Zoanthid colony responsible for a severe respiratory reaction, collected from a home aquarium in 2008. [Referred to as VAZOA in Figure 2A and Virginia zoanthid in text]. Sample was found 

Contexts in source publication

Context 1

... 48 agarose gels (Invitrogen Catalog No. G8008-01) according to manufacturer’s protocols with the E-Base H Integrated power supply (Invitrogen Catalog No. EB-M03). Gels were run for 15 min and then visualized using a Bio-Rad gel documentation system (Gel Doc 2000, Hercules, CA). The gel was also photographed with this instrument and the picture retained for records. Successfully amplified products were purified with Exosap-IT (USB, Cleveland, OH) by adding 2 m l of Exosap-IT to 5 m l of PCR product, incubating at 37 u C for 15 minutes followed by 15 minutes at 80 C. At least two sequencing reactions (in the forward and reverse direction) were run on each sample at the Smithsonian Institution’s Laboratories of Analytical Biology (Suitland, MD, USA). Sequencing reactions consisted of 0.5 m l of BigDye H Terminator v3.1 (Applied Biosystems), 1.75 m l of 5X sequencing buffer (Applied Biosystems), 0.5 m l of primer (either M13F-29 or M13R), 6.25 m l of molecular grade water for a total of 9 m l to which 1 m l of cleaned up PCR product was added. Sequencing reactions were conducted on an Eppendorf Mastercycler H ep gradient S Thermocycler with the following conditions with the following conditions: 95 u C for 2 min, 30 cycles of 95 u C for 30 sec, 50 u C for 15 sec, and 60 u C for 4 min, followed by a 10 u C hold. Reactions were cleaned using a sephedex column, and then dried down. Finally 10 m l of Hi-Di formamide (Applied Biosystems # 4311320) was added to each sample and heated at 95 u C for 2 min. Samples were then sequenced on an ABI 3730xL sequencer (Applied Biosystems, Foster City, CA), which was maintained according to manufacturer’s specifications. Resulting sequences were edited in Sequencher 4.9 (Gene Codes Corp., Ann Arbor, MI). Phylogenetic Analysis. Nucleotide sequences of 16S and COI from samples were manually aligned with previously published 16S and COI sequences from various zoanthid species representing the genera Palythoa , Zoanthus and Isaurus . These published sequences were the top matches to the unknown samples using the BLAST algorithm [16] to search the National Center for Biotechnology Information (NCBI) database located at Out-group sequences for both 16S and COI trees were from the genera Hydrozoanthus (within suborder Macrocnemina). All alignments were inspected by eye and manually edited. All ambiguous sites of the alignments were removed from the dataset for phylogenetic analyses. Consequently, two alignment datasets were generated: 1. 817 sites of 30 sequences (16S) and 2. 461 sites of 26 sequences (COI). The alignment data are available on request from the corresponding author, and also at the webpage For the phylogenetic analyses of the two alignments, the same methods were applied independently. Alignments were subjected to analyses with maximum-likelihood method (ML) with PhyML [17] and neighbor-joining (NJ) method with CLC Free Work- bench 3 [18]. PhyML was performed using an input tree generated by BIONJ with the general time-reversible (GTR) model [19] of nucleotide substitution incorporating invariable sites and a discrete gamma distribution (eight categories) (GTR + I + C ). The proportion of invariable sites, a discrete gamma distribution, and base frequencies of the model were estimated from the dataset. PhyML bootstrap trees (1000 replicates) were constructed using the same parameters as the individual ML tree. The distances were calculated using a Kimura’s 2-parameter model [20]. Support for NJ branches was tested by bootstrap analysis [21] of 1000 replicates. Bayesian trees were reconstructed by using the program MrBayes 3.1.2 under the GTR model [22]. One cold and three heated Markov chain Monte Carlo (MCMC) chains with default- chain temperatures were run for 20 million generations, sampling log-likelihoods (InLs), and trees at 100-generation intervals (20,000 InLs and trees were saved during MCMC). The first 15% of all runs were discarded as ‘‘burn-in’’ for all datasets. The likelihood plots for both datasets also showed that MCMC reached the stationary phase by this time (PSRF = 1.000 for both data sets). Thus, the remaining 170,000 trees (17 million generations) of 16S and COI were used to obtain posterior probabilities and branch- length estimates, respectively. After our initial finding of a highly toxic Palythoa sp. in a home aquarium in Virginia [sample ‘‘Virginia zoanthid’’, described in Deeds and Schwartz [8], (Fig. 1)], we visited a local aquarium store known to carry zoanthids (aquarium store # 1) and purchased every visually distinct variety of zoanthid they had for sale ( n = 7). These included specimens visually consistent with both Palythoa/Protopalythoa spp. and Zoanthus spp. (according to [23]). Of these, only one proved to be as toxic as the sample from the Virginia home aquarium (sample 305.11.2) (Table 3). Among the specimens purchased from aquarium store # 1, only sample 305.11.2 was morphologically similar to the Virginia zoanthid sample (Fig. 1, Fig. 2C). With this information, two additional aquarium stores were visited (aquarium store # 2 and # 3), targeting specimens that were visually consistent with our two toxic Palythoa sp. samples. These stores were identified as the most likely in the area to carry zoanthids based on discussions with local marine aquarium hobbyists. Between the two stores, eight additional specimens were acquired. Of these, three were found to be highly toxic (Table 3). The first sample consisted of a few individual polyps growing on a small fragment of coral at the bottom of a large zoanthid display tank (sample 306.37.3) (Fig. 2D) in aquarium store # 2. The small specimen was located at the bottom of the tank, under a display rack, but was retrieved by the store owner and sold for a minimal price. The last two samples, acquired from aquarium store # 3, were either growing as disperse polyps among a colony of another zoanthid species (sample 306.39.2), or as a few random polyps growing on the side of a small piece of plastic pipe sitting on the bottom of another display tank (sample 306.39.3) (Fig. 2E,F respectively). Using high performance liquid chromatography (HPLC) with UV detection, with quantification against a PLTX standard, we found that all four of the purchased zoanthids that were visually consistent with the Virginia home aquarium specimen were themselves highly toxic (approx. 500-3500 m g crude toxin/g wet zoanthid) (Fig. 3, Table 3). In our previous study, we found the Virginia home aquarium sample to contain approx. 600 m g crude toxin/g wet zoanthid [8]. For comparison, the original Hawaiian P. toxica collections in the 1960’s yielded approximately 300 m g pure toxin/g wet zoanthid [2]. Using electrospray ionization liquid chromatography mass spectrometry (ESI/LC/MS) with a high resolution mass spectrometer, for confirmation of PLTX type, four of the samples were shown to contain only palytoxin (Virginia zoanthid, 305.11.2, 306.39.2, 306.39.3) while the last (306.37.3) contained primarily deoxy-palytoxin with a lesser amount of palytoxin (Table 2). Minor PLTXs could not be quantified due to a lack of chromatographic resolution. None were found to contain 42-hydroxy-palytoxin, the primary toxin recently reported from samples collected in the 1990s from the original Hana tidepools from, presumably, P. toxica [24]. As expected, our PLTX standard, isolated from P. tuberculosa , contained primarily palytoxin with a minor amount of 42-hydroxy-palytoxin [22] (Table 2). Several deoxy-palytoxins have been previously described: one as a minor toxin from P. tuberculosa [25], and more recently, several from cultures of the dinoflagellate Ostreopsis cf. ovata and Ostreopsis ovata (Rossi et al. [26], therein called ovatoxin-b, and Ciminiello et al. [27], therein called ovatoxins d and e, respectively). This is the first report of a deoxy-palytoxin being the primary toxin from any zoanthid. 42-hydroxy-palytoxin has been shown to be similar in potency with palytoxin [24]. None of the described deoxy- palytoxins have been assessed for potency. In this study, the position of deoxygenation was not determined, therefore the exact relationship of the deoxy-palytoxin described here with those previously described could not be established. All of the other samples collected ( n = 11), both additional Palythoa spp. and other specimens visually and genetically (see below) consistent with Zoanthus spp., were either non-toxic or weakly-toxic (i.e. shown to have an extractable hemolytic component with properties consistent with PLTX but too low in concentration to confirm as PLTX by other chemical means – detailed methods can be found in [8]). Mitochondrial 16S and COI markers were successfully generated for 10 of the 16 specimens from this study (Table 3). For the remaining six specimens either 16S or COI sequences were generated. All failed sequences were attempted at least twice. Phylogenetic trees generated using both markers showed the same result that all of our toxic samples ( n = 5) formed a closely related clade that was distinct from the additional non- weakly-toxic Palythoa spp. ( n = 3) and the non- weakly-toxic Zoanthus spp. ( n = 8) (Fig. 2A,B). The most closely related species to our toxic specimens based on available molecular data is Palythoa heliodiscus . It should be noted that relatively few of the described species of zoanthids have been analyzed genetically. To our knowledge, no molecular data ...

Context 2

... 0.062 m l of 10 mM dNTPs (New England Biolabs Catalog No. N0447L), and 0.060 m l of Platinum Taq (5 U/ m l, Invitrogen Catalog No. 10966- 034) and 1 m l of undiluted DNA template per reaction, according to Ivanova et al. [15]. An Eppendorf Mastercycler H ep gradient S Thermocycler was used for the PCRs with the following conditions: 95 u C for 1 min, 35 cycles of 95 u C for 1 min, 50 u C for 40 sec (for 16S, 40 u C for COI), and 72 u C for 1.5 min, with a final extension at 72 u C for 7 min. All products were verified using pre-cast 1% E-gel 48 agarose gels (Invitrogen Catalog No. G8008-01) according to manufacturer’s protocols with the E-Base H Integrated power supply (Invitrogen Catalog No. EB-M03). Gels were run for 15 min and then visualized using a Bio-Rad gel documentation system (Gel Doc 2000, Hercules, CA). The gel was also photographed with this instrument and the picture retained for records. Successfully amplified products were purified with Exosap-IT (USB, Cleveland, OH) by adding 2 m l of Exosap-IT to 5 m l of PCR product, incubating at 37 u C for 15 minutes followed by 15 minutes at 80 C. At least two sequencing reactions (in the forward and reverse direction) were run on each sample at the Smithsonian Institution’s Laboratories of Analytical Biology (Suitland, MD, USA). Sequencing reactions consisted of 0.5 m l of BigDye H Terminator v3.1 (Applied Biosystems), 1.75 m l of 5X sequencing buffer (Applied Biosystems), 0.5 m l of primer (either M13F-29 or M13R), 6.25 m l of molecular grade water for a total of 9 m l to which 1 m l of cleaned up PCR product was added. Sequencing reactions were conducted on an Eppendorf Mastercycler H ep gradient S Thermocycler with the following conditions with the following conditions: 95 u C for 2 min, 30 cycles of 95 u C for 30 sec, 50 u C for 15 sec, and 60 u C for 4 min, followed by a 10 u C hold. Reactions were cleaned using a sephedex column, and then dried down. Finally 10 m l of Hi-Di formamide (Applied Biosystems # 4311320) was added to each sample and heated at 95 u C for 2 min. Samples were then sequenced on an ABI 3730xL sequencer (Applied Biosystems, Foster City, CA), which was maintained according to manufacturer’s specifications. Resulting sequences were edited in Sequencher 4.9 (Gene Codes Corp., Ann Arbor, MI). Phylogenetic Analysis. Nucleotide sequences of 16S and COI from samples were manually aligned with previously published 16S and COI sequences from various zoanthid species representing the genera Palythoa , Zoanthus and Isaurus . These published sequences were the top matches to the unknown samples using the BLAST algorithm [16] to search the National Center for Biotechnology Information (NCBI) database located at Out-group sequences for both 16S and COI trees were from the genera Hydrozoanthus (within suborder Macrocnemina). All alignments were inspected by eye and manually edited. All ambiguous sites of the alignments were removed from the dataset for phylogenetic analyses. Consequently, two alignment datasets were generated: 1. 817 sites of 30 sequences (16S) and 2. 461 sites of 26 sequences (COI). The alignment data are available on request from the corresponding author, and also at the webpage For the phylogenetic analyses of the two alignments, the same methods were applied independently. Alignments were subjected to analyses with maximum-likelihood method (ML) with PhyML [17] and neighbor-joining (NJ) method with CLC Free Work- bench 3 [18]. PhyML was performed using an input tree generated by BIONJ with the general time-reversible (GTR) model [19] of nucleotide substitution incorporating invariable sites and a discrete gamma distribution (eight categories) (GTR + I + C ). The proportion of invariable sites, a discrete gamma distribution, and base frequencies of the model were estimated from the dataset. PhyML bootstrap trees (1000 replicates) were constructed using the same parameters as the individual ML tree. The distances were calculated using a Kimura’s 2-parameter model [20]. Support for NJ branches was tested by bootstrap analysis [21] of 1000 replicates. Bayesian trees were reconstructed by using the program MrBayes 3.1.2 under the GTR model [22]. One cold and three heated Markov chain Monte Carlo (MCMC) chains with default- chain temperatures were run for 20 million generations, sampling log-likelihoods (InLs), and trees at 100-generation intervals (20,000 InLs and trees were saved during MCMC). The first 15% of all runs were discarded as ‘‘burn-in’’ for all datasets. The likelihood plots for both datasets also showed that MCMC reached the stationary phase by this time (PSRF = 1.000 for both data sets). Thus, the remaining 170,000 trees (17 million generations) of 16S and COI were used to obtain posterior probabilities and branch- length estimates, respectively. After our initial finding of a highly toxic Palythoa sp. in a home aquarium in Virginia [sample ‘‘Virginia zoanthid’’, described in Deeds and Schwartz [8], (Fig. 1)], we visited a local aquarium store known to carry zoanthids (aquarium store # 1) and purchased every visually distinct variety of zoanthid they had for sale ( n = 7). These included specimens visually consistent with both Palythoa/Protopalythoa spp. and Zoanthus spp. (according to [23]). Of these, only one proved to be as toxic as the sample from the Virginia home aquarium (sample 305.11.2) (Table 3). Among the specimens purchased from aquarium store # 1, only sample 305.11.2 was morphologically similar to the Virginia zoanthid sample (Fig. 1, Fig. 2C). With this information, two additional aquarium stores were visited (aquarium store # 2 and # 3), targeting specimens that were visually consistent with our two toxic Palythoa sp. samples. These stores were identified as the most likely in the area to carry zoanthids based on discussions with local marine aquarium hobbyists. Between the two stores, eight additional specimens were acquired. Of these, three were found to be highly toxic (Table 3). The first sample consisted of a few individual polyps growing on a small fragment of coral at the bottom of a large zoanthid display tank (sample 306.37.3) (Fig. 2D) in aquarium store # 2. The small specimen was located at the bottom of the tank, under a display rack, but was retrieved by the store owner and sold for a minimal price. The last two samples, acquired from aquarium store # 3, were either growing as disperse polyps among a colony of another zoanthid species (sample 306.39.2), or as a few random polyps growing on the side of a small piece of plastic pipe sitting on the bottom of another display tank (sample 306.39.3) (Fig. 2E,F respectively). Using high performance liquid chromatography (HPLC) with UV detection, with quantification against a PLTX standard, we found that all four of the purchased zoanthids that were visually consistent with the Virginia home aquarium specimen were themselves highly toxic (approx. 500-3500 m g crude toxin/g wet zoanthid) (Fig. 3, Table 3). In our previous study, we found the Virginia home aquarium sample to contain approx. 600 m g crude toxin/g wet zoanthid [8]. For comparison, the original Hawaiian P. toxica collections in the 1960’s yielded approximately 300 m g pure toxin/g wet zoanthid [2]. Using electrospray ionization liquid chromatography mass spectrometry (ESI/LC/MS) with a high resolution mass spectrometer, for confirmation of PLTX type, four of the samples were shown to contain only palytoxin (Virginia zoanthid, 305.11.2, 306.39.2, 306.39.3) while the last (306.37.3) contained primarily deoxy-palytoxin with a lesser amount of palytoxin (Table 2). Minor PLTXs could not be quantified due to a lack of chromatographic resolution. None were found to contain 42-hydroxy-palytoxin, the primary toxin recently reported from samples collected in the 1990s from the original Hana tidepools from, presumably, P. toxica [24]. As expected, our PLTX standard, isolated from P. tuberculosa , contained primarily palytoxin with a minor amount of 42-hydroxy-palytoxin [22] (Table 2). Several deoxy-palytoxins have been previously described: one as a minor toxin from P. tuberculosa [25], and more recently, several from cultures of the dinoflagellate Ostreopsis cf. ovata and Ostreopsis ovata (Rossi et al. [26], therein called ovatoxin-b, and Ciminiello et al. [27], therein called ovatoxins d and e, respectively). This is the first report of a deoxy-palytoxin being the primary toxin from any zoanthid. 42-hydroxy-palytoxin has been shown to be similar in potency with palytoxin [24]. None of the described deoxy- palytoxins have been assessed for potency. In this study, the position of deoxygenation was not determined, therefore the exact relationship of the deoxy-palytoxin described here with those previously described could not be established. All of the other samples collected ( n = 11), both additional Palythoa spp. and other specimens visually and genetically (see below) consistent with Zoanthus spp., were either non-toxic or weakly-toxic (i.e. shown to have an extractable hemolytic component with properties consistent with PLTX but too low in concentration to confirm as PLTX by other chemical means – detailed methods can be found in [8]). Mitochondrial 16S and COI markers were successfully generated for 10 of the 16 specimens from this study (Table 3). For the remaining six specimens either 16S or COI sequences were generated. All failed sequences were attempted at least twice. Phylogenetic trees generated using both markers showed the same result that all of our toxic samples ( n = 5) formed a closely related clade that was distinct from the additional non- weakly-toxic Palythoa spp. ( n = 3) and the non- weakly-toxic Zoanthus spp. ( n = 8) (Fig. 2A,B). The most closely related species to our toxic specimens based on available molecular data is Palythoa heliodiscus . It should be noted that relatively few of the described species of zoanthids have been analyzed genetically. ...

Context 3

... the late 1950’s, Dr. Albert H. (Hank) Banner began a program at the University of Hawaii to search for the elusive cause of ciguatera fish poisoning [1]. Dr. Philip Helfrich, a young researcher hired by Banner, began this search by investigating an entry in the Hawaiian dictionary for the ‘‘ limu-make-o-Hana ’’ (rough translation deadly seaweed of Hana) [2]. This legend dates back to Hawaiian antiquity with tales of Shark Gods, sacred pools, and a seaweed when applied to a warrior’s spear would ‘‘bring sure death’’ to their enemies. The pool became ‘‘kapu’’ or taboo to the local Hawaiians and it was said that an ill fate would befall anyone who disturbed the sacred site. Every legend holds some basis in fact, and in 1961, Helfrich, accompanied by graduate student John Shupe, tracked down the fabled pool near the village of Mu 9 olea on the island of Maui and introduced to the world a new species of cnidarian zoanthid (colonial anemone) known as Palythoa toxica [3,4]. This research led to the discovery of palytoxin (PLTX), one of the most potent natural products ever discovered [5]. Although much of the structural elucidation of palytoxin would be determined from less toxic, but far more abundant, species such as Palythoa tuberculosa [6,7] none were ever found to be as potent as the samples collected from the tidepools at Mu 9 olea. Now the legendary limu appears to be exacting its ancient curse once again, but this time upon unsuspecting marine home aquarists. In 2007, we assisted the Georgia poison center in an investigation into a potential dermal intoxication with palytoxin from zoanthids in a home aquarium [8]. During the course of the investigation, we learned of another marine aquarium hobbyist in Virginia who had recently experienced a severe respiratory reaction while trying to eradicate brown non-descript colonies of zoanthids that had arrived as hitchhikers with live rock and were overgrowing more desirable organisms in the tank (Fig. 1) [account can be found at: http:// www.reefcentral.com/forums/showthread.php?t = 1083843 – ac- cessed 12/09/10]. A sample (here called Virginia zoanthid or VAZOA) was found to contain high levels of palytoxin (600 m g crude toxin/g wet zoanthid). Zoanthids are a notoriously problematic taxonomic group [9–11] and it proved difficult at the time to identify to the species level the sample from the home aquarium in Virginia. Histologic evaluation of preserved polyps could only confirm that it was a Palythoa sp. (D. Fautin, University of Kansas, personal communication). In an attempt to determine the prevalence of palytoxin in aquarium store zoanthids, we purchased several colonies from local aquarium stores and analyzed them for palytoxin. We further performed a molecular analysis in an attempt to identify the colonies to the species level. Fifteen zoanthid colonies visually consistent with Palythoa/ Protopalythoa spp. or Zoanthus spp. were purchased live from 3 aquarium stores in the Washington DC metro area. Samples, with associated seawater, were placed in individual glass beakers under fluorescent light for 24 hours in an attempt to induce the opening of individual polyps for photographic documentation. After photographic documentation, several polyps from each colony were removed from their supporting substrate using forceps and fine dissecting scissors and placed in 10 ml of 80% ethanol for 24 hours at 4 u C to extract PLTXs. Some of the larger polyps were split in half to facilitate extraction but no additional homogeni- zation of tissues was performed. After extraction, polyps were removed, blotted dry, weighed, and placed in fresh 95% ethanol for DNA analysis. The total amount of tissue extracted varied depending on the size of the colony purchased and the size of the individual polyps, but the range for all samples extracted for toxin analysis in this study was 0.2–0.8 g/specimen. Palytoxin Quantification. High performance liquid chromatography (HPLC) analysis was performed using an Agilent 1200 series HPLC system (Agilent, Wilmington, DE) with UV detection following Ciminiello et al. [12]. Samples were diluted 1:10 using a solution of 80% acetonitrile in HPLC grade H 2 O containing 30 mM acetic acid and 30 m l was injected onto a Gemini C18 column (5 m m, 110 A ̊ , 150 mm 6 2 mm) (Phenomenex, Torrance, CA). The sample was eluted using a gradient of 20% solvent A/80% solvent B to 100% solvent A over 10 min at 250 m l/min. at 30 u C [Solvent A: 95% acetonitrile in HPLC grade water, Solvent B: HPLC grade water, 30 mM acetic acid added to both]. Beginning and ending gradient conditions were maintained for 5 min. before and after the gradient run, respectively. PLTX-like compounds were detected at 263 nm and quantified through linear regression of a PLTX standard (2680 Da. from P. tuberculosa purchased from Wako Pure Chemical Industries, Ltd., Japan, dissolved in 50% ethanol according to manufacturer’s instructions) at concentrations of 10, 5, 2.5, 1.25, 0.625, 0.3125 m g/ml. Linear regression analysis was performed using GraphPad Prism software (ver. 4.03, GraphPad Software, Inc., San Diego, CA). Determination of palytoxin type. High resolution liquid chromatography mass spectrometry was used to distinguish palytoxin from 42-hydroxy-palytoxin and deoxy-palytoxin. A Thermo Exactive mass spectrometer coupled with an Accela liquid chromatography system (ThermoScientific, Waltham, MA) was used for this analysis. LC conditions were as follows: 10 ul injections were made onto a Gemini C18 column (5 m m, 110 A ̊, 150 mm 6 2 mm) (Phenomenex, Torrance, CA) and eluted with a linear gradient, as described previously, at a flow rate of 200 m l/ min. Eluent was electrosprayed via the heated probe (HESI-II) into the mass spectrometer (MS). MS conditions were as follows: MS was operated in the full scan positive ion mode from 150–3000 daltons at the ultrahigh resolution setting of 100,000 @ 1 Hz. Other MS conditions are listed in Table 1. PLTX type was confirmed if the total ion chromatogram contained each of the calculated masses in Table 2 as the dominant ion in an isotopic cluster, out to two decimal places. The only predicted ion never detected for any toxin was (M + 3H) 3 + . Minor PLTXs were considered present if select ions for that type were also present as the dominant ion in an isotopic cluster, typically (M + 2H- H 2 O) 2 + and (M + 2H + K) 3 + , the most abundant ions present for each PLTX type in all samples. Molecular Methods. A small piece of tissue ( 10 mg) from one representative polyp from each of the 16 zoanthid samples (15 collected here plus the Virginia zoanthid sample from Deeds and Schwartz [8]) was removed using flame sterilized (with ethanol) tweezers and/or scissors and added to a sterile 1.5 ml microcentrifuge tube. DNA was extracted from tissue through the use of a DNeasy Blood and Tissue Kit (Qiagen # 69506). Reagent volumes were reduced to a quarter of the volume listed in ...