When CFPS Is Not the Right Expression Platform

Cell-free protein synthesis is not the right expression platform when protein function, production volume, or workflow requirements exceed what an open reaction can support. Sepia Bio’s OpenCFPS™ fits rapid screening and milligram-scale research production, but complex post-translational modifications and sustained large-scale output often favor cell-based expression.

What CFPS actually does and does not do inside the reaction tube

CFPS runs protein synthesis in vitro using cell extracts, energy substrates, amino acids, and a DNA or mRNA template. A DNA template needs transcription before translation, while mRNA is ready to serve as the translation template. No living host is required, so researchers can control reaction inputs without maintaining cell viability. Common extract sources include E. coli, wheat germ, rabbit reticulocyte lysate, and insect cells. These sources differ in their modification capabilities, yield profiles, and sensitivity to additives.

Removing the living cell also removes sustained cellular homeostasis and the intact organization of native organelle machinery. The reaction retains useful components of the cell with only some of the functions that made those components work together. That distinction explains why direct access and rapid turnaround coexist with finite energy supplies and incomplete protein processing. Understanding this architecture is therefore a prerequisite for choosing an extract, rather than a troubleshooting exercise after expression fails.

Why post-translational modification requirements make cell-based expression non-negotiable for many proteins

The first decisive boundary is whether a protein needs post-translational modifications to function. Prokaryotic CFPS lacks the eukaryotic machinery needed to reproduce many native modification profiles. It may therefore produce a relatively homogeneous protein that is less stable, less active, or unsuitable for investigating the biology of interest. When a missing modification determines activity, the structural mismatch requires more than increased expression or reagent optimization.

Glycosylation and correct disulfide bonding make the distinction concrete. Many vaccine antigens depend on specific glycosylation patterns and conformational structures to maintain immunogenicity. Producing the amino acid chain without the required folding and processing does not establish that the antigen is suitable. Those folding and processing needs are harder to meet in a cell-free reaction with larger constructs or proteins that need disulfide bond formation.

Eukaryotic CFPS offers a genuine but partial workaround. Rabbit reticulocyte lysate, HeLa, and HEK293-derived systems can provide core glycosylation or some phosphorylation, with completeness depending on the lysate and target. They are not equivalent to an intact mammalian cell with its Golgi apparatus and complete secretory pathway. If an assay requires a native PTM profile, researchers should establish that the selected system can deliver it before committing. For prokaryotic CFPS, the default assumption should be that it cannot.

How protein size and multi-domain architecture degrade CFPS performance

CFPS can be a poor fit as proteins get larger and harder to fold, even when their PTM needs are simple. Full-length antibodies and multi-subunit enzymes need help with folding and assembly that an open reaction cannot give in the same way as an intact cellular system. Expression can therefore fall as molecules get more complex, with failures that follow this pattern as well as isolated failures that are hard to predict.

The choice is less clear for membrane proteins because CFPS can help with their synthesis but may still leave problems with their folding. Reported outcomes include low yields, high yields dominated by aggregated protein, and synthesis into detergents or liposomes outside the natural folding pathway. Incorporation into nanodiscs during synthesis is also achievable. These engineered approaches have conditional value, but incorporation alone does not demonstrate function. A functional membrane protein claim depends on what the resulting material has actually been shown to do.

Sepia Bio’s OpenCFPS™ E. coli Core Kit documentation states that yields may be reduced for proteins with two or more disulfide bridges or membrane domains. For full-length monoclonal antibodies, mammalian cell expression remains the safer starting point. CFPS can still serve the earlier screening stage through simpler fragments, including scFv, VHH, and nanobody constructs.

Why batch-mode reaction kinetics and energy depletion set a hard ceiling on yield

Protein complexity is not the only constraint because even a favorable target depends on the reaction’s energy supply. Traditional batch CFPS is active for approximately one hour, depending on the extract, additives, and techniques used. Transcription and translation need large amounts of energy substrates. As high-energy phosphate chemicals and secondary energy sources deplete, a closed batch cannot continuously replenish them. The reaction therefore terminates rather than sustaining production as cell-based cultures can for days or weeks.

This limit belongs to the reaction format and should not automatically be interpreted as poor reagent quality. Continuous-exchange cell-free formats extend reaction duration by replenishing substrates through a dialysis membrane. However, the additional equipment and operation add cost and complexity, partly offsetting the simplicity that makes batch CFPS attractive.

For a small screening reaction, a short active window can be sufficient. The purpose may be to compare variants. When high yields must be kept up over time, how to replenish energy becomes a key part of designing the reaction. That shift changes both the technical task and its economics.

Where the CFPS cost argument inverts: reagent economics at production scale

CFPS can compete on cost for milligram-scale research protein but may cost more for manufacturing. At small scale, rapid turnaround and many parallel reactions make its format useful. Sepia Bio positions OpenCFPS™ E. coli Core Kit pricing to support high-throughput screening rather than let per-reaction reagent cost restrict the number of variants tested. That is a credible economic role, but it does not establish that the same advantage extends to production volumes.

As reaction volume grows, energy-regeneration components and purified enzymes account for much of the cost. Cell-based production has a different cost structure and benefits from decades of infrastructure investment in stable expression, particularly CHO and HEK293. Mammalian cell processes already support commercial recombinant protein production and monoclonal antibody yields in the range of grams per liter. The comparison therefore changes when the objective moves from exploring candidates to making substantial quantities of a selected protein.

The crossover is not a universal price or volume. Researchers should work out cost per milligram from the yield they get, the number of reactions they run, any need for continuous exchange, and downstream purification. For a difficult protein that fails in cell-based expression, CFPS may still be the necessary choice. That is a technical argument for access to material, with any general cost advantage for CFPS at scale still unproven.

The scalability ceiling: why CFPS yield profiles and bioreactor physics limit production volume

Larger CFPS reactions must overcome physical constraints as well as reagent costs. Most systems are optimized for small reaction volumes, and enlarging the vessel does not guarantee that yield will scale with it. Heat dissipation, oxygen management, and component mixing become harder to control at bioreactor volumes. These changes can reduce protein yield and destabilize the system. Difficulties producing consistent lysate at scale add another barrier, particularly for eukaryotic CFPS.

There have nevertheless been meaningful advances. The tobacco BY-2 lysate system commercialized as ALiCE® by LenioBio demonstrated linear scalability from 0.1 to 100 mL reaction volumes. That result establishes progress across a defined range, with equivalence to manufacturing-scale mammalian bioreactors still unproven. It also addresses a different point on the scale curve from a research workflow centered on screening and milligram preparation.

OpenCFPS™ is explicitly designed for screening and scale-up to milligram quantities. CFPS can provide material for assays, characterization, and work on difficult targets, including milligram-scale preparations for studies that support approval to begin clinical trials. When gram-scale amounts or commercial production are needed, established cell-based bioreactor processes are still the practical choice.

Reproducibility and lot variability as a practical barrier, not just a theoretical concern

Even with a suitable target and reaction volume, results can vary from one extract lot to another. Crude extracts contain undefined component ratios, background proteins, and potentially contaminating nucleases and proteases. Activity can vary between batches because of how they are prepared, how much their components break down, and the reaction conditions. Such variability can arise from the complexity of the reagent system even with skilled laboratory work. Without information on each lot, it is hard to tell whether a problem comes from the protein or the extract.

Template degradation is one concrete example. Endogenous endo- and exonucleases can degrade DNA and reduce yield, with linear templates particularly vulnerable. Defined, reconstituted approaches such as the PURE system reduce this uncertainty by eliminating proteases and nucleases from the formulation and allowing more precise control. However, they raise reagent costs and do not solve the PTM requirements of complex eukaryotic proteins.

Reagent transparency addresses a different part of the problem. In its OpenCFPS™ documentation, Sepia Bio shares genome and buffer composition, QC data for each lot, and Certificates of Analysis that anyone can read. These records give researchers information for troubleshooting rather than requiring them to treat the extract as a black box. They do not abolish biological variability or establish regulatory equivalence. Product quality must still be consistent, especially for therapeutic uses, where cell-based systems have been used for decades.

Where CFPS genuinely wins and cell-based expression is the slower, worse choice

For rapid exploration, CFPS is often the better tool. Reactions run in plates allow high-throughput screening of variants in 96-, 384-, or 1536-well formats with robotic liquid handling. Systems that use extracts can express dozens of open reading frames in a single day. Starting from a PCR product can produce a protein signal in hours without cloning, transformation, colony selection, or culture optimization. When the task is to choose the next construct, those avoided steps directly shorten the design-build-test-learn cycle.

Antimicrobial peptides, proteases, and membrane-disrupting proteins can kill or arrest cells before enough product accumulates. With CFPS, that host no longer has to be kept alive, though the target must still work with the reaction. The open environment also supports direct addition of modified aminoacyl-tRNAs and unnatural amino acids for site-specific labeling. Labels can include fluorophores, biotin, radiolabels, or heavy atoms, without the elaborate genetic engineering required for comparable work in intact cells.

For these research workflows, Sepia Bio offers OpenCFPS™ compatibility with linear and plasmid DNA, low DNA concentrations, plate formats, and standard incubators. Its screening and milligram-production modes let researchers move from testing variants to making the research proteins they have chosen. That fit does not require CFPS to remain the platform for every subsequent stage. Use it for exploratory, combinatorial, or toxicity-constrained work, then reassess expression when the candidate needs PTM-complete, gram-scale, or regulatory-precedented material.

A practical decision framework for choosing between CFPS and cell-based expression

The choice follows four linked requirements, namely PTM profile, folding complexity, production scale, and throughput. If downstream interpretation depends on complex glycosylation or other native processing, default to eukaryotic cell-based expression unless the required profile has been demonstrated in the proposed CFPS system. Eukaryotic CFPS can be an intermediate option for some targets, but its modifications need validation. Detectable expression is not enough when the experiment depends on a specific functional state.

If the target is small, single-domain, and cytosolic, CFPS is a strong candidate. As the target moves toward multi-domain scaffolds, full-length antibodies, or complex membrane topology, optimized mammalian or insect-cell systems become better starting points. Milligram-scale assays and characterization fit CFPS, while gram-scale development and manufacturing favor cell-based bioreactors.

If the immediate task is to screen a large variant library, plate-based CFPS can be faster and more economical than expressing every candidate in cells. Once a candidate is chosen, the next choice should depend on the processing it needs and how it will be produced. A scientist who screens variants with CFPS and moves the manufacturing candidate into CHO expression has not abandoned the platform. They have used it correctly.