ARTICLE pubs.acs.org/est
Identification of Trichotoxin, a Novel Chlorinated Compound Associated with the Bloom Forming Cyanobacterium, Trichodesmium thiebautii Tracey B. Schock,† Kevin Huncik,‡ Kevin R. Beauchesne,‡ Tracy A. Villareal,§ and Peter D. R. Moeller*,†,‡ †
Department of Marine Biomedicine and Environmental Sciences, Medical University of South Carolina, Charleston, South Carolina 29425, United States ‡ Toxin/Natural Product Chemistry, National Ocean Service, Hollings Marine Laboratory, 331 Fort Johnson Road, Charleston, South Carolina 29412, United States § Marine Science Institute, University of Texas at Austin, 750 Channel View Drive, Port Aransas, Texas 78373, United States ABSTRACT: Trichodesmium is a suspected toxin-producing nonheterocystous cyanobacteria ubiquitous in tropical, subtropical, and temperate seas. The genus is known for its ability to fix nitrogen and form massive blooms. In oligotrophic seas, it can dominate the biomass and be a major component of oceanic primary production and global nitrogen cycling. Numerous reports suggest Trichodesmiumderived toxins are a cause of death of fish, crabs, and bivalves. Laboratory studies have demonstrated neurotoxic effects in T. thiebautii cell extracts and field reports suggest respiratory distress and contact dermatitis of humans at collection sites. However, Trichodesmium toxins have not been identified and characterized. Here, we report the extraction of a lipophilic toxin from field-collected T. thiebautii using a purification method of several chromatographic techniques, nuclear magnetic resonance (NMR), mass spectroscopy (MS), and Fourier transformed-infrared spectroscopy (FT-IR). Trichotoxin has a molecular formula of C20H27ClO and a mass of 318 m/z and possesses cytotoxic activity against GH4C1 rat pituitary and Neuro-2a mouse neuroblastoma cells. A detection method using liquid chromatography/mass spectrometry (LC/MS) was developed. This compound is the first reported cytotoxic natural product isolated and fully characterized from a Trichodesmium species.
’ INTRODUCTION Trichodesmium thiebautii is a globally significant filamentous cyanobacterium distributed throughout pelagic, oligotrophic tropical, and subtropical waters. It is known for its ability to fix nitrogen, without heterocysts, under fully aerobic conditions.1 Trichodesmium is also noted for seasonal blooms that dominate kilometers of open ocean and the phytoplankton population.2 The magnitude and density of these blooms yield a substantial quantity of fixed nitrogen, and thus, Trichodesmium is considered a crucial component of oceanic productivity.3 In addition to the biogeochemical importance of Trichodesmium, the two most common species T. erythraeum and T. thiebautii have been identified as toxic. These species have been linked to several toxic events including the death of fish, oysters, and crabs.4,5 T. thiebautii is toxic to calanoid and cyclopoid copepods and brine shrimp;6 marine mammal deaths off the coast of North Carolina were suggested to be associated with a Trichodesmium spp. bloom;7 and T. thiebautii extracts were neurotoxic in laboratory mice.8 In 1963, the first reported human illness associated with Trichodesmium spp. occurred in Brazil when hundreds of inhabitants of the coastal region were diagnosed with “Tamandare Fever”, exhibiting symptoms of respiratory distress, high fever, muscular pain, and a rash on the thorax and arms.9 The causative agent at the time, however, was not directly linked to a source. Since this episode, no r 2011 American Chemical Society
other harmful human illness events associated with Trichodesmium spp. have been formally reported, although biologists have anecdotally described symptoms of respiratory distress and contact dermatitis in the presence of Trichodesmium spp. Recent work with T. erythraeum along the Brazilian coast has shown antimitotic activity in sea urchin larvae but with no acute toxicity in mouse bioassays.10 Sudek et al. have identified Trichamide, a natural cyclic peptide from Trichodesmium erythraeum,11 with innovative genome mining techniques using the biosynthetic pathway of a previously described cytotoxic cyanobacterial product, patellamide. Trichamide was found to be neither cytotoxic nor neurotoxic, and the biological and ecological functions have yet to be revealed. Consequently, the toxic component(s) of pelagic Trichodesmium spp. remains unknown. In this report, we have isolated and purified a cytotoxic compound, trichotoxin, from environmental T. thiebautii samples. Chromatography, nuclear magnetic resonance (NMR), and mass spectrometry (MS) methods were developed in the purification of the first Received: March 28, 2011 Accepted: July 8, 2011 Revised: June 15, 2011 Published: July 08, 2011 7503
dx.doi.org/10.1021/es201034r | Environ. Sci. Technol. 2011, 45, 7503–7509
Environmental Science & Technology cytotoxically active secondary metabolite, trichotoxin, associated with field collected T. thiebautii samples.
ARTICLE
Table 1. Elution Scheme for Trichodesmium thiebautii Toxin Separation Using a Gemini C18 Column (100 2 mm, 5 μ) (Phenomenex) with a 0.5 mL/min Flow Ratea
’ MATERIALS AND METHODS Algal Collection. Mass culturing of Trichodesmium thiebautii has proven very difficult; therefore, it was necessary to rely on wild sample collections. Field samples of T. thiebautii were netcollected (202 μm mesh meter nets towed at 0 10 m) from the RV Longhorn at 27.3° N 95.8° W to 27.2° N 95.5° W in the western Gulf of Mexico during the summer of 2002. Large samples of T. thiebautii were again collected in July and August of 2005 at approximately 27.85° N latitude and 94.59° W longitude in the western Gulf of Mexico. On this trip, tows were conducted for twenty minutes at depths ranging from 5 to 30 m. The resulting cell mass was concentrated and frozen at 20 °C until processed in the laboratory. T. thiebautii was identified by light microscopy using standard morphological characteristics. Isolation and Purification. Approximately 1500 2000 g wet weight of Trichodesmium thiebautii cell mass was extracted with 6 L of acetone in a Soxhlet extraction apparatus. The acetone was rotary evaporated to reduce the volume, and the remaining aqueous liquid was back-extracted and subsequently partitioned with methylene chloride. Both the aqueous and nonaqueous layers were dried under vacuum, and the residues were tested on a colorimetric cytotoxicity assay (described below). Cytotoxic activity was found in both polar and lipophilic extracts; however, we are only reporting on the purification of the lipophilic components because of the difficulty in purifying aqueous natural products from marine matrices. Due to the large quantity of material necessary for extraction, the initial bulk purification was done by flash chromatography through 500 g of a 60 μm (μ) silica gel (Iatrobeads, Iatron, Tokyo, Japan) packed column eluting compounds with hexane, 95% hexane/5% ethyl acetate (EtoAc), 90% hexane/10% EtoAc, and 100% EtoAc. This was followed by a series of preparative centrifugal chromatographic steps using a Chromatotron (Harrison Research, Palo Alto, CA). The first separation was conducted via a 4 mm silica plate and an eluent scheme of 100 mL of 100% hexane, 150 mL of 95% hexane/5% EtoAc, 200 mL of 90% hexane/10% EtoAc, 100 mL of 100% EtoAc, and 100 mL of 100% methanol. Each 200 drop fraction was dried with nitrogen gas, placed under high vacuum for two hours, and then tested for cytotoxic activity. At the same time we utilized a thin layer chromatography (TLC) method in efforts to visualize the active component throughout the purification process, thus reducing the amount of compound lost during cytotoxicity assays as well as reducing the wait time for assay results. The TLC was carried out on a silica card and developed in 80% benzene/20% acetonitrile (ACN). The cytotoxic compound could be visualized at an Rf of 0.87 by staining with a modified molybdic acid (5% phosphomolybdic acid in absolute ethanol +15% sulfuric acid). The second centrifugal chromatographic purification was conducted on a 2 mm silica plate using the same eluent scheme mentioned above. The active material, an orange colored oil, was further purified with preparatory TLC on Whatman Silica Gel 60A preparative TLC glass-backed plate in 80% benzene/20% acetonitrile. The active compound was scraped from the plate and eluted from the silica with methylene chloride. A 1H and 13C NMR (discussed below) screening of the compound provided evidence that the compound was not fully purified. Mass spectral analysis (discussed below) also revealed the presence of more
time (min)
a
% H2O w/0.1% acetic acid
% MeOH w/0.5% acetic acid
0
80
20
1
40
60
2
40
60
4
20
80
7 8
20 0
80 100
12
0
100
12.1
80
20
18
80
20
Purified trichotoxin eluted at 11 min.
than one compound. These data guided the next HPLC (Agilent 1100 series) purification step, a reverse phase Gemini C18 column (100 2 mm, 5 μ) (Phenomenex), with an injection volume of 20 μL, a flow rate of 0.5 mL/min, and a step gradient over 18 min from 80% water/20% methanol to 100% methanol (Table 1). The oily sample was separated into two major compounds. Both compounds were visible by UV at 210 and 254 nm. At this point, large volumes of active sample needed purification. The separation method above was modified and scaled up to 250 μL injections on a Gemini C18 column (250 10 mm, 5 μ) (Phenomenex) with a flow rate of 4 mL/min using an isocratic method of 80% methanol in water. Trichotoxin was further purified by a Luna C18 column (150 3 mm, 3 μ) (Phenomenex) in 10 μL injections at 0.5 mL/min with an isocratic method of 80% methanol (0.01% acetic acid) in acidified water. Compounds that coeluted with trichotoxin prior to this final purification step were also cytotoxically active, but the small quantities collected made structural determination difficult. Cytotoxicity. To determine and follow biological activity throughout the purification process a bioassay guided fractionation scheme was utilized to identify fractions of interest. For this assay, GH4C1 rat pituitary cells (ATCC, CCL-82.2) and Neuro 2A mouse neuroblastoma cells (ATCC, CCL-131) were used and were plated at 3 104 cells/well (200 μL) in a 96-well plate format. The cells were allowed to adhere at 37 °C in 5% CO2 for a minimum of 4 h. Test fractions (in methanol) from the separations mentioned above and solvent controls were added (4 μL), and the cells were incubated for at least 18 h. Cell viability was assessed through an MTT (3-(4,5-cimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide) colorimetric assay.12 After the 18 h toxin incubation, 15 μL of MTT dye (5 mg/mL) was added. Within 4 h, living cells take up the dye in the mitochondria and metabolize the yellow MTT into a blue formosan. The cells are lysed with SDS (100 μL) (0.01% (v/v) HCl and 10% (w/v) SDS), and the dark blue crystals are solublized yielding a purple color. Wells exhibiting cell death remain characteristically yellow. The plates were read on an automated scanning spectrophotometer (Molecular Devices, Spectramax Plus 384) using a wavelength of 570 nm (maximum absorbance of formosan dye). Dose response curves were plotted as % cell viability versus concentration of toxin in parts per million (ppm). Trichotoxin lethal concentration, 50%, (LC50) was estimated for each cell line using probit analysis. Structural Elucidation. NMR. Structural elucidation of trichotoxin was carried out utilizing a Bruker Avance 2 700 MHz NMR equipped with a 5 mm triple nucleus gradient probe 7504
dx.doi.org/10.1021/es201034r |Environ. Sci. Technol. 2011, 45, 7503–7509
Environmental Science & Technology
ARTICLE
Figure 1. Carbon NMR of trichotoxin. Lock solvent chloroform-d (CdCl3). Inset: aromatic and vinyl carbon chemical shift peaks.
(TopSpin 2.1 software) and a Bruker DMX 500 MHz NMR equipped with a 5 mm triple nucleus gradient probe (XWIN NMR software). Compounds were dissolved in d3-ACN, benzene-d6, methanol-d4, or deuterated methylene chloride as the lock solvent. One dimensional proton (1H), carbon (13C), Attached Proton Test (APT), and Distortionless Enhancement by Polarization Transfer (DEPT) experiments were carried out. The following two-dimensional experiments were also carried out to establish molecular structure: Double Quantum Filtered Correlation Spectroscopy (DQF-COSY), Heteronuclear Single Quantum Correlation (HSQC), Heteronuclear Multiple Bond Correlation (HMBC), and Total Correlation Spectroscopy (TOCSY). FT-IR. Functional groups were confirmed with a Nicolet Magna-IR Spectrometer 550 (Nicolet Thermo) using Omnic ESP 5.2A software. A drop of trichotoxin in methylene choride was placed on a sodium chloride plate. The methylene chloride was allowed to evaporate leaving a thin film of trichotoxin. Spectra were collected in a range of 4000 400 cm 1. Mass Spectrometry. Mass data were collected with a triple quadrupole mass spectrometer (TSQ Quantum Access, Thermo Finnigan) with an atmospheric pressure chemical ionization (APCI) source operating in positive ion mode. A front end Agilent series 1100 HPLC was used to concentrate and separate samples prior to ionization. The instruments were controlled by Xcaliber software. Samples were injected at a volume of 20 μL onto a Gemini C18 column (100 2 mm, 5 μ) at 1 mL/min with a step gradient over 18 min from 80% water/20% methanol to 100% methanol (Table 1). Accurate mass measurement of trichotoxin was obtained by AccuTOF DART direct analysis in real time time-of-flight mass
spectrometer (JEOL). The correct mass assignment was confirmed by the addition of NH4 to form an ammonium adduct for direct comparison. Detection Method. LC/MS. Using the MS method described above, detection limits were defined by producing a standard curve by injecting serially diluted samples from 100 ppm to 10 ppb. Trichotoxin was monitored by masses 301 (M+ OH) and 265 (M+ OH-Cl) m/z with selective- ion monitoring (SIM).
’ RESULTS AND DISCUSSION Due to the lack of established laboratory cultures at the time of this research, understanding the biology, ecology, and biogeochemical processes of this globally important organism has spanned several decades.13,14 These are small scale preparations, however, and large quantities of biomass are typically needed for natural products research. For this reason, our results relied on the collection of T. thiebautii from environmental blooms. The collected T. thiebautii cell mass was extracted with acetone and partitioned between methylene chloride and water. Cytotoxicity bioassays guided the fractionation and both organic and aqueous fractions exhibited activity. The research reported here focused only on the nonpolar fraction due to the difficulty in purifying highly polar compounds from marine samples. The methylene chloride fraction was purified over silica and RP-C18 gels to yield trichotoxin. Structural Elucidation. Trichotoxin, a colorless oil, has a molecular formula C20H27ClO, demonstrating seven degrees of unsaturation. The 13C NMR spectrum contained 18 resonances resulting from two methyl, five methylenes, eight methines, and three quaternary carbons (Figure 1). The splitting pattern and 7505
dx.doi.org/10.1021/es201034r |Environ. Sci. Technol. 2011, 45, 7503–7509
Environmental Science & Technology
ARTICLE
Figure 2. Proton NMR of trichotoxin. Lock solvent chloroform-d (CdCl3). Inset: hydrogens bound to aromatic carbons.
Table 2. NMR Spectral Data for Trichotoxin in CD3CN at 700 MHz 13C δ (m)
1H δ (m)
COSY
HMBC (1H-13C)
1 2
138.7 (q) 128.9 (s)
7.22 (d)
H-3, H-4, H-5
3
128.4 (s)
7.33 (m)
H-2, H-4, H-6
4
126.4 (s)
7.26 (m)
H-2, H-3, H-5, H-6
5
128.4 (s)
7.33 (m)
H-2, H-4, H-6
6
128.9 (s)
7.22 (d)
H-3, H-4, H-5
C-4, C-7
7
40.5 (d)
3.41 (s)
H-9
C-2, C-6, C-9, C-11
8
143.1 (q)
9 10-Cl
113.2 (s)
5.99 (s)
11
27.9 (d)
2.06 (m)
H-12
C-7, C-9, C-12, C-13
12
34.5 (d)
1.25 (m), 1.42 (m)
H-11, H-13
C-11, C-13, C-14, C-16, C-17
13
31.9 (s)
2.34 (m)
H-12, H-14, H-16
C-11, C-12, C-14, C-16
14
20.2 (t)
0.94 (d)
H-13
C-12, C-13, C-16, C-17
15
136.3 (q)
16
131.8 (s)
5.09 (d)
H-13, H-17
C-12, C-13, C-14, C-17, C-18
17 18
10.7 (t) 76.7 (s)
1.56 (s) 3.93 (m)
H-16 H-19, H-20
C-16, C-18 C-16, C-17, C-19, C-20, C-21
2.68 (d)
H-18
20
39.8 (d)
2.23 (m)
H-14, H-18, H-21
21
135.8 (s)
5.75 (m)
H-20, H-22
C-18, C-20, C-22
22
115.8 (d)
4.96 (d), 5.05 (d)
H-21
C-20
19-OH
C-2, C-5, C-7 C-4, C-7 C-2, C-6
C-7, C-9, C-11, C-12 C-7, C-11
C-13, C-17, C-19
integration of vinylic protons [δH 7.33 (2H), 7.26 (1H), and 7.22 (2H)] in the 1H spectrum indicated an aromatic functionality with symmetry (Figure 2), providing evidence of two additional carbon resonances [δC 128.9, C-2, C-6; δC 128.4, C-3, C-5]. The IR spectrum showed absorption bands due to OH (3405.84 cm 1), and an oxygenated carbon [δC 76.7, C-18] was confirmed by 13C
C-18, C-19, C-21, C-22
and APT NMR spectra. The 1H 1H-DQF-COSY and HSQC spectra indicated the presence of the structural fragments CH2CH2CHCH3 (C-11-C-12-C-13-C-14), dCHCH3 (C-16-C-17), and CHO HCH2CHCH2 (C-18-C-20-C-21-C-22) (Table 2). HMBC analysis correlated the methylene protons H-7 [δH 3.41 (2H)] to the quaternary carbon of the aromatic ring (δC 138.7, 7506
dx.doi.org/10.1021/es201034r |Environ. Sci. Technol. 2011, 45, 7503–7509
Environmental Science & Technology
Figure 3. Structure of trichotoxin. Arrows show HMBC correlations. Only relative stereochemistry is reported.
C-1). Further correlation of these protons provided couplings to δC 27.9 (C-11) and δC 113.2 (C-9), an olefinic chlorinated methine. The HMBC correlations of the hydroxyl proton H-19 [δH 2.68 (1H)] to δC 136.6 (C-15), δC 76.7 (C-18), and δC 39.8 (C-20) concluded the remaining correlations. Therefore, trichotoxin was chemically elucidated as (5Z,9E)-9-benzyl-10-chloro5-ethylidene-6-methyldeca-1,9-dien-4-ol (Figure 3). The assigned structure has a molecular formula of C20H27ClO with a mass of 318.88. The accurate mass data, however, proposed a molecular formula of C20H25Cl with a mass of 301.17 exhibiting a characteristic isotopic pattern for a singly chlorinated compound (M+2). The difference between the formulas is a loss of H2O. As molecular ions are often unstable and tend to dissociate into smaller fragments upon ionization, it is typical that molecular ions will not be visible. In the case of alcohols, like trichotoxin, the intensity of the molecular ion peak in the mass spectrum is usually rather low. A common fragmentation of alcohols involves dehydration due to the hot surfaces of the inlet system before the compound comes in contact with the electrons from the ionization source. Cytotoxicity. Previous reports have demonstrated a neurotoxic component in crude T. thiebautii cell extracts administered to mice, 8 but the causative agent was never identified. The objective of the present research was to isolate and purify the compound(s) that may correlate the neurotoxic activity previously observed. For this reason, mammalian neuronal cell lines, rat pituitary cells (GH4C1), and mouse neuroblastoma cells (Neuro-2a) were used to assess a cytotoxic dose response of purified trichotoxin. These cell lines have proven useful in assessing toxicity of other lipophilic marine toxins such as brevetoxin,15,16 as they possess voltage-sensitive sodium (Neuro-2a) and calcium (GH4C1) channel receptor sites. It has been well established that a number of marine toxins produce their effects through perturbations of the voltage-gated sodium channels.15 19 The dose response curves of trichotoxin assays are displayed as % cell viability versus concentration of toxin in parts per million (ppm) (Figure 4). Trichotoxin lethal concentrations at 50% (LC50) were estimated at 33.7 ppm (33.1 34.4 ppm; 95% confidence limit) for Neuro-2a cells and 11.7 ppm (11.4 12.1 ppm) in GH4C1. Morphologically, cells killed by trichotoxin lost membrane rigidity, swelled, and subsequently died (data not shown). Since we have obtained less than 1 mg (mg) total of the purified trichotoxin, we were unable to repeat these experiments in efforts to ensure reproducibility. Comparatively, other phycotoxins appear more potent than trichotoxin. With Neuro-2a cells, saxitoxin can be detected at 20 ppb, ciguatoxin has an inhibitory dose that reduces cell viability to 50% of treated controls (ID50) at 1 picogram (pg), and brevetoxin has an ID50 of 0.25 nanograms (ng).15
ARTICLE
Figure 4. Dose response of trichotoxin on mammalian sodium (Neuro-2a) and calcium (GH4C1) channel sensitive cell lines. Serial dilutions of trichotoxin were incubated per cell line and cell viability was assessed with an MTT colorimetric assay measured at 570 nm, n = 1. Trichotoxin LC50 was estimated by probit analysis for Neuro-2a cells (33.7 ppm (33.1 34.4 ppm; 95% confidence limit)) and for GH4C1 (11.7 ppm (11.4 12.1 ppm)).
Although trichotoxin may not be as toxic as other cyanotoxins, it is important to consider that Trichodesmium forms extensive blooms tropically and subtropically worldwide, and it may be possible for trichotoxin to accumulate in nature. With limited sample, other potential biological activities of trichotoxin have not been characterized. We can speculate that allelopathy is the inherent bioactivity of trichotoxin, in that it is synthesized as a chemical means of defense against predators or competing macrophytes, algae, and microbes.20 In an oligotrophic, pelagic habitat where food sources are limited, one would reason that a massive algal bloom would be a dietary component for a variety of organisms. Interestingly, Trichodesmium blooms generally lack grazers.21 Ninety percent of the oceanic copepod community does not consume T. thiebautii.6 Trichodesmium bloom residing zooplankton remain unaffected within healthy intact colonies, but upon disruption or cell lysis, exposed copepods perish from the ingestion of released intracellular toxins.7 The copepod Macrosetella gracilis utilizes T. thiebautii as a major food source, suggesting that it is insensitive to toxins this cyanobacterium produces, giving it an advantage in the food chain. Exploring M. gracilis’s apparent immunity to trichotoxin may help explain the mechanism of toxicity. It has also been observed that schools of tuna intentionally avoid Trichodesmium blooms,22 so whether the action is because of sheer bloom size, release of toxic metabolites or lack of nutritional value is unknown. Chemical defenses against predation would reduce losses and permit development of the large blooms that are typical of Trichodesmium and enhance its role as a major global fixer of nitrogen. Detection Method. To meet increasing demands for protection of public health, aquaculture and natural resources, considerable effort has been directed toward the development of innovative cyanotoxin detection techniques. Because T. thiebautii has been implicated in causing harmful effects to marine flora and fauna as well as to humans,4 9 there was a necessity to develop a means for environmental monitoring and for assessing health risks, although we are not suggesting that trichotoxin is the causative agent for the deleterious effect to these organisms. Possessing a tool to measure this compound can aid in assessing its biological actions. An HPLC-MS detection method was developed for analysis of trichotoxin in the field, as this is an excellent method to identify individual toxins. Detection limits were determined by employing selective-ion monitoring (SIM) for both m/z 301 (M+ OH) 7507
dx.doi.org/10.1021/es201034r |Environ. Sci. Technol. 2011, 45, 7503–7509
Environmental Science & Technology and 265 (M+ OH Cl) fragments of trichotoxin. The advantage of using SIM over full scan monitoring is an increase in sensitivity. Also by scanning for two diagnostic ions, the selectivity is increased. The standard curve of trichotoxin produced an R2 value of 0.9947, and the detection limit was determined to be between 1 and 10 ppb. Toxin detection and quantitation from marine matrices may complicate toxin measurements and requires further optimization, though we have attempted to alleviate this problem by coupling HPLC with MS detection. For rapid detection in field studies, sample preparation and extraction is the limiting step. A collaborator is currently addressing this issue for trichotoxin. Although T. thiebautii is a pelagic species, toxin detection at very low concentrations is important for risk assessment and public health safety. For instance, different strains of harmful algae have been found to exhibit differences in toxicity from one geographic location to another.18 It is essential that we understand the impact of the ocean and ocean life on human health. The reports of harmful algal blooms have been increasing globally. This may be due, in part, to the development of new monitoring technologies, i.e. the accurate identification of species using molecular biology, the determination of bloom locations via satellite imagery, and faster detection of harmful toxins using chromatographic and mass spectrometric (MS) methods. Techniques such as MS are often used as a monitoring tool for toxin accumulation in seafood and water systems. Additionally, we can detect and monitor the fate and transport of toxins. Changes in the chemistry of the aquatic environment, such as anthropogenic increases in concentration of nitrogen and phosphorus, appear to promote rapid and massive growth of toxic cyanobacteria and encourage formation of dense blooms.23 With increased blooms, the likelihood of exposure to toxins and the risk of fish and mammal die-offs as well as chronic and acute human health problems will also increase. Thus, it is of utmost importance to animal, human, and ecosystem health that there are analytical methods of detection for each harmful algal toxin. The research presented here has resulted in the isolation of lipophilic toxins from T. thiebautii field samples. We have purified and structurally elucidated the first naturally occurring lipophilic toxin from this cyanobacterium. Detection methods for environmental monitoring were developed, and studies into the biological activity of this toxin were initiated. We cannot confirm that trichotoxin is the cause of the reported maladies, but we believe that T. thiebautii possesses a suite of unique biologically active secondary metabolites warranting further examination.
’ AUTHOR INFORMATION Corresponding Author
*Phone: 843 762-8867. E-mail:
[email protected]. Corresponding author address: 331 Ft. Johnson Rd., Charleston, SC 29412.
’ ACKNOWLEDGMENT We thank Dr. Steve Morton of NOAA for assisting in initiating this project; Dr. Dan Bearden of the National Institutes of Standards and Technology, Hollings Marine Laboratory, for his assistance with the 700 MHz NMR; JEOL for accurate mass data; and Kevin Crawford currently of the University of Wisconsin Oshkosh for FT-IR instrument time while at the Citadel. NOAA disclaimer: This publication does not constitute an
ARTICLE
endorsement of any commercial product or intend to be an opinion beyond scientific or other results obtained by the National Oceanic and Atmospheric Administration (NOAA). No reference shall be made to NOAA, or this publication furnished by NOAA, to any advertising or sales promotion which would indicate or imply that NOAA recommends or endorses any proprietary product mentioned herein, or which has as its purpose an interest to cause the advertised product to be used or purchased because of this publication.
’ REFERENCES (1) Saino, T.; Hattori, A. Diel variation of nitrogen fixation by marine blue-green alga, Trichodesmium thiebautii. Deep Sea Res. 1978, 25, 1259–1263. (2) Karl, D.; Michaels, A.; Bergman, B.; Capone, D.; Carpenter, E.; Letelier, R.; Lipschultz, F.; Paerl, H.; Sigman, D.; Stal, L. Dinitrogen fixation in the world’s oceans. Biogeochemistry 2002, 57 58, 47–98. (3) Davis, C. S.; McGillicuddy, D. J. Transatlantic abundance of the N2-fixing colonial cyanobacterium Trichodesmium. Science 2006, 312, 1517–1520. (4) Chidambaram, K.; Unny, M. M. Note on the swarming of planktonic alga, Trichodesmium erythraeum in the Pamban area and its effect on the fauna. Curr. Sci. 1944, 13, 263. (5) Chellam, A.; Alagarswami, K. Blooms of Trichodesmium thiebautii and their effect on experimental pearl culture at Veppalodai. Indian J. Fish. 1978, 25, 237–239. (6) Hawser, S. P.; O’Neil, J. M.; Roman, M. R.; Codd, G. A. Toxicity of blooms of the cyanobacteria Trichodesmium to zooplankton. Appl. Phycol. 1992, 4, 79–86. (7) Guo, C.; Tester, P. A. Toxic effect of the bloom-forming Trichodesmium sp. (cyanophyta) to the copepod Acartia tonsa. Nat. Toxins 1994, 2, 222–227. (8) Hawser, S. P.; Codd, G. A.; Capone, D. G.; Carpenter, E. J. A neurotic factor associated with the bloom-forming cyanobacterium Trichodesmium. Toxicon 1991, 29, 277–278. (9) Sato, S.; Paranagua, M. N.; Eskiniazi, E. On the mechanism of red tide of Trichodesmium in Recife, northeastern Brazil, with some considerations of the relation to the human diseases “Tamandare fever”. Trab. Inst. Oceanogr. Univ. Recife 1963, 6, 7–49. (10) Proenca, L. A. O.; Tamanaha, M. S.; Fonseca, R. S. Screening the toxicity and toxin content of blooms of the cyanobacterium Trichodesmium erythraeum (Ehrenberg) in northeast Brazil. J. Venomous Anim. Toxins Incl. Trop. Dis. 2008, 15 (2), 204–215. (11) Sudek, S.; Haygood, M. G.; Youssef, D. T. A.; Schmidt, E. W. Structure of Trichamide, a cyclic peptide from the bloom-forming cyanobacterium Trichodesmium erythraeum, predicted from the genome sequence. Appl. Environ. Microbiol. 2006, 72 (6), 4382–4387. (12) Mosmann, T. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J. Immunol. Methods 1983, 65, 55–63. (13) Chen, Y. B.; Zehr, J. P.; Mellon, M. Growth and nitrogen fixation of the diazotrophic filamentous nonheterocystous cyanobacterium Trichodesmium sp. IMS 101 in defined media: Evidence for a circadian rhythm. J. Phycol. 1996, 32, 916–923. (14) Bell, P. R. F.; Uwins, P. J. R.; Elmetri, I.; Phillips, J. A.; Fu, F.; Yago, A. J. E. Laboratory culture studies of Trichodesmium isolated from the Great Barrier Reef Lagoon, Australia. Hydrobiologia 2005, 532, 9–21. (15) Manger, R. L.; Leja, L. S.; Lee, S. Y.; Hungerford, J. M.; Hokama, Y.; Dickey, R. W.; Granade, H. R.; Lewis, R.; Yasumoto, T.; Wekell, M. M. Detection of sodium channel toxins: directed cytotoxicity assays of purified ciguatoxins, brevetoxins, saxitoxins, and seafood extracts. J. AOAC Int. 1995, 78 (2), 521–527. (16) Truman, P.; Keyzers, R. A.; Northcote, P. T.; Ambrose, V.; Redshaw, N. A.; Chang, F. H. Lipophilic toxicity from the marine dinoflagellate Karenia brevisulcata: use of the brevetoxin neuroblastoma 7508
dx.doi.org/10.1021/es201034r |Environ. Sci. Technol. 2011, 45, 7503–7509
Environmental Science & Technology
ARTICLE
assay to assess toxin presence and concentration. Toxicon 2005, 46 (4), 441–445. (17) Uchitel, O. D. Toxins affecting calcium channels in neurons. Toxicon 1997, 35 (8), 1161–1191. (18) Pierce, R. H.; Kirkpatrick, G. J. Innovative techniques for harmful algal toxin analysis. Environ. Toxicol. Chem. 2001, 20 (1), 107–114. (19) French, R. J.; Terlau, H. Sodium channel toxins receptors targeting and therapeutic potential. Curr. Med. Chem. 2004, 11, 3053–3064. (20) Berry, J. P.; Gantar, M.; Perez, M. H.; Berry, G.; Noriega, F. G. Cyanobacterial toxins as allelochemicals with potential applications as algaecides, herbicides and insecticides. Mar. Drugs 2008, 6, 117–146. (21) Sheridan, C. C.; Steinberg, D. K.; Kling, G. W. The microbial and metazoan community associated with colonies of Trichodesmium spp.: a quantitative survey. J. Plankton Res. 2002, 24 (9), 913–922. (22) Qasim, S. Z. Some characterisitics of a Trichodesmium boom in the Laccadives. Deep-Sea Res. 1970, 17, 655–660. (23) Patocka, J. The toxins of cyanobacteria. Acta Med. 2001, 44 (2), 69–75.
7509
dx.doi.org/10.1021/es201034r |Environ. Sci. Technol. 2011, 45, 7503–7509
