Ben Moree is principal scientist at Quantum-Si.

Ben Moree

It’s no secret that proteins dominate the world of biology. Human cells have more than 20,000 proteins and over one million proteoforms, orchestrating countless mechanisms within the body (1). Looking beyond a natural physiological role, biologics are a mainstay in the biopharma industry’s repertoire of therapeutic modalities. Monoclonal antibodies (mAbs) and other protein-based biologics can be found in most modern therapeutic development pipelines. The industry’s collective knowledge of these biomolecules and their interactions is vast, yet there is still much to learn. Small variations can have significant consequences in the clinic. An accurate understanding, down to each amino acid and small variation, is key to the discovery and development process, requiring broad access to single-molecule protein sequencing technology.

Remarkable advancements have been seen in molecular biology and engineering in the past three decades, resulting in the modern revolution in medicine today (2). Protein-based therapeutics offer a host of benefits when compared to traditional small molecules, including the potential for greater target specificity and improved pharmacokinetics (PK) (3).

The prevailing role for proteomics in drug discovery and development is from an analytical and quality control (QC) perspective. Working with biologics requires a sequence-level understanding of the molecule at each stage of development in order to assess and instill the desired properties in the investigational drug. Their molecular complexity, however, poses a challenge.

With complexity comes opportunity for variation, which can be introduced during transcription and/or translation, by misincorporation at various levels including the DNA, RNA, and amino acid level, and by the introduction of co-translational modifications (CTMs) or post-translational modifications (PTMs) (4). Post-translational modifications such as methylation, acetylation, glycosylation, ubiquitination, phosphorylation, or other modifications can change the way proteins stabilize, fold, function, and localize (4). For example, among the long list of possibilities that can alter the behavior of a biologic in clinically significant ways is errant replacement of an amino acid in amino acid misincorporation. Variations such as these may alter a biologic’s charge and size, reduce or alter target affinity, or lead to unintended aggregation (4–7).

Given the significant opportunity for deviation from the intended product that can occur within the synthesis and larger manufacturing process, these molecules and their variants must be closely monitored and tested in the preclinical setting as they are characterized and optimized as biological drug candidates and, later, in the clinical setting as their PK and therapeutic concentrations are closely scrutinized.

 In early drug discovery, fundamental sequence understanding is key, particularly as multiple variants of the biologic are intentionally produced. The overarching goal is to find the optimal biologic drug candidate or variant, including considerations such as function, specificity, stability, and safety, among others. Understanding of the investigational asset at single molecule amino acid resolution provides information regarding how specific changes within the sequence of that biologic drug impacts the function of that molecule and provides a detailed mechanistic understanding of what is driving biological interactions. Proteomics workflows and biophysical techniques are used together to characterize sequence and function, which are often directly related to one another.

Clinical trials. As molecules move into clinical trials in humans, they are exposed to entirely new parameters. Protein activity can be affected by interactions with small molecules, cofactors, metals, nucleic acids, and other proteins, which may change with variations in protein sequence. Variations can also impact target-binding affinity or promote off-target binding. These factors make a detailed understanding of structure important (8–9). Having confidence that the biologic drug is being made consistently with low batch variation may allow more confidence and insight in the interpretation of clinical trial data and ensure continuity in effect when molecule production is scaled up for commercial availability.

 In manufacturing, lot purity must be maintained as cell-line mutations and amino acid misincorporations can cause major supply chain problems. Significant effort is put into maintaining the integrity of cell lines with regard to biologics that are recombinantly expressed; however, there are still opportunities for deviation. Being able to look at a manufacturing product at amino-acid-level resolution provides insight into the degree of consistency within the product over time and batches.

Taken together, these considerations make proteomics tools indispensable to scientists and technical personnel at each stage of the therapeutic lifecycle.

Current sequencing and mapping techniques

Biologic drugs are complex molecules produced by living cells through sophisticated manufacturing processes. As one of the fastest growing segments in the bio/pharmaceutical industry, biologics have the potential to provide targeted therapy and treat diseases that were once considered untreatable (10). However, successful biologic discovery comes with many challenges, such as high costs, workflow complexities, and end-product safety considerations (11).

Biologics are more complex due to their larger structures and complicated processes in their analysis, development, and manufacturing. When developing a biologic, critical quality attributes such as structural elucidation, physiochemical, immunological, biological properties, as well as impurities and contaminants (12), must be defined and well documented to ensure successful therapy and delivery. Characterization of these attributes are performed with bioanalytical tools that include chromatographic, electrophoretic, spectroscopic, and electrochemical techniques (12).

 Rising to the top of these methods is liquid chromatography–mass spectrometry (LC–MS). LC–MS is a key tool in a scientist’s toolbox and often the available method of choice for protein identification at varying levels of detail (13). The sample of interest is digested into protein fragments that will be in-range for the instrument’s sensitivity and applied to a liquid chromatography column where particles are separated by size and other biophysical properties. Reverse-phase liquid chromatography is the most common choice and uses a hydrophobic stationary phase with separation via hydrophobicity with a polar mobile base. The sample is then ionized and sent over a detector, yielding a molecular mass to charge ratio (m/z).

While LC–MS is an extremely powerful technique, it is equally as complex, with a multitude of possible workflows and expertise required in sample preparation, instrument operation, and output analysis. The resultant spectrum must be deconvoluted into the appropriate peptide fragments. Available automated tools make this process easier; however, subtle changes in m/z and deviations from what may be expected within the workflow can result in misinterpretation of the data. Further, results depend on the existence of a m/z difference between the samples and the ability to separate those samples on the LC column, which is not always possible. Some peptides may not even elute from the column due to strong hydrophobicity. In instances of PTMs, where there may not be a mass change or a mass change on the order of Daltons, LC–MS may not offer the sensitivity required. In instances where there are multiple PTM possibilities within a sequence, the results may indicate those PTMs exist but not their location, obscuring insight into the frequency and heterogeneity of PTMs that can be critical for the function of biological drugs.

 Immunoassays, such as Western blot and enzyme-linked immunoassay (ELISA), are also within the scientist’s toolbox, owing to their ability to quickly identify proteins in a sample. Immunoassays could confirm the presence of PTMs or that the biologic drug is interacting with the target of interest, although they don’t provide sequence-level information. Western blot separates proteins by molecular weight through gel electrophoresis, which are then moved to a membrane and analyzed via antibody binding. ELISA also uses antibodies to detect a protein sample; it is easy to use with the potential for higher throughput as compared to Western blot. Conversely, Western blot can analyze a more complex protein mixture.

Both of these methods are easily accessible tools that offer high degrees of sensitivity and low sample input. However, they require customized reagents, such as high-quality antibodies, for each and every target of interest and additional analysis in order to get protein sequence or individual amino acid information. Further, challenges can arise in the reproduction of results due to biological variation (i.e., polyclonal vs. monoclonal antibodies) leaving a need for enhanced QC (14).

Each of these methods involve trade-offs between ease of use and level of detail in protein identification data, highlighting the need for technological advancements that make detailed proteomics more accessible and targeted toward the field’s evolving needs.

Next-generation sequencing in genomics catapulted science forward. It enabled cost-effective parallel sequencing of millions to billions of DNA fragments and has been applied in investigating molecular-level variation in cancer, rare genetic diseases, and complex microbiomes (15). Single-molecule protein sequencing has the potential to be just as revolutionary. Understanding of biologics on a single amino acid level can provide deeper insights into what may be driving biologic function against a target, allowing optimization on a level that otherwise is unattainable. In larger-scale production, single-molecule protein sequencing may allow more precise quantification of changes occurring from batch to batch, enabling close monitoring of the types of events that are impacting product quality.

Overall, single-molecule protein sequencing technology has the potential to decrease cycle time while facilitating improved quality of biologic drugs available through improved stability and less batch-to-batch variation. Having tools available that enable deeper insights can reduce development time and ultimately get new drugs to patients faster.

The industry is facing a critical challenge within the field: elevating analytical techniques to match the sensitivity that is now required within the drug discovery and development process. There is a great need for sensitive and unambiguous amino acid detection at a scale and ease accessible to broader populations of scientists. If more researchers can “see” proteins with this level of clarity, then there is a tremendous opportunity to transform the way they apply proteins in research and medicine.

1. Aebersold, R.; Agar, J.N.; Amster, I. J.; et al. How Many Human Proteoforms are There? 

 (3), 206–214. DOI:10.1038/nchembio.2576
2. Chen, Z.; Wang, X.; Chen, X.; et al. Accelerating Therapeutic Protein Design with Computational Approaches Toward the Clinical Stage. 

, 2909–2926. DOI:10.1016/j.csbj.2023.04.027
3. Oo, C.; Kalbag, S. S. Leveraging the Attributes of Biologics and Small Molecules, and Releasing the Bottlenecks: A New Wave of Revolution in Drug Development. 

 (6), 747–749. DOI:10.1586/17512433.2016.1160778
4. Jefferis, R. Posttranslational Modifications and the Immunogenicity of Biotherapeutics. 

, 5358272. DOI:10.1155/2016/5358272
5. Samodova, D.; Hosfield, C. M.; Cramer, C. N.; et al. ProAlanase is an Effective Alternative to Trypsin for Proteomics Applications and Disulfide Bond Mapping. 

 (12), 2139–2157. DOI:10.1074/mcp.TIR120.002129
6. Wong, H. E.; Huang, C. J.; Zhang, Z. Amino Acid Misincorporation in Recombinant Proteins. 

 (1), 168–181. DOI:10.1016/j.biotechadv.2017.10.006
7. Bradbury, A. R. M.; Trinklein, N. D.; Thie, H.; et al. When Monoclonal Antibodies are Not Monospecific: Hybridomas Frequently Express Additional Functional Variable Regions. 

 (4), 539–546. DOI:10.1080/19420862.2018.1445456
8. Po, A.; Eyers, C. E. Top-Down Proteomics and the Challenges of True Proteoform Characterization. 

 (12), 3663–3675. DOI: 10.1021/acs.jproteome.3c00416
9. Mateus, A.; Savitski, M. M.; Piazza, I. The Rise of Proteome-Wide Biophysics. 

 (7), e10442. DOI:10.15252/msb.202110442
10. Pharmaceutical Technology. High Biologics Demand Spurs Need for Greater Contract Manufacturing. 

pharmaceutical-technology.com/sponsored/biologics

, April 24, 2023.
11. Olsen, C. Challenges in Biologics R&D. 

, Nov. 7, 2022.
12. Alhazmi, H. A.; Albratty, M. Analytical Techniques for the Characterization and Quantification of Monoclonal Antibodies. 

 (2), 291. DOI:10.3390/ph16020291
13. Song, J. G.; Baral, K. C.; Kim, G. L.; et al. Quantitative Analysis of Therapeutic Proteins in Biological Fluids: Recent Advancement in Analytical Techniques. 

 (1), 2183816. DOI:10.1080/10717544.2023.2183816
14. Baker, M. Reproducibility Crisis: Blame It on the Antibodies. 

, 274–276. DOI.org/10.1038/521274a
15. Satam, H.; Joshi, K.; Mangrolia, U.; et al. Next-Generation Sequencing Technology: Current Trends and Advancements. 


Volume 48, No. 3
March 2024
Pages 27–29

When referring to this article, please cite it as Moree, B. Biologics Quality Control: The Growing Need for Accessible Proteomics. 

Articles

There is a great need for sensitive, precise, and easily accessible analytical detection techniques for protein sequencing.