164 , No. 3 f Gene Expression in Salmonella typhimurium : n Osmotically Induced Betaine Transport System Osmoregulation o proU Encodes a JOHN CAIRNEY ,1 IAN R. BOOTH ,2 AND CHRISTOPHER F. HIGGINS ' * Department of Biochemistry , University of Dundee , Dundee DDI 4HN ,1 and Department of Microbiology , Marischal College , University of Aberdeen , Aberdeen AB9 ] AS ,2 Scotland Received 17 June 1985/Accepted 22 August 1985 Previous evidence has indicated that a gene , proU , is involved in the response of bacterial cells to growth at high-osmolarity .
Using Mu-mediated lacZ operon fusions we found that transcription of the proU gene of Salmonella typhimurium is stimulated over 100-fold in response to increases in external osmolarity .
Our evidence suggests that changes in turgor pressure are responsible for these alterations in gene expression .
Expression of proU is independent of the ompR gene , known to be involved in osmoregulation of porin expression .
Thus , there must be at least two distinct mechanisms by which external osmolarity can influence gene expression .
We show that there are relatively few genes in the cell which are under such osmotic control .
The proU gene is shown to encode a high-affinity transport system ( Km = 1.3 , M ) for the osmoprotectant betaine , which is accumulated to high concentrations in response to osmotic-stress .
Even when fully induced , this transport system is only able to function in medium of high-osmolarity .
Thus , betaine transport is regulated by osmotic pressure at two levels : the induction of expression and by modulation of activity of the transport proteins .
We have previously shown that the proP gene encodes a lower-affinity betaine transport system ( J. Cairney , I. R. Booth , and C. F. Higgins , J. Bacteriol. , 164:1218 -1223 , 1985 ) .
In proP proU strains , no saturable betaine uptake could be detected although there was a low-level nonsaturable component at high substrate concentrations .
Thus , S. typhimurium has two genetically distinct pathways for betaine uptake , a constitutive low-affinity system ( proP ) and an osmotically induced high-affinity system ( proU ) .
Enteric bacteria such as Escherichia coli and Salmonella typhimurium are able to adapt to large fluctuations in the osmolarity of the environment in which they are growing .
However , little is known about the molecular mechanisms by which this adaptation is achieved .
In E. coli , variations in osmotic pressure are known to regulate the expression of certain specific genes .
Thus , the relative expression of the two porins , OmpF and OmpC , which form diffusion channels through the outer membrane can be altered by varying the osmotic pressure of the medium in which the cells are grown ( 10-12 ) .
Expression of the kdp operon , encoding the Kdp potassium transport system , also depends upon osmotic pressure .
Regulation of both porn and kdp expression is at the transcriptional level .
Synthesis of membrane-derived oligosaccharides , which are believed to maintain an osmotic balance in the periplasmic space , is reduced by osmotic pressure ( 15 ) .
In this case , however , osmotic regulation is probably not transcriptional .
No other genes whose expression is clearly dependent upon medium osmolarity have yet been identified .
It has recently been demonstrated that E. coli , S. typhimurium , and related enterobacteria can accumulate proline or betaine or both to high intracellular concentrations in response to high osmotic pressure ( 1 , 5 , 18 , 23 ) .
Proline and betaine are known to serve as osmoprotectants in many diverse species , including higher plants .
Accumulation of these osmoprotectants not only restores turgor pressure across the cell membrane but can also protect enzymes from inactivation at high ionic strength ( 19 , 20 , 28 ) .
Although the proline transport systems of S. typhimurium , and to a lesser extent those of E. coli , have been characterized in some detail , little is known about the pathways for betaine uptake or of the mechanisms by which transport of these compounds is regulated in response to changing osmotic pressure .
Three genes have been implicated in proline uptake by S. typhimurium .
The major proline permease , PP-I , is encoded by the putP gene .
In putP strains , some proline uptake still occurs via the minor permease , PP-II ; this residual uptake component is eliminated by mutations in the proP gene ( 21 , 24 ) .
In the accompanying paper , we have shown that PP-II transports betaine as well as proline ( 1 ) .
Indeed , the kinetics of betaine uptake via PP-Il suggest that betaine , rather than proline , is the primary physiological substrate for this tranls-port system .
Thus , PP-II plays an important role in .
osmoprotection by both proline and betaine ( 1 ) .
The third gene implicated in proline uptake is proU .
In putP proP strains , the toxic proline analog L-azetidine-2-carboxylic acid ( AC ) can only enter the cell and exert its toxic effects when cells are grown at high osmotic pressure .
Mutants selected as resistant to this analog at high osmotic pressure have a defect in a gene , proU , which has been mapped to 58 min on the S. typhimurium chromosome ( 6 ) .
It has been suggested that proU encodes a third , osmotically induced proline transport system , PP-Ill ( 6 ) .
However , proline uptake via PP-III has not been demonstrated directly , and the mechanism by which transport is increased at high-osmolarity is unknown .
In this paper we show that proline uptake via PP-Ill is negligible , at least when compared with uptake via the other proline permeases .
Rather , proU encodes a high-affinity transport system with betaine as a major substrate .
S. Thus , typhimurium possesses two genetically distinct betaine transport systems encoded by proP and pro U. Expression of the proU gene is regulated at the transcriptional level , increasing over 100-fold with increasing osmotic pressure .
The osmotic induction of proU expression is independent of the ompB locus , which is involved in the osmotic regulation of porn expression , implying the existence of at least two independent pathways for the osmoregulation of gene expression .
122 J. R. Roth J. R. Roth J. R. Roth J. R. Roth J. R. Roth J. R. Roth J. R. Roth K. Hughes ( 13 ) K. Hughes ( 13 ) K. E. Sanderson a All strains are derivatives of S. typhimurium LT2 .
b The ompRlOOl : : TnS mutation was originally isolated as the tripeptide permease regulatory mutant tppA66 : : TnS ( 9 ; Gibson and Higgins , submitted ) .
MATERIALS AND METHODS Bacterial strains and media .
Table 1 lists the genotypes and construction of the bacterial strains used .
All strains are derivatives of S. typhimurium LT2 .
Cells were grown in LB-medium ( 22 ) at 30 °C ( to prevent induction of Mu dl lysogens ) with aeration unless otherwise stated .
LC medium is LB-medium containing 2 mM CaCl2 , 0.1 % glucose , 0.001 % thymidine , and 10 mM MgSO4 .
MacConkey plates were used as described by Miller ( 22 ) .
Minimal glucose agar plates were based on the E medium of Vogel and Bonner as described by Roth ( 25 ) .
Tetracycline , carbenicillin ( an analog of ampicillin ; designated as Ap ) , and kanamycin were used at 15 , 50 , and 50 , ug ml - ' , respectively .
LOM is a low-osmolarity minimal salts medium described previously ( 1 ) .
Transport of [ 14C ] proline or [ ' 4C ] betaine was assayed as described previously ( 1 , 2 ) .
For growth at low-osmolarity , LOM-glucose was used .
To induce cells for proU expression at high-osmolarity , cells were grown to an optical density of 0.5 at 600 nm in LOM-glucose , NaCl was added to a final concentration of 0.3 M , and the cells were grown for a further hour .
Transport assays were carried out at low-osmolarity ( LOM-glucose ) or at high-osmolarity ( LOM-glucose containing 0.3 M NaCl ) as described previously .
Because of the variations in the volume of cells grown at different osmolarities , the amount of protein in cell extracts was determined directly , and transport rates are expressed as nanomoles per minute per milligram of protein .
Protein levels in sonicated cell extracts were determined by the method of Bradford , using bovine serum albumin as a standard .
P-Galactosidase activity in whole-cells was determined as described by Miller ( 22 ) .
Cells were permeabilized by the chloroform-sodium dodecyl sulfate procedure .
Roth J. R. Roth J. R. Roth J. R. Roth J. R. Roth J. R. Roth J. R. Roth J. R. Roth J. R. Roth K. Hughes ( 13 ) K. Hughes ( 13 ) K. E. Sanderson a All strains are derivatives of S. typhimurium LT2 .
b The ompRlOOl : : TnS mutation was originally isolated as the tripeptide permease regulatory mutant tppA66 : : TnS ( 9 ; Gibson and Higgins , submitted ) .
Isolation of Mu dl and Mu dl-3 insertions .
Phage Mu dl ( Apr lac cts62 ) can be used to isolate operon fusions , placing lacZ under control of any chromosomal proihoter ( 3 ) .
Because Mu is normally unable to infect S. typhimurium , we used a Mu-Pl hybrid helper phage which confers P1 host range and thus allows infection of galE strains of S. typhimurium ( 7 ) .
Mu dl lysates prepared by induction were of the lysogenic strain JL3473 as described previously ( 1 ) .
At least 104 independent Ampr colonies containing Mu dl insertions randomly distributed around the chromosome were pooled and washed twice in minimal-medium before spreading appropriate on selective plates .
Because of the potential instability of Mu dl , strains containing Mu dl insertions were tested genetically before each set of experiments to ensure that transposition had not occurred ; instability was never observed .
Because of zygotic induction , Mu dl-mediated lac fusions can not easily be transduced between strains without transposition occurring .
Thus , for some experiments a recently constructed , conditionally transposition-defective ( and therefore more stable ) derivative of Mu dl was used ( 13 ) .
This phage , Mu dl-8 , contains an amber mutation in the transposase gene and is stable unless in a sup genetic background .
Random Mu dl-8 insertions into the S. typhimurium chromosome were obtained by P22-mediated transduction of the Mu derivative from strain TT7674 ( pncA2J2 : : Mu dl-8 ) into TT7610 ( supD ) , selecting for Ampr transductants ( 13 ) .
Mu dl-8 fusions initially selected and isolated in the sup strain ( TT7610 ) were transduced into an appropriate sup ' background before further characterization .
Isolation of TnlO insertions .
Random TnWO insertions in the S. typhimurium chromosome were isolated by using the phage P22-TnJO system ( 16 ) as described previously ( 2 ) .
Once selected , TnlO insertions were transduced into a `` clean '' genetic background to preclude the possibility of any secondary mutations having arisen during the construction of the TnIO insertion .
Transductions were carried out with a high-transducing derivative of phage P22 int4 as described by Roth ( 25 ) .
galE strains were grown in LB-medium supplemented with 0.2 % glucose and 0.2 % galactose when used as donors or recipients for transduction , to ensure efficient synthesis of the P22 receptors .
After transduction of TnS , TnJO , or Mu dl-8 insertions from one strain to another , the correct location of the transposon and the presence of just a single insertion in the transductant were verified phenotypically and genetically by marker rescue .
All strains used for transport and expression studies are derivatives of CH223 and are therefore isogenic except for the introduced mutations .
Conjugations were carried out as described previously ( 9 ) , using streptomycin to select against the donor HFr .
Prototrophic conjugants were purified and tested on minimal glucose-tetracycline plates to determine the percent coinheritance of the proU1697 : : TnlO marker .
Identification of genotypes and phenotypes .
Mutations in the proline transport genes were identified by their patterns of resistance and sensitivity to the two toxic proline analogs AC and 3,4-dehydro-DL-proline ( DHP ) as described previously ( 1 , 2 ) ( Table 2 ) .
AC ( 1 mM ) AC AC DHP ( 40 ptg ( 80 F.g ml ) ml - ) ( 40 Lg in the Genotype ie ten presence NaCI presence of0 .3 M NaCi s s s s s r putP + proP + proU + putP proP + proU + putP proP proU ' putP proP proU r s s r r s r r r r RESULTS Isolation of TnlO and Mu dl insertions in proU .
Mutations in proU can be selected in a putP proP background by their aResistance ( r ) and sensitivity ( s ) were determined on minimal glucose plates as described in Materials and Methods .
resistance to 1 mM AC on minimal glucose plates containing 0.3 M NaCl ( 6 ) .
Thus , a pool of approximately 10,000 random and independent TnWO insertions into the chromosome of strain CH638 ( AputPA proP : : Tn5 ) was plated on minimal glucose plates containing tetracycline , 0.3 M NaCl , and 1 mM AC .
which appeared after 48 h were purified and characterized further .
To demonstrate that each Tetr ACr derivative contained just a single transposon insertion and that this transposon was responsible for the ACr phenotype , the transposon was transduced back into CH638 , and each Tetr transductant was tested for resistance to AC at high osmotic pressure .
All ( 100 % [ 24/24 ] ) of such transductants were ACr .
One of these transductants , CH710 ( proUJ697 : : TnIO ) , was used for all further experiments described here .
However , six other independently derived proU : : TnIO insertions ( CH709 to CH714 ) showed identical phenotypes .
Mu dl-mediated lacZ-fusions to proU were isolated from a pool of random insertions in strain CH638 using a selection procedure identical to that described above .
Because of zygotic induction , Mu dl insertions can not be transduced into a Mu-free genetic background .
Thus , to demonstrate that each strain carried just a single Mu dl insertion and that this was in the proU gene , each Ampr ACr derivative was transduced to Tetr with a P22 lysate grown on strain CH710 ( proUJ697 : : TnlO ) .
Most ( > 90 % ) of the Tetr derivatives became Amps Lac - , showing that each strain contains just a single Mu dl insertion which is very closely linked to proU : : TnJO .
Each of six independently isolated TnWO and seven independent Mu dl insertions in proU ( CH639 to CH645 ) were mapped against each other in this manner , and all were found to be closely linked by transduction ( > 90 % ) .
Thus , all ACr mutations selected by this procedure were located at a single chromosomal locus , although this locus may , of course , consist of more than one gene .
Map location of TnlO and Mu dl insertions .
Csonka ( 6 ) showed that proU mutations , selected in a manner similar to that described above , are located at 58 min on the S. typhimurium chromosome .
To establish that we had selected mutations at the same locus , the proU : : TnlO insertion from CH710 was transferred into various HFr strains , and the frequency of cotransfer of Tetr with a number of auxotrophic markers was scored .
These data ( Table 3 ) show that the transposon in strain CH710 is closely linked to cysC , the same map location as that described by Csonka ( 6 ) .
Since we used a selection similar to that described by Csonka , and since 14 independent TnWO or Mu dl insertions were all located at the same position on the chromosome , it seems very probable that mutations at only a single locus ( whic may , of course , consist of more than one gene ) can give rise to the ProU phenotype .
Osmoregulation of proU expression .
It has been suggested that proU might encode an osmotically induced proline transport system ( 6 ) .
To determine whether this is indeed the case and whether such regulation is transcriptional , we assayed 3-galactosidase expression from Mu dl-mediated lacZ-fusions in which P-galactosidase expression is under control of the proU promoter .
Figure 1 shows the expression of 3-galactosidase as a function of salt concentration .
Clearly , expression from the proU promoter increased mark-edly with increasing salt concentration .
This is an osmotic effect , rather than a specific effect of NaCl , as essentially identical results were obtained if choline-chloride , KCI , or sucrose was used in place of NaCl ( Table 4 ) .
As controls , we have shown that expression of Mu dl ( Apr lac ) fusions to a number of other promoters ( tppB , ompR , oppA , pepT ) do not vary with osmotic pressure ( M. M. Gibson and C. F. Higgins , submitted for publication ; unpublished data ) .
Interestingly , there is a threshold osmolarity below which proU induction does not occur .
Thus , the cell is able to accommodate moderate increases in osmolarity ; only under more severe osmotic-stress is the expression of proU induced .
In contrast to other solutes , glycerol did not stimulate proU expression .
Glycerol is known to be freely permeable across the cell membrane and hence would not be expected to be osmotically active .
This implies that turgor pressure differences across the membrane , rather than solute concentration per se , are responsible for the induction of pro U. Figure 2 shows the kinetics of induction of proU expression .
There was a slight lag of about 10 min immediately after an increase in osmotic pressure .
This was presumably due to a transient inhibition of protein synthesis caused by the osmotic shock ( 13 ) .
Subsequently , proU expression increased rapidly , reaching a maximum of about 40 min after the osmolarity of the medium had been increased .
The absolute level of expression achieved is dependent upon the magnitude of the osmotic upshock .
Interestingly , betaine , and to a lesser extent proline , both substantially reduced the expression of proU at high-osmolarity ( Table 4 ) .
No other protein amino-acid had this effect ( data not shown ) .
This is almost certainly an indirect effect due to the restoration of turgor pressure which results from the accumulation of these two amino-acids in response to osmotic-stress .
marker location ( min ) TR5656 proA36 7 0 TR5657 purE8 11 2 TR5658 purC7 23 12 TR5660 pyrF146 33 17 TR5663 purF145 47 17 TR5665 cysC519 60 73 TR5666 serA13 62 41 TR5667 cysG639 72 21 TR5668 cysE396 79 18 a The HFr strain SA722 was transduced to proU : : TnJO ( with CH710 as transduction donor ) , and this strain was used as the conjugation donor with each of the recipients listed above .
Prototrophic recombinants ( 100 of each ) were scored for Tetr to determine the percent cotransfer of proU : : TnJO with each of the auxotrophic markers .
Two other HFr 's with different points of origin were also used and gave essentially identical results .
r s s r r s r r r r The ompR locus is not required for osmoregulation of proU expression .
The ompR and envZ genes are involved in the osmotic pressure-dependent expression of the ompF and ompC genes ( 10-12 ) .
To ascertain whether either of these genes is required for the osmoregulation of proU expression , a Tn5 insertion in ompR , which is polar on envZ , was introduced into CH946 which carries a proU : : IacZ operon fusion .
The resultant strain ( CH831 ) was shown to be OmpR-by its resistance to the toxic peptide alafosfalin ( 9 ; Gibson and Higgins , submitted ) and to be OmpF-OmpCby examination of the outer membrane proteins by polyacrylamide gel electrophoresis ( data not shown ) .
The ompR mutation ompR1001 : : Tn5 was originally isolated as being tripeptide transport defective ( tpp ) and only subsequently was shown to be the same gene as ompR ( 9 ; Gibson and Higgins , submitted ) .
Eliminating ompR and envZ function had no effect on the expression or osmoregulation of proU ( Table 4 ) .
Mutations in the other proline transport genes , putPA and proP , also had no effect on proU expression ( Table 4 ) .
Although the data shown here are only for one proU : : lacZ strain , identical results were obtained for strains CH832 and CH834 .
Are other genes regulated by osmolarity in a manner similar to proU ?
To estimate the number of genes which are regulated by osmotic pressure in a manner similar to proU , we again took advantage of lacZ-fusions .
lacZ-fusions to proU are uninduced and therefore grow as white colonies on MacConkey-lactose plates .
However , if 0.3 M NaCl is added to the plate proU expression is induced , , B-galactosidase is produced by the fusion strains , and the colonies are red .
We therefore used this screen to isolate lacZ-fusions to any chromosomal gene which is osmotically induced .
A pool of 16,000 random and independent Mu dl-8-mediated lacZ-fusions to the S. typhimurium chromosome was made in strain TT7610 ( supD ) as described in Materials and Methods .
The Mu dl-8 fusions were plated onto MacConkey-lactos plates and subsequently replicated onto MacConkey-lactose plates containing 0.3 M NaCl .
Colonies which were white in the absence of salt but red on plates to which salt had been added were purified and characterized further .
Six independent osmotically induced fusions ( each isolated from a separate pool of insertions ) were obtained by this method ( CH826 to CH829 , CH946 , CH947 ) .
Unexpectedly , all six were found to be fusions to the proU gene by two criteria : each was transductionally linked ( 90 % ) to the proU : : TnJO of CH710 and each conferred ACr on high-salt plates when introduced by transduction into strain CH638 ( putP proP ) .
Each of the proU : : Mu dl-8 fusions obtained by this procedure was also regulated in a manner similar to the pro U : : Mu dl fusions described above .
Thus , it seems that there are few , if any , other genes whose expression is regulated by osmotic pressure in a manner similar to that of pro U. Proline uptake through proU .
Mutations in proU were originally isolated on the basis of resistance to the proline analog AC ( 6 ) .
In addition , proU mutations reduced the ability of proline auxotrophs to utilize exogenous proline to support growth .
Thus , it was suggested that proU encodes a third transport system for proline , PP-III .
However , proline uptake via PP-III was never measured directly , so its real significance could not be assessed .
We set out to measure proline uptake through PP-III and to determine whether uptake is increased by osmotic pressure as would be predicted from the expression data ( see above ) .
Figure 3 shows the results of such an experiment .
In wild-type cells ( LT2 ) grown at low-osmolarity , rapid proline uptake was observed .
When a putP mutation was introduced ( CH627 ) eliminating PP-I , uptake was severely reduced .
The residual uptake was completely eliminated by introducing a mutation in proP ( strain CH638 ) .
Thus , strains lacking both PP-I and PP-II show no detectable proline uptake .
Even at substrate concentrations of up to 300 p.M proline we were unable to detect any proline uptake by a putP proP strain , showing that PP-III plays no significant role in proline uptake , at least under these conditions .
Because proU expression is induced by growth at high osmotic pressure , we also determined proline uptake for cells grown under such conditions ( Fig. 3 ) .
Cells of strain CH627 ( putP proP + proU + ) were grown in medium of high-osmolarity ( LOM-glucose containing 0.3 M NaCl ) , and proline uptake was assayed at both high and low-osmolarity .
As has been demonstrated previously , proline uptake by CH627 was stimulated about threefold by growth at high osmotic pressure ( 1 ) .
However , all proline uptake under these conditions was via PP-II and was completely abolished by mutations in proP .
Even at 300 p.M proline proU-dependent proline uptake could not be detected ( data not shown ) .
Thus , we were completely unable to demonstrate significant proline transport through PP-III , even in cells grown under conditions in which expression of proU is fully induced .
Proline is therefore unlikely to be the primary substrate for PP-III .
proU encodes a betaine transport system .
Because expression of proU is osmotically induced , it seemed reasonable to suppose that the gene encodes a transport system for a solute required at high osmotic pressure .
Betaine seemed a possible substrate , both because it is accumulated by cells at high osmotic pressure ( 1 , 23 ) and also because , as an N-substituted amino-acid , it bears at least some structural and resemblance to both proline AC .
We have shown in the accompanying paper that the proP gene of S. typhimurium encodes a transport system with betaine as its primary substrate but which also shows affinity for proline ( 1 ) .
We also found that cells have an osmotically induced betaine uptake component which is independent of proP .
To deter mine whether this transport component is dependent upon proU , we measured betaine uptake in strains variously mutated for proP and proU ( Fig. 4 ) .
Cells of CH627 ( proP + proU + ) , CH638 ( proP proU + ) , and CH640 ( proP proU ) were grown in medium of high-osmolarity ( LOM-glucose containing 0.3 M NaCl ) to ensure full induction of proP and proU expression , and betaine uptake was measured in medium of high-osmolarity .
Clearly , strains mutated for proP had reduced rates of betaine uptake ; the remaining uptake component was almost completely abolished by mutations in proU .
Thus , proP and proU encode genetically independent betaine transport systems which , between them , are responsible for essentially all detectable betaine uptake .
At high substrate concentrations ( 300 , uM ) a low rate of residual betaine uptake was detectable in proP proU strains ( data not shown ) .
However , this uptake component only becomes significant at higher substrate concentrations and probably represents a nonsaturable uptake pathway .
The above experiments were performed with cells grown at high-osmolarity to induce proU expression .
When cells of CH627 ( proP + proU + ) were grown at low-osmolarity but transport was still assayed at high-osmolarity , betaine uptake was considerably reduced ( Fig. 4 ) .
Betaine uptake by cells grown under these conditions was wholly dependent on the proP-encoded uptake system as no uptake was detected for strain CH638 ( proP proU + ) .
This is in agreement with the fact that the proU gene is not expressed under such growth-conditions .
Transport assays on induced cells were also carried out in medium of low-osmolarity to determine the effects of osmolarity on the transport process itself ( as distinct from induction ) .
Regardless of whether cells of CH638 ( proP proU + ) were grown at high or at low-osmolarity , betaine uptake could not be detected when transport was assayed at low-osmolarity .
Thus , proU-dependent betaine uptake is regulated at two levels : the proU gene is only expressed at high-osmolarity , and in addition , once induced , the transport system will only function if the cell is under osmotic-stress .
Kinetics of betaine uptake .
To determine the kinetic constants of proU-dependent betaine uptake , we measured transport in strain CH638 ( putP proP ) at a variety of betaine concentrations ( Fig. 5 ) .
Clearly , the ProU system has high affinity for betaine .
Because of the low specific activity of the [ 14C ] betaine it was not possible to measure transport accurately at concentrations below 1 , uM .
However , when the minor , pro U-independent uptake component was subtracted and kinetic constants were calculated from Eadie-Hoffstee or Lineweaver-Burk transformations of the data in Fig. 5 , the following kinetic paramaters for pro U-dependent betaine uptake were obtained : Km = 1.3 , uM ; Vmax = 12.5 nmol min - ' mg of protein - ' .
500 S Soo 0 400 ' U 4300 u-6 a-200 bc 0 a ¬ 0 05 1 0-2 0 3 04 NaCl concentration ( M ) FIG. 1 .
Osmotic induction of proU expression .
Strain CH946 ( proU : : Mu dl-8 ) was grown in LOM-glucose containing the indiconcentrations cated of NaCl .
Cells were grown to an optical density at 600 nm of about 0.5 , and 3-galactosidase activity was assayed .
f-Galactosidase units were calculated and expressed by the method of Miller ( 22 ) , measuring the cellular protein content directly rather than using turbidity to estimate the protein content of each cell culture .
Essentially identical results were also obtained for CH640 ( proU : : Mu dl ) .
Mapping of proUa Map Auxotrophic Recipient Cotransfer TABLE 4 .
Regulation of proU expression ' Relevant genotype Additives to growth medium proUJ702 : : Mu dl-8 None proUl702 : : Mu dl-8 0.3 M NaCl proUJ702 : : Mu dl-8 0.3 M KCI proUJ702 : : Mu dl-8 0.3 M choline-chloride proUJ702 : : Mu dl-8 0.44 M sucrose proUJ702 : : Mu dl-8 0.5 M glycerol proUJ702 : : Mu dl-8 0.3 M NaCl + 1 mM proline proU1702 : : Mu dl-8 0.3 M NaCl + 1 mM betaine AputPA proP : : Tn5 proU : : Mu dl None AputPA proP : : TnS proU : : Mu dl 0.3 M NaCl proUJ702 : : Mu dl-8 ompRJOOl : : TnS None proUJ702 : : Mu dl-8 ompRlOOl : : TnS P-Galactosidase ( U ) 4 513-486-328-532 6 391 180 6 929 5 529 Strain CH946 CH946 CH946 CH946 CH946 CH946 CH946 CH946 CH640 CH640 CH831 CH831 a Cells were grown in LOM-glucose with additives as indicated , and , B-galactosidase was assayed .
, - Galactosidase units are expressed per milligram of protein .
U 300 ( U 0 i 200 ' 0 bc 0 - * - * .
0 0 2-0A 10 A 60 A A A 4 120 20 0 100 60 80 Time ( min ) FIG. 2 .
Induction of proU expression .
Cells of strain CH946 ( proU : : Mu dl-8 ) were grown in LOM-glucose to an optical density at 600 nm of about 0.5 .
NaCl was added to the appropriate concentration , samples were removed after various periods of induction , and,-galactosidase activity was determined .
PGalactosidase units were calculated after measuring the protein content of each cell culture directly .
Symbols : * , 0.3 M NaCl ; * , 0.16 M NaCl ; A , 0.06 M NaCl .
16 14 ~ 1 2 x T C lo0 0 .
a 0 ~ E C ¬ x O. c 0 -0 S = ai 6 x 0 41 - ¬ 0 0 / 2 0 a : ` km o LL 1 1 4 5 2 3 TIME ( mins ) DISCUSSION When subjected to osmotic-stress , bacterial cells must be able to restore turgor pressure by increasing internal osmolarity and , in addition , must protect sensitive enzymes from the potential deleterious effects of high intracellular ionic strength .
This is most frequently achieved by accumumutants .
Betaine transport by proP and proU strain CH627 ( putP [ A , A ] ) , CH638 ( putP proP [ 0 , 0 , * , EU ) , or CH640 ( putP proP proU [ V ] ) were grown in LOM-glucose in the absence ( open symbols ) or presence ( closed symbols ) of 0.3 M NaCl , and betaine transport was assayed at high ( 0 , 0 , A , A , V ) or low ( U , L ) osmolarity 2 3 TIME ( mins ) FIG. 3 .
Proline uptake through the various proline permeases .
Cells of strains LT2 ( putP ' proP + proU + [ xI ) , CH627 ( putP proP + proU + [ open symbols ] ) , or CH638 ( putP proP proU + [ closed symbols ] ) were grown in LOM-glucose in the presence or absence of 0.3 M NaCl as indicated , and proline transport at 20 , uM was assayed at high or low-osmolarity .
Symbols : x , 0 , 0 , cells grown and assayed at low-osmolarity ; A , A , cells grown at high-osmolarity and assayed at high-osmolarity ; * , O , cells grown at high-osmolarity but assayed at low-osmolarity .
A 80 [ T a , 0 cL60 A E - cS a , 0 = 240 aa , , c - A / Ai A ~ ~ ~ ~ ~ 5 1 4 2 3 TIME ( mins ) 10 15 20 25 BetB nce trati n e aine cotnoanocn ( PM ) netration FIG. 5 .
Kinetics of proU-dependent betaine uptake .
Strain CH638 ( putP proP proU + ) was grown in LOM-glucose of high-osmolarity ( 0.3 M NaCl ) to fully induce proU , and betaine transport was measured at a variety of concentrations .
The inset shows a Lineweaver-Burk transformation of these data .
At the concentrations of betaine used in these experiments any nonsaturable , proU-dependent uptake negligible .
was lating high intracellular concentrations of compatible solutes such as proline or betaine which can serve both functions ( 19 , 20 , 28 ) .
Betaine is known to play an important osmoprotective role in several bacterial species ( 14 , 18 ) and to be accumulated to high concentration by E. coli in response to osmotic-stress ( 23 ) , although little is known about the mechanisms of betaine accumulation .
In a previous paper we have shown that the proP gene of S. typhimurium encodes a betaine transport system ( 1 ) .
However , in proP strains , there is a considerable betaine uptake component which is independent of proP function .
We show here that this additional transport component is dependent upon the proU gene .
Mutations in proU were originally identified as conferring resistnce to the proline analog AC at high osmotic pressure , and the gene was inferred to encode an osmotically induced proline uptake system , PP-Ill ( 6 ) .
However , we were completely unable to detect significant rates of proline uptake through PP-III .
On the other hand , proU mutations reduced betaine transport rates considerably .
Thus , proU encodes a transport system with betaine as its primary substrate .
Presumably , the transport system lacks absolute specificity and can transport proline at low rates ( although insignificant compared with transport via PP-I and PP-II ) , which accounts for the proU-dependent growth stimulation of proline auxotrophs reported by Csonka ( 6 ) .
It should be noted that we have not formally excluded the possibility that the proU gene is actually a positive regulator of the betaine transport system , rather than encoding the transport system per se .
However , this seems unlikely for two reasons .
First , expression of a regulatory gene would not be expected to be highly regulated , and second , operon fusions indicate a high level of proU expression which is atypical of regulatory genes .
In addition , we have recently identified the proU protein , which is produced in very large amounts after osmotic induction ( unpublished data ) .
The kinetics of proU-dependent betaine transport show it to be a high-affinity transport system ( Km = 1.3 , uM ) , particularly when compared with uptake via the ProP system = 44 , uM ) .
In proP proU strains betaine uptake is essentially eliminated , although at high substrate concentrations a nonsaturable component to uptake can be detected .
Thus , there are two active betaine transport systems in S. ium .
proP encodes a low-affinity uptake system which is constitutively expressed ( although expression is slightly enhanced by increased osmotic pressure ) and which relatively broad specificity , also transporting proline and analogs DHP and AC .
proU , on the other hand , encodes a high-affinity and rather more specific transport system .
Whether these two transport systems are energized by different mechanisms , as is common for different transport systems for a single substrate , remains to be determined .
These studies were undertaken with S. typhimurium .
However , betaine can be actively accumulated by an osmotically induced transport system ( s ) in E. coli ( 23 ) .
The kinetic parame coli are very similar of the values we obtained for the ProP and ProU systems in ters for uptake by E. to the sum S. typhimurium .
In addition , a gene equivalent to proP , at least in terms of proline uptake , has been identified in E. coli ( 26 ) .
Thus , it seems likely that similar betaine uptake systems will also be present in E. coli .
We showed , using Mu dl ( Apr lac ) fusions , that transcription of the proU gene is strongly dependent upon medium osmolarity .
Expression is increased over 100-fold by increasing the osmolarity of the medium .
Induction is rapid and is independent of the particular osmolyte used .
The molecular mechanism by which changes in external osmolarity are sensed is obscure .
It has been suggested that expression of the kdp operon is regulated in response to turgor pressure rather than to external osmolarity per se ( 17 ) .
Our results strongly support this , providing evidence that turgor pressure is responsible for proU induction .
First , while many solutes induce proU expression , glycerol has no effect .
Glycerol is known to permeate freely across the membrane and therefore , unlike the other solutes used , would elicit no effect on turgor pressure .
Second , betaine and proline , but not other amino-acids , substantially reduce expression of proU at high osmotic pressure .
This is almost certainly an indirect effect .
Thus , at high-osmolarity these amino-acids are accumulated to restore turgor pressure .
If proU expression is mediated by turgor pressure , rather than by osmotic pressure per se , then such accumulation of betaine and proline would be expected to reduce expression ; this is precisely what is observed .
Third , after a shift to high-osmolarity proU expression initially increases rapidly , and the rate of synthesis is subsequently reduced as a steady state is approached .
This again is explained if turgor pressure is the regulatory signal and after osmotic shift turgor pressure is partially , but never fully , restored ( possibly by K + uptake [ 8 ] ) .
Finally , the induction profile for proU expression shows that , up to a certain threshold concentration of external solute , proU expression is not induced .
While a variety of solutes can serve as osmoprotectants at intermediate osmolarities , betaine is known to be the preferred compatible solute at high-osmolarity ( 28 ) .
It therefore seems likely that , at these relatively low osmolarities , the cell is able to adjust turgor pressure by alternative means such as increased K + uptake ( 8 ) or proline synthesis ( 5 ) .
Above a critical value , turgor pressure can no longer be maintained by these means and proU is induced .
The most reasonable mechanism by which changes in turgor pressure might be sensed is that structural changes resulting from membrane shrinkage cause a conformational change in a membrane-bound regulatory protein ( 12 , 17 ) .
proU-dependent betaine uptake is regulated at two levels 15 - -LCB- Km E E E12 ~ t him yp 1.121 a ) u r / / O. E 9 X * < h E  / a s / it s X 0 j ° 08 .6 t / / o X 04 C3 -4 -2 4 6 -.8 -6 * 2 s 8 5 30 Not only is proU transcription increased by increasing osmolarity , but even when induced , function of the transport system is only detected at high osmotic pressure .
Regulation by osmotic pressure of both transcription and function is also found for the proP betaine-proline transport system ( 1 ) .
Presumably the transport proteins only adopt an active conformation suggesting at high-osmolarity , again a rather specific regulatory role for changes in membrane conformation .
It is of interest to ascertain the number of genes which are under osmotic control and whether a single global regulatory mechanism is in operation .
Besides proU , only the ompF and ompC genes and the kdp potassium transport operon have previously been shown to be osmotically regulated ( 10-12 , 17 ) .
To identify any other genes which might be similarly regulated , we isolated six independent lacZ-fusions to the S. typhimurium chromosome whose expression was osmoregulated .
Each of these fusions was found to be in the proU gene .
It must be considered that the screen adopted here would not detect genes which , unlike proU , are expressed above a certain level even at low-osmolarity ( e.g. , ompC ) or genes whose function is essential to the cell .
Also , fusions to kdp would not have been detected since , in the medium used , K + concentrations were such that kdp expression is induced even at relatively low external osmolarity .
However , despite these reservations it seems likely that only a very limited number of genes ate regulated in response to osmotic pressure .
This is in agreement with the observation that the relative abundance of only a very few cellular proteins changes in response to increased osmolarity ( 4 ; unpublished data ) .
Only one genetic locus has been identified which plays a role in the osmoregulation of gene expression , the adjacent ompR and envZ genes which are involved in the osmoregulation of ompF and ompC porin expression ( 10-12 ) .
Interestingly , we have recently found that ompR also plays a role in the expression of genes which are not osmoregulated ( Gibson and Higgins , submitted ) .
The characteristics of proU and porin expression are quite different .
Thus , it is the ratio of the two porins that is affected by medium osmolarity , and in addition , many other environmental factors can also influence their expression .
Expression of proU , on the other hand , seems to respond solely to turgor pressure .
Expression of proU is effectively zero at low-osmolarity , only being induced above a certain threshold level .
The kdp operon seems to be regulated in a manner similar to proU ( 17 ) .
We have shown here that ompR and envZ mutations do not affect the expression or osmoregulation of proU .
Thus , there must be at least two separate mechanisms by which transcriptional regulation in response to osmotic pressure is achieved .
We thank Hugh Ainsworth for assistance with determining the map location of the proU gene .
We are grateful to Kelly Hughes and John Roth for providing us with phage Mu dl-8 before publication .
Financial support was provided by a grant from the Medical Research Council to C.F.H. and I.R.B. C.F.H. is a Lister Institute Research Fellow .
ADDENDUM Since this manuscript was submitted a paper has been published ( V. J. Dunlop and L. N. Csonka , J. Bacteriol .
163:296 -304 , 1985 ) which also demonstrates that osmotic induction of proU expression is transcriptional .
Cairney , J. , I. R. Booth , and C. F. Higgins .
Salmonella typhimurium proP gene encodes a transport system for the osmoprotectant betaine .
Cairney , J. , C. F. Higgins , and I. R. Booth .
Proline uptake through the major tranport system of Salmonella typhimurium is coupled to sodium ions .
Casadaban , M. J. , and S. N. Cohen .
Lactose genes fused to exogenous promoters in one step using a Mu-lac bacteriophage : in-vivo probe for transcriptional control sequences .
Clark , D. , and J. Parker .
Proteins induced by high osmotic pressure in Escherichia coli .
Proline over-production results in enhanced osmotolerance in Salmonella typhimurium .
A third L-proline permease in Salmonella typhimurium which functions in media of elevated osmotic strength .
Csonka , L. N. , M. M. Howe , J. L. Ingraham , L. S. Pierson , and C. C. Turnbough .
Infection of Salmonella typhimurium with coliphage Mu dl ( Apr lac ) : construction of pyr : : lac gene fusions .
Epstein , W. , and S. G. Schultz .
Cation transport in Esch-erichia coli .
Regulation of cation content .
Gibson , M. M. , M. Price , and C. F. Higgins .
Genetic characterization and molecular cloning of the tripeptide permease ( tpp ) genes of Salmonella typhimurium .
Hall , M. N. , and T. J. Silhavy .
The ompB locus and the regulation of the major outer membrane porin proteins of Escherichia coli K12 .
Hall , M. N. , and T. J. Silhavy .
Genetic analysis of the ompB locus in Escherichia coli K12 .
Hall , M. N. , and T. J. Silhavy .
Genetic analysis of the major outer membrane protein of Escherichia coli .
Hughes , K. T. , and J. R. Roth .
Conditionally transposi-tion-defective derivative of Mu dl ( Ap lac ) .
Imhoff , J. F. , and F. Rodriguez-Valera .
Betaine is the main compatible solute of halophilic eubacteria .
Osmotic regulation and the biosynthesis of membrane-derived oligosaccharides in Escherichia coli .
Kleckner , N. , R. Chen , and B.-K .
Mutagenesis by insertion of a drug-resistance element carrying an inverted repetition .
Laimins , L. A. , D. B. Rhoads , and W. Epstein .
Osmotic control of kdp operon expression in Escherichia ccli .
Le Rudulier , D. , and L. Bouillard .
Glycine betaine , an osmotic effector in Klebsiella pneumoniae and other members of the Enterobacteriaceae .
Le Rudulier , D. , A. R. Strom , A. M. Dandekar , L. T. Smith , and R. C. Valentine .
Molecular biology of osmoregulation .
Le Rudulier , D. , and R. C. Valentine .
Genetic engineering in agriculture : osmoregulation .
Menzel , R. , and J. Roth .
Identification and mapping of a second proline permease in Salmonella typhimurium .
Experiments in molecular genetics .
Cold Spring Harbor Laboratory , Cold Spring Harbor , N.Y. 23 .
Perroud , B. , and D. Le Rudulier .
Glycine betaine transport in Escherichia coli : osmotic modulation .
Ratzkin , B. , and J. Roth .
Cluster of genes controlling proline degradation in Salmonella typhimurium .
Genetic techniques in studies of bacterial metabolism .
Stalmach , M. E. , S. Gruthe , and J. M. Wood .
Two proline porters in Escherichia coli K-12 .
van Alphen , W. , and B. Lugtenberg .
Influence of osmolarity and the growth medium on the outer membrane protein pattern of Escherichia coli .
Yancey , P. H. , M. E. Clark ; S , .
C. Hand , R. D. Bowlu , and G. N. Somero .
Living with water stress : evolution of osmolyte systems .