Osmotic Regulation of L-Proline Transport in Salmonella typhimurium VIRGINIA JOAN DUNLAP AND LASZLO N. CSONKA * Department of Biological Sciences , Purdue University , West Lafayette , Indiana 47907 Received 10 December 1984/Accepted 15 April 1985 The transport of proline is important for the adaptation of Salmonella typhimurium to osmotic-stress because exogenous proline permits the growth of the organism in media of elevated osmotic strength that would otherwise be toxic .
Measurements of the rate of [ 3H ] proline transport in S. typhimurium indicated that the organism has two distinct proline permeases , the ProU and the ProP systems , whose activities increase more than fivefold as a consequence of growth in media containing 0.3 M NaCl or 0.47 M sucrose .
Transport via a third proline permease , the PutP system , is not affected by the osmotic strength of the medium .
We constructed strains that carry fusions of lacZ to proU or proP , genes that are required for the two osmotically stimulated proline transport systems .
Assays of,-galactosidase revealed that the transcription of the proU gene is increased more than 10-fold as a result of exposure of the cells to 0.3 M NaCl , 0.47 M sucrose , or equivalent concentrations of other solutes that are not freely diffusible across the cytoplasmic membrane .
Increased transcription can not be triggered by methanol , ethanol , and glycerol , substances that are freely diffusible across the membrane , suggesting that the signal for increased transcription might be an osmotic gradient across the cytoplasmic membrane .
The proP gene does not show transcriptional regulation of sufficient magnitude to account for the stimulation of [ 3H ] proline transport .
Thus , the osmotic stimulation of the ProP system might be mediated by some posttranscriptional event .
In addition to serving as a carbon or nitrogen source and a constituent of proteins , L-proline has a third , curious biological-function .
It is accumulated to high intracellular levels under conditions of osmotic-stress in bacteria ( 6 , 30 ) , plants ( 19 , 37 , 38 , 42 ) , and animals ( 20 ) .
It has been suggested that the accumulation of proline may be beneficial for adaptation to osmotic-stress because proline might be an osmotic balancer , acting to forestall the loss of water from the cells ( 19 ) .
In Salmonella typhimurium and other enteric bacteria , the synthesis of proline is not stimulated by osmotic-stress ( 13 , 33 ) , but these organisms can accumulate this imino acid to high levels during osmotic-stress by transport from the medium ( 6 , 9 , 10 , 14 ) .
The importance of the transport of proline for osmoregulation has been underscored by the observation that exogenous proline specifically stimulates the growth of some bacteria in media of inhibitory osmotic strength ( 9 , 10 , 13 ) .
S. typhimurium possesses three proline transport systems .
The first , the PutP system , is required for the utilization of proline as a carbon or nitrogen source , and it is subject to catabolite-repression and induction by proline ( 36 ) .
This transport system probably does not play a major role in the accumulation of proline during osmotic-stress because mutants that lack it show normal stimulation of growth by proline under conditions of osmotic-stress ( 14 ) .
The second proline permease , the ProP system , is important for osmoregulation because mutations inactivating it cause approximately a twofold reduction in the stimulatory effect of proline in media of elevated osmotic strength ( 14 ) .
The residual stimulation by proline under osmotic-stress in mutants lacking the ProP system is due to the third proline permease , the ProU system .
This system does not function in cells grown in media of normal osmolarity , but only under conditions of osmotic-stress ( 14 ) .
The existence of the ProU system has been inferred from the patterns of sensitivity to proline analogs of mutants lacking the PutP and the ProP systems .
Such double mutants are resistant to the proline analogs L-azetidine-2-carboxylate ( Azt ) and 3,4-dehydro-DL-proline ( Dhp ) in the normal media , but they become sensitive to both in media of elevated osmotic strength ( 14 ) .
It was suggested that this increased sensitivity is the result of the stimulation of the activity of the ProU system under conditions of osmotic-stress ( 14 ) .
However , the presence of this transport system has not been demonstrated by direct assays of proline transport activity .
In this publication , we present results showing that the rate of proline transport in S. typhimurium is stimulated by osmotic-stress and that this stimulation is due to the enhancement of two proline permeases , the ProP and the ProU transport systems .
The stimulation of transport could result from increased transcription of the genes of the two permeases or from some posttranscriptional event .
To distinguish between these two possibilities , we constructed strains carrying phage Mu dl ( Ampr lac ) insertions ( 8 ) such that the lacZ gene within the Mu dl phage has been placed under the transcriptional control of the proP or proU promoters .
We also present results of measurements of the,-galactosidase levels in these strains grown under various conditions of osmotic-stress .
Our results imply that although osmotic-stress causes a rapid increase in the transcription of the proU gene , it does not greatly affect the transcription of the proP gene .
Thus , the osmotic stimulation of the activity of the ProU and ProP transport systems may be brought about by two very different mechanisms .
MATERIALS AND METHODS Bacterial media and culture conditions .
The rich-medium used was LB ( 17 ) .
The minimal-medium was M63 ( 11 ) .
Culture conditions used were described in reference 14 , except that strains lysogenic for phage Mu dl ( Ampr lac ) were grown at 30 °C .
When used , the Azt concentration was 1 mM ; the Dhp concentration was 0.3 mM ; the 5-bromo-4-chloro-3-indolyl - , - D-galactosidase concentration was 40 mg/liter ; the sodium ampicillin concentration was 25 mg/liter , the chloramphenicol concentration was 12.5 mg/liter ; the tetracycline concentration was 15 mg/liter ; and the kanamy-cin sulfate concentration was 75 mg/liter .
The bacterial strains used are presented in Table 1 .
Generalized transductions were carried out with phage P22 HT105/1 int-201 , as described in reference 17 .
Phage Mu dl ( Ampr lac ) lysogens were constructed by infecting phage P1-sensitive ( P1P ) S. typhimurium strains with defective phage Mu dl that was grown with the helper phage Mu cts62 hP1 # 1 , having the host range of phage P1 ( 15 ) .
PlS S. typhimurium mutants were obtained by plating 4 x 108 cells of the parental P1-resistant strains on a lawn of109 PFU of phage FO on galactose MacConkey ( Difco Laboratories , Detroit , Mich. ) plates containing 5 mM CaC12 .
Mutants resistant to phage FO that are also unable to fernment galactose ( galE or galU ) are sensitive to phage P1 ( 35 ) .
These mutants were identified by their white colony color .
Strains TL200 and T1212 , thus obtained from strains TL145 and TL135 , respectively , proved to be sensitive to phages P1 and Mu cts62 hP1 # 1 .
The isolation of strains with proP : : Mu dl insertions was based on the fact that in M63 , proP + AputPA strains are sensitive to the proline analog Dhp , but proP derivatives are resistant ( 14 ) .
Strain TL212 ( AputPA galE or galU ) was infected with a Mu dl/Mu cts62 hP1 # 1 lysate , and ampicillin-resistant ( Ampr ) transductants were replica plated to M63 glucose plates containing Dhp and to M63 glucose plates .
After 16 to 24 h , colonies resistant to Dhp were detectable at the approximate frequency of 2 per 104 Ampr colonies .
Twenty independent colonies ( strains TL356 to TL375 ) , picked from the M63 glucose master plates , were saved .
As a preliminary test of whether these strains carried a Mu dl-induced proP mutation , we checked whether the mutations causing Dhpr were linked to the melA361 : : TnIO insertion , which has been shown to be 20 to 40 % cotransducible with previously characterized proP mutations ( 31 ) .
Strains TL356 to TL375 were transduced to Tetr by phage P22 grown on strain TT16721 ( melA361 : : TnJO ) , and 20 transductants obtained with each recipient were tested for Dhps and Amps .
Depending on the recipient , 25 to 70 % of the transductants became Dhp5 Amps , indicating that strains TL356 to TL375 carry a single Mu dl insertion in the proP gene or in a nearby gene of associated function .
The basis of the construction of proU : : Mu dl insertions was the observation that proUf proP AputPA strains , which are resistant to the proline analogs Dhp and Azt in M63 , are sensitive to both in M63 containing .0.3 M NaCl , but proU derivatives are resistant to both analogs even in media of elevated osmolarity ( 14 ) .
For the selection , strain TL200 ( proU AputPA galE or gaIU ) was infected with Mu dl , and Ampr colonies were replica plated to M63 glucose containing Azt and 0.3 M NaCl and to M63 glucose .
Colonies resistant to Azt in the presence of 0.3 M NaCl appeared after 3 days at the approximate frequency of 2 per 104 lysogens .
Eighteen independent mutants that were Aztr with 0.3 M NaCl ( strains TL333 to TL350 ) were picked from the M63 glucose master plates and saved .
The location of the Mu dl insertion was determined by mapping against the proUJ655 : : TnJO insertion , but the resident TnOO in the strains had to be removed first ( 4 ) .
A Tets derivative obtained from each strain was transduced to Tetr by P22 grown on strain TL187 ( proUJ655 : : TnJO ) .
With each recipient , 10 transductants that were tested proved to have become Amps , indicating that the Mu dl insertions in strains TL333 to TL350 are probably in the proU gene or in a nearby gene of related function .
Because the Mu dl carries a temperature-sensitive mutation ( cts62 ) in the repressor gene , lysogens have a growth defect above 37 °C , and the phage transposes at a high frequency to new sites , even at 30 °C .
Both of these problems can be ameliorated by the introduction of a Tn9 insertion into the phage B gene ( 3 ; P. Blum , L. Blaha , R. Shand , D. Holzschu , and S. Artz , Abstr .
Strains TL384 , TL386 , and TL388 ( proP : : Mu dl B : : Tn9 ) and strains TL378 , TL380 , and TL382 ( proU : : Mu dl B : : Tn9 ) are Cmlr transductants of strains TL361 , TL370 , TL372 , TL333 , TL343 , and TL346 , respectively , by P22 grown on strain AZ1430 [ F : : Mu dl B : : Tn9 ( Ampr Cmlr lac ) ] .
The B : : Tn9 insertion restored normal growth to the lysogens at 42 °C , and as was noted by Bake ' r et al. ( 3 ) , it reduced but did not completely eliminate the ability of the phage to transpose to new sites ( see below ) .
To confirm further that Mu dl B : : Tn9 insertions in the above transductants were in fact in the proP or proU genes , the fusions were backcrossed into new proP + or proU + strains , respectively .
Phage P22 grown on strains TL384 , TL386 , and TL388 ( proP : : Mu dl B : : Tn9 ) was used to transduce strain TL135 ( AputPA ) to Ampr .
In each case , > 99 % of the transductants were also Cmlr and able to grow at 42 °C .
However , faithful inheritance of the proP : : Mu dl B : : Tn9 insertion was rare .
Only 0.3 to 1 % of the transductants also became Dhpr .
When tested on lactose MacConkey plates , the Ampr Dhps transductants exhibited a range of phenotypes from Lac-to fully Lac ' , indicating that the phage had transposed to new loci during the transduction .
Nevertheless , with each of the donor strains TL384 , TL386 , and TL388 , we obtained Dhpr transductants , designated as strains TL423 to TL425 , respectively .
In P22 transduction , the Dhpr , Ampr , and Cmlr characters in strains TL423 , TL424 , and TL425 were lost simultaneously at the approximate frequency of 80 % upon the inheritance of the meIA361 : : TnJO allele .
The proU : : Mu di B : : Tn9 insertions in strains TL378 , TL380 , and TL382 were transduced with similar complications into strain TL179 ( proP AputPA ) .
The desired proU : : Mu dl B : : Tn9 proP AputPA transductants , strains TL390 , TL391 , and TL393 , were identified by their Aztr phenotype in the presence of 0.3 M NaCl .
The Ampr and Cmlr phenotypes in the latter three strains were lost at > 98 % frequency upon the inheritance of the proUJ655 : : TnlO allele in P22 transduction .
Since faithful transduction of the proP or proU : : Mu dl B : : Tn9 insertions proved to be a rare event , we were concerned whether a deletion or other rearrangement fusing the lacZ gene to an unknown promoter might not have occurred during the construction of the few proP or proU transductants that were obtained .
This is an important consideration , especially in the case of the presumed proU : : Mu dl B : : Tn9 mutants , because the synthesis of P-galactosidase in these strains appears to be nearly constitutive ( see below ) .
To ascertain that the mutation in these strains was due to a simple insertion of phage Mu dl B : : Tn9 into the proP gene , we made use of the fact that the B : : Tn9 insertion enables the reversion of Mu dl insertions by a perfect excision of the phage ( P. Blum , L. Blaha , R. Shand , D. Holzchu , and S. Artz , Abstr .
Proline auxotrophic strains that are also deficient in th 29 297 298 DUNLAP AND CSONKA TABLE 1 .
Bacterial strains used '' Genotype Strain Others Source Suc proP + proU + putPA + AZ1430 pyrFJ46 A ( trp ) -43 leu-500 P. H. Blum recAl srIC2 : : TnIO/F + : : Mu dl BX : : Tn9 ( Ampr Cmlr lac ) + Wild-type S. typhi-murium LT2 Ames , from J. L. Ingraham zcc-628 : : TnS This laboratory ( 14 ) zcc-628 : : Tn5 zjd-27 : : TnlO This laboratory ( 14 ) recAl srlC2 : : TnIO This laboratory ( 15 ) ga1E496 metA22 metE55 ilv ( ?
Tn9 A ( putPA ) 557 zcc-628 : : Tn5 TL478 , TL480 , TL4 ,2 proP + IproP1894 , proP1903 , proPl905 : : Mu dl B : : Tn9 A ( putPA ) 557 + zcc-628 : : TnS mel ' I meIA361 : TnJO P22 * TT1672 TL423 , TL424 , TL425 = Tetr Mel ' Dhps Ampr Cmlr Lac ' ; strains carry duplications of the melA-proP region ; see text ( 1 ) .
+ mneIA361 : : TnlO J. R. Roth ( 31 ) * x-s y = a , b , c denotes that a phage P22 lysate grown on strain x was used to trans ¬ TT1672 + a All strains are derived from S. typhimurium LT2 .
The abbreviations P22 ducq strain y , selecting phenotype a , and screening phenotypes b , c PutP and the ProP systems require abnormally high concentrations ( -0.5 mM ) of proline for rapid growth , whereas derivatives with a functional proline transport system grow well with 0.1 mM proline ( 14 , 36 ) .
We introduced a proB : : TnJO mutation into strains TL423 to TL425 ( proP : : Mu dl B : : Tn9 AputP ) and used the resultant transductants to select derivatives that were able to grow on M63 5-bromo-4-chloro-3-indolyl - , - D-galactosidase tetracycline plates containing 0.1 mM proline .
Such colonies appeared at the frequency of 10-6 .
Of these , only 0.1 to 1 % were white ( Lac - ) , but we obtained Lac-derivatives with each strain tested .
These were restreaked on the selective medium and tested for growth-on-LB ampicillin and LB chloramphenicol with the result that each was Amps and Cmls .
These results show that it is possible to revert the proP1894 : : Mu dl B : : Tn9 , proPJ903 : : Mu dl B : : Tn9 , and proPJ905 : : Mu dl B : : Tn9 insertions by perfect excision of the phage .
Thus , these proP alleles are simple Mu dl B : : Tn9 insertions , and therefore , the lacZ gene in strains TL423 to TL425 is transcribed from the proP promoter .
( We did not analyze the majority class of strains that formed blue colonies in the above selection .
These strains were still Ampr Cmlr , so presumably they were still proP : : Mu dl .
They might be strains that express the ProU system constitutively at a high level [ 18 ; J. Druger-Liotta and L. N. Csonka , unpublished data ] , strains with alterations in the specificity of some permease enabling the rapid uptake of proline , or strains carrying suppressors of the proline auxotrophy [ 22 ] ) .
By an analogous procedure , we confirmed the structure of the presumed proU : : Mu dl B : : Tn9 insertions in strains TL390 , TL391 , and TL393 .
We transduced the strains to proB : : TnJO and selected derivatives able to grow on M63 5-bromo-4-chloro-3-indolyl-3-D-galactoside tetracycline me-dium containing 0.1 mM proline and 0.3 M NaCI .
Such derivatives arose at the approximate frequency of 10-6 .
About 0.1 % of the colonies that appeared were proU + strains that lost the Mu dl phage by proper excision , as judged by the fact that they were Lac-Amps Cmls .
( The remainder of the colonies that were Lac ' might be strains with suppressors of the proline biosynthetic on transport defects or revertants of the proP mutation .
) Strains heterozygous for proP + and proP : : Mu dl insertions were obtained as derivatives with a chromosomal duplication in the melA-proP region .
A small fraction of cells in cultures of S. typhimurium carry spontaneous duplications of various segments of the chromosome ( 1 ) .
Strains TL423 to TL425 ( proP : : Mu dl ) were transduced to Tetr on LB tetracycline plates by phage P22 grown on strain TT1672 ( melA : : TnlO ) , and the transductants obtained were replica plated to M63 melibiose plates .
About 1 % of the progeny proved to be Mel ' .
These may be meIA : : TnJOImeIl heterozygotes that carry a duplication , or they may be transductants in which the TnWO has moved to a new site ( 1 ) .
Strains which carried a duplication that included the proP gene and which coinherited the proP + allele with the melA : : TnJO were recognized by being simultaneously Dhp5 Ampr Lac ' .
The frequency of such strains was ca. 5 % of that of the Mel ' Tetr transductants .
One characteristic of chromosomal duplications is their instability due to homologous recombination ( 1 ) .
At a frequency of 0.1 to 1 % , the presumed meIA : : TnlO proP + lmel + proP : : Mu dl heterozygotes segregated derivatives that were Lac-Amps Dhps ( proP + haploid ) , Dhpr Ampr Lac ' ( proP : : Mu dl haploid ) , or Tets Mel ' ( mel + haploid ) .
Because of this instability , the cells were grown before assays in M63 tetracycline and ampicillin with melibiose as the carbon source .
Strains were grown overnight in M63 glucose , diluted to the density of 5 x 107 cells per ml into the various media to be tested , and grown to the density of 4 x 108 cells per ml .
The cells were sedimented by centrifugation at room temperature and taken up to the original volume in Z buffer ( 32 ) , and 3-galq # tosidase was assayed at 24 °C as described by Miller ( 32 ) , except that the results were expressed as nanomoles of product formed per minute per milligram of protein , with the protein concentration determined by the method of Lowry et al. ( 28 ) .
We used flow dialysis ( 12 ) to measure the rate of proline uptake .
We carried out the assays as described by Dankert et al. ( 16 ) with some minor modifications .
The cells were grown overnight in M63 plus 20 mM glucose and the indicated concentrations of NaCl or sucrose , and then they were subcultured at a 10-fold dilution into fresh medium of the same composition and allowed to double 1.5 times to an approximate cell density of 109 cells per ml .
For the assay , 0.96 ml of the cultures was introduced into the upper chamber of the flow dialysis apparatus , and unless stated otherwise , chloramphenicol was added to a final concentration of 100 p.g / ml .
After 5 min , 5 , uCi of [ 3H ] proline ( ICN Radiochemicals , Inc. , Irvine , Calif. ) together with carrier proline at a final concentration of 20 pLM was added , bringing the final volume to 1.00 ml .
The assays were done at 24 °C , with the cells aerated by vigorous stirring .
The volume of the lower chamber of the flow dialysis apparatus was 2.8 ml .
Medium , which was always identical to that in which the cells were grown and assayed in , was pumped through the lower chamber at the rate of 4.8 ml/min .
Effluent fractions were collected at 0.5-or 1-min intervals .
A 0.4-ml sample from each fraction was mixed with scintillation fluid , and the radioactivity was measured by a scintillation counter .
The radioactivity in each sample was plotted against time , generating a curve that rises sharply to a peak during the first 4 to 5 min , due to the lag time required to pump the sample from the flow dialysis apparatus into the fraction collector tubes .
After the peak , the curve decreases linearly for at least the next 5 min , due to the uptake of substrate from the medium by the cells ( see Fig. 1 ) .
The transport rate is proportional to the slope of the curve during the linear portion of the decay , with the constant of proportionality being dependent on the specific activity of the substrate and the permeability properties of the particular dialysis membrane used .
To take the latter into account , the slope of the linear portion of the curve was divided by the height of the curve at the peak , which is a measure of the initial rate of diffusion of proline across the membrane .
The transport rate was obtained by multiplying the value thus obtained by the initial amount of proline in the upper chamber , 20 nmol .
The results were expressed as nanomoles of substrate taken up from the medium per minute per milligram of protein , with the protein measured by the method of Lowry et al. ( 28 ) .
RESULTS Osmotic stress stimulates [ 3H ] proline uptake via the ProP and ProU transport systems .
We measured the rate of [ 3H ] proline transport by cells grown in M63 or M63 containing 0.3 M NaCI ( Table 2 ) .
In strains that are proficient for all three proline permeases ( TL1 and TL106 ) , the rate of proline transport increases four-to sixfold as a consequence of growth in the presence of 0.3 M NaCl .
Osmotic stimulation of proline transport occurs in mutants that possess the ProP system ( strain TL192 ) , the ProU system ( strain TL179 ) , o both ( strain TL135 ) .
Proline transport is reduced to background levels in the mutant that lacks all three transport systems ( strain TL195 ) both in M63 and in M63 plus 0.3 M NaCl .
These results imply that S. typhimurium has two distinct proline permneases , the ProP and the ProU , whose activity is enhanced by osmotic-stress .
The third proline permease , the PutP system , is not regulated by osmotic-stress ; the transport activity of mutants possessing only that system ( strains TL199 and TL201 ) is unaffected by the presence of 0.3 M NaCl ( Table 2 ) .
The putA gene product , which has been proposed to be the regulatory protein for the transcription of the putP gene ( 29 ) , is not involved in the osmotic control of the proline transport systems , because the rate of proline transport in the presence of 0.3 M NaCl is not affected substantially by a putA mutation .
Since the technique of flow dialysis measures transport indirectly , the question might arise whether the apparent increase in proline uptake in the proP + or proU + strains in the presence of 0.3 M NaCl is truly due to the stimulation of two active transport systemns or whether it could be explained by some experimental artifact , such as a change in the diffusion properties of the dialysis membrane .
To address this point , we performed transport assays in the presence of 0.3 M NaCl and 20 , uM carbonyl-cyanide mchlorophenylhydrazone added 10 or 13 min after the start of the assay ( Fig. 1 ) .
The rapid exit of [ 3H ] proline from the cells could be detected within 3 min after the addition of carbonylcyanide m-chlorophenylhydrazone .
The proP and ProU transport systems are also sensitive to KCN .
At 2 mM , this poison inhibited transport via the former system by ca. 50 % and via the latter system by ca. 95 % ( data not shown ) .
Thus , the uptake of proline via both systems is an active transport process driven by the proton motive force .
The results in Table 2 were obtained in assays in which the proline concentration was 20 , uM .
A limited number of measurements were carried out with the substrate concentration at 200 , uM .
At this higher substrate concentration , the transport rate in the wild-type strain TL1 was 3.5 nmol/min per mg in M63 and 27.8 nmol/nmin per mg in M63 containing 0.3 M NaCi .
In strain TL179 ( proU + proP putP ) it was 15.2 nmol/min per mg in M63 plus 0.3 M NaCI .
Since in the flow dialysis tecchnique the transport rate is calculated from the relative rate of disappearance of the substrate , and since TL 192 ( proPproU puP ) TL 179 ( PrL / .
prOPpuP ) w-i 0o0o a. 600 ( E I ) ci Z 400 a w 0 .
PI-Z 200 C U 00000 0 00 * o * ccc v CCCP o 0 4 9 o. 0 .
The stimulation of proline transport in S. typhimurium by 0.3 M NaCl Mean proline transport activity ( nmol/min per mg of protein ) ± SD witha : Genotype Strain M63 NaCi putP proP proU putA TL106 + + + - + ¬ ¬ + + 0.78 ± 0.08 0.57 ± 0.06 0.40 ± 0.15 ( 5 ) 0.66 ± 0.03 ( 3 ) 0.62 ± 0.03 0.27 ± 0.05 4.78 ± 0.25 4.80 ± 0.20 2.14 ± 0.49 ( 9 ) 3.68 ± 0.59 ( 5 ) 0.57 ± 0.16 0.28 ± 0.01 TL135 TL179 TL192 TL199 TL195 putA + TL1 + + + 1.14 ± 0.04 4.54 ± 0.36 TL201 + -- 0.76 ± 0.01 0.63 ± 0.11 a Transport assays were done as described in the text .
The results are the averages of two independent determinations , except in cases where indicated by the numbers in parentheses .
The sensitivity of the ProP and ProU transport systems to the uncoupler carbonyl-cyanide-m-chlorophenylhydrazone ( CCCP ) .
The proline transport activity of strains TL192 ( proP + proU putP ) and TL179 ( proU + proP putP ) was determined in the presence of 0.3 M NaCl as described in the text .
Symbols : ( 0 ) of 20 , uM CCCP added at 10 min in results obtained in the presence the case of strain TL192 and at 13 min in the case of strain TL179 ; ( 0 ) results in the control experiment without CCCP .
small changes in the diffusion rate are more difficult to measure when the diffusion rates are high , the scatter in the data points tended to be greater at higher substrate concentrations .
For that reason , most of the assays were performed with 20 , uM proline , even though this concentration could be below the Km of the proline transport systems ( 2 ) .
In the experiments described above , only NaCl was used to increase osmolarity of the growth medium .
The transport of some metabolites , including proline. , is known to be dependent on Na + or other alkali ions ( 7 , 24 ) .
Therefore , the stimulation of proline transport might be due to the presence of Na + ( or Cl - ) ions that are not normally in M63 .
We dismiss this possibility on the basis of an experiment in which we used 0.47 M sucrose ( whose osmotic strength is approximately equal to that of 0.3 M NaCI ) to increase the osmolarity of M63 .
The transport activity of strain TL192 ( proP + proU putP ) in M63 plus 0.47 M sucrose was 3.06 ± 0.32 nmol/min per mg , and that of TL179 ( proU ' proP putP ) was 2.60 ± 0.15 nmol/min per mg .
Within experimental error , these values agree with those obtained with 0.3 M NaCl ( Table 2 ) .
Strains carrying proP : : Mu dl or proU : : Mu dl insertions .
To detect possible transcriptional regulation of the proP and proU genes , we constructed strains that carry phage Mu dl insertions in these loci .
We had obtained a total of 11 strains in which the lacZ gene is transcribed from the proP promoter and 12 strains in which the lacZ gene is transcribed from the proP promoter , as judged by their blue colony color on M63 containing 0.3 M NaCl and 5-bromo-4-chloro-3-indolyl-p-D-galactoside ( see above ) .
The effect of osmotic-stress on the 0-galactosidase levels of three representative proP-lacZ fusion strains and three representative proU-lacZ fusion strains is shown in Table 3 .
The three proP-lacZ strains ( and eight similar others ; data not shown ) exhibited some increase in the P-galactosidase levels upon exposure to 0.3 M NaCl , but since these increases were at most two-to threefold and not observed consistently , they are not likely to account for the stimulation of [ 3H ] proline transport re ported in Table 2 .
In contrast , the proU-lacZ fusion strains displayed a consistent increase in the P-galactosidase levels in response to 0.3 M NaCl .
Although there was some variation in the magnitude of the increase ( ranging from 7-fold in strain TL343 to more than 70-fold in strain TL346 ) , these results indicate that osmotic-stress stimulates the transcription of the proU gene .
The data in Table 3 were obtained with cells grown overnight to exhaustion of glucose in the indicated media , but similar results were observed in exponentially growing cells ( data not shown ) .
The phage Mu dl insertions in the strains were made heat resistant and more stable against transposition by the introduction of a Tn9 insertion in the phage B gene , and the resultant proP : : Mu dl B : : Tn9 or the proU : : Mu dl B : : Tn9 alleles were backcrossed into strains TL135 ( AputP ) or TL179 ( proP AputP ) , respectively ( see above ) .
The , B-galactosidase levels in the resultant transductants were , within experimental error , identical to those in the respective parental strains under various conditions of osmotic-stress ( data not shown ) .
A detailed analysis of the regulation of the proP or proU genes was carried out with these transductants .
Assays of [ 3H ] proline transport activities of strains carrying proP : : Mu dl : : Tn9 or proU : : Mu dl : : Tn9 insertions .
To verify that the presumed proP : : Mu dl B : : Tn9 and proU : : Mu dl B : : Tn9 insertions inactivated the relevant proline transport systems , we measured the proline transport activities of strains carrying these mutations .
In accordance with the expected results , the proline transport activity of these strains in the presence of 0.3 M NaCl is 5-to 10-fold lower than in the appropriate proP + or proU + control strains ( Table 4 ) .
Expression of j-galactosidase in proP-lacZ fusion strains .
Since the P-galactosidase levels in strains carrying proP-lacZ-fusions were not greatly affected by growth in 0.3 M NaCl ( Table 3 ) , we tested whether exposure to more extreme osmotic-stress might uncover the possible transcriptional regulation of the proP gene .
The result was again negative .
The , B-galactosidase levels in strains TL423 to TL425 ( proP : : Mu dl B : : Tn9 ) ranged from 29 to 64 nmol/min per mg of protein in cells grown in M63 containing 0.65 M NaCl , as compared with a range of 41 to 47 nmol/min per mg of protein in cells grown in M63 .
One possible reason for our inability to detect transcriptional regulation of the proP gene could be either that the proP + gene product might be required for normal regulation or that the regulatory protein might be part of the operon that contains the proP gene , and the Mu dl phage might have polar effects on its transcription .
To investigate this point , we constructed strains carrying chromosomal duplications , such that they were heterozygous for proP + and the proP1894 : : Mu dl B : : Tn9 , proPJ8903 : : Mu dl B : : Tn9 , or proPl905 : : Mu dl B : : Tn9 insertions ( see above ) .
The P-galactosidase activities in these strains displayed similar nearly constitutive expression ( 61 to 75 nmol/min per mg in M63 and 66 to 94 nmol/min per mg in M63 containing 0.3 M NaCI ) as was seen in the corresponding strains carrying only the proP : : Mu dl insertions ( Table 3 ) , indicating that the lack of transcriptional regulation is not due to loss of proP + gene product or a regulatory protein .
The above results suggest that the stimulation of transport activity via the ProP system is probably the result of some posttranscriptional regulation , such as translational control , or modification of the transport protein brought about by osmotic-stress .
Anderson et al. reported that the rate of proline transport via the ProP system is stimulated as a result of starvation for proline or other amino-acids ( 2 ) .
To test whether this stimulation is due to increased transcription of the proP gene , we introduced into strain T1425 ( proP : : Mu dl ) mutations blocking the proline , histidine , or leucine biosynthetic pathways and attempted to grow the resultant auxotrophs under the respective amino-acid-limiting conditions that Anderson et al. found to enhance the activity of the ProP system ( 2 ) .
We did not find any significant changes in the,-galactosidase levels of these strains ( data not shown ) , indicating that the enhancement of the ProP activity upon amino-acid starvation is not due to a transcriptional control .
Regulation of proU gene expression by osmotic-stress .
The most important result presented in this paper is that the ¬ ¬ ¬ + ¬ + ¬ ¬ ¬ Mean proline transport activity ( nmol/min per mg of protein ) ± SDb TL192 ( proP + proU : : TnIO/putP ) ... 4.90 ± 0.21 TL472 , TL474 , TL476 ( proP : : Mu dl proU : : TnJO AputP ) ... 0.34 +0.13 TL179 ( proU + proP AputP ) ... 2.26 ± 0.25 TL390 , TL391 , TL393 ( proU : : Mu dl proP AproP ) ... 0.26 ± 0.14 TL450 ( proP + proU : : Mu dl AputP ) ... 4.28 ± 0.05 TL423 ( proU + proP : : Mu dl AputP ) ... 1.70 ± 0.24 a The Mu dl insertions used in the strains in this experiment all carry Tn9 in the phage B gene .
The proP allele in strains TL179 , TL390 , TL391 , and TL393 is the spontaneous mutation proPJ654 .
b Proline transport activity was determined as described in the text .
The results are the averages of two independent measurements , with the deviation from the mean indicated , except for the set of isogenic strains TL472 , TL474 , and TL476 and strains TL390 , TL391 , and TL393 , for which the result reported is the average of the results for the set of the three strains , two independent measurements with each , with the error being the largest deviation from the average synthesis of,3-galactosidase in the proU : : Mu dl strains is elevated in response to osmotic-stress ( Table 3 ) , implying that the ProU transport system is regulated at the level of transcription .
Since the regulation of synthesis of 3-galactosidase in all proU-1acZ fusion strains was very similar , we carried out further characterization of the regulation of the proU gene with strain TL393 , whose Mu dl prophage has been stabilized with the B : : Tn9 insertion .
First , we determined the relationship between the osmotic strength of the medium and the expression of the proU gene ( Fig. 2 ) .
The steady-state transcription of this gene increased linearly over a 14-fold range , as the NaCl concentration was varied from 0 to 0.5 M. No further increase in the expression of the gene was brought about by NaCl concentrations above 0.5 M ( data not shown ) .
Increased transcription of the proU gene occurred very rapidly upon osmotic shock .
There was a detectable increase in the synthesis of , B-galactosidase in the proU : : Mu dl strain within 9 to 12 min after exposure to 0.3 M NaCl ( Fig. 3 ) .
These experiments were carried out in the absence of proline , indicating that exogenous proline is not required for the osmotic regulation of the proU gene .
The question might arise whether increased transcription of the proU gene is triggered specifically by NaCl or whether it is the result of a generalized osmotic-stress .
Therefore , we grew strain TL393 ( proU : : Mu dl ) in media whose osmotic strength has been increased by solutes other than NaCI and measured the levels of , B-galactosidase ( Table 5 ) .
Several of these solutes , both ionic [ ( NH4 ) 2SO4 and KH2PO4 ] and nonpolar ( sucrose and polyethylene-glycol 6000 ) , are able to induce the proU gene .
As the with NaCl , the level was case of expression of the proU gene , measured with various concentrations of polyethylene-glycol 6000 ranging from 0 to 200 g/liter , was proportional to the osmotic strength of the medium ( data not shown ) .
However , there were a number of TABLE 3 .
The effect of osmotic-stress on the , B-galactosidase activity in proP : : Mu dl and proU : : Mu dl strains,3-Galactosidase activity ( nmol/min per mg of protein ) withb : ~ ~ ~ Ratio of activity M3 03 M63 M63 +.3 ( with NaCI ) M NaCI without NaCI ) TL361 ( proP1894 : : Mu dl ) 107 214 2.0 TL370 ( proP1903 : : Mu dl ) 128 287 2.2 TL372 ( proP1905 : : Mu dl ) 137 174 1.3 TL333 ( proU1871 : : Mu dl ) 29 483 17 TL343 ( proU1188 : : Mu dl ) 37 258 7 T1346 ( proU1884 : : Mu dl ) 7 545 78 a The strains used are representatives of all proP-lacZ or proU-IacZ fusion strains we obtained which express f-galactosidase , i.e , form blue colonies on M63 containing 0.3 M NaCI and 5-bromo-4-chloro-3-indolyl - , - D-galactoside ( see text ) .
In eight other similar proP-lacZ fusion strains , the ratio of 1Bgalactosidase ( nanomoles per minute per milligram ) in the presence of 0.3 M NaCl/absence of NaCl ranged from 257:160 to 158:54 .
In nine other similar proU-lacZ strains the ratio of , B-galactosidase in the presence/absence of NaCl ranged from 318:34 to 181:4 .
In addition , we obtained a number of proP : : Mu dl or proU : : Mu dl strains that formed white colonies on M63 containing 0.3 M NaCl and 5-bromo-4-chloro-3-indolyl - , - D-galactoside .
These strains probably carry Mu dl in the orientation that is incorrect for the transcription of the lacZ gene from the proP and proU promoters .
The , B-galactosidase levels in these strains were low ( < 7 nmol/min per mg ) and were not influenced by the presence of 0.3 or 0.65 M NaCl .
b The , B-galactosidase activity was determined as described in the text , except that the cells were grown overnight in the indicated media to the exhaustion of glucose .
Proline transport activity in strains carrying proP : : Mu dl or proU : : Mu dl insertions Straina Soo TL393 ( pro U : : Mud-1 ) 700 +0.3 M NaCI - 600 L1J C 500 QE 400 ipE E .5 30 200 100 0 60 5 1 0 15 20 25 30 INDUCTION TIME 0 200 TL 393 ( pro U : : Mud-1 ) 160 TABLE 5 .
The effect of various solutes on the expression ofgalactosidase in a proU : : Mu dl B : : Tn9 strain P-galactosidase omolarityb activity Medium ( nmoVl min per mg ) None 0.22 19 NaCl 0.30 0.83 451 ( NH4 ) 2 S04 0.26 1.06 383 KH2PO4 0.33 0.75 455 Polyethylene 0.034 0.72 205 glycol 6000 Sucrose 0.47 1.05 316 Methanol 0.56 0.95 7 Ethanol 0.54 0.93 4 Glycerol 0.52 0.97 11 a Strain TL393 ( proU1884 : : Mu dl B : : Tn9 ) was grown in M63 with the indicated additions , and f3-galactosidase activity was assayed as described in the text .
b The osmolarity of the media was measured with a dew point psychrometer , except in the case of methanol , ethanol , and glycerol , in which it was determined from freezing point depression .
The osmolarity is expressed as the moles per liter of an ideal solute that would produce the water potential that was determined 1 ) 2 120 4 a. o E Addition to M63 ' Concn ( M ) E / j E 40 ¬ 0 o 0.1 0.2 03 0 NoCI CONCENTRATION 4 0 .
The expression of the proU gene as a function of the concentration of NaCl in the growth medium .
Strain TL393 ( proU : : Mu dl ) was grown in the media with the indicated NaCl concentration and assayed as described in the text .
( MINUTES ) of the Strain TL393 FIG. 3 .
The kinetics of induction proU gene .
( proP : : Mu dl ) grown exponentially in M63 glucose was subcultured into M63 glucose ( -- - ) or M63 glucose containing 0.3 M NaCl ( - ) , and the 3-galactosidase activity was assayed in aliquots removed at the indicated times .
solutes ( methanol , ethanol , and glycerol ) that did not cause an induction of the proU gene , even though they were present at osmolarities that were approximately equal to those of the other solutes .
It could be argued that the solutes that do not cause an induction of the proU gene may interfere with the induction rather than being merely unable to elicit it .
This possibility can be dismissed because normal induction of the gene by 0.3 M NaCl took place even in the presence of 0.52 M glycerol ( data not shown ) .
The difference between the compounds that can induce the proU gene and those that can not may reside in the fact that the latter compounds are lipid-soluble substances that are freely diffusible across the cell membrane ( 5 , 21 , 39 ) , whereas the former compounds are either nondiffusible or are taken up by active transport systems .
These observations suggest that the signal for the increased transcription of the proU gene might be an osmotic gradient across the cell membrane that can not be produced by freely diffusible substances .
We tested whether the proU gene might be subject to regulation by catabolite-repression or the utilization of proline as a nitrogen source .
These experiments were carried out with a putP + A + proP + derivative of strain TL393 ( proU : : Mu dl ) .
Growth with citrate instead of glucose as the carbon source or with proline instead of NH4 ' as the nitrogen source affected neither the basal nor the increased level of expression of the proU gene obtained in the presence of 0.3 M NaCl ( data not shown ) .
We also investigated whether proline starvation affected the expression of the proU gene .
A mutation conferring proline auxotrophy was transduced into strain TL393 ( proU : : Mu dl proP AputP ) .
Because of the defects in proline transport , the growth-rate of this strain can be controlled by varying the external proline concentration ( 14 , 36 ) .
We found that the expression of the proU gene was not altered by proline limitation or excess in either M63 or M63 containing 0.3 M NaCl ( data not shown ) .
Thus , the only environmental factor we found to control the expression of the proU gene is the osmotic strength of the medium .
DISCUSSION We have presented evidence that the transport of [ 3HJproline by S. typhimurium is stimulated by growth in media of elevated osmotic strength .
This enhanced proline uptake is due to the stimulation of two independent permeases , which we designated as the ProP and the ProU systems .
By constructing lacZ-fusions to genes proP and proU that are required for the functioning of the two systems , we carried out experiments to investigate the transcriptional regulation of the two systems .
We found that exposure of the cells to osmotic-stress causes a rapid , 10-fold increase in the transcription of the proU gene .
We were unable to detect a comparable transcriptional regulation of the proP gene .
It may be that the transcription of the proP gene itself is not regulated by osmotic-stress , but there are other unknown genes encoding components of the ProP system whose transcription is regulated .
Alternatively , the component ( s ) of the ProP system may be synthesized constitutively , and the stimulation by osmotic-stress is the consequence of some posttranscriptional event .
Thus , our results suggest that there may be interesting differences between the mechanisms of osmoregulation of the ProP and the ProU systems .
Proline transport via the ProP and the ProU systems is sensitive to the inhibitors of oxidative phosphorylation , carbonyl-cyanide-m-chlorophenylhydrazone or KCN , indicating that both systems accumulate proline actively against a concentration gradient .
In the case of the ProP system , the initial rate of uptake in the presence of 0.3 M NaCl was 3.68 nmol/min per mg ( Table 2 ; strain TL192 ) and was linear for at least the first 5 min ( Fig. 1 ) .
The cell concentration used in the assay ( -109 / ml ) corresponds to a protein concentration of 0.35 mg/ml .
So , in S min , a total of 3.68 x 5 x 0.35 or -6 nmol of substrate was taken up by the cells .
The total internal volume of 109 cells is ca. 10 - ' ml ( 39 ) , so the internal concentration of proline after 5 min is 6 mM .
Since after 5 min the external proline concentration is 20-6 , uM , or 14 , M , the internal concentration of proline that is accumulated by the ProP system in only the first 5 min was ca. 400-fold greater than the external concentration .
By a similar calculation , we can arrive at the result that after 5 min the ProU system could maintain a 200-fold difference between the internal and external concentrations of proline .
It has been proposed that proline , accumulated under conditions of osmotic-stress , functions as an osmotic balancer and thus forestalls the dehydration of the cytoplasm ( 19 ) .
Normally , 0.5 mM external proline is required to elicit the optimal protection against osmotic-stress ( 14 ) .
If the two transport systems together could maintain a 500-to 1,000-fold concentration gradient in the presence of 0.5 mM proline , then the internal concentration of proline would be of sufficient magnitude to function as an osmotic balancer .
Stimulation of proline transport by osmotic-stress has been observed in Escherichia coli by Britten and McClure ( 6 ) and C. A. Weaver ( Ph.D. thesis , University of Illinois , Champaign-Urbana , 1981 ) and in Streptomyces griseus by Killham and Firestone ( 26 ) .
However , these experiments did not address the question of how many transport systems are involved or how they are regulated .
Kaback and Deuel reported that the uptake of proline by cell-free vesicles of E. coli membranes was stimulated by their exposure to buffers of elevated osmotic strength ( 23 ) .
Since protein synthesis could not occur in the vesicles , it is likely that the stimulation of transport was the result of some posttranslational modification activating the ProP system of E. coli .
Fusions of lacZ to an uncharacterized gene whose transcription is regulated by osmotic-stress have been isolated in E. coli K-12 ( R. T. Vinopal , personal communication ) .
This gene is in the analogous region of the chromosome where the proU gene has been located in S. typhimurium , and thus it may be the proU gene of E. coli .
Laimins et al. ( 27 ) have constructed E. coli strains carrying lacZ-fusions to the kdp operon that encodes components of a potassium transport system .
These researchers found that the synthesis of,-galactosidase in these strains was induced by osmotic-stress .
However , in contrast with our results that osmotic-stress causes an increase in the steady state transcription of the proU gene , Laimins et al. ( 27 ) found that osmotic-stress caused only a burst of increased transcription of the kdp operon lasting for -30 min , after which the transcription returned to basal levels .
Besides the transport systems for proline and potassium , there are a number of other functions that respond to osmoregulation in bacteria .
These include the synthesis of glutamate and glutamine ( 41 ) , the intracellular content of polyamines ( 33 ) , the synthesis of a membrane-derived oligosaccharide found in the periplasmic space ( 25 ) , and the synthesis of two outer membrane porin proteins encoded by the ompC and ompF genes ( 34 , 40 ) .
It is not clear to what extent these cellular responses are regulated by a common mechanism .
Laimins et al. proposed a model to account for the osmotic control of the transcription of the kdp operon ( 27 ) .
According to this model , the regulatory protein is at least partially membrane bound , and its conformation , which determines its capacity to bind to the kdp operator , is influenced by the physical force of the turgor pressure exerted on the cytoplasmic membrane .
Since our results suggest that an osmotic gradient across the membrane might be the signal for the regulation of the proU gene , it is possible that the sensory protein regulating the transcriptio of the proU gene may also be located in the cytoplasmic membrane .
We acknowledge W. A. Cramer , S. B. Gelvin , S. G. Kustu , and C. A. Lindquist for constructive critique of the manuscript and J. R. Roth for helpful discussions .
We are grateful to W. A. Cramer and V. L. Davidson for teaching us the flow dialysis technique and for the use of their apparatus .
We thank R. A. Bressan and J. M. Clithero for help with the measurement of the osmolarity of the media .
This work was supported by Public Health service grant 1 R01 GM 3194401 from the National Institutes of Health .
Anderson , P. , and J. Roth .
Spontaneous tandem duplications in Salmonella typhimurium arise by unequal recombination between rRNA ( rrn ) cistrons .
Anderson , R. R. , R. Menzel , and J. M. Wood .
Biochemistry and regulation of second L-proline a transport system in Salmonella typhimurium .
Baker , T. A. , M. M. Howe , and C. A. Gross .
Mu dX , a derivative of Mu dl ( lac Apr ) which makes stable lacZ-fusions at high-temperature .
Bochner , B. R. , H.-C .
Huang , G. L. Schieven , and B. N. Ames .
Positive selection for loss of tetracycline resistance .
Bovell , C. R. , L. Packer , and R. Helgerson .
Permeability of Escherichia coli to organic-compounds and inorganic-salts scattering .
Acta 25 : measured by light 257-266 .
Britten , R. J. , and F. T. McClure .
The amino-acid pool in Escherichia coli .
Cairney , J. , C. F. Higgins , and I. R. Booth .
Proline uptake through the major of Salmonella typhimurium transport system is coupled to sodium ions .
Casadaban , M. J. , and S. N. Cohen .
Lactose genes Mu-lac bacterio-to exogenous promoters in one step using a phage vivo probe for transcriptional control Proc .
Christian , J. H. B. 1955 .
The influence of nutrition the water on relations of Salmonella oranienburg .
Christian , J. H. B. 1955 .
The relations of growth and water respiration of Salmonella oranienburg at 30 °C .
Cohen , G. N. , and H. V. Rickenberg .
Concentration specifique reversible des amino-acides chez E. coli .
Colowick , S. P. , and F. C. molecules by macromolecules : rapid measurement by rate of dialysis .
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 III , and C. L. Turnbough , Jr. 1981 .
Infection of Salmonella typhimurium with coliphage Mu dl ( Apr lac ) : construction of pyr : : lac gene fusions .
Dankert , J. , S. M. Hammond , and W. A. Cramer .
Reversal by trypsin of the inhibition of active transport by Colicin El .
Davis , R. W. , D. Botstein , and J. R. Roth .
Cold Spring Harbor Laboratory , Cold Spring Harbor , N.Y. 18 .
Dunlap , V. J. , and L. N. Csonka .
Regulation of the osmotically stimulated transport of proline and glycinebetaine in Salmonella typhimurium , p. 115-128 .
In J. L. Key and T. Kosuge ( ed .
) , Cellular and molecular biology of plant stress .
Alan R. Liss , Inc. , New York .
Flowers , T. J. , P. F. Troke , and A. R. Yeo .
The mechanism of salt tolerance in halophites .
Holoeuryhalinity and its mechanism in a cirriped crustacean , Balanus improvisus .
Heller , K. B. , E. C. C. Lin , and T. H. Wilson .
Substrate specificity and transport properties of the glycerol facilitator of Escherichia coli .
Itikawa , H. , S. Baumberg , and H. Vogel .
Enzymatic basis for genetic suppression : accumulation and deacylation of N-acetylglutamic-y-semialdehyde in enterobacterial mutants .
Kaback , H. R. , and T. F. Deuel .
Proline uptake by disrupted membrane preparations from Escherichia coli .
Kayama , Y. , and T. Kawasaki .
Stimulatory effect of lithium ion on proline transport by whole-cells of Escherichia 128:157 -164 .
Osmotic regulation and biosynthesis of membrane-derived in Escherichia coli .
Killham , K. , and M. K. Firestone .
Proline transport increases growth efficiency in salt-stressed Streptomyces griseus .
L. Laimins , A. , D. B. Rhoads , and W. Epstein .
Osmotic control of kdp operon expression in Escherichia coli .
H. , N. Rosebrough , A. L. Farr , and R. Randall .
Protein measurement with the Folin J. Biol .
Maloy , S. R. , and J. R. Roth .
Regulation of proline utilization in Salmonella typhimurium : characterization of put : : Mu d ( Ap , lac ) operon fusions .
Role of amino-acids in osmoregulation of nonhalophilic bacteria .
Menzel , R. , and J. Roth .
Identification and mapping of a second proline permease in Salmonella typhimurium .
Miller , J. Experiments in molecular genetics .
Cold Harbor Spring Laboratory , Cold Spring Harbor , N.Y. 33 .
G. K. J. Munro , F. , Herkules , Morgan , and W. Sauerbier .
Dependence of the putrescine content of Escherichia coli on the osmotic strength of the medium .
K. Nara , F. , Inokuchi , S.-I .
Matsuyama , and S. Mizushima .
Mutation causing reverse osmoregulation of synthesis of OmpF , a major outer membrane protein of Escherichia coli .
Ornellas , E. P. , and B. A. D. Stocker .
Relation of lipo-polysaccharide character to P1 sensitivity in Salmonella typhimurium .
Ratzkin , B. , M. Grabnar , and J. Roth .
Regulation of the major proline permease in Salmonella typhimurium .
The influence of water stress on the me-tabolism of diatoms .
Proline accumulation under different conditions of stress and light .
Stewart , G. R. , and J. A. Lee .
The role of proline accumulation in halophytes .
Stock , J. B. , B. Rauch , and S. Roseman .
Periplasmic space in Salmonella typhimurium and Escherichia coli .
Taylor , R. K. , M. N. Hall , and T. J. Silhavy .
Isolation and characterization of mutations altering expression of the major outer membrane porin proteins using the local anesthetic procaine .
Tempest , D. W. , J. L. Meers , and C. M. Brown .
Influence of environment on the content and composition of microbial free amino-acid pools .
Tully , R. E. , A. D. Hanson , and C. E. Nelson .
Proline accumulation in water-stressed barley leaves in relation to translocation and the nitrogen budget .