Charge-Variant Profiling of Biopharmaceuticals

Biopharmaceuticals, particularly monoclonal antibodies (mAbs), often show the presence of distinct variants, especially because of differences in apparent or real charges versus pH. These all form a part of the overall mAb, and thus must be as fully characterized as possible using modern methods of analysis. This detailed analysis becomes even more crucial as the development of biopharmaceuticals moves from innovator products to biosimilars. This column is devoted to reviewing some methods that have evolved for such characterizations of biopharmaceuticals.

It is quite clear that an increasing number of submittals to global regulatory agencies are biopharmaceutically derived. That is, they are derived from mammalian, plant, and other cellular systems and include proteins, fusion proteins, and monoclonal antibodies (mAbs). In general, most of these newer biopharmaceuticals (biopharmas) are mammalian derived, such as from Chinese hamster ovary (CHO) or similar cell systems, using recombinant biotechnology methods. The final proteinaceous products are often mixtures of structures, all forming the drug substance, in varying percentages and structures, usually with small differences and sometimes charge differences. Hundreds of biopharmas are under development at any given time, with few going to the actual regulatory agencies for approvals. Some of these products were first introduced in the 1990s and are now coming off patents; thus biosimilars are being approved around the world for marketing and competing with the innovator drugs (1–3).

Therefore, an increasing number of proprietary biopharmas and biosimilars will be developed and submitted for approval in the coming years, demanding suitable and sufficient analytical characterization and analytical method validation (4–9). In fact, many international organizations such as the International Conference on Harmonization (ICH), Asia-Pacific Economic Cooperation (APEC) Regulatory Harmonization Steering Committee (RHSC), World Health Organization (WHO), and others have recognized this fact. These agencies have started developing centers of excellence to ensure harmonization of drugs, including biopharmaceuticals, which rely heavily on analytical characterization of these biopharmaceuticals. Here, we emphasize certain aspects of mAb characterization, and to a lesser extent other biopharmaceuticals, related to charge-variant analysis and profiling, which is but one aspect of developing well characterized biopharmaceuticals, as described for decades by the United States Food and Drug Administration (US FDA).

There are any number of original and review articles dealing with this particular aspect of biopharmaceuticals, mAbs, and characterization, and there are any number of excellent, industrial papers coming from those firms making high performance liquid chromatography (HPLC) columns and detectors, especially mass spectrometry (MS) detectors. This review does not emphasize the MS aspects for characterizing these charge variants, but rather some of the chromatographic approaches that are now possible, specifically using variations of ion-exchange chromatography (10–22). In general, weak cation exchange approaches using pH gradients (in general, salt gradients are not online MS compatible), have been more successfully interfaced with online electrospray ionization (ESI)-MS than others. There are other, quite popular ion-exchange approaches described in the literature, including weak anion exchange, with elutions using pH or salt gradients.

What Is Desirable to Learn About the Major Variants of mAbs?

Biopharmaceutical analytical testing encompasses numerous aspects of mAb structural features, and these have become numerous (10,12,13,20–22). Intact protein analysis workflows have been described in many places and are nicely summarized online (23). There are several reasons for such variants, and because mAbs are so large and have high molecular weights, these are even more numerous. Of these, charge-variant analysis plays a larger role than most types of analysis, often because of the presence or absence of a simple amino acid, lysine (Lys), at the C-termini (13). Since mAbs have two heavy chains, and either of these can have a Lys variant, there are thus three major variants present: 0 Lys, 1 Lys, or 2 Lys. It is important to recognize that the ratios of these are highly variable and can be influenced by numerous factors in the expression vectors used, as well as the mammalian cell line or others. In addition, they are readily separated, identified, and quantitated by being able to resolve one from the next, and to integrate each of these three peaks by HPLC–ultraviolet-diode array detection–mass spectrometry (UV–DAD–MS) (Figure 1) (7). Figure 1 uses DAD, with a specific pH gradient elution profile and buffers to generate each gradient (10). Of course, each Lys variant may have other variants within it, such as deamidation, oxidation, and so forth.

Figure 1: Typical HPLC-UV chromatograms of mAb Lys variants. (Adapted with permission from reference 10. Copyright 2017 Thermo Fisher Scientific Inc. All Rights Reserved.)

The ratios of each variant and absolute amounts can also be influenced by the chemical nature of the cell system, and what has been added to change these ratios. There is little medical evidence that any one of these variants is more of an effective medical agent against disorders such as rheumatoid arthritis (or other illnesses) than others. They all recognize tumor necrosis factor (TNF), which is the natural target for these mAbs, and their removal leads to temporary medical benefits such as the release from pain. However, none of these variants alone are true cures; they are almost all palliative, which supports the sales of adalimubab (Humira, 2016, $16 billion worldwide). The differences in binding constants for these variants are minimal because the complementarity determining region (CDR) is at the opposite end of the mAb to where the Lys variants are found.

However, the ability to manufacture different lots that have the same therapeutic effectiveness has been related to the ratios of these Lys variants present in any given batch of the final mAb, and to the ability of the proprietary or biosimilar manufacturers to produce the same batch characteristics day-to-day or year-to-year. Thus, an analytical method that can accurately, precisely, and reproducibly determine the relative ratios of these three Lys variants has become crucial for regulatory approvals (13). The best methods to demonstrate such ratios has been shown to be cation ion-exchange HPLC or capillary isoelectric focusing (CIEF). Interfacing cation-exchange HPLC with MS or fraction collection from the cation-exchange HPLC and manual analysis by off-line MS, has conclusively identified each of the three variants, as shown in Figure 1. In the beginning, perhaps 20 years ago, it was not tenable to directly interface cation-exchange HPLC with MS, given that most mobile phases had high salt contents using salt gradients and fewer using pH gradients. As will become evident, pH gradients make direct interfacing of HPLC with MS a simple matter. However, others have also been able to interface online salt gradients in size-exclusion chromatography (SEC) (and other HPLC formats) with MS. Here, the elution buffer is changed before the mass spectrometer to remove such inorganic salts, using a separate buffer-exchange column (for example, reversed-phase LC) just before the MS system. Ion suppressors, long used in ion chromatography, should work just as well when interfacing salt gradient ion exchange to MS for biopharmas. However, this approach apparently has not yet been described in the literature. We will come back to this issue of ease of operating ion-exchange HPLC–ESI-MS with pH or salt gradients shortly. Many routine users believe that pH gradients are more advantageous because they do not require a change of the elution buffer and salt removal before MS (10). However, if complete characterization of each Lys variant is not required in routine usage, then simpler HPLC–UV using salt gradients would be sufficient. The question is really which approach gives the better peak shapes, resolution, and peak heights for qualitative and quantitative purposes.

Beyond the Major Variants of mAbs: Other Modifications That Can Affect Charge

Charge analysis is considered a critical quality attribute (CQA) when developing a biopharmaceutical. Regulatory agencies expect to see information on charge analysis in a dossier submission, especially for a biosimilar. Thus, the biopharmaceutical companies must control potential artifacts introduced during production and analysis-that is, the less manipulation of the product the better. We now briefly consider other modifications that can affect charge.

The major variants of mAbs, as described above, are typically Lys variants; however, other such variants that affect charge (acidic or basic) may exist. These variants usually arise during process development and formulation, as mentioned previously. It is also important to note, when considering charged variants, that theoretical isoelectric points (pIs) typically appear to represent a homogenous mixture. On the other hand, we know that biopharmaceuticals, especially mAbs, are heterogenous and made up of a distribution of charges that can affect elution (24).

The term acidic species refers to a negatively charged species; using cation exchange these variants are eluted before the main peak, and when using anion exchange they are eluted after the main peak. Modifications that form acidic species (25) include

• Sialic acid

• Deamidation

• Nonclassical disulfide linkage

• Trisulfide bonds

• High mannose

• Thiosulfide modification

• Glycation

• Modification by maleuric acid

• Cysteinylation

• Reduced disulfide bonds

• Nonreduced species

• Fragments

The term basic species refers to positively charged species; using cation-exchange HPLC, these variants are eluted after the main peak, and when using anion exchange these variants are eluted before the main peak (25). Modifications that form basic species (25) include

• C-terminal Lys

• N-terminal Glu

• Isomerization of Asp

• Succinimide

• Met oxidation

• Amidation

• Incomplete disulfide bonds

• Incomplete removal of leader sequence

• Mutation from Ser to Arg

• Aglycosylation

• Fragments

• Aggregates

A first approach to characterization is often to remove the acidic (such as sialic acid) or basic (such as Lys, from Lys variants) regions, and then compare the biopharmaceutical before and after removal of the additional charge. Another approach is to just directly use ion-exchange chromatography (24,25). When ion-exchange HPLC is interfaced with online ESI-MS, the characterization of each chromatographic peak becomes fairly routine, requiring appropriate software to convert fragmentation patterns into structural features, as well as intact and fragment mass-to-charge ratios. Obviously, pH gradients are much more readily tolerated by ESI-MS requirements than salt gradients, which is why almost all biopharmaceutical characterizations today routinely use this type of ion-exchange HPLC online with ESI-MS, as described in the literature (10,17,18,24). MS has never been very compatible (in direct mode) with most HPLC buffer salt levels, but those required for almost all ion-exchange resolutions are really not compatible or tolerated. The salts accept the ESI charges before the organic analytes, and weak or no analyte signals are realized.

Why Is Ion-Exchange HPLC Better to Characterize Biopharmaceuticals and mAbs?

Ion-exchange chromatography has long been used to separate inorganic ions (the very foundation of ion chromatography), but it also works very well for organic ions, zwitterions, or very acidic or basic species, such as proteins. Ion exchange has been the traditional HPLC approach to use first for larger proteins such as mAbs. It is a resolution technique, but does not identify the specific nature of each peak, which remains the purview of MS. Hence, HPLC separation approaches must now be compatible with MS for online separation, detection, and identification as well as relative or absolute quantitations. For related species, such as proteins or mAb variants that are charged or with dipoles, ion-exchange LC has long outperformed reversed-phase, size-exclusion, or hydrophilic liquid chromatography (HILIC) modes of HPLC. However, the drawback in using ion-exchange chromatography, whether in cationic or anionic modes, is its need for a salt gradient to elute and resolve all species injected.

Salt or pH Gradient for Intact Protein Separations?

Fekete and colleagues suggested that the complexity of biopharmaceuticals (mAbs) requires a quality-by-design (QbD) approach to analyze charge-state variants (26). Numerous literature reports already exist that describe the basis of QbD, and such guidance should be followed here as well as in most HPLC applications for biopharmaceuticals.

Salt gradients, whether with inorganic or organic salts, provide much improved resolution of similar structures, even if these differ by only one or more Lys moieties. It is the pI differences-isoelectric points-between the mAb variants that allow it to be finely resolved by ion-exchange chromatography but (in the past) usually not by pH gradients (20). Gradient elution ion-exchange LC has shown superb resolving abilities for mAb variants because the salt gradient can be fine tuned to resolve almost any two species with any reasonable pI (isoelectric) differences. Again, previously this resolution capability was not usually true for pH gradient approaches. It is of interest that most of the more recent papers or application notes describing the use of pH gradient ion-exchange HPLC–MS has come from a major manufacturer of MS instrumentation (10,14–23). The immediate advantage of pH gradients, and it is a significant one, is its compatibility with online MS for detection and absolute identification purposes. However, given that using salt gradients requires buffer exchange before the MS step, which almost always involves a separate desalting column (or ion suppression device, as in IC) online with the HPLC elution, this only complicates the overall ion-exchange HPLC–MS arrangements and requires longer delay times. And, it increases the total time per analysis beyond ion-exchange HPLC–UV methods, for the very same mAbs. Given that ion-exchange HPLC–MS is usually only needed one time to characterize each variant peak, there is really no disadvantage in using ion-exchange HPLC–UV-DAD for routine mAb assays, quality control, relative or absolute quantitations, and so forth.

However, there are those who today believe that pH gradients are superior to salt gradients, and so we should discuss that premise (10,14–23). A very recent review article nicely summarizes the arguments of some of the major proponents of pH gradients, who have made an extensive comparison of both approaches with the same mAbs or proteins (10). There are also extensive discussions of these topics and ideas on various websites that should be reviewed (14–23). Recently, a typical characterization of a National Institute of Standards and Technology (NIST) mAb (RM 8671) C-terminal, lysine truncation variants using a Thermo Fisher MAbPac SCX-10 column and pH gradient mode was demonstrated (14). Figure 2 shows two chromatograms from this work (in duplicate), one with both scales expanded, each containing the NIST mAb reference material 8671 (black), and the other the same reference material but after carboxypeptidase treatment (blue) (14). It was obvious that the only change was in certain peaks, those that first contained a C-terminal Lys variant before and after carboxypeptidase treatment. This has been a standard approach for identifying precisely which peaks in such mixtures of variants of the starting mAb were because of C-terminal Lys variants (13). Perhaps the most recent books to have appeared in dealing with the state-of-the-art and emerging technologies for therapeutic mAb characterizations have been those produced by NIST and the American Chemical Society (ACS) (24).

Figure 2: Characterization of NIST mAb (RM 8671) C-terminal lysine truncation variants, using a 250 mm × 4.0 mm, 10-µm MAbPac SCX-10 column, and pH gradient mode. (Adapted with permission from reference 14. Copyright 2017 Thermo Fisher Scientific Inc. All Rights Reserved.)

Head-to-head comparisons of fully optimized ion-exchange chromatography by the two approaches on the same mAbs does not appear to exist yet in the literature. However, nevertheless there are some compelling reasons to favor the pH approach compared to the salt gradient approach. What are these?

Salt Versus pH Gradients or a Combination of Both?

To some degree, that which follows has come from various blogs by Thermo Fisher scientists, especially Cross and others, who have probably explored the use of gradient pH methods for mAbs more than most (10,20–23). Recall that in the late 1990s, the vast majority of separations for mAbs utilized salt gradient elutions because these were not yet routinely interfaced with MS online (13,19). Now that everything involving mAb HPLC has been or will be routinely interfaced with MS, there is a significant drive to make the HPLC elution compatible. In a typical ion-exchange approach, in the cation-exchange mode, the analyte mAb is in its positively charged mode, as a function of the pH of the salt gradient, and the positively charged mAb competes with sodium ions for the ion-exchange sites on the stationary phase. As the salt concentration is increased in its gradient, its sodium ions outcompete the mAb for the ion-exchange sites, and the mAb begins to be eluted. Binding and elution occur throughout this process of salt gradient elution, as a function of the steepness (and salt concentration) of its gradient, from low to higher and higher concentrations. Eventually, the mAb will be eluted as a function of the gradient, and the number of ion-exchange sites it possesses, until each variant has been eluted. Resolution is a function of the steepness of the salt gradient, the nature and concentration of that salt, the temperature, the column length, and the flow rate. Peak shapes are usually quite symmetrical, but resolution of peak variants does not appear to compete well with pH gradients, which came along years after everyone had already applied salt gradients for the same separations (12,13,19). It remains questionable whether pH gradients really provide vastly improved peak shapes, plate counts, peak capacities, or resolutions of variants present, when compared with more recent, modern salt gradients in ion-exchange LC. Head-to-head comparisons just have not appeared in the literature yet.

On the other hand, pH gradients have more recently been emphasized because of the need to directly interface mAb–protein separations with MS, as mentioned above. There is a need for both approaches because, in general, ion-exchange HPLC does not always require MS as the preferred detection mode after each peak has already been fully characterized. In the case of pH gradients, the mAb binds by ion–ion interactions, as for salt gradients, but then an increasing pH gradient is applied. And, when the net charge on the mAb becomes neutral, usually at its pI, the mAb will release and be eluted. No change in the salt concentration is needed. No further interactions with the stationary phase happen after that pH has been reached. Thus, pH gradients can be narrowed (faster elution times) over salt gradients, with outstanding resolutions possible (see Figure 1). Such resolutions do appear superior to those described in the earlier literature using salt gradient elution (13).

We had envisioned, perhaps decades ago, a possible combination of both salt and pH gradients being simultaneously used to further improve peak shapes, resolutions, and peak capacities, but that approach has not been described yet. Given what pH gradients can now accomplish (Figure 1), it does not appear that such a more convoluted elution gradient is really needed (10,14).

Approaches to Modern Ion-Exchange HPLC–ESI-MS Using Efficient pH Gradient Elution of mAbs and Other Proteins

In a recent series of poster and online presentations, as well as at various symposia, what appears to be the state-of-the-art in interfacing native ion exchange chromatography directly to an orbital trap mass spectrometer has been described (14,17,18). A goal in these studies was the use of organic salts to generate the pH gradient, together with a weak cation exchange column, a modern ultrahigh-pressure liquid chromatography (UHPLC) system, an orbital trap MS system, and mass informatics software. One particular application note (14) illustrates the specific buffers, column, flow rate, and gradient used to generate the final chromatogram, leaving the mAb in its native state, presumably with its native conformations and ionization states (different than for denatured mAbs or proteins). There are certain, real advantages in using native MS over traditional, denatured states of proteins, including a reduced degree of ionization, more intense peak heights because of fewer ionized states, and lowered limits of detection. In theory, during native MS, the protein remains in its native or natural state and is not denatured as in most of the previous HPLC–MS studies of proteins before the introduction of native MS. There is not yet any evidence that native MS leads to higher sequencing abilities of proteins, as in top-down sequencing (TDS) routines, but this possibility remains an attractive thought.

Finally, what about the chromatogram in Figure 2 (14)? What was of interest was the presence of an early peak coming from the buffer salts before elution of the protein variants was induced by increasing the percentage of buffer B. That peak was detected early in the LC–MS run, long before the mAb peaks appeared at longer and longer retention times. Each of these peaks presumably represents a different variant of the mAb, which could now be fragmented and sequenced by other MS routines, as desired, or just quantitated as relative or absolute percent compositions versus authentic external standards. The overall MS chromatogram then defines each batch of the mAb after each peak has been identified and characterized, as well as quantitated. Thus, one could have both online DAD and MS to obtain traditional ion-exchange HPLC–UV for relative or absolute quantitations, on a routine basis, as well as MS identification of each peak on a semiroutine or quality control (QC) basis. Obviously, being able to quantitate and identify, with 100% assurance, the nature of each peak in this "typical" weak cation exchange HPLC–MS method has serious advantages compared to using just weak cation exchange HPLC–UV-DAD for research, routine QC, or plant monitoring of expression–batch cleanup methods, before packaging. The elution order, retention times, peak areas, and absolute peak quantitation options all then become a unique picture of each batch of product that is not possible to obtain using most QC methods now in practice for research, expression, or production purposes.

Perhaps to emphasize the above points, Figure 3 illustrates a somewhat older chromatogram using salt gradient elution for a different mAb, the predecessor to adalimubab (Humira), with conditions as indicated in the literature (13). It is clear that, though state-of-the-art in 1999, peak shapes, peak symmetry, peak resolutions, peak capacity, and other variants present or obvious do not compare favorably with either Figures 1 or 2.

Figure 3: Analysis of antibody D2E7 by cation-exchange HPLC: (a) the native mAb, with peaks as described, (b) the CPB digested mAb, with only the main peak of C-terminal 0-Lys remaining, as expected (13, Figure 3). Reprinted by permission of the copyright holder and RightsLink, Inc.

One final point about the relative broadness of the mAb variant peaks and incomplete resolutions: This broadness is perhaps typical of the ion-exchange mechanisms of separation, and the on-off-on-off nature of such separations. It is, really, inherent with all ion-exchange separations, unless one uses UHPLC, elevated temperatures, and low-viscosity eluents to reduce the viscosity of the mobile phase and improve separation kinetics. The final chromatographic picture of the mAb, Figure 2, does not yet show 100% baseline resolution or separations of all variants, nor the ability to quantitate each and every variant. However, it is well on its way to such goals (27). If we just compare batch-to-batch, chromatographic profiles with computer comparisons, it should still be possible to conclude batch-to-batch replicability and reproducibility within experimental variability.

Conclusions

The analytical characterization of recombinant antibodies and biopharmaceuticals has become a science unto itself, with symposia, books, journals, courses, scientific meetings, and conferences devoted to just that topic. It is an incredibly broad and complex subject, as evidenced by the two, very recent volumes just published by the ACS and written by several individuals within NIST (24). It involves different, analytical instrumentation, far beyond just HPLC and MS, and involves higher order structure determinations, biological activity measurements, and binding constant studies, as well as hosts of biological assays, before any new biopharma is even ready for animal or human clinical testing. The determination of major variants, differing in both primary and secondary structures, is commonly done by a combination of HPLC, high performance capillary electrophoresis, MS, and other detection schemes today. All of these methods and many others are now being used in making direct comparisons between biosimilars and innovator biopharmas. The use of ion-exchange HPLC, with or without MS online, has undergone significant improvements and changes in just the past two to three decades. We are at the point of being able to chromatographically resolve most, if not all, of the common variants present, depending on how much oxidation, disulfide reduction, rearrangements, deglycosylations, reductions, and other chemical changes have occurred between expression and structural determinations, using ion exchange and many other separation schemes (14,24–27). It appears that the ability to separate and identify the major structural isomers in a typical mAb will always be needed in the future, until we are able to identify and quantify all post-translational modifications in the expressed biopharmaceutical.

Acknowledgments

We acknowledge with gratitude and appreciation the interest and encouragements shown by J. Bones during the preparation of this column.

References

(1) "Samsung, Biogen version of AbbVie's Humira approved in Europe," Health News (Reuters), August 24, 2017. https://www.reuters.com/article/us-biogen-humira/samsung-biogen-version-of-abbvies-humira-approved-in-europe-idUSKCN1B5041.

(2) "FDA approves Amjevita, a biosimilar to Humira," FDA News Release, September 23, 2016, https://www.fda.gov/newsevents/newsroom/pressannouncements/ucm522243.htm.

(3) "Biopharmaceutical Research Industry, Pharmaceutical Research and Manufacturers of America, 2016 Biopharmaceutical Research Industry Profile,"

(PhRMA, Washington, D.C., 2016).

(4) I. Apostol, I. Krull, and D. Kelner, Anal. Chem. I.S. Krull, Ed. (InTech- Open Access Publishers, Rijeka, Croatia, 2012). http://www.intechopen.com/articles/show/title/analytical-method-validation-for-biopharmaceuticals.

(5) R.J. Duff, T. Eris, D. Kelner, and I.S. Krull, Journal of Validation Technology 22(6), 1–8 (2016).

(6) R.J. Duff, T. Eris, D. Kelner, and I.S. Krull, Journal of Validation Technology 23(1), 1–12 (2017).

(7) R.J. Duff, T. Eris, D. Kelner, and I.S. Krull, Journal of Validation Technology 23(2), 1–17 (2017).

(8) R.J. Duff, T. Eris, D. Kelner, and I.S. Krull, Journal of Validation Technology 23(3), 1–12 (2017).

(9) D. Wilkinson, A.S. Rathore, and I.S. Krull, LCGC North Am. 35(1), 36–44 (2017).

(10) K. D'Silva, K. Cook, and S. Cubbon, The Column 13(9), 28–32 (2017).

(11) M. DiPaola, BioPharm Intl. 30(8), 38–43,46 (2017)

(12) L.C. Santora, Z. Kaymakcalan, P. Sakorafas, I.S. Krull, and K. Grant, Anal. Biochem. 299, 119–129 (2001).

(13) L.C. Santora, I.S. Krull, and K. Grant, Anal. Biochem. 275, 98–108 (1999).

(14) S. Milan, A. Farrell, and J. Bones, "Using the NIST mAb Reference Standard to Demonstrate a Simple Approach to Charge Variant Analysis," Thermo Scientific, Application Note 21684.

(15) M. Singh and S. Pandey, "The Advantage of pH Gradient Buffers Is Demonstrated by the Ion Exchange Separation of Charge Variants of Denosumab," Thermo Fisher Scientific Application Note, 2016, AN21565-EN1116S.

(16) "Advances in Biopharmaceutical Characterization: Charge Variants and Aggregates," June, 2017, Thermo Fisher Scientific.

(17) A.O. Bailey, G. Han, W. Phung, P. Gazis, J. Sutton, W. Sandoval, and J.L. Josephs, "Native Ion Exchange Chromatography Directly Coupled to Orbitrap Mass Spectrometry Allows Surface Charge Discrimination and Online Detection of Intact Proteins," poster presented at 65th ASMS Conference on Mass Spectrometry and Allied Topics, Indianapolis, Indiana (2017)

(18) A.O. Bailey, "Unlocking Combined Biologic Characterization Strategies using Peptide Mapping and Native Intact Analysis on the Thermo Scientific Q Exactive BioPharma MS," presentation at the September 21, 2017, Thermo Users Meeting, CaSSS MS in Biotechnology 2017, Park Plaza Hotel, Boston, Massachusetts, 2017.

(19) D. Farnan and G. Moreno, Anal. Chem. 81(21), 8846–8857 (2009).

(20) T. Cross, "Salt or pH Gradient for Intact Protein Separations?" Thermo Fisher Scientific, Inc., blog (2017). analyteguru.com/salt-or-ph-gradient-for-intact-protein-separations/.

(21) R. Moore, "The day I Met 'Charge Variant Profiling' – Turns Out It's Pretty Basic," Thermo Fisher Scientific, Inc., blog (2017). analyteguru.com/the-day-i-met-charge-variant-profiling-turns-out-its-pretty-basic/.

(22) http://www.thermofisher.com/us/en/home/industrial/pharma-biopharma/biopharmaceutical-analytical-testing.html.

(23) https://www.thermofisher.com/us/en/home/industrial/pharma-biopharma/biopharmaceutical-analytical-testing/intact-protein-analysis-workflows.html.

(24) D.A. Michels, A.Y. Ip, T.M. Dillo, K. Brorson, S. Lute, B. Chavez, K.M. Prentice, L.J. Brady, and K.J. Miller, in State of the Art and Emerging Technologies for Therapeutic mAb Characterization, Volume 2, (Biopharmaceutical Characterization: The NIST mAb case study, American Chemical Society, Washington, D.C., 2015). Chap. 5, pp. 236–284.

(25) Y. Du, A. Walsh, R. Ehrick, W. Xu, K. May, and H. Liu, mAbs 4(5), 578–585 (2012).

(26) S. Fekete, A. Beck, J. Fekete, and D. Guillarme, J. Pharm. Biomed. Anal. 102, 33–44 (2015).

(27) J. Bones, NIBRT, unpublished results, October, 2017, Dublin, Ireland.

ABOUT THE AUTHORS

Ira S. Krull

Ira S. Krull is a Professor Emeritus with the Department of Chemistry and Chemical Biology at Northeastern University in Boston, Massachusetts, and a member of LCGC's editorial advisory board.

Jared R. Auclair

Jared R. Auclair is currently the Director of Executive Training and Biotechnology Programs in the Department of Chemistry and Chemical Biology at Northeastern University. In addition to being Director of Biotechnology, Dr. Auclair also directs the Biopharmaceutical Analysis Training Laboratory and the Asia-Pacific Economic Cooperation Center of Regulatory Excellence in Biotherapeutics. This latter appointment allows Dr. Auclair to collaborate with both academic researchers and industry in the area of biopharmaceutical development and analysis. He has expertise in molecular biology, protein biochemistry, analytical chemistry, protein crystallography, and biological mass spectrometry. He is interested in understanding the molecular mechanisms of neurodegenerative diseases as well as advancing diagnostics for women's health.

Anurag S. Rathore

Anurag S. Rathore is a professor in the Department of Chemical Engineering at the Indian Institute of Technology in Delhi, India.