Most recombinant proteins expressed in E. coli are in inclusion bodies. Are there any mild refolding protocols?

If you don't want to run a column, you can first do small-scale screening plates. Take a small amount of dissolved protein and test conditions such as pH 7.5/8.5/9.5, 0.3/0.5 M arginine, 0.1% PEG 3350, and 0.5 M proline or sucrose separately in a 24-well plate. Use 0.5 mL per well, leave at 4°C overnight, and check for precipitation visually and by SDS-PAGE. Commercial refolding screening kits can save time on buffer preparation, but the principle is similar. After screening out the condition with the least precipitation, scale it up; this saves more sample than repeatedly testing a single system directly.

I understand that you're avoiding the column because you're worried about loss, but refolding loss mainly comes from the refolding process, not the column. You can consider the batch method: bind the solubilized protein to Ni-NTA under denaturing conditions, then gradually exchange into refolding buffer in the tube. Once the protein is immobilized, it can reduce mutual aggregation, and finally elute with buffer containing 0.5 M imidazole. This way, the volume is small, and after elution you can directly measure activity. If you insist on not using a column, add Ni-NTA to the denaturing solution for batch binding, then perform gradient dilution refolding in the tube, centrifuge to collect the resin, and then elute. The loss is usually controllable.

This yield issue may not be solely a refolding problem; you should also keep a backup plan on the expression side. You can first try low-temperature induction on a small scale: 18–25°C, 0.1–0.5 mM IPTG overnight; or switch to strains such as Rosetta, Origami B, or Shuffle, which favor disulfide bond formation or rare codons. If the cytokine has disulfide bonds, the reducing environment of the BL21 cytoplasm is inherently unfavorable. Once soluble expression is obtained, large-volume refolding is no longer necessary. If you still want to salvage the current batch of inclusion bodies, choose 1–2 of the mild conditions above and test them, and do not use up the entire sample at once.

Don't rush into complex refolding yet—washing the inclusion bodies thoroughly is key. I typically wash twice with 50 mM Tris pH 8.0, 2 M urea, 0.5–1% Triton X-100, and 1 M NaCl, then wash once more with a Triton-free buffer. For solubilization, use 6 M guanidine hydrochloride plus 20 mM DTT, pH 8.0, stir at 4°C for 1–2 h, then centrifuge and collect the supernatant. Although guanidine hydrochloride smells worse than urea, it dissolves more completely, and when followed by rapid dilution or gradient dialysis, it tends to reduce precipitation more effectively than direct dialysis from 8 M urea.

A gentle approach can be attempted using rapid dilution or pulse refolding: slowly drip the dissolved protein into 20–50 mL of pre-chilled refolding buffer, controlling the final protein concentration at 20–50 μg/mL. The refolding buffer may consist of 50 mM Tris pH 8.5, 0.5 M L-arginine, 5% glycerol, 1 mM GSH/0.2 mM GSSG, and 1 mM EDTA. Stir slowly at 4°C overnight; do not stir vigorously. After centrifugation, first measure the soluble target protein before concentrating. If there is a lot of precipitate, continue reducing the final concentration to 10 μg/mL and slow the drip rate.