Mutations in
dif-1 result in maternal-effect embryonic lethality. Embryos die after the completion of gastrulation with little or no tissue differentiation or morphogenesis, though markers for cell fate specification (i.e., CeMyoD and
lin-26 )are expressed normally. The development of
dif-1 mutant embryos through the end of gastrulation is indistinguishable from that of wild-type in cell division time and pattern. One allele (
dif-1 (
e2591))is cold sensitive viable (lives at 25, dies at 15); the temperature sensitive period is from the end of gastrulation (when
dif-1 mutants arrest) to the three fold stage, during which time the majority of morphogenesis occurs. A shift to the restrictive temperature from the three-fold stage onward reveals no additional requirement for
dif-1 activity. The Tsp and phenotypic analysis suggests
dif-1 product is needed for three hours of embryogenesis to complete the differentiation of all tissues. I cloned
dif-1 by transformation rescue. Cosmids T08G12 and RO5B1 have rescuing activity; this was narrowed to a 3.3kb fragment, which was sequenced. The sequence shows significant similarity to a family of mitochondrial transporters. This family contains the ATP/ADP transporter that puts ATP in the cytosol in exchange for ADP and other solute transporters (including the phosphate transporter, oxoglutarate/malate transporter, and the brown fat uncoupling protein). The family members are transmembrane proteins found in the inner mitochondrial membrane. The sequence similarity of these genes is low, but the structure is well conserved. The proteins consist of three tandemly repeated 100 amino acid sequences, each of which have certain conserved residues. Each repeat is predicted to have two membrane spanning regions. Comparison of
dif-1 to this family does not show a stronger similarity to one type than to another, suggesting it may be a transporter with a different solute specificity than those that have been sequenced. However, the location of this family of proteins in mitochondria suggests that
dif-1 is required for the function of mitochondria. One speculative possibility is that
dif-1 acts to promote or maintain high levels of ATP during morphogenesis. A high ATP level might be involved in triggering the switch into morphogenesis, or morphogenesis may require more abundant ATP than other processes. John Walker here at the MRC and his colleagues are trying to determine the substrate specificity of
dif-1 using in vitro analyses.
dif-1 RNA is expressed at all stages of development and is also found in the adult soma. This is surprising given that the tsp extends for only three hours of embryogenesis. However, it is possible that all of the alleles isolated thus far are hypomorphs and none is null (though none is stronger over a deficiency). Also, there may be a specific or greater requirement for
dif-1 activity during morphogenesis, or there may be another gene with redundant activity at other times of development. I am currently sequencing
dif-1 mutations to try to determine whether any are null.