ranjini ppr

1 23

Applied Microbiology andBiotechnology ISSN 0175-7598Volume 99Number 5 Appl Microbiol Biotechnol (2015)99:2339-2349DOI 10.1007/s00253-015-6411-x

A new fluorimetric method for the detectionand quantification of siderophores usingCalcein Blue, with potential as a bacterialdetection tool

Ranjini Sankaranarayanan,AlagiachidambaramAlagumaruthanayagam & KrishnanSankaran

1 23

Your article is protected by copyright and

all rights are held exclusively by Springer-

Verlag Berlin Heidelberg. This e-offprint is

for personal use only and shall not be self-

archived in electronic repositories. If you wish

to self-archive your article, please use the

accepted manuscript version for posting on

your own website. You may further deposit

the accepted manuscript version in any

repository, provided it is only made publicly

available 12 months after official publication

or later and provided acknowledgement is

given to the original source of publication

and a link is inserted to the published article

on Springer's website. The link must be

accompanied by the following text: "The final

publication is available at link.springer.com”.

METHODS AND PROTOCOLS

A new fluorimetric method for the detection and quantificationof siderophores using Calcein Blue, with potential as a bacterialdetection tool

Ranjini Sankaranarayanan &

Alagiachidambaram Alagumaruthanayagam &

Krishnan Sankaran

Received: 22 October 2014 /Revised: 12 January 2015 /Accepted: 15 January 2015 /Published online: 30 January 2015# Springer-Verlag Berlin Heidelberg 2015

Abstract The presence of microorganisms in biologicalfluids like urine and blood is an indication of vulnerability toinfections. Iron is one of the important micronutrients requiredfor bacterial growth. In an iron-deficit environment, bacteriarelease high-affinity iron-chelating compounds calledsiderophores which can be used as non-invasive target mole-cules for the detection of such pathogens. However, only lim-ited reagents and procedures are available to detect the pres-ence of these organic molecules. The present study aims atdetecting the presence of siderophores in the iron-depletedmedia, exploiting the reversible quenching of Calcein Blueand iron(III) complex. The fluorescence of Calcein Blue isknown to be quenched in the presence of iron(III); if a strongerchelator removes this ion from the fluorophore, the fluores-cence of the fluorophore is regained. This behaviour of thefluorophore was exploited to detect and quantify siderophoresdown to 50 and 800 nM equivalent of standard siderophore,deferroxamine mesylate (desferal) in Dulbecco’s PBS andsiderophore quantification (SPQ) medium, respectively. Thesiderophores released by pathogens, equivalent to standarddesferal, were in the range of 1.29 to 5.00 μM and those fornon-pathogens were below 1.19 μM. The simple, sensitiveand cost-effective method performed in a 96-well plate wasable to detect and quantify iron chelators within 7–8 h ofincubation.

Keywords Bacteria . Calcein Blue . Fluorescence assay .

Non-invasive . Siderophores

Introduction

Iron is an essential Bmicronutrient^ because it is a cofactor inmany vital metabolic processes including the respiratory elec-tron transport chain of most microbes (Husain 2008). A re-view by Meithek and Maraheil (2007) considers iron to be avital Brivet^ for the structural and functional integrity of var-ious proteins. As the availability of free iron in the environ-ment, and particularly in biological medium, is severely lim-ited due to its high reactivity with atmospheric oxygen (Joshiet al. 2006) and sequestration by chelators respectively, mi-crobes are compelled to produce small, high-affinity iron-che-lating compounds called siderophores to acquire it.

Meithke and Maraheit classified siderophores based on thechemical nature of the oxygen donating moieties for Fe(III)coordination as catecholates (with functional catecholgroups), phenolates (with functional phenol groups),hydroxamates (with functional hydroxylamine groups) andcarboxylates (with functional α-hyroxy carboxylates or keto-hydroxy carboxylates). Integration of the chemical feature ofat least two classes into one molecule, namely catecholate-hydroxamates , phenola te-hydroxamates , c i t ra te-hydroxamates and citrate-catecholates is also possible and isreferred as Bmixed type^. The association and dissociation ofiron depend on the pKa of the siderophores which is deter-mined by the protonation of its functional groups (Miethkeand Marahiel 2007).

R. Sankaranarayanan :A. Alagumaruthanayagam :K. Sankaran (*)Centre for Biotechnology, Anna University, Sardar Patel road,Guindy, Chennai 600 025, Indiae-mail: [email protected]

K. Sankarane-mail: [email protected]

Appl Microbiol Biotechnol (2015) 99:2339–2349DOI 10.1007/s00253-015-6411-x

Author's personal copy

A study by Vagarali et al. (2008) revealed that siderophoreproduction is more frequent (about 98%) in pathogenic strainswhen compared with non-pathogenic strains (about 2 %). It isalso evident from literature that production of the virulentsiderophore pyoverdine regulates the production of at leastthree other virulence factors (Lamont et al. 2002) initiatedby the ferri-pyoverdine complex interacting with the cell sur-face receptor protein. An important adaptation of pathogens isto induce subtle changes in the siderophore molecule (like theaddition of a glucose molecule) to circumvent the host defencestrategies (Murugappan et al. 2012). This underlies our needto detect the presence of siderophores, and thus pathogenicbacteria, using a simple and sensitive method.

Among the colorimetric assays available to detectsiderophores (chrome azurole sulphonate (CAS) (Schwynand Neilands 1987); Csaky (Csaky 1948); Arnow (Arnow1937)), the standard and most commonly used technique isthe CAS assay. It is based on the colour change that the dyeundergoes when iron is removed from it, indicating the pres-ence of iron-chelating molecules (Schwyn and Neilands1987). However, these assays have limitations; the CAS assaycan detect the presence of siderophores qualitatively or in asemi-quantitative manner and utilizes numerous chemicals,some of which are toxic, for the detection of siderophores;Csaky and Arnow assays are specific for hydroxamatesiderophores and catecholate siderophores, respectively. Thisprompted us to develop a generic method for better sensitivityand quantitative analysis of these compounds.

Experimentally, fluorescence-based detection assays arepreferred over colorimetric methods owing to their high sen-sitivity, rapid response rate and relative low cost (Marencoet al. 2012). In this regard, 4-methylumbelliferone-8-methyliminodiacetic acid, commonly referred to as CalceinBlue, was chosen for this study. The fluorophore is knownto be quenched by iron (Glickstein et al. 2005; Huitink et al.1974), and siderophores can remove the iron bound by the dyeand make it fluoresce. Here, this property has been exploitedfor siderophore detection and quantitation.

Materials and methods

Collection and identification of strains

& Standard strains:The standard strains used in the study, Shigella flexneri

(2), Salmonella paratyphi (1), Salmonella entericasubspp. (1), Escherichia coli (4), Staphylococcus aureus(3), Klebsiella pneumonia (1), Proteus mirabilis (2),Staphylococcus haemolyticus (1), Klebsiella oxytoca (1),Pseudomonas aerugenosa (2), Streptococcus pneumonia(1), Lactobacillus acidophilus (1), Lactobacillus casei (1),Lactobacillus fermentum (1) and Lactobacillus lactus (1)

were obtained from American Type Culture Collection(ATCC), USA, National Collection of Type Cultures(NCTC), UK, National Culture Diary Collection (NCDC),India, and Microbial Type Culture Collection (MTCC),Chandigarh, India.

Laboratory-engineered strains of E. coli—DH5α andBL21 (DE3)—were obtained from Invitrogen, San Diego,USA, and GJ1158 was procured from Genei, Bangalore,India.

& Clinical isolates:The clinical isolates of E. coli (10), Klebsiella spp. (3),

P. mirabilis (3), Pseudomonas spp. (1) and Pseudomonasaerugenosa (1) were collected fromM/s Lister MetropolisLaboratory, Chennai, Tamil Nadu, India. The strains wereconfirmed by standard microbiological (selective mediaand motility tests), biochemical tests (methyl red/Voges-Proskauer (MR/VP) test, urease, catalase, triple sugar ironand indole tests) and 16S rRNA sequencing.

Details of the standard and clinical isolates are provided inTable 2.

Preparation of CB dye

A stock concentration of 5 mM 4-methylumbelliferone-8-methyliminodiacetic acid, commonly known as Calcein Blue(CB; Sigma-Aldrich, USA) was prepared in 0.1 M potassiumhydroxide (Merck, India), and the pH was neutralized using0.1 N hydrochloric acid (HCl; Merck, India). A 0.1-M solu-tion of Dulbecco’s phosphate-buffered saline (DPBS) wasprepared by adding 2.7 mM potassium chloride (KCl; Merck,India), 1.5 mM dihydrogen potassium phosphate (KH2PO4;Merck, India), 136.9 mM sodium chloride (NaCl; Merck, In-dia) and 8.9 mM disodium hydrogen phosphate (Na2HPO4·7H2O; Merck, India) to 1 L of distilled water, and the pH wasadjusted to 7.2 (Seto et al. 2012). The CB stock prepared wasdiluted to 200 μM (working stock) in DPBS and stored forfurther use.

Preparation of CB reagent for siderophore detection

A concentration of 10 μMCBwas prepared from the workingstock by diluting in DPBS. From this, 200 μL was added to a96-well plate and the excitation/emission (Ex./Em.) maximumwas determined using the micro-plate reader (PerkinElmer,Enspire multimode plate reader, USA).

From a stock of 10 mM ferric chloride (FeCl3; Merck,India), the quenching was studied with various concentrationsof FeCl3 in CB such that the final ratios of 200 μM CB: Ironwas 1:1, 1:5, 1:10, 1:12, 1:15 and 1:20. These mixtures werethen incubated overnight at room temperature. Two hundredmicolitres of these mixtures were then added to a 96-wellplate, and the relative fluorescence unit (RFU) of these

2340 Appl Microbiol Biotechnol (2015) 99:2339–2349


samples was read at the determined Ex./Em. wavelength in themicro-plate reader. The ratio for the best quenching efficiencywas determined and was further used.

Stern-Volmer analysis of quenched dye

In order to find the nature of quenching, CB: Iron was mixedin 1:1, 1:2, 1:3, 1:4, 1:5, 1:6, 1:7, 1:8, 1:9 and 1:10 ratios.These mixtures were incubated for 90 min and 200 μL of eachwere added to a 96-well plate, and their RFU was measured atthe determined Ex./Em. wavelength. The Stern-Volmer graphwas then plotted with the ratio of the initial (F0) and final (F)RFU at the end of 90 min against quencher concentrations.

Quantification of standard siderophore by regainof fluorescence in CB reagent

The standard siderophore, deferroxamine mesylate (desferal)obtained from Sigma-Aldrich, USA, was prepared at a con-centration of 1 mM by dissolving 3.2 mg in 2 mL of distilledwater. A standard graph was generated by varying final con-centration of desferal from 50 nM to 10 μM in 200 μL ofDPBS containing CB reagent. The plate was incubated inroom temperature for 15 min, and the RFU was measured atthe determined Ex./Em. wavelength.

Various medium experimented for siderophore detection

Luria-Bertani (LB) broth was prepared by dissolving 10 g oftryptone (Himedia, India), 5 g of yeast extract (Himedia, In-dia) and 10 g of NaCl (Merck, India) in 1 L of distilled water,and the pH was adjusted to 7.2 with 1 M sodium hydroxide(NaOH; Himedia, India) and autoclaved at 121 °C and 15 psifor 20 min.

Nutrient broth (NB) was prepared by dissolving 13 g NBpowder (Himedia, India) in 1 L of distilled water andautoclaved at 121 °C and 15 psi for 20 min.

The modified form of LB medium, known as antibiogram(AB) medium (Alagumaruthanayagam and Sankaran 2012),was used in this study. The medium was prepared by dissolv-ing 1 g tryptone (Himedia, India), 1-g yeast extract (Himedia,India), 1 g casein acid hydrolysate (Himedia, India) and 5 gNaCl (Merck, India) in 1 L of distilled water. AB mediumwasfurther modified to deplete the iron by treating with 0.8 Mcalcium carbonate (CaCO3; Merck, India) and keeping inrocking agitation for 30 min. This was then centrifuged at8000 rpm for 10 min, and the supernatant was carefully aspi-rated from the tubes. The pH of the medium was adjusted to6.3 using 0.1 N HCl. The medium was then autoclaved at121 °C and 15 psi for 20 min and stored for further studies.This CaCO3-treated AB medium will be further referred assiderophore quantification (SPQ) medium. A standard graphwas generated using the protocol mentioned in the previous

section by varying final concentration of desferal from800 nM to 10 μM with SPQ medium. The standard CASliquid assay was also performed with this medium using vary-ing concentrations of desferal (10 to 100 μM), as described bySchwyn and Neilands (1987). After reaching equilibrium, theabsorbance of the samples was read at 630 nm.

Elemental analysis for iron content

The amount of iron present in the medium before and afterCaCO3 treatment was analysed using atomic absorption spec-trometry (AAS) (Varian Spectra AA 220 Atomic AbsorptionSpectrophotometer) by measuring at 372 nm. A similar con-dition was stimulated in double-distilled water with FeCl3such that the final concentrations of iron were 2.18, 0.71, 0.5and 0.1 mg/L. One hundred seventy microlitres of these solu-tions were taken in a 96-well plate, and 20 μL of 60 μMdesferal was added to all the wells such that the final concen-tration of the siderophore was 6 μM. Tenmicrolitres of the CBreagent was then added to these solutions, and the RFU wasmeasured at the determined Ex./Em. wavelength after 15 min.

Siderophore production by different number of cells

The siderophore production by various dilutions of standardstrains (S. flexneri, Sa. paratyphi, St. aureus, K. pneumoniaand P. mirabilis) grown overnight in SPQ medium was seri-ally diluted as two batches, in triplicate, in a 96-well plate suchthat each well contained 190 μL of 108,107, 106, 105, 104, 103,102 and 10 cells. One batch was incubated at 37 °C (orbitalshaker) and 100 rpm for 7 h. Following 7-h incubation, 10 μLof the CB reagent was added to the cultures, incubated in roomtemperature for 15 min, and the RFU at the determined Ex./Em. wavelength was read. To the second batch, 10 μL of CBreagent was directly added to the diluted culture, and the RFUwas read after 15 min.

Fluorescence-based detection of siderophores

The assay was validated using a variety of standard and clin-ical isolates. In a 96-well plate, 20 μL of 104 bacterial cellswere inoculated in 170 μL of SPQ medium and incubated at37 °C, 100 rpm for 7 h, and the optical densities of the cultureswere measured at 600 nm in the micro-plate reader. The assaywas performed by adding 10 μL of CB reagent to all the wellsand was incubated at room temperature for 15 min. The ex-periment was repeated twice, with each repeat including threereplicates per bacterial strain. The signal for each sample wasmeasured at the determined Ex./Em. wavelength. The concen-tration of siderophores released, as equivalent of desferal, wasestimated using the standard graph of desferal in medium.

Appl Microbiol Biotechnol (2015) 99:2339–2349 2341


Statistical analysis

The significance of the amount of siderophores released indifferent groups of bacteria was calculated by one-wayANOVA, using the SPSS 20.0 software (IBM SPSS Statistics20). Tukey’s post hoc analysis (honestly significant differencetest) was used to identify the homogeneous subsets. A p valueof <0.05 was considered significant.

Results

Siderophores are vital to the survival of pathogenic bacteria,which have to colonize a severely iron-depleted environment.Presence of siderophores therefore serves as a marker for path-ogens, but the existing colorimetric method, CAS assay, hasserious limitations like complex reagent preparation and lowsensitivity. A simpler fluorimetric assay for easier estimationof siderophores has been developed, and the results are givenbelow.

Evaluation of CB dye suitability for siderophore detection

The reversal of quenching of CB by iron as a function ofsiderophore concentration formed the basis of this method.In literature, the Ex./Em. λmax of CB was reported as 330/455 and 360/455 nm at pH 4.0 and 10.0, respectively (Huitinket al. 1974). We found that Ex./Em. of CB under the growthconditions of bacteria around neutral pH in DPBS (pH 7.2)was 350/431 nm.

Once the Ex./Em. λmax was fixed in DPBS, titrating for themaximum quenching of the dye (at various concentrations)using iron showed that quenching increases with time whenobserved on hourly basis and could be completed only whenthe mixture was left overnight. Figure 1a shows the quenchingof 200 μM CB (concentration for saturating fluorescence)incubated overnight with 200 μM–4 mM of iron. Above2 mM (ten times that of the dye), iron precipitated out andthe quenching remained the same. Hence, to avoid excess-freeiron in the reagent, the 1:10 ratio of CB (200 μM)/Fe3+

(2 mM), which gave the maximum quenching of the dyewas used for further studies; this reagent was found to bestable for more than a year. A 20 times dilution of the abovequenched dye reagent with the composition CB (10 μM)/Fe3+

(100 μM), whose RFU was 28,000 compared to 330,000 ofthe normal dye at 10 μM, was used for further experiments(Fig. 1b).

Stern-Volmer analysis of CB quenching for Fe(III)

Iron quenches CB by forming a 1:1 coordination complex(Thomas et al. 1999). According to Stern-Volmer, thequenching of fluorophores can be static (by complex

formation), dynamic (by collision) or a combination of both(mixed) (Lakowicz 2006). The fluorescence recorded for CBat 431 nm prior to (F0) and after (F) the addition of iron wasplotted against quencher concentrations to determine the na-ture of quenching. Figure 2 shows the Stern-Volmer plot forquenching of CB. From the relationship, it was concluded thatthe quenching was of mixed type due to the deviation of thecurve towards the y-axis.

CB reagent for detection of siderophores in DPBS

Desferal, a commercially available siderophore, was used totest and standardize the maximum reversal of fluorescenceunder bacterial growth conditions. When the regain of fluo-rescence from the quenched reagent was tested with varyingconcentrations of desferal, as shown in Fig. 3a, it was linearfrom 50 nM to 10 μM. The sensitivity of 50 nMwas adequatefor quantifying siderophores.

Fig. 1 Optimized Calcein Blue quenching with Fe3+ for CB assay. a Thedecrease in fluorescence of Calcein Blue (200 μM) with increasing con-centrations of iron (in the ratio of 1:1; 1:5; 1:10; 1:12; 1:15 and 1:20). Aniron concentration ten times that of the dye was required for effectivequenching. b The emission spectra (exCitation 350 nm) of the fluorescentand quenched form of Calcein Blue. The vast difference in the emissionof fluorescence between the two forms is shown



After testing the assay using standard desferal in DPBS, itwas standardized in the bacterial culture medium (Fig. 3b).

Medium selection for CB reagent performance and efficientbacterial growth

The popular LB and NB media showing effective bacterialgrowth were tested for detection of siderophores released byvarious uropathogens using quenched CB reagent. Thoughthe growth was found to be superior in these media, LB orNB by itself showed a high background fluorescence in therange of 80,000–90,000 RFU in the Ex./Em. range of the dye.Hence, these media were not suitable for our study due toa u t o - f l u o r e s c e n c e . H e n c e , t h e AB me d i um(Alagumaruthanayagam and Sankaran 2012), a depleted andsupplemented LB medium developed in our lab for slightlyretarded growth rate but suitable for fluorescence studies, wastested and the auto-fluorescence was found to be lower (50,000 RFU) than LB or NB. Inorganic medium like M9 was notsuitable due to slow growth rate. Further studies were doneusing the AB medium to obtain a signal (expressed as RFU)suitable for siderophore measurement.

Free iron content analysis in medium and conditionoptimization for CB assay

The assay relies on the removal of iron from CB-iron complexby the siderophores released by bacteria. Hence, the quantifi-cation and removal of free iron from the AB medium were anessential step. The AAS analysis showed that AB mediumcontained 2.18 mg/L of iron. When it was treated with0.8 M CaCO3 to remove free iron, there was a threefold re-duction of free iron to 0.71 mg/L (SPQ medium). To estimatethe interference of free iron with the assay, different concen-trations of iron (2.18, 0.71, 0.5 and 0.1 mg/L) in double-

distilled water was prepared and the competition between freeiron and the iron in the quenched reagent was studied. Theresults showed that the free iron interference with quenchingof dye was distinct above 0.71 mg/L and not significant in thelower range (Table 1). The CaCO3 treatment also helped inreducing the auto-fluorescence of the medium in the Ex./Em.range of CB from 50,000 RFU to 40,000 RFU. When the CBreagent was added in the SPQ medium (CB (10 μM)/Fe3+

(100 μM)), the initial RFU again increased to 55,000.The standard curve for re-fluorescing of quenched dye with

desferal in the SPQ medium was generated and found to besensitive up to 800 nM (Fig. 3b) of the siderophore. The re-duced sensitivity in medium compared to DPBS (50 nM) isdue to medium auto-fluorescence. The sensitivity of the CASassay was also determined in the SPQ medium, and the min-imum amount of desferal required for effective change in ab-sorbance was found to be 20 μM (Fig. 3b).

Estimation of siderophore production in different celldensities immediately and after 7-h growth

The production of siderophores by various cell densities ofdifferent bacteria was tested using the developed assay. Theovernight cultures of five organisms grown separately in SPQmedium were serially diluted from 108 to 10 (in the order of10) in 200 μL of the medium in two sets. When one set wastreated with the CB reagent immediately and the regainedfluorescence was read after 15 min, 106–107 cells were re-quired for siderophore detection (Fig. 4a). When the secondset was allowed to grow for 7 h and then the assay was per-formed, it showed detectable levels of siderophore even for aninoculum size of ten cells for S. flexneri, Sa. paratyphi, St.aureus and P. mirabilis, while slow growing organisms likeK. pneumonia (as in this case) showed a detectable signal onlyfor an inoculum size of 105 cells (Fig. 4b). Seven hours of

Fig. 2 Stern Volmer Plotbetween F0/F and quencherconcentration. The Stern-Volmerplot shows a parabolic relation-ship between Calcein Blue andiron(III), characteristic of mixedtype of interaction between themin which the quenching is due toboth complex formation andcollision



growth was found to be effective for measuring detectableamounts of siderophore from a variety of organisms.

Validation and statistical significance of the CB assay usingdifferent standard strains and uropathogens

After confirming with the set of five strains, the assay wasvalidated with three laboratory E. coli strains including popu-lar recombinant hosts (DH5α, BL21DE3, GJ1158), 4Lactobacillus spp. and 36 different isolates (2 Shigella, 2Salmonella, 14 E. coli, 3 Staphylococcus, 5 Klebsiella, 5Proteus, 4 Pseudomonas, 1 Streptococcus), and the resultsare shown in Table 2. The strains with an initial inoculum of104 grown in SPQ medium showed higher fluorescencevalues for pathogenic strains (21,000 to 85,000 RFU), where-as for non-pathogenic and engineered strains, it was in the

range of 9,000 to 20,000 RFU. There was a statistically sig-nificant difference between the groups as determined by theone-way ANOVA (F(42,215)=14.717, p=0.0001). The anal-ysis showed four distinct subsets which were based on theirmean RFU values. These subsets were categorized as non-producers of siderophores (with RFU expressed as mean±standard error of mean in the range −5,382.33±3,757.95 to−2,907.50±1,461.97), low producers of siderophores (withRFU expressed as mean±standard error of mean in the range11,120.50±2,729.80 to 15,349.83±1,306.62), intermediate-producers of siderophores (with RFU expressed as mean±standard error of mean in the range 19,355.83±4,150.75 to46,431.00±1,683.35) and high producers of siderophores(with with RFU expressed as mean±standard error of meanin the range 51,125.83±7824.85 to 80,253.50±2324.36). Thisclassification based on siderophore production is highlightedin the Table 2.

Discussion

The correlation between the production of bacterial virulencefactors like siderophores and host response has been predictedby West and Buckling (2002). Siderophore production byGram-negative bacteria like E. coli, Proteus spp.,M. tuberculosis and Legionella spp. (Adler et al. 2014;Himpsl et al. 2010; Wells et al. 2013) and Gram-positive bac-teria like Bacillus spp., St. aureus (Dale et al. 2004;Zawadzkaa et al. 2009) have been studied in detail. Therefore,we chose siderophores as markers for detecting the presenceof bacterial pathogens. However, currently available methodsbased on the functional groups of these molecules did notpermit their quantitative estimation irrespective of chemicaltypes. The CB-based fluorescence assay reported here had tobe devised based on the common function of iron sequestra-tion by siderophores. Fluorescence has been used as a sensi-tive indicator of this sequestration. Thus, this assay is the firstof its kind in siderophore estimation, and the quantitative re-sults are given as equivalents of desferal.

The choice of the medium for bacterial growth is crucial, asit should support growth as well as siderophore production ofa variety of bacteria. Minimal media such as the M9 or anyother type of inorganic medium (Armstrong and Clements1993; Rondon et al. 2004) are routinely used to studysiderophore production in bacteria. These media facilitatesiderophore production, but variability in growth due to itsminimal nature (Liles et al. 2000) results in the time-consuming process with most studies reporting well over24 h of growth (Sayyed et al. 2005; Vesper et al. 2000). Ad-dition of desferal or 2,2′-bipyridyl to LB medium hinders thegrowth of certain organisms (Himpsl et al. 2010), though stud-ies can be performed with overnight cultures (Rondon et al.2004). Other iron chelators like EDTA and EDDA are also

Fig. 3 Standard graph for regain of fluorescence from quenched CalceinBlue by desferal in DPBS and SPQ medium. a The increased gain influorescence from the quenched Calcein Blue (the starting RFU of thedye has been set to zero) as a function of desferal (commerciallyavailable siderophore) concentration with linearity ranging from 50 nm to10 μMof desferal. The inset shows the relationship at lower concentrationsof desferal. b The relationship in SPQ medium developed for siderophoredetection. The precision of the method was found to be accurate from800 nM of desferal. The inset shows the result of the CAS assay withabsorbance at 630 nm expressed as a function of siderophore concentration



used for the removal of iron from the growth medium, buttheir use is limited owing to their inhibitory effect on bacterialgrowth (Marcelis et al. 1978; Qiu et al. 2010). In this study,availability of free iron has been limited by the addition ofCaCO3 to the medium. Thus, SPQ medium is a new rarefiedform of LB medium, where the nutrients required for micro-bial growth, except for iron, are present adequately.

Central to this method is the reversible conversion of CBbetween its fluorescent and non-fluorescent forms dependingupon whether it is free or bound to iron. Since the bindingaffinity of the molecule for iron is moderate, it is a good carrierof iron from which siderophores can sequester the iron. It hasalso been studied as a sensor for iron owing to its specificityand suggested to be a useful tool for monitoring cytosolic ironand assessing the dynamics of intracellular iron in living cells(Glickstein et al. 2005; Seto et al. 2012). Since its fluorescenceintensity is pH-dependant (Wilikins 1960), conditions havebeen standardized at neutral medium pH at which growth isoptimal and siderophore detection is efficient.

While iron is a known quencher of CB, very few studiesemphasize on the nature of this quenching. Thomas et al.(1999) reported that the CB forms an equimolar complex withiron. The coordinate complex formation with ferric iron takesplace by the associative donation and the dissociative inter-change of a coordinated water molecule with phenolate oxy-gen. This is followed by the rapid donation of the aminonitrogen and carboxylate oxygen(s) to complete the(tetradentate) iron(III) complex (Yoshida et al. 1993). Cyto-solic iron sensors based on CB have indicated that CB bindsiron with a stoichiometry of 2:1 to 3:1 (Glickstein et al. 2005).Our results show that effective quenching occurred only atiron concentrations ten times that of the fluorophore. Thequenching of fluorescence studied using the Stern-Volmer plotshowed that quenching is a mixed type, both by complexformation and by collision.

Quantitative nature of the assay with the ability to detectsiderophores at nanomolar levels is useful in getting a fair ideaabout not only the low amount of siderophores present in asample but also the number of bacterial cells, since the in-crease in fluorescence was found proportional to the numberof cells. After 7 h of growth, even as few as ten cells in theinoculum could be detected. This is advantageous when

Table 1 Interference of free iron in medium with desferal quantification using CB reagent

Condition Blank Control Different concentrations of iron in double-distilled water

0.1 mg/L 0.5 mg/L 0.7 mg/L 2.18 mg/L

Initial difference in RFU 0 12,083 11,499 11,803 11,198 4,138

Difference in RFU after 15 min incubation 0 44,366 35,625 34,575 30,460 5,986

Blank double-distilled water with CB reagent, Control double-distilled water with CB reagent and 6-μM desferal, RFU relative fluorescence unit,Samples double-distilled water with CB reagent, iron (different concentrations) and 6-μM desferal

Fig. 4 Sensitivity of the method as a function of number of differentcommonly encountered pathogenic bacteria in the culture (a) and aftergrowth for 7 h (b). The detection sensitivity of the method in a seriallydiluted batch of cultured bacteria adjusted to an initial concentration of108 cells/mL is shown. As can be seen, the method is generally sensitiveto detect bacteria with cell densities above 106 cells/mL (a). However,when the cultures were grown from their inocula for 7 h, as can be seenfrom b, some of the bacteria like S. flexneri, Sa. paratyphi, St. aureus andP. mirabilis could be detected from a low abundance of 10 cells/mL,while K. pneumonia required 105 cells/mL of initial inocula reflectingthe possible influence of growth rate, siderophore production and thepotency of siderophores in iron acquisition from Calcein Blue



Tab

le2

Validationof

CBassayusingdesferal,standardandclinicalisolates

S.no.

Strain/standard

RFU

-1(m

ean

oftriplicate

values)

Concentratio

nof

siderophore

equivalent

todesferal(μM)

RFU

-2(m

ean

oftriplicate

values)

Concentratio

nof

siderophoreequivalent

todesferal(μM)

1La

ctobacillus

acidophilus(N

CDC15)c

−3657

–−4

425

–

2La

ctobacillus

casei(NCDC17)c

−4217

–−6

548

–

3La

ctobacillus

ferm

entum(M

TCC903)

c−4

059

–−1

756

–

4La

ctobacillus

lactus

(MTCC440)

c−6

800

–−5

24–

5BL21(D

E3)

d18355

1.08

12944

0.76

b

6GJ1158d

15072

0.89

15628

0.92

7Pseudom

onas

aerugenosa

(ATCC27853)

d12544

0.74

b10785

0.64

b

8Pseudom

onas

aerugenosa

(ATCC10415)

d12758

0.75

b13289

0.78

b

9DH5α

e9297

0.55

b20357

1.19

10Shigella

flexneri(ATCC29508)

e33189

1.95

52602

3.09

11Shigella

flexneri(MTCC9543)e

46743

2.75

37063

2.18

12Salmonella

paratyphi(MTCC3220)e

40067

2.36

41411

2.43

13Salmonella

enterica

subspp.(MTCC3231)e

33193

1.95

28382

1.67

14Escherichia

coli(ATCC25922)

e30039

1.77

30420

1.79

15Staphylococcus

aureus

(NCTC3750)e

30673

1.81

48438

2.85

16Klebsiella

pneumonia

(ATCC13883)

e41359

2.44

26963

1.59

17Staphylococcus

aureus

(MTCC6908)e

54998

3.24

47544

2.8

18Staphyloccoushaem

olyticus

(MTCC8924)e

26073

1.54

21158

1.25

19Escherichia

coli(ATCC8739)e

22863

1.35

31904

1.88

20Escherichia

coli(M

TCC9537)e

35065

2.07

21890

1.29

21Escherichia

colia

(21595)e

34652

2.04

24550

1.46

22Escherichia

colia

(25922)e

42357

2.49

25499

1.50

23Escherichia

colia

(339971)

e23172

1.36

28885

1.70

24Escherichia

colia

(340046)

e34678

2.04

27059

1.60

25Escherichia

colia

(111406070)e

45077

2.65

34296

2.02

26Proteus

mirabilisa(307316)

e25433

1.50

26414

1.56

27Proteus

mirabilisa(3401488)e

31706

1.87

36876

2.17

28Proteus

mirabilisa(5155)

e51751

3.05

37454

2.21

29Pseudom

onas

aerugenosa

a(2653)

e28530

1.68

37456

2.20

30Escherichia

colia

(406612)

e25218

1.49

26310

1.55

31Escherichia

coli(ATCC13534)

f63161

3.72

44791

2.63

32Proteus

mirabilis(ATCC7002)f

85249

5.02

75258

4.43

33Proteus

mirabilis(ATCC29906)

f61354

3.61

74911

4.41



Tab

le2

(contin

ued)

S.no.

Strain/standard

RFU

-1(m

ean

oftriplicate

values)

Concentratio

nof

siderophore

equivalent

todesferal(μM)

RFU

-2(m

ean

oftriplicate

values)

Concentratio

nof

siderophoreequivalent

todesferal(μM)

34Klebsiella

oxytoca(M

TCC2275)f

54045

3.18

48731

2.87

35Streptococcuspneumoniae(ATCC33400)

f67085

3.95

61520

3.62

36Escherichia

colia

(21728)f

65404

3.85

53170

3.13

37Escherichia

colia

(21748)f

63471

3.74

52625

3.10

38Escherichia

colia

(21784)f

61986

3.65

60565

3.56

39Klebsiella

spp.a(340053)

f68251

4.02

47140

2.78

40Escherichia

colia

(318253)

f65490

3.86

36762

2.17

41Klebsiella

spp.a(340053)

f77291

4.55

53197

3.13

42Klebsiella

spp.a(4483)

f66419

3.91

44453

2.62

43Pseudom

onas

spp.a(27853)f

59980

3.53

55633

3.28

RFUrelativ

efluorescence

unit

aM/sListerMetropolis

Laboratory

bThese

estim

ates

fallbelowthereliablesensitivity

levelsof

theassay

cNon-producersof

siderophores

dLow

producersof

siderophores

eInterm

ediateproducersof

siderophores

fHighproducersof

siderophores



dealing with early infections, when infection is suspected orfor screening. The developed assay also shows sensitivity thatis 25 times superior to the CAS assay in the developed medi-um. However, the relative sensitivity of the different types ofsiderophores in the new assay could not be studied for want ofpurified siderophores. As the method relies on only the ironacquisition property of the siderophore from the fluorophore,it is presumed that the types of siderophore released by thebacteria might not play a role here. However, from the com-parable values of regained RFU using a variety of bacteria (forexample, E. coli and Shigella spp. known to producecatacholate and hydroxamate siderophores respectively(Payne 1980)), it appeared that our premise could be correct.

The limited survey of pathogenic and non-pathogenicstrains reported here suggested that the former generally se-crete more siderophores than the latter. Vagarali et al. (2008)have reported that pathogenic E. coli produces moresiderophores than their non-pathogenic varients. Studies per-formed with pathogenic microbes like Pseudomonas spp.,E. coli, B. anthracis and Staphylococcus spp. (Adler et al.2012, 2014; Dale et al. 2004; Himpsl et al. 2010; Lamontet al. 2002) have emphasized on the alternative roles ofsiderophores in pathogens, justifying the need for their exces-sive production. In contrast, Lactobacilli are not known toproduce siderophores (Krewulak and Vogel 2008), andBorrelia burgdorferi has evolved to substitute manganese inits metal requiring enzymes (Skaar 2010), to overcome theirdependence on iron. This correlation with the reports is en-couraging as, with more intensive studies and larger surveys,parameters based on siderophore production can be defined toidentify the presence of pathogens.

In conclusion, the new method of bacterial siderophoredetection and quantification is applicable universally, irrespec-tive of the bacteria and the type of siderophore and the type ofsample. Apart from its analytical advantage, it is simple, sen-sitive and cost-effective than the existing methods to know ifbacteria are present in a sample. It has the potential to detectthe presence of pathogenic strains by virtue of their ability toproduce higher levels of siderophores under iron stress condi-tions compared to non-pathogens.

Acknowledgments We are grateful to Mr. Suresh Lingham, M/sTrivitron Pvt Ltd. for clinical samples, Dr. Sridhar, Dept. ofMicrobiology,Sri Ramachandra University, for providing standard bacterial cultures andDr. J. Saibaba, National Agro Foundation, Chennai for the analytical data.We acknowledge the financial support from Centre with Potential forExcellence in Environmental Science (CPEES) of University GrantsCommission.

Conflict of interest There is no conflict of interests for the authors forsubmitting this article. This work was supported by the Centre with Po-tential for Excellence in Environmental Science (CPEES) of UniversityGrants Commission, India. They have no involvements in the study de-sign, in the collection, analysis and interpretation of data, in the writing ofthe article and in the decision to submit the article for publication.

References

Adler C, Corbalán NS, Seyedsayamdost MR, Pomares MF, De CristóbalRE, Clardy J, Kolter R, Vincent PA (2012) Catecholate siderophoresprotect bacteria from pyochelin toxicity. PLoS One 7(10):e46754

Adler C, Corbalan NS, Peralta DR, Pomares MF, De Cristóbal RE,Vincent PA (2014) The alternative role of Enterobactin as an oxida-tive stress protector allows Escherichia coli colony development.PLoS One 9(1):e84734

Alagumaruthanayagam A, Sankaran K (2012) High-throughput fluores-cence-based early antibiogram determination using clinicalEscherichia coli isolates as case study. Microb Drug Resist 18(6):586–596

Armstrong SK, Clements MO (1993) Isolation and characterization ofBordetella bronchiseptica mutants deficient in siderophore activity.J Bacteriol 175(4):1144

Arnow LE (1937) Colorimetric determination of the components of 3,4-dihydroxyphenylalanine tyrosinemixtures. J Biol Chem 118:531–537

Csaky TZ (1948) On the estimation of bound hydroxamine in biologicalmaterials. Acta Chem Scand 2:450–454

Dale SE, Doherty-Kirby A, Lajoie G, Heinrichs DE (2004) Role ofsiderophore biosynthesis in virulence of Staphylococcus aureus:identification and characterization of genes involved in productionof a siderophore. Infect Immun 72(1):29

Glickstein H, El RB, Shvartsman M, Cabantchik ZI (2005) Intracellularlabile iron pools as direct targets of iron chelators: a fuorescencestudy of chelator action in living cells. Blood 106(9):3242–3250

Himpsl SD, Pearson MM, Arewång CJ, Nusca TD, Sherman DH,Mobley HL (2010) Proteobactin and a Yersiniabactin-relatedsiderophore mediate iron acquisition in Proteus mirabilis. MolMicrobiol 78(1):138–157

Huitink GM, Poe DP, Diehl H (1974) On the properties of Calcein Blue.Talanta 21(12):1221–1229

Husain S (2008) Effect of ferric iron on siderophore production andpyrene degradation by Pseudomonas fluorescens 29 L. CurrMicrobiol 57:331–334

Joshi F, Archana G, Desai A (2006) Siderophore cross-utilizationamongst Rhizospheric Bacteria and the role of their differential af-finities for Fe3+ on growth stimulation under iron-limited conditions.Curr Microbiol 53:141–147

Krewulak KD, Vogel HJ (2008) Structural biology of bacterial iron up-take. Biochim Biophys Acta 1778:1781–1804

Lakowicz JR (2006) Principles of fluorescence spectroscopy. Springer,New York

Lamont IL, Beare PA, Ochsner U, Vasil AI, Vasil ML (2002)Siderophore-mediated signaling regulates virulence factor produc-tion in Pseudomonas aeruginosa. PNAS 99(10):7072–7077

Liles MR, Scheel TA, Cianciotto NP (2000) Discovery of a nonclassicalsiderophore, legiobactin, produced by strains of Legionellapneumophila. J Bacteriol 182(3):749–757

Marcelis JH, Daas-Slagt HJD, Hoogkamp-Korstanje JAA (1978) Ironrequirement and chelator production of staphylococci,Streptococeus faecalis and Enterobacteriaceae. Antonie VanLeeuwenhoek 44(3–4):257–267

Marenco MJC, Fowley C, Hyland BW, Galindo-Riaño D, Sahoo SK,Callan JF (2012) A new fluorescent sensor for the determinationof iron(III) in semi-aqueous solution. J Fluoresc 22:795–798

Miethke M, Marahiel MA (2007) Siderophore-based iron acquisition andpathogen control. Microbiol Mol Biol Rev 71(3):413–451

Murugappan RM, Karthikeyan M, Aravinth A, Alamelu MR (2012)Siderophore-mediated iron uptake promotes yeast–bacterial symbi-osis. Appl Biochem Biotechnol 168:2170–2183

Payne SM (1980) Synthesis and utilization of siderophores by Shigellaflexneri. J Bacteriol 143(3):1420–1424



Qiu DH, Huang ZL, Zhou T, Shen C, Hider RC (2010) In vitro inhibitionof bacterial growth by iron chelators. FEMS Microbiol Lett 314(2):107–111

Rondon MR, Ballering KS, Thomas MG (2004) Identification and anal-ysis of a siderophore biosynthetic gene cluster from Agrobacteriumtumefaciens C58. Microbiology 150:3857–3866

Sayyed RZ, Badgujar MD, Sonawane HM, Mhaske MM, ChinchlokarSB (2005) Production of microbial iron chelators (siderophores) byfluorescent Pseudomonads. Indian J Biotechnol 4:484–490

SchwynB,Neilands JB (1987) Universal chemical assay for the detectionand determination of siderophores. Anal Biochem 160(1):47–56

Seto D, Maki T, Soh N, Nakano K, Ishimatsu R, Imato T (2012) A simpleand selective fluorometric assay for dopamine using a Calcein Blue–Fe2+ complex fluorophore. Talanta 94:36–43

Skaar EP (2010) The battle for iron between bacterial pathogens and theirvertebrate hosts. PLoS Pathog 6(8):e1000949

Thomas F, Serratrice G, Béguin C, Aman ES, Pierre JL, Fontecave M,Laulhère JP (1999) Calcein as a fluorescent probe for ferric iron: appli-cation to iron nutrition in plant cells. J Biol Chem 274:13375–13383

Vagarali MA, Karadesai SG, Patil CS, Metgud SC, Mutnal MB (2008)Haemagglutination and siderophore production as the urovirulencemarkers of uropathogenic Escherichia coli. Indian J Med Microbiol26(1):68–70

Vesper SJ, Dearborn DG, Elidemir O, Haugland RA (2000)Quantification of siderophore and hemolysin from Stachybotryschartarum Strains, including a strain isolated from the lung of achild with pulmonary Hemorrhage and Hemosiderosis. ApplEnviron Microbiol 66(6):2678–2681

Wells RM, Jones CM, Xi Z, Speer A, Danilchanka O, Doornbos KS, SunP, Wu F, Tian C, Niederweis M (2013) Discovery of a siderophoreexport system essential for virulence of Mycobacteriumtuberculosis. PLoS Pathog 9(1):e1003120

West SA, Buckling A (2002) Cooperation, virulence andsiderophore production in bacterial parasites. Proc R Soc B270(1510):37–44

Wilikins DH (1960) Calcein Blue—a new metalfluorechromic indicatorfor chelatometric titrations. Talanta 4(3):182–184

Yoshida H, Ozawa T, Jitsukawa K, Einagat H (1993) Kinetics and mech-anism of complex formation reaction of iron(III) with 4-methyl-s-(carboxymethyl)aminomethyl umbelliferrone and related ligands:role of substituent groups on nitrogen donor atoms on the reaction.Polyhedron 12(11):1319–1328

Zawadzkaa AM, Kimb Y, Maltsevab N, Nichiporuka R, Fanb Y,Joachimiakb A, Raymonda KN (2009) Characterization of aBacillus subtilis transporter for petrobactin, an anthrax stealthsiderophore. PNAS 106(51):21854–21859



ranjini ppr

Documents