Polluenten in paling - Actuele verspreiding van de zwemblaasparasiet Anguillicola crassus in Vlaanderen : ontwerpverslag

{

CQ^-*JL

hrlqrr;^^

@

?

?

t{}5

Katholieke

Universiteit

Leuven

Ministerie

van

de

Vlaamse Gemeenschap

Instituut voor

Bosbouw en

V/ildbeheer

r./r"eiding

van

de

zwemblaasparasiet

,rlrrorin

Vlaanderen

+

Audenaert

V.',

Huyse

T.',

Belpaire

C.2,

Volckaert

F.A.M.l

I

Katholieke

Universiteit

Leuven,

Laboratorium voor

Aquatische

Ecologie,

Ch.

de

Bériotstraat

32,3000

Leuven

'

Ministerie

van

de

Vlaamse Gemeenschap,

Instituut voor Bosbouw

en

V/ildbeheer,

A.

Duboislaan

14,

1560

Hoeilaart

Onderzoeksopdracht

TWOL

AMINALIBG/IB

W/2

00

0

-

5

Ontwerpverslag

08 november

2001

Polluenten

in

paling

_-

Actuele

Anguillicola

r)

Ltr

The

temporal

dynamics

of

Anguillicola

crossus

and

the

adaptation

of

its

host,

Anguilla

anguilla,

in

Flanders.

\F'Y

b,!J**

Introduction

Anguillicola

crassus Kuwahara,

Niimi

and

Itagaki,

_{1974 §ematoda,}

Dracunculoidea,

Anguillicolidae)

is

a

parasitic

nematode

that lives

in

the

swimbladder

of

the

European

eel

Anguilla anguilla

L.

(Belpaire et

al.,

1989).

This

parasite

was introduced

in

Europe

in

the

early

1980s

through

uncontrolled

intercontinental

transfer

of

live

eels

from

Taiwan for

consumption

(Koie,

1991

_{; Paggi et}

al.,

1982; Neuman,

1985).

Since

then

it

has spread

rapidly

among

European

eel

populations

(Koie, l99L;

Peters

and Hartman, 1986; Koops

and

Hartmann

,

1987;

Kennedy and

Fitch, 1990).

This quick

expansion

is

due

to

both

a

human

assisted

dispersion

of

the

final

host

and

the

efficient

dispersion

mechanisms

of

the

parasite

itself

using

non-specific

intermediate hosts (Belpaire et

aL.,1990;

Kennedy and

Fitch,

1990).

The

eels become

infected

with Anguillicola

crasszs

through

the food

chain.

The

female

nematodes

produce a large amount

of

eggs

containing

first-

or

second-stage larvae

that

leave

the

swimbladder

through the

pneumatic duct

and pass

with

the

faeces

into

the

water.

The

larvae

use

freshwater

copepods as intermediate hosts

in

which they

develop

into

third-stage

larvae.

Some

fish

species

may act

as

paratenic hosts where

the

parasite doesn't

develop

any

further

and

which,

together

with

the plankton, belong

to

the

eels

diet.

In

the

final

host

the

third-stage

larvae migrate through

the

intestinal

wall

and

the

body cavity

into

the

swimbladder

wall

where

they

develop

into

fourth-stage larvae

and later

into

adults, when

reaching

the lumen

of

the swimbladder.

At

a

temperature

of

20oC

the

development

of

A.

crassus eggs

into

adults

can

be

completed

in

less than

two

months (Belpaire

et

al.,

1989; De

Charleroy

et

al.,

1990; Kennedy and

Fitch,

1990)

The

genus

Angutlla includes

16

species

of

which

only

three occur

in

the

Northern

hemisphere:

the

European,

American

and

Japanese

eel.

The natural habitat

of

the

European

eel reaches

from

the

North

European

to

the

North African

coastal

countries.

Eels present

in

Central

and

Eastern Europe are

assumed

to

have

been

mainly

imported

from

these coastal

areas

(O.V.B.,

1988).

Sexually mature

European

eels migrate

to

the

Sargasso

Sea

for

spawning

after

which

the

larvae start migrating back

to

the

continental

waters. During

the

latter migration period the

young eels

grow

and mature coupled

with

great

internal

and

external

physical

(Belpaire and

Ollevier,'1987;

Degani and Gallagher,

1995). I

+

known

that

migrating

eels

show

a

vertical migration

pattern

for

which

the swimbladder

is

of

great importance.

In

contrast

to

the

Japanese

eel, the nematodes

original

host, the European eel can

suffer

severe

damage

by

the

parasite

due

to

its

higher susceptibility

(Koie, l99l

_).

Damage

to

the

gas

glands

in

the

swimbladder

wall

can

cause a

serious drop

in

swim

activity

of

the eel

preventing

the

fish

from migrating,

but

infection with ,4.

crossus

also

benefits

secundary

bacterial

infections

which

are

often lethal,

causing mass

mortalities

in

eel populations

(Boon et

al.,

1990;

Mólnar

et

al.,

l99l).

The

eels seem

to

show

some

kind of

immunological reaction to

this parasite.

Many

dead and disintegrated adult nematodes

as

well

as

encapsulated

larvae

are

seen

that

are the

result

of

a concentration

of

fagocytes. Moreover,

infected swimbladders can

be

encapsulated

by

connective

tissue.

Heavy

infection

lead

to

very

thickened swimbladder

walls,

creating

poor

conditions

for

reinfection (Hartmann

and Peters,

1989).

Nowadays

eels

can easily be

treated

with

nematicides

and bactericides

(which

affect

the symbiotic

bacteria

Wolbachia)

or

chemicals

that

affect

the

intermediate

host

of

the

nematodes

to

avoid

completion of it's

life

cycle

(Taraschewsky et

al.,

1988;

Kamstra,

1989).

This

study

analyses

population

data

of

Anguillicola

crassus

collected

in

Flanders

in

the

year

2000

and

compares

the

results

with historic

data

from

1987

and

1997

(Huyse,

1999).

The

aims are

to

detect an

evolutionary

pattern

in

the parasite characteristics and

to

find

indications

for

a

growing

equilibrium

in

the parasite-host relationship.

Materials

and methods

From

May

2000

untill

October

2000,

1084 eels were sampled

from

141

different

sites

and

11

..

river

basins

in

Flanders

(Addendum

t

).

The eels were

measured

and weighted

and

swimbladders

\ryere

stored at

4Yo

formol.

For

dissection

the

swimbladders rvere transferred

to

70%

ethanol. Adult

nematodes

were

sexed

and counted

macroscopically

whereas

the

larvae

were identified and

counted

using

a

binocular

with

transmitted

light, easily

detected by

pressing

the swimbladder

wall

between

two

glass

plates.

The thickness

as

well

as

rupture

of

the

swimbadder

wall

and

the

presence

and colour

of

fluid

in

the

lumen was

registrated.

According to

the thickness

of

the

wall,

the

swimbladders

were

divided into

three categories.

In

the

absence

of

parasites

(adults

as

well

as

larvae and

capsules),

swimbladders

with

thin

walls

(<

I

mm)

were recorded

to

be

not

infected, where

as

thick

walls (thicker than

I

mm

and

thinner than

3 mm)

were

assumed

to

be

infected''due

to

the

presence

of

great amounts

of

connective tissue and

if

tissue

proliferation could

be

observed. Very

thick walls (> 3

mm)

L

t.

pointed

towards

former

infection.

Brown

fluid in

the

swimbladder lumen results

from

the

desintegration

of

adult

nematodes and

thus

indicates

infection.

Rupture

of

the swimbladder

can be

the result

of

a

severe

infection

but

can

also be due

to

damage

during dissection

of

the

eel.

Again

the thickness

of

the swimbladder

wall

v/as considered.

Prevalence,

mean

intensity

of

infection

and

abundance rvere estimated

over

all

sites

where

more

than

7

eels

were

captured,

as

well

as

per

basin.

Statistics were carried

out

by

SAS

Results

Pregel-enge..+nd.intenqitv-gf

_.ipfectipnof

lr?gn-iiligolír.crg§{usinElqpders

The mean prevalence

in

Flanders

in

2000 is estimated

at

88,

L

% witha

rninimum

of

A

Yo

(

at a

single

site along the

Boudewijnhanaal

in

the Brugse

Polders)

and a ma*ximum

of

100%.

The

mean

intensity

of

infection is

about 5,5 ranging

from

0

to

13,3.

The

mean

prevaience,

intensity

and abundance per basin are given

in

the table below.

Table

1:

The

mean prevalence

{o/o),

m€an

intensity

of

infection

and

abundance per

river

basin

of

e.

crassu§.

Ns

:

number

of

sites saurpletl,

Ne

:

number

of

eels collected.

River

basin

Ns

Ne

Prevalence

{%)

Intensiry

Abundance

Bekhen yan de Brugse Polders

Bekken

van

de Gentse Kanalen

Benedenscheldebekken

Bovenscheldebekken

Demerbekken

Denderbekken

Dijle-

en Zennebekken

Ijzeibekken

Leiebekken

Maasbekken

Netebekken

11

11

1l

14

J

2

4

7

r1

I8

I

t07

104

103

138

28

17

42

69

98

174

74

72,7

93,2

82,9

90,0

89,3

100,0

86,2

97,1,

gg,2

92,8

83,4

418

4,5

5,9

5,7

6,7

7,6

718

6,9

5,8

4,3

5,8

3r7

$r3

5,2

5,1

6,2

716

7rl

6,,'T

5,2

4,0

4,8

The mean

prevalence

in

the Brugse Polders

is

rather

low which

might

be explained

by

the

vsry

low

prevalences

at the three

sites

that were

sampled

nlong the Boudewijnkanaal, whish

arc

20Ya,20% and

07q

with

an intensity

of

respectively

11

_,5,

t

and 0 (Addendum

2).

\ilhen

excluding

this

canal

the

mean prevalence

would

rise

to

95% and

the

mean

intensity

and

abundance

to

5,0 and

4,8.

In only

one basin, the Denderbekken, the mean prevalence reached

100?4,

but

this

is due to the very

low

number

of

sites

(5) that

has been sampled,

of

which only

two

contained enough samples to calculate the chara,stEristics mentioned above.

q

J.

The

second

highest

prevalence

was obtained

in

the

ljzerbekken, where

the

intensity

was

neitJrer

very

high

nor

low.

The

greatest

intensity

was

found

in

the

Dijle-

and

Zennebekken

(7,8),

where the prevalence reached

86,20/0.

A Kruskal-lilallis

test

showed a

significant

difference

in

the

total

arnount

of

parasites

that

is

due

to

a

difference

in

the

nurnber

of

adult

nematodes

and

not in

the

rrumber

of

larvae or

capsules

(Table

2).

A

lVilcoxon

test

on

the totat

number

of

parasites was

carried

out to

see

whic,h basins

differed

the

most.

The Demer-

and

Denderbekken

are

left

out

in this

analysis

because

of

too few observations.

The results are shown

in

Table

3.

Table

2:

Kruskal-Wallis

test

for

significant

differences

of

the

total

number

of

nernatodes and

on

the different

developmental

stages

between basins;

o

:

lo/o.

Signifïcant

values

are

irdicated

in

bold.

p-value

0,5116

0,5044

0r{x}49

010021

Larvae

Capsules

Adults

Total

Tabts 3:

lVilcoxon

test

for

differences

between

specific

basins, based

on the totat

number

of

parasites;

only

p-values

signiÍicant

at

cÍ.:

17o

are

glven.

In

general,

all

basins are

quite

similar

in

the

infection af

Anguilla

unguilla with Anguillieala

crussus.

Most differences

occur between ttre Brugse Polderbekken and the other

basins. This

is

probably due

to

the

Boudewijnkanaal where

the

abundance

was

very

low.

Another

difference

ís

seen

between the Netebekken anri the Leiebekken.

Thg-tg.lalive p.ropg#.ons

pf

the

different

dpvqlop"rUefrtalstages

and.the nat$ral.distrib'ution

of

Ary#tiÍ{ilplg

cr,as*W

in

Anguillq

ang,uilla

Table

3:

The

abundan§.e, standard

deviation,

minimum, maximum,

variance and

dispersal

coëffïc,iënt per developmental

stage

of

Anguilliuia

{.:ruàslÍ'u*Í'

in

Flanders 2000.

Abundance

§td. Dev.

Min.

Ma,x

Var.

Disp.coëff.

I

L3

L4

Caps

Pre*adult

M

Vr

Rernnants

TOT

54,O0

17,00

32,00

20,00

19,00

15,00

3,flo

87,00

5,94

3,43

5,32

2,29

3,45

2,,42

0,06

52,52

0,68

I,01

1,05

0,71

0,89

0,75

0,04

5,?7

2,44

tr,85

2,31

1,5

I

1,86

1,56

0,24

7,25

0,00

0,00

0,00

0,00

0,00

0,00

0,00

0,00

8,?6

3,41

5,06

3,21

3,85

3,20

1,39

9,10

As

can be

seen

in figwe

I

_o

which

gives

the relative

proportíoru

of all

developmeutal skges,

the

greatest

amount

of

nernatodes belongs

to

the

trarval stages,

present

in tlre

swimbladder

wall.

The

maximun

number

of

thirdstage larvae

is

almost

twice

the

number

of

male

and

female adults

(34).

Fourth-stage

larvae

and encapsulated

larvae,

which

are

equally divided,

are

even

rnore a,bundant,

however

their

maxima

are less

high.

The proportions

of

male and

female

adults are

nearly

equal and there

\Mere

alrnost

no

renulants

found.

A

Spearman

Rank

é

The

nurnber

of

parasites

per

eel

investigated (abundance)

was

calculated

for

each

developmental

stage

of

the

nematode,

together

with

the

range and

coefficient

of

dispersal.

The latter

is

calculated

by

divitting

the

variance

by

the

mean.

Coefficients

greater

than

one

índicare overdispersion.

correlation

test

showed

positive

correlations between

the

larvae,

capsules

anrJ

adults

(0,35,

0,29 and 0,16) ttmt are

all

weak,

but significant

at a

signiÍicance level

of

5%.

1aío 2lolo

Bl5-stages

inthewall

Bl4-sl"tgcs in

fCapnrÍed

Ïanae

tlÀ4ail

dults

tr&sintrgatrd

Fig

1:

Relative proportions

of

ths dif[erent

developmental

stages

of

Anguillicala

crasrsus

in

tlrs

swimbtadder

of

European eels

from

Flanders 2000.

The

last

column

of table

3

shows

that

all

developrnental

stages

in the

swimbladders

of

eÈls

sre

chaÍaÉdËnzed

by

overdispersion. Fig.

2

illusfiates tlrat most eÈl

populations

are

hardly

infected

while only

a

sffill

number

is

heavily furfected.

About

&3,7Y+

of all

eels are

infected

with

less

than

10

parasites,

while only

14 upon

961

swimbladders contained mor€

tltan

30

The

swimblacÍders

with

thickened

walls

greater

than

3 mm

and

c.ontaining no

paÍasitÉs ïvere not considered here.

À^^

r5%

tff/6

_V.r

tza

í00

Fig. 2:

Frequency

2000.

!t

Eao

è

Ë60

₂

o 2 4 6 C ÍO12Í4reÍ820?2242A2E

30323417 4Í5667

Number oÍ nematodes

fr?1

tr»

t

I

13111r1

dis§bution of

nematodes

in

g6l

eels

from

141

different

sites

in

Flanders

in

Host

resistance

The

great number

of

encapsulated

larvae

in

Fig.

2

indicates

àn

increased resistance

of

the

host.

Fig.

3 shows

a

decrease

of

the uumber

of

larvae as

well

as the number

of

capsules

and

adult

nenratodes

in

swimbladders

with

a

thíckened

wal1.

o 5

I

Fig.

3:

Proportion of

all

developmerrtal

stages

of

the nematode

Anguillicalu

crasse*r

for

swimbladders

with

a

different wall

thickness.

A

Kruskal-IVallis

test

for

significant

differenoes

(o:

0,01)

in

the

abunrJance

of

seÍnatodes

in

swimbladders

*ith

a

different raall

thickness showsd

a

highly

significant

p-value

(p

.

0,000001). The

Wilcoxon

test results

(er

₌

0,0U

are

given

in

table

4.

The

swimbladders

with

a

thicknees

<

1

mm differed

from

those

with

a

thickness between

1

and

3

mm

only in

the

arnount

of

larvae

(p

:

0,0080)

but not

in

the amount

of

capsules

(p

:

0,10)

or

adults

(p

:

0,23).

Tlre number

of

adults does

differ

between ths swirnbladders

with

thickness

<

lmm

and

betwesn

I

and 3 mrn at the one hand and swimbladders

with

a

thickness

between

t

and 3

mm

and

sthickness <

3

mmattheother(respectivelyp:0,00001I

andp:0,00019).

Since the

presffIce

of

larvae

or

capsules

in

very

thic.kened

swimbladder

walls

was hard

to

verífy,

tro

comparison could be

made

for

these

developmental stages

in

swimbladders

with

a thickened

wall

geater than 3 mm.

fahle

4: ïWilcoxon

test

for

significant differences

at

u

:

0,01

betwern swimbladdsrs, infected

with

Anguilliwla

Í;rutlstt§,

with

a

different wall

thickness.

Larvae

Capsules

Adutts

Bel+tioqshig_betrryeen the ifrfection.and

thelglÈrh

and vyeight.gf

their

hqst

The

_ryean

lengh and weight

of

the captur«Í

eels

are

given

in

table

5

together

with

their

ranges"

A

frequency

distribution of

the eels

lengÍh

is

given

in

Fig.

4

where

the graph

in

the

upper

dght

corner allowed

us

to identifr two

individuals

with

measurement

errors

for

the

weight

or the

length

Those

individuals

were excluded

frorn further

analysis

Table 5: The

mean, §tandaÍd

deviation, minimurn and

ma"xirnum

of

the length and weight

of

1084 European eels

sampl«Í

from 14ï

sites

in

Flanders

from lvÍay

untill

October 2000.

Gern

Stdev

Var

Min

Ài{ax

Lemgte

(cm)

Gewicht (g)

4t,74

139,87

9,24

114,57

67,82

13126,29

21,70

21,40

81,00

e663A

B

F

0

20 &

60 80

fie

&

t9

I

400

s50

t00

250

na

Í50

6

{,

g

*-o

tr

:,

z,

,pP

sq9

.oP tq?

,trl9

,.P

4xP

d# 69

".9

,rà?

rs$

.p§'

_6P'

àÈ9'

6S' pS'

op$'

pS'

4c$'

6$' 6§'

,1sS'

nt9'

Length

_{cm}

Fig, 4:

Frequency

distribution

of

the length

of

European eels

in

Flanders 2000.

There

was

only

a

weak

but

still

significant positive

correlatí""&ï*

at

a

signifïcance level

of

5olo)

between

the

length

of

the

eels

and

the

number

of

adult nematodes.

This

could

be

partly

explained

by the fact

that adult

nematodes

can

only

develop

in

suffrcientely

large

swimbladders, however

adult

nematodes

fit

their

lenglh

to the

space

available (Banning

and

fIaeRen,

1990;

Moravec,1994).

The

mair

cause

however

is the

diet

of

larger eels

containing

t

@

*r-rl"-c

hosts

for

Anguillicalu

*usst&s(Molruír

er

al.,

1994),

{"

{cqn,

^

{+cnfu,a*,

/lgfl/

larger

Íish

that

serve

as

EAdulb

ElCapsules

]ï-awaa

I

7

6

4

.'

2

I

Abundance

nI?

.,,q?

_"É

.p?

op-$

B?

+?

"*?

6F

E?

tn? o§

,s§'

4,§'-+n'

+§'

§!'

#§'

oF9'

og9'

6,?'

&?'nt§'

^b§'

Length

(cml

Fig.

5:

Frequency

distribution

of

all

developemental stages

of

Anguitticaíu

cïa#su# ín

Europan

eel

in

Flanders 2000.

llo

Discussion

The current spatial pattern of

Anguillicola

cfqssu§

rnJlanderË

This

study shows that Anguillicola

crassus

has invaded

all

Flemish basins since

its

introduction

in

1986.

Eel populations

are

infected

at

all

l4l

sites included

in

the

study;

in

only four

of

the

l4l

sites

did

the prevalence

not

reach

25%.

Three

of

these sites are located

along

the

Boudewijn

Canal,

which

belongs

to

the basin

of

the

Brugse

Polders.

The

other

site

is

located at the

Fort

of

Oelegem

(Vriesel), which

belongs

to

the basin

of

the Benedenscheldt.

Statistical analysis

showe{

a

general

similarity

in

abundance

of Anguillicola

crassus

in

all

basins

in

Flanders.

The

main significant

difference occurred between the Brugse Polders and

the basins

of

the

Gentse

kanalen, Benedenschel

dt,ljtzer,

Leie

and

Maas.

These are

the result

of

the

low

prevalence and mean

intensity

of

infection in

the

Boudewijnkanaal.

The

cause

of

these

low

values

is

still

unknown

since

this

canal

is

connected

with

many

other waters

of

the

Brugse

Polder-s.

Therefore

it

might

be interesting

to look for

correlations

with

differences

in

water

temperature

or

salinity and

pH

in

an

attempt

to

explain

the

causality.

V/ater

temperature is

known to

be an

important

external

factor influencing

the

hatching, sunrival

and

transmission

of

the infective

stages

of Anguillicola

crassus

(Thomas

and

Ollevier,

1993; De

Charleroy

et

al.,

1989; Kennedy and

Fitch, 1990).

Höglund et

al.

(1991)

observed

higher

prevalences

at

Swedish sites

with

thermal

discharges

from

pourer

plants.

Since

Anguillicola

species

are adapted

to

tropical or

subtropical

conditions,

their

natural

distribution

is

restricted

to

those

areas

and infected

eels observed elsewhere

in

Sweden are

the result

of

restocking.

Salinity

may influence

the presence

of

copepods

that

are necessary

for

the completion

of

the

life

cycle

of

Anguitticola

rrassus.

Kirk

and

Kennedy (2000) quantified the

suvivorship

and

transmission

of

the parasite

in

50% seawater.

The copepod

Eurytemora affinÍs is

a

dominant

species

in

most

estuaries

of

the Northern

hemisphere

and

is very

susceptible

to

infection

by

Anguillicola

crassus.

However

the infection level

in high

salinity

waters

is

lower

(Dekker

and

Van ïWilligen,

1989).

From

field

observations

they

concluded

that

eels

infected

in

marine

or

brackish

seawater

ïuere probably

originally infected

in

freshwater.

Huyse

(1998)

concluded

that the

CaCOs

content probably

does

not

restrain

the

parasite

distribution in

The evolution

of

the

Ansuillicola

crassus

infection

of

E

uropean

eel during the last

two

Anguillicola

crassus

was

fïrst

detected

in

Flanders

in

the early

80's.

The

first

systematic

investigation (Belpaire

et

a|.,1987),

only

monitored the presence

of

adult nematodes.

ln

1997

a

second

study lvas

carried

out

by

Huyse,

who

included

the larval

stages

embedded or

encapsulated

in

the swimbladder

wall

(Huyse,

1998).

Table

6

shows the results

of

the

mean

prevalence and mean

intensity

of

infection in

Flanders

of

the situation

since

the introduction

of

A.

crdssus,

in

1987, 1997 and 2000.

Table

6:

The

evolution

of

mean prevalence and mean

intensity

of

Anguillicola

crassus

over

the last

two

decades

in

Flanders

Year

Number of

eels Prevalence Range Intensity

Range

before

1986

(all

stages)

1987

(adutrts)

1997 (adults)

1,997

(all

stages)

2000

(adults)

2000

(all

stages)

0

0

88,1

34,1

62,5

86,2

0

0-l

00

45-83

64-1

00

0-1

00

5,45

3,91

7,15

5,50

0

0

l-14,7

2,5-6,!

"

3,3-11,7

,:f

0,2-13,3

266

266

9s6

The

mean prevalence and mean

intensity

in

1997

and 2000 were

calculated

based

on

all

developmental

stages

(to obtain

a

more

correct result

of

the

current

state

of infection)

and the

adults

only, to

compare the results among

all

years sampled.

The mean

intensity

increased faster than the prevalence and reached

a

maximum f,rrst.

At

the

next

point

of

investigation (1997),

prevalence

had

increased

considerably

while the

mean

intensity

started

to

level

off.

Three

years

later

this

trend

is still

visible.

Such

a

rise

of

the

prevalence

and

decrease

of

the

mean

intensity

is

also

observed

by

Ashworth (1995)

in

a

longitudinal study

of

five

years and

by

Thomas

(1993)

in

a

one

yeaÍ

investigation.

It

is

possible

that

in

Flanders

both the

prevalence and mean

intensity

\

dll

stabilize

due

to

density

dependent

regulation

of

the

parasite

population and

overdispersed

distribution

of

the

nematodes

in

the

host

population,

which

is

a strong

stabilizing factor

(Ashworth, 1995).

Still

some

slight

annual fluctuations as

observed

in

the

lJsselmeer

and

the

Waddenzee

in

the

Netherlands

(Haenen

et

al.,

1994) can be

expected. The

quick

rise

in

prevalence

in

Flanders

might be

explained

by

restocking

of

non-infected area

with

infected

eelso

followed

by

a

4L

T

t

naturan spread

of

_{"4rugulJJfcsJm}

Ërfl,r§a{.§

in

opËn

waters.

This

was also

the

sase

in

the

Netherlands (Van Banning

eÍ sJ.,

986) and Dennaark urhere the

distrihutiun

of,

tLre

nematodes

seemed

tr

he a

ËomsequenËe

of

trade and eel restoaking than

of

matural

spreading (tsnetius,

l

eEe).

The reaction of

Anwtilla

angui,

lla

on

Ansuillicola

crassus

In

1979 Egusa

warned

for

the

damage

tha/'

Anguillicola

crassus

affects

to

the

European eel

and

stated

that

the

introduction

of

the

parasite

into

Europe should

be

prevented.

Unfortunately the introduction

\r/as

a

fact

in

the

early

80's

and

the

spread

over

Europe

was

inevitably

enhanced

by

eel trading

and

restocking. Already

in

1985

the

first

report

of

Banning

et

al.

in

the

Netherlands showed

the reduction

of

body

weight and condition

of

heavily

infected

eels.

Together

with

the

destruction

of

the swimbladder, the eel's

chances

to

reach

the

Sargasso Sea

for

spawning

was reduced

considerably.

In

Lake Balaton (Hungary)

mass

mortalities of

eels were recorded

in

1991

by Mólnar

et

al..

However,

a

few

years later,

in

1994,

the

development

of

fïbrotic

swimbladder

walls

\Mas

observed

in

the

same

eel

population

(Mólnar

et

al,

1994),

which

reflected an

immunological

reaction

to

the nematodes.

Fibrosis

of

the

swimbladder

wall is

due

to

repeated

infection

and reflects

a

severe stage

of

anguillicolosis

(Hartmann and Peters, 1989;

Möller

et

al.,

l99l;

Molnar

et

a|.,1994).

Also

in

our

study

the relative

proportions

of

the

various developmental

stages

in

Flanders

point

to

a

large

increase

in

the

number

of

encapsulated

larvae.

This

number

is, unlike

the

number

of

adult

nematodes,

not

correlated \,'rith

the length

of

the

eels.

The large amount

of

swimbladders

with

a

thickness

greater

than

3 mm

where

no

lumen

is

present, also points

towards

an increased resistance

of

the host, since

poor conditions

for

reinfection

aÍe generated

(Hartmann

and Peters,

l9S9).

Furthermore there seems

to

be

a tendency

towards a

decrease

of

abundance

in

eels

with

thickened

swimbladder

walls

in

all

developmental

stages,

however

an increase

of

encapsulated

larvae could be expected.

Treatment

of

aneuillicolosis

First

of

all,

anguillicolosis

can be prevented

by affecting the

intermediate host

to

intemrpt

the

life-cycle

of

the

parasite.

As

numerous intermediate hosts

of

Anguillicola

crassus have been

reported (Thomas, 1993),

this

is

not

very effective

in

nature.

Another

approach

is

the

chemical treatment

of

the

parasite. Grisez (1988) tested several

concentrations

of

diflubenzuron,

which

appeared

effective.

However;the

effect

on the host should also be taken

anthelmintics.

The

drug should

be

proven effective

against bloodsucking

nematodes.

Furthermore

it

should be resorbed easily

by

the

fish

without

leaving

residues and causing any

toxic effects. An

easy

way

to

treat infected

fish is

by

bathing them

in

water

with

a

certain

concentration

of

the

drug,

because

the

fish

do not

have

to

feed

to

absorb

it.

Taraschewsky eÍ

al.

(1988)

tested several

chemicals,

including

HCl,

levamisole

and

metrifonate

of

which

the

latter

two

seemed

the most effective.

In

general a

non-specific morphological reaction

in

the

nematodes

irreversibly

resulted

in

death.

A

third

strategy

is to

treat

the symbiotic

bacteria

Wolbachia in

A.

crassrzs. The presence

of

the bacteria

is

essential

for

the survival

of

the

host

(Casiraghi

et

al.,

2001).

Hence treatment

with

bactericides

(antibiotics)

effectively kills

symbiont

and the Àost parasite (Hermans et a1.,2001).

However

easy

the

treatment

of

anguillicolosis

might be on

a

small

scale,

it

is

far

more

difficult to

treat complete eel populations

in

ponds and

river basins.

The

best

way

to

deal

with

such a massive

infection

is

to

prevent the parasite

frorn spreading. This

includes a sound

'

management

of

eel restocking.

Acknowledgements

The

ministry

of the

Flemish

Community

(AMINAL

contract

TV/OL

2000

AMINAL/BG/IB\I//2000-5)

funded

the project.

G.

Goemans

kindly

assisted

with the

data

management.

References

(incomplete)

Casiraghi

M,

Anderson

TJC,

Bandi C,Bazzocchi

C, Genchi C

(2001)

A

phylogenetic

analysis

of

filarial

nematodes: comparison

with

the

phylogeny

of

Wolbachia

endosymbionts.

PARASITOLOGY,

I

2?:

93-103

Hermans PG,

Hart

CA,

Trees

AJ

(2001)

In vitro activity of

antimicrobial

agents against the

endosymbiont V/olbachia pipientis.

JOURNAL

OF

ANTIMICROBIAL

CHEMOTHERAPY,

47: (5)

6s9-663.

A1

o

@

o

o

\

(

\

I

\

o

f

,[."J

v

eqrÈ?

l-

2-f

_

LG

F

?>

'#'nd

qo

o

₄

trt

Appendix

2:

Prevalence, intensity and abundance oÍ Anguillicola crassus per basin

in

Flanders

2000.

" indicates the

number

of samples that

are infected

Basin Sampling site Code

Date N

Prev.

G.l.

Bekken van de Brugse Polders Bekken van de Brugse Polclers Bekken van de Brugse Polders Bekken van de Brugse Polders Bekken van de Brugse Polders Bekken van de Brugse Folders Bekken van de Brugse Polders Bekken van de Brugse Polders Bekken van de Brugse Polders Bekken van de Brugse Polders Bekken van de Brugse Polders Bekken van de Brugse PoÍders Bekken van de Brugse Polders Bekken van de Gentse Kanalen Bekken van de Gentse Kanalen Bekken van de Gentse Kanalen Bekken van de Gentse Kanalen Bekken van de Gentse Kanalen Bekken van de Gentse Kanalen Bekken van de Gentse Kanalen Bekken van de Gentse Kanalen Bekken van de Gentse Kanalen Bekken van de Gentse Kanalen Bekken van de Gentse Kanalen Bekken van de Gentse Kanalen Bekken van de Gentse Kanalen Bekken van de Gentse Kanalen Bekken van de Gentse Kanalen Benedenscheldebekken Benedenscheldebekken Benedenscheldebekken Benedenscheldebekken Benedenscheldebekken Benedenscheldebekken Benedenscheldebekken Benedenscheldebekken Benedenscheldebekken Benedenscheldebekken Benedenscheldebekken Benedenscheldebekken Benedenscheldebekken Benedenscheldebekken Benedenscheldebekken Benedenscheldebekken Bovenscheldebekken Bovenscheldebekken Bovenscheldebekken Bovenscheldebekken

Damse Vaart (Apertje) Damse Vaart - Oostkerke brug Noordgeleed - Oudenburg Noordede - KlemskErk+De Haan Elankenbergse vaart - Zuienkerke tr-0ssewegevaart - Lissewege Waggelwater-Brugge

Hoge Dijken - Ettelgem-Oudenburg Lippensgoed - Bulskampveld Boudewijnkanaal Boudewijnkanaal Boudewijnkanaal Leopoldkanaal Hollandergatkreek - St. Laureins Zuídlede - Mendonk

Klaverbladvijvers (afiratering) - Wachtebeke Bosdamvijver - Wachtebeke Moervaart Daknam B[aarnreersen - Drongen Watersportbaan - Drongen Malem-stadsv'rjvers - Gent Bourgoyen - Gent Boerekreek - §t.

Jan-in-Eremo È

Durme-Lokeren Leopoldkanaal Het Broek-vijver 1 Het Broek-vijver 3 Het Broek+ijver 4 Hazewinkel-roeivijver -

Willebroek

O

Fort uan Oetegem -

Vriesel-Oelegem

r]

I

Rivierenhof-grote hengelvijver - Deurne

Galgen'*eel -,4ntwerpen L.O.

Appendix

1: Glossary

Prevalence,

expressed

as a

percentage, is the number

of individuals

of

a host species infected

with

a

particular

parasite species

divided by

number

of

hosts examined.

Intensity

is

defïned

as

the

number

of

individuals

of

a

particular

parasite species

in

each

infected

host.

Mean

intensity

is the

total

number

of

parasites

(of

one

particular

species)

in

a sample

of

host

species

divided by

the number

of

infected individuals

of

the host

species

in

that

sample, i.e.

the mean number

of

parasites

per infected host.

Ahundance

is defined

as

the

total

number

of

a

particular

parasite species

in

a

sarmple

of

hosts

divided

by

the total

number

of

individuals

of

the host

species

examined, infected

and

uninfected.

If

the prevalence

is

l}}yo,the

mean

intensity

equals the abundance.

The dispersal coefÍicient

is

calculated as

the division

of

the

variance

by

the

rnean

and

provides

an iridex

for

the

degree

of

overdispersion

of

the

parasite

species

within its

host

Leiebekken Maasbekken Maasbekken Maasbekken Maasbekken Maasbekken Maasbekken Maasbekken Maasbekken Maasbekken Maasbekken Maasbekken Maasbekken Maasbekken Maasbekken Maasbekken Maasbekken Maasbekken Maasbekken Maasbekken Maasbekken Maasbekken Maasbekken Netebekken Netebekken Netebekken Netebekken Netebekken Netebekken Netebekken Netebekken Netebekken Netebekken Netebekken Netebekken Netebekken Netebekken Netebekken Lei+St. Martens-Leerne Zuid-Wllemsvaart - Rekem Zuid-\Mllemsvaart - Lanklaar Zuid-Willemsvaart - Rotem-Dilsen Zuid-\Mllemsvaart - Bree Zuid-\Mllemsvaart - Bocholt

Den Aerd-Eí0-put - Minderhout

Warmbeek - Achel-Kluis Dommel- Overpelt Dommel- Neerpelt Abeek - Bocholt Abeek - Kinrooi Itterbeek - Kinrooi Bosbeek - Opoeteren Maas - ltteren Maas - Meeswijk Maas - Stevensweert Maas - Lixhe Albertkanaal - Vroenhoven Albertkanaal- Briegden Benltrinne - Moelingen HochterbampÈNeerharen

Steenberg tss plas 1 & 2- Kessenich Fort van Walem - Walem

Spildoomvijver

Kleine Nete - Dessel Kleine Nete - Olen Kleine Nete - Bouwel Netekanaal

Grote Nete - Hulsen-Meerhout Grote Nete - Westerlo

Grote Nete - Bevel Beneden Nete - Duffel

Bovenscheldebekken Bovenscheldebekken Bovenscheldebekken Bovenscheldebekken Bovenscheldebekken Bovenscheldebekken Bovenscheldebekken Bovenscheldebekken Bovenscheldebekken Bovenscheldebekken Bovenscheldebekken Bovenscheldebekken Bovenscheldebekken Bovenscheldebekken Demerbekken Demerbekken Demerbekken Denderbekken Denderbekken Denderbekken Denderbekken Denderbekken Dijle- en Zennebekken Dijle- en Zennebekken Dijle- en Zennebekken Dijle- en Zennebekken ljzerbekken ljzerbekken ljzerbekken ljzerbekken tjzerbekken ljzerbekken ljzerbekken ljzerbekken ljzeöekken ljzerbekken ljzerbekken ljzerbekken ljzerbekken Leiebekken Leiebekken Leiebekken Leiebekken Leiebekken Leiebekken Leiebekken Leiebekken Leiebekken Leiebekken Leiebekken Leiebekken Leiebekken Leiebekken Oude Schelde-Heuvel Oude Schelde-Nederzwalm Oude Schelde-Grootmeers Oude Schelde-Meilegem Oude Schelde-Meilegem Oude Schelde-Semmezake Oude Schelde-Melsen Oude Scheld+Zwijnaarde Oude Scheld+Zevergem Kalkevaart-Kalken D riesesloot-Schel le belle Bellebeek-Uitbergen Schelde-Pottes Schelde - Oudenaarde Albertkanaal-Lan gerlo Albertkanaal-Hasselt Albertkanaal-Tervant De Gavers-Geraardsbergen Dender-ldegem Dender-Ninove Dender-Liederkeke Dender - Appels Ganzepoot Groenendaal Ganzepoot Groenendaal

Hen gelvijver Groenendaal Meer van Weerde

Steengracht - Steenkerke Proostdijkvaart - Booitshoeke Grote Beverd'rjk - LoReninge

Grote Beverdijk - Stuivekenskerke Yzer - Nieuwpoort Yzer - Diksmuide Yzer - Roesbrugge leperkanaal- leper Groot Geleed-Zevekote Kanaal Nieuwpoort-Plassendale Kanaal Nieuwpoort-Flassendale Kreek van Nieuwendamme - Nieuwpoort

Madslovaart - Keiem

Oude Leie-St. Ellois-vijve

Oude Leie-St. Baafs-vijve

Oude Leie-Bavikhove

Oude Leie-Wevelgem

Oude Leie-Machelen

Oude Leie-Gottem

Oude Leie - \Melsbeke

Appendix

3:

Complete dataset

oi Anguillicola

crassus.

Length of

the eel

in cm,

Weigth of

the

eel

in g; Wal! =

thickness

of

the

swimbladder wall where 1=

<

I

mm,

2=

1-3 mm, 3= >

3

mm; Moist = presence and colour of the moisture in the lumen; L2

second-stagelarvae; L3=third-stagelarvae;Caps=encapsulatedlarvae; P=pre-adults;

M=maleadults;

F=femaleadults;

Rem

=

remnants;

Total

=

total

number

of

nematodes

where

*

indicates

infection.

Sample Basin

tength Weight

Rupture

Moist L2 L3

L4

Caps P M

V Rem