Faster gene-to-GMP with integrated E. coli GMP manufacturing – saves up to 3 months

Webinar Recap: From gene to GMP: making better E. coli production decisions earlier

Webinar Date: September 30, 2026 | Organised by: enGenes Biotech and NorthX Biologics

Speakers:

Webinar Synopsis:

Good science still fails on manufacturability.

Many programs lose months not because the science fails, but in the handovers between expression development, process development, and GMP manufacturing. In this webinar, Ola Tuvesson and Juergen Mairhofer explore how connecting these steps earlier can create a faster, more scalable route from gene to GMP.

To learn more about how a fully integrated E. coli development and manufacturing pathway can help reduce bottlenecks and create a more efficient route from gene to GMP, watch the recording below

Key takeaways from the gene to GMP webinar

• 40% of delayed INDs trace back to CMC issues

• 74% of FDA complete response letters cite manufacturing or quality deficiencies

• 3+ months saved by removing the handover between strain developer and CDMO

• Handovers are where time is lost. Moving from an early-stage developer to a CDMO typically takes 3 to 6 months, and the sponsor carries the risk of information lost in translation.
• Shake flask data can mislead. In one transferred process, a protein that was soluble in shake flasks became almost entirely insoluble inclusion bodies in a stirred-tank bioreactor.
• Test several host and vector combinations. Following a single combination can hide the best option and lead to the wrong decision.
• Titer is not the only metric. Downstream steps, refolding yield, cost at scale, and process robustness all affect which route is best.
• Set up analytics early. Methods usually sit on the critical path. Being able to confirm biological activity in process samples keeps optimization pointed in the right direction.
• Keep documentation and cell bank history complete. Regulators expect a clear, unbroken history, and a results report without the “how” (raw materials, water quality, media) is hard to reproduce.

What the session covered:

Why gene to GMP manufacturing programs stall between early development and the clinic

Typical scale-up traps include batch centrifugation, sonication, ultracentrifugation, density gradients, and size exclusion chromatography (which loads only about 2% of column volume). A process built for the bench often cannot be run at manufacturing scale. Media also have to move from complex to defined, with compliant raw materials available in the quantities needed.

The enGenes expression technology

The host is an E. coli B strain close to BL21. A phage-derived RNA polymerase inhibitor (GP2) is integrated into the genome. When L-arabinose and IPTG are added together, the inhibitor binds the host RNA polymerase, host mRNA production stops, and cell division halts. At the same time, T7 RNA polymerase drives very high transcription of the gene of interest, so the non-dividing cells fill up with recombinant protein.

• A clear separation between a fed-batch growth phase and a production phase removes problems such as plasmid loss.
• No antibiotics; mineral salt media with glucose as the carbon source.
• Soluble expression, optional secretion to the extracellular space for simpler downstream processing, and tunable expression with low basal levels.
• Options for genetic code expansion, antibiotic-free plasmid maintenance, and continuous processes that have run for up to 40 days.
• One host for many modalities, including antibody fragments, nanobodies, enzymes, and vaccine antigens.
• Peer-reviewed secreted titers of up to 16 g/L, and on average up to five-fold higher yields than reference E. coli.

The 4 to 8-week decision workflow

The client supplies the DNA or amino acid sequence, product specifications, and any historical data. Then:

• High-throughput screening in 48-well plates run like small fed-batch bioreactors (enzymatic glucose release from dextrin, online biomass monitoring). Variables include codon optimization, signal peptides, solubility tags, chaperones, host strain versions, inducer level, and temperature. Samples are split into soluble, insoluble, and extracellular fractions.
• Benchtop fermentation of the best clones in four 1 L bioreactors in parallel, with the same probes as the stainless steel bioreactor at NorthX. Online data every 5 seconds, 20 samples in total, plus host DNA in the supernatant and biomass centrifugability.

Case study 1: a disulfide-rich receptor ectodomain

The question was whether a hard-to-fold protein could be made in soluble form rather than as inclusion bodies. One codon-optimized sequence plus seven periplasmic leaders and solubility tags gave eight constructs, tested in two host strain versions (16 host-vector combinations) under 40 conditions and 8 controls, with a decision point in two weeks. Six plasmids did not grow in the first host version but did in the second, pointing to toxicity from basal expression.

Inclusion bodies look better on titer alone, but refolding at low protein concentration can turn 1 L of fermentation into 30 L or even 600 L, and adds four unit operations. By mass, refolding yield must exceed about 19% to beat the soluble route, and close to 40% once cost is included. The client is now measuring the achievable refolding yield, which shows how good data raises the next question to answer.

Case study 2: a high-dose protein for a chronic disease

The product needs a very low cost of goods because of the dose and patient population, with manufacture planned at 10 to 100 cubic meter scale, where cell disruption is expensive. The team ran a design of experiments (plates first, then 4 x 1 L bioreactors) to find what drives secretion.

• Client reference with BL21: 9.4 g/L.
• Secretion with the enGenes process: 10.7 g/L, without a cell disruption step.
• Cytoplasmic route with the enGenes process: 16.8 g/L, nearly double the client’s titer.

The commercial model

The enGenes technology is included in the NorthX collaboration with no license fees or royalties, providing a transparent cost structure and full portability. Analytical method development at NorthX can start early, with material from the early-stage work.

Frequently asked questions

What is decision-grade data?

It is data that lets you choose an expression route with confidence before you commit to a process. The most important decision is the route itself, because a program lives with it through the whole life cycle. The highest soluble titer is not always the best choice: quality, downstream steps, and cost at scale all count.

Can I change expression system after early development?

It depends on the data you already have. Once clinical data exists, moving to a different system becomes hard. The safer approach is to start early with a high-performing system and generate data under conditions close to real manufacturing, such as bioreactors rather than shake flasks. Spending earlier tends to save money later.

What yields can I expect?

The target is grams per liter. The outcome depends on the intrinsic properties of the protein, so no platform solves every case. Published results include secreted titers of up to 16 g/L, and in the case studies above, 10.7 g/L (secreted) and 16.8 g/L (cytoplasmic) for a client protein that reached 9.4 g/L with a reference BL21 strain.

Is a higher titer always better than a robust process?

No. Titers need to be high enough for a product that is sustainable in cost, quality, and supply, but predictability matters just as much. A robust process that reliably delivers enough is often worth more than a spectacular but fragile one. Growth-decoupled expression aims to deliver both

Are there license fees or royalties?

No. The enGenes technology is offered inside the NorthX collaboration with no license fees or royalties, which gives a transparent cost structure and full portability. Get in touch to discuss what is included for your project.

When should I approach a CDMO?

As early as possible. You do not need to be ready for the clinic, and early contact helps guide technology and process choices. Once clinical trials are on the horizon, you should be in contact with a CDMO that can manufacture for you. How long does it take to get from a new strain to the first GMP batch?

Plan for at least a year from gene to GMP drug substance. In the integrated model, decision-grade data takes 4 to 8 weeks, drug substance about 12 months, and drug product 15 to 18 months.

Why set up analytical methods early?

Analytics usually sit on the critical path. Titer only matters if you can show the product is biologically active, so being able to confirm activity in process samples early keeps optimization pointed in the right direction. Methods also need qualification and a reference standard to show comparability between sending and receiving sites.

How are screening and scale-up results analyzed?

Results are first assessed by SDS-PAGE across the soluble, insoluble, and secreted fractions. HPLC and other methods can be added where available, and early purification is possible at extra cost. The emphasis is on a standardized workflow rather than a research project for each molecule, with multivariate data analysis used where it adds value.

What does the program cost?

Cost depends on the molecule and the scope of work, so there is no standard price. Discuss your specific case with us to get a a clearer cost estimate.