Transmission of the RNAi effect to the F1 progeny of injected animals has been shown for many genes including zygotic ones (
unc-22,
unc-54 ) and maternal ones (
apx-1,
mom-2,
pos-1 etc.). When interference is applied to zygotic genes like
unc-22, transmission of the effect to the F1 generation is obvious as all animals have the twitching phenotype. In the case of interference with essential maternal genes most of the F1 progeny of the injected animal are dead, but some escape interference and mature to produce 100% dead progeny themselves. Those are considered to be the carriers of the effect . We observed previously that hermaphrodites mated before the injection can produce F1 carrier-males which transmit the effect to their daughters if the males are mated with non-injected hermaphrodites. Thus, the effect of RNA interference is seen in the F2 progeny of the injected animal in this case and is transmitted with the sperm of F1 males. We wanted to know if this inheritance might depend on dominant gene-linked imprinting or a chromatin effect present in the sperm of the carrier. To test this idea we injected mated hermaphrodites bearing a deficiency covering the
pos-1 and
unc-42 loci with
pos-1-specific double stranded RNA (dsRNA). We then asked if sperm from F1 progeny males carrying the
pos-1 deficiency chromosome were able to transmit the RNAi effect. We crossed F1 males with
dpy-11 unc-42 uninjected hermaphrodites and identified progeny carrying parental deficiency chromosome covering
pos-1 and
unc-42 by the unc-phenotype. Those F2 animals matured to form hermaphrodites that produce 100% dead eggs with
pos-1 mutant phenotype. Thus, expression of the
pos-1 gene inherited from the uninjected mother in the F2 animals was effected by the RNAi element carried by the paternal sperm bearing the
pos-1 deficiency. This result clearly indicates that the targeted locus is not needed for the transmission of the interference and therefore a dominantly acting extrachromosomal element appears to mediate this process. We are currently conducting experiments to identify the RNAi interfering element in affected F1 animals. The inheritance of the interference described above provides some insight into the mode of RNAi effect on gene expression. When males carrying
pos-1 RNA interfering element are crossed with non-affected hermaphrodites both maternal
pos-1 mRNA and the paternal RNAi element co-exist in the zygote but expression of
pos-1 is not effected as live progeny arise from the zygote. The effect of paternally derived RNAi takes place later, probably during oocyte development in the progeny animals, and makes them produce all dead eggs with
pos-1 mutant phenotype. This might indicate that RNA interference occurs not on the translational level but effects transcription or processing steps of gene expression. The above interpretation of the effect of RNAi on gene expression is consistent with our studies of the endogenous mRNA levels in the affected animals. We detected reduced levels of
pos-1 mRNA in injected animals and their F1 progeny by RT-PCR. Also, using in situ hybridization we detected reduction in the endogenous mRNA level in F1 progeny of animals injected with
unc-22 specific dsRNA. To ask whether RNAi directed to different sequences of the same gene have different effects we compared the effects of promoter-proximal and promoter-distal sequences of
emb-9 dsRNA in producing
emb-9 embryonic lethal mutant phenotype. Injection of the dsRNA specific to the proximal region of the gene caused the phenotype closely matching
emb-9 while dsRNA specific to the distal region caused no visible phenotype.