USF Libraries
USF Digital Collections

Vectors of brevetoxins to marine mammals

MISSING IMAGE

Material Information

Title:
Vectors of brevetoxins to marine mammals
Physical Description:
Book
Language:
English
Creator:
Flewelling, Leanne J
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Karenia brevis
Red tide
Food web
Brevetoxicosis
Florida manatee
Dissertations, Academic -- Marine Science -- Doctoral -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: Mass mortalities of Florida manatees (Trichechus manatus latirostris) and bottlenose dolphins (Tursiops truncatus) have been attributed to brevetoxins produced by the Florida red tide dinoflagellate Karenia brevis. The multiple routes through which marine mammals can be exposed to brevetoxins have complicated efforts to understand the mechanisms that lead to mass mortality events. In spring of 2002, 34 endangered Florida manatees died in southwest Florida, and in spring of 2004, 107 bottlenose dolphins died in the Florida Panhandle. These events provided unique opportunities to make clear connections between ingested brevetoxins and marine mammal mortalities without the confounding issues of concurrent exposure through direct contact or inhalation. Prior to 2002, the accumulation of brevetoxins on or in seagrass had never been previously reported, and the delayed or chronic exposure of manatees to brevetoxins through seagrass was not recognized as a threat.Brevetoxins were shown to persist in association with seagrass at high levels for weeks and at lower levels for months in the absence of K. brevis. Analyses of the epiphytes and detritus on the surface of the seagrass leaves as well as of the cleaned seagrass leaves and rhizomes revealed that during a K. brevis bloom as much as half of the toxin present in the seagrass may be associated with the leaves themselves, while after a bloom, the majority of the toxin present is associated with the epiphytes. The 2004 mass mortality of bottlenose dolphins in the Florida Panhandle clearly indicated that fish have the potential to vector brevetoxins to higher tropic levels. Analyses of fish collected live from St. Joseph Bay and southwest Florida revealed that brevetoxin accumulation in fish is a common occurrence.Planktivorous clupeid fish are capable of accumulating high concentrations of brevetoxins within their viscera, and their movement can result in spatial separation of a bloom and animal exposure. Sciaenid species and pinfish also accumulated brevetoxins but to a lower extent. These fish, as well as other omnivorous and piscivorous species, may retain brevetoxins in their tissues at significant concentrations after a bloom has dissipated, which may lead to temporal separation of blooms and animal exposure.
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2008.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Leanne J. Flewelling.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 143 pages.
General Note:
Includes vita.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002004829
oclc - 352824117
usfldc doi - E14-SFE0002675
usfldc handle - e14.2675
System ID:
SFS0026992:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 002004829
003 fts
005 20090527083658.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 090527s2008 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002675
035
(OCoLC)352824117
040
FHM
c FHM
049
FHMM
090
GC11.2 (Online)
1 100
Flewelling, Leanne J.
0 245
Vectors of brevetoxins to marine mammals
h [electronic resource] /
by Leanne J. Flewelling.
260
[Tampa, Fla] :
b University of South Florida,
2008.
500
Title from PDF of title page.
Document formatted into pages; contains 143 pages.
Includes vita.
502
Dissertation (Ph.D.)--University of South Florida, 2008.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
520
ABSTRACT: Mass mortalities of Florida manatees (Trichechus manatus latirostris) and bottlenose dolphins (Tursiops truncatus) have been attributed to brevetoxins produced by the Florida red tide dinoflagellate Karenia brevis. The multiple routes through which marine mammals can be exposed to brevetoxins have complicated efforts to understand the mechanisms that lead to mass mortality events. In spring of 2002, 34 endangered Florida manatees died in southwest Florida, and in spring of 2004, 107 bottlenose dolphins died in the Florida Panhandle. These events provided unique opportunities to make clear connections between ingested brevetoxins and marine mammal mortalities without the confounding issues of concurrent exposure through direct contact or inhalation. Prior to 2002, the accumulation of brevetoxins on or in seagrass had never been previously reported, and the delayed or chronic exposure of manatees to brevetoxins through seagrass was not recognized as a threat.Brevetoxins were shown to persist in association with seagrass at high levels for weeks and at lower levels for months in the absence of K. brevis. Analyses of the epiphytes and detritus on the surface of the seagrass leaves as well as of the cleaned seagrass leaves and rhizomes revealed that during a K. brevis bloom as much as half of the toxin present in the seagrass may be associated with the leaves themselves, while after a bloom, the majority of the toxin present is associated with the epiphytes. The 2004 mass mortality of bottlenose dolphins in the Florida Panhandle clearly indicated that fish have the potential to vector brevetoxins to higher tropic levels. Analyses of fish collected live from St. Joseph Bay and southwest Florida revealed that brevetoxin accumulation in fish is a common occurrence.Planktivorous clupeid fish are capable of accumulating high concentrations of brevetoxins within their viscera, and their movement can result in spatial separation of a bloom and animal exposure. Sciaenid species and pinfish also accumulated brevetoxins but to a lower extent. These fish, as well as other omnivorous and piscivorous species, may retain brevetoxins in their tissues at significant concentrations after a bloom has dissipated, which may lead to temporal separation of blooms and animal exposure.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
590
Co-advisor: Gabriel A. Vargo, Ph.D.
Co-advisor: Joseph J. Torres, Ph.D.
653
Karenia brevis
Red tide
Food web
Brevetoxicosis
Florida manatee
690
Dissertations, Academic
z USF
x Marine Science
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.2675



PAGE 1

Vectors of Brevetoxins to Marine Mammals by Leanne J. Flewelling A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy College of Marine Science University of South Florida Co-Major Professor: Gabriel A. Vargo, Ph.D. Co-Major Professor: Joseph J. Torres, Ph.D. Susan S. Bell, Ph.D. Richard H. Pierce, Ph.D. Jan H. Landsberg, Ph.D. Date of Approval: October 24, 2008 Keywords: Karenia brevis red tide, food web, brevetox icosis, Florida manatee, bottlenose dolphin Copyright 2008, Leanne J. Flewelling

PAGE 2

Dedication This work is dedicated to my family with much love, especially to my husband Michael for his constant support and encourag ement, to my mom for hanging on to that graduation card for all these years, an d to my dad whom I miss every day.

PAGE 3

Acknowledgments There were many individuals whose gui dance, support, and assistance were integral to the work presented here. I am grateful to Dr. Karen Steidinger who brought me to the foot of this path and gave me a hard shove and to Dr. Gabriel Vargo and my committee members for their patience and inpu t. I owe a special thanks to Dr. Jan Landsberg for her support and for her ceasel ess efforts to get me to look and think beyond boundaries. For providing samples and assisting with field collections, I am greatly indebted to FWC-FWRI staff members in the Marine Mammal Pathobiology Laboratory, Fisheries Independent Monitoring, Marine Fisherie s Biology, Fish and Wildlife Health, and Harmful Algal Blooms groups; the NMFS Ma rine Mammal Health and Stranding Response Program; the Florida Department of Environmental Protection (St. Joseph Bay Aquatic Preserve); FDACS Division of Aqu aculture; and Damon Gannon at Mote Marine Laboratory. I am especially grateful to FWC-FW RI HAB staff member s Karen Atwood, April Granholm, Jay Abbott, Sheila O Dea, and fo rmer staff member Dan Hammond for all of the technical and analytical support they provided in the fi eld and the lab. I am also thankful to FWC-FWRI HAB taxonomists, Jennifer Wolny and Dr. Earnest Truby, for providing Karenia brevis cell concentrations, and to Dr Richard Pierce and Michael Henry at Mote Marine Laboratory for provi ding LC-MS analyses. This work could not have been performed without the assistance of Dr. Jerome Naar at UNC Wilmington who developed the assay used here and who pr ovided training and much advice over the years. This work was funded in large part by the Florida Fish and Wildlife Conservation Commission and also by grants awarded to J. Landsberg and J. Naar from the CDC and FDOH (U50-CCU423360-01) and to J. N aar and L. Flewelling through the NOAA Oceans and Human Health Initiative (NA05NOS4781246). Funds provided by the NOAA CSCOR HAB Event Response Fund enab led offshore water collections during the 2004 dolphin mortality.

PAGE 4

Note to Reader: The original of this document contains colo r that is necessary for understanding the data. The original dissertation is on file with the USF library in Tampa, Florida.

PAGE 5

i Table of Contents List of Tables iii List of Figures v Abstract viii Chapter 1. Introduction 1 Harmful Algal Blooms 1 Karenia brevis and Brevetoxins 2 Metabolism of Brevetoxins 7 Effects of Brevetoxins on Human Health 8 Brevetoxin Impacts on Fish and Wildlife 11 Brevetoxin-related Mari ne Mammal Mortalities 12 Chapter 2. Confirmation of Manatee Exposure to Brevetoxins During the 2002 Manatee Mortality Event 17 Introduction 17 Methods 25 Results 28 Discussion 35 Chapter 3. Brevetoxin Accumulation and Persistence in Seagrass 41 Introduction 41 Methods 42 Results 46 2002 Collections 46 2005 Collections 51 Discussion 56 Concentration of Brevetoxins in the Epiphytes 58 Concentration of Brevetoxins in the Leaves and Sheaths 59 Concentration of Brevetoxins in the Rhizomes and Roots 59 Food Web Implications 61 Chapter 4. Confirmation of Bottlenose Dolphin Exposure to Brevetoxins During the 2004 Dolphin Mortality Event 64 Introduction 64 Methods 71 Results 73

PAGE 6

ii Discussion 79 Chapter 5. Brevetoxin Accumulation in Live Fish 84 Introduction 84 Methods 86 Results 89 2004 St. Joseph Bay Collections 89 2005-2006 St. Joseph Bay Collections 91 Southwest Florida Collections 97 Discussion 104 Chapter 6. Conclusions 113 Literature Cited 119 Appendices 137 Appendix A. 2002 red tide-suspect mana tees tested for brevetoxins by ELISA. 138 Appendix B. 1996 red tide-suspect mana tees for which archived tissues were tested for brevetoxins by ELISA. 139 Appendix C. 2002 southwest Florida mana tees killed by watercraft collision that were tested for brevetoxins by ELISA. 141 Appendix D. 2002 east coast Florida ma natees tested for brevetoxins by ELISA. 142 Appendix E. 2004 dolphins tested for brevetoxins by ELISA. 143 About the Author End Page

PAGE 7

iii List of Tables Table 2.1 Brevetoxin concentrations (ng PbTx-3 eq./g or ng PbTx-3 eq./mL) measured in tissues from 2002 suspec ted red tide-related mortalities of manatees. 30 Table 2.2 Spearman rank correlation coe fficients (and p values) for brevetoxin concentrations measured in differe nt sample types of 2002 red tidesuspect manatees. 31 Table 2.3 Brevetoxin concentrations (ng PbTx-3 eq./g) measured in archived samples from 1996 suspected red tide-re lated mortalities of manatees. 32 Table 2.4 Brevetoxin concentrations (ng PbTx-3 eq./g or ng PbTx-3 eq./mL) measured in tissues from southw est Florida manatees killed by watercraft collision. 34 Table 3.1 Spearman rank correlation coefficients (and p-values) for brevetoxin concentrations in seawater, sediment, and seagrass components (ns = not significant). 56 Table 4.1 Brevetoxin concentrations (ng PbTx-3 eq./g or ng PbTx-3 eq./mL) measured in tissues from the 2004 dolphin mortality. 75 Table 4.2 Brevetoxin concentrations (ng PbTx-3 eq./g) measured in tissues from rough-toothed dolphins ( Steno bredanensis ) that stranded on Floridas Atlantic coast in August 2004. 77 Table 4.3 Spearman rank correlation coe fficients (and p values) for brevetoxin concentrations measured in differe nt sample types of 2004 dolphins. 77 Table 4.4 Brevetoxin concentrations (ng Pb Tx-3 eq./g) measured in prey items retrieved from the stomach contents of 2004 dolphins. 78 Table 4.5 Domoic acid concentrations (ng DA/g or ng DA/mL) measured in tissues from the 2004 dolphin mortality. 79 Table 5.1 Range of brevetoxin concentrations mesasured in fish collected from St. Joseph Bay between March 28 and May 5, 2004 (during and immediately following the dolphin mortality event). 93

PAGE 8

iv Table 5.2 Range of brevetoxin concentrations measured in fish collected from St. Joseph Bay between February 2005 and November 2006. 94 Table 5.3 Brevetoxin concentrations measur ed in fish collected from southwest Florida between June 2004 and December 2007. 98

PAGE 9

v List of Figures Figure 1.1 Brevetoxin backbone structur es and congeners identified from K. brevis 6 Figure 2.1 Causes of manatee deat hs in Florida from 1974 to 2001. 18 Figure 2.2 Karenia brevis cell concentrations in 2002 seawater samples collected A) February 10March 1, B) March 2-9, C) March 10-16, and D) March 17-23. 22 Figure 2.3 Locations of manatee carca sses reported in Southwest Florida between March 10 and May 1, 2002. 23 Figure 2.4 Number of manatee carcasse s reported per week in Manatee, Sarasota, Charlotte, Lee, and Co llier counties from March 3 to May 4, 2002. 24 Figure 3.1 Sites of environmental sample collections in southwest Florida in 2002 (Sites 1, 2, 3, and Forked Creek) and 2005 (PH1 and PH2). 44 Figure 3.2 Brevetoxin concentrations meas ured by ELISA in composite seagrass samples, sediments, and seawater; and K. brevis cell densities observed in samples collected at: A) site 1, B) site 2, C) site 3, and D) Forked Creek. 47 Figure 3.3 Brevetoxin concentrations measured by ELISA in seagrass components (leaf scrapings, leaves and sheaths, and rhizomes and roots) and K. brevis cell densities observed in samples collected at: A) site 1, B) site 2, C) site 3, and D) Forked Creek. 49 Figure 3.4 Brevetoxin concentrations meas ured by ELISA in leaf scrapings and leaves and sheaths as a fraction of the total brevetoxin measured in the shoot at: A) site 1, B) site 2, and C) site 3. 50 Figure 3.5 Brevetoxin concentrations meas ured by ELISA in composite seagrass samples, sediments, and seawater; and K. brevis cell densities observed in samples collected at: A) PH1 and B) PH2. 52

PAGE 10

vi Figure 3.6 Brevetoxin concentrations measured by ELISA in seagrass components (leaf scrapings, leaves and sheaths, and rhizomes and roots) and K. brevis cell densities observed in samples collected at: A) PH1 and B) PH2. 54 Figure 3.7 Brevetoxin concentrations meas ured by ELISA in leaf scrapings and leaves and sheaths as a fraction of the total brevetoxin measured in the shoot at: A) PH1 and B) PH2. 55 Figure 4.1 Stranding locations of bottlenose dolphin ca rcasses recovered between March 10 and April 13, 2004. 68 Figure 4.2 Number of dolphin strandings in the Florida Panhandle per day between March 10 and April 13, 2004. 69 Figure 4.3 Sites and dates of water co llections between March 11 and May 6, 2004. 70 Figure 4.4 A) Sites of colle ction of seawater analyzed for brevetoxins and brevetoxin concentrations measured ; and B) sites of collection of seawater analyzed for domoic acid. 74 Figure 5.1 Locations of fish colle ctions in southwest Florida. 88 Figure 5.2 Maximum K. brevis cell densities and brev etoxin concentrations measured in seawater samples co llected throughout St. Joseph Bay between September 2003 and December 2006. 90 Figure 5.3 Average brevetoxin concentrations measured in fish (all species) collected throughout St. Jose ph Bay between 2004 and 2006. 95 Figure 5.4 Prevalence of brevetoxin contamination in fish (all species) collected throughout St. Jose ph Bay between 2004 and 2006. 96 Figure 5.5 Brevetoxin concentrations measured in A) muscle and B) viscera of clupeid fish collected in southwest Florida. 101 Figure 5.6 Brevetoxin concentrations measured in A) muscle and B) liver or viscera of pinfish collected in southwest Florida. 102 Figure 5.7 Brevetoxin concentrations measured in A) muscle and B) liver of red drum and spotted seatrout collected in Tampa Bay. 103

PAGE 11

vii Figure 5.8 Karenia brevis cell concentrations in seawater samples and brevetoxin concentrations measur ed in the livers of seatrout collected between June and November 2007. 110 Figure 6.1 Average brevetoxin concentrations measured in tissues and stomach contents of manatees from the 2002 mortality event and dolphins from the 2004 mortality event. 115 Figure 6.2 Causes of manatee deat hs in Florida from 1974 to 2001 (top, n = 4,368) and from 2002 to 2007 (bottom, n = 2,091). 117

PAGE 12

viii Vectors of Brevetoxin s to Marine Mammals Leanne J. Flewelling ABSTRACT Mass mortalities of Florida manatees (Trichechus manatus latirostris ) and bottlenose dolphins (Tursiops truncatus ) have been attributed to brevetoxins produced by the Florida red tide dinoflagellate Karenia brevis The multiple routes through which marine mammals can be exposed to brevetox ins have complicated efforts to understand the mechanisms that lead to mass morta lity events. In spring of 2002, 34 endangered Florida manatees died in southwest Florida, and in spring of 2004, 107 bottlenose dolphins died in the Florida Panhandle. Thes e events provided unique opportunities to make clear connections between ingested br evetoxins and marine mammal mortalities without the confounding issues of concurrent exposure th rough direct contact or inhalation. Prior to 2002, the accumulation of brevetoxins on or in seagrass had never been previously reported, and the delayed or chr onic exposure of manatees to brevetoxins through seagrass was not recognized as a thr eat. Brevetoxins were shown to persist in association with seagrass at high levels for weeks and at lower leve ls for months in the absence of K. brevis Analyses of the epiphytes and de tritus on the surface of the seagrass leaves as well as of the cleaned seagrass l eaves and rhizomes revealed that during a K. brevis bloom as much as half of the toxin pres ent in the seagrass may be associated with the leaves themselves, while after a bloom, the majority of the toxin present is associated with the epiphytes. The 2004 mass mortality of bottlenose dolphi ns in the Florida Panhandle clearly indicated that fish have th e potential to vector brevetoxi ns to higher tropic levels. Analyses of fish collected live from St. Jo seph Bay and southwest Florida revealed that

PAGE 13

ix brevetoxin accumulation in fish is a common occurrence. Planktivorous clupeid fish are capable of accumulating high con centrations of brevetoxins within their viscera, and their movement can result in spatial separation of a bloom and animal exposure. Sciaenid species and pinfish also accumulated brevetox ins but to a lower extent. These fish, as well as other omnivorous and piscivorous specie s, may retain brevetoxins in their tissues at significant concentrations after a bloom has dissipated, which may lead to temporal separation of blooms and animal exposure.

PAGE 14

1 Chapter 1. Introduction Harmful Algal Blooms Algal blooms, or higher than normal c oncentrations of phytoplankton, occur when environmental conditions allow phytoplankton gr owth to exceed losses due to mortality, sinking, grazing, etc.; or when conditions result in the physical concentration of phytoplankton cells. Harmful algal blooms (HABs) are phytoplankton blooms that are in some way deleterious to humans or the envi ronment. Algal species may be considered harmful for many different reasons, and the harmful effects are not necessarily dependent on high biomass or cellular abunda nces (Smayda 1997b). In some instances, a bloom of a species not typically categorized as a HAB species can have incidental, but harmful, effects. For example, hypoxic or a noxic water conditions caused by a bloom of otherwise benign species can result in non-disc riminating kills of fish and invertebrates (Hallegraeff 1993), and water discol oration from an algal bloom can result in severe light attenuation leading to loss of aquatic vegetation (Dennis on et al. 1989). The typical HAB species have more specific harmful eff ects. Ambush predation by the dinoflagellate Pfiesteria piscicida on fish is a very specific and uni que mode of harmful action (Smayda 1997b). Another example is the diatom Chaetoceros which possesses heavily silicified setae that can clog or pierce th e gills of fish and crustacean s causing irritation and even death (Hallegraeff 1993; Smayda 1997b). Many HAB species produce toxins that result in human poisonings, fish kills, and toxicity to other animals. The reasons for microalgal toxin production are not defi nitively known. Proposed roles in clude a variety of cellular functions, inhibition of competitors, grazer dete rrents, chemical signals, and evolutionary relics (Van Dolah 2000). However, in the case of the toxic diatom Pseudo-nitzschia spp., recent research has identified a role for th e production and release of the neurotoxic amino acid domoic acid (DA) as a part of a high efficiency iron uptake system (Wells et

PAGE 15

2 al. 2005). Continued research is likely to id entify similar functions for toxins produced by other species. The past several decades have witnessed an apparent increase in the frequency, intensity, and geographic distri bution of HABs worldwide. In discussing this issue, Hallegraeff (1993) and Anderson (1994) consid er several possible explanations. They suggest that these increases could, to some extent, be perceived increases due to the heightened awareness of HABs among the scien tific community as well as the growth of aquaculture activities that has result in in creased monitoring of both the waters and the fisheries products. Among their possible e xplanations for actual increases are the introduction of HAB species to non-nativ e waters via ballast water, coastal eutrophication, and climate change. Whether or not the apparent trend of increasing HABs is real, there has been a very real in crease in our understandi ng of activities that may lead to increases in HABs and in our knowledge of the risks that HABs can pose to both human and environmental health. Karenia brevis and Brevetoxins Dinoflagellates account for 75% of all HAB species (Smayda 1997a). In Florida, the dominant HAB species is the F lorida red tide dinoflagellate, Karenia brevis (Davis) G. Hansen et Moestrup (formerly Gymnodinium breve and Ptychodiscus brevis). Karenia brevis blooms almost annually in the eastern Gulf of Mexico (Tester and Steidinger 1997; Steidinger et al. 1998b) and repeatedly imp acts southwest Florida in particular. While K. brevis is usually restricted to the Gulf of Me xico, blooms can be entrained in the Loop Current and may then be transported north by the Gulf Stream. As a result, blooms of K. brevis occur occasionally on Floridas east coast, and in 1987 a bloom was transported as far north as North Carolina (Tester et al. 1991). Florida red tides t ypically occur in the late summer and fall months, but have b een documented in every month (Tester and Steidinger 1997; Stei dinger et al. 1998b). Observations of historical blooms suggest that K. brevis blooms are initiated 1874 km offshore on the West Florida Shelf (Steidinger 1975; Steidinger and Haddad 1981; Tester and Steidinger 1997). In the oligotrophic offshore Gulf of Mexico waters, potential

PAGE 16

3 nutrient sources for K. brevis include decaying blooms of the nitrogen-fixing cyanophyte Trichodesmium and nutrients from upwelled water (W alsh and Steidinger 2001; Walsh et al. 2006). Blooms can be transported by current s and winds to shallow nearshore waters, where likely sources of nutrients include decaying fish, phosphorus from fossil deposits, and coastal runoff (Vargo et al. 2004; Walsh et al. 2006; Vargo et al. 2008). Several other Karenia spp. have been identified (Hay wood et al. 2004), and increasingly Karenia blooms in the Gulf of Mexico are being recognized as assemblages of multiple Karenia species, including K. mikimotoi K. selliformis and K. papilionacea (Heil et al. 2004; Heil et al. 2006; Haywood et al. 2007). Karenia brevis is harmful because it produces a family of neurotoxins known as brevetoxins (PbTxs). These lipid-soluble pol yether toxins exert their neurotoxicity by binding to a specific site (designated as si te 5) on voltage-sensitive sodium channels (VSSCs) (Poli et al. 1986), with half-saturation of binding sites (Kd) at nanomolar concentrations. Binding results in opening of sodium channels at normal resting potential, a longer mean open time for the sodium channel, and inhibition of sodium channel inactivation, which can lead to repetitive firi ng in nerves (Catterall and Gainer 1985; Poli et al. 1986; Baden et al. 1995). Although suspected in other Karenia species, so far brevetoxin production has only been conclu sively demonstrated from one other Karenia species, Karenia concordia, isolated from New Zealand waters (Chang et al. 2006). Brevetoxin production was recently documente d in a raphidophyte tentatively named Chattonella cf. verruculosa (Bourdelais et al. 2002) and subsequently renamed Chloromorum toxicum (Giner et al. 2008; Tomas et al. submitted). Production of brevetoxins has also been reported, but not structurally confirmed, in several other raphidophytes including Chatonella marina (Ahmed et al. 1995), Chatonella antiqua (Haque and Onoue 2002), Heterosigma akashiwo (Khan et al. 1997), and Fibrocapsa japonica (Khan et al. 1996). However, the overwhelming majority of documented brevetoxin-related impacts are the result of K. brevis blooms. At least eleven congeners of brevetoxins (PbTx-1, -2, -3, -5, 6, -7, -9, -10, -11, 12, and -tbm) have been isolat ed and characterized from K. brevis (Baden et al. 2005) (Fig. 1.1). Three additional congeners (PbTx-8, 13, and -14) are believed to be artifacts

PAGE 17

4 of the extraction process, not brevetoxins produced by K. brevis. All brevetoxins are composed of a series of trans -fused polyether rings with a lactone function on the A-ring, a rigid tail region formed by the last four ri ngs, and a spacer region separating the lactone from the rigid region (Baden and Adams 2000; Baden et al. 2005). PbTx-1 (10 rings) and PbTx-2 (11 rings) are believed to be the two parent toxins from which all other brevetoxins are derived via various substitutions on the terminal ring (Fig. 1.1). Subsequently, brevetoxins are grouped in to two types according to their backbone structure. Brevetoxins possessing the 10-ring backbone of PbTx-1 are termed A-type, Type 2, or PbTx-1-type toxins. Those po ssessing the 11-ring backbone of PbTx-2 are called B-type, Type 1, or PbTx-2-type toxins To maintain clarity, PbTx-1-type and PbTx-2-type will be used in this disc ussion. Although PbTx-1-type toxins are more potent, the PbTx-2-type toxins are much more abundant (Risk et al. 1979; Baden and Mende 1982; Shimizu et al. 1986; Baden a nd Tomas 1988; Baden 1989). Acute toxicity to mice expressed as the 24 hour LD50 following intraperitoneal (i.p.) injection for PbTx2 and PbTx-3, the most abundant br evetoxins, are 0.20 mg/kg and 0.17 mg/kg, respectively (Baden and Mende 1982). The rece nt discoveries in bot h culture and bloom material of oxidized forms of PbTx-1 and Pb Tx-2 (Plakas et al. 2004) as well as several brevetoxins with open A-rings (Abraham et al. 2006) have further increased the number of known brevetoxin structures. Brevetoxins were first de scribed as a ciguatera-lik e poison by McFarren et al. (1965) as a result of investig ations into human illnesses th at occurred during a December 1962 southwest Florida red tide. This char acterization was based on the symptoms observed in mouse bioassays of extracted sh ellfish. The toxins were first successfully purified and characterized by two groups in the late 1970s (Baden et al 1979; Risk et al. 1979). Earlier reports were put forth by vari ous research groups on the properties of K. brevis toxins, but contained conf licting descriptions as well as inconsistent molecular masses and elemental compositions (refs. cited in Baden et al. 1979, Baden 1983, and Risk et al. 1979). Theses discrepancies were possibly the result of multiple compounds with differing bioactive properties (ichthyot oxic, hemolytic), multiple toxins, different protocols extracting differen t brevetoxins or modifying the structure of unstable

PAGE 18

5 compounds, or impurities in the preparation (Baden et al. 1979; Risk et al. 1979). Ultimately, using thin layer chromatography (T LC) fractionation and high pressure liquid chromatographic separation (HPLC) coupled with fish bioassays, Ri sk et al. (1979) and Baden et al. (1979) succeeded in purifying th ree brevetoxins. Risk et al. (1979) reported two toxic fractions from (then) Gymnodinium breve designated T46 (PbTx-1) and T47 (PbTx-2), while Baden et al. (1979) described two toxins designated T17 (PbTx-3) and T34 (PbTx-2). By 1981, structural determina tion of one brevetoxin (brevetoxin B, T47, T34, or PbTx-2) was accomplished (Lin et al. 1981), and a report of the structures of several more PbTx-2-type toxins quickly fo llowed (Chou and Shimizu 1982; Golik et al. 1982; Chou et al. 1985). The stru ctural identificati on of the second backbone structure exhibited by brevetoxin A (T47 or PbTx-1) was reported by Shimizu et al. in 1986. Additional congeners were id entified in the 1980s and were renamed PbTx-1 through PbTx-9 for Ptychodiscus brevis toxins, in accordance with the nomenclature for K. brevis at that time (Poli et al. 1986). This naming convention has been retained as shorthand for polyether brevetoxins but has not been unive rsally adopted. Some other abbreviations (e.g. GB toxin, BTX) have been or are still used. In addition to parent toxins PbTx-1 a nd PbTx-2 and their natural derivatives found in seawater and culture, K. brevis also produces brevenal, a five-ring trans -fused polyether molecule that is a brevetoxin an tagonist (Bourdelais et al. 2004). Brevenal suppress the toxic effects of brevetoxins by in hibiting brevetoxins from binding to site 5 on VSSCs, presumably as a result of competitio n for the binding site (Bourdelais et al. 2005). Prasad and Shimizu (1989) also described three four-ring trans -fused polyethers, termed hemi-brevetoxins, which are similar to brevenal but have not been reported to be brevetoxin antagonists. Brevenal and hemi-b revetoxins may be incomplete products of the brevetoxin synthesis pathway (Ramsdell 2008).

PAGE 19

O O O CH3 O O CH3 O O O O O O OH CH3 CH3R A B C D E F G H I JH I J K O O O O OH R CH3 O O CH3 CH3 O O O O O CH3 CH3 CH3 O CH3 A B C D E F GPbTx-1-Type (A-type) backbone:ToxinR = PbTx-1 CH2C(=CH2)CHO PbTx-7CH2C(=CH2)CH2OH PbTx-10CH2CH(CH3)CH2OHPbTx-2-Type (B-type) backbone:ToxinR = PbTx-2 CH2C(=CH2)CHO PbTx-3CH2C(=CH2)CH2OH PbTx-5PbTx-2, C-37 OAc PbTx-6PbTx-2, H-ring epoxide PbTx-9CH2CH(CH3)CH2OH PbTx-11*CH2CH=CH2PbTx-12*CH2COCH2CH2CH2CH3PbTx-tbm*H O O O CH3 O O CH3 O O O O O O OH CH3 CH3R A B C D E F G H I JH I J K O O O O OH R CH3 O O CH3 CH3 O O O O O CH3 CH3 CH3 O CH3 A B C D E F GPbTx-1-Type (A-type) backbone:ToxinR = PbTx-1 CH2C(=CH2)CHO PbTx-7CH2C(=CH2)CH2OH PbTx-10CH2CH(CH3)CH2OHPbTx-2-Type (B-type) backbone:ToxinR = PbTx-2 CH2C(=CH2)CHO PbTx-3CH2C(=CH2)CH2OH PbTx-5PbTx-2, C-37 OAc PbTx-6PbTx-2, H-ring epoxide PbTx-9CH2CH(CH3)CH2OH PbTx-11*CH2CH=CH2PbTx-12*CH2COCH2CH2CH2CH3PbTx-tbm*H Figure 1.1 Brevetoxin backbone structur es and congeners identified from K. brevis *Newly identified brevetoxins (Bourdela is and Baden 2004; Baden et al. 2005). 6

PAGE 20

7 Metabolism of Brevetoxins As brevetoxins pass from K. brevis seawater, and aerosols into other living organisms, the more reactive congeners are transformed into brevetoxin metabolites. Most research to date on the metabolism of brevetoxins has focused on the modifications that occur in bivalve shel lfish. Metabolites found in bivalves are the products of reduction, oxidation, and conjugation to adducts including taurine, cysteine, cysteine sulfoxide, amino acids and fatty acids, with close to 30 metabolites identified thusfar from shellfish tissue (Morohashi et al. 1995; Murata et al. 1998; Morohashi et al. 1999; Nozawa et al. 2003; Ishida et al. 2004; Plakas et al. 2004; Wang et al. 2004; Ishida et al. 2006). Most modifications to the molecule occu r at the side chain on the terminal ether ring that differentiates the brevetoxin conge ners, resulting in a somewhat simplified assortment of conjugates with either a Pb Tx-1-type or PbTx-2-type of backbone. These structural modifications are be lieved to be part of a detoxification strategy. Information on potency is lacking for most metabolites. However, acute toxicity to mice has been reported for some, and brevetoxin metabolite s have been shown to contribute to NSP toxicity (Ishida et al. 1996; Morohashi et al. 1999; Poli et al. 2000; Nozawa et al. 2003; Plakas et al. 2004). A minimum LD50 (i.p.) of 0.306 mg/kg was reported for cysteinePbTx-B-sulfoxide (BTX-B2) (Murata et al. 1998 ). Ishida et al. (2004) reported a similar minimum LD50 of 0.3-0.5 mg/kg for oxidized PbTx-2 (PbTx-COOH or BTX-B5). BTXB3 isolated from greenshell mussels ( Perna canaliculus ) in New Zealand was found to be nontoxic at a dose of 0.3 mg/kg (Morohashi et al. 1995). A few shellfish metabolites have demonstrated higher toxicities than PbTx-3. The mouse lethality by i.p. injection of BTXB4, also isolated from greenshell mussels, was estimated to be 0.1 mg/kg (Morohashi et al. 1999), and the N-taurine conjugate BTX-B1 isolated from the New Zealand cockle ( Austrovenus stutchburyi ) exhibited a minimum lethal dose of 0.05 mg/kg (Ishida et al. 1995). Recently, investigators have begun wo rking with semi-synthesized brevetoxin derivatives to try to address the questions regarding their potencies. LD50s for semisynthesized cysteine-PbTx-B a nd cysteine-PbTx-B-sulfoxide (BTX-B2) were reported to be 0.211 mg/kg and 0.400 mg/kg, respectively (Selwood et al. 2008). The latter value suggests a lower potency than that reported by Murata et al (1998), and the investigators

PAGE 21

8 suggest it may be the result of differences in the fasting state of the mice used. Using RBA and cytotoxicity assays, lo wer potencies than that of Pb Tx-3 have been reported for semi-synthesized cysteine-PbTx-B and cysteine -PbTx-B-sulfoxide (Bottein Dechraoui et al. 2007). Similar in vitro comparisons of the open A-ri ng derivatives of PbTx-2 and PbTx-3 determined their potencies were great ly reduced compared to their closed A-ring counterparts (Roth et al. 2007). Relative potencies assessed by mouse bioassay and such alternate in vitro methods do not always agree, sugges ting that factors other than sodium channel affinity also contribute to toxicity in vivo (Selwood et al. 2008). Knowledge about metabolism of brevet oxins in other organisms including mammals is also progressing. By exposing ra ts via intravenous (i.v.) administration to radiolabelled PbTx-3, Poli et al. (1990) provided the first demonstration of in vivo brevetoxin metabolism in mammals. Using TLC to analyze rat feces, they found several fractions with radiolabel that did not correspond to PbTx-3, and suggested that the liver was the major organ of metabolism with bili ary excretion as an important route of elimination. Similar results we re obtained by Kennedy et al. (1992) when they exposed gulf toadfish ( Opsanus beta ) to radiolabelled PbTx-3. Since then, in vivo and in vitro studies on fish and rats have found some of the metabolites initia lly identified from shellfish (Radwan et al. 2005; Radwan and Ramsdell 2006; Naar et al. 2007). Currently this is a very active branch of brevetoxin research, with recent studies proposing new mammalian brevetoxin metabolite structures and pathways (Radwan et al. 2005; Radwan and Ramsdell 2006; Abraham et al. 2008), adding to the already complex picture of the brevetoxins. Effects of Brevetoxins on Human Health During a bloom, many species of filter-feeding bivalves ingest K. brevis cells and efficiently concentrate brevetoxins in their ti ssues. Ingestion of cont aminated shellfish by humans can result in Neurotoxic Shellfish Poisoning (NSP) (McFarren et al. 1965; Music et al. 1973; Hemmert 1975). To protect the public from NSP, shellfish harvesting areas in Florida coastal waters are closed when K. brevis cell counts exceed 5,000 cells per liter, and are reopened only after counts have dropped below this threshold and shellfish

PAGE 22

9 toxicity assessed by regu latory mouse bioassay has been de termined to be less than 20 mouse units per 100 g of meat, the level cons idered safe for human consumption (NSSP 2005). Since the state of Florida began its monitoring program in 1970, no cases of NSP have been recorded from consumption of le gally harvested shellfis h. Of course, illegal harvest does take place and results in peri odic outbreaks of this seafood poisoning, as occurred in 2006 during which 17 NSP cases were documented (Watkins et al. 2007). NSP symptoms typically begin 3-6 hours afte r ingestion, but can be sooner, and include nausea, diarrhea, paresthesi as, tingling of lip s or tongue, muscle ache, lack of coordination, temperature revers al (cold feels hot), and vertigo (Morris et al. 1991; Baden et al. 1995). In severe cases, a feeling of constricti on in the throat may occur. Recovery is usually complete within days. Although th ere have been no recorded human deaths attributed to brevetoxin exposure, a small number of patients su ffering from NSP have been put on ventilatory suppor t (Hemmert 1975; Watkins et al. 2008) and might have died without adequate supportive care. The possibility of hum an death due to brevetoxin exposure can not be ruled out. Brevetoxins can also affect humans who i nhale the toxins. Repor ts of respiratory irritation concurrent with discolored wate r and fish kills are well documented. Even before the toxins had been identified, Woodcock (1948) demonstrated a link between vapors of water containing the red tide orga nism and reversible respiratory irritation. Karenia brevis is an unarmored dinoflagellate, and it is generally thought that waves, wind, and other turbulence in the ocean lyse K. brevis cells, resulting in brevetoxincontaminated particles (Pierce 1986). Bubbles help to transport brevetoxins associated with cell fragments or other particles in the water to the sea surface, where they can be aerosolized as the bubbles burst (Pierce et al. 1990) and then transported onshore by winds (Pierce 1986; Cheng et al. 2005). Brevetoxin aerosol concentr ations are primarily dependent on the toxin concentrations in th e seawater and the wind speed and direction (Cheng et al. 2005). When expressed as the co ncentration in the a queous component of aerosol, brevetoxin concentrations can be enriched as much as 50 times compared to the seawater (Pierce et al. 1990), but brevetoxin concentrations per volume of air are much lower than the concentrations per volume of seawater from which the aerosol was

PAGE 23

10 generated. For example, Pierce et al. (1989) reported a concentra tion of PbTx-2+3 in K. brevis bloom water of 200 g/L of seawater and a correspon ding aerosol concentration of 160 ng/m3 of air. Similarly, in a more recent study, they reported concentrations of PbTx1+2+3 of up to 14 g/L in seawater and corresponding br evetoxin concentrations of up to 33 ng/m3 of air (Pierce et al. 2005). In many cas es, symptoms are alleviated soon after leaving the beach area or entering ai r conditioning. However, a recent study by Kirkpatrick et al. (2006) documented an in crease in emergency room admissions for respiratory-related illnesses in coastal residents during a 3-month K. brevis bloom period compared to the same time frame in a year without a bloom, suggesting that the effects of respiratory exposure to brevetoxins may be longer-lived than pr eviously thought. More severe symptoms are possible in sensitive popul ations (e.g. infants or elderly people) or individuals with lung disease such as asthma or chronic obstructive pulmonary disease (Fleming et al. 2005). The neurotoxic mode of action of brevet oxins is the best-studied, but effects on other systems in the body have been observed and additional receptors are suspected. Because pulmonary effects are observed at br evetoxins concentrations 1,000 times lower than what is required to produce neuronal effects, a separate pulmonary receptor is suspected (Abraham et al. 2005; Baden et al. 2005). Also various an alogues of the semisynthetic brevetoxin antagonist -naphthoyl-brevetoxin, bound to rat brain synaptosomes with similar affinities yet resulted in very different bronchoconstrictor effects in sheep, further suggesting a separate binding site in lung tissues (Michelliza et al. 2007). Impacts on the immune system and the mechanisms by which brevetoxins may impair immune function are also being studie d. Brevetoxins have been shown to inhibit cathepsins (Sudarsanam et al. 1992; Baden et al. 2005). Cathepsins are proteases responsible for carrying out the first step of the immune response, processing antigens to antigenic epitopes (Katunuma et al. 2003), and interference may lead to suppression of the immune system. Suppressed antibody pr oduction by splenic lymphocytes (but no neurotoxic effects) was obser ved in rats repeatedly expos ed via inhalation to crude extracts of K. brevis and to pure PbTx-3 over periods of days to weeks (Benson et al. 2004a; Benson et al. 2004b; Benson et al. 2005). Benson et al. (2005) suggested

PAGE 24

11 cathepsin inhibition as a likely mechanism as there was no evidence of overt cytotoxicty to the splenocytes. In vitro experiments using leukemic T cells (Jurkat) demonstrated decreases in cellular metabolic activity and viability, and an in crease in apoptosis in cells exposed to brevetoxins PbTx-2 and PbTx-6 (Walsh et al. 2008). These cellular effects were either decreased or absent in cells exposed to PbTx-3. Potential genotoxic effects are currently be ing investigated as well. Sayer et al. (2005) found that brevetoxins can cause single-stranded and double-stranded DNA breaks in human lymphocytes, which could al so presumably impair the immune system. In this study, DNA damage was prevented as well as reversed by treatment with the brevetoxin antagonist, brevenal. Chromosoma l aberrations and a decrease in cell proliferation were documented in Chinese hamster ovary cells (CHO-K1-BH4) exposed to brevetoxins (Sayer et al. 2006). These effects were also partially mitigated by treatment with brevanal. Recently in vitro studies by Radwan and Ramsdell (2008) have demonstrated that PbTx-2 is capable of bi nding to nucleic acids and producing nucleotide adducts, a common initiation step for carcinogenesis. Brevetoxin Impacts on Fish and Wildlife Brevetoxins have profound impacts on aqua tic animals. The most affected are fish, with massive fish kills documented as far back as 1844 (Ingersoll 1882) and observed almost annually during K. brevis blooms for over 40 years in the Gulf of Mexico (Landsberg 2002). Often this is the only noticeable effect of K. brevis blooms. The extensive fish kills that occur during re d tides are indeed the reason that red tides were so well documented in the Gulf of Mexico a full century before the causative organism was ever identified (Feinstein et al. 1955). The pr ofound ichthyotoxicity is also the primary characteristic of brevetoxins us ed by researchers to guide them to its purification (Baden et al. 1979; Risk et al. 1979). Baden (1989) listed 24-hr fish bioassay LD50s of brevetoxin congeners ranging from 3-5 nM for PbTx-1-type brevetoxins to 1037 nM for PbTx-2-type brevetoxins. The ichthyo toxicity of brevetoxins varies with life stage (Landsberg 2002) and routes of expos ure (Starr 1958; Naar et al. 2007). Although

PAGE 25

12 not yet addressed in any studies, it is possibl e that there are different tolerance levels among fish species. In addition to fish kills, brevetoxins have also been associated with mortalities of most aquatic animals. Although most species of bivalves appear unaffected by K. brevis blooms, bay scallops ( Argopecten irradians ) seem less impervious to the effects of red tides. Bay scallop mortalities have occurred during red tides (Shu mway 1990; Landsberg 2002), and an almost total recruitment failure of bay scallops was documented during the only K. brevis bloom to impact North Carolina coas tal waters (Summerson and Peterson 1990). Leverone (2006) demonstrat ed dramatically reduced survival of bivalve larvae (hard clams, Mercenaria mercenaria ; oysters, Crassostrea virginica ; and bay scallops, A. irradians ) upon exposure to K. brevis concentrations of 1,000-5,000 cells/mL. Mortalities of other invertebrates such as shrimp, sponges, sea urchins, and crabs have occurred as well (Steidinger et al. 1973; Landsberg 2002), but it is not known whether these mortalities reflect toxicity of brevetoxins to the organisms or are the result of hypoxic or anoxic conditions resulting from the bloom. Many records exist recounting bird mortalities during K. brevis red tides (Quick and Henderson 1974; Forrester et al. 1977; Kreuder et al. 2002; Landsberg et al. 2007; Landsberg et al. in press), as well as increased mortalities and strandings of sea turtles (Landsberg 2002; Redlow et al. 2002; Fauquier et al. in prep; Landsberg et al. in press). Brevetoxin-related Marine Mammal Mortalities Despite a long history of red tide and fish kills in the Gulf of Mexico, records of marine mammal mortalities occurring coincide nt with a red tide bloom are comparatively scarce. At least 28 species of marine mamma ls are found in the Gulf of Mexico, 15 of which are classified as common (Wrsi g et al. 2000). The most often encountered species along the Florida Gulf coast are Florida manatees ( Trichechus manatus latirostris a subspecies of the West Indian manatee) and bottlenose dolphins ( Tursiops truncatus ). Other common marine mammals in Florida Gulf waters include Atlantic spotted dolphins ( Stenella frontalis ), pygmy sperm whales ( Kogia breviceps ), and dwarf sperm whales ( K. sima ) (Reynolds and Wells 2003).

PAGE 26

13 The first record of marine mammal mortali ties hypothesized to be a result of red tide in the Gulf of Mexico is from 1946-47 (Gunter et al. 1948). Fr om November 1946 to August 1947, an intense bloom persisted off the southwest coast of Florida and the Florida Keys. Mass mortalities of marine animals were associated with this bloom including an estimated 50 million fish, bivalves, crabs, shrimp, barnacles, and bottlenose dolphins (Gunter et al. 1948; Landsberg 2002). The earliest suggestion of possible red tide-related manatee mortalitie s was made by Layne (1965) in his reference to the deaths of seven manatees in Lee Count y in late March and early Ap ril and an eighth in early May, 1963. These mortalities occurred during a re d tide that also resu lted in a number of human illnesses (McFarren et al. 1965). More extensive and better documented ma rine mammal mortalities have occurred within the last three decades. In 1982, between February and April, 39 manatees died in Lee County, and numerous sick manatees were observed (O'Shea et al. 1991). A K. brevis bloom was present in the area at the onset of the event, but the mortality event continued for another three w eeks after the bloom subsided in late March. Incidental ingestion of toxic filter-feedi ng tunicates attached to seagrass was believed to be the primary source of brevetoxin (O'Shea et al. 1991). A similar, but much larger-scale event, occurred during a red tide bloom in 1996. In th is case there were fe wer reports of sick manatees, but an unprecedented 149 manatee d eaths were recorded between March and May. Manatees were recovered in areas with high cell concentrations, and the end of the mortality event closely followed the dissipati on of the bloom (Landsberg and Steidinger 1998). The stomach contents of several animals were found to be toxic (Van Dolah et al. 2003), but brevetoxin-specific immunohistochemi cal techniques suggested that inhalation was also an important route of exposure (Bossart et al. 1998). Since the identification of the red tide or ganism and its toxins, there have also been two notable mass mortalities of bottlenose dolphins that were suspected to be red tide-related. Brevetoxins were hypothesized to have played a role in an extensive and prolonged stranding event along the U.S. Atla ntic coast in 1987-1988 involving more than 740 bottlenose dolphins (Geraci 1989). This event occurred during the year in which a red tide was transported out of the Gulf of Mexico via the Loop Current and carried

PAGE 27

14 northward in the Gulf Stream (Tester et al. 1991). Numerous other physiological stressors and high contaminants were also documented in the dolphins, resu lting in a lack of consensus with regard to the ultimate cause of death and the extent to which brevetoxins may have been involved. Nonetheless, the National Marine Fisheries Service (NMFS) estimated that this mortality represented a decline of more than 50% of the coastal migratory stock of bottlenose dolphins al ong the U.S. mid-Atlantic coast, and subsequently designated this population as d epleted according to the Marine Mammal Protection Act (58 FR 17789, April 6, 1993) a status that remains unchanged. More direct evidence for the involvemen t of brevetoxins wa s obtained during the investigation into a mass mortality of more than 120 bottlenose dolphins in the Florida Panhandle between August 1999 and February 20 00. In this case, two peaks of dolphin strandings occurred coincident with K. brevis blooms in the area (V an Dolah et al. 2003). Brevetoxins were measured in multiple tissu es as well as stomach contents of some dolphins, but only 29% of the stranded animals tested were positive for brevetoxins (Van Dolah et al. 2003). As no indications of a ny other potential contributing factors (e.g. infectious disease) were identified, inves tigators concluded that this event was most likely the result of brevetoxin intoxication with the route of exposure most likely through both inhalation and inge stion (Mase et al. 2000). Acceptance of the concept that brevetoxins could be lethal to marine mammals has not been easily won, in part because ther e have been no documented cases of human deaths attributed to brevetoxin exposure. Another reason is that blooms occur almost annually along Floridas southwest coast, ye t mass marine mammal mortalities have been extremely rare. Following the 1996 manat ee mortality and 1999-2000 dolphin mortality, resistance to this notion has eased. However, an understanding of the mechanisms that lead to these unusual events has been complicated by the multiple ways marine mammals can be exposed to brevetoxins. Routes of exposure for marine mammals include direct exposure to bloom water, ingestion of brevetoxins in bloom water or in other organisms in which the toxin has bioaccumulated, or inhalation of aerosolized toxins. Human intoxications are characteris tically the result of inges tion of bivalves that have concentrated brevetoxins in their tissues. Th e typical diets of herbivorous manatees and

PAGE 28

15 piscivorous dolphins do not include species well-known to accumulate brevetoxins. In 1982, the abundance of filter-feed ing tunicates in the gastro intestinal tracts of the recovered manatees and in seagrass beds in the area where mortalities were occurring provided a vector for brevetoxins to the mana tees and an explanation for the continued mortalities after the bloom had dissipated. While tunicates collected in the area were toxic by NSP mouse bioassay, this hypothesis could not be proven at the time with laboratory experiments exposing tunicates to K. brevis (FWC-FWRI, formerly FDNR Marine Research Lab, unpublished data). Dolphi ns are piscivores, and fish are killed en masse during K. brevis blooms. During the investigation into the 1999-2000 dolphin mortality, the stomach contents (fish) of some stranded animals were found to contain brevetoxins, but toxins were only detected in 29% of the 24 animals tested (Van Dolah et al. 2003) and the presence of the K. brevis bloom during the event allowed for other potential routes of exposur e. Although hypothesized in 1987-88, very few samples were analyzed, dietary exposure was confirmed in onl y one dolphin, and the ability of fish to accumulate high and potentially lethal le vels of brevetoxins was not proven. The primary obstacle to establishing th e connection between marine mammal mortality events and K. brevis blooms has been a lack of appropriate brevetoxin testing methods. The methods used to investigate the 1982 manatee mortality and the 1987-88 dolphin mortality (i.e. mouse bioassay, fish bioassay-guided TLC, and in 1987-88 HPLCUV) were the best available at the time but were not definitive for brevetoxins. Advancements in detection methods occurred in the period prior to the 1996 manatee mortality. That investigation benefited gr eatly from the newly-developed receptor binding assay (RBA) (Poli et al. 1986) and immunohistochemical technique (Bossart et al. 1998). The latter was then used to re trospectively examine tissues from the 1982 manatee mortality and provided evidence in support of brevetoxins as the causative agent (Bossart et al. 1998). By the 1999-2000 dol phin mortality, high throughput RBAs (Van Dolah et al. 1994) and liquid chromatographymass spectrometry (LC-MS) methods (Hua et al. 1995; Hua et al. 1996) improved both se nsitivity and specificity, allowing for more definitive confirmation of marine mammal exposure to brevetoxins. However, both

PAGE 29

16 dolphin mortality investigations were weakened by a lack of adequate samples and an insufficient number of analyses to provide convincing results. Nevertheless, as each of these events ha s occurred, evidence both circumstantial and, more recently, direct has accumulated to support the connection between red tides and marine mammal mortalities. The developmen t of rapid and sensitive toxin assays has increased both the quality of the testing and the quantity of animals and tissues that can be tested. The recent development of a brev etoxin-specific ELISA (Naar et al. 2002) has provided an improvement over the receptor-bind ing assay, yielding bette r sensitivity with fewer matrix effects. Both RBAs and ELISAs have become standard tools in the HAB laboratory of the Florida Fish and Wildlife Conservation Commissions Fish and Wildlife Research Institute (FWC-FWRI). Having these capabilities within the agency responsible for conserving the wildlife resources in Fl orida has enabled the direction of more intensive efforts towards investigations of ma rine mammal mortality events occurring in Florida waters and more thorough environmenta l investigations. Such efforts are largely responsible for the data presented in this study. Since the dolphin mortality of 1999-2000, tw o separate mass mortality events (of manatees in 2002 and bottlenose dolphins in 2004) have occurred that have provided unique opportunities to make a clear connect ion between ingested brevetoxins and marine mammal mortalities without the confounding issues of concurrent exposure through direct contact or inhalation. In the following chapters, I will present data from these two events that implicate ingested br evetoxins as the causative agent, and the environmental evidence that illustrates how marine mammals can be exposed to lethal levels of brevetoxins even when separated in time and space from the K. brevis bloom that was the source of the toxins.

PAGE 30

17 Chapter 2. Confirmation of Manatee Exposure to Brevetoxins During the 2002 Manatee Mortality Event Introduction The endangered Florida manatee ( Trichechus manatus latirostris ) is a subspecies of the West Indian manatee ( T. manatus Linnaeus 1758) that is native to Floridas coastal and riverine habitats (Reynolds and Powell 2002). The distribution of the Florida manatee is heavily influenced by temperature. They are widely dispersed in warm weather, and can occur as far north as Virgin ia and as far west as Texas in the summer months. In cold weather, they migrate south a nd seek areas of natural or artificial warm water, where they aggregate in large numbers (Reynolds and Powell 2002). Thermal stress in manatees occurs when they are exposed to low temperature waters, and prolonged exposure to temperatures below 20C can cause death (Bossart 2002; Reep and Bonde 2006). Manatees are at risk from a number of anthropogenic factors and natural events. The greatest human contributor to manatee mortalities is watercraft collisions (FWC 2007b), accounting for 24.5 % of determinable manatee deaths between 1974 and 2001 (Fig. 2.1). Other human-related causes of mana tee mortalities include entrapment in water control structures, cr ushing or asphyxiation in flood gates and canal locks, entanglement, incidental inge stion of debris, and loss of warm water habitat (FWC 2007b; Runge et al. 2007b). Natural hazards incl ude hurricanes and disease, but the most consistent naturally occurring threat is extrem e cold weather, with cold stress blamed for 4% of manatee mortalities in the last thr ee decades (Fig. 2.1). A combination of risk factors may also result in ne gative synergistic effects. For example, the effects of one stressor may result in increased susceptibility to another stressor. Often the cause of death is undetermined for the largest percentage of manatee mortalities.

PAGE 31

1974-200124.5% 3.9% 2.7% 20.6% 4.2% 5.7% 9.4% 29.1% Watercraft Collision Flood Gate/Canal Lock Human, other Perinatal Cold Stress Red Tide Natural, other Undetermined Figure 2.1. Causes of manatee deaths in Florida from 1974 to 2001. Graph was created using data provided by the FWC-FWRI Marine Mammal Pathobiol ogy Laboratory (n = 4,368). Cold stress includes 10 perinatal deat hs in which cold stress was suspected. 18

PAGE 32

19 Brevetoxins produced by Karenia brevis the Florida red tide dinoflagellate, have also been responsible for larg e numbers of manatee mortalities in southwest Florida. In 1982, from February through April, 39 manatees died in the Caloosahatchee River and its estuary in southwest Florida. Several othe rs were found sick, e xhibiting neurological symptoms such as disorientation, inability to submerge or maintain a horizontal position, spasms, lip flaring, and difficulty breathi ng (O'Shea et al. 1991) High mortality and morbidity of double-crested cormorants was also noted in the area at the same time. High concentrations of K. brevis were present in the area from January through March, but the mortality event continued for th ree weeks after the dissipation of the bloom. In almost all manatees necropsied, stomachs and gastrointes tinal tracts were full, and most contained filter-feeding ascidians (tunicates). Incidental ingestion of toxic tunicates attached to seagrass was believed to be the primary s ource of brevetoxin expos ure, but definitive analytical evidence to suppor t this hypothesis was not obtained (O'Shea et al. 1991). In 1996, coincident with a K. brevis bloom, the still-unsurpassed largest mass mortality of manatees occurred in the same region of southwest Florida. In this case, there was a strong correla tion between the appearan ce and disappearance of K. brevis at high concentrations and the be ginning and the end of the mo rtality event (Landsberg and Steidinger 1998). Fewer sick manatees we re noted, but an unprecedented number of manatees died (n = 149). Manatees were reco vered in areas with hi gh cell concentrations and environmental conditions conducive to aer osolization of brevet oxins (Landsberg and Steidinger 1998). Brevetoxins we re detected in the stomach contents of several animals (Van Dolah et al. 2003). Brevetoxin meas ured in the lung tissue and results of brevetoxin-specific immunohistochemical tec hniques suggested that inhalation was an additional route of exposure (B ossart et al. 1998). The e nd of the 1996 mortality event more closely followed the dissipation of the bloom. These two mortality events alone represent three-fourths of the presumed red tiderelated manatee deaths and account for more than 4% of the total number of manatee deaths between 1974 and 2001 (Fig 2.1). Landsberg and Steidinger (1998) recognized the ecological and environmental factors that were common to these mass mortality events as well as seven manatee deaths in 1963 (Layne 1965). Allowi ng for the possibility of

PAGE 33

20 isolated manatee deaths any time a K. brevis red tide occurs, they hypothesized that there is a high risk of manatee mortalities on an epizootic scale if K. brevis is present at bloom concentrations in the months of February th rough April as the Gulf of Mexico warms and manatees begin to disperse in large number s from their winter congregations. A second critical factor is the salinity within intracoastal waters. Since K. brevis typically prefers a salinity range typical of coas tal or oceanic waters, i.e. >24 (Steidinger et al. 1998b), salinity must be sufficiently high to allow K. brevis to move inshore. In 1996, K. brevis concentrations were low (<105 cells/L) at salinities below 27.9 (Landsberg and Steidinger 1998). The confluence of these two events an inshore K. brevis bloom and the end of manatee over-wintering has pr efaced the manatee mortality events documented in 1963, 1982, and 1996. These conditions were again realized in 2002. The K. brevis bloom of early 2002 was actually the continuation of a persistent bloom that had been impacting southwest Florida since late August 2001. Throughout January 2002, K. brevis was present both inshore and offshore between Manatee and Collier counties. Karenia brevis cell concentrations queried from FWCs Harmful Algal Bloom database ranged from low (< 105 cells/L) to high (> 106 cells/L), with high concentrations localized to an area approxima tely 15 miles offshore, west of Charlotte Harbor. By mid-February, medium (105 106 cells/L) to high ce ll concentrations were widespread in nearshore waters with populat ions moving into the Intracoastal waterways and the north end of Charlotte Harbor (Fig. 2.2A). In early March 2002, cell concentrations dropped from high levels and K. brevis was present at medium levels around Charlotte Harbor and Pine Island S ound (Fig. 2.2B). By March 10th the bloom had dissipated to a large extent and K. brevis was present primarily at low levels (Figs. 2.2C, D). In mid-March 2002, manatee deaths greatly increased along the coasts of Sarasota, Charlotte, Lee, and Collier countie s, with mortalities continuing throughout April. Between March 10 and May 1, a tota l of 43 manatees deaths were recorded between Manatee and Collier counties (Fig. 2.3), with the majority (n = 31) occurring between March 10 and April 6 (Fig. 2.4). Th ree of these manat ees were killed by

PAGE 34

21 watercraft collision, three were the victims of cold stress, and one died at birth. For the remaining 36, the cause of death coul d not be immediately determined. Necropsies were conducted at the FWC-FWRI Marine Mammal Pathobiology Laboratory on most of the recovered manatee carcasses. A nu mber of non-specific gross findings such as congestion of the nasophar yngeal mucosa; wet heavy lungs with mucus or froth-filled bronchi; and bleeding on cut surfaces of lungs, liver, and kidney had also been noted in the past events and were c onsistent with brevetoxicosis (FWC 2007a). Despite the history of mass mortalities of manatees in this area associated with K. brevis blooms, these findings were initially confounding as the K. brevis bloom dissipated at the same time the mortalities began (Figs. 2.2C, D) and was temporally separated from the majority of manatee carcasses. A newly-developed enzyme linked immunsorbent assay (ELISA) for brevetoxins (Naar et al. 2002) was used to investigate whether, and to what extent, these manatees had been exposed to brevetoxins.

PAGE 35

22 B A Manatee De Soto Charlotte Lee Collier Sarasota Hardee Manatee De Soto Charlotte Lee Collier Sarasota HardeeCharlotte HarborManatee De Soto Charlotte Lee Collier Sarasota Hardee Manatee De Soto Charlotte Lee Collier Sarasota HardeeCharlotte Harbor Caloosahatchee RiverCaloosahatchee River Gulf of MexicoGulf of Mexico Figure 2.2. Karenia brevis cell concentrations in 2002 seawater samples collected A) February 10March 1, B) March 2-9, C) March 10-16, and D) March 17-23. Cell concentrations were queried from FWCs Harmful Algal Bloom database, which contains all K. brevis statewide monitoring data. Manatee De Soto Charlotte Lee Collier Sarasota Hardee Manatee De Soto Charlotte Lee Collier Sarasota HardeeCharlotte Harbor Caloosahatchee River Gulf of Mexico < 103cells L-1103-105cells L-1105-106cells L-1>106cells L-1 < 103cells L-1103-105cells L-1105-106cells L-1>106cells L-1 C < 103cells L-1103-105cells L-1105-106cells L-1>106cells L-1 < 103cells L-1103-105cells L-1105-106cells L-1>106cells L-1 < 103cells L-1103-105cells L-1105-106cells L-1>106cells L-1 < 103cells L-1103-105cells L-1105-106cells L-1>106cells L-1 Manatee De Soto Charlotte Lee Collier Sarasota HardeeCharlotte Harbor Caloosahatchee River Gulf of Mexico < 103cells L-1103-105cells L-1105-106cells L-1>106cells L-1 Manatee De Soto Charlotte Lee Collier Sarasota Hardee Manatee De Soto Charlotte Lee Collier Sarasota HardeeCharlotte Harbor Caloosahatchee River Gulf of Mexico < 103cells L-1103-105cells L-1105-106cells L-1>106cells L-1 Gulf of Mexico < 103cells L-1103-105cells L-1105-106cells L-1>106cells L-1 D 25025Kilometers 25025Kilometers 25025Kilometers 25025Kilometers

PAGE 36

Charlotte Harbor Caloosahatchee River Manatee De Soto Charlotte Lee Collier Sarasota Hardee Manatee De Soto Charlotte Lee Collier Sarasota Hardee Watercraft collisionGulf of Mexico Cold stress Perinatal Other Cause of Death Watercraft collisionGulf of Mexico Cold stress Perinatal Other Cause of Death Gulf of Mexico Cold stress Perinatal Other Cause of Death 25025Kilometers Figure 2.3. Locations of manatee carcasses re ported in Southwest Florida between March 10 and May 1, 2002. Map was created using data provided by the FWC-FWRI Marine Mammal Pathobiology Laboratory. 23

PAGE 37

3/3-3/93/10-3/163/17-3/23 3/24-3/303/31-4/64/7-4/134 /14-4/204/214/274/28-5/4Week 0 2 4 6 8 10Number of manatee deaths Figure 2.4. Number of manat ee carcasses reported per week in Manatee, Sarasota, Charlotte, Lee, and Collier counties from March 3 to May 4, 2002. Graph was created using data provided by the FWC-FWRI Marine Mammal Pa thobiology Laboratory. 24

PAGE 38

25 Methods Frozen and individually-stored manatee ti ssues were provided by the FWC-FWRI Marine Mammal Pathobiology Lab and were frozen at -20C until extracted and analyzed. For each manatee analyzed the sex, length, location, and the date the carcass was reported are detailed in Appendices AD (data provided by the FWC-FWRI Marine Mammal Pathobiol ogy Laboratory). Selected tissues and stomach contents from 26 red tide-suspect manatee carcasses recovered between March 15 and April 30, 2002 were analyzed for brevetoxins (Appendix A). Tissue types varied but for most animals included liver (n = 24), kidney (n = 22), urine (n = 22), and stomach contents (n = 26). Lung tissue was collected from 25 manatees, and in 10 cases included both cauda l and cranial lung tissue. Fat tissue was collected from 21 manatees, and in eight cases included abdominal, epicardial, and subdermal fat. Less-regularly sampled tissue ty pes included heart (n = 11) and brain (n = 12). A limited number of similar tissue samp les archived from the 1996 mortality event were available and were analyzed for comparison. Sample types included liver (n = 2), lung (n = 4), and stomach contents (n = 13). A large number of samples of duodenal contents (n = 50) from 1996 were available as well (Appendix B). These samples were also tested, but no comparable samples from the 2002 manatees were available. In total, samples from 61 manatees from the 1996 mortality were tested. To assess potential body burdens or backgr ound levels of brevetoxins in manatees from southwestern Florida, tissues from 11 wa tercraft-related mortalities were analyzed (Appendix C). Three of these manatees were recovered during the mo rtality event (April) and eight were recovered in subsequent mont hs (MaySeptember). Liver (n = 11), kidney (n = 7), lung (n = 8), fat (n = 3), stomach contents (n = 11), and urine (n = 5) were analyzed. Tissues and stomach contents from eight manatees recovered on Floridas east coast, where K. brevis blooms are historically rare, served as negative controls to evaluate sample matrix effects (Appendix D). Sample t ypes included liver (n = 8), kidney (n = 7), lung (n = 8), urine (n = 6), stom ach contents (n = 6), heart (n = 3), fat (n = 3), and brain (n = 2).

PAGE 39

26 Acetone (ACS grade), hexane (ACS gr ade), sulfuric acid, and SuperBlock blocking buffer (Pierce) were purchased fr om Fisher Scientific (Pittsburgh, PA). O phenylenediamine (OPD), phosphate buffe red saline (pH 7.4, PBS), Tween 20, and gelatin were purchased from Sigma-Aldrich (St. Louis, MO). ELISA reagents (PbTx-3 positive control, bovine serum albumin-linked brevetoxin, goat anti-brevetoxin antibodies, biotinylated rabbit anti-goa t secondary antibodies, and horseradish peroxidase-linked streptavidin) were purchased from the University of North Carolina at Wilmington, and were diluted as in structed with PBS prior to use. Manatee tissues and stomach contents were thawed, homogenized, and a single subsample of each (2.5 g) was weighed into a 25-mL centrifuge tube (BD Falcon). Brevetoxins were immediately extracted using 100% acetone (4 mL/g tissue). Homogenates were centrifuged ( 10 minutes at 3,200 x g), the s upernatants were retained, and the pellets were extracted a second time in the same manner. The supernatants were pooled and evaporated to dryness. The extract s were then re-dissolved in 80% aqueous methanol and partitioned twice with 100% hexa ne (1:1 v:v). The methanol fraction was evaporated to dryness and re-dissolved in 100% methanol. Urine samples ( 1 mL) were applied to C18 SPE cartridges (Supelco, 500 mg, 3 mL) that had been pre-conditioned with 6 mL of 100% methanol and then 6 mL of deionized (DI) water. After passing the urine through the columns, the columns were washed with 6 mL DI water and toxins were eluted with 5-6 mL 100% methanol. The methanol extract was then evaporated to dryness and re-dissolved in 3 mL of 100% methanol. Total PbTx-2-type brevetoxin was quantifie d in all samples using a competitive enzyme-linked immunosorbent assay (ELISA) performed according to Naar et al. (2002). Flat-bottom polystyrene 96-well plates (N unc MaxiSorp Immunoplates) were first coated with protein-linked br evetoxin (100 L per well). After 60 minutes, the wells were emptied and were rinsed three times (300 L per well) with phosphate buffered saline (PBS, pH 7.4). Any remaining sites on the we lls were then blocke d by incubating with SuperBlock blocking buffer (200 L per well ) for 30 minutes. The blocking buffer was then poured from the wells, and each well was rinsed three times (300 L per well) with PBS containing 0.1% Tween 20 (PBS-Tween). Anti-brevetoxin antibodies (goat

PAGE 40

27 polyclonal antibodies, 100 L per well) were then incubated in the wells with dilutions of samples or PbTx-3 positive controls (100 L per well in duplicate wells). Samples and positive controls were seria lly diluted in PBS-Tween c ontaining 0.5% gelatin. After 60 minutes, the wells were emptied, and an tibodies not attached to the plate-bound brevetoxin were washed out using three rinses (300 L per well) with PBS-Tween. Antibodies immobilized on the pl ate were detected through subsequent steps linking them to biotinylated secondary antibodies (rabbit anti-goat antibody, 100 L per well, 45 minute incubation), and then horseradish pero xidase (HRP)-linked streptavidin (100 L per well, 30 minute incubation), with three PBS-Tween rinses between each step. The HRP substrate, ortho-phenylenediamine (OPD), was then added to the wells (100 L per well). After five minutes, the reaction was stopped by addition of 0.5 M sulfuric acid (50 L per well). Absorbance of the wells was read in a SpectraCount microplate reader (Packard) at 492 nm. The lower limit of quan tification by ELISA as performed here was approximately 10 ng/g for stomach contents a nd liver; 5 ng/g for lung, kidney, and fat; and 2 ng/mL for urine. Results are expressed as ng PbTx-3 equivalents per gram or per mL, and reflect the overall conc entration of PbTx-2-type brevetoxins and brevetoxin-like compounds present in the sample. All statistics were perfor med on log-transformed data using GraphPad Prism version 5.01 for Windows, GraphPad Software (San Diego California USA, www.graphpad.com). In those instances where more than one fat or lung tissue from an individual manatee was analy zed, brevetoxin concentrations from all fat or lung tissues for that animal were averaged prior to sta tistical analyses. In most cases, non-parametric analyses were used due to the presence of assay results that were below the limit of detection. For statistical analyses, and to calculate means, results that were below the limit of detection were assigned a value of half the limit of detection (2.5 ng PbTx-3 eq./g for kidney, lung, and fat; 5 ng PbTx-3 eq./g for liver and stomach contents; and 1 ng PbTx-3 eq./mL for urine).

PAGE 41

28 Results Results of brevetoxin ELISA analyses of manatee tissues from 26 red tide-suspect carcasses are presented in Table 2.1. Brevetoxins were detected in multiple tissues of all 26 manatees. Highest concentr ations were measured in stomach contents and liver. Stomach contents tested positive in 25 of 26 samples. Concentrations measured were highly variable, ranging from 61 to 1,132 ng Pb Tx-3 eq./g, and averaged 458 ng PbTx-3 eq./g. All 24 liver samples contained measurab le concentrations of brevetoxins with an average concentration of 152 ng PbTx-3 eq./g and a range of 58 to 300 ng PbTx-3 eq./g. Lung tissues (caudal, cranial, or both) of 24 out of 25 manatees tested were positive but were comparatively much lower than stomach and liver tissues, ranging from 6 to 64 ng PbTx-3 eq./g. The average concentration measured in lung tissues was 18 ng PbTx-3 eq./g. For 10 manatees, tissue was tested fr om both the caudal and the cranial region of the lung. Although the relative concentrations were low compared to liver and stomach contents in this small subset of manatees, br evetoxin concentrations in the cranial lung were higher and significantly di fferent from concentrations in the caudal lung (matched ttest, p = 0.0193). Brevetoxins were measured in all 22 kidney samples with an average concentration of 35 ng PbTx-3 eq./g and a maximum concentration of 83 ng PbTx-3 eq./g. All 22 urine samples also tested positive and were consistently between 3 to 20 ng PbTx-3 eq./mL (mean = 10 ng PbTx-3 eq./mL). Brain tissue from 12 manatees was also analyzed. All 12 brain samples were positive with consistently low concentrations measured (9 to 22 ng PbTx-3 eq./g). Similarly, all 11 heart tissue samples tested positive, ranging from 19 to 44 ng PbTx-3 eq./g. Brevetoxin was detected in the fat of 20 out of 21 manatees (in 34 of 37 fat samples) tested. In eight cases, three different types of fat were sampled (abdominal, subdermal, epicardial). The average brevetoxin concentration was highest in abdominal fat (58 ng PbTx-3 eq./ g), followed by subdermal fat (40 ng PbTx-3 eq./g), and was lowest in epic ardial fat (32 ng PbTx-3 eq./g) However, within this small sample set, the brevetoxin concentrations measured in the three fat types did not significantly differ from one a nother (Friedman's test). The Spearman rank correlation coefficient was used to assess the correlation of brevetoxin concentrations in the different sample types tested for the 2002 manatees

PAGE 42

29 (Table 2.2). The only statisti cally significant (p < 0.05) correlations found were weak positive correlations between stomach contents and kidney (rs = 0.525, p = 0.012) and stomach contents and lung (rs = 0.513, p = 0.010). Brevetoxins were detected in all animal s and in 68 out of 69 of the archived tissues from the 1996 manatee mortality (Table 2.3). Concentrations of brevetoxins were highest in the duodenal contents followed by stomach contents and then liver. Brevetoxin concentrations in duodenal contents ranged from 48 to 1,738 ng PbTx-3 eq./g (mean = 493 ng PbTx-3 eq./g) and in stomach contents ranged from 79 to 1,042 ng PbTx-3 eq./g (mean = 363 ng PbTx-3 eq./g). Concentrations measured in the two liver samples (174 ng PbTx-3 eq./g and 304 ng PbTx-3 eq./g) were high but still fell within the range measured in the 2002 animals. Brevetoxin was detected in three out of the four lung samples from 1996 animals but at levels close to the limit of detection (4 9 ng PbTx-3 eq./g). Results of brevetoxin ELISAs of 11 sout hwestern Florida manatees killed by collisions with watercraft during and after th e mortality event are presented in Table 2.4. Three of the 11 manatees tested were killed during the mortality event. Brevetoxin was measured in the liver of all three with an average concentration of 26 ng PbTx-3 eq./g and in the stomach contents of one (62 ng PbTx-3 eq./g). Lung, kidney, and urine were all negative. Of the eight manatees that were killed in subsequent months, low levels of brevetoxin were measured only in the stom ach contents of two animals (22 and 29 ng PbTx-3 eq./g). Brevetoxin was not dete cted in any other tissues. Brevetoxin concentrations measured in the liver and stomach contents from watercraft-related manatee deaths were much lower and we re significantly different (p < 0.0001, Mann Whitney test) from levels measured in the 2002 red tide-suspect manatees. Brevetoxin was not detected in any negativ e control samples from eight manatees recovered on Floridas east coast between March and August of 2002.

PAGE 43

Table 2.1. Brevetoxin concentrations (ng PbTx-3 eq./g or ng PbTx -3 eq./mL) measured in tissues from 2002 suspected red tide-rel ated mortalities of manatees. Field IDLiverKidney Lung unspecified Lung caudal Lung cranialUrine Stomach contentsHeart Fat unspecified Fat abdominal Fat epicardial Fat subdermalBrain MSW0239 87219--------33652658---------------MSW0240 2238319------------1,132-----------------------MSW0242 174209--------3182----69---------------MSW0243 1315519--------12488----101---------------MSW0244 224----------------10351-----------------------MSW0245 5831----8673652937------------13 MSW0246 300----------------9403-----------------------MSW0248 11141----911747930----62473810 MSW0249 --------44------------267----------------
PAGE 44

Table 2.2 Spearman rank correlation coefficients (and p values) for brevetoxin concentrations measured in different samp le types of 2002 red tide-suspect manatees. LiverKidneyLung Stomach contentsFatUrine Liver Kidney 0.170 (0.449) Lung 0.350 (0.110)0.182 (0.417) Stomach contents0.080 (0.710)0.525 (0.012)0.513 (0.010) Fat 0.193 (0.414)0.024 (0.920)-0.080 (0.730)-0.006 (0.981) Urine 0.035 (0.877)0.221 (0.349)0.037 (0.878)-0.030 (0.894)-0.259 (0.284) Heart 0.509 (0.114)0.405 (0.216)0.036 (0.915)-0.100 (0.769) 0.360 (0.277)0.324 (0.331) 31

PAGE 45

Table 2.3. Brevetoxin concentrations (ng PbTx -3 eq./g) measured in archived samples from 1996 suspected red tide-rela ted mortalities of manatees. Field IDLiverLung Stomach contents Duodenal contentsBlubber MSW9647---9 ----462---MSW9649-------298------MSW9650304-------329 16 MSW9655----------122---MSW9656------79 120---MSW9664-------101------MSW9667----------532---MSW9673----------148---MSW9671---------48 ---MSW9672----------126---MSW9674----------597---MSW9675-------1042------MSW9676----------364---MSW9681----------186---MSW9684----------178---MSW9698----------1034---MSW9699----------173---MSW96100----------222---MSW96105----------596---MSW96106----------464---MSW96107---4 ----102---MSW96111----------917---MSW96116----------545---MSW96120----------359---MSW96122----------478---MSW96124----------692---MSW96126174-------414 25 MSW96127----------563---MSW96129----------493---MSW96130----------418---MSW96134----------623---MSW96131----------566---MSW96133----------534 23 MSW96135----------511---MSW96136----------365---MSW96137----------383---MSW96142----------849---MSW96144----------342---MSW96145----------431---32

PAGE 46

33 Table 2.3 (Continued). Field IDLiverLung Stomach contents Duodenal contentsBlubber MSW96146----------1,73827 MSW96147----------247---MSW96148----------716---MSW96149----------605---MSW96150----------607---MSW96152----------427 38 MSW96153----------803---MSW96158----------963---MSW96165----------465---MSW96167---
PAGE 47

Table 2.4. Brevetoxin concentrations (ng PbTx-3 eq./g or ng PbTx-3 eq./mL) measured in tissues from southwest Florida manatees killed by watercraft collision. Field IDLiverKidney Lung unspecified Lung cranialUrine Stomach Contents Fat abdominal Fat epicardial Fat subdermal MSW026633
PAGE 48

35 Discussion Exposure to brevetoxins was confirme d by ELISA in all 26 red tide-suspect manatees tested. No evidence of trauma wa s found in these animals, and gross and histopathological findings c onsistent with brevetoxicosis were noted (FWC 2007a). At the time of this investig ation, the only comparative data documenting brevetoxin concentrations in manatee tissues were the re sults of analyses performed in response to the two previous events. Different me thods were employed in each of these investigations. The investig ation of the 1982 event relied heavily on mouse bioassays, which lack both specificity and sensitivit y. In 1996, a combinati on of receptor-binding assays (RBA) and immunohistochemical examination for visual confirmation of brevetoxins was key to connecting the mortalities to the red tide bloom, but the total number of animals tested was small. Samp les from approximately ten manatees were tested by RBA and mouse bioassay (stomach contents only) (FDEP 1996), and preserved tissues from 15 manatees were examined using immunohistochemistry (Bossart et al. 1998). The specific manatees that comprised the latter subset were not identified in the report of the findings (Bossart et al. 1998), so the degree to which these subsets of animals overlapped could not be determined. Comparison of results from the previous events with those obtained during the 2002 investigation is difficult because the ELI SA functions on the basis of structural recognition of the brevetoxin molecule, while the RBA detects binding activity on nerve cells. Results of immunohistochemical analysis while valuable, were not quantitative. By analyzing archived 1996 tissues with the ELISA used in the 2002 investigation, a more meaningful comparison of brevetoxins in the manatees killed in these events can be made. In both 2002 and 1996, stomach contents were composed of seagrass with a noted absence of filter-feeding t unicates (FDEP 1996; FWC 2007a). The concentrations of brevetoxins measured in the stomach cont ents of the 2002 manatees averaged 458 ng PbTx-3 eq./g. Given that most of the mortalities in 2002 occurred after the bloom had dissipated, these results suggest that the lethal route of exposure was ingestion. The lack of tunicates or any other filter-feeding organi sms in the stomach contents and G.I. tracts

PAGE 49

36 of these manatees lead to an environmental i nvestigation into the s ource of the ingested brevetoxins. This investigation is discussed in detail in Chapter 3. Results of the retrospective ELISA analys es of stomach contents from 1996 may be affected by the age of the samples, which were stored at -20C for 7-9 years before being extracted. No information exists on the long-term stability of brevetoxins in biological matrices. Nonetheless, the concentr ations of brevetoxins and brevetoxin-like compounds measured in the 1996 stomach conten ts (average 363 ng PbTx-3 eq./g) were of similar magnitude to those measured in the manatees from the 2002 event, indicating that ingestion of brevetoxins likely contributed significantly to that mortality event as well. The brevetoxin composition of the stomach contents (parent toxins vs. metabolites) in both events is unknown, but the stomach contents of 9 out of 10 manatees tested in 1996 were acutely toxic to mi ce using the NSP mouse bioassay (FDEP 1996). Extrapolating from levels toxic to humans to the body size of a manatee is complicated by their exceptionally slow metabolism (Irvine 1983). However, in 1996 it was hypothesized, based on dose-dependent brevetox in levels and symptoms of illness in humans, that as little as 500 mouse units (e quivalent to 2 mg PbTx-3) could potentially make a manatee ill (FDEP 1996). Since manatees consume 5-10% of their body weight in plant material daily (Reep and Bonde 2006) the average value measured in the 2002 manatee stomach contents could translate to a daily intake of 7-14 mg for a 300 kg manatee. The toxicity of this dose woul d, however, be affected by the forms of brevetoxins ingested and subs equent modification of brevet oxins in the stomach of a manatee. Brevetoxin concentrations in manatee orga n tissues were consistently highest in the liver. These findings are not surprising as studies exposing rats and toadfish to PbTx3, both intravenously and orally, have demonstrated that liver is the principal site of brevetoxin concentration and metabolism (P oli et al. 1990; Kennedy et al. 1992; Cattet and Geraci 1993). Concentrations of brevet oxins measured in two liver samples from 1996 manatees were high as well, and were within the range measured in the 2002 animals.

PAGE 50

37 Brevetoxins were measured at low levels (6 64 ng PbTx-3 eq./g) in the lung of all of the 2002 manat ees tested, despite th e dissipation of the K. brevis bloom at the onset of the event. However, brevetoxins measured in lung tissue are not necessarily the result of exposure via inhalation. Studies exposing ra ts to brevetoxins through ingestion as well as i.v. administration report widespread distri bution to all organs including the lungs (Poli et al. 1990; Cattet and Geraci 1993). Because br evetoxins are lipid soluble and able to cross cell membranes, they can be distributed to any profused tissue, regardless of the route of exposure. It has been shown that br evetoxins remain at measurable levels in blood for days to weeks in rats exposed vi a ingestion, intratracheal instillation, or intraperitoneal administration (Cattet and Geraci 1993; Benson et al. 1999; Radwan et al. 2005). Since concentrations of brevetoxins that induce respiratory eff ects are three orders of magnitude lower than thos e that induce neurological symptoms (Abraham et al. 2005; Baden et al. 2005), it is unclear whether a level that would affect manatees would even be measurable. During a K. brevis bloom, inhalation and ingest ion of brevetoxins are likely to co-occur, and the degree to which each form of exposure contributes to the total body burden detected in tissues is not distinguishable. However, si nce most of the mortalities in 2002 occurred after the bloom had dissipate d, inhalation of brevetoxins was not a likely possibility. Only very low concentrations were meas ured in the archived lung tissues from 1996, but an insufficient number of samples we re available for a meaningful comparison with 2002 data. Brevetoxin concentrations in lung tissue tested in 1996 by RBA were also low, ranging from 4 to 29 ng/g in seven manatees tested, with the notable exception of one much higher level (158 ng/g) measured in the lung tissue of an eighth manatee (FDEP 1996). Brevetoxins were detected in 100% of the kidney tissue and urine samples tested in 2002 animals. Excretion of brevetoxins in experimentally-exposed rats occurs through both the renal system and the bi liary system. Each varies in importance which appears to depend on both the route of toxin administra tion (Poli et al. 1990; Cattet and Geraci 1993) and the forms of brevetoxins to whic h they are exposed (Radwan et al. 2005). Cattet and Geraci (1993) concl uded that rats administered PbTx-3 by ingestion excreted

PAGE 51

38 brevetoxins equally in urine and feces, while Poli et al. (1990) found that elimination occurred primarily through feces in rats e xposed to PbTx-3 by i.v. administration. In a recent study, Radwan et al. (2005) exposed rats to both PbTx-2 and PbTx-3 by i.p. injection and demonstrated that the two toxi ns were processed differently, with PbTx-2 being rapidly metabolized to less pharmacologically active (and presumably less toxic) polar metabolites that were cleared primarily through urine within 24 hours. In this study, only blood and urine samples were analyzed, an d the fate of PbTx-3 was not confirmed. However, given the findings of Poli et al (1990), PbTx-3 was probably eliminated through feces. Bile and feces were not tested in 2002 manatees, but they have been tested in subsequent red tide-related manatee illnesse s and deaths and can c ontain high levels of brevetoxins, indicating biliary excretion is important to manatees as well (Flewelling et al. unpublished data). Excreti on of bile into the duodenum may account for the higher levels of brevetoxins measured in the duode nal contents of the 1996 manatees (mean = 493 ng PbTx-3 eq./g) as compared to the stomach contents (mean = 363 ng PbTx-3 eq./g). The ELISA results reported here demons trate that brevetoxins can accumulate in multiple tissues of manatees. The duration of their persistence in manatees that are exposed to sublethal levels of brevetoxins is unknown. However, the results of the watercraft-killed manatees from southwest Florida suggest that, despite almost annual K. brevis blooms in the area, manatees in southw estern Florida do not carry long-term body burdens of brevetoxin that are measurable by ELISA. Of the 43 manatees recovered between Ma rch 10 and May 1, the cause of death remained undetermined for five. In all, 34 manatee mortalities between January and May 2002 were ultimately attributed to brevetoxicosi s. This event reinforced the concept of Landsberg and Steidinger (1998) that manatees are at high risk in the southwest region of Florida when a K. brevis bloom occurs in February-April, salinities are higher than 27, and manatees are still congregated after overwintering. The mechanisms that lead to manatee d eath as a result of brevetoxicosis remain unresolved. Given that experi mentation on these endangered animals is prohibited, what constitutes a lethal dose of brevetoxin for a manatee may never be definitively known but

PAGE 52

39 may not be as relevant as dosages that induce illness. In these animals, physiologically sub-lethal effects of brev etoxicosis may have lethal consequences due to their vulnerability as air breathers in an aquatic e nvironment. In fact, of the 37 manatees that have been rescued during red tides (up thr ough 2007), only three have died: one at the scene of the rescue, one in tr ansit, and one at the rehabilitation facility (FWC-FWRI MMPL database). Although the cause is no longe r questioned, efforts to implicate brevetoxins as the responsible agent in the previous red tide-su spected manatee mortality events in 1982 and 1996 were limited by the methods of analysis th at were available at the time (especially in 1982), and by the need to rely on scientists outsid e of the investigating state agency to perform the analyses. The investigation of the 2002 mortality event benefited from the recent development of a rapid, sensitive, and specific ELISA for brevetoxins (Naar et al. 2002) and the implementation of this method by scientists within the Florida Fish and Wildlife Conservation Commissions Fish and Wildlife Research Institute. Although the production of antibodies agains t brevetoxins in the 1980s (Baden et al. 1984) lead to the report of a direct ELISA method for brevetoxins in 1991 (Trainer and Baden 1991), this method did not become widely used among red tide researchers. Despite my best efforts to perform this ELI SA as described by Trainer and Baden, I was unsuccessful in obtaining a functional assay (Flewelling, unpublished data). Naars reformatted ELISA coupled a protein-conjugated brevetoxin with a seri es of amplification steps in an indirect competitive format, pr oducing a vastly improved assay that could be applied to complex biological matrices with lit tle interference. Since its development, this ELISA has been shown to be an effectiv e tool to assess brev etoxin and brevetoxin metabolite contamination in a wide range of biological and environmental matrices (Naar et al. 2002; Dickey et al. 2004; Plakas et al. 2004; Cheng et al. 2005; Pierce et al. 2005; Pierce et al. 2006). Although the affinity of th e polyclonal antibodies used in this assay for brevetoxin metabolites is unknown, good correlation has been observed between results from ELISA and HPLC-MS analyses of shellfish tissues (Dickey et al. 2004; Plakas et al. 2004; Pierce et al. 2006), indicating a similar a ffinity of these antibodies for the brevetoxins and their derivatives.

PAGE 53

40 The inability to distinguish between diffe rent forms of brevetoxins that share a similar backbone structure is the main dr awback to this method. ELISA results of samples that are likely to c ontain a mixture of brevetoxins and/or brevetoxin metabolites do not accurately estimate toxicity. However, this property of the assay is also its greatest strength in an investigation such as this. The metabolism of brevetoxins in mammals is not yet well understood. The ability to detect th e entire suite of the most abundant type of brevetoxins as well as their metabolites at lo w levels enables confirmation of brevetoxin exposure in mammals that could otherwise go undetected. The availa bility of this new tool to FWC-FWRI enabled brevetoxin analysis of tissue samples from the majority of the manatees that died, and a more thorough investigation into the contribution of K. brevis and brevetoxins than was possibl e in previous mortality events.

PAGE 54

41 Chapter 3. Brevetoxin Accumulation and Persistence in Seagrass Introduction In mid-March 2002 manatee deaths greatly increased along the coasts of Sarasota, Charlotte, Lee, and Collier counties and continued throughout April. A dense bloom of Karenia brevis (>106 cells/L) had been present in s outhwest Florida in early February. However, by mid-March the bloom had largel y dissipated with lo wer populations (<105 cells/mL) observed immediately prior to the onset of the mortalities. Analyses of tissues and stomach contents from 26 recovered car casses revealed high levels of brevetoxin (Chapter 2). On the basis of the brevetoxi n analyses and gross necropsy findings that were non-specific but consiste nt with findings in earlier red tide-related manatee mortalities, a total of 34 ma natee deaths were attributed to brevetoxicosis between January and May 2002. The absence of a K. brevis bloom together with the high levels of brevetoxin measured in the stomach contents of recove red carcasses indicated th at ingestion was the probable lethal route of exposure. Manatees are herbivores, relying on aquatic plants as their sole food source (Reynolds and Powell 2002; Reep and Bonde 2006). During past mortality events, incidental ingestion of brevetoxins while feeding was hypothesized as a possible route of exposure for manatees. In the 1982 manatee mo rtality event, an abundance of filter-feeding tuni cates found in the gastrointe stinal (GI) tracts of the manatees and attached to seagrasses in th e area where the mortality was occurring were suspected to be brevetoxin vectors (O'Sh ea et al. 1991; Landsberg and Steidinger 1998). In 1996, investigators hypothesized that the toxin detected in the stomach contents of 10 manatees tested by mouse bioassay was from K. brevis bloom water that was incidentally ingested while feeding (FDEP 1996).

PAGE 55

42 No tunicates or any other filter-feeding organisms were noted in the stomach contents or GI tracts of any manatees exam ined during the 2002 event, and the absence of a K. brevis bloom during the period when most of the manatees died made ingestion of water containing K. brevis cells an unlikely mechanism of intoxication. The field collections and analyses presented in this chapter were conducted to identify the source of brevetoxins detected in the manatee stomach contents and to monitor the persistence of brevetoxins in the manatees environment. Environmental samples (seawater, sediments, and seagrasses) were collected by FWC-FWRI staff from seagrass meadows in southwest Florida throughout and after the 2002 mortality event. Since this survey was not initiated until after the K. brevis bloom had dissipated, additional samples were collected during and after a K. brevis bloom in 2005. Methods Based on the location of mana tee carcasses recovered earl y in the mortality event, four sites (sites 1, 2, and 3 in Charlotte County and Forked Creek in Sarasota County) were selected to investigate if residual brevetoxin remained in the system and to identify the source of brevetoxin in the manatee stom ach contents (Fig. 3.1). Seawater, sediment and seagrass samples were collected by FW C-FWRI HAB staff with the assistance of FWC-FWRI Fisheries Independent Monitoring staff. Collections were made bimonthly from March 28 through August 29, 2002 at site s 1, 2, and 3, and through June 20, 2002 at Forked Creek. A fifth site (Bokeelia, Fi g, 3.1) was sampled once (April 25) for comparison because there had been high cell densities of K. brevis at that site just prior to the mortality event, and, although manatees were observed congregat ing and feeding at that site, carcasses were not found there. Similar collections were made by FWC-FW RI HAB staff at two sites in Placida Harbor in the same area (PH1 and PH2, Fi g. 3.1) bimonthly between February 9 and December 27, 2005. A mechanical problem with the boat on August 23 prevented sample collection at site PH2. Seagra ss collected during non-bloom pe riods at two sites in the Tampa Bay area (Ft. DeSoto and North S hore Park) served as matrix controls.

PAGE 56

43 Seawater was collected by hand from a depth of 0.5 m in 1-L polyethylene bottles that were pre-rinsed with ambient water. Sediment samples were collected by scraping a 250-mL glass jar along the top 2-3 cm of se diment. Ambient seawater was added to sediment jars that were 3/4 filled with sediment. Sediment samples were collected in duplicate at all sites between March 28 and June 6, 2002. Both water and sediment samples were held at ambient temperature during transport. Seagrass was collected by hand from water that was 1-1.5m in depth. The species of seag rass collected were Thalassia testudinum at sites 1, 2, and 3 in 2002 and at PH1 and PH2 in 2005; Halodule wrightii at Forked Creek in 2002; and both T. testudinum and Syringodium filiforme at Bokeelia in 2002. Individual plant sections (one shoot and a s ection of attached rhizome, 10 per site) were stored in se parate 1-gallon plastic bags wi th a small amount of ambient seawater, and placed in a cooler on top of ne wspaper-insulated ice packs for transport. In 2002, salinity was measured using a hand-held refractometer. For 2005 collections, water quality parameters (salin ity, conductivity, temperat ure, pH, dissolved oxygen) were measured using a YSI QS 600 seri es sonde. For both years, enumeration of K. brevis cells in seawater was performed by FWC-FWRI HAB taxonomists Jennifer Wolny and Ernest Truby by examining unpreserved aliquots of s eawater under a Leica stereomicroscope (Steidinger and Melton Pent a 1999). Brevetoxins were extracted from seawater within 24 h of collection by passi ng 0.5 L through an Empore Si-C18 disc that had been preconditioned with 20 mL of me thanol and then 20 mL of DI water. Brevetoxins were eluted from the discs into clean glass test tubes using 2 10-mL washes of methanol. The combined eluents were evapor ated to dryness and re-dissolved in 2 mL of methanol. Seawater extracts were stored at -20C until analyzed. From each site, five of the ten plant sections were combined into a single composite sample (referred to as whole seagra ss in this discussion). On May 23 and June 6, 2002 at sites 2 and 3 an additional ten pl ant sections were collected, and whole seagrass was extracted and analyzed in triplicat e. Using a thin stainless steel spatula, the leaves from the remaining five plant sections were scraped for all epiphytes (organisms living on the surface of seagrass leaves and shea ths) and detritus, and the scrapings were processed as a separate sample. For several dates in 2002, and for all collections in 2005,

PAGE 57

Sarasota CharlotteCharlotte Harbor 05 Kilometers 5 05 Kilometers 5 2002 Sites 2005 Sites 2002 Sites 2005 SitesGulf of Mexico Area shown PH1 PH2 Site 3 Site 2 Site 1 Forked Creek Bokeelia Sarasota CharlotteCharlotte Harbor 05 Kilometers 5 05 Kilometers 5 2002 Sites 2005 Sites 2002 Sites 2005 SitesGulf of Mexico Area shown PH1 PH2 Site 3 Site 2 Site 1 Forked Creek Bokeelia Figure 3.1 Sites of environmental sample colle ctions in southwest Florida in 2002 (Sites 1, 2, 3, and Forked Creek) and 2005 (PH1 and PH2). 44

PAGE 58

45 the scraped seagrass sections were rinsed vi gorously with tap water, and the leaves and sheaths were separated from the rhizomes a nd roots. Epiphytes and detritus could not be completely removed from the more the delicate blades of Halodule wrightii collected at Forked Creek, and so only whole grass, rhizomes, and scrapings were analyzed from that site. Composite whole seagrass samples, rinsed seagrass components, leaf scrapings, and sediments were dried overnight in an oven at 45C, weighed, and extracted for brevetoxins. On May 11, 2002 a subset of 10 lim pets collected from the seagrass leaves at site 3 were also extracted and analyzed separately. Dried seagrass and sediment samples were extracted twice using acetone (1:3 w:v). The two acetone extracts were combined and evaporated to dryness. The residue was re-dissolved in 80% aqueous methanol and partitioned twi ce with hexane. The hexane fraction was discarded. Extracts were then evaporated to dryness and re-dissolved in 100% methanol at 0.5-1 g dry weight per mL. Extracts were stored at -20C until analyzed. Brevetoxins and brevetoxin-like com pounds possessing a PbTx-2-type backbone were quantified in all samples using a competitive ELISA performed according to Naar et al. (2002). For samples collected in 2002, the ELISA protocol followe d that outlined in Chapter 2. Two modifications to the ELISA were adopted in 2003. Biotinylated rabbit anti-goat secondary antibody and HRP-linked streptavidin were replaced by a single HRP-linked rabbit anti-goat secondary anti body. Also the HRP substrate, OPD, was replaced by another HRP substrate, 3,3 ,5,5-tetramethylbenzidine (TMB). These changes reduced assay time and improved reproducibility between duplicate wells, but did not alter the sens itivity or specificity of the assay. Samples collected during 2005 were analyzed using the revi sed protocol. Results are expressed in terms of ng PbTx-3 eq./g (dry weight) of seagrass and sediment and g PbTx-3 eq./L of seawater. The lower limit of detection of the ELISA as performed here was approximately 10 ng/g in seagrass, 2.5 ng/g in sediment, and 0.01 g/L in seawater. The total brevetoxin present in the leaf scrapings and the cleaned leaves and sheaths and the degree to which each of th ese components contri buted to the total brevetoxin present on the shoot were calculated using measured brevetoxin

PAGE 59

46 concentrations and total dry weights. All statistics were performed on log-transformed brevetoxin data using GraphPad Prism vers ion 5.01 for Windows, GraphPad Software (San Diego California USA, www.graphpad.com ). For data sets that included assay results that were below the li mit of detection, a value of ha lf the limit of detection was assigned and non-parametric analyses were used. Results 2002 Collections Throughout the sampling period in 2002, sa linities ranged between 30 and 36 at sites 1, 2, and 3, and between 31 and 34 at Forked Creek. Karenia brevis was absent at Forked Creek on all sampling dates and was eith er absent or present only at very low levels (max. 2,670 cells/L) at sites 1, 2, and 3 from March 28 to August 15 (Fig. 3.2). On the last sampling date (August 29), K. brevis was not observed at site 2 but was present at low levels at sites 1 (7,670 cells/L) and 3 (13,700 cells/L). Brevetoxin concentrations in seawater ra nged from below the limit of detection (
PAGE 60

0 2 4 6 8 10 12 14K. brevis (cells/mL) and seawater (g PbTx-3 eq./L) 0 200 400 600 800 1,000 1,200 1,400Seag r ass / sediment (ng PbTx-3 eq. / g d r y wt.) 3/28/02 4/11/02 4/25/02 5/9/02 5/23/02 6/6/02 6/20/02 7/11/02 8/1/02 8/15/02 8/29/02 3/28/02 4/11/02 4/25/02 5/9/02 5/23/02 6/6/02 6/20/02 7/11/02 8/1/02 8/15/02 8/29/02 3/28/02 4/11/02 4/25/02 5/9/02 5/23/02 6/6/02 6/20/02 7/11/02 8/1/02 8/15/02 8/29/02 3/28/02 4/11/02 4/25/02 5/9/02 5/23/02 6/6/02 6/20/02 0 2 4 6 8 10 12 14K. brevis (cells/mL) and seawater (g PbTx-3 eq./L) 0 200 400 600 800 1,000 1,200 1,400Seag r ass / sediment (ng PbTx-3 eq. / g d r y wt.) 3/28/02 4/11/02 4/25/02 5/9/02 5/23/02 6/6/02 6/20/02 7/11/02 8/1/02 8/15/02 8/29/02 3/28/02 4/11/02 4/25/02 5/9/02 5/23/02 6/6/02 6/20/02 7/11/02 8/1/02 8/15/02 8/29/02 3/28/02 4/11/02 4/25/02 5/9/02 5/23/02 6/6/02 6/20/02 7/11/02 8/1/02 8/15/02 8/29/02 3/28/02 4/11/02 4/25/02 5/9/02 5/23/02 6/6/02 6/20/02 0 2 4 6 8 10 12 14K. brevis (cells/mL) and seawater (g PbTx-3 eq./L) 0 200 400 600 800 1,000 1,200 1,400Seag r ass / sediment (ng PbTx-3 eq. / g d r y wt.) 3/28/02 4/11/02 4/25/02 5/9/02 5/23/02 6/6/02 6/20/02 7/11/02 8/1/02 8/15/02 8/29/02 3/28/02 4/11/02 4/25/02 5/9/02 5/23/02 6/6/02 6/20/02 7/11/02 8/1/02 8/15/02 8/29/02 3/28/02 4/11/02 4/25/02 5/9/02 5/23/02 6/6/02 6/20/02 7/11/02 8/1/02 8/15/02 8/29/02 3/28/02 4/11/02 4/25/02 5/9/02 5/23/02 6/6/02 6/20/02 0 2 4 6 8 10 12 14K. brevis (cells/mL) and seawater (g PbTx-3 eq./L) 0 200 400 600 800 1,000 1,200 1,400Seag r ass / sediment (ng PbTx-3 eq. / g d r y wt.) 3/28/02 4/11/02 4/25/02 5/9/02 5/23/02 6/6/02 6/20/02 3/28/02 4/11/02 4/25/02 5/9/02 5/23/02 6/6/02 6/20/02 3/28/02 4/11/02 4/25/02 5/9/02 5/23/02 6/6/02 6/20/02 3/28/02 4/11/02 4/25/02 5/9/02 5/23/02 6/6/02 6/20/02 Seagrass Sediment K. brevis Seawater Seagrass Sediment K. brevis Seawater AB Site 2 Site 1 CSeagrass Sediment K. brevis Seawater Seagrass Sediment K. brevis Seawater D Site 3 Forked Creek Figure 3.2 Brevetoxin concentra tions measured by ELISA in composite seagrass samples, sediments, and seawater; and K. brevis cell densities observed in samples collected at: A) site 1, B) site 2, C) site 3, and D) Forked Creek. Error bars on sediments = standard deviation of duplicate collections. Error bars on seagrass (5/23/02 and 6/6/02, sites 2 and 3 only) = standard deviation of triplicate seagrass collections. 47

PAGE 61

48 (73 ng PbTx-3 eq./g) at site 1 (Fig. 3.2). In general, concentr ations initially decreased and then increased slightly at sites 1, 2, and 3 in May. Brevetoxin concentrations increased greatly at these sites in June when maximum concentr ations of 1,138, 1,263, and 562 ng PbTx-3 eq./g were measured at sites 1, 2, and 3, respectively (Fig. 3.2A-C). At Forked Creek, concentrations dropped after the first seagrass collect ion on April 11 and increased slightly on June 20, the last sa mpling date at that site (Fig. 3.2D). Brevetoxins were detected in all seagra ss components, but concentrations were typically higher by one or two orders of magnitude in the scra pings than in the leaves and sheaths or the rhizomes (Fig. 3.3). Brevetoxi n concentrations range d from
PAGE 62

49 0 500 1,000 1,500 2,000 2,500 3,000Seag r ass (ng PbTx-3 eq. / g d r y wt.) 4/11/02 4/25/02 5/9/02 5/23/02 6/6/02 6/20/02 7/11/02 8/1/02 8/15/02 8/29/02 0 2 4 6 8 K bre v is (cells / mL) 0 500 1,000 1,500 2,000 2,500 3,000 3,500Seag r ass (ng PbTx-3 eq. / g d r y wt.) 4/11/02 4/25/02 5/9/02 5/23/02 6/6/02 6/20/02 7/11/02 8/1/02 8/15/02 8/29/02 0 1 2 3 K bre v is (cells / mL) 0 500 1,000 1,500 2,000 2,500Seag r ass (ng PbTx-3 eq. / g d r y wt.) 4/11/02 4/25/02 5/9/02 5/23/02 6/6/02 6/20/02 7/11/02 8/1/02 8/15/02 8/29/02 0 4 8 12 16 K bre v is (cells / mL) AB Site 1 C D Site 3 Forked CreekScrapingsS i t e 2 Leaves/sheaths Rhizomes/roots K. brevis (cells/mL) Scrapings Leaves/sheaths Rhizomes/roots K. brevis (cells/mL) Scrapings Leaves/sheaths Rhizomes/roots K. brevis (cells/mL) 0 100 200 300 400Seag r ass (ng PbTx-3 eq./g d r y wt.) 4/11/2002 4/25/2002 5/9/2002 5/23/2002 6/6/2002 6/20/2002 0 1 2 3K. bre v is (cells/mL) Scrapings Rhizomes/roots K. brevis (cells/mL) Figure 3.3 Brevetoxin concentra tions measured by ELISA in seagrass components (leaf scrapings, leaves and sheaths, and rhizomes and roots) and K. brevis cell densities observed in samples collected at: A) site 1, B) site 2, C) site 3, and D) Forked Creek.

PAGE 63

Site 1 0 200 400 600 800 1,000 1,200Seag r ass (ng PbTx-3 eq. / g d r y wt.) 4/11/02 4/25/02 5/9/02 5/23/02 6/6/02 BScrapings Leaves/sheaths 0 200 400 600 800 1,000 1,200Seag r ass (ng PbTx-3 eq. / g d r y wt.) 4/11/02 4/25/02 5/9/02 5/23/02 6/6/02 AScrapings Leaves/sheaths 0 200 400 600 800 1,000 1,200Seag r ass (ng PbTx-3 eq. / g d r y wt.) 4/11/02 4/25/02 5/9/02 5/23/02 6/6/02 Scrapings Leaves/sheaths C Site 2 Site 3 Figure 3.4 Brevetoxin concentra tions measured by ELISA in leaf scrapings and leaves and sheaths as a fraction of the total brevetoxin measured in th e shoot at: A) site 1, B) site 2, and C) site 3. 50

PAGE 64

51 2005 Collections As in 2002, salinity of samples collecte d in 2005 ranged from 30 to 36. Sample collections began on February 9, during a K. brevis bloom. On that day, K. brevis was present at low levels at both PH1 (1.7 x 104 cells/L) and PH2 (4.2 x 104 cells/L). Two weeks later the K. brevis cell density had increased to high levels (1.8-2.8 x 106 cells/L), and medium cell concentrations were pres ent at both sites on March 7 (1.6-3.3 x 105 cells/L). After March 7, no K. brevis cells were observed at PH1 until August. At PH2, K. brevis was not observed (except for b ackground concentrations of <103 cells/L on June 1) between March 22 and June 14. Low levels were noted on June 28, but K. brevis was not present in samples collected in July. A sma ller bloom affected Placi da Harbor beginning in late August, and K. brevis cell concentrations reached medium levels (2.7-6.0 x 105 cells/L) on September 6. This bloom quickly moved out of the area, and no K. brevis was seen in samples collected at either site from October through December (Fig. 3.5). Brevetoxin concentrations in seawater were highest (83-104 g PbTx-3 eq./L) when K. brevis was present at high concentrations. During the less intense bloom in early September, seawater concentrations reached 13-17 g PbTx-3 eq./L. In the absence of K. brevis brevetoxin concentrations in seawater we re typically < 1 g PbTx-3 eq./L. The only exceptions to this were in the week s immediately following the bloom when no K. brevis cells were seen yet concentrations rangi ng from 1 to 4 g PbTx-3 eq./L were measured (Fig. 3.5). Similarly, brevetoxin concentr ations in sediments at bot h sites were also highest (109-201 ng PbTx-3 eq./g) when K. brevis was present at high concentrations in February. In the weeks directly following the bloom, sediment brevetoxin concentrations were still elevated, ranging from 60-170 ng PbTx -3 eq./g. Otherwise, in the absence of K. brevis brevetoxin concentrations in sediments ranged from
PAGE 65

0 100 200 300 400 500 600 1,500 1,750 2,000K. brevis (cells/mL) and seawater (g PbTx-3 eq./L) 0 200 400 600 800 1,000 1,200 1,400 1,600 1,800 2,000 2,200 2,400Seagrass/sediment (ng PbTx-3 eq./g dry wt.) 2/9/05 2/23/05 3/7/05 3/22/05 4/4/05 4/19/05 5/3/05 5/17/05 6/1/05 6/14/05 6/28/05 7/13/05 7/26/05 8/9/05 8/23/05 9/6/05 9/20/05 10/4/05 10/18/05 11/1/05 11/15/05 11/30/05 12/14/05 12/27/05 2/9/05 2/23/05 3/7/05 3/22/05 4/4/05 4/19/05 5/3/05 5/17/05 6/1/05 6/14/05 6/28/05 7/13/05 7/26/05 8/9/05 8/23/05 9/6/05 9/20/05 10/4/05 10/18/05 11/1/05 11/15/05 11/30/05 12/14/05 12/27/05 0 100 200 300 2,500 3,000K. brevis (cells/mL) and seawater (g PbTx-3 eq./L) 0 200 400 600 800 1,000 1,200 1,400 1,600 1,800 2,000 2,200 2,400Seagrass/sediment (ng PbTx-3 eq./g dry wt.) 2/9/05 2/23/05 3/7/05 3/22/05 4/4/05 4/19/05 5/3/05 5/17/05 6/1/05 6/14/05 6/28/05 7/13/05 7/26/05 8/9/05 8/23/05 9/6/05 9/20/05 10/4/05 10/18/05 11/1/05 11/15/05 11/30/05 12/14/05 12/27/05 2/9/05 2/23/05 3/7/05 3/22/05 4/4/05 4/19/05 5/3/05 5/17/05 6/1/05 6/14/05 6/28/05 7/13/05 7/26/05 8/9/05 8/23/05 9/6/05 9/20/05 10/4/05 10/18/05 11/1/05 11/15/05 11/30/05 12/14/05 12/27/05 Seagrass Sediment K. brevis Seawater Seagrass Sediment K. brevis Seawater A PH1 B PH2 Figure 3.5 Brevetoxin concentra tions measured by ELISA in composite seagrass samples, sediments, and seawater; and K. brevis cell densities observed in samples collected at: A) PH1 and B) PH2. 52

PAGE 66

53 decreased to <100 ng/g by mid-April. On ly a small increase in the brevetoxin concentration measured in whole seagrass was seen at PH1 with the return of K. brevis in August and September, while a more substantial increase (up to 700 ng PbTx-3 eq./g) was observed at site PH2 (Fig. 3.5). During the peak of the K. brevis bloom in Placida Harbor in early 2005 concentrations in all seagrass components su rpassed those measured in 2002. As in 2002, brevetoxin concentrations were highest in the leaf scrapings collected from the Thalassia blades. Maximum brevetoxin concentrations m easured were 9,536 ng PbTx-3 eq./g in the scrapings, 5,327 ng PbTx-3 eq./g in the clean ed blades and sheaths, and 307 ng PbTx-3 eq./g in rhizomes (Fig. 3.6). Unlike 2002, no limpets were noted on the leaves. The contribution of the scrapings to the total toxi n in the shoots was more variable than in 2002, ranging from 41% to 100% (Fig. 3.7). Spearman rank correlation analysis was used to assess the correlation of brevetoxin concentrations in seagrass components with each other and with the brevetoxin concentration in seawater and sediments (Table 3.1). Weak but significant (p < 0.05) correlations were observed in all case s with the exception of the rhizomes and sediments, which did not significantly correlate. No brevetoxin was detected in seagrass matrix control samples collected at Ft. DeSoto Park in January 2003 and North Shor e Park in April and May 2004 during nonbloom periods.

PAGE 67

0 2,000 4,000 6,000 8,000 10,000Seag r ass (ng PbTx-3 eq./g d r y wt.) 2/9/05 2/23/05 3/7/05 3/22/05 4/4/05 200 400 600 4/19/05 5/3/05 5/17/05 6/1/05 6/14/05 6/28/05 7/13/05 7/26/05 8/9/05 8/23/05 9/6/05 9/20/05 10/4/05 10/18/05 11/1/05 11/15/05 11/30/05 12/14/05 12/27/05 0 1,000 2,000 3,000K. brevis (cells/mL) 1,000 2,000 3,000 4/19/05 5/3/05 5/17/05 6/1/05 6/14/05 6/28/05 7/13/05 7/26/05 8/9/05 8/23/05 9/6/05 9/20/05 10/4/05 10/18/05 11/1/05 11/15/05 11/30/05 12/14/05 12/27/05 0 400 800 1,200 1,600 2,000K. bre v i s (cells/mL) 0 4,000 8,000 12,000 16,000Seag r ass (ng PbTx-3 eq./g d r y wt.) 2/9/05 2/23/05 3/7/05 3/22/05 4/4/05 Scrapings Leaves/sheaths Rhizomes/roots K. brevis Scrapings Leaves/sheaths Rhizomes/roots K. brevis B PH2 A PH1 Figure 3.6 Brevetoxin concentra tions measured by ELISA in seagrass components (leaf scrapings, leaves and sheaths, and rhizomes and roots) and K. brevis cell densities observed in samples collected at: A) PH1 and B) PH2. 54

PAGE 68

0 2,000 4,000 6,000 8,000Seag r ass (ng PbTx-3 eq./g d r y wt.) 2/9/2005 2/23/2005 3/7/2005 3/22/2005 4/4/2005 40 80 120 160 200 4/19/2005 5/3/2005 5/17/2005 6/1/2005 6/14/2005 6/28/2005 7/13/2005 7/26/2005 8/9/2005 8/23/2005 9/6/2005 9/20/2005 10/4/2005 10/18/2005 11/1/2005 11/15/2005 11/30/2005 12/14/2005 12/27/2005 0 2,000 4,000 6,000 8,000 10,000Seagrass (ng PbTx-3 eq./g dry wt.) 2/9/05 2/23/05 3/7/05 3/22/05 4/4/05 200 400 600 4/19/05 5/3/05 5/17/05 6/1/05 6/14/05 6/28/05 7/13/05 7/26/05 8/9/05 8/23/05 9/6/05 9/20/05 10/4/05 10/18/05 11/1/05 11/15/05 11/30/05 12/14/05 12/27/05 B PH2 A PH1Scrapings Leaves/sheaths Scrapings Leaves/sheaths Figure 3.7 Brevetoxin concentra tions measured by ELISA in leaf scrapings and leaves and sheaths as a fraction of the total brevetoxin measured in the shoot at: A) PH1 and B) PH2. 55

PAGE 69

Table 3.1 Spearman rank correlation coe fficients (and p-values) for brevetoxin concentrations in seawater, sediment, and seagrass components (ns = not significant). SeawaterSediment Leaves and Sheaths Roots and Rhizomes Seawater Sediment 0.58 (p < 0.0001) Leaves and Sheaths 0.63 (p < 0.0001)0.39 (p < 0.01) Roots and Rhizomes 0.34 (p < 0.02)0.09 (ns)0.44 (p < 0.005) Leaf Scrapings 0.76 (p < 0.0001)0.48 (p < 0.0006)0.66 (p < 0.0001)0.33 (p < 0.05) Discussion From late March through August 2002, water, sediment and seagrass samples were collected repeatedly at four sites in southwest Florida to identify the source of brevetoxins detected in the manatee stomach contents and to monito r the persistence of brevetoxins in the manatees environment. Karenia brevis cells were either absent or present at very low levels ( 2,670 cells/L) until the final da te of the sampling period, when low levels of cells were observed at tw o sites. Brevetoxin concentrations in the water column were typically below 1 g PbTx -3 eq./L, and toxin in sediments averaged 12-17 ng PbTx-3 eq./g. These results suggest that incidental ingestion of residual brevetoxin in seawater or in sediments was not the mechanism of exposure for the manatees. However, composite samples of Thalassia testudinum and Halodule wrightii tested positive for brevetoxins at high leve ls, ranging from 73-501 ng PbTx-3 eq./g in initial samples. No K. brevis or brevis -like organisms were obs erved on the seagrass (A. Haywood, personal communication), nor were any tunicates or la rge filter-feeding organisms found on the seagrass or in any part of the digestive tracts of any of the manatees necropsied. This investigation is th e first to report brev etoxin accumulation in association with seagrasses, and to impli cate ingestion of brevetoxin-contaminated seagrasses as a route of brev etoxin exposure for manatees. 56

PAGE 70

57 Six species of seagrass occur al ong the west coast of Florida: Thalassia testudinum Halodule wrightii Syringodium filiforme Ruppia maritime Halophila engelmanii and Halophila decipiens (Zieman and Zieman 1989). Of these, T. testudinum Halodule wrightii and S. filiforme are the dominant species. The primary species collected in this study was T. testudinum with the exception of Forked Creek in 2002 (H. wrightii ). The maximum brevetoxin concentra tions measured in composite whole seagrass samples were higher in 2005 (up to 2,249 ng PbTx-3 eq./g) than the maximum levels measured in 2002 (up to 1,263 ng PbTx-3 eq./g). However, collections in 2002 began several weeks afte r the dissipation of a K. brevis bloom (Chapter 2, Fig. 2.2), while the highest brevetoxin concentrations measured in seagrass in 2005 were in samples collected during a K. brevis bloom (Fig. 3.5). During nonK. brevis bloom periods, brevetoxin concentrations were generally hi gher in seagrasses colle cted in 2002 than in samples collected in 2005. On March 7, 2005, the date on which the maximum concentrations in scrapings and the cleaned leaves and sheaths were measured, the brevetoxin measured in the composite whole seagrass collected from PH1 was comparatively very low (165 ng PbTx-3 eq./g). I believe this reflects analytical error. Unfortunately, no sample remains for re-extraction or analysis. Possible mechanisms of accumulation of brevetoxins in association with seagrasses include uptake by filter-feeding ep iphytes, concentration on the epiphytes and detritus through passive adsorption, concentratio n in epifauna that feed on brevetoxincontaminated detritus, passive adsorption ont o the surface of the le aves or partitioning into the wax cuticle, active uptake by the seagrass through the roots and rhizomes or across the leaves, or incorporat ion into the rhizomes through benthic bacterial processes. To determine where the brevetoxins were located, seagrass samples were separated into three fractions. Leaves were scraped, and 1) scrapings, 2) rinsed l eaves and sheaths, and 3) rinsed rhizomes and roots were each ex tracted and analyzed separately to obtain information about the mechanism(s) of accumulation.

PAGE 71

58 Concentration of Brevetoxins in the Epiphytes Highest concentrations were consisten tly measured in the leaf scrapings, accounting for 74-99% of the total toxin measured in the shoot in 2002. In 2005, brevetoxins persisted in the sc rapings at high concentrations for several weeks after the bloom ended. The broad strap-like leaves of T. testudinum provide substrate for epiphytes and epifauna and are effec tive traps for detritus. The T. testudinum seagrass meadows sampled in this study were sheltered and hosted a rich epiphytic community. Although the narrow-leafed H. wrightii at the Forked Creek site ini tially contained high brevetoxin concentrations (408 ng PbTx-3 eq./g), the le vels decreased more quickly than at the T. testudinum sites. Syringodium filiforme was only collected and tested once as it was only encountered at one site that was not revisited, and so no conc lusions can be drawn in this study about the potential for brevetoxin accumula tion in association with this species. However, S. filiforme has narrow tubular leaves and is frequently found lining natural channels with high velocity water flow (Zie man and Zieman 1989). This species is less likely to be substantially c overed by detritus and epiphytized and therefore probably less likely to accumulate high levels of brevetoxin. On some dates in 2002, limpets were abundant in the epiphytic community at sites 1, 2, and 3. Samples of these limpets also te sted positive for brevetoxin, but whether the brevetoxin was recovered from the surface of the organisms or extracted from the tissues could not be determined. Brevetoxin concentrat ions persisted at higher levels at these sites than at Forked Creek where no limpets were observed, and for a longer period in the absence of K. brevis than at sites sampled in 2005. In 2002 at sites 2 and 3, brevetoxins on seagrass persisted at levels greater than 100 ng PbTx-3 eq./g for two months, increasing between collections on June 6 and June 20. This increase was likely in response to an intrusion of K. brevis into Placida Harbor, but water collected at these sites on June 20 and throughout southwest Florida between June 6 and June 20 contained only very low or no K. brevis cells. In contrast, in 2005, brevetox in concentrations at both sites decreased to less than 100 ng PbTx -3 eq./g within a month after K. brevis cells were no longer present. Although the composition and abundance of epiphytic species was not determined in this study, this observation sugge sts that the epiphytic species present may

PAGE 72

59 influence the duration of brevet oxin persistence in associa tion with seagrass. Given the large extent to which the toxin was often lo calized in the scrapings, future research should examine whether brevetoxins are diffe rentially partitioned in the detritus or among the different types of epiphytes present. Concentration of Brevetoxins in the Leaves and Sheaths Brevetoxin concentrations in the cleaned leaves and sheaths were typically much lower than those observed in the scrapings However, for a brief period in 2005 when K. brevis was present at high levels (>106 cells/L), brevetoxin concentrations in the leaves and sheaths reached 5,327 ng PbTx-3 eq./g, a nd contributed up to half of the toxin present in the shoot (Fig. 3.7A). These high conc entrations in the leaves and sheaths were short-lived, decreasing quickly as the bloom ended. Whether this toxin was remnant brevetoxin not removed from the surface of the leaf in the cleaning process or was actually taken up by the seagrass a nd located in the seagrass flui ds or tissue is not easily determined. Harsher methods of cleaning the le aves using detergents or organic solvents may compromise the integrity of th e seagrass tissue. Application of immunohistochemical techniques to visualize the toxin in sec tions of the plant tissue may help address this question. Concentration of Brevetoxins in the Rhizomes and Roots Brevetoxins were measured in rhizomes and roots from below the sediment surface at concentrations up to 300 ng PbTx-3 eq./g. The rhizomes and roots undoubtedly contributed to the brevetoxin measured in composite whole seagrass samples, but rhizomes were omitted from calculations to assess the contribution of the scrapings to the total measured because the amount of material collected varied. The part of the shoot that includes only the leaves and the sheath are easil y identified discrete units. The horizontal rhizomes analyzed here are shared between shoots, and the quantity of rhizome segment collected was not standardized. Brevetoxin concentrations in rhizomes were not significantly correlated to concentrations m easured in sediments (Table 3.1). However they were weakly, but significantly, correlated to brevetoxin concentrations in seawater

PAGE 73

60 (rs = 0.344 p = 0.019) and to brevetoxin in the cleaned leaves and sheaths (rs = 0.440, p = 0.002). This may suggest that brevetoxin wa s transported to the rhizomes from the leaves or taken up from sediment pore waters (not analyzed). The presence of brevetoxins in all three fractions suggests that multiple mechanisms of accumulation may be occurring. Passive adsorption onto the surface of epiphytes, detritus, and the grass itself is highly probable. The hydrophobic nature of brevetoxins leads to their enrichment in mari ne aerosol particles and sea foam (Pierce et al. 1990), and is the basis of current rese arch using polymer-based passive sampling devices for brevetoxins and other similar lipoph ilic algal toxins (MacKenzie et al. 2004; Shea et al. 2006). Such adsorptive accumulation may be expected to result in persistence of toxins for a period of time after a bloom has ended. LC-MS analyses of a subset of whole grass and scrapings from 2002 collectio ns performed by Michael Henry at Mote Marin Laboratory confirmed that PbTx-3 was present was at an aver age ratio of 35.6% of the total brevetoxin measured by ELISA (unpubl ished data). These findings may support the theory of adsorption since PbTx-3 is the most abundant form of brevetoxin found dissolved in water (Pierce et al. 2008). The more rapid drop in brevetoxin con centrations that was observed in the cleaned leaves and sheaths after the dissipation of a bloom may suggest another mechanism is occurring at times when brevet oxin concentrations in the water are high. Although brevetoxin uptake by macrophytes has never been documented, uptake of some other algal toxins by plants has been reporte d. The cyanobacterial toxins microcystins are a family of heptapeptide toxins that inhib it protein phosphatases 1 and 2A (Yoshizawa et al. 1990) and can also act as tumor promoter s (Falconer 1991). Uptake of microcystins has been demonstrated in severa l aquatic macrophytes including Ceratophyllum demersum Elodea canadensis Vesicularia dubyana and Phragmites australis ; with leaves, stem and rhizomes all apparently cap able of taking up micr ocystin (Pflugmacher et al. 1999; Pflugmacher et al 2001). Adverse effects on te rrestrial plants and plant seedlings upon exposure to microcystins have also been documented (Kurki-Helasmo and Meriluoto 1998; McElhiney et al. 2001). Okadaic acid is a marine algal toxin that acts in

PAGE 74

61 a similar manner to microcystins in that it is both a potent in hibitor of protein phosphatases and a tumor promoter; but it is a lipophilic polyether toxi n (Vieytes et al. 2000), making it structurally more similar to brevetoxin. Both microcystins and okadaic acid have molecular weights similar to brev etoxins, and are capable of entering spinach leaves and changing the phosphoryla tion state of proteins in the leaves (Siegl et al. 1990). These studies lend support to the notion that brevetoxins can be taken up into seagrass fluids or tissue. Regardless of which mechanisms of accumulation are occurring, how long brevetoxins persist in association with s eagrass will be limited largely by the turnover time of the leaves and the epiphytic commun ity. The majority of epiphytic species are sessile and are replaced as the leaves grow (Zieman and Zieman 1989). In the results presented here, brevetoxin concentrations in se agrass generally persisted at levels greater than 100 ng PbTx-3 eq./g for one to two months after K. brevis cells were no longer present. This is in agreement with average reported turnover times for T. testudinum leaves on the order of 25-60 days (Zie man and Zieman 1989; Tomasko and Lapointe 1991; Tomasko et al. 1996). The continued dete ction of brevetoxins at lower levels may result from background concentrations of K. brevis and brevetoxins in the water column or from release of brevetoxins recycled from the sediments, leaf litter, or biota in the seagrass meadow. Food Web Implications The accumulation of brevetoxins in associ ation with seagrasses has significant environmental implications. Seagrass meadows are highly productive ecosystems that support very diverse faunal assemblages serv ing as nursery habitat and refuge for many species as well as a food source for herbivor es including fish, water fowl, sea urchins, green turtles, and manatees (Hemminga a nd Duarte 2000). Seagrass represents a large fraction of the diet of some omnivorous fish, such as pinfish (Montgomery and Targett 1992). Crabs and isopods also consume seagrass, including roots and rhizomes. Epiphytes are an important component of th e seagrass ecosystem, and may account for up to 54% of the total biomass (Penhale 1977) and 56% of th e total productivity (Morgan

PAGE 75

62 and Kitting 1984). Grazers on s eagrass epiphytes include fis h, gastropods, polychaetes, amphipods, isopods, and other crustaceans (van Montfrans et al. 1984; Zieman and Zieman 1989; Greenway 1995). De trivores that may consume the detritus deposited on the leaves or the decaying leaves themselves (either principally or as a part of their diet) include fish, shrimp, marine worms, hol othurians, gastropods, isopods and amphipods (Zieman and Zieman 1989; Greenway 1995). Th e diversity of the seagrass community and the complexity of the food web it sustains present a multitude of previously unconsidered routes for brevetoxin transfer from seagrass into the coastal food web. In this study, the accumulation and persis tence of brevetoxins associated with seagrass appeared to be largely due to concen tration of brevetoxins on or in the epiphytes and detritus on the surface of the leaves. As a direct vector of brevetoxin to large herbivorous animals such as manatees th e distinction of whether the toxin is on or in the seagrass is not as important since the anim als ingest the whole seagrass. It may be relevant, however, in assessing the potentia l risk of brevetoxin exposure for the green turtle ( Chelonia mydas), which preferentially consumes younger leaves and leaf parts, avoiding epiphytized older leaves (Hemminga and Duarte 2000). Most research to date on persistence of brevetoxins in the marine environment has focused on bivalve shellfish, which can rema in toxic for several months (Morton and Burklew 1969; Steidinger et al. 1998a; Plakas et al. 2002; Plakas et al. 2004; Pierce et al. 2006). Persistence of brevetoxins in other comp artments of the marine environment after a K. brevis bloom has not been closely examined. Prior to 2002, the accumulation of brevetoxins on or in seagrass had never been previously reporte d, and the delayed or chronic exposure of manatees to brevetoxi ns through seagrass was not recognized as a threat. The presence of PbTx-3 in extr acts of whole seagra ss and scrapings was confirmed. The toxicity of this congener is well-establis hed (Baden and Mende 1982), verifying the presence of a harmful form of brevetoxins in the typical manatee food source. No other forms of brevetoxin were found. However, the LC-MS analyses were performed in 2003. Multiple additional brevetoxins and brevetoxin-like compounds have been identified since then, incl uding some more polar derivati ves that may not have been

PAGE 76

63 extracted by the method used here. Also, the MS detection method used to analyze these samples (MS Scan) was much less sensitiv e than the ELISA a nd smaller amounts of individual congeners may not have been detected. Brevetoxins were associated with seagra sses at high levels for weeks and at low levels for months in the absence of K. brevis cells. These findings confirm the stability of brevetoxins in the environmen t in the absence of an ongoing K. brevis bloom and reveal the possibilities of acute lethal exposur e through ingestion of seagrass during a K. brevis bloom or in the weeks immediately following, ch ronic dietary exposure to low levels of brevetoxins, and transfer of brevetoxins through herbivores and grazers into coastal marine food webs.

PAGE 77

64 Chapter 4. Confirmation of Bottlenose Dolphin Exposure to Brevetoxins During the 2004 Dolphin Mortality Event Introduction Bottlenose dolphins ( Tursiops truncatus ) are the most commonly observed marine mammals in Florida coastal waters (Barros a nd Odell 1990). This spec ies is not listed as threatened or endangered under the Endangere d Species Act, but bot tlenose dolphins are protected (as are all marine mammals) under the Marine Mammal Pr otection Act of 1972. Bottlenose dolphins face a number of risks, both human-related and naturally-occurring. Loss of habitat, changes in availability and quality of food due to human fishing efforts, vessel strikes, noise, pollutants, toxicants, and entanglemen t or bycatch by commercial and recreational fishing gear a nd debris are some examples of anthropogenic contributors to dolphin mortality; while naturally-occurri ng threats to dolphins include predation by large sharks, disease, parasites, and exposure to naturally occurring biotoxins (Reynolds et al. 2000). Brevetoxins produced by Karenia brevis have been the principal biotoxin implicated in dolphin mortality events. Be tween June 1987 and February 1988, more than 740 bottlenose dolphins died along the U.S. A tlantic coast from New Jersey to Florida (Geraci 1989). The mortalities began in the northeast and moved s outh during the fall and winter. This event occurred during the year in which a red tide was tr ansported out of the Gulf of Mexico via the Loop Current and carried northward in the Gulf Stream (Tester et al. 1991). It was documented as far north as North Carolina where it resulted in several human illnesses (Morris et al. 1991). Despite the overwhelming number of dol phins that stranded during the 1987-1988 event, the only samples tested for brevetoxins were liver tissues from 17 animals and the stomach contents from a single animal. Accord ing to the final report, this was the only

PAGE 78

65 suitable sample of stomach contents availabl e in the entire collec tion (Geraci 1989) and was composed of menhaden ( Brevoortia sp.) and weakfish ( Cynoscion regalis ). Additionally, seven species of live-caught fish were also tested. These included four species from the Virginia Beach area (bluefis h, croaker, spot, and red drum) collected in August 1987, and also two silver seatrout, th ree Spanish mackerel, and three menhaden that were collected off of Vero Beach, FL in late February 1988. Samples were extracted using procedures that were optimized for rec overy of PbTx-2 and were analyzed using a combination of thin layer chromatography (TLC) and fish bioassay. TLC fractions that were toxic to Gambusia affinis were subjected to a dditional chromatographic fractionation. Fractions that were toxic after three successive purification steps were then analyzed by HPLC with UV detection. Concen trations and toxin identification were determined by comparison with co-eluting brevetoxin standards. Using these methods, eight of the 17 livers were positive for brevetoxins, ranging in concentration from 83 to15,829 ng/g. None of the 17 control dolphin liv ers also analyzed this way yielded positive results. The viscera of the menhaden taken from the one suitable sample of stomach contents was positive for brevetoxins (level not reported). However, brevetoxins were not detected in the flesh of the menha den, the weakfish from the stomach contents, or the liver tissue of the same dolphin. Among the seven species of live caught fish also analyzed, only the viscera of the menhad en collected in February 1988 contained brevetoxins at approxim ately 200 g per fish. In the final report of this investigation, Geraci (1989) cited a biological toxin as the proximate cause. The dolphins appa rently were poisoned by brevetoxinThe dolphins were eventually infected with a host of bacterial and viral pathogens which produced an array of begu iling clinical signs. However, the numerous other physiological stressors and high contaminants documented in the dolphins and the fact that mortalities began much further north th an the bloom was documented prevented this diagnosis from being universally accepted. In 1994, Lipscomb et al reported detecting morbillivirus in more than half of the 79 animals from this event that they tested, concluding that this was a better candida te for the cause of the mortalities and permanently casting a shadow of doubt ove r the red tide hypothesis. Unfortunately,

PAGE 79

66 whether brevetoxins played a primary role as the cause of death or as an agent that weakened the dolphins immunodefenses, or a se condary role as an added stressor to animals that already had decr eased immune capabilities due to morbillivirus, or whether brevetoxins contributed at all, will probably remain unknown. Nonetheless, the National Marine Fisheries Service (NMFS) estimated th at this mortality represented a decline of more than 50% of the coastal migratory stoc k of bottlenose dolphins along the U.S. midAtlantic coast, and subsequently designated th is population as deple ted according to the Marine Mammal Protection Ac t (58 FR 17789, April 6, 1993), a status that remains unchanged. A somewhat less equivocal conclusion was drawn in the investigation into the stranding of more than 120 bottlenose dolphins in the Fl orida Panhandle between August 1999 and February 2000. The majority of th e strandings occurred in two peaks that coincided with two pulses of K. brevis red tide in the same general area (Van Dolah et al. 2003). The animals appeared to have been in good physical condition prior to their death, but due to the advanced state of decomposition of most of the carcasses, only 24 animals were tested. Highest levels of brevetoxins as measured by receptor binding assay (RBA) were found in stomach contents (0-474 ng/g), followed by liver (0-163 ng/g), and kidney (0-4.8 ng/g). Brevetoxin was not measured in lung or spleen by RBA. Chemical confirmation of brevetoxin was obtained by LC -MS/MS. However, despite the benefit of new and improved methods of analysis, only 29% of the stranded animals tested were positive for brevetoxins (Van Dolah et al. 2003). The same region of the Florida Panhandle was also the site of a more recent mass mortality of bottlenose dolphins. In 2004, between March 10 and Ap ril 13, 107 bottlenose dolphin carcasses were recovered along the northwest Florida Gulf coast, primarily in St Joseph Bay (Fig 4.1). Although the geogra phical area was similar to the 1999-2000 mortality event, the temporal distribution of the stranding was much more condensed, with most of the dolphins stra nding within the first two week s, and all but six stranding within the first three weeks (Fig. 4.2). An extensive multi-agency investigation led by staff of the NMFS Marine Mammal Health and Stranding Response Program (MMHSRP) was launched to

PAGE 80

67 determine the cause of the strandings (NMFS 2004). Necropsies indicated otherwise healthy animals at the time of death. Most of the dolphins examined had full stomachs containing partially digested fish suggesting recent feeding prior to their death. Possible mortality factors investigated included infec tious disease, acoustic trauma, environmental contaminants, and natural biotoxins (algal toxins). To assess the potential involvement of K. brevis or other HAB species, more than 230 water samples from 115 coastal and offshor e stations were collected between March 11 and May 5, 2004 by FWC-FWRI, the Florida Department of Environmental Protection (FDEP), and the Florida Department of Ag riculture and Consumer Services (FDACS) (Fig. 4.3). Microscopic examination of samples for phytoplankton was conducted by FWC-FWRI taxonomists. Karenia brevis was absent from all samples. However, a moderate bloom of the diatom Pseudo-nitzschia delicatissima (up to 464,400 cells/L) was present in St. Joseph Bay (NMFS 2004). Pseudo-nitzschia delicatissima has been documented to produce domoic acid (DA), a neur otoxic excitatory amino acid (Smith et al. 1990; Bates et al. 1998), and DA has been im plicated in mortality events of marine birds and mammals in other parts of the U.S. (Work et al. 1993; Scholin et al. 2000). Although DA-producing diatoms are present in the Gulf of Mexico (Parsons et al. 1999; Pan et al. 2001), no suspected or confirmed DA-related human illnesses or animal mortalities have been documented in Gulf waters. Dolphin tissue samples collected during this investigation by staff of the FWCFWRI Marine Mammal Pathobiology Labor atory and NMFS Marine Mammal Health and Stranding Response Program and seawater samples collected by the agencies listed above were sent to FWC-FWRI and to th e NOAA Center for Coastal Environmental Health and Biomolecular Resear ch (CCEHBR) for analyses of algal toxins. This chapter reports the results of dolphin tissue and seawater analyses for brevetoxins and DA that were performed at FWC-FWRI.

PAGE 81

Gulf of Mexico EscambiaSanta Rosa OkaloosaWalton Washington Bay Gulf Franklin Wakulla Holmes Jackson Gadsen Calhoun Liberty St. Joseph Bay Bay Gulf St. Joseph Bay Bay Gulf 50 0 50Kilometers Figure 4.1 Stranding locations of bottlenose dolphin carcasses rec overed between March 10 and April 13, 2004. Map created with data provided by NMFS. 68

PAGE 82

3/10/04 3/11/04 3/12/04 3/13/04 3/14/04 3/15/04 3/16/04 3/17/04 3/18/04 3/19/04 3/20/04 3/21/04 3/22/04 3/23/04 3/24/04 3/25/04 3/26/04 3/27/04 3/28/04 3/29/04 3/30/04 3/31/04 4/1/04 4/2/04 4/3/04 4/4/04 4/5/04 4/6/04 4/7/04 4/8/04 4/9/04 4/10/04 4/11/04 4/12/04 4/13/04 0 2 4 6 8 10 12 14Numbe r o f dolphin st r andings Figure 4.2 Number of dolphin strandings in the Florida Panhandle per day between March 10 and April 13, 2004. Graph crea ted with data provided by NMFS. 69

PAGE 83

Area Shown Gulf of Mexico Atlantic Ocean Pensacola Gulf of Mexico St. Joseph Bay Panama City 3/11-3/19 (FDACS, FDEP, FWRI) 3/20, 4/9, 4/28 (volunteer) 4/7-4/8 cruise (FMRI) Sample date 3/22-3/24 cruise (NOAA CSCOR HAB Event Response Fund, FWRI) 3/25, 3/31 (FDEP, FWRI) 4/13, 4/16, 4/19 (FDACS, FDEP, FWRI) 4/25-4/29, 5/6 (FWRI) Figure 4.3 Sites and dates of water collec tions between March 11 and May 6, 2004. Map courtesy of FWC-FWRI. 70

PAGE 84

71 Methods Tissues and stomach contents from 18 stranded dolphins were provided by the FWC-FWRI Marine Mammal Pathobiology Lab and NMFS MMHSRP. (The interim report describing this investigation [NM FS 2004] incorrectly cites the number of dolphins tested by FWRI as 11). Tissue t ypes included liver (n = 13), lung (n = 10), kidney (n = 13), cerebellum (n = 13), blood (n = 12), urine (n = 2), and stomach contents (n = 18). After completion of stomach content identification, samples of gastric juices (n = 25) and ingested fish (n = 29), including samples taken from seven additional dolphins were provided by Nelio Barros at Mote Marine Laboratory. All sa mples were frozen at -20C until extracted and analyzed. Sex, le ngth, location, and stranding date for each dolphin are detailed in Appe ndix E (data provided by NMFS). Tissues from 16 rough-toothed dolphins ( Steno bredanensis ) that stranded on Hutchinson Island on Florida s Atlantic coast, where K. brevis blooms are historically rare, were also analyzed to assess sample matrix effects. Tissue t ypes included liver (n = 16), lung (n = 16), kidney (n = 15), urine (n = 2), and brain (n = 10) These dolphins were recovered in August 2004 by staff at the Ha rbor Branch Oceanographic Institution and were necropsied at the FWC-FWRI Marine Mammal Pathobiology Laboratory. Brevetoxins in dolphin tissues and stom ach contents were extracted using 100% acetone as described in Chapter 2. Brevetox ins were extracted from blood and urine by solid phase extraction (SPE) using Supelco Discovery DSC-18 SPE cartridges (500 mg, 3 mL). Blood and urine (2 mL) were applied to cartridges that ha d been pre-conditioned with 6 mL of 100% methanol and then 6 mL of deionized water. Columns were washed with 6 mL of 25% aqueous methanol, and brev etoxins were eluted from the columns with 4 mL of 100% methanol. Brevetoxins in seawater were extr acted by passing 500 mL through a C18 SPE disk (3M Empore) that had been preconditione d with 8 mL of methanol and 8 mL of deionized water. The disk was then rinsed twice with 10 mL of deionized water, and brevetoxins were eluted with 20 mL of 100% methanol. The methanol extracts were evaporated to dryness and re-disso lved in 2 mL of 100% methanol.

PAGE 85

72 Total PbTx-2-type brevetoxin was quantif ied in all samples using a competitive ELISA as described in Chapter 2, with the modifications outlined in Chapter 3. The lower limit of quantification by ELISA as perf ormed here was approximately 10 ng/g for stomach contents and liver; 5 ng/g for l ung, kidney, and brain; 2 ng/mL for blood and urine; and 0.005 g/L for seawater. Results are expressed as ng Pb Tx-3 equivalents per gram of tissue, ng PbTx-3 equivalents per mL of body fluid, and g PbTx-3 equivalents per L of seawater, and reflect the overall c oncentration of PbTx-2-type brevetoxins and brevetoxin-like compounds present in the sample. Tissues from nine dolphins were also analyzed for domoic acid (DA). Tissue types included liver (n = 6), lung (n = 6), ki dney (n = 6), muscle (n = 2), blood (n = 6), urine (n = 2), and stomach contents (n = 8). Domoic acid was extracted by homogenization in 50% aqueous methanol (4 mL/g). Homogenates were centrifuged (10 minutes at 3,200 x g), the supernatants were retained, the pellets we re extracted a second time in the same manner, and the supernatants were pooled. Domoic acid was quantified in tissue extrac ts and whole (not extracted) seawater samples using the Direct cELISA ASP assay (Biosense Laboratories, Bergen, Norway). This pre-titrated commercial ELISA kit is a competitive assay in which free DA in samples or controls compete with plat e-bound DA for binding to specific anti-DA antibodies. The anti-DA antibodies are labele d with horseradish peroxidase (HRP) and the amount bound in the wells is measured by addition of the HRP substrate TMB (3,3 ,5,5 -Tetramethylbenzidine). Absorbance of the wells was read in a SpectraCount microplate reader (Packard) at 450 nm. The lower limit of quantification by ELISA as performed here was approximately 1 ng/g for st omach contents and tissues (liver, kidney, lung, muscle); 0.1 ng/mL for blood and urine; and 0.2-0.4 g/L for seawater. Results are expressed as ng DA per gram of tissue, and ng DA per mL of body fluid. All statistics were perfor med on log-transformed brevet oxin data using GraphPad Prism version 5.01 for Windows, GraphPad Software (San Diego California USA, www.graphpad.com). In most cases, non-parame tric analyses were used due to the presence of assay results that were below th e limit of detection (
PAGE 86

73 analyses, and to calculate means, results that were below the limit of detection were assigned a value of half the limit of detection. Results Of the more than 230 water samples that were collected between March 11 and May 5, 2004, 82 were extracted and analyzed fo r brevetoxins, and 16 were analyzed for domoic acid. Samples tested for DA were pr imarily from inside St. Joseph Bay where elevated concentrations of Pseudo-nitzschia (up to 454,000 cells/L) were observed. Brevetoxins were not detected in any samples collect ed offshore (Fig 4.4A). Low levels of brevetoxin ranging from 0.1 to 2 g/L were measured in alongshore samples collected in March 2004, predominantly in samples from inside St. Joseph Bay (Fig. 4.4A). Domoic acid was not detected in any of the samples analyzed (Fig 4.4B). Brevetoxins were detected in all stomach contents and multiple tissues of all bottlenose dolphins tested (n = 25, Table 4.1). Highest concentrations were measured in stomach contents and gastric fluids. Brevet oxin concentrations m easured in partially digested stomach contents were highly variable, ranging from 401 to 6,176 ng PbTx-3 eq./g, with an average value of 2,102 ng PbTx-3 eq./g. Gastric fluids from all 25 dolphins tested were similarly very high ranging from 137 to 5,535 ng PbTx-3 eq./mL (mean = 1,631 ng PbTx-3 eq./mL). Brevetoxin concentrat ions measured in dolphin tissues were less variable. All 13 liver samp les contained measurable conc entrations of brevetoxins with an average concentration of 51 ng PbTx-3 eq./g and a range of 37 to 104 ng PbTx-3 eq./g. Lung tissues were positive in 10 of 13 dolphins tested, with measured concentrations ranging from 9 to a maximum concentration of 72 ng PbTx-3 eq./g. The average concentration measured in lung ti ssues was 17 ng PbTx-3 eq./g. All 13 kidney samples tested positive ranging from 14 to 50 ng PbTx-3 eq./g (mean = 26 ng PbTx-3 eq./g), and both urine samples tested we re positive (16-34 ng PbTx-3 eq./mL). Brevetoxins were not detect ed in the brain of seven of 13 dolphins tested, and concentrations were low in the other six (9-12 ng PbTx-3 eq./g). Blood from 12 of 13 dolphins tested were positive for brevetoxins with measured concentrations ranging from 3 to 16 ng PbTx-3 eq./mL (mean = 7 ng PbTx-3 eq./mL).

PAGE 87

74
PAGE 88

Table 4.1 Brevetoxin concentrations (ng PbTx-3 eq./g or ng PbTx -3 eq./mL) measured in tissues from the 2004 dolphin mortality. StomachGastric Field ID LiverKidne y Lun g UrinecontentsfluidsCerebellumBlood Hubbs-0415-Tt----------------6631,570-------Hubbs-0416-Tt39 50 72----1,0883,316
PAGE 89

76 Tissue samples from 16 dolphins that stra nded on Floridas Atlantic coast were analyzed as matrix controls. Tissue types in cluded liver (n = 16), lung (n = 16), kidney (n = 15), urine (n = 2), and brain (n = 10). No brevetoxin was det ected in any of the Atlantic coast dolphin tissues tested (Table 4.2). The Spearman rank correlation coefficient was used to assess the correlation of brevetoxin concentrations in the different sample types tested for the 2004 dolphins (Table 4.3). The only statisti cally significant (p < 0.05) relationship found was a positive correlation between stomach contents and cerebellum (rs = 0.699, p = 0.008). However, the sample size was very small, and seven of the 13 brain tissues tested were below the limit of detection for brevetoxin. When th ese samples were excluded there was no relationship in brevetoxin concentr ations between the two tissues (rs = 0.029, p = 1.000). Several intact or only slightly digest ed prey items retrieved from the stomach contents of 20 dolphins were also analyzed including 17 clupeid fish identified by Nelio Barros (Mote Marine Labor atory) as menhaden ( Brevoortia sp.), two silver perch ( Bairdiella chrysoura ), three spot ( Leiostomus xanthurus) one pinfish ( Lagodon rhomboides ), one kingfish ( Menticirrhus sp.), one puffer (species unknown), and four unidentified teleosts. In most cases the w hole fish was homogenized and extracted. All prey items contained high concentrations of brevetoxins with the hi ghest levels measured in the menhaden (Table 4.4). Brevetoxins in whole menhaden ranged from 844 to 15,361 ng PbTx-3 eq./g and averaged 3,903 ng PbTx-3 eq./g. Viscera analyzed separately for three menhaden contained an average concentration of 23,351 ng PbTx-3 eq./g, and brevetoxins in menhaden muscle aver aged 424 ng PbTx-3 eq./g (n = 6). Tissues from nine dolphins were analyzed for DA (Table 4.5). No DA was detected in any of the liver (n = 6), lung (n = 6), kidney (n = 6), or muscle (n = 2) tissues tested. Low concentrations of DA were detectab le in six of eight stomach contents tested, ranging from 2 to 9 ng DA/g. Both urine samples were positive for DA by ELISA (0.5-2 ng DA/mL), and five out of six blood samples tested contained low concentrations of DA, ranging from 0.1 to 0.6 ng DA/mL.

PAGE 90

Table 4.2 Brevetoxin concentratio ns (ng PbTx-3 eq./g) measured in tissues from roughtoothed dolphins ( Steno bredanensis ) that stranded on Florida s Atlantic coast in August 2004. Field ID LiverKidneyLungUrineBrain SFNEFL0410HHN
PAGE 91

Table 4.4 Brevetoxin concentratio ns (ng PbTx-3 eq./g) measured in prey items retrieved from the stomach contents of 2004 dolphins Prey provided and identified by Nelio Barros (Mote Marine Laboratory). Field ID Prey item PbTx-3 eq. (ng/g) Hubbs-0415-TtUnidentified teleost 2,713 Unidentified teleost 496 Hubbs-0416-TtMenhaden ( Brevoortia sp.) 4,261 Hubbs-0418-TtMenhaden 1,442 Menhaden 4,624 Hubbs-0419-TtMenhaden muscle 778 Hubbs-0420-TtMenhaden muscle 206 Hubbs-0422-TtMenhaden 1,059 Menhaden 2,300 Hubbs-0423-TtMenhaden muscle 446 Menhaden viscera 31,822 Hubbs-0424-TtMenhaden 5,327 Menhaden 844 Hubbs-0425-TtSilver perch ( Bairdiella chrysoura )1,092 Hubbs-0426-Tt Spot ( Leiostomus xanthurus ) 2,048 Unidentified teleost 828 Hubbs-0427-TtSpot 536 Hubbs-0428-TtMenhaden 15,361 Hubbs-0430-TtSilver perch 3,572 Hubbs-0431-TtMenhaden 3,812 Puffer sp. muscle 439 Hubbs-0432-TtMenhaden muscle 476 Hubbs-0436-TtMenhaden muscle 437 Menhaden viscera 5,046 Hubbs-0437-Tt Kingfish ( Menticirrhus sp.) 1,085 Unidentified teleost 671 SJP0317-15Menhaden 939 SJP0318-17Menhaden 2,968 Pinfish ( Lagodon rhomboides ) 405 SJP0325-32Menhaden muscle 200 Menhaden viscera 33,185 Spot muscle 181 Spot viscera 555 78

PAGE 92

Table 4.5 Domoic acid concentrations (ng DA/g or ng DA/mL) measured in tissues from the 2004 dolphin mortality. 79 Stomach Field ID MuscleLiverKidne y Lun g UrinecontentsBlood Hubbs-0422-Tt
PAGE 93

80 100% of the dolphins that were tested, including a ll 25 dolphins tested at FWRI and at least 11 others tested by NOAA CCEHBR, supporting brevetoxico sis as the cause for this mortality event. A comparison of the brevetoxin concentrations measured in this event to levels measured in the 1999-2000 event is not straight forward as different methods of detection were employed. Receptor-binding assays (RBAs) were used in the 1999-2000 investigation. As discussed in Chapter 2, th ese assays assess toxin concentration of a sample by measuring the pharmacological activ ity (i.e. binding of the toxins in the sample to sodium channels). The ELISA is an assay based on structural recognition of antibodies to a certain portion of the brevetoxin molecule. Given th e different ways the assays detect brevetoxins and the multip le known forms and varying toxicities of brevetoxins, difference between the results obtained using such assays are expected. Nevertheless, during the 2004 event a minimum of 11 additional dolphins were analyzed by NOAA CCHEBR using RBA (as well as ra dioimmunoassay and LC-MS) and were all positive at similar levels to those reported he re. Results of all 36 dolphins were combined and average values were repor ted in Flewelling et al. (2005, supplementary information). The average concentration in stomach and esophag eal contents (partially digested fish) of eight dolphins from the 2004 event analyzed by RBA was 1,426 ng PbTx-3 eq./g ( 2,104), and was similar to the average value of stomach contents measured by ELISA (2,102 ng PbTx-3 eq./g, 1,654). Regardless of the method used, the brevetoxin concentrations measured in the stomach contents of the 2004 dolphins greatly exceed those measured in the 1999-2000 dolphin mortality event (range: < ld 474 ngPbTx-3 eq./g) by RBA (Van Dolah et al. 2003). LC-MS analyses of the dolphin stomach contents and six of the undigested menhaden were performed at Mote Marine La boratory and identified PbTx-2 and -3 and lower concentrations of brevetoxin metabo lites (Flewelling et al. 2005). The parent brevetoxin PbTx-2 is highly reactive and is quickly metabolized when K. brevis cells are lysed (Pierce et al. 2001; Abraham et al. 2006; Pierce et al. 2008) or consumed by oysters (Plakas et al. 2002; Plakas et al. 2004; Wang et al. 2004). The presence of PbTx-2 in the dolphin stomach contents and the menhaden provide strong evidence that the menhaden,

PAGE 94

81 planktivorous fish, had been f eeding on an unobserved bloom of K. brevis shortly before being preyed upon by dolphins. The high concentrations of brevetoxins m easured in undigested prey items (Table 4.4) clearly demonstrate that fish were the source of the brevetoxins measured in the dolphins. The ability of live fish to accumula te brevetoxins to concentrations lethal for marine mammals has not widely been cons idered a threat for dolphins as it is conceptually contrary to th e well-established ichthyotoxicity of brevetoxins, a property that is repeatedly demonstrated via massive fish kills during K. brevis blooms (Landsberg 2002). Whether the fish eaten by these dolphins were alive or dead at the time of feeding can not be known, but in either case the fish were able to achieve extremely elevated concentrations of brevetoxins in their visc era, demonstrating a capacity for brevetoxin tolerance in fish that has been previously unrecognized. Following this event and the analyses of the fish recovered from the stomach contents, a field study was performed to investigate whether brevetoxin accumulation in live fish is a common occurrence and is described in Chapter 5. Following the 2004 dolphin mortality event, Fi re et al. (2007) recognized a lack of information on brevetoxin body burdens in dolphins that strand during K. brevis blooms but that are not associated with ma ss mortality events, and also a lack of information on baseline brevetoxin body burdens in dolphins that ar e frequently exposed to K. brevis blooms. They measured brevetoxin concentrations in bottlenose dolphins from Sarasota Bay in southwest Florida that stranded during periods of elevated K. brevis cell concentrations but were not associated w ith a large-scale stranding event, as well as in dolphins that stranded during nonK. brevis bloom periods. Samples were analyzed using the same ELISA assay employed he re, allowing for mean ingful comparison. Brevetoxins were detected in 16 out of 19 dolphins that stranded during K. brevis blooms with average concentrations of 30 ng PbTx-3 eq./g in liver, 16 ng PbTx-3 eq./g in kidney, 11 ng PbTx-3 eq./g in lung, 37 ng PbTx-3 eq./ mL in urine, and 1,142 ng PbTx-3 eq./g in stomach contents (Fire et al. 2007). These le vels are similar in magnitude to those measured in the 2004 mortality even t, and suggest the impact of K. brevis blooms on bottlenose dolphins may be underestimated.

PAGE 95

82 Brevetoxins were also measurable, but were significantly lower and much less prevalent, in dolphins that stranded during nonK. brevis bloom periods. Brevetoxins were detected in 5 out of 13 livers (mean = 13 ng PbTx-3 eq./g), in 2 out of 12 kidneys (mean = 2 ng PbTx-3 eq./g), and were not measured in either of two lung samples (Fire et al. 2007). No stomach contents were anal yzed for dolphins that stranded during nonbloom periods. An extremely puzzling question is why have these brevetoxin-related mass mortalities of dolphins occurred in the Flor ida Panhandle but not in southwest Florida, where K. brevis blooms are more frequent? One possibi lity is that the dolphins in the northern Gulf of Mexico are physiol ogically or behaviorally nave to K. brevis blooms (Van Dolah et al. 2003). Anot her possible reason may relate to the feeding habits of bottlenose dolphins in the northern Gulf of Mexico. Dolphins are often observed in association with shrimp boats in the Gulf (Fertl and Leatherwood 1997), and reports of observations as well as the composition of pr ey identified in the stomachs of dolphins from the northern Gulf suggest that feedi ng on bycatch discarded by shrimp boats does occur (Barros and Odell 1990). Clupeid fish are not one of the mo st common types of bycatch, but they are occasionally caught by sh rimp trawlers in the Gulf (Clucas 1997). More typically, clupeid fish are caught in purse-seine nets. Alt hough purse-seine vessels were observed returning to port in St. Jose ph Bay in April and May 2004, the boats were not actively fishing or discarding bycatc h when they were observed (NMFS 2004). In 2005, the first red tide-rela ted dolphin mortality event in southwest Florida did occur during a particularly persistent K. brevis bloom. One theory for why this unusual mortality event occurred is that the prolonged K. brevis bloom resulted in decreased availability of typical dolphi n prey items, and led to shif ts in dolphin feeding behaviors (D. Gannon, Mote Marine Laboratory, unpublished data). Despite the presence of a moderate bloom of the diatom Pseudo-nitzschia delicatissima in St. Joseph Bay, no domoic acid was detected in bloom samples. Pseudonitzschia blooms are common in the Gulf of Me xico (Dortch et al. 1997; Parsons and

PAGE 96

83 Dortch 2002). Pseudo-nitzschia delicatissima and other Pseudo-nitzschia species found in the Gulf of Mexico have been documente d to produce domoic acid (Smith et al. 1990; Dortch et al. 1997; Bates 1998; Pan et al. 2001), but unlike K. brevis the presence of these species does not necessarily mean that domoic acid is bei ng produced. Domoic acid production can be variable among toxic species a nd can also vary in individual strains in response to changes in environmental condi tions (Bates 1998). The analyses by FWCFWRI and NOAA CCEHBR as a part of this inve stigation represent the first analyses of domoic acid performed on ma rine mammal tissues from the Gulf (NMFS 2004). The concentrations measured in the dolphins we re several orders of magnitude lower than those measured in another domoic acid-relate d marine mammal mortality event in the Pacific (Scholin et al. 2000), and the levels present in dolphin stomach contents were several orders of magnitude lower than levels that are considered safe for human consumption. Thus there is considerable uncerta inty that domoic acid played any role in this event (NMFS 2004). Nevertheless, the de tection of domoic acid in these dolphins reveals that dolphin exposure to domoic acid does occur to some extent in the Gulf of Mexico and serves to alert us to the possibi lity of future domoic acid-related poisoning events in the Gulf.

PAGE 97

84 Chapter 5. Brevetoxin Accumu lation in Live Fish Introduction Fish kills in the eastern Gulf of Mexico have been documented as far back as 1844 (Ingersoll 1882) and have been observed almost annually during Karenia brevis blooms for over 40 years (Steidinger et al. 1973; Landsberg 2002). The profound ichthyotoxicity of brevetoxins produced by K. brevis was, in fact, the primary characteristic of the toxins used by researcher s to guide them to their purification in the late 1970s (Baden et al. 1979; Risk et al. 1979). The ichthyotoxici ty of the different brevetoxins varies, and Baden (1989) listed 24-hr fish bioassay LD50s of brevetoxin congeners ranging from 3-5 nM for PbTx-1-type brevetoxins to 10-37 nM for PbTx-2type brevetoxins. In just a seven year period (2000-2006) more than 1,400 red tiderelated fish kill reports were received by the FWC-FWRI Fish Kill Hotline, a 24-hour telephone and web-based reporting system ma intained by the FWRI Fish and Wildlife Health section. Due to the well-established toxicity of brev etoxins to fish, the ability of live fish to accumulate brevetoxins to concentrations dangerous for marine mammals has not widely been considered a threat. In 2004, however, a mass mortality of bottlenose dolphins ( Tursiops truncatus ) in the Florida Panhandle clearly indicated that fish have the potential to vector brev etoxins to higher tropic levels (C hapter 4; Flewelling et al. 2005). High levels of brevetoxins were measured in multiple tissues of all dolphins examined, and in tissues from prey fish retrieved from the stomach contents of 20 dolphins. Menhaden ( Brevoortia sp.) recovered from the stomach contents contained excessive concentrations of brevetoxins, reachi ng 33,185 ng PbTx-3 eq./g in viscera and 15,361ng PbTx-3 eq./g in fish analyzed whole (Chapt er 4, Table 4.4), and parent brevetoxins PbTx2 and PbTx-3 were confirmed by LC-MS (Flewe lling et al. 2005; N aar et al. 2007). These

PAGE 98

85 results demonstrated that planktivorous fish can vector dangerous concentrations of brevetoxins to highe r trophic levels. Despite the frequent occurrence of K. brevis blooms in the Gulf of Mexico, red tide-related mass mortalities of dolphins have been extremely rare. In the 2004 event, brevetoxin-laden clupeid fish (planktivores) were shown to be the vector of brevetoxins to the dolphins (NMFS 2004, Chapter 4). Howeve r, clupeid fish have not been found to be a dominant prey item for dolphins in th e Gulf. In a quantita tive study of the food habits of bottlenose dolphins, the most impor tant prey fish by bot h number and frequency of occurrence for dolphins in the northern Gulf of Mexico an d along the southwest Florida coast were sciaenid fish (drums, cr oakers, seatrout) (Barros and Odell 1990). In a similar study examining prey items from 16 Sarasota Bay dolphins, pinfish ( Lagodon rhomboides ), pigfish ( Orthopristis chrysoptera), striped mullet (Mugil cephalus ) and spot ( Leiostomus xanthurus) comprised more than 80% of the dolphins diet in terms of numerical abundance (Barros and Wells 1998). Although these results differ somewhat from the earlier Barros and Ode ll study, the authors point out that most samples from the 1990 study came from animals that stranded on Gulf beaches and areas north and south of Sarasota Bay, and so may not be representa tive of the resident dolphins in Sarasota Bay. Nonetheless, in both studies clupeid fish were only a minor contributor to the dolphin diet, which may partia lly explain the infrequency of red tide-related dolphin mortalities. This chapter presents data on concentrati ons of brevetoxins measured in various species of fish collected live from St. Joseph Bay in the Florida Panhandle and from southwest Florida coastal waters. The field co llections from St. Joseph Bay were initially conducted to determine whether brevetoxin accu mulation in live fish in the area where the 2004 dolphin mortality occurred extended beyond clupeid fish, and continued in an effort to determine how long brevetoxins c ould persist in the ti ssues of live fish. Additional analyses of fish collected from southwest Florida where red tides occur more frequently were conducted to compare clupeids with more typical dolphin prey species, and to assess whether these species can also accumulate and vector potentially dangerous concentrations of brevetoxins.

PAGE 99

86 Methods Fish included in the study were obtai ned from numerous sources, with many provided from collections made as a part of other investigations. Fish from St. Joseph Bay were mainly collected by staff of the FWC-FWRI Harmful Algal Bloom (HAB) section. Some were collected by the FWCFWRI Fisheries Independent Monitoring (FIM) section, and also obtained from recrea tional fishermen. Fish were collected during and immediately after the dolphin mortality (M arch-May 2004) as well as in February, June, September, and November of 2005, a nd in May, August, and November of 2006. Here the objective was to sample across a br oad range of species a nd trophic levels, and all species collected were analyzed. Fish from southwest Florida were si milarly collected by FWC-FWRI HAB or FIM staff and recreational fishermen, as well as by Damon Gannon at Mote Marine Laboratory. Most samples were collected fr om inshore coastal waters of Tampa Bay, Sarasota Bay, or Placida Harbor (Fig. 5.1) between June 2004 and December 2007. Fish were collected using seine nets, cast nets, or hook-and-line gear. Species of interest for this study were those in the Clupeidae and Sciaeneidae families as well as pinfish ( Lagodon rhomboides ). All fish were frozen at -20C until they were prepared for analysis. Whenever possible, weight and total length were recorded. Fish collected from St. Joseph Bay in 2004 were primarily separated into muscle and viscera (organs including gastrointestinal [GI] tract). In some cases, liver and stomach contents were sampled separately. Undigested prey items found in the stomachs of a few fish were analyzed separately from the rest of the GI contents. For all other collections, muscle, liver, and stomach or GI contents were sampled separately when possible. Gill was also collected in fish from southwest Florida. In smaller fish where the target tissues were less than 0.1 g each, either the tissues of multiple fish of the same species that were collected t ogether were pooled or the visc era (organs including GI tract) of individual fish was collected. Brevetoxins in fish were extracted by homogenization in either 100% acetone (4 mL/g tissue) or 80% aqueous methanol (2.5 mL/g tissue). Homogenates were centrifuged

PAGE 100

87 (10 min at 3,200 x g) and the supernatants we re retained. The pellets were extracted a second time in the same manner and the supern atants were pooled. Acetone extracts were evaporated to dryness and re-dissolved in 80% aqueous methanol. All extracts were partitioned twice with 100% hexane (1:1 v:v). The methanol fraction was retained for analysis. At sites where specimens were collec ted by HAB staff, both surface and bottom water samples were also collected. Aliquots were preserved with Lugols iodine and enumeration of K. brevis cells was performed by FWRI HAB taxonomists Jennifer Wolny and Earnest Truby. Brevetoxins were extracted from unpreserved seawater samples by passing 0.5 L through an Empore Si-C18 disc that had been preconditioned with 20 mL of methanol and th en 20 mL of DI water. Brevetoxins were eluted from the discs into clean glass test tubes using 2 10-mL washes of methanol. The combined eluents were evaporated to dryness and re-di ssolved in 2 mL of methanol. All extracts were stored at -20oC until analyzed. Total PbTx-2-type brevetoxin was quantifie d in all samples using a competitive ELISA as described in Chapter 2, with the modifications outlined in Chapter 3. The lower limit of quantification by ELISA as performed here was approximately 5-10 ng/g for fish tissues and 0.01 g/L in seawater. Additional information on area K. brevis cell concentrations was obtained by querying the FWC-FWRI Harmful Algal Bloom database, which contains all K. brevis monitoring data for the state of Florida. Each fish analyzed from southwest Florida was designated as either red tide or non-red tide based on area cell counts. For Tampa Bay fish, K. brevis cell count data from samples co llected in Pinellas, Hillsborough, and Manatee counties were used; for Sarasota Bay fish, K. brevis cell count data from Manatee and Sarasota counties were used; and for Placida Harbor fish, K. brevis cell count data from southern Sarasota county as well as from Lee and Charlotte counties were used (Fig 5.1). Fish were consid ered red tide specimens when maximum K. brevis cell densities in these areas w ithin a 6-week time window (30 days prior to collection and two weeks after) exceeded 5,000 cel ls/L. This criteria was inte nded to compensate for the

PAGE 101

movement of fish as well as the extremely patchy nature of K. brevis blooms and the discrete nature of water samples collected for red tide monitoring. All statistics were performe d on log-transformed brevetox in data using GraphPad Prism version 5.01 for Windows, GraphPad Software (San Diego California USA, www.graphpad.com). In most cases, non-parame tric analyses were used due to the presence of assay results that were below th e limit of detection (
PAGE 102

89 Results 2004 St. Joseph Bay Collections Maximum K. brevis cell densities observed in St. Joseph Bay each month from September 2003 through December 2006 are sh own in Figure 5.2. In addition to the water samples obtained during each fish collection, Figure 5.2 includes K. brevis cell counts for five fixed sites in St. Joseph Bay that are sampled every month by the Florida Department of Environmental Protecti on (FDEP) and sent to FWC-FWRI for phytoplankton identification. A small bloom of K. brevis was present in St. Joseph Bay in late 2003, three months before the dolphin mortality event began. Between March 2004 (during the dolphin mortality) and July 2005, K. brevis was not present with the exception of background concentrations noted in a few samples collected in April and May 2005. Elevated concentrations of K. brevis were next noted St. Joseph Bay in August 2005, and by September 2005, K. brevis cell concentrations in the bay reached as high as 2.5 million cells/L. Karenia brevis cell densities remained elevated in the bay until late October. Water samples collected in mid-November 2005 contained only low concentrations of K. brevis with maximum observed densities of 5,300 cells/L on November 17. After these collections, no K. brevis cells were found in subsequent samples until one year later (late October 2006), when very low concentrations (1,600 cells/L) were noted at only a few sites. Brevetoxin concentrations measured in St. Joseph Bay water samples ranged from less than 1 to 80 g PbTx-3 eq./L during the K. brevis bloom, and were often measurable but low (typi cally < 1 g PbTx-3 eq./L) in the absence of K. brevis cells (Fig. 5.2). Between March 28 and May 5, 2004, 101 live fish were collected from St. Joseph Bay. Of these, 64 fish, representing 10 species, were analyzed for brevetoxins at FWCFWRI (Table 5.1). The remaining 37 fish were analyzed at the University of North Carolina at Wilmington using the same protoc ols, and the combined results for all 101 fish are reported in Naar et al. (2007). Half of the fish collected in 2004 (n = 32) were separated into muscle and vis cera for brevetoxin analysis. Liver was collected from 11 fish, and stomach contents were collected from nine fish. For 20 Spanish mackerel

PAGE 103

Sep-03 Nov-03 Jan-04 Mar-04 May-04 Jul-04 Sep-04 Nov-04 Jan-05 Mar-05 May-05 Jul-05 Sep-05 Nov-05 Jan-06 Mar-06 May-06 Jul-06 Sep-06 Nov-06 Jan-07 0 5,000 10,000 15,000 20,000 1,000,000 1,500,000 2,000,000 2,500,000Maximum K. bre v i s (cells/L) 0 2 4 6 8 10 40 60 80 100Seawate r (PbTx-3 eq. g/L) K. brevis Seawater Figure 5.2 Maximum K. brevis cell densities and brevetoxin concentrations measured in seawater samples collecte d throughout St. Joseph Ba y between September 2003 and December 2006. 90

PAGE 104

91 ( Scomberomorus maculatus) collected in April 2004, only muscle tissue was provided for analysis. Brevetoxins were measured in all of the fish collected in March 2004, including those analyzed at UNC Wilmington. Among the March 2004 fish results reported here, highest concentrations were measured in th e muscle (414 ng PbTx-3 eq/g.) and stomach contents (3,955 ng PbTx-3 eq./g) of a spotted seatrout ( Cynoscion nebulosus). In general, concentrations were lower in fish collected in April 2004, but high concentrations were measured in the liver (mean = 2,334 ng PbTx-3 eq./g) and stomach contents (mean = 228 ng PbTx-3 eq./g) of four sheepshead ( Archosargus probatocephalus ). Similar concentrations were measured in the liver (mean = 2,369 ng PbTx-3 eq./g) and stomach contents (mean = 251 ng PbTx-3 eq./g) of f our spotted seatrout from early May 2004. An undigested clupeid fish was recovered from the stomachs of one of these seatrout, and analyzed separately. This prey fish contai ned brevetoxin concentrations of 376 ng PbTx-3 eq./g in the muscle and 9,075 ng Pb Tx-3 eq./g in the viscera. 2005-2006 St. Joseph Bay Collections From February 2005 to November 2006, 189 fish were collected from St. Joseph Bay, representing 38 species and various trophic guilds (Table 5.2). The average concentrations measured in muscle, liver, and stomach contents of all fish for each collection is shown in Figur e 5.3, and the prevalence of br evetoxins, expressed as the percentage of brevetoxin-positive samples of each tissue type, is shown in Figure 5.4. Although no K. brevis was noted in the bay until Augus t 2005, brevetoxins were detected in fish collected in February and June 2005. In general, average levels were highest during and immediately followi ng the red tide in Septem ber and November 2005, after which they slowly decreased. The presence of brevetoxins in fish mu scle was closely tied in time to the K. brevis bloom (Fig. 5.3). In February 2005, no toxi n was measured in the muscle of any fish. Low brevetoxin levels (13-20 ng PbTx-3 eq./g) were measured in the muscle of a few fish collected in June 2005. During the K. brevis bloom, brevetoxins were measurable in the muscle of 96% of the fish tested, with concentra tions ranging from 9 to

PAGE 105

92 581 ng PbTx-3 eq./g. In November 2005 after the K. brevis bloom, brevetoxin was found in the muscle of 88% of the fish collect ed, with a maximum concentration of 116 ng PbTx-3 eq./g. By May 2006, both the proportion of positive muscle samples (64%) and the concentrations measured in the musc le had decreased. With the exception of a hardhead sea catfish (Ariopsis felis ), brevetoxin was not found in the muscle of any fish collected in August or November 2006. Overa ll, the highest concen tration measured in muscle (581 ng PbTx-3 eq./g) was from a Spanish sardine ( Sardinella aurita ) collected during the September 2005 red tide. Liver samples consistently yielded the greatest proportion of brevetoxin-positive results. In February 2005, brevetoxin concentrations in fish li vers ranged from not detectable (in 30% of the fish collected) to 81 ng PbTx-3 eq./g. Average liver concentrations increased in subsequent coll ections reaching peak levels in November 2005, just after the red tide ended. The highe st liver concentration measured was 16,483 ng PbTx-3 eq./g in a red snapper (Lutjanus campechanus) collected in November 2005. In May 2006, the average concentration in th e livers had decreased but was still high (598 ng PbTx-3 eq./g), and brevetoxin was detected in 100% of the livers tested. In August and November 2006, the averages continued to decline, but br evetoxin was still found in the livers of more than ha lf of the fish collected. Brevetoxin concentrations in stomach a nd GI contents displayed the greatest variability. In February and June 2005, brev etoxins were measured in the stomach contents of fish at only low levels (maximum of 43 and 144 ng PbTx-3 eq./g, respectively) and in only approximately half of the fish collected. In September and November 2005, brevetoxins were measured in all fish stomach contents. The average concentrations in stomach contents in fi sh collected in September and November 2005 and May 2006 were surprisingly not signifi cantly different from each other (KruskalWallis test with Dunn's multiple comparis on post-test). The highest concentration measured in fish GI contents was 2,839 ng PbTx-3 eq./g in a scaled sardine (Harengula jaguana) collected in May 2006.

PAGE 106

Table 5.1 Range of brevetoxin concentrations m easured in fish collected from St. Joseph Bay between March 28 and May 5, 2004 (durin g and immediately following the dolphin mortality event). Common Name Genus species Main DietNMuscleLiverGI Content s Viscera Scaled sardine Harengula jaguana Plankton 118 185 Sheepshead A rchosargus probatocephalus Benthic invertebrates 4
PAGE 107

Table 5.2 Range of brevetoxin concentrations m easured in fish collected from St. Joseph Bay between February 2005 and November 2006. Range (ng PbTx-3 eq./g) Common Name Genus species NMain Diet MuscleLiver GI Contents Scaled sardine Harengula jaguana 12 Plankton
PAGE 108

2004Feb 05Jun 05Sep 05Nov 05May 06Aug 06Nov 06 1 10 100 1,000 10,000ng PbTx-3 eq./g Muscle Liver Stomach Contents Viscera
PAGE 109

2004Feb 05Jun 05Sep 05Nov 05May 06Aug 06Nov 06 0 20 40 60 80 100% PbTx-positive Muscle Liver Stomach Contents Viscera n = 64 n = 10 n = 29 n = 22 n = 17 n = 23 n = 44 n = 39 Figure 5.4 Prevalence of brevetoxin contamin ation in fish (all species) collected throughout St. Joseph Bay between 2004 and 2006. Prevalence is expressed as the percentage of fish containing detectable le vels of brevetoxins (> 5-10 ng/g). Red bar indicates K. brevis bloom period. 96

PAGE 110

97 Southwest Florida Collections From June 2004 to December 2007, 464 fish were collected from coastal waters along Pinellas, Manatee, Sarasota and Charlotte counties, with the majority of specimens collected in Tampa Bay, Sarasota Bay, and Pl acida Harbor (Fig. 5.1). Species included pinfish ( Lagodon rhomboides), at least eight sciaenid spec ies, and four clupeid species. Ranges of brevetoxin concentrations measured in each species during both K. brevis bloom periods and non-bloom periods are listed in Table 5.3. Clupeid fish were mainly collected from Sarasota Bay and Placida Harbor. Of the 178 clupeid fish analyzed, 156 were collected at times when K. brevis was present at above background levels in proximate coastal waters. As a group, brevetoxin concentrations in clupeid muscle averaged 83 ng PbTx-3 eq./g dur ing bloom periods and 24 ng PbTx-3 eq./g during non-bloom periods, and in gill tissue averaged 634 ng PbTx-3 eq./g during red tides and 327 ng PbTx-3 e q./g during non-bloom pe riods. The highest brevetoxin concentration meas ured in clupeid muscle was 647 ng PbTx-3 eq./g in an Atlantic thread herring ( Opisthonema oglinum ). Brevetoxin concentrat ions in viscera of clupeid fish collected during red tides averaged 4,367 ng Pb Tx-3 eq./g (with a maximum value of 69,047 ng PbTx-3 eq./g in a Spanish sard ine) and were significantly higher (p = 0.0005, Mann Whitney test) than those collect ed during non-bloom periods (mean = 319 ng PbTx-3 eq./g). The average concentr ation of 18 clupeids collected during a K. brevis bloom and analyzed whole was 195 ng PbTx-3 eq./g, and was higher than five whole clupeids collected during non-bloom periods (mean = 42 ng PbTx-3 eq./g). Brevetoxin concentrations measured in the muscle and viscera of individual Atlantic thread herri ng, scaled sardines, and Spanish sardines collected between January 2006 and March 2007 are plotted by date of collec tion in Figures 5.5A and B. Most of the specimens were collected during K. brevis bloom periods, with highe st concentrations in all species observed in fish collected in September 2006. Karenia brevis was not present at more than 5,000 cells/L in waters near these clupeid fish collection sites between February and June 2006. Brevetoxins were still measurable in both muscle and viscera in most fish collected in February and were lo wer in the few fish collected in March and June 2006.

PAGE 111

Table 5.3 Brevetoxin concentrations measured in fish collected from southwes t Florida between June 2004 and December 2007. 98 NnMeanRangenMeanRangenMeanRangenMeanRangenMeanRange Yellowfin menhaden RT221311-1621,98743-3,9322534461-6062221204-238 Brevoortia smithi non-RT0 Menhaden RT115,000 cells/L in area within 30 days prior to and 14 days after collection); non-RT = non-red tide;
PAGE 112

99 A total of 60 pinfish were collected from Sarasota Bay and Placida Harbor, with 49 collected during K. brevis bloom periods. Brevetoxin conc entrations measured in the muscle, liver, and viscera of pinfish colle cted between January 2005 and March 2007 are plotted in Figures 5.6A and B. During K. brevis bloom periods, brevetoxin concentrations averaged 42 ng PbTx-3 eq./g in muscle, 401 ng PbTx-3 eq./g in gill, 1,962 ng PbTx-3 eq./g in liver, and 1,194 ng PbTx-3 eq./g in stomach contents. Mean brevetoxin concentrations in pinfish collected during nonbloom periods were lo wer in muscle (5 ng PbTx-3 eq./g), gill (41 ng PbTx-3 eq./g), and stomach contents (526 ng PbTx-3 eq./g), yet were higher in liver (2,864 ng PbTx-3 eq./ g). Significant differences in brevetoxin concentrations were found between red tide and non-red tide pinfish muscle (p < 0.0001) and gill (p < 0.05), but not stomach contents or liver. In smaller specimens collected during red tides, brevetoxin concentrations in viscera averaged 1,092 ng PbTx-3 eq./g and were significantly higher (p = 0.0033) than viscera of non-bloom specimens (266 ng PbTx-3 eq./g). The highest brevetoxin concentrations in pinfish muscle (197 ng PbTx-3 eq./g) and viscera (5,188 ng PbTx-3 eq./g) were observed in late August 2006, while highest liver concentration (5,144 ng PbTx-3 e q./g) was measured in a pinfish collected in February 2006, just after the dissipation of a K. brevis bloom (Fig. 5.6). Of 226 sciaenid fish analyzed from Tamp a Bay, Sarasota Bay, and Placida Harbor were analyzed, 101 were collected during K. brevis bloom periods. Brevetoxin concentrations during bloom periods averag ed 17 ng PbTx-3 eq./g in muscle, 137 ng PbTx-3 eq./g in gill, 647 ng PbTx-3 eq./g in liver, and 386 ng PbTx-3 eq./g in stomach contents. Similar to what was observed in the pinfish, mean brevetoxin concentrations in sciaenids collected during non-bl oom periods were lower in muscle (4 ng PbTx-3 eq./g), gill (19 ng PbTx-3 eq./g), and stomach contents (185 ng PbTx-3 eq./g), yet were higher in liver (1,114 ng PbTx-3 eq./g). Significant di fferences were found between red tide and non-red tide sciaenid muscle (p < 0.0001), gi ll (p < 0.0001), and stomach contents (p < 0.05), but not liver. In sma ller specimens collected during red tides, brevetoxin concentrations in viscera averaged 589 ng PbTx-3 eq./g and were significantly higher (p < 0.05) than viscera of non-bloom specimens (45 ng PbTx-3 eq./g).

PAGE 113

100 Red drum ( Sciaenops ocellatus) and spotted seatrout were the sciaenid species analyzed in greatest numbers. Brevetoxin co ncentrations measured in the muscle and liver of red drum and spotted seatrout collected in Ta mpa Bay between August 2006 and December 2007 are plotted by date of collecti on in Figures 5.7A and B. In late 2006, a K. brevis bloom was present in the area. Duri ng the bloom period, brevetoxin was only detected in the muscle of one red drum out of 21 tested and wa s present in low but measurable concentrations in the muscle of nine of 12 seatrout. In both species, elevated concentrations of brevetoxins were detect ed in some livers while remaining low or undetectable in others. During 2007, no bloom concentrations of K. brevis were recorded near to the fish collection sites, and br evetoxin was detectable (only at very low concentrations) in the muscle of only one of 24 red drum and one of 74 seatrout analyzed, yet highest liver concentrations for both spot ted seatrout (15,015) and red drum (1,951 ng PbTx-3 eq./g) were measured in May and September 2007, respectively. A Kruskal-Wallis test with Dunn's multip le comparison post-test was used to compare brevetoxin concentrations during K. brevis bloom periods in tissues of the species that were analyzed in greatest numbers: Atlantic th read herring, scaled sardines, pinfish, red drum, and spotted seatrout. In a comparison of brevetoxin concentrations in the muscle of all six species, red drum musc le differed significantly from the muscle of thread herring, scaled sardine, and pinfish (p < 0.001 for all). Spotted seatrout muscle also differed significantly from the muscle of thread herring (p < 0.05) scaled sardine (p < 0.001), and pinfish (p < 0.05). In a comparis on of brevetoxin concentrations in the livers of Atlantic thread herring, pinfish, red drum, and spotted seatrout, red drum liver differed significantly from A tlantic thread herring and pi nfish (both p < 0.001), and from spotted seatrout (p < 0.05). No significant differences were found in a comparison of brevetoxin concentrations in the viscera of Atlantic thread herring, scaled sardines, and pinfish.

PAGE 114

1/1/06 2/1/06 3/1/06 4/1/06 5/1/06 6/1/06 7/1/06 8/1/06 9/1/06 10/1/06 11/1/06 12/1/06 1/1/07 2/1/07 3/1/07 0 2,000 4,000 6,000 20,000 40,000 60,000 80,000PbTx-3 eq. (ng/g) Atlantic thread herring Scaled sardine Spanish sardine 1/1/06 2/1/06 3/1/06 4/1/06 5/1/06 6/1/06 7/1/06 8/1/06 9/1/06 10/1/06 11/1/06 12/1/06 1/1/07 2/1/07 3/1/07 0 100 200 300 400 500 600 700PbTx-3 eq. (ng/g) Atlantic thread herring Scaled sardine Spanish sardine A B Figure 5.5 Brevetoxin concentra tions measured in A) muscle and B) viscera of clupeid fish collected in southwest Florida. Red bars indicate K. brevis bloom periods. 101

PAGE 115

1/1/05 2/1/05 3/1/05 4/1/05 5/1/05 6/1/05 7/1/05 8/1/05 9/1/05 10/1/05 11/1/05 12/1/05 1/1/06 2/1/06 3/1/06 4/1/06 5/1/06 6/1/06 7/1/06 8/1/06 9/1/06 10/1/06 11/1/06 12/1/06 1/1/07 2/1/07 3/1/07 4/1/07 0 40 80 120 160 200PbTx-3 eq. (ng/g) 1/1/05 2/1/05 3/1/05 4/1/05 5/1/05 6/1/05 7/1/05 8/1/05 9/1/05 10/1/05 11/1/05 12/1/05 1/1/06 2/1/06 3/1/06 4/1/06 5/1/06 6/1/06 7/1/06 8/1/06 9/1/06 10/1/06 11/1/06 12/1/06 1/1/07 2/1/07 3/1/07 4/1/07 0 2,000 4,000 6,000PbTx-3 eq. (ng/g) Liver Viscera A B Figure 5.6 Brevetoxin concentratio ns measured in A) muscle and B) liver or viscera of pinfish collected in southwes t Florida. Red bars indicate K. brevis bloom periods. 102

PAGE 116

8/1/06 9/1/06 10/1/06 11/1/06 12/1/06 1/1/07 2/1/07 3/1/07 4/1/07 5/1/07 6/1/07 7/1/07 8/1/07 9/1/07 10/1/07 11/1/07 12/1/07 1/1/08 0 2,000 4,000 6,000 12,000 14,000 16,000PbTx-3 eq. (ng/g) Red drum Seatrout 8/1/06 9/1/06 10/1/06 11/1/06 12/1/06 1/1/07 2/1/07 3/1/07 4/1/07 5/1/07 6/1/07 7/1/07 8/1/07 9/1/07 10/1/07 11/1/07 12/1/07 1/1/08 0 20 40 60 80 100PbTx-3 eq. (ng/g) Red drum Seatrout A BFigure 5.7 Brevetoxin concentra tions measured in A) muscle and B) liver of red drum and spotted seatrout collected in Tampa Bay. Red bars indicate K. brevis bloom periods. 103

PAGE 117

104 Discussion Collection and analyses of fish from St. Joseph Bay in 2004 was initiated in response to the ongoing dolphin mortality event and was based on the gross findings that the dolphins had stomachs packed with fish (NMFS 2004) as well as the initial ELISA results demonstrating extremely high brevetox in concentrations in the dolphin stomach contents (Chapter 4). These collections provide d a cursory look at whether live fish in the bay were contaminated with brevetoxins. A lthough concentrations were lower than those measured in fish recovered from the dolphi n stomachs, the consistent detection of elevated concentrations of brevetoxins in all fish collected in March 2004 was unexpected at the time. Further collectio ns throughout 2005 and 2006 were conducted to examine whether this phenomenon was widespr ead among fish species and to assess how long brevetoxins could persist in tissues of live fish. The average levels of brevetoxins were highest during and just after a two-month K. brevis bloom. Since this was the only time dur ing the two-year co llection period that bloom concentrations of K. brevis were observed in water samples collected from St. Joseph Bay, the data presented here may suggest that brevetoxins continue to be present in the livers of fish for more than a year after the cessation of a bloom (Fig. 5.3). Brevetoxins are known to accumulate in she llfish where they can persist at levels dangerous for human consumers for weeks to months after a K. brevis bloom has ended (Morton and Burklew 1969; Stei dinger et al. 1998a; Pierce et al. 2006). Analyses of contaminated shellfish have shown that the parent toxin PbTx-3 is largely eliminated from shellfish within weeks after a red tide has ended, and that brevetoxins persist for months in the form of metabol ites (Plakas et al. 2002; Plak as et al. 2004; Pierce et al. 2006). However, the surprising levels of brevet oxins measured in the stomach contents of the majority of fish, especially in clupeid species (scaled sardines and Atlantic thread herring), collected in May 2006 when no red tide bloom was present in St. Joseph Bay suggests the more likely explanation is that there were undetected s ources of brevetoxins. The low levels of K. brevis (333-4,000 cells/L) found in tw o water samples collected at the mouth of the bay in May 2006 may be evid ence of a bloom that remained undetected

PAGE 118

105 outside of the bay. Water sampling for K. brevis in northwest Florid a is largely eventdriven, and with the exception of monthly sampling by the FDEP in St. Joseph Bay, little data is available on K. brevis cell densities along the norther n Gulf coat of Florida unless a bloom is suspected and red tide monitoring activities increase. Additionally, since brevetoxins persist in fish food sour ces after dissipation of a red tide, chronic post-bloom exposure of omnivor ous and piscivorous fish to toxins in the marine food web is likely, further complicat ing determination of how long brevetoxins may persist in tissues of fish in the wil d. Regardless of the source, the brevetoxins detected in the stomach contents of fish in May 2006 would certainly have contributed to the persistence of toxins noted in the livers over th e next several months. The data on brevetoxin concentrations in fish collected from St. Joseph Bay demonstrate that the accumulation of brevetoxin in fish in th e wild is a common occurrence and that brevetoxins can persist in fish tissues for an extended period of time. Further transfer of brevetoxi ns through the fish food we b was well-illustrated by the presence of a small contaminated clupeid fish in the stomach of a larger piscivorous seatrout collected in May 2004. Similarly, three cl upeids taken from the stomach of a dead red drum and four clupeids taken from the stomach of a dead hardhead sea catfish collected in St. Joseph Bay during the dolphi n mortality event contained brevetoxin concentrations in their muscle ranging from 495 to 1,356 ng PbTx-3 eq./g. These fish were not included in the fish pr esented here as they were not alive when collected, but the stomach contents of these fish also demons trate trophic transfer of brevetoxins through the fish food web. Prior to the 2004 dolphin mortality, info rmation on brevetoxin accumulation and persistence in fish from cont rolled experiments was very li mited. Vectorial transport of brevetoxins to fish through copepods was demonstrated by Te ster et al. (2000) in shortterm feeding experiments (2-25h). In this study, juvenile fish were fed toxic copepods that had been exposed to K. brevis cultures. They accumulated PbTx-2 and PbTx-3 in both the muscle and viscera, to maximum levels in muscle of approximately 40 ng/fish

PAGE 119

106 and in viscera of 120 ng/fish. The persistence of brevetoxins in the fish over time was not assessed. However, the evidence gathered in the investigation of the 2004 dolphin mortality has spurred broader interest in the topic of brevetoxin accumulation in fish. By exposing juvenile planktivorous striped mullet ( Mugil cephalus ) to K. brevis and omnivorous pinfish and croaker (Micropogonias undulatus ) to brevetoxin-contaminated clams, Naar et al. (2007) further demonstrat ed that fish can accumulate ichthyotoxic brevetoxins when exposed through their diet and th at brevetoxins can be transf erred into the fish food web by feeding on toxic prey. Here, the levels of brevetoxin accumulati on observed in the exposure of juvenile mullet to K. brevis were lower than those achieved by the fish feeding on toxic shellfish, but the exposure duration for the former was restricted to 24 hours due to the conditions requi red to prevent lysing of the K. brevis cells (no filtration or aeration), while the latter was carried out over two weeks. Maximum concentrations of 2,675 ng/g in viscera and 1,540 ng/g in muscle we re measured in pinfish on the last day of exposure. After two weeks without expos ure, the experiment was terminated, but viscera and muscle tissue still containe d 15% and 40% of the maximum observed concentrations, respectively. Transfer of brevetoxins into fish during short-term exposures to K. brevis cultures was also demonstrated by Woofter et al. (2005) who noted immediate uptake of brevetoxin with measurable blood levels as soon as one hour af ter exposing juvenile striped mullet to K. brevis Over the course of the 24hour experiment, maximum blood levels were found after 8-12 hours of exposure, with a drop to 50% of maximum levels after 24 hours. Tissue concentr ations were not determined, but slow elimination was observed with brevetoxin measurable in blood several days after exposure. More recently, and in direct response to the findings of the 2004 dolphin mortality event investigation, longer-term depurat ion studies were performed using two planktivorous fish, juvenile stri ped mullet and Atlantic menhaden ( Brevoortia tyrannus ), exposed to cultures of K. brevis (Hinton and Ramsdell 2008) Depuration of brevetoxins was followed in the mullet over two months and in the menhaden over one month. The investigators analyzed indi vidual organs of the mullet after exposure and found that

PAGE 120

107 brevetoxin content of the muscle and intestine decreased significantly after one week. The toxin content of the live r increased to a maximum af ter one month post-exposure, and after two months was still higher than when the depuration period began. Menhaden were separated into head, body and viscera. Maximum brevetoxin levels measured were 169 ng/g in the body and 1,858 ng/g in viscera. Th e investigators observed a decrease in toxin of at least 50% in the body and head after 17 days, but found no change in the viscera. The persistence of high levels of toxin in the viscera was attributed to redistribution of toxin from the intestine and muscle to the liver, where it is metabolized and re-introduced through bile into the intestines prior to elimination. Toxin content in the whole fish was reduced by 65% after 30 days and a half-time of complete elimination was calculated to be 24 days. Many of the species tested from St. Joseph Bay were not caught during the bloom in the fall of 2005, and in several cases only one fish per species wa s tested, making this data set unsuitable to compare species or to draw conclusions regarding the ability for brevetoxins to accumulate in particular fish species. Analyses of fish collected from southwest Florida where red tides occur mo re frequently were conducted to examine more typical dolphin prey species, and to compare these species with clupeids. Sciaenid species and pinfish, important in the diet of bottlenose dolphins in the Gulf of Mexico (Barros and Odell 1990; Barros and Wells 1998), were capable of accumulating brevetoxins, although to levels generally much lower than those measured in clupeids. However, given their greater size, they may still represent a significant dose of brevetoxin. These fish are not planktivorous, but they are directly exposed to dissolved brevetoxins during blooms and indirectly exposed to brevetoxins through their food sources. For example, pinfish consume large quantities of seagrass throughout much of their life (Montgomery and Targett 1992) and, as described in Chapter 3, brevetoxins can accumulate to high levels in association with seagrass. The highest concentration measured in the GI contents of the pinfis h analyzed here was 5,537 ng PbTx-3 eq./g in a fish collected from a seagrass meadow in Placida Harbor during a bloom.

PAGE 121

108 Although most of the red drum and spotted seatrout were collected from the same general area, mean and maximum brevetoxin concentrations were higher in spotted seatrout for all tissue types. Both of these sc iaenid species both consume a variety of prey items. As juveniles, the diet of both species consists primarily of invertebrates with an increasing importance on fish as they grow larger (Peters and McMi chael 1987; Llanso et al. 1998; Murphy 2003). In a study of these two species at two sites in upper Tampa Bay, small fish were the main prey item for spotted seatrout >200 mm standard length (SL) at both sites, while fish prey gained importanc e for red drum >200 mm but polychaetes and decapod crustaceans remained the main food it ems (Llanso et al. 1998). Most standard lengths of spotted seatrout a nd red drum collected here, as calculated from total lengths using length-length relations hips reported for each sp ecies (Murphy and Taylor 1990; Murphy 2003), were >200 mm. The mean SL of spotted seatrout collected during red tides (256 mm) was similar to the mean SL of those collected during non-bloom periods (283 mm). Only five spotted seatrout were < 200 mm SL and most were >250 mm. However, the mean SL of red drum collected during red tides (540 mm) was almost twice the mean SL of the red drum collected dur ing non-bloom periods (288 mm). Nine red drum were <200 mm SL, and seven of thes e were collected duri ng non-bloom periods. A shift in prey preferen ces with size may contribute to differences seen between red tide and non-red tide red drum. Interestingly, high concentra tions of brevetoxins were measured in the livers of spotted seatrout collected from Tampa Bay in the second half of 2007 in the absence of an inshore bloom (Fig. 5.7). Between June and November, 2007, red tide monitoring along the southwest Flor ida coast detected no K. brevis above background concentrations ( 1,000 cells/L) until late September when lo w concentrations were first noted in samples collected from Lee county (Fig. 5.8) However, two research cruises conducted one month apart, in late September and la te October of 2007, detected and repeatedly sampled a bloom of K. brevis located 5-10 km off of Lee county (Fig. 5.8). When this K. brevis bloom began is unknown. Results from th e seatrout collected from Tampa Bay between June and November 2007 revealed that they had been exposed to brevetoxins. Recent direct exposure to an undetected insh ore or nearshore bloom seems unlikely given

PAGE 122

109 the number of red tide monitoring samples that ar e routinely collected in this part of the state year-round. Direct exposure to the bloom in Lee county is also unlikely as seatrout do not typically migrate great distances (Hendon et al. 2002; Bortone 2003) and the bloom was located approximately 130 km s outh of where the fish were caught. The remaining possibilities are that these concen trations were remnant brevetoxins from exposure six or more months prior, or that these fish were more recently exposed by ingesting prey fish that had migrated north from the area of the bloom. Given that the concentrations measured in seatrout from the second half of 2007 were generally much higher than those measured in seatrout coll ected during the bloom earlier in the year, the latter explanation seems more likely. Brevetoxin was also measured in the GI contents of the seatrout collected between June and November 2007, but overall was low, with an average concentration of 45 ng PbTx-3 e q./g and a maximum value of 209 ng PbTx-3 eq./g. Brevetoxin concentrations in clupeids from southwest Florida we re variable but in some cases were extremely high, demonstrati ng that the excessive toxin concentrations present in the menhaden retrieved from the dolphin stomachs in 2004 were not unique occurrences. Brevetoxin levels exceeded 10,000 ng PbTx-3 eq./g in the viscera of 12 clupeids (representing 3 species), with three fish containing highe r levels than the maximum measured in the menhaden take n from the 2004 dolphin stomachs (up to 69,047 ng PbTx-3 eq./g). In the few instan ces where clupeid liver was collected, concentrations were lower, reaching a maximum of 3,256 ng PbTx-3 eq./g, suggesting that the extreme concentrations of brevetoxi ns found in some clupeid viscera is probably due to the concentration of K. brevis cells in the GI tract whil e feeding in a bloom. Since the gut passage time for planktivorous fish is on the order of hours (Friedland et al. 2005), the highest danger to dolphi ns would likely result from feeding on planktivorous fish within hours of the prey fish feeding on K brevis The persistence of high levels of toxin in the viscera of experimentally exposed menhaden led Hinton and Ramsdell (2008) to suggest that menhaden can remain a vector of lethal levels of brevetoxi ns to dolphins for several weeks after a bloom. However, to measure brevetoxins they employed antibody-b ased assays similar to the ELISA used

PAGE 123

50 0 50Kilometers < 103cells L-1103-105cells L-1105-106cells L-1>106cells L-1 < 103cells L-1103-105cells L-1105-106cells L-1>106cells L-1 Gulf of MexicoManatee Charlotte Sarasota Hardee Pinellas De Soto 1,000 ng/g Liver (PbTx-3 eq.) 1,000 ng/g Liver (PbTx-3 eq.) 50 0 50Kilometers < 103cells L-1103-105cells L-1105-106cells L-1>106cells L-1 < 103cells L-1103-105cells L-1105-106cells L-1>106cells L-1 Gulf of MexicoManatee Charlotte Sarasota Hardee Pinellas De Soto 1,000 ng/g Liver (PbTx-3 eq.) 1,000 ng/g Liver (PbTx-3 eq.)Lee Figure 5.8 Karenia brevis cell concentrations in seaw ater samples and brevetoxin concentrations measured in the livers of s eatrout collected between June and November 2007. The two water samples coded in yellow offshore of Pinellas and Manatee counties (within the area queried for the Tampa Bay seatrout) contained only 1,300 and 2,000 cells/L, and were collected in Sept ember and October 2007, respectively. 110

PAGE 124

111 here, and the composition of brevetoxin c ongeners present in the viscera was not determined. It is important to note that th e ELISA used in these analyses and antibodybased brevetoxin assays used in other st udies do not differentiate the individual congeners and metabolites of brevetoxins, and actual toxicity can not be inferred from the results. Toxin profiles determined by LC-MS fo r some fish collected live from St. Joseph Bay were mainly metabolized brevetoxins with lesser amounts of PbTx-3, while the menhaden from the dolphin stomachs containe d parent brevetoxins PbTx-2 and PbTx-3 with lesser amounts of metabol ites (Naar et al. 2007). Our findings of the presence of the rapidly metabolized parent toxin, PbTx2, in the menhaden taken from the dolphin stomachs in 2004 strongly suggest that the menha den involved in that mortality event had been very recently feeding on a K. brevis bloom. Assessing the risk to dolphins resulting from exposure to brevetoxin metabolites is currently not possible. The toxicity of the numerous brevet oxin metabolites is largely unknown. While they are generally believed to be products of detoxification pathways, it has been shown that some brevetoxin metabolites contribute to NSP toxi city (Ishida et al. 1996; Morohashi et al. 1999; Poli et al. 2000; Nozawa et al. 2003; Plakas et al. 2004). Proper toxicology studies have not yet been c onducted due to a lack of pure brevetoxin metabolites. Indeed, accurate analytical quant ification of known metabol ites in biological samples is also currently prevented by th e lack of standards. Unless pure reference materials are used, brevetoxin metabolite concentrations are suitable for relative comparisons at best. Overall, brevetoxin accumulation in fish was found to be widespread and was documented in fish representing more than 30 different species. Planktivorous clupeid fish are capable of accumulating extremely high concentrations of brevetoxins within their viscera, and their movement can result in spatial separation of a bloom and animal exposure. The window of time for which they present the biggest th reat to dolphins is probably quite small, and acute poisoning events such as the 2004 mortality in the Florida Panhandle are rare most likely because clupeid fish do not typically comprise a

PAGE 125

112 significant portion of the dolphin diet. Sciaen id species and pinfish also accumulated brevetoxins but to a lower extent, which ma y be a result of their indirect exposure through an additional trophic level. These fish, as well as other omnivorous and piscivorous species, may retain brevetoxins in their tissues at significant concentrations after a bloom has dissipated, and may lead to temporal separation of blooms and animal exposure. Karenia brevis blooms are a regular occurrence in southwest Florida and the contamination of typical dolphin prey appears to be as well. The data presented here suggest that dolphins in sout hwest Florida are routinely e xposed to brevetoxins through their diet. Little is known about the effect s of chronic exposure to brevetoxins, but evidence indicates it may lead to suppression of the immune system (Bossart et al. 1998; Benson et al. 2004a; Benson et al. 2004b; Benson et al. 2005; Saye r et al. 2005), which could increase the susceptibility of mari ne mammals to other pathogens. Virtually nothing is known about the effects of e xposure to brevetoxin metabolites. More information is needed on the how brevetoxins are metabolized and on the toxicity of the metabolites to better understand the risks to dolphins of feeding on brevetoxincontaminated fish. (Baden 1989)

PAGE 126

113 Chapter 6. Conclusions The multiple routes through which marine mammals can be exposed to brevetoxins (i.e., direct exposure to bloom wate r, ingestion of brevetoxins in bloom water or in other organisms in which the to xin has bioaccumulated, and inhalation of aerosolized toxins) have complicated efforts to understand the mechanisms that lead to mass mortality events. The 2002 manatee mortality and 2004 bottlenose dolphin mortality provided unique opportunities to ma ke clear connections between ingested brevetoxins and marine mammal mortalities wi thout the confounding issues of concurrent exposure through direct c ontact or inhalation. Prior to 2002, the accumulation of brevetoxins on or in seagrass had never been previously reported, and the delayed or chr onic exposure of manatees to brevetoxins through seagrass was not recognized as a th reat. The 2002 manatee mortality event was the first time that a significant number of mana tee deaths have been positively related to the ingestion of brevetoxin via seagrass subseq uent to a red tide event. Analyses of the epiphytes and detritus on the surface of the seagrass leaves as well as of the cleaned seagrass leaves and rhizomes revealed that during a K. brevis bloom as much as half of the toxin present in the seagrass may be asso ciated with the leaves themselves, while after a bloom, the majority of the toxin present is associated with the epiphytes. Brevetoxins were shown to pers ist in association with seagrass at high levels for weeks and at lower levels for months in the absence of K. brevis Given that K. brevis blooms can be extremely patchy and that discrete water samples can easily misrepresent the overall bloom intensity, brevetoxin levels in seagrass may provide a useful new parameter that can be monitored in addition to K. brevis cell concentrations for assessing the risk of manatees during and following K. brevis blooms.

PAGE 127

114 The 2004 mass mortality of bottlenose dolphi ns in the Florida Panhandle clearly indicated that fish have th e potential to vector brevetoxi ns to higher tropic levels. Analysis of fish collected live from St. Jo seph Bay and southwest Florida revealed that brevetoxin accumulation in fish is a co mmon occurrence. Brevetoxin accumulation was documented in fish representing more than 30 different species. Planktivorous clupeid fish are capable of accumulating extremely high concentrations of brevetoxins within their viscera, and their movement can result in spatial separation of a bloom and animal exposure. Sciaenid species and pinfish also accumulated brevetoxins but to a lower extent. These fish, as well as other omni vorous and piscivorous species, may retain brevetoxins in their tissues at significant concentrations af ter a bloom has dissipated and may lead to temporal separation of blooms and animal exposure. These findings confirm the stability of br evetoxins in the marine environment in the absence of an ongoing K. brevis bloom and reveal the possi bilities of acute lethal exposure for manatees and dolphins through in gestion of seagrass a nd fish, respectively. Additionally, persistent lower levels of brevetoxins were documented in the food sources for manatees and dolphins, suggesting that marine mammals in southwest Florida are chronically exposed to low levels of brevet oxins and brevetoxin metabolites. The effects of chronic dietary exposure to these compounds are unknown, and research is needed to determine whether such exposure can compromise the health of these animals leading to disease or even death. Brevetoxins can accumulate in multiple ti ssues of both manatees and dolphins. Brevetoxin distribution in the 2004 dolphin tissues was similar to what was seen in the 2002 manatees, with highest concentrations m easured in the liver, followed by the kidney and then the lung (Fig 6.1). A Mann-Whitney rank sum test was used to compare brevetoxin concentrations measured in the stomach contents, liver, kidney, and lung of the 2004 dolphins with those measured in the 2002 manatees. Brevetoxins in the stomach contents of the 2004 dolphins were significantly higher than those measured in the 2002 manatee mortality (p < 0.0001). In organ tissu es, brevetoxin concentrations were lower and significantly different in the dolphi n livers (p < 0.0001) and kidneys (p = 0.0357). Brevetoxin concentrations measured in lung tiss ue were not significantly different in the

PAGE 128

n=24 n=13 1 10 100 1,000 10,000 Liver Kidney LungStomach contentsPbTx-3 eq. (ng/g) 2002 Manatees 2004 Dolphinsn=22 n=13 n=24 n=13 n=26 n=18 p < 0.0001 p < 0.0001 p = 0.0357 Figure 6.1 Average brevetoxin concentrations m easured in tissues and stomach contents of manatees from the 2002 mortality event a nd dolphins from the 2004 mortality event. Error bars = standard deviation. 115

PAGE 129

116 two events. Higher stomach contents and lower organ tissue concentrations may reflect the more acute nature of the dolphin mortality (i.e. a briefer period of exposure to higher doses of brevetoxins). Since the marine mammal mortality events discussed here, there have been red tide-related mass mortalities of manatees in 2003, 2005, 2006, and 2007, as well as an additional dolphin mortality event in the Fl orida Panhandle and the first red tide-related mass mortality of dolphins in southwest Florida, both in 2005-2006. The FWC-FWRI documentation of red tide-related manatee mort alities now also includes manatee deaths that occur outside of mass stranding events. ELISA analyses are routinely run on manatee carcasses recovered from southwest Florida when K. brevis is present, when necropsy findings are consistent with brevetoxicosis or when cause of death is unclear. The information reported here and cu rrently being accumulated on brevetoxin levels in manatee and dolphin mortalities associated with K. brevis blooms, as well as background body burdens carried by marine ma mmals in red tide-endemic southwest Florida, will help natural resource ma nagers better understa nd the impacts of K. brevis red tides on marine mammals. In the Manate e Core Biological M odel (CBM) currently used to analyze the viability of this endangered species, the occasional non-mass stranding red tide-related manatee mortalities that occur during red tides are built into annual survival estimates (Runge et al. 2007a). Mass mortality events associated with K. brevis blooms are modeled as natural catastr ophes and are presently estimated in the model as occurring with a 22% frequency in southwest Florida. This estimate is based on the occurrence of five mass mortalit y events since 1980 (1982, 1996, 2002, 2003, and 2005). However, the CBM modelers acknowledg e that the model may be underestimating the effect of this threat on the manatee popul ation if the recent increase frequency of mortality events indicates a tr end (Runge et al. 2007b). They also recognize that inclusion of annual non-catastrophic red tid e manatee mortality would furt her increase the effect of this threat (Runge et al. 2007b). Between 1974 and 2001, red tide was blamed for 5.7% of manatee deaths in Florida. Since then, due to the occurrence of multiple mass mortality events and improved recognition and documenta tion of brevetoxicosis in manatees, the fraction of red tide-related mortalities has increased to 16.5% (Figure 6.2).

PAGE 130

1974-2001 24.5% 3.9% 2.7% 20.6% 4.2% 5.7% 9.4% 29.1% 2002-2007 23.0% 1.1% 1.9% 17.9% 10.0% 16.5% 7.8% 21.8% Watercraft Collision Flood Gate/Canal Lock Human, other Perinatal Cold Stress Red Tide Natural, other Undetermined Figure 6.2 Causes of manatee deaths in Florida from 1974 to 2001 (top, n = 4,368) and from 2002 to 2007 (bottom, n = 2,091). Graphs were created using data provided by the FWC-FWRI Marine Mammal Pathobiology Labora tory. Cold stress includes 10 perinatal deaths between 1974 and 2001 and 26 perinata l deaths between 2002 and 2007 in which cold stress was suspected. 117

PAGE 131

118 The data presented here on brevetoxin concentrations in marine mammal stomach contents and in their food sources, and simila r data from subsequent and future events, may help researchers arrive at estimates for wh at represents a lethal dose of brevetoxins to manatees and dolphins. However, what constitutes a lethal dos e of brevetoxins in marine mammals may not be as relevant as dosages that induce illness, since sufficient physiological insult due to brevetoxicosis ma y have lethal consequences due to their vulnerability as air breathers in an aquatic environment. The investigations of the 2002 manat ee mortality event and the 2004 bottlenose dolphin mortality event benefite d greatly from the development of a rapid, sensitive, and specific ELISA for brevetoxins (Naar et al. 2002) and from the implementation of this method by scientists within FWC-FWRI. The av ailability of this tool enabled rapid brevetoxin analysis of tissue samples from multiple animals and more thorough environmental investigations than were possi ble in previous mortality events. The main limitation of the method is the inability to distinguish between different forms of brevetoxins. Biological samples are likely to contain a mixture of brevetoxins and/or brevetoxin metabolites, and ELISA results do not accurately estimate toxicity. While this is a drawback when investigating brevetoxi ns in marine mammals food sources where the toxicity of the compounds present are extrem ely relevant, this property of the assay is also its greatest strength in the analyses of tissues from exposed animals. The metabolism of brevetoxins in mammals is not yet well understood. The ability to detect the entire suite of the most abundant type of brevetoxins as well as their metabolites at low levels enables confirmation of brev etoxin exposure in mammals that would otherwise go undetected. As the number of brevetoxin me tabolites continues to expand, so does the need for toxicological studies on their poten cy. While this information is lacking, the coupling of rapid sensitive screening assays with structural identification via mass spectrometry can provide a better, but still very incomplete, understanding of the risks posed to marine mammals by food web transfer of brevetoxins.

PAGE 132

119 Literature Cited Abraham, A., Plakas, S.M., Wang, Z., Jester E.L.E., El Said, K.R., Granade, H.R., Henry, M.S., Blum, P.C., Pierce, R.H. a nd Dickey, R.W. (2006) Characterization of polar brevetoxin derivatives isolated from Karenia brevis cultures and natural blooms. Toxicon 48, 104-115. Abraham, A., Plakas, S.M., Flewelling, L.J., El Said, K.R., Jester, E.L.E., Granade, H.R., White, K.D. and Dickey, R.W. (2008) Biomarkers of Neurotoxic Shellfish Poisoning. Toxicon 52, 237-245. Abraham, W.M., Bourdelais, A.J., Ahmed, A ., Serebriakov, I. and Baden, D.G. (2005) Effects of inhaled brevetoxins in allergic airways: toxi n-allergen interactions and pharmacologic intervention. Environmental Health Perspectives 113, 632-637. Ahmed, M.S., Arakawa, O. and Onoue Y. (1995) Toxicity of cultured Chatonella marina. In: P. Lassus, G. Arzul, E. Erard-LeDenn, P. Gentien and C. MarcaillouLeBaut (Eds), Harmful Marine Alga l Blooms, Lavoisier, Paris, pp. 499-504. Anderson, D.M. (1994) Red tides. Scientific American 271, 52-58. Baden, D.G., Mende, T.J. and Block, R.E. (1979) Two similar toxins isolated from Gymnodinium breve. In: D.L. Taylor and H.H. Selig er (Eds), Toxic Dinoflagellate Blooms, Elsevier, New York, pp. 327-334. Baden, D.G. and Mende, T.J. ( 1982) Toxicity of two toxins from the Florida red tide marine dinoflagellate, Ptychodiscus brevis Toxicon 20, 457-461. Baden, D.G. (1983) Marine food-borne dinofla gellate toxins. Intern ational Review of Cytology 82, 99-149. Baden, D.G., Mende, T.J., Walling, J. and Schultz, D.R. (1984) Specific antibodies directed against toxins of Ptychodiscus brevis (Florida's red tide dinoflagellate). Toxicon 22, 783-789. Baden, D.G. and Tomas, C.R. (1988) Variat ions in major toxin composition for six clones of Ptychodiscus brevis Toxicon 26, 961-963. Baden, D.G. (1989) Brevetoxins: unique polyeth er dinoflagellate toxins. FASEB Journal 3, 1807-1817.

PAGE 133

120 Baden, D.G., Fleming, L.E. and Bean, J.A. ( 1995) Marine toxins. In: F.A. DeWolf (Ed), Handbook of Clinical Neurology: Intoxications of the Nervous System. Part H. Natural Toxins and Drugs, Elsevi er Press, Amsterdam, pp. 141-175. Baden, D.G. and Adams, D.J. (2000) Brevet oxins: chemistry, mechanisms of action, and methods of detection. In: L.M. Botana (Ed), Seafood and Freshwater Toxins: Pharmacology, Physiology, and Detection, Marcel Dekker, Inc., New York, pp. 505-566. Baden, D.G., Bourdelais, A.J., Jacocks, H., Mich elliza, S. and Naar, J. (2005) Natural and derivative brevetoxins: historical background, multiplicity, and effects. Environmental Health Perspectives 113, 621-625. Barros, N.B. and Odell, D.K. (1990) F ood habits of bottlenose dolphins in the southeastern United States. In: S. Leatherwood and R.R. Reeves (Eds), The Bottlenose Dolphin, Academic Press, San Diego, CA, pp. 309-328. Barros, N.B. and Wells, R.S. (1998) Prey a nd feeding patterns of resident bottlenose dolphins ( Tursiops truncatus ) in Sarasota Bay, Florida. Journal of Mammalogy 79, 1045-1059. Bates, S.S. (1998) Ecophysiology and meta bolism of ASP toxin production. In: D.M. Anderson, A.D. Cembella and G.M. Hallegraeff (Eds), Physiological Ecology of Harmful Algal Blooms, SpringerVerlag, Heidelberg, pp. 405-426. Bates, S.S., Garrison, D.L. and Horner, R. A. (1998) Bloom dynamics and physiology of domoic-acid-producing Pseudo-nitzschia species. In: D.M. Anderson, A.D. Cembella and G.M. Hallegraeff (Eds), Physiological Ecology of Harmful Algal Blooms, Springer-Verlag, Heidelberg, pp. 267-292. Benson, J.M., Tischler, D.L. and Baden, D.G. (1999) Uptake, tissue distribiution, and excretion of brevetoxin 3 administered to rats by intratracheal instillation. Journal of Toxicology and Environmental Health, Part A 56, 345-355. Benson, J.M., Hahn, F.F., March, T.F., McDonald J., Sopori, M., Seagrave, J., Gomez, A., Bourdelais, A.J., Naar, J., Zaias, J., Bossart, G. and Baden, D.G. (2004a) Inhalation toxicity of brevetoxin 3 in rats expose d for 5 days. Journal of Toxicology and Environmental Health, Part A 67, 1443-1456. Benson, J.M., Hahn, F.F., Tibbetts, B.M., Bowen, L.E., March, T.F., Langley, R.J., Murray, T.F., Bourdelais, A.J., Naar, J., Zaias, J. and Baden, D.G. (2004b) Florida red tide: inhalation toxicity of Karenia brevis extract in rats. In: K.A. Steidinger, J.H. Landsberg, C.R. Tomas and G.A. Vargo (Eds), Harmful Algae 2002, Florida Fish and Wildlife Conservation Commission, Florida Institute of Oceanography, and Intergovernmental Oceanographic Commission of UNESCO, St. Petersburg, FL USA, pp. 502-504.

PAGE 134

121 Benson, J.M., Hahn, F.F., March, T.H., McDona ld, J.D., Gomez, A.P., Sopori, M.J., Bourdelais, A.J., Naar, J., Zaias, J ., Bossart, G.D. and Baden, D.G. (2005) Inhalation toxicity of brevetoxin 3 in rats exposed for twenty-two days. Environmental Health Perspectives 113, 626-631. Bortone, S.A. (2003) Biology of the spotted seatrout. Marine Biol ogy Series, CRC Press, Boca Raton, FL. Bossart, G.D., Baden, D.G., Ewing, R.Y., Roberts, B. and Wright, S.D. (1998) Brevetoxicosis in manatees ( Trichechus manatus latirostris ) from the 1996 epizootic: gross, histologic, and immunohistochemical features. Toxicologic Pathology 26, 276-282. Bossart, G.D. (2002) Pathological features of the Florida manatee cold stress syndrome. Aquatic Mammals 29, 9-17. Bottein Dechraoui, M.Y., Wang, Z. and Rams dell, J.S. (2007) Intrinsic potency of synthetically prepared brevetoxin cystei ne metabolites BTX-B2 and desoxyBTXB2. Toxicon 50, 825-834. Bourdelais, A.J., Tomas, C.R ., Naar, J., Kubanek, J. and Baden, D.G. (2002) New fishkilling alga in coastal Delaware produces neurotoxins. Environmental Health Perspectives 110, 465-470. Bourdelais, A.J. and Baden, D.G. (2004) T oxic brevetoxin complexes are in aqueous solutions. Abstract. Toxicologist 78(1-S), 807. Bourdelais, A.J., Campbell, S., Jacocks, H., N aar, J., Wright, J.L.C., Carsi, J. and Baden, D.G. (2004) Brevenal is a natural inhibito r of brevetoxin action in sodium channel receptor binding assays. Cellular and Molecular Neurobiology 24, 553-563. Bourdelais, A.J., Jacocks, H.M., Wright, J.L., Bigwarfe, P.M., Jr. and Baden, D.G. (2005) A new polyether ladder compound produced by the dinoflagellate Karenia brevis Journal of Natural Products 68, 2-6. Catterall, W.A. and Gainer, M. (1985) Intera ction of brevetoxin A with a new receptor site on the sodium channel. Toxicon 23, 497-504. Cattet, M. and Geraci, J.R. (1993) Distribution and elimina tion of ingested brevetoxin (PbTx-3) in rats. Toxicon 31, 1483-1486. Chang, F.H., Bourdelais, A., Baden, D.G., Ga ll, M., Hulston, D. and Webb, V. (2006) Karenia concordia (Dinophyceae) as a brevetoxin-producer and comparison with two closely related species K. brevisulcata and K. mikimotoi 12th International Conference on Harmful Algae, Cope nhagen, Denmark, Abstract, pp. 51-52.

PAGE 135

122 Cheng, Y.S., Zhou, Y., Irvin, C.M., Pierce, R ., Naar, J., Backer, L.C., Fleming, L.E., Kirkpatrick, B. and Baden, D.G. (2005) Ch aracterization of marine aerosol for assessment of human exposure to brevetoxi ns. Environmental Health Perspectives 113, 638-643. Chou, H.-N. and Shimizu, Y. (1982) A new polyether toxin from Gymnodinium breve Davis. Tetrahedron Letters 23, 5521-5524. Chou, H.-N., Shimizu, Y., VanDuyne, G. and Cl ardy, J. (1985) Isolati on and structure of two new polycyclic ethers from Gymnodinium breve Davis (= Ptychodiscus brevis ). Tetrahedron Letters 26, 2865-2868. Clucas, I. (1997) A study of the options for utilization of bycatch and discards from marine capture fisheries. FAO Fisheries Circular. No. 928. Rome, FAO. Dennison, W.C., Marshall, G.J. and Wigand, C. (1989) Effect of brown tide shading on eelgrass ( Zostera marina L.) distributions. In: E.M. Cosper, V.M. Bricelj and E.J. Carpenter (Eds), Novel Phytoplankton Blooms: Causes and Impacts of Recurrent Brown Tides and Other Unusual Bloo ms, Springer-Verlag, Berlin, pp. 675-692. Dickey, R.W., Plakas, S.M., Jester, E.L.E., El Said, K.R., Johannessen, J.N., Flewelling, L.J., Scott, P., Hammond, D.G., Dolah, F.M.V., Leighfield, T.A., Dachraoui, M.Y.B., Ramsdell, J.S., Pierce, R.H., Henry, M.S., Poli, M.A., Walker, C., Kurtz, J., Naar, J., Baden, D.G., Musser, S.M., White, K.D., Truman, P., Miller, A., Hawryluk, T.P., Wekell, M.M., Stirling, D., Quilliam, M.A. and Lee, J.K. (2004) Multi-laboratory study of five methods fo r the determination of brevetoxins in shellfish tissue extracts. In: K.A. Stei dinger, J.H. Landsberg, C.R. Tomas and G.A. Vargo (Eds), Harmful Algae 2002, Fl orida Fish and Wildlife Conservation Commission, Florida Institute of Oceanography, and Intergovernmental Oceanographic Commission of UNESCO, St. Petersburg, FL USA, pp. 300-302. Dortch, Q., Robichaux, R., Pool, S., Milsted, D., Mire, G., Rabalais, N.N., Soniat, T.M., Fryxell, G.A., Turner, R.E. and Parsons, M.L. (1997) Abundance and vertical flux of Pseudo-nitzschia in the northern Gulf of Me xico. Marine Ecology Progress Series 146, 249-264. Falconer, I.R. (1991) Tumor pr omotion and liver injury cau sed by oral consumption of cyanobacteria. Environmental Toxicology and Water Quality 6, 177-184. Fauquier, D., Flewelling, L.J., Maucher, J., Manire, C., Socha, V., Kinsel, M., Stacy, B., Gannon, J., Ramsdell, J. and Landsberg, J. (in prep) Brevetoxin levels in blood and tissues of naturally exposed sea turtles from Florida. FDEP (1996) A draft summary of selected FMRI activities du ring the 1996 manatee epizootic. Department of Environmenta l Protection, Florida Marine Research Institute.

PAGE 136

123 Feinstein, A., Ceurvels, A.R., Hutton, R.F. and Snoek, E. (1955) Red tide outbreaks off the Florida west coast. Florida State Bo ard of Conservation Report 55-15, Marine Laboratory, University of Miami. Fertl, D. and Leatherwood, S. (1997) Cetacean interaction with tr awls: a preliminary review. Journal of Northwest A tlantic Fishery Science 22, 219-248. Fire, S.E., Fauquier, D., Flewel ling, L.J., Henry, M., Naar, J., Pierce, R. and Wells, R.S. (2007) Brevetoxin exposure in bottlenose dolphins (Tursiops truncatus ) associated with Karenia brevis blooms in Sarasota Ba y, Florida. Marine Biology 152, 827-834. Fleming, L.E., Backer, L.C. and Baden, D.G. (2005) Overview of aerosolized Florida red tide toxins: exposures and effects. Envi ronmental Health Perspectives 113, 618620. Flewelling, L.J., Naar, J.P., Abbott, J.P., Baden, D.G., Barros, N.B., Bossart, G.D., Bottein, M.-Y.D., Hammond, D.G., Haubold, E.M., Heil, C.A., Henry, M.S., Jacocks, H.M., Leighfield, T.A., Pierce, R.H., Pitchford, T.D., Rommel, S.A., Scott, P.S., Steidinger, K.A., Truby, E. W., Van Dolah, F.M. and Landsberg, J.H. (2005) Red tides and marine mamma l mortalities. Nature 435, 755-756. Forrester, D.J., Gaskin, J.M., White, F.H., Thompson, N.P., Quick, J.A., Jr., Henderson, G.E., Woodard, J.C. and Robertson, W.D. (1977) An epizootic of waterfowl associated with a red tide episode in Fl orida. Journal of Wildlife Diseases 13, 160-167. Friedland, K.D., Ahrenholz, D.W. and Haas, L.W. (2005) Viable gut passage of cyanobacteria through the filter-f eeding fish Atlantic menhaden, Brevoortia tyrannus Journal of Plankton Research 27, 715-718. FWC (2007a) Florida manatee ( Trichecus manatus latirostris ) red tide mortality event in southwestern Florida March-April 2 002, Final Report to the Working Group on Marine Mammal Unusual Mortality Ev ents. Florida Fish and Wildlife Conservation Commission, Fish and Wildlife Research Institute. FWC (2007b) Florida manatee management pla n. Florida Fish and Wildlife Conservation Commission, Fish and Wildli fe Research Institute. Geraci, J.R. (1989) Clinical investigation of the 1987-1988 mass mortality of bottlenose dolphins along the U.S. central and sout h Atlantic coast. Final report to the National Marine Fisheries Se rvice and U.S. Navy,Offi ce of Naval Research and Marine Mammal Commissi on, Ontario Veterinary College, Guelph, Ontario. Giner, J.-L., Zhao, H. and Toma s, C. (2008) Sterols and fatty acids of three harmful algae previously assigned as Chattonella Phytochemistry 69, 2167-2171.

PAGE 137

124 Golik, J., James, J.C., Nakanishi, K. and Li n, Y.Y. (1982) The structure of Brevetoxin-C. Tetrahedron Letters 23, 2535-2538. Greenway, M. (1995) Trophic relationships of macrofauna within a Jamaican seagrass meadow and the role of the echinoid Lytechinus variegatus (Lamarck). Bulletin of Marine Science 56, 719-736. Gunter, G., Williams, R.H., Davis, C.C. and Smith, F.G.W. (1948) Catastrophic mass mortality of marine animals and coin cident phytoplankton bloom on the west coast of Florida, November 1946 to August 1947. Ecological Monographs 18, 309-324. Hallegraeff, G.M. (1993) A review of harm ful algal blooms and their apparent global increase. Phycologia 32, 79-99. Haque, S.M. and Onoue, Y. (2002) Variati on in toxin compositions of two harmful raphidophytes, Chattonella antiqua and Chattonella marina, at different salinities. Environmental Toxicology 17, 113-118. Haywood, A.J., Steidinger, K.A., Truby, E. W., Bergquist, P.R., Bergquist, P.L., Adamson, J. and Mackenzie, L. (2004) Comparative morphology and molecular phylogenetic analysis of three new species of the genus Karenia (Dinophyceae) from New Zealand. Journal of Phycology 40, 165-179. Haywood, A.J., Scholin, C.A., Marin, R., Steidi nger, K.A., Heil, C. and Ray, J. (2007) Molecular detection of the br evetoxin-producing dinoflagellate Karenia brevis and closely related species using rRNAtargeted probes and a semiautomated sandwich hybridization assay. Journal of Phycology 43, 1271-1286. Heil, C.A., Mullholland, M.R., Bronk, D.A., Bernhardt, P. and O'Neil, J.M. (2004) Bacterial and size fractionated pr imary production within a large Karenia brevis bloom on the West Florida Shelf. In: K. A. Steidinger, J.H. Landsberg, C.R. Tomas and G.A. Vargo (Eds), Harmfu l Algae 2002, Florida Fish and Wildlife Conservation Commission, Florida Institute of Oceanography, and Intergovernmental Oceanographic Commi ssion of UNESCO, St. Petersburg, FL USA, pp. 38-40. Heil, C.A., Truby, E., Wolny, J., Pigg, R., Ri chardson, R.W., Garrett, M., Haywood, A., Petrik, K., Flewelling, L., Cook, S., Stone E., Steidinger, K.A. and Landsberg, J.H. (2006) The multi-species nature of the 2005 Karenia bloom: implications for management and monitoring in Florid a. 12th International Conference on Harmful Algae, Copenhagen, Denmark, Abstract, pp. 79-80. Hemmert, W.H. (1975) The public health implications of Gymnodinium breve red tides, a review of the literature a nd recent events. In: V.R. LoCicero (Ed), Proceedings of the First International Conference On Toxic Dinoflagellate Blooms,

PAGE 138

125 Massachusetts Science and Technol ogy Foundation, Wakefield, MA, pp. 489497. Hemminga, M.A. and Duarte, C.M. (2000) Seagrass Ecology, Cambridge University Press, Cambridge. Hendon, J.R., Warren, J.R., Franks, J.S. a nd Buchanan, M.V. (2002) Movements of spotted seatrout (Cynoscion nebulosus ) in Mississippi coasta l waters based on tagrecapture. Gulf of Mexico Science 2, 91-97. Hinton, M. and Ramsdell, J.S. (2008) Brevet oxin in two planktivorous fishes after exposure to Karenia brevis : implications for food-web transfer to bottlenose dolphins. Marine Ecology Progress Series 356, 251-258. Hua, Y., Lu, W., Henry, M.S., Pierce, R. H. and Cole, R.B. (1995) On-line highperformance liquid chromatography-elec trospray ionization mass spectrometry for the determination of brev etoxins in "red tide" alg ae. Analytical Chemistry 67, 1815-1823. Hua, Y., Lu, W., Henry, M.S., Pierce, R. H. and Cole, R.B. (1996) On-line liquid chromatography-electrospray ionization mass spectrometry for determination of the brevetoxin profile in natural "red tide" algae blooms. Journal of Chromatography A 750, 115-125. Ingersoll, E. (1882) On the fish mortality in the Gulf of Mexico. Proceedings of the United States National Museum 4, 74-80. Irvine, A.B. (1983) Manatee metabolism and its influence on distri bution in Florida. Biological Conservation 25, 315-334. Ishida, H., Nozawa, A., Totoribe, K., Murama tsu, N., Nukaya, H., Tsuji, K., Yamaguchi, K., Yasumoto, T., Kaspar, H., Berkett, N. and Kosuge, T. (1995) Brevetoxn B1, a new polyether marine toxin from the New Zealand shellfish, Austrovenus stutchburyi Tetrahedron Letters 36, 725-728. Ishida, H., Muramatsu, N., Nukaya, H., Ko suge, T. and Tsuji, K. (1996) Study on neurotoxic shellfish poisoning involving the oyster, Crassostrea gigas in New Zealand. Toxicon 34, 1050-1053. Ishida, H., Nozawa, A., Hamano, H., Naoki, H., Fujita, T., Kaspar, H.F. and Tsuji, K. (2004) Brevetoxin B5, a new brevetox in analog isolated from cockle Austrovenus stutchburyi in New Zealand, the marker for monitoring shellfish neurotoxicity. Tetrahedron Letters 45, 29-33. Ishida, H., Nozawa, A., Nukaya, H., Rhodes, L., McNabb, P., Holland, P.T. and Tsuji, K. (2006) Brevetoxin metabolism in shellfish associated with neurotoxic shellfish poisoning. In: H. Njapeu, S. Trujillo, H.P. van Egmond and D.L. Park (Eds),

PAGE 139

126 Mycotoxins and Phycotoxins: Advan ces in Determination, Toxicology, and Exposure Management, Wageningen Academic Publishers, The Netherlands pp. 297-307. Katunuma, N., Matsunaga, Y., Himeno, K. and Ha yashi, Y. (2003) Insights into the roles of cathepsins in antigen processing and presentation revealed by specific inhibitors. Biological Chemistry 384, 883-890. Kennedy, C.J., Schulman, L.S., Baden, D.G. a nd Walsh, P.J. (1992) Toxicokinetics of brevetoxin PbTx-3 in the gulf toadfish, Opsanus beta following intravenous administration. Aquatic Toxicology 22, 3-13. Khan, S., Arakawa, O. and Onoue, Y. (1996) Neurotoxin production by a chloromonad Fibrocapsa japonica (Raphidophyceae). Journal of the World Aquaculture Society 27, 254-263. Khan, S., Arakawa, O. and Onoue, Y. (1997) Neurotoxins in a toxic red tide of Heterosigma akashiwo (Raphidophyceae) in Kagoshima Bay, Japan. Aquaculture Research 28, 9-14. Kirkpatrick, B., Fleming, L.E., Backer, L.C ., Bean, J.A., Tamer, R., Kirkpatrick, G., Kane, T., Wanner, A., Dalpra, D., Reich, A. and Baden, D.G. (2006) Environmental exposures to Florida red tides: Effects on emergency room respiratory diagnoses admissions. Harmful Algae 5, 526-533. Kreuder, C., Mazet, J., Bossart, G.D., Carpenter, T., Holyoak, M., Elie, M. and Wright, S.D. (2002) Clinicopathologic features of suspected brevetoxicosis in doublecrested cormorants (Phalacrocorax auritus ) along the Florida Gulf coast. Journal of Zoo and Wildlife Medicine 33, 8-15. Kurki-Helasmo, K. and Meriluoto, J. (1998) Microcystin uptake inhibits growth and protein phosphatase activity in mustard ( Sinapis alba L.) seedlings. Toxicon 36, 1921-1926. Landsberg, J.H. and Steidinger, K.A. (1998) A historical review of Gymnodinium breve red tides implicated in mass mortalities of the manatee ( Trichechus manatus latirostris ) in Florida, USA. In: B. Reguera, J. Blanco, M.L. Fernandez and T. Wyatt (Eds), Harmful Algae, Xunta de Galicia and Intergovernmental Oceanographic Commission of UNESCO, pp. 97-100. Landsberg, J.H. (2002) The effects of ha rmful algal blooms on aquatic organisms. Reviews in Fisheries Science 10, 113-390. Landsberg, J.H., Vargo, G.A., Flewelling, L.J. and Wiley, F. (2007) Algal biotoxins. In: N.J. Thomas, D.B. Hunter and C.T. Atkins on (Eds), Infectious Diseases of Wild Birds, Blackwell Pub., Ames, Iowa, pp. 431-455.

PAGE 140

127 Landsberg, J.H., Flewelling, L.J. and Naar, J. (in press) Karenia brevis red tides, brevetoxins in the food web, and imp acts on natural resources: decadal advancements. Layne, J.N. (1965) Observations on marine ma mmals in Florida waters. Bulletin of the Florida State Museum 9, 131-181. Leverone, J.R., Blake, N.J., Pierce, R.H. and Shumway, S.E. (2006) Effects of the dinoflagellate Karenia brevis on larval development in three species of bivalve mollusc from Florida. Toxicon 48, 75-84. Lin, Y.Y., Risk, M., Ray, S.M., Van Engen, D., Clardy, J., Golik, J., James, J.C. and Nakanishi, K. (1981) Isolati on and structure of Brevetoxin B from the "red tide" dinoflagellate Ptychodiscus brevis ( Gymnodinium breve). Journal of the American Chemical Society 103, 6773-6775. Lipscomb, T.P., Schulman, F.Y., Moffett, D. and Kennedy, S. (1994) Morbilliviral disease in Atlantic bottlenose dolphins (Tursiops truncatus ) from the 1987-1988 epizootic. Journal of W ildlife Diseases 30, 567-571. Llanso, R.J., Bell, S.S. and Vose, F.E. ( 1998) Food habits of red drum and spotted seatrout in a restored mangrove impoundment. Estuaries 21, 294-306. MacKenzie, L., Beuzenberg, V., Holland, P., Holland, P., McNabb, P. and Selwood, A. (2004) Solid phase adsorption toxin tracki ng (SPATT): a new monitoring tool that simulates the biotoxin contamination of filter feeding bivalves. Toxicon 44, 901918. Mase, B., Jones, W., Ewing, R., Bossart, G., Van Dolah, F., Leighfield, T., Busman, M., Litz, J., Roberts, B. and Rowles, T. (2000) Epizootic in bottlenose dolphins in the Florida panhandle: 1999. In: C.K. Baer (Ed), Proceedings of the American Association of Zoo Veterans and Intern ational Association for Aquatic Animal Medicine, pp. 522-525. McElhiney, J., Lawton, L.A. and Leifert, C. (2001) Investigations into the inhibitory effects of microcystins on plant growth, and the toxicity of plant tissues following exposure. Toxicon 39, 1411-1420. McFarren, E.F., Tanabe, H., Silva, F.J., Wils on, W.B., Campbell, J.E. and Lewis, K.H. (1965) The occurrence of a ciguatera-l ike poison in oysters, clams, and Gymnodinium breve cultures. Toxicon 3, 111-123. Michelliza, S., Abraham, W.M., Jacocks, H.M., Schuster, T. and Baden, D.G. (2007) Synthesis, modeling, and biological evalua tion of analogues of the semisynthetic brevetoxin antagonist beta-napht hoyl-brevetoxin. Chembiochem 8, 2233-2239.

PAGE 141

128 Montgomery, J.L. and Targett, T.E. (1992) The nutritional role of seagra ss in the diet of the omnivorous pinfish Lagodon rhomboides (L.). Journal of Experimental Marine Biology and Ecology 158, 37-57. Morgan, M.D. and Kitting, C.L. (1984) Productivity and utilization of the seagrass Halodule wrightii and its attached epiphytes. Limnology and Oceanography 29, 1066-1076. Morohashi, A., Satake, M., Murata, K., Naoki H., Kaspar, H.F. and Yasumoto, T. (1995) Brevetoxin B3, a new brevetoxin analog isolated from the greenshell mussels Perna canaliculus involved in neurotoxic shellf ish poisoning in New Zealand. Tetrahedron Letters 36, 8995-8998. Morohashi, A., Satake, M., Naoki, H., Kasp ar, H.F., Oshima, Y. and Yasumoto, T. (1999) Brevetoxin B4 isolated from greenshell mussels Perna canaliculus, the major toxin involved in neurotoxic she llfish poisoning in Ne w Zealand. Natural Toxins 7, 45-48. Morris, P.D., Campbell, D.S., Taylor, T.J. and Freeman, J.I. (1991) Clinical and epidemiological features of Neurotoxic Shellfish Poisoning in North Carolina. American Journal of P ublic Health 81, 471-474. Morton, R.A. and Burklew, M.A. (1969) Flor ida shellfish toxicity following blooms of the dinoflagellate Gymnodinium breve. Technical Series No. 60. Florida Department of Natural Res ources, Marine Research Lab. Murata, K., Satake, M., Naoki, H., Kaspar, H. F. and Yasumoto, T. (1998) Isolation and structure of a new brevetoxin analog, br evetoxin B2, from greenshell mussels from New Zealand. Tetrahedron 54, 735-742. Murphy, M.D. and Taylor, R.G. (1990) Reproduction, growth, and mortality of red drum Sciaenops ocellatus in Florida waters. Fishery Bulletin 88, 531-542. Murphy, M.D. (2003) A stock asse ssment of spotted seatrout Cynoscion nebulosus in Florida: status of the stocks through 2001. Florida Fish and Wildlife Conservation Commission, Florida Marine Research Institute. Music, S.I., Howell, J.T. and Brumback, L.C. (1973) Red tide: its public health implications. Florida Medical Journal 60, 27-29. Naar, J., Bourdelais, A., Tomas, C., Kubanek, J., Whitney, P.L., Flewelling, L.J., Steidinger, K.A., Lancaster, J. and Ba den, D.G. (2002) A competitive ELISA to detect brevetoxins from Karenia brevis (formerly Gymnodinium breve) in seawater, shellfish, and mammalian body fluid. Environmental Health Perspectives 110, 179-185.

PAGE 142

129 Naar, J.P., Flewelling, L.J., Lenzi, A., A bbott, J.P., Granholm, A., Jacocks, H.M., Gannon, D., Henry, M., Pierce, R., Baden, D.G., Wolny, J.L. and Landsberg, J.L. (2007) Brevetoxins, like cigua toxins, are potent ichthyot oxic neurotoxins that accumulate in fish. Toxicon 50, 707-723. NMFS (2004) Interim report on the bottlenose dolphin ( Tursiops truncatus ) unusual mortality event along the Panhandle of Florida, March-April 2004, National Oceanic and Atmospheric Administration, Na tional Marine Fisheries Service and the Florida Fish and Wildlife Conservation Commission. Nozawa, A., Tsuji, K. and Ishida, H. (2003) Implication of brevetoxin B1 and PbTx-3 in neurotoxic shellfish poisoning in New Zealand by isolation and quantitative determination with liquid chromatogr aphy-tandem mass spectrometry. Toxicon 42, 91-103. NSSP (2005) National Shellfish Sanitation Progr am, Guide for the Control of Molluscan Shellfish. US FDA/CFSAN & Interstate Shellfish Sanitation Conference. O'Shea, T.J., Bonde, R.K., Buergelt, C.D. and Od ell, D.K. (1991) An epizootic of Florida manatees associated with a dinoflage llate bloom. Marine Mammal Science 7, 165-179. Pan, Y., Parsons, M.L., Busman, M., Moeller, P.D.R., Dortch, Q., Powell, C.L. and Doucette, G.J. (2001) Pseudo-nitzschia sp. cf. pseudodelicatissima --a confirmed producer of domoic acid from the north ern Gulf of Mexico. Marine Ecology Progress Series 220, 83-92. Parsons, M.L., Scholin, C.A., Miller, P.E., D oucette, G.J., Powell, C.L., Fryxell, G.A., Dortch, Q. and Soniat, T.M. (1999) Pseudo-nitzschia species (Bacillariophyceae) in Louisiana coastal waters: molecular pr obe field trials, genetic variability, and domoic acid analysis. Journal of Phycology 35, 1368-1378. Parsons, M.L. and Dortch, Q. (2002) Sedi mentological evidence of an increase in Pseudo-nitzschia (Bacillariophyceae) abundance in response to coastal eutrophication. Limnology and Oceanography 47, 551-558. Penhale, P.A. (1977) Macrophyte-epiphyte bi omass and productivity in an eelgrass ( Zostera marina L.) community. Journal of E xperimental Marine Biology and Ecology 26, 211-224. Peters, K.M. and McMichael, R.H., Jr. (1987) Early life history of the red drum Sciaenops ocellatus (Pisces: Sciaenidae) in Tampa Bay, Florida. Estuaries 10, 92107. Pflugmacher, S., Codd, G.A. and Steinberg, C. E.W. (1999) Effects of the cyanobacterial toxin microcystin-LR on detoxification enzy mes in aquatic plants. Environmental Toxicology 14, 111-115.

PAGE 143

130 Pflugmacher, S., Wiegand, C., Beattie, K.A ., Krause, E., Steinberg, C.E.W. and Codd, G.A. (2001) Uptake, effects, and metabo lism of cyanobacterial toxins in the emergent reed plant Phragmites australis (Cav.) Trin. ex Steud. Environmental Toxicology and Chemistry 20, 846-852. Pierce, R.H. (1986) Red tide ( Ptychodiscus brevis ) toxin aerosols: a review. Toxicon 24, 955-965. Pierce, R.H., Henry, M., Boggess, S. and Ru le, A. (1989) Marine toxins in marine bubble-generated marine aerosols. In: E. Monahan and M. VanPatten (Eds), The Climate and Health Implications of Bubble-Mediated Sea-Air Exchange, Connecticut Sea Grant Publications (CT-56-89-06), Groton, CT, pp. 27-42. Pierce, R.H., Henry, M.S., Proffitt, L.S. and Hasbrouck, P.A. (1990) Red tide toxin (brevetoxin) enrichment in marine aerosol. In: E. Granli, B. Sundstrm, L. Edler and D.M. Anderson (Eds), Toxic Marine Phytoplankton, Else vier, New York, pp. 397-402. Pierce, R.H., Henry, M., Blum, P.C. and Payne, S. (2001) Gymnodinium breve toxins without cells: intra-cellula r and extra-cellular toxins. In: G.M. Hallegraeff, S.I. Blackburn, C.J. Bolch and R.J. Lewi s (Eds), Harmful Algal Blooms 2000, International Oceanographic Commission of UNESCO, Paris, pp. 421-424. Pierce, R.H., Henry, M.S., Blum, P.C., Hame l, S.L., Kirkpatrick, B., Cheng, Y.S., Zhou, Y., Irvin, C.M., Naar, J., Weidner, A., Fleming, L.E., Backer, L.C. and Baden, D.G. (2005) Brevetoxin composition in wa ter and marine aerosol along a Florida beach: assessing potential human exposure to marine biotoxins. Harmful Algae 4, 965-972. Pierce, R.H., Henry, M.S., Blum, P.C., Plakas S.M., Granade, H.R., Jester, E.L.E., ElSaid, K.R., Dickey, R.W., Steidinger, K.A., Scott, P.S., Flewelling, L.J. and Wright, J.L.C. (2006) Comparison of met hods for determination of brevetoxins and their metabolites in NSP-toxic biva lved molluscs. In: K. Henshilwood, B. Deegan, T. McMahon, C. Cusack, S. Keav eney, J. Silke, M. O'Cinneide, D. Lyons and P. Hess (Eds), Proceedings of the 5th Interna tional Conference on Molluscan Shellfish Safety, Galway Ireland, June 14-18, 2004, The Marine Institute, Rinville, Oran more, Galway, Ireland, pp. 37-42. Pierce, R.H., Henry, M. and Blum, P. ( 2008) Brevetoxin abundance and composition during ECOHAB-Florida field monitoring cruises in the Gulf of Mexico. Continental Shelf Research 28, 45-58. Plakas, S.M., El-Said, K.R., Jester, E.L., Gran ade, H.R., Musser, S.M. and Dickey, R.W. (2002) Confirmation of brevetoxin metabolism in the Eastern oyster ( Crassostrea virginica ) by controlled exposures to pure toxins and to Karenia brevis cultures. Toxicon 40, 721-729.

PAGE 144

131 Plakas, S.M., Wang, Z., El-Said, K.R., Jester E.L.E., Granade, H.R., Flewelling, L., Scott, P. and Dickey, R.W. (2004) Brevetoxin metabolism and elimination in the Eastern oyster (Crassostrea virginica ) after controlled exposures to Karenia brevis Toxicon 44, 677-685. Poli, M.A., Mende, T.J. and Baden, D.G. (1986) Brevetoxins, unique activators of voltage-sensitive sodium channels, bind to specific sites in rat brain synaptosomes. Molecular Pharmacology 30, 129-135. Poli, M.A., Templeton, C.B., Thompson, W.L. and Hewetson, J.F. (1990) Distribution and elimination of brevetoxin Pb Tx-3 in rats. Toxicon 28, 903-910. Poli, M.A., Musser, S.M., Dickey, R.W., Eilers, P.P. and Hall, S. (2000) Neurotoxic shellfish poisoning and brevetoxin me tabolites: a case study from Florida. Toxicon 38, 981-993. Prasad, A.V.K. and Shimizu, Y. (1989) The st ructure of hemibrevetox in-B: a new type of toxin in the Gulf of Mexico red tide organism. Journal of the American Chemical Society 111, 6476-6477. Quick, J.A., Jr. and Henderson, G.E. (1974) Effects of Gymnodinium breve red tides on fishes and birds: a preliminary repor t on behavior, anatomy, hematology, and histopathology. In: R.L. Amborski, M.A. Hood and R.R. Miller (Eds), Proceedings of the Gulf Coast Region al Symposium on Diseases of Aquatic Animals, Louisiana Sea Grant: Loui siana State University, pp. 85-113. Radwan, F.F.Y., Wang, Z. and Ramsdell, J.S. (2005) Identification of a rapid detoxification mechanism for brevetoxin in rats. Toxicological Sciences 85, 839846. Radwan, F.F. and Ramsdell, J.S. (2006) Ch aracterization of in vitro oxidative and conjugative metabolic pathways for brevet oxin (PbTx-2). Toxicological Sciences 89, 57-65. Radwan, F.F. and Ramsdell, J.S. (2008) Br evetoxin forms covalent DNA adducts in rat lung following intratracheal exposure. E nvironmental Health Perspectives 116, 930-936. Ramsdell, J.S. (2008) The molecular and integrative basis to brevetoxin toxicity. In: L.M. Botana (Ed), Seafood and Freshwater Toxins: Pharmacology, Physiology, and Detection. Second Edition, CRC Press, Boca Raton, pp. 519-550. Redlow, T., Foley, A. and Singel, K. (2002) Sea turtle mortality associated with red tide events in Florida. In: W.N. Witzell (Ed), 22nd Annual Symposium on Sea Turtle Biology and Conservation, Miami, Florida, USA, p. 272.

PAGE 145

132 Reep, R.L. and Bonde, R.K. (2006) The Florid a Manatee, University Press of Florida, Gainesville. Reynolds, J.E., Wells, R.S. and Eide, S.D. (2000) The Bottlenose Dolphin, University Press of Florida, Gainesville. Reynolds, J.E., III and Powell, J.A. (2002) Manatees. In: W.F. Perrin, B. Wursig and J.G.M. Thewissen (Eds), Encyclopedia of Marine Mammals, Academic Press, San Diego, pp. 709-720. Reynolds, J.E., III and Wells, R.S. (2003) Dolphi ns, whales, and manatees of Florida: a guide to sharing their world, Univers ity Press of Florida, Gainesville. Risk, M., Lin, Y.Y., Sadagopa Ramanujam, V.M., Smith, L.L., Ray, S.M. and Trieff, N.M. (1979) High pressure liquid chromat ographic separation of two major toxic compounds from Gymnodinium breve Davis. Journal of Chromatographic Science 17, 400-405. Roth, P.B., Twiner, M.J., Wang, Z., Bottein De chraoui, M.Y. and Doucette, G.J. (2007) Fate and distribution of brevetoxin (PbTx) following lysis of Karenia brevis by algicidal bacteria, including analysis of open A-ring derivatives. Toxicon 50, 1175-1191. Runge, M.C., Sanders-Reed, C.A. and Fonnesbeck, C.J. (2007a) A core stochastic population projection model for Florida manatees ( Trichechus manatus latirostris ), Open-File Report 2007-1082. U.S. Department of the Interior, U.S. Geological Survey. Runge, M.C., Sanders-Reed, C.A., Langtimm, C.A. and Fonnesbeck, C.J. (2007b) A quantitative threats analysis for the Florida manatee ( Trichechus manatus latirostris ), Open-File Report 2007-1086. U.S. Department of the Interior, U.S. Geological Survey. Sayer, A., Hu, Q., Bourdelais, A.J., Baden, D.G. and Gibson, J.E. (2005) The effect of brevenal on brevetoxin-induced DNA damage in human lymphocytes. Archives of Toxicology 79, 683-688. Sayer, A.N., Hu, Q., Bourdelais, A.J., Baden, D.G. and Gibson, J.E. (2006) The inhibition of CHO-K1-BH4 cell prolif eration and induction of chromosomal aberrations by brevetoxins in vitr o. Food and Chemical Toxicology 44, 10821091. Scholin, C.A., Gulland, F., Doucette, G.J., Benson, S., Busman, M., Chavez, F.P., Cordaro, J., DeLong, R., De Vogelaere, A., Harvey, J., Haulena, M., Lefebvre, K., Lipscomb, T., Loscutoff, S., Lowenstin e, L.J., Marin, R., 3rd, Miller, P.E., McLellan, W.A., Moeller, P.D., Powell, C.L., Rowles, T., Silvagni, P., Silver, M.,

PAGE 146

133 Spraker, T., Trainer, V. and Van Dolah, F.M. (2000) Mortality of sea lions along the central California coast linked to a toxic diatom bloom. Nature 403, 80-84. Selwood, A.I., Ginkel, R., Wilkins, A.L., Munday, R., Ramsdell, J.S., Jensen, D.J., Cooney, J.M. and Miles, C.O. (2008) Se misynthesis of S-desoxybrevetoxin-B2 and brevetoxin-B2, and assessment of their acute toxicities. Chem Res Toxicol 21, 944-950. Shea, D., Tester, P., Cohen, J., Kibler, S. and Varnam, S. (2006) Accumulation of brevetoxins by passive sampling devices. African Journal of Marine Science 28, 379-381. Shimizu, Y., Chou, H.N., Bando, H., Van Duyne, G. and Clardy, J.C. (1986) Structure of brevetoxin A (GB-1 Toxin), th e most potent toxin in the Florida red tide organism Gymnodinium breve ( Ptychodiscus brevis). Journal of the American Chemical Society 108, 514-515. Shumway, S.E. (1990) A review of the e ffects of algal blooms on shellfish and aquaculture. Journal of the Wo rld Aquaculture Society 21, 65-104. Siegl, G., MacKintosh, C. and Stitt, M. (1990) Sucrose-phosphate synthase is dephosphorylated by protein phosphatase 2A in spinach leaves. Evidence from the effects of okadaic acid and microcystin. FEBS Letters 270, 198-202. Smayda, T.J. (1997a) Harmful algal blooms: their ecophysiology a nd general relevance to phytoplankton blooms in the sea. Limnology and Oceanography 42, 11371153. Smayda, T.J. (1997b) What is a bloom? A commentary. Limnology and Oceanography 42, 1132-1136. Smith, J.C., Odense, P., Angus, R., Bates, S.S., Bird, C.J., Cormier, P., de Freitas, A.S.W., Leger, C., O'Neil, D., Pauley, K. and Worms, J. (1990) Variation in domoic acid levels in Nitzschia species: implications for monitoring programs. Bulletin of the Aquaculture A ssociation of Canada 90, 27-31. Starr, T.J. (1958) Notes on a toxin from Gymnodinium breve Texas Reports on Biology and Medicine 16, 500-507. Steidinger, K.A., Burklew, M.A. and Ingle, R.M. (1973) The effects of Gymnodinium breve toxin on estuarine animals. In: D.F. Ma rtin and G.M. Padilla (Eds), Marine Pharmacognosy, Academic Press, New York, pp. 179-202. Steidinger, K.A. (1975) Basic factors influencing red tides. In: V.R. LoCicero (Ed), Proceedings of the First Internati onal Conference on Toxic Dinoflagellate Blooms, Massachusetts Science and Technology Foundation, Wakefield, Massachusetts, pp. 153-162.

PAGE 147

134 Steidinger, K.A. and Haddad, K. (1981) Biologic and hydrographic aspects of red tides. BioScience 31, 814-819. Steidinger, K.A., Carlson, P., Baden, D., Rodriguez, C. and Seagle, J. (1998a) Neurotoxic shellfish poisoning due to t oxin retention in the clam Chione cancellata In: B. Reguera, J. Blanco, M.L. Fernandez and T. Wyatt (Eds), Harmful Algae, Xunta de Galicia and Intergovernmental O ceanographic Commission of UNESCO, pp. 457-458. Steidinger, K.A., Vargo, G.A., Tester, P.A. and Tomas, C.R. (1998b) Bloom dynamics and physiology of Gymnodinium breve with emphasis on the Gulf of Mexico. In: D.M. Anderson, A.D. Cembella and G.M. Hallegraeff (Eds), Physiological Ecology of Harmful Algal Blooms, Sp ringer-Verlag, Heidelberg, pp. 133-153. Steidinger, K.A. and Melton Penta, H.L. (1999) Harmful microalgae and associated public health risks in the Gulf of Mexico: Florida Marine Research Institute, Department of Environmental Protec tion EPA Grant #MX004729-95-0 [US EPA: Gulf of Mexico Program]. Sudarsanam, S., Virca, G.D., March, C.J. a nd Srinivasan, S. (1992) An approach to computer-aided inhibitor desi gn: application to cathepsin L. Journal of ComputerAided Molecular Design 6, 223-233. Summerson, H.C. and Peterson, C.H. (1990) Recruitment failure of the bay scallop, Argopecten irradians concentricus during the first red tide, Ptychodiscus brevis outbreaks recorded in North Carolina. Estuaries 13, 322-331. Tester, P.A., Stumpf, R.P., Vukovich, F.M., Fowler, P.K. and Turner., J.T. (1991) An expatriate red tide bloom: transport, di stribution, and persistence. Limnology and Oceanography 36, 1053-1061. Tester, P.A. and Stei dinger, K.A. (1997) Gymnodinium breve red tide blooms: initiation, transport, and consequences of surface circulation. Limnology and Oceanography 42, 1039-1051. Tester, P.A., Turner, J.T. and Shea, D. (2000) Vectorial transport of toxins from the dinoflagellate Gymnodinium breve through copepods to fish. Journal of Plankton Research 22, 47-61. Tomasko, D.A. and Lapointe, B.E. (1991) Productivity and biomass of Thalassia testudinum as related to water column nutrient availability an d epiphyte levels: field observations and expe rimental studies. Marine Ecology Progress Series 75, 9-17. Tomasko, D.A., Dawes, C.A. and Hall, M. O. (1996) The effects of anthropogenic nutrient enrichment on turtle grass ( Thalassia testudinum ) in Sarasota Bay, Florida. Estuaries 19, 448-456.

PAGE 148

135 Trainer, V.L. and Baden, D.G. (1991) An enzyme immunoassay for the detection of Florida red tide brevetoxins. Toxicon 29, 1387-1394. Van Dolah, F.M., Finley, E.L., Haynes, B.L., Do ucette, G.J., Moeller, P.D. and Ramsdell, J.S. (1994) Development of rapid and sensitive high throughput pharmacological assays for marine phycotoxins. Natural Toxins 2, 189-196. Van Dolah, F.M. (2000) Marine al gal toxins: origins, health effects, and their increased occurrence. Environmental Health Perspectives 108 (S1), 133-141. Van Dolah, F.M., Doucette, G.J., Gulland, F.M.D., Rowles, T.L. and Bossart, G.D. (2003) Impacts of algal toxins on marine mammals. In: J.G. Vos, G.D. Bossart, M. Fournier and T.J. O'Shea (Eds), T oxicology of Marine Mammals, Taylor & Francis, New York, pp. 247-269. van Montfrans, J., Wetzel, R.L. and Orth, R. J. (1984) Epiphyte-grazer relationships in seagrass meadows: consequences for seagrass growth and production. Estuaries 7, 289-309. Vargo, G.A., Heil, C.A., Ault, D.A., Neely, M.B., Murasko, S., Havens, J., Lester, K.M., Dixon, L.K., Merkt, R., Walsh, J., Weisber g, R. and Steidinger, K.A. (2004) Four Karenia brevis blooms: a comparative analysis. In: K.A. Steidinger, J.H. Landsberg, C.R. Tomas and G.A. Vargo (Eds), Harmful Algae 2002, Florida Fish and Wildlife Conservation Commission, Fl orida Institute of Oceanography, and Intergovernmental Oceanographic Commi ssion of UNESCO, St. Petersburg, FL USA, pp. 14-16. Vargo, G.A., Heil, C.A., Fanning, K.A., Dixon, L.K., Neely, M.B., Lester, K., Ault, D., Murasko, S., Havens, J., Walsh, J. and Bell, S. (2008) Nutrient availability in support of Karenia brevis blooms on the central West Fl orida Shelf: What keeps Karenia blooming? Continental Shelf Research 28, 73-98. Vieytes, M.R., Louzao, M.C., Alfonso, A., Cabado, A.G. and Botana, L.M. (2000) Mechanism of action and toxicology. In: L.M. Botana (Ed), Seafood and Freshwater Toxins: Pharmacology, Phys iology, and Detection, Marcel Dekker, Inc., New York, pp. 239-256. Walsh, C.J., Leggett, S.R., Strohbehn, K., Pierce, R.H. and Sleasman, J.W. (2008) Effects of in vitro brevetoxin exposure on apoptosis and cellular metabolism in a leukemic T cell line (Jurkat). Marine Drugs 6, 291-307. Walsh, J.J. and Steidinger, K.A. (2001) Sa haran dust and Florida red tides: The cyanophyte connection. Journal of Geophysical Research 106, 11597-11612. Walsh, J.J., Jolliff, J.K., Darrow, B.P., Lene s, J.M., Milroy, S.P., Remsen, A., Dieterle, D.A., Carder, K.L., Chen, F.R., Vargo, G.A., Weisberg, R.H., Fanning, K.A., Muller-Karger, F.E., Shinn, E., Steidinger, K.A., Heil, C.A., Tomas, C.R.,

PAGE 149

136 Prospero, J.S., Lee, T.N., Kirkpatrick, G.J., Whitledge, T.E., Stockwell, D.A., Villareal, T.A., Jochens, A.E. and Bontempi P.S. (2006) Red tides in the Gulf of Mexico: Where, when, and why? Journal of Geophysical Research 111, C11003, doi:11010.11029/12004JC002813. Wang, Z., Plakas, S.M., Said, K. R.E., Jester, E.L.E., Granade, H.R. and Dickey, R.W. (2004) LC/MS analysis of brevetoxin metabolites in the Eastern oyster ( Crassostrea virginica ). Toxicon 43, 455-465. Watkins, S.M., Reich, A., South, R., Te rzagian, R., Hammond, R., Blackmore, C., Flewelling, L., Plakas, S.M., Abraham, A., Dickey, R.W. and Landsberg, J. (2007) Features of neurotoxic shellfish poisoning from recreationally harvested clams in Florida, 2006: Epidemiologic a nd clinical factors. American Public Health Association 135th Annual M eeting and Expo, Washington D.C., November 3-7, 2007. Watkins, S.M., Reich, A., Fleming, L.E. and Hammond, R. (2008) Neurotoxic Shellfish Poisoning. Marine Drugs 6, 431-455. Wells, M.L., Trick, C.G., Cochlan, W.P., Hughes, M.P. and Trainer, V.L. (2005) Domoic acid: the synergy of iron, copper, and the toxicity of diatoms. Limnology and Oceanography 50, 1908-1917. Woodcock, A.H. (1948) Note concerning human respiratory irritation associated with high concentrations of plankton and mass mo rtality of marine organisms. Journal of Marine Research 7, 56-62. Woofter, R.T., Brendtro, K. and Ramsdell, J.S. (2005) Uptake and elimination of brevetoxin in blood of striped mullet (Mugil cephalus ) after aqueous exposure to Karenia brevis Environmental Health Perspectives 113, 11-16. Work, T.M., Barr, B., Beale, A.M., Fritz, L ., Quilliam, M.A. and Wright, J.L.C. (1993) Epidemiology of domoic acid poisoning in brown pelicans ( Pelecanus occidentalis ) and Brandt's cormorants ( Phalacrocorax penicillatus ) in California. Journal of Zoo and Wildlife Medicine 24, 54-62. Wrsig, B., Jefferson, T.A. and Schmidly, D.J. (2000) The marine mammals of the Gulf of Mexico, Texas A&M Universi ty Press, College Station. Yoshizawa, S., Matsushima, R., Watanabe, M. F., Harada, K., Ichihara, A., Carmichael, W.W. and Fujiki, H. (1990) Inhibition of protein phosphatases by microcystins and nodularin associated with hepatotoxi city. Journal of Cancer Research and Clinical Oncology 116, 609-614. Zieman, J.C. and Zieman, R.T. (1989) The eco logy of the seagrass meadows of the west coast of Florida: a community profile. U.S. Fish and Wildlife Service Biological Report 85(7.25).

PAGE 150

137 Appendices

PAGE 151

Appendix A. 2002 red tide-suspect manatees tested for brevetoxins by ELISA. Data provided by FWC-FWRI Marine Mammal Pathobiology Laboratory. Carcass Report Date Field IDSexLength (cm)CountyWaterway 3/15/2002MSW0239M321CharlotteStump Pass 3/17/2002MSW0240M268Lee Pine Island Sound 3/19/2002MSW0242M315Lee Pine Island Sound 3/19/2002MSW0243F220CharlotteAinger Creek 3/21/2002MSW0244M254Lee Pine Island Sound 3/22/2002MSW0245M344SarasotaBlackburn Bay 3/22/2002MSW0246M287Lee Charlotte Harbor 3/24/2002MSW0248M265CharlotteLemon Bay 3/24/2002MSW0249F250Co llierJ ohnson Bay 3/24/2002MSW0250F236CharlotteLemon Bay 3/24/2002MSW0251M276SarasotaForked Creek 3/25/2002MSW0252M260SarasotaForked Creek 3/25/2002MSW0253M327SarasotaLemon Bay 3/25/2002MSW0254M295CharlotteGulf of Mexico 3/25/2002MSW0255M300Lee San Carlos Bay 3/25/2002MSW0256M348SarasotaGulf of Mexico 3/26/2002MSW0257F302CharlotteStump Pass 4/5/2002MSW0261F264SarasotaForked Creek 4/6/2002MSW0262M274Lee Matlacha Pass 4/6/2002MSW0263F281SarasotaForked Creek 4/7/2002MSW0264M247Lee Pelican Bay 4/7/2002MSW0265M294Lee Pine Island Sound 4/8/2002MSW0267F269CharlotteLemon Bay 4/15/2002MSW0269M262CharlotteCharlotte Harbor 4/16/2002MSW0270F229Lee Pine Island Sound 4/30/2002MSW0272M231SarasotaSarasota Bay 138

PAGE 152

Appendix B. 1996 red tide-suspect manatees for which archived tissues were tested for brevetoxins by ELISA. Data provided by FW C-FWRI Marine Mammal Pathobiology Laboratory. Carcass Report Date Field IDSexLength (cm)CountyWaterway 3/12/1996MSW9647 F297CollierCaxambas Pass 3/12/1996MSW9649 M302Lee Pine Island Sound 3/12/1996MSW9650 M289Lee Charlotte Harbor 3/13/1996MSW9655 F243Lee Orange River 3/13/1996MSW9656 F306Lee Orange River 3/14/1996MSW9664 M321Lee Matlacha Pass 3/15/1996MSW9667 M288SarasotaLemon Bay 3/15/1996MSW9673 M307Lee Matlacha Pass 3/16/1996MSW9671 F268Lee Caloosahatchee River 3/16/1996MSW9672 M285Lee San Carlos Bay 3/16/1996MSW9674 F295Lee Pine Island Sound 3/16/1996MSW9675 M244CharlotteCharlotte Harbor 3/16/1996MSW9676 M284Lee Pine Island Sound 3/17/1996MSW9681 M304Lee San Carlos Bay 3/17/1996MSW9684 M293Lee Caloosahatchee River 3/22/1996MSW9698 F288Lee Caloosahatchee River 3/22/1996MSW9699 F273CollierSanctuary Sound 3/22/1996MSW96100 M266CollierRoberts Bay 3/23/1996MSW96105 F290Lee Charlotte Harbor 3/24/1996MSW96106 F308CollierJohnson Bay 3/24/1996MSW96107 M323CharlotteGasparilla Sound 3/25/1996MSW96111 M321SarasotaBlackburn Bay 3/26/1996MSW96116 F255CollierGulf of Mexico 3/28/1996MSW96120 F238Lee Hendry Creek 3/30/1996MSW96122 M301Lee Gulf of Mexico 3/30/1996MSW96124 M306Lee Pine Island Sound 3/30/1996MSW96126 F295CollierTarpon Bay 3/30/1996MSW96127 M286CharlotteLemon Bay 4/1/1996MSW96129 F299CollierCaxambas Pass 4/1/1996MSW96130 M293CharlottePlacida Harbor 4/1/1996MSW96134 M214CollierRoberts Bay 4/2/1996MSW96131 M255Lee Pine Island Sound 4/2/1996MSW96133 F276CollierSmokehouse Bay 4/2/1996MSW96135 M207CollierNaples Bay 4/2/1996MSW96136 M274Lee Estero Bay 4/3/1996MSW96137 F273CollierCoon Key Pass 4/3/1996MSW96142 M328CharlotteCharlotte Harbor 4/4/1996MSW96144 M297CollierGullivan Bay 4/4/1996MSW96145 M312Lee Matlacha Pass 139

PAGE 153

Appendix B (Continued). Carcass Report Date Field IDSexLength (cm)CountyWaterway 4/5/1996MSW96146 F302CollierJ ohnson Bay 4/5/1996MSW96147 M219CollierGor don Pass 4/5/1996MSW96148 F291CollierBarfield Bay 4/6/1996MSW96149 F316CharlotteLemon Bay 4/6/1996MSW96150 F239Lee Imperial River 4/7/1996MSW96152 F284CharlotteGasparilla Sound 4/7/1996MSW96153 M254Lee Matlacha Pass 4/9/1996MSW96158 M307CollierCollier Bay 4/11/1996MSW96165 F298CollierSanctuary S ound 4/13/1996MSW96167 M310Lee Charlotte Harbor 4/13/1996MSW96169 F288CollierGullivan Bay 4/14/1996MSW96170 M252CollierGulf of Mexico 4/14/1996MSW96171 F215CollierGullivan Bay 4/16/1996MSW96173 F274Lee Pine Island Sound 4/17/1996MSW96175 F309Lee Caloosahatchee River 4/19/1996MSW96176 M268Lee Pine Island Sound 4/20/1996MSW96177 F224Lee Estero Bay 4/22/1996MSW96181 F275Lee Caloosahatchee River 4/25/1996MSW96185 M295Lee San Carlos Bay 4/27/1996MSW96187 M223CollierGullivan Bay 4/27/1996MSW96189 M246Lee Matlacha Pass 5/9/1996MSW96191 M265CollierSmokehouse Creek 140

PAGE 154

Appendix C. 2002 southwest Florida manatees killed by watercraft co llision that were tested for brevetoxins by ELISA. Data provided by FWC-FWRI Marine Mammal Pathobiology Laboratory. Carcass Report Date Field IDSexLength (cm)CountyWaterway 4/8/2002MSW0266M322SarasotaMyakka River 4/15/2002MSW0268M267CharlotteCharlotte Harbor 4/25/2002MSW0271F211SarasotaBlackburn Bay 5/5/2002MSW0274F329Co llierJ ohnson Bay 5/26/2002MSW0278F286CharlotteLemon Bay 5/27/2002MSW0279F307Lee Pine Island Sound 7/6/2002MSW0286M323SarasotaRobert's Bay 7/20/2002MSW0289M315CharlotteCharlotte Harbor 7/21/2002MSW0290M304Lee San Carlos Bay 8/9/2002MSW0291F309Lee Bull Bay 9/8/2002MSW0293M323Lee Caloosahatchee River 141

PAGE 155

Appendix D. 2002 east coast Florida manatees tested for brevetoxins by ELISA. Data provided by FWC-FWRI Marine Mammal Pathobiology Laboratory. Carcass Report Date Field IDSexLength (cm)County Waterway 2/24/2002MSE0215M215MonroeTarpon Basin 3/17/2002MEC0220M283Miami-DadeBlue Lagoon 3/26/2002MNE0211 F283DuvalSt. Johns River 3/26/2002MSE0222 F201Palm BeachLake Worth 3/29/2002MSE0223 M283Palm BeachLake Worth Creek 4/7/2002MEC0223 M290BrevardPort Canaveral Barge Canal 6/19/2002MNE0218 M302St. JohnsRobinson Creek 8/18/2002MEC0244 M277BrevardIndian River 142

PAGE 156

143 Appendix E. 2004 dolphins tested for brevetoxins by ELISA. Data provided by NMFS. Strand Date Field IDSexLength (cm)CountyWaterway 3/10/2004Hubbs-0415-TtF255GulfSt. Joseph Bay 3/10/2004Hubbs-0430-TtM184GulfSt. Joseph Bay 3/10/2004Hubbs-0431-TtF247GulfSt. Joseph Bay 3/10/2004Hubbs-0433-TtF228GulfSt. Joseph Bay 3/11/2004Hubbs-0416-TtM200GulfSt. Joseph Bay 3/11/2004Hubbs-0418-TtM225.5GulfSt. Joseph Bay 3/11/2004Hubbs-0419-TtM246.5GulfSt. Joseph Bay 3/11/2004Hubbs-0420-TtF244GulfSt. Joseph Bay 3/11/2004Hubbs-0421-TtM225GulfSt. Joseph Bay 3/11/2004Hubbs-0422-TtM241GulfSt. Joseph Bay 3/11/2004Hubbs-0423-TtM229GulfSt. Joseph Bay 3/11/2004Hubbs-0424-TtF243GulfSt. Joseph Bay 3/11/2004Hubbs-0429-TtM254.2GulfSt. Joseph Bay 3/11/2004Hubbs-0432-TtM222GulfSt. Joseph Bay 3/12/2004Hubbs-0425-TtF243GulfSt. Joseph Bay 3/12/2004Hubbs-0426-TtM176GulfSt. Joseph Bay 3/12/2004Hubbs-0427-TtM221GulfSt. Joseph Bay 3/12/2004Hubbs-0428-TtM236GulfSt. Joseph Bay 3/12/2004Hubbs-0436-TtM228GulfSt. Joseph Bay 3/12/2004Hubbs-0437-TtF247GulfSt. Joseph Bay 3/16/2004PCNMFS 04-8M277GulfSt. Joe Beach 3/17/2004SJP0317-15M249GulfSt. Joseph Bay 3/18/2004SJP0318-17F207GulfSt. Joseph Bay 3/25/2004SJP0325-32M268GulfSt. Joseph Bay 3/25/2004SJP0326-34F245GulfSt. Joseph Bay

PAGE 157

About the Author Leanne Joseph Flewelling received a B achelor of Arts in Philosophy from the University of Massachusetts at Boston in 1989. In 1994, she returned to the University of Massachusetts to study phytoplankton ecology and in 1997 received a Master of Science in Environmental Sciences. In 1996 while wr iting her thesis, Lea nne relocated from Boston, Massachusetts to Tampa, Florida and be gan her career at what is now the Florida Fish and Wildlife Conservation Commissions Fi sh and Wildlife Research Institute in St. Petersburg. In her current position as an A ssistant Research Scientist in the Harmful Algal Blooms section, Leanne investigates the production of algal toxins by multiple species of phytoplankton found in Florida wa ters, their fate and persistence in the environment, and their impacts on fish and wildlife.