NFCC chemist details LC‑MS proteomics used to identify counterfeit, contaminated biotherapeutics

Dr. David Skelton, a chemist at the FDA’s National Forensic Chemistry Center (NFCC), outlined how liquid chromatography–mass spectrometry (LC‑MS) proteomic techniques are used to identify and characterize peptides and proteins in samples submitted as part of regulatory and criminal investigations. The presentation, given at an FDA Grand Rounds session (date not specified in the transcript), included step‑by‑step descriptions of sequencing by MS2 fragmentation and several case studies demonstrating how the NFCC applies those methods to suspected counterfeit or adulterated biotherapeutics.

Skelton framed proteomics as the analysis of peptides and proteins and emphasized why sequence matters: "the sequence of amino acids in the protein determine[s] the structure of the protein, and it's the structure of the protein that determines its function," he said. He explained the laboratory workflow—chromatographic separation, ionization, mass analysis and MS2 fragmentation—then described how predictable fragment ions (B and Y ions) are matched to theoretical masses to assign amino‑acid sequences.

Using a real example, Skelton showed how an unlabeled white powder submitted to the lab produced a single dominant peptide peak; fragmentation and database comparison identified the peptide as semaglutide. He explained that semaglutide contains a synthetic alpha‑aminoisobutyric acid (AIB) at position 2 and a modified lysine side chain, and that the NFCC confirmed the sequence and localized modifications by matching observed fragment ions with predicted B and Y ions and comparing results with a certified reference standard. "We are able to confirm the presence of semaglutide in this unknown powder," Skelton said.

Skelton stressed that beyond identity, the NFCC performs impurity profiling to detect synthesis byproducts (oxidation products, N‑terminal formylation, acetaldehyde adducts), inserted or missing residues, aggregation (dimers) and other anomalies that can indicate improper synthesis or incomplete purification. He said such impurities can both help trace the source of counterfeits and pose health risks because some impurities "can also potentially elicit immunogenic responses in the patients."

Shifting to larger proteins, Skelton used human serum albumin (HSA; 585 amino acids) to illustrate the bottom‑up approach: digest the protein with trypsin into predictable peptides, analyze those peptides by LC‑MS, and map peptide sequences back onto the parent protein to assess sequence coverage. He reported achieving over 99% sequence certainty for an HSA sample after comparison with a certified reference standard.

In an untargeted search example, Skelton described peptides that did not map to HSA and how the NFCC searches large protein libraries (e.g., UniProt) to find unique peptide matches. In one sample, a large fraction of peptides matched collagen (type III alpha‑1 chain): 53 peptides mapped to the collagen protein and 33 of those peptides were unique to that collagen, which Skelton said provided "very strong evidence" the sample had been contaminated or adulterated. He noted that "there is no FDA approved formulation for this particular biotherapeutic that allows collagen proteins to be present," and that many such samples are tied to Office of Criminal Investigations casework.

Skelton also described analysis of glycosylated biotherapeutics and monoclonal antibodies: locating a glycosylation site on a specific asparagine residue, identifying nine distinct glycans mapped to that single site, and comparing glycan structures and fractional abundances against an authentic sample obtained from the manufacturer to confirm identity. However, he noted the NFCC sometimes observes peptides in the suspect that do not appear in the authentic product. In one case, sequence similarity searching revealed a single‑amino‑acid substitution (valine→threonine) present in approximately 40% of the suspect material; Skelton warned that such substitutions "can have effects on the structure of the biotherapeutic" and could "elicit an immunogenic response in the patient."

Skelton closed by summarizing NFCC workload and capabilities: the center has analyzed more than 50 different biotherapeutic peptides, characterized the structures of over 20 non‑natural amino acids and synthetic modifications, and determined sequences for more than seven different active pharmaceutical ingredients for GLP‑1 agonists (five FDA‑approved and two not FDA‑approved). He said the NFCC is expanding its database of non‑natural amino acids and impurity profiles and will pivot method development based on incoming samples; the lab is also working on bacterial prototyping using LC‑MS for rapid bacterial identification. He credited NFCC analysts Robert Chapman, Rachel Dickon and Samuel Seibert for assistance with the work.

The NFCC presentation underscores how proteomic LC‑MS can both confirm authentic products and reveal adulteration or manufacturing anomalies that may have public‑health and law‑enforcement implications. Skelton said the laboratory routinely provides these analytical results and impurity profiles to special agents and investigators to support sourcing and case development.