Death of mouse embryos that lack a functional gene for glucose phosphate isomerase.

A null allele of the Gpi-1s structural gene, that encodes glucose phosphate isomerase (GPI-1; E.C. 5.3.1.9), arose in a mutation experiment and was designated Gpi-1sa-m1H. The viability of homozygotes has been investigated. No offspring homozygous for the null allele were produced by intercrossing two heterozygotes, so the homozygous condition was presumed to be embryonic lethal. Embryos were produced by crossing Gpi-1sa/null heterozygous females and Gpi-1sb/null heterozygous males. Homozygous null embryos were identified at different stages of development by electrophoresis and staining either for GPI-1 alone or GPI-1 plus phosphoglycerate kinase (PGK) activity. At 6 1/2 and 7 1/2 days post coitum homozygous null embryos were present at approximately the expected 25% frequency (37/165; 22.4% overall) although at 7 1/2 days the homozygous null embryos tended to be small. By 8 1/2 days most homozygous null embryos were developmentally retarded and had not developed significantly further than at 7 1/2 days; some were dead or dying. By 9 1/2 days the homozygous null conceptus was characterised by a small implantation site that contained trophoblast and often a small amount of extraembryonic membrane. Surviving trophoblast tissue was also detectable at 10 1/2 days. Previous studies have shown that oocyte-coded GPI-1 persists only until 5 1/2 or 6 1/2 days. Survival of homozygous null embryos to 7 1/2 or 8 1/2 days and survival of certain extraembryonic tissue to 10 1/2 days suggests that the homozygous null condition may not be cell-lethal although it is certainly embryo-lethal. Mutant cells that are deficient in glycolysis may use the pentose phosphate shunt to bypass the block in glycolysis created by the deficiency of glucose phosphate isomerase, and/or might be rescued by the transport, from the maternal blood, of energy sources other than glucose (such as glutamine). Either strategy may only permit slow cell growth that would not be adequate to support normal embryogenesis. Transport of maternal nutrients would be more efficient to the trophoblast and extraembryonic membranes and this may help to explain why these tissues survive for longer than the embryo itself. The morphological similarity between homozygous nulls and androgenetic conceptuses, where the trophoblast also survives better than the embryo, is discussed.