The Assumption That Derails Programs
The promise of mRNA therapeutics is elegant: deliver genetic instructions, let cells produce therapeutic proteins, and bypass traditional drug manufacturing. But there’s a critical gap between mRNA design and therapeutic outcome. Your construct may be producing proteins you never intended and your standard assays might miss them entirely.
Most mRNA therapeutic developers follow a familiar workflow. Design an optimized sequence encoding your target protein. Package it in lipid nanoparticles. Dose your model system. Run an ELISA or Western blot to confirm protein production. Check the box and move forward.
The problem is that these targeted immunoassays only detect what you’re looking for. If your mRNA generates splice variants, truncated forms, or unexpected modifications, you won’t see them. The antibody recognizes its epitope, gives you a signal, and one assumes everything is working as designed.
This assumption has consequences. Consider a recent case where an mRNA construct designed to produce a full-length enzyme consistently underperformed in functional assays despite strong ELISA signals. The company spent months troubleshooting formulation and dosing before unbiased proteomics revealed the issue: their mRNA was producing both the intended enzyme and a truncated variant lacking the catalytic domain. The ELISA antibody, raised against an N-terminal epitope, couldn’t distinguish between functional and non-functional forms.
Why Unintended Proteins Happen More Than You Think
mRNA translation is remarkably complex, and optimization strategies can introduce unexpected outcomes. Codon optimization alters RNA secondary structure, potentially exposing cryptic splice sites or alternative start codons. Modified nucleotides like pseudouridine improve stability and reduce immunogenicity, but they can also affect ribosome processivity and frameshifting rates.
UTR sequences added to enhance expression may contain regulatory elements that create alternative translation initiation sites. Even poly(A) tail length influences ribosome loading and can affect which protein isoforms dominate. The cell’s translational machinery doesn’t always read your synthetic mRNA the way you designed it on paper.
Post-translational modifications add another layer of uncertainty. Is your therapeutic protein properly folded? Are critical disulfide bonds formed? If the protein requires glycosylation or phosphorylation for activity, are these modifications occurring correctly? Antibody-based assays typically can’t distinguish between modified and unmodified forms unless you have modification-specific antibodies, which often don’t exist for novel therapeutic proteins.
The Discovery Proteomics Solution
This is where unbiased discovery proteomics becomes essential. Unlike immunoassays that only detect predetermined targets, liquid chromatography-mass spectrometry (LC-MS) based proteomics provides a comprehensive, unbiased view of every protein in your sample.
When you analyze mRNA-treated cells or tissues with discovery proteomics, you’re asking a fundamentally different question. Instead of “Is my target protein present?” you’re asking “What proteins are actually being produced?” This is a distinction that matters enormously.
Discovery proteomics identifies protein sequences based on peptide mass fingerprints and tandem MS fragmentation patterns. This approach detects truncated variants, splice isoforms, and sequence modifications without requiring specific antibodies. You can distinguish between your intended full-length product and shorter forms that may lack functional domains. You can identify unexpected fusion proteins or readthrough products.
Critically, discovery proteomics also reveals post-translational modifications. Phosphorylation sites, glycosylation patterns, acetylation, methylation, these modifications directly affect protein function but are invisible to standard immunoassays. For enzyme replacement therapies, confirming that your mRNA-produced enzyme carries the correct activating phosphorylation can be the difference between a functional therapeutic and an expensive failure.
When to Use Discovery vs. Targeted Approaches
The optimal strategy uses both approaches at different stages. During mRNA construct optimization, start with discovery proteomics. Panome Bio’s discovery proteomics platform uses label-free bottom-up untargeted proteomics to provide an unbiased assessment of the proteome across species and sample types. This approach identifies exactly what proteins your mRNA constructs produce – including variants and modifications you weren’t expecting.
Once you’ve confirmed your construct produces the intended protein without significant off-target products, you can develop targeted assays for routine QC and dose-response studies. ELISA becomes appropriate when you know what you’re measuring. But skip the discovery phase, and you risk optimizing a construct that fundamentally doesn’t produce what you think it does.
For post-translational modification analysis, phosphoproteomics offers unparalleled insight. Rather than requiring modification-specific antibodies for each site of interest, phosphoproteomics can map thousands of phosphorylation sites across your proteome in a single experiment. This matters particularly for enzyme replacement therapies where phosphorylation status determines catalytic activity.
The Cost of Not Looking
The expense of comprehensive proteomics analysis seems significant until you compare it to the alternatives. Running a six-month animal study with a construct producing unintended proteins wastes far more resources than upfront analytical characterization. Discovering in Phase 1 that your protein has unexpected modifications explains disappointing efficacy, but discovering this during construct design prevents the problem entirely.
Companies have abandoned promising mRNA programs due to inexplicable failures in functional assays, only to later discover the issue was protein quality, not the underlying therapeutic concept. Unbiased proteomics early in development prevents these costly false negatives.
Making Discovery Proteomics Routine
The technology for comprehensive protein characterization now operates at sufficient throughput and cost-effectiveness to be practical for mRNA therapeutic development. Modern LC-MS platforms combined with sophisticated bioinformatics can analyze dozens of samples per week, making iterative construct optimization feasible.
The key is integrating discovery proteomics into your workflow at the right stage. Use it when designing and selecting lead constructs. Apply it when modifying sequences for codon optimization or regulatory element changes. Deploy it when formulation changes might affect protein expression patterns. Once your construct is locked and you’re scaling toward clinical development, transition to validated targeted assays for consistency and regulatory purposes. But that transition should follow comprehensive characterization, not replace it.
Conclusion
The elegance of mRNA therapeutics – directing cells to produce therapeutic proteins – only delivers value if the produced proteins match your design. ELISA tells you whether your epitope is present. Discovery proteomics tells you what you actually made. For novel therapeutic modalities where each construct represents a unique molecular entity, the question isn’t whether you can afford comprehensive characterization but whether you can afford not to know what your mRNA is really producing.