Discovering Unusual Biological Product Characteristics in Antibodies with Next-Generation Sequencing
March 10, 2022
Marie-Ange Kouassi
Understanding and controlling glycosylation patterns is essential for ensuring the quality, safety, and consistency of therapeutic antibodies produced in CHO cells. Biological molecules provide promising treatment options allowing to alter disease mechanisms with precision. Yet, the development of these molecules can be challenging. During the development process, it is critical to check for changes that may occur and affect the functionality, safety, and efficacy of these therapeutic products.
Sialylation patterns play a critical role in determining the safety and efficacy of therapeutic antibodies. Among these terminal glycans, NGNA (N‑glycolylneuraminic acid) can appear in recombinant protein production systems and is closely monitored due to its potential to trigger unwanted immunogenic responses in patients. In contrast, NANA (N‑acetylneuraminic acid) is the predominant human sialic acid and represents the expected, biologically compatible glycosylation pattern on therapeutic antibodies, supporting normal receptor interactions and favorable pharmacokinetics.
Because NANA is the desired human‑like sialylation profile, the unexpected detection of NGNA instead of NANA in CHO‑derived antibodies immediately signals a potential quality concern—one that warrants deeper molecular investigation to protect product consistency and patient safety.
The Importance of NGS and Genomic Analysis Platforms in Biologic Troubleshooting
High‑throughput sequencing has become an essential tool for rapidly diagnosing unexpected phenotypes in biologics development. By integrating NGS with advanced genomic analysis, biopharma teams can uncover the molecular root causes of product deviations with far greater accuracy and speed than traditional analytical methods. This data‑rich approach not only accelerates issue resolution but also strengthens regulatory confidence by providing a comprehensive, traceable understanding of cell line behavior.
Modern bioinformatics platforms further enhance this workflow by managing large, complex datasets and transforming them into actionable insights. With unified environments for genome assembly, mutation profiling, and transcriptomic analysis, scientists can quickly pinpoint the drivers of manufacturing challenges and make informed decisions that keep development timelines on track.
The Challenge of NGNA Sialylation Detected Instead of NANA in CHO Cell-Derived Antibodies
Successful cell line development requires routine screening to ensure continuous generation of the desired product. As these biological molecules are purposely intended as treatment, it is critical to ensure they do not change in a way that makes them ineffective in targeting disease, or detrimental to patient health. By screening production cell lines at Boehringer Ingelheim, Fischer et al. identified an unusual characteristic of monoclonal antibodies produced by their proprietary Chinese hamster ovary (CHO) cells. The antibody generated in one cell line exhibited a high level of sialylation with NGNA (N-glycolylneuraminic acid) instead of the typical NANA (N-acetylneuraminic acid). Knowing this could have adverse effects in patients by inducing an elevated immunogenic response, the team decided to investigate the reason for this unusual phenotype. They used Next-Generation Sequencing (NGS) applications to delve deep into the root cause and identified a new microRNA (miR-111). They also identified a single point mutation that altered a binding site allowing the binding of a different transcription factor and causing the downregulation of miR-111. The lack of miR-111 allowed an unusual expression of cytidine monophosphate-N‐acetylneuraminic acid hydroxylase (CMAH), leading to increased NGNA antibody sialylation.
To investigate the phenotypic differences between the two clones (one with normal NANA sialylation and another with high NGNA sialylation), Fischer et al. generated a cell-specific reference genome knowing its importance in serving as an accurate reference for all sequencing data analysis performed. Genedata Selector® was used to assemble and annotate the CHO cell genome which was enhanced with genomic DNA and RNA data specific to Boehringer Ingelheim’s cell line. Following NGS, mutation profiles were generated, and gene expression patterns were analyzed. Advanced statistical analyses (e.g., principal component and correlation analyses) were performed using Genedata Selector to obtain a deeper biological understanding of the underlying data.
Experimental Validation Using CRISPR Gene Editing
To functionally validate the molecular drivers identified through NGS, Fischer et al. applied targeted CRISPR gene editing to dissect the regulatory mechanisms behind the abnormal NGNA sialylation phenotype. Knocking out CMAH, the enzyme responsible for converting CMP‑NANA to CMP‑NGNA, immediately eliminated NGNA incorporation and restored the expected human‑like sialylation pattern. This confirmed that CMAH activity was directly responsible for the phenotype observed in the affected clone.
To further investigate why CMAH was aberrantly expressed, the team performed differential expression analysis of small non‑coding RNAs and identified miR‑111 as a key regulatory microRNA that was significantly downregulated. Reintroducing miR‑111 into the high‑NGNA clone suppressed CMAH expression and reduced NGNA levels, demonstrating its role as a negative regulator of CMAH. Genedata Selector® was then used to design the single‑guide RNAs for CRISPR editing and to predict transcriptional binding sites, revealing that miR‑111 resides within an intron of the SDK1 gene and is co‑regulated by its promoter.
Genome sequencing data uncovered a single point mutation in the SDK1 promoter that created a novel binding site for HINF‑P, a repressor highly expressed in CHO cells. This mutation silenced both SDK1 and miR‑111, releasing CMAH from its normal repression and driving the unexpected NGNA sialylation. Together, these CRISPR‑based experiments provided precise, mechanistic confirmation of the NGS‑derived hypothesis and demonstrated how integrated omics and gene editing can rapidly resolve complex cell‑line quality issues.
Genedata Selector’s Role in Accelerating Biologic Characterization
Genedata Selector® played a central role in enabling the rapid, data‑driven investigation of the NGNA sialylation phenotype. By providing a unified environment for genome assembly, mutation profiling, transcriptomic analysis, and regulatory element prediction, the platform allowed researchers to seamlessly integrate multi‑omics datasets and pinpoint the molecular drivers behind the abnormal glycosylation pattern.
The ability to design CRISPR single‑guide RNAs directly within the platform further streamlined experimental validation, ensuring that hypotheses derived from NGS data could be tested quickly and accurately. This end‑to‑end workflow—spanning data processing, interpretation, and functional follow‑up—demonstrates how Genedata Selector accelerates biologic characterization and supports confident decision‑making during cell line development.
To find out more, read the publication, or watch the talk delivered by the first author of this paper, Simon Fischer (Ph.D.). In the webinar he explains in detail how his team at Boehringer Ingelheim interrogated the differences between the two clones by leveraging NGS for troubleshooting. Simon is a Senior Principal Scientist and Head of Cell Biology at Boehringer Ingelheim and presented this talk at our last biosafety Open Forum.