Identification and Characterization of Starvation-Regulated Genetic Loci in Salmonella typhimurium by Using Mu d-Directed lacZ Operon Fusions MICHA EL P. SPECTOR , t YONG K. PARK , SIMIN TIRGARI ,4 TANIA GONZALEZ , AND JOHN Marshall University School of Medicine , Huntington , West Virginia 25704 Received 24 April 1987/Accepted 23 October 1987 W. FOSTER § * We used the technique of Mu d-directed lac operon fusion formation in an effort to identify loci in Salmonella typhimurium which are transcriptionally regulated by nutrient-starvation conditions .
We identified lacZ operon fusions in eight genetic loci , all of which exhibited increased transcription when starved for two or more of the following nutrients : nicotinate , phosphate , ammonium , glucose , and sulfqte .
The loci have been designated stiA to stiH for starvation-inducible loci .
Mutations in two sti loci ( stiC and stiD ) significantly decreased cell viability during prolonged periods of nicotinate starvation .
sti4 and stiD are linked and map at 30 min .
The stiC , stiE , stiG , and stiH loci mapped at approximately 77 , 43 , 88 , and 56 min , respectively , on the S. typhimurium linkage map .
In nature , Salmonella typhimurium can encounter a variety of environmental conditions .
Some of these conditions permit optimal or nearly optimal growth , whereas others can cause mild-to-severe metabolic physiological stress , creating feast-or-famine-like situations .
It is reasonable to assume that physiological mechanisms have evolved to enable bacteria to survive periods of suboptimal growth-conditions or physiological stress .
Several different genetic systems which respond to various environmental stresses have been studied in Escherichia coli , including the heat-shock regulon ( 16 ) , the SOS regulon ( 27 ) , and the phosphate-starvation stimulon ( 28-30 ) .
The phosphate-starvation stimulon of E. coli appears to be analogous to the system described in this communication for S. typhimurium .
Wanner and McSharry ( 30 ) identified several genes regulated under phosphate-starvation conditions , which were designated psi for phosphate-starvation induc-ible .
The psi loci were classified into different genetic and physiological types on the basis of the extent of their regulation by known pho regulatory genes ( phoM , phoR , and phoB ) ( 29 , 30 ) .
Since glucose , ammonium , and phosphate are important for NAD biosynthesis and because NAD ( P ) participates in hundreds of enzymatic reactions ( both anabolic and cata-bolic ) , it was reasoned that nicotinate ( NA ) starvation might result in numerous stress signals which , in turn , would trigger the induction of many genes associated with the maintenance or restoration of balanced growth .
We report the identification of a system in S. typhimurium which appears to be analogous to the psi system of E. coli ( 30 ) but in which the initial selection was for NA starvation-inducible genetic loci .
We identified and characterized , using Mu ddirected lac operon fusion construction ( 1 , 2 ) , eight loci t Present Department of Microbiology , University of Austin , Austin , TX Texas 78712-1095 .
at t Present address : Department of Health and Mental Hygiene , Baltimore , MD 21201 .
§ Present address : Department of Microbiology and Immunology , University of South Alabama College of Medicine , Mobile , AL 36688 .
which respond to starvation of two or more of the following : NA , phosphate , ammonium , glucose , and sulfate .
We refer to the genes which make up this stimulon as sti for starvation inducible .
The locus stiA was originally reported as sinA ( 6 ) .
The genetic designation has been changed to avoid potential confusion with a previously used mnemonic .
MATERIALS AND METHODS Bacterial strains , phage , and transductions .
The strains used in this study were all derivatives of S. typhimurium LT-2 and are listed in Table 1 .
Transductions were performed by using a high-transducing derivative of S. typhi-murium bacteriophage P22 , P22 HT 10511-int ( 3 ) .
Culture media and chemicals .
The following minimal basal media were used for starvation studies , depending on the growth-condition being tested .
These included minimal E medium ( 26 ) and modifications of minimal MOPS ( morpholinepropanesulfonic acid ) and minimal M9 media ( 15 ) .
The medium used for carbon source utilization studies was NCE medium ( 5 ) .
The complex media used were LB-medium ( 5 ) , NB medium , MacConkey agar medium , and MacConkey base medium .
Antibiotics were present at the following amounts per milliliter : 30 , ug of ampicillin , 10 , ug of tetracycline , 50 Rg of kanamycin , 1 mg of streptomycin .
5-Bromo-4-chloro-3-indolyl - , - D-galactopyranoside ( X-Gal ) was used at a final concentration of 40 , ug/ml .
Isolation of sti-lac fusion strains of S. typhimurium .
An appropriate ampicillin-sensitive ( Aps ) strain was transduced with an HT lysate propagated on a strain lysogenic for either Mu dl ( 1 , 2 ) or Mu dl-8 ( 11 ) as described by Holley and Foster ( 9 ) .
Initial selection was for the transfer of Apr on LB-medium containing ampicillin .
Apr transductants were then replicated onto minimal E or MOPS medium supplemented with 0.4 % glucose , 40 p.g of X-Gal per ml , ampicillin , and either nonlimiting nicotinate ( NA ; 100 F.M ) or limiting NA ( 0.5 to 1.0 , uM ) .
Colonies which appeared to be a more intense blue on limiting NA medium than on nonlimiting NA medium were selected and purified for further testing of the starvation-inducible character .
Derivatives which lacked P22 prophage were identified with the green medium of Chan et al. ( 3 ) and checked for sensitivity to P22 H5 phage .
To test the effects of NA , phosphate , ammonium , glucose , and sulfate starvation on sti-lac fusions , each fusion strain was grown overnight in 5 ml of nonlimiting medium .
For NA , phosphate , glucose , and ammonium starvation , cells were grown overnight in minimal MOPS containing 0.4 % glucose , 13.2 mM phosphate , 9.5 mM ammonium , 280 , uM sulfate , and 100 p.M NA ( all nonlimiting concentrations ) .
Before sulfate starvation , each strain was grown overnight in minimal M9 medium containing 0.4 % glucose , 64 mM phosphate , 19 mM ammonium , 100 puM NA , and 280 puM sulfate ( all nonlimiting concentrations ) .
Subsequently , cultures were washed in 10 ml of MOPS buffer and suspended in 5 ml of the same .
Washed cells were then grown under nonlimiting and limiting ( starvation ) growth-conditions .
Cells were inoculated into nonlimiting MOPS or MOPS supplemented with either 0.5 , uM NA for NA starvation or 0.132 mM phosphate for phosphate limitation .
For sulfate starvation , cells were inoculated into nonlimiting M9 or M9 containing 2.8 , uM sulfate ( limiting ) and 8 mM L-methionine .
Methionine was added to decrease the severity of this starvation and permit some growth .
Ammonium and glucose-starvation followed growth under nonlimiting conditions ( MOPS medium ) to a cell density of 2 x 108 to 3 x 108 cells per ml .
Cells were collected by centrifugation and suspended in MOPS containing either 0.08 % glucose for glucose limitation or 0.6 mM ammonium plus 8 mM L-arginine for ammonium limitation ( 28 ) .
Pyrimidine , purine , amino-acid , and thiamine starvation required uracil , adenine , isoleucine-valine - , or thiamine-requiring derivatives of sti-lac strains .
Starvation for these nutrients was as described previously ( 6 ) .
Growth in all cases was monitored as a function of the increase in optical density at 600 nm .
, - Galactosidase activities were assayed as described by Miller ( 14 ) .
Cell viability over prolonged starvation .
The effects of sti mutations on long-term viability were determined by using appropriate parent strains and sti deletion or stabilized Mu insertion mutants .
sti deletion mutations were constructed as Ts ' Aps Lac-strains by the method of Maloy and Roth ( 13 ) .
Stabilized Mu dl-8 insertions were constructed by transferring the Mu dl-8 insertion into a nonsuppressor background via P22 HT transduction .
Samples ( 2 ml ) of overnight cultures ( minimal E ) were inoculated into 100 ml of limiting minimal E glucose containing 1 , uM NA and grown at 30 °C with aeration for up to 21 days .
At designated time intervals , cell densities were re-corded and 10 - , ul samples of each culture were removed and used to make serial 10-fold dilutions .
Samples of appropriate dilutions were plated on LB agar and incubated overnight at 30 °C .
The number of colonies per plate was used to estimate viable counts for each culture .
Data were then plotted as the log % of the maximum viability determined for each culture versus time ( 18 ) .
Construction of TnlO insertions near sti loci .
TnWO insertions near sti + loci were identified by crossing sti : : Mu d fusion strains with an HT lysate propagated on a pool of cells containing random chromosomal TnWO insertions ( 5 ) .
Tcr transductants were replicated to nonlimiting MOPS glucose plates containing 0.5 pLM NA and tetracycline .
Colonies which appeared white , indicating repair of a Mu d-lac insertion , were restreaked and purified for further study .
Mapping of various sti loci on the S. typhimurium genetic map .
Three methods were used to map the sti loci on the S. typhimurium linkage map ( 20 ) .
The first involved transducing a known Hfr strain ( 19 ) to Tcr with an HT lysate grown on a strain containing a TnWO insertion near a particular sti locus ( see above ) .
Subsequent Tcr Hfr strains were purified and tested for the ability to repair known auxo-trophic markers and to transfer Tcr to a known F-Smr recipient .
Tcr Hfr strains exhibiting a good ability to transfer desired markers were used to map the TnWO insertion , as well as to repair known auxotrophic mutations via interrupted mating experiments ( 19 ) .
The second mapping method was that of TnJO-directed Hfr formation as described by Chumley et al. ( 4 ) .
For this , an F'ts lac TnJO plasmid was transferred via conjugation to a strain containing a TnJO insertion near a specific sti + locus .
Temperature-resistant Lac ' derivatives were isolated and used as Hfr donor strains .
Alternatively , lac-directed Hfr formation was used to map some loci ( 13 ) .
This method is similar to the TnJO-directed Hfr formation method .
However , instead of using the TnWO insertions present on the chromosome and plasmid as areas of homology , the lac sequences of a stabilized Mu d fusion ( Mu dl-8 fusions in nonsuppressive backgrounds ) and the lac region on the donor plasmid were used as areas of homology in the formation of subsequent ts + Tcr Hfr strains .
RESULTS Identification of starvation-inducible loci .
NA-requiring strains ( JF235 , JF560 , and JF715 ) were transduced with HT phage lysates propagated on either a Mu dl or a Mu d1-8 ( Apr lac ) lysogen of S. typhimurium as described in Materials and Methods .
Initial screening was for the transfer of Apr on LB-ampicillin medium .
Approximately 40,000 Apr colonies were screened for increased 0-galactosidase production under NA starvation with X-Gal as an indicator of P-galacto-sidase activity .
Colonies which appeared more intensely blue on minimal E glucose medium containing a limiting concentration of NA than on the same media containing a nonlimiting concentration of NA were selected for further study .
All potential starvation-inducible ( sti ) operon fusions were then screened in-vitro for NA starvation induction .
Characterization of sti loci based on expressions under various starvation conditions .
Starvation-inducible fusion strains were subsequently characterized as to their expression under several starvation conditions including NA , phosphate , ammonium , sulfate , glucose , amino-acid , purine , pyrimidine , and thiamine starvation .
Table 2 presents the base,-galactosidase activities under nonlimiting conditions and the fold inductions obtained under NA , phosphate , ammonium , glucose , and sulfate limitation .
Table 2 also presents the results of isoleucine-valine , adenine , uracil , and thiamine limitation of the appropriate derivative strains .
The values represent the average activities calculated from at least three separate experiments run under similar conditions .
The stiB-lac fusion ( JF450 ) was induced under NA , phosphate , and glucose-starvation to various degrees but was unaffected by ammonium or sulfate limitation .
Induction ranged from approximately sevenfold under NA starvation to less than fourfold under glucose-starvation .
The stiC-lac ( JF519 ) fusion was induced under NA , phosphate , ammonium , and glucose , but not sulfate , limitation to various degrees , ranging from 13-fold induction under phosphate and NA starvation to 3.3-fold induction under glucose-starvation .
The stiD-lac ( JF595 ) fusion exhibited significant induction only under NA and phosphate-starvation , showing consistent and reproducible induction ranging from 1.5-to 2-fold under both of these conditions .
The stiE-lac ( JF759 ) fusion was significantly induced under NA , phosphate , and ammonium limitation but not under glucose or sulfate starvation .
Its induction ranged from approximately 2.6-fold under NA and phosphate-starvation to about 1.9-fold under ammonium starvation .
The stiG-lac ( JF1074 ) fusion was consistently induced under NA , phosphate , and ammonium starvation but not under glucose or sulfate limitation .
Induction ranged from 2.8-fold under NA limitation to just over 1.6-fold under ammonium starvation .
The stiH-lac ( JF1075 ) fusion showed consistent induction during NA , phosphate , and glucose-starvation but did not show significant induction under either ammonium or sulfate limitation .
Induction ranged from 3.9-fold under NA limitation to about 1.9-fold under phosphate-starvation .
None of these fusions were induced by sulfate limitation .
One fusion strain ( JF746 ) was identified initially as sulfate starvation inducible ( fivefold ) .
stiF was also expressed during phosphate ( 2.6-fold ) and ammonium ( 1.8-fold ) limitation ( Table 2 ) .
Several of the sti-lac fusion strains were further tested for the regulatory effects of purine ( adenine ) , pyrimidine ( uracil ) , amino-acid ( isoleucine and valine ) , and thiamine limitation .
For Ile-Val starvation , an ilvA : : TnlO insertion ( SF129 ) was introduced via transduction into each of the sti-lac fusion strains to be tested so that each resulting strain required Ile-Val to grow .
For adenine and uracil starvation , purE : : TnlO ( SF127 ) and pyrD : : TnJO ( SF32 ) insertions were similarly introduced into each of the sti-lac fusion strains .
For thiamine starvation , a thi : : TnlO insertion ( SF131 ) was introduced .
The stiB-lac fusion was significantly induced only under thiamine limitation but not by any of the other secondary conditions tested , exhibiting a 1.7-fold induction , whereas the stiD-lac fusion was not significantly induced under any of the secondary starvation conditions tested .
As noted for stiA-lac ( 6 ) , both the stiC-lac and stiE-lac fusions were significantly induced under Ile-Val , adenine , and thiamine , but not uracil , starvation .
The stiC induction ranged from 2.9-fold under thiamine limitation to greater than sixfold under adenine starvation .
The stiE induction ranged from approximately 1.7-fold under both Ile-Val and adenine starvation to 2.8-fold under thiamine limitation .
Several NA starvation-inducible loci were found to be induced by NA limitation only .
These genes were associated with NAD metabolism and were described earlier ( 10 , 24 ) .
Effect of carbon source shifts on sti-lacZ expression .
Shifting cells from a preferential carbon source such as glucose to one that is more difficult to metabolize ( i.e. , an energy source downshift ) could provide some insight into the mechanisms controlling some of these genes .
Preliminary experiments on solid media suggested that stiA , stiC , stiH , and , to a lesser extent , stiG all respond to energy downshifts .
These strains were subsequently examined in a more quantitative manner ( Table 3 ) .
The data indicate that transcription of stiA , stiC , and stiH increased dramatically after shifts from glucose to acetate or succinate .
As one might expect , all three of these loci also increased P-galactosidase activity in response to glucose-starvation ( Table 2 ) .
This correlation does not hold for all of the sti genes , however .
For example , stiB , which does respond to glucose-starvation , was not noticeably affected by energy downshift .
Also , whereas stiG responded somewhat to energy downshift , it was unaffected by glucose-starvation .
Thus , whereas common signals probably occur after glucose-starvation and carbon source downshift , there must be additional signals produced which are unique to each of these stresses .
It is unlikely that catabolite or glucose repression is solely responsible for controlling the expression of these loci , since ( i ) these genes are regulated in minimal salts glucose medium , ( ii ) relA is involved in the responses of several sti loci , and ( iii ) the correlation between cAMP and classic glucose repression does not hold ( data not shown ) .
Point iii can be made by using stiA or stiC .
Data on cya : : TnJO derivatives of stiA or stiC suggest that cAMP is involved with maintaining repressed levels of expression ( nonlimiting ) .
However , growth on acetate and succinate when intracellular cAMP levels are high results not in decreased expression but in increased expression of both loci .
Effects of sti mutations on cell viability during prolonged NA starvation .
Since the sti loci respond to a variety of starvation conditions , it seemed reasonable to assume that some of these genes might play a role in maintaining cell viability during prolonged starvation .
sti deletion mutants or nonsuppressor strains containing sti : : Mu dl-8 insertions ( JF1142 , JF1145 , JF1146 , and JF1148 ) and the parent strain ( JF235 ) were inoculated into minimal E glucose containing 1 , uM NA ( limiting ) .
Samples were removed periodically and treated as described in Materials and Methods .
The results ( Fig. 1 ) indicate that at least two of the sti loci participated in maintaining cell viability , at least over the 24-day period in which the experiment was run .
Both the stiD and stiC mutations significantly affected viability over the period of the experiment ( each starvation was conducted two or three times ) .
The stiD mutation ( JF1148 ) decreased viability as a function of the log % of the maximum viability achieved for each culture , down to approximately 0.29 % as compared with 1.2 % for the wild-type culture after a 24-day period .
A stiC mutation caused an even more dramatic effect .
After 24 days , the log % of the maximum viability of the stiC deletion ( JF1145 ) culture was approximately 0.013 % , compared with 1.2 % for the wild-type cells .
Thus , stiC and , to a lesser extent , stiD both appear to be involved with maintaining cell viability over prolonged starvation periods .
The stiC-lac fusion mutation generated several additional phenotypic effects .
Loss of stiC function caused the morphology of individual colonies to become wrinkled , resulting in an almost square appearance .
This mutation also made the resulting strain very resistant to transduction by P22 HT phage as compared with its parent strain or a strain repaired for this insertion .
Thus , the function of the stiC gene product appears to be pleiotropic in nature , affecting viability , colony morphology , and P22 sensitivity .
Effects of nutrient limitations on starvation-inducible ( sti ) locus expression sti-lac P-Galactosidase Fold increase in P-galactosidase activity after starvation for '' : fusion activity ( nonlimiting ) '' NA P04 Ammonium Glucose S04 Ile-Val Uracil Adenine Thiamine stiA 25 3.9 6.3 5.4 2.1 0.7 3.6 1.0 1.7 3.8 stiB 23 7.0 5.0 1.5 3.8 0.8 1.0 1.0 1.2 1.7 stiC 7.5 12.9 13.3 7.1 3.3 0.8 3.4 1.1 6.3 3.0 stiD 97 1.6 1.9 1.1 0.9 0.4 0.4 1.0 1.0 1.2 stiE 162 2.6 2.6 1.9 1.2 0.5 1.6 0.9 1.7 2.8 stiF 29 1.4 2.7 1.8 NDC 5.0 ND ND ND ND stiG 38 3.0 2.4 1.7 1.0 0.7 ND ND ND ND stiH 26 3.9 1.9 1.3 2.8 0.6 ND ND ND ND a Starvation conditions and f-galactosidase assays were as described in Materials and Methods .
3-Galactosidase activity was measured in Miller units .
b ilvA : : TnIO derivatives were starved for Ile-Val , pyrD : : TnJO derivatives were starved for uracil , purE : : TnJO derivatives were starved for adenine , and thi : : TnJO derivatives were starved for thiamine .
c ND , Not determined TABLE 3 .
Effect of carbon source on sti-lac expression Strain , B-Galactosidase activity ina : ( fusion genotype ) 0.4 % glucose 40 mM glycerol 0.4 % acetate 0.6 % succinate JF420 ( stiA-lacZ ) JF519 ( stiC-lacZ ) JF1074 ( stiG-lacZ ) JF1075 ( stiH-lacZ ) 20.3 4.9 54.6 23.2 48.4 11.6 60.0 87.5 232.0 88.0 84.9 188.0 202.0 84.8 53.7 278.5 a Cells were grown to logarithmic phase ( optical density at 600 nm , 0.2 ) in NCE medium containing 0.4 % glucose and centrifuged , and the pellet was suspended in an equal volume of NCE medium containing the carbon source indicated .
Each culture was grown to an optical density of 0.6 to 0.7 before assay = 1 .
o 0.1 Time ( days ) FIG. 1 .
Cell viability curves for wild-type and stiB-stiE mutant strains .
Cultures were grown under limiting NA conditions as described in Materials and Methods .
Cell viability was plotted as a function of the log % of the maximum viability achieved for each culture versus time in days .
Symbols : * , JF235 ( wild type ; parent * , strain ) ; JF1142 ( AstiB ) ; 0 , JF1145 ( AstiC ) ; 0 , JF1148 ( stiD5 : : Mu dl-8 sup ' ) ; JF1146 ( AstiE ) .
x , Mapping of sti loci on the S. typhimurium linkage map .
Several of the sti loci have been approximately mapped on the S. typhimurium linkage map ( 20 ) .
The stiD , stiG , and stiH genes were all mapped by using the technique of lac-directed Hfr formation ( 13 ) , with JF991 , JF1128 , and JF1129 as recipient strains .
One subsequent Hfr , MU37 , was used to map stiD at approximately 30 min on the S. typhi-murium linkage map ( Fig. 2 ) .
Cotransduction experiments indicated that the stiDS : : Mu dl-8 fusion is 80 % cotransduc-ible with a TnJO insertion ( zde-6025 : : TnlO ) located near stiA + ( JF428 ; reference 6 ) .
The Hfr strain MU47 was used to map the stiGO060 : : Mu dl-8 fusion in the region of 83 to 90 min on the linkage map , and MU48 was used to map the stiHO061 : : Mu dl-8 fusion in the region between 52 and 60 min on the S. typhimurium linkage map ( Fig. 2 ) .
By this technique , both the stiD-lac and stiH-lac fusions determined to be transcribed in the were clockwise direction ( Fig. 2 ) .
The stiG-lac fusion was found to be transcribed in the counterclockwise direction ( Fig. 2 ) .
Construction of the lac-directed Hfr strain MU52 revealed that transcription of the stiA locus is clockwise .
The stiC locus was mapped by using the technique of TnJO-directed Hfr formation ( 4 ) .
For this , a TnJO strain ( JF918 ) containing an insertion near stiC + was isolated and used to construct the Hfr strain MU41 described in as Materials and Methods .
This Hfr strain was used to map the stiC locus in the region between 73 to 79 min the S. on typhimurium linkage map ( Fig. 2 ) .
The stiE locus was mapped by moving a TnJO insertion found to be 90 % cotransducible with stiE + ( JF788 ) to two known Hfr strains , SA722 and SA464 ( 19 ) .
Subsequent Tcr Hfr strains exhibiting good ability to transfer both known auxotrophic markers and Tcr to recipient strains were then used to map the TnJO insertion near stiE + via interrupted mating experiments as discussed in Materials and Methods .
These experiments indicate that the location of the stiE locus is in the region between 40 and 45 min on the S. typhimurium linkage map ( Fig. 2 ) .
DISCUSSION Unlike Escherichia coli , S. typhimurium is suited for long periods of nutrient deprivation such as those encountered in various environmental situations .
This communication describes a systematic study of starvation-regulated gene expression in S. typhimurium .
These studies with Mu d-lacZ operon fusion methodology , along with simultaneous studies using O'Farrell two-dimensional gel electrophoresis , help illustrate how S. typhimurium responds to various nutrient limitations ( 23 ) .
Table 4 presents an overview of the physiological and genetic data .
The sti loci were placed into five classes based on their response to the primary starvation conditions but , as is obvious after examination of the information in Table 4 , there are overlapping controls between groups .
For example , whereas the members of class I responded in an identical manner over all conditions , one member of class II responded to alternate carbon sources in a fashion more consistent with a class I gene .
Table 4 also shows that several sti loci require relA + for full induction and cya + for full repression .
The significance of this phenomenon is not apparent .
The sti genes and the psi loci ( induced only under phosphate-starvation ; 7 ) may be analogous to the phosphate-starvation-inducible ( psi ) stimulon of E. coli ( 17 , 28-31 ) .
The phosphate-starvation stimulon of E. coli was also shown to be affected by a variety of different starvation conditions , including phosphate , nitrogen ( ammonium ) , and carbon ( glucose ) starvation .
The loci were designated psi regardless of whether they were induced by other types of starvation , because their initial selection was based on phosphate-starvation .
We favor the sti designation since it reflects the general nature of the starvation response .
It is preferable to marker isno non stiH C , sA FIG. 2 .
Linkage map of S. typhimurium .
Arcs represent the approximate map positions of the gene indicated .
Arrows indicate the directions of transcription of the genes indicated .
See the text for an explanation of the mapping procedures used for individual genes .
An asterisk indicates that the map positions relative to adjacent markers is not known use designations such as psi when referring to genes which only respond to one condition .
Wanner ( 30 ) and Wanner and McSharry ( 31 ) have shown that several of the E. coli psi loci are under complete or partial control by the various pho regulatory loci ( 22 , 25 , 28 ) .
Schlesinger and Olsen ( 21 ) and Kier et al. ( 12 ) have reported that , when a plasmid-borne phoA ( alkaline phosphatase ) gene is placed into S. typhimurium , phoA is regulated by phosphate availability as in E. coli .
This suggests that a type of phosphate regulation similar to that reported in E. coli may exist in S. typhimurium .
However , no subsequent study characterizing such a system in this organism has been reported .
Consequently , we are unable to evaluate the potential role that pho regulatory loci may have in controlling sti gene expression .
The system we describe differs from the E. coli system dramatically as to the extent of induction for specific genes .
Whereas the S. typhimurium sti loci are regulated 2-to 15-fold , the psi loci of E. coli are induced 10-to 200-fold .
The difference could be that our initial selection involved NA starvation , which might uncover a set of genes which show only modest induction ratios .
However , we have also performed experiments which initially screened for phosphate-starvation-inducible genes ( 7 ) .
Several S. typhimurium psi genes were discovered , but none were affected by other forms of starvation .
Studies on the synthesis of starvation-inducible proteins as detected by two-dimensional polyacrylamide gel electrophoresis have also been conducted in our laboratory ( 23 ) .
Evidence that the fusion studies can be correlated with the earlier protein studies was found in that the SIN-8 polypeptide was missing in all of the stiA mutants tested ( data not shown ) .
Recent studies by Groat et al. ( 8 ) examined carbon-starvation proteins in E. coli with findings similar to those obtained with S. typhimurium ( 23 ) .
As proposed by Groat et al. ( 8 ) , the process of molecular realignment at the onset of starvation with both of these organisms may be analogous to what occurs during sporulation in bacilli .
This is not surprising , since the physiological needs of starving organisms are much different from those of cells in balanced growth .
It is also evident that not all sti products are essential for starvation survival , as noted for the stiB and stiE deletion mutants ( Fig. 1 ) .
pheN '' / pROA nodC paimmtpodA so0 20A ' 9 / stic yr > 60 40 relA thietosA oftasrpino h ee niae .
See th tetZ o + + + + + + NDC ¬ ¬ ¬ ¬ ¬ r ¬ ¬ + + + ¬ ¬ ND ND ND ND i + + + + + + + i + + + + stiH III stiE + + - + + ¬ + + ¬ ¬ ¬ ¬ r o + ¬ + + ¬ ¬ ¬ stiG ND ND ND ND ¬ ND ND IV stiF - + - + + + ND ND ND ND-ND ¬ ND ND V stiD + + -------- ND-a + , Detectable induction ; + + , moderate induction ; + + + , strong induction ; + + + + , very strong induction ; - , no induction .
b r , Regulator involved in maintaining repressed levels during NA excess ; i , regulator required for full induction during NA limitation ; o , no effect .
c ND , Not done .
r o The complexity of overlapping regulatory circuits which control the various sti loci is evident when the effects of different nutrient limitations on sti gene expression are noted .
The results suggest that the promoter-operator regions of these genes respond to a variety of regulatory signals .
What constitutes these regulatory signals remains unknown .
The modest response of some of the fusions could reflect increased stability of specific messages during starvation rather than a direct effect on transcription .
This would be an intriguing way for a cell to cope with starvation stress .
The elucidation of specific starvation response mechanisms will be important , not only from an evolutionary standpoint but also for gaining insight into how other cells cope with starvation .
We thank A. Matin and J. Shultz for helpful discussions and A. G. artwork .
We also the excellent secreMoat for expert appreciate tarial assistance of C. Wolfe .
This work was supported by Public Health Service grant GM34147 from the National Institutes of Health .
Transposition and fusion of the lac genes to selected promoters in Escherichia coli .
Casadaban , M. , and S. N. Cohen .
Lactase genes fused to exogenous promoters in one step using a Mu-lac bacteriophage : in-vivo probe for transcriptional control sequences .
Chan , R. K. , D. Botstein , T. W. Watanabe , and Y. Ogata .
Specialized transduction of tetracycline resistance by phage P22 in Salmonella typhimurium .
Properties of a high frequency transducing lysate .
Chumley , F. G. , R. Menzel , and J. R. Roth .
Hfr formation directed by TnIO .
Davis , R. W. , D. Botstein , and J. R. Roth .
A manual for genetic engineering : advanced bacterial genetics .
Cold Spring Harbor Laboratory , Cold Spring Harbor , N.Y. 6 .
Identification and characterization of a relA-dependent starvation-inducible locus ( sin ) in Salmonella typhimurium .
Foster , J. W. , and M. P. Spector .
Phosphate starvatio regulon of Salmonella typhimurium .
Groat , R. G. , J. E. Shultz , E. Zychlinsky , A. Bockman , and A. Matin .
Starvation proteins in Escherichia coli : kinetics of synthesis and role in starvation survival .
Holley , E. A. , and J. W. Foster .
Bacteriophage P22 as a vector for Mu mutagenesis in Salmonella typhimurium : isolation of nad-lac and pnc-lac gene fusions .
Holley , W. A. , M. P. Spector , and J. W. Foster .
Regulation of NAD biosynthesis in Salmonella typhimurium : expression of nad-lac gene fusions and identification of a nad regulatory locus .
Hughes , K. T. , and J. R. Roth .
Conditionally transposi-tion-defective derivative of Mudl ( Amp Lac ) .
Kier , L. D. , R. Weppelman , and B. N. Ames .
Regulation of two phosphatases and a cyclic phosphodiesterase of Salmonella typhimurium .
Maloy , S. R. , and J. R. Roth .
Regulation of proline utilization in Salmonella typhimurium : characterization of put : : Mu d ( Ap , lac ) operon fusions .
Miller , J. H. ( ed .
Experiments in molecular genetics .
Cold Spring Harbor Laboratory , Cold Spring Harbor , N.Y. 15 .
Neidhardt , F. C. , P. L. Bloch , and D. F. Smith .
Culture medium for enterobacteria .
Neidhardt , F. C. , R. A. Van Bogelen , and V. Vaughn .
The genetics and regulation of heat-shock proteins .
Overbeeke , N. , H. Bergmans , F. Van Mansfeld , and B. Lugtenberg .
Complete nucleotide sequence of phoE , the structural gene for the phosphate limitation inducible outer membrane pore protein of Escherichia coli K-12 .
Reeve , C. A. , A. T. Bockman , and A. Matin .
Role of protein degradation in the survival of carbon-starved Esche-richia coli and Salmonella typhimurium .
Sanderson , K. E. , H. Ross , L. Ziegler , and P. H. Mfikela .
F ' , Hfr , and F ' strains of Salmonella typhimurium and Salmo-nella abony .
Sanderson , K. E. , and J. R. Roth .
Linkage map of Salmonella typhimurium , edition VI .
Schlesinger , M. J. , and R. Olsen .
Expression and localization of Escherichia coli alkaline phosphatase synthesized in Salmonella typhimurium cytoplasm .
Shinagawa , H. , K. Makino , and A. Nakata .
Regulation of the pho regulon in Escherichia coli K-12 .
Genetic and physiological regulation of the positive regulatory gene phoB .
Spector , M. P. , Z. Aliabadi , T. Gonzalez , and J. W. Foster .
Global control in Salmonella typhimurium : two-dimensional electrophoretic analysis of starvation - , anaerobiosis - , and heat-shock-inducible proteins .
Spector , M. P. , J. Hill , E. A. Holley , and J. W. Foster .
Genetic characterization of pyridine nucleotide uptake mutants of Salmonella typhimurium .
Tommassen , J. , P. deGuess , B. Lugtenberg , J. Hackett , and P. Reeve .
Regulation of the pho regulon of Escherichia coli : cloning of the regulatory genes phoB and phoR and identification of their gene products .
Vogel , H. J. , and D. M. Bonner .
Acetyl ornithinase of Escherichia coli : partial purification and some properties .
Mutagenesis and inducible responses to deoxyribonucleic acid damage in Escherichia coli .
Overlapping and separate controls on the phosphate regulon in Escherichia coli .
Wanner , B. L. , and P. Latterell .
Mutants affected in alkaline phosphatase expression .
Evidence for multiple positive regulators of the phosphate regulon in Escherichia coli .
Wanner , B. L. , and R. McSharry .
Phosphate-controlled gene expression in Escherichia coli using Mudl-directed lacZ-fusions .
Wanner , B. L. , S. Wieder , and R. McSbarry .
Use of bacteriophage transposon Mu dl to determine the orientation for three proC-linked phosphate-starvation-inducible ( psi ) genes in Escherichia coli K-12 .