158 , No. 1 of Growth Temperature on the Acquisition of Iron Salmonella typhimurium and Escherichia coli by PATRICIA L. WORSHAM AND JORDAN KONISKY * Department of Microbiology , University of Illinois , Urbana , Illinois 61801 We have examined the effect of growth temperature on three systems normally induced under conditions of iron limitation : synthesis of the siderophore enterochelin ( enterobactin ) , transport of ferric enterochelin , and production of the outer membrane protein which serves as the colicin I receptor .
We found that although Salmonella typhimurium produces less enterochelin when grown at 42 °C , synthesis of this siderophore was not diminished in Escherichia coli grown under the same conditions .
Growth at 42 °C under a condition of iron stress led to a reduction in the ability of cells to transport ferric enterochelin in both organisms .
A two-to threefold decrease in the number of colicin I receptors was observed in cells of E. coli or S. typhimurium grown at 42 °C as compared with the number of receptors observed in cells grown at 37 °C .
The colicin I receptor was shown not to be inherently unstable at 42 °C .
By using a cir-lacZ operon fusion , it was shown that at least part of the decrease in receptor levels found in cells grown at high-temperature was the result of decreased transcription of cir , the receptor structural gene .
The effect of growth temperature on these systems was shown to be independent of fur , a regulatory element which mediates their enhanced production in response to iron stress .
We suggest that a second regulatory element common to gene products involved in iron sequestration may be responsible for temperature regulation of these systems .
When deprived of iron , Salmonella typhimurium and Escherichia coli synthesize the high-affinity iron chelator enterochelin ( enterobactin ) , as well as produce a specific transport system for this siderophore ( 25 ) .
In E. coli , the biosynthesis of enterochelin and the production of the 81,000-dalton outer membrane protein that functions as receptor for ferric enterochelin are coordinately regulated by the amount of intracellular iron ( 22 , 25 ) .
The levels of two other outer membrane polypeptides appear to be similarly regulated : an 83,000-dalton protein with no known function and the 74,000-dalton cir gene product which is the receptor for colicins Ta and lb ( 20 , 22 , 25 ) .
The colicin I receptor is not required for transport offerric enterochelin ( 29 ) .
This protein may serve as receptor for a siderophore which has yet to be identified .
Proteins of a similar size are observed in outer membranes prepared from cultures of S. typhimurium grown in low-iron medium ( 3 , 9 ) .
The regulation of synthesis of these polypeptides and of enterochelin is at the level of transcription ( 16 , 33 ) .
In both E. coli and S. typhimurium , regulatory mutants have been isolated which constitutively produce enterochelin , as well as the three outer membrane proteins normally observed only in cells grown under iron limitation ( 9 , 16 ) .
Such mutants have been designated fur ( ferric uptake regulation ) .
Since the loci affected by this mutation are located in different regions of the chromosome ( 2 ) , it seems likely that some regulatory element under fur control is a diffusible gene product .
Whether the fur gene product itself is a diffusible regulatory element has not been established .
Evidence for a relationship between growth temperature and the ability to sequester iron has been presented by several groups .
Many gram-negative bacteria do not grow well at elevated temperatures unless the growth medium is supplemented with iron ( 6 , 11-13 , 17 ) .
However , the nature of this apparently widespread defect in iron acquisition at high-temperatures has not been elucidated .
In S. typhimurium , the requirement for additional iron is due , at least in part , to decreased biosynthesis of the phenolate siderophore enterochelin at the elevated temperature ( 12 ) .
Temperature-sensitive synthesis of a hydroxymate siderophore has also been reported in a fluorescent pseudo-monad ( 11 ) .
Thus , the defect in the acquisition of iron is not peculiar to the biosynthetic pathway of either phenolate or hydroxymate siderophores .
We initiated this study to define further the relationship between growth temperature and three elements known to be regulated by the amount of intracellular iron : excretion of the siderophore enterochelin , the rate of ferric enterochelin uptake , and the level of colicin I receptors .
( Preliminary reports of these data were presented at the 81st Annual Meeting of the American Society for Microbiology , Atlanta , Ga. , 7 to 12 March 1982 , and at the 82nd Annual Meeting of the American Society for Microbiology , New Orleans , La. , 6 to 11 March 1983 .
) MATERIALS AND METHODS Bacterial strains .
Strains used in these experiments are listed in Table 1 .
LB and LB agar have been previously described ( 23 ) .
Minimal medium M63 was prepared as described by Miller ( 23 ) , except that FeSO4 was omitted .
Chelex 100 resin ( 200 to 400 mesh ; Bio-Rad Laboratories ) was used to remove iron from the medium by the method of Murphy ( 24 ) .
The resin was removed by filtration through Sterilfil D-65 filters ( 0.22 pum ; - Millipore Corp. ) .
Casein hydrolysate ( 0.15 % ; Difco Laboratories ) was added to the medium before adding the resin .
Glycerol ( 0.4 % ) was used as the carbon source for all experiments .
Thiamine ( 1 , ug/ml ) and nicotinic acid ( 3 ILg/ml ) were also added to the minimal-medium .
Where necessary , tryptophan ( 40 , ug/ml ) was added to allow-growth of Trp-strains .
Glassware was washed with 2 N HCI for 24 h and rinsed with 4 volumes of glass-distilled water before use .
Where possible , polystyrene or polycarbonate tubes , flasks , and pipettes were used .
We have examined the effect of growth temperature on three systems normally induced under conditions 16 Where indicated , ethylenediamine-di ( orthohydroxyphenyla-cetic acid ) ( EDDA ; Sigma Chemical Co. ) was added to deplete the medium of free iron .
EDDA was deferrated by the method of Rogers ( 28 ) , and stock solutions were prepared as previously described by Ong et al. ( 26 ) .
Colicin I binding assays ; Procedures for purification and radioiodination of colicin la have been described previously ( 19 ) .
Unless otherwise indicated , binding assays were performed on whole-cells harvested in mid-to late-exponential phase as previously described ( 19 ) .
Smooth strains of E. coli and S. typhimurium are only slightly sensitive to colicin Ia , and binding of the colicin to these strains is poor .
Therefore , a rough strain of S. typhimurium and a K-12 strain of E. coli were chosen for experiments involving colicin binding .
Sensitivity to colicin Ia .
Approximate colicin titers were evaluated by the serial dilution method of Guterman ( 15 ) .
Overnight cultures of strains PW401 and JK1 which had been grown at 37 °C were diluted to an absorbance at 600 nm of 0.25 .
The cell suspension ( 100 , ul ) was then used to inoculate flasks containing 10 ml of the same medium which had been prewarmed to 37 Qr 42 °C .
After six generations of growth , the cultures were assayed for 3-galactosidase activity as described by Miller ( 23 ) , using chloroform and sodium dodecyl sulfate to make the cells permeable to the substrate .
Isopropyl-p-D-thiogalactopyran-oside ( 1 mM ; Sigma ) was used to induce,-galactosidase in strain JK1 .
The Arnow reaction ( 1 ) was used as an assay for enterochelin .
2,3-Dihydroxybenzoic acid ( Sigma ) was used as the standard .
Culture supernatants were collected from stationary-phase cultures .
For each individual strain , all cultures were harvested at the same turbidity .
However , from strain to strain the turbidity varied from an absorbance at 600 nm of 1.5 to 4.0 .
Purification of enterochelin has been described previously ( 30 ) .
`` Fe-enterochelin was prepared as follows .
A ferric enterochelin solution ( 10 ml ; 320 nmol ) was acidified to pH 1.5 with concentrated HCl and then extracted twice with 10 ml of ethyl acetate .
Nitrogen gas was bubbled through the ethyl acetate solution until the volume was reduced to 1 ml .
Carrier-free 55FeCl ( 320 nmol ; 1.44 Ci/mol ; New England Nuclear Corp. ) was added next , and the pH of the solution was adjusted to 7.0 with 0.5 N NaOH .
The solution was then dried under vacuum , resuspended in 1.0 ml of M56 uptake medium which contained 100 , uM EDDA , and incubated at 37 °C for 6 h. Nine milliliters of M56-100 FtM EDDA was added next , and the solution was filtered through a 0.45 - , um Millex-HA filter ( Millipore Corp. ) .
The transport medium , M56 , was prepared by the method of Langman et al. ( 21 ) , except that 100 , g of EDDA per ml was added to chelate free iron .
Cells were harvested during lateexponential-growth ( absorbance at 600 nm , 0.9 ) and spun at 10,000 rpm in a Sorvall SS31 rotor for 15 min at 4 °C .
The pellet was washed three times with ice-cold M56 containing 9.2 % glucose .
Finally , the cells were resuspended in this same buffer .
They were kept on ice no longer than 30 min before use .
Immediately before each assay , the cell suspension and the solution containing 55Fe-enterochelin ( 1.5 x 105 cpm/nmol ) were warmed to 26 °C .
Each time point consisted of a separate reaction mixture which was prepared by mixing equal volumes ( 100 , ul ) of cell suspension and 5Fe-entero-chelin solution .
Transport was terminated at the appropriate time by the addition of 3 ml of ice-cold LiCl ( 0.1 M ) - EDTA ( 100 , uM ) solution before the sample was rapidly filtered through 0.45 - , uM nitrocellulose filters ( type BA85 ; Schlei-cher & Schuell ) .
Reaction tubes were rinsed three times with 3 ml of LiCI-EDTA solution , and the rinse fluids were a Growth of cells and determination of enterochelin are described in the text applied to the filter .
The filters were dried , and radioactivity associated with the filters was determined by liquid scintillation counting .
Before use , filters were soaked for at least 2 h at room temperature in LiCl-EDTA solution containing 50 , uM FeCl3 and 1 FtM ferric enterochelin .
Immediately before use , filters were rinsed twice with 3 ml of LiCI-EDTA solution .
These steps decreased nonspecific association of radioactivity with the filter .
RESULTS Effect of growth temperature on production of enterochelin .
It has been reported that enterochelin biosynthesis , in both E. coli and S. typhimurium , is inhibited at temperatures greater than 40 °C ( 12 , 18 ) .
We were unable to reproduce these results in E. coli when using K-12 strains , a recently isolated strain of human colonic origin ( CR ) , or several clinical isolates ( LL2 , LL3 , LL5 , LL6 , LL7 ) ( Table 2 ) .
Under the same growth-conditions , the culture fluids of S. typhimurium exhibited decreased levels of enterochelin at 42 °C .
Enterochelin synthesis in both Fur + and Fur-strains of S. typhimurium was diminished at 42 °C , indicating that the effect of growth temperature on enterochelin production is independent of the regulatory element Fur which mediates the response to iron .
Effect of growth temperature on the transport of ferric enterochelin .
To determine whether growth temperature affects the ability of cells to transport ferric enterochelin , cultures of E. coli MC4100 and S. typhimurium G30 ( Fur + ) and RB338 ( Fur - ) grown in M63 medium at 37 and 42 °C were assayed for ability to transport this siderophore .
The initial rates of 55Fe-enterochelin uptake in cells grown at these temperatures are illustrated in Fig. 1 .
In E. coli MC4100 and in S. typhimurium G30 ( Fur ' ) , we observed no difference in the rate of 55Fe-enterochelin transport in cells grown at the two temperatures .
However , in S. typhimurium RB338 ( fur ) the rate of transport in cells grown at 42 °C is approximately one-half of that in cells grown at 37 °C .
S. typhimurium RB338 , which carries the fur mutation , involved in iron expresses gene products transport in a manner analogous to a Fur ' strain which is under iron limitation ( 3 , 9 ) .
Only the Fur-strain exhibited a decreased rate of 55Fe-enterochelin transport in cells grown at 42 °C , suggesting that in cells grown under conditions of iron stress , ferric enterochelin transport is affected by growth temperature ( Fig. 1 ) .
To test this hypothesis , we carried out similar transport assays but used 2 , M EDDA in the growth medium to decrease the amount of available iron .
The difference in initial transport rates of cells grown at 42 versus 37 °C observed in the fur strain RB338 was also found in E. coli MC4100 and S. typhimurium G30 when the organisms were grown in medium treated to remove much of the free iron ( Fig. 2 ) .
Effect of growth temperature on the colicin I receptor .
Although the function of the colicin I receptor in iron metabolism is unknown , the production of this protein is coordinately regulated with the biosynthesis of enterochelin and the ferric enterochelin receptor ( 20 , 25 ) .
Therefore , we examined the amount of this receptor in cells grown at 37 and 42 °C .
In S. typhimurium , we found that the level of colicin I receptors is relatively low compared with that of E. coli K-12 , unless the cells are grown in medium which has a Growth of cells and determination of P-galactosidase are described in the text .
been treated to reduce the concentration of iron .
However , due to decreased synthesis of enterochelin at elevated temperatures , S. typhimurium does not grow rapidly at 42 °C under conditions of iron limitation .
To minimize the difference in growth-rate between cultures incubated at 37 and at 42 °C , we used strain PW734 , afur strain of S. typhimurium for this experiment , and grew the organism in M63 medium which had not been treated to lower the concentration of iron .
This strain produces high levels of the receptor even in growth medium that is not depleted of iron ( 9 ) .
The effect of growth temperature on the number of colicin I receptors per cell is shown ( Fig. 3 ) .
In both E. coli and S. typhimurium , there was a two-to threefold decrease in binding of colicin Ta to cells grown at 42 °C as compared with binding to cells grown at 37 °C .
A corresponding decrease in sensitivity to the colicin was observed ( data not shown ) .
The decrease in receptor level in the Fur-strain RB338 indicates that the effect of growth temperature on the number of colicin I receptors , like that on enterochelin synthesis and transport , is independent of the fur gene product .
It is important to point out that since growth-rates at 37 and at 42 °C were similar , the differences in receptor levels observed in cultures grown at the two temperatures are not growth-rate dependent .
To determine whether the receptor was thermolabile , cells of either E. coli or S. typhimurium grown at 37 °C were shifted to 42 °C in the presence of chloramphenicol , so as to inhibit protein synthesis and cell growth , and then assayed for receptor .
We did not observe a significant decrease in the amount of receptor activity in either strain ( data not shown ) .
Thus , the colicin receptor is not inherently unstable at 42 °C .
We have previously described operon fusions of cir , in which expression of,-galactosidase is controlled by the cir promoter ( 33 ) .
The Mu phage used in the construction of cirlac fusion E. coli PW400 has a temperature-sensitive repressor , thus the strain is not stable at temperatures greater than 33 °C ( 5 , 33 ) .
To observe temperature effects on the regulation of cir in E. coli K-12 PW400 , we introduced plasmid pJB4J1 , which carries a Mu phage with a wild-type repressor ( 4 ) .
E. coli PW401 , a derivative of PW400 which carries this plasmid , is thermostable , as evidenced by the fact that a culture of PW401 grown at 30 °C , diluted , and plated on LB produced the same number of colonies on LB plates incubated at 30 or 42 °C .
Furthermore , the generation time of this strain at 42 °C is identical to that of the parental strain MC4100 .
The effect of growth temperature on,-galacto-sidase levels in strain PW401 is shown in Table 3 .
When incubated at 42 °C , PW401 exhibited approximately one-half the amount of,-galactosidase activity that was observed in the culture incubated at 33 or 37 °C .
P-Galactosidase levels in the control cir + lac + E. coli JK1 did not respond to growth temperature under these conditions .
This experiment sug gests that the influence of growth temperature on receptor production is at least partially mediated at the level of transcription .
Bacterial strains used in this study Strain characteristics Reference E. coli JK1 MC4100 W3110 K-12 X rpsL K-12 araD139 AlacUJ69 thi rpsL MC4100 ( F ( cir-lac ) PW400 ( pJB4J1 ) Clinical isolates This laboratory ( 5 ) PW400 PW401 ( 33 ) ( 33 ) LL2 through LL7 CR J. Lembkea Human colon isolate This laboratory S. typhimurium G30 LT2 galE ( 9 ) RB338 LT2 fur ( 9 ) PW734 Rough derivative of This laboratory RB338 TN1073 zbi-812 : : TnlO trp43 galE K. G. Sandersonb a School of Veterinary Medicine , University of Illinois , Urbana .
b Salmonella Genetic Stpck Center .
Enterochelin production in cells grown at various temperatures Phenolate GrowthCPhconcn temp COC ) ' / i Strain E. coli K-12 MC4100 37 42 18 17 W3110 37 42 20 29 E. coli natural isolates CR 37 42 15 18 LL2 37 42 22 18 LL3 37 42 7 10 LL5 37 42 17 18 LL6 37 42 8 8 LL7 37 15 14 42 S. typhimurium G30 ( LT2 derivative ) 37 20 42 4 RB338 ( LT2 fur ) 37 48 42 14 TN1073 37 23 42 12 A 0 20 ) 0 E E 160 E CL .
S-c 4 ) 120 U 0 ) o C w 80 ) 0 in o 4 ) 0c 40 0 I I 90-180-270-360 450 Time ( sec ) FIG. 1 .
Effect of growth temperature on transport of 5Fe-enterochelin .
Cells were harvested and transport assays were performed as described in the text .
Growth temperatures used were 42 ( U ) and 37 °C ( 0 ) .
Strains used were ( A ) E. coli MC4100 , ( B ) S. typhimurium G30 , and ( C ) S. typhimurium fur RB338 .
¬ - B ¬ c A E 8000 4000 E ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ E 6400 3200 E 2400 X 4800 1600 ~ 3200 ¬ U - ~ ~ 1600 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ -800 A 90 90 180-270-360-450 180-270-360-450 Time ( sec ) FIG. 2 .
Effect of growth temperature on transport of `` Fe-enterochelin in cells grown under iron limitation .
Cells were grown in M63 medium containing 2 , uM EDDA .
Cells were harvested and transport assays were carried out as described in the text .
Growth temperatures used were 42 ( U ) and 37 °C ( 0 ) .
Strains used were ( A ) E. coli MC4100 and ( B ) S. typhimurium G30 TABLE 3 .
cir promoter activity in cells grown at 37 and 42 °C Growth P-Galactosidase Strain Strain ~ ~ ~ temp ( ` C ) a ( U ) a PW401 4 ( cir-lac ) 37 331 42 180 JK1 cir + lac + 37 1,100 42 1,186 12 0 a 10 \ C ~ 0 \ O 0 6 0 3 0 32 34 36 38 40 4 Growth Temperature ( IC ) 2 FIG. 3 .
Effect of growth temperature on the level of colicin I receptors in strains PW734 and JK1 .
Overnight cultures grown in M63 at 33 °C were diluted to a turbidity at 600 nm of 0.3 .
and then diluted 150-fold further into M63 medium which had been prewarmed to the desired temperature .
After eight generations of growth at the appropriate temperature , the cells were harvested and assayed for binding of 125I-labeled colicin Ia as described in the text .
The values shown represent the amount of colicin bound at saturation .
Symbols : ( 0 ) , S. typhimurium PW734 ; ( - ) , E. coli JK1 .
DISCUSSION Although we were able to reproduce the finding of Gari-baldi ( 12 ) in regard to diminished enterochelin production by S. typhimurium at elevated temperatures , we did not observe this phenomenon in E. coli K-12 or in several clinical isolates of this species .
Our data conflict in this regard with a previous report ( 18 ) , which purported to demonstrate reduced enterochelin synthesis by an E. coli strain at elevated temperatures .
However , interpretation of these published experiments is subject to some ambiguity since production of enterochelin was not evaluated by biochemical means .
We can not rule out the possibility that some strains of E. coli do exhibit temperature-sensitive production of enterochelin .
Growth at 42 °C of E. coli or S. typhimurium under a condition of iron stress also led to a decrease in the initial rate of ferric enterochelin transport .
The same phenomenon was observed in a fur mutant of S. typhimurium grown in medium not depleted of iron .
We suggest , therefore , that the gene product ( s ) which determines the rate of ferric entero-chelin transport in cells under iron stress is affected by growth at elevated temperatures .
Whether this is due to an effect of growth temperature on the activity or level of a component of the ferric enterochelin transport system remains to be established .
Whatever the cause , it is unrelated to growth-rate , since nearly identical growth-rates were observed for E. coli grown at 37 and at 42 °C , as well as for the S. typhimurium fur mutant .
It is interesting that there was no difference in initial transport rates of cells grown at 37 and at 42 °C when the cells were grown in medium not depleted of iron .
It is possible that the element which determines the rate of ferric enterochelin transport in cells grown under iron limitation is distinct from that which determines the transport rate in cells which are not grown under iron stress .
Growth temperature also affected the level of colicin I receptors .
When grown at 42 °C , cells of both E. coli K-12 and S. typhimurium LT2 had one-half to one-third the number of receptors observed in cells grown at 37 °C .
This decrease was not the result of thermal lability of the receptor at the higher temperature .
Although modulation of 3-galac-tosidase levels in strains carrying cir-lac operon fusions may not accurately reflect all aspects of the regulation of cir , the decrease in galactosidase synthesis in such a strain grown at 42 °C suggests that at least part of the decrease in receptor level is due to reduced transcription at this temperature .
However , the possibility that a posttranscriptional regulatory mechanism may play a part in this phenomenon is not excluded .
Since the regulatory elementfur , which appears to control the transcription of these genes in response to levels of iron ( 16 ) , affects the same systems which are subject to regulation by temperature , it was reasonable to entertain the possibility that the temperature effect is mediated by the fur gene product .
However , the temperature regulation observed in a fur strain of S. typhimurium , which constitutively produces enterochelin , the transport system for ferric enterochelin , and the colicin Ia receptor , rules this out .
It is tempting to speculate , therefore , that a second regulatory factor common to gene products involved in iron sequestration is affected at elevated temperatures .
The observation that many gram-negative bacteria are unable to sequester iron efficiently when grown at higher temperatures suggests that such an element may be widespread and highly conserved .
It has been suggested that fever may act as a host defense mechanism by depriving invading organisms of the iron necessary for growth ( 31 , 32 ) .
Since the ability to acquire iron does appear to be a virulence factor in some bacteria ( 7 , 8 , 10 , 27 , 28 , 34 ) , it would be interesting to examine the efficiencies of iron sequestration systems at elevated temperatures in a variety of pathogens .
An organism which is capable of efficient production and transport of siderophores at febrile temperatures is likely to be a more successful pathogen .
The reduced initial rate of ferric enterochelin uptake in E. coli does not appear to inhibit growth at 42 °C .
We do have E. coli strains in our collection that do not grow well at 42 °C .
However , growth of these strains at this temperature is not stimulated by the addition of ferric citrate or ferric chloride to the medium and appears to be unrelated to acquisition of iron .
In contrast , the decrease in enterochelin production exhibited by S. typhimurium does lead to an inhibition of growth at elevated temperatures under conditions of iron limitation .
The effect that a two-to threefold decrease in enterochelin synthesis , transport of ferric enterochelin , and levels of the colicin Ia receptor has on growth of the organism in-vivo is uncertain .
Enterochelin appears to be essential for the rapid growth of E. coli and S. typhimurium in-vivo ( 28 , 34 ) .
Furthermore , pathogenic strains of E. coli growing in-vivo produce high levels of two outer membrane proteins : an 81,000-dalton protein , likely to be the ferric enterochelin receptor , and a 74,000-dalton protein , which may correspond to the colicin I receptor ( 14 ) .
These data suggest that these outer membrane proteins may play a role in the adaption of the organism to growth in-vivo .
Thus , the inhibition of enterochelin synthesis and transport and the decrease in levels of the colicin I receptor at febrile temperatures could conceivably prove detrimental to the in-vivo growth of both E. coli and S. typhimurium .
P.L.W. is a recipient of National Institutes of Health predoctoral traineeship grant GM 7283 .
We thank L. Lembke for the clinical isolates of E. coli used in this study .
LITERATURE CITED marine fish pathogen Vibrio anguillarum specifies an ironsequestering system .
Ernst , J. F. , R. L. Bennett , and L. I. Rothfield .
Constitutive expression of the iron-enterochelin and ferrichrome uptake systems in a mutant strain of Salmonella typhimurium .
Finkelstein , R. A. , and R. J. Yancey .
Effect of sidero-phores on virulence of Neisseria gonorrhoeae .
Influence of temperature on the iron metabolism of a fluorescent pseudomonad .
Influence of temperature on the biosynthesis of iron transport compounds by Salmonella typhimurium .
Grieger , T. A. , and M. J. Kluger .
Fever and survival : the role of serum iron .
Griffiths , E. , P. Stevenson , and P. Joyce .
Pathogenic Escherichia coli express new outer membrane proteins when growing in-vivo .
Colicin B : mode of action and inhibition by enterochelin .
Regulation of ferric-iron transport in Esche-richia coli K12 : isolation of a constitutive mutant .
Kluger , M. J. , and B. A. Rothenberg .
Fever and reduced iron : their interaction as a host defense response to bacterial infection .
Role of iron in the regulation of nutritional immunity , p. 55-77 .
In K. N. Raymond ( ed .
American Chemical Society , Washington , D.C. 19 .
Konisky , J. , and B. S. Cowell .
Interaction of colicin Ia with bacterial cells .
Konisky , J. , S. Soucek , K. Frick , J. K. Davies , and C. Hammond .
Relationship between the transport of iron and the amount of specific colicin Ia membrane receptors in Escherichia coli .
Langman , L. , I. G. Young , G. E. Frost , H. Rosenberg , and F. Gibson .
Enterochelin system of iron transport in Esche ¬ richia coli : mutations affecting ferric-enterochelin esterase .
McIntosh , M. A. , and C. F. Earhart .
Coordinate regulation by iron of the synthesis of phenolate compounds and three outer membrane proteins in Escherichia coli .
Experiments in molecular genetics .
Cold Spring Harbor Laboratory , Cold Spring Harbor , N.Y. 24 .
Murphy , J. R. , J. L. Michel , and M. Teng .
Evidence that the regulation of diphtheria toxin production is directed at the level of transcription .
Microbial envelope proteins related to iron .
Ong , S. A. , T. Peterson , and J. B. Neilands .
Agrobactin , a siderophore from Agrobacterium tumefaciens .
Payne , S. M. , K. K. Holmes , and R. A. Finkelstein .
Role of iron in disseminated gonococcal infections .
Iron-binding catechols and virulence in Escherichia coli .
Soucek , S. , and J. Konisky .
Normal iron-enterochelin uptake in mutants lacking the colicin I outer membrane receptor protein of Escherichia coli .
Wayne , R. , and J. B. Neilands .
Evidence for common binding sites for ferrichrome compounds and bacteriophage 080 in the cell envelope of Escherichia coli .
Roles of iron in fever and infectious disease , p. 105-110 .
In B. Cox , P. Lomax , A. S. Milton , and E. Schonbaum ( ed .
) , Thermoregulatory mechanisms and their therapeutic implications .
Worsham , P. L. , and J. Konisky .
Use of cir-lac operon fusions to study transcriptional regulation of the colicin Ia receptor in Escherichia coli K-12 .
Yancy , R. J. , S. A. L. Breeding , and C. E. Lankford .
Enterochelin ( enterobactin ) : virulence factor for Salmonella typhimurium .