When the growth temperature of an exponential culture of Escherichia coli is abruptly decreased from 37 to 10°C, growth stops for several hours before a new rate of growth is established. During this growth lag the number of proteins synthesized is dramatically reduced, and at one point only about two dozen proteins are made; 13 of these are made at differential rates that are 3 to 300 times increased over the rates at 37°C. The protein with the highest rate of synthesis during the lag is not detectably made at 37°C. The identities of several of these cold shock proteins correlate with previous observations that indicate a block in translation initiation at low temperatures.Escherichia coli can maintain balanced growth in rich medium over the temperature range from 10 to 49°C. The growth rate varies with temperature as if cellular growth were a simple chemical reaction with a temperature characteristic of 12,000 to 14,000 cal/mol (50,200 to 58,600 J/mol) between 20 and 37°C (7). Raising the temperature above 400C or lowering it below 20°C results in progressively slower growth, until growth ceases at the maximum temperature of growth, 49°C, or the minimum, 8°C.The steady-state levels of most cellular proteins do not change greatly within the normal temperature range, but many exhibit significantly increased or decreased levels near the temperature extremes (6). On the other hand, temperature shifts of only a few degrees within the normal range result in immediate, transient changes in the rates of synthesis of most cellular proteins (9). The synthesis of about two dozen proteins accelerates dramatically after a shift from 28 to 42°C, constituting the well-characterized heat shock response of E. coli (12).In this work, we asked whether there is an analogous "cold shock" response upon a shift of E. coli from 37 to 10°C. The synthesis of individual cellular proteins was monitored through the use of two-dimensional gel electrophoresis. We found that several proteins are induced during cold shock, and we have termed this induction the cold shock response. The identification of some cold shockinduced proteins has allowed us to speculate on the function of this response.MATERIALS AND METHODS Bacterial strains. E. coli strain W3110 has been described previously (18).Media and bacterial growth. The rich medium used was composed of defined MOPS medium (11) ,uCi/ml) for 5 min before the shift or for 30 min at various times after the shift. Samples were mixed with cells from the reference culture. Extracts were prepared and processed by O'Farrell two-dimensional polyacrylamide gel electrophoresis (15). Individual protein spots were excised, processed by using a Packard sample oxidizer, and counted in a scintillation counter. The 3H/14C ratios of the individual polypeptides were determined and divided by the ratios of the unfractionated total protein extract. The resulting values were the differential rates of synthesis of the individual proteins and were normalized to 1.0 prior to the shift.Identification of polypeptides. ...

Nearly all cells respond to an increase in temperature by inducing a set of proteins, called heat shock proteins (HSPs). Because a large number of other stress conditions induce the HSPs (or at least the most abundant ones), this response is often termed the universal stress response. However, a careful study ofconditions that truly mimic a temperature shift suggested that these proteins are induced in response to a change in the translational capacity of the cell. To test this directly, Escherichia coli cells were treated with antibiotics that target the prokaryotic ribosome. Two-dimensional gels were used to evaluate the ability of these drugs to alter the rate of synthesis of the HSPs. One group of antibiotics induced the HSPs, whereas a second group repressed the HSPs and induced another set of proteins normally induced in response to a cold shock. Depending on the concentration used, the induction of the heat or cold shock proteins mimicked a mild or severe temperature shift. In addition, antibiotics of the cold shock-inducing group were found to block high temperature induction of the HSPs. The results implicate the ribosome as a prokaryotic sensor for the heat and cold shock response networks, a role it may serve in eukaryotes as well.The heat shock response, first described in Drosophila, is now recognized as a nearly universal cellular response to a shift up in temperature (1, 2). The components of the response, and the general nature of its induction (increased initiation of transcription of heat shock genes), have been conserved across the boundaries of the prokaryotic, eukaryotic, and archaebacterial kingdoms, strongly suggesting some important role(s) for the response in cell physiology (1, 2).Many stress conditions have been reported to induce the heat shock response, leading to the idea that there may be multiple cellular targets, or sensors, that can generate the inducing signal (2). Close analysis of the effects of these stress conditions has revealed that in most instances the response is weak and involves only a subset of the heat shock proteins (HSPs) (3). Furthermore, this weak, partial response constitutes, for most of these stress conditions, a trivial portion of the cell's overall response (3). Only two stress conditions have been reported in Escherichia coli that induce the heat shock response exclusively and in a temperature-like manner-treatment with ethanol (3) and depletion of 4.5S RNA (4). Both of these stresses alter the translational capacity of the cell.Paradoxically, some conditions that decrease the translational capacity of the cell cause repression of the HSPs (5-7). One of these conditions is a shift to low temperature. A shift ofE. coli from 370C to 10'C leads to a characteristic response, the most prominent feature of which is the induction of 14 proteins, 13 described previously (6) plus one newly discovered (Fig. 2), several of which are involved in translation (6), a finding that meshes with earlier reports that cold temperature restricts bacterial growth by int...

Heat and various inhibitory chemicals were tested in Escherichia coli for the ability to cause accumulationi of adenylylated nucleotides and to induce proteins of the heat shock (htpR-controlled), the oxidation stress (oxyR-controlled), and the SOS (lexA-controlled) regdlons. Under the conditions used, heat and ethanol initiated solely a heat shock response, hydrogen peroxide and 6-amino7-chloro-5,8-dioxoquinoline (ACDQ) induced primarily an oxidation stress response and secondaily an SOS response, nalidixic acid and puromycin induced primarily an SOS and secondarily a heat shock response, isoleucine restriction induced a poor heat shock response, and CdCI2 strongly induced all three stress responses. ACDQ, CdCl2, and H202 each stimulated the synthesis of approximately 35 proteins by factors of 5-to 50-fold, and the heat shock, oxidation stress, and SOS regulons constituted a minor fraction of the overall cellular response. The pattern of accumulation of adenylylated nucleotides during these treatments was inconsistent with a simple role for these nucleotides as alarmones sufficient for triggering the heat shock response, but was consistent with a role in the oxyRlmediated response.

A group of nine proteins of Escherichia coli KI2 vary in steady state level with growth temperature, and are particularly abundant above 4O°C. The identities of most of these HTP (high temperature production) proteins are unknown; they are primarily recognizable on two-dimensional polyacrylamide gels by their very high rates of synthesis during the ten-minute period following a shift-up in temperature. This stimulation, as much as 20-fold for some HTP proteins, is abolished by a conditionally lethal nonsense mutation in a chromosomal gene located at 75 minutes. Evidence suggests that this regulatory gene, htpR, makes an activator protein that is required for heat induction of HTP proteins.

The pattern of proteins synthesized in Escherichia coli during steady-state growth in media with ample inorganic phosphate (P i ), upon limitation for P i (without an alternative phosphorous compound), and during steady-state growth in media containing phosphonate (PHN) as the sole P source was examined by twodimensional gel electrophoresis. Of 816 proteins monitored in these experiments, all those with differential synthesis rates greater than 2.0 or less than 0.