-
i
Impacts of traditional husbandry practices on exploitable levels
of
genetic diversity in cultured ‘Tra’ catfish (Pangasianodon
hypophthalmus) in the Mekong Delta, Vietnam’
Bui, Thi Lien Ha
B.Sc, University of Natural Sciences, Vietnam
Biogeosciences
Faculty of Science and Technology FaST
Queensland University of Technology
Brisbane, Australia
Submitted in fulfillment of the requirement of the degree of
Master of Science
September 2011
-
ii
Abstract
Sutchi catfish (Pangasianodon hypophthalmus) – known more
universally by the
Vietnamese name ‘Tra’ is an economically important freshwater
fish in the Mekong
Delta in Vietnam that constitutes an important food resource.
Artificial propagation
technology for Tra catfish has only recently been developed
along the main
branches of the Mekong River where more than 60% of the local
human population
participate in fishing or aquaculture. Extensive support for
catfish culture in general,
and that of Tra (P. hypophthalmus) in particular, has been
provided by the
Vietnamese government to increase both the scale of production
and to develop
international export markets. In 2006, total Vietnamese catfish
exports reached
approximately 286,602 metric tons (MT) and were valued at 736.87
$M with a
number of large new export destinations being developed. Total
value of production
from catfish culture has been predicted to increase to
approximately USD 1 billion
by 2020. While freshwater catfish culture in Vietnam has a
promising future,
concerns have been raised about long-term quality of fry and the
effectiveness of
current brood stock management practices, issues that have been
largely neglected
to date.
In this study, four DNA markers (microsatellite loci: CB4, CB7,
CB12 and CB13) that
were developed specifically for Tra (P. hypophthalmus) in an
earlier study were
applied to examine the genetic quality of artificially
propagated Tra fry in the Mekong
Delta in Vietnam. The goals of the study were to assess: (i) how
well available levels
of genetic variation in Tra brood stock used for artificial
propagation in the Mekong
Delta of Vietnam (breeders from three private hatcheries and
Research Institute of
Aquaculture No2 (RIA2) founders) has been conserved; and (ii)
whether or not
genetic diversity had declined significantly over time in a
stock improvement
program for Tra catfish at RIA2. A secondary issue addressed was
how genetic
markers could best be used to assist industry development. DNA
was extracted
from fins of catfish collected from the two main branches of the
Mekong River inf
Vietnam, three private hatcheries and samples from the Tra
improvement program
at RIA2.
-
iii
Study outcomes:
i) Genetic diversity estimates for Tra brood stock samples were
similar to, and
slightly higher than, wild reference samples. In addition, the
relative contribution by
breeders to fry in commercial private hatcheries strongly
suggest that the true Ne is
likely to be significantly less than the breeder numbers used;
ii) in a stock
improvement program for Tra catfish at RIA2, no significant
differences were
detected in gene frequencies among generations (FST=0.021,
P=0.036>0.002 after
Bonferroni correction); and only small differences were observed
in alleles
frequencies among sample populations.
To date, genetic markers have not been applied in the Tra
catfish industry, but in the
current project they were used to evaluate the levels of genetic
variation in the Tra
catfish selective breeding program at RIA2 and to undertake
genetic correlations
between genetic marker and trait variation. While no
associations were detected
using only four loci, they analysis provided training in the
practical applications of
the use of molecular markers in aquaculture in general, and in
Tra culture, in
particular.
-
iv
TABLE OF CONTENTS
Chapter 1. GENERAL INTRODUCTION
.............................................................................
1 1.1. Status of world fisheries and aquaculture
....................................................... 1 1.2. Role
of genetics and population genetics in aquaculture
................................ 4 1.3. Genetic improvement -
moving from wild to improved culture strains ............. 8 1.4.
The Tra catfish culture industry in Vietnam
................................................... 10 1.5. Tra fry
cohort production in private hatchery practices in the Mekong Delta
. 12 1.6. Genetic marker applications in aquaculture
.................................................. 15 1.7. Specific
aims of the current study
.................................................................
18
Chapter 2. MATERIALS AND METHODS
........................................................................
19 2.1. The Tra catfish selective breeding program at RIA2
..................................... 19 2.2. Propagation of fry at
3 private hatcheries in the Mekong Delta ..................... 20
2.3. Sample collection
..........................................................................................
20 2.4. Genomic DNA extraction
...............................................................................
22 2.5. Genotyping procedures
.................................................................................
22 2.6. Data
analysis.................................................................................................
23
Chapter 3. RESULTS
........................................................................................................
25 3.1. Part A - Characterisation of genetic variation in cultured
and wild populations of Tra catfish
........................................................................................................
25
3.1.1 Pair wise linkage disequilibrium
.............................................................. 26
3.1.2. Conformation to HWE results
................................................................ 26
3.1.3 AMOVA analysis of hierarchical differentiation within and
among populations
.......................................................................................................
31 3.1.4. Genetic characterization of sampled Tra catfish culture
stocks .............. 32
3.2. Part B - Assessment of the relative contribution by
breeders to fry cohorts in three private hatcheries in the Mekong
Delta .......................................................
33
3.2.1. Estimation based on the number of males and females in
the brood stock
..........................................................................................................................
34 3.2.2. Estimated number of males and females actually
contributing to offspring, based on the results of pedigree
analyses using CERVUS v3.0 software. ....... 35
Chapter 4. EVALUATION OF THE LEVELS OF GENETIC VARIATION IN THE
TRA CATFISH SELECTIVE BREEDING PROGRAM AT RIA2 AND GENETIC
CORRELATIONS BETWEEN GENETIC MARKER AND TRAIT VARIATION (% FILLET
YIELD).
..............................................................................................................................
39
4.1. Introduction
...................................................................................................
39 4.2. Methods and Materials
..................................................................................
41 4.3. Results
..........................................................................................................
41
4.3.1. Levels of genetic variation in 3 generations in the Tra
catfish selective breeding program
.............................................................................................
41 4.3.2. Genetic correlations between genetic markers and a
production trait (% fillet yield)
..........................................................................................................
43
Chapter 5. DISCUSSION
..................................................................................................
47 5.1. Characterisation of genetic variation in cultured and wild
populations of Tra catfish
...................................................................................................................
47 5.2. The relative contribution by breeders to fry cohorts in
three private hatcheries in the Mekong Delta
.............................................................................................
50
-
v
5.3 The levels of genetic variation in the Tra catfish selection
breeding program at RIA2 and genetic correlations between genetic
marker and trait variation (% fillet yield)
.................................................................................................................................
51 Chapter 6. GENERAL CONCLUSIONS
............................................................................
52
6.1. Genetic Diversity in selected Tra catfish lines in cultured
and a reference wild populations in the Mekong Delta of Vietnam
........................................................ 52 6.2.
Individual brood stock contribution to fry cohorts in three private
hatcheries in the Mekong Delta
.................................................................................................
53 6.3. Assessment of possible correlations among genotypes and
trait quality ...... 54
REFERENCES
..................................................................................................................
55 APPENDICES
...................................................................................................................
62
-
vi
List of figures
Figure 1: Diagram of the structure of the catfish selective
breeding program at RIA219 Figures 2: Gelscan images of genetic
diversity in RIA2 34 brood fish samples (2a)
and from 3 private hatcheries (n=31) (2b) at locus CB 12;
gelscan images of allelic diversity in high fillet yield
individuals (2c) and of low fillet yield individuals (n= 24) (2d)
at locus CB 12.
.................................................... 24
Figure 3: Map of Mekong Delta identifying the main areas where
Tra catfish are cultured (grey colour)
................................................................................
34
Figure 4: Ne estimate for Hau brood fish contributions to
offspring on day one and day two for individuals confidently
assigned to specific parental pairs. ..... 37
Figure 5: Relative allele frequencies at the CB4, CB7, CB12 and
CB13 loci in 3 generations of Tra catfish used in the selective
breeding program. .......... 42
-
vii
List of tables
Table 1: Sample name, sample size, collection date and source of
samples for the whole study.
...............................................................................................
21
Table 2: Primer sequence details of the four microsatellite loci
screened ............... 23 Table 3: The potential for null alleles
for each locus by sample detected using
MICRO-CHECKER.....................................................................................
25 Table 4: Observed and expected heterozygosities (Obs. And Exp,
respectively),
probability value (P-value) and standard deviation (sd).
Significant deviations from HWE indicated as heterozygote
deficiency (def), heterozygote excess (excess) or not significant
(ns) after Bonferroni correction.
..................................................................................................
27
Table 5: Microsatellite polymorphism in 7 sample populations of
wild and cultured Tra catfish populations.
..............................................................................
30
Table 6: The statistical significance of FST values of
population differentiation ...... 31 Table 7: Statistical
significance of FST values (the significance of population)
among
sample pairs. P values that were significant after Bonferroni
correction (α (Bonf) = 0.05/number of test = 0.05/ 21 = 0.002) are
highlighted ............... 32
Table 8: Estimation of Ne in three private hatcheries based on
the number of male and female brood stock
..............................................................................
35
Table 9: Parentage assignment rate of 3 groups of brood fish and
3 groups of fry . 36 Table 10: Genetic correlations between genetic
marker and % fillet yield phenotypic
classes
.......................................................................................................
44
-
viii
Statement of Original Authorship
The work contained in this thesis has not been previously
submitted for a degree or
diploma at any other higher education institution. To the best
of my knowledge and
belief, the thesis contains no material previously published or
written by another
person except where due reference is made.
Signed:
Date: 09/ 09 /2011
-
ix
Acknowledgements:
Special thanks to my supervisors Professor Peter Mather and Dr
David Hurwood
(Discipline of Biogeosciences, Queensland University of
Technology) for their
mentorship and kindness. Thank you to Vincent Chand for
laboratory assistance
and technical support. I especially would like to thank my good
friend Eleanor
Adamson for all her help and encouragement throughout my time in
Australia. I
would also like to thank all my friends in Biogeosciences at
QUT.
-
1
Chapter 1. GENERAL INTRODUCTION
1.1. Status of world fisheries and aquaculture
Fish and other aquatic organisms produced in aquaculture have
become a major
world food production system that has been called on
increasingly to fill the gap
between demand for, and supply of, seafood products for human
consumption
(Knibb 2000; Lymbery 2000). Fish produced from culture provide
an important
human food resource and can reduce pressure on wild fish stocks
in the natural
environment (Lymbery 2000; Primmer, 2005). Aquaculture moreover,
is now a key
industry and plays an increasingly important role in global fish
production and to
meet rising demand for fish and seafood as many wild fish stocks
have declined
over recent decades (Dunham 2000). Under pressure for increased
production of
seafood and the need to develop more productive culture strains,
many aquaculture
industries are trialling new stock management approaches and
attempting to
improve their stock quality (Subasinghe et al. 2000).
Development of sustainable
aquaculture production systems that are economically viable and
that provide
incomes and livelihoods for poor people in many parts of the
globe, is an important
goal in many developing regions around the world.
A number of approaches are being implemented worldwide currently
to assist
aquaculture development and to promote a move away from a major
reliance on
wild fish resources. They include application of traditional
stock improvement
practices (domestication, crossbreeding, hybridisation and
artificial selection) that
can deliver more productive culture stocks. This change has
required application of
a variety of new technologies including identification of
Quantitative Trait Loci
(QTLs), Marker Assisted Selection (MAS) and development and
application of
genetic markers among others. The combination of application of
traditional
breeding practices with appropriate molecular technologies has
been demonstrated
in some aquaculture species to deliver significant productivity
increases for culture
industries (Dunham 2000). For example, gene maps are now
available for some
important aquaculture species, notably. Channel catfish, tiger
shrimp, Japanese
flounder, rainbow trout and Atlantic salmon (Liu 2004).
-
2
Aquaculture Genetics is a relatively new science in many parts
of the world, but
where it has been applied it is transforming the performance
traits of many cultured
organisms and moving the industry towards enhanced
sustainability. For example,
development of the Kansas strain of Channel catfish (Ictalurus
punctatus) provides
an excellent example of the potential that applied genetics can
offer. This fish is the
oldest domesticated strain of Channel catfish having spent more
than a century in
culture. During domestication, growth rates of this strain have
increased 3 to 6% per
generation in culture while growth rates of crossbred strains of
both Channel catfish
and rainbow trout are 55% and 22% higher, respectively (Dunham
2000). Other
examples include: positive heterosis identified in carp
crossbreeds in Israel,
Vietnam, China and Hungary (Moav et al. 1964; Moav and Wohlfarth
1974; Nagy et
al. 1984; Wohlfarth 1993; Hulata 1995); Common carp crossbred
lines in Hungary
have shown almost 20% improvement in growth rate compared with
pure lines while
a Vietnamese x Hungarian common carp crossbreed is now
particularly popular,
due to fast growth and high survival rates under a variety of
production
environments; in Bangladesh, a crossing program was instituted
for three carp
strains referred to as “Bangladesh”, “Thailand” and “Indonesian”
with growth rates
of females from six inter-strain crosses reported to be 23%
higher for average
growth rate compared with parental strains (Dunham 2000).
A stock improvement program for European catfish, Silurus
glanis, has produced a
culture strain that can tolerate warm water conditions and that
can accommodate
mixed diet feeding systems that was achieved via a crossbreeding
approach
(Krasznai and Marian 1985). Another example where improved
culture lines have
been developed is for walking catfish, Clarias macrocephalus,
where improved
tolerance of Aeromonas hydrophila infections was developed by
careful application
of a breeding program directed at genes that influence
resistance traits (Prarom
1990).
The earliest modern genetic selection program directed at an
aquatic species was to
improve survival rate in brook trout (Salvelinus fontinalis), a
species that was
susceptible to endemic furunculosis and this program was
initiated in the 1920s.
The outcome of this program was to increase survival rate from
2% to 69% after
three generations of artificial selection. Later in 1932, simple
selection approaches
were applied to improve the growth rate and fecundity of rainbow
trout
-
3
(Oncorhynchus mykiss) in culture (Donaldson and Paul 1957).
After 35 years of
direct individual selection, this strain is now widely cultured
in the USA and more
widely in other regions across the world (Parsons 1998 cited in
Hulata 2000). In
Norway, large stock improvement programs were initiated for
Atlantic salmon and
rainbow trout in 1971 and these programs have achieved genetic
gains estimated to
be 10 to 15% per generation over the first two generations.
Following this, growth
rate and age at sexual maturation traits were targeted from the
fifth generation, and
in later generations, disease resistance and meat quality traits
were subjected to
artificial selection (Gjedrem 2000). The selected Atlantic
salmon culture strain in the
fourth generation showed 77% faster growth rate than control
wild fish from the
Namsen River. In 1993, Kirpichnikov reported the outcome of a
breeding program in
common carp that employed mass selection to improve resistance
to dropsy and
that increased growth rate in Krasnodar in the former Soviet
Union (Hulata 2001).
A genetic selection program on gilthead sea bream (Sparus
aurata) was also
successful with single-pair offspring groups (full- and
half-sibs) employed to improve
production traits (Knibb et al. 1997a). The most widespread fish
cultured in Asia
today is a genetically improved strain, the GIFT tilapia
(Genetic Improvement of
Farmed Tilapia) and increasingly male hybrid tilapia stocks are
also produced widely
(Subasinghe 2000). Several studies have reported, for tilapia
strains (Oreochromis
mossambicus, red tilapia, O. aureus and O. niloticus) that mass
selection can
improve body weight significantly. Family selection for improved
growth rate in the
GIFT Nile tilapia has achieved 77% to 123% improvement with an
11% genetic gain
achieved per generation (Padi 1995).
Carcass quality and percent fillet recovery traits have also
been targeted for
improvement in salmonids and catfish (Dunham 1996a). In
Thailand, selection for
improved growth rate and disease resistance are currently being
trialled for a
number of important native and exotic culture species including
pangasiid
freshwater catfish (Pangasius sutchi, syn. of P. hypophthalmus),
rohu (Labeo
rohita), Thai walking catfish (Clarias macrocephalus), Java barb
(Barbodes
gonionotus), bighead carp (Aristichthys nobilis) and Asian rock
oyster (Saccostrea
cucullata) (Dunham 2000). In Australia, genetic improvement
programs have been
trialled recently for Pacific oyster (Crassostrea gigas) and
Sydney rock oyster
(Saccostrea glomerata). Haley et al. (1975) reported on
selection in a related oyster
-
4
species, C. virginica, where adult oysters showed an apparent
strong response to
mass selection to improve growth rate (Dunham 2000).
More recently, selection responses reported for body weight at
harvest size using a
mass selection approach conducted on Channel catfish (I.
punctatus) were high with
genetic gains of 29% for the Kansas strain and 21% for the
Marion strain with
associated heritability values of 0.16 and 0.23, respectively
(Rezk et al. 2003).
Over the years however, traditional approaches for improving
culture stocks have
faced a number of serious challenges with both a decline in the
quality of important
production traits and erosion of the response to selection
becoming significant
issues for the industry (Knibb 2000; Lymbery 2000). Of
increasing interest also are critical concerns about the threats to
genetic diversity levels in cultured fish
populations and why there is high risk associated with erosion
of exploitable
variation in culture. Understanding the unique attributes of
many aquatic species
that make them potentially much more vulnerable to rapid loss of
genetic diversity in
culture compared with their terrestrial live stock counterparts
will be critical to
developing better breed improvement programs that can assist new
aquaculture
industries in many parts of the world.
1.2. Role of genetics and population genetics in aquaculture
Genetic variation or genetic diversity in a population consists
of the heritable
information contained in the genome of any population of a
species (Kottelat and
Whitten 1996). It describes the diversity of different alleles,
or alternative forms of a
given gene that can be found in the target population. Genetic
variation implies
presence in individuals in a population of different alleles
that if expressed, may
produce a variety of phenotypes. In theory this variation will
reflect a population’s
ability to adapt to changes in its environment (Gjedrem 2005).
While individuals in a
population cannot predict future environmental change, the more
variation that
exists in their collective genomes, the better placed a
population will be to adapt to
change if and when it occurs. Thus variable populations will
generally respond
better than non-variable ones to environmental change because
more exploitable
variation remains.
-
5
In aquaculture, where relatively small brood stock populations
are often maintained
separately, the genetic constitution of farmed fish stocks can
change rapidly over
just a few generations and where populations are small they will
tend to lose genetic
variation rapidly (Gjedrem 2005). This is a critical issue in
aquaculture, because
achieving genetic gains from a selective breeding program will
ultimately depend on
there being sufficient genetic diversity present in the original
brood stock (Yu and Li
2007). The amount of genetic variability present in a target
population will ultimately
be influenced by diversity levels in the parental stock and the
mating system
employed. Genetic variation is essential for population
persistence because
populations with higher levels of genetic diversity have greater
adaptive potential
and hence provide the best resources for selective breeding
programs (Kottelat and
Whitten 1996, Mable and Adam 2007).
Many studies conducted on a variety of different cultured
aquatic species including
brown trout – Salmo trutta (Aho et al. 2006), catla – Catla
catla (Alam and Islam
2005), rainbow trout – Oncornhynchus mykiss (Pante et al. 2001),
black tiger shrimp
– Penaeus monodon (Xu et al. 2001), sole – Solea senegalensis
(Porta et al. 2006)
and Pacific white shrimp – Penaeus vannamei (Moss et al. 2008)
have reported
ongoing declines in genetic variation over generations during
the very early stages
of domestication with some even reporting complete population
homozygosity. In
some instances, virtually all of the natural levels of variation
can be lost as a result
of poor stock management practices, so this has become an
important issue for
aquaculture.
Aquatic species, unlike most terrestrial farmed livestock
species, are particularly
vulnerable to rapid loss of genetic diversity because of
inherent characteristics that
are very different to their terrestrial counterparts. First,
most aquatic species used in
culture are highly fecund, producing large numbers of gametes
per individual with
females often capable of producing millions of eggs in a single
mating event. While
survival of fertilised eggs in the wild is usually extremely low
(90%) in culture. Thus large numbers of offspring can be
generated form even a single mated pair in culture situations.
This is important for
hatchery managers because it means that they often require only
a few breeders to
generate all the fry or larvae they will need to meet demand.
This immediately
creates problems with conserved genetic diversity levels because
while fry/larvae
-
6
numbers may be very high (a good outcome for breeders due to the
reduced costs
of producing them), diversity can be low compared with
equivalent wild-spawned
offspring cohorts where individual survival rates are often very
low.
Where effective population size (Ne) is low, genetic drift and
inbreeding can rapidly
erode population levels of genetic diversity and change gene
frequencies without
regard to their adaptive potential. Ne is defined as "the number
of breeding
individuals in an idealized population that would show the same
amount of
dispersion of allele frequencies under random genetic drift or
the same amount of
inbreeding as the population under consideration” (Hallerman
2003). Another way of
understanding the concept is that effective population size
refers to the number of
individuals that contribute genes in equal proportions to the
next generation
(Bensten and Gjerde 1994, Doyle et al. 2001, Bensten and Olesen
2002, Yu and Li
2007). Effective population size is usually considered to be a
significant parameter
in many population genetics models and in practice effective
population size is often
much smaller than observed population size (N) (Brown et al.
2005, Primmer 2005).
As an example, while there are estimated to be 2000 mature
individuals of winter
chinook salmon (Oncorhynchus tshawytscha) in the Sacramento
River in California,
the Ne of this population has been estimated to be as low as 85
breeding individuals
each reproductive cycle (Bartley et al. 1992). Japanese
flounder, Paralichthys
olivaceus, provides another illustration of Ne
-
7
Population genetics is a field of biology that studies the
genetic composition of
biological populations (allele frequency distributions) and the
changes in genetic
composition that result from evolutionary forces of which
natural selection, genetic
drift, mutation and gene flow are considered to be the major
ones that influence
gene frequencies in both natural and cultured populations (Hartl
and Clark 1997,
Hallerman 2003). Important applications of population genetics
in fisheries and
aquaculture have included delineation of wild stock structure,
identification of
breeding units, generating theoretical estimates of effective
population size,
assessment of inbreeding rate and documenting levels of genetic
variation in target
populations (Tave 1993; Gjedrem 2005).
Population genetic methodologies when applied appropriately in
aquaculture can
help to address issues associated with negative impacts of
animal husbandry
practices that have the potential to severely impact levels of
genetic diversity and
hence cause loss of fitness in cultured aquatic populations. The
genetic risks
associated with the domestication process employed to produce
fish artificially has
also become an important issue in recent times. Many farmed fish
strains have
relatively low levels of genetic variation compared with their
wild progenitors
(Hansen et al.1997; Yu and Li 2007). If this causes a decline in
stock productivity,
then inbreeding depression (where alleles of low fitness may
accumulate in the
stock due to combined impacts of genetic drift and inbreeding)
has often been
implicated as a major causal factor. The use of only a limited
number of broodstock
can potentially lead to high population levels of inbreeding and
consequently cause
rapid declines in population genetic diversity levels. This can
be reflected in random
changes in frequency or even total loss of critical alleles
responsible for important
production traits (Gaffney 2006). The importance of maintaining
high levels of
genetic variation in any brood stock is now appreciated more
widely and is
considered to be essential for a stock’s long-term
sustainability and productivity
(Mustafa 2003). Small effective population size, line breeding
or close relative
mating, are all factors that can contribute to increased levels
of inbreeding and
ultimately may result in inbreeding depression; therefore, to
limit this potential effect,
where possible animal breeders should maintain pedigree records
for their brood
stock. This practice, will allow development of strategies that
contribute to
maintenance of a sustainable base population for culture
(Reisenbichler and Rubin
1999).
-
8
Understanding the individual attributes of aquatic animal
species used in culture,
especially their genetic characteristics, will support science
and aquaculture to
provide better solutions to current problems. Thus traditional
stock improvement
approaches in aquaculture and modern genomics can benefit from
population
genetic theory and methodologies that identify, monitor and
conserve genetic
diversity in culture lines.
1.3. Genetic improvement - moving from wild to improved culture
strains
Genetic stock improvement programs provide a powerful means for
enhancing both
preferred qualitative and quantitative traits in any cultured
population (Eknath et al.
1998). Stock enhancement in essence, aims to increase the
biomass of the target
species with little or no disadvantageous impacts on native gene
pools (Ward 2006).
Maintaining genetic quality in fish stock management is now
therefore considered to
be a priority for many aquaculture sectors (Lymbery 2000). While
in general, stock
improvement can be used to enhance the majority of desirable
traits in seed stocks
(Allan 1999), in reality, however, this practice is often
difficult and requires long-term
effort from a large number of scientists with access to
appropriate technologies,
facilities and adequate financial support (Hulata 2001; Knibb
2000). Genetic
improvement involves making decisions about which individuals to
mate (selection)
and how to mate them in the most optimal way (mating systems)
(Allan 1999), while
stock improvement is the ability to select brood stock
individuals with an appropriate
combination of superior breeding values for selected economic
traits (William 1991).
This outcome will result from fully understanding the genetic
variation present in a
stock and applying sound breeding practices. Developing
genetically improved
stocks however, requires a complex link between theory and
practice and so
requires understanding both an individual’s biological and
genetic backgrounds. The
objective of selection is to improve the stock genetically by
increasing the frequency
of desirable genes (alleles) while decreasing the frequency of
less-desirable ones
(Allan 1999). This needs to occur without significantly eroding
exploitable genetic
variation levels to a point where any future response to
selection may be
compromised. Additionally, while the focus is on optimizing
favourable alleles, this
needs to be achieved without increasing the inbreeding rate to a
point where
inbreeding depression may result (Davis and Hetzel 2000).
-
9
Genetic improvement programs in farmed aquatic species around
the world for the
most part, have been very successful (Knibb 2000). The power of
genetic programs
is to change a cultured stock attributes to fit a purpose or
production environment.
Records of genetic enhancement activities were evident in
terrestrial agriculture
from the time early humans first made the transition from being
hunter-gatherers to
farmers and producers. These achievements not only solved
problems associated
with obtaining necessary food resources but also reduced the
risk of loss of wild
genetic diversity (Booke 1999). In particular, genetic
enhancement can provide
solutions to ongoing emergence of pathogens and parasites in
high-density culture
that often result in serious new disease outbreaks. Good genetic
management and
selection of strains that are resistant to important pathogens
have the potential to
address this emerging problem (Chevassus and Dorson 1990).
Furthermore, stock
enhancement can open up new profit opportunities for companies
and export
industries.
In fisheries and aquaculture, genetic improvement and stock
enhancement efforts
continue to create new opportunities (Liao et al. 2004).
Increased demand for
aquatic products together with the significant contribution that
aquaculture can make
to world food resources makes the role of genetic stock
enhancement an emerging
practice in aquaculture, worldwide (Mustafa 2003). A decade ago,
genetically
improved fish and shellfish contributed only around 1% to total
global aquaculture
production (Gjedrem 2000); of which only a few species made the
major contribution
to this component e.g. 75% of Penaeus japonicus farm stock in
Australia were
genetically improved (Hulata 2001). According to well-documented
records (Hulata
2001), effective aquatic stock enhancement programs are still in
the pioneering
stage and only a limited number of species have been addressed
intensively to
date, mostly in developed countries especially Atlantic salmon
in Norway and
Channel catfish in the USA.
More recently, a long term and large-scale breeding program was
initiated for
Channel catfish (Ictalurus punctatus) and after three
generations of selection the
program achieved a 10-20% gain in growth rate per generation
(Mahmound et al.
2003). Genetic stock improvement via artificial selection has
also been successful in
several other breeding programs including in a cultivated strain
of Penaeus
(Litopenaeus) vannamei (Donato et al. 2005), that was then used
for stock
enhancement of depleted wild populations (Davenport et al.
1999), There is also a
-
10
long history of stock enhancement in Japan for three finfish
species: Japanese
flounder (Paralichthys olivaceus), red sea bream (Pagrus major)
and black sea
bream (Acanthopagrus schlegeli) (Fushimi 2001). Moreover, some
aquaculture
stock improvement programs in developing countries have achieved
positive
outcomes over the past few decades for silver barb (Barbodes
gonionotus) selected
for better growth performance in Bangladesh (Hussain et al.
2002) and a
commercial crossbreeding program for common carp in Vietnam
(Thien and Thang
1993). Thus, the evidence is clear, where well-designed stock
improvement
programs have been implemented for aquatic species, genetic
gains can be rapid
and can enhance industry development. Genetic gains from
breeding programs for
aquatic species often produce much more rapid gains than
equivalent programs in
terrestrial species because genetic variation levels in
broodstock are often quite
high as they have only recently been sourced from the wild while
terrestrial species
have been domesticated for 100s if not 1000s of generations by
humans.
In the current study Pangasianodon hypophthalmus referred to as
Tra catfish was
the target species and this species has become a key export
industry in Vietnam.
While the Tra catfish culture industry in Vietnam is quite
young, it now contributes
significantly to export revenue there, but also the industry is
a significant provider of
employment for many poor farmers most importantly, in the Mekong
Delta region.
1.4. The Tra catfish culture industry in Vietnam
Two Vietnamese catfish species (Pangasianodon hypophthalmus –
Tra) and
(Pangasius bocourti – Basa) are native freshwater fishes that
are common and
widely distributed across the Mekong Delta in Vietnam. Both
species constitute very
important food fishes in the region (Trong 2007) and occur
naturally in both main
branches of the Mekong River. P. hypophthalmus and P. bocourti
provide major
protein resources and livelihoods for many rural households
particularly during the
flood season. Originally, catfish culture in Vietnam was
small-scale and was
generally poorly organized such that most fish were used only
for family or local
domestic consumption. Only low quantities were produced in
culture and
management of quality was essentially absent. Over a number of
decades, a
number of long-term trials were undertaken to close the life
cycle of both species in
hatcheries and practices were developed simultaneously by a
number of
Vietnamese research institutes, in the south of Vietnam (Mekong
Delta) in particular
-
11
by Trinh et al. (2002) that allowed production of catfish fry in
sufficient quantities to
support development of a large culture industry. From this
simple beginning, culture
of P. hypophthalmus and P. bocourti has gradually become a
significant economic
driver and a key income generator for thousands of poor
households in the Mekong
Delta.
Covering approximately 40,000 square km and possessing a large
young population
of approximately 17 million, the region accounts for 21% of
Vietnam’s population
and more than 60% of people participate in fishing there. In
2006, the Vietnamese
fisheries sector accounted for an estimated 6.1% of Gross
Domestic Product (GDP)
producing US$ 3.4 billion in export revenue (Nortvedt 2007).
Recently, P.
hypophthalmus and P. bocourti were selected to become Vietnam’s
aquaculture
product trademark, as cultured catfish has become a major export
product.
Following a recent trade war however, the Vietnamese government
adjusted
aquaculture export strategies and supported policies for
protecting local fish
farmers, processing and export plants. The Vietnamese government
has also
recently contributed significantly to, and promoted expansion
of, catfish aquaculture
in the Mekong Delta and now sees the industry as playing a major
role in social and
economic development for the nation (Monti et al. 2006).
Over recent years, Vietnamese catfish producers have
successfully established and
expanded export markets that ensure annual export targets are
met. In parallel, the
industry has decreased dependence on the US export market, and
has diversified
the variety of available catfish products (Tung et al. 2004). In
2006, the largest
export destinations for Tra and Basa catfish were European
countries with export of
123,212 metric tons (MT) worth approximately US$ 343.4 million
per annum,
followed by Russia with 42,779 MT estimated value at US$ 83.2
million per annum,
while the American market accepted only 24,281 MT producing US$
72.9 million per
annum while Australia consumed 10,149 MT valued at US$ 31.0
million per annum.
Thus total catfish exports in 2006 were approximately 286,602 MT
valued at a total
of US$ 736.87 million per annum (Nortvedt 2007). Increasing
popularity of
Vietnamese catfish in the international aquaculture market
reflects recent significant
improvements in product quality. Today more than 60 processing
plants have been
developed across the Mekong Delta that produce a variety of
catfish products
compared with only eight plants that existed in 1997 (Monti et
al. 2006). As a
-
12
consequence, of all fish cultured in Vietnam, catfish now
accounts for approximately
80% of exported fish product. Vietnamese catfish production has
shown a 36-fold
increase in production since 1997 and this is linked to an
increase of 40% in export
of frozen fillets. This now results in a total production of
over 825,000 MT
(Lobegeiger 2007), most now going to high value consumers in
developed countries
outside Asia.
1.5. Tra fry cohort production in private hatchery practices in
the
Mekong Delta Results of a survey by Yen and Trieu (2008) of 30
catfish hatcheries in the Mekong
Delta showed that the majority of brood stock used to produce
fry were sourced
from first or second generation domesticated stock. Most
hatchery owners possess
only low educational levels rarely above high school level.
Initially, most hatchery
owners employed staff with only simple technical training in
hatchery techniques
that had been learnt from government extension training
programs. From this basic
starting point, practices used in hatcheries were established
and now most hatchery
owners undertake their own fry production. In general, two to
ten people are
employed in most hatcheries, with the majority being family
members. The number
of brood stock held by each hatchery ranges from approximately
100 to over 1000
individuals depending on the hatchery size. For example, one
large hatchery in the
Mekong Delta holds approximately 1,700 brood fish (estimated 4
kg each) with a
capacity to produce more than 200 million fry per year (Hung et
al. 2008). 47% of
the 30 hatcheries surveyed had collected their brood fish from
the wild while 30%
had obtained brood fish from other hatcheries and 23% routinely
obtained their
brood fish from both sources. In terms of broodstock management
practices,
approximately 73% of hatcheries regularly sourced their brood
fish from commercial
grow out ponds, 17% kept their own fry to become brood fish and
only 10% of
hatcheries used wild fish to replenish their brood stock
supplies (Yen et al. 2008).
While brood stock age varied, most were less than 7 years old
while 70% of brood
fish were younger than 5 years because the best reproductive age
is 3 to 5 year old
fish with an average weight of 3 to 5 kg. With this approach,
brood stocks are
usually replaced every 2 to 3 years. 40% of hatcheries replace
brood fish every
year, 60% of the remaining hatcheries have more than two
generations of brood fish
while a single hatchery maintains 4 generations of brood fish
(Yen and Trieu 2008).
-
13
Fish used as brood stock are tagged on their head with a mark
and kept separately
in small concrete ponds while two of the 30 hatcheries screened
did not tag their
brood stock (Yen and Trieu 2008). Artificial propagation of Tra
catfish is carried out
by each farmer where they use sperm from a single male to
fertilize 3-5 females
each cycle with an average estimated 120,000 egg/kg female. Some
hatcheries mix
the sperm from multiple males to increase their fertilization
rate. Almost all of the
hatcheries surveyed produce fry for commercial sale directly,
but a few also keep up
to 12.5% for nursing to a larger size before sale. Number of
hatchery runs (batches)
varies from 17 to 19 times per year and each brood stock
individual is spawned two
to four times per year using a sex ratio of one male to four
females. When asked to
comment on fry quality, 70% of Tra hatchery owners considered
that fry from
artificial propagation were better or at least of equal quality
to fry spawned from wild
brood fish. This is the major reason why 60% of hatchery owners
routinely source
their brood fish from their own ponds while only 10% of owners
believe that use of
wild fish contributes to better quality fry. A diversity of
different breeding practices
have been adopted by hatchery owners from use of only a single
pair of fish to
combining batches of fry from multiple crosses to sourcing fry
from neighbouring
provinces (Yen and Trieu 2008, Hung et al. 2008). Hatchery
owners were surveyed
about issues associated with levels of inbreeding in their
hatcheries, with only 40%
responding that they knew that mating related fish can increase
the inbreeding rate
and this can affect fry quality; 56% said it was not necessary
to consider genetic
relationships among brood fish when producing fry for
culture.
During field sampling of brood stock from three private
hatcheries in the Hong Ngu
district of Dong Thap Province, specific details of practices in
hatcheries were
documented. Two hatchery owners possessed high school level
training and the
third was a local aquaculture official. Each produced fry in
their hatcheries
themselves and employed 3-5 of their relatives (to assist).
Assistants in hatcheries
were also employed elsewhere as teachers or rice farmers or were
otherwise
unemployed. Each hatchery had approximately 500 to 1,000 brood
stock cages
along the river or brood stock were held in earthen ponds behind
houses. No
information was available however, about the number of brood
stock used regularly
for spawning, or whether all fish were used to produce fry.
Common practice was to
initiate artificial propagation using 40-80 female and 5-10 male
brood fish for each
cycle. Before each batch, brood stocks were selected based on
visual inspection; if
-
14
they possessed good phenotypes and were healthy. For the initial
spawning in each
batch, hatchery owners used about 10-15 quality fish with an
average weight of 2-3
kg/individual that were ready to spawn. All eggs were collected
from 10-15 females
into a plastic basket and combined with milt from 2-3 males. A
single female was
used for 3 to 4 repeat spawns per year and after 2-3 years;
individuals were
replaced. In some instances, hatchery managers employed brood
fish from other
hatcheries or even wild fish obtained from Cambodia. Most
hatchery owners were
quite secretive about their practices and often will not
cooperate with government
officials or volunteer information because they think it is not
essential and do not like
their business practices being scrutinised. Only a very few
owners were interested
in genetic or inbreeding issues with their stocks or saw the
relevance of these
issues to their operations.
Provision of appropriate support can allow cultured catfish from
Vietnam to become
as well-known an aquaculture brand as has been achieved for
Norwegian salmon.
Recent strategies employed in the Mekong Delta industry have
increased the catfish
breeding area to 10,200 ha with 1,900 fish cages to be
introduced by late 2010 with
a total capacity now exceeding 860,000 MT in output. Export
targets predict a
potential production of 230,000 MT of Tra and Basa catfish
fillet, earning 600 million
USD in 2010 and this is forecast to increase to 460,000 MT with
an estimated
turnover of 1.2 billion USD by 2020 (TheFishSite). There are
many reasons why the
catfish culture industry in Vietnam has expanded so rapidly
including: the industry
has addressed specific demands from foreign consumers, Pangasius
catfish culture
makes an important contribution to household incomes and
Vietnam’s export
income has grown and increased employment opportunities in the
Mekong Delta.
Improving the quality and productivity of Tra and Basa
aquaculture is now seen as a
significant opportunity for Vietnam and if issues are addressed
appropriately this will
allow the industry to continue to expand. Currently, virtually
nothing is known
however, about how existing management practices of cultured
catfish stocks in
Vietnam are impacting levels of genetic diversity. Understanding
what impacts (if
any) have occurred will allow better breeding practices to be
designed in the future,
if they prove necessary.
-
15
1.6. Genetic marker applications in aquaculture
Exploiting advanced genetic marker technologies can be useful at
three levels for
management of aquatic species: i) for detecting simple inherited
traits; ii) finding loci
that affect variation in quantitative traits; and iii) for
assisting with optimizing
selective breeding outcomes based on marker assisted selection.
Examples where
markers have applications in aquaculture include: gene mapping,
identification of
sex, individual identification and parentage assignment,
population genetic
applications, detecting effects of selection and trans-genesis
(Lo Presti 2009).
For example, the basic objective of gene mapping studies is to
elucidate the location
of functional genes responsible for important performance and
production traits
(Davis and Hetzel 2000; Liu 2006). The position of gene loci
encoding simple
inherited characteristics can be located in the genome of a
target species by
analysing correlations between allelic variation in families
that co-segregate with
markers after linkage analysis (Georges 1998 cited in Davis and
Hetzel 2000). Once
identified, gene markers allow screening of parents and progeny
and development
of a diagnostic test such as the simple genetic markers used
widely for human
disease diagnostics (Hartl and Clark 1997). Many genomics
projects have used
microsatellite markers to develop aquaculture databases for
target species. For
example, in Europe, there are a number of genetic improvement
programs that have
developed more than 100 microsatellite markers for species such
as common carp
(Cyprinus carpio), approximately 1,700 microsatellites for
Atlantic salmon (Salmo
salar), and more than 250 microsatellites for European sea bass
(Dicentrarchus
maximus). Other important species in European aquaculture
include flat oyster,
lobster, and Atlantic cod where more than 50 microsatellite
markers are available for
each species (Blohm et al. 2006). A very large and long-term
project is applying
genetic markers (based on microsatellites) to developing a
saturated linkage map
for Channel catfish (Ictalus punctatus). Several hundred
microsatellite markers have
been developed over the recent decade for both Channel and blue
catfish families
by Liu et al. (2006) and these will contribute to the resulting
linkage map. Another
study of Channel catfish developed 293 microsatellite markers to
add to the growing
genetic linkage map (Geoffrey et al. 2001). To date, two genetic
linkage map
frameworks have been published for the species. The first map
developed by
-
16
Waldbieser et al. (2001) was established using Channel catfish
intra-specific
resource families, and the other was initiated by Liu et al.
(2003) using Channel
catfish X blue catfish inter-specific families (Liu 2006).
Linkage maps for salmonids
are also under construction with over 228,000 bacterial
artificial chromosome (BAC)
clones fingerprinted from Atlantic salmon (Liu 2006). Tilapia,
oysters, shrimps and
striped bass are other species where concentrated efforts are
being made to map
each species’ entire genome (Liu 2006).
Identification of sex of cultured individuals can also be
important in some aquatic
species because one sex may perform better when produced in
monosex culture
stocks, an example being culture of all-male freshwater prawn
(Macrobrachium
rosenbergii) (Sagi et al. 1998) or Nile tilapia (Oreochromis
niloticus niloticus)
(Angienda et al. 2000). Discrimination of phenotypic sex can be
quite difficult in the
early larval stages for many aquatic species, but molecular
genetics has proven to
be very effective at addressing the problem via detection of
sex-linked DNA markers
in different species (Devlin and Nagahama 2002).
For individual identification and parentage assignment, genetic
markers can not only
solve problems associated with physical tags that are often
difficult or even
impossible to apply in juveniles or farmed molluscs etc. but
also significantly
decrease the cost and time required to keep different families
in separate ponds, a
process that also limits the number of animals available for
selection (Lymbery
2000, Bentsen and Olesen 2002, Liu and Crodes 2004). In
parentage analysis,
genetic markers can identify individuals effectively using
distinct genotypes from
allelic diversity and allele frequency data. Once genetic
information is available for
parental pairs (sires and dams) and their offspring, breeders
can construct simple
pedigrees (Martinez 2007). Individual identification combined
with pedigree data,
allow the researcher to identify individuals and their genetic
relationships to select
those with the best breeding values (Bentsen and Olesen 2002).
This can help to
estimate selection response and to optimise breeding parameters
(Zhang et al.
2006).
In selection programs, molecular markers can be used to identify
genetically
superior individuals; this process is referred to as marker
assisted selection (MAS).
-
17
MAS or genome wide marker assisted selection (G-MAS) is a
selection process in
which specific candidate genes are identified based on genotypes
using molecular
markers (Liu and Crodes 2004) and results from the use of gene
markers linked to
QTLs in genetic improvement programs (Davis and Hetzel 2000). To
employ MAS,
researchers need to have a clear and high-density linkage map
and to understand
comprehensively the number of QTLs that affect each phenotypic
characteristic,
their mode of inheritance and potential interactions of
different QTLs on traits and
the economic characteristics of the traits studied (Poompuang
and Hallerman 1997
cited in Liu and Crodes 2004). MAS in general, is applied mostly
to traits that are
otherwise difficult or expensive to measure and QTLs are loci
that exert a major
influence on important quantitative traits. Many commercially
important traits that
show continuous variation (quantitative traits) are generally
influenced by a group of
genes of small additive effect (Falconer 1989). In genetic
improvement programs,
QTLs are evaluated as complementary to breeding value estimates
of genetic merit
(Davis and Hetzel 2000). QTLs are usually detected by combined
analysis of
phenotypes with linked marker maps and they have been applied in
a number of
aquaculture stock improvement programs because of their capacity
to assist animal
breeders to reach specific breeding goals.
Some examples where molecular genetic approaches have been
broadly applied in
genetic improvement programs include Appleyard and Ward (2005),
who reported
that 8 microsatellite markers were useful in a mass selection
program for Pacific
oyster (Crassostrea gigas) in Australia and New Zealand.
Similarly, MacAvoy et al.
(2008) published 49 microsatellite primer sets for a selective
breeding program in
the New Zealand GreenshellTM mussel (Perna canaliculus). Asian
aquaculture
researchers have also contributed recently to world genetic
databases. Eleven
microsatellites were developed to estimate kinship for brood
stock management in
Japanese flounder (Paralichthys olivaceus) to minimise risk of
inbreeding (Sekino et
al. 2004), while Wang et al. (2006) used 240 microsatellites to
screen 24
chromosomes in the karyotype of Asian sea bass (Lates
calcarifer), known locally
as Barramundi in Australia. They mapped 5 significant and 24
putative QTLs that
influenced individual body weight. A four-way tilapia cross in
Israel also initiated a
linkage map for QTL studies of this important culture species.
20 microsatellites
were found to be associated with two significant QTL traits, one
for cold tolerance
and the other for individual growth rate (Moen et al. 2004).
-
18
Currently, much effort and cooperation worldwide is directed at
developing QTL
maps or MAS for specific aquaculture species. In Channel
catfish, rainbow trout and
tilapias, QTL markers for growth, feed conversion efficiency,
tolerance to bacterial
disease, spawning time, embryonic developmental rates and cold
tolerance have all
been reported (LaPatra et al. 1993, 1996). In trout and salmon,
a candidate DNA
marker linked to infectious haematopoietic necrosis (IHN)
disease resistance was
also identified recently (Houston et al. 2008). Thus, genetic
markers can assist
animal breeders to improve the quality of their culture
stocks.
1.7. Specific aims of the current study
There were four aims in the current study:
i. To assess the genetic status of cultured Tra catfish
(Pangasianodon
hypophthalmus) populations in the Mekong Delta of Vietnam,
ii. To evaluate the levels of genetic variation in the Tra
catfish selective breeding
program at RIA2 after 3 generations and to estimate the
effective population size
and other related genetic parameters in this population.
iii. To assess the percent contribution by individual breeders
to fry cohorts and to
estimate effective population size and inbreeding coefficients
in three private
hatcheries in the Mekong Delta,
iv. To trial genetic correlations between genetic markers and an
important
production trait (fillet yield).
-
19
Chapter 2. MATERIALS AND METHODS
2.1. The Tra catfish selective breeding program at RIA2
A selective breeding program for Tra catfish commenced at the
Research Institute
for Aquaculture No 2 (RIA2), Ho Chi Minh City, in 2001. Brood
stock used were
obtained as fry produced from wild parents sourced from three
private hatcheries–
Truong, Duong, and Ro (WP – TDR). After brood stock were raised
for 3 years
(2004), fish were mated as families and tagged for breeding
selection at RIA2. The
parental generation was referred to as the ‘P’ generation.
The selective breeding program at RIA2 was initiated to improve
fillet yield in 2004.
In 2005, when F1 individuals were 8 months old, the average time
to market size, all
individuals were measured. Each fish was measured for total
length, weight, and
body depth. Following this, 30 individuals were chosen at random
from a sample of
100 fish from each family and individuals were euthanized by
professional filleters at
the Angiang catfish-processing plant to assess fillet metrics
per family. After filleting,
each individual was weighed for total fillet yield and remaining
body components.
Figure 1: Diagram of the structure of the catfish selective
breeding program at RIA2
-
20
2.2. Propagation of fry at 3 private hatcheries in the Mekong
Delta
In private hatcheries in the Mekong Delta, broodstock are
selected as breeders
based on their individual growth performance. In some instances
brood stock are
also sourced directly from the wild or are domesticated fish
obtained from the
hatchery. Breeders are selected at approximately 2 years of age
when they weigh
approximately 3 kg (Yen and Trieu 2008). Of the three hatcheries
examined here,
two-employed wild caught brood stock from a local river (Khanh
and Hau) while the
third hatchery employed only domesticated stock (Nam). For each
fry propagation
cycle, approximately 40 females and 10 males were separated to
become brood
stock for artificial spawning. Sperm from two or three males
were employed
commonly to fertilize eggs combined from 8 females. Spawns were
then pooled
together into a few large batches (depending on the number of
eggs produced).
After hatching, newly spawned fry resulting from multiple
batches were combined
automatically into a large single circular nursery tank. Samples
for the current study
were then collected randomly on the first and second days after
fry had hatched. On
the second day, remaining fry were sold to farmers. Samples of
fins from all brood
stock individuals and a random sample of whole larvae from
brood-tanks were
collected and stored in 70% ethanol at 4°C prior to DNA
extraction and genetic
analysis.
2.3. Sample collection
Two groups of samples were available:
Group A (RIA 2 group): The first group comprised individuals
from the catfish
selective breeding program at RIA2. This group consisted of 48
founder individuals
collected form the wild in 2001 (WP called TDR); 34 offspring
from P (F1) that
contributed to reproduction events in 2005; and 120 selected
individuals (F2) that
included 48 high fillet yield individuals, 48 low fillet yield
individuals and 24 random
individuals (Controls) in 2006. In addition, 27 individuals were
collected from the wild
as a reference sample of wild diversity levels and these
individuals were sourced
from two branches of the lower Mekong River (HW: Hau River Wild
= 20 fingerlings
and TW: Tien River Wild = 7 samples) in 2008 and 2009,
respectively (see Table 1).
-
21
A small piece of fin tissue was removed from each adult fish and
samples stored in
70% ethanol at 4°C prior to DNA extraction.
Group B (Private hatcheries group): The second group comprised
individuals from
three private hatcheries: Hau, Nam and Khanh owned by
smallholders in one of the
two main branches of the Mekong River and (Tien River) that
produced fry cohorts
for commercial culture that were chosen for the study in 2008.
Sampling was
conducted during fry propagation periods. Eggs were mass hatched
in 100 litre
containers and fry were transferred to a 3000 litre tank for
nursing; 200 larvae were
sampled on the first and second days (100 each day) after
hatching and samples
fixed in 70% ethanol at 4°C prior to DNA extraction. A total of
31 samples of
parental individuals and 600 fry from 3 hatcheries were sampled
for the study.
Details of the sampling (Table 1) are presented below.
Table 1: Sample name, sample size, collection date and source of
samples for the
whole study.
Sample name Abbreviation Place
Year
n
RIA 2 group (A)
Wild Parents TDR Tien River - wild 2001 47
F1 RIA2 RIA2 - domestic 2005 34
F2 - High fillet H RIA2 - selection 2006 48
F2 - Low fillet L RIA2 - selection 2006 48
F2 - Random R RIA2 - selection 2006 24
Wild reference HW Hau River – wild offspring 2009 20
Wild reference TW Tien River – wild adult 2008 7
Hatcheries group (B)
Hau brood HB Tien River 2008 10
Hau offspring H1 and H2 196
Nam brood NB Domestic 2008 10
-
22
Nam offspring N1 and N2 196
Khanh brood KB Tien River 2008 11
Khanh offspring K1 and K2 196
2.4. Genomic DNA extraction
Total genomic DNA was extracted from all mature brood stock
individuals from
approximately 50mg of caudal fin tissue using a standard salt
procedure, as
described in the QUT Ecological genetics laboratory manual, a
procedure adapted
from Miller et al. (1988).
Total genomic DNA was also extracted from first day, and second
day old larvae
using a Chelex procedure (QUT Ecological genetics laboratory
manual 2004). This
procedure required that whole larvae were digested overnight at
55°C in 100 μl of
10% Chelex 20mg/ml Proteinase K.
2.5. Genotyping procedures Multi-locus microsatellite genotypes
were obtained for each sample individual via
polymerase chain reaction (PCR) amplification using four
microsatellite primer sets
purchased from a Thai commercial company (DNA Technology
Laboratory, BIOTEC
– Kasetsart University - Thailand) with support from RIA2.
Primers were specifically
designed for P. hypophthalmus. Four dinucleotide repeat loci
(CB4, CB7, CB12 and
CB13, Table 2) of the ten loci available for Tra catfish were
polymorphic and were
screened in the samples available here. PCR amplifications were
performed in 10 μl
volumes reaction mixtures containing 1 μl approximately 50 ng of
extracted P.
hypophthalmus DNA template, 1 μl of 10X reaction buffer [500 mM
KCl, 200 mM
Tris-HCl (pH 8.4)], 1.5 mM MgCl2, 2.5 mM of each DNTP, 5 pM of
each primer, 0.5
units of Taq DNA polymerase (Promega, Madison, WI). Thermal
cycling was carried
out as follows: initial denaturation at 95oC for 4 min, followed
by 30 cycles consisting
of 30 sec denaturising at 95oC, 30 sec annealing at the
optimized annealing
temperature (see Table 2), 30 sec extension at 72oC, with a
final extension of 10
min at 72oC.
-
23
Table 2: Primer sequence details of the four microsatellite loci
screened
Locus Sequence Primer Annealing
Temp (oC) Forward (5’…………………………→3’) Reverse (5’……………………→3’)
CB4 CCA CAT CCT TAT CAC CCT GAA C ACA ATA CAG AGA AAT CCC CAA GG
55
CB7 GAA CAT CCA CAA ACA CAT CAC AC ACT TTC CCG GAG TAA TCG TTG
55
CB12 GCG ATA GAG ACA GAG AGT CAT GG ATC TGG GTC AAA ATG ATT GGA
AC 55
CB13 GTG TGT CAA GTT GGG ATC ATG G CTC CAT TTA CAG ACC ATC CGT
AG 55
2.6. Data analysis PCR amplified products were screened in
acrylamide gels using a Gel-Scan-3000
(Corbet Research) genotyper. Genotypes were scored using One
D-scan version
2.05 software (Scanalytics, Inc., 1998). Data were then stored
in MSExcel format
(2003). For the pre-data analysis step, allelic data were
checked for presence of null
alleles (signified by an excess of homozygotes), large allele
drop out (preferential
amplification of small alleles) or incorrect scoring due to
stutter bands (created by
slippage during PCR extension) using MICROCHECKER software
version 2.2.3
(Van Oosterhout et al. 2004). Microsatellite polymorphisms were
quantified by
assessing genetic diversity parameters and how diversity was
partitioned within and
among samples: observed (Ho) and expected (He) heterozygosity;
inbreeding
coefficients (FIS) and genetic differentiation among samples
(FST), using ARLEQUIN
v3.1 software (Schneider et al., 2000). Statistical significance
of F statistics was
determined using a non-parametric permutation process
incorporating 100
iterations. Allelic richness (An) was estimated using FSTAT
v2.9.3 (Goudet 1995).
Exact P-values that test for conformity of genotypes to
Hardy–Weinberg proportions
and linkage equilibrium were estimated using a Markov chain
method (1000
dememorization steps, 1000 batches, 1000 iterations per batch)
using ARLEQUIN.
Estimates of levels of genetic variation in three generations of
Tra catfish used in
the selective breeding program at RIA2 and pedigree analysis of
hatchery juveniles
were undertaken using CERVUS v3.0 (Kalinowski 2007) software
employing
10000 cycles. In all analyses, levels of significance for
multiple tests were corrected
using Bonferroni adjustment (Rice, 1989). With exact P value for
all experimental
tests set at α = 0.05; after Bonferroni, α (Bonf) = 0.05/ number
of tests.
http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6T4D-4SGKB98-3&_user=62921&_coverDate=08%2F01%2F2008&_rdoc=1&_fmt=full&_orig=search&_cdi=4972&_sort=d&_docanchor=&view=c&_searchStrId=1173002344&_rerunOrigin=scholar.google&_acct=C000005418&_version=1&_urlVersion=0&_userid=62921&md5=d2e514892737afdee954b175a7408245#bib33
-
24
Effective population size (Ne) of hatchery stocks from three
rivers were estimated
using two methods: i) estimation based on the number of males
and females in the
brood stock (Hartl and Clark, 1997) and ii) number of males and
females that
actually contributed to offspring, based on results from the
pedigree analysis
(parentage assignment) using CERVUS v3.0 software. Assessment of
genetic
correlations between genetic markers and production trait (%
fillet yield) were
assessed using GENECLASS v2.0. software (Piry et al. 2004).
Representative
examples of microsatellite Gelscan images are presented in
Figures 2 a-d.
Figures 2: Gelscan images of genetic diversity in RIA2 34 brood
fish samples (2a)
and from 3 private hatcheries (n=31) (2b) at locus CB 12;
gelscan
images of allelic diversity in high fillet yield individuals
(2c) and of low
fillet yield individuals (n= 24) (2d) at locus CB 12.
2a 2b
2c 2d
http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6T4D-4SGKB98-3&_user=62921&_coverDate=08%2F01%2F2008&_rdoc=1&_fmt=full&_orig=search&_cdi=4972&_sort=d&_docanchor=&view=c&_searchStrId=1173002344&_rerunOrigin=scholar.google&_acct=C000005418&_version=1&_urlVersion=0&_userid=62921&md5=d2e514892737afdee954b175a7408245#bib14
-
25
Chapter 3. RESULTS
3.1. Part A - Characterisation of genetic variation in cultured
and wild
populations of Tra catfish
The pairwise linkage disequilibrium results from the sampled
populations are
presented in Appendix 1. Initial analysis of the raw data
assessed presence of null
alleles, linkage disequilibrium and sample conformation to Hardy
Weinberg
Equilibrium (HWE). The four microsatellite loci (CB4, CB7, CB12
and CB13)
screened in the Tra catfish samples showed amplified fragments
that ranged in size
from 195 to 277bp. Individual sample amplification success
ranged from 88 to 95%
across the four loci with observed number of alleles per locus
ranging from 2 (Locus
CB7) to 10 (Locus CB12) MICROCHECKER analysis indicated that
null alleles were
present in high frequency at two loci; CB4 and CB 12 (>50%)
(Table 3).
Table 3: The potential for null alleles for each locus by sample
detected using
MICRO-CHECKER.
locus CB4 CB7 CB12 CB13
population
Wild no no yes no
TDR no no yes yes
RIA2 yes no yes yes
Fillet yes yes no yes
Hau yes no yes no
Nam yes no yes yes
Khanh yes no yes yes
Average expected number of homozygotes at locus CB4 and CB12
were 14.5 and
7.5, respectively while average observed homozygotes at these
loci were 24.8 and
16.7, respectively. In most instances, visualization of gel
images of the two loci
-
26
suggested that null alleles were a likely reason for higher than
expected
homozygotes numbers at the two loci in most sampled
populations.
3.1.1 Pair wise linkage disequilibrium
Pair wise linkage disequilibrium analysis using an exact P value
(α = 0.05), after
Bonferroni correlation (α Bonf = 0.05/75 = 0.0007) showed no
evidence for
significant linkage disequilibrium among the four loci (P<
0.0007). This result
indicates that the four loci screened in the Tra samples
examined here provided
independent assessments of genetic diversity in the sample
populations. A
complete statistical summary of the linkage disequilibrium
results are presented in
table 4.
3.1.2. Conformation to HWE results
Table 4 presents expected and observed heterozygosity estimates
and exact P
values for Hardy Weinberg Equilibrium (HWE) tests (α = 0.05),
after Bonferroni
correction (α Bonf = 0.05/60 = 0.0008). A substantial number of
tests revealed
significant deviations from HWE. In most cases this was in the
form of heterozygote
deficiency, however several instances of heterozygote excess
were also observed.
While these data could be used to infer a serious problem with
null alleles, there is
strong evidence from the different broodstock samples and wild
population sample
that this is unlikely to be the case. Out of the 16 HWE tests
for H Brood, N Brood, K
Brood and the wild sample, only two indicated heterozygote
deficiency. It is more
likely that the result seen here for the offspring reflect
non-random mating in the
broodstock (a function of the breeding protocols used in the
hatcheries), differential
contribution of breeders and/or differential survival of fry
from particular crosses.
(Tave 1994).
-
27
Table 4: Observed and expected heterozygosities (Obs. And Exp,
respectively),
probability value (P-value) and standard deviation (sd).
Significant
deviations from HWE indicated as heterozygote deficiency
(def),
heterozygote excess (excess) or not significant (ns) after
Bonferroni
correction. Pop/Locus Obs.Het Exp.Het P-value s.d Dev. from
HWE
H1
CB4 0.52083 0.72818 0.00006 0.00002 def
CB7 0.00000 0.60929 0.00000 0.00000 def
CB12 0.88542 0.67828 0.00000 0.00000 excess
CB13 0.67708 0.67294 0.00000 0.00000 excess
H2
CB4 0.71875 0.65211 0.09747 0.00079 ns
CB7 0.62500 0.52285 0.06969 0.00080 ns
CB12 0.42708 0.53916 0.00000 0.00000 def
CB13 0.72917 0.70370 0.14440 0.00122 ns
H Brood
CB4 0.50000 0.78947 0.11528 0.00082 ns
CB7 0.60000 0.51053 0.43256 0.00140 ns
CB12 0.50000 0.84737 0.01231 0.00031 ns
CB13 0.80000 0.78947 0.17228 0.00145 ns
N1
CB4 0.25532 0.65588 0.00000 0.00000 def
CB7 0.54255 0.58078 0.00000 0.00000 def
CB12 0.51685 0.68152 0.00000 0.00000 def
CB13 0.75532 0.72193 0.00000 0.00000 excess
N2
CB4 0.62637 0.67245 0.00386 0.00019 def
CB7 0.63441 0.58983 0.18024 0.00106 ns
CB12 0.58889 0.79963 0.00000 0.00000 def
CB13 0.74468 0.68819 0.00000 0.00000 excess
N Brood
-
28
Pop/Locus Obs.Het Exp.Het P-value s.d Dev. from HWE
CB4 0.50000 0.66842 0.26167 0.00124 ns
CB7 0.70000 0.59474 0.49094 0.00151 ns
CB12 0.60000 0.62105 0.19733 0.00088 ns
CB13 0.80000 0.60526 0.81148 0.00100 ns
K1
CB4 0.63441 0.64940 0.79725 0.00139 ns
CB7 0.62766 0.67306 0.11780 0.00085 ns
CB12 0.57778 0.79081 0.00000 0.00000 def
CB13 0.68085 0.61230 0.03387 0.00064 ns
K2
CB4 0.21277 0.61617 0.00000 0.00000 def
CB7 0.65625 0.62778 0.79531 0.00130 ns
CB12 0.35065 0.79611 0.00000 0.00000 def
CB13 0.69565 0.78326 0.00000 0.00000 def
K Brood
CB4 0.90909 0.65801 0.13736 0.00089 ns
CB7 0.63636 0.49784 1.00000 0.00000 ns
CB12 0.54545 0.73593 0.07103 0.00080 ns
CB13 0.90909 0.71429 0.56921 0.00135 ns
Wild
CB4 0.25926 0.47799 0.00151 0.00011 ns
CB7 0.59259 0.64570 0.28614 0.00140 ns
CB12 0.22222 0.72816 0.00000 0.00000 def
CB13 0.66667 0.81551 0.00769 0.00023 ns
TDR
CB4 0.57143 0.70711 0.01556 0.00037 ns
CB7 0.54545 0.59953 0.12716 0.00121 ns
CB12 0.50000 0.85829 0.00000 0.00000 def
CB13 0.55556 0.79526 0.00000 0.00000 def
Ria2
-
29
Pop/Locus Obs.Het Exp.Het P-value s.d Dev. from HWE
CB4 0.14706 0.64311 0.00000 0.00000 def
CB7 0.47059 0.58824 0.00376 0.00015 ns
CB12 0.43750 0.66964 0.00358 0.00016 ns
CB13 0.35294 0.77217 0.00000 0.00000 def
High
CB4 0.39535 0.57346 0.00396 0.00020 ns
CB7 0.52174 0.71261 0.00402 0.00019 ns
CB12 0.64444 0.87241 0.00000 0.00000 def
CB13 0.64103 0.76190 0.00060 0.00007 def
Low
CB4 0.27083 0.63136 0.00000 0.00000 def
CB7 0.66667 0.64320 0.97791 0.00045 ns
CB12 0.65909 0.76959 0.04750 0.00026 ns
CB13 0.48936 0.78609 0.00000 0.00000 def
Random
CB4 0.08696 0.53816 0.00000 0.00000 def
CB7 0.43478 0.61836 0.04401 0.00057 ns
CB12 0.52381 0.83624 0.00000 0.00000 def
CB13 0.34783 0.76715 0.00000 0.00000 def
-
30
Table 5: Microsatellite polymorphism in 7 sample populations of
wild and cultured Tra catfish populations.
Group/ Locus
A An
CB4 P A An
CB7 P A An
CB12 P A An
CB13 P Average gene
diversity Average
number of alleles
Hau Brood (n=10)
5 5.00 0.13 3 3.00 0.42 7 7.00 0.01 6 6.00 0.17 0.73 +/- 0.45
5.30 +/- 1.48
Nam Brood (n=10)
4 4.00 0.26 4 4.00 0.48 6 6.00 0.18 4 4.00 0.82 0.62 +/- 0.39
4.50 +/- 0.87
Khanh Brood (n=10)
4 3.91 0.13 3 2.99 1.00 6 5.72 0.07 4 3.91 0.56 0.65 +/- 0.40
4.25 +/- 1.09
Wild 1
(n=20)
3 3.54 0.01 4 3.36 0.65 6 4.64 0.00 6 5.68 0.01 0.64 +/- 0.38
4.75 +/- 1.29
Wild 2
(n=7)
4 3.54 0.09 2 3.36 0.44 3 4.64 0.02 4 5.68 0.46 0.63 +/- 0.40
3.25 +/- 0.83
Founder
(TDR n=45)
3 3.75 0.02 4 2.99 0.12 10 7.58 0.00 6 5.55 0.00 0.68 +/- 0.47
5.75 +/- 2.68
RIA2 Brood (n=34)
3 2.99 0.00 6 4.02 0.00 6 4.53 0.00 5 4.74 0.00 0.67 +/- 0.42
5.00 +/- 1.23
-
31
3.1.3 AMOVA analysis of hierarchical differentiation within and
among
populations
ARLQUIN 3.1 software was used to assess hierarchical
differentiation among
sample populations at two levels; within and among populations
employing analysis
of molecular variance (AMOVA). AMOVA allows estimation of the
statistical
significance of FST values between sample pairs, i.e. the
significance of population
differentiation. The following settings; 100 permutations for
significance with 10000
steps were employed in the Markov chain.
Table 6: The statistical significance of FST values of
population differentiation
Source of variation d.f Percentage of variation
Among samples 6 7.41
Within samples 267 92.59
FST = 0.0741, P < 0.0000 +/- 0.0000
The results show that the majority of variation present was
evident within
populations (92.6 %) while only 7.4% variation was evident among
populations.
Variation among populations however, was highly significant
(P
-
32
Table 7: Statistical significance of FST values (the
significance of population) among
sample pairs. P values that were significant after Bonferroni
correction (α
(Bonf) = 0.05/number of test = 0.05/ 21 = 0.002) are
highlighted
P values →
FST ↓ Hau
Brood
Nam
Brood
Khanh
Brood
Wild Wild 2 TDR RIA2
Hau Brood - 0.000 0.0270 0.000 0.0270 0.1351 0.0811
Nam Brood 0.0668 - 0.000 0.0090 0.000 0.000 0.000
Khanh Brood 0.0477 0.1009 - 0.000 0.0090 0.000 0.000
Wild 0.1098 0.0623 0.1404 - 0.000 0.000 0.000
Wild 2 0.7060 0.1706 0.0599 0.1184 - 0.0090 0.0180
TDR 0.0236 0.0819 0.0557 0.0951 0.0776 - 0.0360
RIA2 0.0406 0.1435 0.0834 0.1232 0.0770 0.0212 -
Genetic variation among parents representing the 6 populations
(average gene
diversity ranged from 0.62 – 0.73 and average number of alleles
per locus ranged
from 4.25 – 5.75). Allelic richness (An) across the sampled
populations was highest
at the CB12 locus (10 alleles) and lowest at the CB7 locus (2
alleles) (Table 5).
3.1.4. Genetic characterization of sampled Tra catfish culture
stocks
Microsatellite polymorphism was quantified by estimating gene
diversity within
samples (FIS) and between samples (FST), observed (Ho) and
expected (He)
heterozygosity, and estimating level of differentiation among
stocks. Seven
populations were available for comparisons of genetic
differentiation among sample
populations (Table 7). Brood stock samples were available from 3
private hatcheries
namely: the Hau and Khanh brood stocks and Nam hatcheries of
which the Nam
hatchery had developed the first domesticated brood stock in the
Mekong Delta
region. The TDR as sourced from wild fish and the first
domesticated generation
from the TDR brood fish at RIA2 were used for the selective
breeding program. Wild
1 and Wild 2 were wild caught individuals collected from the
Tien and Hau River as
a wild reference for comparison of genetic diversity levels in
culture stocks.
-
33
Parameters for genetic variation (Table 5) in 6 sample
populations of Tra catfish in
culture and wild are presented as mean number of alleles per
locus (A), allelic
richness (An), observed heterozygosity (Ho), expected
heterozygosity (He) and
average gene diversity. For wild Tra catfish, these values
showed A = 3.25–4.75;
allelic richness, An = 3.3–5.6; observed heterozygosity, Ho =
0.15–0.71; expected
heterozygosity, He = 0.3–0.82 and average gene diversity = 0.64,
and hatchery
samples showed A = 4.25–5.75; An = 2.99–7.58; Ho = 0.14–0.91; He
= 0.49–0.82.
3.2. Part B - Assessment of the relative contribution by
breeders to fry
cohorts in three private hatcheries in the Mekong Delta
The majority of Tra brood stock examined in the current study
was sampled from the
main Mekong River in Dong Thap Province (RIA2 founder and 3
private hatcheries).
Dong Thap Province in the Mekong River Delta is recognised as
the premier region
(An Giang, Can Tho and Vinh Long and Hau Giang) for Tra catfish
culture in
Vietnam. P. hypophthalmus culture has become a major industry in
the south of
Vietnam because there is abundant supply of fresh water
available year-round. In
2006, this province produced 300,000 MT of catfish that
contributed 37.5% to the
total production of catfish from aquaculture in Vietnam. The
industry in this province
also employs approximately 40,000 people of the 1.6 million
people who live in the
region (Phuong et al. 2007). In 2007, 87 catfish hatcheries were
operational in Dong
Thap Province, producing more than 4.4 billion fry per year
constituting the majority
of catfish seed supplied to the industry across the whole Mekong
Delta.
-
34
Figure 3: Map of Mekong Delta identifying the main areas where
Tra catfish are
cultured (grey colour)
The current experiment was designed to estimate the effective
population size (Ne)
of fish in a closed system and follow them from spawning to
hatching. For example,
at the Hau hatchery for the first spawn, eggs from 8 female fish
and sperm from 2
males were pooled and hatched as a single composite cohort. On
the first day, 100
fry were sampled randomly followed by a second sample of fry (n
= 100) on day two
before remaining fry were sold to farmers. The design employed
to estimate Ne had
to be suitable to fit into commercial hatchery practices
employed on farms while
allowing molecular assessment and analysis of the relative
contributions by parents
that potentially contributed to the fry cohorts.
3.2.1. Estimation based on the number of males and females in
the
brood stock
According to Hartl and Clark (1997), to ensure the highest Ne
during spawning,
under ideal conditions all brood stock should contribute equally
to offspring.
However, Ne will be substantially less than N if there is
unequal representation of
the sexes and Ne will be impacted more by the rare sex. A simple
statistic for
estimating Ne is where:
http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6T4D-4SGKB98-3&_user=62921&_coverDate=08%2F01%2F2008&_rdoc=1&_fmt=full&_orig=search&_cdi=4972&_sort=d&_docanchor=&view=c&_searchStrId=1173002344&_rerunOrigin=scholar.google&_acct=C000005418&_version=1&_urlVersion=0&_userid=62921&md5=d2e514892737afdee954b175a7408245#bib14
-
35
with Nm and Nf being the numbers of male and female brood
stock,
respectively.Table 8 provides estimates of Ne for the three
private hatcheries using
this method. While the Ne estimates are all less than the total
numbers of brood
stock in each case, there is no insight into the relative
contributions of each breeder.
Unequal contribution will lead to a further decline in
approximation of the true Ne
and therefore needs to be assessed in order to provide a more
realistic estimate.
Table 8: Estimation of Ne in three private hatcheries based on
the number