Effects of light deprivation on RNA synthesis, accumulation of guanosine 3'(2')-diphosphate 5'-diphosphate, and protein synthesis in heat-shocked Synechococcus sp. strain PCC 6301, a cyanobacterium

The rate of total RNA synthesis, the extent of guanosine 3'(2')-diphosphate 5'-diphosphate (ppGpp) accumulation, and the pattern of protein synthesis were studied in light-deprived and heat-shocked Synechococcus sp. strain PCC 6301 cells. There was an inverse correlation between the rate of total RNA synthesis and the pool of ppGpp, except immediately after a temperature shift up, when a parallel increase in the rate of RNA synthesis and accumulation of ppGpp was observed. The inverse correlation between RNA synthesis and ppGpp accumulation was more pronounced when cells were grown in the dark. Heat shock treatment (47 degrees C) had an unexpected effect on ppGpp accumulation; there was a fairly stable level of ppGpp under heat shock conditions, which coincided with a stable steady-state rate of RNA synthesis even in the dark. We found that the pattern of dark-specific proteins was altered in response to heat shock. The transient synthesis of several dark-specific proteins was abolished by an elevated temperature (47 degrees C) in the dark; moreover, the main heat shock proteins were synthesized even in the dark. This phenomenon might be of aid in the study of cyanobacterial gene expression.

Growing bacteria are strongly influenced by their nutritional, chemical, and physical environment, and they adjust their growth to alterations in these conditions in their habitat (19). Procaryotic organisms respond to unfavorable alterations of these factors by a reduction of their growth rate and by a concomitant accumulation of guanosine 3'(2')diphosphate 5'-diphosphate (ppGpp), the compound implicated as a key substance for the stringent control system (10,16,20,22). The accumulation of ppGpp under shift down conditions is generally considered to be a signal for the curtailment of several metabolic processes. In heterotrophic bacteria, the intracellular level of ppGpp is inversely correlated with the rate of synthesis of stable RNA species, suggesting that the regulation of cell metabolism is directly or indirectly under the control of the effector molecule ppGpp (for reviews, see references 8, 12, 13, and 19). Recently it has been reported that a temperature shift up of Escherichia coli cells leads to a transient accumulation of ppGpp (10,16,20,22). Ryals et al. (22), as well as Mackow and Chang (16), demonstrated that an inverse correlation exists between ppGpp level and the rate of total RNA synthesis in heat-shocked E. coli, which is consistent with this view.
In cyanobacteria, the mode of accumulation of ppGpp is comparable with that reported for heterotrophic bacteria (17). However, there are conflicting results for the correlation between nitrogen starvation and ppGpp accumulation (1,2,11,17,25) and the effect of light (hence, energy) deprivation, which does not seem to be comparable with the nutritional step-down of heterotrophic bacteria (3,17,25). In studies dealing with the light-dependent regulation of gene expression in Synechococcus sp., an obligate autotrophic cyanobacterium, the pleiotropic nature of regulatory responses has been emphasized (9,23).
Indeed, under energy deprivation (dark) conditions, some specific polypeptides are synthesized (23). The heat shock phenomenon has recently been reported in Synechococcus sp. (6), and the response of the cells to elevated temperatures has been found to involve the reduction of growth rate and induction of a specific set of polypeptides, the so-called heat shock proteins. Since the reduction of growth during heat shock in Synechococcus and the nutritional or energy shift down in various cyanobacterial systems may have much in common with heterotrophic shift down systems, one would also expect a role for ppGpp in the temperatureinduced regulation of stable RNA and protein synthesis in cyanobacteria.
Here we describe the effects of heat shock on the synthesis of RNA and the accumulation of ppGpp upon transfer of Synechococcus cultures from light to darkness. We relate these data to the pattern of proteins synthesized under heat shock (6) and in the dark (energy starvation [9]). Such studies are of interest not only in their own right, but from the viewpoint of light-dependent cyanobacterial gene expression as well.

MATERIALS AND METHODS
Growth conditions. Synechococcus sp. strain PCC 6301 (Anacystis nidulans ATCC 27144) was grown in the liquid medium of Allen (4) with jacketed glass culture vessels as described earlier (6).
Heat shock conditions. Exponentially growing Synechococcus cultures (0.500 to 0.700 A800 units) were heat shocked as described earlier (6), with a minor modification, i.e., in the series of experiments, 10-ml jacketed glass culture vessels were used to study ppGpp accumulation and RNA synthesis at normal (39°C) and elevated (47°C) temperatures.
Isotope-labeling conditions and analysis of RNA synthesis. Measurement of ppGpp and guanosine 3'(2')-triphosphate 5'-diphosphate pools was done in formic acid extracts, which were chromatographed as described by Borbdly et al. (5) by using polyethylenimine thin-layer chromatography plates (Macherey and Nagel Co., Duren, Federal Republic of Germany).
Polyacrylamide gel electrophoresis of Synechococcus proteins. Exponential-phase Synechococcus cultures (1-ml portions) were labeled with 0.5 MBq of U-_4C-labeled protein hydrolysate for 1.5 h in the light and 3.0 h in the dark under both normal (39°C) and heat shock (47°C) conditions, unless otherwise stated. After labeling, the samples were processed for 10 to 18% linear sodium dodecyl sulfate-polyacrylamide gradient gel electrophoresis and autoradiography (6).

RESULTS
Effect of elevated temperature on RNA synthesis of Synechococcus sp. In Synechococcus cells, a shift in growth temperature from 39 to 47°C resulted in an increased rate of accumulation of radioactive uracil for a 20-min period (Fig.  1). This initial increase was followed by a decrease in the rate of accumulation. A minimum for the uracil accumulation was observed between 20 and 30 min after the temperature increase to 47°C. After this time, an increase in RNA accumulation, as determined by the uptake of [3H]uracil, began and continued essentially at the same rate during the remainder of the heat shock (Fig. 1). Since the mere accumulation of radioactive uracil under these conditions may not be a true measure of RNA synthesis, we also measured the rate of total RNA synthesis in heat-treated cells by [3H]uracil pulse-labeling of aliquots of Synechococcus cultures that had been continuously labeled with [14C]uracil.
When the rate of total RNA synthesis was measured, an immediate increase in RNA synthesis was observed after the shift to an elevated temperature ( Fig. 2A). A subsequent gradual decrease was noted in the rate of total RNA synthesis between 10 and 30 min. Subsequently, the rate of RNA synthesis gradually increased again and then reached a new steady-state rate for the period of heat treatment. In cultures shifted back to the initial growth temperature (39°C), the rate of incorporation of the radioactive RNA precursor decreased by some 30 to 40% for 60 min, after which the rate of total RNA synthesis returned to that characteristic of normal cells not stressed by heat treatment.
When the rate of total RNA synthesis of heat-shocked cells was measured in the dark, there was an immediate decrease in the rate of RNA synthesis followed by a slight, transient increase (at about 40 min), and afterwards, a steady state was maintained for the remainder of the heat shock period; this rate was notably lower than that observed under light conditions (Fig. 2C).
To analyze more closely the effect of different down shifts (heat shock and darkness) on RNA synthesis, we examined the consequences of different sequences of down shifts. Cells heat shocked in light responded to darkness under both normal (39°C) and elevated (47°C) temperatures with a rapid decrease in the rate of total RNA synthesis (Fig. 2B). This rate was slightly higher in cells maintained at 47°C than in those returned to the initial growth temperature (39°C);  with a rapid increase in RNA synthesis (Fig. 2C). Cells illuminated at 47°C displayed the more dramatic, immediate increase in RNA synthesis, followed by a gradual decrease which ultimately reached a new steady state of total RNA synthesis (Fig. 2C). Cells reincubated under the initial growth conditions showed a less pronounced initial response to reillumination, but one which seemed more evenly sustained over the time period. Cells which were maintained in the dark during and after heat shock and were shifted to the lower (growth) te'mperatuire did not synthesize RNA at the The culture was shifted from normal (39°C) to elevated (47°C) temperature at time zero. One hour later, the cells were light deprived at 47°C. (B) The culture was shifted from normal (39°C) temperature to both elevated (47°C) temperature and darkness at time zero. In the time zero sample, the basal level of ppGpp was 33 pnmol per A800 unit. The relative concentration of nucleotides was calculated as explained in the legend to Fig. 3.
increased rate characteristic of similar, but illuminated, cells (Fig. 2D). Instead a second rapid decrease in synthesis, but of a more limited magnitude, typified the cells, which then established a moderate rate of RNA synthesis.
Darkness per se, rather than incubation temperature, was the imposed variable affecting the rapidity and magnitude of the cellular response vis-'a-vis RNA synthesis when a treatment of heat plus illumination followed the initial 2-h stress period (cf. Fig. 2C and E). The kinetics of RNA synthesis in heat-shocked cells seemned to be similar, independent of the preceding fate of cells ( Fig. 2A through D). Cells heat shocked in the dark respond to light shift up at the normal temperature with an initially rapid and then a gradual increase in the rate of RNA synthesis (Fig. 2C). However, the kinetics of RNA synthesis differed between the cultures heat shocked in light versUs in the dark; the cells heat shocked in light responded to normal temperature and light first with a decrease and then an increase in the rate of RNA synthesis ( Fig. 2A), while cells heat shocked in the dark responded to normal temperature and light only with an increase in the rate of RNA synthesis (Fig. 2C). These kinetics of RNA synthesis were the same as those seen when a culture grown in the dark at normal temperature was shifted to light (Fig.  2E). A decrease in temperature of a culture both heat shocked and kept in the dark resulted in a slight but significant decrease in the rate of total RNA synthesis. In addition, there was an immediate inhibition of rRNA synthesis in heat-shocked Synechococcus cells in the light, and the elevated temperature seemed to inhibit the lightdependent in vivo postmaturational cleavage of 23S rRNA (data not shown).
ppGpp Accumulation after temperature upshifts under light or dark regimens. In illuminated Synechococcus sp., after a temperature increase to 47°C, there was a rapid accumulation of ppGpp to about 6to 10-fold over the basal level (Fig.  3). The accumulation of this nucleotide reached a peak after 10 min of heat shock in light, but this level was substantially lower than that noted in the dark (energy starvation) condi- tions at the initial growth temperature (390C). The decline of ppGpp level was fairly slow under either condition; cells subjected to a temperature increase maintained their ihcreased ppGpp content at a rather stable level for at least 2 h, which was similar to the behavior of cells at 39°C in the dark (5, 17; Fig. 3). To study whether the increase in ppGpp level of heat-shocked cells was due to changes in synthesis or degradation or both, chloramphenicol (100 ,ug ml-') was added to the culture (7,15,23). After the addition of chloramphenicol, there was an immediate dramatic decrease in the pool size of ppGpp in heat-shocked Synechococcus sp. (Fig. 4), as would be expected if the effect of synthesis on the ppGpp level were more significant than was a change in the rate of degradation of ppGpp (7). The ppGpp level remained unexpectedly high for a longer period of time if light deprivation and heat stress regimens were simultaneous (Fig. 5). Under conditions of constant illumination, a shift up in temperature was followed by an increase in ppGpp level, and a subsequent return to the initial temperature resulted in an equally rapid decrease to a near-basal level of ppGpp (Fig.  6A). When the shift up was instead followed by darkness, there was another, even more significant increase in ppGpp levels under both normal (growth) and continued heat stress conditions (Fig. 6B).
When heat stress was applied concomitantly with dark incubation, the increase in ppGpp was 10-fold greater (Fig.  6C) than that seen under conditions of heat and illumination (Fig. 6A) of ppGpp (Fig. 6C), but the return to the basal, pre-heat stress level was much slower than seen in Fig. 6A, whether the heat stress continued or a shift down to growth temperature ensued. When a regimen of simultaneous darkness and heat was followed by darkness (but at the initial, lower growth temperature), the ppGpp content decreased (Fig.  6D), but more slowly and to a much lesser extent than it did under comparable but illuminated conditions (Fig. 6C). When the initial stress was darkness instead of heat, the level of ppGpp in cells rose sharply. When the growth temperature was maintained, however, both the duration and extent of the increase were less extensive than when photosynthesis was occurring (Fig. 6E). A shift from darkness at 39°C to a combination of light and 47°C resulted first in a sharp fall in the nucleotide level and then a rise to about 10 times the basal level for an extended time period. In contrast, when the dark, 39°C treatment was followed by heat stress in the absence of light, an initial decrease of ppGpp was followed by a comparatively slow but sustained increase (Fig. 6F).
Effect of darkness on the pattern and kinetics of the synthesis of heat shock proteins. To examine how darkness (energy deprivation) alters the response of Synechococcus sp. to heat shock treatment, the control (390C) and heatshocked (47°C) cultures were simultaneously put in the dark and pulse-labeled with 14C-labeled protein hydrolysate. In these cells, the same major heat shock polypeptides were synthesized as we have described elsewhere (6) for cells heat shocked in the light, albeit at a lower rate (Fig. 7). The specific pattern of polypeptide synthesis seen in cells in the dark at normal temperature was altered at the elevated temperature (Fig. 7); the synthesis of several dark-specific proteins with unknown functions was abolished in the dark, elevated-temperature conditions. The molecular masses of these polypeptides are indicated in Fig. 7. The disappearance of the 80.0-, 17.8-, 15.4-and most obviously, the 23.0-kilodalton (kDa) dark-specific proteins was characteristic.
To compare the kinetics of polypeptide synthesis in lightdeprived cells (39°C) with cells both light deprived and heat shocked (47°C), samples were pulse-labeled for 30 min with 14C-labeled protein hydrolysate over a 2.5-h period. Some polypeptides (e.g., the 80.0-, 23.0-, 17.8-, and 15.4-kDa proteins; Fig. 8A) were synthesized transiently solely in the dark, while others (e.g., the 44.5-, 31.7-, and 26.2-kDa proteins; Fig. 8B) were synthesized transiently in the dark while heat stressed. On these two autoradiograms, the above-mentioned phenomenon, i.e., the lack of several dark-specific proteins in heat-shocked cells in the dark, is even more obvious (Fig. 8A and B). The results are summarized in Table 1.

DISCUSSION
In cyanobacteria, experimental perturbation of growth alters the rate of stable RNA accumulation (9,17). Although few studies have been done on the phenomenon in cyanobacterial systems, it is clear that darkness (energy deprivation) may control stable RNA synthesis, as is known for heterotrophic systems in which availability of charged tRNA is altered by inhibitors or by nutritional step-down procedures (3,9,24,25). Since we recently demonstrated that heat stress (elevated temperature [47°C]) reduces growth rate and protein synthesis in the cyanobacterium Synechococcus sp. (6), we expected changes in RNA synthesis and in the pool size of highly phosphorylated nucleotides to accompany this physiological stress as well. Indeed, after a temperature increase to 47°C, the RNA accumulation increased during the early period of stress treatment, subsequently decreased, and again increased near the end of the temperature shock period. The explanation of this oscillation in the rate of RNA accumulation is not known, but recent observations on E. coli (16,21,22) may be relevant. In E. coli, an inverse correlation exists between ppGpp content and the rate of total RNA synthesis during temperature shifts up and down except when the elevated temperature directly affects RNA chain elongation.
To analyze more precisely the effect of elevated temperatures on RNA synthesis, we measured the rate of total RNA synthesis in Synechococcus sp. during transitions from a normal growth temperature (39°C) to an elevated one (47°C) and from light (photosynthesis) to the dark (energy deprivation). By measuring the rate of total RNA synthesis rather than only RNA accumulation, an immediate increase VOL. 169, 1987 go 638 SURANYI ET AL. in RNA synthesis was observed after temperature stress (shift up). A subsequent decrease in the rate of total RNA synthesis (at 10 to 30 min) preceded a gradual increase and a new steady-state rate of RNA synthesis during the latter period of heat stress. These alternations in the rate of total RNA synthesis were correlated with dramatic changes in ppGpp pool size. When the ppGpp level rose, cessation in RNA synthesis was observed. Under heat shock conditions, the higher level of ppGpp coincided with a lower rate of RNA synthesis. This phenomenon was even more apparent if the ppGpp level and the rate of total RNA synthesis were compared in the dark. The slight decrease in ppGpp content appeared to trigger a subsequent increase in the rate of total RNA synthesis; this seemed to be true not only for the heat shock period, but also for the recovery period, when the temperature was returned to normal (39°C), except for a lag period. Thus, this study extends to an obligate photosynthetic cyanobacterium the inverse correlation between ppGpp levels and the rate of RNA synthesis, a phenomenon which thus far had been known only for heat-stressed heterotrophic organisms (16,21,22). Under light, the primary difference observed was in the early period of heat shock, when the expansion of the ppGpp pool was accompanied by an increase, in the rate of total RNA synthesis. The explanation of this phenomenon was not clear in our system. Although we have not measured the elongation rate of RNA chains in heat-shocked cyanobacterial cells, we see no reason to suppose that the elevated temperature has different effect on RNA chain elongation in Synechococcus sp. than in E. coli (21,22). A similar intriguing irregularity existed between RNA synthesis and ppGpp pool size when heatshocked cells were returned to normal (growth) temperature either in light or in the dark. Under these conditions of decreasing RNA synthesis, a lag period preceded the new steady-state rate of RNA synthesis. This irregular, concomitant oscillation of RNA synthesis and ppGpp pool size was seen only at the beginning of temperature shifts, when changes of temperature may have had an immediate and direct effect on RNA chain elongation rates. For heterotrophic bacteria, the control of RNA and protein synthesis during transitions of growth rate is well understood in two systems, i.e., in situations irq which alterations are due to restricted tRNA aminoacylation and in systems in which changes are the consequence of manipulations of carbon or energy sources (19). However, unlike the situation in heterotrophs, in which the sources of carbon and energy are usually one and the same, for this cyanobacterium, the source of energy (light) differs from the source of carbon (bicarbonate); thus, the use of heat stress may be a valuable experimental approach for the analysis of potential differential effects of energy versus carbon starvation on RNA accumulation. In heterotrophic organisms, ppGpp plays a physiologically important role in the regulation of translation as well as of RNA synthesis (14,18,20).
Accordingly, this study is of interest, at least qualitatively, for the consequences of darkness on the cyanobacterial heat shock proteins. Singer and Doolittle (23) demonstrated that J. BACTERIOL. ,k, the lack of light provokes specific changes in the pattern of protein synthesis in Synechococcus sp. Our more detailed analysis of this phenomenon at normal growth temperature thus extends and refines these observations by showing that there are indeed three classes of proteins that can be observed: (i) proteins transiently synthesized in the dark, (ii) polypeptides accumulated only in the dark, and (iii) proteins synthesized (but not necessarily at the same rate) both in light and in the dark. Nonetheless, we are left with the necessity of explaining what may be the function(s) of dark-specific proteins and now of the heat shock proteins in cyanobacteria. These proteins may provide an experimental system for a better understanding of dark-specific cyanobacterial gene expression. Indeed, we infer that the synthesis of heat-shock (stress) proteins in the dark indicate a significant role for these proteins in cell homeostasis. In cultures simultaneously light deprived and heat shocked (47°C), we observed (i) the main heat shock proteins, (ii) several transiently synthesized dark-specific proteins (of 44.5, 31.7, and 26.2 kDa), and (iii) polypeptides synthesized in the cells under each condition (light, dark, and heat shock in the dark). However, 80.0-, 23.0-, 17.8-, and 15.4-kDa proteins, which seem to be dark specific under normal growth conditions (39°C) (23), are abolished by heat. Thus, these proteins may have a specific role(s) in cells in the dark; the phenomenon needs further study in both obligate as well as facultative autotrophic cyanobacteria.
Our consideration of the results reported here leads us to the expectation that the high and more or less stable level of ppGpp in heat-shocked and light-deprived Synechococcus sp. may have pleiotropic effects on cyanobacterial metabolism.