Alexandrium peruvianum (Balech and Mendiola) Balech and Tangen a new toxic species for coastal North Carolina

Harmful Algae 17 (2012) 54–63

Contents lists available at SciVerse ScienceDirect

Harmful Algae journal homepage: www.elsevier.com/locate/hal

Alexandrium peruvianum (Balech and Mendiola) Balech and Tangen a new toxic species for coastal North Carolina Carmelo R. Tomas a,*, Ryan van Wagoner a,1, Avery O. Tatters a,2, Kevin D. White b,3, Sherwood Hall b, Jeffrey L.C. Wright a a b

Center for Marine Science, University of North Carolina Wilmington, 5600 Marvin K. Moss Lane, Wilmington, NC 28409, USA Center for Food Safety and Applied Nutrition, Food & Drug Administration, 5100 Paint Branch Pkwy, College Park, MD 20740, USA

A R T I C L E I N F O

A B S T R A C T

Article history: Received 21 April 2009 Received in revised form 21 February 2012 Accepted 28 February 2012 Available online 7 March 2012

Routine sampling of the water quality stations in the New River Estuary (Jacksonville, North Carolina, USA) during November 2004 revealed the presence of a previously unidentified dinoflagellate. Preliminary observations of its morphology suggested it to be consistent with that of Alexandrium peruvianum (Balech et Mendiola) Balech et Tangen. Observations using brightfield, epifluorescence and scanning electron microscopy confirmed the diagnostic thecal plates to be those of A. peruvanium. Clonal cultures established from cells isolated from the New River Estuary samples were also used for further studies of morphology and for the presence of toxins. Thecal morphology was consistent with that described by Balech clearly separating it from the sister species Alexandrium ostenfeldii. Three classes of toxins were detected from these cultures. An erythrocyte lysis assay (ELA) was used to confirm the presence of hemolytic toxins in A. peruvianum cultures. A cellular EC50 for lysis was 1.418  104 cells, well within the range the maximal cells densities found in the New River and more potent when compared on a cellular basis with Prymnesium parvum. Another toxin class detected in A. peruvianum cultures was the fast acting 13-desmethy C and D spirolides also produced by the sister species A. ostenfeldii. The last toxin type detected in the A. peruvianum cultures was the paralytic shellfish toxins, GTX 2, 3, B1, STX and C1,2. These findings expand the geographic range of occurrence for A. peruvianum in the U.S. to be much greater than previously considered. The morphological characters agreed with previously reported molecular data in separating A. peruvianum from A. ostenfeldii. It is also the first confirmed report that this species produces PSP toxins, spirolides and naturally occurring hemolytic substances. In light of these findings additional attention is needed for the detection of Alexandrium species in all coastal waters of the U.S. This added effort will enhance the evaluation of the relative impacts of the species to shellfish safety and bloom surveillance. ß 2012 Elsevier B.V. All rights reserved.

Keywords: Alexandrium peruvianum A. ostenfeldii 13-Desmethyl spirolide C, D STX GTX2 3 B1 STX C1,2 Hemolytic activity New River North Carolina

1. Introduction Members of the dinoflagellate genus Alexandrium are well recognized as producers of potent neurotoxins and causative agents of the human syndrome called paralytic shellfish poisoning (PSP). The global distribution of this genus and its impacts on marine systems and human health were recently reviewed (Anderson et al., 2012) where 31 species were listed along with

* Corresponding author. Tel.: +1 910 962 2385; fax: +1 910 962 2410. E-mail address: [email protected] (C.R. Tomas). 1 Present address: Department of Medicinal Chemistry, University of Utah, 201 South Presidents Circle, Room 201, Salt Lake City, UT 84112, USA. 2 Present address: Department of Biological Sciences, University of Southern California, 3616 Trousdale Parkway, AHF 107A, Los Angeles, CA 90089-0371, USA. 3 Retired. 1568-9883/$ – see front matter ß 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.hal.2012.02.011

toxin types. Among them, the two species Alexandrium peruvianum (Balech et Mendiola) Balech et Tangen and Alexandrium ostenfeldii (Paulsen) Balech et Tangen, were noted. Both appeared similar in general morphology with subtle differences in plate structures but varied in toxin type and production. These two species produced the novel class of cyclic imine neurotoxins called spirolides. Spirolides were discovered in the digestive glands of shellfish from the southeastern Nova Scotia (Hu et al., 1995, 2001a), as well as in shellfish and water column populations in Norway (Hu et al., 2001b; Aasen et al., 2005). A. ostenfeldii was confirmed as the biological source of the seven spirolide congeners isolated from shellfish, natural populations and cultures (Cembella et al., 1998, 1999, 2000). Some forms of the spirolides were bioactive (A–D, G; Hu et al., 2001a) while others (E and F; Hu et al., 1996) were inactive metabolites in shellfish (Richard et al., 2001; Gill et al., 2003). A. peruvianum was reported from the

C.R. Tomas et al. / Harmful Algae 17 (2012) 54–63

Mediterranean as vegetative cells and cysts (Figueroa et al., 2008) that were also found to produce spirolides (Franco et al., 2006). The Mediterranean A. peruvianum appeared identical to A. ostenfeldii based on molecular ITS sequences data (Penna et al., 2008) but had subtle differences in sulcal plates. In Malaysian waters, A. peruvianum was reported (Lim et al., 2005) to contain PSP toxins only with no mention of spirolides. Most recently, A. peruvianum isolated from North Carolina was shown to have both a novel gymnodimine congener (12-methylgymnodimine) as well as 13desmethylspirolide C (Van Wagoner et al., 2011). Originally A. peruvianum was first identified from Callao, Peru in 1976 as Gonyaulax peruviana (Balech and Rojas de Mendiola, 1977), changed by Taylor (1979) to Protogonyaulax peruviana (Balech and Mendiola) and subsequently renamed A. peruvianum based on material collected in Oslofjorden, Norway (Balech and Tangen, 1985). Balech (1995) mentions A. peruvianum from Mineola, NY, whose location was recently confirmed to be Hempstead Harbor located on the south shore of Long Island, New York, USA (A. Freudenthal, per. comm.). This is the southernmost extent of A. peruvianum reported to date. Due to the difficulties of observing thecal plates and quantifying variations among them, identification led to some taxonomic confusion. Recent molecular studies of the of the 18S rDNA gene (SSU, ITS spacers and partial LSU) of several New River clones including AP0411 conducted by Schwarz (2011) of this isolate showed it to be identical to other A. peruvianum sequences. The GenBank accession numbers for this clone are JF921179, JF921180 and JF921181. The difficulty of identifying A. peruvianum by morphology alone may have resulted in misidentification or ambiguities in routine plankton samplings. The fact that this species produces several types of toxins (Van Wagoner et al., 2011) prompts greater investigations into its presence in local waters.

55

2. Materials and methods 2.1. Field collections and strain information Routine monthly samplings during 2004 in the New River, North Carolina, USA (Fig. 1) obtained from the North Carolina Department of Environment and Natural Resources (NC DENR) were observed as live and Lugol’s preserved samples using a Nikon Diaphot Inverted Microscope. The toxic strain of A. peruvianum used in these studies was isolated from a surface sample (upper 0.5 m) at the Frenchs Creek Station (34.639 N, 77.34 W) of the New River estuary in Southeastern, North Carolina, USA (Fig. 1) on 5 November 2004. This culture is presently deposited at the Center for Marine Science, Toxic Algal Culture Collection identified as TACC AP0411. Isolation were made using an Olympus CK40 (Olympus America, Center Valley, PA, USA) inverted tissue culture microscope and hand held micropipette. Single cells were first grown in filter sterilized Frenchs Creek water at 20 psu and a temperature of 15 8C. Once growth was evident, additions of L1 medium (Guillard and Hargraves, 1993) modified by the elimination of silica, were gradually introduced until a stable culture could be maintained in full strength modified L1. All growth studies were conducted in an EGC 8 (Chargin Falls, OH, USA) temperature regulated incubator having cool white fluorescent light with a fluence rate of 60 mmol photons m2 s1 and a 14:10 h light:dark period. 2.2. Morphology Morphology of A. peruvianum cells from culture and field samples were studied using brightfield, epifluorescence and scanning electron microscopy (SEM). Live and Lugol’s preserved cells were observed with a Zeiss Axio Imager II (Carl Zeiss,

Fig. 1. New River Estuary, North Carolina with the location of Frenchs Creek where A. peruvianum cells were initially found in surface samples.

56

C.R. Tomas et al. / Harmful Algae 17 (2012) 54–63

Hemolytic assays were conducted using a modified erythrocyte lysis assay (ELA) (Eschbach et al., 2001) and performed as described by Tatters et al. (2010). This method employed human erythrocytes, osmotically adjusted natural whole samples, cell pellets and supernatants obtained using a Hermle Z383 (ICS Kaysville, UT, USA) refrigerated centrifuge. Cells were tested as whole cultures, a centrifuged concentrate and diluted series with ELA buffer. All tests were conducted on 8 replicates in V-bottom 96-well microtiter plates containing human red blood cells (RBCs) that were previously washed 3 times with ELA buffer. Microtitre plate assays had controls as RBCs + ELA buffer, and RBCs + ELA buffer + saponin (20 mg 125 mL1) analyzed along with samples incubated in the dark for 24 h at 4 8C. After incubation, the V-bottom plates were centrifuged at 2500  g for 10 min at 4 8C after which aliquots of 150 mL from each V-shaped well were transferred into clean flat-bottom microtiter plates. Optical density (OD) of each well was measured at 415 nm with a BiotekTM Powerwave X (Winooski, VT, USA) microtiter plate reader and processed with KC JuniorTM software. The OD of each well was corrected by subtracting the corresponding control wells and results expressed as % of the saponin or saponin equivalents (mg mL1) control or normalized per cell numbers. The EC50 values were calculated from the results of the ELA dilution series at different A. peruvianum concentrations using Graphpad Prism1. A similar analysis was conducted on a Prymnesium parvum clone (TACC PP0019) for comparison on a cellular basis of the EC50 of its hemolytic capability. No attempt was made to correct for cell volume differences between these two species.

80% aqueous methanol, absolute methanol and acetone. All samples with solvents underwent an initial 5 min treatment in a sonicating water bath. The first extraction was performed overnight with subsequent extractions performed during 4 h intervals. At each step, the organic extract was recovered by vacuum filtration and concentrated by rotary evaporation. The dried residue was dissolved in 90% aqueous methanol (50 mL), and washed three times with hexane (50 mL). The percentage of organic solvent was reduced to 70% by the addition of water (12 mL), and the solution extracted with 50, 40, and 30 mL portions of dichloromethane. The methanol remaining in the aqueous fraction was removed by rotary evaporation and the resulting residue was extracted with 10, 7, and 3 mL of 1-butanol. Based on the previous isolation of 13-desmethyl spirolide C from this organism (Van Wagoner et al., 2011), the dichloromethane extract was examined for the presence of additional spirolides. The samples in methanol (1 mL) and 5.0 mL of a 1:2 dilution of the stock were injected for analysis. High-resolution mass spectrometry was carried out using an Applied Biosystems Qstar XL mass spectrometer with an IonSpray source. The sample was introduced by direct infusion of a solution containing sodium trifluoroacetate as an internal calibrant (Moini et al., 1998). A chromatography method, based on a previously reported method (MacKinnon et al., 2006) used a flow rate of 600 mL/min and a gradient going from 80% A (50 mM formic acid, 2 mM ammonium formate, 0.02% TFA) and 20% B (95% acetonitrile, 50 mM formic acid, 2 mM ammonium formate, 0.02% TFA) to 60% B over 7.0 min; from 60% B to 100% B over 0.10 min; holding at 100% B for 0.90 min; from 100 to 20% B over 0.50 min; then holding at 20% B for 2.50 min. The column was maintained at 40 8C. The mass spectrometer was configured with a source temperature of 140 8C, a desolvation temperature of 300 8C, a capillary voltage of 3.5 kV, and a cone voltage of 30.0 V. Scans were performed for the molecular weight range m/z 100–1000 in both positive and negative ion modes. Single ion recordings were performed for 0.100 s for each ion in the following ion pairs (each representing [M+H]+ and [M+Na]+ ions) for known spirolides. These were 692.50 and 714.50; 694.50 and 716.50; 706.50 and 728.50; 708.50 and 730.50; 710.50 and 732.50; 712.50 and 734.50. LC/MS/MS experiments were performed on an Applied Biosystems (Foster City, CA, USA) QTrap mass spectrometer interfaced with an Agilent (Santa Clara, CA, USA) 1100 HPLC with binary pump. The chromatographic method was identical to that described in the preceding paragraph except that the column was maintained at ambient temperature. The injection size was 5.0 mL as described above. The source was heated to 450 8C. A capillary voltage of 4500 V, a declustering potential of 5.0 V, an entrance potential of 10.0 V, and a collision cell entrance potential of 17.3 V were used. An enhanced mass spectrum (EMS) scan over the mass range m/z 200– 1000 was collected with a scan rate of 4000 amu/s. Spirolides were detected using multiple reaction monitoring (MRM) scans based on a previously reported method (Ciminiello et al., 2006) detected the following transitions with a collision energy of 30.0 V and dwell times of 100 ms: 692.8 > 674.3, 692.8 > 444.3, 694.8 > 676.3, 695.8 > 444.3, 706.8 > 788.5, 706.8 > 458.3, 708.8 > 790.5, 708.8 > 458.3. To confirm the presence of spirolides and provide further information on the compounds present, enhanced product ion (EPI) scans were performed using a scan rate of 4000 amu/s, a collision energy of 60.0 V, and a mass range of m/z 50–700. Based on a survey scan, EPI scans were configured to measure the products of m/z 692.8 and 694.8.

2.5. Spirolide extraction and detection

2.6. Saxitoxin extraction and detection

For spirolides, A. peruvianum cells (2.78 g wet weight) were extracted at room temperature sequentially with 50 mL each of

A 10.7 g wet weight aliquot of A. peruvianum cell pellet was mixed with 1 M aqueous acetic acid and frozen at 80 8C. For

Thornwood, NY) as whole cells or disassociated thecae using a Calcifluor method (Fritz and Triemer, 1985). For SEM, cultured cells were fixed with a combined 0.2% gluteraldehyde and cacodylate buffer fixative at room temperature and concentrated on a 3.0 mm Poretic filter (Osmonics, Inc.). For some preparations, cells were swollen by pretreating them in 40% salinity water prior to fixing. After fixation the cells were rinsed 3 times with ultra pure DI water and dehydrated with an alcohol/water series of 30%, 50%, 70%, 90%, 100% and again with 100% EtOH with 15 min dehydration intervals at each step. After treatment with absolute alcohol, cells were processed in a Poloron 3100 critical point dryer (Quorum Technologies, East Sussex, UK) prior to being mounted on aluminum SEM stubs and coated with gold palladium using a Cressington 208 HR Sputter Coater (Ted Pella, Inc., CA, USA). All mounts were observed using a Phillips XL 30S FEG (Hillsboro, OR, USA) scanning electron microscope. 2.3. Cells for toxin analysis Cells for extraction and detection of spirolides and PSP toxins in A. peruvianum were grown in an IKA BR photobioreactor (IKA Works Inc., Wilmington, NC, USA) and cells harvested using a Sorvall RC2B centrifuge equipped with a Kendro continuous flow head. Cell pellets were harvested, frozen to 80 8C and stored either as moist pellets or pellets to which 4 mL of glacial acetic acid was added. The latter were required for analysis of saxitoxins. All pellets remained stored at 80 8C until extractions and purification for spirolides and PSP toxins were accomplished. Cultures for hemolytic activity were grown in 3 L fernbach flasks in EGC growth chambers as above and cells were analyzed from late log or early stationary phase cultures. 2.4. Hemolytic toxin assays

C.R. Tomas et al. / Harmful Algae 17 (2012) 54–63

analysis, the frozen pellet was thawed and rinsed with water yielding a 34.56 g suspension (0.310 g packed cells/g suspension). The suspension was frozen, thawed, and mixed. A 1.33 g aliquot having an estimated 0.412 g of packed cells was removed, sonicated briefly and centrifuged. Two 0.45 mL portions of the supernatant were removed. One portion was acidified with 0.3 mL 1 M HCl, was mixed and placed in a boiling bath for 5 min (Hall et al., 1980; Hall, 1982). The mixture was cooled and pH adjusted to 5–5.6 with 1 M ammonium hydroxide and 0.1 mL of 1 M acetic acid to a final volume 1.1 mL. The other portion received 0.65 mL of water to final volume 1.1 mL. Each portion received 1.1 mL acetonitrile to a nominal final volume of 2.2 mL. One milliliter of this was derived from 0.063 g packed cells. Initial toxin separations were performed by liquid chromatography electrospray-ionization multiple reactions monitoring mass spectrometry (LC/ESI/MRM/MS) according to Negri et al. (2003). The LC system used was an Agilent 1100 series vacuum degasser, binary pump, autosampler, and column oven (Agilent, Wilmington, DE, USA). The chromatographic separation was achieved using a TosoHaas TSK-GEL Amide-80 HILIC column; 250 mm  2 mm inner diameter packed with 5 mm particles (Tosoh Bioscience,

57

Montgomeryville, PA, USA). Toxins were eluted with an isocratic mobile phase consisting of 2 mM ammonium formate and 3.6 mM formic acid in 65:35 (v,v) acetonitrile/water pumped at 0.3 mL/ min. Mass spectrometry was performed with an API5000 triple quadrupole mass spectrometer (Applied Biosystems/MDS Sciex, Framingham, MA, USA) equipped with a turbospray ionization source and operated in positive ion mode. The following instrument parameters were used; 550 8C turbo heater temperature; curtain gas (CUR), 35 L N2/h; nebulizer gas (GS1), 40 L N2/h; turbo heater gas (GS2), 50 L N2/h; spray voltage of 5000 V. Multiple reactions monitoring (MRM), in which the precursor ion for each toxin is fragmented and specific fragments characteristic of that compound are monitored, was used for measurement of the toxins. Sixteen collisionally activated decomposition (CAD) fragments were chosen to monitor which provided sufficient evidence to prove and quantify sixteen congeners of paralytic shellfish poison (PSP) if present in matrix extracts. The CAD parameters were set as follows: 110 V declustering potential; 10 V entrance potential; 9 psi nitrogen CAD gas; 11 V collision chamber exit potential. Using certified reference materials obtained from National Research Council (NRC, Halifax, Canada) the MRM conditions

Fig. 2. A. peruvianum (A and B). Whole cells in brightfield, scale = 20 m; (C) calcofluor stained cells observed with epifluorescence, scale = 20 m; (D) whole cell SEM indicating the apical pore complex (APC), fist apical plate (10 ), the ventral pore (VP) and anterior sulcal plate (Sa) scale = 10 m; (E) whole cell SEM showing the 6th precingular (600 ) and anterior sulcal plate (Sa) scale = 10 m; (F) apical pore complex with ridged margin, scale = 5 m; (G) apical pore complex with hook shaped opening with closing plate and pores, scale = 5 m.

58

C.R. Tomas et al. / Harmful Algae 17 (2012) 54–63

were optimized to produce the maximum intensity or signal for the ions monitored. Dwell time for each reaction was 100 ms. 3. Results 3.1. Environmental samples A. peruvianum cells were first observed from a surface sample at Frenchs Creek, New River, NC, USA (Fig. 1) in November 2004 when the temperature was 18 8C and salinity 23 psu. Cell populations during the November through February period varied from a few to 105 cells L1, after which A. peruvianum was rarely seen. This species appeared during the subsequent years (2005–2008; data not presented) and was recognized from archived Lugol’s preserved dating from 2000 to 2004. Single cells isolated using a micropipette into separate wells of a 96-well microtiter plate containing 0.2 mL of sterile filtered water grew to form stable cultures. These cultures were maintained in full strength modified L1 medium in 3 L Fernbach flasks for the duration of the growth study. For toxin analyses, A. peruvianum grown in the photobioreactor routinely reached densities exceeding 90,000 cells mL1. 3.2. Morphology – brightfield microscopy – SEM Microscopic examination of the cultures confirmed these rapidly moving cells to be thecate dinoflagellates with distinct plates. They were reddish brown in appearance with a well defined central nucleus and chloroplasts with pyrenoids (Fig. 2A and B). The general morphology indicated these golden brown cells conformed to a typical Alexandrium ‘‘type’’ dinokont dinoflagellate. The epitheca was tall and domed while the hypotheca was rounded. The plate morphology as seen with Calcofluor stained cells (Fig. 2C) clearly showed an elongated first apical plate with a large ventral pore, 6th precingular plate and an ‘‘A’’ shaped anterior sulcal plate typical of A. pervuianum. When observed in SEM preparations these critical features were more evident. The elongated 10 plate always had a prominent large pore and terminated with a flat margin as it made contact with the Sa plate near the cingulum (Fig. 2D and E). The anterior sulcal plate (Sa) (Fig. 2D and E) which in A. peruvianum is ‘‘A’’ shaped differs from that of A. ostenfeldii which is wider than tall and ‘‘door latch’’ shaped. The apical pore complex (APC) when seen using SEM (Fig. 2F and G) consisting of an oval plate with a comma shaped covered pore fringed by smaller pores which was ridged by abutting plates (Fig. 2F) or could be smooth. The other areas of variation in cells observed by SEM were that the 6th precingular (600 ) which was described wider than tall could vary (Fig. 2D and E) and was not a consistent feature. One notable feature of A. peruvianum was that the base of the first apical plate (10 ) was always flattened as it reached the Sa plate. This differed from what is normally found in A. ostenfeldii.

Fig. 3. Hemolytic analyses for EC50 of (A) A. peruvianum and (B) P. parvum.

3.3. Hemolytic activity Cells of A. peruvianum proved to contain one or more hemolytic factors as determined by the ELA assay. Fractions containing whole cells and media or cell pellets alone consistently gave elevated hemolytic results compared to the cell free supernatant. While the supernatant gave weekly positive results, it was rarely found to give values from which effective concentrations could be accurately measured. When potency of the hemolytic activity was examined, a maximum lysis representing 80% of the saponin lysis control was found resulting in an EC50 of 1.44  104 cells (Fig. 3A). Using this same assay, a highly hemolytic P. parvum culture (TACC PP0019) gave EC50 values of 1.74  106 cells (Fig. 3B).

Fig. 4. Molecular structure and congeners with functional groups R of spirolide A through D and 12-desmethyl C and D. The m/z is expressed as the molecular weight plus hydrogen [M+M]+.

C.R. Tomas et al. / Harmful Algae 17 (2012) 54–63

3.4. Spirolides LC/MS analysis of A. peruvianum organic extracts performed on the protonated molecular ion at m/z 692.4532 was consistent with a molecular formula identical to 13-desmethyl spirolide C or spirolide A (calculated for C42H62NO7, 692/4527. D + 0.8 ppm) confirming the presence of spirolides (Fig. 4). A preliminary survey using single ion recording scans targeted against known spirolides revealed several possible candidates (Fig. 4). In particular, a large peak at m/z 692.5 ([M+H]+) was detected corresponding to the protonated molecular ion 13-desmethyl spirolide C (Hu et al., 2001a,b). Another much smaller peak at m/z 694.5 ([M+H]+) corresponded to 13-desmethyl spirolide D (Hu et al., 1995; Sleno et al., 2004a). A less intense peak at m/z 708.5, was consistent with the protonated molecular ion of spirolide D. Further characterization of the spirolides was obtained from mass spectral fragmentation studies. Multi-reaction monitoring (MRM) scans incorporating fragmentation transitions specific for the various individual spirolides were used to differentiate isomeric species such as spirolide A and 13-desmethyl spirolide C as described by Ciminiello et al. (2006) (Fig. 5A). Product ion

59

scans for m/z 692.5 (Fig. 5B) and 694.5 (Fig. 5C) provided additional information on the locations of the various methyl groups. The presence of fragments at m/z 444, 204, and 164 for both compounds suggested the presence of two methyl groups in the seven-membered imine ring as found in 13-desmethyl spirolides C and D (Hu et al., 2001a,b; Sleno et al., 2004b). Comparing the product ion spectrum (Fig. 6B) to that of 13-desmethyl spirolide C (Sleno et al., 2004b) indicated the two spectra to be nearly identical confirming 13-desmethyl spirolide C to be the most abundant spirolide in A. peruvianum. The lower abundance spirolide m/z 694.5 was assigned as 13-desmethyl spirolide D based on the mass spectral data. These observations were consistent with the previous study on A. ostenfeldii showing 13-desmethyl spirolide C and 13-desmethyl spirolide D to co-occur in the same organism, and 13-desmethyl spirolide C to constitute >90% of the total spirolide content (Sleno et al., 2004a,b). 3.5. Saxitoxins Of the known congeners of saxitoxins, 6 were found in the extracts from A. peruvianum (Figs. 7 and 8). These consisted of

Fig. 5. Single ion LC/MS chromatograms from the dichloromethane-soluble material from an extract of A. peruvianum: (A) m/z 692.5; (B) m/z 694.5; (C) m/z 706.0; (D) m/z 708.5; (E) m/z 710.5; (F) m/z 712.5. The spirolides elute between 1.8 min and 3.0 min by this chromatographic method.

60

C.R. Tomas et al. / Harmful Algae 17 (2012) 54–63

Fig. 6. LC/MS/MS analysis of the dichloromethane partition of A. peruvianum extracts. (A) Summed ion current for the MRM transitions 692.8 > 674.3, 692.8 > 444.3, both representative of 13-dimethyl spirolide C; (B) summed ion current for the RMR transitions 694.8 > 676; (C) the EPI spectrum for m/z 692.8; (D) the IPI spectrum for m/z 694.8. For both B and D, a collision energy of 60.0 eV was used and the most abundant fragment occurred at m/z 164, suggesting methylation of the seven membered imine ring similar to that of spirolide C.

Fig. 7. General structure of saxitoxin and list of congeners with associated four functional groups R.

congeners bearing 21-sulfo and 11-dydroxysulfate groups, both separately as B1 and GTX 2,3 and in combination as C1,2 (Fig. 8). It is noteworthy that there was no evidence of N-1-hydroxy saxitoxins and other forms in our analyses. 4. Discussion From the analyses using the ELA, cells of A. peruvianum proved to contain hemolytic factors. Fractions of whole cell cultures (cells

and media), cell pellet and cell supernatant were assayed. The whole culture and particularly cell pellets had the highest hemolytic activity while the supernatant was slightly positive suggesting that the hemolytic agents were most likely intracellular or membrane bound. When potency of this hemolytic activity was examined, a maximum lysis activity representing 80% of the saponin lysis control for A. peruvianum gave an EC50 of 1.418  104 cells (Fig. 3A). When compared strictly on a cellular density level, the potent hemolytic culture of P. parvum (CMSTAC

C.R. Tomas et al. / Harmful Algae 17 (2012) 54–63

Fig. 8. LC/MS chromatograms of A. peruvianum extracts with defined intensity levels of C1, C2, GTX 2, 3, B1 and STX with elution time.

PP0019) ELA results gave an EC50 = 1.74 106 cells (Fig. 3B). Although this comparison suffers from not being corrected for biomass, it does serve to indicate that A. peruvianum was clearly hemolytic and on a cellular basis in natural samples does represent activities two orders of magnitude greater than P. parvum cells to give an equivalent EC50. Both densities are not uncommon for the New River estuary (Tomas, 2002). 4.1. Investigations to determine the specific nature of these hemolytic is presently ongoing The confirmation of a novel toxic species from North Carolina coastal waters represents a new awareness of HAB’s in the mid Atlantic states. Fish killing events and blooms from NC were described for both the Neuse River and New River regions (Burkholder and Glasgow, 1997; DENR Fish Kill Reports, 2008; Tomas, 2002). While the suspected species attributed to these mortalities centered on the presence of Pfiesteria or Pfiesteria-like organisms (PLOs), fish kills often occurred without the presence of these suspected HAB species. Also mortalities occurred during periods when lesser known suspects were found. For a number of years, the highly hemolytic species Karlodinium veneficum, included as a ‘‘Pfiesteria-like organism’’ (PLO) in species counts, was not taken into account in the fish mortalities. This species produces a series of highly hemolytic toxins ‘‘karlotoxins’’ that caused fish kills in South Carolina and the Chesapeake Bay areas (Kempton et al., 2002; Deeds et al., 2002; Van Wagoner et al., 2008). The confirmed presence of an Alexandrium species producing spirolides adds to the complexity of the coastal HAB community. While A. ostenfeldii was responsible for spirolide production in the cooler waters of Canada (Cembella et al., 1998, 1999, 2000), Norway and Denmark (Aasen et al., 2005) it has rarely been reported south of the Gulf of Maine (Gribble et al., 2005) and to our knowledge, it was never previously observed in temperate Atlantic waters as far south as North Carolina. An annual recurrence of A. peruvianum in the New River is also not surprising given the report of its complex life cycle including two forms of cysts (Figueroa et al., 2008). The obligate cyst dormancy in the sexual resting cyst in excess of 3 months for this species suggests that the cysts form can be an effective means for over wintering periods. As for the potential of A. peruvianum blooms to cause fish distress in the New River, no confirmed mortalities were reported during the known periods when A. peruvianum blooms occurred. A partial reason for this may be that A. peruvianum is difficult to

61

identify and as it was previously unknown from this region, it may have been easily overlooked. It is unlikely that routine monitoring techniques would identify this species even if it were in high abundances. At present, there are no molecular probes for this species as there are for A. ostenfieldii (John et al., 2003). A retrospective examination of archived Lugol’s preserved samples did allow the identification of this species in past samples that remained undetected. The presence each in archived samples spanning 4 years suggests that it has annually been a regular component of the coastal HAB community. In addition, the hemolytic EC50 determination at 104 cells of A. peruvianum was well within the range of abundances found in the archived samples (Tomas, unpublished data). This suggests that based on hemolytic activity alone, A. peruvianum has the potential to cause stress and eventual mortality of fish exposed to them. As for the precise hemolytic toxin composition in this species, it still remains unanswered. Recent evidence from a study involving another Alexandrium species (A. taylori) suggests that this species excretes a protenaceous hemolytic compound with a molecular weight > 10 kDa (Emura et al., 2004). Highly hemolytic polyunsaturated fatty acids were found in the raphidophytes Fibrocapsa japonica (Fu et al., 2004a,b) and Chattonella marina (Marshall et al., 2003). None of these types of compounds were reported from A. peruvianum. Hemolytic toxins were not the only ones detected in the clonal cultures. A. peruvianum from warmer Mediterranean Sea waters contained spirolide toxins as detected by LC/MS with an average composition of 90% 13-desmethyl spirolide C, 6% spirolide B, 2.7% spirolide D, 2% 13-desmethyl spirolide D, and traces of spirolide C (Franco et al., 2006). Our clone of A. peruvianum from the New River was shown previously to contain 13-desmethylspirolide C in addition to 12-methylgymnodimine (Van Wagoner et al., 2011). It was thus not surprising to find 13 desmethylspirolide C and D in the present study. The spirolides, with an unusual sevenmembered spiro-linked cyclic imine moiety, were termed ‘‘fast acting’’ toxins causing convulsions, respiratory distress and subsequent death within minutes upon intraperitoneal injections into laboratory mice (Richard et al., 2001; Gill et al., 2003; Christian et al., 2008). Furthermore, studies of spirolides with vicinal methyl groups on the imine ring survived enzymatic conditions within shellfish (Christian et al., 2008). Histopathological examination of the brains of mice injected with 13-desmethyl spirolide C indicated that the neurological mode of action involved the hippocampus and brain stem as the primary affected organs (Pulido et al., 2001; Gill et al., 2003). The findings of Christian et al. (2008) as well as those of Gill et al. (2003) and Pulido et al. (2001) suggest that accumulation of spirolide toxins in shellfish could pose a potential human health hazard vectored through seafood. For the New River, regions where A. peruvianum blooms were detected are permanently closed to shell fishing due to poor water quality. However, other adjacent regions of this river estuary system further south are commonly used in harvesting oysters and may require further surveillance. Other A. peruvianum clones were noted for the production of paralytic shellfish toxins (PST). Lim and Ogata (2005) reported the presence of PST’s in an A. peruvianum clone from Malaysia. That clone, ApKS01, was reported to produce paralytic shellfish toxins constantly throughout a salinity range of 5–30 (Lim et al., 2005). No spirolides were mentioned for the Malaysian clones. Conversely, the clone of A. peruvianum from the Mediterranean was reported as the first indication of spirolides in that region (Franco et al., 2006). No mention of paralytic shellfish toxins was made in this study. The present study showed A. peruvianum from the New River to produce 6 of the 12 known saxitoxin congeners. It is important that shellfish regulatory agencies be aware of this situation and provide appropriate measures as needed to protect

62

C.R. Tomas et al. / Harmful Algae 17 (2012) 54–63

human health. Further studies on A. peruvianum from North Carolina are ongoing. 5. Conclusion A spirolides producing species A. peruvianum is confirmed for the first time from U.S. coastal mid Atlantic waters. The range for A. peruvianum is now expanded to include coastal North America and its abundance is seasonal but documented from limited observations. The species does contain potent hemolytic compounds, well defined spirolides and paralytic shellfish toxins in its toxin arsenal. Having such a cadre of toxins defines a new toxin paradigm where single toxic HAB species can have multiple toxins that presumably can act individually or synergistically on susceptible organisms. The impact of these toxins on humans is presently unknown, however, evidence of the resistance of the spirolides and PST’s to hydrolysis in shellfish suggests that if introduced into the food chain, they can potentially be harmful to humans ingesting mollusks. Expanded surveillance, particularly in monitoring phytoplankton in coastal waters is needed to give accurate species identification essential in understanding the role this toxic species in our coastal waters. In this regard, a molecular rRNA probe for A. peruvianum, as the one for A. ostenfeldii, would be most helpful. Along with those probes for routine identification, confirmation of this and other harmful species will have to rely on the continued use of traditional microscopic methods involving plate morphology. Acknowledgements We thank Ms. Stephanie Garrett of the North Carolina Department of Environment and Natural Resources, Division of Water Quality, for critical assistance in collecting samples from the New River. Mr. Robert York assisted with the SEM preparations, Ms. Brooke Stuercke gave invaluable assistance with maintaining cultures and Ms. Melissa D. Smith assisted with the preparation of the figures. Dr. Anita Freudenthal kindly provided clarification of sample locations in Long Island, NY. This work was supported by the NC Water Resources Research Institute award # 50337, and Centers for Disease Control through NC DHHS Grant # 01505-07 awarded to C. Tomas and funds from State of North Carolina Marine Biotechnology Program (MARBIONC) (C. Tomas, R. Van Wagoner and J.L.C. Wright). [TS] References Anderson, D.M., Alpermann, T.J., Cembella, A.D., Collos, Y., Masseret, E., Montresor, M., 2012. The globally distributed genus Alexandrium: multifaceted roles in marine ecosystems and impacts on human health. Harmful Algae 14, 10–35. Aasen, J., MacKinnon, S.L., LeBlanc, P., Walter, J.A., Hovgaard, P., Aune, T., Quilliam, M.A., 2005. Detection and identification of spirolides in Norwegian shellfish and plankton. Chemical Research in Toxicology 18, 509–515. Balech, E., 1995. The Genus Alexandrium Halim (Dinoflagellata). Sherkin Island Marine Station, Cork, Ireland, 151 pp. Balech, E., Rojas de Mendiola, B., 1977. Un nuevo Gonyaulax productor de Hemotalasia en Peru´. Neotropica 23, 49–54. Balech, E., Tangen, K., 1985. Morphology and taxonomy of toxic species in the Tamarensis Group Dinophyceae): Alexandrium excavatum (Braarud) comb. nov. and Alexandrium ostenfeldii (Paulsen) comb. nov. Sarsia 70, 333–343. Burkholder, J.M., Glasgow Jr., H.B., 1997. Pfiesteria piscicida and other Pfiesteria-like dinoflagellates. Behavior, impacts and environmental controls. Limnology and Oceanography 42, 1052–1075. Cembella, A., Lewis, N., Quilliam, M., 2000. The marine dinoflagellate Alexandrium ostenfeldii (Dinophyceae) as a causative organism of spirolide shellfish toxins. Phycologia 39, 67–74. Cembella, A., Quilliam, M., Lewis, N., Baude, A., Wright, J.L.C., 1998. Identifying the plankton origin and distribution of spirolides in coastal Nova Scotian waters. In: Reguera, B., Blanco, J., Ferna´ndez, M.L., Wyatt, T. (Eds.), Harmful Algae. Xunta de Galicia and Intergovernmental Oceanographic Commission, UNESCO, Santiago de Compostela, Spain, pp. 481–484.

Cembella, A.D., Lewis, N., Quilliam, M.A., 1999. Spirolide composition of microextracted pooled cells isolated from natural plankton assemblages and from cultures of the dinoflagellate Alexandrium ostenfeldii. Natural Toxins 7, 197–206. Ciminiello, P., Dell’Aversano, C., Fattorusso, E., Magno, S., Tartaglione, L., Cangini, M., Pompei, M., Guerrini, F., Boni, L., Pistocchi, R., 2006. Toxin profile of Alexandrium ostenfeldii (Dinophyceae) from the Northern Adriatic Sea revealed by liquid chromatography–mass spectrometry. Toxicon 47, 597–604. Christian, B., Below, A., Dressler, N., Scheibner, O., Luckas, B., Gerdts, G., 2008. Are spirolides converted in biological systems?—a study. Toxicon 51, 934–940. Deeds, J.R., Terlizzi, D.E., Adolf, J.E., Stoecker, D.K., Place, A.R., 2002. Toxic activity from cultures of Karlodinium micrum (=Gyrodinium galatheanum) (Dinophyceae)—a dinoflagellate associated with fish mortalities in an estuarine aquaculture facility. Harmful Algae 1, 169–189. DENR Fish Kill Reports, 2008 and previous, North Carolina Department of Environment and Natural Resources. Web Site. http://www.esb.enr.state.nc.us/ Fishkill/fishkillmain.htm. Emura, A., Matsuyama, Y., Oda, T., 2004. Evidence for the production of a novel proteinaceous hemolytic exotoxin by dinoflagellate Alexandrium taylori. Harmful Algae 3, 29–37. Eschbach, E., Scharsack, J., John, U., Medlin, L., 2001. Improved erythrocyte lysis assay in microtitre plates for sensitive detection and efficient measurement of haemolytic compounds from ichthyotoxic algae. Journal of Applied Toxicology 21, 513–519. Figueroa, R.I., Bravo, I., Garce´s, E., 2008. The significance of sexual versus asexual cyst formation in the life cycle of the noxious dinoflagellate Alexandrium peruvianum. Harmful Algae 7, 653–663. Franco, J.M., Paz, B., Riobo, P., Pizarro, G., Figueroa, R., Fraga, S., Bravo, I., 2006. First report of the production of spirolides by Alexandrium peruvianum (Dinophyceae) from the Mediterranean Sea. In: 12th International Conference on Harmful Algae, Copenhagen, Denmark, September 4–8, 2006 (Programs and Abstracts), p. 174. Fritz, L., Triemer, R.E., 1985. A rapid simple technique utilizing Calcofluor white M2R for the visualization of dinoflagellate thecal plates. Journal of Phycology 21, 662–664. Fu, M., van Rijssel, M., Liebezeit, G., 2004a. Hemolytic compounds from Fibrocapsa japonica (Raphidophyceae). In: Steidinger, K.A., Landsberg, J.H., Tomas, C.R., Vargo, G.A. (Eds.), Harmful algae 2002 Florida Fish and Wildlife Conservation Commission and Intergovernmental Oceanographic Commission. UNESCO, St. Petersburg, FL, pp. 207–209. Fu, M., Koulman, A., van Rijssel, M., Lu˝ tzen, A., de Boer, M.K., Tyl, M.R., Liebezeit, G., 2004b. Chemical charaterisation of three haemolytic compounds from the microalgal species Fibrocapsa japonica (Raphidophyceae). Toxicon 43, 355–363. Gill, S., Murphy, M., Clausen, J., Richard, D., Quilliam, M.A., MacKinnon, S., LaBlanc, P., Mueller, R., Pulido, O., 2003. Neural injury biomarkers of novel shellfish toxins spirolides: a pilot study using immunochemical and transcriptional analysis. Neurotoxicology 24, 593–604. Gribble, K.E., Keafer, B.A., Quilliam, M.A., Cembella, A.D., Kulis, D.M., Manahan, A., Anderson, D.M., 2005. Distribution and toxicity of Alexandrium ostenfeldii (Dinophyceae) in Gulf of Maine, USA. Deep-Sea Research II 52, 2745–2763. Guillard, R.R.L., Hargraves, P.E., 1993. Stichochrysis immobilis is a diatom, not a chrysophyte. Phycologia 32, 234–236. Hall, S., Reichardt, P.B., Neve, R.A., 1980. Toxins extracted from an Alaskan isolated of Protogonyaulax sp. Biochemical and Biophysical Research Communications 97, 649–653. Hall, S., 1982. Toxins and Toxicity of Protogonyaulax from the Northeast Pacific. University of Alaska, Fairbanks, AK, 196 pp. Hu, T., Curtis, J.M., Oshima, Y., Quilliam, M.A., Walter, J.A., Watson-Wright, W., Wright, J.L.C., 1995. Spirolides B and D two novel macrocycles isolated from the digestive glands of shellfish. Journal of the Chemical Society 20, 2159–2161. Hu, T., Curtis, J.M., Walter, J.A., Watson-Wright, W., 1996. Characterization of biologically inactive spirolides E and F: identification of the spirolide pharmacophore. Tetrahedron Letters 37, 7671–7674. Hu, T., Curtis, J.M., Oshima, Y., Quilliam, M.A., Walter, J.A., Watson-Wright, W., Wright, J.L.C., 2001a. Spirolides B and D, two novel macrocycles isolated from the digestive glands of shellfish. Journal of Natural Products 64, 308–312. Hu, T., Burton, I.W., Cembella, A.D., Curtis, J.M., Quilliam, M.A., Walter, J.A., Wright, J.L.C., 2001b. Characterization of spirolides A, C, and 13-desmethyl C new marine toxins isolated from toxic plankton and contaminated shellfish. Journal of Natural Products 64, 308–312. John, U., Cembella, A.D., Hummert, C., Elebrachter, M., Groben, R., Medlin, L., 2003. Discrimination of the toxigenic dinoflagellates Alexandrium tamarense and A. ostenfeldii in co-occurring natural populations from Scottish coastal waters. European Journal of Phycology 38, 25–40. Kempton, J.W., Lewitus, A.J., Deeds, J.R., Law, J.M., Place, A.R., 2002. Toxicity of Karlodinium micrum (Dinophyceae) associated with a fish kill in a South Carolina brackish retention pond. Harmful Algae 1, 233–241. Lim, P.T., Ogata, T., 2005. Salinity effect on growth and toxin production of four tropical Alexandrium species (Dinophyceae). Toxicon 45, 699–710. Lim, P.T., Usup, G., Leaw, C.P., Ogata, T., 2005. First report of Alexandrium taylori and Alexandrium peruvianum (Dinophyceae) in Malaysia waters. Harmful Algae 4, 391–400. MacKinnon, S.L., Walter, J.A., Quilliam, M.A., Cembella, A.D., LeBlanc, P., Hardstaff, I.W., Burton I W.R., Lewis, N.I., 2006. Spirolides isolated from Danish strains of the toxigenic dinoflagellate Alexandrium ostenfeldii. Journal of Natural Products 69, 983–987.

C.R. Tomas et al. / Harmful Algae 17 (2012) 54–63 Marshall, J.A., Nichols, P.D., Hamilton, B., Lewis, R.J., Hallegraeff, G.M., 2003. Icthyotoxicity of Chattonella marina (Rapahidophyceae) to samsel fish (Acanthochromis polycanthus): the synergistic role of reactive oxygen species and free fatty acids. Harmful Algae 2, 273–281. Moini, M., Jones, B.L., Rogers, R.M., Jiang, L., 1998. Sodium trifluoroacetate as a tune/ calibration compound for positive- and negative-ion electrospray ionization mass spectrometry in the mass range of 100–4000 Da. Journal of the American Society for Mass Spectrometry 9, 977–980. Negri, A.M., Stirling, M., Quilliam, M.A., Blackburn, S., Bolch, C., Burton, I., Englesham, G., Thomas, K., Walter, J., Willis, R., 2003. Three novel hydroxybenzoate saxitoxin analogues isolated from the dinoflagellate Gymnodinium catenatum. Chemical Research in Toxicology 16, 1029–1033. Penna, A., Fraga, S., Maso´, M., Giaccobbe, M.G., Bravo, I., Garce´s, E., Vila, M., Bertozzini, F.A., Luglie´, A., Cristiano, V., 2008. Phylogenetic relationships among the Mediterranean Alexandrium (Dinophyceae) species based on sequences of 5 8 S gene and internal transcript spacers of the rRNA operon. European Journal of Phycology 43, 163–178. Pulido, O., Richard, D., Quilliam, M.A., Clausen, J., Murphy, M., Smyth, P., Mueller, R., Gill, S., 2001. Toxicological neuropathy from domoic acid to spirolides—the Health Canada Experience. In: Hallegraff, G.M., Blackburn, S.I., Bolch, C.J., Lewis, R.J. (Eds.), Harmful Algal Blooms, 2000. Intergovernmental of Oceanographic Commission of UNESCO, pp. 383–386. Richard, D., Arsenault, E., Cembella, A.D., Quilliam, M.A., 2001. Investigations into the toxicology and pharmacology of spirolides, a novel group of shellfish toxins. In: Hallegraef, G.M., Blackburn, S.I., Bolch, C.J., Lewis, R.J. (Eds.), Harmful Algal Blooms, 2000. Intergovernmental of Oceanographic Commission of UNESCO, pp. 383–386.

63

Schwarz, E.N., 2011. Molecular and morphological characterization of Alexandrium species (Dinophyceae) from the East Coast, USA. M.S. Thesis, Department of Biology and Marine Biology, University of North Carolina Wilmington, 65 pp. Sleno, L., Windust, A.J., Volmer, D.A., 2004a. Structural study of spirolide marine toxins by mass spectrometry. Part I. Fragmentation pathways of 13-desmethyl spirolide C by collision-induced dissociation and infrared multiphoton dissociation mass spectrmetry. Analytical and Bioanalytical Chemistry 378, 969–976. Sleno, L., Chalmers, M.J., Volmer, D.A., 2004b. Structural study of spirolide marine toxins by mass spectrometry. Part II. Mass spectrometric characterization of unknown spirolides and related compounds in a cultured phytoplankton extract. Analytical and Bioanalytical Chemistry 378, 977–986. Tatters, A.O., Muhlstein, H.I., Tomas, C.R., 2010. The hemolytic activity of Karenia selliformis and two clones of Karenia brevis throughout a growth cycle. Journal of Applied Phycology 22, 435–442. Taylor, F.J.R., 1979. The toxigenic gonyaulacoid dinoflagellates. In: Taylor, D.L., Seliger, H.H. (Eds.), Toxic Dinoflagellate Blooms. Elsevier, New York, pp. 47–56. Tomas, C.R., 2002. The Presence of Toxic Marine Phytoplankton in the Cape Fear and New River Estuaries. North Carolina Water Research Resources Institute, NCSU, Report, 68 pp. Van Wagoner, R.M., Deeds, J.R., Masayuki, S., Ribero, A.A., Place, A.R., Wright, J.L.C., 2008. Isolation and characterization of karlotoxin 1 a new amphipathic toxin from Karlodinium veneficum. Tetrahedron Letters 40, 6457–6461. Van Wagoner, R.M., Misner, I., Tomas, C.R., Wright, J.L.C., 2011. Occurrence of 12 methylgymnodimine in a spirolide-producing dinoflagellate Alexandrium peruvianum and the biogenic implications. Tetrahedron Letters 52, 4243–4246.