🧬 biology

Escherichia coli–based Nanobody Production: High-Cell-Density Process Development and Benchmarking of Strain-Plasmid Combinations

This study establishes a robust high-cell-density fed-batch process in *E. coli* that, through the optimization of induction timing, temperature, and host-strain combinations, achieves yields exceeding 2.5 g/L of functional anti-CD45 nanobodies, with periplasmic secretion proving superior to cytoplasmic production in terms of yield, thermal stability, and folding uniformity.

Original authors: Jan-Angelus Meyer, Marie Beneke, Sadija Altmüller, Alexander Reeb, Jannis Künzl, Nathalie Opilo, Theo Weise, Christos Gouloudis, Rebekka Biedendieck, Rainer Krull

Published 2026-07-08

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Jan-Angelus Meyer, Marie Beneke, Sadija Altmüller, Alexander Reeb, Jannis Künzl, Nathalie Opilo, Theo Weise, Christos Gouloudis, Rebekka Biedendieck, Rainer Krull

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

The Big Picture: Building Tiny Antibody Robots in a Factory

Imagine Nanobodies as tiny, super-specialized "robots" designed to grab onto specific targets in the body (in this case, a marker called CD45 found on immune cells). These robots are much smaller and simpler than full-sized antibodies, making them easier to mass-produce.

The scientists in this study wanted to build a highly efficient factory to make these nanobody robots using E. coli bacteria. Think of the bacteria as the workers, and the factory as a large, controlled bioreactor (a giant high-tech tank).

The goal was simple: Make as many high-quality robots as possible, as fast as possible, without breaking them.

Part 1: Tuning the Factory Settings (Process Optimization)

Before building the robots, the team had to figure out the perfect settings for their factory. They used a method called "Design of Experiments" (DoE), which is like a chef tasting a soup at different stages to find the perfect balance of salt, heat, and cooking time.

They tested three main "knobs" on their machine:

When to start the assembly line (Induction Timing): Should they start making robots when the bacteria are young and small, or when the tank is packed with bacteria?
- The Discovery: They found that waiting until the tank was very crowded (a high density of bacteria) before starting production was best. It's like waiting until a construction crew is fully assembled and ready before handing out the blueprints.
How hot to keep the factory (Temperature): Should it be hot and fast, or cool and careful?
- The Discovery: A cooler temperature (around 21–22.5°C) worked best. Think of it like baking a delicate cake; if the oven is too hot, the cake burns or collapses. A cooler environment helped the nanobodies fold into their correct shape without getting "jumbled."
How much "starter" to add (Inducer Concentration): They tested adding different amounts of a chemical trigger (IPTG).
- The Discovery: Surprisingly, it didn't matter much how much trigger they added, as long as it was enough to start the process. This is great news because the trigger is expensive; they can save money by using less of it.

The Result: By setting the factory to start late (when the tank is full) and keep it cool, they achieved a massive output: over 2.5 grams of pure nanobodies per liter of liquid. That's a lot of tiny robots!

Part 2: Choosing the Right Workers and Blueprints (Strain & Plasmid Benchmarking)

Next, the team asked: "Does it matter which bacteria we use or how we give them the instructions?"

They tested five different combinations of bacterial strains (the workers) and plasmids (the instruction manuals). They looked at two main strategies for where the robots should be built:

Strategy A: Inside the Cell (Cytoplasmic). The robot is built inside the main body of the bacteria.
Strategy B: Outside the Cell (Periplasmic). The robot is built and pushed into the "vestibule" or outer room of the bacteria.

The Findings:

The "Best" Worker: One specific strain called NEB Express® combined with a specific instruction manual (using a signal peptide called PelB) was the champion. It produced the most nanobodies per hour.
The "Runner-Up": Another strain called SHuffle® T7 Express was also excellent, especially for building robots inside the cell.
The Losers: Some common bacteria strains (like standard BL21) were terrible at this specific job, producing almost nothing.

Key Lesson: It's not just about the worker; it's about the teamwork between the worker and the instruction manual. A great worker with the wrong manual fails, and a great manual with the wrong worker fails.

Part 3: Quality Control (Are the Robots Good?)

Making a lot of robots is useless if they are broken or don't work. The team had to check the quality of their products.

1. The "Folding" Test (Thermal Stability):
Imagine the nanobody is a piece of origami. If it folds correctly, it's strong. If it's crumpled, it falls apart easily when heated.

The Periplasmic Robots (Strategy B): These were like perfectly folded origami. When heated, they stayed strong and uniform until a very high temperature (around 73–74°C). They were consistent and stable.
The Cytoplasmic Robots (Strategy A): These were a bit more "messy." They showed two different melting points, meaning some parts of the population were folded well, while others were slightly crumpled. They started falling apart at lower temperatures (around 51–55°C).
Why? The "vestibule" (periplasm) is a natural environment for folding these specific robots, while the inside of the cell is a chaotic, crowded place that makes folding harder, even for the "super-strains" designed to help.

2. The "Grip" Test (Functionality):
Finally, they tested if the robots could actually grab their target. They used a flow cytometer (a machine that counts cells) to see if the nanobodies stuck to CD45-positive cells.

The Result: Every single robot that was made, regardless of whether it was built inside or outside the cell, worked perfectly. They all grabbed their targets tightly.

The Final Conclusion

The paper concludes that to make these nanobody robots efficiently in a factory setting, you need to:

Wait until the bacteria are very crowded before starting production.
Keep it cool to ensure they fold correctly.
Pick the right team: The combination of the NEB Express bacteria with the PelB instruction manual (building outside the cell) was the absolute winner for speed and quality.

While building inside the cell (using SHuffle strains) also worked well and produced functional robots, building them in the "vestibule" (periplasm) resulted in a more uniform, stronger, and more stable product.

In short: The scientists built a high-speed, high-quality factory line for nanobodies, proving that with the right timing, temperature, and bacterial team, you can produce massive amounts of functional medical tools.

Technical Summary: Escherichia coli–based Nanobody Production: High-Cell-Density Process Development and Benchmarking of Strain-Plasmid Combinations

Problem Statement
Nanobodies (NBs), or VHH single-domain antibody fragments, possess significant diagnostic and therapeutic potential due to their compact size, stability, and solubility. While microbial production in Escherichia coli is attractive for scalability, achieving high yields of correctly folded, functional NBs under controlled high-cell-density cultivation (HCDC) conditions remains insufficiently characterized. Key challenges include the formation of intramolecular disulfide bonds (often hindered by the reducing cytoplasm), the metabolic burden of expression, and the complex interplay between process parameters (induction timing, temperature), host strain background, and subcellular targeting (cytoplasmic vs. periplasmic). Existing datasets linking HCDC process optimization with host-strain selection and product quality (folding uniformity and thermal stability) are limited.

Methodology
The study employed a systematic, two-stage approach to develop a robust fed-batch process for an anti-human CD45 nanobody in E. coli:

Process Optimization via Design of Experiments (DoE):
- Platform: 2 L stirred-tank bioreactors operated in fed-batch mode.
- Strategy: A fractional factorial screening design followed by a response-surface methodology (RSM) using an "onion D-optimal design."
- Variables: Induction optical density (OD $_{600nm}$ ), post-induction temperature, and IPTG concentration.
- Strain: Initial screening utilized E. coli SHuffle T7 Express with a cytoplasmic expression construct (pT7-Cyto).
- Goal: Identify the optimal operating window for maximizing post-purification nanobody concentration.
Benchmarking of Strain-Plasmid Combinations:
- Selection: Five distinct host–plasmid systems were selected based on preliminary shake-flask screening. These included cytoplasmic production (using E. coli SHuffle T7 Express and SHuffle Express) and periplasmic secretion (using E. coli NEB Express® with OmpA or PelB signal peptides). E. coli BL21(DE3) served as a reference.
- Conditions: All cultivations were performed under strictly standardized HCDC conditions derived from the DoE optimization (induction at OD $_{600nm}$ ~150, temperature shift to 22.5 °C, 20 h production phase).
- Metrics: Volumetric productivity, yield on biomass (Y $_{P/X}$ ), and yield on substrate (Y $_{P/S}$ ).
Product Quality Characterization:
- Purification: Strep-tag affinity chromatography.
- Identity & Processing: SDS-PAGE and Western blotting to assess molecular weight and C-terminal integrity.
- Thermal Stability: Nano differential scanning fluorimetry (nanoDSF) to evaluate unfolding transitions and folding homogeneity.
- Functionality: Flow cytometry using CD45-positive Jurkat cells to confirm antigen-binding activity.

Key Results

Process Optimization: The DoE identified the induction OD $_{600nm}$ and post-induction temperature as the primary drivers of yield, while IPTG concentration (within the tested range of 80–650 µM) had no significant effect.
- Optimal Window: Late induction at OD $_{600nm}$ 125–150 combined with a production temperature of 21–22.5 °C defined a robust operating window.
- Yield: Under these conditions, purified nanobody concentrations exceeded 2.5 g L⁻¹.
Strain-Plasmid Benchmarking:
- Top Performers: E. coli NEB Express® with the PelB signal peptide (periplasmic) achieved the highest volumetric productivity (0.122 g L⁻¹ h⁻¹) and a final titer of ~2.43 g L⁻¹. E. coli SHuffle T7 Express (cytoplasmic) followed closely with ~2.19 g L⁻¹.
- Host Dependency: Performance was heavily dependent on the specific host–construct combination. For instance, E. coli BL21(DE3) with the same cytoplasmic construct yielded negligible amounts (<0.1 g L⁻¹), highlighting the importance of host background over expression architecture alone.
- Signal Peptide Specificity: Within the NEB Express® host, the PelB signal peptide significantly outperformed OmpA for periplasmic secretion.
Product Quality and Folding:
- Thermal Stability: NanoDSF revealed a distinct difference between production routes. Periplasmic preparations (NEB Express®) exhibited a single, sharp unfolding transition at high temperatures (73.27–74.05 °C), indicating a homogeneous, thermally stable population. In contrast, cytoplasmic preparations (SHuffle strains) showed biphasic unfolding profiles with lower first transitions (51.62–55.31 °C) and second transitions (68.30–71.53 °C), suggesting a more heterogeneous population with varying folding states.
- Functionality: Despite differences in thermal stability, flow cytometry confirmed that all tested preparations (cytoplasmic and periplasmic) retained functional binding to CD45-positive Jurkat cells.

Significance and Claims
The authors claim that this study establishes a controlled, scalable platform for nanobody production in E. coli that achieves gram-per-liter titers. The primary contribution is the demonstration that high-level production is not solely a function of the host strain but is governed by the integrated optimization of induction timing, temperature, host background, and subcellular targeting.

Crucially, the paper argues that product quality is route-dependent. While cytoplasmic production in engineered strains (SHuffle) can yield high titers, periplasmic secretion (specifically using NEB Express® with PelB) produces nanobodies with superior thermal uniformity and stability, likely due to the oxidizing environment facilitating correct disulfide bond formation. The study concludes that for applications requiring high conformational homogeneity, periplasmic secretion is the preferable route, whereas cytoplasmic production remains a viable, high-yield alternative if folding heterogeneity is managed. The work provides a data-driven framework for selecting production strategies based on both yield and product quality metrics.

The Big Picture: Building Tiny Antibody Robots in a Factory

Part 1: Tuning the Factory Settings (Process Optimization)

Part 2: Choosing the Right Workers and Blueprints (Strain & Plasmid Benchmarking)

Part 3: Quality Control (Are the Robots Good?)

The Final Conclusion

More like this