Skip to main content
SearchLoginLogin or Signup

Protein Expression Autoinduction in a Cold-Shock Expression System in Escherichia coli

Keywords: autoinduction, cold-shock expression, Escherichia coli, recombinant proteins

Published onApr 29, 2024
Protein Expression Autoinduction in a Cold-Shock Expression System in Escherichia coli


The cold-shock expression system in Escherichia coli was developed on a manual induction approach using optical density at 600 nm (OD600) measurements and isopropyl β-D-1-thiogalactopyranoside (IPTG) addition. In this study, we show that cold-shock expression performs equally well using an autoinduction approach wherein OD600 measurements and IPTG addition may be eliminated. We further demonstrate that cold-shock expression with autoinduction can better facilitate high-throughput experiments.

ADDRESS CORRESPONDENCE TO: Naoto Isono, Faculty of Bioresources, Mie University, Tsu 514-8507, Japan (Phone: +81-59-231-9613; Email: [email protected])

Financial Support: This work was supported in part by Japan Society for the Promotion of Science KAKENHI Grant Numbers 15K07386 and 19K05826.

Conflict of Interest: The authors declare no conflicts of interest.

Keywords: autoinduction, cold-shock expression, Escherichia coli, recombinant proteins


Recombinant protein expression is crucial for protein and enzyme studies and for the efficient production of various useful materials.[1] A cold-shock expression system employing pCold vectors is a widely used method for producing proteins in Escherichia coli.[2] This system utilizes the promoter of the cspA (cold-shock protein A) gene from E. coli, inducing protein production through temperature reduction.[2],[3] This method offers benefits, including the ability to obtain soluble proteins that may otherwise remain insoluble or unexpressed in alternative systems.

In pCold vectors, a lac operator sequence is placed downstream of the cspA promoter to prevent leaky target protein expression.[2],[3] Hence, in addition to lowering the culture temperature, isopropyl β-D-1-thiogalactopyranoside (IPTG) must be added to obtain protein expression. The manufacturer’s standard protocol[3] involves incubation at 37 °C until OD600 reaches 0.4 to 0.5 followed by rapid cooling at 15 °C for 30 min. IPTG is then added to 0.1 to 1.0 mM, and the culture is incubated overnight at 15 °C. However, when conducting high-throughput experiments to simultaneously evaluate the functions of various expressed proteins from small-scale cultures, measuring OD600, rapid cooling, and adding IPTG become cumbersome.

Other expression systems, including the pET system, occasionally utilize autoinduction (AI) media, eliminating the need for IPTG addition.[1] AI media contains balanced glucose and lactose. E. coli preferentially initially consume the glucose. Lactose uptake occurs once glucose is depleted, abrogating suppression of expression by the lac operator. IPTG addition therefore becomes unnecessary for protein expression in AI medium. In this study, we examined the feasibility of using AI medium for cold-shock expression.

Several bacterial glycosyl hydrolase (GH) family enzymes were expressed as models using pCold II–derived plasmids (see Methods for details). HhGH161 and HhGH30_1 genes were obtained from Halobacillus halophilus. PbGH30_3 and PbGH94 were obtained from Paenibacillus borealis. E296G is an HhGH30_1 mutant. We initially expressed HhGH161 in 15-mL tubes containing 3 mL of AI medium using 2 shakers set to different temperatures. To investigate the effect of timing on temperature changes, transformants were cultured at 37 °C for 2 (0.17), 3 (0.50), 4 (0.85), and 5 h (1.03) followed by 22, 21, 20, and 19 h at 15 °C, respectively (numbers in parentheses indicate OD600 at temperature changes). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis confirmed the expression of HhGH161 in the soluble form under all conditions, with excellent expression, especially in recombinants cooled after the incubation at 37 °C for 3 to 4 h (Figure 1A). These findings suggest that, while cell density affects the level of protein expression, measuring the OD600 of every sample may not be necessary because protein expression can still be induced across a broad range of OD600 values at the initiation of cooling. We subsequently performed experiments using 4 additional proteins. Transformants were cultured in 15-mL tubes for 4 h at 37 °C and then for 20 h at 15 °C, resulting in the expression of all proteins in soluble form (Figure 1B). Finally, transformants were cultured in 96-deep-well plates (1.5-mL AI medium per well) to assess applicability for high-throughput experiments. For this experiment, a single incubator was used, programmed to automatically cool to 15 °C after 2.5 h of incubation at 37 °C. Cooling from 37 °C to 15 °C took approximately 30 min. All proteins were expressed in soluble form (Figure 1C).

Figure 1

SDS-PAGE analysis of soluble proteins from transformants cultured in AI medium. (A) Expression of HhGH161 using 15-mL tubes. Transformants were cultured for 2 to 5 h at 37 °C and then at 15 °C for a total of 24 h. E. coli harboring empty pCold II was cultured as a negative control. (B) Expression of various GH family enzymes using 15-mL tubes by culturing for 4 h at 37 °C and then for 20 h at 15 °C. (C) Expression of various GH family enzymes using a 96-deep-well plate by culturing for 2.5 h at 37 °C and then for 21.5 h at 15 °C. Arrows indicate recombinant protein bands.

In conclusion, the cold-shock expression system can be used to express various proteins without adding IPTG when using AI medium. We found that measuring the OD600 would not be essential and that target proteins could be expressed without rapid cooling.



Information regarding the plasmids used in this study is available at All plasmids were derived from pCold II (Takara Bio Inc., Japan). E. coli BL21(DE3) (Merck, Germany) cells were used as expression hosts. BL21(DE3) cells harboring empty pCold II were used as negative controls. Lysogeny broth (LB) medium (Nacalai Tesque, Japan) and AI medium (#AIMLB0101, Formedium, UK) containing 100 µg/mL ampicillin were used for preculture and experimental culture, respectively.

Culture of transformants

Transformants were cultured in 1 or 2 temperature-controlled shakers (BR-23FP; TAITEC, Japan). OD600 was measured when necessary.

In precultures, 3 mL of LB medium and transformants inoculated from glycerol stocks were added to 15-mL conical tubes and incubated overnight at 37 °C.

For expression in 15-mL conical tubes, 3 mL of AI medium and 50/OD600 µL precultured bacteria were added. After incubating at 37 °C with 200 rpm shaking for a specified period (2-5 h), tubes were transferred to the incubator set at 15 °C and incubated with 200 rpm shaking. Total incubation time was 24 h.

For expression in 96-deep-well plates (#780270, Greiner Bio-One, Austria), 1.5 mL of AI medium and 25/OD600 µL precultured bacteria were added. After incubating at 37 °C with 300 rpm shaking for 2.5 h, the temperature was shifted to 15 °C using the shaker’s programmable function followed by 21.5 h of incubation.

Analysis of expressed proteins

Bacterial precipitates were suspended in an appropriate amount of lysis buffer (50 mM 3-morpholinopropanesulfonic acid [pH 7.5], 150 mM NaCl, and 1 mM ethylenediaminetetraacetic acid). A 1/10 volume of CelLytic B Cell Lysis Reagent (Merck) was added to the suspension and incubated at 25 °C for 20 min. Supernatants obtained after centrifugation were analyzed using SDS-PAGE.


We thank the reviewers for their helpful comments. We would like to thank Editage ( for English language editing.

No comments here
Why not start the discussion?