Revolutionizing protein refolding

The production of therapeutic proteins is a cornerstone of modern medicine, enabling treatments for conditions from diabetes to cancer. However, a significant bottleneck persists in the manufacturing process: protein misfolding. When bacteria like E. coli are used to produce recombinant proteins, they often accumulate as inactive, aggregated clumps known as inclusion bodies. While this simplifies initial purification, refolding these proteins into their correct, active conformation is notoriously difficult. Traditional refolding strategies, such as dilution or dialysis, are complex, time-consuming, and generate low yields with limited scalability, driving up the cost of biologic drugs [1].

A groundbreaking study presents a novel solution to this decades-old problem using the power of Covalent Organic Frameworks (COFs). This research offers a glimpse into a future where high-yield, scalable protein refolding could become a standard, efficient industrial process.

To understand the breakthrough, one must first understand the tool. COFs are a class of highly ordered, porous crystalline materials constructed from organic building blocks linked by strong covalent bonds. Their key advantages are their precise tunability and permanent porosity. Scientists can rationally design COFs with specific pore sizes, shapes, and surface chemistries (e.g., hydrophobic, hydrophilic, hydrogen-bonding) for a target application [2]. This makes them ideal candidates for creating a tailored nano-environment to guide a protein through the complex energy landscape of folding.

A team led by senior authors Professor Yao Chen, PhD, of Nankai University, and Professor Wen Chen, PhD, of the South China University of Technology, identified the critical COF properties integral to successful protein folding. They determined that pore size, hydrophobicity, pi-pi conjugation, and hydrogen bonding are critical to regulating protein conformation [3].

Armed with this knowledge, the team developed a continuous, solid-phase refolding system. They synthesized a mesoporous COF called NKCOF-122 and packed it into a column. Instead of a slow, batch-based process, their method involves gradually pumping a solution of misfolded lysozyme (a model protein) through the column. In a single step, the protein interacts with the COF's engineered interior, refolds correctly, and is eluted. This process achieved a remarkable refolding yield of approximately 100% and the column could be used for at least 30 cycles of refolding, elution, and regeneration without significant loss of performance.

The study meticulously detailed how each COF property contributes to refolding:

• Pore Size is Paramount: The researchers found a strong correlation between COF pore size and refolding yield. NKCOF-122, with a pore size of 4.1 nm (similar to lysozyme's 4.5 nm diameter), achieved a 95% refolding rate. A larger pore (5.4 nm) dropped the yield to ~53%, while a smaller pore (2.7 nm) drastically reduced yield to 18%, as the protein could only interact with the external surface, not the optimized internal pore environment.

• Hydrophobic and Aromatic Interactions: The internal hydrophobicity of the COF and pi-pi conjugation (stacking interactions between aromatic rings in the COF and aromatic amino acids in the protein) were shown to be crucial. The researchers hypothesize that controlled hydrophobic interactions help segregate partially folded protein chains, preventing them from aggregating and guiding them toward the correct pathway.

• The Role of Hydrogen Bonding: Hydrogen bonding between the COF and the protein backbone likely helps stabilize key folding intermediates, potentially lowering the activation energy required to achieve the native state.

This "designer environment" effectively acts as a nano-chaperone, mimicking the function of natural chaperone proteins in cells that prevent misfolding and aggregation.

The system's success was not limited to lysozyme. The team demonstrated high refolding efficiency (>70%) for other enzymes like trypsin, nattokinase, and papain, suggesting the platform could be broadly applicable. The common thread was that as the hydrophobicity of the COF decreased, so too did the refolding efficiency for these proteins.

This indicates a path forward: customizing the COF microenvironment for different classes of proteins could yield optimized processes for specific therapeutics.

While exceptionally promising, the technology is still in its experimental stage. As first author Jinbiao Guo, a PhD candidate at Nankai University, stated, the next steps involve applying the platform to more complex, therapeutically relevant proteins and real-world inclusion body samples.

The future roadmap, as outlined by the researchers, includes:

1. Testing on complex proteins like monoclonal antibodies.

2. Developing a scalable column design that can integrate seamlessly with existing biopharmaceutical manufacturing pipelines.

3. Further reengineering and optimization of COF structures for specific protein targets.

If successful, this COF-directed refolding strategy could transform biopharmaceutical manufacturing, making the production of life-saving drugs more efficient, scalable, and cost-effective.

[1] Singh, A., Upadhyay, V., Upadhyay, A. K., Singh, S. M., & Panda, A. K. (2015). Protein recovery from inclusion bodies of Escherichia coli. Microbial Cell Factories, 14(1), 41. https://doi.org/10.1186/s12934-015-0222-8

[2] Diercks, C. S., & Yaghi, O. M. (2017). The atom, the molecule, and the covalent organic framework. Science, 355(6328), eaal1585. https://doi.org/10.1126/science.aal1585

[3] Guo et al. (2025) Precise modulation of protein refolding by rationally designed covalent organic frameworks Nature Communications 16:4122 https://doi.org/10.1038/s41467-025-59368-z