To the editor:

Growth and multiplication of specific cells and organisms occurs within narrow physico-chemical conditions. Despite the fundamental importance of one (or at most two) cellular functions that determine the growth range of a cell or an organism, in most cases we have little idea of their identity. Here, we report the finding that chaperonins determine growth at lower temperatures of the bacterium Escherichia coli K-12. The finding has implications for the use of bacteria in environmental biotechnology, biochemical engineering and recombinant protein production.

E. coli is a mesophilic bacterium able to grow well in the temperature range from 21 °C to 49 °C, with an optimum at about 37 °C. The growth rate of E. coli strain XLOLR drops rapidly as incubation temperatures decrease from 20 °C, and the minimum for measurable growth is around 7.5 °C (ref. 1; Fig. 1a). Interestingly, the ability of the E. coli chaperonins GroEL and GroES to fold denatured proteins also rapidly decreases below 15 °C (ref. 2; Fig. 1b). These chaperones promote the folding and/or assembly of over 30% of cellular proteins, are required for bacteriophage morphogenesis and have a role in protein secretion3,4. The question thus arises of whether the vital role of chaperonins is the function that determines the lower temperature limit of E. coli growth.

Figure 1: In vivo and in vitro properties of the chaperonins of Oleispira antarctica.
figure 1

(a) Effect of expression of the O. antarctica chaperonins on the growth of E. coli at different temperatures. (b) In vitro refolding activities of O. antarctica Cpn60/10 and E. coli GroEL/ES chaperonins at different temperatures. Data are not fitted to any model. For details see Supplementary Methods online.

We recently isolated a new psychrophilic bacterium, Oleispira antarctica strain RB-8 T (DSMZ14852 T), from Antarctic seawater5 and characterized its chaperonin Cpn60 and co-chaperonin Cpn10 (Ferrer, M., Lünsdorf, H., Chernikova, T.N., Yakimov, M.M., Golyshin, P.N. & Timmis, K.N., unpublished data; Swiss-Prot accession numbers Q8KM30 and Q8KM31, respectively). Both chaperonins show high protein refolding activities in vitro at temperatures of 4–12 °C (16-fold higher than at 30 °C; Fig. 1b). We reasoned that if the cold-sensitive GroEL and GroES chaperonins of E. coli determine its lower growth temperature, and if the cold-adapted Cpn60 and Cpn10 chaperonins of O. antarctica can assume the roles of GroEL and GroES in E. coli, then introduction of the corresponding genes into, and their expression in, E. coli should extend its temperature range of growth by decreasing its lower temperature limit.

We therefore cloned and expressed the O. antarctica genes cpn60 and cpn10, encoding the two chaperonins, under the control the Plac promoter in E. coli strain XLOLR and examined the growth characteristics of the transgene after induction of expression with isopropyl-d-galactopyranoside (IPTG; Fig. 1a). The strain bearing the construct grows much faster than the parental strain at low temperatures: 3-fold faster than the parental strain at 15 °C, 36-fold faster at 10 °C and 141-fold faster at 8 °C (growth rate of parental E. coli 0.002 h−1; that of the transgenic strain 0.282 h−1). No growth of the parental E. coli was detected below 8 °C, whereas the transgenic strain grew at temperatures below 4 °C As determined using the square-root growth model of Ratkowsky et al.6, the theoretical minimum temperatures for the parental and transgenic E. coli would be 7.5 °C and −13.7 °C, respectively (see Supplementary Methods online).

To rule out the possibility that hyperexpression of chaperones per se lowers the growth limit of E. coli, we also expressed the GroEL and GroES chaperonins to similar cellular levels— 160 μg GroEL/ES per milligram of protein versus 120 μg Cpn60/10 per milligram of protein, using plasmids pBB528 and pBB541 (kindly provided by E. Betiku and U. Rinas (GBF)), in which the chaperonins are expressed from the same Plac promoter (for details, see Supplementary Fig. 1 online). The growth characteristics of E. coli at temperatures below 15 °C were not influenced by hyperexpression of the homologous chaperonins (data not shown). This demonstrates that the depression of the lower limit of growth of E. coli by Cpn60 and Cpn10 is due to a qualitative change in cellular function effected by the psychrophilic chaperonins and not to a quantitative change in chaperonin level.

We have thus established causality between increased growth rates of E. coli at lower temperatures and depression of its minimum temperature for growth, on one hand, and recruitment of the O. antarctica cold-adapted chaperones, on the other. This, in turn, demonstrates that the chaperones of E. coli are the rate-limiting cellular determinant of growth at lower temperatures. As the principal function of chaperones is protein folding, and we have shown a correlation between protein folding ability and growth at lower temperatures, it is highly probable that the cellular function determining growth of E. coli at lower temperatures is protein folding. Nevertheless, we cannot presently exclude the possibility that another chaperonin-mediated cellular function is responsible for the altered growth characteristics.

Two related questions arise from this finding: how widespread is chaperone-determined growth among other organisms and under different environmental conditions, and will it be possible to extend the temperature ranges of growth of other cells by recruiting chaperones that have the required properties (e.g., can the temperature ranges of growth of psychrophiles and mesophiles be extended by recruitment of chaperones from mesophiles or thermophiles)?

Whatever the case, the finding presented here has implications for biotechnology. One important strategy for developing new biocatalytic processes is to mine biodiversity by creating genomic libraries of DNA resources in E. coli and screening them for desired activities7. Enzymes from psychrophiles are particularly interesting for certain enzymatic bioconversions8, but some cannot be produced in an active form in E. coli because they are denatured in vivo at the temperatures used in cultivating this bacterium9. Thus, screens for psychrophilic enzyme activities would clearly benefit from growth of such E. coli libraries at low temperatures, and subsequent production of identified enzymes will also require low-temperature growth of the host organism. Use of an E. coli host producing the O. antarctica chaperonins, or other cold-tolerant chaperones, will permit both lower growth temperatures and efficient folding of the psychrophilic proteins produced.

It is noteworthy that, if the temperature ranges of growth of organisms generally prove to be modifiable by recruitment of heterologous chaperones, this could become a generic means of altering their biogeography and of making them more robust either for a wide range of environmental applications that are subject to climate-related fluctuations in temperature (waste treatment, bioremediation, microbially mediated plant growth promotion and protection, retting, biomining, etc.) or for biotech-nological processes at temperatures that are stressful for the organisms used. It may also have other interesting implications for agriculture if the cold adaptation strategy we have developed for E. coli is also applicable to plants and can be used to increase their robustness to weather conditions and extend their growth windows in time (length of growing season) and space (latitude growth range).

Note: Supplementary information is available on the Nature Biotechnology website.