Article pubs.acs.org/ac
Chromatographic Determination of Nanomolar Cyanate Concentrations in Estuarine and Sea Waters by Precolumn Fluorescence Derivatization Brittany Widner,* Margaret R. Mulholland, and Kenneth Mopper† Department of Ocean, Earth and Atmospheric Sciences, Old Dominion University, 4600 Elkhorn Ave. Norfolk, Virginia 23529, United States ABSTRACT: Recent studies suggest that cyanate (OCN−) is a potentially important source of reduced nitrogen (N) available to support the growth of aquatic microbes and, thus, may play a role in aquatic N cycling. However, aquatic OCN− distributions have not been previously described because of the lack of a suitable assay for measuring OCN− concentrations in natural waters. Previous methods were designed to quantify OCN− in aqueous samples with much higher reduced N concentrations (micromolar levels) than those likely to be found in natural waters (nanomolar levels). We have developed a method to quantify OCN− in dilute, saline environments. In the method described here, OCN− in aqueous solution reacts with 2-aminobenzoic acid to produce a highly fluorescent derivative, 2,4-quinazolinedione, which is then quantified using high performance liquid chromatography. Derivatization conditions were optimized to simultaneously minimize the reagent blank and maximize 2,4-quinazolinedione formation (>90% reaction yield) in estuarine and seawater matrices. A limit of detection (LOD) of 0.4 nM was achieved with only minor matrix effects. We applied this method to measure OCN− concentrations in estuarine and seawater samples from the Chesapeake Bay and coastal waters from the mid-Atlantic region. OCN− concentrations ranged from 0.9 to 41 nM. We determined that OCN− concentrations were stable in 0.2 μm filtered seawater samples stored at −80 °C for up to nine months. flagellate Prorocentrum donghaiense,9 and some heterotrophic bacteria10−12 have been cultured in media containing cyanate as the sole N source, attesting to the bioavailability of this compound. Potential sources of cyanate to natural waters include industrial wastewater discharges,13 in situ release of cyanate by the microbial community as a byproduct of cellular metabolism,14 cyanate release through cell lysis or “sloppy feeding” by grazers, spontaneous decomposition of the metabolic intermediate carbamoyl phosphate15 and other organic cellular metabolites, herbicide runoff,16 and urea runoff from urban and agricultural settings17 followed by spontaneous aqueous decomposition to cyanate.18 Despite the growing evidence that cyanate is bioavailable to aquatic microorganisms and there are many potential sources of cyanate to aquatic systems, it is unclear whether cyanate contributes to the N demands of microorganisms in nature because there are no measurements of cyanate in aquatic systems. The distributions of nutrient elements provide clues as to their chemical and biological reactivity in natural systems, and quantifying cyanate distributions is a necessary first step toward understanding the cycling of this compound in natural aquatic systems. In many estuarine and seawater samples,
I
n the marine environment, N is often the nutrient that limits primary productivity by phytoplankton. However, because dissolved N is stable in a variety of chemical forms and oxidation states in aquatic environments, the N cycle is complex and involves feedbacks between various dissolved N pools and the microbes, including phytoplankton and bacteria, that mediate their production, consumption, and transformation. Phytoplankton and bacteria take up both organic N and inorganic N compounds,1 and the genetic capability for uptake and assimilation of this diverse N pool has recently been confirmed within individual microbes and microbial communities.2,3 On the basis of recent genomic and physiological evidence, it has been hypothesized that cyanate (OCN−), a reduced N compound, contributes to the N and C requirements of marine microbial communities.4−6 Genes encoding a cyanate-specific transporter as well as an enzyme catalyzing intracellular cyanate decomposition, cyanase, have been identified in strains of the globally important marine cyanobacterial groups, Prochlorococcus and Synechococcus.4 Because these two genera are thought to account for up to two-thirds of oceanic primary production and one-third of global primary production,7 cyanate could be a quantitatively significant component of the marine N cycle that has not yet been examined. Genes related to cyanate metabolism have also been identified in other marine microorganisms.6,8 In addition, Synechococcus WH8102,6 Prochlorococcus MED4,6 the dino© 2013 American Chemical Society
Received: February 1, 2013 Accepted: June 6, 2013 Published: June 6, 2013 6661
dx.doi.org/10.1021/ac400351c | Anal. Chem. 2013, 85, 6661−6666
Analytical Chemistry
Article
Potassium cyanate (KOCN, Sigma-Aldrich, 96%) was stored in a desiccator to slow decomposition of OCN− to NH4+ and CO2.32 Primary KOCN standards were prepared in deionized water (DI) and were stored at 4 °C for up to one month. Working standards were prepared fresh in artificial seawater of the same salinity as the samples. Standard curves were prepared by derivatizing standards and samples simultaneously using the same reagents for each set of samples. Samples were derivatized as follows. A 1 mL, filtered (0.2 μm) sample or standard solution was combined with 0.4 mL of 30 mM 2-aminobenzoic acid in a combusted 4 mL amber borosilicate glass vial (Fisher Scientific) with a polyproplene “top hat” cap (Sigma-Aldrich, PTFE/silicone septum). The vials were placed in a 35 °C water bath for 30 min. Immediately upon removal from the water bath, 1.4 mL of 12N HCl (reagent grade) was added to each sample (6N HCl final concentration). Samples were then run immediately on an HPLC equipped with a refrigerated autosampler (4 °C). HPLC Conditions. A modular Shimadzu HPLC was used to measure cyanate concentrations in derivatized aqueous samples. The pumps were model LC-10ATvp. The degasser was a DGU14A. The autosampler was a SIL-10advp. The fluorescence detector was a RF-10AXL, and the system controller was a SCL-10Avp. Shimadzu CLASS-VP VP1 software was employed for peak enumeration and integration. Mobile phase components were HPLC grade methanol (99.9%; Fisher Scientific), HPLC grade trifluoroacetic acid (TFA) (97%, Fisher Scientific), and nanopure deionized water from a Barnstead system. The mobile phase was 60:40 5% TFA/ 100% MeOH, and the flow rate was 100 μL/min. We used a poly(styrene-divinylbenzene) column with broad pH stability (Hamilton, PRP-1, 2.1 × 150 mm, 5 μm). Cyanate (as 2,4-quinazolinedione) was quantified using a fluorescence detector set at excitation and emission wavelengths of 312 and 370 nm, respectively. The sample injection volume was 100 μL, and the run time was 20 min. The PRP-1 column is stable for at least two years (∼120 L mobile phase); however certain HPLC components, specifically the autosampler, were easily damaged by the concentrated HCl (6 N) in the samples. To minimize instrument damage and disruption of analyses, we recommend use of acid-tolerant components, preinjection neutralization, or frequent autosampler preventative maintenance and consumable replacement. Separation was first achieved using an isocratic mobile phase of 86:14 5% TFA/acetonitrile (ACN) and a flow rate of 110 μL/min. We later determined that use of methanol (MeOH) as the organic solvent was more cost-effective than ACN. Subsequently, the method was altered so that the mobile phase was 60:40 5% TFA/100% MeOH, and the flow rate was 100 μL/min. Although lower column backpressure and shorter retention times were observed using ACN as the organic modifier, both sets of conditions yielded similar precision and accuracy. We recommend use of MeOH; however the results presented here from the Chesapeake Bay Mouth (CBM) were analyzed using ACN. Statistical Calculations and Recovery. Concentrations calculated from both peak area and peak height were very accurate, but at low cyanate concentrations (
