Vol. 151, No. 1JOURNAL OF BACTERIOLOGY, July 1982, p. 288-2940021-9193/82/070288-07$02.00/0

Active Transport of Ferric Schizokinen in Anabaena sp.

PETER J. LAMMERSt AND JOANN SANDERS-LOEHR*

Department of Chemistry, Portland State University, Portland, Oregon 97207

Received 27 October 1981/Accepted 13 February 1982

The cyanobacterium Anabaena sp. strain ATCC 27898 was found to utilize thesiderophore schizokinen to accumulate iron from the environment. This organismhad previously been shown to produce schizokinen under low-iron conditions,and we observed that the iron-transport capability is also increased in response toiron limitation. Uptake activity was specific for ferric schizokinen and did notoccur with ferrioxamine B. The uptake of ferric schizokinen displayed kineticstypical of a protein-mediated process with an apparent Km of 0.04 ,uM andsaturation at high concentrations of substrate. Light-driven transport was blockedby uncouplers and by ATPase inhibitors. Transport in dark-adapted cells was

additionally blocked by inhibitors of respiration. We conclude that ATP serves as

an energy source for the cellular uptake of ferric schizokinen.

The efficient utilization of iron by microorga-nisms requires the production of high-affinitychelating agents known as siderophores to solu-bilize iron in a biologically available form (16,21). A wide variety of bacteria and fungi producesiderophores as well as membrane transportproteins that facilitate uptake of hydrophilicsiderophore complexes across nonpolar cellmembranes (21, 23). Most siderophore transportsystems require the input of cellular energy foriron uptake into cells (8, 9, 20, 22).

Cyanobacteria also produce siderophores tosatisfy their iron requirements. Field experi-ments by Murphy et al. showed that under ironlimitation, Anabaena flos-aquae produces a hy-droxamate siderophore which suppresses thegrowth of eucaryotic algae and promotes"bloom" formation (19). The production of asiderophore by Anabaena flos-aquae has beenconfirmed in an independently isolated axenicculture and extended to Anabaena cylindrica(17). In addition to freshwater species, the ma-rine cyanobacterium Agmenellum quadruplica-tum has shown siderophore activity (2). Further-more, the experiments of Lange on the ability ofmicroorganisms to grow in the absence of addedchelator suggest that many other cyanobacteriamay eventually be found to produce sidero-phores (15).The only cyanobacterial siderophore which

has been structurally characterized is schizo-kinen, a dihydroxamate produced by Anabaenasp. strain ATCC 27898 (24). Schizokinen is amember of the citrate-hydroxamate family ofsiderophores, which includes aerobactin and

t Present address: Department of Biophysics and Theoreti-cal Biology, University of Chicago, Chicago, IL 60637.

arthrobactin (16). The present study was under-taken to elucidate the role of schizokinen infacilitating iron uptake in Anabaena sp. strainATCC 27898. We have been able to show thatthis organism contains an iron uptake systemwhich has a high affinity for ferric schizokinenand whose activity increases in response to ironlimitation. In addition, we have found that thisiron transport requires metabolic energy. It hasbeen possible to probe the nature of this energyrequirement by using a variety of inhibitors.

MATERIALS AND METHODSGrowth of Anabaena sp. Anabaena sp. strain ATCC

27898 was maintained in axenic culture at 20°C and 800lx on 1.5% agar slants of BG-11 medium (26) contain-ing (per liter): 1.5 g of NaNO3, 40 mg of K2HPO4, 20mg of Na2CO3, 75 mg of MgSO4-7H2O, 36 mg ofCaCl2-2H20, 6 mg of citric acid, 6 mg of ferric ammo-nium citrate, 1 mg of EDTA, 2.9 mg of H3BO3, 1.8 mgof MnCl2 4H2O, 0.2 mg of ZnSO4 7H2O, 0.05 mg ofCo (NO3)2-6H2O, 0.08 mg of CuSO4 5H2O, and 0.4 mgof Na2MoO4 2H2O (pH 7.1). Liquid cultures weregrown on BG-11 medium without ferric ammoniumcitrate, and iron was added from a stock solution of 0.5mM FeCl3 in 10 mM HNO3, unless otherwise stated.The inoculum for the cultures was grown in 0.1 F.Miron. Cultures were incubated at 34°C and continuouslight (2,500 lx) from cool-white fluorescent lights (Syl-vania Electric Products Inc.). Growth was monitoredin a Klett-Summerson colorimeter fitted with a red no.66 filter (100 Klett units = 0.46 mg [dry weight] perml).Growth media were sterilized by passage through

0.22-p.m filters (GS, Millipore Corp.). All glasswarewas soaked in 6 N HCI and rinsed in glass-distilleddemineralized water (Milli-Q system, Millipore) tominimize iron contamination. Milli-Q water was alsoused in the preparation of growth media. Liquid BG-11culture medium with no added iron was analyzed by

288

on February 3, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

FERRIC SCHIZOKINEN TRANSPORT IN ANABAENA 289

atomic absorption spectroscopy and found to containless than 0.08 tLM iron. Anabaena cultures wereperiodically checked for contaminating bacteria byinoculation into nutrient agar and also by microscopicexamination of aged cultures.

Purification of schizokinen. Schizokinen was isolatedfrom culture supernatants of Bacillus megateriumATCC 19213. Cultures were grown in 500-ml batchesin 2-liter Erlenmeyer flasks in the medium describedby Mullis et al. (18). The medium was purged of ironby passage through a column of Chelex 100 (Bio-RadLaboratories) before the addition of calcium and mag-nesium salts. Maximum yields were obtained when10'- M FeCI3 was added to the medium. Cultureswere harvested after incubation for 72 h at 37°C withheavy aeration.

Schizokinen was extracted from 10 liters of superna-tant fluid after rotary evaporation to 250 ml as previ-ously described (18). The aqueous extract was appliedto a column (2.5 by 20 cm) of Bio-Rad AG-2-X1O in theacetate form (pH : 6.0), rinsed with 100 ml of distilledwater, and eluted with a linear gradient of 0.1 to 1.0 MNH4Cl. Schizokinen was detected by absorbance at490 nm after 0.1 ml of each fraction was added to0.9 ml of FeCI3 in 0.1 M HC104 (4). Peak fractionswere lyophilized. Subsequent gel filtration on Bio-GelP2 (1.5 by 90 cm) produced pure schizokinen in thefirst half of the ferric ion reactive peak. After lyophili-zation and dissolution in D20, this material yielded an1H nuclear magnetic resonance spectrum which wasidentical to those previously published for schizokinen(18, 24).

Schizokinen concentrations were determined by ti-trating a sample with a standardized solution of ferricchloride in dilute HCI (followed by immediate read-justment to pH 7.0) to an endpoint in the absorbance at390 nm. The titration was performed at neutral pHbecause schizokinen acetyl groups are labile uponprolonged exposure to acid.

Siderophore assays. The production of schizokinenby Anabaena sp. was measured by two differenttechniques. Hydroxamate groups in culture filtrateswere assayed by the method of Csaky (7) with N-(1-naphthyl)ethylenediamine used as the color-formingagent (27) after a 4-h acid hydrolysis at 120°C (12). Thepresence of intact siderophores was measured using abioassay with Arthrobacter flavescens JG-9 (ATCC25091), an auxotrophic mutant requiring siderophoreactivity for growth. Tube cultures of JG-9 were grownin the medium described by Estep et al. (10) at 30°C ona rotary shaker set at 200 rpm. Growth was measuredafter 24 h and found to be proportional to the concen-tration of added purified schizokinen. The half-maxi-mal response of 0.04 to 0.05 ,ug of schizokinen per mlwas identical to that reported by Lankford for schizo-kinen (16).

Iron transport assay. Anabaena filaments were har-vested from the exponential phase (0.2 to 0.4 mg [dryweight] per ml) by collection on 0.45-pm filters (HA,Millipore). The filaments were washed with, and sus-pended to a similar cell density in, Chelex 100-treateduptake medium containing (per liter): 1.10 g ofNaNO3, 50 mg of NH4C1, 250 mg of K2HPO4, 531 mgof MgSO4-7H2O, and 58 mg of CaCl2 2H2O (pH 7.0).Stock solutions of [55Fe]ferric schizokinen were pre-pared by mixing "5FeCl3 in 0.1 M HCI (New EnglandNuclear Corp.; 25.9 mCi/mg) with a twofold molar

excess of schizokinen from B. megaterium, adjustingthe pH to 7.0 with NaOH, and passing the solutionthrough a 0.45-,um filter. Assays were initiated by theaddition of ferric schizokinen (10 to 20 nM in "5Fe) andincubated at 34°C and 2,500 Ix unless otherwise noted.Samples were removed at timed intervals, rapidlyfiltered through 0.45-,um filters (HA, Millipore), andwashed with 1 volume of ice-cold uptake medium.Samples of filtrates were dispersed in Ready-Solv HPscintillation fluid (Beckman Instruments, Inc.) andcounted in a Beckman LS-9000 liquid scintillationcounter. Uptake was calculated from the loss ofcounts relative to filtrates of cell-free solutions con-taining the same initial concentration of [55Fe]ferricschizokinen. For Km determinations, the iron concen-tration was increased by adding unlabeled iron, andthe resultant uptake values were corrected for thedilution of the radioactive label. The twofold molarexcess of schizokinen to iron was utilized to ensurethat all of the added iron was coordinated. Even a 10-fold excess of schizokinen caused no significant de-crease in uptake activity.

Chemicals. Desferal (methane sulfonate salt of de-ferriferrioxamine B) was purchased from CIBA Phar-maceutical Co. N,N'-Dicyclohexylcarbodiimide(DCCD) and carbonyl cyanide-m-chlorophenyl hydra-zone (CCCP) were obtained from Sigma Chemical Co.

RESULTSGrowth and siderophore production. Ana-

baena cultures were exposed to a range of ironconcentrations to study the behavior of thisorganism in response to iron limitation. Growthcurves for these cultures are shown in Fig. 1. Incultures receiving no added iron (i.e., with lessthan 0.08 ,uM iron present) or 1 p.M added iron,an initial period of reduced growth was ob-

200-

0

0 2 4 6 a 10 12 14

TIME (Days)

FIG. 1. Growth of Anabaena sp. in BG-11 mediumas a function of added iron. Iron was added from afreshly prepared solution of filter-sterilized FeSO4 asfollows: E, 1.0 mM;U, 100 p.M;@, 10 t.M; A, 1.0>M;and 0, no added iron.

VOL. 151, 1982

on February 3, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

290 LAMMERS AND SANDERS-LOEHR

served, indicating an iron-limited state. Howev-er, once growth commenced, the rates approxi-mated those observed with the 10 and 100 ,Miron cultures, and a higher cell density was

actually achieved with the iron-limited cultures.Only the 1 mM iron culture exhibited lack ofgrowth due to iron toxicity.

It has been previously demonstrated by Simp-son and Neilands (24) that growth of Anabaenasp. under iron-limiting conditions results in theproduction of Csaky assay-detectable (7) extra-cellular hydroxamates and that the hydroxa-mate-containing siderophore schizokinen is pre-

sent in the supematant of iron-limited cultures.To obtain a more direct measure of siderophoreproduction in response to iron availability, we

determined the siderophore activity of Ana-baena filtrates by using the siderophore auxo-troph Arthrobacter flavescens JG-9 as a testorganism. We found that siderophore concentra-tions for cells grown on BG-11 medium minusEDTA and citrate increased linearly during days1 through 5 of growth and that the amount ofsiderophore released was inversely proportionalto the amount of iron in the Anabaena growthmedium. After 5 days of growth, all of theAnabaena cultures showed a marked drop in theArthrobacter assay response. A similar mid-log-phase drop in the production of Arthrobacter-active siderophore has been reported for themarine cyanobacterium Agmenellum quadrupli-catum (2), which was also grown on nitrate-containing medium.Subsequent experiments on Anabaena fil-

trates showed that Arthrobacter growth factoractivity could be restored by organic extractionof the schizokinen or by omitting nitrate fromthe Anabaena growth medium (Table 1). Thus,siderophore levels continued to increase in log-phase Anabaena sp., and they were inverselyproportional to the amount of iron in the growthmedium. The apparent drop in growth factoractivity after 5 days was most likely due to theproduction of a substance which was inhibitoryto Arthrobacter flavescens JG-9, which was

eliminated during an organic extraction, andwhich was particularly produced by cells grownin nitrate. Although nitrite would be a candidatefor this inhibitor, we found that the concentra-tion of nitrite was negligible according to Csakyassays performed without the addition to iodine.

Greatly enhanced growth factor activity was

observed for Anabaena sp. grown on BG-11medium minus EDTA with 0.1 mM citrate. Forcells grown in 0.1 ,uM added iron, the schizo-kinen concentration in extracted filtrates in-creased to 42 ,uM by the end of the log phase.Medium blanks containing this amount of citratehad no significant growth factor activity withArthrobacterflavescens JG-9.

Csaky assays also showed increased hydroxa-mate production in low-iron media. However, inthis case the apparent siderophore concentra-tions from cells grown on N2, NH4', or NO3were approximately 10 times higher than thoseobtained by the growth factor assay on extractedfiltrates. When filtrates for the Csaky assay wereextracted as described in Table 1, there wassome reduction (ca. over twofold) in the Csakyresponse. Since schizokinen recoveries of about80% are generally observed after organic extrac-tion, the excess Csaky assay-positive material inunextracted filtrates must be due to a substanceother than schizokinen. Since this Csaky assay-positive material was also present at high levelsin cells grown in 10 ,uM iron, it is unlikely to be asiderophore. These results indicate that a pre-liminary purification of siderophores by organicextraction or column chromatography is re-quired to obtain meaningful data on siderophorelevels in the growth media of cyanobacteria withthe above assays.

Specificity of iron transport. The ability ofAnabaena sp. to take up iron from the mediumwas studied by exposing cells to "Fe underconditions in which all iron in the assay mediumwas bound to schizokinen. To ensure that theAnabaena cells were responding primarily toadded [55Fe]ferric schizokinen, we used an up-take medium which was devoid of iron chelatorsand depleted of endogenous iron, and we added2 mol of schizokinen per mol of "5Fe. Afterexposure to [55Fe]ferric schizokinen, the cellswere filtered and washed to remove looselybound iron. The amount of radioactivity remain-ing in the filtrate was then determined. The

TABLE 1. Concentration of schizokinen inAnabaena filtratesaSchizokinenconcn (~LM) with added iron:

Nitrogen source' No Fe 1.0b.M Fe

Filtrate Extractc Filtrate Extractc

N2 0.9 2.1 0.2 0.8NH4Cl, 1 mM 3.6 5.1 0.1 0.3NaNO3, 13 mM 0.5 6.6 0.1 2.2

a The schizokinen concentration was determined bygrowth factor activity for Arthrobacterflavescens JG-9 relative to a standard curve obtained with purifiedschizokinen. Errors were ±30%6 for values above 3and ±70%o for values below 2.

b Anabaena sp. grown to the late log phase on BG-11 medium (minus citrate, EDTA, and NaNO3) withnitrogen added as indicated.

c Anabaena filtrates (1 to 2 ml) were extracted threetimes with 5 ml of chloroform-phenol (1:1 by weight);then 80 ml of ethyl ether was added, and the schizo-kinen was back-extracted twice with 5 ml of distilledwater. The water layer was lyophilized, and the mate-rial was redissolved in the original sample volume.

J. BACTERIOL.

on February 3, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

FERRIC SCHIZOKINEN TRANSPORT IN ANABAENA

uptake offerric schizokinen by iron-starved cellswas linear with time until the substrate becamelimiting (Fig. 2). The fraction of iron taken upwas generally 75 to 90% of the amount added inthe concentration range studied (10 to 50 nMFe). The ability of Anabaena sp. to accumulateiron against a concentration gradient indicatesthat a transport system is present and that it isdriven by metabolic energy. Furthermore, thetransport system was found to be selective foriron bound to schizokinen. Another sidero-phore, ferrioxamine B, showed no support ofiron uptake in concentrations of up to 0.055 ,uMin Fe and 0.11 FjM in deferriferrioxamine B.

Control of iron transport. Cells grown in thepresence of three different concentrations ofiron were suspended to identical cell densitiesand tested for their ability to transport [55Fe]fer-ric schizokinen. Ferric schizokinen was takenup by all cultures, but the rates of uptake weresignificantly faster in the iron-starved cultures(Fig. 3). The lower the concentration of iron inthe growth medium, the greater the uptakerate. This pattern of inducible uptake activityindicates that the schizokinen-dependent ironuptake system is subject to regulatory control.

Iron transport kinetics. Further evidence forthe involvement of specific carriers in the trans-port offerric schizokinen was obtained by study-ing the kinetics of iron uptake as a function ofthe concentration of ferric schizokinen. Resultsdemonstrate that uptake occurs via a saturatablesystem (Fig. 4, inset). Double-reciprocal plots asin Fig. 4 yield an apparent Km of 0.04 F.M and a

12 I

10

E

0

E

wCL

I-

In

0 2 4 6 8 10

TIME (Min)

FIG. 3. Rate of uptake of ferric schizokinen as afunction of the iron concentration in the growth medi-um. Anabaena sp. was grown in 0 (0), 0.1 ,uM (O),and 1.0 ,uM (A) added iron. Cells were suspended to 52Klett units in uptake medium containing 0.014 ,uM55Fe and 0.028 F.M schizokinen.

Vma, of approximately 10 nmol/min per g of cells(dry weight) (for Anabaena sp. exposed to 0.1F.M Fe). The Km value is similar to reportedvalues for the uptake of ferrienterobactin and

oI

Ee

E

6w

4_I-

2

0 5 10 20

TIME (Mi)

FIG. 2. Rate of uptake of 55Fe by Anabaena cells.Anabaena sp. was grown in 0.1 F.M added iron andsuspended to 96 Klett units in uptake medium contain-ing 0.012 PM 55Fe and 0.024 ,uM schizokinen (0) and0.008 p,M "Fe and 0.016 F.M Desferal (A).

[Fe - SK] ( )

FIG. 4. Kinetics of uptake of ferric schizokinen.Anabaena sp. was grown in uptake medium containing0.1 ,uM added iron and suspended to 47 Klett units inuptake medium containing a twofold excess of schizo-kinen (SK) to Fe.

o _

VOL. 151, 1982 291

on February 3, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

292 LAMMERS AND SANDERS-LOEHR

ferrichrome by Escherichia coli (20, 28), and itgives an indication of the extremely high affinityof the transport system for its substrate.Energy dependence. Once it was shown that

Anabaena sp. is capable of accumulating ferricschizokinen against a concentration gradient, itwas of interest to determine the energy require-ments of the process. Since Anabaena sp. is anobligate phototroph, light is the ultimate sourceof all its metabolic energy. To determine wheth-er ferric schizokinen uptake is energized directlyby light-associated processes, uptake activity inthe light and in the dark was measured. Wefound that uptake activity was only slightlydecreased in the dark, ranging from 80 to 95% oflight controls; dark adaptation periods from 15min to 2 h had little effect.To discriminate between possible sources of

energy for iron uptake, the effect of variousinhibitors on iron transport was tested (Table 2).Both light uptake and dark uptake were sensitiveto the uncoupler CCCP, which dissipates thetransrhembrane, electrochemical hydrogen iongradient. In addition, DCCD, an inhibitor of H+-translocating ATPases in bacteria, mitochon-dria, and chloroplasts (1), caused almost com-plete inhibition of iron uptake both in the lightand in the dark. The sensitivity of iron uptake toDCCD implies that ATP is required for thisprocess. The inhibition of iron uptake by arse-nate, which inhibits ATP synthesis by acting as asubstrate analog, supports this conclusion. Thefact that ferric schizokinen uptake in the light isless sensitive than uptake in the dark to arsenateinhibition may possibly be due to a higher over-all rate of ATP formation in the light.The inhibition of ferric schizokinen uptake by

cyanide in the dark strongly suggests that dark

TABLE 2. Effect of metabolic inhibitors on ferricschizokinen uptake in Anabaena sp.

Uptake activityInhibitor Concn (%)

Light Dark

CCCP 8 ,uM 10 0DCCD 40 ,uM 5 5Na2HAsO4 20 mM 50 15KCN 100 ,LM 90 20

a Uptake activity relative to light or dark control inthe absence of an inhibitor. Values are ±5% forCCCP, DCCD, and Na2HAsO4 and ±10% for KCN.For dark uptake, cells were dark adapted for 10 minbefore adding the inhibitor. The total time of exposureto the inhibitor before the addition of [55Fe]ferricschizokinen was 20 min for CCCP and DCCD and 50min for Na2HAsO4 and KCN. Anabaena cultureswere grown in uptake medium containing 0.1 ,uMadded iron.

uptake in Anabaena sp. is driven by respiratoryelectron transport. Cyanide is a well-knowninhibitor of electron transport in bacteria andmitochondria due to strong binding of CN- tothe oxygen-binding site in the terminal cyto-chrome oxidase of the electron transport chain.The lack of significant cyanide inhibition ofuptake in the light indicates that Anabaena sp.depends on photophosphorylation for light up-take and on oxidative phosphorylation for darkuptake.

DISCUSSIONActive transport of nutrients across bacterial

cell membranes is generally facilitated by mem-brane permeases, periplasmic binding proteins,or intracellular modifications leading to grouptranslocation (30). In the case of iron uptake inE. coli, ferrichrome transport has been shown tooccur in inner membrane vesicles via a per-mease system which responds to the electricalcomponent (Aip) of the membrane potential (20),possibly via a symport process with Ca2' orMg2' bound to the ferrichrome molecule toproduce a divalent cation (14). Ferrienterobactinuptake also requires an energized membranestate (22). Ferrienterobactin receptors with pos-sible permease activity are present in both theouter and the inner membranes (23), with theinner membrane system presumably accountingfor the AIH+ dependence. In addition, thereappears to be an ATP requirement for ferrienter-obactin uptake in E. coli. However, in thisheterotrophic organism, it is difficult to rule outthe possibility that the primary target of ATPinhibitors is glucose uptake rather than ironuptake (22).

Conversely, with the autotroph Anabaena sp.we have been able to demonstrate unambig-uously that ferric schizokinen uptake is depen-dent on the availability of ATP. The completeinhibition of light and dark ferric schizokinenuptake by the ATPase inhibitor DCCD, which isnot expected to affect electron transport or thedevelopment of a membrane potential (5), clear-ly shows that the energized membrane statealone is not sufficient to drive the transport ofiron. Similar results were obtained with arse-nate, which blocks ATP formation. Becausemembrane level phosphorylation is the onlyknown mechanism for ATP synthesis in cyano-bacteria (25), it has not been possible to deter-mine whether the inhibition caused by uncou-plers and electron transport inhibitors is dueonly to interference with ATP formation orwhether a membrane potential is also requiredfor iron transport.

It is perhaps significant that ferric schizokinenand ferrienterobactin are monovalent and triva-lent anions, respectively, at physiological pH.

J. BACTERIOL.

on February 3, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

FERRIC SCHIZOKINEN TRANSPORT IN ANABAENA

As a result, both must be transported against theelectrical component of the membrane potential(interior negative). The ATP requirement shownby both systems suggests that transport is morecomplicated than a simple permease-type sys-tem, yet whether either system will fit the bind-ing protein model or the group translocationmodel (or some combination ofthem) remains tobe determined. The involvement of a periplas-mic binding protein appears to have been ruledout in the case of ferrienterobactin uptake in E.coli (22). A group translocation process acting atthe level of intracellular release of iron fromferric Desferal has been implicated in B. mega-terium (6, 8).

Siderophore transport systems tend to behighly specific for a given ligand, so that if morethan one siderophore is transported by an orga-nism, separate transport systems exist for eachone (3, 11). One of the most versatile organismsis E. coli, which utilizes not only the previouslymentioned ferrichrome and enterobactin sys-tems, but also an inducible ferric citrate system(23), a low-affinity Fe(III) uptake system (23),and a plasmid-mediated aerobactin uptake sys-tem (29). B. megaterium has separate transportsystems that recognize ferrioxamine B and ferricschizokinen, but will not accept ferriaerobactin,which is structurally similar to ferrischizokinen(6, 13). Anabaena sp. appears to be somewhatmore selective than B. megaterium in that fer-rioxamine B is not utilized as an iron transportmediator.The affinity of siderophore transport systems

for their substrates is generally higher in bacteri-al than in fungal systems. For comparison, Neu-rospora crassa accumulates iron via its ownsiderophore, coprogen, with a Km of 20 ,uM (31).The ferrichrome and ferrienterobactin systemsin E. coli are characterized by Km values of 0.15to 0.25 and 0.10 to 0.36 ,uM, respectively (11, 20,28), whereas the apparent Km for ferric schizo-kinen uptake in Anabaena sp. is 0.04 ,uM. Thehigh affinity of the Anabaena siderophore trans-port system may help to counteract the largepotential for dilution of cyanobacterial extracel-lular products in an aquatic environment.

ACKNOWLEDGMENTS

We are grateful to Jack Myers, Mary Taylor, and JonAbramson for their advice and discussions. We thank AldenPritchett for his valuable and good-natured technical assist-ance.

This work was supported in part by Public Health Servicegrant GM 18865 from the National Institutes of Health.

LITERATURE CITED1. Abrams, A. 1976. Structure and function of membrane-

bound ATPase in bacteria, p. 67-69. In A. N. Martonosi

(ed.), The enzymes of biological membranes, vol. 3.Plenum Publishing Corp., New York.

2. Armstrong, J. E., and C. Van Baalen. 1979. Iron transportin microalgae: the isolation and biological activity of ahydroxamate siderophore from the blue-green alga Ag-menellum quadruplicatum. J. Gen. Microbiol. 111:253-262.

3. Aswell, J. E., A. H. Haydon, H. R. Turner, C. A. Dawkins,J. E. L. Arceneaux, and B. R. Byers. 1977. Specificity ofsiderophore receptors in membrane vesicles of Bacillusmegaterium. J. Bacteriol. 130-.173-180.

4. Atkin, C. L., and J. B. Neilands. 1968. Rhodotorulic acid,a diketopiperazine dihydroxamic acid with growth-factoractivity. Biochemistry 7:3734-3739.

5. Bohler, M. C., and A. Binder. 1980. Photosynthetic activi-ties of a membrane preparation of the thermophilic cyano-bacterium Mastigocladus laminosus. Arch. Microbiol.124:155-160.

6. Byers, B. R., J. E. L. Arceneaux, A. H. Haydon, and J. E.Aswell. 1977. Siderophore transport in Bacillus megater-ium, p. 371-378. In E. B. Brown, P. Aisen, J. Fielding,and R. R. Crichton (ed.), Proteins of iron metabolism.Grune & Stratton, New York.

7. Csaky, T. 1948. On the estimation of bound hydroxyl-amine. Acta Chem. Scand. 2:450-454.

8. Davis, W. B., and B. R. Byers. 1971. Active transport ofiron in Bacillus megaterium: role of secondary hydrox-amic acids. J. Bacteriol. 107:491-498.

9. Emery, T. 1971. Role of ferrichrome as a ferric ionophorein Ustilago sphaerogena. Biochemistry 10:1483-1488.

10. Estep, M., J. E. Armstrong, and C. Van Baalen. 1975.Evidence for the occurrence of specific iron (III)-bindingcompounds in near-shore marine ecosystems. Appl. Mi-crobiol. 30:186-188.

11. Frost, G. E., and H. Rosenberg. 1973. The induciblecitrate-dependent iron uptake system in Escherichia coliK12. Biochim. Biophys. Acta 330:90-101.

12. GUilam, A. H., A. G. Lewis, and R. J. Andersen. 1981.Quantitative determination of hydroxamic acids. Anal.Chem. 53:841-844.

13. Haydon, A. H., W. B. Dais, J. E. L. Arceneaux, and B. R.Byers. 1973. Hydroxamate recognition during iron trans-port from hydroxamate-iron chelates. J. Bacteriol.115:912-918.

14. Hider, R. C., A. F. Drake, R. Kuroda, and J. B. Neilands.1980. Symport transport of ferrichrome-type sidero-phores. Naturwissenschaften 67:136-139.

15. Lange, W. 1974. Chelating agents and blue-green algae.Can. J. Microbiol. 20:1311-1321.

16. Lankford, C. E. 1973. Bacterial assimilation of iron. Crit.Rev. Microbiol. 2:273-331.

17. McKnight, D. M., and F. M. M. Morel. 1980. Coppercomplexation by siderophores from filamentous blue-green algae. Limnol. Oceanogr. 25:62-71.

18. MullLs, K. B., J. R. Pollack, and J. B. NelUands. 1971.Structure of schizokinen, an iron-transport compoundfrom Bacillus megaterium. Biochemistry 10:4894-4898.

19. Murphy, T. P., D. R. S. Lean, and C. Nalewajko. 1976.Blue-green algae: their excretion of iron selective chela-tors enables them to dominate other algae. Science192:900-902.

20. Negrln, R. S., and J. B. Neilands. 1978. Ferrichrometransport in inner membrane vesicles of Escherichia coliK12. J. Biol. Chem. 253:2339-2342.

21. Neilands, J. B. 1973. Microbial iron transport compounds(siderochromes), p. 167-200. In G. L. Eichhom (ed.),Inorganic biochemistry, vol. 1. Elsevier Scientific Pub-lishing Co., New York.

22. Pugsley, A. P., and P. Reeves. 1977. Uptake offerrientero-chelin by Escherichia coli: energy-dependent stage ofuptake. J. Bacteriol. 130:26-36.

23. Silver, S. 1978. Transport of anions and cations, p. 243-324. In B. P. Rosen (ed.), Bacterial transport. MarcelDekker, Inc., New York.

24. Simpson, F. B., and J. B. Neilands. 1976. Siderochromes

VOL. 151, 1982 293

on February 3, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

294 LAMMERS AND SANDERS-LOEHR

in Cyanophycea: isolation and characterization of schizo-kinen from Anabaena sp. J. Phycol. 12:44 48.

25. Stanier, R. Y., and G. Cohen-Bazlre. 1977. Phototrophicprokaryotes: the cyanobacteria. Annu. Rev. Microbiol.31:225-274.

26. Stanier, R. Y., R. Kunisawa, M. Mandel, and G. Cohen-Bazlre. 1971. Purification and properties of unicellularblue-green algae. Bacteriol. Rev. 35:171-205.

27. Strickland, J. D. H., and R. T. Parson. A practical guide toseawater analysis. Bull. Fish. Res. Board Can. 167:95-98.

28. Wang, C. C., and A. Newton. 1971. An additional step in

the transport of iron defined by the tonB locus of Esche-richia coli. J. Biol. Chem. 246:2147-2151.

29. Warner, P. J., P. H. Williams, A. Binderelf, and J. B.Neilands. 1981. ColV plasmid-specified aerobactin synthe-sis by invasive strains of Escherichia coli. Infect. Immun.33:540-545.

30. Wilson, D. B. 1978. Cellular transport mechanisms. Annu.Rev. Biochem. 47:933-965.

31. Winkelmann, G. 1974. Metabolic products of microorga-nisms: uptake of iron by Neurospora crassa. Arch. Micro-biol. 98:39-50.

J. BACTERIOL.

on February 3, 2020 by guest

http://jb.asm.org/

Dow

nloaded from