MODIFIED PROCEDURES AND PRELIMINARY RESULTS IN THE ANALYSIS OF POSSIBLE IRON ?%NSPORT CGMPOUND SYNTHESIS BY BACTERIA _ Thesisfw‘theADeg‘reeofMfi. j wcmm Isms unmansmi RGBERT'D.._.RERRY ms “fists .! ! _ ‘IA- 9 HIBWIK BIMTRY III} .1 My em ms! ABSTRACT MODIFIED PROCEDURES AND PRELIMINARY RESULTS IN THE ANALYSIS OF POSSIBLE IRON TRANSPORT COMPOUND SYNTHESIS BY BACTERIA By Robert D. Perry The synthesis of phenolate and/or hydroxamate com- pounds during iron deficiency was investigated in 51gb— siella pneumoniae, Salmonella enteriditis, Serratia mar- cescens, Shigella sonnei, and Pseudomonas aeruginosa. A minimal medium and method of iron extraction de- scribed by Waring and Werkman (1942) were employed. Commercially prepared penassay broth was also deferrated (iron-extracted) by this method. A study with E, Bagg— moniae indicated that the deferri-penassay broth may contain more residual iron than the deferri-dextrose broth. The original bathophenanthroline iron assay procedure, as described by Diehl and Smith (1960), was found to be unsatisfactory for the analysis of low iron concentrations in media and water. Extensive modifications of this pro- cedure were necessary to insure adequate reduction of ferric ions. Incomplete reduction occurred unless samples were steam-heated for one hour in the presence of 0.1 m1 of concentrated HCl, and L-ascorbate was substituted for hydroxylamine. Minor modifications of assays for the de— tection of phenolate and hydroxamate compounds were test- ed for their sensitivity and reliability. Deferri-dextrose cultures of K, pneumoniae and Sal. enteriditis incubated at 25 C and 37 C synthesized pheno— late compounds. At 25 C Ser. marcescens also produced phenolate compounds. At 37 C Eén aeruginosa synthesized hydroxamate compounds during iron deficiency. In deferri— dextrose broth cultures at 25 C and 37 C, §h, sonnei made both phenolate and hydroxamate compounds. _ In deferri-penassay broth, Pseudomonas fluorescens synthesized hydroxamate compound(s) while Shigella boydii produced both phenolate and hydroxamate compounds. Four strains of Staphylococcus aureus, two strains of Yersinia pestis, four species of Proteus, and Yersinia pseudotuber- culosis grew well in deferri-penassay broth without syn- thesizing detectable levels of phenolate or hydroxamate compounds. MODIFIED PROCEDURES AND PRELIMINARY RESULTS IN THE ANALYSIS OF POSSIBLE IRON TRANSPORT COMPOUND SYNTHESIS BY BACTERIA By I? Robert DT‘Perry A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Microbiology and Public Health 1975 ACKNOWLEDGEMENTS The author wishes to thank Dr. C. L. San Clemente for his support and encouragement during the course of this study; Dr. R. R. Brubaker and Dr. T. R. Corner for their helpful discussions and guidance; Dr. E. D. Weinberg, Dr. J. A. Garibaldi, and Dr. E. J. Wawszkiewicz for their suggestions and information; Steven Glenn for his assis— tance and constructive criticism; and Sue Rose for her assistance with the figures appearing in this paper. ii TABLE OF CONTENTS LIST OF TABLES. . . . . . . . . . . . LIST OF FIGURES . . . . . . . . INTRODUCTION. . . . . . . . . . . . . . . . . REVIEW OF LITERATURE. . . . . . . . . Iron Metabolism of Microorganisms. . . . Iron Requirements . . . . . . . . . Iron Transport Compounds (ITC's). . Mechanisms of Bacterial Iron Transport. Iron Transport by Enterochelin . Iron Transport by Hydroxamates Iron Transport by Mycobactin . Metabolic Pathways for the Synthesis of Microbial ITC's. . . . . . . . Biosynthesis of Enterochelin and Salicylate . . . . . . . Biosynthesis of Hydroxamates . Biosynthesis of Mycobactins. . Conditions for the Synthesis of Microbial ITC's . . . . . . . . . Host Iron Metabolism during Infection. . The Effect of Iron Availability upon Virulence. . . . . . . . . . . . . . . The Antibiotic Activity of Sideromycins. MATERIALS AND METHODS . . . . . . . . . . . . Microorganisms . . . . . . . . . . . . . Equipment. . . . . . . . . . . . . . . . Assays for ITC's . . . . . . . . . . . . The Arnow Assay . . . . . . . . . . The Csaky Assay . . . . . . . . . . iii Page vi \OkDLflvbuh-blfl 11 13 15 15 16 16 18 19 20 21 23 23 23 24 24 24 TABLE OF CONTENTS (cont.) The FeC13-ITC Assay . . . . . . . The Bathophenanthroline Iron Assay . . Preparation of In Vitro Iron-deficient Conditions . . . . . . . . . . . . . Preparation of Deferri-glassware. Preparation of Deferri-water. . . Preparation of Deferri-media. . . In Vitro Growth Study Procedures . . . RESULTS . . . . . . . . . . . . . . . Procedural Results . . . . . . . . . . Assays for ITC's. . . . . . . . . The Arnow Assay. . . . . . . The Csaky Assay. . . . . . . The FeCl3-ITC Assay. . . . . The Bathophenanthroline Iron Assay. Preparation of Iron-deficient Conditions. . . . . . . . . . . Preparation of Deferri-water . Preparation of Deferri-glassware Preparation of Deferri-media Growth Study Results . . . . . . . . . Preliminary Studies . . . . . . . Growth Studies in Deferri-dextrose Broth . . . . . . . . . . . . . Growth Study in Deferri—penassay Broth . . . . . . . . . . . . . DISCUSSION. . . . . . . . . . . . . . . . . LITERATURE CITED. . . . . . . . o . . . . . iv Page 25 26 27 27 27 27 3O 32 32 32 32 36 38 44 46 46 46 47 49 49 50 6O 62 66 LIST OF TABLES Table Page 1. Composition of dextrose broth. . . . . . . . 28 2. Composition of penassay broth. . . . . . . . 29 3. Correlation between incubation time and color intensity (Arnow Assay). . . . . . . 35 4. Correlation between boiling time and release of unbound desferal (Csaky assay) . . . . . . . . . . . . . . . . . . 39 Figure 1. 10. 11. 12. 13. LIST OF FIGURES Page Structures of some iron transport compounds. . . . . . . . . . . . . . . . . 6 Structures of chemically defined species of mycobactins . . . . . . . . . . . . . . 8 A model for iron transport by enterochelin in E. coli. . . . . . . . . . 10 A model for iron transport by mycobactin and salicylate in mycobacteria . . . . . . l4 Metabolic pathways for the synthesis of microbial ITC's . . . . . . . . . . . . 17 Arnow assay standard curves with DHB . . . . 34 Cséky assay standard curves with desferal in dextrose and penassay broths. . . . . . 37 Csaky assay standard curves with desferal in dextrose broth obtained at various boiling times. . . . . . . . . . . . . . . 4O Csaky assay standard curves with desferal in penassay broth obtained at various bOiling times. 0 O O O O O O O O I O O O O 41 Csaky assay standard curves with desferal in dextrose broth obtained at various incubation times . . . . . . . . . . . . . 42 FeCl -ITC assay standard curves with de feral and DHB . . . . . . . . . . . . . 43 Bathophenanthroline iron assay standard curves with FeCl3 and FeSO4-7 H20. . . . . Ultraviolet absorption spectra of ferri- dextrose broth containing 8—HQ (a), de- ferri-dextrose broth diluted to volume with doubly distilled deionized water (b), and deferri-dextrose broth diluted to volume with doubly distilled deion- ized water further demineralized with 8-HQ (C) . . . . . . . . . . . . . . . . . 48 45 vi LIST OF FIGURES (cont.) Figure 14. 15. 16. 17. 18. 19. 20. Growth curves of iron-sufficient and iron-deficient dextrose broth cultures of 5, pneumoniae at 25 C and 37 C. . . Phenolate accumulation curves of iron- deficient dextrose broth cultures of E, pneumoniae at 25 C and 37 C . . . . Growth curves and phenolate accumulation curves of iron-sufficient and iron- deficient dextrose broth cultures of Sal. enteriditis at 25 C and 37 C. . . Growth curves and phenolate accumulation curves of iron-sufficient and iron- deficient dextrose broth cultures of Ser. marcescens at 25 C. . . . . . . . Growth curves and hydroxamate accumulation curves of iron-sufficient and iron- deficient dextrose broth cultures of Igg. aeruginosa at 37 C . . . . . . . . Growth, phenolate, and hydroxamate curves of iron-sufficient and iron- deficient dextrose broth cultures of .gg. sonnei at 25 C and 37 C. . . . . . Growth, phenolate, and hydroxamate curves of iron-sufficient and iron- deficient penassay broth cultures of sg, sonnei at 30 C . . . . . . . . . . Vii Page 51 52 55 56 57 59 61 INTRODUCTION Trace amounts of metallic elements such as iron, man- ganese, zinc, copper, cobalt, and molybdenum are essential for microbial growth. The concentration ranges of these trace metals necessary for maximal microbial growth are greatly affected by a number of synthetic and natural metal-binding compounds. Secondary metabolism and dif- ferentiation occur at trace metal concentration ranges which are generally narrower than those ranges allowing growth. For bacteria, iron appears to be the trace metal most critically affecting growth, longevity, and secondary metabolism, while zinc is important in yeasts and molds. The genus Bacillus is an exception since manganese appar- ently is the most important factor. Manganese is also a key factor for some fungi (Weinberg, 1970; 1972). For a number of these trace metals, evidence of their mediated transport into microbial cells has been described. Although a biological requirement for chromium has not been established, some yeast strains produce a chromium trans- port compound (Weinberg, 1970). Cells of Escherichia coli have been shown to accumulate manganese on the one hand by an active transport system (Silver and Kralovic, 1969) and magnesium on the other hand by a temperature dependent process inhibited by dinitrophenol and cyanide (Silver, 1969; Lusk and Kennedy, 1969). Cells of Salmonella typhi— murium grown on limiting sulfur sources synthesize a pro— tein capable of binding sulfate ions (Pardee, 1968). Numerous microorganisms are autosequesteric; they produce small organic compounds which solubilize and transport exogenous iron into the cell (Weinberg, 1974a). These small microbial compounds have been variously termed iron transport compounds (ITC's), ionophores, sideramines, and siderophores. The synthesis of these ITC's has been found to be inhibited by iron concentrations which are adequate for growth (Weinberg, 1974a). The wide variety of microbial species possessing iron transport systems suggests that microorganisms often must exist where iron is not readily available. Although iron is the fourth most abundant element on earth, it is almost entirely present in forms unavailable for biological use. Due to its chemical properties, ferric iron is extremely insoluble at room temperature and at alkaline and neutral pH's. In sea water the iron present (0.5 to 5.0 ppb) is mainly in a particulate form and the free ionic concentra- tions are very low. Although abundant in soil and rock, iron is present almost entirely as insoluble aggregates and precipitates (Lankford, 1973). In mammals, internal con- ditions are extremely iron limiting because of the presence of a complex system of ligand proteins which sequester iron (Weinberg, 1974a). Generallx iron need not be added to complex synthetic media since iron contamination alone meets microbial growth requirements (Weinberg, 1971). Evidence is accumulating that the iron transport sys- tems of some bacteria play an important role in pathogene- sis. In—depth studies of these systems have been limited to a relatively small number of bacteria. Analysis of iron metabolism in other bacteria (especially pathogens) will further clarify this biological role of bacterial iron transport systems. The purpose of this investigation was (1) to improve and develop procedures necessary for the investigation of ITC synthesis and growth by both pathogenic and sapro— phytic bacteria during iron deficiency, and (2) to study the effect of iron concentration and temperature upon bacterial ITC synthesis. The test organisms were select- ed either because their growth patterns suggest ITC syn— thesis, or because their capacity for ITC synthesis is undetermined. In preliminary eXperiments, Perry and Wein- berg (1973) observed that Pseudomonas aeruginosa, Staphyl- ococcus aureus, Shigella sonnei, Shigella alkalescens, and Proteus vulgaris as well as other bacteria competed with orthophenanthroline (OP) for iron. REVIEW OF LITERATURE Iron Metabolism of Microorganisms Iron Requirements In order to function, numerous microbial enzymes re— quire the presence or direct incorporation of ferrous or ferric ions. The active sites of metapyrocatechase and 3,4-dihydroxyphenylacetate-Z,3-oxygenase contain ferrous ions. Pyrocatechase and pryocatechuate-B,4,-oxygenase con- tain ferric iron at their active sites (Hayaishi, 1966). Iron deficiency in Aerobacter indologenes severely reduced the activities of catalase, peroxidase, formic hydrogen- 1yase, and hydrogenase (Waring and Werkman, 1944). Theo- dore and Schade (1965) have proposed that Staph. aureus enzymes which oxidize pyruvate, formate, acetate, succinate, malate, and citrate require iron for activity. At low pH's iron promotes increased production of penicillinase by Staph. aureus (Leitner and Cohen, 1962). Staph. aureus is unable to produce coagulase under conditions of iron defi- ciency (Schade 25 $1., 1968). Rosenberg and Gefter (1969) found that E. £91; cells grown under iron-deficient condi- tions synthesized abnormal transfer RNA for the amino acids phenylalanine, leucine, serine, tyrosine, and cys- teine. There are many other examples in the literature on the effect of iron deficiency upon bacterial metabolism (see Weinberg, 1970). Of course iron is extremely impor- tant in many of the respiratory enzymes (e.g., cytochromes). The synthesis of many bacterial toxins is also affected by iron concentrations (e.g., tetanus toxin and staphylococcal enterotoxin require moderate iron concen— trations for synthesis). However, in some instances, low iron concentrations facilitate toxin production (e.g., cx-toxin lecithinase of Clostridium perfringens, neurotox— in of Shigella shigae, and diphtheria toxin of Coryne- bacterium diphtheriae) (Weinberg, 1966). Because of its established role in bacterial metabo- lism, the iron concentration of the environment greatly in- fluences the growth and longevity of many bacterial species (e.g., Perry and Weinberg, 1973; Sword, 1966; Davis e; .31., 1971; Garibaldi, 1970; Garibaldi, 1972). Iron Transport Compounds (ITC's) The various ITC's fall primarily into two groups: phenolate compounds and hydroxamate compounds. The iron metabolism of bacteria has been primarily studied in Sal. typhimurigm, E. coli, Enterobacter aerogenes, Bacillgs subtilis, Bacillus megaterium, and the mycobacteria. The first four organisms produce phenolate compounds (Peters and Warren, 1968a; Pollack and Neilands, 1970; Young 23 _§1., 1971), whereas Ent. aerogenes and B. megaterium produce a hydroxamate compound (Davis and Byers, 1971; Gibson and Magrath, 1969). Mycobacteria synthesize a more complex ITC called mycobactin, whose chemical structure varies from strain to strain but always contains both phenolate and hydroxamate moieties (Snow, 1970). A number of microorganisms produce phenolate com- pounds when grown under iron-deficient conditions. .E- subtilis produces 2,3-dihydroxybenzoylglycine (Peters and Warren, 1968a). Enterochelin is synthesized by E-.SQL£: Sal. typhimurium, and Ent. aerogenes (Pollack and Neilands, 1970; Young 33 a1., 1971). Cultures of Azotobacter vine— landii synthesize 2-N, 6-N-di-(2,3-dihydroxy—benzoy1)-L- lysine (Lankford, 1973). These structures are depicted in Fig. 1. ()ll ()1! ll() on on no can/I'm\c=o\ so c=o :o /C2 II I)" ll ) ' \O i (C 2 4 ill—NH .n2 00H (XDH 2,3-Dihydroxybcn- 2—N,6-N Di-- 0:). CI! zoylglycinc (2,3-Dihydroxy-benzoyl)- 2 L-lysino OH Oils. c=\o /m/|c \NH ‘\ . C'll3 CHJ H :0 0:? 3—0" HO\ [/0 80-1! cu: /C=O ( '112)2 o c. ('3‘ ( 112)2 \O HZC— —Nll-—C— CI!2*C‘012“'C‘N““C”2 Enterochelin $n Schizokincn CH CH CH 95 I3 :3 I3 C=O 01C c=o 0:": , I 21-10}: 0 Ho\ 0 "01:“ ) (2:10}: o Ho\c’o o "0131! ) (' 2 4 " c NH (.112 4 2 4 n 2 4 — I __ __ I... — fl — -— — ‘ — — — — .._ .— v .— "('2 M! C C112 ' C 2 ' H2C NH C C112 C- CH2 C M! C1!2 COOH OH COOH H Aerobactin Arthrobactin (Terrcgcns Factor) u , :/\z . OI / I \ . \ (CH2), *Q\ CH2 )5 cl;",); rerrloxanlnc I ICII / I I'm / N ( u ) \ \V (c """ l /I, /C 2 J (,0 / n u’c- \o 0 .o/ o. -—(cn )e-c'o—firc’l-‘o- -R-(<‘N,);—"H on 5, “K 2 2 ole 0'0 I N o: 0,] 0 mn / °\c _ ’ u’ \ we cu "J c c ' c—cn, \u . J \u 1, c1020" ' . am 01: um oar-alf—cuf—m u", a I c.o , a - "\ / ' 2 mac: c u 1 1 cu -—c u £\‘l‘ I'M [on oxa— no . 5- “c1- fl 2 rottlchronc Ford-yen: A. Figure l.--Structures of some iron transport com- pounds. Adapted from Lankford (1973). Hydroxamate-producing microorganisms are more diver- gent and numerous than those producing phenolate compounds. Various species of Streptomyces produce ferrioxamines and ferrimycins (Neilands, 1967; Nuesch and Knusel, 1967). Desferal (CIBA, Summit, New Jersey) is the commercial name for the iron-free (deferri) form of ferrioxamine B. Ferri- chrome is synthesized by the smut fungus Ustilagp sphaero- 2222 (Neilands, 1957). A number of fungal species pro- duce hydroxamate compounds (Lankford, 1973). Arthrobacter pascens synthesizes arthrobactin (also called terregans factor) (Demain and Hendlin, 1959). Schizokinen and aero- bactin are synthesized respectively by B. megaterium and Ent. aerogenes (Davis and Byers, 1971; Gibson and Magrath, 1969; Lankford 23 al., 1966). The structures of a number of these compounds are diagramed in Fig. l. Chemically defined species of mycobactin are shown in Fig. 2. Halmann and Mager (1967) have partially purified a growth-initiating substance (GIS) from Pasteurell§_ tularensis which is of low molecular weight and complexes with iron and copper ions. The requirement for GIS can be replaced only by iron salts and by high levels of cer- tain hydroxamates. Although the role of ornithine as a precursor for GIS suggests that it may be a hydroxamate compound, assays for hydroxamate compounds yielded nega- tive results (Halmann and Mager, 1967). Straube and Fritsche (1973) have suggested that ribo- flavin serves as an iron transport compound for Candida guilliermondii. Overproduction, excretion, and later up- take of riboflavin by E, gpilliermondii under conditions of iron deficiency have been noted. These observations and the observed formation of a complex between iron and riboflavin are the only data which presently implicate riboflavin in iron transport (Straube and Fritsche, 1973). .Amomav mugrz 6cm 30cm was Amomav 30cm cam mugs: scum omummoa+ .musuoouum m>fiumucma x. dam . .ma .mH m A AC m I I IV m mu m m Hm .mm..mH .mH a A AC m a A AC m mmo m m AqumgoC om .ma..mm..aa m a m m a a AV mmo mmmo m mmo I cqumHo mm..mm..oa .vH m a Av m a a m m mmo mmo mmo mam..wm. m 4 Av m I I I m mmo mmo m mm..wa .va smw .qoa m I I I I I I m mmo m mmo ch a w m m m m m m m m. m a N 3 .CTCHEHTDTUCD coflumusmflwcou .I uuwucmu oauuwesxmm Gm mxomH .Av “wumucmo Uauuwesxmm mum mum mHOQENm .UTDMUHGCH mum mocon THQsOU czocx can a an pmucmwmummu mum mcflmnu pmaocmuncs .pwcflaumpcs mum comm >HCOEEOU umOE mcflmnu mpam .BOHTQ c303m mEOum conumo mo HTQESC Hmuou wnu mCHCHmucou mmooum Hmum mum mafimno mpflmam was .mcofluomnooxe mo wmflommm pmcflmmp maamoflsmno mo mmusuosuumII.m musmflm + m mo mm m _ o m a m\\o Z m m \ w _ m u a z I mz.oo.m .mo. oo.mo.mz.oo w m U m N _ .m mwu mo SOIIZIIMm Recently, a citrate-dependent iron transport system has been identified in E, ggli_K12 (Frost and Rosenberg, 1973). Induction of this system requires new protein syn— thesis and occurs within twenty minutes of exposure to sodium citrate. In E, ggli_mutants unable to utilize the enterochelin transport system, citrate has been shown to stimulate iron uptake. This citrate-dependent system appears to be less widely distributed than other iron transport systems, even among other strains of E, 2211. Studies with various mutants have indicated that the iron uptake observed in the citrate system and the enterochelin system equal the sum of iron uptake observed in wild-type E, 221; strains endowed with both systems. These results indicate that the citrate system is wholly independent of the enterochelin system and may operate simultaneously in wild-type strains of E, 221;. In Egg. aerogenes citrate has been shown to repress 2,3-dihydroxybenzoate (DHB) bio- synthesis and may therefore serve as a less energetically expensive method for the transport of soluble chelated iron from the environment into the cell. A citrate-dependent transport system has also been reported in Neurospora crassa, which produces coprogen, a hydroxamate (Frost and Rosenberg, 1973). Mechanisms of Bacterial Iron Transport Iron Transport by Enterochelig. Langman g; 31. (1972) have proposed a mechanism of iron chelation and utilization in which external iron is chelated by enterochelin and transported into the cell (see Fig. 3). The iron in the ferri-enterochelin complex can— not be utilized until it is broken down by the cell into 2,3-dihydroxybenzoylserine (DHBS) subunits. By this .33: 30353 scum 63mm? .28 .m fl 3623 Ioumucm an uuommcmup come you proE 2m s9 uuommcmuu COHH How HmpoE rHoI m m .«m .mmmumruasm mumoucmnmxowmxgaoI m mIouomnHoI m m . m “mmMDmnucwm wumHSUAHMm . m “mmMDTSucmm mumsmAHOAUOmH . IMm musmflm now maonfimm .m.UBH amenouuwfi mo mammnucmm may now whmznumm UHHonmumZII m Tasmam .Amhmav whom mszeHzmomxomnwm mszaHzmo oH64 oHADmoeooomm IzIaIuwemoaIzIs waomcwmIzIw mszeHzmo woo woo woo m _ m _ m . m2- om $2. or szom +m m I+m m +m N _ zIommoTII AmovTII Amovxlll :18 . . mmuI o- 2 mo TOIz moI mmm +mmz _ 0 0m OuU m _ mo Q cflamsooumucm m flflmmHvam . oHom ngonqam Ammav aflfluxo ¢ 5 Im mooo .’.A w my a. «m a on 984 093.305 am mm Hm m mooo IIIIIIII. mooo IIIIIIII mooo «IMIII mooo z ++ m0 0 p momWIIJ m o m o m o m o o m m m oaz m m _ m m _ m moooIoI mo moooIou mo aHo¢ oHomzmm w oHoa oHomzmm wxommmmHaIm.m waomnmeoIm.m . baa IomommHQIm N o oHoa onmHmomoomH mxmxrumm meoua umruo 18 Mycobactins are unusual in that their two hydroxamate moieties have different acyl functions. Acylation may be the first step in the incorporation of lysine into myco- bactins. Hydroxylation of the acylated-lysine compound would follow. These steps are reversed in the biosynthe- sis of ferrichrome (Tateson, 1970). Either salicylate or 6-methylsalicylate is incorpor- ated into mycobactin depending upon the type of mycobactin synthesized by the organism (Hudson and Bentley, 1970). Conditions for the Synthesis of Microbial ITC's It is evident that ITC's act as survival factors by sequestering iron in an environment of limited iron concen- tration; when iron concentrations are sufficient, these iron transport systems are necessary. Indeed, numerous studies have shown that the iron transport systems of a variety of bacteria are repressed by high iron concentra- tions (e.g., O'Brien 2; $1., 1971; Garibaldi, 1971; Byers §£_§l., 1967; Pollack and Neilands, 1970; Peters and Warren, 1968a; Gibson and Magrath, 1969). The structural genes for the enterochelin-iron transport system of E. 221; are located together on the bacterial genome (YOung .g£.§1., 1971; Luke and Gibson, 1971), and it has been suggested that these genes form an operon which is regu- lated by iron (Luke and Gibson, 1971; Bryce and Brot, 1971). In most studies where various iron concentrations have been tested, concentrations of iron ranging from 10-5 M to 10-4 M have been found to inhibit synthesis of ITC's. Temperature is a second important factor regulating the synthesis of ITC's in a fluorescent pseudomonad and strain Tm—l of Sal. typhimurium (Garibaldi, 1971; 1972). The ability to synthesize ITC's at elevated temperatures 19 is reduced or completely lost in these organisms. Reduced synthesis of enterochelin has been observed in thirteen species of Salmonella (Garibaldi, personal communication). Host Iron Metabolism during Infection The concentration of free ionic iron in the human body is extremely low since most of the metal is complexed with protein ligands such as transferrin, lactoferrin, hemoglobin, and ferritin. The invading pathogen must either produce competitive iron chelating compounds or des- troy the host iron ligands. The association constant of transferrin for iron is about 1030. Likewise, many micro- bial ITC's have association constants for iron of approxi- mately 1030 ferrin by mycobactins and enterochelin has been demon- strated (Weinberg, 1974a). Thus it has been proposed that ITC's should be considered virulence factors (Kochan, 1973; Rogers, 1973; Weinberg, 1971). At the onset of a bacterial infection there is a and the removal of iron complexed with trans— prompt reduction in the level of serum iron. This reduc— tion is achieved by a transfer of serum iron to the liver and by blockage of intestinal absorption of iron (Wein- berg, 1974a). This serum hypoferremia can be detected before the onset of any clinical symptoms and the inten- sity of the hypoferremia can be directly correlated with the severity of the infection. If an acute infection develops, a second lowering of serum iron occurs (Wein- berg, 1971; Pekarek g5 a1., 1969; Weinberg, 1974b). Fever may play an important role in host defense against microbial infection since fever is invariable accompanied by serum hypoferremia. The rise in temper- ature may also favor the host by restricting the ability of the pathogen to synthesize its ITC's (Garibaldi, 1972; 20 Weinberg, 1974b). This hypothesis is supported by the in vitro temperature studies on various Salmonella species (Garibaldi, 1972; personal communication). However, fever may come into play only when the other defense mechanisms of the host have failed to halt the invasion, and should be viewed as a stop-gap measure to help contain the in- fection while resistance is being built up (Weinberg, 1974b). The Effect of Iron Availabilitypupon Virulence Various studies have shown that the addition of small molecular weight iron compounds enhances the growth and virulence of Klebsiella pneumoniae, Ps. aeruginosa, E, coli, Staphylococcus epidermidis, Mycobacterium fortuitum, and Listeria monocytogenes in mice and rats (Martin gt 21,, 1963; Fletcher and Goldstein, 1970; Sword, 1966). Infec- tion of iron-treated mice with a Yersinia_pestis strain of reduced virulence greatly increases the severity of the infection (Jackson and Burrow, 1956; Brubaker g5 al., 1965). When injected into iron-treated mice, an avirulent strain of L, monocytogenes, which.was incapable of growth in non-treated mice, multiplied and caused some deaths (Sword, 1966). However, injection of avirulent strains of ,X. pestis and mycobacteria into iron-treated animals does not promote infection(Kochan, 1973). Bullen.g£.al. (1967; 1968) demonstrated that injection of iron compounds into experimental animals abolished the protective effects of antisera against Yersinia septica and Clostridium welchii. In mammals, a state of hypoferremia seems to be ad— vantageous to the host and detrimental to the invading pathogen. Animals rendered hypoferremic with endotoxin are more resistant to some bacterial infections than un- treated animals (Kochan, 1973). Furthermore, the degree 21 of iron saturation of endogenous transferrin can be di- rectly related to the survival of an invading pathogen. In humans, normally 30.0% of the endogenous transferrin is complexed with iron, and the serum is tuberculostatic. The addition of iron to human serum destroys tuberculo- stasis. This tuberculostasis can be reconstituted by the addition of quantities of transferrin which thereby lower the degree of iron saturation of transferrin to normal levels. Similarly, guinea pig serum, which contains trans- ferrin that is 84.4% iron-saturated, supports the growth of mycobacteria (Kochan, 1973). Therefore the addition of transferrin to untreated serum should result in decreased virulence of the invading pathogen. Indeed, addition of transferrin causes growth inhibition of Staph. aureus in blood samples (McFarlane gg‘al., 1972), and affords a slight protection to rats and mice infected with E. Egggr moniae and fig. aeruginosa (Martin 25 a1., 1963). Like- wise, animals made hypoferremic with desferal show an enhanced resistance to infection with L, monocytogenes (Sword, 1966). Desferal also inhibits the multiplication of mycobacteria in guinea pig serum (Kochan g£.al., 1969). However, desferal enhances the growth of mutants of Sal. typhimurium and B. megaterium and counteracts the actions of sideromycins in Staph. aureus (Arceneaux and Lankford, 1966; Luckey 353;” 1972; KnI'Isei g; 11,, 1969). The growth of Bacillus stearothermophilus is not affected by desferal (Cram and Reiter, 1968). The Antibiotic Activity of Sideromycins Naturally occurring secondary hydroxamates that che- late iron have been divided into three groups; sideromycins, sideramines, and siderochromes. The sideramines are ITC's, while sideromycins, which are only synthesized by a few 22 species of actinomycetes, display antibiotic activity against a number of microorganisms. The biological role of siderochromes is undetermined (Naesch and Knusel, 1967). The antibiotic activity of sideromycins is competitively antagonized by sideramines with similar structures. This antagonism probably occurs as a result of competition for receptor sites on the cell surface (Knusel 33 al., 1969). The mechanism of action of the sideromycins has not been definitely established. While sideromycins neither disrupt hemin biosynthesis nor inactivate enzymes contain- ing hemin, there is some evidence suggesting that carbo- hydrate metabolism may be affected (Naesch and Knusel, 1967). There is no evidence, however, that the mode of action of all sideromycins is the same. In cell-free sys- tems (obtained from Staph. aureus SG511), albomycin had no effect on poly U-directed incorporation of phenylalanine; ferrimycin A1 increased incorporation; while danomycin and antibiotic A22765 inhibited incorporation. Thus sidero- mycins should be considered a homogeneous group only in regard to their competition with sideramines for transport into the cell (Knasel g; 31., 1969). Resistance is the major problem in the application of sideromycins as antibiotics. Although many sideromycins are active against Gram-positive bacteria, certain Gram- negative organisms (e.g., Shigella dysenteriae, Salmonella typhi, and PE, aeruginosa) are resistant to them. In addition, susceptible bacteria rapidly develop resistance to sideromycins (Nuesch and Knusel, 1967). 23 MATERIALS AND METHODS Microorganisms The following microorganisms were examined for pos- sible synthesis of iron transport compounds: Klebsiella pneumoniae ATCC 13883, Proteus mirabilis ATCC 9240, Proteus morganii ATCC 8019, Proteus rettgeri ATCC 9918, Proteus vulgaris ATCC 13315, Pseudomonas aeruginosa, Pseudomonas _§1uorescensl, Salmonella enteriditis ATCC 13076, Serratia marcescens ATCC 13880 (a nonpigmented mutant), Shigella boydii ATCC 9207, Shigella sonnei ATCC 9290, four clini- cally isolated strains of Staphylgcoccus aureus, Yersinia pestis Kuma P+VW', Yersinia pestis Kuma P‘VW', and Egg: sinia pseudotuberculosis PB 1/+, All cultures were routinely maintined in a glycerol- phosphate storage medium at -28 C. The storage cultures were prepared by pipetting 4.0 m1 of 0.033 M KH2P04 onto a 24 hour sloped agar culture and mixed until the bacteria were suspended in the buffer. The suspension was then pipetted into a screw-cap tube containing 6.0 m1 of sterile glycerol, mixed thoroughly, and stored at -28 C (Brubaker, personal communication). Equipment All growth study cultures were incubated in a New Brunswick gyrotory water bath shaker (model G76). Opti- cal density measurements were determined on a Perkin-Elmer double beam spectrophotometer (Coleman 124). A potentio— metric recorder (Sargent-welch, model SRG) was attached to 1This organism was isolated and identified as the con- taminant found in a preparation of deferri-penassay broth. 24 the spectrophotometer for ultraviolet scanning of the pre- pared media. During growth studies, cell-free supernatant samples were obtained by centrifugation in an Internation- al centrifuge (model SBU). Distilled water was demineral- ized by passage through a Bantam demineralizer (model BD-l). Assays for ITC's The Arnow Assay Phenolate compounds were routinely assayed by the procedure of Arnow (1937). This procedure has been slightly modified by the deletion of the final step (ad- dition of 1.0 m1 of water). Optical densities were de- termined at a wavelength of 525 nm. The Csaky Assay Hydroxamate ITC's may be assayed by the procedure of Csaky (1948). The modified procedure used in this study is outlined below. (1) A volume of 1.0 m1 of 1% (w/v) sulfanilamide (Sigma, St. Louis, M0.) in 30 % (v/v) glacial acetic acid was added to 1.0 m1 of cell—free supernatant in a test tube. (2) After the addition of 0.5 ml of 6 M H2S04, the sample was capped and boiled in a 100 C water bath for one to five hours. (3) The sample was removed from the water bath and 3.0 m1 of 35%»(w/v) sodium acetate were added. (4) A three to five minute reaction time was ob- served between the addition of 0.5 m1 of 1.3% (w/v) iodine in glacial acetic acid and the 25 subsequent addition of 0.5 m1 of the sodium arsenite reagent. (5) The color intensity was allowed to develop for five hours after the addition of 1.0 m1 of 0.3% (w/v) N-1-naphthy1ethylenediamine dihydrochloride (Sigma) in 30% (v/v) glacial acetic acid. (6) The optical density was measured at 500 nm. The sample was mixed after the addition of each reagent. The modifications from the original procedure were as follows: (1) sulfanilamide and N—1-naphthy1ethylenediamine dihydrochloride were respectively substituted for sulfa- nilic acid andcx-naphthylamine; (2) the volume of the sul- fanilamide reagent added to the sample was increased from 0.5 ml to 1.0 ml; (3) the boiling time and time allowed for color development were changed; and (4) the sample was not diluted to 10.0 ml with water. The 1.3% (w/v) iodine reagent was prepared by dis- solving 3.0 g of iodine and 4.58 g of KI in 500 m1 of glacial acetic acid. The sodium arsenite reagent was pre- and 16.0 g of NaHCO to 3 3 500 ml of water. The solution was heated to boiling with pared by adding 6.0 g of A520 steam under a hood and continued until the compounds had dissolved and CO2 evolution had ceased. The reaction is As O + 6 NaHCO -——-%- 6 C0 + 3 H 0 + 2 Na AsO 2 3 3 2 2 3 3 (Partington, 1933). The FeCl -ITC Assay 3 This procedure may be used to detect both reported types of ITC's (Gibson and Magrath, 1969). The modified procedure consisted of the addition of 1.0 m1 of cell- free supernatant to 3.0 ml of a 2%»(w/v) FeCl3 5 mM HCl. The minor modification made was the addition of solution in 26 1.0 ml of cell-free supernatant instead of 0.1 ml. The Bathophenanthroline Iron Assay Several modifications of the bathophenanthroline iron assay method of Diehl and Smith (1960) were tested for their effectiveness in measuring the low iron concentra- tions present in culture media, water, and glassware. The modification subsequently used in all iron determinations is outlined below. (1) Samples (100 ml) were heated with steam for one hour after the addition of 4.0 m1 of 15%I(w/v) L-ascorbate and 0.1 m1 of concentrated HCl. (2) After the samples had cooled to room temperature, they were transferred to separatory funnels; 5.0 m1 of 10% (w/v) sodium acetate were added and the samples were vigorously shaken. (3) Subsequently, 4.0 m1 of 1 mM bathophenanthroline (Sigma) in 50% (v/v) ethanol were added, follow- ed by vigorous shaking of the samples. (4) After a reaction time of thirty minutes, 10.0 ml of isoamyl alcohol were added. (5) The samples were shaken vigorously and allowed to stand overnight until the aqueous and alco- hol phases had completely separated. (6) The lower aqueous layers were drawn off and dis- carded. (7) The optical densities of the alcohol layers were measured at 533 nm. Contaminating iron was partially extracted from the L-ascorbate and sodium acetate reagents by the same proce- dure used in the assay itself, but the hour of steaming was omitted. The ethanol was redistilled before use as a solvent for the bathophenanthroline. No treatment of the 27 concentrated HCl and reagent grade isoamyl alcohol was necessary. Preparation of In Vitro Iron—deficient Conditions Preparation of Deferri-glassware Contaminating iron was removed from glassware by im- mersion in 50% (v/v) HCl for approximately 24 hours. Traces of HCl were removed by thorough rinsing with dis- tilled deionized water. Preparation of Deferri-water Distilled water used in the preparation of iron—de- ficient media was demineralized by passage through a de- mineralizing column (Bantam standard cartridge) followed by redistillation. Preparation of Deferri-media A liquid medium described by Waring and Werkman (1942) and a commercially prepared penassay broth (Difco, Detroit, Mi.) were used in all growth studies. The compositions of these two media are denoted in Tables 1 and 2. The medium described by Waring and Werkman (1942) was slightly modi- fied by the addition of 10 mg of nicotinamide (Sigma) per liter and will be hereafter referred to as the dextrose broth. Contaminating iron was extracted from all growth media by the method of Waring and Werkman (1942). In this pro- cedure, the iron is chelated with 8—hydroxyquinoline (8—HQ) (Sigma) and the chelated complex is removed by repeated 28 Table l.--Composition of dextrose brotha Substance Quantity per liter Dextrose 20,0 9 KZHPO4 4.0 g KH2P04 1.0 g (NH4)2504 1.0 g MgSO4-7H20 [20% (w/v) solutiorfl 0.5 m1 Nicotinamide (10 mg/ml solution) 1.0 m1 Trace mineral solutionb 0.1 ml aAdapted from Waring and Werkman (1942L bA 100 ml solution containing 44 mg of ZnSO4;7H20, 40 mg of CuSO4-5H O, 41 mg of MnSO4-4H20, and 42 mg of KI. 2 29 Table 2.--Composition of penassay brotha Substance Quantity per liter Bacto-beef extrace 1.5 g Bacto-yeast extract 1.5 g Bacto-peptone 5.0 g Bacto-dextrose 1.0 9 NaCl 3.5 g KZHPO4 3.68 g KH2P04 1.32 g aAdapted from Difco, Detroit Mi. 30 extractions with chloroform. Residual chloroform was re- moved by heating the extracted media under a hood. In order to avoid carmelization, the dextrose broth was ster- ilized by membrane filtration (Millipore Filter, type HA, pore size 0.45 pm, Millipore Corp., Bedford, Mass.). The media were examined for traces of chloroform and 8—HQ by ultraviolet scanning spectroscopy. In Vitro Growth Study Procedures Sloped nutrient agar cultures were used to inoculate 25.0 ml of the ferri-dextrose broth in an 125 m1 Erlen- meyer flask. After 24 hours of incubation at 25 C, 0.2 m1 of the culture was pipetted into a second 125 ml Erlen— meyer flask containing 25.0 ml of ferri-dextrose broth. This culture was incubated at 37 C on a shaker. During exponential phase, cells from 5.0 m1 of the culture were harvested by centrifugation. The cell-free supernatant fluid was discarded and the cell pellet resuspended in deferri-dextrose broth at a concentration of 7.5 x 108 colony forming units (CFU)/m1. One tenth milliliter of this suspension was used as the inoculum in growth stud- ies. Growth studies were performed at three different iron concentrations in the deferri-dextrose broth. A11 eXperi- mental cultures contained 70.8 ml of the deferri-dextrose broth. A 10'4 M iron concentration was prepared by asep- tically pipetting 4.2 m1 of a sterile solution (0.1 mg Fe3+/m1) into the 70.8 ml of deferri—dextrose broth. An iron concentration of approximately 10-7 M was prepared by aseptically adding 4.2 m1 of a sterile solution (0.1 pg Fe3+/ml) to the deferri-dextrose broth. The third culture contained only iron which had not been extracted from the deferri-dextrose broth and was prepared by the aseptic 31 addition of 4.2 ml of sterile doubly distilled deionized water to the deferri-dextrose broth. In growth studies using the penassay broth, ferri- penassay broth and deferri-penassay were substituted for ferri-dextrose broth and deferri-dextrose broth, respec— tively. Although the 10'4 M iron and residual iron con- centration cultures were prepared as previously described, a third iron-limiting culture was prepared by aseptically adding 4.2 ml of 10-5 M orthophenanthroline (OP) (Sigma) to 70.8 ml of the deferri-penassay broth. The final con- M. For growth studies in deferri-dextrose broth, viable centration of OP under this condition was 5.6 x 10- counts and ITC assays were performed at four-hour inter— vals over a period of 24 hours. To assure that contamin- ation of the stock cultures had not occurred, subcultures were biochemically identified before inoculation into experimental flasks. At the end of the experiment, a sample from each flask was subcultured on a nutrient agar plate to screen for contamination. Isolated colonies from each subculture were identified biochemically. For growth studies in deferri-penassay broth, viable counts and ITC assays were performed at regular intervals for a period of three days. Cultures were screened for contaminants as above at 24 hour intervals during the experiment. 32 RESULTS Procedural Results Assays for ITC's The Arnow Assay A comparison was made of the standard curves generat- ed by use of fresh reagents, one-month old (aged) reagents, and the original procedure of Arnow (1937) with aged rea- gents (Fig. 6). In all cases 2,3-dihydroxybenzoate (DHB) (Aldrich Chem. Co., Milwaukee, Wis.) was used as the stand— ard. The standard curve obtained with fresh reagents has a line equation of y = (0.018)x - (0.00272), a correlation coefficient (r) of 0.9996, and a standard error of the estimate (SEE) of :_0.011. The same reagents one month later gave the following results: (1) r = 0.9994; (2) y = (0.01687)x + (0.00617): and (3) SEE = :_o.009. The unmodified Arnow assay procedure includes the addition of 1.0 ml of water as the final step and yielded the follow— ing results: (1) y = (0.01238)x - (0.00427); (2) r = 0.99 r = 0.9999; and (3) SEE = i_0.004. Since the original assay procedure did not produce a standard curve signifi- cantly more reliable, the modified procedure was used in all subsequent growth studies. The aged reagents were used in all growth studies since a reliable standard curve was still obtained (see Fig. 6). During the course of preliminary experiments it was noted that the color intensity of the Arnow reaction di- minished with time. Accordingly, a comparison was made of the standard curves obtained with DHB at various time in- tervals after the addition of the last reagent. The results are reported in Table 3 and Fig. 6. Only three 33 Figure 6.--Arnow assay standard curves with DHB. Modified procedure standard curves with zero incubation time using fresh reagents (OJ and one-month old (aged) reagents (09. Original procedure standard curve with zero incubation time using aged reagents (‘9. Modified procedure standard curves using aged reagents with in- cubation times of zero hours 6*). three hours (.9. and seven hours (a) . 34 _ 0.8 0.7 0.6 p p P _ 3 2 cl 0 0 0 0 s o o I: nun u. 532:5 uuuuuno 50 40 )1; of NIB/o1 Figure 6 35 .0 .ma& CH memummfip mum mmEHu coflumnsocw mmwcu mo mCOHumzvm TCHH pwumHsuHmu mnam eso.o.H seam.o meemo.o I xfimmmso.oo I s we sso.o.H Seem.o meemo.o I xlemeso.ov n » m.m mao.o H. ommm.o Homso.o I xlsoeso.ov u s s sHo.o.H mmsm.o emmeo.o I xlmmeso.ov u x mm mHo.o.H msmm.o neoeo.o I xlsmmso.ov I a N mHo.o_H mmmm.o mmmoo.o I xasmmso.ov u a mo mmm u coflumsvm TCHA n mEHu mCOADMHsonu HMUHumHumum coflquSUCH .ASMmmm 3ocuflm m mcaumcflswam an pmcamuno Buoun mmonuxmp new .¢¢o shown mummmcmm .eow nuoun mmonuxmp CH m0>uso pHMUGMum mmmmmcmm 0cm mmouuxmp a“ Hmumwmmp nuw3 mm>uso pumocmum Summm MMWmoII.n musmfim o- OOH 3:30:09 mo M1. on co oa ON d - n m u — I - q 0 “.0 N O n I O 8.0 an ooq an Kalcuaq [cajado In C 0.0 «.0 .mruoun 38 Optimal boiling times for release of unbound hydrox- amates in penassay and dextrose broths were determined and are reported in Table 4, Fig. 8, and Fig. 9. The optimal boiling time in penassay broth was five hours, while one hour was the optimal boiling time in dextrose. These boiL— ing times were used in the preparation of the standard curves reported in Fig. 7 and in later growth studies. Since the color intensity of the reaction increases with time, optical density readings of the standards in the dextrose broth were taken at one-half, two and one- half, and five hour intervals after the addition of the last reagent. A significant difference in intensity was noted between the samples allowed to stand for one-half hour and those allowed to stand for two and one-half and five hours (see Fig. 10). The five hour time interval was used in the determination of standard curves as well as in later growth studies, thereby eliminating conflicts with other procedures being performed concurrently. The Csaky assay is at best a rough estimate of the quantity of hydroxamates present. Although this assay is sensitive to small quantities of hydroxamates, the harsh treatment necessary to cleave the secondary hydroxamate and release the unbound form partially destroys the com- pound. The large differences in color intensity caused by small differences in the time intervals between the addi— tion of various reagents further reduce the accuracy of this assay. The FeCl -ITC Assay 3 The standard curves generated by this assay using DHB and desferal are represented in Fig. 11. DHB yielded a line equation of y = (0.000965)x - (0.03429), an r = 0.9978, and an SEE = :_0.013. Desferal gave a line equation of 39 .m .mwm ca memum IMHo mum nuonn mammmcmm CH mmosu “m .mHm CH pmemummap mum Queen mmouu Ixmp Tau ca meowumsww mcwa one .ucmamoam>mp noHoo How ©030aam mm3 pcmm Immu umma may mo coauappm Tau Hmumm Hm>umuCH mafia masses muuflsu ¢m meo.o.H ommm.o Heoeo.o I xlmseoo.ov I a smmmmsmm o.m 830.0.H osmm.o someo.o I xamosoo.ov u » mummmemm 0.8 mao.o.H ommmIO moooo.o + onmmoo.ov u s smmmmrmm o.m Hao.o.H emmmIO smflmo.o + rammoo.ov I a mmosuxma o.m mae.o.H mmmm.o Hmeoo.o + xlmamoo.oo I » amourxmn o.H 800.0.“ Hmmm.o enamo.o + xfimomoo.ov n » mmouuxmn m.o 830.0.“ smmm.o mosso.o + xflmemoo.ov u S mmouran m~.o mum H COHumsvm mafia Esapmz TEA» mcoeumssuamo amusemeumum mcsaeom . A>Mmmm ameov Hmummmmp masons: mo mmmwamn cam meu mcwaaon cmw3uwn coaumamuuooll.v magma 4O .mcfi—fion mo mason 03”. [4| new “mcflawon no “so: mco 9? 59.132“ .05 noon m.o .IOI «mcwawon no use: mNIO .IOI ”macaw 603.3 mcflafion 3.9.58.5 um 008.3qu nuoun mmonuxmc cw Hmuwmmmp £ua3 mo>uso pnmocmum hmwmm mxmeII.m Tasman Ha\.nouommon «0 0R on." 0v." on." OOH 00 09 Orv on . _ I _. . _ . _ 1 _ I _ . q I“ «MIL. \ L 10 .II\\.\ 6 I «6 . «\t l ".0 D. o l v.0 L moo m 005 we Runner! I‘D'ndo 41 .mceawon mo mason 0.33 r... can umcwafloa mo mason “sou SAY “maeawon no muses 00mg» cl. «maonE>m .m0Eep mcwawon msoanm> um o0cflmu Ino Buoys mmmmmc0m CH H0H0mm0p sues 00>uso oumocnum manna mxmeII.m 0usmflm HE\Hnu0umoo mo ml. one ova one ooa om oo oe on 1 _ q _ _ . — a fi . q q a e — .w‘..I Il> O I ". O mu 009 :9 Katsuaa 1901360 42 Icowumnsocfl mo 0.30: 03mm :0. can «coaumnsocw mo mason mamnl0co can 03» ..O.. “coaumnaoca mo m0u5cfi= huuflnu C1 “maogm 005.3 cowumnsocw. msowume, um 00 Icwmuno nuoun 0mouuxmp CH HOH0mm0p £uw3 00>Hsu CHMUGMDm >Mmmm MMWnUII.oa musmflh , aa\-uouaoa «0 M4 0.1 on” 03 on 8 00 on «II — . — s _ s _ . _ a a a _ . I 4 mu 00g 10 £11300“ IOJIIdO 43 .9550 pumpcmum mun ...OI ps0 10>uso pumcfium Hmu0mm0p .IOI «mHOQEmm .mma 0cm H0u0mm0p £uw3 m0>uso pHMOGMDm momma UBHImaU0mII.aH 0usmflm nlxul coo con cow can con 00" mu 9;; 10 KJISUJQ [cayjdo 44 y = (0.00114l)x — (0.008535), an r = 0.9991, and an SEE = :,0.002. With the standards used, this assay is at best accu- rate down to 20 yg/ml, and therefore has little value in the monitoring of ITC synthesis during growth studies. If enterochelin had been available for use as a standard for phenolate compounds, the accuracy might have been improved slightly due to the higher association constant of entero- chelin as compared to DHB. This assay was used in preli- minary studies after 24 and 48 hours of growth to show the presence of a compound or compounds which chelate iron. The Bathophenanthroline Iron Assay A number of modifications of the bathophenanthroline iron assay procedure of Diehl and Smith (1960) were tested. The original procedure and all modified procedures (other than that reported in the section on methods) failed to reduce iron adequately from the ferric to the ferrous state. When iron is not reduced to the ferrous state, incomplete analysis of the iron content occurs since batho- phenanthroline reacts only with ferrous ions. The var- ious concentrations of hydroxylamine did not affect iron reduction. Ascorbate reduced iron only when 0.1 ml of con- centrated HCl was added and the sample was steamed for one hour. Reducing either the HCl concentration or the steam- ing time decreased the reduction of ferric ions. The mod- ification reported in the section on methods was signifi— cantly more effective in reducing ferric ions. In order to assure that this reduction was occurring, standard curves were prepared using FeCl3 and FeSO4° 7 H20 stand- ards (Fig. 12). Analysis of these standard curves indi- cates that the modified procedure reduces approximately 60% of the iron present as ferric ions. Statistical 45 .00 @500 0umowamso scum 005Hu> mo c00E 03¢ mud0m0um0u unwom noun 00 CG“ 6.842538% 5. can $.88 .6. N as. ao0m sues 00>uso cumpcuum momma some 0swaou£unuc0£monummIIINH 0usmem 1 8:: so «a 0 I" 0 In Q n N .— _ . W (cs 3' hI-uoa IWHdO 2: m Ommh 46 analysis of the FeSO4-7 H20 standard curve yielded a line equation ofyy = (0.0795)x +(0.0315), an r = 0.9997, and an SEE =.: 0.005; whereas the FeCl3 standard curve yielded a y = (0.0545)x + (0.0009), an r = 0.9998, and an SEE = :_0.002. It is impossible to determine the amount of iron add- ed to the sample from the reagents used in this assay. This introduced a maximal error of 9.33 x 10-8 M in all determinations of total iron content. Preparation of Iron-deficient Conditions Preparation.nyDeferri-water A comparison was made of the iron content of doubly distilled deionized water and water further treated by the 8-HQ method of Waring and Werkman (1942). The doubly dis- tilled deionized water had an iron content of not more than 1.18 x 10-7 M, whereas the 8-HQ treated water had an iron content of not more than 1.15 x 10'7 M. Since this difference is insignificant, doubly distilled deionized water was used in the preparation of all media and rea- gents. Preparation 9f Deferri-glassware A comparison of acid-treated flasks and flasks fur- ther treated with 8-HQ showed that differences in contam- inating iron concentrations were insignificant. The iron concentrations were respectively 9.33 x 10"8 M and 1.15 x 10_7 M. An iron concentration of 1.1 x 10'6 M was detect- ed with washed, untreated flasks. This analysis of resi- dual iron in glassware was made by simply running iron assays in the variously treated glassware. 47 Preparation g£_Deferri;media The deferri-media, prepared by the method of Waring and Werkman (1942) were analyzed for iron content. The deferri-dextrose broth had a residual iron concentration of 4.25 (i 0.933) x 10‘7 had a residual iron concentration (as detected by the bathophenanthroline iron assay) of 4.68 (i 0.933) x 10'.7 M. M, and the deferri-penassay broth However, it is possible that iron-protein complexes pre- sent in the deferri-penassay broth were not detected by this method. This hypothesis is supported by the obser- vation that after 24 hours of growth at 25 C in deferri- dextrose broth, 5, pneumoniae had accumulated 9.05‘pg/m1 of a phenolate compound. Under similar conditions in deferri-penassay broth, this organism had accumulated only 2.7‘pg/ml of a phenolate compound after 24 hours of growth. Two substances used in the iron extraction process, chloroform and 8-HQ, inhibit bacterial growth. While chloroform kills bacteria by disrupting cell membranes, the toxic effect of 8—HQ is more complex and not merely a simple metal deficiency death. The compound appears to penetrate the cell wall and membranes and to exert its toxic effect intracellularly, especially in Gram-positive organisms (Rubbo gt a1., 1950). Therefore, complete re- moval of chloroform and 8-HQ from the deferri-media must be accomplished. All deferri-media were analyzed for traces of chloro— form and 8-HQ by ultraviolet scanning spectroscopy before use in growth studies. In ferri-dextrose broth, the addi- tion of 8—HQ produces an absorption peak at 240 nm, whereas chloroform has an absorption peak at 200 nm. The absorp— tion spectra are shown in Fig. 13. It was noted that when the concentrated solution of deferri-dextrose broth was 48 .on omIm Apes o0~wa0u0cws0p u0nuusm u0um3 pouwcoa0p p0aawumwo mHASOp apes 0ESH0> ou p0usawp auoun 0mouux0cIHuu0m0p paw .Anv H0u03 p0uwc0fl0p o0HHHu Imwo mansoo :uw3 0EsHo> 0» p0usawo nuoun 0moqu0pIHuu0m0p .Amv omIm mcwcfimu Icou ruoun 0mouux0pIHuu0m mo muuo0mm coaumuomnm u0a0fi>wuuabll.ma 0usmwm Aacv guaco~o>cu com on oow com own I1 ‘ T q ‘I I I‘ J ‘ II I‘ and cow o~n com 1‘ dI ‘ III I Q d) a ‘ T 1 aallu)nlumw11 3110310,] oo— 49 diluted to volume with doubly distilled deionized water a small absorption peak appears near 270 nm (Fig. 13b). This peak is eliminated by the dilution of the concentrated solution with doubly distilled deionized water which had been further demineralized by the method of Waring and Werkman (1942) (Fig. 13c). Because a small peak occurs at 200 nm even in the absence of chloroform (Fig. 13a), the large peaks at 200 nm found in deferri-dextrose broth (Fig. 13b and Fig. 13c) probably do not represent a high concentration of chloroform. When penassay broth is deferrated by the method of Waring and Werkman (1942) a whitish precipitate occurs. This precipitate probably contains chloroform-soluble com- ponents such as lipids, fats, and hydrophobic short-chain polypeptides. Growth Study Results Preliminary Studies Preliminary studies at 25 C in deferri-penassay broth of E, mirabilis, E, morganii, g. rettgeri, 3, vulgaris, X. pseudotuberculosis, Staph. aureus BB, Staph. aureus ONT 6, Staph. aureus UNH 10, Staph. aureus Smith's diffuse, and the two strains of X, pestis showed viable counts in these cultures of approximately 109 CFU/ml after 24 hours of in- cubation. With all these organisms, assays performed at 24 and 48 hours of incubation for hydroxamate and pheno- late compounds were negative. For phenolate compounds, the assay was positive only for identical cultures of 5. pneumoniae, Sal. enteiidipig, Ser. marcescens, §h, sonnei, and Sh, boydii. For identical cultures of Sh, sonnei, Sh. boydii, gs, aeruginosa, and fig. fluorescens the Cséky assay yielded positive results. In all cases when either or 50 both the Arnow and Cséky assays yielded positive reac- tions, the FeClB-ITC assay detected the presence of a compound or compounds which chelated iron. Growth Studies in Deferri-dextrose Broth E, pneumoniae, Pg, aeruginosa, Sal. enteriditis, Ser. marcescens, and Sh. sonnei were all tested for growth patterns and synthesis of phenolates and/or hydroxamates in deferri-dextrose broth. In all cases detectable accu- mulation of these compounds was repressed in cultures con- taining an iron concentration of 10.4 M. The two iron- deficient cultures (4.25 x 10-7 M iron and 5.25 x 10.7 M iron) showed no differences in the synthesis of suspected ITC's except in the case of §h, sonnei. In all cases the phenolate compounds were not detected until cultures had reached populations of 2 x 108 CFU/ml. All recorded re- sults in this section are averages of duplicate cultures. The concentrations of phenolate or hydroxamate compounds are reported as DHB or desferal equivalent micrOgrams per 108 CFU, respectively. Iron-deficient cultures contain- ing 4.25 x 10"7 M iron were tested only at 25 C and not at 37 C. The curves of growth and phenolate synthesis of K. pneumoniae incubated at 25 C and 37 C are represented in Fig. 14 and Fig. 15. There were no significant differ- ences in the growth patterns between iron-sufficient and iron-deficient cultures at 25 C; however, after sixteen hours of incubation at 37 C in the iron-deficient culture, large numbers of organisms began to die. Production of phenolate compounds at 37 C reached much higher levels than at 25 C. Neither differences in the growth patterns nor in the levels of phenolate compounds produced by Sal. 51 5 u 10 F 1 x 10 51:107 Colony Forming Units/ml I x 10 S x 106 I 1 l J 12 lb 20 24 Hour- ot Incubation Figure 14.--Growth curves of iron-sufficient and iron— deficient dextrose broth cultures of E, pneumoniae at 25 C and 37 c. Symbols: -0-, 1 x 10"4 M Fe culture at 37 c; +, 5.25 x 10'7 M Fe culture at 37 c; +, 1 x 10‘4 M Fe culture at 25 C; -<>5 5.25 x 10'7.M Fe culture at 25 C; and +, 4.25 x 10'7 M Fe culture at 25 c. 52 5 x )0 F l x )0 ‘ 5 3 10° L DHB Equivalents 1n ya/IOa CFU l x 100 ” 5 x 10. _ 4 8 12 16 20 24 "our: of Incuhntlon Figure 15.--Phenolate accumulation curves of iron- deficient dextrose broth cultures of E, Eneumoniae at 25 C and 37 C. Symbols: +, 5.25 x 10-7 M Fe culture at 37 C; -4}7 5.25 x 10"7 M Fe culture at 25 C; and -01 4.25 x 10-7 M Fe culture at 25 C. 53 enteriditis were noted between the two temperatures. How- ever, at 25 C iron-deficient cultures containing 4.25 x 10'7 M iron attained lower cell populations (a full log unit) than did the iron-deficient cultures containing 5.25 x 10.7 M iron (Fig. 16). Cultures of Ser. marcescens were incubated only at 25 C and grew equally well in iron- deficient and iron—sufficient conditions (Fig. 17). For fig. aeruginosa, iron-sufficient and iron-deficient cultures showed equivalent growth patterns at 37 C (Fig. 18). Accumulation of significant levels of hydroxamate compounds by the iron-deficient culture was delayed until sixteen hours of incubation when the culture had been at a popu- 8 CFU/ml for four hours. Within twenty hours of incubation at 25 C and in the lation exceeding 5 x 10 presence of 4.25 x 10‘? M iron, cultures of gg, sonnei exhibited a significantly lower population than the iron— sufficient and 5.25 x 10-7 M iron cultures. In this in- stance phenolate compounds were detected four hours ear- lier (at twelve hours of incubation) in cultures contain- ing 4.25 x 10-7 M iron compared to those with 5.25 x 10-7 M iron. However, at both iron-deficient concentrations, the accumulation of phenolate compounds eventually reached nearly equal levels (Fig. 19). At 25 C hydroxamate com- pounds were detected at both iron-deficient concentrations, but only after 24 hours of incubation when the cultures with 4.25 x 10-7 M iron had reached populations of 5 x 108 CFU/ml four hours earlier, and in the case of cultures 7 M iron, a minimum of eight hours containing 5.25 x 10' earlier. The growth patterns of iron-sufficient and iron- deficient cultures at 37 C were similar. At 37 C the levels of phenolate compounds accumulated were higher than at 25 C, while hydroxamate compounds were detected after twelve hours of incubation at 37 C; the latter eventually reached a level not significantly higher than those 54 Figure l6.--Growth curves and phenolate accumulation curves of iron-sufficient and iron-deficient dextrose broth cultures of Sal. enteriditis at 25 C and 37 C. Symbols for growth curves: 4}, l x 10"4 M Fe culture at 37 C;-4y; 5.25 x 10'7 M Fe culture at 37 C;‘4}y 1 x 10'4 M Fe cul- ture at 25 C; 49; 5.25 x 10.7 M Fe culture at 25 C; and -On 4.25 x 10".7 M Fe culture at 25 C. Symbols for pheno- late accumulation curves: q‘n 5.25 x 10'7 M Fe culture at 37 C; 4, 5.25 x 10-7 M Fe culture at 25 C; and +, 4.25 x 10-7 M Fe culture at 25 C. Colony Forming Units/m1 £555 O O ‘\ l l l l )2 )6 20 24 "our: of lneuhntlon Figure 16 I x 10 s x lo" 1 x lo' 5 x 10' DHB Equivalents in )13/108 CPU 56 o 5 x 10 Colony Forming Units/m1 C a l x 10 d 5 x 10- l x 10 - DHB Equivalents mpg/108 C‘FU ! l I l to a )2 I6 20 24 Hours of Incubation Figure l7.--Growth curves and phenolate accumulation curves of iron-sufficient and iron-deficient dextrose broth cultures of Ser. marcescens at 25 C. Symbols for growth curves: +, l x 10“4 M Fe culture; -O—, 5.25 x 10-7 M Fe culture; and -<>, 4.25 x 10.7 M Fe culture. Symbols for phenolate accumulation curves: «in 5.25 x 10"7 M Fe cul- 7 ture; and -'3 4.25 x 10' M,Fe culture. 57 Colony Forming Units/n1 H‘XIO ‘ 5 x 10 Desferal Equivalents lnlpg/IO8 CPU l l I l I l 5 8 12 16 20 24 "our: of Incubation Figure 18.--Growth curves and hydroxamate accumula- tion curves of iron-sufficient and iron-deficient dextrose broth cultures of fig. aeruginosa at 37 C. Symbols: ‘43-, growth curve of l x 10"4 M Fe culture; +, growth curve of 5.25 x 10'7 M Fe culture; and «b, hydroxamate accumu- lation curve of 5.25 x 10-7.M Fe culture. 58 Figure l9.——Growth, phenolate, and hydroxamate curves of iron—sufficient and iron-deficient dextrose broth cul- tures of Sh, sonnei at 25 C and 37 C. Symbols for growth curves: -O-, l x 10"4 M Fe culture at 37 C; '9'. 5.25 x 10'7 M Fe culture at 37 C; 49; l x 10-4 M Fe culture at 25 c; -o-, 5.25 x 10'7 M Fe culture at 25 c; and +, 4.25 x 10‘7 M Fe culture at 25 C. Symbols for phenolate ac- cumulation curves: 4., 5.25 x 10-7 M Fe culture at 37 C; -¥3 5.25 x 10-7 M Fe culture at 25 C; and a», 4.25 x 10-7 M Fe culture at 25 C. Symbols for hydroxamate accumula- tion curves: 4k, 5.25 x 10'7 M Fe culture at 37 C;-¢», 5.25 x 10‘7 M Fe culture at 25 c; and -v-, 4.25 x 10'7 M Fe culture at 25 C. Colony Forming Units/ml IO 10 )0 10 IO 59 Q 0 x \ .A b r A D A I I L J 12 I6 20 2!» "our! of lncuhntlon Figure 19 5 I 10 l x 10 Son' DHB and Desferal Equlvalents Inipallo8 CFU 60 detected at 25 C (Fig. 19). A comparison of phenolate accumulation curves (Fig. 15, 16, 17, and 19) shows that at 25 C 5, pneumoniae reaches the highest level of accumulation followed by §g£. ‘marcescens,‘§h. sonnei, and Sal, enteriditis. At 37 C the differences between E, pneumoniae, gg. sonnei, and §2l, enteriditis are more pronounced while maintaining the same order of phenolate concentrations observed at 25 C. A comparison of the hydroxamate accumulation curves (Fig. 18 and Fig. 19) indicates that at 37 C hydroxamate compounds are detected four hours after populations of both,§h. 8 CFU/ml , or greater. The level of hydroxamate compounds accumulated by sonnei and fig. aeruginosa had reached 5 x 10 ,gg. aeruginosa is clearly larger than that accumulated by .Sh. sonnei. Growth Study in Deferri:penassay Broth Only'§h, sonnei (incubated at 30 C for three days) was examined for possible ITC synthesis in the deferri- penassay broth (Fig. 20). The growth curves of iron-suffi- cient and iron-deficient cultures were nearly identical as were the hydroxamate and phenolate accumulation curves of the two iron-deficient cultures. Phenolate compounds were detected six hours after the iron-deficient cultures had attained populations of 2 x 108 CFU/ml or greater, while hydroxamate compounds were detected two hours after the appearance of phenolate compounds. Hydroxamate compounds quickly reached much greater concentrations than those attained by phenolate compounds (Fig. 20). By the third day of incubation, there was an unexplained disappearance of the phenolate compounds (Fig. 20). Colony Forming Cnita/nl 5 x IOQ F 9 leO r- 5 x lo8 - " I x lo I x 10 - 7 5x10. 5 x l0 " l x 107 ~ ,I‘\ Ion' I DHB and Desferal Equivalents in‘yg/IO8 CFU JSon’ 1 x 10 ~ 5 x 10 — '0 I l i I I I 10 2O 30 40 50 60 70 72 Houra of incubation Figure 20.--Growth, phenolate, and hydroxamate curves of iron-sufficient and iron-deficient penassay broth cul- tures of §h. sonnei at 30 C. Symbols for growth curves: -0-, 1 x lo‘4 M Fe culture; +, 4.68 x 10‘7 M Fe culture; and-01 5.6 x 10-7 M OP culture. Symbols for phenolate accumulation curves: -‘-, 4.68 x 10.7 M Fe culture; and -4F, 5.6 x 10'7 M 0P culture. Symbols for hydroxamate accumulation curves: -4¥-, 4.68 x 10'7 M Fe culture; dv-, 5.6 x 10'7 M 0P culture. 62 DISCUSSION The simultaneous production of phenolate and/or hy- droxamate compounds and the occurrence of iron-chelating compounds during iron deficiency suggests that the pheno- late and hydroxamate compounds synthesized by the bacteria studied herein are ITC's. Definite proof of their involve- ment in the transport of iron requires that these compounds first be isolated and characterized. Subsequent studies 59Fe will clarify their role in the uptake of iron by iron-defi- employing 3H-labelled compounds complexed with cient cells. Previous studies have shown that increasing amounts of iron decrease ITC production and will completely halt ITC synthesis at concentrations above 10'4 M to 10"5 M (Young and Gibson, 1969; Lankford, 1973; Brot £3.2l,, 1966; Emery, 1965; Emery, 1971b). Only in the case of‘gg, sonnei were any differences detected between the two iron-deficient cultures (4.25 x 10'”7 M iron and 5.25 x 10"7 This result is not unexpected because of the minute differ- M iron). ence between the two iron-deficient concentrations. Later investigations will use iron concentrations of approximate- ly lo"6 M, 10'7 M, and lo"8 M. Decreased synthesis of ITC's at elevated temperatures has been noted only in a fluorescent pseudomonad and a number of species of Salmonella (Garibaldi, 1971; 1972; personal communication). In the case of gal. typhimurium Tm—l synthesis was decreased at 36.9 C and no growth was observed at 40.3 C (Garibaldi, 1972). Garibaldi (person- al communication) also observed decreases in synthesis of phenolates by a number of species of Salmonell§_when the incubation temperature was raised from 30.6 C to 34<3 and 38.6 C. This effect was not observed in any of the bac- teria employed in this study. On the contrary 5. 63 pneumoniae and gg. sonnei accumulated higher levels of phenolate compounds at 37 C than at 25 C. This increased accumulation may be partially due to the decreased gener- ation time at 37 C and consequently the higher total num— ber of organisms which have grown, synthesized phenolate compounds, and died. This may be an important factor in the observed slight increase in phenolate accumulation at 37 C by gg. sonnei. However, its importance in the case of E, pneumoniae is probably negligible in comparison to the large increases observed at 37 C. These throw into doubt the hypothesis that decreased synthesis of bacter- ial ITC's at elevated temperatures is a generally occurr- ing phenomenon. With E, pneumoniae, the increased accumulation of the phenolate compound(s) did not allow longevity at 37 C com- parable to the culture with 10'4 M iron. After sixteen hours of incubation at 37 C, the iron-deficient culture had begun to show a significant death phase whereas the iron-sufficient culture remained in stationary phase. The increased phenolate accumulation observed at 37 C compared to 25 C could have been partially due to increased synthe- sis when the available iron had become limiting enough to initiate a death phase. In the deferri-penassay broth, the observed lack of synthesis of either phenolate or hydroxamate compounds, coupled with good growth by species of Proteus, Yersinia, and Staphylococcus may be explained by two possibilities. Firstly, these species may synthesize an ITC which does not contain either the phenolate or hydroxamate moiety. An- other explanation might be that the deferri-penassay broth contained sufficient contaminating iron complexed with pro- teins to repress the synthesis of any ITC's. The observa- tion by Knusel gghgl. (1969) that the growth of Staph. . aureus is inhibited by some sideromycins suggests that 64 this organism possesses the ability to transport hydrox- amate compounds if not to synthesize them. The deferri-dextrose broth is a minimal medium which will not support the growth of many bacterial species. For example, Proteus, Staphylococcus and Yersinia species would not grow in the ferri-dextrose broth even with the addi- tion of various cofactors and vitamins. Since ITC synthe- sis is linked to amino acid biosynthesis, hydroxamate and phenolate synthesis might be increased by growth in a more complex deferri-broth containing exogenous amino acid com- ponents. In any case a complex deferri-broth will be re- quired to continue studies with fastidious organisms. A number of alternate procedures are available for the ex- traction of contaminating iron from more complex media (Donald §£_§l,, 1952; Wawszkiewicz 23 al., 1971). Passage of penassay broth through a Chelex 100 column resulted in a residual iron concentration in this medium of approxi- mately 10"8 M (Wawszkiewicz, personal communication). Che- lex 100 treatment of media or chelation of contaminating iron by addition of transferrin would be preferred alter- natives to the method of Waring and Werkman (1942) used in this study. A successful pathogen will mimic as many of the bio- logical characteristics of its host as possible, thereby reducing the number of specific bacterial metabolic reactions which would be susceptible to attack by host defense factors. Indeed, the major goal in therapeutic medicine is to achieve maximal damage to the pathogen and none to the host. To achieve this end, one needs to iden- tify a unique metabolic reaction, absolutely essential for the survival of the pathogen, which is absent in or of min- imal importance to the host. . The extreme importance of iron in the metabolism of microorganisms and mammals is evidenced by the extra- 65 ordinary iron chelating and transport systems which both groups possess. Whereas microbial iron transport systems utilize low molecular weight organic compounds for the sequestering of iron, such sequestering compounds in mam- malian systems are large proteins such as transferrin, lactoferrin, and ferritin. The in yiyg survival of several pathogens has been correlated with their capa- city for ITC synthesis (Kochan, 1973; Lankford, 1973). Extension of this observation to include a wide variety of pathogens will serve to define a novel area of bacterial metabolism critical in establishing an infection. Exploit- ing the vast differences in the chemical natures of micro- bial and mammalian iron sequestering compounds as well as their distinct biosynthetic pathways and control mechan- isms may be an approach to an antimicrobial therapy with practical applications. ' 66 LITERATURE CITED Akers, H.A., and J.B. Neilands. 1973. A hydroxamic acid present in Rhodotorula pilimanae cultures grown at low pH and its metabolic relation to rhodotorulic acid. Biochemistry 12: 1006-1010. Arceneaux, J.B.L., W.B. Davis, D.N. Downer, A.H. Haydon, and B.R. Byers. 1973. Fate of labelled hydroxamates during iron transport from hydroxamate-iron chelates. J. Bacteriol. 115: 919-927. Arceneaux, J;L., and C.E. Lankford. 1966. A schizokinen (siderochrome) auxotroph of Bacillus megaterium in- duced with N-methyl-N'-nitro—N-nitrosoguanidine. Biochem. BiOphys. Res. Commun. 24: 370-375. Arnow, L.E. 1937. Colorimetric determination of the com- ponents of 3,4-dihydroxyphenylalanine-tyrosine mix- tures. J. Biol. Chem. 118: 531-537. Atkin, C.L., and J.B. Neilands. 1968. Rhodotorulic acid, a diketopiperazine dihydroxamic acid with growth-fac- tor activity. I. Isolation and characterization. Biochemistry 1: 3734-3739. Brot, N., J. Goodwin, and H. Fales. 1966. .;g vivo and in vitro formation of 2,3-dihydroxybenzoylserine by Escherichia coli K 12. Biochem. Biophys. Res. Commun. ‘25: 454-461. Brubaker, R.R., E.D. Beesley, M.J. Surgalla. 1965. Pas- teurella pestis: role of pesticin I and iron in ex- perimental plague. Science 149: 422-424. Bryce, G.F., and N. Brot. 1971. Iron transport in Escher- ichia coli and its relation to the repression of 2,3-dihydroxy-N—benzoyl-L-serine synthetase. Arch. Biochem. Biophys. 142: 399-406. Bryce, G.F., and N. Brot. 1972. Studies on the enzymatic synthesis of the cyclic trimer of 2,3—dihydroxy-N— benzoyl—L-serine in Escherichia coli. Biochemistry ‘11: 1708-1715. Bullen, J.J., G.H. Cushnie, and H.J. Rogers. 1967. The abolition of the protective effect of Clostridium welchii type A antiserum by ferric iron. Immunology _12: 303-312. 67 Bullen, J.J., A.B. Wilson, G.H. Cushnie, and H.J. Rogers. 1968. The abolition of the protective effect of Pasteurella septica antiserum by iron compounds. Immunology 14: 889-898. Burnham, B.F., and J.B. Neilands. 1961. Studies on the metabolic function of the ferrichrome compounds. J. biol. Chem. 236: 554-559. Burton, M.O. 1957. Characteristics of bacteria requiring the terregens factor. Can. J. Microbiol. 3; 107-112. Byers, B.R., M.V. Powell, and C.E. Lankford. 1967. Iron- chelating hydroxamic acid (schizokinen) active in initiation of cell division in Bacillus megaterium. J. Bacteriol. 235 286-294. Csaky, T.Z. 1948. On the estimation of bound hydroxyl- amine in biological materials. Acta Chem. Scand. g; 450-454. Davis, W.B., and B.R. Byers. 1971. Active transport of iron in Bacillus megaterium: role of secondary hy- droxamic acids. J. Bacteriol. 107: 491—498. Davis, W.B., MWJ. McCauley, and B.R. Byers. 1971. Iron requirements and aluminum sensitivity of an hydrox— amic acid-requiring strain of Bacillus megaterium. J. Bacteriol. 125: 589-594. Demain, A,L., and D. Hendlin. 1959. 'Iron transport' com- pounds as growth stimulators for Microbacterium sp. J. Gen. Microbiol. 21: 72-79. Diehl, H., and G.F. Smith. 1960. The iron reagents: bathophenanthroline, 2,4,6-tripyidyl-s-triazine, phenyl-Z—pyridyl ketoxime. The G. Frederick Smith Chem. Co., 50 p. Donald, C., B.I. Passey, and R.J. Swaby. 1952. A compar- ison of methods for removing trace metals from micro- biological media. J. Gen. Microbiol. 1: 211-220. Emery, T. 1965. Isolation, characterization, and proper- ties of fusarinine, a thydroxamic acid derivative of ornithine. Biochemistry 4: 1410-1417. Emery, T. 1971a. Hydroxamic acids of natural origin. Adv. Enzymol. 35: 135-185. 68 Emery, T. 1971b. Role of ferrichrome as a ferric iono- phore in Ustilago sphaerogena. Biochemistry 10: 1483-1488. Emery, T., and L. Emery. 1973. The biological activity of some siderochrome derivatives. Biochem. Biophys. Res. Commun. g9: 670-675. Fletcher, J., and E. Goldstein. 1970. The effect of par- enteral iron preparations on experimental pyelone- phritis. Brit. J. Exp. Pathol. 51: 280-285. Frost, G.E., and H. Rosenberg. 1973. The inducible ci- trate-dependent iron transport system in Escherichia coli K12. Biochim. Biophys. Acta 330: 90-101. Garibaldi, J.A. 1970. Role of microbial iron transport compounds in the bacterial spoilage of eggs. Appl. Microbiol. .29: 558-560. Garibaldi, J.A. 1971. Influence of temperature on the iron metabolism of a fluorescent pseudomonad. J. Bacteriol. 105: 1036-1038. Garibaldi, J.A. 1972. Influence of temperature on the biosynthesis of iron transport compounds by Salmonella typhimurium. J. Bacteriol. 110: 262-265. Gibson, F., and D.I. Magrath, 1969. The isolation and characterization of a hydroxamic acid (aerobactin) formed by Aerobacter aerogenes 62-I. Biochim. Biophys. Acta 122: 175-184. Golden, C.A., I. Kochan, and D.R. Spriggs. 1974. Role of mycobactin in the growth and virulence of tubercle bacilli. Infect. Immunity'g: 34-40. Halmann, M., and J. Mager. 1967. An endogenously pro- duced substance essential for growth initiation of Pasteurella tularensis. J. Gen. Microbiol. 42: 461-468. Hayaishi, O. 1966. Crystalline oxygenases of pseudo- monads. Bacteriol. Rev._39: 720-731. Haydon, A.H., W.B. Davis, J.B.L. Arceneaux, and B.R. Byers. 1973. Hydroxamate recognition during iron transport from hydroxamate-iron chelates. J. Bacteriol. 112: 912-918. 69 Hudson, A.T., and R. Bentley. 1970. Utilization of shi- kimic acid for the formation of mycobactin S and sal- icylic acid by Mycobacterium smegmat1g. Biochemistry 2: 3984-3987. Jackson, 8.. and T.W. Burrows. 1956. The virulence-en- hancing effect of iron on non-pigmented mutants of virulent strains of Pasteurella pestis. Brit. J. Exp. Pathol.-11: 577-583. Knusel, F., B. Schiess, and W. Zimmermann. 1969. The influ- ence exerted by sideromycins on poly-U—directed in- corporation of phenylalanine in the S-30 fraction of Staphylococcus aureus. Arch. Mikrobiol. 98: 99-106. Kochan, I. 1973. The role of iron in bacterial infec- tions: with special consideration of host-tubercle bacillus interaction. Curr. Topics Microbiol. Immunol. 69: 1-30. Kochan, I., C.A. Golden, and J.A. Bukovic. 1969. Mechan- ism of tuberculostasis in mammalian serum. II. In- duction of serum tuberculostasis in guinea pigs. J. Bacteriol. 129: 64-70. Langman, L., I.G. Young, G.E. Frost, H. Rosenberg, and F. Gibson. 1972. Enterochelin system of iron transport in Escherchia coli: mutations affecting ferric- enterochelin esterase. J. Bacteriol. 112: 1142- 1149. Lankford, C.E. 1973. Bacterial assimilation of iron. Crit. Rev. Microbiol. 1: 273-331. Lankford, C.E., J.R. Walker, J.B. Reeves, N.H. Nabbut, B.R. Byers, and R.J. Jones. 1966. Inoculum-depen- dent division lag of Bacillus cultures and its rela- tion to an endogenous factor(s) ("schizokinen"). J. Bacteriol. 21: 1070-1079. Leitner, F., and S. Cohen. 1962. Effect of pH and inor- ganic salts on penicillinase formation in Staphylo- coccus aureus. J. Bacteriol. 81: 314-323. Luckey, Mt, J.R. Pollack, R. Wayne, B.N. Ames, and J.B. Neilands. 1972. Iron uptake in Salmonella typhi- murium: utilization of exogenous siderochromes as iron carriers. J. Bacteriol. 111: 731-738. 70 Luke, R.K.J., and F. Gibson. 1971. Location of three genes concerned with the conversion of 2,3-dihydroxy- benzoate into enterochelin in Escherichia co11_K-12. J. Bacteriol. 107: 557-562. Lusk, J.B., and E.P. Kennedy. 1969. Magnesium transport in Escherichia coli. J. Biol. Chem. 244: 1653-1655. Martin, C.M., J.B. Jandl, and M. Finland. 1963. Enhance- ment of acute bacterial infections in rats and mice by iron and their inhibition by human transferrin. J. Infect. Dis. 11;; 158-163. McFarlane, H., M. Okubadejo, and S. Reddy. 1972. Trans- ferrin and Staphylococcus aureus in Kwashiorkor. Amer. J. Clin. Pathol. 51: 587-591. Morrison, N.E., A.D. Antoine, and E.E. Dewbry. 1965. Synthetic metal chelators which replace the natural growth-factor requirements of Arthrobacter terregens. J. Bacteriol. 82: 1630. Neilands, J.B. 1957. Some aspects of microbial iron meta- bolism. Bacteriol. Rev..g1: 101-111. Neilands, J.B. 1967. Hydroxamic acids in nature. Science 156: 1443-1447. Nuesch, J., and F. Knusel. 1967. sideromycins, p. 499— 541. 1g: Gottlieb, D., and P.D. Shaw (eds.). Anti- biotics I. Mechanism of action. Springer-Verlag, New YOrk Inc., New York. O'Brien, I.G., G.B. Cox, and F. Gibson. 1971. Entero- chelin hydrolysis and iron metabolism in Escherichia coli. Biochim. Biophys. Acta 237: 537-549. Ong, D.E., and T.F. Emery. 1972. Ferrichrome biosyn- thesis: enzyme catalyzed formation of the hydroxamic acid group. Arch. Biochem. Biophys. 148: 77-83. Oram, J.D., and B. Reiter. 1968. Inhibition of bacteria by lactoferrin and other iron-chelating agents. Biochim. Biophys. Acta 170: 351-365. Pardee, A.B. 1968. Membrane transport proteins. Science 162: 632-637. 71 Partington, J.R. 1933. A text-book of inorganic chemistry for university students. MacMillan and Co., limited, London, 1062 p. Pekarek, R.S., K.A. Bostian, P.J. Bartelloni, F.M. Calia, and W.R. Beisel. 1969. The effects of Francisella tularensis infection on iron metabolism in man. Amer. J. Clin. Sci. 258: 14-25. Perry, R.D., and E.D. Weinberg. 1973. Effect of iron deprivation and temperature upon bacterial survival. Microbios 8: 129-135. Peters, W.J., and R,A.J. Warren. 1968a. Itoic acid syn- thesis in Bacillus subtilis. J. Bacteriol..g§: 360- 366. Peters, W.J., and R.A.J. Warren. 1968b. Phenolic acids and iron transport in Bacillus subtilis. Biochim. Biophys. Acta 165: 225-232. Pollack, J.R., and J.B. Neilands. 1970. Enterobactin, an iron transport compound from Salmonella typhimurium. Biochem. Biophys. Res. Commun. 18: 989-992. Ratledge, C., and M.J. Hall. 1972. Isolation and proper- ties of auxotrophic mutants of Mycobacter1gm_smegmatis requiring either salicylic acid or mycobactin. J. Gen. Microbiol. 22; 143-150. Ratledge, C., and B.J. Marshall. 1972. Iron transport in Mycobacterium smegmatis: the role of mycobactin. Biochim. Biophys. Acta 279: 58-74. Rogers, H.J. 1973. Iron-binding catechols and virulence in Escherichia coli. Infect. Immunity 1: 445-456. Rosenberg, A.H., and M.L. Gefter. 1969. An iron-dependent modification of several transfer RNA species in Escherichia coli. J. Mol. Biol. 46: 581-584. Rubbo, S.D., A. Albert, and M.I. Gibson. 1950. The influ- ence of chemical constitution on antibacterial activ- ity. Part V: The antibacterial action of 8-hydroxy- quinoline (oxine). Brit. J. Exp. Pathol. 11: 425— 441. Schade, A.L., N.H. Myers, and R.W. Reinhart. 1968. Car- bohydrate metabolism and production of diffusible active substances by Staphylococcus aureus grown in 72 serum at iron levels in excess of siderophilin iron— saturation and below. J. Gen. Microbiol. 51; 253- 260. Silver, 8. 1969. Active transport of magnesium in Escher- ichia coli. Proc. Nat. Acad. Sci. 62: 764-771. Silver, 8., and M.L. Kralovic. 1969. Manganese accumula- tion by Escherichia coli: evidence for a specific transport system. Biochem. Biophys. Res. Commun. 14; 640-645. Snow, G.A. 1970. Mycobactins: iron-chelating growth factors from mycobacteria. Bacteriol. Rev. 14: 99-125. Snow, G.A., and A.J. White. 1969. Chemical and biolo- gical properties of mycobactins isolated from various mycobacteria. Biochem. J. 115: 1031-1045. Straube, G., and W. Fritsche. 1973. The influence of iron concentration and temperature on growth and riboflavin overproduction of Candida guilliermondii. Biotechnol. Bioeng. Symp. fig: 225-231. Sword, C.P. 1966. Mechanisms of pathogenesis in Listeria monocytogenes infection. I. Influence of iron. J. Bacteriol. 21: 536-542. Tateson, J.E. 1970. Early steps in the biosynthesis of mycobactins P and S. Biochem. J. 118: 747-753. Theodore, T.S., and A.L. Schade. 1965. Carbohydrate meta- bolism of iron-rich and iron-poor Staphylococcus aureus. J. Gen. Microbiol. 49: 385-395. Waring, W.S., and C.H. Werkman. 1942. Growth of bacteria in an iron-free medium. Arch. Biochem. 1: 303-310. Waring, W.S., and C.H. Werkman. 1944. Iron deficiency in bacterial metabolism. Arch. Biochem. 4: 75-87. Wawszkiewicz, B.J., H.A. Schneider, B. Starcher, J. Pollack, and J.B. Neilands. 1971. Salmonellosis pacifarin activity of enterobactin. Proc. Nat. Acad. Sci. gs: 2870-2873. Weinberg, E.D. 1966. Roles of metallic ions in host- parasite interactions. Bacteriol. Rev. 10: 136-151. 73 Weinberg, E.D. 1970. Biosynthesis of secondary meta- bolites: roles of trace metals. Adv. Microbiol. Physiol. 4: 1-44. Weinberg, E.D. 1971. Roles of iron in host-parasite in- teractions. J. Infect. Dis. 124: 401-410. Weinberg, E.D. 1972. Infectious diseases influenced by trace element environment. Ann. N.Y. Acad. Sci. 199: 274-284. Weinberg, E.D. 1974a. Iron and susceptibility to infec- tious disease. Science 184: 952-956. Weinberg, E.D. 1974b. Roles of temperature in host-path- ogen interactions. Proc. Second Int. Confer. Trace Metal Metabol. in Animals 1974: 241-257. White, A.J., and G.A. Snow. 1969. Isolation of mycobac- tins from various mycobacteria. The properties of mycobactins S and H. Biochem. J. 111: 785-792. Young, I.G., and F. Gibson. 1969. Regulation of the en- zymes involved in the biosynthesis of 2,3-dihydroxy- benzoic acid in Aerobacter aerogenes and Escherichia coli. Biochim. Biophys. Acta 111: 401-411. Young, I.G., L. Langman, R.K.J. Luke, and F. Gibson. 1971. Biosynthesis of the iron-transport compound entero- chelin: mutants of Escherichia coli unable to synthe- size 2,3-dihydroxybenzoate. J. Bacteriol. 126: 51- 57. MICHIGAN STATE UNIVERSITY LIBRARIES ll IHII IHHIHI 3 1293 03103 8494