Characterizing toxic activity from Heterosigma akashiwo: a … Savage/Adjudication/Exhibits/Tesoro/Exhibit... · Romberg Tiburon Center for Environmental Studies, San Francisco State

82 A. Lincoln MacKenzie [Ed]. Marine and Freshwater Harmful Algae 2014. Proceedings of the 16th International Conference on

Harmful. Algae. Cawthron Institute, Nelson, New Zealand and the International Society for the Study of Harmful Algae (ISSHA)

Characterizing toxic activity from Heterosigma akashiwo: a tale of two assays

Vera L. Trainer 1*

, Leslie Moore

1, Bich-Thuy L. Eberhart

1, Brian D. Bill

1, William P. Cochlan

2, Christopher

Ikeda2, Mark L. Wells

3, John Incardona

1, Tiffany Linbo

1, Christopher O. Miles

4 and Charles G. Trick

5

1*Marine Biotoxins Program, Northwest Fisheries Science Center, National Marine Fisheries Service,

National Oceanic and Atmospheric Administration, 2725 Montlake Blvd. East, Seattle, WA 98112 USA,

[email protected]; 2Romberg Tiburon Center for Environmental Studies, San Francisco State

University, Tiburon, CA USA; 3University of Maine, Orono, ME USA;

4Norwegian Veterinary Institute,

Oslo, Norway,5Western University, London, Ontario, Canada

Abstract

Blooms of the raphidophyte, Heterosigma akashiwo (Y. Hada) Y. Hada ex Y. Hara et M. Chihara, have

caused severe economic damage to fish farms in the inland waterways of Washington State, USA, and

British Columbia, Canada, and are believed to be increasing in frequency and severity. In our study, two

laboratory tests were used to characterize H. akashiwo toxicity - a modified rainbow trout gill cell assay and

embryonic and larval zebrafish exposures. The gill assay demonstrated that the H. akashiwo toxin is

primarily intracellular, highly soluble in methanol and ethyl acetate, and pH stable, with no loss of activity

upon storage at -20oC. Stationary phase extracts from H. akashiwo culture were used to characterize the

toxin‘s specific cellular targets on the development of zebrafish. At 48-hour postfertilization (hpf), intrinsic

and specific effects to cardiomyocytes included reduced heart rate and atrial dilation, leading to pericardial

edema. Zebrafish heart chambers formed normally, suggesting that the H. akashiwo toxin does not affect

early cardiac development but is a physiological poison. In summary, the non-labile toxin from H. akashiwo

is a largely intracellular, medium-to-low polarity organic compound that causes impairment of cardiac

function in fish, possibly through impacts on cellular Ca2+

homeostasis.

Keywords: Heterosigma akashiwo, raphidophyte, toxicity, fish kill, zebrafish assay, gill cell assay

Introduction

Recurring threats from the raphidophyte,

Heterosigma akashiwo have caused extensive

devastation ($2-6 million USD per episode) to

wild and net-penned fish of Puget Sound,

Washington. The toxic activity of H. akashiwo

has been attributed to the production of reactive

oxygen species, brevetoxin-like compounds,

excessive mucus, or hemolytic activity; however

these mechanisms are not expressed consistently

in all fish-killing events or cultured strains (e.g.

Yang et al. 1995; Khan et al. 1997; Oda et al.

1997; Twiner and Trick, 2000). The difficulty of

conducting research with active, toxin-producing

field populations of H. akashiwo has resulted in

conflicting findings from those obtained in lab

culture studies, thereby limiting the ability of fish

farmers to respond to these episodic blooms.

However repeated studies have suggested that a

neurotoxin is produced which causes hypoxic

conditions of the blood and tissues as well as

asphyxiation (e.g., Oda et al. 1997; Twiner and

Trick 2000; Twiner 2002; Marshall et al. 2005;

Khan et al. 1997). The goal of this study was to

identify the primary neurotoxic element(s)

associated with fish-killing H. akashiwo blooms,

and thereby, to provide managers with the

fundamental tools needed to monitor the toxicity

associated with these harmful events.

Material and Methods

Heterosigma cell isolation and culture: Culture

methods followed those of Guillard (1995).

Single H. akashiwo cells were isolated from net

tow samples. Cultures were grown in filter-

sterilized natural seawater, enriched with nutrients

according to Berges et al. (2001, 2004), with

modifications following Cochlan et al.

(2008). Nitrate, the sole nitrogen source, was

reduced from 550 to 300 µM and silicate was not

added. Cells were grown to mid-exponential

phase through multiple transfers before testing for

toxicity using the gill cell assay. H. akashiwo

isolate NWFSC-513 was harvested during early

stationary phase when H. akashiwo toxicity was

highest (Cochlan et al., 2014) and used for solvent

extraction experiments.

EX5097-000001-TRB

83

Solvent extraction of cells: Cells of H. akashiwo

were pelleted by centrifugation at 1000 g for 10

min. Pellets were extracted in 4 mL methanol, the

supernatants SpeedVacTM

concentrated to dryness

and reconstituted with distilled H2O (dH2O) to a

final 1% methanol solution. This aqueous solution

was partitioned using a separatory funnel through

a series of immiscible solvents in the following

order: hexane, ethyl acetate, dichloromethane

(Fig. 1). Each extract (including the final dH2O

extract) was assayed using gill cell assay and

compared to a control (no cell, solvent only) and

the initial methanol extract.

Gill cell assay: Rainbow trout gill cells (RTgill-

W1) from the American Type Culture Collection

(CRL-2523) were exposed to NWFSC-513

extracts using the method of Dorantes-Aranda et

al., (2011) with the following modifications:

NWFSC-513 was pelleted and extracted in 100%

methanol, speed vacuumed to dryness and

reconstituted in L15/ex medium, then applied on

plated RTgill-W1 cells. Extracts resulting from

liquid-liquid partitioning were processed similarly

and resuspended in L15/ex prior to testing in gill

cell assays.

Cell pumping system: A bloom of H. akashiwo

at Cap Sante Marina Marina (48° 31' 5" N, 122°

36' 1"W) in Anacortes, northern Puget Sound,

WA was collected in June 2014 using a

polyaromatic adsorbent DIAIONTM

resin (HP-20,

Mitsubishi Chemical Corp., Tokyo, Japan) in a

large-scale pumping system described by

Rundberget et al. (2007). Seawater was pumped

continuously for 54h at a flow rate of 360 L h-1

.

An average of 50 x 106 H. akashiwo cells L

-1 were

collected over the period of the bloom, resulting

in ca. 1 x 1012

cells per g resin. During the

pumping period, H. akashiwo was the dominant

species, with Prorocentrum lima and small

unidentified flagellates present at <100 cells L-1

.

Solvent extraction of resin: Exposed resin (2g)

was rinsed with 30 mL H2O at 1mLmin-1

to

remove salts then eluted with 6 mL methanol at a

rate of 0.5 mL min-1

. The extract was dried via

speed vacuum and N2 stream then diluted with 18

mL dH2O to a final 1% methanol solution. The

extract was partitioned with hexane and ethyl

acetate as describe above (solvent extract of cells

section), evaporated under a stream of N2,

resuspended in 1% methanol in zebrafish water

H2O H2O

H2O

H2O

Hexane

EtAc

DCM

0

20

40

60

80

100

62,500 125,000 250,000 500,000

% T

oxic

ity

Heterosigma Cell Concentration (cells/mL)

Hexane

0

20

40

60

80

100

62,500 125,000 250,000 500,000

% T

ox

icit

y

Heterosigma Cell Concentration (cells/mL)

H2O

0

20

40

60

80

100

62,500 125,000 250,000 500,000

% T

oxic

ity

Heterosigma Cell Concentration (cells/mL)

DCM

0

20

40

60

80

100

62,500 125,000 250,000 500,000

% T

oxic

ity

Heterosigma Cell Concentration (cells/mL)

Ethyl Acetate

0

10

20

30

40

50

60

70

80

90

100

250,000

% T

ox

icit

y

Heterosigma Cell

Concentration (cells/mL)

Fig. 1. Liquid-liquid partitioning of H. akashiwo strain NWFSC-513 harvested during

stationary phase. Error bars on plots indicate ± 1 SD (n = 3) of replicate determinations.

EX5097-000002-TRB

84

(deionized and reverse osmosis filtered water

amended with Instant Ocean Salt to a conductivity

of 1500 μS cm-1

) for zebrafish exposure.

Concentrations used for exposures (in triplicate)

ranged from 4-15 x 106 H. akashiwo cell-

equivalents per treatment.

Zebrafish exposures: Zebrafish embryos (n=15)

at 256-cell stage (2.5 hpf) were placed in 5 ml

zebrafish water in 6-well glass plates (Corning,

Costar 3516) and kept in the dark at 28.5°C with

water replacement every 24 h until 72 hpf.

Controls were zebrafish water containing 1%

methanol and zebrafish water alone. Digital video

clips of embryos were collected from each

treatment using a Nikon SMZ800

stereomicroscope or a Nikon Eclipse E600 and a

Unibrain Fire-I800 camera with BTV Carbon Pro

software as described in Incardona et al. (2014).

For each treatment, both heart rate and percentage

of zebrafish with edema were recorded (n=15).

Results and Discussion

The ethyl acetate extract of NWFSC-513 culture

resulting from liquid-liquid partitioning showed

the greatest dose-dependent toxicity in the gill cell

assay compared to other fractions. About 25% of

the activity observed in the original methanol

extract was retained in the ethyl acetate fraction

(Fig. 1). Typically, when extracted sequentially in

this manner: the hexane fraction contains very

lipophilic compounds (e.g. fatty acids); the ethyl

acetate fraction contains the less polar uncharged

molecules (e.g. pectenotoxins); the

dichloromethane fraction contains the more polar

organic molecules (e.g. okadaic acid); and, salts

and highly polar or ionic organic compounds (e.g.

yessotoxins) remain in the water fraction. Resin

extracts from the high density Heterosigma bloom

(June 2014) also showed the highest activity in the

ethyl acetate fraction and, in general, showed

more reliable toxicity than cultures, in particular

because the latter appeared to lose their toxicity

after ~1 year. For this reason, resin extracts were

used for zebrafish exposure experiments.

Zebrafish embryos exposed to the HP20 resin

ethyl acetate extract showed severe pericardial

edema (Figs. 2, 3), atrial dilation, and

hydrocephalus shown by a clear space in the 4th

ventricle of the brain (Fig. 2). The incidence of

edema and decrease in heart rate was dose-

dependent with exposure to HP-20 resin extract

(Fig. 3). Together, these data suggest that H.

akashiwo produces a stable toxin that is

Fig. 3. Dose dependent cardiac effects of H.

akashiwo ethyl acetate extracts on zebrafish

pericardial edema (A), atrial dilation (B), and

heart rate (C). Control heart rate is 211 beats per

minute (bpm) and highest exposed is 132 bpm at

48 hpf. Average std. dev. in (A) = 2 bpm (n=15).

A

B

C

A

V

Treated Control

*

Figure. 2. Zebrafish exposed to H. akashiwo extract

2-48 hpf shows severe edema, hydrocephalus (*),

and atrial (A) dilation. V=ventricle

0

20

40

60

80

100

120

control 4 million 8 million 15 million

% w

ith

ed

ema

Pericardial edema

48h

60h

72h

0

50

100

150

200

250

control 4 million 8 million 15 million

Hea

rt r

ate

(b

eats

/min

)

Heterosigma extract (cell equivalents)

Heart rate

48h

60h

0

20

40

60

80

100

120

control 4 million 8 million 15 million

Atr

ium

wid

th (

mm

)

Atrial dilation

48 hr

60 hr

EX5097-000003-TRB

85

moderately lipophilic. It is proven to be a

chemical entity because toxic activity is recovered

after drying down in methanol and resuspended in

assay buffer. Solubility in ethyl acetate suggests

that it may be a relatively non-polar, neutral

polyether-like compound. The toxin causes

bradycardia, edema and atrial dilation has a

negative chronotropic effect in zebrafish.

Because the cardiac-specific effects occurred by

48h exposure and the heart chambers were formed

normally, the effect is intrinsic to heart

development and not a secondary effect, such as

at the acetylcholine receptor which appears at a

later stage in zebrafish development. This

evidence suggests that the H. akashiwo toxin is a

physiological poison.

Twiner (2002) proposed that H. akashiwo extracts

had an effect on Ca2+

release by muscarinic-1

transfected sf9 insect cells pretreated with

lanthanum, a Ca2+

channel blocker (Twiner 2002).

The effects of H. akashiwo extracts on cardiac

function that we observed also point to the

possibility that H. akashiwo impairs Ca2+

homeostasis. The negative chronotropic effects

shown by H. akashiwo extracts on zebrafish are

consistent with Ca2+

blockage during the plateau

phase of the action potential. However, this study

is not a confirmation of H. akashiwo impacts

solely on Ca2+

homeostasis as the zebrafish

exposures resulted in mixed phenotypes,

suggesting other toxic mechanisms. Head-first

hatching which is characteristic of brevetoxin

effects on zebrafish embryos (Kimm-Brinson &

Ramsdell 2001), was not observed, suggesting

that H. akashiwo toxins are not brevetoxins, as has

been suggested (Khan et al. 1997).

Acknowledgements

We thank Kevin Bright, American Gold Seafood,

for alerting us to H. akashiwo blooms. We

appreciate the assistance of the Cap Sante marina

personnel. This work is a result of ECOHAB

research funded by the U.S. National Oceanic and

Atmospheric Administration Center for Sponsored

Coastal Ocean Research Project No.

NA10NOS4780160 awarded to the University of

Maine (MLW) and San Francisco State University

(WPC), an internal grant to the NWFSC (VLT),

with a subcontract to Western University and a

Canadian NSERC Discovery Grant awarded to

CGT. This is ECOHAB publication No. 813.

References

Berges, J.A., Franklin, D.J., & Harrison, P.J.

(2001). J. Phycol. 37: 1138-1145.

Berges, J.A., Franklin, D.J.,& Harrison, P.J.

(2004). J. Phycol. 40: 619.

Cochlan, W.P. Herndon, J., & Kudela, R.M.

(2008). Harmful Algae 8: 111-118.

Cochlan W.P., Trainer, V.L., Trick, C.G. et al.

(2014). In: Kim, H.G., B. Reguera, G.M.

Hallegraeff et al. (eds). Harmful Algae 2012,

Proceedings of the 15th International Conference

on Harmful Algae, 2014, pp. 203-206.

Dorantes-Aranda, J.J., Waite, T.D., Godrant, A.,

et al. (2011). Harmful Algae 10(4):366-373.

Guillard, R. R. L. (1995). Culture methods. In:

Manual on Harmful Marine Microalgae,

Hallegraeff G. M., Anderson D. M., Cembella A.

D. (eds), UNESCO IOC Manuals and Guides No.

33. 551pp.

Khan, S., Arakawa, O., & Onoue, Y. (1997).

Aquac. Res. 28: 9-14.

Kimm-Brison, K.L. & Ramsdell, J.S. (2001).

Environ. Health Persp. 109: 377-381.

Marshall, J.A., de Salas, M., Oda, T., &

Hallegraeff, G. (2005). Mar.Biol. 147: 533-540.

Oda, T., Nakamura, M., Shikayama, M. et al.

(1997). Biosci. Biotech. Biochem. 61: 1658-1662.

Rundberget, T., Sandvik, M., Larsen, K. et al.

(2007). Toxicon 50: 960-970.

Twiner, M.J. & Trick, C.G. (2000). J. Plankton

Res. 22: 1961-1975.

Twiner, M.J. (2002). Bioactive extracellular

metabolites from Heterosigma akashiwo. Ph.D.

Thesis, The University of Western Ontario,

London, Ontario, 127 pp.

Yang, C.Z., Albright, L.J., & Yousif, A.N. (1995).

Dis. Aquat. Org. 23: 101-108.

EX5097-000004-TRB