a digestion solution under conditions that allow digestion of the sample, and contacting the digested sample with a quench solution under conditions that stop digestion of the sample.

[0062] In some embodiments, preparing the sample includes releasing glycans from the sample using an enzyme or chemical reaction, and labeling the released glycans with a fluorescent label or reducing the released glycans using a reducing agent.

[0063] In some embodiments, N-glycans can be released from glycoproteins using PNGase F. In some embodiments, O-glycans are released from glycoproteins by a basic chemical. The released N-glycans can react with a RapiFluor fluorescent label. The released N-glycans and O-glycans can be reduced by sodium borohydride or conjugated to PROCA or 2-AB by incubation with acetic acid and sodium cyanoborohydride.

[0064] In some embodiments, the sample is a monoclonal antibody and the digestion solution contains one or more proteases such as trypsin. In some examples, the method is used to characterize / analyze glycopeptides such as those obtained from monoclonal antibodies, such as antibodies digested with one or more proteases. In some embodiments, the antibody in the sample can be processed and prepared by reduction, enzymatic degradation, denaturation, or fragmentation before contacting the resulting sample with the support. For example, the method can be used to characterize the glycosylation of proteins, such as monoclonal antibody (mAb) therapeutics, by fragment and peptide-level LC-MS such as HILIC-MS analysis. In certain embodiments, the sample at any intervening step may be concentrated, diluted, desalted, etc.

[0065] Glycopeptides are obtained from glycosylated proteins such as monoclonal antibodies. Glycosylated monoclonal antibodies can be prepared by reduction, enzymatic digestion, denaturation, fragmentation, chemical cleavage, and combinations thereof. The methods disclosed herein are applicable to any antibody isotype, such as IgG1, IgG2, IgG3, IgG4, or mixed isotypes. Reduction is the reduction of disulfide bonds to two thiols in a three-dimensional protein such as a monoclonal antibody. Reduction can be performed in the presence of a reducing agent, such as TCEP-HCl, by heat denaturation, addition of a surfactant, or addition of a denaturing agent, such as guanidine HCl (6M). Enzymatic digestion is the digestion of a protein by a protease, such as trypsin or Achromobacter protease I (Lys-C). In addition, glycoproteins can be denatured by heat or chemicals, or combinations thereof. Fragmentation involves cleaving the protein portion of a single or multiple subunit protein, such as a monoclonal antibody, by physical, biological, or chemical methods.

[0066] In some embodiments, the separation column is a liquid chromatography (LC) separation column. Liquid chromatography, including HPLC, can be used to analyze structures such as peptides including glycopeptides. Various forms of liquid chromatography can be used to study these structures, including anion exchange chromatography, reverse-phase HPLC, size exclusion chromatography, high-speed anion exchange chromatography, and normal-phase (NP) chromatography (including NP-HPLC) (see, for example, Alpert et al., J. Chromatogr. A 676:191-202 (1994)). Hydrophilic interaction chromatography (HILIC) is a variant of NP-HPLC that can be performed using a partially aqueous mobile phase and enables normal-phase separation of peptides, carbohydrates, nucleic acids, and many proteins. The elution order in HILIC is from the least polar to the most polar, opposite to that in reverse-phase HPLC. HPLC can be performed using an HPLC system from, for example, Waters (e.g., Waters 2695 Alliance HPLC system), Agilent, Perkin Elmer, Gilson, etc.

[0067] NP-HPLC, preferably HILIC, is a particularly useful form of HPLC that can be used in the methods described herein. NP-HPLC separates analytes based on polar interactions between the analyte and the stationary phase (e.g., the support). Polar analytes associate with the polar stationary phase and are retained by the polar stationary phase. The adsorption strength increases with increasing analyte polarity, and the interaction between the polar analyte and the polar stationary phase (relative to the mobile phase) increases the elution time. The use of a more polar solvent in the mobile phase decreases the retention time of the analyte, while a more hydrophobic solvent tends to increase the retention time.

[0068] In NP-HPLC, various types of carriers such as silica, amino, amide, cellulose, cyclodextrin, and polystyrene carriers can be used for column chromatography. For example, examples of useful carriers that can be used in column chromatography include polysulfoethyl aspartamide (e.g., manufactured by PolyLC), sulfobetaine carriers such as ZIC®-HILIC (e.g., manufactured by SeQuant), POROS® HS (e.g., manufactured by Applied Biosystems), POROS® S (e.g., manufactured by Applied Biosystems), polyhydroxyethyl aspartamide (e.g., manufactured by PolyLC), Zorbax 300 SCX (e.g., manufactured by Agilent), PolyGLYCOPLEX® (e.g., manufactured by PolyLC), Amide-80 (e.g., manufactured by Tosohaas), TSK GEL® Amide-80 (e.g., manufactured by Tosohaas), polyhydroxyethyl A (e.g., manufactured by PolyLC), Glyco-Sep-N (e.g., manufactured by Oxford GlycoSciences), and Atlantis HILIC (e.g., manufactured by Waters). In some embodiments, the disclosed method includes a column that utilizes one or more of the following functional groups: a carbamoyl group, a sulfopropyl group, a sulfoethyl group (e.g., poly(2-sulfoethyl aspartamide)), a hydroxyethyl group (e.g., poly(2-hydroxyethyl aspartamide)), and an aromatic sulfonic acid group.

[0069] The column temperature can be maintained at a constant temperature throughout the chromatography run, for example, using a commercially available column heater. In some embodiments, the column is maintained at a temperature of about 18°C to about 70°C, such as about 30°C to about 60°C, about 40°C to about 50°C, for example, about 20°C, about 25°C, about 30°C, about 35°C, about 40°C, about 45°C, about 50°C, about 55°C, about 60°C, about 65°C, or about 70°C. In some embodiments, the column temperature is about 40°C.

[0070] The flow rate of the mobile phase can be from about 0 to about 100 ml / min. For analytical purposes, the flow rate is typically in the range of 0 to 10 ml / min, and for preparative HPLC, flow rates exceeding 100 ml / min can be used. For example, the flow rate may be about 0.5, about 1, about 1.5, about 2, about 2.5, about 3, about 3.5, about 4, about 4.5, or about 5 ml / min. Replacing a column of the same packing material and the same length with a column having a smaller diameter requires a decrease in the flow rate to maintain the same retention time and resolution for the peaks seen with a wider diameter column. In some embodiments, a flow rate corresponding to about 1 ml / min in a 4.6×100 mm, 5 μm column is used.

[0071] In some embodiments, the run time can be from about 15 to about 240 minutes, for example, from about 20 to about 70 minutes, from about 30 to about 60 minutes, from about 40 to about 90 minutes, from about 50 minutes to about 100 minutes, from about 60 to about 120 minutes, from about 50 to about 80 minutes.

[0072] NP-HPLC can be adjusted to be performed on the nanoscale, for example, using a column having an inner diameter of about 75 μm (see, for example, Wuhrer et al., Anal. Chem. 76:833-838 (2004); Wuhrer et al., Internat. J. Mass. Spec. 232:51-57 (2004)).

[0073] In certain embodiments, the separation column is a hydrophilic interaction (HILIC) separation column, and molecules such as glycopeptides are then eluted from the HILIC separation column using, for example, a mobile phase gradient to resolve the individual species of glycopeptides, thereby purifying and / or separating the glycopeptides in the sample. In certain examples, the glycopeptides eluted from HILIC are separated into one or more fractions. Such fractions can be used for subsequent analysis such as MS analysis. In certain embodiments, the method includes identifying molecules such as glycopeptides and / or glycans present in one or more of the fractions. In certain embodiments, the glycan is an N-glycan or an O-glycan. In some embodiments, the method further includes detecting the glycopeptides, for example, using a UV signal from the peptide portion of the glycopeptide. This can be done on fractions of the sample and enables the selection of specific fractions for further analysis, such as mass spectrometry (MS) analysis. In some embodiments, the method includes detecting the glycans using an FLR signal from a fluorescent label linked to the glycan.

[0074] In some embodiments, performing mass spectrometry on the eluted sample components includes applying electrospray ionization to generate charged ions from the eluted sample components and measuring the generated charged ions.

[0075] In the application of mass spectrometry for the analysis of biomolecules, molecules are transferred from a liquid or solid phase to a gas phase and a vacuum phase. Since many biomolecules are large and fragile (proteins being the main example), two of the most effective methods for their transfer to the vacuum phase are matrix-assisted laser desorption ionization (MALDI) or electrospray ionization (ESI). Generally, ESI is more sensitive and MALDI is faster. Importantly, some peptides ionize better in MALDI mode than in ESI mode and vice versa (Genome Technology, June 2002, p52). ESI is performed by mixing the sample with a volatile acid and an organic solvent and injecting it through a conductive needle charged at a high voltage. The charged droplets sprayed (or ejected) from the needle tip are induced into the mass spectrometer and dried by heat and vacuum as they fly. After the droplets are dried, the remaining charged molecules are directed towards the mass detector by an electromagnetic lens and mass-analyzed. In one embodiment, the eluted sample is directly deposited from the capillary to the electrospray nozzle, for example, the capillary functions as a sample loader. In another embodiment, the capillary itself functions as both an extraction device and an electrospray nozzle. In some embodiments, the method enhances the mass spectral signal, shown by an increase on average of about 2 - 27 fold, for example about 5 - 14 fold and / or approximately about 2 - 1000 fold, in highly charged state species (e.g., z ≧ 3). In some embodiments, at 1 mM glycine, when the sample loading amount is 10 μg, the fold change is about 5. In some embodiments, the spectral signal increases approximately 14 fold and / or approximately 1000 fold in highly charged state species. It is contemplated that the boost factor of glycine may depend on the sample loading amount and glycine concentration. For example, the higher the glycine concentration, the higher the boost produced compared to a lower glycine concentration for different sample loading amounts, and the boost generally increases with a decrease in the sample loading amount. There are certain points where no significant change (<10%) in the boost factor due to the loading amount can be observed (as demonstrated in FIGS. 1A, 1B and 2).

[0076] In some embodiments, other ionization modes are used, such as turbo spray ionization mass spectrometry, nano spray ionization mass spectrometry, thermo spray ionization mass spectrometry, sonic spray ionization mass spectrometry, SELDI-MS, and MALDI-MS. Generally, the advantage of these methods is that they enable "just in time" purification of the sample and direct introduction into the ionization environment. Note that the various ionization modes and detection modes introduce their own constraints with respect to the nature of the desorption solution, and it is important that the desorption solution be compatible with both. For example, sample matrices in many applications need to have a low ionic strength or be within a specific pH range. In ESI, salts in the sample can interfere with detection by reducing ionization or by clogging the nozzle. This problem is addressed by presenting the analyte in low salt and / or by using volatile salts.

[0077] In some embodiments, the support is prepared for sample addition by washing, such as a pre-wash step. In some embodiments, the support is washed prior to contact with the glycopeptide sample. In various embodiments, the support is contacted with a sample containing a biomolecule, such as a glycopeptide, for concentration. With respect to the sample solution, the sample solution contains a biomolecule, such as a glycopeptide, dissolved in a solvent in which the biomolecule, such as a glycopeptide, is soluble and the biomolecule, such as a glycopeptide, binds to the support. Preferably, the binding is strong, resulting in binding of a substantial portion of the biomolecule, such as more than 50% of the biomolecule, such as more than 50% of the glycopeptide. In some cases, substantially all, more than 95% of the biomolecule, such as a glycopeptide, is bound. In various embodiments, the solvent is an aqueous solution and typically contains a buffer, salt, and / or surfactant to solubilize and stabilize the biomolecule, such as a glycopeptide. In some embodiments, the biomolecule sample, such as a glycopeptide sample, is a solution having a low pH of less than about 6.5, such as less than about 6.0, 5.5, 5.0, 4.5, 4.0, 3.5, or 3.0.

[0078] In one particular embodiment, a method for enhancing a mass spectrometry signal includes denaturing and reducing a monoclonal antibody. For example, the monoclonal antibody may be denatured and reduced by heating in the presence of TCEP-HCl with acetic acid (e.g., 5 mM acetic acid) for a time sufficient for denaturation and reduction to occur (e.g., 10 minutes at 80° C.). After denaturation and reduction, the sample is alkylated. In some examples, the sample is first diluted and then alkylated. For example, after dilution with 100 mM Tris-HCl (pH 7.5) containing 8M urea, it can be alkylated with iodoacetamide in the dark at room temperature for 30 minutes. After alkylation, the sample is further diluted to reduce the urea concentration, for example, diluted with 100 mM Tri-HCl (pH 7.5) to reduce the urea concentration to less than 1M. The sample is then digested with a protease. For example, the sample is treated with trypsin at an enzyme-to-substrate ratio of 1:20 (w / w) at 37° C. for 4 hours. The digestion is stopped, for example, by quenching the sample with TFA such as 10% TFA at the desired time. The digested sample is then subjected to online LC-MS analysis. For example, the trypsin digested (reduced / alkylated) sample is loaded at a sufficient concentration (e.g., 0.25 μg), and the mobile phase gradient A (MP-A) contains TFA in H2O containing 1-2 mM glycine, e.g., TFA in H2O containing 2 mM glycine, followed by mobile phase gradient B of TFA in ACN containing 1-2 mM glycine, e.g., 0.05% TFA in 80% ACN and 20% H2O containing 2 mM glycine.

Example

[0079] The following examples are presented to provide those skilled in the art with a complete disclosure and description of how to make and use the methods of the invention and are not intended to limit the scope of what the inventors regard as their invention. Efforts have been made to ensure the accuracy with respect to the numerical values used (e.g., amounts, temperatures, etc.), but some experimental errors and deviations should be accounted for. Unless otherwise indicated, parts are parts by weight, molecular weights are average molecular weights, temperatures are in degrees Celsius, room temperature is about 25° C., and pressures are at or near atmospheric pressure.

[0080] Example 1: A New Method for ESI-MS Signal Boosting Using Glycine Additive Trypsin Digestion of NISTmAb: 100 μg of NISTmAb was denatured and reduced at 80 °C for 10 min in 5 mM acetic acid in the presence of 5 mM TCEP-HCl. After denaturation and reduction, the sample was diluted with 100 mM Tris-HCl (pH 7.5) containing 8 M urea and alkylated with iodoacetamide in the dark at room temperature for 30 min. After alkylation, the sample was further diluted with 100 mM Tri-HCl (pH 7.5) to reduce the urea concentration to less than 1 M. The sample was incubated with trypsin at an enzyme-to-substrate ratio of 1:20 (w / w) at 37 °C for 4 h. The digested sample was quenched by the addition of 10% TFA to stop trypsin digestion and then subjected to online LC-MS analysis. The NISTmAb trypsin digest (reduced / alkylated) was loaded in different amounts of 0.05 - 10 μg MP-A: 0.05% TFA in H2O, 0.0625 - 5 mM glycine in MP-B: 0.05% TFA, 80% ACN and 20% H2O. The column was ACQUITY UPLC Peptide BEH C18, 130 Å, 1.7 μm, 2.1 mm × 150 mm (Waters), and the LC conditions were 0.25 mL / min and a column temperature of 40 °C.

[0081] To investigate different small molecule reagents for MS boosting (as shown in Figure 3), the NISTmAb trypsin digest (reduced / alkylated) was loaded at a concentration of 0.25 μg, MP-A: 0.05% TFA in H2O, MP-B: 0.045% TFA in ACN. The column was ACQUITY UPLC Peptide BEH C18, 130 Å, 1.7 μm, 2.1 mm × 150 mm (Waters), the LC conditions were 0.25 mL / min and a column temperature of 40 °C, and different small molecule solutions of 125 mM in 50:50 H2O / ACN were delivered at 10 μL / min using a syringe pump and mixed with the eluate from the LC column, and the final small molecule concentration was approximately 5 mM.

[0082] Table 1 provides %A and %B at various time points.

Table 1

[0083] Figures 1A and 1B show the effects of various concentrations of glycine on MS boost and the effects of various sample loading amounts on MS boost. Figures 1A and 1B were generated by directly adding glycine to the TFA mobile phase bottles (MPA: 0.05% TFA and glycine in water, MPB: 80% ACN and 20% water and 0.05% TFA in glycine), rather than by a post-column syringe pump configuration. In Figure 1A, an obvious dependence of the glycine boost on the sample loading amount is shown. Higher glycine concentrations were required for higher loading amounts in order to obtain the same boost factor as low glycine concentrations at low loading amounts.

[0084] Figure 1B shows that an increase in the sample loading amount overall decreased the boosting power of glycine on the peptide response. 2 mM glycine showed a higher boost than 1 mM for different sample loading amounts. Also, as the sample loading amount decreased, the increasing trend of the boost factor gradually became slower. The smallest change in the boost factor (<10%) due to the loading amount was observed when the sample loading amount was less than 0.2 and 0.5 μg at 1 and 2 mM glycine concentrations, respectively.

[0085] A decrease in the boost factor was observed with an increase in the sample loading amount, but the mass spectrometry response (i.e., the EIC peak area of the peptide) was still higher at higher loading amounts than at lower loading amounts, even at high boost factors.

[0086] Figure 2 provides the chemical structures of 10 amino acids based on the different chemical properties of their side chains, including alanine and valine (hydrophobic side chains), serine (hydrophilic side chain), proline (cyclic side chain), methionine (sulfur-containing side chain), glutamine (amide-containing side chain), glutamic acid (acidic side chain), lysine and arginine (basic side chains) and histidine (aromatic side chain), as well as side chains derived from amines (e.g., N,N-dimethylglycine, N-acetylglycine, N-methylglycine and propionic acid), glycine modified on the side chain (e.g., 3,3,3-trifluoro-DL-alanine) or carboxyl group (e.g., ethylamine), or glycine with additional carbon atoms inserted to extend the distance between the amine group and the carboxyl group (e.g., β-alanine and γ-aminobutyric acid). Figure 3 is a graph showing the average MS1 boost factor of NISTmAb tryptic peptides with different small molecular weight reagents (2 mM). As shown in Figure 3, glycine showed the highest average boost. Glycine is also the molecule with the lowest molecular weight. Figure 4 is a trace showing the minimal change in the chromatographic separation of peptides after adding glycine to the TFA buffer. The study conditions were 2.5 μg loading of NISTmAb tryptic digest (reduced / alkylated), MP-A: 0.05% TFA in H2O with or without 2 mM glycine, MP-B: 0.05% TFA in 80% ACN and 20% H2O with or without 2 mM glycine. The column was Acquity UPLC Peptide BEH C18, 130A, 1.7 μm, 2.1 mm × 150 mm (Waters, LC conditions 0.25 mL / min, 40 °C column temperature). Note that MP-B was changed from the conventional condition of 0.05% TFA in ACN to 0.05% TFA in 80% ACN and 20% H2O due to the solubility problem of 2 mM glycine in 100% ACN.Table 2 below shows the UPLC performance metrics for eight representative peptide peaks (P1 - P8) selected in the UV chromatogram across the entire LC gradient (see Figure 4), showing nearly equivalent UPLC performance metrics between two mobile phase systems with a retention time difference < 0.06 minutes, a peak width % difference < 1.4%, and a peak area % difference < 1.9%. The peak width is the corresponding width of the base of the peak obtained by drawing a tangent at 50% peak height. Table 3 provides %A and %B at various time points.

Table 2

Table 3

[0087] Figure 5 is a bar graph showing the reproducibility of the boost by 2 mM glycine in TFA buffer containing 0.25 μg of loaded NISTmAb, with each sample repeated 10 times. A peptide-dependent response boost was observed, and the boost factors for different tryptic peptides ranged widely from 2 to 27-fold, with an average of approximately 13.2-fold. Most peptides showed a response boost of at least 5-fold, and more than half of the peptides showed a boost of more than 10-fold. For all peptides, the average boost was 13.2-fold, and the % relative standard deviation (RSD) was less than 5%. Figure 6 is a total ion current (TIC) plot showing the signal-to-noise ratios observed for the TFA control and TFA with glycine, with a 2.5 μg NISTmAb loading. Ten representative tryptic peptides with different retention times across the entire LC gradient were selected to compare their signal-to-noise ratios in the total ion chromatograms between the TFA mobile phase containing the glycine additive and the TFA mobile phase without it. The signal-to-noise ratios were indeed boosted for these 10 tryptic peptides with the glycine additive. Similar to the peptide-dependent response boost, the selected peptides also showed a boost in signal-to-noise ratios that varied over a wide range of more than one order of magnitude. The glycine additive boosted the mass spectrometry response of the tryptic peptides, but no boost was observed for the background noise derived from small molecule contaminants generated from the ambient environment during electrospray ionization. This boost feature can lead to an improvement in the signal-to-noise ratio of the tryptic peptides, while the absolute response was boosted.

Table 4

[0088] Figures 7, 8A, and 8B are bar graphs showing the charge state shifts in the presence of glycine. %RSD was less than 0.4% for all peptides (10 replicates). The average charge state shift was peptide-dependent, and both upward and downward shifts were observed. Figure 9 is a bar graph showing PTM quantification (%) in the presence and absence of 2 mM glycine (TFA buffer, 0.25 μg NISTmAb loading) (10 replicates). Figure 10 is a bar graph showing the boost factor of glycine in the presence of TFA or FA or FA alone compared to the TFA control (3 replicates). As shown in Figure 10, TFA in the presence of 2 mM glycine resulted in a 13.2-fold boost compared to the 12.3-fold boost observed with FA or the 6.1-fold boost with FA in the presence of 2 mM glycine.

[0089] Figure 11 is a graph showing the boost factor of individual amino acids by 2 mM glycine (Column: Discovery HS F5-3, 15 cm × 2.1 mm, 3 μm (Supelco); 1 nmol loading of each amino acid). Figure 12 shows the effect of trace sodium in a commercially available glycine preparation.

[0090] The ESI-MS signal boost by glycine additives for protein quantification was evaluated. Figure 13 provides a comparison of the quantification of the STSGGTAALGCLVK peptide (SEQ ID NO: 22) using PRM and full scan. The same three-digit response linear range was observed for TFA and FA with a glycine mobile phase having a regression slope similar to that of the TFA and FA control mobile phases, showing a consistent response boost of the glycine additive over a wide protein concentration. Due to the response boost of the glycine additive in the TFA and FA mobile phases, the response linear range also reached lower protein concentrations compared to the TFA and FA control mobile phases. This feature may have the potential to expand the opportunity to consider glycine-based mobile phases as an alternative for quantifying low-abundance proteins in complex matrices with increased sensitivity.

[0091] Example 2: ESI-MS Signal Boost Using Glycine Additive: IP-HILIC-MS Based Glycopeptide Identification and Quantification Figure 14 shows the advantages of HILIC-based liquid chromatography over reversed phase for glycopeptide analysis. Peptides were prepared by either of the routinely performed reduction or non-reduction peptide mapping methods. As shown in Figure 14, better separation of glycopeptides was obtained with HILIC, while clustering of glycopeptides was observed with reversed phase. For example, an increase in the separation of isomers such as G1F was seen with IP-HILIC versus reversed phase. Furthermore, the glycopeptide elution order on the HILIC column correlated with the size of the glycan attached to the peptide, and it was possible to easily distinguish between artifacts generated by in-source fragmentation of large glycans and actual peaks (separate elution times without overlap).

[0092] Figure 15 shows that signal boost was observed in glycine-containing buffer for both MS1 and MS2. The sample was a trypsin digested pool VEGF trap IND lot, the column was Waters BEH-Amide, and the mobile phases were A: 0.1% TFA and B: 80% ACN + 0.1% TFA. Figure 16 shows that high concentrations of glycine prevented binding.

[0093] Figure 17 provides a schematic illustration of an exemplary sample preparation workflow for glycopeptide analysis by HILIC-based LC. Figure 18 shows the effect of desalting by a SepPak 30 cartridge in eliminating the elevated baseline (see the arrow). The elution of peptides with mobile phase B (0.1% TFA in 80% acetonitrile) eliminated the drying step after desalting, thereby saving time and increasing the efficiency of the process. Also, studies have revealed that metal adducts can be reduced by using a polyethylene solvent container (such as a polyethylene bottle made by Belart) in the mobile phase as compared to a glass bottle. When using a plastic bottle compared to a glass bottle during the mobile phase, a 2.7-fold intensity boost and a 1.7-fold intensity boost were observed. The plastic bottle reduced the heterogeneity caused by sodium, thereby boosting the MS signal. Furthermore, soaking overnight in 60% isopropyl alcohol removed leachable impurities / hardening agents. Figures 19 and 20 provide tables showing the effect of glycine. Figure 19 shows that glycine increased the peak area of all glycoforms in IGG1 by 20-fold (*the glycoforms in this table were identified only based on tandem MS in any run. **For glycoforms without tandem MS data, the peaks were determined based on the precursor mass and retention time identified in other runs (shown in the column labeled Peptide 1_Control)). Figure 20 provides the peak area of all glycoforms in the IGG4 molecule (*the glycoforms in this table were identified only based on tandem MS in any run. **For glycoforms without tandem MS data, the peaks were determined based on the precursor mass and retention time identified in other runs (shown in the column labeled Peptide 2 Control). ***The fold change of glycoforms with an abundance below the LLOQ in the control run gives an overestimated number (shown in parentheses)). Figure 21 provides the peak area of all glycoforms in the evaluated mAb.The relative distribution of all glycoforms remained the same in the presence of glycine. Figure 22 shows the enhanced fragmentation ability due to the charge state shift in the presence of 1 mM glycine. The major charge state of the glycopeptide shifted from +2 to +3, and the latter tends to fragment easily. A signal boost was observed upon addition of glycine (1 mM) to VEGF TRAP. Glycine increased the number of peptide spectrum matches (PSMs) and glycoforms in VEGF TRAP (see Figure 23).

[0094] Example 3: ESI-MS Signal Boost Using Glycine Additive: Sequence Variant Analysis This example demonstrates the ability of glycine in TFA to increase the number of sequence variants. Figure 24 shows a comparison of the number of sequence variants identified when using TFA + glycine after Byologic verification with FA.

[0095] Example 4: ESI-MS Signal Boost Using Glycine Additive: Glycan Analysis When released from glycoproteins, N-glycans can often be labeled with various fluorescent tags that also enhance the MS response. Labeled or reduced N-glycans can be analyzed by HILIC using a mobile phase containing a salt, such as ammonium formate. 1 mM glycine in the mobile phase was found to boost (more than 3 - 50-fold) the MS signal of PROCA-labeled N-glycans derived from human serum (see Figures 25 and 26A - 26C). Glycine had a stronger effect (>10-fold) on high-mannose and large acidic N-glycans. Using an ammonium formate mobile phase without glycine, the MS profile of PROCA-labeled N-glycans showed some differences from the FLR profile widely used for glycan quantification (see Figures 27 and 28A - 28C). By comparing the MS and FLR profiles, large N-glycans with high mannose and three sialic acids had relatively low MS signals, while bisecting N-glycans had relatively high MS signals. By adding glycine to the mobile phase, not only did the MS signal become higher, but it was also more comparable to FLR (see Figures 27 and 28A - C). Figure 29 shows a scatter plot of serum N-glycans with PROCA-labeled EIC peak areas against FLR peak areas (log scale) under control and glycine-containing conditions. A smaller R-squared value for the linear regression for the glycine condition compared to that for the control condition demonstrates that glycine can improve the accuracy of N-glycan quantification using the MS signal. A strong signal-boosting effect of glycine on PROCA-labeled N-glycans can also be observed using a mobile phase containing 0.1% FA (3 - 38-fold) (see Figures 30A - 30C). Glycine also showed a moderate signal-boosting effect (1 - 3-fold) on RapiFluor-labeled N-glycans and small reduced N-glycans using an ammonium formate mobile phase (see Figures 31A - 31C and 32). An inhibitory effect on larger reduced N-glycans was observed (Figure 32).The relative levels of large RapiFluor-labeled N-glycans based on MS peak areas and FLR peak areas were equivalent to those under control conditions and were comparable to those using the glycine condition (Figure 32). Figure 33 provides exemplary LC-FLR-MS conditions for the analysis of N-glycans.

[0096] Unlike N-glycans, there is no satisfactory method for O-glycan release because of the lack of releasing enzymes and problems with O-glycan reducing-end cleavage by release efficiency and chemical release. Figure 34 provides an exemplary method for O-glycan preparation and analysis by LC-MS and fluorescence (FLR) detection.

[0097] Figure 35 shows the results of PROCA-labeled O-glycans from 2 μg of bovine submandibular mucin (BSM) on column, fluorescence, and optimized reductive amination. Figure 36 demonstrates that the PROCA-labeled O-glycan FL and MS TIC profiles are nearly identical. The PROCA-labeled O-glycans released from BSM were run on HILIC (equivalent to 2 μg on column). Low MS background was observed. 1 mM glycine in the mobile phase was found to boost the effect on PROCA-labeled BSM O-glycans purified from beads (2 μg on column, Figures 37A - B). Figures 37A - 37B provide the results of comparing O-glycan FLR and MS peak intensities with and without glycine in the mobile phase. Compared to FLR, the PROCA signal significantly decreases with increasing glycan size. Glycine compensated for this effect and made the MS signal more equivalent to the FLR intensity (Figures 37A - 37B). Glycine had a moderate boosting effect (1 - 3 fold) on small (1 - 2 sugar rings) O-glycans (see Figures 38A - 38C). Glycine had a strong boosting effect (greater than 3 - 35 fold) on larger and / or acidic glycans labeled with PROCA (see Figure 38A). This effect was strong for those with two sialic acids (>35 fold). Glycine was observed to have a moderate boosting effect (<3 fold) on 2AB-labeled BSM O-glycans (see Figure 38B). The MS signal of small reduced O-glycans can be boosted up to 12 fold by glycine (see Figure 38C). However, an inhibitory effect on larger reduced O-glycans was observed. Figure 39 provides exemplary LC-FLR-MS conditions for the analysis of PROCA-labeled or reduced O-glycans. Figure 40 shows that monovalent glycans provide insufficient MS2 in the absence of glycine. Figure 41 shows that doubly charged ions from samples boosted with glycine provide significantly enhanced MS2 information.

[0098] Example 5: Facilitation of Site-Specific Glycosylation Profiling of Biologics by Ion-Pairing Hydrophilic Interaction Chromatography-Mass Spectrometry Using a Glycine Additive in Trifluoroacetic Acid-Containing Mobile Phase Many biotherapeutics, such as monoclonal antibodies (mAbs) and Fc domain fusion proteins, contain heterogeneous glycan contents at one or more glycosylation sites. Site-specific glycan profile characterization has been important for monitoring the quality of these molecules during different stages of drug development. Ion-pairing hydrophilic interaction chromatography (IP-HILIC) as an orthogonal separation method to reversed-phase liquid chromatography (RPLC) can achieve better separation between individual glycoforms and the identification of glycopeptides from non-glycosylated peptides. However, online IP-HILIC coupled with mass spectrometry detection may suffer from mass spectrometry signal suppression during electrospray ionization due to trifluoroacetic acid (TFA), which is generally used as an ion-pairing agent. In this example, optimized conditions for IP-HILIC-MS are reported, in which glycine is added to the TFA-containing mobile phase to eliminate the ion suppression effect of the ion-pairing agent, enhancing the MS detection sensitivity of glycopeptides by up to approximately 50-fold while still retaining excellent separation capabilities. With enhanced detection sensitivity, IP-HILIC-MS can identify an increase in the number of site-specific N-linked glycans for IgG1 and IgG4 mAbs as well as Fc domain fusion proteins (containing five N-glycosylation sites), and it is demonstrated that comparable quantitative results can be achieved compared to conventional methods using RPLC mass spectrometry (RPLC-MS). It is also demonstrated that low-level O-glycosylation and non-consensus N-glycosylation on mAbs can be identified using IP-HILIC-MS without concentration prior to LC-MS analysis.

[0099] I. Introduction Glycosylation is an important quality attribute of biotherapeutics, including monoclonal antibodies (mAbs) and Fc domain (fragment crystallization domain) fusion proteins. The Fc domain glycosylation profile at the conserved asparagine-297 site is strongly associated with effector functions such as antibody-dependent cell cytotoxicity (ADCC) and complement-dependent cytotoxicity (CDC), which can affect drug efficacy in oncological therapies. Fc N-glycosylation is not directly involved in the interaction with the target, but N- or O-linked glycans at non-standard sites, which are often located in the Fab (antigen-binding fragment) region of the mAb or the functional domain of the Fc-domain fusion protein, can have an adverse effect on the binding affinity to the target. Glycosylation in biotherapeutics also correlates with pharmacokinetic and pharmacodynamic profiles as well as other molecular properties such as charge heterogeneity, stability, and immunogenicity. Glycosylation of mAbs or Fc domain fusion proteins often exhibits diverse profiles in different protein expression systems, manufacturing processes, and protein sequences. Therefore, comprehensive characterization of site-specific glycan profiles involves a series of extremely important tasks, which include demonstrating the comparability of glycan profiles between different sample lots or investigating the root causes of glycosylation-related issues during the non-clinical development of biologics.

[0100] N-linked glycosylation profiling can be performed by successively releasing glycans from proteins via exoglycosidase treatment, labeling the reduced termini with a fluorescent reagent, and analyzing the released glycan mixture using hydrophilic interaction chromatography (HILIC-FLR-MS) combined with fluorescence and mass spectrometry detection. However, when a protein contains multiple glycosylation sites, information regarding site-specific glycosylation cannot be provided. Instead, direct analysis of intact glycopeptides can reveal site-specific glycosylation profiles for both N-linked and O-linked glycans. In a typical workflow for glycopeptide identification, proteins are digested with proteases and analyzed using reversed-phase chromatography (RPLC) combined with mass spectrometry. This is an approach used for the characterization of biopharmaceuticals to confirm the protein amino acid sequence and provide site-specific quantification during post-translational and chemical modifications, including glycosylation (often referred to as "peptide mapping"). The relative abundance of individual glycans can be quantified based on the peak areas of the extracted ion chromatograms (EICs) of the corresponding glycopeptides. However, RPLC-based glycopeptide separation depends mainly on the amino acid sequence since the glycan composition contributes little to the differences in hydrophobicity. As a result, (1) glycoforms from peptides with the same amino acid sequence elute as clusters of peaks with similar retention times, and (2) glycopeptides cannot be sufficiently distinguished from other non-glycosylated peptides when their peptide sequences have similar hydrophobicities. For example, the MS signals of low-abundance glycopeptides from non-canonical N-linked or O-linked glycosylation sites can be greatly suppressed in the presence of co-eluting high-abundance interfering species from other glycopeptides or non-glycosylated peptides due to limitations in the MS detection dynamic range and column loading capacity.

[0101] In contrast to RPLC, HILIC performs effective separation of glycans with different compositions and structural isomers. HILIC combined with mass spectrometry has been used to analyze N-glycans released from a single biotherapeutic agent or for glycomics analysis in different types of complex samples. Intact glycopeptide analysis using HILIC-MS performed under mild acidic conditions has also been reported for antibodies and other glycoproteins, which showed excellent separation for different glycoforms including glycan isomers. Glycopeptides can be better separated from non-glycosylated peptides by simply changing the HILIC mobile phase by adding 0.1% TFA, which provides an acidic environment (pH approximately 2) and strong ion-pairing properties (TFA anion). Under these conditions, the charged groups of the whole peptide can be neutralized, and the hydrophilicity of non-glycosylated peptides decreases significantly, while glycans are rich in uncharged polar moieties such as hydroxyl groups, so glycopeptides are less affected. Ion-pairing HILIC (IP-HILIC) separation has become a standard technique for offline glycopeptide enrichment in many glycoproteomics studies, but due to the harmful signal suppression caused by the ion-pairing agent TFA, it is rare to directly combine IP-HILIC with MS detection for glycopeptide identification, which can cause a 5- to 10-fold decrease in the MS signal compared to the same concentration of formic acid. Therefore, restoring unwanted ion suppression is important for expanding the application scope of the IP-HILIC-MS approach involving TFA.

[0102] Many efforts have been made over the past decades to reduce TFA-related MS signal suppression, ranging from the introduction of post-column "fixing solutions" to replacing TFA with other "weaker ion-pairing" agents. By directly adding glycine to the TFA-containing mobile phase, a method was discovered that resulted in a significant improvement of approximately one order of magnitude in the signal-to-noise ratio (S / N) in peptide mapping of mAbs using a pre-column C18 column. The glycine additive was introduced in front of the column, but did not affect the performance of peptide separation on the C18 column. In this example, it is also demonstrated that this solution can be applied to IP-HILIC-MS, thereby enabling the construction of a sensitive and sustainable platform for unbiased site-specific glycosylation profiling of monoclonal antibodies or Fc domain fusion proteins. A unique example is also disclosed where this platform can selectively enhance the possibility of identifying extremely low-abundance O-linked and non-consensus N-linked glycosylation in the Fab region of mAbs without the need for additional offline enrichment prior to LC-MS analysis.

[0103] II. Materials and Experiments Chemicals and Materials Four IgG4 monoclonal antibodies (mAb1, mAb2, mAb3, mAb4), an IgG1 monoclonal antibody (mAb5) and an Fc domain fusion protein (fP1) were manufactured at Regeneron (Tarrytown, NY). Ultra-high purity glycine (J.T. Baker brand), trifluoroacetic acid (sequencing grade), formic acid (sequencing grade) and acetonitrile (LC-MS grade) were purchased from Thermo Fisher Scientific (Waltham, MA). Ammonium formate (99%) was purchased from Acros Organics™. PNGase F was purchased from New England Biolabs (Ipswich, MA). GlycoWorks™ Rapid Deglycosylation Kit was purchased from Waters (Milford, MA). Ultra-pure water was generated by a Milli-Q System (Millipore, Burlington, MA). All other chemicals were purchased from Sigma-Aldrich (St. Louis, MO) unless otherwise specified.

[0104] Trypsin digestion To prepare the protein digest, the monoclonal antibody was denatured and reduced by heating at 80 °C for 10 min in a solution containing 5 mM acetic acid and 5 mM tris(2-carboxyethylphosphine hydrochloride). Each sample was then neutralized in 100 mM Tris buffer (pH 8.0) containing 15 mM iodoacetamide, followed by trypsin digestion at 37 °C in the dark for 2 h at an enzyme-to-carrier ratio of 1:20 (w / w). For the Fc domain fusion protein, the protein was reduced and denatured by heating at 80 °C for 10 min in the presence of 8 M guanidine-HCl and 5 mM dithiothreitol, followed by alkylation with 15 mM iodoacetamide. The sample was buffer-exchanged to 100 mM Tris, pH 8.0 using a NAP-5 Sephadex 5-25 column (GE Healthcare, Chicago, IL) and then digested under the same digestion conditions as the monoclonal antibody. The digested peptides were further purified through a Sep-Pak C18 cartridge (Waters) according to the vendor-provided protocol. The sample was dried under vacuum and reconstituted in 80% ACN (for HILIC-MS) or water (for RPLC-MS).

[0105] LC-MS analysis of glycopeptides All LC-MS experiments were performed using an Acquity UPLC I-Class System (Waters) coupled to a Q-Exactive Plus Hybrid Quadrupole-Orbitrap mass spectrometer equipped with a heated electrospray ionization (HESI) source (Thermo Fisher Scientific). To minimize the formation of alkali adducts, all samples were transferred to polypropylene vials made for injection, the mobile phase solutions were prepared, stored in polyethylene bottles, and the LC lines were thoroughly pre-washed. Mobile phase A (MPA) is a pure aqueous phase containing 100% water (v / v), 0.1% TFA or FA (v / v), and 1 mM glycine. Mobile phase B (MPB) was composed of 80% acetonitrile (v / v), 20% water (v / v), 0.1% TFA or FA (v / v), and 1 mM glycine. Glycine-free versions of MPA and MPB were also prepared using the exact same recipe with equal volumes of water instead of adding glycine.

[0106] Before the first sample injection after switching the mobile phase between the glycine-containing and glycine-free versions, the LC system was conditioned for at least 1 hour and the intensity of the MS peak of protonated glycine at 76.07 m / z was monitored for quality control purposes (scan range 50 - 750 m / z). For the glycine-free and glycine-containing mobile phases, stable MS signals were expected to reach normalized intensities at 1e 6 and 1e 9 respectively.

[0107] For HILIC-MS analysis, 6 μg (or more if annotated) of desalted tryptic digested peptides were loaded onto a Waters Acquity UPLC Glycan BEH Amide column (130 Å, 1.7 μm, 2.1 mm × 150 mm). The flow was initiated with 99.9% MPB at 0.2 mL / min, and glycan-containing peptides were eluted and separated when the proportion of MPB decreased from 90% to 62.5%. For RPLC-MS analysis, the sample was loaded onto a Waters Acquity UPLC BEH C18 column (130 Å, 1.7 μm, 2.1 mm × 150 mm). The mobile phase settings for RPLC-MS analysis were exactly the same as those for HILIC-MS, but conversely, the flow was initiated with 99.9% MPA, and the peptides were eluted when the proportion of MPB increased from 0.1% to 40%. Full MS scans were collected at 500 - 2000 m / z to avoid the glycine signal (resolution = 70,000, AGC target = 1e 6 , maximum IT = 100 ms, sheath gas = 40, auxiliary gas = 10, sweep gas = 0, spray voltage = 3.8 kV, capillary temperature = 350 °C, auxiliary gas heater temperature = 250 °C, S-lens RF level = 50). The five most abundant precursors were selected for data-dependent MS2 scans, where NCE = 27, resolution = 17,500, AGC target = 5e 5 , and maximum IT was set to 250 ms.

[0108] Data processing for glycopeptide identification Glycopeptide identification was performed by searching raw files against a protein sequence and a built-in glycan database containing 132 human N-linked glycans or 70 common O-linked glycans using the Byonic software of the Protein Metrics suite. For non-consensus N-glycan searches, the same N-glycosylation database was customized to remove site restrictions according to the vendor-provided technical notes. A preliminary list of unique glycopeptides was generated by filtering against a 1% FDR. The list of precursors and the original search results as a spectral library were then imported into Skyline Daily software (University of Washington, WA) for final ID verification and quantification based on full-scan by automated feature extraction and peak integration.

[0109] HILIC LC-MS Analysis of Derivatized Glycans To prepare samples for released N-linked glycan analysis, proteins were denatured and reduced by heating at 80 °C for 10 min in a solution containing 0.1% RapiGest™ SF (Waters) and 4.2 mM tris(2-carboxyethyl)phosphine hydrochloride (TCEP-HCl). Each sample was then deglycosylated by the addition of PNGase F at an enzyme-to-substrate ratio of 1:5 (w / w) and incubation at 45 °C for 25 min to release oligosaccharides, followed by derivatization of the released glycans with the RapiFluor™-MS reagent (Waters) fluorescent tag by incubation at 45 °C for 25 min. The derivatized samples were diluted into a final solution containing 25% N,N-dimethylformamide and 53% acetonitrile (v / v).

[0110] Data acquisition was performed using an Acquity UPLC I-Class System (Waters) coupled to a Q-Exactive Plus Hybrid Quadrupole-Orbitrap mass spectrometer (Thermo Fisher Scientific). 1 μg of the released and derivatized glycan was loaded onto an Acquity UPLC Glycan BEH Amide column (130 Å, 1.7 μm, 2.1 mm × 150 mm) (Waters). Mobile phase A was a pure aqueous phase containing 50 mM ammonium formate in water, with pH = 4.4. Mobile phase B was a pure organic phase (100% acetonitrile). The gradient started at 25% mobile phase A and was subsequently increased to 32.2% of mobile phase A in order to elute all the derivatized glycans. The MS parameters were set as follows: full scan m / z range = 650 - 2000, ACG target = 1e 6 , maximum IT = 100 ms, resolution = 70,000, source temperature = 350 °C, spray voltage = 4.0 kV, auxiliary gas heater temperature = 250 °C, S-lens RF level = 50. The five most abundant precursors were selected for data-dependent MS2 scans, where ACG target = 1e 5 , maximum IT = 250 ms, stepped NCE = 13, 20, resolution = 17,500.

[0111] The monosaccharide composition of the glycans was assigned based on the experimental masses measured for each glycan. The structure of the glycans was assigned based on the match of the MS / MS fragmentation spectra to the theoretical fragmentation patterns predicted by the glycan structures in the UniCarbKB database.

[0112] III. Results and Discussion Identification and relative quantification of the glycoforms of monoclonal antibodies.

[0113] Trypsin digestion of IgG4 molecules yielded conserved N-glycosylation sites with variable glycans (

Chem.

[0114] A similar chromatographic profile is observed for another IgG4 molecule (mAb2). As shown in FIGS. 42C1-42C3, the peak resolution in the glycopeptide region can be independently improved by extending the linear gradient (from 92% to 73% of mobile phase B). The recovered S / N in the presence of glycine also ensures the quality of the tandem mass spectra (exemplified in FIGS. 42D and 42E) and enables the reliable assignment of glycopeptides without referring to other assays such as released glycan analysis. As shown in FIG. 42F, the extracted ion chromatograms of the 18 most abundant glycoforms of mAb1 from the data obtained with the optimized gradient are uniformly distributed within 40 minutes. Each EIC peak representing a different glycan composition has a highly predictable elution time. For example, the addition of galactose (Gal) always caused a 7-minute increase in elution time regardless of its glycan composition. It can be a fucosylated core structure or a non-fucosylated core structure with any number of branched N-acetylglucosamine (GlcNAc). In addition to the well-resolved separation of glycans with different compositions, baseline separation was also observed for the structural isomers resulting from the binding of Gal to branched GlcNAc (e.g., FA2G1, A2G1) via either a 1,3 or 1,6 linkage. Notably, the elution profiles observed from intact glycopeptides are very consistent with the elution profiles of the released glycans obtained using the same HILIC column under nearly neutral pH, except for the two sialic acid-containing glycans FA2G2S1 and FA2G1S1 (FIG. 42G), and these relative elution times are most likely due to the protonation of carboxylic acid groups in the presence of TFA.

[0115] In contrast to the wide elution time range in IP-HILIC, the EIC peaks of the same glycopeptides from the RPLC-MS dataset elute only within a 1.5-minute window, even though the entire linear range of the gradient is 80 minutes (Figures 42H1 - 42H2). Due to the high scanning speed and sensitivity in current state-of-the-art mass spectrometers, such deficiencies in peak separation can still be used for glycopeptide identification, but improved separation always provides advantages in reducing the dynamic range of co-eluting peptides and achieving more accurate peak integration. Additionally, artificial glycopeptides generated during in-source fragmentation have the same retention time as the larger glycoforms from which they are generated, so they can be easily excluded for the same existing glycopeptides in IP-HILIC because of different retention times, but it can be difficult in RPLC due to inappropriate separation (Figures 42H1 - 42H2).

[0116] The level of S / N enhancement for glycine can vary for different peptide sequences or different peptide-to-glycine molar ratios. For a given sequence (EEQFNSTYR; SEQ ID NO: 45), all glycopeptides show a similar S / N improvement independent of glycan composition and the relative abundance of individual glycoforms, as shown by the excellent linearity (R 2 = 0.998) when plotting the EIC peak areas for the individual glycoforms of mAb1 under two conditions (Figure 43A). The consistency in signal boost results in approximately equal percentage levels for different glycoforms quantified under two conditions, as shown in Figures 43B and 43C. This result suggests that TFA suppresses the signal mainly by forming ion pairs between TFA anions and primary amines, which can only be found in amino acids rather than in sugar chains.

[0117] The average S / N boost for glycopeptides in IgG4 is determined from the intercept of the plot in Figure 43A and is approximately 19.4-fold. Glycopeptides derived from IgG1 were also analyzed, and the peptide sequence (

Chemical formula

[0118] Characterization of the Site-Specific N-Glycosylation Profile of Fc Domain Fusion Proteins Containing Multiple Glycosylation Sites

[0119] The fusion protein fP1 contains five N-glycosylation sites, including one conserved Fc glycosylation site (site 1) corresponding to N297 from the mAb and four additional sites (sites 2-4) located in the functional domains. For the IP-HILIC-MS analysis of the digested peptides, the enhanced signal-to-noise ratio in the presence of glycine is crucial for improving glycopeptide identification, which depends on the yield of fragment ions in the MS2 spectrum (see FIGS. 43F and 43G for examples of tandem mass spectra). To systematically evaluate how the glycine additive improves the identification of site-specific glycopeptides for fP1, as shown in FIG. 44A (the upper two panels), the search results from two conditional runs were combined and filtered using relatively stringent parameters (score > 150, |LogProb| > 2, #PSM ≧ 2), and it was shown that the numbers of both high-confidence PSMs and identified glycopeptides were improved with the glycine additive. The glycosylation profiles were observed to be different at each site, which was also verified by relative quantification using the EIC peak areas (Table 5).

Table 5-1

Table 5-2

Table 5-3

[0120] Similar to mAbs, signal boosting is highly consistent for all glycoforms from the same glycosylation site (Figure 44C), and the relative ratios of the quantified individual glycoforms are equivalent with and without glycine additive (Table 5). It is also worth noting that for glycopeptides containing different glycosites, signal boosting varies in the range of 2-fold (site 5) to 50-fold (site 2). This indicates that the levels of TFA-based signal suppression or glycine-based TFA relaxation vary greatly for different peptide sequences.

[0121] In addition to a sufficient signal-to-noise ratio, sufficient LC separation is also advantageous for site-specific glycosylation identification in highly glycosylated proteins. All glycopeptide peaks can be well resolved from non-glycosylated peptides by IP-HILIC separation and are revealed by the EIC peaks of oxonium signature ions in the MS2 spectra from glycopeptides (Figure 44D). To examine how these glycopeptides separate from each other, the EIC peaks of four representative glycoforms (FA2G1, FA2G2, FA2G2S1, and FA2G2S2) are extracted for all seven glycosite-containing peptides as the seven peak “envelopes” shown in Figure 45A. The elution profiles within one envelope are nearly equivalent to those within other envelopes, including the elution order of glycoforms and the retention time differences between each peak. And the retention time differences between two envelopes of glycopeptides only correlate with the amino acid sequence. This is probably due to the frequency of polar residues, as suggested in multiple models for predicting peptide retention in HILIC. Descriptively, the sequence of the peptide at site 4 (

Chemical formula

[0122] Such features observed in IP-HILIC separations using TFA may not be reproduced when using weaker acids that do not have strong ion-pairing properties such as formic acid. First, HILIC / FA-based separations can result in incomplete separation of non-glycosylated peptides from glycopeptides, and the same extracted representative peaks showed other early-eluting subpopulations indicative of the heterogeneous hydrophilicity of different glycopeptides due to the lack of an ion-pairing reagent (Figure 44D). This defect in the LC performance of HILIC / FA compared to HILIC / TFA separation results in fewer glycopeptide identifications, as shown in the third panel of Figure 44A and the Bengraph of Figure 44B, but the addition of glycine can still achieve an average signal boost of about 5-fold, and the identified glycopeptides have a greater number of PSMs compared to the TFA dataset. HILIC performance in a FA-containing mobile phase can be obtained using other HILIC stationary materials that may further improve glycopeptide identification when combined with HILIC / FA with a glycine additive. RPLC / TFA using a C18 column retains the orthogonal resolving power to separate glycopeptides from multiple sites. Glycopeptides from multiple sites are well separated from each other as distinct clusters (Figure 36D). They are not completely isolated from other non-glycosylated peptides, but the identification of glycopeptides by RPLC-MS remains comparable to the IP-HILIC-MS method because there are few non-glycosylated peptides present in this single glycoprotein fP1 eluting in these glycopeptide clusters (Figure 44A, last two panels and Figure 44B).

[0123] Discovery of low-abundance O-glycosylation and non-standard N-glycosylation in monoclonal antibodies using IP-HILIC-MS.

[0124] Since IP-HILIC has shown the ability to characterize glycopeptides from multiple glycosites in fP1, it should also have the potential to identify glycopeptides from non-standard glycosites for mAbs. The sample injection volume was slightly increased to improve the detection ability of low-abundance glycopeptides, and thus high-quality tandem mass spectra were obtained. In the study of mAb3 from the early stage of drug development, miscleaved N297-containing glycopeptides (e.g., TKPREEQFNSTYR; SEQ ID NO: 44), as well as glycopeptides containing non-standard glycosites of N91 located in the VL domain and N163 located in the CH1 domain (see examples of tandem mass spectra in FIGS. 45B and 45C), a plurality of low-abundance glycopeptides with different amino acid sequences have been reliably identified. Different from the standard Fc N297 site mainly occupied by glycans with a fucosylated branched complex structure, three high-mannose type glycans were identified at light chain N91, and only one glycan FA2G2 was identified at heavy chain N163, and both sites have low glycan occupancies of 0.4% and 0.07%, respectively. By injecting a large amount of sample, the possibility of identifying these low-abundance glycopeptides can be obtained, but due to their signal saturation, the abundance of non-glycosylated peptides may be underestimated. When the sample injection volume is low, for each individual glycan, a (presumably more accurate) occupancy level is quantified at a low level (FIG. 45D). In another study, when searching against the most frequently occurring mammalian O-linked glycan database (see examples of tandem mass spectra in FIGS. 45E and 45F), the low-abundance mucin-type O-linked glycan HexNAc(1)Hex(1)NeuAc(2) is identified in the VH domain of one arm for the bispecific antibody mAb4, but the level of this O-linked glycopeptide is quantified to be less than 0.05%. The presence of these rare glycans on the Fab region was also confirmed at the intact protein level using strong cation exchange chromatography combined with native mass spectrometry, and no other glycans were found on the Fab region.

[0125] These rare glycans cannot be discovered independently using RPLC-MS, but there are not yet enough cases to statistically demonstrate the advantage of using IP-HILIC-MS to identify low-abundance non-standard N-glycans or O-glycans. Instead, in the investigation of low-abundance glycopeptides, insights into the effects from potential interferences may be obtained. The detection of very low-abundance (glyco)peptides is usually difficult, especially for data-dependent acquisition in an Orbitrap mass spectrometer, in the presence of very abundant co-eluting interferences. This is because the C-trap can be rapidly filled by very abundant ions, and low-abundance ions may not be sufficiently accumulated within the shortened injection time and may not generate signals above the threshold for triggering tandem MS2 for identification. Even when neither HILIC nor RPLC can completely remove the interfering signals, the main causes and levels of interference may be different. For RPLC-MS, non-glycosylated peptides are widely distributed throughout the gradient and can potentially cause interference to any low-abundance glycopeptides (Figure 46A), while for IP-HILIC-MS, there is probably no interference from non-glycosylated peptides, and the entire population is compressed to the initial elution time due to the ion pairs of TFA, and the only source of interference is from highly abundant canonical Fc glycopeptides (Figure 46B). It is hypothesized that this Fc glycopeptide interference should have a lower negative effect compared to non-glycosylated interference in RPLC-MS for low-abundance glycopeptide identification, as (1) the number of Fc glycopeptides is much less than that of non-glycosylated peptides, and (2) the intensity of Fc glycopeptides is usually the lowest among all peptides digested from mAbs. Indeed, only two or three major glycoforms such as FA2, FA2G1, and FA2G2 in mAb3 (relative abundance >10%) can be identified as true interferences, and the remaining Fc glycoforms have minimal concerns due to their low abundance.

[0126] As shown in Fig. 46C, out of the representative 6-minute window (67 - 73 minutes) in RPLC, a total of 2.7 minutes has MS signals from high-abundance peptides (67.5 - 68 minutes, 69 - 70 minutes, 70.4 - 70.6 minutes, and 71 - 72 minutes), creating a high risk of co-elution with any low-abundance glycopeptides. It is observed that the entire cluster of three high-mannose glycoforms in HC N91 co-elutes at a peak of 250-fold intensity (69.0 - 69.4 minutes). This situation can occur anywhere in the 80-minute chromatogram. In IP-HILIC, there are two glycopeptides (Man6 and Man7) from N91 that partially co-elute with the Fc glycopeptide, but the intensity of interference from the Fc glycopeptide is only 50-fold and 8-fold respectively. Out of the entire 10-minute glycopeptide elution region (12 - 22 minutes), only 1 minute (the total peak width for the three major glycoforms) has strong interference peaks, meaning that 90% of the region contains a "valley" region (12.5 - 16.5 minutes in Fig. 46B) where only background noise and other low-abundance glycopeptides are observed, and low-abundance glycopeptides cannot have severe interference.

[0127] Similarly, the O-linked glycopeptides on the VH domain of mAb4 also elute in the same region in IP-HILIC-MS. In contrast, the EIC peak cannot be extracted from the RPLC-MS dataset, probably due to an increase in the number of non-glycosylated peptides for the bispecific mAb (Fig. 46E). Since O-glycans usually contain fewer monosaccharide units compared to N-glycans, O-glycopeptides are less hydrophilic and are likely to elute earlier than N-glycopeptides. Any other O-glycopeptides from mAbs are assumed to elute in a "valley" region with a minimal interference signal from either non-glycosylated peptides or Fc-glycosylated peptides.

[0128] This reduction of interference makes IP-HILIC-MS a unique approach for the identification of low-abundance glycopeptides, highlighting the irreplaceable roles of TFA as the ion-pairing reagent and glycine as the signal-boosting reagent in this application. This approach can be used as a means for rapid screening of rare glycosylation during early-stage biopharmaceutical discovery without the need for second-dimension glycopeptide enrichment. The performance can be further optimized by using a long gradient. Additionally, since most of the critical quality attributes (CQAs) with unpredictable molecule dependencies can be located in the Fab domain rather than the Fc domain, the focus is on the Fab region isolated by complete removal of the Fc domain and Fc N-glycopeptide interference, resulting in further improved detection sensitivity.

[0129] In this study, it was demonstrated that adding glycine to the TFA-containing mobile phase can significantly solve the sensitivity deficiency in TFA-based IP-HILIC-MS without adversely affecting the LC performance of peptide separation. This method is based on a conventional flow pump, shows excellent stability and robustness, and enables site-specific glycosylation profiling for numerous different types of biotherapeutics and glycosylated functional proteins. For mAbs, IP-HILIC-MS generates a bias-free glycan profile at the intact glycopeptide level compared to released glycan analysis, suggesting that this approach can be complementary or alternative to the released glycan assay in standard protein characterization. Additionally, IP-HILIC-MS can be compatible with MRM- or PRM-based methods for glycopeptide quantification, and due to the wide elution time range of glycopeptides, a relatively small number of precursors need to be scheduled simultaneously. These targeted or untargeted IP-HILIC-MS methods can be smoothly transplanted into multi-attribute monitoring workflows or high-throughput analysis platforms, which is promising for the advancement of analytical science for the biopharmaceutical industry.

[0130] Glycine additives can be used in RPLC-MS or HILIC-MS with weak ion pair formation to boost MS signals and increase overall site-specific glycosylation identification. From the disclosed results regarding Fc domain fusion proteins, it is demonstrated that the IP-HILIC-MS method can be nearly equivalent (at the same signal intensity) to the RPLC-MS method with respect to mapping glycosylation profiles from multiple glycosylation sites. Due to the high scan speed and high dynamic range of current state-of-the-art mass spectrometers, interference from non-glycosylated peptides can be well tolerated in RPLC-MS. However, for low occupancy glycosites such as non-standard sites in mAbs, IP-HILIC still shows the advantage of improving their detectability even without additional enrichment steps. Also, it is reasonable to assume that IP-HILIC-MS functions much better than RPLC-MS when applied to samples containing a large amount of non-glycosylated peptide background. Selectively extending the linear gradient in the glycopeptide elution region in IP-HILIC can efficiently increase glycopeptide separation that was hardly achievable in RPLC-MS. Therefore, targeted or untargeted methods for rapid and easy glycoform screening on the glycoproteome scale, such as in human serum, can be implemented using the IP-HILIC-MS (UPLC) platform. In addition to neutral HILIC stationary phase materials (amides), this study reports that other charged stationary phases such as zwitterionic materials can be evaluated.

[0131] The invention should not be limited in scope by the specific embodiments described herein. Indeed, various modifications of the invention in addition to those described herein will be apparent to those skilled in the art from the foregoing description and the accompanying drawings. Such modifications are intended to be included within the scope of the appended claims.

Claims

1. A method for enhancing a mass spectrometry signal, comprising: contacting a sample with a separation column under conditions that allow sample components to bind to a carrier; applying a first mobile phase gradient to the separation column, the first mobile phase gradient comprising trifluoroacetic acid (TFA) and a small molecule additive or formic acid (FA) and a small molecule additive; the small molecule additive in the first mobile phase being selected from glycine, alanine, serine, valine, N-acetylglycine, methionine, β-alanine, aspartic acid, or N-methylglycine; applying a second mobile phase gradient to the separation column, the second mobile phase gradient comprising TFA in acetonitrile (ACN) and a small molecule additive or FA in ACN and a small molecule additive; the small molecule additive in the second mobile phase being selected from glycine, alanine, serine, valine, N-acetylglycine, methionine, β-alanine, aspartic acid, or N-methylglycine; performing mass spectrometry on the eluted sample components; and a method comprising the steps of:

2. The method according to claim 1, wherein the small molecule additive in the first mobile phase is selected from glycine, alanine, serine, or valine.

3. The method according to claim 2, wherein the small molecule additive in the first mobile phase is glycine.

4. The method according to any one of claims 1 to 3, wherein the concentration of the small molecule additive is 1 mM to 2 mM.

5. The method according to claim 3, wherein the glycine concentration is 1 mM.

6. The method according to claim 3, wherein the glycine concentration is 2 mM.

7. The method according to claim 1, wherein the small molecule additive in the second mobile phase is selected from glycine, alanine, serine, or valine.

8. The method according to any one of claims 1 to 7, wherein the small molecule additive in the second mobile phase is glycine.

9. The method according to claim 7 or 8, wherein the concentration of the small molecule additive is 1 mM to 2 mM.

10. The method according to claim 8, wherein the glycine concentration is 1 mM.

11. The method according to claim 8, wherein the glycine concentration is 2 mM.

12. The TFA concentration in the first mobile phase is 0.05% to 0.1% TFA in H 2 O, or the FA concentration in the first mobile phase is 0.1% FA. The method according to any one of claims 1 to 11.

13. The TFA concentration in the second mobile phase is 0.05% TFA in 80% ACN and 20% H 2 O or 0.1% TFA in 80% ACN and 20% H 2 O. The method according to any one of claims 1 to 12.

14. The sample contains a peptide or a nucleotide. The method according to any one of claims 1 to 13.

15. The method according to claim 14, wherein the peptide is a glycopeptide.

16. The method according to claim 15, wherein the glycopeptide is obtained from a monoclonal antibody.

17. The method according to claim 16, wherein the monoclonal antibody is of isotype IgG1, IgG2, IgG3, IgG4, or a mixed isotype.

18. The method according to any one of claims 1 to 17, further comprising preparing the sample before contacting the sample with the separation column under conditions that allow the sample components to bind to the carrier. **Claim 19** Preparing the sample comprises: contacting the sample with a denaturing and reducing solution under conditions that allow denaturation and reduction of the sample; contacting the denatured and reduced sample with an alkylating solution under conditions that allow alkylation of the sample; contacting the alkylated sample with a digestion solution under conditions that allow digestion of the sample; contacting the digested sample with a quenching solution under conditions that stop digestion of the sample, the method according to claim 18. **Claim 20** The method according to claim 19, wherein the sample is a monoclonal antibody and the digestion solution contains a protease. **Claim 21** The method according to claim 20, wherein the protease contains trypsin. **Claim 22** The method according to any one of claims 1 to 21, wherein the separation column is a liquid chromatography column. **Claim 23** The method according to claim 22, wherein the liquid chromatography (LC) separation column includes a hydrophilic interaction (HILIC) liquid chromatography column. **Claim 24** Performing mass spectrometry on the eluted sample components includes applying electrospray ionization to generate charged ions from the eluted sample components and measuring the generated charged ions, the method according to any one of claims 1 to 23. **Claim 25** The method according to any one of claims 1 to 24, enhancing the mass spectrum signal, as indicated by an average increase of 5 to 14-fold and / or 2 to 1000-fold at high charge state species (z≥3). Claim 26 The mass spectral signal obtained for the eluted sample component is enhanced 2 to 50 times relative to the mass spectral signal obtained for a control sample in the absence of the small molecule additive, the method according to any one of claims 15 to 17. Claim 27 The glycopeptide is an O-glycan-containing glycopeptide, the method according to claim 26. Claim 28 The glycopeptide is an N-glycan-containing glycopeptide, the method according to claim 26. Claim 29 The O-glycan or N-glycan is linked to a label, the method according to claim 27 or 28. Claim 30 The label is procainamide, the method according to claim 29. Claim 31 The small molecule additive is glycine, the method according to any one of claims 26 to 30.

Images