Role of the spoT gene product and manganese ion in the metabolism of guanosine 5'-diphosphate 3'-diphosphate in Escherichia coli.

Addition of divalent ion chelating agents picolinic acid, 1,10-phenanthroline, or quinoline-2-carboxylic acid to wild type, relA, or relX, but not spoT strains of Escherichia coli increases the levels of guanosine 5'-diphosphate 3'-diphosphate (ppGpp). Poorly chelating analogs of these agents and a larger and more highly charged chelating agent, ethylene glycol bis(beta-amino-ethyl ether) N,N,N',N'-tetraacetic acid are ineffective. Mn2+ reverses the increase in ppGpp. The increase in ppGpp in wild type cells can be explained by an inhibition of degradation. In spoT cells the response is more complex; ppGpp does not increase although degradation is completely inhibited. The lack of increase in spoT cells suggests a role for spoT in synthesis of ppGpp in addition to its known role in degradation. Growth of both spoT+ and spoT cells is inhibited following chelator addition. This suggests that growth inhibition is through a mechanism not directly involving ppGpp. The results of this study provide evidence in intact cells for a role for Mn2+ and the spoT gene product in ppGpp degradation, and provide further evidence for an involvement of spoT and possibly divalent ions in ppGpp synthesis.


Addition
of divalent ion chelating agents picolinic acid, l,lO-phenanthroline, or quinoline-2-carboxylic acid to wild type, relA, or relX, but not SPOT strains of Escherichia coli increases the levels of guanosine 5'diphosphate 3'-diphosphate (ppGpp). Poorly chelating analogs of these agents and a larger and more highly charged chelating agent, ethylene glycol bis@aminoethyl ether) N,N,N',N'-tetraacetic acid are ineffective. Mn2+ reverses the increase in ppGpp. The increase in ppGpp in wild type cells can be explained by an inhibition of degradation.
In SPOT cells the response is more complex; ppGpp does not increase although degradation is completely inhibited.
The lack of increase in SPOT cells suggests a role for SPOT in synthesis of ppGpp in addition to its known role in degradation.
Growth of both SPOT+ and SPOT cells is inhibited following chelator addition. This suggests that growth inhibition is through a mechanism not directly involving PPGPP.
The results of this study provide evidence in intact cells for a role for Mn2+ and the SPOT gene product in ppGpp degradation, and provide further evidence for an involvement of SPOT and possibly divalent ions in ppGpp synthesis.
Exponentially dividing Escherichia coli maintain low levels of the nucleotides pppGpp' and ppGpp. However, when wild type cells are deprived of a source of an amino acid (1, 2), energy (3)(4)(5)(6), sulfur (3), or nitrogen (3), ppGpp or both nucleotides increase. The biochemical mechanisms that regulate metabolism of these nucleotides are complex and have not been clearly defined. At least five gene products are required. relA (7), relB (8), and reZC (9) gene products are involved in synthesis in a complex consisting of ribosomes, mRNA, and codon-specific uncharged tRNA. The gene product of another locus, SPOT, is important in maintaining pppGpp and ppGpp levels (10,ll also a trace metal, which is probably Mn'+, in the metabolism of ppGpp: 1) Breakdown in "cold-shocked" SPOT but not in SPOT' cells requires added MnP+ (13). 2) ppGpp decay in SPOT cells is inhibited by tetracycline (11); this inhibition is overcome by Mnz+ (14). 3) Under conditions "in vitro" in which extracts from wild type cells degrade ppGpp, essentially no degradation is detectable with SPOT cell extracts (15,16); this function is partially restored by Mn')+ (15). 4) The rate of synthesis is altered by the SPOT mutation (17)(18)(19).
To further understand regulation of ppGpp metabolism and the role of SPOT, we have added divalent ion chelating agents to E. coli. We found that after addition of certain hydrophobic chelating agents ppGpp increases in wild type, relA, and relX, but not SPOT cells, and the increase is overcome by Mn')+. A preliminary report of these results has been published.'

EXPERIMENTAL PROCEDURES
Cell Lines-The cell lines used in this study and their sources are listed in Table I (Table I) with the ion chelator picolinic acid (22) increases ppGpp. With strain N720, a 15-fold increaSe is maintained for at least 3% h (Fig. 1). No pppGpp is detectable before or after addition of picolinic acid. Other chelating agents, l,lO-phenanthroliie and quinoline-2-carboxylic acid (22), also increase pp6pp (Table II). Nicotinic acid and 1,5phenanthroline, analogs of picolinic acid and 1, Sphenanthroline, respectively, which are poor ion chelators (22), do not. Picolinic acid is ineffective at 0.7 mm. Since the chelating agents are relatively hydrophobic and may be able to enter the ceil, chelation leading to the increase in ppGpp could be either from medium or within the cell. To distinguish between these two possibilities, a larger and more highly charged &elating agent, EGTA, was tested. We found that it is ineffective in elevating ppGpp levels (Table II), even though it chelates transition metal ions and other divalent ions (23) more effectively than do the other agents that increase ppepp-This suggests t&t chelation within the cell is important. The effect of chelating agents on ATP and GTP levels was also measured. In the experiment shown in Fig. 1, ATP levels remained con&ant during the 3%-h incubation.
GTP levels decreased about 20% within 5 to 10 min and then remained constant. In some other experiments in which similar km-eases in ppGpp were observed, both ATP and GTP decreased. These results show that increases in ppGpp are not necessarily correlated with alterations in ATP.
Effect in Mutant Cells-To understand the mechanism for the increase in ppGpp, we tested the effects of picolinic acid in cells with mutations in genes which influence ppGpp metabolism.
We,fmt tested the requirement for a functional reZA product and found increases in ppGpp in relA strains treated with picolinic acid (Table I). The ppGpp responses in one strain and in its isogenic parent are shown in Fig. 1 (inset).
A r&A independent increase in ppGpp is characteristic of an energy shiftdown (4). We therefore tested the effect of relX, a gene whose product is required for ppGpp synthesis during a glucose-to-succinat shiftdown ( 12) e ppGpp increases within 15 ruin after addition of piculinic acid to relX+ or relX cells grown on glucose (Fig. 2), Thus, the increase in ppGpp is independent of both relA and rekX. We then teested the effect of SPOT (10,11). Surprisingly, spoT cells show no increase in ppGpp after chelator addition (Fig. 2). This is true for all SIT mutant strains tested, regardies of accompanying genotypes (Table I).
Effect 0-n Degradaticua and Synthesis-The increase in ppGpp could be due to changes in synthesis, or degradation, or both.
Chloramphenicol addition, to block synthesis of ppGpp (1,5), was used to examine degradation, In wild type cells, picolinie acid treatment increases the half-time of degradation from less than 20 aec to about 3 min. Essentially identical results are found with AT3 (Fig. 3) (Fig. 1). Degradation is much slower in spoT than it is in SPOT+ cells (11, Fig. 3). When picolinic acid is added to SPOT cells, ATP .'* degradation is reduced to the point that no decay of ppGpp is detectable for 24 min (Fig. 3). This decrease in degradation cannot completely explain the effects of picolinic acid on ppGpp in these mutant cells. With an inhibition of degradation, a large increase in ppGpp would be expected. However, ppGpp does not increase; therefore, synthesis must be shut off as well. The half-time of decrease is 5 min with 1 pM and 2.5 min with 10 pM (Fig. 4). Fe2+, Zn"+, and Co2+ (100 pM) are ineffective (Fig. 4). Mn2+ does not alter basal or isoleucine-starved levels of ppGpp when picolinic acid is not added.
Growth Inhibition-Growth of AT3 (reZA'spoT+) cells is quickly but only transiently inhibited following addition of picolinic acid; growth inhibition is reversed by Mn2+ (Fig. 5A).
This growth pattern is similar to that previously observed in N720 cells3 Since ppGpp increases after picolinic acid addition ( Fig. 1) and decreases after subsequent Mn2+ addition (Fig.  4), ppGpp may be instrumental in growth inhibition. To analyze this possibility, we tested the effect of picolinic acid on growth of AT16 (reZA+spoT) where ppGpp does not in- i crease. Growth of these cells is also inhibited (Fig. 5B). Thus a ppGpp increase is not correlated with growth inhibition.
In fact, the lag is even longer in SPOT cells, and Mn'+ added at 3.5 pM is not as effective in overcoming growth inhibition. (Although resumption of SPOT cell growth without added Mn2' was not detected in this experiment (Fig. 5), growth does resume with longer incubations.) The longer period of growth inhibition in SPOT cells is surprising and is not easily explained by alterations in ppGpp metabolism, the only known defect in SPOT cells. This suggests that the SPOT mutation may exert pleotropic effects on cells. We previously proposed that the primary cause for growth inhibition by chelator treatment is reduced RNA synthesis." In uitro, l,IO-phenanthroline inhibits RNA polymerase by chelation of essential metal ions on RNA polymerase (24). In uiuo inhibition could be due to a similar mechanism. Alter- natively, an increase in ppGpp could be responsible for RNA synthesis inhibition (25,26). However, growth inhibition in SPOT cells, where ppGpp does not increase following picolinic acid addition, suggests that an elevation of ppGpp is not responsible.
The longer lag period of growth inhibition after chelator addition (Fig. 5) and the decreased rate of ppGpp degradation (11, Fig. 3 According to this scheme, pppGpp is formed by a pyrophosphoryl transfer from ATP to GTP (7) (Reaction 1). pppGpp is dephosphorylated to ppGpp (Reaction 2), and ppGpp is degraded (Reaction 3) to GDP (15,16). Under normal growth conditions, E. coli maintains: low, but detectable levels of ppGpp. Synthesis of this basal ppGpp is independent of relA mutations (25) and is not clearly understood. Presumably, relX is required for basal synthesis. The ppGpp increase during an energy shiftdown is due to a decrease in degradation and not to an increase in synthesis (25,26). Mutant reZX cells have low basal ppGpp levels and are unable to increase ppGpp during a glucose-to-succinate shiftdown (12). reZX may directly catalyze pppGpp (or ppGpp) formation, or it may regulate basal activity of the "stringent factor" in some unknown manner independent of relA mutations. spoT cells have elevated basal ppGpp levels (18, 19,30). This increase is due to a decrease in the rate of degradation (11); however, the increase in ppGpp is less than that expected for the decrease in degradation.
It follows that the rate of basal synthesis is also decreased in apoT cells (17)(18)(19). Extracts from SPOT cells contain much less ppGpp degrading activity than do extracts from wild type cells (15,16). This decreased activity is partially restored by MnZ+ (15). These in vitro results suggest a direct role for the SPOT gene product and also for Mn"+ in ppGpp degradation.
The observed effects on synthesis have not been explained.
When wild type cells under normal growth conditions are treated with chelating agents, ppGpp increases. This effect is presumably due to chelation of trace metals that are essential for degradation of ppGpp: 1) the increase can be explained by the measured inhibition of ppGpp decay; 2) structurally similar, poorly chelating analogs of these agents are ineffective; 3) a divalent ion, Mn'+, overcomes the action of the chelators. It should be emphasized that these results demonstrate that Mn2+ can activate degradation, but it does not necessarily mean that Mn2+ is the normally required ion.4 ' Mnz+ is poorly chelated by l,lO-phenanthroline (22). To the best of our knowledge, the ability of picolinic acid to chelate Mn*' has not been analyzed.
Our results suggest that Mn'+ is poorly chelated by picolinic acid. Therefore, small amounts of Mn'+ are sufficient to overcome the ion deficiency induced by chelator addition. TIME ,hr, ppGpp binds Mn"+ (31). Thus the possibility that the complex ppGpp . Mn'+ is the substrate for degradation should be considered.
Alternatively, the affinity of ppGpp for Mn'+ could aid binding of ppGpp to an enzyme. Mn'+ complex.
Genetic and metabolic studies show that the increase in ppGpp is separable from reZA-dependent ppGpp synthesis. Identical responses are observed in relA+ and relA cells. Isoleucine does not decrease ppGpp in picolinic acid-treated reZA+ cells starved for isoleucine before picolinic acid addition." Also, the increase following chelator addition occurs while RNA synthesis is inhibited"; relA-dependent ppGpp synthesis requires mRNA in vitro (7) and is inhibited in viva by rifampicin (32). Our results demonstrate that chelator addition to intact cells inhibits ppGpp decay. This is expected from previously reported requirements for MnZf in the degradation of ppGpp (13)(14)(15)(16). In addition, our results show a complexity in ppGpp metabolism that has not previously been observed. ppGpp increases in relA relX SPOT' cells after chelator addition. This raises the question of how ppGpp is synthesized in these cells. Even more intriguing is the fact that there is no increase in reZA+ reZX+ SPOT cells when similarly treated. These results suggest that the SPOT gene product either influences synthesis (17)(18)(19) through some undetermined regulatory mechanism, or that the SPOT gene itself catalyzes pppGpp (or ppGpp) formation.