Physiological Consequences of the Complete Loss of Phosphoryl - Transfer Proteins HPr and FPr of the Phosphoenolpyruvate : Sugar Phosphotransferase System and Analysis of Fructose ( fru ) Operon Expression in Salmonella typhimurium DAVID A. FELDHEIM , t A. MICHAEL CHIN , : CATHERINE T. NIERVA , BRIGITTE U. FEUCHT , YONG WEI CAO , YOU FENG XU , SARAH L. SUTRINA , AND MILTON H. SAIER , JR. * Department of Biology , University of California , San Diego , La Jolla , California , 92093 Mutants of Salmonella typhimurium defective in the proteins of the fructose operon IfruB ( MH ) KA ] , the fructose repressor ( fruR ) , the energy-coupling enzymes of the phosphoenolpyruvate : sugar phosphotransferase system ( PTS ) ( ptsH and ptsl ) , and the proteins of cyclic AMP action ( cya and crp ) were analyzed for their effects on cellular physiological processes and expression of the fructose operon .
The fru operon consists of three structural genes : fruB ( MH ) , which encodes the enzyme IIIFru-modulator-FPr tridomain fusion protein of the PTS ; fruK , which encodes fructose-l-phosphate kinase ; and fruA , which encodes enzyme IlFru of the PTS .
Among the mutants analyzed were TnlO insertion mutants and lacZ-transcriptional-fusion mutants .
It was found that whereas afruR : : TnlO insertion mutant , severalfruB ( MH ) : : Mu dJ andfruK : : Mu dJ fusion mutants , and several ptsHI deletion mutants expressed the fru operon and I8-galactosidase at high constitutive levels , ptsH point mutants andfruA : : Mu dJ fusion mutants retained inducibility .
Inclusion of the wild-typefru operon in trans did not restore fructose-inducible 0-galactosidase expression in thefru : : Mu dJ fusion mutants .
cya and crp mutants exhibited reduced basal activities of all fru regulon enzymes , but inducibility was not impaired .
Surprisingly , fruB : : Mu dJ crp or cya double mutants showed over 10-fold inducibility of the depressed I-galactosidase activity upon addition of fructose , even though this activity in thefruB : : Mu dJ fusion mutants that contained the wild-type cya and crp alleles was only slightly inducible .
By contrast , jI-galactosidase activity in a fruK : : Mu dJ fusion mutant , which was similarly depressed by introduction of a crp or cya mutation , remained constitutive .
Other experiments indicated that sugar uptake via the PTS can utilize either FPr-P or HPr-P as the phosphoryl donor , but that FPr is preferred for fructose uptake whereas HPr is preferred for uptake of the other sugars .
Double mutants lacking both proteins were negative for the utilization of all sugar substrates of the PTS , were negative for the utilization of several gluconeogenic carbon sources , exhibited greatly reduced adenylate cyclase activity , and were largely nonmotile .
These phenotypic properties are more extreme than those observed for tight ptsH and ptsl mutants , including mutants deleted for these genes .
A biochemical explanation for this fact is proposed .
Recent studies have revealed that bacteria utilize many diverse mechanisms to regulate gene expression ( 25 , 37 ) , including antitermination ( 29 ) , enhancement ( 52 ) , competitive repression-activation ( 55 ) , and protein phosphorylation ( 1 , 11 ) .
Most of these mechanisms are believed to have their counterparts in eucaryotes ( 23 , 43 ) .
They have been found to operate in a variety of combinations in regulons encoding proteins of the bacterial phosphoenolpyruvate : sugar phosphotransferase system ( PTS ) ( 40 , 41 ) , which frequently possess multiple promoters or transcriptional terminators within single operons ( 10 , 33 , 34 , 53 , 55 ) .
In previous reports , we demonstrated that the fructose-specific catabolic system in many bacteria exhibits unique features .
First , it is the only carbohydrate catabolic system of the PTS that in enteric bacteria possesses its own HPr-like activity ( 49 , 50 , 54 ) and its own enzyme I-like protein ( 4 , 27 ; R. Geerse , Ph.D. thesis , University of Amsterdam , Amsterdam , The Netherlands , 1989 ) .
Second , the protein possessing HPr-like activity ( the FruBMH protein ) proved to be a t Present address : Department of Biochemistry , University of California , Berkeley , Berkeley , CA 94720 .
t Present address : Applied Biosystems , San Jose , CA 95131 .
fusion protein encoded by the fruB ( MH ) gene and apparently consisting of three domains , an N-terminal enzyme IIIFru ( IIIFru ) domain ( FruB ) , a C-terminal HPr-like domain termed FPr ( FruH ) , and a central modulator domain lacking sequence identity with any known PTS protein but possessing sequence similarity to part of the receiver module of bacterial two-component regulatory systems ( FruM ) ( 18 , 26 , 54a ) .
Third , it was shown that the fructose system apparently plays a central role in the regulation of gluconeogenesis ( 4 , 5 , several bacterial species were found to 19 , 20 ) .
Fourth , possess a fructose-specific PTS , and the fructose PTS is more widespread among bacterial species than any other sugar-specific PTS permease ( 8 , 12 , 46 , 54a ) .
Finally , it was proposed , on the basis of several indirect lines of evidence , that the complex , current-day PTS in numerous eubacteria evolved from a primordial fructose-specific PTS of relatively simple structure ( 39 , 47 , 51 ) .
All of these observations and postulates led to the possibility that the fructose regulon in enteric bacteria might possess unusual and interesting features .
In this report , we present the results of our genetic , physiological , and biochemical studies concerned with the consequences of the loss offru regulon expression as well as with the transcriptional regulation of the fru regulon in 545 Salmonella typhimurium .
We have isolated and characterized transposon insertion mutants , lacZ-transcriptional-fusion mutants , and point mutants in the fru regulon .
The induction characteristics of the fructose catabolic enzymes in these mutants and of 3-galactosidase in fru-lacZ fusion mutants are described .
The results of this and a previous study ( 6 ) show that expression of the fru regulon is influenced by a variety of genetic alterations as follows : ( i ) fruR mutations render expression of the five known activities of thefru regulon high-level constitutive ; ( ii ) fruB ( MH ) : : Mu dJ andfruK : : Mu dJ mutants synthesize P-galactosidase noninducibly ( constitutively ) although fruA : : Mu dJ mutants still exhibit inducibility for lacZ expression ; ( iii ) unlike all other carbohydrate catabolic enzyme systems thus far studied , crp and cya mutations reduce basal activities of fru regulon but do reduce the of induction ; and enzymes not ( iv ) extent fruB : : Mu dJ double crp or cya mutants exhibit greatly reduced basal P-galactosidase activity with restored fructose inducibility , butfruK : : Mu dJ crp and cya mutants which also exhibit reduced 3-galactosidase activity do not exhibit restored fructose inducibility .
Finally , we show that the simultaneous loss of both HPr and FPr activities results in a more severe phenotype than any genetic defect in the PTS studied thus far .
lacZ-transcriptional-fusions were constructed by mutagenesis with the Mu dJ transposable element ( 24 ) .
P22 was grown on strain TT10288 , and the resulting lysate was used to transduce SB1873 to kanamycin resistance ( 13 ) .
The Mu dJ transposon does not contain the Mu A or B gene and therefore can not by itself transpose .
However , when TT10288 ( 24 ) is the donor , the Mu A and B functions can be supplied transiently by the Mu A and B genes encoded by the adjacent helper bacteriophage present in strain TT10288 .
Many independent , kan-amycin-resistant , fructose-negative or fructose-sensitive transductants were isolated , and representatives of each phenotypic class were assayed for IlFru , IllFru , and fructose-1-phosphate kinase .
Strains LJ1182 , LJ2371 , LJ2375 , LJ2383 , LJ2391 , and LJ2394 ( Table 1 ) were obtained by this procedure .
All Mu dJ mutants in a wild-type background exhibited normal fermentation properties on mannitol , maltose , xylose , and galactose eosin-methylene blue plates and normal or slightly elevated levels of enzyme I and HPr when assayed in-vitro .
The Mu dJ markers in strains LJ1182 , LJ2371 , LJ2375 , LJ2383 , and LJ2394 were shown to be > 90 % cotransducible with a TnJO insertion mutation in the fru operon ( in strain LJ1120 , which lacked detectable IIFn , and IllFru activities ) when phage P22 was used as the carrier .
Thus , the acquisition of tetracycline sensitivity accompanied the acquisition of kanamycin resistance in greater than 90 % of the 30 to 60 transductants examined in each cross .
Thefru : : Mu dJ fusion mutations were transferred from the wild-type background into the genetic backgrounds of two fruR strains by transduction with phage P22 as the carrier and selection for kanamycin resistance .
These genetic backgrounds were LJ2301 ( fruRSi : : TnlO ) and LJ716 ( manA56 ptsHJSfruRSJ : : TnJO ( Table 1 ) .
In the latter background , the presence of the FPr domain of the FruBMH protein correlated with mannitol fermentation , since HPr was lacking as a result of the presence of the ptsHJ5 mutation .
In addition , any fructose fermentation or phosphoenolpyruvate-depen-dent phosphorylation observed could be attributed exclusively to the fructose-specific PTS , since the mannose PTS ( which also phosphorylates fructose with low affinity ) was lacking as a result of the manA56 mutation .
These fusion mutations were assigned to specific cistrons in the fru operon as follows ( Table 2 ) : fruB57 : : Mu dJ and fruB58 : : Mu dJ , fructose fermentation negative in the presence of the manA56 mutation , mannitol fermentation negative in the presence of the ptsHJ5 and fruRSI mutations , FPr , fructose-i-phosphate kinase , and IjFru activities absent , and lljFru activity present in fruB57 : : Mu dJ but no fruB58 : : Mu dJ ; fruK59 : : Mu dJ , fructose sensitive , mannitol fermentation positive in a genetic background containing the ptsHJ5 and fruR51 mutations , fructose-i-phosphate kinase and 11Fru activities absent , and 111Fru and FPr present ; fruA62 : : Mu dJ , fruA63 : : Mu dJ , and fruA64 : : Mu dJ , fructose fermentation negative in the presence of the manA56 mutation , mannitol fermentation positive in the presence of the ptsHJ5 and fruR51 mutations , and IIFru activity absent but IIIFru , FPr , and fructose-l-phosphate kinase activities present .
To study the effect of the fru operon in trans on lacZ expression in the various fru : : Mu dJ mutants , the Mu C ' gene ( ampicillin resistant ; 3 ) was transduced into the strains of interest , using strain SL4213 Mu C + amp # 1 ( E. A. Groisman , personal commnunication ) as the donor and phage P22 as the carrier .
The resultant strains were transformed with plasmid pMC3 carrying the entire fru operon ( 4 ) .
Selection was for resistance to neomycin ( 200 , uglml ) on LB plates .
Strains transformed were LJ1182 , LJ1237 , LJ2375 , LJ2391 , LJ2394 , LJ721 , and LJ723 ( Table 1 ) , and these transformed strains all proved to be capable of fructose fermentation .
The fruA60 : : Tn5 mutant was isolated by mutagenizing LJ2301 with TnS and screening for fructose fermentation .
TnS mutagenesis was accomplished by using TT3416 as the donor in P22-mediated crosses .
This strain possesses a temperature-sensitive F ' plasmid into which the Tn5 is inserted .
Since there is no Salmonella DNA in this replicon , incorporation of kanamycin resistance to homologous recombination can not occur .
The transductions were performed at 42 °C to prevent replication of the F ' plasmid in the transductants .
LJ2538 was tentatively designated fruA60 : : Tn5 on the basis of its in-vitro enzyme activities .
The fru : : TnJO mutant ( strain LJ1120 ) used in the mapping experiment described above was isolated as described previously ( 4-6 ) LJ724 LJ2537 LJ2538 SB1475 LJ1190 LJ2046 LJ1216 PP994 LJ1202 SB1477 LJ703 LJ1193 LJ1211 PP1002 PP1037 LJ1223 LJ1224 LJ1225 LJ1226 LJ1219 LJ1245 E. coli LU181 LU124 50 50 9 50 7 P. Postma P. Postma 22 F-lacI3 glpR galE crp-20b F-lacI3 22 pro cya-8306 glpR gaiR rpsL araC ,67 '' Additional strains are described in Table 8 .
Properties of Mu dJ insertion mutants specifically defective forfru operon expression and function in S. typhimuriuma Property Genotype and Fermentation ofb : strain Utilization of ' : Enzyme activityf 13-Galactosidase activitye FPr - Fructose + Fructose Fold induction activity IIIFru phosphate IIFru kinase Fructose Mannitol Fructose Citrate Acetate Lactate fruB58 : : Mu dJ LJ719 LJ2371 LJ2396 fruB57 : : Mu dJ LJ720 LJ1182 LJ1188 fruK59 : : Mu dJ LJ721 LJ2375 LJ2398 fruA62 : : Mu dJ LJ722 LJ2383 LJ2400 fruA63 : : Mu dJ LJ723 LJ2391 LJ2403 fruA64 : : Mu dJ LJ724 LJ2394 LJ2404 ¬ ¬ < 3 + + + + 1.4 1.9 1.3 < 1 < 0.5 < 3 ¬ ¬ < 1 + + + + 2.5 3.2 1.3 7 < 1 < 5 S + 32 S + + + 1.6 1.8 1.1 7 < 0.5 < 3 ¬ + 34 + + + + 0.07 0.4 6 7 85 < 1 + ¬ 30 + + ¬ + 0.14 0.7 5 11 41 < 1 ¬ + 39 + + + + 0.09 0.57 6 13 59 < 1 a Three genetic backgrounds were used in the analyses reported , as indicated in footnotes b tof ( see Table 1 ) .
b Fermentation responses on eosin-methylene blue-sugar plates containing the sugar at 0.5 % and growth responses on minimal-medium 63 plates containing the carbon source at 0.2 % were recorded as follows : + , strong ; ± , weak ; - , no response .
Kanamycin ( 50 , ug/ml ) was present in the media for both the fermentation and growth studies .
Fermentation responses were performed with strains possessing the genetic background of LJ716 ( manA56 ptsH15 fruRSl : : TnlO ) .
Consequently , negative fermentation of fructose indicates that at least one of the fructose catabolic enzymes is impaired , while negative fermentation of mannitol indicates that the fructose-inducible HPr-like protein , FPr , is impaired .
' Cells of the LJ716 genetic background were grown and extracts were prepared as described in footnote f. Activities are expressed as percentages of the activity of the isogenic parental strain , LJ716 ; 100 % FPr activity ( for strain LJ716 ) corresponds to 10 nmol of fructose-i-phosphate produced per min per mg of protein .
d Growth studies on minimal plates were conducted in the genetic background of strain LT2 ( SB1873 , wild type ) , as were the P-galactosidase assays .
eCells were grown at 37C in LB-medium with or without fructose ( 0.5 % ) , harvested in the exponential-growth-phase , washed three times , resuspended in medium 63 to 10 Klett units ( green filter , 16 , ug [ dry weight ] of cells per ml ) , toluene treated ( 2 drops of toluene per ml ) with agitation , and exposed to 2.5 mM p-nitrophenyl-p-galactoside at 37 °C for up to 45 min ; 3 ml of a 0.2 M Na2CO3-10 mM EDTA solution was then added , and the A410 was determined .
Absolute specific activities are expressed in nanomoles of substrate hydrolyzed per minute per milligramn of protein .
f Cells ( fruRS : : TnJO genetic background ) were grown in LB-medium , washed , suspended in 50 mM potassium phosphate ( pH 7.4 ) containing 1 mM dithiothreitol , disrupted by passage through a French pressure cell , centrifuged at low speed , and assayed as described in Materials and Methods .
Values are expressed in percentages of the activity found in the isogenic parental strain , LJ2301 ( fruRSJ : : TnlO ) .
Absolute specific activities for LJ2301 were as follows : 111 `` Ft , 36 nmol of fructose-i-phosphate produced per min per mg of protein ; llFra , 40 nmol of fructose-i-phosphate produced per min per mg of protein ; fructose-i-phosphate kinase , 2.6 pLmol of fructose 1,6-bisphosphate produced per min per mg of protein .
`` 4C-sugar or [ ' 4C ] citrate uptake rates in-vivo were measured as described previously ( 50 ) , with the radioactive substrate present at a concentration of 10 , uM ( specific activity , 5 , uCi / , umol ) .
Uptake of a metabolizable substrate represents the sum of transport and metabolism and therefore corresponds to a steady-state metabolic rate .
For the experiments conducted in Table 6 , 1.6 g of nutrient broth per liter and 0.2 % DL-lactate were present in the uptake medium as energy sources .
Assays for sugar phosphorylation in-vitro were performed as described by Saier et al. ( 50 ) except that the concentrations of constituents in the in-vitro assay mixture were as follows : `` 4C-sugar , 10 p , M ; phosphoenolpyruvate , 5 mM ; phosphate buffer , pH 7.4 , 50 mM ; MgCl2 , 12.5 mM ; KF , 25 mM ; and dithiothreitol , 2.5 mM .
The incubation time was usually 20 min at 37 °C .
IIFrU was assayed by inclusion of excess quantities of the soluble fraction of LB broth-grown LJ2301 ( fruRSl : : TnJO ) in the assay mixture .
I11Fru was assayed by inclusion of excess quantities of the soluble fraction of strain LJ144 [ cpd401 cysAJJSO ( F ' 198 pts + crr + cysA + ) ] ( 44 ) as well as butanol-urea-extracted membranes ( 45 ) from LJ2301 grown in LB-medium .
Net FPr-plus-HPr activity was assayed by inclusion of a crude extract from LB broth-grown SB1467 ( ptsH6 ) cells ( 50 ) to the assay mixture .
In a similar fashion , net enzyme I plus enzyme I * activity ( 4 ) was assayed by inclusion of a crude extract of strain SB1477 ( ptsI18 ) ( 50 ) in the assay mixture .
The radioactive sugar used to assay IIF , U and IIIF , was [ 14C ] fructose .
[ 14C ] fructose or [ ' 4C ] mannitol was used to assay the HPr/FPr and enzyme I/enzyme I * activities with the knowledge that the relative contributions to the total activity , due to the presence of the two proteins , HPr and FPr , would be different when the two substrates were used .
Fructose-i-phosphate kinase assays were performed as described by Fraenkel ( 16 ) .
, B-Galactosidase was assayed by the procedure of Miller ( 30 ) .
Protein was determined by the method of Lowry et al. ( 28 ) .
Net cyclic AMP production ( cells plus medium ) was measured after growth of cells from a small inoculum to the stationary-phase ( 12 h ) in medium 63 containing 0.5 % galactose essentially as described previously ( 14 , 44 ) .
Cultures were then brought to 100 °C in a boiling water bath ; after a 5-min incubation period , cells were removed by centrifugation .
The cyclic AMP concentration in the supernatant was measured by the cyclic AMP-binding protein method as described previously ( 14 , 44 ) .
Relative motility in semisolid agar .
Freshly grown wildtype and mutant colonies on LB plates were stabbed with toothpicks into semisolid agar medium consisting of 0.13 % tryptone , 0.7 % NaCl , 0.35 % Bacto-Agar ( Difco Laboratories ) , and a sugar ( when present ) at 1 % .
Plates were incubated at 30 °C for 16 h , and swarm diameters were measured at 2-h intervals thereafter .
Alternatively , 1 % tryptone , 0.5 % yeast extract , 1 % NaCl , and 0.35 % Bacto-Agar were used , and the freshly grown bacteria were stabbed into the agar with a straight platinum wire and incubated at 37 °C .
Swarm diameters were measured periodically thereafter .
Results obtained with these two media were similar .
Cyclic AMP ( 2 to 10 mM ) was included in the agar medium in some experiments .
All radioactive sugars were purchased from Dupont , NEN Research Products .
Nonradio-active sugars and reagents were purchased from the Sigma Chemical Co. .
LB-medium was as described by Miller ( 30 ) .
When a sugar was included , its concentration was 0.5 % .
The minimal-medium used was medium 63 without iron ( 50 ) .
Tetracycline , kanamycin , and ampicillin were used at concentrations of 20 , 25 or 50 , and 50 , uglml , respectively .
Fermentation was recorded on eosin-methylene blue plates without lactose .
Sugars were included at a concentration of 0.5 or 1 % .
RESULTS Characterization offru : : Mu dJ fusion mutants .
Six Mu dJ insertion mutants expressing,-galactosidase and specifically resulting in the loss of fructose utilization were isolated and characterized ( Table 2 ) .
In a wild-type genetic background , all of these mutants fermented and utilized fructose weakly ; in the genetic background of a manA mutant ( 1IMan ) , they were negative for fructose utilization ( except strain LJ2375 , which was sensitive to fructose ) .
They nevertheless fermented other sugars ( mannitol , maltose , xylose , and galactose ) normally unless the ptsH mutation was also present , in which case the two fruB mutants did not ferment mannitol .
All of the mutants except strain LJ2391 utilized citrate , acetate , and lactate at wild-type rates ( Table 2 ) .
, - Galactosidase in strains LJ2371 , LJ1182 , and LJ2375 was synthesized at high , nearly constitutive levels , but this enzyme in strains LJ2383 , LJ2391 , and LJ2394 exhibited low basal activity which could be enhanced about sixfold by inclusion of fructose in the growth medium ( Table 2 ) .
These Mu dJ insertion mutations were transferred by transduction to a fruRSl : : TnJO genetic background ( see Materials and Methods ) .
Strain LJ2396 exhibited low to negligible activities of the three fructose-specific proteins ( Il1FrU , fructose-1-phosphate kinase , and IIF , ) ; LJ1188 exhibited IIIFIi activity but nondetectable kinase and 1IFru activities ; LJ2398 similarly exhibited nondetectable kinase activity ; and strains LJ2400 , LJ2403 , and LJ2404 exhibited low IIIFnI activity , high fructose-l-phosphate kinase activity , and nondetectable IIFru activity ( Table 2 ) .
The fruBS8 : : Mu dJ mutation evidently abolished the activities of all fru operon products , including FPr ( measured in a ptsH genetic background ) , but the fruB57 : : Mu dJ mutation apparently left the IIIFm domain of the FruBMH protein intact while abolishing the FPr and downstream activities .
On the basis of these results , it was concluded that the Mu dJ insertions in strains LJ2371 and LJ1182 were in the B and M or H domains of the fruB ( MH ) gene , respectively ; the insertion in strain LJ2375 was in the fruK gene , and those in strains LJ2383 , LJ2391 , and LJ2394 were in the fruA gene .
Strains bearing the chromosomal fru : : Mu dJ mutations in the wild-type genetic background ( LT2 ) and also bearing plasmid pMC3 carrying the entire fru operon ( 4 ) were constructed as described in Materials and Methods .
These strains were then examined for,-galactosidase activities after growth in complex medium with or without fructose as described for the experiments presented in Table 2 .
The presence of the fru operon in trans had little or no effect on the induction of 13-galactosidase in the fruA , fruB , or fruK mutants .
The consequences of mutations in the fruR , crp , cya , and pts genes on expression of f-galactosidase in some of the Mu dJ fusion mutants were examined ( Table 3 ) .
Synthesis of,-galactosidase in the fruB58 : : Mu dJ fusion mutant , which was only slightly inducible , was increased threefold by inclusion of the fruRS : TnJO mutation .
crp and cya mutations lowered basal , B-galactosidase activity in this fusion mutant about 10-fold , but surprisingly , this activity was restored by inclusion of fructose in the growth medium .
Similar results were obtained with the fruB57 : : Mu dJ mu a Cells were grown in LB-medium with or without 0.5 % fructose , washed three times , suspended in 50 mM potassium phosphate buffer ( pH 7.4 ) containing 1 mM dithiothreitol , and passed through a French press at 10,000 lb/in2 .
, B-Galactosidase activity was measured as described in Materials and Methods .
In all strains analyzed , fructose-i-phosphate kinase activity was low to negligible and was not inducible by inclusion of fructose in the growth medium .
Assays were conducted in duplicate or triplicate with reproducibility of ± 30 % .
The value of 1.0 corresponds to an absolute specific activity of 1.4 per mg of protein .
nmol of substrate hydrolyzed per min tant , although the fruR mutation had a lesser effect on the uninduced level of P-galactosidase .
Introduction of the ptsH15 or A ( cysK-ptsHI41 ) mutation , which like the cya mutation reduces cellular cyclic AMP levels ( 14 , 44 ) , similarly reduced basal activity and restored induction by fructose .
By contrast , the constitutive expression of P-galactosi-dase activity in the fruK59 : : Mu dJ mutant , which was likewise depressed by the cya or crp mutation , was not rendered inducible by fructose when cyclic-AMP-receptor-protein ( CRP ) or adenylate cyclase function was destroyed .
Effects of mutations in the pts operon on fru regulon expression .
Table 4 summarizes the effects of mutations in the pts operon ( 10 , 38 , 49 , 50 ) on expression of the four activities associated with the three structural genes of thefru operon in S. typhimurium as measured by enzyme assays in-vitro .
In the wild-type strain , FPr , IlFru , and 11IFru were apparently about 10-fold inducible , whereas fructose-i-phos-phate kinase seemed to be about 25-fold inducible .
Two ptsH mutants were examined for their effects on fru operon expression .
The ptsH15 mutation is a UGA nonsense mutation in the structural gene encoding HPr , and consequently it is somewhat leaky for ptsH expression ( 50 ) .
The ptsH6 mutation is a missense mutation in the structural gene for HPr and exhibits a tighter phenotype ( 50 ) .
Both mutations appeared to depress the basal and induced levels of fru operon expression , possibly because of the reduction in adenylate cyclase activity in these strains ( 14 , 41 , 44 ; see also Tables 5 and 8 ) .
Similar behavior was observed for the tight ptsl mutant , ptsIJ8 ( data not shown ) , which resulted from the introduction of an ochre nonsense mutation in the structural gene for enzyme I ( 50 ) .
The A ( cysK-ptsHI41 ) mutation exhibited a similar effect , but induction was abolished or markedly reduced .
Similar results were obtained when the cells were grown in minimal-medium rather than complex medium ( data not shown ) .
Elimination of IIGc ' c function by introduction of a TnJO insertion mutation into the crr gene ( 9 , 48 ) did not appear to appreciably affectfru regulon expression either by itself ( Table 4 ) or in the presence of pts mutations ( data not shown ) .
Dependency of fru operon expression on cyclic AMP and CRP .
The dependency of [ 14C ] fructose uptake on cyclic receptor protein ( CRP ) is shown in Table 5 .
AMP and its Wild-type cells effectively accumulated radioactive metabolites from [ 14C ] fructose when grown in a nutrient medium lacking fructose , and inclusion of fructose in the growth medium enhanced the uptake rate about threefold .
By contrast , cya or crp mutants , which lack adenylate cyclase or CRP , respectively , accumulated radioactive metabolites from exogenous [ 14C ] fructose poorly when grown in the absence of fructose or cyclic AMP ( uptake rate , about 25 % of the uninduced wild-type rate ) .
Surprisingly , the uptake rates were enhanced about 10-fold by inclusion of fructose in the growth medium ( Table 5 ) .
The induced rate of [ 14C ] fructose uptake was essentially the same in the mutants as in the wild-type strain .
Inclusion of cyclic AMP in the growth medium lacking fructose restored uptake by the cya mutant , but not by the crp mutant , to the level observed for the wild-type strain when grown without fructose .
The unusual dependency of fructose uptake on the intact cya and crp genes suggested that basal synthesis of the proteins of the fru operon was dependent on cyclic AMP and CRP but that induction by fructose was not .
This conclusion was confirmed by assaying the fru enzymes in-vitro ( Table 5 ) .
The basal activities of 11Fru , 111Fru , and fructose-i-phosphate kinase were all reduced in the cya and crp mutants relative to the wild-type parental strain .
Inclusion of fructose in the growth medium enhanced the activities of all of these enzymes to about the same extent as in the parental strain .
The fully induced level of each enzyme was somewhat less than that observed for the parental strain , however .
Comparison of the levels of enzyme activities with the uptake rates suggested that for both the wild-type and mutant strains , the pair was rate limiting for uptake of [ 14C ] fructose in the uninduced state but not after induction by fructose .
The effects of the loss of the CRP on the uptake of glucose and fructose after growth under different conditions were determined ( Table 6 ) .
In the wild-type strain , glucose uptake was mildly inducible by growth in the presence of glucose , in agreement with previous results ( 36 ) , and the presence of fructose in the growth medium depressed this activity only slightly .
Addition of a 100-fold excess of fructose to the cell suspensions during transport inhibited glucose uptake only 5 to -28 % ( data not shown ) .
In the crp mutant , glucose uptake activity was not induced by inclusion of a sugar in the growth medium .
In fact , the activity was slightly depressed .
Effects of fructose induction and various relevant mutations on P-galactosidase synthesis infruB : : Mu dJ andfruK : : Mu dJ strains of S. typhimuriuma Relative sp act-Fructose + Fructose LJ2371 fruB58 : : Mu dJ 1.0 2.6 LJ2396 fruB58 : : Mu dJfruR51 : : TnJO 3.0 1.8 LJ1224 fruB58 : : Mu dJ crp : : TnJO 0.14 1.3 LJ1226 fruB58 : : Mu dJ cya : : TnJO 0.10 1.2 LJ1182 fruB57 : : Mu dJ 1.5 2.4 dJfruR5J : : TnlO 2.4 1.2 LJ1188 fruB57 : : Mu LJ1245 fruB57 : : Mu dJ crp : : TnJO 0.09 1.8 LJ1219 fruB57 : : Mu dJ cya : : TnJO 0.17 2.2 LJ1190 fruB57 : : Mu dJ ptsH15 0.3 2.7 LJ1193 fruB57 : : Mu dJ A ( cysK-ptsHI41 ) 0.6 2.9 LJ2375 fruK59 : : Mu dJ 1.8 1.2 LJ1223 fruK59 : : Mu dJ crp : : TnJO 0.15 0.21 LJ1225 fruK59 : : Mu dJ cya : : TnJO 0.15 0.19 Strain Genotype TABLE 4 .
Effects of mutations in the pts operon on expression of the genes of the fru operon in S. typhimuriuma Enzyme activityb FFFructos Fr ina e IIIeu-i1-phoskpshate FPr ( apparent ) IIFru kinase ± + - + - + SB1873 Wild type + 1.0 10 4 1.0 7 1 1.0 10 ± 4 1.0 25 ± 13 SB1475 ptsHJ5 + 0.4 ± 0.2 3.0 ± 2.1 0.6 ± 0.1 4 ± 1 0.7 4.1 1.2 ± 0.6 22 ± 7 LJ2046 ptsH6 + ND ND 0.5 3.8 0.7 3.3 1.1 ± 0.7 12 ± 7 LJ703 A ( cysK-ptsHI41 ) - ND ND 0.7 0.4 1.1 1.1 0.9 ± 0.4 2.3 ± 1.3 PP994 crr-307 : : TnJO + ND ND 0.9 9.3 1.7 5.7 2.1 ± 1.7 23 ± 12 a Cells were grown in LB-medium with ( + ) or without ( - ) 0.5 % fructose and assayed for the various enzyme activities as described in Materials and Methods .
b The value of 1.0 corresponds to absolute specific activities ( in nanomoles of product formed per minute per milligram of protein ) of 2 for FPr plus HPr , 5 for IIFI , 4 for IIFI , and 100 for fructose-i-phosphate kinase .
When error values are provided ( in standard deviations ) , the experiments were conducted at least three times .
ND , Not determined Fefremrenutation of fruc o cttsoen Strain Genotype Enzyme activity IIIF Fructose uptakeb iiFr Strain Genotype Fructose-i-phosphate kinase - + cAMP + Fru - + Fru - + Fru - + Fru SB1873 Wild type 1.0 1.0 ± 0.2 2.6 ± 0.6 1.0 7.1 ± 0.6 1.0 11 ± 8 1.0 25 ± 13 PP1002 cya : : TnlO 0.2 ± 0.1 1.0 ± 0.2 2.8 ± 0.7 0.5 ± 0.1 5.3 ± 2.0 0.5 ± 0.2 6.9 ± 4.1 0.5 ± .03 15 ± 8 PP1037 crp : : TnJO 0.3 ± 0.1 0.3 ± 0.1 2.4 ± 0.5 0.5 ± 0.1 6.0 ± 4.0 0.3 ± 0.2 7.5 ± 3.5 0.5 ± 0.13 12 ± 6 a All assays were performed as described in Materials and Methods .
Most values were determined three times .
Error values are expressed in standard deviations .
Absolute specific enzyme activities are reported in Table 4 .
The value of 1.0 for fructose uptake corresponds to 4 , umol of [ 14C ] fructose taken up per min per g ( dry weight ) of cells at 37 °C .
All strains fermented fructose .
b - , Cells were grown in LB-medium without fructose ; + cAMP or + Fru , cells were grown in the same medium in the presence of 5 mM cyclic AMP or 0.5 % fructose , respectively , for 3 h before harvesting during exponential-growth .
The low degree of apparent induction in the fructose uptake experiment ( measured in-vivo ) relative to the phosphorylation experiments ( measured in-vitro ) is believed to reflect the fact that enzyme I and HPr rather than are rate limiting for transport in the fully induced state .
Excess amounts of these energy coupling proteins were added for the in-vitro assays .
No sugar Fructose Glucose Glucose + fructose [ ' 4C ] fructose uptake after growth in presence of : No sugar 26 7 Fructose 100 100 Glucose 37 13 Glucose + fructose 55 71 a Cells were grown in LB-medium containing the sugar ( s ) indicated at a concentration of 0.5 % .
The uptake experiment was conducted at 37 °C , with the radioactive sugar at a concentration of 10 , uM ( 5 p.CiIpmol ) as described in Materials and Methods with washed cells suspended to a density of 10 Klett units ( green filter ; 16 , ug [ dry weight ] of cells per ml ) in medium 63 containing 1.6 g of nutrient broth per liter and 0.2 % DL-lactate as energy sources .
Additional experiments revealed that 1 mM nonradioactive fructose only weakly inhibited [ `` 4C ] glucose uptake ( 5 to 28 % inhibition ) but that glucose more strongly inhibited [ 14CJfructose uptake ( 30 to 62 % inhibition ) under the same standard conditions in both the wild-type and the crp mutant strains .
Absolute rates of uptake , corresponding to 100 % , were 12 and 10 , umol of [ 14C ] glucose and [ ' 4C ] fructose taken up per min per g ( dry weight ) of cells at 37 °C .
presence of a 100-fold excess of fructose depressed glucose uptake by the crp mutant only 5 to 10 % ( data not shown ) .
In contrast , fructose uptake was markedly induced by growth in the presence of fructose , both in the wild-type and in the crp mutant strains .
Glucose , when present as the sole sugar in the growth medium , had a slight inductive effect , but it repressed when fructose was also present ( Table 6 ) .
Inhibition of fructose uptake by a 100-fold excess of glucose was similar in the wild-type and crp mutant strains ( 30 to 62 % inhibition ) .
Thus , catabolite-repression and catabolite inhibition are both apparently operative in the absence of CRP .
Fructose uptake and the activities of the fructose-specific enzymes of the PTS ( IIFrU and lljFru ) were assayed in a wild-type E. coli strain , a cya deletion mutant , LU124 , and a crp deletion mutant , LU181 ( 22 ) , as described in Table 5 .
The induction properties of these E. coli strains were very similar to those reported for the corresponding Salmonella strains ( data not shown ) .
Properties of mutants lacking HPr or FPr of the PTS .
Growth of the wild-type cells in the presence of fructose induced fructose uptake activity ( Table 7 ) .
Cells defective for HPr ( ptsH15 ; SB1475 ) or components of the fru operon including FPr ( fruB57 : : Mu dJ ; LJ2529 ) took up [ ' 4C ] fructose poorly after growth in complex medium lacking fructose .
Of these two mutants , only the ptsH15 mutant took up fructose appreciably after growth in the presence of fructose , and negligible activity was observed for the ptsH15 fruB57 double mutant ( Table 7 ) .
[ 14C ] mannitol uptake activity showed very different behavior ( Table 7 ) .
In the wild-type strain , this activity was specifically induced by growth in the presence of mannitol .
In uninduced cells , the fruB57 : : Mu dJ mutant took up [ 14C ] mannitol as well as the wild type , whereas the ptsH15 mutant exhibited exceptionally low activity which was only slightly greater than that of the ptsH15 fruB57 double mutant .
Growth of the ptsHJ5 mutant in the presence of fructose ( but not mannitol ) induced uptake activity , presumably due to FPr induction ( 49 , 50 ) , as suggested by the fact that the double mutant was essentially inactive .
When [ 14C ] glucose replaced [ 14C ] mannitol as the radioactive substrate , similar results were obtained ( data not shown ) .
The sums of the activities of HPr and FPr , here designated HPr plus FPr , as well as the enzyme I activities in the same strains , were measured in-vitro .
When assayed with [ 14C ] mannitol as the sugar substrate ( Table 7 ) or [ 14C ] methyl-a-glucoside as the sugar substrate ( data not shown ) , somewhat higher than wild-type HPr activity was observed in the uninduced fruB57 : : Mu dJ mutant , whereas the ptsH15 mu tant and the double mutant showed low activities .
After fructose induction , the FPr activity increased dramatically in the ptsH15 mutant strain , but no appreciable increase in activity was observed in the fruB57 : : Mu dJ mutant or the ptsH15 fruB57 double mutant .
In fact , the latter mutant lacked measurable activity .
Enzyme I activities were low in the ptsH15 and ptsH15 fruB57 mutants relative to the wild-type strain as a result of the polarity of the UGA nonsense mutation in the mutant ptsH15 gene ( 50 ) .
Correlation of citrate uptake and motility with cyclic AMP production .
Whereas the ptsH15 mutant produced 50 % less cyclic AMP than the parental strain , and the fruB57 : : Mu dJ mutant produced 50 % more cyclic AMP than the parent , the double mutant produced very little cyclic AMP ( Table 7 ) .
These findings correlated with the induction of citrate uptake , motility in semisolid agar medium , and the utilization of various gluconeogenic substrates by the different mutant strains ( Table 7 ) .
Earlier observations had suggested that citrate utilization in mutants defective for the proteins of the PTS could serve as an indicator of adenylate cyclase activity ( 43 ; M. H. Saier , Jr. , unpublished observations ) .
The same appears to be true of alanine and lactate utilization .
Since ribose and xylose could still be utilized by the double mutant , it was clear that some basal level of cyclic AMP synthesis occurred .
Motility in semisolid agar medium also correlated with cyclic AMP production ( Table 7 ) .
The motile behavior of many mutant strains defective for the proteins of the PTS , various catabolic enzymes , and the proteins of cyclic AMP action are summarized in Table 8 .
It can be seen that leaky pts mutants , mutants defective for the enzymes II of the PTS , and mutants lacking specific enzymes of intermediary metabolism exhibited normal motility in the semisolid medium used .
Tight pts mutants , lacking HPr or enzyme I , and crr mutants , defective for IIIGc , exhibited depressed motility , whereas cya mutants were essentially nonmotile .
These results correlate with the known effects of the mutations studied on adenylate cyclase activity in-vivo ( 14 , 44 ) .
crp mutants were also nonmotile .
The fruRSi : : TnJO mutant , which synthesizes normal levels of cyclic AMP ( 5 ) , exhibited normal motility .
As expected , motility of a cya mutant was restored by addition to the semisolid medium of 5 mM cyclic AMP ( Table 8 ) .
The same was partially true of pts mutants and the ptsHJ5fruB57 : : Mu dJ double mutant ( Table 8 and unpublished experiments ) .
Table 8 also shows the motile behavior of a number of Escherichia coli strains .
Once again , in the pts , crr , and cya mutants studied , motility correlated well with adenylate cyclase activities measured previously ( 14 , 41 ) .
The partial restoration of motility by the crr-13 mutation in the ptsI31 - + cAMP + Fru - + Fru - + Fru - + Fru SB1873 Wild type 1.0 1.0 ± 0.2 2.6 ± 0.6 1.0 7.1 ± 0.6 1.0 11 ± 8 1.0 25 ± 13 PP1002 cya : : TnlO 0.2 ± 0.1 1.0 ± 0.2 2.8 ± 0.7 0.5 ± 0.1 5.3 ± 2.0 0.5 ± 0.2 6.9 ± 4.1 0.5 ± .03 15 ± 8 PP1037 crp : : TnJO 0.3 ± 0.1 0.3 ± 0.1 2.4 ± 0.5 0.5 ± 0.1 6.0 ± 4.0 0.3 ± 0.2 7.5 ± 3.5 0.5 ± 0.13 12 ± 6 a All assays were performed as described in Materials and Methods .
Most values were determined three times .
Error values are expressed in standard deviations .
Absolute specific enzyme activities are reported in Table 4 .
The value of 1.0 for fructose uptake corresponds to 4 , umol of [ 14C ] fructose taken up per min per g ( dry weight ) of cells at 37 °C .
All strains fermented fructose .
b - , Cells were grown in LB-medium without fructose ; + cAMP or + Fru , cells were grown in the same medium in the presence of 5 mM cyclic AMP or 0.5 % fructose , respectively , for 3 h before harvesting during exponential-growth .
The low degree of apparent induction in the fructose uptake experiment ( measured in-vivo ) relative to the phosphorylation experiments ( measured in-vitro ) is believed to reflect the fact that enzyme I and HPr rather than are rate limiting for transport in the fully induced state .
Excess amounts of these energy coupling proteins were added for the in-vitro assays .
Effect of carbon source present during-growth on relative [ 14C ] glucose and [ ' 4C ] fructose uptake activities in wild-type and crp mutant strains of S. typhimuriuma Relative rate in strain : SB1873 PP1037 ( wild type ) ( crp ) Determination [ ' 4C ] glucose uptake after growth in presence of : 62 72 100 80 67 67 100 89 TABLE 7 .
Properties of S. typhimurium mutants defective for HPr , FPr , or botha Value LJ1182 ( fruB57 : : Mu dJ ) Property Inducer SB1873 SB1475 ( ptsH15 ) LJ1190 ( wild type ) ( fruB57 : : Mu dJ ptsH15 ) Uptake ofb : [ 14C ] fructose None 2.4 ± 0.7 0.6 ± 0.1 0.1 ± 0.03 Fructose 9 ± 4 6 ± 3 0.1 ± 0.03 0.2 ± 0.04 0.1 ± 0.03 ND 0.2 ± 0.2 ND 0.1 ± 0.03 0.06 ± 0.04 0.02 ± 0.01 0.10 ± 0.04 0.11 + 0.03 0.2 0.1 0.2 0.2 0.2 Mannitol 1.9 ± 0.6 0.5 ± 0.3 0.1 ± 0.03 None 4.3 ± 1.0 0.4 ± 0.2 5.0 ± 2.0 Fructose 4.7 ± 1.7 4.5 ± 3.2 5.9 ± 1.9 Mannitol 12 ± 5 0.4 ± 0.2 13 ± 3 None 1.0 0.10 ± 0.01 1.6 ± 0.6 Fructose 2.3 ± 1.0 0.56 ± 0.10 1.2 + 0.3 None 1.0 0.05 ± 0.01 1.8 ± 0.3 Fructose 1.5 ± 0.5 0.2 ± 0.02 1.3 + 0.1 None 1.0 0.5 1.5 None 1.0 0.4 1.3 None 1.0 0.4 1.0 Fructose 0.8 ND 1.1 Mannitol 0.9 ND 1.0 Utilization ofg : Citrate Citrate + + + Alanine Alanine + + + Lactate Lactate + + + + a Strains are listed in Table 1 , and experimental procedures were as described in Materials and Methods .
All strains fermented galactose , xylose , and arabinose normally .
Maltose was fermented by strains SB1873 and LJ1182 but not by strains SB1475 and LJ1190 .
Error values are expressed in standard deviations ( three or more determinations ) .
b Determined with cells grown in LB-medium with or without the sugar inducer indicated .
Cells were harvested in the logarithmic growth phase , washed three times with medium 63 , and resuspended in the same salts medium at a cell density of 0.04 to 0.16 mg ( dry weight ) of cells per ml .
Values are expressed in micromoles of ` IC-sugar accumulated per minute per gram ( dry weight ) of cells at 37 °C .
The uptake experiments were conducted three times , each time with a distinct cell preparation .
c The value of 1.0 corresponds to a value of 30 nmol of [ 14CJmannitol phosphorylated at 37 ' C per min per mg of protein when HPr or FPr was measured and 140 nmol of [ ' 4C ] mannitol phosphorylated per min per mg of protein when enzyme I was measured .
Assays were conducted as described in Materials and Methods .
The phosphorylation assays were conducted three times , each time with a distinct extract .
The lower rates of phosphorylation in-vitro as compared with the in-vivo uptake rates are consistent with previous reports and are believed to be due to disruption of a more active complex of PTS enzymes during cell disruption ( 17 , 42 ) .
d The value of 1.0 corresponds to 0.31 , umol of cyclic AMP produced and secreted into the medium per g ( dry weight ) of cells during-growth to the stationary-phase as described in Materials and Methods ( 14 ) .
e The value of 1.0 corresponds to a value of 4 pLmol of [ ' 4C ] citrate taken up per min per g ( dry weight ) of cells .
Uptake was quantitated as described in Materials and Methods .
f The value of 1.0 corresponds to a swarm diameter of 25 mm in assays conducted as described in Materials and Methods .
g Estimated on minimal plates consisting of medium 63 ( 49 ) containing Bacto-Agar ( 1.5 % ) and the carbon source at a concentration of 0.5 % .
The diameter of the swarm of a given bacterial strain is a measure of the motility of that strain under the conditions used .
genetic background correlates with increased cyclic AMP production as reported previously ( 44 ) .
Increased adenylate cyclase activity is an unusual characteristic of crr mutants in this E. coli genetic background .
The results generally suggest that motility in semisolid agar medium can be used to estimate levels of cyclic AMP production in mutants altered for the regulation of adenylate cyclase activity .
DISCUSSION We have constructed a double mutant lacking both HPr and FPr activities , and this mutant has proven to possess drastically altered physiological properties .
It synthesizes extremely low levels of cyclic AMP , utilizes many carbon sources poorly , and is poorly motile ( Table 7 ) .
An effect of fru mutations on cyclic AMP production has been noted by Gershanovitch et al. ( 21 ) .
It is reasonable to suppose that these extreme physiological properties result from the strong inhibition of adenylate cyclase by the PTS ( 41 ) .
It is interesting to note that tight point and deletion ptsI mutants are not as .
deficient for cyclic AMP synthesis , for motility , or for gluconeogenic carbon utilization as the ptsHJS fruBS7 : : Mu dJ double mutant described here .
This fact may be attributed to the residual enzyme I activity of tight ptsI mutants as a result of the presence of a nearly cryptic gene encoding a second enzyme I that is negatively regulated by the fruR gene product ( 6 ) .
The fruR gene product , suggested but not yet conclusively shown to be the repressor of the fru regulon , exerts a pleiotropic effect on the expression of genes encoding proteins involved in gluconeogenesis , the glyoxylate stunt , oxidative metabolism , and Salmonella virulence ( 4-6 , 18 , 19 ; B. Hoertt , R. L. Warren , A. M. Chin , and M. H. Saier , Jr. , unpublished results ) .
Mutations in the fruR gene also give rise to elevated activities of enzyme I , HPr , and some of the enzymes II ( 4 ) .
Interestingly , loss of FPr due to the fruB57 : : Mu dJ mutation apparently exerts a similar effect ( Table 7 ) .
A product of the fru operon can possibly interact with or phosphorylate a component of the transcriptional apparatus controlling pts operon expression .
The simultaneous loss of HPr and FPr functions results in the loss of PTS-mediated transport function ( Table 7 ) .
In the fruB57 : : Mu dJ mutant lacking FPr but possessing wild-type levels of HPr , fructose uptake was also reduced to near basal activity .
Evidence published previously led to the conclusion that HPr could substitute for FPr in-vitro ( 54 ) .
The dramatic loss of transport function in the fruB57 : : Mu dJ mutant ( Table 7 ) can be explained either by the imprecise cleavage of FPr from the IIIFm-FPr fusion protein or by the loss of 11Fru due to a polar effect of the Mu dJ insertion element .
Mutations in the crp or cya gene that eliminate activity of CRP or adenylate cyclase , respectively , and pts mutations that reduce rates of cyclic AMP synthesis ( 14 , 41 , 42 ) resulted in depressed basal fru operon expression , as reflected by both in-vivo and in-vitro assays ( Tables 5 and 6 ) a well as in-vitro assays of P-galactosidase activities of frulacZ-transcriptional-fusions ( Table 3 ) .
This situation is reminiscent of that observed previously for synthesis of functional 11G1C and IlMan ( 36 ) .
Unlike the latter systems , however , the crp or cya mutations did not interfere with fructose-promoted induction of fru operon expression .
To the best of our knowledge , this is a unique finding .
The other regulons or operons which encode transport proteins or enzymes that initiate the catabolism of exogenous carbohydrates are not appreciably inducible by the inducer or substrates in tight crp or cya mutants .
The in-vitro results reported therefore explain the in-vivo transport results ( Table 7 ) as well as the positive fructose fermentation responses which have long been known to be characteristic of these mutants ( 50 ) .
Lowered cyclic AMP levels can not explain the lack offru operon induction in a pts deletion strain , A ( cysK-ptsHI41 ) ( Table 4 ) .
In this case , an insufficient supply of the presumed inducer , fructose-i-phosphate , may provide an explanation .
However , if fructose-i-phosphate is the true inducer , then the 3-galactosidase induction infruA : : Mu dJ mutants ( Table 2 ) is anomalous .
Two distinct mechanisms of induction , one by fructose and a second by fructose-i-phosphate , may be operative .
It is possible that even free fructose can not enter the cytoplasm of the pts deletion strain to a concentration sufficient to effect induction .
Alternatively , a cryptic fru gene encoding a second , poorly expressed IIFru may be present on the Salmonella chromosome ( Y. W. Cao and M. H. Saier , Jr. , unpublished results ) .
Three independently isolated Mu dJ insertion mutants , specifically defective for fructose utilization , lacked I1Fru activity but possessed measurable IIIFru and fructose-i-phosphate kinase activities ( Table 2 ) .
On this basis , the insertion elements were presumed to be in thefruA gene , the last known gene in the fru operon ( 19 , 35 ) .
However , the activities of fructose-i-phosphate kinase were high , whereas those of IIIFW were low .
Considering that the gene order is believed to be fruB ( MH ) KA , the noncoordinate expression of the fruB and fruK genes is anomalous .
Interestingly , this observation correlates with the low fruB gene expression and high fruK gene expression in a fruA : : Tn5 insertion mutant of Rhodobacter capsulatus ( 8 ) .
Since the gene order in R. capsulatus is fruB ( HI ) KA ( 54a ) ( similar to that in Salmonella species ) , this result correlates with those reported here but is not easy to interpret .
This observation contrasts with the report of Geerse et al. ( 19 ) , who found that a TnJO insertion mutation in S. typhimurium gave rise to loss of IIFni activity but retention of full lI1Fru and fructose-i-phosphate kinase activities .
Since the fruA : : Mu dJ mutants could ferment mannitol in a ptsHfruR genetic background , the fruB ( MH ) gene product must have been synthesized in appreciable amounts , in accordance with the report of Geerse et al. ( 19 ) .
Degradation of IllFru in the absence of IlFru or the presence of a second promoter preceding the fruH gene segment may provide an explanation .
Geerse et al. showed with a galK fusion that the AluIHinfl fragment in the region preceding the fruB gene ( nucle-otides 235 to 405 ) contains a fru operon regulatory region that is responsive to the presence of FruR , but they could not find a promoter sequence ( 18 ) .
Computer analyses using five different programs ( 31 , 32 ) have revealed a single -35 region and a single -10 region within this fragment ( TTGCGT at position 304 and CAGACT at position 327 , respectively ) .
The -10 region is the same as that for the E. coli p.fkB gene .
The sequence TGAAACGTTTCA ( positions 336 to 347 ) forms a hairpin structure , the only such structure anywhere within this region , and FruR might bind to this hairpin structure , thereby blocking transcription by RNA polymerase .
With regard to transcriptional regulation of thefru operon , the results summarized here and previously ( 6 ) , as well as unpublished results ( Cao and Saier ) suggest ( i ) that the fru regulon in S. typhimurium is complex ; ( ii ) that it may consist of at least three or four operons , one encoding the IIIFm-FPr protein , fructose-i-phosphate kinase , and IIFru , a second encoding a fructose-inducible enzyme I-like protein , a third encoding the fructose repressor , FruR , and possibly a fourth encoding a cryptic 11Fru ( Cao and Saier , unpublished results ) ; ( iii ) that it is unique in its response to the loss of the cyclic AMP-CRP complex as a result of mutations in the cya , crp , or pts gene ; ( iv ) that it may be regulated at the transcriptional level by two or more distinct mechanisms ( possibly involving two distinct inducers , fructose-i-phos-phate and fructose ) which together account for induction by fructose , or that low-level expression of cryptic fru genes accounts for the anomalous fru operon induction behavior ; and ( v ) that the proteins of the fructose-specific PTS may play a direct or indirect role in transcriptional regulation .
Some of these observations are reminiscent of transcriptional regulation in other operons encoding PTS proteins .
For example , the E. colipts operon , the nag regulon , and the gut operon have multiple promoters ( 10 , 33 , 34 , 55 ) , the E. coli bgl and Bacillus subtilis sac operons are controlled by PTS-mediated antiterminator phosphorylation ( 1 , 2 , 15 , 29 , 52 ) , and the E. coli gut operon is regulated by antagonistic transcriptional repression-activation ( 55 ) .
A concise discussion of transcriptional regulation involving PTS proteins and the PTS genetic apparatus has recently appeared ( 40 ) .
Thus , while the modes of fru regulon transcriptional regulation have not yet been delineated , mechanisms to be considered include ( i ) regulation involving PTS-mediated protein phosphorylation , ( ii ) antitermination , ( iii ) initiation at multiple promoters , ( iv ) classical repression or activation , and ( v ) an unusual ( if not unique ) interaction between the cyclic AMP-CRP complex and fructose-specific transcriptional regulatory proteins .
Further studies will be required to establish the mechanistic bases for the complex expressional patterns reported in this communication .
The diameter of the swarm of a given bacterial strain is a measure of the motility of that strain under the conditions used .
We thank Eduardo Groisman for assistance with some of strain constructions , Holman Massey and Gideon Sonstag , with 7 and 8 were whom some of the motility studies reported in Tables C. O'Neill for analyzing the fru operon operaconducted , Michael tor-promoter region , and Mary Beth Hiller for help with preparation of the manuscript .
This work was supported by Public Health Service grants 5 RO1 AI 21702 and 2 RO1 AI 14176 from the National Institute of Allergy and Infectious Diseases .
Amster-Choder , O. , F. Houman , and A. Wright .
Protein phosphorylation regulates transcription of the j-glucoside utili-zation operon in E. coli .
Aymerich , S. , and M. Steinmetz .
Cloning and preliminary of the sacS locus from Bacillus subtilis which characterization controls the regulation of the exoenzyme levansucrase .
Berg , C. M. , D. E. Berg , and E. A. Groisman .
Transposable elements and the genetic engineering of bacteria , p. 879-925 .
In D. E. Berg and M. M. Howe ( ed .
American Society for Microbiology , Washington , D.C. 4 .
Chin , A. M. , D. A. Feldheim , and M. H. Saier , Jr. 1989 .
Altere transcriptional patterns affecting several metabolic pathways in strains of Salmonella typhimurium which overexpress the fructose regulon .
Chin , A. M. , B. U. Feucht , and M. H. Saier , Jr. 1987 .
Evidence for regulation of gluconeogenesis by the fructose phosphotransferase system in Salmonella typhimurium .
Chin , A. M. , S. Sutrina , D. A. Feldheim , and M. H. Saier , Jr. 1987 .
Genetic expression of enzyme I activity of the phosphoenolpyruvate : sugar phosphotransferase system in ptsHI deletion strains of Salmonella typhimurium .
Cordaro , J. C. , and S. Roseman .
Deletion mapping of the genes coding for HPr and enzyme I of the phosphoenolpyruvate : sugar phosphotransferase system in Salmonella typhimu-rium .
Daniels , G. A. , G. Drews , and M. H. Saier , Jr. 1988 .
Properties of a Tn5 insertion mutant defective in the structural gene ( fruA ) of the fructose-specific phosphotransferase system of Rhodobacter capsulatus and cloning of the fru regulon .
den Blaauwen , J. L. , and P. W. Postma .
Regulation of cyclic AMP synthesis by enzyme IIIGIC of the phosphoenolpyruvate : sugar phosphotransferase system in crp strains of Salmo-nella typhimurium .
De Reuse , H. , and A. Danchin .
The ptsH , ptsI , and crr genes of the Escherichia coli phosphoenolpyruvate-dependent phosphotransferase system : a complex operon with several modes of transcription .
Deutscher , J. , and M. H. Saier , Jr. 1988 .
Protein phosphorylation in bacteria-regulation of gene expression , transport function , and metabolic processes .
Dills , S. S. , A. Apperson , M. R. Schmidt , and M. H. Saier , Jr. 1980 .
Carbohydrate transport in bacteria .
Ely , B. , R. M. Weppleman , H. C. Massey , Jr. , and P. E. Hartman .
Some improved methods in P22 transduction .
Feucht , B. U. , and M. H. Saier , Jr. 1980 .
Fine control of adenylate cyclase by the phosphoenolpyruvate : sugar phosphotransferase systems in Escherichia coli and Salmonella typhi-murium .
Fouet , A. , M. Arnaud , A. Klier , and G. Rapoport .
Genetics of the phosphotransferase system of Bacillus subtilis .
The phosphoenolpyruvate-initiated pathway of fructose metabolism in Escherichia coli .
A new assay of the phosphotransferase system in Escherichia coli .
Geerse , R. H. , F. Izzo , and P. W. Postma .
The PEP : fructose phosphotransferase system in Salmonella typhimu-rium : FPr combines Enzyme IIIFni and pseudo-HPr activities .
Geerse , R. H. , C. R. Ruig , A. R. J. Schuitema , and P. W. Postma .
Relationship between pseudo-HPr and the PEP : fructose phosphotransferase system in Salmonella typhimurium and Escherichia coli .
Geerse , R. H. , J. van der PluiJm , and P. W. Postma .
The repressor of the PEP : fructose phosphotransferase system is required for the transcription of the pps gene of Escherichia coli .
Gershanovitch , V. N. , T. N. Boishakova , M. L. Moichanova , A. M. Umyarov , 0 .
Y. Dobrynina , Y. A. Grigorenko , and R. S. Erlagaeva .
Fructose-specific phosphoenolpyruvate dependent phosphotransferase system of Escherichia coli : its alterations and adenylate cyclase activity .
Groisman , E. A. , and M. J. Casadaban .
Cloning of genes from members of the family Enterobacteriaceae with mini-Mu bacteriophage containing plasmid replicons .
Heldin , C.-H. , C. Betsholtz , L. Claesson-Welsh , and B. Westermark .
Subversion of growth regulatory pathways in malignant transformation .
Hughes , K. T. , and J. R. Roth .
Transitory cis complementation : a method for providing transposition functions to defective transposons .
Jacob , F. , and J. Monod .
Genetic regulatory mechanisms in the synthesis of proteins .
Kofoid , E. C. , and J. S. Parkinson .
Transmitter and receiver modules in bacterial signaling proteins .
Fructose transport by Escherichia coli .
H. , N. J. Rosebrough , A. L. Farr , and R. J. Randall .
Protein measurement with Folin phenol reagent .
Mahadevan , S. , and A. Wright .
A bacterial gene involved in transcription antitermination : regulation at a Rho-indepen-dent terminator in the bgl operon of E. coli .
Experiments in molecular genetics .
Cold Spring Harbor Laboratory , Cold Spring Harbor , N.Y. 31 .
I. Consensus as it relates to spacing class , specificity , repeat substructure , and three-dimensional organization .
O'Neill , M. C. , and F. Chiafari .
A spacing class-dependent promoter search protocol .
Peri , K. G. , and E. B. Waygood .
Sequence of cloned Enzyme 11NAG of the phosphoenolpyruvate : N-acetylglucos-amine phosphotransferase system of Escherichia coli .
Sequence of the nagBACD operon in Escherichia coli K12 and pattern of transcription within the nag regulon .
Prior , T. I. , and H. L. Kornberg .
Nucleotide sequence of fruA , the gene specifying Enzyme IIfm of the phosphoenolpyru-vate-dependent sugar phosphotransferase system in Escherichia coli K12 .
Rephaeli , A. W. , and M. H. Saier , Jr. 1980 .
Regulation of genes coding for enzyme constituents of the bacterial phosphotransferase system .
Reznikoff , W. S. , D. A. Siegele , D. W. Cowing , and C. A. Gross .
The regulation of transcription initiation in bacteria .
Saffen , D. W. , K. A. Presper , T. L. Doering , and S. Roseman .
Sugar transport by the bacterial phosphotransferase system .
Saier , M. H. , Jr. 1985 .
Mechanisms and regulation of carbohydrate transport in bacteria .
Academic Press , Inc. , New York .
Saier , M. H. , Jr. 1989 .
Involvement of the bacterial phosphotransferase system in diverse mechanisms of transcriptional regulation .
Saier , M. H. , Jr. 1989 .
Protein phosphorylation and allosteric control of inducer exclusion and catabolite-repression by the bacterial phosphoenolpyruvate : sugar phosphotransferase system .
Saier , M. H. , Jr. , D. F. Cox , B. U. Feucht , and M. J. Novotny .
Evidence for the functional association of Enzyme I and HPr of the phosphoenolpyruvate-sugar phosphotransferase system with the membrane in sealed vesicles of Escherichia coli .
Saier , M. H. , Jr. , G. A. Daniels , P. Boerner , and J. Lin .
Neutral amino-acid transport systems in animal cells : potential targets of oncogene action and regulators of cellular growth .
Saier , M. H. , Jr. , and B. U. Feucht .
Coordinate regulation of adenylate cyclase and carbohydrate permeases by the phosphoenolpyruvate : sugar phosphotransferase system in Salmo-nella typhimurium .
Saier , M. H. , Jr. , B. U. Feucht , and W. K. Mora .
Sugar phosphate : sugar transphosphorylation and exchange group translocation catalyzed by the Enzyme II complexes of th bacterial phosphoenolpyruvate : sugar phosphotransferase system .
Saier , M. H. , Jr. , B. U. Feucht , and S. Roseman .
Phos-phoenolpyruvate-dependent fructose phosphorylation in photosynthetic bacteria .
Saier , M. H. , Jr. , F. C. Grenier , C. A. Lee , and E. B. Waygood .
Evidence for the evolutionary relatedness of the proteins of the bacterial phosphoenolpyruvate : sugar phosphotransferase system .
Saier , M. H. , Jr. , and S. Roseman .
The crr mutation : its effect on repression of enzyme synthesis .
Saier , M. H. , Jr. , R. D. Simoni , and S. Roseman .
The physiological behavior of Enzyme I and heat-stable protein mutants of a bacterial phosphotransferase system .
Saier , M. H. , Jr. , R. D. Simoni , and S. Roseman .
Properties of mutant bacteria defective in proteins of the phosphoenolpyruvate : sugar phosphotransferase system .
Saier , M. H. , Jr. , M. Yamada , B. Erni , K. Suda , J. Lengeler , R. Ebner , P. Argos , B. Rak , K. Schnetz , C. A. Lee , G. C. Stewart , F. Breidt , Jr. , E. B. Waygood , K. G. Peri , and R. F. Doolittle .
Sugar permeases of the bacterial phosphoenolpyruvate ¬ dependent phosphotransferase system : sequence comparisons .
Schnetz , K. , and B. Rak .
Regulation of the bgl operon of Escherichia coli by transcriptional antitermination .
Schnetz , K. , C. Toloczyki , and B. Rak .
P-Glucoside ( bgl ) operon of Escherichia coli K-12 : nucleotide sequence , genetic organization , and possible evolutionary relationship to regulatory components of two Bacillus subtilis genes .
Sutrina , S. L. , A. M. Chin , F. Esch , and M. H. Saier , Jr. 1988 .
Purification and characterization of the fructose-inducible HPrlike protein , FPr , and the fructose-specific enzyme III of the phosphoenolpyruvate : sugar phosphotransferase system of Sal-monella typhimurium .
Wu , L.-F. , J. M. Tomich , and M. H. Saier , Jr. 1990 .
Structure and evolution of a multidomain multiphosphoryl transfer protein .
Nucleotide sequence of the fruB ( HI ) gene in Rhodobacter capsulatus and comparisons with homologous genes from other organisms .
Yamada , M. , and M. H. Saier , Jr. 1988 .
Positive and negative regulators for glucitol ( gut ) operon expression in Escherichia coli .