Production of Androgenetic Zebrafish ( Danio rerio ) Graham

To help investigate the evolutionary origin of the imprinting (parent-of-origin mono-allelic expression) of paternal genes observed in mammals, we constructed haploid and diploid androgenetic zebrafish (Danio r h o ) . Haploid androgenotes were produced by fertilizing eggs that had been X-ray irradiated to eliminate the maternal genome. Subsequent inhibition f the first mitotic division of haploid androgenotes by heat shock produced diploid androgenotes. The lack of inheritance of maternal-specific DNA markers (RAF'D and SSR) by putative diploid and haploid androgenotes confirmed the androgenetic origin of their genomes. Marker analysis was performed on 18 putative androgenotes (five diploids and 13 haploids) from six families. None of 157 maternal-specific RAF'D markers analyzed, some of which were apparently homozygous, were passed on to any of these putative androgenotes. A mean of 7.7 maternal-specific markers were assessed per family. The survival of androgenetic zebrafish suggests that if paternal imprinting occurs in zebrafish, it does not result in essential genes being inactivated when their expression is required for development. Production of haploid androgenotes can be used to determine the meiotic recombination rate in male zebrafish. Androgenesis may also provide useful information about the mechanism of sex determination in zebrafish. Z EBRAFISH (Danio rho , formerly known as Brachydanio rerio; MEYER et al. 1993) are an important model for studying vertebrate development and are amenable to genetic analysis (STREISINGER et al. 1981; KIMMEL 1989; BARINAGA 1990, 1994; NUSSLEIN-VOL HARD 1994; CONCORDET and INGHAM 1994; DRIEVER et al. 1994; KAHN 1994). Their use as a model organism for genetic analysis is facilitated by a linkage map of DNA markers (POSTLETHWAIT et al. 1994) and large scale screens for mutations are underway (DRIEVER et al. 1994; KAHN 1994; MULLINS et al. 1994). To investigate imprinting in vertebrates and to help develop zebrafish as a genetic system, we constructed haploid and diploid androgenotes. Construction of individuals with uniparental inheritance can facilitate genetic analysis. Haploid and diploid gynogenotes have been produced by fertilizing zebrafish eggs with sperm irradiated to eliminate the paternal genome (STREISINGER et al. 1981; HORSTGEN-SCHWARK 1993). Haploid gynogenotes were used to produce a zebrafish linkage map based on rates of meiotic crossing over in oocytes (POSTLETHWAIT et al. 1994). Haploid gynogenotes complete embryogenesis and arrest as larvae. Thus, haploid embryos can be used for F1 mutant screening (KIMMEL 1989) : mutations in the stem cells of the maternal germ line are introduced by fertilization with mutagenized sperm or by mutagenesis of early embryos. Such screens require maintenance of considerably fewer progeny to recover an interesting recessive mutation from the maCorresponding author Graham E. CorleySmith, Institute of Molecular Biology and Biochemistry, Simon Fraser University, Burnaby, B.C. V5A 1S6, Canada. E-mail: [email protected] Genetics 142: 1265-1276 (April, 1996) ternal stock than conventional diploid screens, which require production of an F3 generation (KAHN 1994; MULLINS et al. 1994). Diploid gynogenotes can be produced by inhibiting extrusion of the second polar body or by inhibiting the first mitotic division of the gynogenote (STREISINGER et al. 1981; HORSTGEN-SCHWARK 1993). In the latter instance, the progeny are homozygous and clonal lines of gynogenetic zebrafish have been produced (STREISINGER et al. 1981; KIMMEL 1989). Androgenetic haploid progeny would result from fertilization of eggs that have been treated to eliminate the maternal genome. A method for production of androgenetic zebrafish has not been reported, but would facilitate determination of rates of meiotic recombination in males, mapping of male specific DNA markers and linkage groups (if any), and analyses of sex determination and genomic imprinting. Use of androgenetic haploids may have some usefulness for F1 mutant screens if interesting mutations can be reliably recovered from cryopreserved milt long after the screen was