Introduction

Mass spectrometry plays an increasingly important role in the characterization and comparison of innovator protein products as well, as biosimilar samples in research and development labs. There is increasing interest in using mass spectrometry in QC labs. The application of Multi-Attribute Methods (MAM) to characterize biotherapeutics, if done in a consistent manner, could supplement or even supersede traditional analyses. Underscoring the interest in this approach, NIST is participating in a cross-industry MAM Consortium which recently produced a new peak detection interlaboratory study (J. Am. Soc. Mass Spectrom. 2021, 32, 913−928). The FDA has also identified MAM as an emerging technology for biological molecules. In general, there are two main routes to establish a QC-ready MAM. Historically, chromatography-based methods were preferred in QC settings. MS feature-based methods are now coming to the forefront. The good news is that both options are available in Protein Metrics Byos® software. Regardless of the monitoring method selected, any MAM consists of two key steps; first, characterization of the molecule to identify Critical Quality Attributes (CQAs) and second, routine monitoring. Before rolling multi-attribute methods out to a QC organization, a strong fundamental knowledge of the molecule is essential. From this knowledge, organizations build the essential understanding of their molecule that not only allows them to identify the CQAs that are to be monitored, but also develop a robust library of known peptides, modifications, artifacts, and other stress or process induced changes that allow them to quickly identify changes in process lots.

 

 

This document (part A of a series), highlights the application of Byos software to the careful characterization of molecules for the purpose of identifying CQAs and building a library of molecular attributes. This library will be key input for later a routine monitoring and new peak detection analyses using Byos software. Part B of this series will cover the use of Byos for feature based, routine CQA monitoring and new peak detection. It is important to note that Byos software is incredibly flexible and if there is a need to perform CQA validation and comparison all in one step, this is possible. In that event, simply skip this document and move directly to Part B of this series.

Summary

• Byos Characterization Workflow (PTM)
• Analytical Notes
• PTM Workflow Configuration
  • Digest Settings
  • Modification, glycan and input settings
  • Advanced settings
• Project review
  • XIC inspection
  • CQA labels
  • CQAs and knowns library exports
• Producing Characterization Reports
• Appendix

Byos MAM CQA Identification Workflow (PTM)

• Click the PTM icon to launch a workflow

 

• Three steps to project creation

Analytical Notes

The first step of the MAM workflow has two parts. First, carefully characterize the molecule. Second, construct a library of CQAs and known peptides. The CQAs are the bellwether peptides that indicate there may be a problem with a new material lot (e.g. increased oxidation or isomerization). A general library of knowns, while optional, can also be very helpful. For example, while over- or under-alkylation, cation adduction, or missed cleavages may indicate a sample processing issue, they generally do not indicate a process problem. However, when sample processing artifacts appear in a QC sample, it is likely better if the analyst quickly distinguishes them from a genuine issue with a new lot. If these known species are added to the CSV library, new peaks can be expeditiously identified even in MS1- only processes. Moreover, including stressed samples in characterization work will highlight peptides susceptible to changing conditions. These peptides may also serve as good indicators of product quality. 

 

PTM Workflow Configuration

Instrument Parameter and Digest Settings

As required in all Byos MS2 workflows, the Instrument and Digest settings should be appropriate for the sample. Select the correct enzyme and specificity to match the digest. Precursor and fragment tolerances should be set according to actual instrument accuracy. If a lock mass is present in the file, do not forget to add it both here and later in the Quant (Byologic) panel in the “Advanced” section. Be sure to select the correct fragmentation type for the sample as well.

 

Modification, Glycan & Input Settings

The PTM workflow Modification and Glycan settings are preconfigured for general searches to include common modifications and artifacts, but they can be modified as needed. Give special attention to the sort of modifications that may later show up when a new lot is being compared. If these types of modifications are added to a library of knowns used in the search and identification, this can eliminate later mysteries for QC staff. Examples of this might include over- or under-alkylation or adducts that might change depending on reagents or LC systems.

 

Advanced Settings

As Protein Metrics introduces new features, to make them more quickly available to our customers, they are sometimes added as “Advanced Configurations”. While the default PTM workflow does not require them, there are several that are worth adding to speed project execution and simplify analysis.

MS2PeptideCombined count permits a summed view of observed MS2 spectra, while AdjustPrimaryProtein assures that peptides from all samples are assigned to the same primary protein (if the peptide occurs in multiple proteins).

 

[Byologic]

MS2PeptideCombinedCount=5
AdjustPrimaryProtein=true

 

Project Review

Upon completion of project execution, Byos will open an inspection tab and a second tab holding a report based on the project. Click the project tab to allow a detailed inspection of the peptide identifications, along with their associated XICs, MS2s, and all the information needed for a thorough review.

The goal of project review is to identify, validate, and label potential CQAs. At the same time, information about all the known product attributes is at hand as well. This data can later be used as an attribute library for annotation of QC samples. The key peptides to inspect include those in CDR regions, and those with glycosylations, deamidations, isomerizations, or oxidations which might be used as CQAs. Protein Metrics offers application notes dedicated to glycosylation, deamidation, and isomerization identification and quantification.  The result of the review should be a project with true peptides validated with XICs confirmed and any unwanted peptides marked as “False Positive” and excluded.

XIC Inspection

Byos peak identification algorithms are quite good, but there will always be challenging cases where the software leaves a few peaks which may need adjustment. This is most common on deamidations and oxidations where there can be confusion with wildtype artifacts. For example, in the “PENNY” peptide shown below, the peaks for the NN, NG, and isoNG deamidation sites area all poorly spaced from one another and the wildtype peak. In situations like these, manual confirmation and adjustment of XICs assures quality results.

Any identified issues are quickly corrected by clicking on that peptide and then simply dragging the black bars that mark the XIC boundary to a new location. Alternatively, it is possible to adjust the same peptide in multiple files by right clicking on the peptide in the peptide table (1) and selecting the “Set XIC time Interval” option (2). From here, a new window will pop up that will allow the begin and end XIC time range to be set for all files. Additionally, a small tolerance can be added to allow automatic adjustments in each file for retention time shifts (3).

 

These updates will immediately yield new area calculations and XIC ratios in the project. In addition to adjusting XIC limits, other information in the XIC can also be quite helpful. For example, do the MSMS hits (red and pink dots) correspond with the XIC? In complex cases like the “PENNY” peptide, assuring the MS2 of a given form matches the RT of the peak can help differentiate unique deamidation sites or isomerizations. Intervening grey dots can show that another species with the same mass is present.

 

Since the XIC of the modified form is automatically shown in tandem with the unmodified form, it is possible to quickly assess coelution with wildtype. Species that exactly coelute with the wildtype are likely to be artifacts that occurred post-column or in-source. See the manual for automated validators that are available to highlight these for rapid identification.

 

 

CQA Label

Once CQA peptides are identified, both the modified and unmodified forms of the peptide should be labeled as such. In addition to allowing for the insertion of CQA labels, Byos allows for the creation of any custom dynamic label. To add the CQA labels (or another label), click on the label manager button in the upper left corner of the Byos window. This will open a dialog that will allow for the addition or update of labels. Once added, simply click the OK button.

 

To label individual peptides (or even groups), click (or Ctrl + click) on the rows to be updated and then left click on the label column to bring up the labeling dialog. In this dialog box, add a check mark for any appropriate labels.

CQAs and Knowns CSV Library Exports

Once CQAs are labeled and any false positives are filtered from the project, exporting the library of knowns and CQAs is a simple matter of a few clicks. Select Edit > In-silico Peptides > Export In-silico To CSV to bring up a file save dialog box. Enter a file name when requested and a CSV library will be exported.

 

The CSV file generated in this step can be used as the input into subsequent evaluations of new materials or process lots using the MAM New Peak Detection workflow described in detail in the second document in this series: Multi-Attribute Monitoring in Protein Metrics Byos Software - B.

 

Byos opens not only the inspection view, but also the corresponding report as an additional tab. Depending on the starting workflow, the default report configuration includes key tabs such as a summary tab, and several tabs showing percent modifications. Regardless of the initial report, it is always possible to switch to another from our library or add a custom one if desired. One key preset report that can be helpful in determining if a particular modification might make a good CQA is the “MAM – Modified Peptides” report which is the second tab of the Blgc_MAM report. To access the report, select File > Presets > Report Presets > Blgc_MAM.rptc.

 

 

Characterization Reports

Once the report opens, the second tab, MAM – Modified Peptides, contains a summary of the relative abundance of all modified peptides across all samples. The report also contains key information such as protein annotation (indicating a CDR peptide), modifications, and digest specificity that might help determine if a candidate peptide form might make a reliable CQA.

 

Note that as updates are made to the project, they will immediately yield new calculations in the project, but in the report, data must be refreshed. To do so, in the report tab menu choose Tabs > Update tab content.

 

For more information on how to customize reports, please see our related videos available at https://www.proteinmetrics.com/resources#videos-tutorials or contact the Customer Success Team at support@proteinmetrics.com.