Zebrafish Breakout Group

Recommendations

The zebrafish is a powerful model system for the genetic analysis of vertebrate embryogenesis, organ development, and disease. Its unique power is its tractable, phenotype driven mutation screens and readily accessible transparent embryos. Because of its facile forward genetics, zebrafish accelerates gene discovery; because of its accessible embryos, it promotes deep understanding of gene function; because of its phylogenetic position, it informs mechanisms of genome conservation. More than 2000 mutants are currently available in more than 600 genes, and new phenotypic screens are uncovering more mutations in many laboratories.

The zebrafish system relies on the ability to clone and characterize mutant genes rapidly. Positional cloning and insertional mutagenesis have been successfully utilized to identify the molecular nature of mutations, but candidate gene approaches have been the most successful to-date. The infrastructure of the community and the recent Zebrafish Genome Initiative have added value to the system by shortening the time to gene isolation, and with the addition of a few more reagents and services, the cloning of mutations could become routine. New investigators are becoming interested in zebrafish and a mission of the current community is to make the system more easily accessible and tractable. The current RFA and implementation of the suggestions listed below for additional support will have significant impact on investigator-driven research in the field.

A. Current RFA

In 1998, the NIH issued a Request for Applications to improve genomic resources for zebrafish. Five projects were funded. These three year projects have begun to vastly increase the number of expressed sequence tags and the number of mapped expressed sequences and anonymous markers, and are providing a collection of chromosomal deletions that cover the entire genome. The funded projects are:

100,000 ESTs 5' and 3' sequences
10,000 ESTs mapped to RH panel
3,000 ESTs mapped meiotically
0.5 cM microsatellite map, 5000 microsatellites
Deletion panel to cover the genome

B. More Infrastructure Needed for Genomics

I. ESTs. A large collection of ESTs benefits all strategies for the molecular identification of mutations. For positional cloning, sample sequencing of clones in the critical interval can reveal sequence identity to ESTs. For insertional mutagenesis, tags isolated from genomic DNA adjacent to inserts can reveal overlap with ESTs, thereby reducing cDNA library screening. For the candidate approach, the greater the number of ESTs to serve as candidates, the greater the likelihood that a mutated gene is represented by a known EST. Knowing the genetic location and expression pattern further allows potential candidate genes to be ruled out rapidly, thereby facilitating cloning of mutations. Therefore, as high priority, our community needs more EST sequences, and more ESTs characterized by genetic location and expression pattern.

Make new, normalized cDNA libraries from multiple stages and tissues. ($150,000/year x 3 years).
Sequence 5' and 3' ends of 100,000 cDNAs. ($400,000/year x 3 years.)
Map the ESTs to compare to mutant map positions. ($160/EST X 5,000/year X 3 years = $2,400,000).
Determine the expression pattern of the ESTs for comparison to mutant phenotypes. $150,000/year X 3 years for 10,000 cDNAs. Make these ready for the Zebrafish Gene Expression Atlas.
Develop and make widely available microarrays of 10,000-20,000 ESTs. ($150,000/year X 3 years.)

II. Stock Center. The power of zebrafish is the collection of mutations isolated on the basis of mutant phenotypes. Currently, laboratories are unable to keep all the mutations they recover for lack of space and funds for maintenance. Requests for the distribution of mutant stocks burdens all laboratories that have generated substantial mutant collections. Most mutations that are not of immediate interest must be maintained as frozen sperm. This hinders research because it requires at least three months to obtain homozygous embryos starting from frozen sperm, and stocks adequate for experimentation can take up to 9 months from the thawing of sperm. Clearly a stock center for the preservation and distribution of mutant stocks is key to the rapid analysis of zebrafish mutants. NIH and State of Oregon funds have been obtained to support the construction of a stock center in Eugene and to partially fund the preservation and distribution of zebrafish stocks. More funds are required to maintain all of the current stocks, not to mention the many new mutations made by new, innovative and directed mutations screens.

Stock Center: Additional personnel needed. $250K/year additionally.

III. Data Base. Ready access of the entire community to genomic and developmental data is required to turn data into understanding. The zebrafish database ZFIN is positioned to become a tool for archiving and retrieving data on zebrafish. Links must be made between the zebrafish database and the database of other organisms. A substantial increase in database support is needed to turn ZFIN into the tool the community requires.

Data editors to annotate expression and phenotypic data from old and current papers. ($500,000/year X 3 years.)
Data base: Need bioinformaticists to expand capabilities to include conservation of syntenies, histological atlas, and Zebrafish Gene Expression Atlas. ($500,000/year X 3 years.)

IV. Mutant mapping. One of the first steps to the molecular characterization of a zebrafish mutation is its genetic mapping. This rules out many potential EST candidates, and "rules in" others, as well as providing an initial stage for positional cloning. It is important to map quickly mutations in each gene mutated in the two large screens, 600 mutations, by a mechanism that guarantees that mutant locations are rapidly made public. Only about 150 of these have been mapped to date.

Mapping mutations. ($2000/Mutation x 500 mutations = $1,000,000 over 3 years.)

V. Physical Map of Genome. DNA pooling methods and the huge number of anonymous polymorphisms available by microsatellites, RAPDs, and AFLPs permit the identification of DNA polymorphisms very closely linked to zebrafish mutations. Chromosome walks from those linked markers requires large insert DNA libraries, and is facilitated by a physical map of the genome. A physical map that includes telomeres furthermore provides reagents for genome sequencing.

BAC libraries, 200 kb clones: ($250,000 for 10-genome coverage.)
BACs contiged and anchored to the genetic map. ($1,000,000/year x 3 years.)
YAC telomere library. Knowing the end of the chromosomes will aid positional cloning of genes in these regions. ($150,000)

VI. Genome Sequencing. Complete knowledge of the sequence of the zebrafish genome would provide a tool to revolutionize our efforts to understand zebrafish biology.

Phase 1, start sequencing of the genome. $3M first year. Initially science driven as to regions to start sequencing, focussing on regions of highest interest and regions orthologous to those already sequenced in other important model systems such as human and pufferfish. This will define repeat structure of the genome and will provide information on the structure of intron-exon boundaries.

Phase 2. Ramping up to sequence the entire genome beginning in 2000 with targeted completion 2008.

VII. Technology development. Although zebrafish has tremendous advantages as a system for forward genetics, the development of several additional methodologies would further improve the system. Key among these needs is gene knockout technology and the construction of more transgenic zebrafish. $1.5m/y

ES cells
transgenesis to provide more GFP lines
Gal4 UAS system to investigate gene interactions
RNAi to eliminate gene functions

C. More Infrastructure Needed for Cell Biology

I. Imaging. The beauty of zebrafish lies in its optically clear embryos and their accessibility to experimental intervention. Some additional descriptive studies exploiting these attributes would significantly enhance our ability to understand mutant phenotypes. Chief among these would be a four dimensional fish, that is, a time-lapse reconstruction of development in real time. Substantial computer assisted imaging must accompany this so that different organ systems, tissues, and cell types can be highlighted and so that the images can be viewed from any orientation desired by the viewer. Furthermore, the gene expression patterns of the ESTs generated in item B1 above and other isolated genes must be visible on this 4D fish. Furthermore, the images must be retrievable from a database in any of a number of ways determined by the user (for example, the expression patterns of all loci that begin expression within 1 hour of the appearance of sonic hedgehog transcript in the floorplate).

4D Fish. X,Y,Z images in real time, including gene expression patterns
More fate maps for later stages. This would help in the interpretation of gene expression patterns.

II. Novel and improved phenotyping. Members of the community perceive that more effort needs to go into developing novel means for phenotyping zebrafish, for example by electrophysiological methods or to assay behaviors, or to detect changes in ions such as calcium. These goals are probably best achieved by RO1's.

III. Protein expression and biochemistry. Although the community feels at this point that infrastructure for genomics should take priority to infrastructure for proteomics, the construction of monoclonal antibody panels from various stages and tissues, and the establishment of tissue-specific cell lines would be useful.

Monoclonal antibodies for various stages
Cell lines from various tissues

D. Budget

Here is listed the budget for the genomics items in part B above. The cell biology items are also needed, but we were less able to put dollar figures on them. To accomplish the aims in part B would cost about $20 million spent over a three year period.

	Items	Per Year	Years	Total
		($ in thousands)		($ in thousands)

ESTs
	cDNA libraries	150	3	450
	End sequences	400	3	1,200
	Map ESTs	800	3	2,400
	Expression	150	3	450
	Microarrays	150	3	450

Stock Center		250	3	750

Data Base
	Data editors	500	3	1,500
	Bioinformaticists	500	3	1,500

Mutant Mapping		333	3	1000

Physical Map
	BAC libraries	250	1	250
	Contigs	1,000	3	3,000
	Telomeric YACs	150	1	150

Genome Sequence
	Phase 1	3,000	1	3,000
	Phase 2	1,500	3	4,500

Total		9,133		20,600