1. Trang chủ
  2. » Luận Văn - Báo Cáo

Aquaculture research, tập 42, số 2, 2011

142 399 0

Đang tải... (xem toàn văn)

Tài liệu hạn chế xem trước, để xem đầy đủ mời bạn chọn Tải xuống

THÔNG TIN TÀI LIỆU

Thông tin cơ bản

Định dạng
Số trang 142
Dung lượng 9,27 MB

Các công cụ chuyển đổi và chỉnh sửa cho tài liệu này

Nội dung

The results showed that the matology parameters, packed cell volume, haemoglo-bin, erythrocyte sedimentation rate, red blood cell,white blood cell, total serum protein, Mg21, Ca21, hae-C

Trang 2

Low-density culture of red abalone juveniles, Haliotis

system and flow-through system

Miroslava Vivanco-Aranda1, Cristian Jorge Gallardo-EscaŁrate2& Miguel AŁngel del R|¤o-Portilla1

1 Laboratorio de Gene¤tica, Departamento de Acuicultura, Centro de Investigacio¤n Cient|¤¢ca y de Ecuacio¤n Superior de Ensenada Ensenada, BC, Me¤xico

2 Departamento de Oceanograf|¤a, Centro de Biotecnolog|¤a, Universidad de Concepcio¤n, Barrio Universitario s/n Casilla 160-C, Concepcio¤n, Chile

Correspondence: M A del R|¤o-Portilla, CICESE, Acuicultura, PO Box 434844, San Diego, CA 92143-4844, USA E-mail: mdelrio@ cicese.mx

Abstract

Commercial abalone culture is carried out using

£ow-through systems with a high water volume

ex-change in Baja California, Mexico The objective of

this work was to compare the growth rate and

survi-val of red abalone cultured in two systems Flow

through (daily water exchange rate of 800%) and

re-circulating systems consisted of a 250 L ¢breglass

tank and constant aeration, but bio¢ltration in the

re-circulating system was provided with a 28 L (1ft3)

bubble-washed bead ¢lter.Water variables were

mea-sured either daily (dissolved oxygen, temperature, pH

and salinity) or three times a week (total ammonia

nitrogen, nitrate-nitrogen, nitrite-nitrogen and

alkali-nity) Shell length was measured every 2 weeks for 18

weeks Only the alkalinity and pH were signi¢cantly

di¡erent due to the addition of sodium bicarbonate to

the recirculating system Abalone growth rate was

26.1 15.96 mm day 1in the recirculating systems

and 22.21 18.69 mm day^1in the £ow-through

sys-tems The ¢nal survival was 78.74% in the

recirculat-ing systems and 71.82% in the £ow-through systems

Signi¢cant di¡erences in the ¢nal size and survival of

the abalones were found between systems (Po0.05)

Therefore, recirculating aquaculture systems is a

fea-sible alternative for juvenile red abalone culture

Keywords: red abalone, Haliotis rufescens,

recircu-lating systems, growth, closed system

Introduction

Abalones, Haliotis, are marine gastropod molluscs,

with a high commercial value There is a high

inter-est in abalone culture for commercial and stock hancement In Baja California, abalone culturestarted in 1984 with the red abalone, Haliotis rufes-cens Swainson 1822 (Mazo¤n-SuaŁstegui, Mucino-D|¤az

en-& Bazu¤a-Sicre 1996), although blue, Haliotis fulgensPhilippi 1845, and yellow, Haliotis corrugata Wood

1828, abalone culture are also of interest The through system is commonly used both commer-cially and for stock enhancement In this system,water is constantly introduced to provide good waterquality to the organisms with an 800% daily waterexchange This practice requires large amounts ofwater to be pumped and, thus, yields a high cost.Recently, the abalone industry has implementedimportant technical changes to reduce costs and toimprove the production procedures, and recirculat-ing systems are considered to be a feasible option toreduce the pumping cost, but without reducing aba-lone growth Commercial recirculating systems havebeen used for nearly three decades (Masser, Rakocy &Losordo 1992) This type of system allows high envir-onmental control and feasible cultivation conditions

£ow-as well £ow-as a reduction of water and land (M£ow-asser et al.1992)

Abalone requires a good water quality to obtainhigh growth rates (Basuyaux & Mathieu 1999) thisinvolves the control and monitoring of di¡erent phy-sical^chemical variables Among the most importantare the temperature, dissolved oxygen, pH, salinity,concentrations of total ammonia nitrogen (TAN), ni-trite-nitrogen (NO2-N), nitrate-nitrogen (NO3-N) andalkalinity In general, control of these variableswill provide good water quality for abalone growthand development (Huchette, Koh & Day 2003) TheAquaculture Research, 2011, 42, 161^168 doi:10.1111/j.1365-2109.2010.02545.x

r 2010 CICESE

Trang 3

objective of this work was to evaluate and compare

the growth rate and survival of red abalone cultured

in a recirculating and in a £ow-through system

Materials and methods

Juvenile abalone

Seven-month-old red abalone (batch 200D, produced

by spawning 33 males and 73 females), were obtained

from the commercial farm ‘Abulones Cultivados S.A

de C.V.’ located in Ejido Ere¤ndira, Baja California,

Me¤x-ico (3111603300N,11612205300W) Juvenile abalones were

separated from substrata using CO2and were

trans-ported on a wet sponge in plastic bags with oxygen at

the Aquaculture Department (Departamento de

Acui-cultura) at the Centre for Scienti¢c Research and

Higher Education in Ensenada (Centro de

Investiga-cio¤n Cient|¤¢ca y de EducaInvestiga-cio¤n Superior de Ensenda,

CICESE) After arriving at our facilities, individual

se-paration was not possible; thus, abalone distribution

was based on the weight of several abalones, which

were then transferred to culture systems In each tank

50 g of abalones were placed, which corresponds to a

low stocking density of 0.070 kg m 2570 g cm 2

(2265 organisms m 2)

Culture systems

One £ow-through system and one recirculating

sys-tem were used in the growth trial Three individual

£ow-through systems, each one with individual

water supply plus three individual recirculating

tems, were used Each one of the recirculating

sys-tems consisted of a cylindrical £at-bottom ¢breglass

tank (250 L capacity) connected to a magdrive 500

pump (Danner Manufacturing, Islandia, NY, USA)

and a 28.31 L (1ft3) bubble wash bead ¢lter

(Aquacul-ture System Technologies, New Orleans, LA, USA) as

a bioclari¢er (Malone & Beecher 2000), with a

back-wash frequency of two to three times a week The

bead ¢lters were acclimated for 60 days at 20 1C

be-fore the experiment as recommended by Malone and

Beecher (2000) In the recirculating systems, the £ow

rate was 30 L min 1 The three individual

£ow-through systems consisted of similar tanks (250 L

ca-pacity) mentioned above A £ow rate of 1.39 L min l

was set in the three tanks for a daily water exchange

rate of 800% The water used was pumped from the

ocean and ¢ltered to 70mm before reaching the

tanks Constant aeration was provided throughout

the experiment to both systems The experiment

lasted 4 months, beginning on 9 October 2003 andconcluding on 13 February 2004 Fifty grams of or-ganisms were randomly allocated to each tank Theinitial mean shell length was 5.88 0.04 mm forabalones maintained in the recirculating system and6.21 0.01mm in the £ow-through system All dataare average standard error Abalones were fed at aratio of 70% of abalone weight per week with Macro-cystis pyrifera according to the alimentation of theabalones in the commercial farms (N Garc|¤a, pers.comm.) Every week, macroalgae were removed andreplaced with fresh ones

Abalone and environmental variablemeasurements

Every 2 weeks, the shell length of 50 randomly lected animals per tank was measured using an elec-tronic vernier calliper (precision to 0.01mm) modelS225 (Fowler Company, Newton, MA, USA) Everyday, temperature and dissolved oxygen were mea-sured using an oxymeter model YSI 55 (YSI, YellowSprings, OH, USA) On 24 November, in both systems,

se-a hese-ater (precision of  0.55 1C) model Pro Heat II

150 W (Won Brother’s, Fredericksburg, VA, USA) wasplaced to maintain the temperature above 16 1C Sali-nity was measured using a temperature-compen-sated refractometer (conventional refractometer)and pH was measured using a Hanna Hi98127 meter(Hanna Instruments,Woonsocket, RI, USA) Alkalinitywas measured two or three times a week with 1.6 N

H2SO4and a bromcresol green-methyl red indicator(Hach Company, Loveland, CO, USA) Because an alka-linity concentration between 80 and 200 mg L 1ofCaCO3is optimal for bacterial survival in the bio¢lters(Loyless & Malone 1997), when alkalinity values de-clined below 100 mg L 1of CaCO3, alkalinity was ad-justed to 150 mg L 1of CaCO3with the addition ofcommercial sodium bicarbonate, Iris quality (Smart

& Final, Tijuana, BCN, Mexico), to the recirculatingsystems according to the method described by Loylessand Malone (1997) The concentrations of TAN, NO2-Nand NO3-N were determined two or three times perweek using a saltwater master liquid test kit (Aqua-rium Pharmaceuticals, Chalfont, PA, USA); for colora-tion of each nutrient, the absorbance was read using aspectrophotometer Shimadzu UV-1201 (ShimadzuScienti¢c Instruments, Columbia, MD, USA)

Statistical analyses

At the beginning of the experiment, abalone shelllengths were compared among tanks with anANOVA

Trang 4

after assumptions of normality and homogeneity of

variances were met At the end of the experiment,

the abalone shell lengths and the survival between

the two culture systems were compared using a

cov-ariance analysis (ANCOVA) to determine the di¡erences

among the slope of the two culture systems against

time Di¡erences between treatment means were

considered to be signi¢cant at Po0.05 Growth rates

[(shell length at a given time shell length at the

time of the previous measurement)/shell length at

the time of the previous measurement 100] were

calculated for every period of measurement Survival

was estimated by counting empty shells at each

mea-surement time For the di¡erent physical^chemical

variables (temperature, dissolved oxygen, pH,

sali-nity,TAN, NO2-N, NO3-N and alkalinity), paired t-tests

were carried out to determine whether there was a

tendency for the data to have a higher value between

culture systems using [recirculating datum]

[£ow-through datum] to calculate the t-value in the t-tests

(paired comparisons, Sokal & Rohlf 1995) Statistical

analyses were carried out usingMINITABforWINDOWS

release 10.2 (Minitab, State Collage, PA, USA)

Results

The initial mean shell length (10 October) was

5.88 0.04 mm for abalones maintained in the

recir-culating system and 6.21 0.01mm in the

£ow-through system (Fig 1) Signi¢cant di¡erences in the

initial abalone size were found between systems

(F(1,298)5 12.25; Po0.01) The ¢nal mean shell length

was 9.17 0.03 mm for organisms in the

recirculat-ing system and 9.01 0.02 mm for organisms in the

£ow-through system (Fig 1) Marginally signi¢cant

di¡erences were found between systems in the

aba-lone growth rate (F(1,2996)5 4.16; P 5 0.04)

There-fore, throughout the experiment, abalones that were

in the recirculating system grew more in comparisonwith the £ow-through system Abalone growthrates ranged from 6.43 to 58.64 (total mean26.1)mm day 1in the recirculating system and from2.3 to 65.87 (total mean 22.21)mm day 1in the £ow-through system (Fig 2) Juvenile survivals (at day126)were 78.74% for organisms in the recirculation sys-tem and 71.82% for organisms in the £ow-throughsystem (Fig 3) No signi¢cant di¡erence was found

in abalone survival (F(1,4)5 3.78, P 5 0.12)

Temperature, salinity and dissolved oxygen did notshow a signi¢cant di¡erence between treatments(Table 1) The values of dissolved oxygen found in thiswork were below the saturation point (72.54% for therecirculating system and 71.62% for the £ow-through system), but never below 5 mg L 1.Although in the £ow-through system, there was

a tendency (paired samples test) to have a highertemperature (n 5128, t 5 13.89, Po0.001), lower

NO2-N concentration (n 5 45, t 5 2.53, P 5 0.015),

Figure 1 Shell length (mm) of juvenile Haliotis rufescens

in two culture systems Mean SE Three £ow-through

system (n 5 50150150 at each time point) Three

recircu-lating systems (n 5 50150150 at each time point)

Figure 2 Growth rate of juvenile Haliotis rufescens intwo culture systems Mean SE Three £ow-through sys-tem (n 5 50150150 at each time point) Three recirculat-ing systems (n 5 50150150 at each time point)

Figure 3 Survival (%) of juvenile red abalones Haliotisrufescens in two culture systems Mean SE Three

£ow-through system (n 5 50150150 at each time point).Three recirculating systems (n 5 50150150 at each timepoint)

Aquaculture Research, 2011, 42, 161^168 Recirculating and £ow through red abalone culture M.Vivanco-Aranda et al.

r 2010 CICESE

Trang 5

lower dissolved oxygen concentration (n 5128,

t 513.51, Po0.001) and higher salinity (n 5128,

t 5 2.45, P 5 0.016)

Alkalinity and pH were signi¢cantly di¡erent

be-tween systems (Table 1) due to the addition of sodium

bicarbonate in the recirculating system, where

alkali-nity and pH were higher (n 5 45, t 511.2, Po0.001;

Abalone growth is in£uenced by environmental

con-ditions, water quality (Leitman 1992; Hoshikawa,

Sa-kai & Kijima 1998; Harris, Maguire, Edwards & Johns

1999), diet type (Viana, Cervantes-Trujano &

Solana-Sansores 1994; Capinpin Jr & Corre 1996; Viana,

Cervantes-Trujano & Solana-Sansores 1996; Haaker,

Parker, Barsky & Chun 1998; Lopez, Tyler & Viana

1998; Bautista-Teruel & Millamena 1999; Capinpin Jr,

Toledo, Encena & Doi 1999), culture density (Day &

Fleming 1992; Mgaya & Mercer 1995; Mgaya, Gosling,

Mercer & Donlon 1995; Clarke & Creese 1998;

Valdes-Urriolagoitia 2000) and abalone size at the beginning

of the experiment (Corazani & Illanes 1998;

Trevel-yan, Mendoza & Buckley 1998; Steinarsson &

Ims-land 2003) Abalone size is a principal factor

a¡ecting the feeding rates of gastropods Generally,

feeding rates per biomass unit are higher in smaller

and faster growing juveniles than in larger abalone

(Marsden & Williams, 1996)

H rufescens fed with arti¢cial diets and di¡erent

types of macroalgae produce similar growth rates

as macroalgae (Table 2) It can be observed that red

abalone growth rate/initial shell length (GR/ISL)

obtained in this study were similar to other GR/ISLvalues Furthermore,Table 3 shows a comparison be-tween studies carried out with arti¢cial diets againstmacroalgae with other abalone species The highestgrowth rates in most of these works were achieved

by feeding abalones with an arti¢cial diet This may

be explained by the fact that both the arti¢cial dietshad higher protein and fat contents and producedthe best growth rates in terms of the total weightand shell length (Capinpin & Corre 1996), which sug-gests that diet type has a direct e¡ect on abalonegrowth rate

In contrast to this study, for other abalone species ofsimilar sizes growth rates of the100mm day 1growthrate of similar abalone size were obtained with aba-lone juveniles of Haliotis tuberculata (Linnaeus, 1758)fed with an arti¢cial diet (Lopez et al 1998) and withJapanese abalone Haliotis discus hannai Ino,1953 (Hos-hikawa et al 1998) fed with diatoms The highestgrowth rates have been reported in juveniles between

10 and 40 mm shell length in other abalone species:Haliotis asinina Linnaeus, 1758 (Capinpin & Corre,1996; Fermin 2002), black abalone Haliotis cracherodiiLeach, 1814 (Leighton & Boolootian 1963), paua aba-lone Haliotis iris Gmelin, 1791 (Clarke & Creese 1998),

H fulgens (Leighton, Byhower, Kelly, Hooker & Morse1981), greenlip abalone Haliotis laevigata Donovan,

1808 (Gilroy & Edwards 1998) and H tuberculata pez et al 1998), all fed with an arti¢cial diet

(Lo-An other aspect that a¡ects the growth rate is theculture density For abalone cultured in £ow-throughsystems, growth is inversely related to density(Mgaya & Mercer1995; Capinpin et al.1999;Valde¤s-Ur-riolagoitia 2000) Culture density has an inverse ef-fect on abalone survival and may a¡ect abalonegrowth directly through competition for foodand space However, in this study, density was not acritical factor for the growth rate because a low

Table 1 Water quality parameters measured in two systems of culture of juvenile red abalones Haliotis rufescens

Water parameters

Recirculating system Flow-through system

Temperature ( 1C) 16.07  0.98 10.40–20.80 16.77  0.88 12.90–20.90 Dissolved oxygen (mg L 1) 7.08  0.51 5.34–8.50 6.89  0.55 5.23–8.28 Salinity (g L 1) 35.35  0.38 35.00–38.00 35.23  0.29 34.00–37.00

pH 8.13  0.04 7.90–8.30 8.01  0.03 7.80–8.20 Alkalinity (mg L  1 ) of CaCO3 139.00  7.99 100.00–165.00 115.11  4.76 95.00–140.00 TAN (mg L  1 ) of CaCO 3 0.11  0.04 0.00–0.36 0.12  0.05 0.01–0.49

NO 2 -N (mg L  1 ) of CaCO 3 0.04  0.05 0.00–0.65 0.01  0.01 0.00–0.05

NO 3 -N (mg L  1 ) of CaCO 3 1.08  1.21 0.02–17.29 0.83  0.33 0.01–4.61 Temperature, dissolved oxygen, salinity and pH (n 5128) Alkalinity, TAN, nitrite-nitrogen and nitrates-nitrogen (n 5 45).

Trang 6

stocking density was used for both culture systems to

eliminate potential complications due to a high

stock-ing rate The low stockstock-ing density used in this study

can explain the highest survival rate obtained

com-pared with the survival rates reported in other

stu-dies (Nie, Ji & Yan 1996, Park, Rho & Song 1995)

Also, the di¡erences may be due to the size of the ganisms and the duration of the experiment There-fore, it is possible to obtain high growth rates of H.rufescens even on feeding with macroalgae if the phy-sical^chemical variables are within the ranges ofgood water quality and low stocking density

or-Table 3 Data of di¡erent relating experimental studies for the growth juvenile of diverse species

Species

Initial shell length (mm)

Growth rate

Haliotis fulgens 10.00 2.71 0.271 Egregia laevigata and

Macrocystis pyrifera

Leighton et al (1981) 25.00 1.74 0.070

42.00 0.71 0.017 Haliotis tuberculata 15.30 1.75 0.114 Palmaria palmata Mgaya and Mercer (1995)

15.20 1.07 0.070 19.60 1.61 0.082 23.80 1.67 0.070 16.80 1.81 0.108 Haliotis asinina 15.80 4.07 0.258 Gracilaria bailinae Bautista-Teruel and Millamena (1999)

15.20 6.67 0.439 Artificial diet 15.80 7.33 0.464 Artificial diet 15.90 7.43 0.467 Artificial diet Haliotis asinina 19.00 4.20 0.221 Gracilaria heteroclada Capinpin et al (1999)

Haliotis rubra 33.91 1.30 0.038 Artificial diet Huchette et al (2003)

34.23 0.90

Present study Haliotis rufescens 5.88 0.78 0.133 Macrocystis pyrifera Recirculating system

6.21 0.67 0.108 Flow through system Initial shell length (ISL) is the average length of the abalone’s shell in the beginning of the experiment and growth rate (GR) is the monthly growth of the abalones shell (evaluated in millimetres).

Table 2 Data from di¡erent studies on the growth juvenile red abalones Haliotis rufescens

Aquaculture Research, 2011, 42, 161^168 Recirculating and £ow through red abalone culture M.Vivanco-Aranda et al.

r 2010 CICESE

Trang 7

Mortality not only depends on the culture density

but also on inadequate handling (stress) of food

transfer (Searcy-Bernal, Salas-Garza &

Flores-Agui-lar 1992) The mortality rate is also associated with

the organism’s size and stage of development

(Ma-zo¤n-SuaŁstegui et al.1996).Water quality variables like

salinity, temperature, dissolved oxygen, pH and

ni-trogen wastes are also important factors governing

the growth of abalones (Harris, Maguire, Edwards &

Hindrum 1998) The results obtained in the present

work for all physical^chemical variables are within

the ranges of good water quality

Special attention should be paid to the pH and

al-kalinity variables, because they were the only two

quality parameters that di¡ered considerably

be-tween the £ow-through and the recirculating system

due to the addition of sodium bicarbonate to the

re-circulating system However, the pH range observed

in the recirculating system was similar to those

shown in other studies with di¡erent abalone species

(Nie et al 1996; Harris, Maguire, Edwards & Hindrum

1997; Harris et al 1998; Basuyaux & Mathieu 1999;

Bautista-Teruel & Millamena 1999) The pH interval

obtained in this study is within the pH values that

promoted the activities of nitrifying bacteria and

pre-vents ammonia toxicity (Loyless & Malone 1997,

Mal-one & Beecher 2000) To our knowledge, there are no

studies on the in£uence of alkalinity on abalone

me-tabolism; however, the alkalinity values in the

recir-culating system (100^165 mg L 1 CaCO3) were

within the ranges recommended for bacterial

survi-val in bio¢lters (Masser et al 1992; Loyless & Malone

1997; Malone & Beecher 2000) Further studies on

the e¡ect of alkalinity on abalone growth could be

carried out; however, it is considered that low

alkali-nity and pH values may have a higher e¡ect on

aba-lone growth than alkalinity values between 100 and

165 L 1CaCO3, because these values were close to

the alkalinity values found in the £ow-through

sys-tem and in nature

In the case of the waste nitrogen compounds, the

TAN, NO2-N and NO3-N concentrations presented in

this study are below the toxic levels reported for

aba-lone (Harris et al 1997; Harris et al 1998) Basuyaux

and Mathieu (1999) found that H laevigata mortality

and growth were a¡ected by a concentration of

1mg TAN L 1; they also found that a concentration

of 1^5 mg NO2-N L 1 does not in£uence abalone

growth, and in contrast, a concentration of

2 mg NO2-N L 1stimulates the growth in H

tubercu-lata On the other hand, they report that toxic levels

are from 8.5 to 15.4 NO2-N L 1 For the NO3-N,

Basuyaux and Mathieu (1999) reported that H culata supports (without a¡ecting growth) a concen-tration range of 100^250 mg of NO3-N L 1 The lowconcentration of the waste nitrogen compounds(TAN, NO2-N and NO3-N) presented in this study wasprobably due to the low density of abalones used andhad no e¡ect on abalone growth

tuber-In the £ow-through system, nearly 15^30% of theproduction costs are associated with the mainte-nance of a high rate of exchange One way to de-crease the cost of production associated withconstantly pumping water through the abalonegrow-out tanks can be with the use of recirculatingsystems (Badillo, Segovia & Searcy-Bernal 2007) For

a recirculating system to be considered a closed tem, it should haveo10% of water exchange per day

sys-of the total volume sys-of the system (Masser et al 1992;Loyless & Malone 1997) In our experiment, the waterexchange was 4.8% of the total volume per day andthe only water replaced was that lost by evaporationand bio¢lter back£ushes; thus, it can be considered as

a closed system

Table 4 shows a cost analysis of the two culturesystems used in this study In the recirculating sys-tem, the high initial costs are due to the equipmentpurchased (bead ¢lter) However, with the saving ofthe rate of water exchange in the recirculating sys-tems, it is possible to recover the initial investment

in a short period of time In this case, the investmentrecovery time is close to 8 months

An economic analysis to the commercial level isnecessary and will be important for the choice of theoptimum stocking density to maximize the produc-

Table 4 Cost analysis of the two systems of culture of nile red abalones Haliotis rufescens

juve-Cost (US$) Recirculating system

Flow-through system

Equipment cost Bubble wash bead filter 571.67 Fibreglass tank 154.30 154.30 PVC pipes 30.86 30.86 Magdrive 500 pump 51.02

Electric energy Pumping cost (per day) 0.09 1.76 Pumping total cost (127 days) 11.43 223.52 Total 819.28 408.68 Difference between systems 410.60 Costs calculated by tank in the experiment time The cost of water pumped from the ocean and the equipment associated was not considered.

Trang 8

tion of H rufescens in recirculation culture systems.

With the present work, it is possible to state that

recir-culating systems are a feasible alternative for the

culture of juvenile red abalone at a low density

Acknowledgments

The CONACYT supported the M.Sc studies of M.V.-A

with a scholarship This study was partially ¢nanced

by means of the CONACYT project ‘Genetic markers

of abalone, Haliotis spp.’ (33018 B) and by CICESE

pro-ject number 655 The authors are particularly

grate-ful to the commercial farm ‘Abulones Cultivados S.A

de C.V.’for providing the abalone used in this research

The authors also thank Marisela Aguilar-JuaŁrez for

the support provided and Oscar B Del R|¤o Zaragoza

for the technical assistance

References

Badillo L., Segovia M & Searcy-Bernal R (2007) E¡ect of two

stocking densities on the growth and mortality of the

pink abalone Haliotis Corrugata in recirculating and

£ow-through systems Journal of Shell¢sh Research 26,

801^807.

Basuyaux O & Mathieu M (1999) Inorganic nitrogen and its

e¡ect on growth of the abalone Haliotis tuberculata

Lin-naeus and the sea urchin Paracentrotus lividus Lamarck.

Aquaculture 174, 95^107.

Bautista-Teruel M.N & Millamena O.M (1999) Diet

develop-ment and evaluation for juvenile abalone, Haliotis asinina:

protein/energy levels Aquaculture 178, 117^126.

Capinpin E.C Jr & Corre K.G (1996) Growth rate of the

Phi-lippine abalone, Haliotis asinina fed an arti¢cial diet and

macroalgae Aquaculture 144, 81^89.

Capinpin E.C Jr, Toledo J.D., Encena V.C II & Doi M (1999)

Density dependent growth of the tropical abalone Haliotis

asinina in cage culture Aquaculture 171, 227^235.

Clarke C.B & Creese R.G (1998) On-growing cultured

aba-lone (Haliotis iris) in northern New Zealand Journal of

Shell¢sh Research 17, 607^613.

Corazani D & Illanes J.E (1998) Growth of juvenile abalone,

Haliotis discus hannai Ino 1953 and Haliotis rufescens

Swainson 1822, fed with di¡erent diets Journal of Shell¢sh

Research 17, 663^666.

Day R.W & Fleming A.E (1992) The determinants and

mea-surement of abalone growth In: Abalone of theWorld

(Biol-ogy, Fisheries and Culture) Proceedings of the 1st

International Symposium on Abalone (ed by S.A Shepherd,

M.J Tegner & S.A GuzmaŁn del Pro¤o), pp.141^168 Fishing

News Books, Oxford, UK.

Fermin A.C (2002) E¡ects of alternative starvation and

re-feeding cycles on food consumption and compensatory

growth of abalone, Haliotis asinina (Linnaeus) ture Research 33, 197^202.

Aquacul-Gilroy A & Edwards S.J (1998) Optimum temperature for growth of Australian abalone: preferred temperature and critical thermal maximum for blacklip abalone, Ha- liotis rubra (Leach), and greenlip abalone, Haliotis laeviga-

ta (Leach) Aquaculture Research 29, 481^485.

Haaker P.L., Parker D.O., Barsky K.C & Chun C.S.Y (1998) Growth of red abalone, Haliotis rufescens (Swainson), at Johnsons Lee, Santa Rosa Island, California Journal of Shell¢sh Research 17, 747^753.

Harris J.O., Maguire G.B., Edwards S.J & Hindrum S.M (1997) E¡ect of nitrite on growth and oxygen consump- tion for juvenile greenlip abalone, Haliotis laevigata Dono- van Journal of Shell¢sh Research 160, 395^401.

Harris J.O., Maguire G.B., Edwards S.J & Hindrum S.M (1998) E¡ect of ammonia on the growth rate and oxygen consumption of juvenile greenlip abalone, Haliotis laevi- gata Donovan Aquaculture 160, 259^272.

Harris J.O., Maguire G.B., Edwards S.J & Johns D.R (1999) Low dissolved oxygen reduces growth rate and oxygen consumption rate of juvenile greenlip abalone, Haliotis laevigata Donovan Aquaculture 174, 265^278.

Hoshikawa H., Sakai Y & Kijima A (1998) Growth teristics of the hybrid between pinto abalone, Hatiotis kamtsckatkana Jonas, and ezo abalone, H discus hannai Ino, under high and low temperature Journal of Shell¢sh Research 17, 673^677.

charac-Huchette S.M.H., Koh C.S & Day R.W (2003) Growth of juvenile blacklip abalone (Haliotis rubra) in aquaculture tanks: e¡ects

of density and ammonia Aquaculture 219, 457^470 Leighton D.L & Boolootian R.A (1963) Diet and growth

in the black abalone Haliotis cracherodii Ecology 44, 227^238.

Leighton D.L., Byhower M.J., Kelly J.C., Hooker G.N & Morse D.E (1981) Acceleration of development and growth in young green abalone (Haliotis fulgens) using warmed ef-

£uent seawater Journal of the World Mariculture Society

12, 170^180.

Leitman A (1992) The e¡ects of gas supersaturation on the behaviour, growth and mortality of red abalone, Haliotis rufescens (Swainson) In: Abalone of the World (Biology, Fisheries and Culture) Proceedings of the 1st International Symposium on Abalone (ed by S.A Shepherd, M.J Tegner

& S.A GuzmaŁn del Pro¤o), pp 75^85 Fishing News Books, Oxford, UK.

Lopez L.M.,Tyler P.A & Viana M.T (1998) The e¡ect of perature and arti¢cial diets on growth rates of juvenile Haliotis tuberculata (Linnaeus, 1758) Journal of Shell¢sh Research 17, 657^662.

tem-Loyless J.C & Malone R.F (1997) A sodium bicarbonate ing methodology for pH management in freshwater-recir- culating aquaculture systems American Fisheries Society

dos-59, 198^205.

Malone R.F & Beecher L.E (2000) Use of £oating bead ¢lters

to recondition recirculating waters in warm water Aquaculture Research, 2011, 42, 161^168 Recirculating and £ow through red abalone culture M.Vivanco-Aranda et al.

aqua-r 2010 CICESE

Trang 9

culture production systems Aquacultural Engineering 22,

57^73.

Marsden I.D & Williams P.M.J (1996) Factors a¡ecting the

grazing rate of the New Zealand abalone Haliotis iris

Mar-tyn Journal of Shell¢sh Research 15, 401^406.

Masser M.P., Rakocy J & Losordo T.M (1992) Recirculating

aquaculture tank production systems Management of

re-circulating systems Louisiana State University

Agricul-tural Center 452, 1–11.

Mazo¤n-SuaŁstegui J.M., Mucino-D|¤az M & Bazu¤a-Sicre L.A.

(1996) Cultivo de abulo¤n Haliotis spp In: Estudio del

Po-tencial Pesquero y Acu|¤cola de Baja California Sur, Vol II

(ed by M Casas-Va´ldez & G Ponce-Dı´az), pp 475–

511 SEMARNAP, Gobierno del Estado de Baja

California Sur, FAO, Instituto Nacional de la Pesca,

UABCS, CIBNOR, CICIMAR, CETMAR Baja

Cali-fornia Sur, Me´xico.

Mgaya Y.D., Gosling E.M., Mercer J.P & Donlon J (1995)

Genetic variation at three polymorphic loci in wild

and hatchery stocks of the abalone Halitios tuberculata

Linnaeus Aquaculture 136,71^80.

Mgaya Y.D & Mercer J.P (1995) The e¡ects of size grading

and stocking density on growth performance of juvenile

abalone, Haliotis tuberculta Linnaeus Aquaculture 136,

297^312.

Nie Z.Q., Ji M.F & Yan J.P (1996) Preliminary studies on

increased survival and accelerated growth of

overwinter-ing juvenile abalone, Haliotis discus hannai Ino

Aquacul-ture 140, 177^186.

Park M.E., Rho S & Song C.B (1995) Density e¡ect on the

growth of juvenile abalones (Haliotis discus hannai) reared

in the closed recirculating water system Bulletin of

Marine Research Institute, Cheju National University 19, 93^102.

Searcy-Bernal R., Salas-Garza A.E & Flores-Aguilar R.A (1992) Research in Me¤xico on the critical stage of abalone (Haliotis spp.) seed production In: Abalone of the World (Biology, Fisheries and Culture) Proceedings of the 1st Inter- national Symposium on Abalone (ed by S.A Shepherd, M.J Tegner & S.A GuzmaŁn del Pro¤o), pp 547^560 Fishing News Books, Oxford, UK.

Sokal R.R & Rohlf F.J (1995) Biometry The Principles and Practice of Statistics in Biological Research State University

of New York at Stony Brook W H Freeman and Company, New York, NY, USA, 887pp.

Steinarsson A & Imsland A.K (2003) Size dependent tion in optimum growth temperature of red abalone (Ha- liotis rufescens) Aquaculture 224, 353^362.

varia-Trevelyan G.A., Mendoza J.L & Buckley B (1998) Increasing the yield of red abalone with the alga, Microcladia coulteri Journal of Shell¢sh Research 17, 631^633.

Valde¤s-Urriolagoitia A.A (2000) Efecto de tres densidades

de cultivo en la sobrevivencia y crecimiento de juveniles

de abulo¤n rojo Haliotis rufescens en un laboratorio comercial Tesis de Licenciatura, UABC Ensenada, Me¤xico, 52pp

Viana M.T., Cervantes-Trujano M & Solana-Sansores R (1994) Attraction and palatability activities in juvenile abalone (Haliotis fulgens): nine ingredients used in arti¢- cial diets Aquaculture 127, 19^28.

Viana M.T., Cervantes-Trujillo M & Solana-Sansores R (1996) The use of silage made from ¢sh and abalone viscera as an ingredient in abalone feed Aquaculture

140, 87^98.

Trang 10

Assay performance during validation of freezing

infected with a Gram-negative bacterium

Julie Bebak1, Craig Shoemaker1, Covadonga Arias2& Phillip Klesius1

1 USDA ARS AAHRU, Auburn, AL, USA

2 Department of Fisheries and Allied Aquaculture, Auburn University, Auburn, AL, USA

Correspondence: J Bebak, USDA ARS AAHRU, 990 Wire Rd., Auburn, AL 36832, USA E-mail: julie.bebak@ars.usda.gov

Abstract

Recovery of bacteria from infected ¢sh during

popula-tion sampling can be a¡ected by factors including the

type of assay, method of specimen preservation and

concentration of bacteria present Consequently, before

use in ¢eld sampling, methods should be validated The

three objectives of this study were, ¢rst, to determine

whether a channel cat¢sh Ictalurus punctatus

(Ra¢n-esque) ¢ngerling classi¢ed as positive for Gram-negative

Edwardsiella ictaluri infection according to bacterial

culture before freezing was also classi¢ed as positive

after freezing, second, to determine how direct culture

from the kidney (DIRECT), culture of homogenate

(HOMOG) and standard PCR (PCR) agree with bacterial

culture in terms of classifying ¢sh as positive or

nega-tive and third, to estimate diagnostic sensitivity (dSe)

and diagnostic speci¢city (dSp) for DIRECT, HOMOG

and PCR In fresh and frozen ¢sh, as bacterial

concen-tration decreased, the ability of each assay to detect

po-sitive ¢sh also decreased, especially when there were

o104colony-forming units per gram (CFU g 1) tissue

HOMOG proved to be the most reliable at correctly

clas-sifying cat¢sh, whether they were subclinically or

clini-cally infected PCR assay was the least reliable Overall,

values for this study population for dSe were 0.66, 0.92

and 0.43, and for dSp were 0.86, 0.91 and 0.95, for

DI-RECT, HOMOG and PCR respectively

Keywords: validation, freezing, Edwardsiella, cat¢sh,

PCR, bacterial culture

Introduction

Studies conducted on ¢eld populations of ¢sh may

require sampling large numbers of individuals

to achieve su⁄cient statistical power For example,

studies that estimate prevalence, incidence or otherepidemiological measures generally require sam-pling hundreds, or even thousands, of ¢sh Assuming100% diagnostic sensitivity (dSe) and speci¢city(dSp), for a prevalence estimate in a population of

100 000 animals, an error of 5%, and con¢dence of95%, the sample size for simple random samplingwill be 138 animals if the expected prevalence is 10%and 384 animals if the expected prevalence is 50%(Thrus¢eld 1995) If the animals sampled are warm-water pond-cultured ¢sh species, then sampling willmost likely occur in an environment with daytimetemperatures that range from an average of about

27 1C to 40 1C, processing ¢sh will proceed slowlyand the distance from laboratory facilities would pre-clude transporting large numbers of dead ¢sh for la-boratory analysis while still maintaining sampleintegrity The transport of live ¢sh back to the labora-tory may be considered, but would likely be con-strained by the need to keep groups of ¢sh separatedand the lack of adequate culture facilities at the desti-nation One possible solution to this problem would

be to place dead ¢sh on dry ice (i.e at least

 78.2 1C), and then transport them to the laboratoryfor storage at  80 1C and processing at a later date.Freezing can result in the inactivation of bacteriaand a loss of infectious units (Sheridan 1997; Archer2004) Suomalainen, Reunanen, Ijas, Valtonen andTiirola (2006) froze enriched Flavobacterium colum-nare from rainbow trout skin mucus at  20 1C for

an unspeci¢ed length of time and then thawed it at

23 1C and found that bacterial DNA could not bedetected with PCR after freezing Electron micro-scopy revealed that frozen then thawed bacterialcells had disintegrated They hypothesized thatbacterial Dnases, lyases and proteases present in

F columnare were responsible for the destruction ofAquaculture Research, 2011, 42, 169^176 doi:10.1111/j.1365-2109.2010.02551.x

r 2010 Blackwell Munksgaard

Trang 11

bacterial cell walls and of the PCR-ampli¢able DNA.

Therefore, before freezing is incorporated as a

sam-pling strategy for pond-based samsam-pling, the method

should be validated

In addition to Suomalainen et al (2006), some

other studies have examined the e¡ect of freezing at

 20 1C and/or at  70 1C on positive or negative

re-sults for ¢sh pathogens isolated from various organs

after injecting ¢sh with a known concentration of

bacteria (Brady & Vinitnantharat 1990; Evans,

Wie-denmayer, Klesius & Shoemaker 2004) However, to

our knowledge, no studies have quanti¢ed the

survi-val of bacteria pre- and post freezing in the same ¢sh

in the same organ Nor have any studies linked the

concentration of bacteria in the organ with the

re-sults of molecular assay

The three main objectives of this laboratory study

were, ¢rst, to determine whether a channel cat¢sh

¢ngerling classi¢ed as positive for Edwardsiella

icta-luri (Enterobacteriaceae, Gram-negative rod)

infec-tion according to bacterial culture before freezing

was also classi¢ed as positive after freezing at

 80 1C for 58^60 days, second, to determine how

di-rect culture from the kidney, culture of homogenate

and standard PCR agree with bacterial culture

[col-ony-forming units per gram (CFU g 1) kidney tissue]

in terms of classifying ¢sh as positive or negative and

third, to estimate dSe and dSp for direct culture,

cul-ture of homogenate and standard PCR

Methods

Speci¢c pathogen-free channel cat¢sh were obtained

from an Alabama farm as sac fry They were kept in

£ow-through dechlorinated fresh water and fed one

to four times per day, depending on age and water

temperature Fish were 120.6 2.1g [mean

weight standard error (SE)] and from 6 to 14

months old when they were used for the study

Before injection with E ictaluri, ¢sh were stocked

individually into 57 L £ow-through tanks and

accli-mated for at least 48 h Water temperature was

29 2 1C.Water £ow rate was about 500 mL min 1

Photoperiod was 12 h light:12 h dark

Edwardsiella ictaluri (AL-93-75), isolated previously

from a naturally occurring outbreak of ESC in

Alaba-ma industry cat¢sh, was the isolate used for the study

(Bader, Shoemaker & Klesius 1998; Shoemaker,

Kle-sius & Bricker 1999) The stock culture was stored in

brain heart infusion (BHI) broth at  80 1C Before

use, the stock culture was thawed, 100mL was added

to 25 mL BHI and grown overnight at 28 1C After22^26 h growth, a 10-fold serial dilution was madeand ¢sh were injected intracoelomically with 100mL

of the dilution needed to infect the ¢sh This dilutionvaried according to whether ¢sh were to be infectedwith low numbers or high numbers of bacteria Therange in numbers of bacteria injected was 102 to

107CFU E ictaluri Numbers were estimated by ing and counting the 10-fold serial dilution in tripli-cate

plat-Fish were injected and processed in groups of 20^

40 animals Fish that died before the sampling daywere placed in 0 1C and processed within 24^48 h.Surviving ¢sh were sampled at 6^9 days post injec-tion to allow the infection to establish itself as clinical

or subclinical, and to allow time to prepare BHI brothand tubes, Edwardsiella isolation medium (EIM)plates and other supplies needed for sampling Onsample day, ¢sh were euthanized by immersion in atleast 250 mg L 1tricaine methanesulphonate (MS-222; Argent Chemical, Redmond,WA, USA) Each ¢shwas weighed, then the anterior and posterior kidneywere exposed A 1mL culture loop was inserted via astabbing technique into the left anterior and then theleft posterior kidney and streaked onto an EIM (selec-tive for E ictaluri; Shotts & Waltman 1990) plate (DI-RECT) These DIRECT samples were collected induplicate on a divided culture plate The left anteriorand left posterior kidney were then removed andplaced in the same preweighed sterile 5 mL tube.The body cavity of the ¢sh was then closed and theremainder of the ¢sh was placed in a plastic bagand frozen at  80 1C The kidney tissue was homo-genized (TissueTearor, Biospec Products, Bartlesville,

OK, USA), and15^25 mg was immediately transferred

to a 1.5 mL microcentrifuge tube for DNA extractionthat same day (PCR) The remaining homogenate wasthen streaked in triplicate onto a divided EIM plate,using a 1mL culture loop each time (HOMOG) Thehomogenate and tube were weighed, and the weightwas recorded Two millilitres of BHI broth was added

to the homogenate and three 10-fold dilution series(i.e in triplicate) were made using 500mL of thehomogenate plus BHI for each dilution series Onehundred microlitres of each of the target dilutionswere plated with a plate spreader onto EIM plates Re-sults were recorded after plates were incubated atroom temperature for 48^72 h, which we had deter-mined previously to be su⁄cient time for colonies to

be visualized, if present For plates that were streaked

in duplicate directly from the kidney or that werestreaked in triplicate from the kidney homogenate,

Trang 12

results were recorded as positive or negative For

plates spread in triplicate from serial dilutions, the

number of bacterial colonies were counted and

recorded and the kidney was classi¢ed as positive or

negative for E ictaluri infection This result was used

as the reference test (gold standard) for comparison

with DIRECT, HOMOG and PCR samples The mean

( SE) analytical sensitivity of plating the dilution

series was 123 ( 23) CFU g 1tissue

The ¢sh were frozen at 80 1C, which was chosen

because it is a typical storage temperature for the

pre-servation of stocks of bacteria To allow for the

amount of time that it may take to process large

num-bers of samples (e.g hundreds), ¢sh were removed

from the freezer after 59^63 days After removal, they

were thawed within about 30 min at room

tempera-ture The right anterior and posterior kidneys were

then assayed as described above

To test the assumption that the left and right sides

of the kidney contain the same log concentration of

bacteria in each ¢sh, 40 ¢sh were injected with

vary-ing concentrations of E ictaluri, then sampled 7^8

days later The left and right kidneys were removed

from each ¢sh, homogenized separately and 2 mL

BHI was added Appropriate dilutions were made

and plated on an EIM plate in triplicate using a plate

spreader Results were recorded after plates were

in-cubated at room temperature for 48^72 h

DNA extraction and ampli¢cation

DNA from the homogenized tissue samples was

extracted and puri¢ed using the animal tissue

spin-column protocol and the Qiagen Dneasy Blood and

Tissue Kit (QIAGEN, Germantown, MD, USA) with

one minor modi¢cation in that after adding Bu¡er

ATL to the sample it was ground with a sterile pestle

Nucleic acid concentration was quanti¢ed using a

nucleic acid spectrophotometer (ND-1000

Spectro-photometer, Nano-drop Technologies, Wilmington,

DE, USA) The ampli¢cation pro¢le for this standard

PCR was the same as used by the Bilodeau,

Waldbie-ser, Terhune, Wise and Wolters (2003) real-time PCR

assay with minor modi¢cations The thermal cycler

(MyCycler Thermal Cycler, Bio-rad Laboratories,

Hercules, CA, USA) was hot-started at 94 1C before

samples were loaded and the ampli¢cation cycle was

started at 94 1C for 5 min instead of 10 min One

posi-tive (75 ngmL 1E ictaluri DNA) and two negative

controls (75 ngmL 1¢sh DNA; master mix with no

template) were included in each batch Each

ampli¢-cation reaction mixture (50mL) contained a DNA

sample (1mL, approximately 75 ng of bacterial and

cat-¢sh genomic DNA); 25mL of 2 Epicentre Bu¡er D;

15 pmol Eict forward primer (ACTTATCGCCCTCGCAACTC) and 15 pmol Eict reverse primer (CCTCTGATAAGTGGTTCTCG) (Bilodeau et al 2003); and 1.0 UTaq polymerase (5 UmL 1) Ampli¢cation was veri¢ed

by electrophoresis of 12mL of the PCR product through1.0% agarose gels Gels were interpreted by an observerwho was blinded to the results of the bacterial culture.For each 1mL template, the analytical sensitivity of thisprocedure for bacterial DNA only was 500 fg to 5 pgand for bacterial DNA plus ¢sh DNA it was 500 fg bac-terial DNA 100 ng ¢sh DNA 1 Consequently, at leastabout100 bacterial genomes must be present for detec-tion by this PCR method

Data analysisThe number of CFU of E ictaluri per gram of kidneytissue was estimated as

CFU g1tissue¼ CFU

inoculumðmLÞ

1dilution

ðg tissue mL1BHIÞwhere CFU is the average colony-forming units perplate, inoculum the volume (ml) of sample inoculatedonto plate, dilution is that which corresponds to platecount, g tissue ml 1BHI the weight of the kidney tis-sue in 2 mL of BHI

For comparison of CFU E ictaluri g 1kidney in theleft and right side of the kidney to test the assumptionthat both sides contain the same log concentration,bacterial numbers were expressed as 10x, where x isthe number of logs for the sample The agreement be-tween the left and right sides of the kidney, based onlog number, was examined by plotting the di¡erence

in log concentration over the average log tion in the left and right side (Altman & Bland 1983;Bland & Altman 1986) To obtain the values betweenwhich 95% of the results should lie, two times thestandard deviation of the di¡erence in the value be-tween the left and right side was estimated and sub-tracted from the mean di¡erence If, for post-freezingsamples, the change in concentration was outsidethis range, then freezing was assumed to have af-fected the concentration of E ictaluri

concentra-Thek statistic was used to estimate the agreementbetween sample sets (Thrus¢eld 1995; StataCorp2005) The interpretation of k values followed thecategorization proposed by Everitt (1989), whereinAquaculture Research, 2011, 42, 169^176 Assay performance during validation of freezing J Bebak et al.

r 2010 Blackwell Munksgaard

Trang 13

k values 40.81 represent an almost perfect agreement,

values from 0.61 to 0.80 represent substantial

agree-ment, values from 0.41 to 0.60 represent moderate

agreement, values from 0.21 to 0.40 represent fair

agreement, values from 0 to 0.20 represent a slight

agreement, and a value of 0 represents poor agreement

Diagnostic sensitivity (proportion of true positives

detected by a test) and dSp (proportion of true

nega-tive detected by a test) (Saah & Hoover 1997) with

95% con¢dence intervals (95% CI) were estimated

for DIRECT, HOMOG and PCR for pre-freezing

com-bined with post-freezing samples The results from

bacterial culture (CFU g 1tissue) of the serial

dilu-tions were used as the reference test to classify ¢sh

as positive or negative

Results

Comparison of right and left kidney

The mean SE body weight of the 40 ¢sh used to

test the assumption that the right and left sides of

the kidney would contain the same log concentration

of E ictaluri was132.7 7.5 g.When bacterial

concen-trations for the left and right side of the kidney were

compared, 24/40 kidneys were within the same log

concentration (i.e 10x, x is the log number of bacteria

present), 15/40 were 1 log di¡erent and one was 2 logs

di¡erent These results indicate that the left and right

kidney contained unequal log concentrations of

bac-teria Therefore, the pre-freezing vs post-freezing

com-parisons that could be made were limited However,

analysis of agreement indicated that if, post freezing,

the right kidney was greater than 101.5CFU g 1tissue

below the pre-freezing results, then the sample could

be classi¢ed as a¡ected by freezing

E¡ect of freezing on log bacterial

concentration in kidney

Thirty of the ¢sh died before the day of sampling

and were placed at 0 1C These ¢sh were processed

within 24^48 h of death Mean weight SE was

120.6 2.1g for the 162 ¢sh sampled pre- and post

freezing Results for the culture of serial dilutions

ranged from 0 to 108CFU E ictaluri g 1 kidney

tissue After freezing for 58^60 days, the proportion

of ¢sh that were still classi¢ed as positive according

to the culture of serial dilutions depended on how

high the log concentration was in the kidney before

Ten out of 57 of the ¢sh with 0 CFU g 1tissue freezing were positive post freezing, that is, they had

pre-an increase in bacterial titre, as did 2/7 of the ¢shthat were 101CFU g 1tissue pre-freezing None ofthe rest of the ¢sh in any other log category had ahigher post-freezing log concentration

Bacterial concentration compared withDIRECT, HOMOG and PCR

Pre- and post-freezing samples were combined for acomparison between bacterial concentration resultsand DIRECT, HOMOG and PCR samples For each as-say type, the per cent correctly classi¢ed as positiveincreased as the log number concentration increased(Table 2) The best performing assay was HOMOG,which correctly classi¢ed490% of the positive sam-ples as positive for all concentrations  101

CFU

E ictaluri g 1tissue The worst performer was PCR,which correctly classi¢ed o65% of the samplesbelow the  105

CFU g 1category When tration was 0 CFU g 1, DIRECT, HOMOG and PCRmisclassi¢ed 14%, 8% and 5% of the samples aspositive respectively

concen-Table 1 Pre-freezing (Pre FC) and post-freezing (Post FC) concentration (CFU E ictaluri g  1 kidney tissue) for indivi- dual channel cat¢sh ¢ngerlings (n 5162)

Number

of fish

Pre FC

Post FC

% Pos

Trang 14

Overall k values for the three assays compared

with the reference test were 0.49 0.05, 0.75  0.05

and 0.34 0.04 for DIRECT, HOMOG and PCR

respectively (test prevalence 5 61%; Table 3) When

kidneys contained o104CFU g 1 tissue (test

pre-valence 5 38%),k was 0.21 0.06, 0.62  0.06 and

0.12 0.05 for DIRECT, HOMOG and PCR

respec-tively When there was  104CFU g 1 tissue

present, most comparison categories were 100% in

agreement (Table 2)

Diagnostic sensitivity varied widely depending on

the assay and whether the samples were pooled or

the cutpoint was set at 104CFU E ictaluri g 1tissue

(Table 4) PCR had the lowest dSe (and the highest

false-negative fractions) for all categories, and was

as low as 0.15 (95% CI 5 0.07, 0.23) for samples with

o104

CFU g 1tissue The highest dSe for PCR was

0.62 for samples  104CFU g 1tissue The HOMOG

assay had consistently the highest dSe values of 0.92

(95% CI 5 0.89, 0.96), 0.80 (95% CI 5 0.71, 0.89)

and 1.00 for all samples combined, o104

and

 104CFU g 1 respectively There was much less

variability in dSp for the three assays The lowest

dSp was 0.86 (95% CI 5 0.80, 0.92) for DIRECT and

the highest was 0.95 (95% CI 5 0.92, 0.95) for PCR

Discussion

Under the conditions evaluated in this study, the

low-er the concentration of E ictaluri in the kidney, the

greater the chance that ¢sh would be misclassi¢ed

post freezing This chance was at least 59% wheno104

CFU g 1tissue was present For studies that tempt to measure population parameters in a cat¢shpopulation that includes subclinically infected ¢sh,freezing may signi¢cantly bias the result towards aninaccurate, low estimate of prevalence There wasalso some likely misclassi¢cation error due to e¡ects

at-of freezing when concentrations were higher, inthe range of 104to 105CFU g 1tissue Edwardsiellaictaluri is a Gram-negative bacterium In general,because of di¡erences in cell wall structure, Gram-positive bacteria are more resistant to freezing thanGram-negative bacteria (Lund, Baird-Parker & Gould2000) Therefore, freezing ¢sh for analysis of a Gram-positive bacterium, such as S iniae, may result in lessloss of bacterial cells

A few ¢sh had an ‘increase’ in bacterial tion while they were in the freezer, even though theywere speci¢c pathogen free and even when the nor-mal variation between left and right sides of the kid-ney was considered Error is inherent in evaluation ofdiagnostic tests and should be allowed to be ex-pressed Therefore, all ¢sh were included in the re-sults This di¡erence may be explained by samplingerror in that the sampling technique did not capturebacteria present at low concentrations while takingthe pre-freezing sample These results also demon-strate the well-known concept that reference tests areoften imperfect, and this one likely is as well (Enoe,Georgiadis & Johnson 2000; Cameron 2002) It is alsopossible that viable bacteria were clustered in macro-phages (Blazer 1991), which could account for this re-sult especially at the very low log concentrations

concentra-If the two sides of the kidney had been the same interms of concentration of E ictaluri, it would be valid

Table 2 Per cent of channel cat¢sh ¢ngerlings with 0^

108CFU g 1kidney tissue that were also positive (% Pos)

for direct culture of Edwardsiella ictaluri from kidney tissue

(DIRECT), culture of homogenate (HOMOG) or PCR (PCR)

Results for samples taken before freezing were combined with

samples taken after freezing.

n, total number of samples included in each group; CFU,

colony-forming unit.

Table 3 k statistics  standard error (SE) for the tion of channel cat¢sh ¢ngerling kidney samples as positive

classi¢ca-or negative by culture (CFU g 1kidney tissue of

Edwardsiel-la ictaluri) compared with direct (DIRECT) culture from the kidney, culture of homogenate (HOMOG) or PCR

j  SE All samples o10 4

CFU g  1 tissue.

Aquaculture Research, 2011, 42, 169^176 Assay performance during validation of freezing J Bebak et al.

r 2010 Blackwell Munksgaard

Trang 15

to compare post-freezing results for DIRECT, HOMOG

and PCR to pre-freezing estimates of E ictaluri

concentration However, as can be seen from Table 2,

a 1 log change in concentration greatly a¡ected the

ability of DIRECT and PCR assays to detect positive

samples Therefore, comparing pre-freezing

con-centration with post-freezing results for DIRECT,

HOMOG and PCR is not valid

Microbiological assay and standard PCR assay

pro-vided less than perfect estimates of infection status

of channel cat¢sh infected with E ictaluri As the

bacterial concentration decreased, the ability of each

assay to detect positive ¢sh also decreased Out of

the three assays tested, only HOMOG consistently

demonstrated an agreement with the results in

CFU g 1tissue Neither DIRECT nor PCR were able

to detect480% of the positive ¢sh until the

concen-tration in the kidney was  104

or 105CFU g 1respectively

Analytical sensitivity is the smallest amount of

substance in a sample that can accurately be

mea-sured by an assay Diagnostic sensitivity is the

pro-portion of individuals who have a given condition

that are identi¢ed as positive for the condition High

analytical sensitivity does not guarantee that dSe

will also be high (Saah & Hoover 1997) This

relation-ship can be observed in the results for both DIRECT

and PCR methods The target bacterium may have

been missing from the sample taken by the culture

loop or from the 15 to 25 mg tissue sample removed

for PCR analysis The dSe of the PCR method used

de-clined abruptly at about 104^105CFU E ictaluri g 1

kidney tissue At these concentrations, even if

bacter-ia were evenly distributed in kidney tissue, there

would be only about 60^125 E ictaluri bacterial cells

present in the 15^25 mg sample used for PCR As

described previously, the analytical sensitivity of theassay was about 100 bacterial genomes Therefore,the results for dSe make sense given the analyticalsensitivity of the PCR assay

DIRECT, HOMOG and PCR also classi¢ed some ¢sh

as positive when, according to culture in CFU g 1sue, they were actually negative As discussed pre-viously, this result could have been due to animperfect reference test and/or sampling error and/

tis-or the presence of bacteria in the phagocytic phages

macro-Thek results corresponded with the other ods used in this study to determine how likely ¢shwould be to be misclassi¢ed.k, the level of agreementbeyond chance for two observations, ranged fromslight to substantial agreement, and was the lowestfor PCR assay and the highest for HOMOG.k esti-mates are a¡ected by the underlying prevalence ofthe trait in question (Sargeant & Martin 1998) Inthis study, the underlying prevalence did not changefor the comparisons being made as long ask valuesfor assays are compared within a cut-o¡ group.The use ofk was appropriate as long as k values for

meth-‘all samples’ are not compared with k values foro104

CFU g 1.Bayesian modelling may also be used to estimatedSe and dSp when the reference test is imperfect orthe true disease state is unknown (Enoe et al 2000;Branscum, Gardner & Johnson 2005) This approachcan provide a useful way to evaluate the diagnostictests in a future work

Sample sizes for the estimation of population meters such as prevalence and incidence should beadjusted to account for the less than perfect dSe anddSp of bacterial culture and PCR Estimates of dSeand dSp can be used to determine how sample sizes

para-Table 4 Estimates of diagnostic sensitivity (dSe) and diagnostic speci¢city (dSp)  95% con¢dence intervals (95% CI) for culture of Edwardsiella ictaluri from serial dilutions of channel cat¢sh kidney tissue compared with direct culture from the kidney (DIRECT), culture from kidney homogenate (HOMOG) and PCR (PCR) results

Concentration in

kidney (CFU g 1) Test prevalence (%) Type dSe 95% CI (lower, upper) dSp 95% CI (lower, upper)

All 61 DIRECT 0.66 0.60, 0.73 0.86 0.80, 0.92

HOMOG 0.92 0.89, 0.96 0.91 0.86, 0.96 PCR 0.43 0.36, 0.50 0.95 0.92, 0.99 o10 4 38 DIRECT 0.32 0.22, 0.43 0.86 0.80, 0.92

HOMOG 0.80 0.71, 0.89 0.91 0.86, 0.96 PCR 0.15 0.07, 0.23 0.95 0.92, 0.99

All samples combined (All), results for o10 4

CFU g 1tissue and results for  10 4

CFU g 1tissue were estimated.

Trang 16

should be adjusted for estimates of prevalence or

in-cidence Additionally, if the dSe and dSp of the test are

known or can be estimated, it is possible to correct for

mistakes, and convert the apparent prevalence to the

true prevalence (Greiner & Gardner 2000; Cameron

2002)

Transferring estimates of dSe and dSp from one

po-pulation to another should be carefully considered

However, the estimates provided by this laboratory

study provide a starting point for sample size

esti-mates for ¢eld studies, if a test prevalence of 61% is

acceptable If the infection status of a population is

unknown, and standard PCR is going to be used as

the assay, then a dSe 5 0.43 (95% CI 5 0.36, 0.50)

and a dSp 5 0.95 (95% CI 5 0.92, 0.99) provide a good

starting point If the population is composed of ¢sh

that are all subclinically infected, then dSe 5 0.15

(95% CI 5 0.07, 0.23) and dSp 5 0.95 (95% CI 5 0.92,

0.99) may be used

If the infection status is to be estimated with the

direct culture of the kidney, then a dSe and dSp of

0.66 (95% CI 5 0.60, 0.73) and 0.86 (95% CI 5 0.80,

0.92), respectively, will be informative for a

popula-tion with unknown status and dSe 5 0.32 (95%

CI 5 0.22, 0.43) and dSp 5 0.86 (95% CI 5 0.80, 0.92)

for a subclinically infected population For assay

using homogenized kidney samples, an overall

dSe 5 0.92 (95% CI 5 0.89, 0.96) and dSp 5 0.91 (95%

CI 5 0.86, 0.96) will provide a starting point For a

subclinically infected population, dSe 5 0.80 (95%

CI 5 0.71, 0.89) and dSp 5 0.91 (95% CI 5 0.86, 0.96)

may be considered

This work demonstrated that, especially at lower

log concentrations, E ictaluri bacterial cell count

may decrease during freezing Brady and

Vinit-nantharat (1990) injected individual cat¢sh

¢nger-lings averaging 4.7 g with 1.8 107

cells of E

ictaluri and found that they could isolate bacteria

from the kidney and liver with an inoculating loop

for no longer than 30 days after freezing at  20 1C

It is possible that bacteria were not viable after 30

days, but it is also possible that as CFU g 1tissue

de-clined in the organs, isolation with an inoculating

loop became a less accurate way of isolating E

icta-luri and the possibility of false negative results

in-creased

As mentioned previously, the results of this study

should be extended to other techniques, ¢sh species

and bacterial species with caution Evans et al (2004)

injected Nile tilapia (Oreochromis niloticus) with

5.5 102

or 4.5 106

CFU Streptococcus agalactiaeper ¢sh, and placed them in the freezer 24 h later

After 180 days at  70 1C, all ¢sh were positive for

S agalactiae infection using kidney culture with asterile swab and overnight culture in enrichmentbroth The concentration in the ¢sh tissue whenfrozen was unknown, but ¢sh injected with 102CFU

S agalactiae probably had a concentration ofo104

CFU g 1tissue when they were frozen The ference in recovery results could have been due to theuse of overnight culture in enrichment broth or thefact that S agalactiae is a Gram-positive bacterium

dif-In conclusion, ¢ngerling cat¢sh kidneys lost a ni¢cant number of E ictaluri cells when they wereplaced in a  80 1C freezer for 60 days Comparedwith direct culture or PCR, culture of homogenizedkidney tissue proved to be the most reliable at cor-rectly classifying ¢sh, whether they were clinically

sig-or sub-clinically infected When estimating ¢shpopulation parameters for E ictaluri infection,sample size estimates should be adjusted to accountfor the error when using these less than perfectassays This study provides some suggested valuesfor dSe and dSp

ReferencesArcher D.L (2004) Freezing: an underutilized food safety technology? International Journal of Food Microbiology 90, 127^138.

Altman D.G & Bland J.M (1983) Measurement in medicine: the analysis of method comparison studies The Statisti- cian 32, 307^317.

Bader J.A., Shoemaker C.A & Klesius P.H (1998) Genomic subtyping of Edwardsiella ictaluri isolated from diseased channel cat¢sh by arbitrarily primed polymerase chain reaction Journal of Aquatic Animal Health 10, 22^27 Aquaculture Research, 2011, 42, 169^176 Assay performance during validation of freezing J Bebak et al.

r 2010 Blackwell Munksgaard

Trang 17

Bilodeau A.L., Waldbieser G.C., Terhune J.S., Wise D.J &

Wolters W.R (2003) A real-time polymerase chain

reaction assay of the bacterium Edwardsiella ictaluri in

channel cat¢sh Journal of Aquatic Animal Health 15,

80^86.

Bland J.M & Altman D.G (1986) Statistical methods for

assessing agreement between two methods of clinical

measurement The Lancet 8, 307^310.

Blazer V.S (1991) Piscine macrophage function and

nutri-tional in£uences: a review Journal of Aquatic Animal

Health 3,77^86.

Brady Y.J & Vinitnantharat S (1990) Viability of bacterial

pathogens in frozen ¢sh Journal of Aquatic Animal Health

2, 149^150.

Branscum A.J., Gardner I.A & Johnson W.O (2005)

Estima-tion of diagnostic-test sensitivity and speci¢city through

Bayesian modeling Preventive Veterinary Medicine 68,

145^163.

Cameron A (2002) Survey Toolbox for Aquatic Animal

Diseases A Practical Manual and Software Package.

ACIAR Monograph No 94, Australian Center for

Inter-national Agricultural Research, Canberra, Australia,

375pp.

Enoe C., Georgiadis M.P & Johnson W.O (2000)

Esti-mation of sensitivity and speci¢city of diagnostic tests

and disease prevalence when the true disease state is

un-known PreventiveVeterinary Medicine 45, 61^81.

Evans J.J.,Wiedenmayer A.A., Klesius P.H & Shoemaker C.A.

(2004) Survival of Streptococcus agalactiae from frozen ¢sh

following natural and experimental infections

Aquacul-ture 233, 15^21.

Everitt R.S (1989) Statistical Methods for Medical

Investiga-tions Oxford University Press/Edward Arnold, New York,

NY, USA/London, UK.

Greiner M & Gardner I.A (2000) Application of diagnostic tests in veterinary epidemiologic studies PreventiveVeter- inary Medicine 45, 43^59.

Lund B.M., Baird-Parker T.C & Gould G.W (2000) The biological Safety and Quality of Food, Vol I Aspen Publish- ers, Gaithersburg, MD, USA.

Micro-Saah A.J & Hoover D.R (1997) ‘‘Sensitivity’’and ‘‘speci¢city’’ reconsidered: the meaning of these terms in analytical and diagnostic settings Annals of Internal Medicine 126, 91^94.

Sargeant J.M & Martin S.W (1998) The dependence of kappa

on attribute prevalence when assessing the repeatability

of questionnaire data Preventive Veterinary Medicine 34, 115^123.

Sheridan J.J (1997) The e¡ect of freezing on the survival of pathogens in di¡erent meat types and the e¡ect of varying lean fat ratios The Society of Food Hygiene and Technology Hygiene Review 1–8 Available at http:// www.sofht.co.uk/isfht/irish_97_freezing.htm.

Shoemaker C.A., Klesius P.H & Bricker J.M (1999) E⁄cacy of

a modi¢ed live Edwardsiella ictaluri vaccine in channel cat¢sh as young as seven days post hatch Aquaculture

176, 189^193.

Shotts E.B & Waltman W.D (1990) A medium for the tive isolation of Edwardsiella ictaluri Journal of Wildlife Diseases 26, 214^218.

selec-StataCorp (2005) Stata Statistical Software: Release 9 Corp LP, College Station,TX, USA.

Stata-Suomalainen L.R., Reunanen H., Ijas R.,Valtonen E.T & ola M (2006) Freezing induces biased results in the mole- cular detection of Flavobacterium columnare Applied and Environmental Microbiology 72, 1702–1704.

Tiir-Thrus¢eld M (1995) Veterinary Epidemiology, 2nd edn well Science Ltd, Oxford, UK, 483pp.

Trang 18

Black-Sperm capacitation of the shrimp Litopenaeus

vannamei

Sirinda Aungsuchawan1, Craig L Browdy2& Boonsirm Withyachumnarnkul3

1

Department of Anatomy, Faculty of Medicine, Chiang Mai University, Chiang Mai,Thailand

2 South Carolina Department of Natural Resources,Waddell Mariculture Center, Charleston, SC, USA

3 Department of Anatomy, Center of Excellence for Shrimp Molecular Biology and Biotechnology (Centex Shrimp), Faculty of Science, Mahidol University, Bangkok,Thailand

Correspondence: B Withyachumnarnkul, Department of Anatomy, Center of Excellence for Shrimp Molecular Biology and Biotechnology (Centex Shrimp), Faculty of Science, Mahidol University, Rama 6 Rd., Bangkok 10400,Thailand E-mail address: boonsirm@yahoo.com

Abstract

Litopenaeus vannamei is one of the most important

spe-cies of farmed shrimp The females have

an‘open’thely-cum Mating is accomplished by attaching the male

spermatophore onto the surface of the thelycum 4^

6 h before spawning During this period, sperm may

have to undergo morphological changes associated

with a capacitation process that has been described

for other shrimp species The objective of this research

was to extend research on sperm capacitation in L

vannamei by ultrastructural and biochemical means

The sperm of L vannamei were divided into those

freshly prepared from the spermatophore (S-sperm),

extracted from the male gonopores, and those

ex-tracted from the female thelycum (T-sperm) Under

transmission electron microscopy, ultrastructural

dif-ferences were detected between the S- and theT-sperm

in the nuclear material, the ¢lamentous meshwork and

the cytoplasmic particles Under scanning electron

mi-croscopy, the di¡erence was observed in the cap and

spike regions Immuno£uorescence using confocal

mi-croscopy to detect tyrosine phosphorylated proteins

revealed di¡erent distribution patterns between

S-and T-sperm The location of phosphorylation activity

changed from the spike in S-sperm, to the ¢lamentous

meshwork in the T-sperm These morphological and

biochemical changes con¢rm that capacitation of L

vannamei sperm takes place following mating

Keywords: Litopenaeus vannamei, sperm,

sperma-tophore, thelycum, capacitation, tyrosine

phos-phorylation

Introduction

In many mammalian species, sperm must undergo acapacitation process in order to be able to fertilizeeggs The process is usually accomplished in thefemale reproductive organ, with complex morphologi-cal and biochemical changes in the sperm (Yanagi-machi 1994; Cooper 1998) The capacitation process

in invertebrates has not been as thoroughly gated as in mammals, and some investigatorshave suggested that invertebrate sperm did not re-quire capacitation before fertilization (Vacquire1998) Nevertheless, a capacitation process hasbeen observed in tick, nematode and shrimp (Oliver

investi-Jr & Shephard 1981; Clark investi-Jr & Gri⁄n 1988; nayake, Uhlinger, Gri⁄n & Clark Jr 1992; Vani-chviriyakit, Kruevaisayawan, Weerachatyanukul,Tawipreeda, Withyachumnarnkul, Pratoomchat,Chavadej & Sobhon 2004) In shrimp, the phenomen-

Wikrama-on has been described in SicyWikrama-onia ingentis (Clark Jr &Gri⁄n 1988; Wikramanayake et al 1992) and Penaeusmonodon (Vanichviriyakit et al 2004), both of whichare close-thelycum species In P monodon, it wasfound that the nuclear material of the sperm takenfrom the female thelycum (T-sperm) is less condensedthan that of the sperm taken from the male terminalampoule of the vas deferens (S-sperm) In addition,the protein pro¢le of the plasma membrane of S- andT-sperm is also di¡erent Certain proteins wereabsorbed while others were removed from theperipheral and integral membrane Tyrosinephosphorylation, an index of sperm capacitation(Visconti, Bailey, Moore, Pan, Olds-Clarke & KopfAquaculture Research, 2011, 42, 188^195 doi:10.1111/j.1365-2109.2010.02579.x

r 2010 The Authors

Trang 19

1995), was also increased in T-sperm, compared with

that of S-sperm These morphological and

biochem-ical modi¢cations were accomplished within 3 days

after the sperm sac was inserted into the female

the-lycum The modi¢cation was accompanied by an

in-crease in the acrosome reaction, a necessary step

before fertilization (Kruevaisayawan,Vanichviriyakit,

Weerachatyanuku, Iamsaard, Withyachumnarnkul,

Basak, Tanphaichitr & Sobhon 2008) Morphological

changes in sperm have also been observed in

S ingentis (Talbot & Chanmanon 1980; Lynn & Clark

1983; Arsenault 1984; Gri⁄n, Shigekawa & Clark Jr

1988; Dougherty 1990)

In the open-thelycum species, capacitation was

also observed in Litopenaeus occidentalis and possibly

in Litopenaeus vannamei (Alfaro, Ulate & Vargas

2007) In L occidentalis, the acrosome reaction was

observed in T-sperm but not in S-sperm In L

vanna-mei, the ultrastructure of the S- and T-sperm is

simi-lar, except for some changes in the ¢lamentous

network that lies between the nucleus and the

hemi-spherical cap On the other hand, in vitro fertilization

with the production of viable nauplii using S-sperm

has been reported, although the fertilization rate is

low (Misamore & Browdy1997) Thus, it is still a

ques-tion whether capacitaques-tion is required for L vannamei

In L vannamei, mating is accomplished by attaching

male spermatophores or sperm sacs onto the surface

of the thelycum and the spermatophores remain

at-tached to the female thelycum about 4^6 h from

mat-ing to spawnmat-ing (Misamore & Browdy 1996) The

purpose of this study was therefore to explore the

morphological changes in sperm of L vannamei at

4 h after mating using transmission electron

micro-scopy (TEM), scanning electron micromicro-scopy (SEM), as

well as the distribution of tyrosine phosphorylated

protein using confocal laser scanning microscopy

The results suggested that sperm capacitation is also

required in this economically important species

Materials and methods

Animals

Male and female broodstock of L vannamei were

ob-tained from commercial hatcheries They were

main-tained under a 12 h:12 h (L:D) photoperiod and given

commercial pellets and fresh feed Water qualities

(temperature, 28 1C; dissolved oxygen, 5 ppm;

sali-nity, 34 ppt; pH, 7.5; alkalisali-nity, 150 ppm; total

ammo-nia nitrogen,o0.1ppm; and total nitrite, o0.1ppm)

were monitored daily to assure an optimal ment for shrimp health

environ-TEMSpermatophores were collected manually from theterminal ampoule of the vas deferens of male brood-stock and from the thelycum of female broodstock at

4 h after mating The spermatophores were cut openusing a sharp knife to obtain S- and T-sperm.The spermwere ¢xed in 2.5% glutaraldehyde in arti¢cial seawater

at pH 8.0 overnight The rest of the spermatophore wascut into small pieces, approximately 1 1mm, andwashed with 0.2 M cacodylate bu¡er, at 4 1C, threetimes They were post-¢xed in 1% OsO4in 0.2 M caco-dylate bu¡er for 30 min, rinsed three times with bu¡erand dehydrated in ethanol series The samples weresubsequently washed in propylene oxide (PO) for

10 min and then 20 min After that, the tissues wereembedded in a series of mixed resin:PO (1:2,1:1 and 2:1)and pure Epon 812 for 1h each, followed by pure Epon

812 overnight On the following day, they were placed

in embedding moulds in fresh resin at 60 1C overnight.The tissue blocks were sectioned using a Reichert-JungUltracut E Ultra-microtome (Capovani Brothers, Scotia,

NY, USA) Thin sections were doubly stained with asaturated solution of uranyl acetate in methanol, fol-lowed by Reynold’s lead citrate and viewed under FEITecnai-20 TEM (Eindhoven, the Netherlands)

SEMS- and T-sperm were processed in the same manner

as in the TEM procedure from the beginning to thedehydration step From that step, they were critical-point dried and mounted on stubs The samples werecoated with 25 nm of platinum and paradium andviewed under a Hitachi S-2500 SEM (Tokyo, Japan)

In addition to the free sperm, S-sperm in the phore were also processed and visualized under SEM

spermato-Confocal laser scanning microscopyThis procedure was applied to localize tyrosinephophorylation, which is one of the indices of spermcapacitation in mammalian species (Visconti et al.1995) and in shrimp (Vanichviriyakit et al 2004).The S- and T-sperm were ¢xed with 4% paraformal-dehyde, washed twice with 0.1 M phosphate-bu¡eredsaline (PBS) and smeared on histoGrip adhesiveslides (Zymed, San Francisco, CA, USA) The smeared

Trang 20

slides were dried at room temperature To

permeabi-lize the plasma membrane of the sperm,Triton X-100

solution (0.1%) was gently dropped onto the slides to

cover the whole smear and incubated for 5 min To

prevent non-speci¢c binding, the slides were

incu-bated with 10% normal bovine serum for 30 min

Subsequently, they were incubated overnight with

monoclonal anti-tyrosine phosphate

immunoglobu-lin G (IgG) (clone P-Try-102: Cell Signaimmunoglobu-ling, Beverly,

MA, USA), followed by a gentle wash with PBS For

£uorescent visualization, the slides were incubated

with Alexa Flours546 goat anti-mouse IgG (H1L)

(Molecular Probe, Invitrogen, CA, USA), 1:100 dilution,

for 30 min and gently washed with PBS to remove

un-bounded antibody Further, the slides were secondarily

¢xed with 4% formaldehyde in PBS for 15 min,

coun-terstained with To-Pro-3 (1:500) for 1h, mounted with

antifade and observed under an Olympus FV-1000

confocal laser scanning microscope (Tokyo, Japan)

Results

Sperm of L vannamei were non-£agellated and thus

non-motile The basic ultrastructures of both S- and

T-sperm were similar, except for details of certaincomponents The features were the same as those de-scribed in Alfaro et al (2007), and their terminologieshave been used here The sperm can be divided intothe spike (S), the hemispherical cap (C), a ¢lamentousmeshwork (FM) and a hemispherical rim of cytoplas-mic particles (CP) (Fig 1a) The hemispherical capcontained acrosome (ACr) and there was a clear zone(CZ) between the ¢lamentous meshwork and the ac-rosome The cap covered about half of the area of thesperm and thus, at both peripheral ends of the cap, adiameter of 4mm could be drawn The nucleus con-tained ¢brillar chromatins, without a nucleolus and

a nuclear membrane The spike, about 3^4mm inlength, contained abundant micro¢laments, align-ing in a parallel fashion (Fig 1b), and on the base ofthe spike, two electron-dense triangular dots wereobserved (Fig 1c)

The most distinctive di¡erence between the structures of the S- and T-sperm was in the nuclearmaterial Chromatin ¢bres in the nucleus of theT-sperm were more de-condensed than that of theS-sperm (Fig 2a vs b) This made the nucleus ofthe T-sperm more vesicular than that of the S-sperm

ultra-In most cases, the ¢lamentous meshwork of the

Figure 1 Transmission electron microscopy of Litopenaeus vannamei sperm taken from the spermatophore (S-sperm) (a)The sperm is composed of an anterior spike (S), a cap region (C) and a posterior main body (MB) The cap region containsthe acrosome (ACr) and the subacrosomal region, which is composed of a clear zone (CZ) and the ¢lamentous meshwork(FM) The posterior main body contains the nucleus (N) and cytoplasmic particles (CP) (b) The spike contains abundantmicro¢laments (c) Two prominent electron-dense triangular dots (arrows) at the base of the spike

Sperm capacitation of L vannamei S Aungsuchawan et al Aquaculture Research, 2011, 42, 188–195

r 2010 The Authors

Trang 21

T-sperm was larger and denser than those of the

S-sperm (Fig 2c vs d), and in a few cases, it was

di⁄-cult to distinguish the di¡erence; however, it has

never been observed that the ¢lamentous meshwork

of the S-sperm is larger than that of the T-sperm

In the hemispherical rim of cytoplasmic particles

containing several small and large electron-dense

granules, the T-sperm occasionally contained largevesicles (V), which were not found in the S-sperm(Fig 2e vs f)

Under SEM, a spiral line on the external surface ofthe spike and a basal circular line on the external sur-face of the cap region of S-sperm, especially thosethat were embedded in the spermatophoric matrix,

Figure 2 Transmission electron microscopy of S- and T-sperm of Litopenaeus vannamei showing ultrastructural ences All pictures use the same scale bar (a, b) The nucleus (N): The chromatin ¢bres of the T-sperm are more de-con-densed than those of the S-sperm (c, d) The ¢lamentous meshwork (FM): The granules in the T-sperm were larger thanthose of the S-sperm (e, f) The cytoplasmic particles: Electron-lucent vesicles (V) were occasionally found in the T- spermbut not in the S-sperm

Trang 22

di¡er-were observed (Fig 3a) The two lines di¡er-were less

pro-minent in T-sperm (Fig 3b), compared with those of

the S-sperm

The immuno£uorescent localization of tyrosine

phosphorylated proteins on intact free S- and

T-sperm revealed a di¡erence in the area staining

positive S-sperm showed positive £uorescent signals

on their plasma membrane and spike, but T-sperm

was positive on the plasma membrane and

¢lamen-tous meshwork, and on the base of spikes of very few

sperm (Fig 4) A bright spot of red £uorescence was

present at a central part of the ¢lamentous meshwork

of many T-sperm, but not in the S-sperm The positive

signal of tyrosine phosphorylated proteins at the

spike was decreased from S- to T-sperm and it

disap-peared completely when the bright red spot apdisap-peared

on the ¢lamentous meshwork

Discussion

The basic ultrastructure of L vannamei sperm was

similar to that described by Alfaro et al (2007)

Simi-lar to the spike of Penaeus aztecus and S ingentis

sperm, abundant micro¢laments run along its length

Figure 3 (a) Scanning electron microscopic of S-spermembedded in the matrix of spermatophore; (b) T-spermisolated from the thelycum of Litopenaeus vannamei Thebasal line (BL) surrounding the base of the spike and thespiral line (SL) on the spike are less prominent in the T-sperm Both pictures use the same scale bar

Figure 4 Immuno£uorescent localization of tyrosine phosphorylation (P-tyr) on S- and T-sperm of Litopenaeus mei Four pictures in (a), as well as in (b), were taken from the same sperm but di¡ered in the staining selection: the red

vanna-£uorescence signal represents P-tyr, green vanna-£uorescence represents the sperm nucleus or DNA, no-colour picture showsthe phase-contrast one and the fourth picture shows the overlapping of P-tyr and DNA stains The P-tyr signal was found

on the sperm membrane surrounding the cap and the main body of both sperm types; however, in S-sperm, it was tionally found on the spike (arrows, a&c), and in T-sperm, on the ¢lamentous meshwork (FM) A few T-sperm had the P-tyrsignal on the base of the spike (d, arrow), which disappeared when a bright red spot of the P-tyr signal appeared at thecentre of FM (d, arrowhead) All pictures use the same scale bar

addi-Sperm capacitation of L vannamei S Aungsuchawan et al Aquaculture Research, 2011, 42, 188–195

r 2010 The Authors

Trang 23

without microtubules or centrioles, and unlike

Macrobrachium rosenbergii sperm, which possessed a

pair of centrioles in closely associated with the base

of the spike (Lynn & Clark 1983) Interestingly, a

pro-minent pair of triangular electron dense spots was

present at the base of spike in both S- and T-sperm

This structure, which was also observed in the TEM

picture by Alfaro et al (2007), may act as an anchor

to ¢x the cap with the spike or may be the centre of

the origin of micro¢laments that run into the length

of the spike, or may control the retraction of spike

during an acrosome reaction Similar to the previous

¢nding (Alfaro et al 2007), the nucleus of L vannamei

has no nuclear membrane, but instead, is surrounded

by an amorphous cytoplasm; this lack of nuclear

en-velop is also shared by S ingentis sperm (Kleve,Yudin

& Clark 1980)

In this study, three basic di¡erences in the

ultra-structure were detected between the S- and the

T-sperm: the nucleus, the ¢lamentous meshwork and

the cytoplasmic particles Alfaro et al (2007)

des-cribed the di¡erence in the ¢lamentous meshwork in

both L vannamei and Litopenaeus stylirostris, which

was larger in T-sperm, and suggested that the change

signi¢ed its role in the acrosome reaction This

di¡er-ence was also detected in our study in that the

¢la-mentous meshwork was not only larger but also

denser in the T-sperm This change was also clearly

observed in the capacitation process in S ingentis

(Wikramanayake et al 1992), where the ¢lamentous

meshwork was termed the granular region and the

increase in this area was termed an extended saucer

In L stylirostris (Alfaro et al 2007) and P monodon

(Pongtippatee, Vanichviriyakit, Chavadej, Plodpai,

Pratoomchart, Sobhon & Withyachumnarnkul

2007), the ¢lamentous meshwork was not observed

in S-sperm, but it appeared in sperm that entered the

early stage of the acrosome reaction in P monodon

(Pongtippatee et al 2007), where it was described as

polymerization of the anterior granulation

The signi¢cant di¡erence between the S- and

T-sperm is in the nucleus T-T-sperm had more

deconden-sation than S-sperm, which was probably a part of

the capacitation process similar to that observed in

P monodon (Vanichviriyakit et al 2004) However, this

di¡erence was not observed by Alfaro et al (2007)

The di¡erence was not distinctive and might become

overlooked, perhaps requiring densitometry to

quan-tify the density of the nuclear material In our TEM

studies, this di¡erence was observable and

consis-tent In S ingentis, it was shown that the nuclear

ma-terial was devoid of basic histones and protamines

and was composed of naked DNA (Kleve et al 1980)

If L vannamei sperm lacks one or both of these cleoproteins, then the more decondensed appearance

nu-in the T-sperm may be due to uncoilnu-ing of its DNA orwater in£ux into the nucleus

At present, we cannot o¡er any explanation for thedi¡erences in cytoplasmic particles, the spiral lineand the basal line between the S- and the T-spermfound in this study Interestingly, the spiral line wasalso observed in S ingentis S-sperm, and was also ab-sent in its T-sperm (Wikramanayake et al 1992).The immuno£uorescent analyses to localize tyro-sine phosphorylation protein revealed striking di¡er-ences between S- and T-sperm The reaction wentfrom the spike to the ¢lamentous meshwork The redspot at the ¢lamentous meshwork could indicate anearly acrosome reaction event similar to the spheri-cal mass formation as shown in P monodon sperm(Pongtippatee et al 2007) Because the spike was stillpresent in the T-sperm (Figs 3 and 4), the absence ofimmuno£uorescence from the spike and its presence

as a red spot on the ¢lamentous meshwork was fore a purely biochemical phenomenon, i.e., tyrosinephosphorylation

there-In rat, di¡erences in tyrosine phosphorylationhave been shown between caput and caudal epididy-mal sperm (Lewis & Aiken 2001) Tyrosine phosphor-ylation was localized in the acrosomal region of thecaput sperm, whereas a majority of caudal sperm ex-hibited a narrow band of tyrosine phosphorylation

at the posterior acrosomal margin These di¡erences

in tyrosine phosphorylation protein expressionbetween caput and caudal epididymal spermatozoaappeared to re£ect the competence of the sperm Inmouse sperm, increased tyrosine phosphorylation inthe £agellum during sperm maturation was detected(Urner, Leppens-Luisier & Sakkas 2001) In the blacktiger shrimp P monodon, tyrosine phosphorelationwas found at a low level in S-sperm A stepwiseincrease in tyrosine phosphorylation markedlyappeared in T-sperm from day 1 to day 2 and its levelwas relatively constant in day 3 (Vanichviriyakit et al.2004) It was not shown, however, where thetyrosine phosphorylation took place, and whetherits location changed within the sperm cell

It is clear in this study that sperm of L vannameishows morphological and biochemical changes afterbeing transferred from the male to the female thely-cum; it is likely that the changes suggest a capacita-tion process of the sperm The questions are howthis process occurs, within 4^6 h, and whetherthe female has any in£uence on these changes

Trang 24

In P monodon, it takes at least 3 days for the

capacita-tion process to be complete, and it is conceivable that

the female may play a signi¢cant role in inducing

these changes because the sperm sacs have to be

in-serted into the female thelycum and surrounded by

components within the thelycal £uid But in L

vanna-mei, the sperm sacs are completely outside the female

body as the shrimp has the open-type thelycum

Therefore, the female may not play any part in the

in-duction of the capacitation process, unless there are

minute pores or channel connecting the sperm sacs

with the inner part of the female Another possibility

is that the sperm of L vannamei may have a

sponta-neous capactation after being released from the male

gonopore, i.e., just being exposed to seawater It is

also possible that the capacitation may be inhibited

by certain factor(s) inside the terminal part of the

vas deferens and disinhibition occurs as the sperm

sac is released out from the male gonopore The latter

would explain the potential for in vitro fertilization

using S-sperm observed by Misamore and Browdy

(1997) These two possibilities are related and are

un-der further investigation In light of the ease of sperm

capacitation of L vannamei, it can be stated that

sperm of this shrimp species are quite di¡erent from

those of P monodon, and this phenomenon may be

one of the basic di¡erences between the closed- and

the open-type thelycum species regarding the

phylo-genetic development of the reproductive function of

the shrimp

Acknowledgments

We would like to thank Eleanor F Shepard and

Laur-inda L Smith for their kind support and technical

help The study was funded by The Commission on

Higher Education and Mahidol University grant in

2007

References

Alfaro J., Ulate K & Vargas M (2007) Sperm maturation and

capacitation in the open thelycum shrimp Litopenaeus

(Crustacea: Decapoda: Penaeoidea) Aquaculture 270,

436^442.

Arsenault A.L (1984) Changes in the nuclear envelope

asso-ciated with spermatid di¡erentiation in the shrimp,

Cran-gon septemspinosa Journal of Ultrastructure Research 86,

294^308.

Clark W.H Jr & Gri⁄n F.J (1988) The morphology and

phy-siology of the acrosome reaction in the sperm of the

dec-apod, Sicyonia ingentis Development, Growth and Di¡erentiation 30, 451^462.

Cooper T.G (1998) Interactions between epididymal tions and spermatozoa Journal of Reproduction and Ferti- lity Supplement 53, 119^136.

secre-Dougherty W.J (1990) Ultrastructural observation on nized sperm in developing and fully formed spermato- phores of male shrimp, Penaeus vannamei In: Pathology

mela-in Marmela-ine Science (ed by JO Jenkmela-ins & TC Cheng), pp 387^

394 Academic Press, London, UK.

Gri⁄n F.J., Shigekawa K & Clark W.H Jr (1988) Formation and structure of the acrosomal ¢lament in the sperm of Sicyonia ingentis Journal of Experimental Zoology 246, 94^102.

Kleve M.G., Yudin A.I & Clark W.H (1980) Fine structure of the unistellate sperm of the shrimp, Sicyonia ingentis (Na- tantia) Tissue and Cell 12, 29^45.

Kruevaisayawan H., Vanichviriyakit R., Weerachatyanuku W., Iamsaard S., Withyachumnarnkul B., Basak A., Tan- phaichitr N & Sobhon P (2008) Induction of the acro- some reaction in black tiger shrimp (Penaeus monodon) requires sperm trypsin-like enzyme activity Biology of Re- production 79, 134^141.

Lewis B & Aiken R.J (2001) Impact of epididymal tion on the tyrosine phosphorylation patterns exhibited

matura-by rat spermatozoa Biology of Reproduction 64, 1545^ 1556.

Lynn J.W & Clark W.H Jr (1983) Morphology of sperm-egg interaction in the eggs of the freshwater prawn Macrobra- chium rosenbergii Biological Bulletin 164, 446^458 Misamore M & Browdy C.L (1996) Mating behavior in the white shrimp Penaeus setiferus and Penaeus vannamei: a generalized model for Penaeus mating Journal of Crusta- cean Biology 16, 61^70.

Misamore M & Browdy C.L (1997) Evaluating hybridization potential between Penaeus setiferus and Penaeus vannamei through natural mating, arti¢cial insemination and in vi- tro fertilization Aquaculture 150, 1^10.

Oliver J.H Jr & Shephard J (1981) Morphogenesis and tion of tick sperm In: Advances in Invertebrate Reproduc- tion (ed by W.H Clark & T.S Adams), pp 243^251 Elsevier, NewYork, NY, USA.

activa-Pongtippatee P., Vanichviriyakit R., Chavadej J., Plodpai P., Pratoomchart B., Sobhon P & Withyachumnarnkul B (2007) Acrosome reaction in the sperm of the black tiger shrimp Penaeus monodon (Decapoda, Penaeidae) Aquacul- ture Research 38, 1635^1644.

Talbot P & Chanmanon P (1980) The structure of sperm from the lobster, Homarus americanus Journal of Ultra- structure Research 70, 275^286.

Urner F., Leppens-Luisier G & Sakkas D (2001) Protein osine phosphorelation in sperm during gamete interac- tion in the mouse: the in£uence of glucose Biology of Reproduction 64, 1350^1357.

tyr-Vacquire V.D (1998) Evolution of gamete recognition teins Science 281, 1995^1998.

pro-Sperm capacitation of L vannamei S Aungsuchawan et al Aquaculture Research, 2011, 42, 188–195

r 2010 The Authors

Trang 25

Vanichviriyakit R., Kruevaisayawan H., Weerachatyanukul

W., Tawipreeda P., Withyachumnarnkul B., Pratoomchat

B., Chavadej J & Sobhon P (2004) Molecular modi¢cation

of Penaeus monodon sperm in female thelycum and its

consequent responses Molecular Reproduction and

Devel-opment 69, 356^363.

Visconti P.E., Bailey J.L., Moore G.D., Pan D., Olds-Clarke P &

Kopf G.S (1995) Correlation between the capacitation

state and protein tyrosine phosphorylation Development

121, 1129^1137.

Wikramanayake A.H., Uhlinger K.R., Gri⁄n F.J & ClarkW.H.

Jr (1992) Sperm of the shrimp Sicyonia ingentis undergo a bi-phasic capacitation accompanied by morphological changes Development, Growth and Di¡erentiation 34, 347^355.

Yanagimachi R (1994) Mammalian fertilization The siology of Reproduction (ed by E Knobil & JD Neill), pp 189^317 Raven Press, NewYork, NY, USA.

Trang 26

Phy-Evaluating the use of Lactobacillus acidophilus as

a biocontrol agent against common pathogenic

bacteria and the effects on the haematology

parameters and histopathology in African catfish

Mohammed Abdullah Al-Dohail1,2, Roshada Hashim1& Mohammed Aliyu-Paiko1

1 Laboratory of Aquafeed and Feeding Development, Aquaculture Research Group, School of Biological Sciences, Universiti Sains Malaysia, Penang, Malaysia

2 Faculty of Environmental Sciences and Marine Biology, Hadhramout University of Science and Technology, Mukalla,Yemen

Correspondence: M A Al-Dohail, Laboratory of Aquafeed and Feeding Development, Aquaculture Research Group, School of Biological Sciences, Universiti Sains Malaysia, 11800 Penang, Malaysia E-mail: aldohial@yahoo.com

Abstract

This study was carried out to evaluate the use of

Lac-tobacillus acidophilus as a biocontrol agent against

some common ¢sh pathogenic bacteria

(Staphylococ-cus xylosus, Aeromonas hydrophila gr.2 and

Streptococ-cus agalactiae) in African cat¢sh, Clarias gariepinus

Eight treatments were designed inclusive of 10 C

gar-iepinus juveniles (mean weight 190 g) per tank, each

in triplicate Four groups of ¢sh were fed a diet

supple-mented with L acidophilus, comprising about 3.01

107colony-forming units per gram of diet (the

probio-tics diet), while the other four groups were fed a diet

not supplemented with probiotics (the non-probiotics

diet) In the ¢rst group, ¢sh were injected with 1mL

physiological saline and fed the non-probiotic

diet (non-probiotic control); in the second, third

and fourth groups, ¢sh were injected with 1mL each

of S xylosus, A hydrophila gr.2 and S agalactiae,

respectively, and were all fed the non-probiotic diet

(designated as non-probiotic treatments; NPsx, NPah

and NPsa respectively) In the ¢fth group, ¢sh were

injected with 1mL physiological saline but fed the

probiotic diet (probiotic control), while ¢sh in the

sixth, seventh and eighth groups were each injected

with 1mL of S xylosus, A hydrophila gr.2 and S

aga-lactiae, respectively, and were all fed the probiotic diet

(and designated as probiotic treatments; Psx, Pah and

Psa respectively) Blood samples were collected for

haematology analysis, while samples of the liver and

kidney were examined for pathohistology after 7 and

21 days of infection The results showed that the matology parameters, packed cell volume, haemoglo-bin, erythrocyte sedimentation rate, red blood cell,white blood cell, total serum protein, Mg21, Ca21,

hae-Cl, glucose, cholesterol and total immunoglobulinconcentrations and the pathohistology of the liverand kidney were better in the challenged ¢sh (in-fected) maintained on the probiotic diet than those

in the groups fed the non-probiotic diet It is cluded, based on these results, that L acidophilus isuseful as a probiotic agent in C gariepinus againstthese pathogenic bacteria (S xylosus, A hydrophilagr.2 and S agalactiae)

con-Keywords: Probiotic, Lactobacillus acidophilus,pathogenic bacteria, blood, histopathology, Clariasgariepinus, ¢sh diseases

IntroductionFish diseases are widely distributed worldwide andare considered to be serious problems in aquaculture(Lim & Webster 2001) The advancement in intensiveaquaculture is usually accompanied by several dis-ease problems, often due to opportunistic pathogens,

as evident from general aquaculture (Abraham, dal & Babu 2008) This is because at higher stockingdensities, high feed inputs and other organic loadsAquaculture Research, 2011, 42, 196^209 doi:10.1111/j.1365-2109.2010.02606.x

Mon-r 2010 Universiti Sains Malaysia

Trang 27

stimulate the selective proliferation of opportunistic

bacteria, leading to ¢sh diseases (Austin, Stuckey,

Ro-bertson, E¡endi & Gri⁄th 1995) In freshwater ¢sh

culture, disease outbreaks due to pathogenic bacteria

are very common, especially in the production of

cat¢sh (Lim & Webster 2001) Among the common

bacteria pathogens, Staphylococcus xylosus,

Aeromo-nas hydrophila gr.2 and Streptococcus agalactiae are

known to seriously infect ¢sh in culture systems,

sometimes causing heavy mortalities (Schperclaus,

Kulow & Schreckenbach 1992)

Blood parameters of ¢sh provide accurate

indica-tions of any changes occurring in the organism as a

result of injuries to organs or tissues related to

infec-tious diseases, similar to those of warm-blooded

ani-mals (Adeyemo 2007) Ranzani-paiva, Ishikawa, Das

Eiras and Felizardo (2000) also reported that the

blood parameters of ¢sh are important tools useful

for detecting abnormalities related to temperature

and dissolved oxygen variations, diseases and other

factors Thrall (2004) and Pimpao, Zampronio and

Silva de Assis (2007) reported that the

haematologi-cal and biochemihaematologi-cal parameters of ¢sh can be used

to evaluate the health condition of the organism

Furthermore, haematological variations are useful

for the clinical diagnosis of ¢sh physiology, to

deter-mine the e¡ects of external stressors and toxic

sub-stances (Cech Jr, Bartholow, Young & Hopkins 1996;

Bonga 1997) For these reasons, many authors have

used haematological parameters to report the e¡ects

of environmental pollutants and pathogenic diseases

causing changes in blood characteristics of ¢sh such

as Clarias gariepinus (Onusiriuka & Ufodike 2000;

Ezeri 2001; Gabriel, Alagoa & Allison 2001)

Probiotics can be de¢ned as living microbial cells

that exert bene¢cial e¡ects on the health and

well-being of their host (Salminen, Ouwehand, Benno &

Lee 1999) The use of probiotic bacteria in

aquacul-ture to enhance the growth performance and to also

improve the quality of water in which ¢sh are

cultured has received considerable attention

(Verschuere, Rombaut, Sorgeloos & Verstraete 2000)

Generally, probiotics are associated with bacteria,

which produce bacteriocins and other compounds

that may inhibit the growth of pathogenic bacteria

(Klaenhammer & Russell 2000) Lactic acid bacteria

have been identi¢ed to possess probiotic properties

and some bene¢cial advantages, as it enhances the

digestive process in its host and could therefore be

useful due to its bene¢cial e¡ects on the health of its

consumers (Schillinger 1999) Strm and Ring

(1993), Stosik and Szenfeld (1996), Rengpipat,

Ruk-pratanporn, Piyatiratitivorakul and Menasaveta(2000) and Vine (2004) all concluded that probioticshave the ability to improve ¢sh health and preventbacterial diseases in ¢sh Consequently, the use ofprobiotics as a new technique to confer protection

in the host ¢sh against pathogenic bacteria in themost economic and environment-friendly manner iscertainly worth evaluating in aquaculture

Information regarding the e¡ects of probiotics ondiseases of C gariepinus is limited; therefore, thisstudy was conducted to evaluate the use of Lactoba-cillus acidophilus as a biocontrol agent against somecommon pathogenic bacteria (S xylosus, A hydrophi-

la gr.2 and S agalactiae) and its e¡ects on the tological parameters and histopathology of Africancat¢sh C gariepinus juveniles

haema-Materials and methods

L acidophilusLactobacillus acidophilus was isolated from a knownsource (Vitagenryoghurt drink, produced by Malay-sia Milk Sdn Bhd Penang, Malaysia), identi¢ed usingthe Biolog plate technique (Choi & Dobbs 1999) andsub-cultured on de Man, Rogosa and Sharpe (MRS)media in the microbiology laboratory, UniversitiSains Malaysia One millilitre of Vitagen yoghurtdrink was added to 99 mL of MRS media broth andincubated at 37 1C for 36 h About 1.0 mL of this mix-ture was serially diluted in 10-fold (10 1^10 10)physiological saline as a diluent to determine the vi-able cell count of L acidophilus in a 10-fold dilution ofthe solution (Koch 1994) Subsequently, 0.1mL ofeach dilution (10 1^10 10) was taken and evenlyspread on MRS agar plates and incubated at 37 1Cfor 48 h to produce L acidophilus colonies The num-bers of colonies produced on the plates were countedmanually

Pathogenic bacteriaThree common pathogenic bacteria of fresh water

¢sh, S xylosus, A hydrophila gr.2 and S agalactiae,were obtained from the National Fish Health Re-search Center Penang, Malaysia The bacteria werecultivated at 30 1C for 24 h in the Microbiology La-boratory, Universiti Sains Malaysia, and maintained

on nutrient agar slants kept in a refrigerator at 4 1Cuntil used The viable cell count of pathogenic bacter-

ia on the plates was also performed manually

Trang 28

Experimental ¢sh and husbandry conditions

This experiment was conducted in the Diseases

La-boratory at the Aquaculture Research Complex of

Universiti Sains Malaysia, in May 2008 Two groups

of juvenile African cat¢sh, C gariepinus, were reared

for12 weeks on a probiotic diet containing 3.01 107

colony-forming units (CFU) per gram and a

non-pro-biotic supplemented diet respectively (refer to

Al-Do-hail, Hashim & Aliyu-Paiko 2009) From these, 120

homogenous juveniles (of mean weight 190 5 g)

were collected from each group and used for the

pre-sent trial Fish were acclimated to laboratory

condi-tions for 2 weeks in a £ow-through water system

before the commencement of this study, feeding on

the same diets as those in the previous experiment

Ten juveniles were randomly distributed into each of

24 glass aquariums (measuring 60 35  30 cm,

L W  H) Three replicate groups of ¢sh were

maintained on each assigned diet and ¢sh were

reared under a natural photoperiod of approximately

12/12 h light/dark cycles Each glass aquarium was

connected to the £ow-through water system, which

was continuously supplied with de-chlorinated tap

water, with the £ow rate set at 1.5 L min 1and also

aerated continuously Water quality parameters

were monitored weekly, and the mean values

re-corded were as follows: ammonia (mg L 1) 0.9

0.10, nitrite (mg L 1) 0.05 0.02, nitrate (mg L 1)

3.07 0.29, temperature ( 1C) 29.47  0.50 and pH

6.52 0.11 At the end of the ¢rst and third weeks of

the trial, three ¢sh from each aquarium were

ran-domly removed for haematology and patho-histology

analysis

Preparation of experimental diets

Two practical diets were formulated to contain 35%

crude protein each, using soybean and ¢sh meals

as the protein sources The ¢rst diet was

supplemen-ted with L acidophilus (probiotic diet), containing

 3.01  107CFU g 1, while the second diet was

not supplemented with probiotic bacteria

(non-pro-biotic diet) The lipid level in the two diets was 10%,

comprising a 1:1 ratio of ¢sh oil to vegetable oil Under

sterile conditions, the probiotic diet (supplemented

with L acidophilus) was prepared by slowly spraying

250 mL of media broth containing about 3.7 109

CFU mL 1of live L acidophilus (grown in MRS media

and1% w/v of skimmed milk as a cryoprotective

solu-tion) into a clean plate containing the dry pellets

while slowly mixing for 3^5 min to prepare1kg of

ex-perimental feedstu¡ The pellets were dried in anoven at 30 1C for 18 h, packed and stored in a freezer

at 20 until used The ¢nal concentration of L ophilus in the probiotic diet was 3.01 107

acid-g 1,whereas the formulation of diets used in the feedingtrial is as described previously in Al-Dohail et al.(2009)

Feeding trialThe feeding trial was performed for eight treatments

in a completely randomized design, each in triplicate.Juvenile C gariepinus were hand fed the assigneddiets to visual satiation three times daily at 08:00,14:00 and 18:00 hours respectively The feeding trialand treatments used for this study are as presented inTable 1 All groups were maintained on the assigneddiets for 3 weeks after infection with the designatedpathogenic bacteria

Challenging ¢sh with pathogensThree di¡erent pathogenic bacteria, S xylosus, A hy-drophila gr.2 and S agalactiae, were used in the pre-sent study Under sterile conditions, the ¢rst and

¢fth groups of ¢sh were injected with physiologicalsaline to serve as the controls However, the secondand sixth groups were infected with S xylosus, thethird and seventh groups were infected with A hy-drophila gr.2 and the fourth and eighth groups wereinfected with S agalactiae

Fish in all groups (except the ¢rst and the ¢fth) wereinjected intraperitoneally with1mL of the pathogenicbacteria, containing  2  106

CFU mL 1 pended in physiological saline Fish in the ¢rst and

sus-¢fth groups were injected intraperitoneally with1mL of physiological saline, using 1mL capacity syr-inges (TERUMOssyringes and needles, Somerset, NJ,USA) The amount of 2 106CFU mL 1of patho-genic bacteria injected into C gariepinus juvenilewas determined to be the infectious dose, in a sepa-rate preliminary study before the current feedingtrial (Al-Dohail, Hashim & Aliyu-Paiko 2008)

In another preliminary study, the in vitro inhibition

of the three pathogenic bacteria by L acidophilus wasassessed and found to be 15.50 mm for A hydrophilagr.2 and 20.33 mm for S xylosus, while the inhibitionzone for S agalactiae was 21.17 mm (Al-Dohail et al.2008)

L acidophilus as a probiotic in African cat¢sh M A Al-Dohail et al Aquaculture Research, 2011, 42, 196–209

r 2010 Universiti Sains Malaysia

Trang 29

Haematological parameters

For the determination of each of the following

hae-matology parameters, blood was collected from the

caudal vein of three randomly selected ¢sh per tank

All the procedures used are as described previously

(Al-Dohail et al 2009)

Haematocrit

Haematocrit was measured according to the

proce-dures of Schperclaus et al (1992) as described

pre-viously The haematocrit value was calculated based

on the following formula:

PCV¼ Height of packed red cells ðmmÞ=

Height of packed red cells and plasmaðmmÞ

 100

where PCV is the packed cell volume

Haemoglobin (Hb) concentration

The Hb content was determined according to the

Cyan-methaemoglobin method (Blaxhall & Daisley

1973) as described previously The Hb concentration

was determined as g dL 1of blood from a standard

graph, using the following equations:

Haemoglobin concentration (g dL 1) 5 Absorbance

of sample/Absorbance of standard concentration of

standard

Mean corpuscular haemoglobin concentration(MCHC) g dL 15Haemoglobin g %/Haematocrit

volume % 100Mean corpuscular haemoglobin (MCH)

pg cell 15Haemoglobin g %/erythrocyte

(millions mm 3) 10Mean corpuscular volume (MCV)mm 35 Haema-tocrit volume/erythrocyte (millions per mm 3) 10

Erythrocyte sedimentation rate (ESR)The ESR was calculated according to the method ofWestergren (Britton1963) as described previously Er-ythrocyte sedimentation rates were expressed as thedistance moved in mm per hour of corpuscles in theblood

Total red blood cell count (erythrocyte or RBC

106mm 3)The total RBC counts were carried out based on theprocedure of EWOS Technology Center (2000) andJohnson, Timmons and Hall (2002) as described pre-viously The RBC count was recorded as millions ofcells per cubic millimetre Total RBC was calculatedaccording to the following formula:

RBCðmm3Þ ¼ ðN  5  10  200Þwhere N is the number of cells in ¢ve squares,5 repre-sents multiplication factor to yield the number ofcells in 1mm2, 10 represents multiplication factor tobring the depth of the chamber from 0.1 to 1mm and

200 represents the dilution factor

Total white blood cells (WBCs)

A leucocyte count (LC) was performed according toStoskopf (1993) and Johnson et al (2002) as describedpreviously The LC was expressed as thousands ofcells in mm3

Leucocyteðmm3Þ ¼ LC  500where LC represents number of cells in 4 mm2squares and 500 represents dilution and volume cor-rection factor

Serum mineral content (Mg21, Ca21and Cl)Serum minerals (magnesium, calcium and chloride)were measured as described previously (Al-Dohail

et al 2009) The serum mineral values were lated and recorded as (mg dL 1) or (mmole L 1)according to the following formula:

calcu-Table 1 De¢nition of ¢sh groups and treatments used for

the feeding trial

Group Treatment Feeding trial descriptions 

1 NPc Fish injected with 1 mL of physiological saline

and fed the non-probiotic diet (control)

2 NPsx Fish injected with 1 mL of Staphylococcus

xylosus and fed the non-probiotic diet

3 NPah Fish infected with 1 mL of Aeromonas

hydrophila gr.2 and fed the non-probiotic diet

4 NPsa Fish infected with 1 mL of Streptocossus

agalactiae and fed the non-probiotic diet

5 Pc Fish injected with 1 mL of physiological saline

and fed the probiotic diet (control)

6 Psx Fish infected with 1 mL of S xylosus and fed

the probiotic diet

7 Pah Fish infected with 1 mL of A hydrophila gr.2

and fed the probiotic diet

8 Psa Fish infected with 1 mL of S agalactiae and

fed the probiotic diet

Fish in all the treatments were injected with pathogenic

bacter-ia or physiological saline after a 3-month rearing period (initbacter-ial

time of starting the experiment).

Trang 30

Serum minerals (mg dL 1) or (mmole L 1) 5

(Absorbance of sample/Absorbance of standard)

 concentration of standard

Cholesterol concentration

The blood cholesterol level was calculated as

described previously (Al-Dohail et al 2009)

Cholesterol (mg dL 1) 5 Absorbance of sample/

Absorbance of standard) concentration of standard

(mg dL 1)

Glucose concentration

Blood glucose was measured as described previously

(Al-Dohail et al 2009) and calculated according to

the following formula:

Glucoseðmg dL1Þ ¼ ðDA sample=DA standardÞ

concentration of standard ðmg dL1Þ

Total protein content

Total serum protein was measured as already

de-scribed (Al-Dohail et al 2009) The total protein value

was expressed as (g dL 1), calculated according to

the following formula:

Proteinðg dL1Þ ¼ ðD Absorbance sample=

D Absorbance standardÞ

concentration of standard ðg dL1Þ

Total immunoglobulin (Ig) concentration

The total Ig concentration was measured according

to the method of Siwicki and Anderson (1993) and

Amar, Kiron, Satoh, Okamoto and Watanabe (2000)

as also described previously The total Ig value was

expressed as (mg mL 1), calculated according to the

following formula:

Total Igðmg mL1Þ ¼ total protein in serum sample

 total protein treated with PEG

Histopathology assessment

Samples of the liver and kidney were collected for a

pathohistological examination and ¢xed in 10%

for-malin Afterwards, the samples were dehydrated in a

series of ethanol solutions: 50%,70%, 80%, 90%, 95%

and 100% for 1, 1, 2, 2, 1.5 and 16 hours respectively

Each of the samples was then transferred into a

xy-lene solution for another 30 min and then embedded

in a solution of xylene wax for 1h A 5-mm-thick

sec-tion was cut using a rotary microtome (A.O Spencer,model Reicheit-Jurg 820, Leica, Germany) and quicklytransferred onto a slide and kept in an oven at 40^

50 1C for 24 h Each sample was observed under acompound light microscope attached with a camera

Statistical analysisThe data for haematology parameters and Ig concen-tration were analysed using one-way analysis of var-iance (ANOVA), and the di¡erences in the meansbetween groups were tested for signi¢cance at the95% con¢dence level P valueso0.05 were consid-ered to be signi¢cant, using Duncan’s multiple rangetest All statistical analyses were performed using the

SPSSsoftware package, version 11.5

ResultsSeven days post infection with pathogenicbacteria

Fish survival rate in the controls and infected groupsfed the probiotic diet was100%, whereas in the groupsinfected with S xylosus, A hydrophila gr.2 and S aga-lactiae fed the non-probiotic diet, ¢sh survival re-corded was 83.3%, 76.6% and 80.0% respectively Theresults for the monitored haematology parameters of

C gariepinus juveniles fed probiotic or non-probioticdiets after 1 week of infection with the three patho-genic bacteria are presented in Table 2 Signi¢cant(Po0.05) di¡erences were observed in the haematol-ogy parameters between the infected groups fed theprobiotic diet and those maintained on the non-pro-biotic diet Overall, the haematocrit (PCV), Hb, RBC,cholesterol, total serum protein, calcium (Ca12), mag-nesium (Mg12), chloride (Cl) and total Ig values fol-lowed a similar pattern; the values were signi¢cantlyhigher in the infected ¢sh maintained on the probioticdiet than in those fed the non-probiotic diet Conver-sely, the ESR, WBC and serum glucose concentrationwere signi¢cantly higher in the infected ¢sh fed thenon-probiotic diet than in those infected with patho-genic bacteria and maintained on the probiotic dietand the two controls respectively

The ESR values in the ¢sh groups infected with S.agalactiae and S xylosus and maintained on the pro-biotic diet did not signi¢cantly di¡er from both theprobiotic and the non-probiotic controls, whereasthe non-probiotic treatment di¡ered signi¢cantlyfrom both controls in their ESR values

L acidophilus as a probiotic in African cat¢sh M A Al-Dohail et al Aquaculture Research, 2011, 42, 196–209

r 2010 Universiti Sains Malaysia

Trang 31

The Hb and PCV values of the probiotic treatments

infected with S xylosus and A hydrophila gr.2 were

similar to those of the non-probiotic control Fish

in-fected with S agalactiae and fed the probiotic diet, on

the other hand, showed Hb and PCV values similar to

those of the probiotic control

The RBC values of the ¢sh infected with S

agalac-tiae and fed the probiotic diet did not di¡er

signi¢-cantly from the non-probiotic control, whereas all

the ¢sh infected with all pathogenic bacteria and

maintained on the non-probiotic diet showed

signi¢-cant di¡erences in RBC between both probiotic and

non-probiotic controls respectively

For the RBC indices, MCHC, MCH and MCV in the

infected groups of ¢sh maintained on the probiotic

diet were signi¢cantly higher than the groups fed the

non-probiotic diet, even though both controls were

si-milar However, MCH and MCV in ¢sh fed either the

probiotic or the non-probiotic diets were signi¢cantlyhigher than both the controls respectively

The WBC level was signi¢cantly higher in theblood of all the infected ¢sh fed both probiotic andnon-probiotic diets than in both the controls, respec-tively, whereas the levels in all the infected ¢sh fed thenon-probiotic diet were signi¢cantly and consistentlyhigher than those in their counterpart groups main-tained on the probiotic diet respectively

The serum total Ig concentration was signi¢cantlyhigher (Po0.05) in the infected ¢sh fed the probioticdiet than in the counterpart groups maintained onthe non-probiotic diet, although infected ¢sh fed theprobiotic diet showed no signi¢cant di¡erences fromthose recorded for the probiotic control group, norwas there any noticeable di¡erence in the Ig levels be-tween the non-probiotic treatments and the non-pro-biotic control

Table 2 Haematological parameters of Clarias gariepinus at 7 days post infection and fed non-probiotic and probiotic diets

Results are expressed as mean  SD.

Mean values in the same row with di¡erent superscripts show signi¢cant di¡erence (P o0.05).

NPc, non-probiotic control; NPsx, ¢sh infected with Staphylococcus xylosus and fed the non-probiotic diet; Npah, ¢sh infected with Aeromonas hydrophila gr.2 and fed the non-probiotic diet; Npsa, ¢sh infected with Streptocossus agalactiae and fed the non-probiotic diet;

Pc, probiotic control; Psx, ¢sh infected with Staphylococcus xylosus and fed the probiotic diet; Pah, ¢sh infected with Aeromonas phila gr.2 and fed the probiotic diet; Psa, ¢sh infected with Streptocossus agalactiae and fed the probiotic diet; WBC, white blood cell, RBC, red blood cell; MCHC, mean corpuscular haemoglobin concentration; MCH, mean corpuscular haemoglobin; MCV, mean corpuscular volume.

Trang 32

hydro-The serum blood protein, cholesterol, Mg12and

Cllevels in all the infected ¢sh groups fed the

pro-biotic diet were signi¢cantly di¡erent from those of

their counterparts maintained on the non-probiotic

diet respectively The serum Ca12concentration was

signi¢cantly lower in infected ¢sh groups fed the

non-probiotic diet than in the other groups, with no

signi¢cant di¡erences recorded between the infected

¢sh groups maintained on the probiotic diet and the

non-probiotic control group

The serum glucose level was signi¢cantly higher in

infected groups fed the non-probiotic diet than in the

groups fed the probiotic diet and both the control

groups respectively

Twenty-one days post infection with

pathogenic bacteria

The results of the haematology parameters

moni-tored for C gariepinus ¢ngerlings fed probiotic or

non-probiotic diets on the third week after infecting

with three di¡erent pathogenic bacteria are

pre-sented in Table 3 At 21 days post infection withpathogens, the levels of Hb, PCV, RBC, total serumprotein, Ca12, Mg12, Clserum glucose and Ig con-centrations in all treatments followed a pattern simi-lar to that recorded at 7 days post infection, althoughthe di¡erences in these parameters between the in-fected groups maintained on probiotic diets com-pared with the groups fed the non-probiotic dietswere clearer and more pronounced after 21 days

HistopathologyMonitoring of the histopathologic e¡ects in the inter-nal organs (liver and kidney) of C gariepinus revealedsevere damage to these organs in the infected groupsmaintained on the non-probiotic diet when the speci-mens were collected at 7 and 21 days post infectionwith the three pathogenic bacteria (Fig 1; images7

NPsx,7NPah and7NPsa and Fig 2; images21NPsx,21

NPah and21NPsa), compared with the very mild fects observed in the infected ¢sh fed the probioticdiet during the same period (Fig 1; images 7Psx,

ef-Table 3 Haematological parameters of clarias gariepinus at 21 days post infection and fed non-probiotic and probiotic diets

Results are expressed as mean  SD.

Mean values in the same row with di¡erent superscripts show signi¢cant di¡erence (P o0.05).

See Table 1 for details of abbreviations.

WBC, white blood cell, RBC, red blood cell; MCHC, mean corpuscular haemoglobin concentration; MCH, mean corpuscular bin; MCV, mean corpuscular volume.

haemoglo-L acidophilus as a probiotic in African cat¢sh M A Al-Dohail et al Aquaculture Research, 2011, 42, 196–209

r 2010 Universiti Sains Malaysia

Trang 33

B A

Figure 1 (a) Livers were stained with haematoxylin and eosin, and shown at  40., bar 200 mm; sampling collection at

7 days post infection with pathogenic bacteria of images:7NPc,7NPsx,7NPah,7NPsa,7Pc,7Psx,7Pah and7Psa Liver ture of non-probiotic control (7NPc) Note (A) blood vessel and (B) hepatocyte NPsx: ¢sh infected with Staphylococcusxylosus and fed the non-probiotic diet Note the acute hepatitis area (arrowheads).7NPah: ¢sh infected with Aeromonoshydrophila and fed the non-probiotic diet Note the acute necrosis area (arrowheads).7NPsa: ¢sh infected with Streptococ-cus agalactiae and fed the non-probiotic diet Note acute hepatitis with the necrosis area (arrowheads).7Pc: probiotic con-trol: (A) blood vessel and (B) hepatocyte.7Psx: ¢sh infected with S xylosus and fed the probiotic diet Note the mild hepatitisarea (arrowheads).7Pah: ¢sh infected with A hydrophila and fed the probiotic diet Note the mild necrosis area (arrow-heads).7Psa: ¢sh infected with S agalactiae and fed the probiotic diet Note mild hepatitis with the necrosis area (arrow-heads) (b) Livers were stained with haematoxylin and eosin, and shown at  40, bar 200 mm; sampling collection at 21days post infection with pathogenic bacteria of images:21NPc,21NPsx,21NPah,21NPsa,21Pc,21Psx,21Pah and21Psa Liverstructure of non-probiotic control (21NPc) Note (A) blood vessel and (B) hepatocyte.21NPsx: ¢sh infected with Staphylococ-cus xylosus and fed the non-probiotic diet Note the acute hepatitis area (arrowheads).21NPah: ¢sh infected with Aeromo-nos hydrophila and fed the non-probiotic diet Note very acute necrosis, swelling, vacuolation and the degeneration area(arrowheads).21NPsa: ¢sh infected with Streptococcus agalactiae and fed the non-probiotic diet Note very acute hepatitis,swelling and the degeneration area (arrowheads).21Pc: probiotic control (A) blood vessel and (B) hepatocyte.21Psx: ¢shinfected with S xylosus and fed the probiotic diet Note the mild hepatitis area (arrowheads).21Pah: ¢sh infected with A.hydrophila and fed the probiotic diet Note the mild necrosis area (arrowheads).21Psa: ¢sh infected with S agalactiae andfed the probiotic diet showed a normal area in hepatocyte (arrowheads)

Trang 34

Haematological characteristics have been studied in

many ¢sh species to determine the normal range,

and any variations from these ranges were indicative

of pathophysiology (Ranzani-paiva et al 2000) In the

present study, at 7 days post infection, all the blood

parameters, Ig, Mg21, Ca21, Cl, total serum protein

and cholesterol levels in the infected ¢sh fed the

pro-biotic diet were signi¢cantly higher (Po0.05) than

those in the infected groups maintained on the

non-probiotic diet, whereas ESR, WBC and glucose were

signi¢cantly higher in the infected ¢sh fed the

non-probiotic diet than in the other groups These

obser-vations, in general, could be because infected ¢sh fed

probiotic diets had a better capacity (and also lowerstressor levels) to resist infection compared with theother infected ¢sh fed the non-probiotic diets Thehigher ESR level noted in the present study could berelated to damage to RBCs, due to the poison released

by pathogenic bacteria during infection This tion is supported by Haney, Hursh, Mix & Winton(1992), Harikrishnan, Rani and Balasundaram(2003) and Scott and Rogers (1981), who all claimedthat the RBC swells during bacterial infection Itcould also be that the probiotic treatments weremuch better at resisting infection than the othergroups, due to induced immune response when Lac-tobacillus was used as a probiotic, as reported by Pa-nigrahia, Kirona, Puangkaewa, Kobayashib, Satohaand Sugitac (2005) and Nikoskelainena, Ouwehanda,Bylundb, Salminena and Lilius (2003) Results simi-lar to those reported in the current study have beenobtained with numerous other ¢sh species, where

L acidophilus as a probiotic in African cat¢sh M A Al-Dohail et al Aquaculture Research, 2011, 42, 196–209

r 2010 Universiti Sains Malaysia

Trang 35

the ESR values were shown to increase with bacterial

infection or toxins (Gabriel, Ezeri and Opabunmi

(2004)

For RBC indices, the mean range reported for

nor-mal MCHC is 32^36 g dL 1, 27^31pg cell 1for MCH

and 80^100mm 3 for MCV (Johnson et al 2002;

George & Parker 2003) The results obtained in the

present study were higher than these values and

were better in the probiotic treatments infected with

all the three pathogenic bacteria than those fed the

non-probiotic diet, possibly due to lower swelling

and hence reduced RBC damage (Scott & Rogers

1981; Haney et al 1992; Harikrishnan et al 2003) in

the probiotic-treated groups

Moreover, serum protein, cholesterol, Mg12, Ca12

and Clwere signi¢cantly decreased in infected ¢sh

fed either probiotics or non-probiotic diets in

compar-ison with both the probiotic and the non probiotic

controls In many ¢sh species, the blood chemistry

parameters were altered throughout the period of

se-vere bacterial infection as reported when rainbow

trout was infected with a combination of Aeromonas

and Streptococcus (Barham, Smith & Schoonbee

1980) and in tilapia infected with Edwardsiella tarda(Benli & Yildiz 2004) In addition, decreased choles-terol, Mg12, Ca12and Clcould be related to the bio-chemical requirements of organic and inorganiccompounds for growth and reproduction or loss ofelectrolytes (Benli & Yildiz 2004) due to the perme-ability of renal tubules (Barham et al 1980)

WBC in the infected ¢sh fed the non-probiotic dietwas signi¢cantly higher than that in the infected ¢shmaintained on the probiotic diet possibly due tothe induction of the non-speci¢c defence system(Panigrahia et al 2005) and/or increased phagocyto-sis and cytotoxic activity (Salinas, Cuesta, Esteban &Meseguer 2005; Picchietti, Mazzini, Taddei, Renna,Fausto, Mulero, Carnevali, Cresci & Abelli 2007) inthe infected ¢sh maintained on the probiotic diet Ser-

um glucose was signi¢cantly the highest in thegroups of ¢sh fed the non-probiotic diet than the othergroups as reported when tilapia was infected with Ed-wardsiella tarda (Benli & Yildiz 2004) andVibrio vulni-

¢cus or Streptococcus iniae (Chen, Wooster & Bowser2004) as well as in Anguilla anguilla infected with A.hydrophilus (Yildiz, Bekans, Benli & Akan 2005)

Figure 2 (a) Kidneys were stained with haematoxylin and eosin, and shown at  200, scale bar 50 mm; sampling tion at 7 days post infection with pathogenic bacteria of images:7NPc,7NPsx,7NPah,7NPsa,7Pc,7Psx,7Pah and7Psa.Kidney structure of non-probiotic control (7NPc) Note (A) glomerulus, (B) proximal convoluted tubules and (C) Bowman’scapsule.7NPsx: ¢sh infected with Staphylococcus xylosus and fed the non-probiotic diet Note glomerular capillaries areextremely dilated and Bowman’s capsule space is blocked (arrowheads).7NPah: ¢sh infected with Aeromonos hydrophilaand fed the non-probiotic diet Note glomerular capillaries are extremely dilated and Bowman’s capsule space is blocked(arrowheads) and glomerular capillaries are degenerated (star).7NPsa: ¢sh infected with Streptococcus agalactiae and fedthe non-probiotic diet Note glomerular capillaries are extremely dilated and Bowman’s capsule space is blocked (arrow-heads); necrosis in the epithelia cell of proximal convoluted tubules (star).7Pc: probiotic control Note (a) glomerulus, (b)rental tubules and (c) Bowman’s capsule.7Psx: ¢sh infected with S xylosus and fed the probiotic diet Note that the glo-merular capillaries are dilated and Bowman’s capsule space is blocked in some areas (arrowheads).7Pah: ¢sh infected with

collec-A hydrophila and fed the probiotic diet Note that the glomerular capillaries are dilated and Bowman’s capsule space isblocked at one part (arrowheads), and slight mesangial cell proliferation (star).7Psa: ¢sh infected with S agalactiae and fedthe probiotic diet Note that the glomerular capillaries are dilated and Bowman’s capsule space is blocked at one side (ar-rowheads) (b) Kidneys were stained with haematoxylin and eosin, and shown at  200., scale bar 50 mm; samplingcollection 21 days post infection with pathogenic bacteria of images:21NPc,21NPsx,21NPah,21NPsa,21Pc,21Psx,21Pah and

21Psa Kidney structure of non-probiotic control (21NPc) Note (A) glomerulus, (B) proximal convoluted tubules and (C)Bowman’s capsule.21NPsx: ¢sh infected with S xylosus and fed the non-probiotic diet Note glomerular capillaries areextremely dilated and Bowman’s capsule space is blocked (arrowheads), and mesangial cell proliferation and degenerated(star).21NPah: ¢sh infected with A hydrophila and fed the non-probiotic diet Note very acute necrosis and the degenera-tion area of architecture kidney structure (circle).21NPsa: ¢sh infected with S agalactiae and fed the non-probiotic diet.Note the expansion of space inside the Bowman’s capsule, shrinking of convoluted tubules (arrowheads) and contraction

of the glomerulus (star).21Pc: probiotic control Note (A) glomerulus, (B) rental tubules and (C) Bowman’s capsule.21Psx:

¢sh infected with S xylosus and fed the probiotic diet Note glomerulusis appears mostly normal (arrowheads), and withsome necrosis in the epithelia cell of proximal convoluted tubules (star).21Pah: ¢sh infected with A hydrophila and fed theprobiotic diet Note the expansion of space inside the Bowman’s capsule with contraction of the glomerulus (arrowheads),thickening of Bowman’s capsule basement membrane (arrowheads-A) and degenerated glomerulus (star).21Psa: ¢sh in-fected with S agalactiae and fed the probiotic diet Note that the glomerulus capillaries are very slight dilated and Bow-man’s capsule space is not blocked (arrowheads)

Trang 36

The increased glucose level in the non-probiotic

treatments observed in the present study could be

be-cause ¢sh require more energy to support all

meta-bolic processes under abnormal conditions

However, during emergencies such as the

stimula-tion of the immune defence system during bacterial

infection, the ¢sh could have also converted several

lipids into glucose, and this may explain the

de-creased cholesterol level observed in this study The

increase in blood glucose may also be attributed to a

slow metabolism or an increase in stress (Smith,

Hat-tingh & Burger 1976; Connors, Schneider, Genoway &

Barraclough 1978; Best, Eddy & Codd 2001)

The observation in the present study that total Ig

was signi¢cantly higher in infected ¢sh fed the

pro-biotic diet was expected and agrees with reports for

trout, Oncorhynchus mykiss, maintained on a

probio-tic-based diet containing Lactobacillus rhamnosus

strains (Nikoskelainena et al 2003; Panigrahia et al

2005)

At 21 days post infection, however, the WBC levels

in infected groups fed probiotic diets were decreased

and became similar to those in the probiotic and

non-probiotic control groups, whereas the level in ¢sh fed

the non-probiotic diet remained higher than in their

probiotic counterparts and both controls These

ob-servations could indicate that the immune system

was successful in controlling the bacterial infection

in the probiotic treatments, returning the WBC count

back to the normal condition This suggestion is also

supported by the similar pattern observed in the

other haematology parameters, which increased or

decreased from 7 days post infection and became

si-milar to the levels of the controls (normal) after 21

days Reports by many authors (Van Vuren, van der

Merwe & Du Preez 1994; Ranzani-paiva et al 2000;

Ezeri 2001; O’Neal & Weirich 2001; Omoregie &

Oye-banji 2002; Gabriel et al 2004) support this opinion,

whereby haematological deviations were indicative

of pathophysiology and health status

In addition to haematology parameters,

histo-pathology of the liver, the main organ for processing

and removal of toxic substances or transforming

them into harmless materials for excretion (Dutta,

Adhikari, Singh & Munshi 1993), was assessed The

liver also has the ability to accumulate xenobiotics,

and bio-transforms and transports them to the bile

for removal (Metcalfe 1998) In the present study, the

histopathology of the liver at 7 days post infection

with the three pathogenic bacteria revealed acute

he-patitis, with large necrosis areas appearing in the

non-probiotic treatments (Fig 1a) and subsequent

worsening conditions at 21 days post infection Thesechanges and damage to the hepatic cells could haveoccurred due to the poisons released by the patho-genic bacteria during infection and the long-term ac-tion of poisonous compounds and their metabolites

on the liver cells This suggestion concurs with theobservations of Bowser, Wooster, Chen and Mo(2004), who reported that hybrid striped bass had ne-crotic foci in the liver when it was infected with Aero-monas salmonicida, Mycobacterium sp and Vibrioanguillarum Stoskopf (1993) also reported thatvacuolar degeneration in the cytoplasm of liver cellswas observed when Atlantic salmon was infectedwith vibirios

In the kidney, capillaries and mesangial cells form important functions in the normal ¢ltration ofblood In the present study, histopathology examina-tion of the kidney at 7 days post infection revealedconditions such as extreme dilatation of the glomer-ular capillaries, blocked Bowman’s capsule space andnecrosis in epithelial cells of the proximal convolutedtubules, when ¢sh were raised on the non-probioticdiet The situation worsened at 21 days post infection,evident by very acute necrosis and degeneration ofareas of the kidney architectural structure, expan-sion of spaces inside the Bowman’s capsule andshrinking of the convoluted tubules (refer to plates

per-in Fig 2b) Kidney cell damage due to pathogenic terial infection such as multiple necrotic foci in hy-brid striped bass, Morone chrysops Morone saxatilis(Bowser et al 2004), congested kidneys in silver pom-fret Pampus argenteus, Euphrasen (Duremdez,Al-marzouk, Qasem, Al-harbi & Gharabally 2004),haemorrhaging and necrosis in turbot Scophthalmusmaximus (Bj˛rnsdo¤ttir, Gudmundsdo¤ttir, Bambir &Gudmundsdo¤ttir 2005; Padr, Zarza, Dopazo, Cuadra-

bac-do & Crespo 2006), lesions, enlarged kidneys andnecrosis in rainbow trout (Hirvel-Koski, Pohjanvir-

ta, Koski & Sukura 2006; Altinok, Balta, Capkin &Kayis 2007) and embolism in capillary and lymphaticvessels in Seriola dumerili (Hagiwara, Takano, Nogu-chi & Narita 2009) have also been reported pre-viously

In conclusion, the results of the present studyclearly demonstrate that the infected groups of ¢shmaintained on probiotic diets yielded signi¢cantlybetter haematology parameters and histopathologythan the infected groups fed with the non-probioticdiets This supports the bene¢cial e¡ects of probioticbacteria (L acidophilus) as a bio-control agent in as-sisting ¢sh (C gariepinus in this case) to better with-stand bacterial infections (especially S agalactiae,

L acidophilus as a probiotic in African cat¢sh M A Al-Dohail et al Aquaculture Research, 2011, 42, 196–209

r 2010 Universiti Sains Malaysia

Trang 37

S xylosus and A hydrophila gr.2) respectively It is

known that the e¡ect of a bacterial toxin depends on

the pathogenic bacteria species from which it is

de-rived Therefore, the results of this study showed that

S agalactiae was the highest inhibited by probiotic

bacteria (L acidophilus), whereas A hydrophila gr.2

was the lowest inhibited in both the in vitro and the

in vivo challenge L acidophilus is therefore useful as a

bio-control agent against bacterial infections in C

gariepinus

Although the market value of cat¢sh is relatively

low per kilo, compared with other species, the

con-sumption, on the other hand, is quite high in Asian

markets Therefore, the use of single probiotic

bacter-ia such as L acidophilus rather than a cocktail of

bac-teria (as has been suggested by other authors) for the

treatment of cat¢sh diseases is more a¡ordable,

trans-lating into higher pro¢ts in the aquaculture industry

Acknowledgments

This study was funded by the Malaysian Government

Probiotic Project number 305/PBIOLOGI/613514 and

supported by the USM fellowship scheme The ¢rst

author was ¢nancially supported by the

Hadhram-out University of Science and Technology (Yemen)

The authors will like to express their gratitude to the

National Fish Health Research Center Penang,

Ma-laysia, for the supply of the pathogenic bacteria The

authors also express their gratitude to Professor

Abdlhakim Alrawi of Baghdad University, who was

a visiting Professor to USM, for his technical

assis-tance during pathohistology analysis and Dr Saeed

Alfadly of Ministry of health, Hadhramount, Yemen,

for his technical assistance during the haematology/

immunology analysis

References

Abraham T.J., Mondal S & Babu C.S (2008) E¡ect of

com-mercial aquaculture probiotic and ¢sh gut antagonistic

bacterial £ora on the growth and disease resistance of

or-namental ¢shes Carassius auratus and Xiphophorus helleri.

E.U Journal of Fisheries and Aquatic Sciences 25, 27^30.

Adeyemo O.K (2007) Haematological pro¢le of Clarias

garie-pinus (Burchell, 1822) exposed to lead Turkish Journal of

Fisheries and Aquatic Sciences 7, 163^169.

Al-Dohail M.A., Hashim R & Aliyu-Paiko M (2008)

Chal-lenges of Probiotic, Lactobacillus acidophilus Against

Pathogenic Bacteria in Vitro International Conference

Marine Organisms and their Biomedical Potentials:

Expecta-tions and RealizaExpecta-tions, Universiti Malaysia Terengganu,

18^20 May 2008.

Al-Dohail M.A., Hashim R & Aliyu-Paiko M (2009) E¡ects

of the probiotic, Lactobacillus acidophilus, on the growth performance, haematology parameters and immunoglo- bulin concentration in African Cat¢sh (Clarias gariepinus, Burchell 1822) ¢ngerling Aquaculture Research 40, 1642^ 1652.

Altinok I., Balta F., Capkin E & Kayis S (2007) Disease of rainbow trout caused by Pseudomonas luteola Aquacul- ture 273, 393^397.

Amar E.C., Kiron V., Satoh S., Okamoto N & Watanabe T (2000) E¡ect of dietary h carotene on the immune re- sponse of rainbow trout Oncorhynchus mykiss Fisheries Science 66, 1068^1075.

Austin B., Stuckey L., Robertson P., E¡endi I & Gri⁄th D (1995) A probiotic strain of Vibrio alginolyticus e¡ective in reducing diseases caused by Aeromonas salmonicida,Vi- brio anguillarum and Vibrio ordalii Journal of Fish Diseases

18, 93^96.

Barham W.T., Smith G.L & Schoonbee H.J (1980) The haematological assessment of bacterial infection in rainbow trout, Salmo gairdneri Journal of Fish Biology

17, 275^281.

Benli A.C.K & Yildiz H.Y (2004) Blood parameters in Nile tilapia (Oreochromis niloticus L.) spontaneously infected with Edwardsiella tarda Aquaculture Research 35, 1388^ 1390.

Best J.H., Eddy F.B & Codd G.H (2001) E¡ects of puri¢ed crocystin-LR and cell extracts of Microcystis strains PCC

mi-7813 and function in brown trout (salmo trutta) alevins Fish Physiology and Biochemistry 24, 171^178.

Bj˛rnsdo¤ttir B., Gudmundsdo¤ttir S., Bambir S.H & mundsdo¤ttir B.K (2005) Experimental infection of turbot, Scophthalmus maximus (L.), by Aeromonas salmonicida subsp achromogenes and evaluation of cross protection in- duced by a furunculosis vaccine Journal of Fish Diseases

Gud-28, 181^188.

Blaxhall P.C & Daisley K.W (1973) Routine haematological methods for use with ¢sh blood Journal of Fish Biology 5, 771^781.

BongaW.S.E (1997) The stress response in ¢sh Physiological Reviews 77, 591^625.

Bowser P.R.,Wooster G.A., Chen C.-Y & Mo R.S (2004) microbic infection of hybrid striped bass (Morone chry- sops  Morone saxatilis) with three bacterial pathogens:

Poly-a cPoly-ase report JournPoly-al of Fish DisePoly-ases 27, 123^127 Britton C.J (1963) Disorders of the Blood, 9th edn I A Churchill, London, UK.

Cech J.J Jr, Bartholow S.D., Young P.S & Hopkins T.E (1996) Striped bass exercise and handling stress in freshwater: physiological responses to recovery environment Trans- actions of the American Fisheries Society 125, 308^320 Chen C., Wooster G.A & Bowser P.R (2004) Comparative blood chemistry and histopathology of tilapia infected with Vibrio vulni¢cus or Streptococcus iniae or exposed to carbon tetrachloride, gentamicin or copper sulphate Aquaculture 239, 421^443.

Trang 38

Choi K.H & Dobbs F.C (1999) Comparison of two kinds of

Biolog microplates (GN and Eco) in their ability to

distin-guish among aquatic microbial communities Journal of

Microbiological Methods 36, 203^213.

Connors T.T., Schneider M.J., Genoway R.G & Barraclough

S.A (1978) E¡ect of acclimation temperature on plasma

levels of glucose and lactate in rainbow trout, Salmo

gaird-neri J Exp Zool 206, 443^449.

Duremdez R., Al-Marzouk A., Qasem J.A., Al-Harbi A &

Gharabally H (2004) Isolation of Streptococcus agalactiae

from cultured silver pomfret, Pampus argenteus

(Euphra-sen), in Kuwait Journal of Fish Diseases 27, 307^310.

Dutta H.M., Adhikari N.K., Singh P.K & Munshi J.S (1993)

Histopathological changes induced by malathion in the

liver of a freshwater cat¢sh, Heteropneustes fossilis (Bloch).

Bulletin of Environmental Contamination andToxicology 51,

895^900.

EWOS (2000) Determination of the packed cell volume

(PCV) of erythrocytes, Ref No M326 and Red cell

count-ing method, Ref No 328, EWOS, Technology

Centre-Laboratory Method Norway.

Ezeri G.N.O (2001) Haematological response of Clarias

garie-pinus to bacteria infection and prophylactic treatment

with antibiotics Journal of Aquaculture Science 16, 22^24.

Gabriel U.U., Alagoa J.K & Allison M.E (2001) E¡ects of

dis-persed crude oil water dispersion on the haemaglobin and

haematocrit of Clarias gariepinus Journal of Aquaculture

Science and Environment Management 5, 9^11.

Gabriel U.U., Ezeri G.N.O & Opabunmi O.O (2004) In£uence

of sex, source, health status and acclimation on the

hae-matology of Clarias gariepinus (Burchell 1822) African

Journal of Biotechnology 3, 463^467.

George G.B & Parker K (2003) Understanding the complete

blood count with di¡erential Journal of Perianesthesia

Nursing 19, 96^114.

Hagiwara H., Takano R., Noguchi M & Narita M (2009)

Study of the Lesions Induced in Seriola dumerili by

Intra-dermal or Intraperitoneal Injection of Streptococcus

dys-galactiae Journal of Comparative Pathology 140, 25^30.

Haney D.C., Hursh D.A., Mix M.C & Winton J.R (1992)

Phy-siological and hematological changes in chum salmon

ar-ti¢cially infected with erythrocytic necrosis virus Journal

of Aquatic Animal Health 4, 48^57.

Harikrishnan R., Rani M.N & Balasundaram C (2003)

He-matological and biochemical parameters in common

carp, Cyprinus carpio, following herbal treatment

forAero-monas hydrophila infection Aquaculture 221, 41^50.

Hirvel-Koski V., Pohjanvirta T., Koski P & Sukura A (2006)

A typical growth of Renibacterium salmoninarum in

sub-clinical infections Journal of Fish Diseases 29, 21^29.

Johnson C.W., Timmons D.L & Hall P.E (2002) Essential

Laboratory Mathematics: Concepts and Applications for the

Chemical and Clinical Laboratory Technician Cengage

Learning, Florence, KY, USA, 268pp.

Klaenhammer T.R & Russell W.M (2000) Lactobacillus

acid-ophilus In: Encyclopedia of Food Microbiology Volume 2

(ed by R.K Robinson, C.A Batt & P.D Patel), pp 1151–1157 Academic Press, San Diego, CA.

Koch A (1994) Growth measurement In: Methods for

Gener-al and Molecular Bacteriology (ed by P Gerhardt, R ray, W Wood & N Krieg), pp 249^277 American Society for Microbiology,Washington, DC, USA.

Mur-Lim C & Webster C.D (2001) Nutrition and Fish Health worth Press Food Products Presss, New York, NY, USA Metcalfe C.D (1998) Toxicopathic responses to organic com- pounds In: Fish Diseases and Disorders,Volume 2: Non-in- fectious Disorders (ed by J.F Leatherland & P.T.K Woo), pp 133–166 CABI Publishing, Oxon, UK Nikoskelainena S., Ouwehanda C.A., Bylundb G., Salminena

Ha-S & Lilius E (2003) Immune enhancement in rainbow trout (Oncorhynchus mykiss) by potential probiotic bacter-

ia (Lactobacillus rhamnosus) Fish and Shell¢sh Immunology

15, 443^452.

Omoregie E & Oyebanji (2002) Oxytetracycline-induced blood disorder in Nile tilapia, Oreochromis niloticus Jour- nal of World Aquaculture Society 33, 377^382.

O’Neal C.C & Weirich C.R (2001) E¡ects of low level salinity

on product and haematology parameters of channel

cat-¢sh, Ictalurus punctatus reared in multi-crop ponds In: Book of abstract Aquaculture 2001 International Triennial Conference ofWorld Aquaculture Society, Disney Coronado Springs Resort, Lake Buena Vista, FL, 21–25 January 2001.

Onusiriuka B.C & Ufodike E.B.C (2000) E¡ects of sublethal concentrations of Akee apple, Bligha sapida and sausage plant Kigella africana on tissue chemistry of the Afr cat¢sh Clarias gariepinus Journal of Aquaculture Science 5, 47^49 Padr F., Zarza C., Dopazo L., Cuadrado M & Crespo S (2006) Pathology of Edwardsiella tarda infection in turbot, Scophthalmus maximus (L.) Journal of Fish Diseases 29, 87^94.

Panigrahia A., Kirona V., Puangkaewa J., Kobayashib T., toha S & Sugitac H (2005) The viability of probiotic bac- teria as a factor in£uencing the immune response in rainbow trout Oncorhynchus mykiss Aquaculture 243, 241^254.

Sa-Picchietti S., Mazzini M.,Taddei A.R., Renna R., Fausto A.M., Mulero V., Carnevali O., Cresci A & Abelli L (2007) E¡ects of administration of probiotic strains on GALT of larval gilthead seabream: immunohistochemical and ultrastructural studies Fish and Shell¢sh Immunology 22, 57^67.

Pimpao C.T., Zampronio A.R & Silva de Assis H.C (2007) fects of deltamethrin on hematological parameters and enzymatic activity in Ancistrus multispinis (Pisces,Teleos- tei) Pesticide Biochemistry and Physiology 88, 122^127 Ranzani-Paiva M.J.T., Ishikawa C.M., das Eiras A.A & Feli- zardo N.N (2000) Haemotological analysis of ‘chara’ Pseu- doplatystoma fasciatum in captivity In: Aqua 2000: Responsible Aquaculture in the New Millennium, Nice, France, 2–6 May 2000 Special publication 28 Eur- opean Aquaculture Society, 590pp.

Ef-L acidophilus as a probiotic in African cat¢sh M A Al-Dohail et al Aquaculture Research, 2011, 42, 196–209

r 2010 Universiti Sains Malaysia

Trang 39

Rengpipat S., Rukpratanporn S., Piyatiratitivorakul S &

Me-nasaveta P (2000) Immunity enhancement in black tiger

shrimp (Penaeus monodon) by a probiont bacterium

(Bacil-lus S11) Aquaculture 191, 271^288.

Salinas I., Cuesta A., Esteban M.A & Meseguer J (2005)

Dietary administration of Lactobacillus delbrˇ eckii and

Bacillus subtilis, single or combined, on gilthead

seab-ream cellular innate responses Fish and Shell¢sh

Immu-nology 19, 67^77.

Salminen S., Ouwehand A., Benno Y & Lee Y.K (1999)

Pro-biotics: how should they be de¢ned? Trends in Food Science

and Technology 10, 107^110.

Schperclaus W., Kulow H & Schreckenbach K (1992)

Fish Diseases, Vol 1, 5th edn A A Balkema/Protterdam,

Rotterdam, 594pp.

Schillinger U (1999) Isolation and identi¢cation of

lactoba-cilli from novel-type probiotic and mild yoghurts and

their stability during refrigerated storage International

Journal of Food Microbiology 47,79^87.

Scott A.L & Rogers W.A (1981) Hematological e¡ects of

pro-longed sublethal hypoxia on channel cat¢sh Ictalurus

punctatus (Ra¢nesque) Journal of Fish Biology18,591^601.

Siwicki A.K & Anderson D.P (1993) Non-speci¢c defence

mechanisms assay in ¢sh: II Potential killing activity

of neutrophils and macrophages, lysozyme activity in

serum and organs and total immunoglobulin level in

serum In: Disease Diagnosis and Prevention Methods.

FAO-Project GCP/INT/JPA, pp 105–112 IFI, Olsztyn,

Poland.

Smith G.L., Hattingh J & Burger A.P (1976) Haematological

assessment of the e¡ect of the anaesthetic MS 222 in

natural and neutralized form in three freshwater ¢sh species; Interspecies di¡erences Journal of Fish Biology

15, 633^643.

Stosik M & Szenfeld J (1996) Use of CERBIO probiotic in carp nutrition MedycynaWeterynaryjna 52, 467^469 Stoskopf M.K (1993) Fish Medicine.W B Saunders, London,

UK, 882pp.

Strm E & Ring E (1993) Changes in bacterial £ora in veloping cod, Gadus morhua L., larvae after inoculation of Lactobacillus plantarum in the water In: Physiological and Biochemical Aspects of Fish Larval Development (ed by B Walther & H.J Fyhn), pp 226^228 University of Bergen, Bergen, Norway.

de-Thrall M.A (2004) Veterinary Haematology and Clinical Chemistry, Vol 19, pp 277^289 Williams and Wilkins cap, Philadelphia, PA, USA.

Van Vuren J.H.J., van der Merwe M & Du Preez H.H (1994) The e¡ect of copper on the on the blood chemistry of Clar- ias gariepinus (Claridae) Ecotoxicology and Environmental Safety 29, 187^199.

Verschuere L., Rombaut G., Sorgeloos P & Verstraete W (2000) Probiotic bacteria as biological control agents in aquaculture Microbiology and Molecular Biology Reviews

64, 655^671.

Vine N.G (2004) Towards the development of a protocol for the selection of probiotics in marine ¢sh larviculture PhD thesis, Rhodes University.

Yildiz H.Y., Bekans S., Benli A.C.K & Akan M (2005) Some Blood parameters in the Eel Anguilla anguilla Sponta- neously Infected wtith Aeromonas hydrophilla Israel Jour- nal of Veterinary Medicine 60, 91^92.

Trang 40

Microsatellite–centromere mapping in walking catfish

gynogenetic diploids

Supawadee Poompuang1& Chantapim Sukkorntong2

1 Department of Aquaculture, Faculty of Fisheries, Kasetsart University, Bangkok,Thailand

2 Center for Agricultural Biotechnology, Kasetsart University, Nakorn Pathom,Thailand

Correspondence: S Poompuang, Department of Aquaculture, Faculty of Fisheries, Kasetsart University, 50 Paholyothin Road, Bangkok 10900,Thailand E-mail: supawadee.p@ku.ac.th

Abstract

Thirty new microsatellite loci were isolated from a

mi-crosatellite-enriched genomic library of walking

cat-¢sh Clarias macrocephalus The CA motif was the most

abundant, although several other motifs were also

isolated Two gynogenetic diploid families, each

con-sisting of a female and 50 o¡spring, were genotyped

at 56 microsatellite loci, including 30 loci developed

in the present study and 26 loci reported in the

previous study Overall, 33 anonymous microsatellites

derived from genomic library and11 EST-linked

micro-satellites were mapped to their centromeres High

levels of chiasma interference were apparent in the

walking cat¢sh genome as indicated by an average

fre-quency of second-division segregation (y) of 0.643

0.248 Twenty-six loci (59%) showed a high

microsa-tellite^centromere recombination, with a frequency

40.67, and three loci had recombination frequencies

40.9 This study demonstrated that gene^centromere

mapping provided a rapid method for the expansion of

the initial linkage map of walking cat¢sh

Keywords: Clarias macrocephalus, centromere, gene

mapping, gynogenetics, microsatellite

Introduction

Gene^centromere mapping is the technique used to

determine the position of a centromere on the genetic

linkage map In species with a small number of

avail-able genetic markers, it is di⁄cult to construct a

ge-netic map, but the distances of these markers from

their centromere can be estimated (Thorgaard,

Allen-dorf & Knudsen 1983; AllenAllen-dorf, Seeb, Knudsen,Thorgaard & Leary 1986) The principle of gene^cen-tromere mapping is based on half-tetrad analysis, forwhich two products of second meiotic division can berecovered by inhibiting the release of the secondpolar body (Zhao & Speed 1998) In ¢sh and shell¢sh,two products of second meiotic division ^ includingthe second polar bodies and the secondary oocytes ^can be recovered using gynogenesis techniques(Thorgaard et al 1983; Streisinger, Singer, Walker,Knauber & Dower 1986)

Walking cat¢sh (Clarias macrocephalus) is an portant food ¢sh species in Thailand.Walking cat¢shhas superior £esh quality and commands high retailprices Although culture of walking cat¢sh has beenwell established in Thailand for over 30 years, the po-tential of the industry is limited by the species’ slowgrowth rate and disease susceptibility (Na-Nakorn,Rangsin & Witchasunkul 1993) A molecular ap-proach could be applied for implementing geneticmarker-assisted selective breeding of walking cat¢sh.The application of this method, however, requires in-formation on its genome including a high-resolutiongenetic map A ¢rst generation of genetic map based

im-on AFLP markers has been generated for walkingcat¢sh, but the marker density was low (Poompuang

& Na-Nakorn 2004) Because the localization ofquantitative trait loci (QTLs) is essential for the appli-cation of marker-assisted selection, QTL mappingstrategies would bene¢t from a higher gene density

on the map (Poompuang & Hallerman 1997) over, the genetic map consisted of 31 linkage groups,whereas walking cat¢sh has 27 chromosomes, indi-cating that there are at least four gaps in the map ToAquaculture Research, 2011, 42, 210^220 doi:10.1111/j.1365-2109.2010.02613.x

More-r 2010 The Authors

Ngày đăng: 19/07/2016, 19:33

TỪ KHÓA LIÊN QUAN

TÀI LIỆU CÙNG NGƯỜI DÙNG

  • Đang cập nhật ...

TÀI LIỆU LIÊN QUAN