Effects of genetic drift and selection at Effects of genetic drift and selection at larval stage resulting from hatchery larval stage resulting from hatchery

practices in the Pacific oyster (practices in the Pacific oyster (Crassostrea Crassostrea gigasgigas))

N. Taris, C. Sauvage, B. N. Taris, C. Sauvage, B. ErnandeErnande*, F. Batista*, F. Batista## & & P. BoudryP. BoudryLaboratoire de GLaboratoire de Géénnéétique et Pathologie, Ifremer, La Tremblade, Francetique et Pathologie, Ifremer, La Tremblade, France**Laboratoire Ressources Halieutiques, Ifremer, Laboratoire Ressources Halieutiques, Ifremer, PortPort--enen--BessinBessin, France, France

## INIAP/IPIMAR, INIAP/IPIMAR, Olhão, + ICBAS, Universidade do Porto, PortugalOlhão, + ICBAS, Universidade do Porto, Portugal

Crassostrea gigas life cycle

Williams 1975

Stages Survival

++++++++++++++

+

+++++++++

+++++

The “elmThe “elm--oyster model”oyster model”

High fecundity and high mortality at early stages

Can specific rearing practices (culling) and/or environmental conditions (high temperature) lead to a

specific genetic adaptation in C. gigas larvae ?

Which consequences of such a life history strategy for hatchery-based aquaculture production ?

☺ Few genitors needed for massive production of juveniles☺ Culling (i.e. size selection)

Low effective population size (Hedgecock et al., 1992)Risks of rapid loss of genetic variability and inbreeding in

closed populations

(Ernande et al., Journal of Evolutionary Biology 2003

Larval traits Metamorphic traits

Survival

Growth

Developmentrate

Size atsettlement

Sucess at metamoprphosis

Spat weight

Growth

Weight after settlement

Survival

Post-metamorphic traits

H2 significantly ≠ 0

Genetic correlationsignificantly positive

Genetic correlationsignificantly negative

Genetic variability of early life traits in C. gigas

Technical constrains often lead to limit the number of families and to rear them in a single environment

Mixed-family approach

Set of PCR- multiplexed markers allowing efficient parental assignment of larvae

125 150 175

female

male

offspring

- More families- More homogeneous rearing conditions among families- Different environments (G x E ?)

(Taris et al., Aquaculture Research 2005)

X

1. Effect of culling

Full factorial cross with equal gametic contribution within each sex

10

3

Which phenotypic and genetic consequences ?

6 tanks

Progressive culling

(day 4 - day 15)

- 50%Control

3 tanks 3 tanks

Mix

culled population control

1.1 Phenotypic effect of culling 50% of the (smallest) larvae

Limited effect on yield:-30 % of ready- to- settle larvae (higher survival of fast growing larvae)-15 % of spat (higher settlement success of fast growing larvae)

Coefficient of variation of larval length

0

2

4

6

8

10

12

14

16

1 3 6 8 10 13 15 17 20Days after fertilization

X 2

Number of pediveliger larvae

0

10 000

20 000

30 000

40 000

50 000

60 000

70 000

20 21 22 23 24 25 26 27 28 29 30 31Days after fertilization

Age at settlement

males

0

10

20

30

40

50

M7 M9 M1M10 M4 M5 M2 M3 M6 M8

D20

B20 21 22 23 24 25 26 27 28 29 30 31

Sampling

1.2 Genetic effect of culling

Sampling

Sampling

0

10

20

30

40

50

M7

M9

M1

M10 M4

M5

M2

M3

M6

M8

0

10

20

30

40

50

M7 M1 M4 M2 M6

D25

D28

8.26.3

Ne =

15.9Ne =

12.315.2Ne =

The effect of culling on genetic diversity is mediated through its effects on the timing of settlement

(Taris et al., JEMBE 2006)

2. Effect of temperature

Estimation of hatching rate at day 1X

Full factorial cross with equal gametic contribution within each sex

12

4M

ix

6 tanks

“Early” and “late” settlement cohorts

26°C 20°C

Individual measurements of 22-day old larvae prior to genotyping

3 tanks3 tanks

Which phenotypic and genetic consequences ?

2.1. « G x E » interaction on larval growth

210

220

230

240

250

260

270

20°C 26°C

Mean HS larval diameter (day 22)

h²(ns) 0,007 ± 0,007 <<< 0,117 ± 0,019

-10

-5

0

5

10

15

20

25

30

2 40 55 58 71 74 89 90 168 179 180 199

(No-

Na)

/100

Significantly different contributions between early and late cohorts reared at 20°C (26°C: similar result)

Late cohort 20°C

Early cohort 20°C

-10

-5

0

5

10

15

20

25

30

2 40 55 58 71 74 89 90 168 179 180 199

Early cohort 20°C

Early cohort 26°C

(No-

Na)

/100

Significantly different contributions between early cohorts reared at 20°C and 26°C (late cohorts: similar result)

2.2. Paternal contributions in spat (day 80)

Temperature significantly affects the genetic composition of the population and its growth (G x E)

LarvalLarval rearingrearing::24°C24°Cno no cullingculling3 3 replicatedreplicated tanks / tanks / progenyprogeny

Oysters from a commercial hatchery

broodstock following 7 generations of closed hatchery matings with high culling and high

temperature

Oysters froma French natural bed

WWildildHHatcheryatchery HHxxWW WWxxHH

3. Comparison of “domesticated” and “wild” larvae

Microsatellite markers :Mean nb. Of allele / locus 10 < 34Observed heterozygosity 0.66 < 0.86Expected heterozygosity 0.77 < 0.96

HatcheryH x WW x HWild

3.1. Larval growth, survival and settlement

89

113

138

150

175

199

230

85

105

125

145

165

185

205

225

245

265

3 6 8 10 13 15 17

days after fertilization

diam

eter

(µm

)

89

113

138

150

175

199

230

85

105

125

145

165

185

205

225

245

265

3 6 8 10 13 15 17

days after fertilization

diam

eter

(µm

)

GrowthGrowth (ns)(ns)

42

57

62

51

30

40

50

60

70

80

90

100

3 6 8 10 13 15 17 20days after fertilization

% s

urvi

val

42

57

62

51

30

40

50

60

70

80

90

100

3 6 8 10 13 15 17 20days after fertilization

% s

urvi

val

SurvivalSurvival ****

0

5

10

15

20

25

30

3 6 8 10 13 15 17

Days after fertilization

coef

ficie

nt o

f var

iatio

n

0

5

10

15

20

25

30

3 6 8 10 13 15 17

Days after fertilization

coef

ficie

nt o

f var

iatio

n

Variance in Variance in GrowthGrowth **

Settlement success : Settlement success : HatcheryHatchery > > HxWHxW > Wild > > Wild > WxHWxH(%) (%) 90,790,7 78,1 72,3 78,1 72,3

68 768 7

Timing to Timing to settlementsettlement **

0

10 000

20 000

30 000

40 000

50 000

60 000

70 000

80 000

90 000

22 24 26 28 30 32

Days post fertilization

Rea

dy to

se

tle la

rvae

0102030405060708090

100

100 115 130 145 160 175 190 205 220 235 250 265 280 295

Wild progeny

0102030405060708090

100

100 115 130 145 160 175 190 205 220 235 250 265 280 295

Hatchery progeny

Distribution oflarval length atDay 15

3.4. Within progeny variation for larval size

205 µm 225 µm **

Response to selection due to culling ?

hh²² = 0.16 (= 0.16 (DDéégremontgremont, 2003, 2003) ) S = 20S = 20µµm (m (Taris et al., 2006Taris et al., 2006))

∆∆ µ µ ~20~20µµm m overover 7 7 generationsgenerations+ + earlierearlier settlementsettlement+ + higherhigher settlementsettlement successsuccess

Inbredlarvae ?

0,012 ± 0,001 0,068 ± 0,005Pairwise relatednessin the broodstocks

HighHigh geneticgenetic loadload ((LauneyLauney andand HedgecockHedgecock, , GeneticsGenetics 2001)2001)

x 6

4. Conclusions

Methodology

- As individual tagging is impossible at early life stages, marker-based parentage analysis of mixed families represents an efficient way to study the genetics of larval traits in oysters.

Unintentional selection at larval stage in hatcheries

- Significant differences are observed between families, confirming the existence of genetic variation for several traits.

- Temperature influences the expression of genetic variability for growth and survival. It therefore is likely to increase the genetic effect of culling.

- Intensive rearing practices can lead to the selection of faster growing larvae and higher settlement rates, despite inbreed depression.

AcknowledgmentsAcknowledgments::Bureau des Ressources GénétiquesBureau des Ressources GénétiquesMinistère de l’Écologie et du Développement DurableMinistère de l’Écologie et du Développement DurableConseil Général de CharenteConseil Général de Charente-- MaritimeMaritime

Oyster ponds along the Seudre estuary, Marennes-Oléron Bay, France