Tissue Distribution of Amino Acid– and Lipid–Brevetoxins after

Article pubs.acs.org/crt

Tissue Distribution of Amino Acid− and Lipid−Brevetoxins after Intravenous Administration to C57BL/6 Mice Tod A. Leighfield, Noah Muha, and John S. Ramsdell* Marine Biotoxins Program, Center for Coastal Environmental Health and Biomolecular Research, NOAA-National Ocean Service, 219 Fort Johnson Road, Charleston, South Carolina 29412, United States ABSTRACT: Brevetoxins produced during algal blooms of the dinoflagellate Karenia are metabolized by shellfish into reduction, oxidation, and conjugation products. Brevetoxin metabolites comprising amino acid- and lipid conjugates account for a large proportion of the toxicity associated with the consumption of toxic shellfish. However, the disposition of these brevetoxin metabolites has not been established. Using intravenous exposure to C57BL/6 mice, we investigated the disposition in the body of three radiolabeled brevetoxin metabolites. Amino acid−brevetoxin conjugates represented by S-desoxyBTX-B2 (cysteine-BTX-B) and lipid−brevetoxin conjugates represented by N-palmitoyl-S-desoxy-BTX-B2 were compared to dihydro-BTX-B. Tissue concentration profiles were unique to each of the brevetoxin metabolites tested, with dihydro-BTX-B being widely distributed to all tissues, S-desoxy-BTX-B2 concentrated in kidney, and N-palmitoyl-S-desoxy-BTX-B2 having the highest concentrations in spleen, liver, and lung. Elimination patterns were also unique: dihydro-BTX-B had a greater fecal versus urinary elimination, whereas urine was a more important elimination route for S-desoxy-BTX-B2, and N-palmitoyl-S-desoxyBTX-B2 persisted in tissues and was eliminated equally in both urine and feces. The structures particular to each brevetoxin metabolite resulting from the reduction, amino acid conjugation, or fatty acid addition of BTX-B were likely responsible for these tissue-specific distributions and unique elimination patterns. These observed differences provide further insight into the contribution each brevetoxin metabolite class has to the observed potencies.

■

INTRODUCTION Brevetoxins are neurotoxic polyether toxins that are produced by the dinoflagellate species Karenia; they accumulate in the marine food web and can have adverse effects on the health of humans and marine organisms.1 Brevetoxins are composed of either 10 fused cyclic ether rings comprising the A-type backbone or 11 fused cyclic ether rings comprising the B-type backbone.2,3 The A-ring of both types of brevetoxin molecules contains a terminal lactone. The other end of the brevetoxin molecules contains three six-membered fused cyclic ether rings with a reactive α,βunsaturated aldehyde addition that is metabolized by shellfish through reduction, oxidation, and conjugation, thus creating multiple metabolites of brevetoxin.4 In shellfish, metabolism of brevetoxins predominately results in a suite of amino acid− brevetoxin and lipid−brevetoxin conjugates that, combined, are responsible for a large portion of observed mammalian toxicity.5,6 Consumption of seafood contaminated with brevetoxin conjugates can result in neurotoxic shellfish poisoning.7,8 Public health protection strategies have historically considered the commonly occurring reduction product of brevetoxin B, dihydro-BTX-B, as the basis for brevetoxin-associated toxicity, yet amino acid−brevetoxin and lipid−brevetoxin conjugates are now known to be the primary drivers of toxicity from brevetoxin contaminated shellfish.9−11 This study investigated the distribution from blood to tissues of three shellfish metabolites of brevetoxin in mice exposed intravenously. The goal of this study was to determine how the tissue uptake of amino acid− and lipid−brevetoxins (S-desoxy-BTX-B2 and N-palmitoyl-S-desThis article not subject to U.S. Copyright. Published 2014 by the American Chemical Society

oxy-BTX-B2, respectively) differs from that of the commonly investigated dihydro-BTX-B. Previous knowledge of brevetoxin tissue distribution comes from studies of murine exposures to the nonconjugated brevetoxin metabolite dihydro-BTX-B. Poli et al.12 intravenously exposed rats to radiolabeled dihydro-BTX-B and found that the majority (96%) distributed to three compartments: the liver, skeletal muscle and intestine. Liver was described as the primary metabolic organ and skeletal muscle was identified as the primary storage pool, containing 70% of the total dose. A broad tissue distribution to all organs was noted and the tissue concentrations decreased with time. The same study showed that feces were found to be the primary route of excretion, accounting for approximately ∼70% of the dose eliminated within 2 days, with some biotransformation to more polar compounds.12 Cattet and Geraci13 administered radiolabeled dihydro-BTX-B by oral gavage to rats. They observed that brevetoxin concentrations were highest in liver and persisted for 8 days after exposure. Distribution was widespread in other tissues, but at low concentrations. Intestine and stomach contained the next highest levels and persisted for approximately 1 day. Elimination occurred through the urine and feces with equivalent amounts in each after 7 days, and the authors concluded that urinary clearance is the primary route of excretion for brevetoxins that are absorbed from the gastrointestinal tract. Received: February 18, 2014 Published: June 20, 2014 1166

dx.doi.org/10.1021/tx500053f | Chem. Res. Toxicol. 2014, 27, 1166−1175

Chemical Research in Toxicology

Article

Figure 1. Radiolabeled brevetoxins used in this study. The brevetoxin B backbone is composed of 11 fused cyclic rings with R group for BTX-B. BTX-B contains an α,β-unsaturated aldehyde (not shown), and the substitutions for the radiolabeled shellfish metabolites dihydro-BTX-B, S-desoxy-BTX-B2, and N-palmitoyl-S-desoxy-BTX-B2 used in this study are shown as R groups.

■

Benson et al.14 administered radiolabeled dihydro-BTX-B to rats via intratracheal instillation as a model for the primary route of human exposure. As measured by body burden, the brevetoxin was distributed within 30 min to skeletal muscle, intestine and, to a lesser extent, liver and lung. Brevetoxin was distributed to, in decreasing order of tissue concentration, the lung, heart, intestines, kidney, and liver. The concentrations were highest in the lung, the site of administration, although after 30 min this represented only 6% of the initial dose. The elevated brevetoxin concentrations present in the gastrointestinal tract after 6 h were presumably associated with bile excretion. Excretion of brevetoxin into urine and feces within 48 hours accounted for approximately 90% of the brevetoxin dose. Tibbetts et al.15 administered radiolabeled dihydro-BTX-B via intratracheal instillation to mice. Using total body burden measurements, this study showed that brevetoxin was cleared quickly from the lungs and was distributed to the liver, gastrointestinal tract, and skeletal muscle within 30 min. The total integrated amount measured in the tissues (area under the concentration with time curve) was highest in liver, with approximately half as much in kidneys, intestines, stomach, lungs, and heart, and a quarter as much in blood and spleen. There was no accumulation in liver over time, with most depurated within 4 days. Three-quarters of the administered dose was recovered in feces and urine after 4 days, with the majority in feces. A better understanding of the contribution that these three brevetoxin shellfish metabolites have on toxicity will influence decisions in the development and application of new assays for the protection of public health, and aid in establishing risk criteria such as action levels for brevetoxin exposures. Determining the distribution of amino acid− and lipid−brevetoxin conjugates is critical in determining the overall toxicity in brevetoxinassociated intoxications or mortalities. To examine the differences in distribution of amino acid− and lipid−brevetoxin conjugates compared to dihydro-BTX-B, mice were exposed to sublethal amounts of a radiolabeled-brevetoxin conjugate, and then the concentration in multiple tissues, urine, and feces was measured using radioactivity as a proxy. We found that amino acid− and lipid−brevetoxin conjugates distributed in unique patterns compared to those of dihydro-BTX-B which may partially account for the differential observed toxicities.

EXPERIMENTAL PROCEDURES

Chemicals. A tritium-labeled amino acid−brevetoxin conjugate ([3H]-S-desoxy-BTX-B2) with a specific activity of 1.0 Ci/mmol and a carbon-radiolabeled lipid−brevetoxin conjugate ([14C]-N-palmitoyl-Sdesoxy-BTX-B2) with a specific activity of 0.2 Ci/mmol were semisynthetically prepared by our laboratory using the materials and methods described in Leighfield et al.16 Tritium-labeled dihydrobrevetoxin B ([3H]-dihydro-BTX-B) was obtained from Amersham (Piscatway, NJ). The stability of the radiolabeled brevetoxins has been addressed in Leighfield et al.16 Animal Care, Exposure and Sample Collection. Adult male C57BL/6 mice weighing 21.4 ± 1.3 g (mean ± SD) (Charles River Laboratories, Wilmington, MA) were group housed in solid bottom polypropylene microisolator cages (Allentown, Inc., Allentown, NJ) with paper chip bedding (PharmaServ, Framingham, MA) in a temperature- and humidity-controlled environment using alternating photoperiod of 12 h. Rodent chow (#5001 PMI Nutrition, Brentwood, MO) and water was available ad libitum during a 4-day acclimation period and throughout the experiment. All animal manipulations were conducted in accordance with the NIH Guide for the Care and Use of Laboratory Animals (NIH Publication No. 86-23, 2011 revision) under the guidance of the laboratory Institutional Animal Care and Use Committee. Prior to intravenous administration of brevetoxin metabolites, mice were fully anesthetized in a chamber supplied with 2.5% isoflurane (Butler Schein Animal Health, Dublin, OH) in surgical oxygen flowing at a rate of 3 L per minute for 2−4 min. Mice were then placed on a surgical board equipped with an anesthetic cone through which maintenance gases containing 1.5% isoflurane in surgical oxygen flowed at a rate of 2 L per minute. A 1 cm ventral skin incision, centered on the clavicle and 1 cm to the right of the midline, was made to expose the right jugular vein. A sterile disposable syringe was then fitted with a 30 gauge needle and loaded with one of the brevetoxin congeners or phosphate buffered saline (PBS) vehicle only. Using the clavicle for support, the syringe needle was carefully inserted into the jugular vein. Slight negative pressure on the syringe plunger resulted in blood flow into the syringe and verified proper needle placement. Animals were administered either 93 ± 3 pmol [3H]-S-desoxy-BTXB2 the amino acid−brevetoxin conjugate, 134 ± 10 pmol [14C]-Npalmitoyl-S-desoxy-BTX-B2 (lipid−brevtoxin conjugate), or 140 ± 5 pmol [3H]-dihydro-BTX-B (Figure 1). This corresponds to 4.5 μg/kg or 4.4 nmol/kg for [3H] S-desoxy-BTX-B2; 8.1 μg/kg or 6.3 nmol/kg for [14C]-N-palmitoyl-S-desoxy-BTX-B2; and 6.0 μg/kg or 6.7 nmol/kg for [3H]-dihydro-BTX-B, which, in terms of radioactivity administered, represents approximately 100,000 cpm per animal. The radiolabeled brevetoxin test solution, prepared as described in Leighfield et al.,16 was 1167

dx.doi.org/10.1021/tx500053f | Chem. Res. Toxicol. 2014, 27, 1166−1175

Chemical Research in Toxicology

Article

investigate metabolic products formed. Extracts of each tissue were prepared by weighing the whole organ which was then chopped prior to solvent extraction. Five mL of acetone was added to each sample and the supernatant collected after centrifugation at 1000 × g. The remaining tissue cake was re-extracted with additional acetone (10 mL total for kidney, and 20 mL total for liver). The tissue cake was then extracted with 5 mL increments of 100% methanol (2 × 5 mL = 10 mL total for kidney and 4 × 5 mL = 20 mL total for liver) until no radioactivity remained as determined by subsampling 10% of the extract volume. All supernatants were combined, 0.45 μm filtered using an Acrodisc GXF/ GHP syringe filter (Pall Corp, Port Washington, NY), and dried under nitrogen. Each sample was resuspended in 0.5 mL 100% methanol and then subjected to chromatographic fractionation. Urine samples were prepared from samples collected from the metabolic cage of either [14C]-N-palmitoyl-S-desoxy-BTX-B2, [3H]-Sdesoxy-BTX-B2, or [3H]-dihydro-BTX-B exposed mice. Prior to fractionation the urine was centrifuged at 2000 × g, and the supernatant was collected and then subjected to liquid chromatographic (LC) fractionation as described below. A 1 gram subsample of feces collected from the metabolic cage of either [14C]-N-palmitoyl-S-desoxy-BTX-B2-, [3H]-S-desoxy-BTX-B2-, or [3H]-dihydro-BTX-B-exposed mice was diluted with 4 mL of water (w/v) and homogenized with a glass tissue homogenizer. This 1 g aliquot was combined with 5 mL acetone, mixed, and centrifuged at 1000g, and the supernatant was collected. The centrifuged pellet was reextracted with 5 mL additional acetone. Acetone supernatants were 0.45 μm filtered using an Acrodisc GXF/GHP syringe filter (Pall Corp, Port Washington, NY) and then dried under nitrogen, and the sample was reconstituted in 0.5 mL of methanol and clarified using a 0.45 μm filter prior to LC fractionation. Fractionation of extracts (liver, kidney, feces) or nonextracted urine was conducted on a Luna C8(2) 100 Å, 5 μm, 150 mm × 2 mm column (Phenomenex, Torrance, CA) using an Agilent Technologies model 1100 LC system with a Gilson model FC203B fraction collector (Middleton, WI). The samples were injected (20 μL × 5) and fractions collected as a series of overlays into the same tubes, which resulted in the collection of 30 samples (2 min each) for each tissue. The mobile phase consisted of water and acetonitrile, with 0.1% formic acid additive at a flow rate of 0.1 mL/min. The chromatographic gradient was: 50% acetonitrile for 0−2 min, then a linear gradient of 50% to 95% acetonitrile from 2 to 20 min, held at 95% acetonitrile from 20 to 40 min, returned to 50% acetonitrile at 41 min, and held for 19 min for column re-equilibration before the next injection. A subsample (0.1 mL) of each fraction was held for ELISA analysis, and the remaining volume (0.9 mL) in the fraction was analyzed for radioactivity. Scintillation cocktail (4 mL) (Scintiverse BD, Fisher Scientific) was added directly to the sample. Samples were held in the dark overnight before being counted in a scintillation counter (TriCarb 3100, PerkinElmer, Waltham, MA). Brevetoxin ELISA. The direct, competitive, brevetoxin ELISA was conducted using the reagents and methods of Maucher et al.19 Briefly, free brevetoxin in the sample competed with horseradish peroxidase (HRP)-labeled brevetoxin B for binding to antibrevetoxin-specific antisera coated onto a 96-well plate. After incubation, unbound brevetoxins were washed from the plate, and the amount of brevetoxin−HRP bound to the plate was visualized through the addition of a dissolved substrate (tetramethylbenzidine) and measured spectrophotometrically. Sample absorbance values were quantified against a standard curve composed of each toxin being investigated. Standards for S-desoxy-BTX-B2 and N-palmitoyl-S-desoxy-BTX-B2 were obtained from previously characterized semisynthesized material.5,6 Dihydro-BTX-B standard was obtained from World Ocean Solutions (Marbionc, Wilmington, NC).

diluted in phosphate buffered saline (or phosphate buffered saline-only for sham dosing) then injected intravenously via the jugular vein as a single subacute bolus dose in a total volume of 0.1 mL. After the dose was administered, the incision was closed with a wound clip. Immediately after dosing, triplicate mice were placed together in a metabolic cage with ad libitum access to food and water for 48 h. Urine and feces were collected continuously on ice from the metabolic cage. Feces and urine were removed from the metabolic cage at 12, 24, 30, 36, and 48 h, weighed, and frozen prior to analysis. Three mice for each time point for each brevetoxin were exposed for 0.5, 3, 12, 24, or 48 h before being humanely euthanized using carbon dioxide asphyxiation prior to terminal sample collection. Tissues were collected by blunt dissection after blood was removed by cardiac exsanguination. Tissues collected included brain, lung, heart, liver, spleen, kidney, lumbar muscle, testes, pancreas, and digestive tract (divided into stomach, duodenum, jejunum−ileum (small intestine), and colon−cecum. The remaining carcass was also collected. All tissue samples were stored at −80 °C until preparation. Preparation of Organs and Carcass. Samples were weighed, then 1 mL Scintigest tissue solubilizer (Fisher Scientific) was added to all organs, except in liver where 2 mL of Scintigest was added, and samples were placed in a 60 °C water bath overnight. To decolor the samples, a hydrogen peroxide solution, (30 wt % in water) was added in 50 μL increments until foaming ceased and the resulting liquid was a lightyellow color. Typically 0.2 mL of the hydrogen peroxide solution was used for each sample. The pH of the resulting solution was then lowered to 6.0 using glacial acetic acid and checked with pH paper. This solution was then transferred to a scintillation vial, the glass vial used in sample preparation was rinsed with water (0.2 mL) and that water transferred also, and then 5 mL of scintillation cocktail (Scintiverse BD, Fisher Scientific) was added directly to the liquefied, decolorized sample. Samples were held in the dark overnight before being counted in a scintillation counter (TriCarb 3100, PerkinElmer). All samples that had radioactivity measurements lower than twice that of the corresponding PBS vehicle-treated preparation for each tissue were considered to be less than the respective limit of quantitation for each brevetoxin. Concentrations of brevetoxin were determined from the radioactivity measured in each tissue. Carcasses comprised the remaining biological material not specifically collected as described previously. The majority of this sample by weight is presumed to be bone (11%), muscle (38%), skin (16%), and dissectible adipose (7%).18 These samples were prepared by adding 50 mL of 2 M sodium hydroxide to the remaining carcass and then were incubated at 60 °C water bath overnight. Triplicate 2 mL aliquots were removed and placed into 20 mL scintillation vials. These were decolorized by adding approximately 400 μL of a hydrogen peroxide solution, (30 wt % in water) until foaming ceased and the resulting liquid was a light-yellow color. The pH of the resulting solution was then lowered to 6.0 using glacial acetic acid, typically ∼100 μL, and checked with pH paper. Preparation of Urine and Feces. Urine (0.5 mL) without modification was combined with 4 mL of scintillation cocktail (Scintiverse BD, Fisher Scientific). Samples were held in the dark overnight before being counted in a scintillation counter (Tri-Carb 3100, PerkinElmer). Feces were prepared for scintillation counting by adding four volumes of HPLC-grade water to the entire sample (w/v), and then vortexed to prepare a slurry. Triplicate 1 g aliquots of aqueous feces were weighed directly into scintillation vials. Scintigest (2 mL) was added, and samples were placed in a 60 °C water bath overnight. To decolor fecal samples, a hydrogen peroxide solution, (30 wt % in water) was added in 100 μL increments until foaming ceased and the resulting liquid was a light-yellow color. Typically 0.4 mL of the hydrogen peroxide solution was used for each sample. The pH was then lowered to 6.0 using glacial acetic acid and checked with pH paper. Fifteen mL of scintillation cocktail (Scintiverse BD, Fisher Scientific) was then added directly to the sample. Samples were held in the dark overnight before being counted in a scintillation counter (Tri-Carb 3100, PerkinElmer). LC Fractionation. A liver from a mouse exposed to [14C]-Npalmitoyl-S-desoxy-BTX-B2 and a kidney collected from a mouse exposed to [3H]-S-desoxy-BTX-B2 were subjected to fractionation to

■

RESULTS This study measured the tissue distribution and elimination of three radiolabeled brevetoxin metabolites in multiple tissues, feces, and urine. Amino acid− and lipid−brevetoxin conjugates were compared to dihydro-BTX-B for each tissue. Tissue concentrations expressed as weight equivalents of each 1168

dx.doi.org/10.1021/tx500053f | Chem. Res. Toxicol. 2014, 27, 1166−1175

Chemical Research in Toxicology

Article

dihydro-BTX-B had measurable concentrations over 48 h, but levels declined over time. Taken together, the gastrointestinal tract (stomach, duodenum, jejunum−ileum, and colon−cecum) all showed delayed peak concentrations either 3 or 12 h after dosing, consistent with an intravenous route of administration. N-palmitoyl-S-desoxyBTX-B2 concentrations in the gastrointestinal tract were comparable to the amounts observed in dihydro-BTX-B-exposed mice, with more N-palmitoyl-S-desoxy-BTX-B2 present in the combined colon−cecum, whereas the dihydro-BTX-B was higher in the jejunum and ileum. The duodenum from Sdesoxy-BTX-B2-treated mice showed the highest levels at 0.5 h followed by the combined colon−cecum at 3 h. Lung tissues showed accumulation of N-palmitoyl-S-desoxyBTX-B2 with a peak at 12 h accounting for nearly 2% of the dose and greater than 5 times the concentration of dihydro-BTX-B at all times. The N-palmitoyl-S-desoxy-BTX-B2 amounts measured in the lung were approximately 2−3 times greater than the whole blood brevetoxin concentrations measured at the same postdose time by Leighfield et al.16 S-desoxy-BTX-B2 and dihydro-BTX-B were found at low concentrations in the lung and were measurable at all collected times. There were no measurable amounts of dihydro-BTX-B in the spleen, and S-desoxy-BTX-B2 was measurable at low levels at only some time points. On the contrary, N-palmitoyl-S-desoxyBTX-B2 in the spleen showed increasing accumulation with time and accounted for up to 3% of the dose. Additionally, there was approximately 2.4 times more N-palmitoyl-S-desoxy-BTX-B2 in the spleen than in the blood at 0.5 h, and at 48 h there was 25 times more N-palmitoyl-S-desoxy-BTX-B2 in the blood measured at the same postdose time as previously described.16 In mice administered S-desoxy-BTX-B2, kidney had the highest brevetoxin concentration of all tissues. The amount of S-desoxy-BTX-B2 associated radioactivity present in kidney accounted for 8−22% of the dose, where dihydro-BTX-B and Npalmitoyl-S-desoxy-BTX-B2 accounted for less than 2% of the dose. The highest concentration of S-desoxy-BTX-B2 in kidney was measured at 3 h with nearly 20 times more brevetoxin than was present in the blood at this time.16 Furthermore, S-desoxyBTX-B2 persisted at high levels throughout the 48 h time period. Low N-palmitoyl-S-desoxy-BTX-B2 levels remained constant in the kidney, whereas dihydro-BTX-B levels dropped over the duration of the experiment. Radiolabel recovered from kidney of S-desoxy-BTX-B2-treated mice eluted at the same time as the administered material (Figure 5). Kidneys from other brevetoxin exposed animals were not subjected to similar fractionation because the low levels of radioactivity present in the samples precluded this analysis. Liver showed two peak concentrations of N-palmitoyl-Sdesoxy-BTX-B2 at 3 and 48 h, and together these two time points accounted for greater than 20% of the dose. During these two peak periods after N-palmitoyl-S-desoxy-BTX-B2 exposure, tissue concentrations were more than 4 times greater than blood concentrations at 3 h and greater than 11 times the blood concentration at 48 h at the same postdose time described previously.16 S-desoxy-BTX-B2 concentrations in liver were low or below the detection limit at all measured times and accounted for