Aquaculture research, tập 41, số 6, 2010

It was found that at a dietary level of 11.7 mg kg 1Se, a total selenium level in the ¢llet of 0.7 mg kg 1 was reached in a relatively short enrichment period of 10 days before harvest..

Enrichment of the African catfish Clarias gariepinus (Burchell) with functional selenium originating from garlic: effect of enrichment period and depuration

on total selenium level and sensory properties

Edward Schram1, Rian A A M Schelvis-Smit1, Jan W van der Heul1& Joop B Luten1,2

1 IMARES,Wageningen UR, IJmuiden,The Netherlands

2 No¢ma Marine,Troms, Norway

Correspondence: E Schram, IMARES, Wageningen UR, PO box 68,1970 AB, IJmuiden,The Netherlands E-mail: edward.schram@wur.nl

Abstract

We wanted to optimize the procedure for the

sele-nium enrichment of farmed African cat¢sh, using

garlic as dietary selenium source In the ¢rst

experi-ment we established the relation between the length

of the selenium enrichment period and the resulting

total selenium level in the ¢llet of the ¢sh It was

found that at a dietary level of 11.7 mg kg 1Se, a total

selenium level in the ¢llet of 0.7 mg kg 1 was

reached in a relatively short enrichment period of 10

days before harvest In the second experiment we

stu-died the e¡ect of depuration on the selenium level in

the ¢llet and the sensory properties of

selenium-enriched African cat¢sh It was found that total

sele-nium levels in the ¢llet were not a¡ected during a

7-day depuration period, while garlic odours and

£a-vours in the raw and cooked ¢llets were signi¢cantly

reduced after 2 days of depuration.We concluded that

selenium enrichment of farmed African cat¢sh can

be obtained by selenium-enriched ¢nishing diets,

while garlic odours and £avours resulting from

diet-ary garlic can be e¡ectively reduced in the ¢llet

dur-ing a short depuration period without negatively

a¡ecting ¢llet levels of total selenium

Keywords: African cat¢sh, functional food,

sele-nium, garlic, depuration, sensory properties

Introduction

Dietary selenium is essential for human health

(Ray-man 2000; Birringer, Pilawa & Flohe 2002) and at

concentrations above dietary requirement, selenium

is reported in several studies as having genic e¡ects in humans (Ip 1998) However in someEuropean countries the average daily intake is lowerthan the recommended daily intake (Rayman 2005).The need for selenium has resulted in an increase inselenium-rich functional foods (Dumont, Vanhaecke

anti-carcino-& Cornelis 2006), including the development of nium-enriched farmed ¢sh

sele-Functional food has been de¢ned in a Europeanconsensus document as ‘a food can be regarded asfunctional if it is satisfactorily demonstrated to a¡ectbene¢cially one or more target functions in the body,beyond adequate nutritional e¡ects, in a way that isrelevant to either improved stage of health and well-being and/or reduction of risk of disease A functionalfood must remain food and it must demonstrate itse¡ects in amounts that can normally be expected to

be consumed in the diet: it is not a pill or a capsule,but part of the normal food pattern’ (Diplock, Aggett,Ashwell, Bornet, Fern & Roberfroid 1999) Aquacul-ture provides an excellent opportunity for the pro-duction of functional seafood products as manyfactors that determine the composition of the edibleportion of ¢sh can be controlled under farming con-ditions

A number of studies have demonstrated the hancement of the selenium concentration in themuscle tissue of ¢sh as a result of elevated levels ofdietary selenium (see Schram, Pedrero, CaŁmara, Vander Heul & Luten 2008 for an overview) None ofthese trials used garlic as source of functional sele-nium species while Ip and Lisk (1996) showed that

en-the anti-carcinogenic e⁄cacy of en-the selenium species

in garlic is superior to selenomethionine and selenite

(in Ip 1998)

In our previous work we took the ¢rst step towards

the development of farmed African cat¢sh enriched

with functional selenium using garlic as selenium

source (Schram et al 2008) We established a dose^

response relation (feed to ¢sh), the growth

perfor-mance in relation to dietary selenium and garlic and

a ¢rst step was taken towards establishing the

reten-tion of funcreten-tional selenium in the edible porreten-tion of

the ¢sh However, to e⁄ciently use the

selenium-enriched garlic resources we also need to know the

minimally required length of the selenium

enrich-ment period to reach target concentrations in the ¢sh

¢llet In addition, the e¡ect of depuration, a

neces-sary procedure to eliminate o¡-£avours (Tucker

2000) during which ¢sh are not fed, on the selenium

level in the ¢llet needs to be established to ensure

tar-geted selenium levels are reached at harvest The use

of garlic as a ¢sh feed ingredient demands evaluation

of the sensory properties of the ¢sh at harvest and in

relation to depuration, because the e¡ects of dietary

garlic on sensory properties of ¢sh are unknown but

likely to be present

Materials and methods

Introduction

Two experiments were performed In the ¢rst

experi-ment ¢sh were fed a selenium-enriched feed for six

di¡erent periods before harvest to investigate the

ef-fect of the length of the enrichment period on the

total selenium level in the ¢llet at harvest Garlic

was used as dietary selenium source Fillet samples

were analysed for total selenium

In the second experiment selenium enrichment

was followed by a depuration period during which

¢sh were sampled at intervals to assess the total

sele-nium level and the sensory properties

Experimental feeds

For both experiments two feeds were used:

selenium-enriched and a control feed Selenium enrichment

was achieved by inclusion of selenium-enriched

gar-lic This garlic was grown and processed by Plant

Research International,Wageningen UR, the

Nether-lands, with selenium forti¢cation of the soil as

de-scribed in Larsen et al (2006), which resulted in

garlic with a high selenium content Garlic was plied as a dry powder and included as such in thefeed The control feed was the same except for theinclusion of garlic Experimental feeds were pro-duced by Research Diets Services, the Netherlands,

sup-as 4 mm steam pellets, with a proximate analysis of46.8% crude protein,13.6% crude fat,1.1% crude ¢breand 10.1% crude ash The formulation of the experi-mental feeds is shown in Table 1

Experiment 1: total selenium levels andselenium retention in relation to the length ofenrichment period

The six treatments consisted of six selenium ment periods of di¡erent length Groups of ¢sh werefed the selenium-enriched feed for 35, 25, 17, 10 or

enrich-5 days before harvest The experimental period lastedfor 35 days for all treatments Fish were fed the con-trol feed (Table 1) before receiving the selenium-enriched feed Table 2 provides an overview of thetreatments

Treatments were randomly assigned to the tanks,with triplicate tanks for each of the six treatments.The experimental set up consisted of 18, 30 L glasstanks placed in two rows of nine tanks Tanks were

£own through with tap water with a selenium level

ofo0.5 mg Se L 1at a rate of 10 L h 1 Tank e¥uentswere discharged as ¢sh are known to take up water-borne selenium across the gills (Hodson et al 1980)and recirculation of water combined with leaching

of selenium from the feeds and excretion by the ¢shcould have resulted in transfer of selenium com-pounds between tanks.Water temperature was main-tained at 25 1C throughout the experimental period.Each tank was stocked with 14 African cat¢sh

Table 1 Formulation of the experimental feeds on the wet weight basis

Feed component Control feed

enriched feed

Selenium enrichment of African cat¢sh part II E Schram et al Aquaculture Research, 2010, 41, 793–803

(Clarias gariepinus) at a sex ratio of one male to one

female African cat¢sh were obtained from

Fleuren-Nooijen BV, the Netherlands, a commercial African

cat¢sh hatchery After stocking, the ¢sh were allowed

to acclimatize to the experimental conditions for

5 days, during which all ¢sh were fed the control feed

(Table 1) The day after the acclimatization period is

referred to as day 1 of the experimental period Fish

were fed by hand to visually observe satiation twice

daily (10:00 and 17:00 hours) The total feed load

was documented for each tank Mortalities were

recorded daily and dead ¢sh were removed

immedi-ately from the tanks after detection On days 1 and

35 (harvest), the total biomass and number of ¢sh

were determined for each tank The mean (SD)

indi-vidual weight of 230 (6.7) g at day 1 was not

signi¢-cantly di¡erent between tanks [analysis of variance

(ANOVA), Po0.05] Before stocking of the experimental

tanks on day 1, three males and three females were

randomly sampled from the total stock On day

35, three males and three females were sampled

ran-domly from each tank Sampled ¢sh were ¢lleted,

¢l-lets were pooled per tank, homogenized and stored

frozen at  70 1C Feed samples were taken at day 1

of the experiment and stored frozen at 70 1C

Experiment 2: e¡ect of depuration on total

selenium level and sensory properties of the

¢llet

Treatments consisted of two di¡erent experimental

feeds: control feed and selenium-enriched feed

(Table 1) The experimental period consisted of two

parts The ¢rst part of the experimental period was a

rearing period of 21 days The second part of the

experimental period was an 8-day depuration period

during which ¢sh were sampled at regular vals The total experimental period lasted for 29 days.Treatments were randomly assigned to the tanks, withtriplicate tanks for each of the two treatments Theexperimental set up consisted of six, 400 L plastictanks placed in two rows of three tanks Tank e¥u-ents were discharged as recirculation could have re-sulted in transfer of selenium compounds betweentanks Each tank was stocked with 38 African cat¢sh(C gariepinus) with a mean (SD) individual weight of

inter-796 (11) g at a sex ratio of one male to one female Themean initial individual weight was not signi¢cantlydi¡erent between tanks (ANOVA, Po0.05) After stock-ing, the ¢sh were allowed to acclimatize to the experi-mental conditions for 5 days, during which all ¢shwere fed a commercial grower diet The day after theacclimatization period is referred to as day 1 of theexperimental period During the ¢rst 21 days ofexperimental period the tanks were £own throughwith 25 1C tap water with a selenium level ofo0.5 mg Se L 1

at a rate of 35 L h 1 At day 22, thestart of the depuration period, the £ow was increased

to 60 L h 1 On days 1 and 22, the total biomass andnumber of ¢sh was determined for each tank Beforestocking of the experimental tanks on day 1, threemales and three females were randomly sampledfrom the total stock On days 22, 23, 24, 25 and 29,concurring with days 0, 1, 2, 3 and day 7 of the de-puration period, three males and three females weresampled randomly from each tank Sampled ¢shedwere ¢lleted Per ¢sh one ¢llet was used for total sele-nium analysis and one ¢llet for sensory analysis.Fillets sampled for total selenium analysis werepooled per tank (six ¢llets per tank per samplingpoint), homogenized and stored frozen at  70 1C

Production parameters

The speci¢c growth rate (SGR) for of the group of ¢sh

in each tank was calculated as follows:

SGR¼ ðlnðWtÞ  lnðW0ÞÞ 100

Twhere SGR is the speci¢c growth rate (%BWday 1),

BW is the body weight, Wtis the average individualweight at harvest (g), W0is the average individualweight at stocking (g) and T is the number of days.The feed conversion rate (FCR) per tank was calcu-lated as follows:

Proportion of total feed load (%)

Presented are for each treatment the experimental days during

which the control (not selenium-enriched) feed and the

sele-nium-enriched feed were fed and the relative proportion of the

selenium-enriched feed load to the total feed load.

where FCR is the feed conversion rate (g g 1), FL is

the total amount of feed administered to a tank

dur-ing the experiment (g), Wtis the average individual

weight at harvest (g), W0is the average individual

weight at stocking (g), Ntis the number of ¢sh per

tank at harvest and N0is the number of ¢sh per tank

½ feedFL

 100%

where R is the selenium retention rate in the ¢sh ¢llet

(% of dietary intake), [Se]tis the total selenium

con-centration in the ¢llet at harvest (mg kg 1),Wtis the

average individual weight at harvest (kg), Ntis the

number of ¢sh per tank at harvest, F is the ¢lleting

yield (%), [Se]0is the total selenium concentration

in the ¢llet at stocking (mg kg 1), W0is the average

individual weight at stocking (kg), N0is the number

of ¢sh per tank at stocking, [Se]feedis the total

sele-nium concentration in feed (mg kg 1) and FL is the

total feed load (kg)

Total selenium analysis

Homogenized samples of the muscle tissue were

mi-crowave digested in mimi-crowave-lined digestion

ves-sels using 70% (m/v) nitric acid (5 mL) and 30%

(m/v) hydrogen peroxide (5 mL) as the oxidant

mix-ture After digestion, the samples were reduced in

the microwave by adding 5 mL 37% (m/v)

hydrochlo-ric acid to convert Se61into Se41 The resulting

solu-tion was diluted to a ¢nal volume of 25 mL using 10%

hydrochloric acid Blanks [5 mL 70% (m/v) nitric acid

and 5 mL 30% (m/v) hydrogen peroxide] were treated

in the same way Total selenium was determined

using hydride generation Flow Injection Analysis

System^Atomic Absorption Spectrometry NaBH4

(0.2%) in NaOH (0.05%) was used to generate the

H2Se The accuracy of the analyses was established

by analysing the selenium content in the

BCR-certi-¢ed Cod-CRM422 reference sample

Sensory analysis

The e¡ect of dietary garlic and the e¡ect of

depura-tion on the sensory properties of the

selenium-enriched African cat¢sh were assessed in a sensory

intensity test in Experiment 2 On days 19, 21, 22 and

26, concurring with days 0, 2, 3 and 7 of the tion period, three males and three females weresampled randomly from each tank, yielding 18 ¢shper treatment From each ¢sh, one ¢llet was used forsensory analysis The ¢sh were ¢lleted on the day ofslaughter and stored in vacuum at  25 1C On theday of testing six randomly selected ¢llets per treat-ment were thawed for 60 min in running cold tapwater, pooled per treatment and homogenized bycooled mincing (type DRC compact 92, France) Sam-ples (approximately 50 g) of the homogenized minced

depura-¢llets were put in small aluminium boxes with athree-digit random code and stored at 0 1C until pre-paration Samples were prepared by placing the alu-minium boxes in 1cm boiling water in a hot air oven(Miele H 216, Miele, Germany) at 160 1C for 7 min.Eight members of the Wageningen IMARES sen-sory panel participated in the sensory analysis of theAfrican cat¢sh samples All panelists were trainedbefore sensory analysis in two1-h sessions using sam-ples of all experimental treatments (seleniumenriched, control) and sampling days (0, 1, 3 and 7days of depuration) in accordance with internationalstandards (ISO 1993) Panelists were trained to detect,recognize, describe and scale the intensity of odoursand £avours of raw and cooked cat¢sh, using an exist-ing sensory evaluation scheme for African cat¢sh as astarting point The resulting vocabulary used to de-scribe odours and £avours (attributes) is listed in Table

3 Intensities of each attribute were scaled using anine-point intensity scale For each of the four sam-pling days, duplicates of the selenium-enriched ¢llet

Table 3 Mean (SD, n 5 3) total selenium levels in African cat¢sh ¢llets and selenium retention (mean1SD, n 5 3) for di¡erent selenium enrichment periods before harvest

No.

Treatment (no.

of selenium enrichment days)

Total selenium level (mg kg 1)

Selenium retention (%)

LSD, least square di¡erence of means at 5% signi¢cance level;

Selenium enrichment of African cat¢sh part II E Schram et al Aquaculture Research, 2010, 41, 793–803

samples were evaluated in random order by each

pa-nelist in a single session A computerized system (FIZZ,

version 2.10a 1994^2000, Biosyste¤mes, Couternon,

France) was used for data recording

Statistics

Statistical analysis was conducted withGENSTAT10.1

Di¡erences in mean values of SGR, feed conversion

rate, weight gain, selenium retention and selenium

levels between treatments were analysed using

selenium enrichment period and the response

vari-ates selenium level in the ¢llets and selenium

reten-tion were analysed with non-linear regression using

a two straight-line model The individual mean body

weights at the sampling days in the depuration period

were tested for signi¢cant di¡erences between and

within treatments using repeated measurements

ANOVA Mean ¢llet weights were tested for signi¢cant

di¡erences among sampling days and between

treat-ments using repeated measuretreat-mentsANOVA In case

of signi¢cance, one-wayANOVAand repeated

mea-surementsANOVAwere followed using the least

signif-icant di¡erence (LSD) post hoc analysis to detect the

di¡erent mean values A signi¢cance level of 5% was

used in all cases

For the sensory analysis, the statistical analyses

were performed inSPSSversion 15.0 Mean values per

sample and per attribute were calculated Di¡erences

in the mean values were tested for signi¢cance using

one-wayANOVA, followed by the least signi¢cant

dif-ference (LSD) post hoc analysis to detect the di¡erent

mean values A signi¢cance level of 5% was used

Pearson’s correlation tests (two tailed) were

per-formed to ¢nd correlations between attributes, for

which a signi¢cance level of 1% was used

Results

Experiment 1: total selenium levels and

selenium retention in relation to the length of

enrichment period

The total selenium levels in the ¢llets were di¡erent

between treatments (ANOVA, Po0.001) and increased

with increasing length of the selenium enrichment

period (Table 4) The two straight-line regression

model accounted for 98% of the variance and

re-sulted in an in£ection point at a selenium

enrich-ment period of 10.3 days (Fig 1) Selenium retention

in the ¢llet was di¡erent between treatments (ANOVA,

Po0.001) (Table 4) The two straight-line regressionmodel accounted for 76% of the variance andresulted in a sharp in£ection point at a seleniumenrichment period of 5.8 days (Fig 2) Seleniumretention ¢rst increased with an increasing length

of the selenium enrichment period up to 5.8 days.Beyond 5.8 days of feeding selenium-enriched feedbefore harvest, selenium retention decreased (Fig 2).Final weight (ANOVA, P 5 0.14), SGR (ANOVA, P 50.17) and FCR(ANOVA, P 5 0.56) were not di¡erentbetween treatments (Table 5)

Table 4 Mean (standard deviation, n 5 3) ¢nal weight, speci¢c growth rate (SGR), feed conversion rate (FCR) for the treatments

No.

Treatment (Se feeding days before harvest)

Final weight (g)

SGR (%BW day 1)

FCR (g g 1)

4020

cat¢sh ¢llets in relation to the length of the selenium richment period before harvest Line 1, selenium level ¢l-

days10.612 In£ection point (10.3, 0.72) The two line regression model accounts for 98% of the variance

straight-Experiment 2: e¡ect of depuration on total

selenium level in the ¢llet

After 21 days of ¢sh rearing, the total selenium levels

in the ¢llets were 0.62 mg kg 1for the

selenium-enriched ¢sh and 0.16 mg kg 1for the ¢sh that ceived the control feed (Table 6) The ¢nal total mean(SD, n 5 38) body weight after 21 days of ¢sh rearingwas 1083 (25) g for the selenium-enriched ¢sh and

re-1071 (19) g for the ¢sh that received the control feed.Mean (SD, n 5 3) SGR was 1.46 (0.07)%BWday 1and1.42 (0.10)%BWday 1 Final weight (ANOVA, P 50.545) and SGR (ANOVA, P 5 0.614) were not di¡erentbetween treatments

The length of the depuration period was found tohave no e¡ect on the total selenium levels in the ¢llets

of both the selenium-enriched ¢sh (simple linearregression, P 5 0.06, Fig 3) and the ¢sh that receivedthe control feed (simple linear regression, P 5 0.31,

Days of feeding selenium enriched feed prior to harvest

Selenium retention in the fillet (% of Se load)

Line2

Line1

Figure 2 Selenium retention (%) in African cat¢sh ¢llets

expressed as the percentage of the total dietary selenium

feed load in relation to the length of the selenium

enrich-ment period before harvest Line 1, selenium retention 5

days10.099 In£ection point (5.81, 0.093) The two

straight-line regression model accounts for 76% of the variance

Table 5 Mean (SD, n 5 3) total selenium levels in the ¢llets

(mg kg 1) and mean (SD, n 5 3) ¢llet weight (g) during the

depuration period for the selenium-enriched ¢sh and the

¢sh that received the control feed

Depuration

day 

Total selenium level in

the fillet (mg kg 1) Fillet weight (g)

Selenium

enriched Control

Selenium enriched Control

Depuration days 0, 1, 2, 3 and 7 concur with days 22, 23, 24, 25

and 29 of the experimental period.

Total selenium levels were equal within treatments for all

sam-pling days during the depuration period for the

P treatment  time 5 0.31).

Table 6 Sensory vocabulary for the selenium-enriched African cat¢sh using dietary garlic as the selenium source

Short name

Sensory attribute Description of attribute

Odour of the raw fillet

Odour of the cooked fillet

potato

Odour of cooked potatoes Flavour of the cooked fillet

potato

Flavour of cooked potatoes

0.000.100.200.300.400.500.600.700.80

Depuration period (Days)

–1)

the ¢llet of African cat¢sh in relation to the length of the

relation is not signi¢cant

Selenium enrichment of African cat¢sh part II E Schram et al Aquaculture Research, 2010, 41, 793–803

Fig 3) Within treatments, the total selenium level

in the ¢llet was equal for all sampling days for the

se-lenium-enriched ¢sh (ANOVA, P 5 0.10) and for the

¢sh that received the control feed (ANOVA, P 5 0.14)

(Table 6)

African cat¢sh lost weight during the

depura-tion period (Fig 4) Mean individual weights were

not di¡erent between treatments for each of the

sampling days during the depuration period

(repeated measurements ANOVA, Ptreatment time5

0.45) Overall mean weights were di¡erent for

sam-pling days 0, 1 and 2 (repeated measurements

period, the overall mean weight loss was 8.7% of the

individual weight at the start of the depuration

per-iod Mean individual ¢llet weights were equal for

all sampling days during the depuration period

between treatments and within treatments

(re-peated measurements ANOVA, P time  treatment5

0.31,Table 6)

Experiment 2: sensory properties of enriched African cat¢sh ¢llets in relation todepuration

selenium-Raw and cooked samples of African cat¢sh ¢llets fedwith selenium-enriched feed (with garlic as seleniumsource) were found to have a rather intense garlicodour and £avour at day 0 of the depuration period,which had signi¢cantly decreased at day 2 of depura-tion period (Table 7, Fig 5) No changes in the inten-sity of these attributes were observed between days 2and 3 of the depuration period The intensity of theodour of both the raw and cooked samples was signi-

¢cantly reduced between days 3 and 7 of the tion period, while the intensity of the £avour of thecooked ¢llet stabilized at a minimal detectable level

depura-in this period (Fig.5) The garlic odour correlates withthe metallic odour and with the low values of sour

Figure 4 Overall mean (n 5 6) individual weight (g) of

African cat¢sh during the depuration period Data points

marked with di¡erent letters di¡er signi¢cantly (repeated

123456789

Depuration period (Days)

Figure 5 Sensory intensity (score 1^9) in relation tothe length of the depuration period for the garlic odour

odour (Table 8) The intensity scores of the other

sen-sory attributes were the same for all sampling days

during the depuration period (Table 7) No o¡ £avours

were detected in the ¢llet samples

Discussion

Total selenium levels and selenium retention

in relation to the length of enrichment period

The daily selenium intake among Europeans is on

average approximately 100mg short of the

recom-mended daily intake of 150mg day 1.We aimed to ¢ll

this gap with one 150 g portion of selenium-enriched

African cat¢sh ¢llet, which demands a ¢llet selenium

concentration of approximately 0.70 mg Se kg 1

This target concentration was reached after an

en-richment period of 10 days before harvest, using a

dietary selenium level of 11.7 mg kg 1 This clearly

demonstrates that selenium enrichment of African

cat¢sh can be achieved by feeding selenium-enriched

feeds for a relatively short period before harvest

We therefore consider the selenium-enriched feed a

¢nishing diet Finishing diets for farmed ¢sh have

been studied for Atlantic salmon (e.g Jobling, Larsen,

Andreassen, Olsen & Sigholt 2002; Bell, Henderson,

Tocher & Sargent 2004) and red seabream (Glencross

et al., 2003), all successfully aiming to boost n-3

poly-unsaturated fatty acid (PUFA) levels before harvest to

compensate for initially reduced n-3 PUFA levels due

to growing ¢sh on feeds containing vegetable oils

Finishing diets aiming to boost levels in farmed ¢sh

of nutrients important in human nutrition other

than n-3 PUFA, such as selenium, have not been

studied to date Apart from the current study, there

are no records of studies aiming to optimize the

enrichment process in terms of the minimal required

feeding period for ¢nishing diets to reach target levels

in the second experiment had reached a mean dual weight over 1kg and a selenium level in the ¢llet

indivi-of 0.63 mg kg 1(Table 5), which is well in line withthe targeted 0.70 mg kg 1Se

The highest selenium retention from feed to ¢lletwas reached when feeding the ¢sh the selenium-enriched feed during 10 days before harvest Thehigher retention of selenium in the treatment groupsfed selenium-enriched feeds as compared with thecontrol diet is most likely related to the presence of

Table 8 Correlation matrix for the intensity scores of the sensory attributes of the ¢llets during the depuration period for the selenium-enriched African cat¢sh

Attribute R O metal R O garlic R O sour C O garlic C O cook potato C T musty C T garlic

C T cook potato

For values marked with the correlation is signi¢cant (the two-tailed Pearson correlation test, Po0.01).

Selenium enrichment of African cat¢sh part II E Schram et al Aquaculture Research, 2010, 41, 793–803

selenomethionine in the selenium-enriched feeds

(Schram et al 2008), which can substitute for

methio-nine in protein synthesis As a result,

selenomethio-nine is more easily incorporated in the muscle tissue

of growing ¢sh than the inorganic selenium in the

control feed (Waschulewski & Sunde 1988) The low

retention of selenium obtained after 5 days of

sele-nium enrichment before harvest as compared with

the other treatment groups that received

selenium-enriched feed can be explained by the delay between

feed intake and actual incorporation of selenium

compounds in the ¢llet Such a delay would have

caused that the selenium intake from the last meals

before harvest were not incorporated in the ¢llet As

this unknown number of last meals before harvest

account for a larger proportion of the total selenium

intake for short enrichment periods, the selenium

retention is underestimated when based on the full

enrichment period of 5 days

The decreasing selenium retention with

increas-ing length of the enrichment period (Fig 2) can be a

re£ection of a £attening of selenium incorporation

in the ¢llet, which is likely to occur as the amount of

selenium that can be incorporated in the ¢llet is

limited (Schram et al 2008)

E¡ect of depuration on total selenium level in

the ¢llet

Depuration was found to have no e¡ect on the total

selenium level in the ¢llet for both the

selenium-enriched ¢sh and the control treatments (Fig 3) As

¢llet weight of the selenium-enriched African cat¢sh

did not decrease during the depuration period, despite

the considerable total body weight loss (Table 5), the

absolute amount of selenium in the ¢llet was also not

a¡ected by depuration

Sensory properties of selenium-enriched

African cat¢sh ¢llets in relation to depuration

The intensity of the garlic odour and £avour of the

African cat¢sh samples were found to decrease

sig-ni¢cantly during the ¢rst 2 days of the depuration

period (Fig 4), showing that depuration is probably

an e¡ective measure to eliminate the e¡ects of

diet-ary garlic on the sensory properties of the ¢nal

pro-duct, without a¡ecting other sensory characteristics

(Table 7)

As we were mainly interested in the changes of the

sensory properties of the garlic-fed African cat¢sh

during the depuration period and due to the absence

of information on the sensory properties of garlic-fed

¢sh, we used a sensory intensity test in this ment and not a sensory di¡erence test nor a sensorythreshold test Using a sensory di¡erence test wouldhave enabled us to detect di¡erences in the sensoryproperties of the garlic-fed and the non-garlic-fedAfrican cat¢sh for each of the individual samplingdays during the depuration period However, wewould not have been able to detect any changes inthe sensory properties during the course of the de-puration period Using a sensory threshold test wewould have been able to determine exactly the re-quired length of the depuration period to eliminategarlic odours and £avours, but for such a test, preli-minary information on the changes in the sensoryintensity during the course of the depuration period

experi-is needed

This means that the current results give a strongindication that it is possible to reduce the garlic £a-vour of the garlic-fed African cat¢sh to minimal sen-sory detection levels within 3 days of depuration, but

a sensory threshold test including samples of bothgarlic-fed and non-garlic-fed ¢sh is needed to con¢rmthis

During the experiment, a strong garlic-like odourcould be smelled in the experimental room, suggest-ing that African cat¢sh excrete compounds causing agarlic odour after garlic consumption, but this re-mains to be con¢rmed by measurements of garlicodour causing compounds Odours resulting fromgarlic consumption have been studied in humans.Garlic odour in human breath following garlic con-sumption is caused by allyl mercaptan, a water solu-ble and indigestible sulphide compound that is taken

up in blood from the intestinal tract, followed by cretion via the skin and lungs (Takeshi, Boku, Inada,Morita & Okazaki 1989) A similar route in ¢sh canexplain our observations in this study Potential ex-cretion routes of compounds causing garlic odour in

ex-¢sh include faeces and urine, but after uptake of thecompounds from the intestinal tract in the blood, ex-cretion (of metabolites) can also take place via thegills

Next to uptake from the intestinal tract, the ing water contaminated by faeces or urine with com-pounds causing garlic £avours could be considered

rear-as a source of these compounds in the ¢llets, rear-as ¢shcan take up contaminants from the surroundingwater via the gills (Streit 1998) Either way, uptake

in the blood of compounds causing garlic odour and

£avour is needed to explain their presence in the

African cat¢sh ¢llets, not necessarily followed by

in-corporation from the blood in ¢llet tissues as the ¢sh

were not bled during slaughter

All in all we consider excretion via the gills of

pounds responsible for garlic odour and £avour,

com-bined with removing their source by cessation of

dietary garlic intake, a plausible explanation for the

reduction in the intensity of these attributes in the

African cat¢sh ¢llets during depuration

Selenium enrichment of market-sized African

cat¢sh

We previously established that African cat¢sh can be

enriched with selenium via the diet (Schram et al

2008) However, these results were obtained in African

cat¢sh of on average 424 g at the end of the experiment

whereas market size of African cat¢sh is generally over

1kg In Experiment 2 of the present study, the ¢nal

mean (SD) body weight was 1080 (10) g and a mean

(SD) total selenium level of 0.63 (0.02) mg kg 1in the

selenium-enriched ¢sh Although a direct comparison

cannot be made as for the di¡erences in enrichment

periods and dietary selenium levels, these current

¢nd-ings show that selenium enrichment can also be

reached in market-sized ¢sh

Conclusions

Based on the current study we conclude that the

length of the selenium-feeding period is an

impor-tant factor a¡ecting the selenium level in the ¢sh

¢l-let and that selenium enrichment during a relatively

short period before harvest is su⁄cient to reach

tar-get levels In addition we conclude that depuration

has no e¡ect on total selenium level in the ¢sh ¢llet,

whereas o¡ £avours associated to dietary garlic are

e¡ectively removed by a short depuration period

Acknowledgments

This study was part of SEAFOODplus, an integrated

research project ¢nancially supported by the

Eur-opean Commission under FP6 (http://www.seafood

plus.org) The authors would like to express their

gra-titude to Chris Kik and Olga Scholten, Plant Research

International, Wageningen UR, the Netherlands, for

the production and supply of garlic, Marion Hoek

van Nieuwenhuizen and Afke Stein for the total

sele-nium analysis, Fleuren-Nooijen, Someren, the

Neth-erlands, for the supply of the experimental ¢sh,

Research Diet Services, Wijk bij Duurstede, the

Netherlands, for the production of the experimentalfeeds, Mercedes Careche, CSIC, Madrid, Spain, for pro-ject management and the members of the sensory pa-nel of Wageningen IMARES for the sensory analysis

References

Bell G., Henderson R.J.,Tocher D.R & Sargent J.R (2004) placement of dietary ¢sh oil with increasing levels of lin- seed oil: modi¢cation of £esh fatty acid compositions in Atlantic Salmon (Salmo salar) using a ¢sh oil ¢nishing diet Lipids 39, 223^232.

Re-Birringer M., Pilawa S & Flohe L (2002) Trends in selenium biochemistry Natural Product Reports 19, 693^718 Diplock A.T., Aggett P.J., Ashwell M., Bornet F., Fern E.B & Roberfroid M.B (1999) Scienti¢c concepts of functional foods in Europe: consensus document British Journal of Nutrition 81(Suppl 1), S1^S28.

Dumont E.,Vanhaecke F & Cornelis R (2006) Selenium ciation from food source to metabolites: a critical review Analytical and Bioanalytical Chemistry 385, 1304^1323 Glencross B., Hawkins W & Curnow J (2003) Restoration of the fatty acid composition of red seabream (Pagrus aura- tus) using a ¢sh oil ¢nishing diet after grow-out on plant oil based diets Aquaculture Nutrition 9, 409^418 Hodson P.V., Spry D.J & Blunt B.R (1980) E¡ects on rainbow trout (Salmo gairdneri) of a chronic exposure to water- borne selenium Canadian Journal of Fisheries and Aquatic Sciences 37, 233^240.

spe-ISO (1993) SensoryAnalysis ^ General Guidance for the tion, Training and Monitoring of Assessors Part 1: Selected Assessors, 8586 -1 The International Organization for Standardization, Geneva, Switzerland.

Selec-Ip C (1998) Lessons from basic research in selenium and cancer prevention Journal of Nutrition 128, 1845^1854.

Ip C & Lisk D.J (1996) The attributes of selenium-enriched garlic in cancer prevention Advances in Experimental Medicine and Biology 401, 179^187.

Jobling M., Larsen A.V., Andreassen B., Olsen R.L & Sigholt

T (2002) In£uence of a dietary shift on temporal changes

in fat deposition and fatty acid composition of Atlantic salmon post-smolt during the early phase of seawater rearing Aquaculture Research 33, 875^889.

Larsen E.H., Lobinski R., Burger-Meyer K., Hansen M., Ruzik R., Mazurowska L., Rasmussen P.H., Sloth J.J., Scholten O & Kik C (2006) Uptake and speciation of sele- nium in garlic cultivated in soil amended with symbiotic fungi (mycorrhiza) and selenate Analytical and Bioanaly- tical Chemistry 385, 1098^1108.

Rayman M.P (2000) The importance of selenium to human health The Lancet 356, 233^241.

Rayman M.P (2005) Selenium in cancer prevention: a review of the evidence and mechanism of action Proceed- ings of the Nutrition Society 64, 527^542.

Schram E., Pedrero Z., CaŁmara C.,Van der Heul J.W & Luten J.B (2008) Enrichment of African cat¢sh with functional Selenium enrichment of African cat¢sh part II E Schram et al Aquaculture Research, 2010, 41, 793–803

selenium originating from garlic Aquaculture Research

39, 850^860.

Streit B (1998) Bioaccumulation of contaminants in ¢sh In:

Fish Ecotoxicology (EXS: 86) (ed by T Braunbeck, D.E.

Hinton & B Streit), pp 353^387 Birkhuser, Basel,

Swit-zerland.

Takeshi M., Boku T., Inada K., Morita M & Okazaki Y.

(1989) Odor components of human breath after the

inges-tion of grated raw garlic Journal of Food Science 54, 763pp.

Tucker C.S (2000) O¡-£avor problems in aquaculture Reviews in Fisheries Science 8, 45^88.

Waschulewski I.H & Sunde R.A (1988) E¡ect of dietary methionine on utilization of tissue selenium from dietary selenomethionine for glutathione peroxidase in the rat Journal of Nutrition 118, 367^374.

Influence of the length of time after hormonal

stimulation on selected parameters of milt of

Beata Irena Cejko1, Radosław Kajetan Kowalski1, Dariusz Kucharczyk2, Katarzyna TargonŁska2,Sławomir Krejsze¡2, Daniel Z_arski2& Jan Glogowski1,3

1 Department of Gamete and Embryo Biology, Institute of Animal Reproduction and Food Research, Polish Academy of Sciences, Olsztyn, Poland

2 Department of Lake and River Fisheries, Faculty of Environmental Sciences and Fisheries, University of Warmia and Mazury, Olsztyn, Poland

3 Department of Ichthyology, Faculty of Environmental Sciences and Fisheries, University of Warmia and Mazury, Olsztyn, Poland

Correspondence: B I Cejko, Department of Gamete and Embryo Biology, Institute of Animal Reproduction and Food Research, Polish Academy of Sciences, 10-747 Olsztyn, Poland E-mail: bcejko@poczta.onet.pl

Abstract

Milt of the Leuciscus idus L was collected from ¢ve

ex-perimental groups, and selected parameters of its

quality were analysed for 36 h (group II), 60 h (group

III), 84 h (group IV) and 108 h (group V), respectively,

after hormonal stimulation with Ovopel

(1gran-ule kg 1of body weight) The control (group I) ¢sh

were not subjected to hormonal stimulation The

highest milt volume was obtained from the ¢sh in

group IV (0.70 0.55 mL), where the largest volume

of milt expressed per kilogram was also obtained

(3.03 1.94 mL kg 1) Signi¢cant di¡erences were

also found in milt volumes obtained between group

I and groups III (Po0.01) and IV (Po0.05) The

high-est percentage of motile spermatozoa was found in

the milt of group IV (59%); signi¢cant di¡erences

were found between group I and groups II (Po0.01)

and III (Po0.001) The value of osmotic pressure of

seminal plasma was the highest in group IV

(203.19 37.63 mOsm kg 1

), and the lowest ingroup I (118.31 41.13 mOsm kg 1

) Parameters termining milt quality and quantity indicate that the

de-period of 60^84 h after hormonal stimulation with

Ovopel is optimal for obtaining milt from ide

Keywords: Leuciscus idus L., milt, hormonal

sti-mulation, seminal plasma, motility of spermatozoa,

Ovopel

Introduction

Ide Leuciscus idus L is a reophileous cyprinid ¢sh thatinhabits the rivers of Central and Eastern Europe aswell as Asia The spawning period of this ¢sh in Po-land occurs from April to May and lasts for up to 10days (Tadajewska 2000) The ide is of little economicsigni¢cance but is an important natural component

of the aquatic ecosystem because it increases the duction potential of rivers by occupying niches thatother ¢sh are unable to (Błachuta 1998) This ¢sh isalso important in recreational ¢sheries and in thepropagation of ornamental ¢sh

pro-The study of the biology of reophileous cyprinid

¢sh reproduction is mainly concerned with the duction of quality stocking material in hatcheriesand developing arti¢cial spawning This is becauseriver regulations and pollution have caused declines

pro-in ide populations pro-in Polish waters (Kruk 2007) males and males of the genus Leuciscus are also sen-sitive to negative changes in antropogenic factors innatural waters The production of stocking materialunder controlled conditions can prevent a dramaticdecline in the number of reophileous ¢sh in the eco-system and signi¢cantly increase the number of en-dangered ¢sh species in open waters

Fe-In the available literature, little attention has beenpaid to the genus Leuciscus males and the rudimen-tary information focuses on the characteristics of the

Aquaculture Research, 2010, 41, 804^813 doi:10.1111/j.1365-2109.2009.02357.x

basic determinants of their milt quality, i.e the

vo-lume of milt obtained, motility and concentration of

spermatozoa (Kucharczyk, Kujawa,

Mamcarz,Wyszo-mirska & Ulikowski 1999; Kucharczyk, Borejko,

Tar-gonŁska, Roz’ ek, Chwaluczyk, Kowalski & Glogowski

2007; Krejsze¡, Kucharczyk, Kupren, TargonŁska,

Mamcarz, Kujawa, Kaczkowski & Ratajski 2008;

Tar-gonŁska, Kucharczyk, Mamcarz, Glogowski, Krejsze¡,

PrusinŁska & Kupren 2008) Only a few publications

have provided information on the cryopreservation

of milt of such ¢sh (Lahnsteiner, Berger, HorvaŁth,

Ur-banyi & Weismann 2000), detailed characteristics of

spermatozoa motility parameters using the

compu-ter-assisted sperm analysis (CASA) system (Kowalski,

Sarosiek, Kucharczyk,TargonŁska & Glogowski 2006),

proteolytic enzyme activity (Kowalski, Glogowski,

Kucharczyk, Goryczko, Dobosz & Ciereszko 2003) or

their inhibitors in seminal plasma (Wojtczak,

Glo-gowski, Kołdras, Kucharczyk & Ciereszko 2003)

During spawning, all reproductive functions of

both female and male ides must be maintained The

temperature noted during the ide spawning period

£uctuates between 8 and 15 1C and this range is also

recommended at hatcheries (Kupren 2005)

Sperma-tozoa in males mature in spermatic ducts and are

hormonally (11-ketotestosterone and 17a,

20b-dihy-droxy-4-pregnen-3-one) controlled; the entire

pro-cess of spermatogenesis is conditioned by

environmental factors and is linked to the

reproduc-tion strategy of the given species The milt produced

by the male is characterized by both individual and

seasonal variability (Kruger, Smit, Van Vuren &

Fer-reira 1984; Munkittrick & Moccia 1987; Aas, Refstie &

Gjerde 1991) This applies to spermatozoa

morphol-ogy as well as to organic components and the

bio-chemical parameters of seminal plasma (Glogowski,

KwasŁnik, Piros, D ˛abrowski, Goryczko, Dobosz,

Ku$-minŁski & Ciereszko 2000) Reproduction success is

in£uenced by the quality of gametes produced by

the ¢sh Based on motility and spermatozoa

concen-tration, we can evaluate the biological quality of milt

and the propensity of spermatozoa for fertilization

Consequently, those parameters are characterized

¢rst and referred to as the basic ones (Kruger et al

1984; Rurangwa, Kime, Ollevier & Nash 2004)

It is impossible to conduct controlled reproduction

of ¢sh from genus Leuciscus without hormonal

sti-mulation (Kucharczyk 2002; Krejsze¡ et al 2008)

Controlled reproduction of those ¢sh is conducted

on the basis of stimulation using substances of

natur-al origin, i.e human chorionic gonadotropin (hCG),

carp pituitary homogenate, pure

gonadotropin-releasing hormone (GnRH, LH-RH) or synthetic dotropin releasing-hormone analogues, i.e sGnRHaand mGnRHa (Kucharczyk et al 1999; Kucharczyk2002; Jamro¤z, Kucharczyk, HakucŁ^Błaz’owska, Krejs-ze¡, Kujawa, Kupren, Kwiatkowski,TargonŁska, ’arski,Cejko & Glogowski 2008; Krejsze¡ et al 2008; Ku-charczyk, TargonŁska, ’arski, Kujawa & Mamcarz2008)

gona-The ¢rst synthetic analogue to the reproduction ofreophileous ¢sh was the Hungarian preparation Ovo-pel (HorvaŁth, Szabo¤ & Burke 1997), a mamalian ana-logue GnRH (D-Ala6Pro9NEt-mGnRH) with adopamine receptor antagonist, i.e metoclopramide.Ovopel has been used in ¢shery practice for over 10years because of its wide application in the controlledreproduction of numerous ¢sh species, high e¡ective-ness (ovulation/spermation), low purchase costs incomparison with hCG and sGnRH in Poland and ease

of preparation (Brzuska 2001; Kucharczyk, bowski,yuczynŁski, Kujawa, Mamcarz,Wyszomirska,Szabo¤ & Ratajski 2001; Brzuska & Białow ˛as 2002;Kowalski, Hliwa, Andronowska, Kro¤l, Dietrich,Wojtc-zak, StabinŁski & Ciereszko 2006; Krejsze¡ et al 2008).Dosages of Ovopel and the results of controlled re-production of females (ovulation, latency time andsurvival of the eyed-egg stage) and males (sperma-tion, milt volume and motility of spermatozoa) be-longing to genus Leuciscus were described byKucharczyk et al (2008) Based on long-time studies,

Szczer-it can be concluded that Ovopel also yields better sults in the controlled reproduction of ide males innatural and out of spawning seasons in comparisonwith CPH and hCG (Kucharczyk et al 1999; Kucharc-zyk 2002) Among GnRH analogue preparations,Ovaprim (a salmon analogue GnRH, D-Arg6Pro9NEt-sGnRH with a dopamine receptor antagonistdomperidone) also yields good results in ide males’maturity stimulations (Kucharczyk et al 2007; Jam-ro¤z et al 2008) However, because of the higher coste¡ectiveness of Ovopel, we decided to use this pre-paration in the experiments presented

re-The type of hormonal preparation, the dose plied and the time between the stimulation per-formed and obtaining the milt signi¢cantlydetermine the in£uence on the e¡ects of reproduc-tion and depend on the species that is subject to thetreatment Spermatozoa of carp Cyprinus carpio L arepresent in the testicles all year round (Kołdras, Bien-iarz & Kime 1990) while in the case of trout Oncorch-ynhus mykiss (Walbaum), they are found during thereproductive season, which occurs in the spring orautumn only (Billard 1986) Also, maturation con-

ap-trolled in both species by similar hormones di¡ers

signi¢cantly in its dynamics In salmonids, the

dis-charge of gonadotropins occurs gradually over

sev-eral days before attaining maturity (Munkittrick &

Moccia 1987), but in case of carp it takes just hours

(Billard, Cosson, Perchec & Linhart 1995) Thus,

hor-monal stimulation of cyprinids can be limited to

sev-eral hours, while in case of salmonids it takes up to

several weeks The temperature of water that holds

the spawners in£uences the dynamics of maturation

Probably, as compared with carp, other cyprinids, i.e

reophileous ¢sh spawning in colder water need

slightly more time from carp stimulation until full

spawning readiness

Considering the poor literature resources on

reo-phileous ¢sh, and in particular the issues related to

the biology of their milt, the goal of this study was to

determine the in£uence of the time after hormonal

stimulation on selected quality parameters of milt of

ide Additionally, we attempted to determine the

mo-lecular weights of seminal plasma protein pro¢les for

that species

Materials and methods

Origin and transport of ¢sh

Male ides originated from the Knieja Fish Farm

situ-ated near Cz˛estochowa and they belonged to a

3-year-old breeding stock In April 2008, the ¢sh

caught from the ponds were transported to the

hatchery of the Department of Lake and River

Fish-ery of the University of Warmia and Mazury in

Olsz-tyn, where they were placed in tanks with water at

10 1C All males were divided into ¢ve experimental

groups, i.e group I ^ control (n 515) and group II

(n 512), from which milt was collected after 36 h;

group III (n 511), from which milt was collected after

60 h; group IV (n 512), from which milt was collected

after 84 h; and group V (n 512), from which milt was

collected after 108 h from hormonal stimulation

After 3 days of adaptation, the water temperature in

the tanks was increased to 12 1C and maintained at a

constant level until the end of the experiments

Hormonal stimulation and manipulations

with spawners

Hormonal stimulation was performed with an

intra-peritoneal injection of a single dose of Ovopel

(Unic-trade, Hungary) at 1granule kg 1

(one granule

con-tains 18^20mgD-Ala6Pro9Net-mGnRH and 8^10 mgmetoklopramide (HorvaŁth et al 1997) After 36 h fromthe commencement of the experiment, males ofgroups I and II were examined and milt was collectedusing sterile syringes by delicately massaging the ab-dominal parts and taking care to avoid contamination

of samples with urea, faeces or blood After 60 h (groupIII) of conducting the stimulation, another group ofmales was obtained from the tanks and milt was ob-tained as described above Further samples were ob-tained after 84 h (group IV) and 108 h (group V) ofadministration of Ovopel to the males The control(group I) consisted of ¢sh that were not treated withany hormonal preparations Before milt collection, in-dividuals of each group were weighed and then thequantity of milt obtained from them was measured.Milt was collected after anaesthesia was administered2-phenoxyethanol (Sigma, St Louis, MO, USA) at0.5 mL L 1water The milt obtained was transported

on ice (14 1C) to the Department of Gamete and bryo Biology, Institute of Animal Reproduction andFood Research of the Polish Academy of Sciences inOlsztyn, where further analyses were carried out

Em-Determination of the basic parameters of milt

Motility of ide spermatozoa was determined by a jective method using a light microscope under

sub- 400 magni¢cation Spermatozoa were activated

by mixing 1mL of milt with 30 mL of activation tion that contained 120 mM NaCl (Sigma) Motilitywas determined immediately after activation by oneobserver and the value of motile spermatozoa wassubjectively estimated in per cent (%) The concentra-tion of spermatozoa was determined using the spec-trophotometric method (Ciereszko & D˛abrowski1993) by diluting the milt with 0.7% NaCl at 1:1000.Absorption was measured on a Beckman DU-640spectrophotometer (Analytical Instruments, LLS,Golden Valley, MN, USA) ata 5 530 nm Absorptionmeasurement results were then applied to the stan-dard curve formula prepared earlier for ide using theBˇrker chamber (cytometric method) and the con-centration values were calculated (109mL 1)

solu-Determination of total protein content andseminal plasma osmotic pressure

Seminal plasma was obtained by the centrifugation

of milt batch (10 000 g) for 10 min and next thesupernatant obtained was transferred into test tubes

Hormonal stimulation of ide males B I Cejko et al Aquaculture Research, 2010, 41, 804^813

and stored at 80 1C until analyses The total

pro-tein content in seminal plasma (mg mL 1) was

deter-mined using the method presented by Lowry,

Rosenbrough, Farr and Randall (1951), and the value

of osmotic pressure (mOsm kg 1) was measured

using a vapor pressure osmometer 5520 (WESCOR,

Logan, UT, USA)

Separation of proteins using denaturing

electrophoresis SDS-PAGE

Electrophoresis was conducted in 12.5%

polyacryla-mide gels using a horizontal electrophoresis unit

Hoefer SE 250 Mighty Small II (Hoefer-Amersham

Biosciences, Piscataway, NJ, USA) Seminal plasma

samples with a known total protein content were

di-luted with 0.85% NaCl (Sigma) to obtain the overall

concentration of protein in the sample at 30mg Next,

10mL of a twice-concentrated staining agent was

added to each sample, which were then boiled for

5 min at 100 1C Electrophoretic separation was

car-ried out in an electrode bu¡er (0.025 M Tris-HCl,

0.192 M glycine, 0.1% SDS) at pH: 8.3 in 200 V and

40 mA for 1.5 h (for two gels)

Staining of gels and documentation of results

Following electrophoresis, the gels were stained

(0.025% Coomassie brilliant blue, 40% methanol and

7% acetic acid) and discoloured (discolouration agent

I: 40% methanol17% acetic acid and discolouration

agent II: 7% acetic acid15% methanol) The gels

pre-pared were then documented using a photo camera

The molecular weights of protein pro¢les were

esti-mated on the basis of protein standard [aprotinin

(6.4 kDa), lysozyme (20 kDa), trypsin inhibitor (28 kDa),

carbonic anhydrose (34.1kDa), ovalbumin (43.0 kDa),

albumin (67.0 kDa), b-galactosidase (120 kDa) andmyosin (203 kDa), Bio-Rad, Hercules, CA, USA] using

insilico.ehu.es/mini_tools/molecular_weight/)

Statistical analysis

The results obtained were characterized using thearithmetic average and standard deviation ( SD).The signi¢cance of the di¡erences between thegroups of ¢sh examined for the analysed characteris-tic was veri¢ed using the non-parametric Kruskal^Wallis test (one-wayANOVA) using theGRAPHPAD PRISM

4 software (GraphPad Software Inc., USA)

Percentage

of motility spermatozoa (%)

Concentration of spermatozoa (109mL 1)

Total protein content (mg mL 1)

Osmotic pressure (mOsm kg 1)

Fish group

Volume

of milt (mL kg 1b.w.)

Total number of spermatozoa

in milt (109)

Number of spermatozoa

in milt (109kg 1b.w.)

ob-tained after 60 h; IV, milt obob-tained after 84 h; V, milt obob-tained after 108 h after hormonal stimulation with Ovopel.

mental group The largest milt volume was obtained

from ¢sh of group IV (0.70 0.55 mL), 84 h after

hor-monal stimulation It was also the largest volume of

milt (3.03 1.94 mL) per kilogram of body weight

of males (Tables 1 and 2) The lowest milt volume was

obtained from males of group I, i.e control

(0.17 0.07 mL) Also in this case the volume of miltper kilogram of body weight was the lowest(0.78 0.36 mL) (Tables1and 2) A statistically signif-icant di¡erence was found in volumes of samples ob-tained between group I and group III (Po0.01) as well

as between groups I and IV (Po0.05) (Fig 1a).When

Figure 1 In£uence of the length of time after hormonal stimulation on selected parameters of milt of ide Leuciscus idus

L analyses group (a) Volume of milt (mL); (b) percentage of motility spermatozoa (%); (c) concentration of spermatozoa

percentile; middle box line: median; upper box line: 75th percentile; upper whisker: maximum) Boxes labelled with ent superscripts are statistically di¡erent from each other (Po0.05) Fish group: I, control; II, milt obtained after 36 h; III,milt obtained after 60 h; IV, milt obtained after 84 h; V, milt obtained after 108 h after hormonal stimulation with Ovopel

di¡er-Hormonal stimulation of ide males B I Cejko et al Aquaculture Research, 2010, 41, 804^813

the volume of milt was expressed per kilogram of

body weight, a di¡erence was showed between group

I and groups IV (Po0.01) andV (Po0.01) (Fig.1f)

The percentage of motile spermatozoa after

activa-tion was low in all groups (Table 1) The highest value

was determined in group IV (59%) and the lowest in

group I (23%) In the other groups, spermatozoa

mo-tility was at the level of 46%, 52% and 48%,

respec-tively, in groups II, III and V (Table 1) There were

signi¢cant di¡erences in the motility of spermatozoa

between groups I and II (Po0.01) and between

groups I and III (Po0 001) (Fig.1b) The highest milt

production expressed as a billion of spermatozoa per

kilogram body weight was found 108 h after

treat-ment of the males with Ovopel, i.e in group V

the control group and groups III (Po0.05), IV

(Po0.001) and V (Po0.01) (Fig.1h)

The concentration of spermatozoa in milt of

groups III, IV and V was at a similar level (11.86

3.02; 11.29 3.29 and 11.04  4.00  109mL 1

re-spectively) and slightly higher in group III Lower

values of this parameter were recorded for groups

I (8.75 5.13  109mL 1) and group II (9.05

1.83 109mL 1), i.e in the groups where milt was

collected 36 h after commencement of the

experi-ment (Table 1) No statistically signi¢cant di¡erences

(P40.05) were found between tested groups of ¢sh inthe concentration of spermatozoa in milt (Fig 1c).The total protein content in seminal plasmareached very similar levels in all the groups, although

in groups II, III and IV it was slightly lower(1.84 1.08, 1.88  0.91 and 1.82  0.82 mg mL 1respectively) than in the other groups (group I:1.97 1.10 mg mL 1

and group V: 1.98 0.88 mg

mL 1) (Table 1) Similar to the concentration of matozoa, no statistically signi¢cant (P40.05) di¡er-ences were found between the groups (Fig 1d).Based on the electrophoretic images obtained, it can

sper-be concluded that the protein pro¢les of tested groupsare characterized by an identical distribution of pro-teins in ide seminal plasma The main seminal plas-

ma proteins of that species are characterized bymolecular weights of 52, 26 and 6 kDa, while pro-teins with a molecular weight of 52 kDa clearly dom-inate (Fig 2)

The higher seminal plasma osmotic pressure valuewas determined in group IV (203.19 37.63 mOsm

kg 1) and the lowest in group I (118.3141.13 mOsm kg 1) Seminal plasma osmolality of theother groups was at similar levels (group II: 171.8350.00 mOsm kg 1, group III: 191.20 39.56 mOsm

kg 1and group V: 176.15 35.50 mOsm kg 1, Table1) Di¡erences in seminal plasma osmotic pressurewere found to be statistically signi¢cant betweengroup I and groups II (Po0.01) and III (Po0 001)(Fig 1e)

con-trol; II, milt obtained after 36 h; III, milt obtained after 60 h; IV, milt obtained after 84 h; V, milt obtained after 108 h afterhormonal stimulation with Ovopel Molecular mass of protein standards electrophoresed in gel are given in the column tothe right of the gel and approximate molecular mass of protein detected are given in the column to the left of the gel

In the studies by Kucharczyk et al (1999), hormonal

stimulation of male ide during the reproductive

sea-son using CPE and Ovopel caused a double increase

in the volume of obtained milt as compared with the

control group Signi¢cant di¡erences were also

re-corded in the motility of spermatozoa between the

control group (52%) and groups hormonally

stimu-lated using hCG (77%), CPE (76%) and Ovopel (75%)

Also, in case of chub Leuciscus cephalus L., treatment

with the same substances resulted in increased milt

volume and better biological quality of ejaculate

ob-tained as compared with the non-stimulated control

group (Krejsze¡ et al 2008) Only in the case of dace

Leuciscus leuciscus L did hormonal stimulation not

cause clear changes in the quantity and quality of

ob-tained milt (Kucharczyk 2002)

In our studies, increased milt volume was found in

each of the test groups after treatment with Ovopel,

and it was the highest 84 h after injection (group IV)

(Table 1) It should also be pointed out that 36 h after

injection, the quantity of milt obtained in group II

was already twice as large as that in the control

group This also applies to the quantity of milt

ob-tained per kilogram of body weight of the males

(Table 2) An increase in the volume of milt

posses-sing better spermatozoa motility parameters is also

characteristic for hormonal stimulation in case of

other ¢sh species including bream Abramis brama

(L.) (Kucharczyk, Kujawa,yuczynŁski, Glogowski,

Ba-biak & Wyszomirska 1997), yellow perch Perca

£aves-cens Mich (D˛abrowski, Ciereszko, Ramseyer, Culver

& Kestemont1994) and European perch Perca

£uviati-lis L (Kucharczyk, Kujawa, Mamcarz, Skrzypczak &

Wyszomirska 1998) From the obtained milt volume

[mL kg 1body weight (b.w.)] and number of

sperma-tozoa per kilogram of spawner’s body weight, it can

be concluded that those values increased over time

(Table 2) This indicates that hormonal stimulation

positively in£uences the maturation of milt of that

¢sh species

The percentage of motile spermatozoa in each

group was low and did not exceed 60% (Table 1)

Low motility values (group I: 23%, group II: 46% and

group V: 47% respectively), which correspond to low

seminal plasma osmotic pressure values (group I:

118.31 41.13, group II: 171.83  50.00 mOsm kg 1

and group V: 176.15 35.50 mOsm kg 1

respec-tively), indicate that milt was contaminated with

ur-ine during collection, which occurs frequently under

controlled conditions in many ¢sh species

(Glogows-ki et al 2000; Bokor, Mˇller, Bercse¤nyi, HorvaŁth, baŁnyi & HorvaŁth 2007) It should also be pointed outthat the seminal plasma osmolality determined waslower for reophileous ¢sh than presented in the lit-erature (Hliwa, Kro¤l, Kowalski & Glogowski 2003; Ko-walski et al 2003; Glogowski, Kowalski & Ciereszko2007; Cejko, Kucharczyk, TargonŁska, Kubiak, Saro-siek & Glogowski 2008)

Ur-During the peak of the reproductive season, miltparameters such as spermatozoa concentration inmilt, total protein content, osmotic pressure andseminal plasma ionic composition assume values op-timal for the given species The spermatozoa concen-tration in genus Leuciscus ¢sh is within a relativelywide range and the values determined in our studydi¡er from those presented by Kowalski et al (2003)for chub (15.26 109

mL 1) and by Glogowski et al.(2007) for ide and dace (5.70 and 4.34 109

respectively) Our results are similar to the data sented by Kucharczyk et al (2007), where, after treat-ment of males with Ovopel, the concentration ofspermatozoa in milt of wild ide and ornamental form

pre-of these ¢sh, i.e Leuciscus idus Auber Orfus was 8.4and 9.8 109mL 1respectively This may be linked(in both cases) to an application of the same sperma-tion-inducing procedure (hormonal preparationOvopel, dose: 1granule kg 1b.w.)

Proteins are the main organic component of ostei ¢sh seminal plasma, and their concentration inthe seminal plasma is low and does not exceed

Tele-3 mg kg 1 (Loir, Labbe, Maisse, Pinson, Boulard,Mourot & Chambeyron 1990) From the literaturedata on the content of proteins in reophileous ¢shseminal plasma, it can be concluded that depending

on the ¢sh species they assume the values within arelatively wide range of 1.25 mg mL 1 for ide,2.28 mg mL 1for chub (Kowalski et al 2003), 1.95^3.50 mg mL 1for asp Aspius aspius (L.) (Cejko et al.2008) and 1.24^1.67 mg mL 1for barbel Barbus bar-bus L (B I Cejko, pers comm.) In this study, thesevalues were within a narrow range from 1.82 to1.98 mg mL 1

Spermatozoa of cyprinids are maintained in an active state in the testes and spermatic ducts due to ahigh (300 mOsm kg 1) seminal plasma osmotic pres-sure (Morisawa, Suzuki, Shimizu, Morisawa & Yasu-

in-da 1983; Redondo-Mˇller, Cosson, Cosson & Billard1991; Perchec, Jeulin, Cosson, Andre¤ & Billard 1995).Movement initiation starts in their case when plasmaosmolality decreases below 160 mOsm kg 1, whichoccurs when milt makes contact with water or otheractivating solutions (Poupard, Paxion, Cosson, Jeulin,

Hormonal stimulation of ide males B I Cejko et al Aquaculture Research, 2010, 41, 804^813

Fierville & Billard 1998) During milt collection,

un-controlled (and unwanted) spermatozoa activation

can occur through their contact with urine released

during spermation of spawners (with milt) from the

common urogenital ori¢ce (Rana, Gupta &

McAn-drew 1992) Most probably, such a situation occurred

in the control group where an osmotic pressure of

118 mOsm kg 1was recorded

Signi¢cant di¡erences in spermatozoa motility

emerged between groups I and II and groups I and

III (Fig 1b), i.e the control and groups from which

milt was obtained 36 and 60 h after stimulation

Also, in case of the seminal plasma osmotic pressure,

signi¢cant di¡erences emerged between groups I and

II and groups I and III (Fig.1e) Our observations

indi-cate that both the value of spermatozoa motility and

seminal plasma osmolality following hormonal

sti-mulation increased until 84 h after treatment of ¢sh

with Ovopel and then decreased gradually, reaching

the motility value of ca 48% and osmolality value of

ca 178 mOsm kg 1after 108 h from commencement

of the experiment (Table 1, Fig 1b and e) Changes in

seminal plasma osmolality recorded in the literature

result from milt hydration that occurs during the

peak of the reproductive season and that results from

hormonal stimulation treatment of ¢sh outside the

season (Redondo-Mˇller et al 1991) In our study,

seminal plasma osmolality increased after Ovopel

in-jection Statistically signi¢cant di¡erences in this

parameter were found between the control group

sulted from contamination of milt with urine in

group I or increased secretion of ions into seminal

plasma after hormonal stimulation

On the basis of our studies, it can be concluded that

within 5 days (108 h) of hormonal stimulation

treat-ment, no di¡erences were observed either in the

con-centration of spermatozoa in the milt or in the

content of seminal plasma total proteins Signi¢cant

di¡erences between groups (in terms of these

para-meters) could indicate the presence of uncontrolled

factors or anomalies and they would indicate the

be-ginning of milt ageing that is observed at the end of

the reproductive season In case of European perch,

with time, changes in spermatozoa motility and

seminal plasma total proteins content were observed

at the beginning (April) and the end (May) of the

re-productive season (Kro¤l, Glogowski, Demska-Zak˛esŁ &

Hliwa 2006) Changes were also observed in the

his-tological image of European perch gonads, where the

absence of spermatides in testes in May indicated theend of the spermatogenesis process after the comple-tion of the reproductive season In our studies,changes in milt quality in the obtained sample vo-lume, spermatozoa motility and plasma osmolalityare the e¡ect of hormonal stimulation treatment anddid not result (because of the short duration of thestudy) from ageing changes

The application of hormonal stimulation in trolled ide reproduction appears to be favourable inthe quantity of milt obtained and percentage of mo-tile spermatozoa (Tables 1 and 2, Fig 1a, b and f), notcausing evident changes in other determinants of itsquality (Table 1, Fig 1c and d) The basic parametersdetermining milt quality, i.e motility and concentra-tion of spermatozoa determined in our study (Fig 1band c) indicate that the period of 60^84 h with hor-monal stimulation treatment is optimal for obtaininggood-quality milt from males of that ¢sh species This

con-is, as expected, a period longer than in the case ofcarp, which might be related to a lower temperature

of water in which the ide reproduces A positive ence of the applied hormonal stimulation procedurealso manifested through an evident increase in thenumber of spermatozoa in milt of ¢sh tested overtime due to stimulation with Ovopel (Table 2)

in£u-References

Aas G.H., RefstieT & Gjerde B (1991) Evaluation of milt ity of Atlantic salmon Aquaculture 95, 125^132 Błachuta J (1998) Function of native reophileous cyprinids in river ecosystems Proceedings of Conference reophileous cyprinids, Brwino¤w 1998, SGGW, pp 17^21, (in Polish) Billard R (1986) Spermatogenesis and spermatology of tele- ost ¢sh species Reproduction Nutrition Development 26, 877^920.

qual-Billard R., Cosson J., Perchec G & Linhart O (1995) Biology

of sperm and arti¢cial reproduction in carp Aquaculture

129, 95^112.

Bokor Z., Mˇller T., Bercse¤nyi M., HorvaŁth L., UrbaŁnyi B & HorvaŁth AŁ (2007) Cryopreservation of sperm of two Eur- opean percid species, the pikeperch (Sander lucioperca) and the Volga pikeperch (S volgensis) Acta Biologica Hun- garica 58, 199^207.

Brzuska E (2001) Arti¢cial spawning of European cat¢sh lurus glanis L.: di¡erences between propagation results after stimulation of ovulation with carp pituitary and Ovopel Aquaculture Research 32, 11^19.

Si-Brzuska E & Białow ˛as H (2002) Arti¢cial spawning of carp, Cyprinus carpio (L.) Aquaculture Research 33,753^765 Cejko B.I., Kucharczyk D.,TargonŁska K., Kubiak D., Sarosiek

B & Glogowski J (2008) Quality parameters and selected

biochemical marker of Asp, Aspius aspius (L.), semen

ob-tained after hormonal stimulation with Ovaprim or

Ovo-pel Archives of Polish Fisheries 16, 179^188.

Ciereszko A & D˛abrowski K (1993) Estimation of sperm

con-centration of rainbow trout, white¢sh and yellow perch

using a spectrophotometric technique Aquaculture 109,

367^373.

D˛abrowski K., Ciereszko A., Ramseyer L., Culver D &

Keste-mont P (1994) E¡ect of hormonal treatment on induced

spermation and ovulation in the yellow perch (Perca

£a-vescens) Aquaculture 12, 171^180.

Glogowski J., KwasŁnik M., Piros B., D ˛abrowski K., Goryczko

K., Dobosz S., Ku$minŁski H & Ciereszko A (2000)

Char-acterization of rainbow trout milt collected with a

cathe-ter: semen parameters and cryopreservation success.

Aquaculture Research 31, 289^296.

Glogowski J., Kowalski R.K & Ciereszko A (2007) Biology

and cryopreservation of ¢sh milt In: Biology of Animal

Re-production Biological Conditionings Males

ReproductionVa-lues (ed by J Strzez ’ ek), pp 265^294 University of Warmia

and Mazury, Olsztyn, Poland (in Polish).

Hliwa P., Kro¤l J., Kowalski R.K & Glogowski J (2003)

Bio-chemical quality parameters of vimba Vimba vimba (L.)

milt In: Predators Fish Reproduction, Breeding, Prevention

(ed by Z Zak ˛esŁ, K Demska-Zak˛esŁ, T Krzywosz &

J Wolnicki), pp 193^199 Inland Fisheries Institute,

Olsz-tyn, Poland.

HorvaŁth L., Szabo¤ T & Burke J (1997) Hatchery testing

of GnRH analogue-containing pellets on ovulation in

four cyprinid species Polish Archives of Hydrobiology 44,

221^226.

Jamro¤z M., Kucharczyk D., HakucŁ^Błaz ’owska A., Krejsze¡

S., Kujawa R., Kupren K., Kwiatkowski M., TargonŁska K.,

’arski D., Cejko B.I & Glogowski J (2008) Comparing the

e¡ectiveness of Ovopel, Ovaprim and LH^RH analogue

used in the controlled reproduction of ide, Leuciscus idus

(L.) Archives of Polish Fisheries 16, 363^370.

Kołdras M., Bieniarz K & Kime E (1990) Sperm production

and steroidogenesis in testes of the common carp,

Cypri-nus carpio L., at di¡erent stages of maturation Journal of

Fish Biology 37, 635^645.

Kowalski R.K, Glogowski J., Kucharczyk D., Goryczko K.,

Do-bosz S & Ciereszko A (2003) Proteolytic activity and

elec-trophoretic pro¢le of proteases from seminal plasma of

teleosts Journal of Fish Biology 63, 1008^1019.

Kowalski R.K., Hliwa P., Andronowska A., Kro¤l J., Dietrich

G.J., Wojtczak M., StabinŁski R & Ciereszko A (2006)

Se-men biology and stimulation of milt production in the

European smelt (Osmerus eperlanus L.) Aquaculture 261,

760^770.

Kowalski R.K., Sarosiek B., Kucharczyk D., TargonŁska K &

Glogowski J (2006) Changes of sperm motility ide

(Leucis-cus idus) and chub (Leucis(Leucis-cus leucis(Leucis-cus) parameters

depen-dence to time after activation In: Reproduction, Breeding,

Prevention Cyprinids Fish and Another Species (ed by Z.

Zak ˛esŁ, K Demska-Zak˛esŁ & J Wolnicki), pp 45^51 Inland Fisheries Institute, Olsztyn, Poland (in Polish).

Krejsze¡ S., Kucharczyk D., Kupren K., TargonŁska K., Mamcarz A., Kujawa R., Kaczkowski Z & Ratajski S (2008) Reproduction of chub, Leuciscus cephalus L., under controlled conditions Aquaculture Research 39, 907^912.

Kro¤l J., Glogowski J., Demska-Zak ˛esŁ K & Hliwa P (2006) Quality of semen and histological analysis of testes in European perch Perca £uviatilis L during a spawning per- iod Czech Journal of Animal Science 51, 220^226 Kruger J.C.De W., Smit G.L., Van Vuren J.H.J & Ferreira J.T (1984) Some chemical and physical characteristics of the semen of Cyprinus carpio L and Oreochromis mossambicus (Peters) Journal of Fish Biology 24, 263^272.

Kruk A (2007) Role of habitat degradation in determining

¢sh distribution and abundance along the lowland Warta River, Poland Journal of Applied Ichthyology 23, 9^18 Kucharczyk D (2002) Arti¢cial reproduction and androgenesis

of chosen cyprinid ¢shes Dissertations and monographs

63, p 81 University of Warmia and Mazury, Olsztyn (in Polish with English summary).

Kucharczyk D., Kujawa R., yuczynŁski M., Glogowski J., biak I & Wyszomirska E (1997) Induced spawning in bream, Abramis brama (L.), using carp pituitary extract and hCG Aquaculture Research 28, 139^144.

Ba-Kucharczyk D., Kujawa R., Mamcarz A., Skrzypczak A & Wyszomirska E (1998) Induced spawning in perch, Perca

£uviatilis L using FSH1LH with pimozide or mide Aquaculture Research 29, 131^136.

metoclopra-Kucharczyk D., Kujawa R., Mamcarz A., Wyszomirska E & Ulikowski D (1999) Arti¢cal spawning of ide (Leuciscus idus) dunder controlled conditions Electronic Jour- nal of Polish Agricultural Universities 2, 2 (available at http://uwm.ejpau.media.pl/volume2/issue2/fisheries/abs- 05.html)

Kucharczyk D., Szczerbowski A., yuczynŁski M.J., Kujawa R., Mamcarz A.,Wyszomirska E., Szabo¤ T & Ratajski S (2001) Arti¢cial spawning of Eurasian perch, Perca £uviatilis L using Ovopel Archives of Polish Fisheries 9, 39^49 Kucharczyk D., Borejko A., TargonŁska K., Roz ’ ek W., Chwa- luczyk R., Kowalski R.K & Glogowski J (2007) Impact of Ovaprim on the reproductive results of ide (Leuciscus idus) In: Reproduction, Rearing, Prophylactics of Cyprinid Fish and Another Species (ed by J Wolnicki, Z Zak ˛esŁ & R Ka- minŁski), pp.31^35 Inland Fisheries Institute, Olsztyn, Po- land (in Polish).

Kucharczyk D., TargonŁska K., ’arski D., Kujawa R & carz A (2008) A review of the reproduction biotechnol- ogy for ¢sh from the genus Leuciscus Archives of Polish Fisheries 16, 319^340.

Mam-Kupren K (2005) Thermal conditions for embryonic ment of ¢sh from the genus of Leuciscus Doctoral thesis, University of Warmia and Mazury, Olsztyn, 95pp (in Polish).

develop-Hormonal stimulation of ide males B I Cejko et al Aquaculture Research, 2010, 41, 804^813

Lahnsteiner F., Berger B., HorvaŁth A., Urbanyi B &

Weis-mann T (2000) Cryopreservation of spermatozoa in

cy-prinid ¢shes Theriogenology 54, 1477^1498.

Loir M., Labbe C., Maisse G., Pinson A., Boulard G., Mourot

B & Chambeyron F (1990) Proteins of seminal £uid and

spermatozoa of rainbow trout (Oncorhynchus mykiss):

partial characterization and variations Fish Physiology

and Biochemistry 8, 485^495.

Lowry O.H., Rosenbrough N.J., Farr A.R & Randall K.J (1951)

Protein measurement with the Folin phenol-reagent.

Journal of Biological Chemistry 193, 265^275.

Morisawa M., Suzuki K., Shimizu H., Morisawa S & Yasuda

K (1983) E¡ects of osmolality and potassium on motility

of spermatozoa from freshwater cyprinid ¢shes The

Jour-nal of Experimental Biology 107, 95^103.

Munkittrick K.R & Moccia R.D (1987) Seasonal changes in

the quality of rainbow trout (Salmo gairdneri) semen:

ef-fect of a delay in stripping on spermatocrit, motility,

vo-lume and seminal plasma constituents Aquaculture 64,

147^156.

Perchec G., Jeulin C., Cosson J., Andre¤ F & Billard R.

(1995) Relationship between sperm ATP content and

motility of carp spermatozoa Journal of Cell Science 108,

747^753.

Poupard G.P., Paxion Ch., Cosson J., Jeulin C., Fierville F &

Billard R (1998) Initiation of carp spermatozoa motility

and early ATP reduction after milt contamination by

ur-ine Aquaculture 160, 317^328.

Rana K.J., Gupta S.D & McAndrew B.J (1992) The relevance

of collection techniques on the quality of manually stripped Atlantic Salmon (Salmo salar) milt.Workshop on Gamete and Embryo Storage and Cryopreservation in Aquatic Organism International workshop on the sto- rage and cryopreservation of gametes and embryos of aquatic organisms Marly le Roy, France Aquaculture News 14, 4.

Redondo-Mˇller C., Cosson M.P., Cosson J & Billard R (1991)

In vitro maturation of the potential for movement of carp spermatozoa Molecular Reproductive Development 29, 259^270.

Rurangwa E., Kime D.E., Ollevier F & Nash J.P (2004) The measurement of sperm motility and factors a¡ecting sperm quality in cultured ¢sh Aquaculture 234, 1^28 Tadajewska M (2000) Ide In: Freshwater Poland Fish (ed by

M BrylinŁska), pp 314^318 Poland Copyright Science, Warsaw, Poland (in Polish).

TargonŁska K., Kucharczyk D., Mamcarz A., Glogowski J., Krejsze¡ S., PrusinŁska M & Kupren K (2008) In£uence

of individual variability in the percentage of motile matozoa and motility time on the survival of embryos of chosen ¢sh species Polish Journal of Natural Sciences 23, 178^187.

sper-Wojtczak M., Glogowski J., Kołdras M., Kucharczyk D & eszko A (2003) Characterization of protease inhibitors of seminal plasma of cyprinids Aquatic Living Resource 16, 461^465.

Cier-Growth, survival and immune activity of scallops,

suspended and bottom culture in Haizhou Bay, China

Zonghe Yu1,2, Hongsheng Yang1, Baozhong Liu1, Qiang Xu1, Kun Xing1,2& Libin Zhang1,2

1 Institute of Oceanology, Chinese Academy of Sciences, Qingdao, China

2 Chinese Academy of Sciences, Graduate University, Beijing, China

Correspondence: H Yang, Institute of Oceanology, Chinese Academy of Sciences, Qingdao 266071, China E-mail: hshyang@ms qdio.ac.cn

Abstract

We examined the growth, survival and immune

re-sponse of the scallop, Chlamys farreri, during a 1-year

period in deep water of Haizhou Bay Scallops were

cultured using two methods: (1) in lantern nets at a

5 m depth and (2) in a bottom culture system (sleeves)

on the seabed at about a 25 m depth Shell heights,

meat dry weight and immune activities in the

hae-molymph (superoxide dismutase and

myeloperoxi-dase) were measured bimonthly or quarterly from

July 2007 to June 2008 Survival was measured at

the end of the study and environmental parameters

in the experimental layers were monitored during

the experiment The growth and immune activities

of scallops were lower when the water temperature

was high, which was consistent with the main

mor-tality occurring in summer The growth and

immu-nity of scallops were higher in the suspended

culture than in the bottom culture during the

experi-ment, with the exception of shell growth during the

last study period Survival of scallops in the

sus-pended culture (54.6 12.3%) was signi¢cantly

low-er than that in the bottom culture (86.8 3.5%) at

the end of this study We conclude from our results

that the high mortality of C farreri can be prevented

by culturing them in a bottom culture system before

November of the ¢rst year, and then transferring

them to a suspended culture to improve scallop

pro-duction

Keywords: Chlamys farreri, aquaculture, seasonal

variation, water temperature, dissolved oxygen

(DO), immune activities

Introduction

The scallop, Chlamys farreri (Jones et Presten), is one

of the main aquatic animals cultured in the coastalwaters of northern China The culture operationsexpanded considerably from the 1980s after the suc-cessful development of new methods of spat collec-tion, with the total production increasing 42-foldbetween 1986 and 1996 (Zhang & Yang 1999a; Guo,Ford & Zhang 1999) After 1996, mass mortality hasoccurred in the summer in most culture areas andhas drastically a¡ected the production of this species(Xiao, Ford, Yang, Zhang, Zhang & Guo 2005) Suchmortality outbreaks have also been reported for otherbivalves like clams, mussels and oysters (Ho & Zheng1994; Tremblay, Myrand, Sevigny, Blier & Guderley1998; Tomaru, Kawabata & Nakano 2001; Patrick,Faury & Goulletquer 2006) Scientists hypothesizedthat the main causes of high mortality in C farreriare inbreeding and environmental stresses (Zhang &Yang 1999a; Wang, Li, Qiu & Zhang 2001; Xiao et al.2005; Ma, Liu, Mai, Deng & Liufu 2006), but the cause

of mass mortality has not been identi¢ed

Many methods have been suggested to avoid mer mortality and enhance the production of thisscallop, including polyculture, maintaining reason-able stocking densities, improving the germ plasm,maintaining healthy seed-stock farms and enhan-cing marine environmental monitoring (Zhang &Yang 1999b; Wang et al 2001; Ma et al 2006) Zhangand Yang (1999b) proposed extending the farmingareas to depths deeper than 20 m to eliminate sum-mer mortality The C farreri is cultured mainly innear shore areaso15 m in depth, which are now

sum-Aquaculture Research, 2010, 41, 814^827 doi:10.1111/j.1365-2109.2009.02358.x

characterized by high pollution and excessive

stock-ing densities Extendstock-ing the farm areas to deeper

water could provide the cultured scallops with more

constant environmental conditions, lower pollution

and better water exchange than shallow areas In

ad-dition, extending culture can reduce the culture

stress on coastal areas, and, would thus contribute

towards environmental protection of the near-shore

areas

Scallops are reared in both suspended cultures

and bottom cultures (Bergh & Strand 2001) In a

sus-pended culture, the animals are cultured in the water

column within lantern nets, cones, cages and other

containers (Emerson, Grant, Mallet & Carver 1994;

Mendoza, Freites, Lodeiros, Lo¤pez & Himmelman

2003) or using the ear-hanging technique (Cano,

Campos & RomaŁn 2000; Hamada, Yamashita,

Wata-nabe & Natsume 2001; Grant, Emerson, Mallet &

Car-ver 2003) The traditional method of bottom culture

is to release scallops directly on the seabed, which

has the advantage of reducing the costs of equipment

and husbandry compared with a suspended

cul-ture (Frishman, Nooman, Naidu & Cahill 1980);

how-ever, scallops on the bottom might be a¡ected more

by benthic predators, such as crabs and seastars

(Arsenault & Himmelman 1996; Wong & Barbeau

2003) Therefore, various methods were suggested

for a bottom culture, such as corrals, pockets, sleeves,

fences and other equipment (Bergh & Strand 2001;

Freites, Himmelman, Babarro, Lodeiros & Ve¤lez

2001; Mendoza et al 2003) The bivalves in a bottom

culture may su¡er less from bad weather and fouling

organisms than those in a suspended culture;

how-ever, growth and survival may be better in a

sus-pended than that in a bottom culture (Emerson et al

1994; Mendoza et al 2003)

In China, C farreri is cultured mainly in lantern

nets on suspended longlines in coastal areas, and

farmers have been growing scallops at increasing

densities within lantern nets and adding more

long-lines in the culture areas (Xiao et al 2005) There are

few published reports on culturing C farreri in

deep-water areas or on the bottom culture of this species in

shallow water (Wang, Lan,Yang & Zhang 1992)

Bivalves always show seasonal varations in growth

and physiological conditions Many immune

para-meters, such as superoxide dismutase (SOD) and

myeloperoxidase (MPO), are promising biomarkers

to evaluate the health conditions of bivalves

(Holm-blad & S˛derhll 1999; Chen, Yang, Delaporte, Zhao

& Xing 2007; Xing, Lin & Zhan 2008) All

oxygen-utilizing organisms have a mechanism to remove

reactive oxygen species (ROS) that are produced bymetabolism (Santovito, Piccinni, Cassini, Irato &Albergoni 2005) Superoxide dismutase can directlydegrade ROS to defend against and repair oxidativedamage (Demple 1999) and is the most important en-zyme for minimizing oxidative damage to host cells

in the immune defence of animals (Downs, Fauth &Woodley 2001; Campa-Co¤rdova, HernaŁndez-Saave-dra & Ascencio 2002) The SOD activity of bivalvesvaries widely with many factors such as season, re-productive cycle, age, tissue type and environmentalstress (Lau, Wong & Garrigues 2004; Chen, Mai, Ma,Wang, Deng, Liu, Xu, Liufu, Zhang, Tan & Ai 2007).Myeloperoxidase is an intracellular enzyme locatedmainly inside primary lysosomes (Austin & Paynter1995); it is able to catalyse the oxidation of chloride

to hypochlorous acid (HClO), a potent bactericidalagent and is involved in oxidative killing (Holmblad

& S˛derhll 1999) The SOD and MPO activities cane⁄ciently re£ect the immune condition of bivalves.The present study examined the growth and survi-val of C farreri in suspended and bottom cultures in

an open sea area, and also evaluated their health bymeasuring the immune biomarkers, SOD and MPO

Materials and methods

We performed our experiment at the site (3510802400

N, 11915403000E) in the northeast of Haizhou Bay ofthe middle Yellow Sea, 48 km to the east of RizhaoHarbor, Shandong province (Fig 1) The study area is

at the southernmost natural distribution boundary

of C farreri (Guo et al 1999), with about 25 m waterdepth, high-quality water, a £at seabed and muddysand bottom sediment

The juveniles used in this study were obtained on1July 2007 from Yantai, the main culture area of

C farreri in China located on the north side of theShandong peninsula The scallops (n 5 30) had aninitial mean shell height of 2.96 0.43 cm and aninitial mean dry soft tissue weight of 0.17 0.06 g.Scallops were acclimated in the lantern nets at about

5 m depth on the longline in the study area for 1 weekbefore the experiment

Lantern nets and a bottom culture system wereused for the experiment Ten lantern nets (eightlayers per net; each layer was 30 cm in diameter and

20 cm in height) were used in the suspended culture,with 30 scallops per layer (425 ind m 2) The lanternnets were suspended on a longline at about 5 m be-low the sea surface One bottom culture system

(sleeves) used in the experiment, as described in

Mae-da-Mart|¤nez, Ormart, Mendez, Acosta and Sicard

(2000), was 16 m in length and sectioned into 16

compartments (1m 0.2 m height) Four hundred

scallops were placed into each compartment and the

sleeves were then anchored on the seabed about 10 m

beside the loneline at about 25 m depth Both types of

enclosures were made with a 1.5 cm mesh net

Environmental conditions

The environmental conditions were monitored at

5 m depth and at the near-bottom water column

every 2^3 months Water temperature ( 1C), salinity,

dissolved oxygen (DO; mg L 1) and chlorophyll

a (mg L 1) were monitored using a Yellow Springs

Instruments (YSI) 650 (Yellow Springs, OH, USA)

Seawater samples were collected at these two depths

using a 5 L Niskin bottle and then pre¢ltered through

a 200mm mesh to eliminate large particles The total

particulate matter (TPM; mg L 1), particulate

organ-ic matter (POM; mg L 1) and the organic fraction

(fPOM) were determined following the study of

Cran-ford, Armsworthy, Mikkelsen and Milligan (2005), by

¢ltering one litre of the seawater samples through

47 mm Whatman GF/C ¢lters that previously had

been ashed (450 1C for 4 h) and weighed previously

After ¢ltration, the ¢lters were rinsed with 20 mL tilled water and frozen for later analysis The ¢lterswere dried at 60 1C for 48 h and weighed to a con-stant weight The dry weight of the retained material

dis-at 60 1C gave the TPM Then, the ¢lters were ated at 450 1C for 4 h; the weight di¡erence between

inciner-60 and 450 1C was de¢ned as the POM The organicfraction (fPOM) was calculated as POM/TPM

Scallop sampling and laboratoryanalysis

The study was conducted from 9 July 2007 to 2 June

2008 After deployment, the scallops were sampledbimonthly or quarterly Four layers of scallops in asuspended culture were collected randomly fromthe lantern nets, and about 100 scallops were takenarbitrarily from one compartment of the sleeves bySCUBA divers In order to reduce the density e¡ect,the compartments were not sampled repeatedly atdi¡erent times Some of the scallops (about 20 ind)were used for haemolymph sampling, and the re-maining living specimens (about 80 ind) were frozen

at 20 1C for growth measurements All living anddead specimens from three di¡erent layers of lanternnets (about 100 ind) and three di¡erent compart-ments of sleeves (about 1000 ind) were counted toFigure 1 Study site on cultured scallops, Chlamys farreri, during 2007^2008 in Haizhou Bay, theYellow Sea, China

Growth, survival and immune activity of scallops Z Yu et al Aquaculture Research, 2010, 41, 814^827

evaluate survival at the end of this study; the time of

death of the scallops was estimated retrospectively

from the shell heights of dead scallops

Growth was determined by measuring the

changes in shell height and meat dry weight of the

previously frozen scallops after removing the fouling

organisms from the shells; the scallops were steamed

for 5 min and partitioned into shell and soft tissue

The shell height was measured using a vernier

cali-per to the nearest 0.02 mm Soft tissue and shell from

each scallop was dried at 60 1C for 72 h to a constant

weight and weighed to the nearest 0.01g The

condi-tion index (CI) was calculated according to the

meth-od described by Orban, Di Lena, Nevigato, Casini,

Marzetti and Caproni (2002):

CI¼ ðMDW=SDWÞ  1000

where MDW is the meat dry weight (g) and SDW is

the shell dry weight (g)

Haemolymph collection and analysis

The haemolymph was sampled immediately after the

scallops were taken from the seawater The

haemo-lymph was sampled from the anterior adductor

mus-cle using a 1mL plastic syringe with a 25 G needle,

placed in 2.0 mL cryogenic vials, immediately frozen

with liquid nitrogen and stored at  80 1C for later

analysis For each treatment, four samples (three to

¢ve scallops) of haemolymph were collected Samples

were then thawed, centrifuged at 825 g for 10 min at

4 1C using a Sigma 3K15 centrifuge (Deisenhofen,

Germany) and the supernatant was separated for

en-zyme activity analysis

All the enzyme activities were determined

accord-ing to the instruction of the assay kits of Nanjaccord-ing

Jian-cheng, China Superoxide dismutase was measured

with the xanthinoxidase method (Sun, Oberley & Li

1988), and MPO activity was measured using the

O-dianisidine method (Harrison & Schultz 1978).

Measurements of the enzymes in the haemolymph

were expressed as speci¢c activity, i.e U mL 1and

U L 1respectively

Statistical methods

Statistical analysis was conducted using a statistical

package (SPSS11.5 for Windows) Data were reported

as mean standard deviation (SD) Data were

com-pared between the two culture methods at the same

sampling time using a Student t-test; data from the

same treatment at di¡erent sampling times were lysed with one-way analysis of variance (ANOVA)usingTukey’s test The signi¢cance of the relationship be-tween enzyme activities and environmental factorswas evaluated using the Pearson’s correlation test.Data as percentages were transformed (arcsine ofthe square root) before analysis, but presented asnon-transformed percentages Statistical signi¢-cance was set at Po0.05

ana-Results

Environmental conditions

The pH and DO values are presented in Table1 At bothexperimental water depths, the pH values were simi-lar and consistent throughout the study period, ran-ging between 8.0 and 8.3 on di¡erent dates Similarly,salinity was about 30 g L 1throughout the study The

DO levels were all nearly saturated or over-saturated

at di¡erent sampling times, and DO decreased with creasing water temperature and depth

in-The water temperatures at the two depths variedover the experimental period (Fig 2) The bottomwater temperature increased from 19.5 1C at the be-ginning of July 2007 to a maximum of 26 1C at theend of August 2007 and then declined, reaching theminimum of 2.8 1C in mid-February 2008; from thistime point onward, the water temperature increased,reaching 14.5 1C at the end of this study The watertemperatures at the surface were higher than that atthe near-bottom from the beginning of June to thebeginning of August 2007 and also from the end ofMay to the beginning of June 2008; by contrast, thetemperatures at both depths were very similarthroughout the remainder of the year

The chlorophyll a concentrations, which re£ectedphytoplankton biomass, were low throughout the ex-

Table 1 Variations in pH and DO at the experimental depths on di¡erent sampling dates

periment at both depths (Fig 3) There were large

var-iations in chlorophyll a on di¡erent sampling dates;

the values increased with decreasing water

tempera-ture from September onward, and the average

con-centrations between September and December were

higher than in other periods There was a trend of

comparatively higher levels of chlorophyll a at the

surface than at the bottom, however, the di¡erence

was not statistically signi¢cant (t-test, t 5 0.769,

df 5 44, P40.05)

The POM and the fPOM at di¡erent depths were

monitored during each sampling time (Fig 4) The

average concentrations of POM ranged from 1.46 to

2.56 mg L 1, and the POM values at 5 m were lower

than at the near-bottom at the same sampling times

POM had a seasonal trend similar to that of phyll a, with the levels in autumn higher than those

chloro-in other seasons The values of fPOM showed a trendopposite to that of POM, indicating that the seston insurface water had a greater organic component thandid near-bottom seston

In this study, heavy fouling by organisms was served on the lantern nets and the shell of scallops inthe suspended culture; however, there was little bio-fouling in the bottom culture The types of fouling or-ganisms di¡ered over time in the suspended culture,with the main types being bryozoans in summer andautumn, algae in winter and mussels and other smallbivalves in spring and early summer in the secondyear (Fig 5)

ob-0510

Figure 2 Water temperatures at the experimental depths (5 m and near bottom) in Haizhou Bay on sampling dates Forclarity, no standard error bars are shown

00.511.522.53

Jul-08May-08

Jul-07Jun-07

standard error bars are shown

Growth, survival and immune activity of scallops Z Yu et al Aquaculture Research, 2010, 41, 814^827

Figure 6 shows the growth of scallops with the two

culture methods Scallop shells grew slowly in

sum-mer but rapidly in autumn and in the beginning of

winter in both culture methods The shell heights in

the suspended culture increased faster than in the

bottom culture, except during the last 3 months of

the experiment and the shell growth rate of the

bot-tom-cultured scallops (55.5mm day 1) was nearly

four times the value of the suspended culture

(13.8mm day 1) during this period The shell heights

of scallops in the suspended culture were

signi¢-cantly greater than those in the bottom culture at

the November and December sampling (t-test,

t 5 3.364, df 5 24, Po0.01 and t 5  2.238, df 5

34, Po0.05 respectively) The scallops in the

sus-pended and bottom cultures achieved commercial

size (6 cm in shell height) at the end of January 2008

and at the beginning of March 2008 respectively

Soft tissue growth showed a trend similar to that of

the shell height in the two culture methods Meat dry

weights in the suspended culture were all cantly higher than in the bottom culture at eachsampling time (t-test, t 5 2.4933, df 5 20, Po0.05,

Jul-08May-08

Jul-07

0.10.150.20.250.30.350.40.45

00.511.522.533.5

Jul-07Jun-07Figure 4 Particulate organic matter (POM) and organic fraction of total particulate matter ( fPOM) at 5 m and near bot-tom in Haizhou Bay on sampling dates For clarity, usually only one standard error bar is shown

Figure 5 Contrast of biofouling on scallop shells tured in (a) suspended lantern nets and (b) a bottom cul-ture system in Haizhou Bay Photographs were taken on 2June 2008

cul-t 5 6.537, df 5 24, Po0.001, t 5  4.890, df 5 24,

Po0.001, t 5  3.086, df 5 38, Po0.01 and t 5

 3.115, df 5 33, Po0.01respectively)

The CI in both the suspended and the bottom

cul-tures declined during the summer of the ¢rst year

and increased after that until the end of this study

(Fig 7) The CIs in the suspended culture were

signi¢-cantly higher than in the bottom culture at each

sam-pling time after deployment (t-test, t 5 2.124, df 5

20, Po0.05, t 5  6.656, df 5 24, Po0.001, t 5

 2.962, df 5 34, Po0.01, t 5  2.964, df 5 38, Po

0.05,and t 5 2.510, df 5 33, Po0.05 respectively)

The CIs in summer were signi¢cantly lower than inother periods in the suspended culture (one-wayANOVA,

F 5 55.182, df 5 85, Po0.001) The CIs in March andJune 2008 were signi¢cantly higher than those inJuly and September 2007 (one-wayANOVA, F 5 55.182,

df 5 85, Po0.001), November and December 2007(one-wayANOVA, F 5 55.182, df 5 85, Po0.01); the CIs

in July and September 2007, November and December

2007, March and June 2008 were not signi¢cantly ferent (one-wayANOVA, F 5 55.182, df 5 85, P40.05).The CIs of scallops in the bottom culture showed atrend similar to those in the suspended culture

Meat dry weight (S) Meat dry weight (B)

Jul-08May-08

Jul-07Jun-07

Figure 6 Growth of shell height and soft tissue of the scallop, Chlamys farreri, in a suspended culture and a bottom

standard error bar is shown

Figure 7 Variation of the condition index (CI) of the scallop, Chlamys farreri, in a suspended culture and a bottom culture

in Haizhou Bay from 9 July 2007 to 2 June 2008 Di¡erent letters above the bars indicate signi¢cant di¡erences among

error bar is shown

Growth, survival and immune activity of scallops Z Yu et al Aquaculture Research, 2010, 41, 814^827

The mortalities of scallops in the two culture

meth-ods were determined at the end of this study (Fig 8)

The survival of scallops in the suspended culture

(54.6 12.3%) was signi¢cantly lower than that in

the bottom culture (86.8 3.5%) (t-test, t 5  4.332,

df 5 4, Po0.01) The densities of live scallops in the

suspended and bottom culture were 232 and 347

in-d m 2, respectively, at the end of this study Dead

scallops in the suspended culture with shell heights

under 3.58 cm (mean in September) were about

29.3% of the total mortality in lantern nets; about

half of the dead scallops had shell heights between

3.58 and 5.30 cm (mean in November) and thus, the

remainder occupied only about 19.5% of the total

mortality This suggested that scallops in the

sus-pended culture mainly died between July and ber, 2007 In the bottom culture, most of the deadscallops were of a similar size as at deployment Thus,

Novem-we speculate that the major mortalities mainly curred before November of the ¢rst year in both cul-ture methods

oc-Immune activityThe SOD activities in scallop haemolymph are shown

in Fig 9 In the suspended culture, minimum SOD tivities were detected in the summer of the ¢rst year,with the value in September 2007 being signi¢cantlylower than at other times (one-way ANOVA, F 5209.099, df 519, Po0.001); however, the SOD value

ac-in March 2008 was signi¢cantly higher than other

Figure 8 Shell heights of the dead scallops, Chlamys farreri, in a suspended culture and a bottom culture in Haizhou Bay

on 2 June 2008 (n 5 30^80) Figure legends are in the same pattern as to the scallops died during the same period

150

180

6-Mar-083-Dec-07

Figure 9 Variation in the superoxide dismutase (SOD) activity of the scallop, Chlamys farreri, in a suspended culture and

a bottom culture in Haizhou Bay from 7 September 2007 to 2 June 2008 Di¡erent letters above the bars indicate

times (one-way ANOVA, F 5 209.099, df 519, Po

0.001), being about four times greater than the

mini-mum There was no signi¢cant di¡erence between

the values of November and December 2007 A

simi-lar variation in SOD was found for scallops in the

bot-tom culture The SOD activities in the suspended

culture were all higher than that in the bottom

cul-ture at the same sampling times, and the di¡erences

were statistically signi¢cant between the two culture

methods in September 2007, December 2007 and

June 2008 (t-test, t 5 3.371, df 5 6, Po0.05, t 5

2.630, df 5 6, Po0.05 and t 5 3.485, df 5 6, Po0.05

respectively) The SOD activities of scallops in the

sus-pended cultured scallop were strongly correlated

with the water temperature (Pearson’s correlation,

r 5 0.902, N 5 5, Po0.05) and DO (Pearson’s

cor-relation, r 5 0.902, N 5 5, Po0.05); however, there

was a signi¢cant correlation only between SOD

activ-ity of bottom-cultured scallops and DO (Pearson’s

correlation, r 5 0.885, N 5 5, Po0.05)

Myeloperoxidase activities also £uctuated

ob-viously with time (Fig 10) Minimum MPO activities

occurred in June 2008 and the maximum values

were observed in December 2007 in both culture

methods The MPO activities of scallops in the

sus-pended culture were all signi¢cantly higher than

in the bottom culture on the same dates, with the

ex-ception of June 2008 (t-test, t 5 6.041, df 5 6, Po0.05,

t 5 3.164, df 5 6, Po0.05, t 5 3.187, df 5 6, Po0.05,

t 5 2.951, df 5 6, Po0.05 and t 51.964, df 5 6, P4

0.05 respectively) There were no signi¢cant tions between MPO activity and the environmentalparameters

correla-Discussion

Bivalves have always shown di¡erences in growthand survival between a suspended and a bottom cul-ture (Ve¤lez, Freites, Himmelman, Senior & Mar|¤n1995; Lodeiros, Pico, Prieto, NarvaŁez & Guerra 2002;Mendoza et al 2003; Uddin, Park, Kang, Park & Choi2007) The growth of bivalves is a¡ected by many fac-tors, such as water temperature, the quantity andquality of food, culture density and fouling (Broom

& Mason 1978; MacDonald & Thompson 1986; sons & Dadswell 1992; Claereboudt, Bureau, Cote &Himmelman 1994; Lodeiros, Rengel, Freites, Morales

Par-& Himmelman 1998; Pilditch Par-& Grant 1999; Grecian,Parsons, Dabinett & Couturier 2000; Navarro, Leiva,Martinez & Aguilera 2000; Maguire & Burnell 2001)

In the present study, the growth, survival and mune activities of C farreri di¡ered in the two culturemethods The di¡erences were probably due to envir-onmental factors, such as water temperature, DO,quantity and quality of seston and other parametersthat di¡ered in the surface and bottom waters at dif-ferent times; the survival may be in£uenced at pHvalues below 7.8 (Yuan, Chen, Chen, Qu, Guo, Li &Cui 2001) The favoured salinity range for the growth

Suspended cultureBottom culture

6-Mar-083-Dec-07

Figure 10 Variation in the myeloperoxidase (MPO) activity of the scallop, Chlamys farreri, in a suspended culture and abottom culture in Haizhou Bay from 7 September 2007, to 2 June 2008 Di¡erent letters above the bars indicate signi¢cant

Growth, survival and immune activity of scallops Z Yu et al Aquaculture Research, 2010, 41, 814^827

of C farreri is 27^32 g L 1(Liang & Zhang 2008) In

our study, pH and salinity were similar throughout

the water column during the experiment and their

values were suitable for the survival and growth of

C farreri

Growth

The present study indicated that scallops in the two

culture methods all grew slowly during summer but

rapidly during autumn The optimum temperature

for the growth of C farreri is between 14 and 22 1C,

with growth nearly absent when the water

tempera-ture is lower than 5 1C (Lou 1991;Yang, Zhang,Wang,

Wang, He & Zhang1999); the metabolism and growth

of C farreri can also be reduced when the water

tem-perature increases beyond 23 1C (Yang et al 1999;

Zhang & Yang 1999a)

Among the numerous parameters regulating the

growth of bivalves, water temperature and food

avail-ability are probably the most important (Broom &

Mason 1978; MacDonald & Thompson 1986; Pilditch

& Grant 1999; Yang et al 1999; Navarro et al 2000)

The growth rate of C farreri increased with water

temperatures between 5 and 23 1C and POM

be-tween 0.9 and 3.67 mg L 1 (Yang et al 1999) In

the present study, the water temperatures in both

surface and bottom waters were very high during

summer, with values between August and September

all higher than 23 1C Chlorophyll a and POM, which

can re£ect the food concentration available for

bi-valves, were very low during this period; thus, the

scallops grew slowly and the CIs were low By

con-trast, water temperature declined to the optimum

for scallop growth during autumn, and the food

con-centration was higher than in summer, all of which

contributed to the fast growth of scallops during

autumn The food concentration was intermediate

during winter and early spring, and the water

tem-perature ranged from 7 to 14 1C between early

December 2007 and mid-January 2008; also, the

scallops had lower metabolism and were less a¡ected

by fouling organisms during winter Although

growth might be absent between the middle of

Janu-ary and the beginning of March 2008 because of the

low water temperature, the average growth was

rela-tively greater than in summer during the whole

peri-od (December 2007 to March 2008)

The water temperature was higher near the

sur-face than the bottom during summer and the food

concentration was lower on the surface than in the

bottom, and both these factors could inhibit thegrowth of C farreri in lantern nets; however, scallops

in the suspended culture grew more quickly thanthose in the bottom culture In addition, the watertemperature was nearly the same in surface as in bot-tom waters during most of the experimental time.Thus, the growth di¡erence between the two culturemethods cannot be explained by di¡erences in watertemperature

In the present study, the food quantity in POM washigher in the bottom than in the surface and the foodquality (in fPOM) showed an opposite trend Thegrowth of scallops generally showed the same trend

as food qualities rather than quantity Bivalves canadjust their digestive processes when exposed to dif-ferent food quantity and quality (Cranford 1995;Bacon, MacDonald & Ward 1998; Ibarrola, Navarro &Iglesias 1998) Food quality has frequently been cited

as a determining factor in the condition of bivalves.Schneider, Madon, Stoeckel and Sparks (1998) sug-gested that food quality may be a better indicator ofgrowth than food quantity for zebra mussels, andthe scope for growth declined with decreased foodquality The percentages of organic matter were posi-tively correlated with growth of the sea scallop, Pla-copecten magellanicus and high concentrations oflow-quality seston near the bottom inhibited itsgrowth (Emerson et al 1994) Low-quality sestonwas also the main reason for reduced growth in thescallop, Patinopecten yessoensis, cultured in near^bottom water (Silina & Zhukova 2007) For C farreri,fPOM was positively correlated with the growth ofshell height and tissue weight (Zhang, Yang, Wang,

concentra-£uence the growth of scallops

Fouling can negatively in£uence the growth ofcultivated bivalves (Claereboudt et al 1994; Taylor,Southgate & Rose 1997) Fouling organisms on the

lantern nets and scallop shells were heavy

through-out the study, but fouling was less in the bottom

cul-ture system, possibly because accumulated silt

decreased the development of fouling organisms In

our study, the growth rates of scallops were high

when the load of fouling organisms was heavy

dur-ing the ¢rst year; however, durdur-ing the last 3 months

of the experiment, scallops in the bottom culture

grew faster than in the suspended culture, even

though the culture density was lower in the lantern

nets than in the bottom culture system during this

time, which was due to the rapid development of

mussels in the lantern nets (unpubl obs.)

We also suggest that the di¡erences in growth

were due to inherent di¡erences in the culture

meth-ods First, the current £ow near the bottom was

slower than at the surface (Huang, Zhang & Ding

1994), and second, the seabed and the sleeves may

re-duce the exchange of water from the outside to the

inside more for the bottom culture system than for

the lantern nets with their fouling organisms during

most of this study Thus, scallops in the bottom

cul-ture would receive less food than scallops in the

sus-pended culture

Survival

In this study, water temperature was likely to be the

most important factor regulating scallop survival in

a suspended culture Deaths mainly occurred in

summer and were associated with a high water

tem-perature The metabolism of C farreri is a¡ected if the

water temperature exceeds 23 1C (Zhang & Yang

1999a), and the mortality of C farreri in the

sus-pended culture mainly occurred when the water

temperature reached 23^25 1C (Xiao et al 2005)

From the end of July to the beginning of October

2007, water temperatures near the surface were

mainly423 1C, and they £uctuated more than near

the bottom, which corresponded with the greater

mortality of scallops in the suspended culture The

majority of dead scallops in the bottom culture

sys-tem were similar in size to those at deployment,

which suggests that mortality might have resulted

from poor acclimation early in the study

Immune activity

Superoxide dismutase and MPO correlate well with

the immune competence of bivalves and are

appro-priate biomarkers to evaluate environmental e¡ects

on C farreri (Pan, Ren & Liu 2005; Chen, Mai, et al.2007; Chen,Yang, et al 2007; Xing et al 2008) In thepresent study, SOD activity varied with the seasonalchanges in the water temperature and DO The highvalues of SOD activities in the suspended culturewere also consistent with the higher DO near the sur-face This agrees with Santovito et al (2005), whoshowed that Mytilus galloprovincialis increasedSOD activities in months when water temperatureswere the lowest and the solubility of oxygen in-creased, and a study of DO on immune responses

of C farreri also indicated that a lower DO level canreduce the SOD activities of C farreri (Chen, Mai,

et al 2007)

Myeloperoxidase can form a defensive system bined with hydrogen peroxide (H2O2) produced bySOD and halogen (chlorine) (Holmblad & S˛derhll1999); it can function independently or in concertwith lysosomal enzymes (Feng 1988; Adema,Van derKnaap & Sminia 1991) The activation of MPO can in-crease the oxygen consumption when haemocytes ofmollusks are stimulated (OrdaŁs, Novoa & Figueras2000) Myeloperoxidase in C farreri increased in re-sponse to a virus challenge (Xing et al 2008) In thepresent study, the lowest MPO was found in June

com-2008, which indicated the poorest defence capabilityagainst disease agents during this period

Our study indicated that the immunity of C farrerideclined with increasing water temperature and de-creasing DO level, and scallops in summer were in apoorer condition than in other seasons The scallops

in the suspended culture were always in a better siological and immunological condition than those

phy-in the bottom culture This is consistent with the ter growth rates of scallops in lantern nets; however,the mortality was higher in the suspended culturebecause water temperature was higher in the surfacethan at the bottom during summer and autumn

bet-Conclusions and recommendations

The results of the present study suggest that the son and culture method a¡ect the growth and immu-nity of scallop, C farreri Scallops grew slowly andmortality was high in summer, which was consistentwith the poor immune conditions in summer Thegrowth and immunity of scallops were better in thesuspended culture, where food quality was higher,than in the bottom culture; however, survival waslower in the suspended culture, which may be due

sea-to the higher water temperature near the surface

Growth, survival and immune activity of scallops Z Yu et al Aquaculture Research, 2010, 41, 814^827

during summer Fouling was heavy in the suspended

culture, and it may have a¡ected the growth of

scal-lop in lantern nets during the last period of this study

In conclusion, the growth, survival and immunity of

C farreri in di¡erent culture methods are

compli-cated and may be in£uenced by many factors, not all

of which are well understood

We recommend culturing C farreri in bottom

cul-ture systems before November of the ¢rst year, and

then transferring them to lantern nets for higher

growth Scallops should be harvested in early spring

of the second year, which would prevent mass

settle-ment of mussels on the scallop shells and culture

equipment

Acknowledgments

This research was supported by the National Science

& Technology Pillar Program (2006BAD09A02) and

the Hi-tech Research and Development Program of

China (No 2006AA100304/2006AA100307) The

authors thank the sta¡ of the Yellow Sea Aquatic

Company of Lanshan for providing assistance during

this study We also thank Sea Pen Scienti¢c Writing

for help in editing our manuscript

References

Adema C.M., Van der Knaap W.P.W & Sminia T (1991)

Mol-luscan haemocyte-mediated cytotoxicity: the role of

reac-tive oxygen intermediates Reviews in Aquatic Sciences 4,

201^223.

Arsenault D.J & Himmelman J.H (1996) Size-related

changes in vulnerability to predators and spatial refuge

use by juvenile Iceland scallop Chlamys islandica Marine

Ecology Progress Series 140, 115^122.

Austin K.A & Paynter K.T (1995) Characterization of

chemi-luminescence measured in haemocytes of the Eastern

oy-ster, Crassostera virginica Journal of Experimental Zoology

273, 461^471.

Bacon G.S., MacDonald B.A & Ward J.E (1998) Physiological

responses of infaunal (Mya arenaria) and epifaunal

(Placo-pecten magellanicus) bivalves to variations in the

concen-tration and quality of suspended particles I Feeding

activity and selection Journal of Experimental Marine

Biology and Ecology 219, 105^125.

Bergh  & Strand  (2001) Great scallop, Pecten maximus,

research and culture strategies in Norway: a review.

Aquaculture International 9, 305^318.

Broom M.J & Mason J (1978) Growth and spawning in the

pectinid Chlamys opercularis in relation to temperature

and phytoplankton concentration Marine Biology 47, 227^285.

Campa-Co¤rdova A.I., HernaŁndez-Saavedra N.Y & Ascencio

F (2002) Superoxide dismutase as modulator of immune function in American white shrimp (Litopenaeus vanna- mei) Comparative Biochemistry and Physiology C 133, 557^565.

Cano J., Campos M.J & RomaŁn G (2000) Growth and tality of the king scallop grown in suspended culture in Malaga, Southern Spain Aquaculture International 8, 207^225.

mor-Chen J.H., Mai K.S., Ma H.M., Wang X.J., Deng D., Liu X.W.,

Xu W.M., Liufu Z.G., Zhang W.B., Tan B.P & Ai Q.H (2007) E¡ects of dissolved oxygen on survival and im- mune responses of scallop (Chlamys farreri Jones et Pre- ston) Fish and Shell¢sh Immunology 22, 272^281 Chen M.Y., Yang H.S., Delaporte M., Zhao S.J & Xing K (2007) Immune responses of the scallop Chlamys farreri after air exposure to di¡erent temperatures Journal of Experimental Marine Biology and Ecology 345, 52^60 Claereboudt M.R., Bureau D., Cote J & Himmelman J.H (1994) Fouling development and its e¡ect on the growth

of juvenile giant scallops (Placopecten magellanicus) in pended culture Aquaculture 121, 327^342.

sus-Cranford P.J (1995) Relationships between food quantity and quality and absorption e⁄ciency in sea scallops Placopecten magellanicus (Gmelin) Journal of Experimental Marine Biology and Ecology 189, 123^142.

Cranford P.J., Armsworthy S.L., Mikkelsen O.A & Milligan T.G (2005) Food acquisition responses of the suspension- feeding bivalve Placopecten magellanicus to the £oc- culation and settlement of a phytoplankton bloom Journal of Experimental Marine Biology and Ecology 326, 128^143.

Demple B (1999) Radical ideas: genetic responses to tive stress Clinical and Experimental Pharmacology and Physiology 26, 64^68.

oxida-Downs C.A., Fauth J.E & Woodley C.M (2001) Assessing the health of grass shrimp (Palaemonetes pugio) exposes to natural and anthropogenic stressors: a molecular bio- marker system Marine Biotechnology 3, 380^397 Emerson C.W., Grant J., Mallet A & Carver C (1994) Growth and survival of sea scallops Placopecten magellanicus: ef- fects of culture depth Marine Ecology Progress Series

108, 119^132.

Feng S.Y (1988) Cellular defense mechanisms of oysters and mussels American Fisheries Society Special Publication 18, 153^168.

Freites L., Himmelman J.H., Babarro J.M., Lodeiros C.J & Ve¤lez A (2001) Bottom culture of the tropical scallop Lyropecten (Nodipecten) nodosus (L.) in the Golfo de Caria- co,Venezuela Aquaculture International 9, 45^60 Frishman Z., Nooman A., Naidu K.S & Cahill F.M (1980) Farming scallop in Newfoundland, Canada: a cost^bene¢t analysis Third International Pectinid Workshop, Port Erin, Isle of Man, 28pp.

Grant J., Emerson C.W., Mallet A & Carver C (2003) Growth

advantages of ear hanging compared to cage culture for

sea scallops, Placopecten magellanicus Aquaculture 217,

301^323.

Grecian L.A., Parsons G.J., Dabinett P & Couturier C (2000)

In£uence of season, initial size, depth, gear type and

stocking density on the growth rates and recovery of sea

scallop, Placopecten magellanius, on a farm-based nursery.

Aquaculture International 8, 183^206.

Guo X.M., Ford S & Zhang F.S (1999) Molluscan

aquacul-ture in China Journal of Shell¢sh Research 18, 19^31.

Hamada T., Yamashita N., Watanabe T & Natsume S (2001)

Drilling position of the ear a¡ects growth and mortality

of scallop (Patinopecten yessoensis, Jay) in ear-hanging

cul-ture Aquaculture 193, 249^256.

Harrison J.E & Schultz J (1978) Myeloperoxidase:

con¢rma-tion and nature of hemebinding inequivalence

Resolu-tion of a carbonyl-substituted heme Biochimica and

Biophysica Acta 536, 341^349.

Ho J & Zheng G.X (1994) Ostrincola koe (Copepoda,

Myicoli-dae) and mass mortality of cultured hard clam (Meretrix

meretrix) in China Hydrobiologia 284, 169^173.

Holmblad T & S˛derhll K (1999) Cell adhesion molecules

and antioxidative enzymes in a crustacean, possible role

in immunity Aquaculture 172, 111^123.

Huang Q., Zhang Y & Ding J (1994) Hydrographic feature

and development plan of the ¢rst three islands in Yellow

Sea Ocean Technology 13, 30^36 (in Chinese with English

abstract).

Ibarrola I., Navarro E & Iglesias J.I.P (1998) Short-term

adaptation of digestive processes in the cockle

Cerastoder-ma edule exposed to di¡erent food quantity and quality.

Journal of Comparative Physiology B 168, 32^40.

Lau P.S.,Wong H.L & Garrigues P (2004) Seasonal variation

in antioxidative responses and acetylcholinesterase

activ-ity in Perna viridis in eastern oceanic and western

estuar-ine waters of Hong Kong Contestuar-inental Shelf Research 24,

1969^1987.

LiangY.B & Zhang F.S (2008) E¡ect of temperature/salinity

on development of embryos and larvae of scallop Chlamys

farreri Oceanologia et Limnologia Sinica 39, 334^340 (in

Chinese with English abstract).

Lodeiros C., Pico D., Prieto A., NarvaŁez N & Guerra A (2002)

Growth and survival of the pearl oyster Pinctada

imbrica-ta (R˛ding 1758) in supended and bottom culture in the

Golfo de Cariaco,Venezuela Aquaculture International 10,

327^338.

Lodeiros C.J.M., Rengel J.J., Freites L., Morales F &

Himmel-man J.H (1998) Growth and survival of the tropical scallop

Lyropecten (Nodipecten) nodosus maintained in suspended

culture at three depths Aquaculture 165, 41^50.

Lou Y.S (1991) Fisheries and aquaculture In: Scallops:

Biol-ogy, Ecology and Aquaculture (ed by S.E Shumway), pp.

809^824 Elsevier, Amsterdam, the Netherlands.

Ma H.M., Liu X.W., Mai K.S., Deng D & Liufu Z.G (2006)

E¡ects of abrupt salinity decline on anti-disease ability of

scallop Chlamys farreri Chinese HighTechnology Letters 16, 746^751 (in Chinese with English abstract).

MacDonald B.A & Thompson R.J (1986) In£uence of perature and food availability on the ecological energetics

tem-of the giant scallop Placopecten magellanicus Marine ogy 93, 37^48.

Biol-Maeda-Mart|¤nez A.N., Ormart P., Mendez L., Acosta B & Sicard M.T (2000) Scallop growout using a new bottom- culture system Aquaculture 189,73^84.

Maguire J.A & Burnell G.M (2001) The e¡ect of stocking density in suspended culture on growth and carbohy- drate content of the adductor muscle in two populations

of the scallop (Pecten maximus L.) in Bantry Bay, Ireland Aquaculture 198, 95^108.

Mendoza Y., Freites L., Lodeiros C.J., Lo¤pez J.A & man J.H (2003) Evaluation of biological and economical aspects of the culture of the scallop Lyropecten (Nodipec- ten) nodosus in suspended and bottom culture Aquacul- ture 221, 207^219.

Himmel-Navarro J.M., Leiva G.E., Martinez G & Aguilera C (2000) Interactive e¡ects of diet and temperature on the scope for growth of the scallop Argopecten purpuratus during reproductive conditioning Journal of Experimental Marine Biology and Ecology 247, 67^83.

Orban E., Di Lena G., Nevigato T., Casini I., Marzetti A & Caproni R (2002) Seasonal changes in meat content, con- dition index and chemical composition of mussels (Mytilus galloprovincialis) cultured in two di¡erent Italian sites Food Chemistry 77, 57^65.

OrdaŁs M.C., Novoa B & Figueras A (2000) Modulation of the chemiluminescence response of Mediterranean mussel (Mytilus galloprovincialis) haemocytes Fish and Shell¢sh Immunology 10, 611^622.

Pan L.Q., Ren J.Y & Liu J (2005) E¡ects of anthene exposure on the biomarkers of scallop Chlamys farreri Comparative Biochemistry and Physiology C 141, 248^256.

benzo(k)£uor-Parsons J.G & Dadswell M.J (1992) E¡ect of stocking density

on growth, production, and survival of the giant scallop, Placopecten magellanicus, held in intermediate culture in Passamaquoddy Bay, New Brunswick Aquaculture 103, 291^309.

Patrick S., Faury N & Goulletquer P (2006) Seasonal changes in carbohydrate metabolism and its relationship with summer mortality of Paci¢c oyster Crassostrea gigas (Thunberg) in Marennes^Ole¤ron bay (France) Aquacul- ture 252, 328^338.

Pilditch C.A & Grant J (1999) E¡ect of temperature tions and food supply on the growth and metabolism of juvenile sea scallops (Placopecten magellanicus) Marine Biology 134, 235^248.

£uctua-Santovito G., Piccinni E., Cassini A., Irato P & Albergoni V (2005) Antioxidant responses of the Mediterranean mus- sel, Mytilus galloprovincialis, to environmental variability

of dissolved oxygen Comparative Biochemistry and siology C 140, 321^329.

Phy-Growth, survival and immune activity of scallops Z Yu et al Aquaculture Research, 2010, 41, 814^827

Schneider D.W., Madon S.P., Stoeckel J.A & Sparks R.E.

(1998) Seston quality controls zebra mussel (Dreissena

polymorpha) energetics in turbid rivers Oecologia 117,

331^341.

Silina A.V & Zhukova N.V (2007) Growth variability and

feeding of scallop Patinopecten yessoensis on di¡erent

bottom sediments: evidence from fatty acid analysis.

Journal of Experimental Marine Biology and Ecology 348,

46^59.

Sun Y., Oberley L.W & Li Y (1988) A simple method for

clin-ical assay of superoxide dismutase Clinclin-ical Chemistry 34,

497^500.

Taylor J.J., Southgate P.C & Rose R.A (1997) Fouling animals

and their e¡ect on the growth of silver-lip pearl oysters,

Pinctada maxima (Jameson) in suspended culture

Aqua-culture 153, 31^40.

Tomaru Y., Kawabata Z & Nakano S (2001) Mass mortality

of the Japanese pearl oyster Pinctada fucata martensii in

relation to water temperature, chlorophyll a and

phyto-plankton composition Diseases of Aquatic Organisms 44,

61^68.

Tremblay R., Myrand B., Sevigny J., Blier P & Guderley H.

(1998) Bioenergetic and genetic parameters in relation to

susceptibility of blue mussels, Mytilus edulis (L.) to

sum-mer mortality Journal of Experimental Marine Biology and

Ecology 221, 27^58.

Uddin M.J., Park K.I., Kang D.H., Park Y.J & Choi K.S (2007)

Comparative reproductive biology of Yezo scallop,

Patino-pecten yessoensis, under two di¡erent culture systems on

the east coast of Korea Aquaculture 265, 139^147.

Ve¤lez A., Freites L., Himmelman J.H., Senior W & Mar|¤n N.

(1995) Growth of the tropical scallop, Euvola (Pecten)

ziczac, in bottom and suspended culture in the Golfo de

Cariaco,Venezuela Aquaculture 136, 257^276.

Wang Y.L., Lan X.L., Yang X.Y & Zhang D.H (1992) Enlarge

the experiment on proliferation of the scallop by sowing

the sea bottom with its seeds Shandong Fish 2, 3^6 (in

Chinese with English abstract).

WangY.L., Li M.Z., Qiu Z.X & ZhangY (2001) On the cause of

large scale death of the cultured scallop and its

counter-measure Transactions of Oceanology and Limnology 3, 63^69 (in Chinese with English abstract).

Wong M.C & Barbeau M.A (2003) E¡ects of substrate on interactions between juvenile sea scallops (Placopecten magellanicus Gmelin) and predatory sea stars (Asterias vulgaris Verrill) and rock crabs (Cancer irroratus Say) Jour- nal of Experimental Marine Biology and Ecology 287, 155^ 178.

Xiao J., Ford S.E., Yang H.S., Zhang G.F., Zhang F.S & Guo X.M (2005) Studies on mass summer mortality of cul- tured Zhikong scallops (Chlamys farreri Jones et Preston)

in China Aquaculture 250, 602^615.

Xing J., Lin T.T & Zhan W.B (2008) Variations of enzyme tivities in the haemocytes of scallop Chlamys farreri after infection with the acute virus necrobiotic virus (AVNV) Fish and Shell¢sh Immunology 25, 847^852.

ac-Yang H.S., Zhang T., Wang J., Wang P., He Y.C & Zhang F.S (1999) Growth characteristics of Chlamys farreri and its relation with environmental factors in intensive raft- culture areas of Sishiliwan bay,Yantai Journal of Shell¢sh Research 18,71^76.

Yuan Y.X., Chen J.F., Chen B.J., Qu K.M., Guo F., Li Q.F & Cui

Y (2001) Study on adaptability of scallop Chlamys farreri

to environmentçe¡ect of salinity and pH on survival, spiration, ingestion and digestion Journal of Fishery Sciences of China 7, 73^77 (in Chinese with English ab- stract).

re-Zhang F.S & Yang H.S (1999a) Analysis of the causes of mass mortality of farming Chlamys farreri in summer in coastal areas of ShanDong, China Marine Science 1, 44^47 (in Chinese with English abstract).

Zhang F.S & Yang H.S (1999b) Strategic and counter sures to resolve mass mortality problems of Chlamys far- reri Marine Science 2, 38^42 (in Chinese with English abstract).

mea-Zhang T., Yang H.S., Wang P., He Y.C & mea-Zhang F.S (2001) The factors a¡ecting meat condition and growth rate of Farreri’s scallop Chlamys (Azumapecten) farreri in Sishili- wan Bay, Yantai Marine Fisheries Research 22, 25^31 (in Chinese with English abstract).

Impact of aquaculture on mangrove areas in the

northern Pernambuco Coast (Brazil) using remote

sensing and geographic information system

Ariana Silva Guimaraes1, Paulo Travassos1, Pedro Wal¢r Martins E Souza Filho2, Fabr|¤cio DiasGoncalves2& Francisco Costa2

1 Laborato¤rio de Ecologia Marinha, Departamento de Pesca e Aqˇicultura, Universidade Federal Rural de Pernambuco, CEP 52171-900, Recife, Pernambuco, Brazil

2

Laborato¤rio de AnaŁlise de Imagens do Tro¤pico UŁmido, Faculdade de Oceanogra¢a, Instituto de GeocieŒncias, Universidade Federal do ParaŁ, CEP 66075-110, Bele¤m, ParaŁ, Brazil

Correspondence: A S Guimaraes, Laborato¤rio de Ecologia Marinha, Departamento de Pesca e Aqˇicultura, Universidade Federal Rural

de Pernambuco, CEP 52171-900, Recife, Pernambuco, Brazil E-mail: guimaraesariana@yahoo.com.br

Abstract

The conversion of mangrove areas into shrimp

farm-ing ponds has been indicated as the main activity

re-sponsible for the reduction in the area of this

ecosystem along the northeastern coast of Brazil

The present study was conducted using remote

sen-sing methods and a geographical information system

with the aim of quantifying the participation of this

activity in the reduction of the mangrove areas along

the northern coast of the State of Pernambuco

(northeast, Brazil), where shrimp farming has been

implanted in last recent years From 1973 to 2005,

there was reduction of about 2.052 ha of mangrove,

197 ha of which were converted into shrimp ponds

Thus, the real contribution of shrimp farming to this

reduction was just 9.6% of the total area Other

anthropogenic activities, such as agriculture, urban

expansion and tourism, contributed greatly to the

re-duction in the mangrove areas along the northern

coast of the State of Pernambuco

Keywords: aquaculture, mangrove, remote

sen-sing, GIS, Brazil

Introduction

The conversion of mangrove areas into ¢sh farming

nurseries is a long-standing activity in Asian

coun-tries and has been reported for over 500 years

(Die-gues 1990) This practice has been recorded since the

17th century in Brazil, when nurseries originatedfrom excavations in mangroves (Cavalcanti, Santana

& Luna 2004) In recent decades, the practice has come the main threat to mangroves in a large num-ber of countries (Stevenson 1997) and is consideredthe principal factor contributing to the negative im-age of aquaculture (Primavera 2006) It is estimatedthat in Ecuador, about one-third of the 30 000 ha ofshrimp nurseries were constructed in mangroveareas, while in the Indo-Paci¢c region, about

be-12 000 000 ha of forest were cut down for the struction of nurseries (Saenger, Hegerl & Davie1983) In southeast Asia, which possesses 35% of the

con-18 000 000 ha of mangrove forests in the world(Spalding, Blasco & Field 1997), the highest rates ofmangrove losses were recorded in the last 30 years,reaching as much as 70^80% inVietnam and the Phi-lippines (Primavera 2006) Over one-third of themangrove forests worldwide have disappeared in thelast 20 years The human activity that has most con-tributed to this fact is shrimp farming, which is re-sponsible for 35% of this decline (Valiela, Bowen &York 2001)

The Brazilian coastline has the second largestmangrove areas in the world, distributed betweenthe Oiapoque River, on the border French Guyanaand the state of AmapaŁ (041300N), to Sonho Beach

in the state of Santa Catarina (281530S), along proximately 6800 km of coastline (Kjerfve & Lacerda1993) totaling nearly 1300 000 ha of mangroveareas (Spalding et al 1997) Although protected by

ap-Aquaculture Research, 2010, 41, 828^838 doi:10.1111/j.1365-2109.2009.02360.x

legislation, this ecosystem is under permanent threat

from a number of human activities that are developed

both along the coast as well as in the interior of

the country and distributed through the

hydro-graphic basins of the rivers Mangrove areas have

un-dergone signi¢cant changes throughout the entire

Brazilian coastline, especially due to direct and

indirect anthropogenic actions associated to urban

expansion such as the port, tourism and agriculture

industries, especially along the northeastern and

southeastern coasts (Vannucci 2002; Monteiro

2005) Some erosion process can be observed in

cer-tain areas along this coast (Souza Filho, Martins &

Costa 2006)

Among the agricultural activities that have

devel-oped most in recent years, marine shrimp farming

merits special attention The constructed nurseries

in this activity increased from 3548 ha in 1997 to

14 824 ha in 2003 (Rocha, Rodrigues & Amorim

2004), presenting a growth rate of 318% in just 6

years The rapid growth of this activity, especially in

the northeastern Brazilian, associated to a history of

converting mangrove areas into ¢sh farming

nur-series, has led aquaculture to be blamed for the

re-duction in mangrove areas

The conversion of mangrove areas into nurseries

has not yet been quanti¢ed in Brazil This fact has

contributed to intensifying con£icts between

entre-preneurs, environmentalists and traditional

commu-nities Such con£icts are based on the apparent,

potentially negative impact of aquaculture,

espe-cially with regard to the conversion of mangrove

areas of multiple use into a support for an activity

with just one use (Primavera 2006), despite of the

generation of 1.89 direct jobs per hectare of nursery

in production Thus, the aim of the present study

was to determine the actual contribution of

aquacul-ture in the diminishing of mangrove areas along the

northern coast of the State of Pernambuco

(north-east Brazil) over the last 30 years, thereby providing

information to generate a databank that can

signi¢-cantly contribute toward the conservation of these

areas and assist the decision makers in drafting

pre-servation actions and management plans

Material and methods

Study area

The study area comprises the mangroves along the

northern coast of the state of Pernambuco (northeast

Brazil), from the left bank of the Santa Cruz Channel

on the ItamaracaŁ/Igarassu and Itapissuma/Igarassuinter-municipal borders to the right bank of the Goi-ana River on the border with the state of Para|¤ba.There are four estuary areas within the study area:the estuary area of the Goiana and Megao¤ Rivers;the estuary area of the Itapessoca River; the estuaryarea of the Jaguaribe River and the estuary area of theSanta Cruz Channel (Fig 1) Based on the K˛ppenclassi¢cation, this is a predominately Class A region,with a humid, tropical climate and no cold season.The climatology shows that the rains in this regionare at about 1.400 mm year 1, but in some locationsthe amount of rain can reach 2.000 mm or more peryear (LAMEPE 2008) This climate occurs at low lati-tudes and characterizes a region with vegetationadapted to constant high temperatures (annualmean 5 26 1C) and abundant rains (SRH 2001; LA-MEPE 2008)

Digital image processingThe following satellites and sensors were used in thepresent study: LANDSAT-1/MSS, LANDSAT-5/TM,LANDSAT-7/ETM1and CBERS-2/CCD, a China^Bra-zil Earth Resources Satellite The digital image pro-cessing was carried out using thePCI GEOMATICSV 9.1and followed the steps described below:

 Atmospheric correction: to eliminate the spheric e¡ect, the gray levels in the images were ex-amined The values of the atmospheric e¡ects ofthe digital level analysis in each band of each im-age were then subtracted using the ARICONSTfunction (Constant Arithmetic)

atmo- Enhancement: adaptive enhancement was applied

to all images through manipulation of the gram Enhancements were then applied to the dif-ferent bands using the Edit LUT function available

histo-on the program

 Geometric correction: geometric correction wasused by polynomial models, employing GroundControl Points GCP obtained from a LANDSAT-7/ETM1image (orbit/point: 214/65, with passingdata from August 4, 2001) available from the Glo-bal Land Cover Facility (GLCF 2006), previouslyorthorecti¢ed in order to minimize the errors ofthe geometric correction The Ortho Engine mod-ule, with a root mean square of 0.46 pixels, wasused for this procedure

 Supervised classi¢cation: in the supervised cation, samples were obtained from 13 classes(ocean water, river water, salt marsh, urban area,

classi¢-mud bank, mangrove, dense forest, sparse forest,

cloud, plantation, exposed soil, cloud shadow and

nursery) During the classi¢cation process these

classes were grouped in seven new classes, as

fol-lows: mangrove, water, urban area, nurseries

(aquaculture), salt marshes and coastal plateau

During the same process the following algorithms

were tested: Maximum Likelihood Classi¢cation,

Parallelepiped Classi¢cation e Minimun Distance

Classi¢cation In the mangrove class, on mangrove

vegetation was considered, whereas the ocean,

riv-ers and estuaries were considered in the water

class As the images were obtained under di¡erent

tide conditions, the mud £ats exposed at low tide

were classi¢ed as water so as to maintain the same

pattern of images obtained at high tide Only cities

of an expressive size and near mangrove areaswere considered for the urban area class Theshoreline was added to this class due to the similar-ity of the spectral signature and because it did notinterfere in the analysis of the present study Nur-series were grouped into one class The marsh classencompassed the brackish transitional area be-tween the mangrove ecosystem and the terrestrialecosystem The seaboard class included all remain-ing areas that did not ¢t in the other speci¢cclasses, such as plantations, rainforest and smallurban developments situated on land

 Filter: Three types of low-frenquency ¢lters (3  3pixels) were tested: Average Filter, Median Filterand Mode Filter These ¢lters evaluate a particu-lar input pixel brightness value, and the pixelsFigure 1 Location of study area along the northern coast of the State of Pernambuco, Brazil

Aquaculture and mangrove in Pernambuco (Brazil) A S Guimaraes et al Aquaculture Research, 2010, 41, 828^838

surrounding the input pixel, and outputs a new

brightness value, that is the mean, median or mode

of this convolution (Jensen 1996)

GIS analisys

maps and spatial data manipulation The raster

images classi¢ed and ¢ltered by thePCI GEOMATICSV

9.1 were initially converted into polygons, generating

a shape ¢le recognized by theARCGISV 9.0 The

poly-gons were classi¢ed as mangrove, water, urban area,

nursery, marsh and seaboard and later manually

edi-ted by direct visual classi¢cation, which in some

cases relied on the use of the Social Environmental

Diagnostic Map of the Northern Coast of

Pernambu-co regarding the use and occupation of land (CPRH

2003) The area of each polygon was then calculated

and added to its respective class, thereby obtaining

the area in hectares for each class

For the analysis of the temporal evolution, the

layers referring to the years 1973 and 2005 were

in-serted into the same ¢le, then proceeding to the

un-ion of equal classes The temporal analysis assessed

the evolution of the mangrove areas and nurseriesusing Boolean analysis, attributing a value of 1 tothe presence of the class in the respective year and azero value for its absence By cross-referencing thisinformation, it was possible to quantify the areas ofincrease (0 in 1973 and 1 in 2005), reduction (1 in

1973 and 0 in 2005) and those that remained stable(1 in 1973 and 1 in 2005) It was therefore possible toassess the conversion of mangrove areas into nur-series by cross-referencing the information on theareas that were mangroves in 1973 and nurseries in

2005 Along with the temporal evolution map (1973/2005), thematic maps were generated for each yearstudied (1973, 1988, 1999, 2001 and 2005), allowingthe temporal conversions of mangrove areas

Results and discussion

Thematic maps quantifying the areas occupied bymangrove vegetation and aquaculture nurserieswere generated on a 1:100.000 scale for each yearstudied (Figs 2 to 6 and Table 1), enabling the tempor-

al evolution analysis and the determination of theactual contribution of aquaculture on the reduction

Figure 2 (a) Image MSS/LANDSAT-1 of year 1973 and composition R3G2B1 (b) Mapping of mangrove and aquacultureponds in the northern coast of Pernambuco (Brazil) in 1973