Production of transgenic tilapia homozygous for a humanized insulin gene

Abstract

Recently we developed transgenic tilapia carrying and expressing a ‘‘humanized’’ tilapia insulin gene (Pohajdak et al. 2004). A mosaic male founder (Wright et al. 2008) was mated to multiple wild-type females and the F1 offspring were screened for the presence of the transgene using PCR. DNA from 13 random transgenic offspring and the founder were also analyzed by Southern hybridization. The genome of all tested fish contained three copies of the humanized gene, suggesting that all three copies were integrated in the same chromosomal locus. Using PCR analysis we determined that inserts are not linked in tandem repeats. Transgenes were expressionally active since [desThr30] human insulin was detected in serum and in clusters of b-cells in F1 offspring (Pohajdak et al. 2004). In the next round of breeding, we crossed two halfsibling transgenic F1’s and obtained 163 offspring. To distinguish homozygous, heterozygous and wildtype fish in this F2 population we estimated relative ratios of total insulin genes (tilapia ? humanized insulin genes) versus tilapia insulin genes using quantitative PCR, similar to our previous approach (Pohajdak et al. 2004). Amplifications were performed on a Rotor Gene 2000 real-time PCR machine using a QuantiFast SYBR Green PCR kit. All data were analyzed using a standard curve method. Considering three integrated transgenes are likely located on the same chromosomal locus, the homozygous: wild-type: heterozygous ratio in the F2 population was expected to be 1:1:2. Indeed, from 163 tested fish we found that 41 fish were homozygous, 39 were nontransgenic and 83 were heterozygous, resulting in an almost perfect Mendelian distribution. Homozygous fish were phenotypically indistinguishable from wild-type and heterozygous fish, suggesting that over-expression of humanized insulin has no obvious influences on fish health. The normal 1:1:2 distribution in the adult transgenic F2 population suggests that the mortality of the homozygous fish at early stages of development was approximately the same as in normal and heterozygous tilapia. Interestingly, even though insulin is a known stimulator of growth and differentiation in fish, we did not observe any morphological differences in the development of transgenic tilapia whose genome contained three or six additional insulin gene copies. Furthermore, crossbreeding of two homozygous siblings resulted in the production of a phenotypically-normal fry population with 100% homozygosity. Thus far, we have noted no O. Hrytsenko B. Pohajdak Department of Biology, Dalhousie University, Halifax, NS B3H 4J1, Canada