Transcript PowerPoint Presentation - No Slide Title

Zebrafish, an alternative to the mouse for
genetic studies of vertebrate development.
Use quotation marks!!!!
Avoid the use of “demonstrate”
or “prove” when describing
research goals.
Zebrafish advantages:
Cheaper to keep in the lab relative to mouse (~ half the cost)
Large broods (usually hundreds, but up to 1000)
Can strip both sperm and eggs from adults and fertilize in vitro
Can get new eggs every few days
Rapid development ( fertilization – hatch 2 days; adult
morphological features can be scored in seven days)
Reproductively capable at 3-4 months
Accessible embryogenesis (external fert. Transparent eggs)
Disadvantages:
New organism - no experimental history
N=25
1.6 x109
Possibly tetraploid
Will not address questions specific to mammals
Development is
rapid
Embryos are
transparent
Embryos develop
outside the
mother
Cells can be
marked and
lineage tracing
can be done
Tissue transplants
are possible
Fish
Frog
“Heterozygosity in diploid
eukaryotes often makes
genetic studies cumbersome.”
G. Streisinger
Making haploids:
UV
sperm
eggs
Haploids develop normally
– up to a point
Making haploids:
UV
+/m
sperm
eggs
+
+
+
m
Making gynogenetic diploids:
UV
+/m
sperm
eggs
Early pressure (EP)
Heat shock(HS)
Late pressure (LPP)
Meiosis II
Mitosis (1st cleavage)
+/+
m/m
+/+
m/m
Kimmel (1989)TIG 5:283
Problems with screening via haploids:
1. mutations that affect processes that are defective in
haploids will not be detected.
2. mutations are difficult to recover (?? Loss of founder
fish??)
Problems screening with diploids:
1. high background of abnormal embryos and
larvae (50-70% that don’t make normal-looking
embryos)
2. recombination masks mutations near the ends of
the chromosomes.
m
m
m
m
+
+
m
+
+
m
+
EP
+
(not a problem for HS or LP)
Nusslein-Volhard decides to throw her hat in the ring.
Decides that these problems with haploids and
gynogenetic diploids are serious barriers to systematic
screens for developmental mutants.
Needs to use another approach:
Using standard genetics requires:
Devising an efficient breeding scheme for producing
homozygous fish
Methods for cleanly culturing many lines of fish
An efficient mutagen
Lots of people
Lots of work
(Need ways for cloning the genes)
Why do they use mutagenized spermatogonia rather than
mutagenized mature sperm?
spermatogonium
Mature sperm
First mitosis
Mosaic embryo
Mature sperm
mature sperm
Pre-meiotic sperm(spermatogonia)
This fish also carries one
mutagenized genome
Cross scheme for
Haffter et al. screen
Background work (Mullins et al. (1994) Current Biology 4: 189)
Efficient system for maintaining large numbers of fish.
Efficient system for mutagenesis:
Need high frequency mutagen and easy means to deliver it
X-rays and gamma rays were used in the past.
How to test a new mutagen?
Specific locus test (4 loci):
ENU was best with 1-3 mutations/locus/1000 haploid genomes
Results of the Haffter et al. screen
(15)
(1.3)
Allele frequencies
Scheme for Driever screen (identical to that of Haffter et al.)
Driever W, Solnica-Krezel L, Schier AF, Neuhauss SC, Malicki J, Stemple DL, Stainier DY, Zwartkruis F,
Abdelilah S, Rangini Z, Belak J, Boggs C.(1996)A genetic screen for mutations affecting embryogenesis in
zebrafish. Development123:37-46.
Results of Driever et al screen:
2383 mutations identified (4264)
695 mutations kept (1557)
577 finished (894)
220 loci (372)
164 single alleles (222)
Did they achieve saturation?
No. Estimates are between 50% and 90% (favoring
the lower number). However, I don’t know the final
results based on complementation tests between the
two screens (still not done).
Using the estimates from the screen, they predict
about 2400 genes that can mutate to embryonic
lethality.
But they did find a boatload of mutants - enough to fill
an entire issue of Development with phenotypic
descriptions.
Limitations of the screens:
1) redundancy
2) maternal effect mutants only by accident
3) missed subtle phenotypes
4) observer bias
(30)*
Driever et al. (39/220 loci = 18%)
*Haffter et al. (24/327 loci = 7%)
(24)* evidence for bias in screening
How do you get the DNA corresponding to the mutant gene?
Cloning:
Map based approach
a) map gene
b) obtain overlapping DNA clones of region
c) high resolution mapping (many progeny)
d) identify gene among small set of candidates
1. Rescue of mutant phenotype
2. Identify mutant allele by sequencing
Candidate gene approach
a) use phenotype to predict gene known
from other organisms
b) find that gene in zebrafish (use ESTs and
genetic map)
c) test as in d above
Total of 145 genes cloned as of 2006 (Amsterdam & Hopkins (2006) Trends in Gen 22: 473 )
Insertional mutagenesis provides
a more rapid way to clone genes.
Identification of 315 genes essential for early
zebrafish development
Adam Amsterdam, Robert M. Nissen, Zhaoxia Sun, Eric C. Swindell, Sarah Farrington, and Nancy Hopkins*
Center for Cancer Research and Department of Biology, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, MA 02139
Contributed by Nancy Hopkins, June 8, 2004
This contribution is part of the special series of Inaugural Articles by members of the National Academy of Sciences elected on April 20, 2004.
1. Mutagenize by injecting retrovirus into fish embryos
2. Virus inserts randomly into genome creating some
gene disruptions.
3. Screen for disruptions of essential genes (lethal
mutations)
4. Outcross the mutations to eliminate extraneous
inserts
5. Use inverse PCR to amplify flanking DNA
6. If no exon, then sequential inverse PCR until an exon
is hit.
Hopkins screen:
mutagenesis
Amsterdam et al. (1999) Genes & Development 13:2713
Amsterdam et al. (2004) PNAS 101: 1293
Amsterdam & Hopkins (2006) Trends in Gen 22: 473 (review)
Inject virus into about 250,000 embyros to
generate about 36,000 founder fish.
Pair-mate founders to produce F1 with
many insertions
Raise 10,000 F1 families and test each
family by PCR to identify fish with > 3
unique inserts.
Interbreed F1’s with inserts to make 10,000
F2 families
Proceed to screen F3 for phenotypes
Results:
One BIG drawback:
Insertional mutagenesis is much less efficient than ENU.
Requires 15x the person hours
Need to use both approaches.
Other insertional agents are being tested
Transposons (Tol2 and Sleeping Beauty)
The SIGNIFICANT plus:
Cloning mutants takes about two weeks!!
ALL mutants affecting embryogenesis are kept (even
pleiotropic ones), thus eliminating biases about
which mutants will be most informative.
Conclusions:
They have identified about 25% of the zygotically expressed
genes that are essential for embryonic development and therefore
they estimate that there are about 1,600 of those.
Evidence: They have insertions in 5/20 tRNA synthase genes, 26
of the 79 ribosomal protein genes, 23/97 genes identified by
chemical mutation and cloned. Also estimates based on the
Poisson distribution give similar numbers.
Haffter paper - 2400 essential genes
Lots of cool genes!
Another potential problem for zebrafish genetics:
Genome duplication.
Evidence from studies of gene families, like the Hox
genes and other pattern formation genes that have
been studied in mammals show that there are twice
as many of these genes as are expected from
mammalian studies.
Is this bad? Maybe. However, in the few cases that
have been examined, the duplicated genes have
different expression patterns suggesting that they
will be mutable.
Two orthologues of the mammalian gene Nodal,
squint and cyclops.
Overlapping but not identical expression
patterns
Differences in single mutant phenotypes
Double mutant has new defect
Conclusion: partial redundancy
Conclusion:
Although forward genetic screens are labor intensive,
they are identifying important developmental regulators.
Insertional mutagenesis makes cloning easier, and
although the mutation rate is low, persistent effort has
paid off for about 25% of the essential genes.
What about the maternal
contribution to development?
Two types of screens:
1) Gynogenetic diploids
2) Four generation screen
Why?
The screen is biased toward genes close to the centromere
+
+
m
Meiosis I
m
EP
m
m
Gynogenetic
diploid
(m/m)
Polar body
+
+
m
Meiosis I
+
m
m
Polar body
EP
Gynogenetic
diploid (m/+)
A pretty messy screen.
Problems:
Inbred lines are required for production of gynognetic diploids; inbred
lines are not as healthy as wild lines.
Because EP suppresses meiosis II, crossing over will result in
heterozygous gynogentic diploids. The frequency of crossing over
increases away from the centromere, so genes that are far away from
the centromere will not get sampled.
Gynogenesis also results in high frequencies of males (Why?)
Alternative approach: Bite the bullet and do a brute force
traditional screen. Mary Mullins, Penn
Pool and
subject to PCR
analysis of
polymorphic
markers;
markers close
to the mutant
will be
homozyogous
Results:
605* genomes screened (400 families)
68 maternal effect mutants
21 were early defects
15 “representative mutants” were propagated
(not clear why the others were not)
7 were able to be mapped by bulk segregant
analysis
at least 12 of the 15 are in separate genes
47 were late defects (at or after MBT)
36 led to massive cell death
13 “representative mutants” were propagated
*This is not 800 because of sampling error.
How do you propagate a maternal effect mutant in the
absence of balancer chromosomes?
Flanking DNA markers.
M1a mutant M2a
M1b +
M2b
M1 and M2 have two alleles in the two starting strains.
Take tail snips for each animal in the family:
m/+
M1a
M2b
M2a
M1b
m/m
+/+
??
Take home lessons:
Even though it is a lot of work, it is possible
to do a standard forward genetic screen to
identify maternal effect mutants.
Most common phenotype is general cell
death - probably general cell viability
functions (housekeeping).
Can obtain Mel mutants with defects before
or after zygotic transcription begins.
Can obtain paternal effect mutants - sperm
has some early contribution.
It is difficult, but not impossible to obtain
mutations affecting the development of zebrafish.