ranjini ppr
TRANSCRIPT
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
Author's personal copy
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
Author's personal copy
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
2342 Appl Microbiol Biotechnol (2015) 99:2339–2349
Author's personal copy
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
Appl Microbiol Biotechnol (2015) 99:2339–2349 2343
Author's personal copy
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
2344 Appl Microbiol Biotechnol (2015) 99:2339–2349
Author's personal copy
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
Appl Microbiol Biotechnol (2015) 99:2339–2349 2345
Author's personal copy
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
2346 Appl Microbiol Biotechnol (2015) 99:2339–2349
Author's personal copy
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
Appl Microbiol Biotechnol (2015) 99:2339–2349 2347
Author's personal copy
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
2348 Appl Microbiol Biotechnol (2015) 99:2339–2349
Author's personal copy
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
Appl Microbiol Biotechnol (2015) 99:2339–2349 2349
Author's personal copy