Separation and Detection of Siderophores Produced by Marine

May 1, 2003 - Iron concentrations in seawater are very low,1 and iron limits productivity and biodiversity and affects CO2 sequestration in many areas...
2 downloads 10 Views 85KB Size
Download PDF
Anal. Chem. 2003, 75, 2647-2652

Separation and Detection of Siderophores Produced by Marine Bacterioplankton Using High-Performance Liquid Chromatography with Electrospray Ionization Mass Spectrometry Paul McCormack, Paul J. Worsfold, and Martha Gledhill*

School of Environmental Sciences, Plymouth Environmental Research Center, University of Plymouth, Plymouth, PL4 8AA, U.K.

High-performance liquid chromatography-electrospray ionization mass spectrometry was applied to the detection of the iron(III) complexes of the hydroxamate siderophores rhodotoluric acid, deferrioxamine B, and deferrichrome. Separation of the iron(III) complexes was obtained using a polystyrene-divinylbenzene stationary phase. The retention and responses of ferrioxamine and ferrichrome were optimal when a gradient elution program with methanol and 0.1% (v/v) formic acid as the mobile phases was used. These conditions were also suitable for the retention and separation of the uncomplexed ligands. Retention of iron(III) rhodotoluate was improved when formic acid was replaced by the ion-pairing reagent heptafluorobutyric acid (0.1%). Detection limits for the ferric complexes, defined as 3 SD of the lowest determined standard, were 26 nM for iron(III) rhodotoluate, 0.23 nM for ferrioxamine, and 0.40 nM for ferrichrome. A protocol for the solid-phase extraction of these hydroxamate siderophores from seawater was developed and applied to the extraction of siderophores from enriched incubated seawater samples.

species of marine bacteria have been shown to produce them in culture.8-12 However, it is not yet known how ubiquitous siderophore production might be, whether it is significant in the oceanic environment, or what effect it might have on iron availability to other organisms. Progress in this area of research has been limited by the lack of sensitive analytical techniques capable of detecting and separating siderophores in complex matrixes. Siderophores are commonly detected in bulk samples using spectrophotometric assays, (e.g., the chrome azural S test, the Arnow assay, and the Csaky test13). However, these assays are not applicable to complex matrixes as they are not sufficiently sensitive and can suffer from matrix interferences.13 Microbial assays for the detection of siderophores have been applied to marine samples,14 but these give no indication of the quantity or range of siderophores produced. To fully understand the behavior and significance of siderophores in the marine environment, a sensitive method that has the capability of rapidly separating and detecting multiple siderophores is required. Recently, highperformance liquid chromatography electrospray ionization-mass spectrometry (HPLC-ESI-MS) has been proposed as a rapid method for screening pyoverdin-type siderophores,15 and furthermore, ESI-MS has been shown to provide sensitive detection of

Iron concentrations in seawater are very low,1 and iron limits productivity and biodiversity and affects CO2 sequestration in many areas of the worlds oceans.2-4 Plankton have developed specialized strategies that allow them to grow in these irondepleted open ocean regions.5 For bacterioplankton, it is likely that a high-affinity iron uptake mechanism is used, which utilizes siderophores as external sequestering agents.5,6 Siderophores are metal chelates with a particularly high affinity for iron,7 and several

(4) Martin, J. H.; Coale, K. H.; Johnson, K. S.; Fitzwater, S.; Gordon, R. M.; Tanner, S. J.; Hunter, C. N.; Elrod, V. A.; Nowickl, J. L.; Coley, T. L.; Barber, R. T.; Lindley, S.; Watson, A. J.; van Scoy, K.; Law, C. S.; Liddicoat, M. I.; Ling, R.; Stanton, T.; Stockel, J.; Collins, C.; Anderson, A.; Bidigare, R.; Ondrusek, M.; Latasa, M.; Millero, F. J.; Lee, K.; Yao, W.; Zhang, J. Z.; Freiderich, G.; Sakamoto, C.; Chavez, F.; Buck, K.; Kolber, Z.; Greene, R.; Falkowski, P.; Chisholm, S. W.; Hoge, F.; Swift, R.; Yungel, J.; Turner, S.; Nightingale, P.; Hatton, A.; Liss, P.; Tindale, N. W. Nature 1994, 371, 123129. (5) Butler, A. Science 1998, 281, 207-210. (6) Granger, J.; Price, N. M. Limnol. Oceanogr. 1999, 44, 541-555. (7) Neilands, J. B. J. Biol. Chem. 1995, 270, 26723-26726. (8) Wilhelm, S. W.; Macauley, K.; Trick, C. G. Limnol. Oceanogr. 1997, 43, 992-997. (9) Trick, C. G.; Kerry, A. Curr. Microbiol. 1992, 24, 241-245. (10) Martinez, J. S.; Zhang, G. P.; Holt, P. D.; Jung, H.-T.; Carrano, C. J.; Haygood, M. G.; Butler, A. Science 2000, 287, 1245-1247. (11) Haygood, M. G.; Holt, P. D.; Butler, A. Limnol. Oceanogr. 1993, 38, 10911097. (12) Reid, R. T.; Live, D. H.; Faulkner, D. J.; Butler, A. Nature 1993, 366, 455457. (13) Neilands, J. B.; Nakamura, K. Handbook of Microbial Iron Chelates; Winkelmann, G., Ed.; CRC Press: Boca Raton, FL, 1991; Chapter 1. (14) Soria-Dengg, S.; Reissbrodt, R.; Horstmann, U. Mar. Ecol. Prog. Ser. 2001, 220, 73-82.

* Corresponding author. E-mail: [email protected]. (1) Johnson, K. S.; Gordon, R. M.; Coale, K. H. Mar. Chem. 1997, 57, 137-161. (2) Boyd, P. W.; Watson, A. J.; Law, C. S.; Abraham, E. R.; Trull, T.; Murdoch, R.; Bakker, D. C. E.; Bowie, A. R.; Buesseler, K. O.; Chang, H.; Charette, M.; Croot, P.; Downing, K.; Frew, R.; Gall, M.; Hadfield, M.; Hall, J.; Harvey, M.; Jameson, G.; LaRoche, J.; Liddicoat, M.; Ling, R.; Maldonado, M. T.; Mckay, R. M.; Nodder, S.; Pickmere, S.; Pridmore, R.; Rintoul, S.; Safi, K.; Sutton, P.; Strzepek, R.; Tanneberger, K.; Turner, S.; Waite, A.; Zeldis, J. Nature 2000, 407, 695-702. (3) Coale, K. H.; Johnson, K. S.; Fitzwater, S. E.; Gordon, R. M.; Tanner, S.; Chavez, F. P.; Ferioli, L.; Sakamoto, C.; Rogers, P.; Millero, F. J.; Steinberg, P.; Nightingale, P.; Cooper, D.; Cochlan, W. P.; Landry, M.; Constantinou, J.; Rollwagen, G.; Trasvina, A.; Kudela, R. Nature 1996, 383, 495-501. 10.1021/ac0340105 CCC: $25.00 Published on Web 05/01/2003

© 2003 American Chemical Society

Analytical Chemistry, Vol. 75, No. 11, June 1, 2003 2647

hydroxamate siderophores.16 HPLC-ESI-MS offers several advantages over currently available techniques. It separates analytes from interfering compounds, is particularly sensitive for the determination of low molecular weight polar compounds, provides structural information after collision-induced fragmentation of parent ions, and gives isotopic distributions that can be used to identify specific elements in molecules. This paper reports a novel HPLC-ESI-MS technique for the determination of siderophores in complex environmental matrixes. Three hydroxamate siderophores, deferrioxamine B, deferrichrome, and rhodotoluric acid were used for method development. The iron(III) complexes were used as it is probable that a significant proportion of the siderophores present in marine samples is complexed.17,18 Optimization of ESI-MS for the detection of these complexes has been reported previously,16 so this paper focuses on the chromatographic separation and solid-phase extraction (SPE) of these siderophores from seawater. Separation of these compounds was investigated using four stationary phases: C18, porous graphitic carbon, amino bonded C8, and polystyrene-divinylbenzene. Method optimization focused on the hydrophobic siderophores, as these are more representative of marine siderophores characterized to date.8-12 Capacity factors were obtained, and a protocol for solid-phase extraction of siderophores from aqueous samples was developed and applied to an incubated seawater sample. Excess carbon, nitrogen, and phosphorus were added to the sample in order to maximize siderophore production, while iron availability was limited by the addition of ethylenediamine-N,N′-diacetic acid (EDDA) as chelating agent. Exchange of the iron in the sample with gallium, which resulted in a characteristic shift in the mass/charge ratio of the parent ion and a new parent ion with a distinctive isotopic ratio, was used to detect unknown siderophores. EXPERIMENTAL SECTION Instrumentation. Mass spectrometry analysis was carried out using an ion trap mass spectrometer fitted with an electrospray interface (ThermoQuest Finnigan Mat LCQ, San Jose, CA). Data were acquired and processed with Xcalibur 1.0 software. Instrument tuning and mass calibration were carried out and checked using the automatic calibration procedure and standard calibration solutions. Instrument optimization was carried out by infusing 1 M sodium acetate at 1 µL min-1 into a 200 µL min-1 eluent flow (0.1% trifluoroacetic acid in methanol and water) from the HPLC system via a built-in syringe pump, a 250-µL syringe (Hamilton 1725N, Reno, CA), and a PEEK Tee union (Upchurch Scientific Ltd., Oak Harbor, WA). The automatic tune function was used on a suitable m/z sodium trifluoroacetate adduct ion. For the positive ion full-scan range (m/z ) 100-1500), tuned on adduct ion m/z 431, the following instrument parameters were used: source voltage, +4.5 kV; capillary voltage +20 V; tube lens offset, +10.00 V; capillary temperature, 220 °C; nitrogen sheath gas flow rate, 60 (arbitrary units), and nitrogen auxiliary gas flow rate 20 (arbitrary units). Mass spectra were recorded in the positive ion mode within m/z 300-1500. Peak areas were measured after extracting (0.5 mass unit of the ions or H+ and Na+ adducts from the total ion current. (15) Kilz, S.; Lenz, Ch.; Fuchs, R.; Budzikiewicz, H. J. Mass. Spectrom. 1999, 34, 281-290. (16) Gledhill, M. Analyst 2000, 126, 1359-1362.

2648

Analytical Chemistry, Vol. 75, No. 11, June 1, 2003

High-Performance Liquid Chromatography. HPLC was carried out using a binary pump (P580A, Dionex-Softron GmbH, Germering, Germany). HPLC phases investigated were as follows: C18 (Xterra, 100 × 2.1 mm, 3 µm, Waters), porous graphitic carbon (100 × 2.1 mm, 5 µm, Hypersil, Runcorn, U.K.), aminobonded C8 (Advance, 150 × 2.1 mm, 3 µm, Hypersil), and polystyrene-divinylbenzene (PRP-1, 100 × 2.1 mm, 3 µm and 50 × 4.1 mm, 5 µm, Hamilton, Reno, NA). For the 4.1-mm-diameter column, a 1 mL min-1 flow rate, split ∼200 µL min-1 (highpressure microsplitter valve, Upchurch Scientific Ltd.) to the mass spectrometer was used. A 150 µL min-1 flow rate direct to the mass spectrometer was used with the 2.1-mm-diameter columns. Five-microliter sample injections were suction loaded into a PEEK sample loop via a metal-free manual injector (9125, Rheodyne). Mobile phases were 0.1% aqueous ammonium acetate (pH 4.3), 0.1% aqueous formic acid, 0.1% aqueous heptafluorobutyric acid, 0.1% aqueous ammonium carbonate (pH 8.1), and methanol. The organic phase (methanol) was modified by the addition of 0.1% of the appropriate aqueous-phase buffer to avoid changes in pH during gradient elution. A standard gradient of 95% aqueous phase/5% methanol to 100% methanol in 20 min, followed by 10min isocratic elution with 100% methanol, was used except where otherwise stated. Chemicals. Ultrapure water was obtained from a Milli-Q water purification system (Millipore, Bedford, MA). All solvents and modifiers used were HPLC grade except heptafluorobutyric acid (99%, Aldrich, Milwaukee, WI), glacial acetic acid (Aristar, VWR, Poole, U.K.) and ∼20 M ammonia (Aristar, VWR). Methanol, formic acid, and iron(III) chloride were obtained from VWR. Deferrioxamine B (as deferoxamine mesylate) and deferrichrome were obtained from Sigma (St. Louis, MO) while rhodotoluric acid was obtained from Frontier Scientific (Logan, UT). All other reagents were purchased from Fisher Scientific (Pittsburgh, PA). Stock Solutions. The 1 mg mL-1 aqueous stock solutions of the hydroxamate siderophores, deferrioxamine, deferrichrome, and rhodotoluric acid, were made up and stored at 4 °C. Iron(III) complexes were prepared by a 2 molar equiv addition of a freshly prepared 10 mM iron(III) chloride standard and diluted to 10 × with water. Working standards were prepared by dilution in water of the required volumes. A solution of 1.4 M gallium in 1% nitric acid (plasma emission standard, VWR) was used to identify siderophores in the sample. Solid-Phase Extraction. Siderophores were extracted from aqueous phases using a polystyrene-divinylbenzene SPE cartridge (Isolute ENV+, 100 mg × 3 mL, International Sorbent Technology Ltd., Hengoed, Mid Glamorgan, U.K.). Cartridges were cleaned with 1 mL of methanol and 1 mL of 11.2 mM ammonium carbonate (pH 7.5) prior to extraction of siderophores from 0.3 to 1 L of sample at a flow rate of 3 mL min-1. Cartridges were washed with 1 mL of 11.2 mM ammonium carbonate and siderophores eluted with 5 mL of 1:20:80 (v/v/v) formic acid/ water/methanol. The eluent was blown down to 40) using water, but iron(III)

Figure 3. (a) Total ion chromatogram for a seawater sample after 1400× preconcentration on an ENV+ SPE cartridge. Preconcentration was carried out as described in the text. (b) Extracted mass chromatograms for the three spiked siderophores iron(III) rhodotoluate (FeRA), ferrioxamine (FeFO), and ferrichrome (FeFC). Extracted masses and chromatographic conditions are the same as in Figure 1. NL, maximum ion count.

rhodotoluate was poorly retained (k ) 3). In ammonium carbonate, all three siderophores were well retained (k > 52). Capacity factors decreased from 1.4, >40, and >40 in formic acid to 0, 0, and 0.1 in 25% formic acid/75% methanol for iron(III) rhodotoluate, ferrioxamine, and ferrichrome, respectively. Capacity factors remained low (0) for iron(III) rhodotoluate and ferrioxamine in methanol, but rose to 0.5 for ferrichrome. Based on these results, a protocol using an ammonium carbonate wash and elution with 1:80:100 (v/v/v) formic acid/water/methanol for SPE extraction of siderophores was adopted. Extraction of Siderophores from Salt- and Seawater. The three siderophores were spiked at concentrations of 3.14, 0.16, and 0.77 nM of iron(III) rhodotoluate, ferrioxamine, and ferrichrome, respectively, into synthetic ocean water and seawater from the English Channel. The total ion chromatogram and extracted mass chromatograms for the seawater extractions are shown in Figure 3. Recoveries from synthetic ocean water were 7.9, 46, and 48% for iron(III) rhodotoluate, ferrioxamine, and ferrichrome, respectively. Recoveries for the seawater sample were lower (4.0, 21, and 37% for iron(III) rhodotoluate, ferrioxamine, and ferrichrome, respectively) either because of interferences from other constituents in the seawater adsorbing onto the column or through losses of the siderophores onto particulate material prior to sample filtration. Recoveries of iron(III) rhodotoluate from both samples were very low. Iron(III) rhodotoluate, in common with other dihydroxamates, is hydrophilic and thus not very amenable to

Table 2. Detection Limits (nM), Sensitivities (Ion Current nM-1), and Linear Ranges Obtained Using HPLC-ESI-MS of Iron(III) Rhodotoluate, Ferrioxamine, and Ferrichromea

iron(III) rhodotoluate ferrioxamine ferrichrome a

extracted masses

detection limit (nM)

sensitivity (ion current nM-1)

linear range (nM)

r2

397.6-398.6 613.8-614.8 and 635.8-636.8 740.8-741.8 and 762.8-763.8

26 0.23

9400 160000

26-3140 0.23-160

0.9997 0.9984

0.40

100000

0.40-8000

0.9912

Experimental conditions are as given in the text. Figures were obtained by extracting masses from the total ion chromatogram.

2650 Analytical Chemistry, Vol. 75, No. 11, June 1, 2003

Figure 4. (a) Total ion chromatogram of an incubated seawater sample. Chromatographic conditions are as described in Figure 1. Sample preconcentration was carried out as described in the text. (b) Extracted mass chromatogram (m/z ) 613.9-614.9 and 671.9-672.9) showing two peaks (A and B) identified as siderophores following exchange of iron in the complex with gallium. (c, d) Mass spectra of peaks A and B, respectively, following exchange with gallium, showing the shift in m/z of the parent ion from 614.4 to 627.4 and 629.4 (peak A) and 672.4 to 685.4 and 687.4 (peak B), and the distinctive isotopic ratio due to the two naturally occurring gallium isotopes. NL, maximum ion counts.

extraction using hydrophobic solid phases.23 However, it is possible that addition of HFBA to the sample may increase retention of this compound. Recoveries of deferrioxamine and rhodotoluric acid from UV-irradiated seawater using similar hydrophobic resins (XAD 16, Biobeads SM2) have been reported to be 79-91%;24 however, to our knowledge there are no reports in the literature of the efficiencies of the hydrophobic resins for the extraction of the iron(III) complexes of these ligands. Identification of Siderophores in Complex Matrixes. In complex matrixes, the presence of multiple peaks from other nonsiderophore-type compounds means that siderophores of unknown mass and structure would be extremely difficult to identify. However, we have found that repeat analysis of the sample after addition of excess gallium allows us to identify unknown siderophores through the distinctive isotopic ratio of gallium (Mr ) 69 and 71, ratio 3:2) and the shift in mass/charge ratio of the parent ion. This is shown in the analysis of an incubated seawater sample, where two siderophores (peaks A, m/z 614.4, and B, m/z 672.4) were identified after the addition of 14.3 mM gallium to the sample (Figure 4). Following MS/MS (see Supporting Information Figures S1 and S2) and comparison with previously published data,16 peak A was identified as ferrioxamine B. Ferrioxamines are excreted by species of Actinomycetes,25 a group of bacteria found ubiquitously in seawater and marine sediments.26,27 This is the first time that

ferrioxamine B has been reported from bacteria growing in seawater although ferrioxamines E and G have previously been identified in a seawater extract.28 The iron(III) complex of ferrioxamine G has a molar mass of 671, which, on protonation, would produce a m/z ratio of 672, the same m/z as peak B. In the absence of standards with which to compare fragmentation patterns, however, we are unable to positively confirm the identity of this peak.

(23) Jalal, M. A. F.; van der Helm, D. In CRC Handbook of Microbial Iron Chelates; Winkelmann, G., Ed.; CRC Press Inc.: Boca Raton, FL, 1991; Chapter 8. (24) Macrellis, H. M.; Trick, C. G.; Rue, E. L.; Smith, G.; Bruland, K. Mar. Chem. 2001, 76, 175-187. (25) Matzanke, B. F. In CRC Handbook of Microbial Chelates; Winkelmann, G., Ed.; CRC Press Inc.: Boca Raton, FL, 1991; Chapter 2.

(26) Mincer, T. J.; Jensen, P. R.; Kauffman, C. A.; Fenical, W. Appl. Environ. Microbiol. 2002, 68, 5005-5011. (27) Ghanem, N. B.; Sabry, S. A.; El Sherif, Z. M.; Abu El-Ela, G. A. J. Gen. Appl. Microbiol. 2000, 46, 105-111. (28) Mucha, P.; Rekowski, P.; Kosakowska, A.; Kupryszewski, G. J. Chromatogr., A 1999, 830, 183-189.

CONCLUSIONS SPE-HPLC-ESI-MS is shown to be a powerful technique for the separation and identification of siderophores in complex matrixes. Advantages over previously published approaches include the rapid identification of both known and unknown siderophores and the ability to obtain structural information on individual siderophores. The technique is sensitive, with detection limits on the order of nanomolar. However, despite this sensitivity, preconcentration is necessary for the detection of these compounds at ambient levels in seawater samples. Construction of a library containing ESI-MS/MS fragmentation patterns for known siderophores would greatly facilitate the identification of siderophores in complex matrixes using this technique. The results presented here demonstrate that marine bacterioplankton are capable of producing ferrioxamines, an important class of hydroxamate-type siderophores. This is the first time that ferrioxamine

Analytical Chemistry, Vol. 75, No. 11, June 1, 2003

2651

B has been identified from bacterioplankton growing in seawater. ACKNOWLEDGMENT This work was funded by NERC Grant NER/B/S/2000/00764. We thank the crew of the R.V. Squilla and S. Madgwick for collecting the seawater sample. SUPPORTING INFORMATION AVAILABLE Figure S1 MS/MS of siderophore ferrioxamine B identified

2652

Analytical Chemistry, Vol. 75, No. 11, June 1, 2003

in the extract of an enriched, incubated seawater sample; Figure S2 MS/MS of an unidentified siderophore present in the extract of an enriched, incubated seawater sample. This material is available free of charge via the Internet at http://pubs.acs.org. Received for review January 6, 2003. Accepted March 26, 2003. AC0340105