GLYCAN OXONIUM ION PROFILING OF GLYCOSYLATED PROTEINS
20210333288 · 2021-10-28
Inventors
Cpc classification
G01N2440/38
PHYSICS
International classification
Abstract
Methods disclosed herein include an improved technique for comparing a glycosylation profile of a first protein (e.g., an innovator protein drug) with a glycosylation profile of a second protein (e.g., a corresponding biogeneric/biosimilar). For example, a method of manufacture can include providing or obtaining a batch of a test glycoprotein drug substance, using mass spectrometry to acquire a test oxonium ion profile from a sample of the test glycoprotein drug substance batch, comparing the test oxonium ion profile to a corresponding target oxonium ion profile of a target glycoprotein drug product, and processing the batch of the test glycoprotein drug substance as a drug product if the difference between the test oxonium ion profile and the corresponding target oxonium ion profile is tolerable, or taking an alternative action if the difference between the test oxonium ion profile and the target oxonium ion profile is not tolerable.
Claims
1. A method of manufacture, comprising: providing or obtaining a batch of a test glycoprotein drug substance; using mass spectrometry to acquire a test oxonium ion profile from a sample of the test glycoprotein drug substance batch; comparing the test oxonium ion profile to a corresponding target oxonium ion profile of a target glycoprotein drug product; and processing the batch of the test glycoprotein drug substance as a drug product if the difference between the test oxonium ion profile and the corresponding target oxonium ion profile is tolerable; or taking an alternative action if the difference between the test oxonium ion profile and the target oxonium ion profile is not tolerable.
2. The method of claim 1, wherein the test glycoprotein drug substance comprises a glycoprotein that has an amino acid sequence that is identical to a glycoprotein of the target glycoprotein drug product.
3. The method of claim 1 or 2, wherein the test glycoprotein drug substance comprises an Fc fusion protein or an antibody.
4. The method of any one of the preceding claims, wherein the target glycoprotein drug product is approved under a primary approval process.
5. The method of any one of the preceding claims, wherein the target glycoprotein drug product is approved under a BLA.
6. The method of any one of the preceding claims, wherein using mass spectrometry comprises digesting the sample to produce a plurality of glycopeptides and/or glycans.
7. The method of any one of the preceding claims, wherein using mass spectrometry comprises performing liquid chromatography-tandem mass spectrometry.
8. The method of any one of the preceding claims, wherein using mass spectrometry comprises performing data-independent mass spectrometry.
9. The method of any one of the preceding claims, wherein using mass spectrometry comprises performing all ion fragmentation.
10. The method of any one of the preceding claims, wherein the test oxonium ion profile comprises one or more MS signals associated with levels of oxonium ion-containing fragments that include the following oxonium ions: (i) HexNAc internal fragment; (ii) Hex; (iii) HexNAc; (iv) sialic acid; (v) sialic acid+H.sub.2O; (vi) Hex+HexNAc; or (vii) combinations thereof.
11. The method of any one of the preceding claims, wherein the test oxonium ion profile comprises one or more MS signals associated with levels of oxonium ion-containing fragments that include oxonium ions having the following m/z values as well as any associated isotope peaks: (i) m/z 138.06 (monoisotopic), 139.06 (+1 isotope), 140.06 (+2 isotope); (ii) m/z 163.06 (monoisotopic), 164.06 (+1 isotope), 165.06 (+2 isotope); (iii) m/z 204.09 (monoisotopic), 205.09 (+1 isotope), 206.09 (+2 isotope); (iv) m/z 274.09 (monoisotopic), 275.10 (+1 isotope), 276.10 (+2 isotope); (v) m/z 292.10 (monoisotopic), 293.11 (+1 isotope), 294.11 (+2 isotope); (vi) m/z 366.14 (monoisotopic), 367.14 (+1 isotope), 368.14 (+2 isotope); or (vii) combinations thereof.
12. The method of any one of the preceding claims, further comprising producing a representation of the comparison of the test oxonium ion profile to the target oxonium ion profile.
13. The method of any one the preceding claims, further comprising using mass spectrometry to acquire a target oxonium ion profile of the target glycoprotein drug product.
14. The method of any one of the preceding claims, wherein the difference between the test oxonium ion profile and the corresponding target oxonium ion profile is tolerable if the difference between a level of at least one oxonium ion-containing fragment derived from the test oxonium ion profile and a level of at least one oxonium ion-containing fragment derived from the corresponding target oxonium ion profile is less than a predetermined value.
15. The method of any one of claims 1-13, wherein the difference between the test oxonium ion profile and the corresponding target oxonium ion profile is not tolerable if the difference between a level of at least one oxonium ion-containing fragment derived from the test oxonium ion profile and a level of at least one oxonium ion-containing fragment derived from the corresponding target oxonium ion profile is greater than a predetermined value.
16. The method of any one of claims 1-13, wherein the test oxonium ion profile comprises one or more MS signals that are each associated with a level of an oxonium ion-containing fragment and the target oxonium ion profile comprises one or more MS signals that are each associated with a level of an oxonium ion-containing fragment.
17. The method of claim 16, wherein the difference between the test oxonium ion profile and the corresponding target oxonium ion profile is tolerable if the difference between 2, 3, 4, 5, or more MS signals of the test oxonium ion profile and 2, 3, 4, 5 or more corresponding MS signals of the target oxonium ion profile is each less than a predetermined value.
18. The method of claim 16, wherein the difference between the test oxonium ion profile and the corresponding target oxonium ion profile is not tolerable if the difference between 1, 2, 3, 4, 5, or more MS signals of the test oxonium ion profile and 1, 2, 3, 4, 5 or more corresponding MS signals of the target oxonium ion profile is each greater than a predetermined value.
19. The method of any one of claims 14-15 and 17-18, wherein the predetermined value is equivalent to the variability in oxonium ion profiles determined for three or more distinct batches of the target glycoprotein.
20. The method of any one of claims 14-15 and 17-18, wherein the predetermined value is 20%, 15%, 10% or 5%.
21. The method of any one of the preceding claims, wherein the alternative action comprises disposing of the batch of the test glycoprotein drug substance, classifying for disposal the batch of the test glycoprotein drug substance, labeling the batch of the test glycoprotein drug substance for disposal, reprocessing the batch of the test glycoprotein drug substance or a combination thereof.
22. The method of any one of the preceding claims, wherein the processing step comprises: (i) formulating the batch of the test glycoprotein drug substance; (ii) combining the batch of the test glycoprotein drug substance with a second component, e.g., an excipient or buffer; (iii) changing the concentration of the batch of the test glycoprotein drug substance in the drug product; (iv) lyophilizing the batch of the test glycoprotein drug substance; (v) combining a first and second aliquot of the batch of the test glycoprotein drug substance to provide a third, larger aliquot; (vi) dividing the batch of the test glycoprotein drug substance into smaller aliquots; (vii) disposing the batch of the test glycoprotein drug substance into a container, e.g., a gas or liquid tight container; (viii) packaging the batch of the test glycoprotein drug substance; (ix) associating a container comprising the batch of the test glycoprotein drug substance with a label (e.g., labeling); (x) shipping or moving the batch of the test glycoprotein drug substance to a different location; or (xi) a combination thereof.
Description
BRIEF DESCRIPTION OF THE DRAWINGS
[0068]
[0069]
[0070]
[0071]
[0072]
[0073]
[0074]
[0075]
[0076]
[0077]
[0078]
[0079]
[0080]
[0081]
[0082]
[0083]
[0084]
[0085]
[0086]
[0087]
[0088]
DETAILED DESCRIPTION
[0089] Protein glycosylation plays an important role in a variety of cellular functions, and many protein-based biotherapeutics contain sites along the protein backbone where heterogeneous glycan moieties reside. (Moremen, K. W., et al., Nature Reviews Molecular Cell Biology (2012) 13, 448-462; Spiro, R. G., Glycobiology (2002) 12, 43R-56R). For example, modulation of effector functions via Fc glycosylation has been shown to affect target cell killing mechanisms such as antibody-dependent cellular cytotoxicity (ADCC) and complement-dependent cytotoxicity (CDC). (Jefferis, R., Nature Reviews Drug Discovery (2009) 8, 226-234; Natsume, A., et al., Drug Des Devel Ther (2009) 3, 7-16; Kellner, C., et al., Methods (2014) 65, 105-113). Because glycosylation is a determinant of the function and efficacy for therapeutic proteins, characterization of glycoforms and glycosylation sites of therapeutic proteins is an important part of drug development.
[0090] Monoclonal antibodies are the most commonly prescribed biotherapeutic, and usually contain only one N-glycosylation site on the Fc-domain of the protein. For analyzing such antibodies, reducing end modification of enzymatically released N-glycans, followed by high performance liquid chromatography (HPLC) remains a popular technique for glycan characterization. Many important classes of biologics, however, contain multiple sites of both N- and O-linked glycosylation. (Zhu, L., et al. (2014) Taylor & Francis, 1474-1485; Houel, S., et al., Anal. Chem. (2013) 86, 576-584; Larsen, C. P., et al., American Journal of Transplantation (2005) 5, 443-453; Balaratnasingam, C., et al., Clinical ophthalmology (Auckland, NZ) (2015) 9, 2355). For example, many of the marketed Fc-fusion proteins contain five or more glycosylation sites. (Zhu (2014); Houel (2013); Larsen (2005); Balaratnasingam (2015); Berry, J. D., Therapeutic Fc-Fusion Proteins, 217-232). The emergence of antibody and antibody-like molecules with multiple glycan attachment sites has rendered glycan analysis increasingly more complicated with the presence of each additional glycosylation site.
[0091] Analyzing multiple glycan attachment sites using established glycopeptide methodologies presents challenges. For instance, by solely characterizing these complex molecules by glycan release-based methods, all glycoforms from all sites become pooled together, thus all information on glycosylation site-specificity is lost. Furthermore, 0-glycans still prove to be difficult to remove from the protein by both enzymatic and chemical procedures. (Takahashi, K., et al., Molecular & Cellular Proteomics (2010) 9, 2545-2557; Wada, Y., et al., Molecular & Cellular Proteomics (2010) 9, 719-727; Christiansen, M. N., et al., Anal. Chem. (2010) 82, 3500-3509).
[0092] In addition, established glycopeptide methodologies have generally utilized a priori knowledge of the glycosylation states of the investigated protein(s), database searching of results generated from data-dependent liquid chromatography-tandem mass spectrometry workflows, and extracted ion quantitation of the individual identified species. The inherent complexity of glycosylation, however, makes predicting all glycoforms on all glycosylation sites extremely challenging. As a result, such analyses are typically limited to known glycans.
[0093] A common methodology for assessing site-specific glycosylation is by liquid chromatography-tandem mass spectrometry (LC-MS/MS). Typically, unknown non-glycosylated peptides and glycopeptides are fragmented separately by various MS/MS techniques (see, e.g.,
[0094] While this experimental workflow can work well in certain applications, complete characterization of all glycoforms and glycosylation sites can be challenging because it was very difficult to predict all glycosylated species, and insufficient fragmentation patterns of glycopeptides hinders direct identification from MS/MS spectra. A substantial number of false positives and true negatives have been observed due to, e.g., the fact that not all glycopeptide species are contained in reference databases and poor fragmentation patterns may be generated, even when more advanced MS/MS techniques, such as Electron Transfer Dissociation (ETD) and ultraviolet photodissociation (UVPD), are employed. (Desaire, H. Molecular & Cellular Proteomics (2013) 12, 893-901; Mechref, Y. Current protocols in protein science (2012) 12.11. 11-12.11. 11; Saba, J., et al., International journal of proteomics (2012); Singh, C., et al., J. Proteome Res. (2012) 11, 4517-4525; Zhang, L. and Reilly, J. P., J. Proteome Res. (2008) 8, 734-742; Madsen, J. A., et al., Anal. Chem. (2013) 85, 9253-9261). Moreover, because conventional MS methodologies often use data-dependent peak-picking, MS is only performed on the most abundant ions. While there are advantages to using such methodologies, these methodologies may not detect less abundant, but still important, ions (e.g., ions associated with unknown or unique glycopeptides). Because of these shortcomings, it can be difficult to assign unambiguously all potential sites of glycosylation and their associated glycans for complicated glycoproteins. In turn, it can be difficult, using standard methods, to compare the glycosylation of some glycoproteins (e.g., test proteins) to the glycosylation of another glycoprotein (e.g., a reference protein, e.g., a target protein).
[0095] Data-independent analysis of glycopeptides has been applied to glycopeptide analysis to circumvent the traditional reliance on data-dependent acquisition of LC-MS/MS data, which relies on peak picking of the most abundant species from the MS1 spectra to initiate MS/MS, a process that can yield incomplete site-specific characterization. (Sanda, M. and Goldman, R. Anal. Chem. (2016) 88, 10118-10125; Sanda, M., et al., Anal. Bioanal. Chem. (2017) 409, 619-627; Pan, K.-T., et al., Anal. Chem. (2017) 89, 4532-4539). Data-independent analysis techniques fragment all precursor ions in a specified m/z range resulting in a potentially more complete and reproducible data procurement. Quantitation of the glycopeptides (identified by any MS means) can be generally performed by ion extraction of individual glycosylated species from the MS1 and/or MS/MS data. Targeting of low-mass, glycan-specific oxonium ions generated by various MS/MS techniques can be an especially useful technique for deciphering glycopeptides from non-glycosylated species. These ions have been utilized in numerous advantageous approaches to qualitatively and quantitatively assess site-specific glycosylation. (Saba, J., et al., International journal of proteomics (2012); Singh, C., et al., J. Proteome Res. (2012)11, 4517-4525; Cao, L., et al., Anal. Biochem. (2014) 452, 96-102; Toyama, A., et al., Anal. Chem. (2012) 84, 9655-9662; Song, E., et al., Rapid Commun. Mass Spectrom. (2012) 26, 1941-1954; Hart-Smith, G. and Raftery, M. J., J. Am. Soc. Mass Spectrom. (2012) 23, 124-140; Halim, A., et al., J. Proteome Res. (2014) 13, 6024-6032; Nasir, W., et al., J. Proteome Res. (2016) 15, 2826-2840). Nonetheless, even with the advancements made in glycoprotein analysis, it remains difficult to unambiguously assign most, if not all, potential sites of glycosylation and their associated glycans for complicated multiglycosylated proteins. Therefore, currently the known glycoforms and glycosylation sites are assessed during biotherapeutic characterization, which can be a potentially problematic analytical strategy, especially when developing drugs such as biosimilars.
[0096] The present disclosure is based, in part, on the discovery that comparisons of glycosylation profiles (e.g., site-specific glycosylation profiles) between glycoproteins can be improved in some circumstances by comparing oxonium ion profiles of the glycoproteins, e.g., when fragmentation of the glycoproteins generates glycopeptides that are difficult to predict and/or identify. The present disclosure describes that MS can be used to generate an oxonium ion profile for a protein (e.g., a glycoprotein). In some embodiments, an oxonium ion profile of a protein can be used to assess biosimilarity, e.g., to manufacture biosimilar proteins (e.g., a protein approved for use in humans by a secondary approval process). For example, in some embodiments, an oxonium ion profile of a sample (e.g., of a test protein) can be compared to an oxonium ion profile of a reference sample (e.g., of a target protein) to, e.g., assess whether the similarity/dissimilarity between the sample and the reference sample is tolerable.
Overview of Oxonium Ion Profiling
[0097] Higher-energy collisional dissociation (HCD) of glycopeptides generates several diagnostic low-mass oxonium ions that can be used for both identification and quantification of glycosylated species. Typical oxonium ions observed include: HexNAc internal fragment (m/z 138), Hex (m/z 163), HexNAc (m/z 204), sialic acid—H.sub.2O (m/z 274), sialic acid (m/z 292), Hex+HexNAc (m/z 366), (among others). The HexNAc (m/z 204) ion is universal to glycopeptides, and can be produced at high abundance under elevated HCD conditions. The sialic acid—H.sub.2O (m/z 274) ion can also be produced with high abundance, and can represent a particularly important acidic sugar that has been shown to alter the anti-inflammatory properties of therapeutic proteins. (Czajkowsky, D. M., et al., EMBO molecular medicine (2012) 4, 1015-1028; Washburn, N., et al., Proceedings of the National Academy of Sciences (2015) 112, E1297-E1306). Therefore, these two ion species can be employed to create the oxonium ion profiles used for biotherapeutic comparisons herein. The structure of the HexNAc and sialic acid ions, and an example mass spectrum of each, can be seen in
[0098] The general methods workflow for oxonium ion profiling can be seen in
[0099] The conceivable benefits of this methodology include, but are not limited to: (1) most or all precursor ions are fragmented collectively by data independent analysis; therefore, peak picking inconsistencies and high ion abundance bias issues are minimized or eliminated, (2) a priori assumptions about the presence of specific glycosylated species are not required, which better ensures that unknown glycopeptides contribute signal to the overall measurement of similarity, and (3) profile similarity can be quickly and directly assessed between samples (e.g., a reference protein compared to a biosimilar, etc.). That is, while it may be impossible to unambiguously predict and/or identify all glycopeptides in a given sample, the unknown glycosylated species can at least be specifically and reproducibly quantified between samples.
Analysis Methods
[0100] Methods described herein utilize mass spectrometry (MS), a general technique known in the art. Mass spectrometry obtains molecular weight and structural information on chemical compounds by ionizing the molecules and measuring either their time-of-flight or a response of the molecular trajectories to electric and/or magnetic fields. The methods of the present disclosure employ conventional mass spectrometry techniques known to those of skill in the art, and any known MS method can be adapted for use in methods of the disclosure. Exemplary MS can include, but are not limited to, tandem MS (MS/MS), LC-MS, LC-MS/MS, matrix assisted laser desorption ionisation mass spectrometry (MALDI-MS), Fourier transform mass spectrometry (FTMS), ion mobility separation with mass spectrometry (IMS-MS), electron transfer dissociation (ETD-MS), and combinations thereof. Such methods are described in, e.g., Pitt, Clin. Biochem. Rev. 30:19-34 (2009). Mass spectrometers are known in the art and are commercially available from, e.g., Agilent Inc., Bruker Corporation, Waters, AB Sciex, Shimadzu, and Thermo Scientific.
[0101] MS can be used to generate an oxonium ion profile, which can include one or more MS signals associated with one or more oxonium ion-containing fragments. The one or more MS signals can represent values (e.g., retention time, relative abundance of the ion, and/or mass-to-charge ratio) obtained by MS for the one or more oxonium containing fragments. An oxonium ion profile can be generated, for example, by creating a chromatogram showing the relative abundance versus retention time as acquired by MS for oxonium ions having a particular mass-to-charge ratio. The mass-to-charge ratio of exemplary oxonium ions observed is provided in Table 1 (below).
TABLE-US-00001 TABLE 1 Oxonium Ion m/z value HexNAc internal fragment 138.06 Hex 163.06 HexNAc 204.09 Sialic acid 274.09 sialic acid + H.sub.2O 292.10 Hex + HexNAc 366.14
Those skilled in the art will recognize that the measured mass-to-charge ratio for an oxonium ion may vary from the above value based on, e.g., the tolerance of the mass spectrometer used and the isotopic distribution. As such, in some circumstances, it may be advantageous to examine MS signals over a range of mass-to-charge ratios. For example, an oxonium ion profile may include MS signals for oxonium ion-containing fragments that include oxonium ions within the following m/z ranges: (i) m/z 138.05-138.07; (ii) m/z 163.05-163.07; (iii) m/z 204.08-204.10; (iv) m/z 274.08-274.10; (v) m/z 292.09-292.11; (vi) m/z 366.13-366.15; or (vii) combinations thereof.
[0102] In some embodiments, a conventional MS spectra is not generated in methods described herein. For example, a plot of the relative abundance of an ion versus the mass-to-charge ratio of the ion is not generated, and therefore, an MS spectra is not used to identify and/or quantify individual glycopeptides. In certain circumstances, an MS spectrum is generated in parallel to the methods disclosed herein.
[0103] In some embodiments, higher-energy collisional dissociation (HCD) of glycopeptides can be used to generate several diagnostic low-mass oxonium ions that can be used for both identification and quantification of glycosylated species. The typical ions observed include those ions shown in Table 1 above. The HexNAc (m/z 204) ion is universal to glycopeptides and can be produced at high abundance under elevated HCD conditions. The sialic acid—H.sub.2O (m/z 274) ion can also be generated from a large number glycopeptides, and represents a particularly important acidic sugar that has been shown to alter the anti-inflammatory properties of therapeutic proteins. (Czajkowsky, D. M., et al., EMBO molecular medicine (2012), 4, 1015-1028; Washburn, N., et al., Proceedings of the National Academy of Sciences (2015) 112, E1297-E1306). These two ion species were employed to create the oxonium ion profiles used for biotherapeutic comparisons herein. The structure of the HexNAc and sialic acid ions, and an example mass spectrum of each, can be seen in
[0104] In some embodiments, data independent MS methodologies are used to acquire oxonium ion profiles. Such methods can increase the likelihood that most, if not all, glycopeptide species are accounted for in an analysis. For example, data independent methodologies (e.g., all ion fragmentation) can be advantageous because such methodologies are less dependent on a priori assumptions about what glycopeptides are present in a sample. Such methodologies can also facilitate comparisons of a sample (e.g., of a test protein) to a reference sample (e.g., of a target protein), e.g., where the sample and/or the reference sample includes an unknown or unique glycopeptide.
[0105] In some methods of the disclosure, a glycosylated protein (e.g., a test protein, e.g., a biotherapeutic) is digested (e.g., using a protease, e.g., trypsin, chymotrypsin, AspN) into a mixture of peptides and glycopeptides (see, e.g.,
[0106] In some instances, higher-order structure of a protein can be assessed by performing MS on a protein (e.g., a sample of a protein preparation, more specifically, a sample of a glycoprotein preparation) to obtain a mass spectrum of relative abundance of ions with a particular mass-to-charge ratio over a given range (e.g., 100 to 2000 amu). Numerous methods for determining amount/abundance of a peptide from an amount/abundance of an ion are known to those of ordinary skill in the art. For example, relative abundance of a given ion may be compared to a table that converts that relative abundance to an absolute amount of a peptide. Additionally or alternatively, external standards may be run with samples, and a standard curve constructed based on ions generated from such standards. Using a standard curve, relative abundance of a given ion may be converted into an absolute amount of a peptide. Methods of generating and using such standard curves are well known in the art, and one of ordinary skill is capable of selecting an appropriate internal standard.
Applications
[0107] In some instances, methods disclosed herein can be used to confirm the identity and/or quality of a protein, e.g., glycoprotein preparation. For example, methods can include assessing preparations (e.g., samples, lots, and/or batches) of a test protein, e.g., to confirm whether a test protein qualifies as a target protein, and, optionally, qualifying a test protein as a target protein if qualifying criteria (e.g., predefined qualifying criteria) are met; thereby evaluating, identifying, and/or producing (e.g., manufacturing) a protein product.
[0108] Methods of the disclosure can have a variety of applications and can include, e.g., quality control at different stages of manufacture, analysis of a protein preparation prior to and/or after completion of manufacture (e.g., prior to or after distribution to a fill/finish environment or facility), and/or prior to and/or after release into commerce (e.g., before distribution to a pharmacy, a caregiver, a patient, or other end-user). In some instances, a protein preparation may be a drug substance (i.e., an active pharmaceutical ingredient or “API”) or a drug product (i.e., an API formulated for use in a subject such as a human patient). In some instances, a protein preparation may be from a stage of manufacture or use that is prior to release to care givers or other end-users; prior to packaging into individual dosage forms, such as syringes, pens, vials, or multi-dose vials; prior to determination that a batch can be commercially released, prior to production of a Certificate of Testing, Material Safety Data Sheet (MSDS) or Certificate of Analysis (CofA) of a preparation. In some instances, a protein preparation may be from an intermediate step in production, e.g., it is after secretion of a protein from a cell but prior to purification of drug substance.
[0109] Evaluations from methods described herein can be useful for guiding, controlling or implementing one or more of a number of activities or steps in a process of making, distributing, and monitoring and providing for a safe and efficacious use of a protein preparation. Thus, in some embodiments, e.g., responsive to an evaluation, e.g., depending on whether a criterion is met, a decision or step is taken. Methods can further include one or both of a decision to take a step and/or carrying out the step itself. For example, a step can include one in which a preparation (or another preparation for which the preparation is representative) is: classified; selected; accepted or discarded; released or processed into a drug product; rendered unusable for commercial release, e.g., by labeling it, sequestering it, or destroying it; passed on to a subsequent step in manufacture; reprocessed (e.g., a preparation may undergo a repetition of a previous process step or subjected to a corrective process); formulated, e.g., into drug substance or drug product; combined with another component, e.g., an excipient, buffer or diluent; disposed into a container; divided into smaller aliquots, e.g., unit doses, or multi-dose containers; combined with another preparation of a protein (e.g., the same protein); packaged; shipped; moved to a different location; combined with another element to form a kit; combined, e.g., placed into a package with a delivery device, diluent, or package insert; released into commerce; sold or offered for sale; delivered to a care giver or other end-user; or administered to a subject. For example, based on a result of a determination or whether one or more subject entities is present, or upon comparison to a reference standard, a batch from which a preparation is taken can be processed, e.g., as just described.
[0110] Methods described herein may include making a decision: (a) as to whether a protein preparation may be formulated into drug substance or drug product; (b) as to whether a protein preparation may be reprocessed (e.g., a preparation may undergo a repetition of a previous process step); and/or (c) that a protein preparation may not be suitable for formulation into drug substance or drug product. In some instances, methods can include: formulating as referred to in step (a), reprocessing as referred to in step (b), or rendering a preparation unusable for commercial release, e.g., by labeling it or destroying it, as referred to in step (c).
Test Proteins and Target Proteins
[0111] Methods described herein can be used to make and/or evaluate a test protein preparation, e.g., a test biologic preparation. In some embodiments, a test protein can be a test biologic being evaluated for similarity to a target protein, e.g., a target biologic. A test biologic may or may not be commercially available. In some embodiments, a test biologic is not commercially available for therapeutic use in humans or animals. In some embodiments, a test biologic has not been approved for therapeutic or diagnostic use in humans or animals. In some embodiments, a test biologic has been approved, e.g., under a secondary approval process, for therapeutic or diagnostic use in humans or animals. In some embodiments, a test protein (e.g., test biologic) has the same primary amino acid sequence as a target protein (e.g., target biologic) or differs by no more than 1, 2, 3, 4, 5, 10, 15, 20, 25, 30 residues and/or has at least 90, 95, 98, 99% or is identical to a target protein sequence (e.g., target biologic sequence). The terms the “same primary amino acid sequence,” “a primary amino acid sequence that differs by no more than 1, 2, 3, 4, 5, 10, 15, 20, 25, or 30 residues,” “sequences that have at least 98% or more sequence identity,” or similar terms, relate to level of identity between a primary amino acid sequence, e.g., of first protein, e.g., a test protein, and a primary amino acid sequence, e.g., of second protein, e.g., a target protein. In some embodiments, a protein preparation or product includes amino acid variants, e.g., species that differ at terminal residues, e.g., at one or two terminal residues. In some embodiments of such cases, a sequence identity comparison is between a primary amino acid sequence of the most abundant (e.g., the most abundant active) species in each of the products being compared. In some embodiments, sequence identity refers to an amino acid sequence encoded by a nucleic acid that can be used to make the product.
[0112] Nonlimiting, exemplary target proteins can include abatacept (Orencia®, Bristol-Myers Squibb), abciximab (ReoPro®, Roche), adalimumab (Humira®, Bristol-Myers Squibb), aflibercept (Eylea®, Regeneron Pharmaceuticals), alefacept (Amevive®, Astellas Pharma), alemtuzumab (Campath®, Genzyme/Bayer), basiliximab (Simulect®, Novartis), belatacept (Nulojix®, Bristol-Myers Squibb), belimumab (Benlysta®, GlaxoSmithKline), bevacizumab (Avastin®, Roche), canakinumab (Ilaris®, Novartis), brentuximab vedotin (Adcetris®, Seattle Genetics), certolizumab (CIMZIA®, UCB, Brussels, Belgium), cetuximab (Erbitux®, Merck-Serono), daclizumab (Zenapax®, Hoffmann-La Roche), denileukin diftitox (Ontak®, Eisai), denosumab (Prolia®, Amgen; Xgeva®, Amgen), eculizumab (Soliris®, Alexion Pharmaceuticals), efalizumab (Raptiva®, Genentech), etanercept (Enbrel®, Amgen-Pfizer), gemtuzumab (Mylotarg®, Pfizer), golimumab (Simponi®, Janssen), ibritumomab (Zevalin®, Spectrum Pharmaceuticals), infliximab (Remicade®, Centocor), ipilimumab (Yervoy™, Bristol-Myers Squibb), muromonab (Orthoclone OKT3®, Janssen-Cilag), natalizumab (Tysabri®, Biogen Idec, Elan), ofatumumab (Arzerra®, GlaxoSmithKline), omalizumab (Xolair®, Novartis), palivizumab (Synagis®, MedImmune), panitumumab (Vectibix®, Amgen), ranibizumab (Lucentis®, Genentech), rilonacept (Arcalyst®, Regeneron Pharmaceuticals), rituximab (MabThera®, Roche), tocilizumab (Actemra®, Genentech; RoActemra, Hoffman-La Roche) tositumomab (Bexxar®, GlaxoSmithKline), trastuzumab (Herceptin®, Roche), and ustekinumab (Stelara®, Janssen).
[0113] Antibodies
[0114] In some instances, test proteins and target proteins described herein are antibodies. As used herein, the term “antibody” refers to a polypeptide that includes at least one immunoglobulin variable region domain, e.g., an amino acid sequence for an immunoglobulin variable region domain or immunoglobulin variable region domain sequence. For example, an antibody can include a heavy (H) chain variable region domain (abbreviated herein as V.sub.H), and a light (L) chain variable region domain (abbreviated herein as V.sub.L). In another example, an antibody can include two heavy chain variable region domains and two light chain variable region domains. The term “antibody” encompasses antigen-binding fragments of antibodies (e.g., single chain antibodies, Fab, F(ab′).sub.2, Fd, Fv, and dAb fragments) as well as complete antibodies, e.g., intact immunoglobulins of types IgA, IgG, IgE, IgD, IgM (as well as subtypes thereof). Light chains, light chain variable regions or fragments thereof can be of types kappa or lambda. In some embodiments, an antibody includes an Fc region. In some embodiments, an antibody is a therapeutic antibody.
[0115] Antibodies described herein can include, for example, monoclonal antibodies, polyclonal antibodies (e.g., IVIG), multi-specific antibodies (e.g., bispecific antibodies), human antibodies, humanized antibodies, camelized antibodies, chimeric antibodies, single-chain Fvs (scFv), disulfide-linked Fvs (sdFv), and anti-idiotypic (anti-Id) antibodies, and antigen-binding fragments of any of the above. Antibodies can be of any type (e.g., IgG, IgE, IgM, IgD, IgA and IgY), class (e.g., IgG1, IgG2, IgG3, IgG4, IgA1 and IgA2) or subclass.
[0116] Antibodies or fragments thereof can be produced by any method known in the art for synthesizing antibodies (see, e.g., Harlow, et al., Antibodies: A Laboratory Manual, (Cold Spring Harbor Laboratory Press, 2nd ed. 1988); Brinkman, et al., 1995, J. Immunol. Methods 182:41-50; WO 92/22324; WO 98/46645). Chimeric antibodies can be produced using methods described in, e.g., Morrison, 1985, Science 229:1202, and humanized antibodies by methods described in, e.g., U.S. Pat. No. 6,180,370.
[0117] Glycoprotein Conjugates
[0118] In some instances, test proteins and target proteins can be glycoprotein conjugates (e.g., Fc regions or Fc fragments containing one or more N-glycosylation sites thereof that are conjugated or fused to one or more heterologous moieties). Heterologous moieties can include, but are not limited to, peptides, polypeptides, proteins, fusion proteins, nucleic acid molecules, small molecules, mimetic agents, synthetic drugs, inorganic molecules, and organic molecules. In some instances, a glycoprotein conjugate can be a fusion protein that includes a peptide, polypeptide, protein scaffold, scFv, dsFv, diabody, Tandab, or an antibody mimetic fused to an Fc region, such as a glycosylated Fc region. A fusion protein can include a linker region connecting an Fc region to a heterologous moiety (see, e.g., Hallewell, et al. (1989), J. Biol. Chem. 264, 5260-5268; Alfthan, et al. (1995), Protein Eng. 8, 725-731; Robinson & Sauer (1996)).
Recombinant Gene Expression
[0119] In accordance with the present disclosure, there may be employed conventional molecular biology, microbiology, and recombinant DNA techniques within the skill of the art. Such techniques are described in the literature (see, e.g., Sambrook, Fritsch & Maniatis, Molecular Cloning: A Laboratory Manual, Second Edition (1989) Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.; DNA Cloning: A Practical Approach, Volumes I and II (D. N. Glover ed. 1985); Oligonucleotide Synthesis (M. J. Gait ed. 1984); Nucleic Acid Hybridization (B. D. Hames & S. J. Higgins eds. (1985)); Transcription And Translation (B. D. Hames & S. J. Higgins, eds. (1984)); Animal Cell Culture (R. I. Freshney, ed. (1986)); Immobilized Cells and Enzymes (IRL Press, (1986)); B. Perbal, A Practical Guide To Molecular Cloning (1984); F. M. Ausubel, et al. (eds.), Current Protocols in Molecular Biology, John Wiley & Sons, Inc. (1994)).
[0120] In some embodiments, a protein described herein can be produced using recombinant methods. Recombinant expression of a gene, such as a gene encoding a polypeptide, such as an antibody described herein, can include construction of an expression vector containing a polynucleotide that encodes a polypeptide. Once a polynucleotide has been obtained, a vector for the production of a polypeptide can be produced by recombinant DNA technology using techniques known in the art. Known methods can be used to construct expression vectors containing polypeptide coding sequences and appropriate transcriptional and translational control signals. These methods include, for example, in vitro recombinant DNA techniques, synthetic techniques, and in vivo genetic recombination.
[0121] An expression vector can be transferred to a host cell by conventional techniques, and transfected cells can then be cultured by conventional techniques to produce polypeptide.
[0122] A variety of host expression vector systems can be used (see, e.g., U.S. Pat. No. 5,807,715). Such host-expression systems can be used to produce polypeptides and, where desired, subsequently purified. Such host expression systems can include microorganisms, such as bacteria (e.g., E. coli and B. subtilis), transformed with recombinant bacteriophage DNA, plasmid DNA or cosmid DNA expression vectors containing polypeptide coding sequences; yeast (e.g., Saccharomyces and Pichia) transformed with recombinant yeast expression vectors containing polypeptide coding sequences; insect cell systems infected with recombinant virus expression vectors (e.g., baculovirus) containing polypeptide coding sequences; plant cell systems infected with recombinant virus expression vectors (e.g., cauliflower mosaic virus, CaMV; tobacco mosaic virus, TMV) or transformed with recombinant plasmid expression vectors (e.g. Ti plasmid) containing polypeptide coding sequences; or mammalian cell systems (e.g., COS, CHO, BHK, 293, NSO, and 3T3 cells) harboring recombinant expression constructs containing promoters derived from the genome of mammalian cells (e.g., metallothionein promoter) or from mammalian viruses (e.g., an adenovirus late promoter; a vaccinia virus 7.5K promoter).
[0123] For bacterial systems, a number of expression vectors can be used, including, but not limited to, an E. coli expression vector pUR278 (Ruther, et al., 1983, EMBO 12:1791); pIN vectors (Inouye & Inouye, 1985, Nucleic Acids Res. 13:3101-3109; Van Heeke & Schuster, 1989, J. Biol. Chem. 24:5503-5509); and the like. pGEX vectors can also be used to express foreign polypeptides as fusion proteins with glutathione 5-transferase (GST).
[0124] For expression in mammalian host cells, viral-based expression systems can be utilized (see, e.g., Logan & Shenk, 1984, Proc. Natl. Acad. Sci. USA 8 1:355-359). Efficiency of expression can be enhanced by inclusion of appropriate transcription enhancer elements, transcription terminators, etc. (see, e.g., Bittner et al., 1987, Methods in Enzymol. 153:516-544).
[0125] In addition, a host cell strain can be chosen that modulates expression of inserted sequences, or modifies and processes a gene product in a specific fashion desired. Different host cells may have characteristic and specific mechanisms for post-translational processing and modification of proteins and gene products. Appropriate cell lines or host systems can be chosen to ensure the correct modification and processing of a polypeptide expressed. Such cells can include, for example, established mammalian cell lines and insect cell lines, animal cells, fungal cells, and yeast cells. Mammalian host cells can include, but are not limited to, CHO, VERY, BHK, HeLa, COS, MDCK, 293, 3T3, W138, BT483, Hs578T, HTB2, BT20 and T47D, NSO (a murine myeloma cell line that does not endogenously produce any immunoglobulin chains), CRL7O3O and HsS78Bst cells.
[0126] For long-term, high-yield production of recombinant proteins, host cells can be engineered to stably express a polypeptide. Host cells can be transformed with DNA controlled by appropriate expression control elements known in the art, including promoter, enhancer, sequences, transcription terminators, polyadenylation sites, and selectable markers. Methods commonly known in the art of recombinant DNA technology can be used to select a desired recombinant clone.
[0127] Once a protein described herein been produced by recombinant expression, it may be purified by any method known in the art for purification, for example, by chromatography (e.g., ion exchange, affinity, and sizing column chromatography), centrifugation, differential solubility, or by any other standard technique for purification of proteins. For example, an antibody can be isolated and purified by appropriately selecting and combining affinity columns such as Protein A column with chromatography columns, filtration, ultrafiltration, salting-out and dialysis procedures (see Antibodies: A Laboratory Manual, Ed Harlow, David Lane, Cold Spring Harbor Laboratory, 1988). Further, as described herein, a glycoprotein can be fused to heterologous polypeptide sequences to facilitate purification. Glycoproteins having desired sugar chains can be separated with a lectin column by methods known in the art (see, e.g., WO 02/30954).
Pharmaceutical Compositions
[0128] A protein (e.g., an antibody) described herein can be incorporated into a pharmaceutical composition. Such a pharmaceutical composition can be useful in prevention and/or treatment of diseases. Pharmaceutical compositions including a polypeptide (e.g., an antibody) can be formulated by methods known to those skilled in the art (see, e.g., Remington's Pharmaceutical Sciences, 20th Ed., Lippincott Williams & Wilkins, 2000). A pharmaceutical composition can be administered parenterally in the form of an injectable formulation including a sterile solution or suspension in water or another pharmaceutically acceptable liquid. For example, a pharmaceutical composition can be formulated by suitably combining a polypeptide with pharmaceutically acceptable vehicles or media, such as sterile water and physiological saline, vegetable oil, emulsifier, suspension agent, surfactant, stabilizer, flavoring excipient, diluent, vehicle, preservative, binder, followed by mixing in a unit dose form required for generally accepted pharmaceutical practices. An amount of active ingredient included in a pharmaceutical preparation can be such that a suitable dose within a designated range is provided.
[0129] Route of administration can be, e.g., parenteral, for example, administration by injection, transnasal administration, transpulmonary administration, or transcutaneous administration. Additionally or alternatively, administration can be systemic or local by intravenous injection, intramuscular injection, intraperitoneal injection, subcutaneous injection.
[0130] A suitable means of administration can be selected based on factors, such as, the age and condition of a patient. A single dose of a pharmaceutical composition containing a polypeptide (e.g., antibody) can be selected from a range of 0.001 mg/kg of body weight to 1000 mg/kg of body weight. On the other hand, a dose can be selected in the range of 0.001 mg/kg of body weight to 100000 mg/kg of body weight, but the present disclosure is not limited to such ranges. Dosages and methods of administration can vary depending on the weight, age, condition, and the like of a patient, and can be suitably selected as needed by those skilled in the art.
[0131] The disclosure is further illustrated by the following examples. The examples are provided for illustrative purposes only. They are not to be construed as limiting the scope or content of the disclosure in any way.
EXAMPLES
Example 1: Glycan Oxonium Profiling Method
[0132] This Example describes an exemplary method of producing an oxonium ion profile for a glycoprotein using an LC-MS/MS analysis of all digestion fragments from a glycoprotein. In one exemplary method, a preparation of a glycosylated model Fc fusion protein was first subjected to protease digestion to produce a mixture of peptides and glycopeptides (see
[0133] The data corresponding to glycan oxonium ions were then extracted with high mass accuracy from the resulting mass spectra to create chromatograms (“oxonium ion profiles”) that were specific to glycan-modified peptides.
[0134] In addition, peaks in the oxonium ion profile were assigned to glycosylation sites on the model Fc fusion protein based on the chromatographic retention time. Thus, as shown in
Example 2: Glycan Oxonium Ion Profiling of Samples of a Model Fc Fusion Protein
[0135] This Example shows that glycan oxonium ions profiles can be used to detect and analyze the similarity/dissimilarity of glycans and glycopeptides obtained from different glycoprotein preparations (e.g., samples or reference samples). Using the methodology described in Example 1, glycan oxonium ion profiling was conducted on samples of a model Fc fusion protein from four different CHO cell line clones expressing the model Fc fusion protein (Sample 1, Sample 2, Sample 3 and Sample 4) and on reference samples from three commercial lots of the model Fc fusion protein (RP1, RP2, and RP3). Exemplary chromatograms are depicted in
[0136] The chromatograms illustrate that glycan oxonium ion profiles, for example as obtained by the method described in Example 1, can be used to discern differences between the glycans and glycopeptides in different glycoprotein preparations. For example, a unique glycan or glycopeptide was detected in Sample 2 that was not detected in any other samples or reference samples. This observation is highlighted in
Example 3: Quantification of Differences in Glycan Oxonium Ion Profiles for Samples of a Model Fc Fusion Protein
[0137] This Example describes a method for quantifying the similarity/dissimilarity of glycans and glycopeptides obtained from different glycoprotein preparations (e.g., samples or reference samples). As demonstrated in Example 2, visual inspection of glycan oxonium ion profiles can be used to detect differences in glycoprotein samples. In some embodiments, statistical measures of similarity/dissimilarity can be used to quantitate the similarity/dissimilarity.
[0138] In one exemplary method, an algorithm that separates each ion profile into specified “bins” based on retention time was created. The algorithm was written in the R programming language. As shown in
Example 4: Comparison of Glycan Oxonium Ion Profiling with Standard Analysis Methods
[0139] This Example demonstrates that oxonium ion profiling can provide information about the glycans present on a glycoprotein sample (including how those glycans are similar/dissimilar to the glycans on a reference sample) that may be different than the information about the glycans that can be determined by standard LC-MS/MS methodology. Three of the four samples used in Example 2 (Sample 1, Sample 2, and Sample 3) and the three reference samples used in Example 2 (RP1, RP2, and RP3) were analyzed using standard LC-MS/MS for assessing glycosylation (see
[0140] The increased resolving power of oxonium ion profiling for assessing similarity/dissimilarity between a sample and a reference sample was demonstrated by comparing the results obtained from oxonium ion profiling to the results obtained from the standard LC-MS/MS method. For example, using the standard LC-MS/MS, Sample 2 had the most similar glycopeptide profile to the reference protein (
Example 5: Glycan Oxonium Ion Profiling Methods
Sample Preparation
[0141] 100 μg of a sample or a reference sample was diluted to 1 mg/mL with 6 M guanidine HCL in 20 mM sodium phosphate, 100 mM sodium chloride, pH 7.0, and denatured for 30 minutes at 37° C. Disulfide reduction was then performed by adding 5 mM Dithiothreitol, and incubating for 80 minutes at 37° C. Proteins in the sample or reference sample were then alkylated with 12 mM N-ethylmaleimide for two hours in the dark at room temperature. Using Amicon 10 k spin filters (EMD Millipore, Billerica, Mass.) the sample or reference sample was buffer exchanged into 50 mM ammonium bicarbonate, and digested with 5 μg of trypsin (1:20 enzyme to substrate ratio) for 19 hours at 37° C. Protease reactions were quenched with 2.5% formic acid.
LC-MS/MS
[0142] Tryptic peptides obtained from the samples and/or reference samples (4 μg) were injected onto a 2.1×50 mm (1.7 μm particle size) AQUITY BEH C18 column (Waters, Milford, Mass.) and heated at 50° C. using a Dionex Ultimate 3000 RSLCnano (Santa Clara, Calif.) system. Eluent A and B were 0.1% formic acid in water and 0.1% formic acid in acetonitrile, respectively. Gradient elution was performed at a flow rate of 50 μL/min as follows: linear gradient of 3% eluent B at zero minutes, 7% eluent B at 3 minutes, 13% eluent B at 20 minutes, 28% eluent B at 40 minutes, 35% eluent B at 60 minutes, 50% eluent B at 70 minutes, 80% eluent B from 80 to 90 minutes, and 3% eluent B from 92 to 110 minutes.
[0143] The liquid chromatography system was coupled to a Thermo Scientific Q Exactive (Bremen, Germany) for mass spectrometric analysis. The instrument was operated to accommodate both data-independent all ion fragmentation and data-dependent analyses in a single run. Data dependent acquisition was performed as follows: MS1 events were comprised of the positive mass scan at a range of 400-2000 m/z followed by one HCD event at 25% normalize collision energy (NCE) on the most abundant ion from the first event. Dynamic exclusion duration was 20 s with a single repeat count, and charged species ≥7 were excluded. Data-independent all ion fragmentation was performed at a range of 200-2000 m/z, and a NCE of 35%. The electrospray ionization voltage was set to 3.32 kV, capillary temperature was 250° C., sheath gas was 15, and the S-Lens RF level was set to 50 for all analyses. Resolution was 35,000 for MS1 scans, and 17,500 for data dependent acquisition MS/MS and data-independent all ion fragmentation scans. The automated gain control was set to 1E6 with a maximum injection time of 250 ms for both MS1 and data-independent all ion fragmentation, and automated gain control was 2E5 with a maximum injection time of 80 ms for data dependent acquisition MS/MS. An isolation window of 4 m/z was used for data dependent acquisition MS/MS scans. MS1, data-independent all ion fragmentation, and data dependent acquisition MS/MS spectra were produced from 5, 3, and 1 microscans, respectively.
Data Analysis
[0144] HexNAc and sialic acid oxonium ion profiles were generated by extracting out the m/z range of 204.08-204.10 and 274.08-274.10, respectively, across the entire data-independent acquisition for each sample or reference sample. The mass accuracy tolerance for ion extraction was ±0.01 Da for the experiments herein, and proved to be sufficiently specific for the targeted oxonium ions as no obvious interferences were observed.
[0145] Conventional targeted glycopeptide analysis was performed by mining LC-MS/MS files of the most abundant known glycoforms and glycosylation sites. Skyline (version 3.5.0.9320) was used to create extracted ion chromatograms (XICs) of each individual species. Results were then reported as a relative abundance percentage for each individual glycosylation site.
[0146] Database searching analysis was performed using Byonic, similarly as previously described. (Bern, M., et al., Anal. Chem. (2007) 79, 1393-1400; Bern, M., et al., Curr. Protoc. Bioinformatics (2012) Chapter 13, Unit 1320; Madsen, J. A., et al., Anal. Chem. (2016) 88, 2478-2488). In short, LC-MS/MS files were searched against a protein database composed of the protein of interest. Trypsin was used as the enzyme with up to two allowable missed cleavages. Mass tolerances of 10 ppm and 0.02 Da were used for MS1 and MS/MS spectra, respectively. N-Ethylmaleimide of cysteine was set as a fixed modification for the searches, and glycan databases consisting of 118 N-glycans and 6 O-glycans were used by Byonic for identifying glycopeptides.
[0147] The automated similarity assessment tool was created in the R programming language. (http://www.R-project.org (2014)). Thermo .RAW files were converted to MGF format and loaded into R using MSnbase library (Gatto, L., Lilley, K. S. Bioinformatics (2011) btr645), and extracted ion chromatograms of m/z 204.08-204.10 and 274.08-274.10 were generated from the data-independent scans. The resulting oxonium ion profiles were then loess smoothed (Cleveland, W. S., et al., Statistical models in S (1992) 2, 309-376) and adjusted for differences in retention time by parametric time warping. (Eilers, P. H. Anal. Chem. (2004) 76, 404-411; Bloemberg, T. G., et al., Chemometrics and Intelligent Laboratory Systems (2010) 104, 65-74). Selection of parameters for loess regression was performed by cross-validation as illustrated in
[0148] The dissimilarities between each pair of samples and reference samples were calculated as the sum of the absolute differences between resulting oxonium ion relative abundances across the entire time course of the experiment (e.g., from 0-80 minutes) or for time intervals corresponding to selected glycopeptide sites. For the site-specific calculations, the sample dissimilarities were calculated over time intervals of 0-20 minutes (glycopeptide site 1), 20-36 minutes (glycopeptide site 2/3), 36-55 minutes (glycopeptide site 4), and 55-80 minutes (glycopeptide site 5). Heat maps were generated from the 2-way comparison of all Fc-fusion protein samples, and were clustered using average linkage. For direct comparisons to the oxonium ion profiling results, sample dissimilarities from conventional targeted analysis were calculated as sums of absolute differences between relative abundances (normalized to the highest area count) for each individual glycopeptide.
Session Information
[0149] Except for
Example 6: Rapid Identification of Fab Glycosylation
[0155] This Example demonstrates that oxonium ion profiling can be used to quickly detect the presence or absence of glycosylation on multiple glycosylation sites. In particular, this Example demonstrates that oxonium ion profiling can be used to quickly detect the presence or absence of Fab glycosylation, as well as Fc glycosylation.
[0156] N-linked Fc glycosylation is well conserved across IgG subclasses; therefore, glycopeptide identification at the Fc region is relatively straightforward using conventional methodologies. Fab glycosylation characterization, conversely, is subject to a number of challenges when conventional methodologies are used. As one example, there are a tremendous number of amino acid combinations that can be present in a Fab variable region; as such, predicting the amino acid sequence would be difficult and predicting the site-specific glycosylation sites would be even harder. As another example, it would be very difficult to predict all the Fab site-specific glycosylation in intravenous immunoglobulin (IVIG) because the IgGs are pooled from thousands of plasma donors. It would also be difficult to foresee how a given expression system might affect Fab glycosylation of recombinantly produced antibodies and/or antibody-like therapeutics. Additionally, O-glycosylation is difficult to predict because a consensus amino acid sequence motif has not been identified. Glycan oxonium profiling can overcome these challenges.
[0157] Glycan oxonium profiles were obtained for two samples, mAb1 (which has Fc glycosylation) and mAb2 (which has both Fc and Fb glycosylation), in accordance with the methods described above in Example 5. In this application, the glycan oxonium profiling was utilized to rapidly detect the presence or absence of Fab glycosylation, as shown in
[0158] A more exhaustive and retention-time targeted database searching procedure was then initiated on the accompanying data dependent analysis spectra revealing that indeed, the detected glycopeptides were from the Fab region (data not shown). In this example, both protein sequences of mAb1 and mAb2 were known; however, for biotherapeutics with less clear Fab sequences and more potential sites of glycosylation, the employment of database searching often becomes significantly less effective.
Example 7: Comparing a Base Peak Chromatogram to Oxonium Ion Profiles
[0159] This Example shows that glycan oxonium ions profiles can be used to determine the number of glycosylation sites present on a glycoprotein in a sample or reference sample. This Example also demonstrates that comparing different glycan oxonium ions profiles (e.g., a HexNAc oxonium ions profile to a sialic acid oxonium ion profile) of a glycoprotein can reveal the relative amounts of the glycans on the glycoprotein and/or at specific glycan sites of the glycoprotein. For this Example, glycan oxonium profiles were obtained for an exemplary Fc fusion protein reference sample in accordance with the methods described above in Example 5.
[0160]
[0161] The HexNAc and sialic acid oxonium profiles showed four distinct areas where the different glycosylation sites elute (
[0162] Thus, this example demonstrates that oxonium ion profiling can provide site-specific information about glycans on glycoproteins. Moreover, because different oxonium ion profiles can provide useful and different information, multiple oxonium profiles should be analyzed (and, e.g., matched) when assessing samples and/or reference samples.
Example 8: Identification of Unknown/Unpredicted Glycosylated Species
[0163] This Example demonstrates that oxonium ion profiling can provide information about the glycans present on a glycoprotein sample (including how those glycans are similar/dissimilar to the glycans on a reference sample) that may be different than the information about the glycans that can be determined by conventional methodologies (e.g., data dependent LC-MS/MS. This Example also demonstrates that oxonium ion profiling can be used to potentially detect unknown and/or unpredicted glycosylated species. To compare to the oxonium ion profiling methodology to a conventional LC-MS/MS methodology, four different Fc fusion protein clones (Sample 1, Sample 2, Sample 3, and Sample 4), as well as the three replicate reference samples (RP1, RP2, and RP3) were analyzed in accordance with the methods described in Example 5.
[0164]
[0165] Turning to the data, unknown glycopeptide species were detected in Sample 2 that were not detected by either conventional analysis or by glycopeptide database searching (data not shown). The unknown glycopeptides could be from glycosylation at an unexpected protein site, unpredictable protein modification, or from a glycopeptide containing an unusual glycan; notably, these species were not observed with sufficient abundance in the reference sample. Conversely, the HexNAc oxonium ion profile of Sample 4 exhibits obvious similarity to the reference sample. In other words, differences were observed between Sample 2 and Sample 4 when glycan oxonium ion profiling was used. This differs from the data obtained from the conventional targeted method (see
[0166]
[0167] In this analysis, Sample 4 had several glycoforms that were substantially different than the reference samples, as well as Sample 2 by conventional targeted analysis (
Example 9: Automated Similarity Assessment from Oxonium Ion Profiles
[0168] This Example describes a method for quantifying the similarity/dissimilarity of glycans and glycopeptides obtained from different glycoprotein preparations (e.g., samples or reference samples). This Example also shows a method for quantifying the similarity/dissimilarity of glycans and glycopeptides obtained from different glycoprotein preparations (e.g., samples or reference samples) can be performed at a globally or site-specifically. For this Example, four different Fc fusion protein clones (Sample 1, Sample 2, Sample 3, and Sample 4), as well as the three replicate reference samples (RP1, RP2, and RP3) were analyzed in accordance with the methods described in Example 5.
[0169] While visually assessing the oxonium profiles may be sufficient for some applications, a rapid and automated assessment of profile similarity between numerous samples will often be warranted. Thus, an algorithm that quantifies the differences of each section of the oxonium profile between samples was developed. All data was directly input into the program, loess smoothed and parametric time warped to eliminate retention time discrepancies between samples. Differences in ion abundances at every second were calculated between a set of profiles.
[0170]
[0171] To compare between methods, a global difference value was calculated from the conventional targeted results (
[0172] The similarity assessment of the conventional targeted data exhibited noticeably less discriminating power between the samples as compared to the oxonium profiles (e.g., conventional analysis yielded data that showed the reference samples were less similar to each other and more similar to the samples). The trends in similarity between the two methods were in agreement for some pair-wise comparisons, and in disagreement for others. For example, Sample 1 and Sample 3 were shown to be highly similar for each method; however, Sample 1 and Sample 4 were dissimilar by HexNAc oxonium profiling and similar by conventional analysis.
[0173] As a final assessment of the oxonium ion profiling method, the automated assessment tool was applied to quantify dissimilarity between samples site-specifically. That is, the HexNAc and sialic acid profiles were divided into sections that represent the various glycosylation sites and the differences between the reference samples and the samples at each site were calculated. The results can be seen in
[0174] In its entirety, the automated similarity assessment results shown in this section illustrate the utility of using oxonium ion profiles (e.g., HexNAc and/or sialic acid oxonium ion profiles, e.g., global and/or site-specific). Targeted conventional analysis can also be used with oxonium ion profiling when a very comprehensive comparison between samples is warranted. Being that the data independent all ion fragmentation scans to generate the oxonium ion profiles can be added directly to conventional data dependent analyses, and that the results can be quantitated automatically, the additional time commitment to produce these fingerprint comparisons is quite low.
EQUIVALENTS
[0175] It is to be understood that while the disclosure has been described in conjunction with the detailed description thereof, the foregoing description is intended to illustrate and not limit the scope of the invention, which is defined by the scope of the appended claims. Other aspects, advantages, and modifications are within the scope of the following claims.